<%BANNER%>

Responses of Plant and Small Mammal Communities to Prescribed Burning in Cedar Key Scrub State Reserve

Permanent Link: http://ufdc.ufl.edu/UFE0022089/00001

Material Information

Title: Responses of Plant and Small Mammal Communities to Prescribed Burning in Cedar Key Scrub State Reserve
Physical Description: 1 online resource (249 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: burning, habitat, mammals, prescribed, scrub, small
Wildlife Ecology and Conservation -- Dissertations, Academic -- UF
Genre: Wildlife Ecology and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Although prescribed burning is an important management tool for ecosystem restoration in Cedar Key Scrub State Reserve, this is the first study that analyzes the effect of prescribed burning on plants and small mammals. In addition, this is the first research carried out on plant community response to prescribed fire in coastal scrub on the west side of Florida, and the 12th study about the effects of prescribed burning on small mammals in Florida. The main objectives were to determine: (a) if there were structural and compositional changes in the plant community after prescribed burning, (b) if small mammals used wetlands as temporal refugia after prescribed fire; and (c) if prescribed burning had a negative effect on the survival of the small mammal species. The experimental design consisted of two treatment and two control sites that were sampled before and after burning from December 2003 to August 2006. Preburn vegetation samples were conducted one time in all sites, and postburn vegetation samples were carried out every three months for a 12 month period. Fifty quadrats (4 m2 each) per site were assessed in each sampling. Resprouting was the main way of surviving and recovering from fire by the majority of the species, and almost all of the dominant species reached preburn levels during the 12 months period. This fast recovery of the vegetation after burning has been reported in the literature but not in one year. The Detrended Correspondence Analysis showed that woody species had structural and compositional changes during the first three months postburn, but there were more compositional than structural changes after that. According to the Multi-response Permutation Procedure, the structural changes were significant; therefore, there were significant changes in absolute densities in treatment sites between pre- and 12 months postburn and between control values and 12 months postburn as a consequence of prescribed burning. A total of 29,340 trapping nights were completed in treatment and control sites. Each site had a grid (100 traps) and a wetland next to it with two transects (10 traps each). Mice were marked to monitor movements between scrub and the vegetation surrounding wetlands during four trapping sessions before and after prescribed burning. A total of 184 individuals of Sigmodon hispidus (cotton rat), Podomys floridanus (Florida mouse), Peromyscus gossypinus (cotton mouse), and Ochrotomys nuttalli (golden mouse) were monitored during this study. In treatment sites, mice were captured mainly in the scrub (75%) before burning, they used the vegetation surrounding wetlands as temporal refugia for 11 months after burning, and they returned to the scrub after that. In control sites, mice were captured mainly in the scrub (91%) during the study. MARK analysis was only carried out on S. hispidus and P. floridanus because of the small sample size obtained for the other two species. MARK indicated that fire did not have a negative effect on the survival of S. hispidus. I cannot state the same for P. floridanus because the ? parameter was not estimable. However, the data indicated that mice moved to wetlands and survived for 11 months. These results will provide guidance to managers in prescribed burning plans to establish a fire return interval according to the recuperation of the vegetation and to maintain viable populations of small mammals.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Tanner, George W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022089:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022089/00001

Material Information

Title: Responses of Plant and Small Mammal Communities to Prescribed Burning in Cedar Key Scrub State Reserve
Physical Description: 1 online resource (249 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: burning, habitat, mammals, prescribed, scrub, small
Wildlife Ecology and Conservation -- Dissertations, Academic -- UF
Genre: Wildlife Ecology and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Although prescribed burning is an important management tool for ecosystem restoration in Cedar Key Scrub State Reserve, this is the first study that analyzes the effect of prescribed burning on plants and small mammals. In addition, this is the first research carried out on plant community response to prescribed fire in coastal scrub on the west side of Florida, and the 12th study about the effects of prescribed burning on small mammals in Florida. The main objectives were to determine: (a) if there were structural and compositional changes in the plant community after prescribed burning, (b) if small mammals used wetlands as temporal refugia after prescribed fire; and (c) if prescribed burning had a negative effect on the survival of the small mammal species. The experimental design consisted of two treatment and two control sites that were sampled before and after burning from December 2003 to August 2006. Preburn vegetation samples were conducted one time in all sites, and postburn vegetation samples were carried out every three months for a 12 month period. Fifty quadrats (4 m2 each) per site were assessed in each sampling. Resprouting was the main way of surviving and recovering from fire by the majority of the species, and almost all of the dominant species reached preburn levels during the 12 months period. This fast recovery of the vegetation after burning has been reported in the literature but not in one year. The Detrended Correspondence Analysis showed that woody species had structural and compositional changes during the first three months postburn, but there were more compositional than structural changes after that. According to the Multi-response Permutation Procedure, the structural changes were significant; therefore, there were significant changes in absolute densities in treatment sites between pre- and 12 months postburn and between control values and 12 months postburn as a consequence of prescribed burning. A total of 29,340 trapping nights were completed in treatment and control sites. Each site had a grid (100 traps) and a wetland next to it with two transects (10 traps each). Mice were marked to monitor movements between scrub and the vegetation surrounding wetlands during four trapping sessions before and after prescribed burning. A total of 184 individuals of Sigmodon hispidus (cotton rat), Podomys floridanus (Florida mouse), Peromyscus gossypinus (cotton mouse), and Ochrotomys nuttalli (golden mouse) were monitored during this study. In treatment sites, mice were captured mainly in the scrub (75%) before burning, they used the vegetation surrounding wetlands as temporal refugia for 11 months after burning, and they returned to the scrub after that. In control sites, mice were captured mainly in the scrub (91%) during the study. MARK analysis was only carried out on S. hispidus and P. floridanus because of the small sample size obtained for the other two species. MARK indicated that fire did not have a negative effect on the survival of S. hispidus. I cannot state the same for P. floridanus because the ? parameter was not estimable. However, the data indicated that mice moved to wetlands and survived for 11 months. These results will provide guidance to managers in prescribed burning plans to establish a fire return interval according to the recuperation of the vegetation and to maintain viable populations of small mammals.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Tanner, George W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022089:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101204_AAAAAP INGEST_TIME 2010-12-04T09:45:08Z PACKAGE UFE0022089_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 58634 DFID F20101204_AAANKK ORIGIN DEPOSITOR PATH silvalugo_j_Page_177.pro GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
39c94075695ab4cd746b8f00c16be019
SHA-1
0ef6b4ed530a95cf6d22c75911f9e8e023fe4260
38853 F20101204_AAAOHD silvalugo_j_Page_030.QC.jpg
9609ce9a2e0745a81b13d016048720b0
529060734a13d4bdc32e3472189edba18d66364e
116937 F20101204_AAAMAG silvalugo_j_Page_159.jpg
d3dd1271bef5050cbe00cea4f8e62d3f
6d4f623732df2c60f77197e4ddf5c14bb4a9b281
3665 F20101204_AAANXV silvalugo_j_Page_049thm.jpg
7eab688493859ff9b0cb77d7c17bc213
c5e13b2b1f37cd73c28c5631d32dccb4c96a3c5b
1051912 F20101204_AAAMNR silvalugo_j_Page_033.jp2
6d6137b8e7a7fcced68af6d134fc2984
a55dc2444afe5390e4c8b88a7f2c94c01c4e4a21
54955 F20101204_AAANKL silvalugo_j_Page_178.pro
b9cd804161ebc5753c01d8bea69ce676
4718610e619fa29072d9512ae5d6031a3f8dc0cb
9704 F20101204_AAAOHE silvalugo_j_Page_030thm.jpg
d7544738e87969ce660223dde2da48e9
35d9e61bc2f9e606079d0ab6e5e13c9ca50f9a3f
15076 F20101204_AAAMAH silvalugo_j_Page_103.jpg
1edb249fbd0966bced65e5923d6a7118
69b6b665dda3f2b8d7b410aa826502291110a8f6
30291 F20101204_AAANXW silvalugo_j_Page_039.QC.jpg
46c08b968fdf8a62c6b798e26ced6d3a
30890361f2284c6b7eff6f0d3f237ab57135841a
1051977 F20101204_AAAMNS silvalugo_j_Page_034.jp2
e094bd7d967b4da29ce47273c4682790
a91a2d76b82258185051c60357bd9ea2126c697e
54719 F20101204_AAANKM silvalugo_j_Page_179.pro
09f459788aec817d9a4de0ef423b0878
0309c3a415111430695a2cf54aff5ba25452976e
25229 F20101204_AAAOHF silvalugo_j_Page_031.QC.jpg
2865bdb3a49ad48a837c78c18894a2dd
65c33e832a3de13aee9d151e1af22d07e5c423c0
50357 F20101204_AAAMAI silvalugo_j_Page_204.pro
1334a64b4189723c1ee4f2149d5d22dc
949cf7d662d26231e2362ddbd0aab76c949d4718
8655 F20101204_AAANXX silvalugo_j_Page_179thm.jpg
3f599101d79f4b78ec6078b8d3538d85
ce08757f4e9992d3df3302a30a731a4a525f208f
1051986 F20101204_AAAMNT silvalugo_j_Page_035.jp2
a3a6dafa5f0447a26b45b72ce30c2e69
3300a710ae9ced5c233f8729d144158ae299c534
55838 F20101204_AAANKN silvalugo_j_Page_180.pro
3426cf6fd24c90a2deb3420ff2e194dc
e62c82bc0260141afcc278e474f1c5e431b533d4
6979 F20101204_AAAOHG silvalugo_j_Page_031thm.jpg
eb3eb387072b7f50c3259c99766fe23e
a18d4e024b718dcaea6b1970f298585cf434ab16
9502 F20101204_AAAMAJ silvalugo_j_Page_239thm.jpg
e6abcf87e00f2c4282d7b128c67cbdd1
ba65b0e601833b1c74834a9af311b6fdfa0db95f
3531 F20101204_AAANXY silvalugo_j_Page_218thm.jpg
a3473ff2673629621d5f301e7fea8b48
a534c3a5d2df75ef2aa6f1d0b5af39cf789009b1
967145 F20101204_AAAMNU silvalugo_j_Page_037.jp2
997007b7765926e0674571c02a8452e1
721a5a8b419ff40f116cf48ac5c940889d566bfa
58203 F20101204_AAANKO silvalugo_j_Page_181.pro
e973d4ed45387a5732604338adb37566
89038b063be0568e0541f485597b17fd9ce8ae17
9915 F20101204_AAAOHH silvalugo_j_Page_033thm.jpg
dde083b2835f49bda58abb30d4ea0bcb
a6c7e087e0c8b40bdc258fb5dbed390721b8bc24
25271604 F20101204_AAAMAK silvalugo_j_Page_078.tif
307dc84497f7074df012ea49185ddd52
85ef16c7ee00c3683c5380a34b4ef06369ea966a
9179 F20101204_AAANXZ silvalugo_j_Page_177thm.jpg
f0bafe6619d202e588b34bfc0772d7a5
729e481a5efe3566d5ae76cd44efa303a83679de
1051921 F20101204_AAAMNV silvalugo_j_Page_038.jp2
f4b158ad957ea69f5cba7f42988a1390
6b021edd8c1ad8ca872e3433a67913f4046337ef
50547 F20101204_AAANKP silvalugo_j_Page_182.pro
b230216adc79cb1e595355a19b8fde2b
73084e2fe5d26340397b447cdba46cbad19aea3c
20218 F20101204_AAAOHI silvalugo_j_Page_034.QC.jpg
cd34f4484827b6dfe0c80dda90c70acf
0f36424e3a923127b38313f762bc5290cbb60f3d
6974 F20101204_AAAMAL silvalugo_j_Page_028thm.jpg
e3f7307d624f68f48c1310166c171d88
9301cffd96f605ed61fc2a74b7a9d2ac2feef05e
994537 F20101204_AAAMNW silvalugo_j_Page_039.jp2
936c2f4385e4a032faeff8a4c03ea035
f9579f37332de78714fd42d252f706d25bfc6ab7
55167 F20101204_AAANKQ silvalugo_j_Page_184.pro
560df6e5223f605a2f3578db20dbbfd8
a6ce317013713aaac52ce2107d91084d513ea6c6
5202 F20101204_AAAOHJ silvalugo_j_Page_034thm.jpg
8900c437bae6bb4c454d4aff6bb61018
4f85c0b481f6bbdfa558a9a0b592a9ae301fbc8c
F20101204_AAAMAM silvalugo_j_Page_182.tif
09f9d4e587eb4424b98a72d3b0c9a01d
1bcc87b28a3fe638d224f9aa7aa9b56a5d17eb2c
F20101204_AAAMNX silvalugo_j_Page_040.jp2
eac83646062b3270a8db03b1324823fb
75da9a602508e1d77704dc340193fff3dc048eea
54081 F20101204_AAANKR silvalugo_j_Page_185.pro
f1984ed93a69cb1580b4b20f79b3ef6d
05e90537e5f8e390a6f18563b4753d1e510134bc
28530 F20101204_AAAOHK silvalugo_j_Page_037.QC.jpg
8997d69233d03778834082ce80e9d2c4
c5c72b0b1ba07d7e43660437f2974e177b030472
15504 F20101204_AAAMAN silvalugo_j_Page_149.QC.jpg
f87da8bbdd5d08450cd5a1ad890a7191
11cb607a630a611ad1d922c49f6d7ccdd25b75f1
1051985 F20101204_AAAMNY silvalugo_j_Page_041.jp2
dd25126a0fce7e7973bfdf0f0af370cd
fd220f07e3de731015018e39c2fb5acb9928ad92
54083 F20101204_AAANKS silvalugo_j_Page_186.pro
d7bec5cca84b2fd0d36243874d1c9090
d702e1014c83ffdfac9fd715b00bf80bf1c9cc9c
7307 F20101204_AAAOHL silvalugo_j_Page_037thm.jpg
457355be946ec8672217e140b9046851
5fff32a3c98dca8470b9d3af068c7d9d279e1dfb
55612 F20101204_AAAMAO silvalugo_j_Page_091.pro
61eef845323d17b04ea4e0e60d18f9d4
ab4ff3369dc969bab4a3f4d1bda4e9cdba3bd0b9
1051975 F20101204_AAAMNZ silvalugo_j_Page_042.jp2
1178cc5211fee14292e3b9c23d62f6b3
7e7de51cdd33f7b6ab1f42d19bca73bcfdaa79cd
55758 F20101204_AAANKT silvalugo_j_Page_187.pro
91342c44f844a55cdfbcf4b6f2ddedb0
ebb5c333607b55f0b40163128aeb3e9e2d795482
8106 F20101204_AAAOHM silvalugo_j_Page_039thm.jpg
74922be6a289661dad9ba650162b536a
d8628f29155303956185cf01b7a73e4be36eb554
31101 F20101204_AAAMAP silvalugo_j_Page_002.jp2
1241c2b893a8178d00d12e3613ac5058
8847a1ff717ebc586c7a067ba6889c21696e7fbc
57437 F20101204_AAANKU silvalugo_j_Page_188.pro
e3b933c528312429d621ea95d2d97c19
a54b6d68ed066a4db94f609937e5de8913d48e43
9275 F20101204_AAAOHN silvalugo_j_Page_040thm.jpg
4aacd1ee9925fee4c1753ad638572cbc
7a43a83a05ce60fbb3ab9f01f971005025d5cd35
34833 F20101204_AAAMAQ silvalugo_j_Page_183.QC.jpg
572a1a8afa8fde3b0e96ec29d708e082
952616265d2d3fd4ee7bfadbe06356180d55a3e1
55799 F20101204_AAANKV silvalugo_j_Page_189.pro
fc624fc5e50606b37f08beb17bd8850f
47c27051df3fefe74de541d7d4514aabd268ecb6
8638 F20101204_AAAOHO silvalugo_j_Page_041thm.jpg
ca911757ca22ef2ed2bba0ac11424646
912776a3fb975f09b2dc06a9390f8caafc8b77be
433454 F20101204_AAAMAR silvalugo_j_Page_155.jp2
57d27ab7a6a3f97dcdc6088a3fdbdc7f
3209310a8f274eada48e68314b5b9bb76c07eb57
54692 F20101204_AAANKW silvalugo_j_Page_190.pro
63bd5489869e18ea63cd3e2734b8afe4
b694aa41057522566675a030c057dabab045a22f
37432 F20101204_AAAOHP silvalugo_j_Page_043.QC.jpg
00891a1d559e02e9763ebf5f7e95b5da
5f4c58cfec3007e4ce736e590eeeda3459627135
2327 F20101204_AAAMAS silvalugo_j_Page_086.txt
e484178ee3767060ab01a86b727fe2b7
ce8acafa704745512b84517984c51fa987e223fe
55157 F20101204_AAANKX silvalugo_j_Page_191.pro
55f283945bf2925bf49f72b9941216ec
b5869590032961366f1101b147c9f0ab74ebc123
34973 F20101204_AAAOHQ silvalugo_j_Page_044.QC.jpg
f4237a56666294a2e7d209279b6b04cc
ba4c9976f97b25ca0a4fe8a3ed6e669dbe2bd837
35206 F20101204_AAAMAT silvalugo_j_Page_170.QC.jpg
7d193dddba2310b1deaf166714937af5
b92aa09a46626b428cf3bb2b60ddfe9df39c2b29
56919 F20101204_AAANKY silvalugo_j_Page_193.pro
c5e3c00aaec417935f7fd3fc478cb393
81088980153908bf61994c0f6046acde6a6eb6ce
4742 F20101204_AAAOHR silvalugo_j_Page_046.QC.jpg
0bb91da2977bf3aa86ce471c1fb2d180
546a2b764f8380619e5fb2ce826f95b6fd46c2bb
46933 F20101204_AAAMAU silvalugo_j_Page_009.QC.jpg
72764a2fefd8793c0feaaf61a4db5aa7
5d6eed505135a35de47836c238c33428ffae9acf
57313 F20101204_AAANKZ silvalugo_j_Page_194.pro
3a836c16594858198c771e5b2b458fde
d31c3b1cb5a73fb131f8c4b5a47746046a864a64
1248 F20101204_AAAOHS silvalugo_j_Page_046thm.jpg
186cab7f2f8ba0b9fbaf2bf952b1e95d
329f23d0288f1087f219e8dc9f77adc6f22ab63f
11897 F20101204_AAAMAV silvalugo_j_Page_156.QC.jpg
9f2f1a4078f266d16f0fa430f9b762ef
8db33c21aa4fd1ad60e6974d695eefba4787d9f8
17164 F20101204_AAAOHT silvalugo_j_Page_047.QC.jpg
b884c89b7af6497849bc683cb85132c8
bec3f2aa356fd82d4acce30e1c05e6028b89978b
1482 F20101204_AAAMAW silvalugo_j_Page_103thm.jpg
2261b143f49ac0bead30b6c2f1ffedb2
2786f0754cb55ed326baf3b92e8890b0c0926613
4367 F20101204_AAAOHU silvalugo_j_Page_047thm.jpg
746fc04d59a7a6c128287a78d1719cce
0c2a6c8ee434eb9140d4cceb247be58e31ebae27
8642 F20101204_AAAMAX silvalugo_j_Page_095thm.jpg
db66f37d9113b09fa2b3437b39e916ca
76d055f8688ab50a728842d38d4be8678af20308
1051959 F20101204_AAAMTA silvalugo_j_Page_179.jp2
5a02a0549d65733644354965f312e3a5
4d5cc6c3a0c35674dbdadfc2450f3038469c6568
22383 F20101204_AAAOHV silvalugo_j_Page_048.QC.jpg
68c12bc5400ccd6cdbae0e50d6052561
1b18436a49f2e0a9dd33f53e463ff1af92618bf1
56369 F20101204_AAAMAY silvalugo_j_Page_195.pro
f0e4c0df62401ba1d914f20905125d00
4eaad933c71197c2b1273edd365920f1ed028710
1051908 F20101204_AAAMTB silvalugo_j_Page_180.jp2
9f6a0c8b5e944364be63bb23ff3883af
496a391203c9f1000b8618adaed7a2502fb350ee
31134 F20101204_AAAOHW silvalugo_j_Page_051.QC.jpg
0b81ea2f464ca8f0a4ad436e0df247f7
972e81364094eb4fe14383c91e09e65c34a71f84
9192 F20101204_AAAMAZ silvalugo_j_Page_158thm.jpg
8d50fd712e888c0a9ba4589a59aa59b7
6cd0775be73baf360fd7187a552d04f24618a4dc
1051974 F20101204_AAAMTC silvalugo_j_Page_181.jp2
390bf48c318bdf1f8af3d56c51034741
ac1fb18fe05382c828258cb7587e284c8015bb07
32390 F20101204_AAAOHX silvalugo_j_Page_052.QC.jpg
534ed30a37591e5a6be7c9a8bfc85df0
59a99dad25ed40da6e775ca87aa0052cecc258ec
1051957 F20101204_AAAMTD silvalugo_j_Page_182.jp2
0517a08cf722fd454ec489c6a73158de
5a3c8cf45fd56109b489fccc7c6a05bc947fe263
4655 F20101204_AAAOHY silvalugo_j_Page_054thm.jpg
81bed0e533152b502e5b6dce309d762a
d1c979916800a0299b746de0d538e045fac1f9b0
F20101204_AAAMTE silvalugo_j_Page_183.jp2
6d992e64e17570b9b3fb1cec0d2348f9
fd8a9a36675b9846b4130d13106336ee0a4a7ad1
7873 F20101204_AAAOHZ silvalugo_j_Page_055thm.jpg
cdb3cefe1991c414268fececd334aacb
243b5d70da5375ee2c74bc60275e2657bf28a9ba
F20101204_AAAMTF silvalugo_j_Page_185.jp2
8f1762a181e361fedd8800e710d9f95a
9bacf2217461b049b110318f8c6a5fabe3ceced0
1051945 F20101204_AAAMTG silvalugo_j_Page_186.jp2
965cc97c37070e4cea2eaa66c3cc551f
3d38f459293f89413429ac622e6bb6011619d56b
2150 F20101204_AAANQA silvalugo_j_Page_087.txt
224ec2f1206c39a4b0ff5eb5948d868b
c1267b031de77c285c5aeffe5452e673a36352e2
1051968 F20101204_AAAMTH silvalugo_j_Page_187.jp2
cdc0f43e26ccf647cae0e8e0462b03e3
cb1840a48773f390c7d29efa43b4c5625bcb70f1
2213 F20101204_AAANQB silvalugo_j_Page_088.txt
c7f25e127c458b4372fc410a6a01eac1
d76eb1512984fa50c68b4ffd1fd7afe9f37f7baa
1051979 F20101204_AAAMTI silvalugo_j_Page_188.jp2
9a919dfda61602d93b9ddbc1ada64f54
7ac8d544ed0710a512e4516f34924475ca8d30dc
2144 F20101204_AAANQC silvalugo_j_Page_089.txt
047b197001f47a0c19bad2d26d04c32b
e5ce59669bedc98a07c734e2353a96a0d7d6306f
F20101204_AAAMTJ silvalugo_j_Page_189.jp2
8aa1b576ac443fb4946352a484bd48fb
24d1c60e6b76cc444c22a9e8ec2a7dadc64ef34e
2207 F20101204_AAANQD silvalugo_j_Page_090.txt
ae16da50a8485d67626e2ae27ac63dfa
889801d12d65439ca6057740c60cd27fb45d97b4
1051914 F20101204_AAAMTK silvalugo_j_Page_190.jp2
0b94e467e74fb5a8a32569f5e2237d3b
db184c2451416ce6cf6381ca0c841333caa5e58c
2232 F20101204_AAANQE silvalugo_j_Page_091.txt
41ad92bfa8076ad3c9dc8d0d484774d5
d52c24100854a428ef07be4cbceaa3b09e92f203
130251 F20101204_AAAMGA silvalugo_j_Page_061.jpg
1841ce097780c881e1e59d6b93d529ce
494c16980285d8fcf000316278801936b5547bdc
1051962 F20101204_AAAMTL silvalugo_j_Page_191.jp2
143180d10d5cf768bac7252263b457fa
ff5894a6532e439ef91550e780b0a120adff2350
2043 F20101204_AAANQF silvalugo_j_Page_093.txt
7c14eac1e9e6accc4a48440084659669
3868716f0e272f42519824a8eab5126fd2f82a32
45484 F20101204_AAAMGB silvalugo_j_Page_062.jpg
152802987402bdf03a66ee29a9dbf99d
53f30b45ea5ee950e11eaeb3eeec5f852d74ae5b
1051971 F20101204_AAAMTM silvalugo_j_Page_192.jp2
123390a2d9a8eb1519f5dc1e9d71d31a
28c868e70ab55c2eddc3e97ec8fdec9d7e4fdc9d
2059 F20101204_AAANQG silvalugo_j_Page_095.txt
a427960944d4dc5649fb9b4d10d0f902
45cad44f45153dda3ff89f34d7a008c381416439
43204 F20101204_AAAMGC silvalugo_j_Page_063.jpg
94be14e3926d526cd3b81a0132cf2860
f37dbffad3214df78118fda92ef2f3e7a2d52437
1051984 F20101204_AAAMTN silvalugo_j_Page_193.jp2
9dba543fffb62a966d9689cde1e6a1b3
ee757d60e5a9f96929e1cbdf8ca6d0445217a793
16794 F20101204_AAAONA silvalugo_j_Page_208.QC.jpg
36ae045302ad3d12f80c80f740d23353
b338ffff05a7862e3bbb592c40cd32a7e971ea30
2174 F20101204_AAANQH silvalugo_j_Page_096.txt
8a849d6e8d0a5e14c51fdb8dd269825c
54ccf2c8b2af9fa819c18c664288f8a96404df99
1051955 F20101204_AAAMTO silvalugo_j_Page_194.jp2
a7ff89d3dc0599436414b1ceed71bee0
b2c823b0e2b084f8201737335992679b5750b57e
32588 F20101204_AAAONB silvalugo_j_Page_211.QC.jpg
f17fccffdfec09ed6382d0ee56c97807
2618ccba946cc8bdb4efa0894bb93596ef068c63
2083 F20101204_AAANQI silvalugo_j_Page_097.txt
ca0321ce863e8c7c466972050184638b
39c3613e9e8e48e5b21ac0eff09813c9ea77b90c
106996 F20101204_AAAMGD silvalugo_j_Page_064.jpg
c47dd509320e8eddfa210804d8983b76
461123ede6f6e3fff9735907445fc4c08ba93339
1051963 F20101204_AAAMTP silvalugo_j_Page_195.jp2
bde5918ce1c903c25399aa118f79b504
7613e27c8ac090476a0c32a07b14fdd24fc9d5f1
18780 F20101204_AAAONC silvalugo_j_Page_213.QC.jpg
c9fc100448154a9096c8acd9ceaafdf8
1933f8eb51f69774420ff5a0d10da811c422c980
2217 F20101204_AAANQJ silvalugo_j_Page_098.txt
1045a1b292ca97fd2cd252b8e85f320f
80642a7f239cbf1e2c68d6ff6d63b8e4a2836934
109461 F20101204_AAAMGE silvalugo_j_Page_065.jpg
c77bbd1648c4338022237bb963635d96
b8bd636dbedc21f5356f7978c1abed173132a4fe
5154 F20101204_AAAOND silvalugo_j_Page_214.QC.jpg
fa83abcba854dc88556d1b4d1e5fd6a0
6656ad4ddc38cd063a3aaf87f65bcdef6228a301
2148 F20101204_AAANQK silvalugo_j_Page_100.txt
e1fb4ce949e87c64783667b3a530984f
374f658ff68672cdc306198d102e85d6a6311c5c
115321 F20101204_AAAMGF silvalugo_j_Page_066.jpg
d6dbd3bad5597f449ab2352d29706e48
5ff70e930277b46d005cb3d51a04901f27bdbe7a
F20101204_AAAMTQ silvalugo_j_Page_196.jp2
3d99b4e8aad62c37c725705f1bca26fb
4e04f4377135be890e54838f526d835abf325885
1711 F20101204_AAAONE silvalugo_j_Page_214thm.jpg
5fcee0b46706a9b6130637278137ea30
36dc5649f4a7b4a4c74c902cafa018ef270c0b05
2114 F20101204_AAANQL silvalugo_j_Page_101.txt
b7f5e1962f514cbb5de0bbcc88948fc6
bbe11b8518abaae476fa268e235058c9c16628c7
109531 F20101204_AAAMGG silvalugo_j_Page_067.jpg
43dcc08b85148a415e6fb65b1fab71c5
9cd95cfcf71acdb5283d7e4c0f771d625a92ca5b
1051965 F20101204_AAAMTR silvalugo_j_Page_197.jp2
654717cec3fd658725598468aa337d43
84f7a80e610d045c52b4a8187e3f511a5274a93c
3021 F20101204_AAAONF silvalugo_j_Page_215thm.jpg
2d1b8507561fe827631dfbc8ec75b160
0fe06b2ea82732718f48ef2891be71a67803a365
112831 F20101204_AAAMGH silvalugo_j_Page_068.jpg
1e16372495fb683a22e044860adbcf27
0a1c53700806365602d62fe12dbc751160185386
F20101204_AAANDA silvalugo_j_Page_218.tif
188261dbcc4c0955b47ce38695ed0abb
fd3e19291af28deaafe94a9f745373fdfd70cae7
F20101204_AAAMTS silvalugo_j_Page_198.jp2
4ed6b4a8552df6616a3dd5940efe887c
04434dc6cefdc4a699354464320561718f3f5ad7
11172 F20101204_AAAONG silvalugo_j_Page_221.QC.jpg
bdd212ed6a968c2213549f75145512b2
4b004e48dd4e8cf1bed417e6db44ca7034ab57b2
2253 F20101204_AAANQM silvalugo_j_Page_102.txt
c93e4c92c3c233b977933518664606b1
961b0745ee5259728698541e549110708a07ad6e
116537 F20101204_AAAMGI silvalugo_j_Page_069.jpg
395bf7397c6d79baece1ebfde178e594
1364cb2f51de126bf3d4de5c57fea99c031643aa
F20101204_AAANDB silvalugo_j_Page_220.tif
9998ee0f37fabda668920a45f61fc1d2
32849927b05e545ce3dd7bb875ae4577f6777a6c
1051941 F20101204_AAAMTT silvalugo_j_Page_200.jp2
24841eece9f08b36837a28f5bc003d00
eec17ca231353cdbaa11a5b53f94f514ed4b645d
3768 F20101204_AAAONH silvalugo_j_Page_224thm.jpg
05df5514e0a114d61effe60d16fe8b87
64fffed38c738cbaa5fead1dd5463c150bfc23f1
95651 F20101204_AAAMGJ silvalugo_j_Page_070.jpg
5484919b64df6b89ff5a6a42b5c25e79
ff7afb6c99908bed7490186f7d92353d7f72966a
F20101204_AAANDC silvalugo_j_Page_221.tif
a91c5556006135538097a3443931cca0
aec547400ff87cef881ea4a965265fa98a05b870
54201 F20101204_AAAMTU silvalugo_j_Page_201.jp2
346595e149d7e43c08654af226f1c627
cfe982c0bf401fd9b9301e2abb435c54ac29d223
241 F20101204_AAANQN silvalugo_j_Page_103.txt
26662ada3b937e62a9c00189ba8452a6
805388f825d091c88a5f61e4a916fd584dfb4ce5
124742 F20101204_AAAMGK silvalugo_j_Page_071.jpg
7772714e9c97edfa04d8817afe31a609
a45770412eec5a285afd8fb3d2f16eeb8cea3191
F20101204_AAANDD silvalugo_j_Page_222.tif
aedd45442008a610a82499fc9d167490
526012977980b40eb9750b825e44d40691f6ade5
28258 F20101204_AAAMTV silvalugo_j_Page_202.jp2
37dc784d0aec86fc28f256f9b5a6b08e
15cd072d9d4e33acc653f70756fbf2441a5b7467
2818 F20101204_AAANQO silvalugo_j_Page_104.txt
d03ef953f023c01e4fef5703b43256a0
96a4f52f3be96f6c684cdf0bcde03be0617753b9
7972 F20101204_AAAONI silvalugo_j_Page_225thm.jpg
f1e288ebf65d2db2d8263db01e2cb210
e026481b833c4b7ba4c0ae37d40a3971e281f623
113500 F20101204_AAAMGL silvalugo_j_Page_072.jpg
22dcd6145e99227e4d79a21b2cf5ae07
f9ecb893f6eeebdc109f2b83e5987a84885e9280
F20101204_AAANDE silvalugo_j_Page_223.tif
12e33ee89dfbe52d50524e388a84df07
c7e4085f283539332bf1c919d03885673a4bf089
747883 F20101204_AAAMTW silvalugo_j_Page_203.jp2
928e93f866707f7b81fc3231e8b2a90f
4939a7f791e5dddba2a94d8327f2157e497e60c7
600 F20101204_AAANQP silvalugo_j_Page_105.txt
39d94ddd07d32131069dfe5622a9c1eb
4918f9ffa43edaf3b68afd2e79066e527d3581fe
36196 F20101204_AAAONJ silvalugo_j_Page_226.QC.jpg
5e4e5fd994b99df571f04d0c767564a0
a02a477d015c8f8385d7ab8e0f0c9071444dddd5
110694 F20101204_AAAMGM silvalugo_j_Page_073.jpg
1f21e20895e2fa5f4e53bdfb274ec80d
85fd04015d6beb68b50fe7fda9574ba5a5a8a20b
F20101204_AAANDF silvalugo_j_Page_224.tif
4ac46ed99a63f7b6f6e6415e516ce451
28b37ab9e22a23046251997dcdb359c3b9f0d850
693638 F20101204_AAAMTX silvalugo_j_Page_205.jp2
667080731d2567821ec6f9cd9bc6dcbf
6c447945243d2ba37542c8cb654fef829c58c3c8
2431 F20101204_AAANQQ silvalugo_j_Page_106.txt
243a27bff32b98485036f7516155caee
3bf0852ce062bde4f155ea38958d06247a378189
9092 F20101204_AAAONK silvalugo_j_Page_226thm.jpg
159284cac86ac48f00567cfbcd8907ea
87404e2a46f7bff137813b6609479bf971686659
113368 F20101204_AAAMGN silvalugo_j_Page_074.jpg
574d15637fa3a7ee7faf467d11cf9a10
fbda1a4f9c4184c601b7c8bfaaecaae6929c36fa
F20101204_AAANDG silvalugo_j_Page_225.tif
4715ee97097b7a3cda8ebdfb89fee0d8
f3a6a82b12d8e94ea1036fe708d6789a7b7ab9f3
F20101204_AAAMTY silvalugo_j_Page_206.jp2
35406d140342332ced06c2594f0024f0
626c427824da069c01e1b276e7afc3627a9a9f34
1230 F20101204_AAANQR silvalugo_j_Page_107.txt
100651542e43fb2fbd015f512bb125cf
b4339d4c9f0d1b550267ae4ffe369032e90638be
36443 F20101204_AAAONL silvalugo_j_Page_228.QC.jpg
446fdd46b8368d40cd40edd26febfc2f
042a0804decb40c5aa0f66c0b88007930d4e83ad
105003 F20101204_AAAMGO silvalugo_j_Page_075.jpg
c863eabd1a8a2c39b7947f428e4e0bd0
39d5ed73d0bcc7090bd061c6a894f12c8d6c8dff
F20101204_AAANDH silvalugo_j_Page_226.tif
e4a1a572ccef66f7b6a4bdb08133b2ab
8bfc0ecf8c8f2ee0b8da758e4868adafaca39724
343228 F20101204_AAAMTZ silvalugo_j_Page_207.jp2
20a819af3183b863fd28b49a454d94fc
b41b6a71c951696b2cefddcd671e6614d81b315a
8144 F20101204_AAAOAA silvalugo_j_Page_035thm.jpg
1eee0cfdd4b139db158075b70bee4a9c
635c4cd552870fc580edbbb1db9029d9a845d7d3
3465 F20101204_AAANQS silvalugo_j_Page_108.txt
eba26483faa12fae0fac4e47974c8f11
8db26844f19250fe02abaee3c42d5286e9c79fe8
37645 F20101204_AAAONM silvalugo_j_Page_229.QC.jpg
04e2887e87a759bea3a9a3600bb31c9e
3da090f52949bfcfeddbf972e014678362e36667
94921 F20101204_AAAMGP silvalugo_j_Page_076.jpg
e138bd628cf4b212710b489cdf099256
af55063eaa4211e7e3ac4bb7c5e0127dc747a3b3
F20101204_AAANDI silvalugo_j_Page_227.tif
2cc70b4c6f8f45f6f5c0e07b8823cfb7
72988af08efbd081bb93fda8ba327558af014d88
3729 F20101204_AAAOAB silvalugo_j_Page_122thm.jpg
625570fa55a4eef8cad694b5db9dace8
6a86b31465c385722f18ec25f32e498faf424048
3989 F20101204_AAANQT silvalugo_j_Page_109.txt
8d1b3973ba9e7d2a613079ea4cf44214
dec676789db65fc46820274cc349014bb1f3b4e6
5249 F20101204_AAAONN silvalugo_j_Page_232thm.jpg
4c34d95557fa08eddf3d0d78af96dc27
1744a1174b2297e738965ee10c575df2b5e4c22d
99442 F20101204_AAAMGQ silvalugo_j_Page_077.jpg
e6ed891e6b67f735ec7a5ead571a963d
51475af0a558183ebf726be0f120137b067e2c66
F20101204_AAANDJ silvalugo_j_Page_228.tif
96041f8fb0e3bf16e14370d7e653d780
9dcb30667377644b53e6328f65eb7546f4ec348f
32378 F20101204_AAAOAC silvalugo_j_Page_165.QC.jpg
a3e7360b109477438a4a6c3576f7c696
60fba9296ff404c9ed1b47a5ea785e79f68c91c3
3990 F20101204_AAANQU silvalugo_j_Page_110.txt
836b5f19f74668daa01d2d22091a7551
c22c0b3ad59bf5a21636a73f9d89f57d92309626
33948 F20101204_AAAONO silvalugo_j_Page_234.QC.jpg
95fb1c6aa075e8dad9bfc84b3df829ef
0a2f105c557999876c3cbd2a3482d116042ed89c
114533 F20101204_AAAMGR silvalugo_j_Page_078.jpg
326ad6475905a666d2203d6998fa5c57
303f0c175f3356d73a805f59909c5d3a57318e10
F20101204_AAANDK silvalugo_j_Page_229.tif
857b11b3ca6c300482e43928fe2e6220
eae00096652c323efd9d7a47aa1588ddbef6eac0
9062 F20101204_AAAOAD silvalugo_j_Page_178thm.jpg
0fc420578885cfca16310985744a56d1
b1b08c99a99920f545890da3f7d5959d57735e74
3545 F20101204_AAANQV silvalugo_j_Page_111.txt
6aef725b417a137e2544398f3ad98c9a
80abe42a641689888fcd191b0496bd7aed91bded
36327 F20101204_AAAONP silvalugo_j_Page_235.QC.jpg
f10554827e987ea65e907f8c0c63a472
d15c10ad6540c4dd0de046e5f84e562aad3ed5d3
112472 F20101204_AAAMGS silvalugo_j_Page_079.jpg
273f7fc14c14214befffb29afc3b96ce
f00d0f3c6617b9df6b41f93fefddd5afae2882d8
F20101204_AAANDL silvalugo_j_Page_230.tif
92a6a7b028d0adda0e845b2a99c31d95
f8de4a8b3edbeb6d8aa87628abf3c2d1eeb53c56
10320 F20101204_AAAOAE silvalugo_j_Page_014thm.jpg
0ca5dbb123134354238eaea02c83bf47
5be7c159997a4bafa551bc1618086b7186316834
3109 F20101204_AAANQW silvalugo_j_Page_112.txt
e1026f5a18cadba7ca83c22212931dfe
997fd1a83efe017327f103d62b1a7ecc3e4e0b54
36724 F20101204_AAAONQ silvalugo_j_Page_238.QC.jpg
04d1edba8ab08d4290817f792c610b45
c275fa1b8944bed8c2bc8f959b390fb8191c4272
108734 F20101204_AAAMGT silvalugo_j_Page_082.jpg
aea0d288c61279a0fbc78aaf76597280
e5a3a7be1c2326c26e637d726e298859a78a3361
F20101204_AAANDM silvalugo_j_Page_231.tif
334e5bc8f2e4b1272da9a80db3d6092a
7e0a10dd0b3d8b6fb3a08a9d6bc65ea981913eae
37599 F20101204_AAAOAF silvalugo_j_Page_195.QC.jpg
3ca16383308ebc0a370b0c95c270d084
0acc286ebed7297cd51c139ff88bb600093ac630
3987 F20101204_AAANQX silvalugo_j_Page_113.txt
8bd293f6e83b86886728e3241ead9eb3
d8dd8ad01b36e24816b6d4f09345e5d4afd59ad0
39309 F20101204_AAAONR silvalugo_j_Page_239.QC.jpg
c540c315986d83baf658771b715934bf
cc926a893d2d81dbd3cf3cb99e4b8c80f17f4c2c
109993 F20101204_AAAMGU silvalugo_j_Page_083.jpg
f4ef797104449c9ab8781e087d5e2430
da021ee8a751d6edd4a8e979da3a460f66d70f31
F20101204_AAANDN silvalugo_j_Page_233.tif
ab27246dcae7ec84ba47c7a888625708
6e4136eef25df1569d9c158cb82ed76841117e3b
35448 F20101204_AAAOAG silvalugo_j_Page_191.QC.jpg
c1313e39eb2579f15f9d557b953d2c15
04ccd87af22ad5c65f292236c3d78b3adbd5c08d
3376 F20101204_AAANQY silvalugo_j_Page_114.txt
76977fc731e1db6de3bda6e1c2c6c2e6
fa9d3e056e8f9a68ef73e13594d1235e0a2d3d5c
38092 F20101204_AAAONS silvalugo_j_Page_241.QC.jpg
d8dbda473966b4cb521759a784f4e085
a3df1731ad90e86245273b87ffb90fe8e4505140
109473 F20101204_AAAMGV silvalugo_j_Page_084.jpg
753d2645493bc3a2bff2267eab7febfa
c0b4b80d6deacf13b11c97f9096904177c7b26cc
F20101204_AAANDO silvalugo_j_Page_234.tif
f423233790c4bd2b69abc3630cba487f
b4278247b3f08ff2df63033e0361aa2a882f4f27
35389 F20101204_AAAOAH silvalugo_j_Page_021.QC.jpg
a7aa39318f3ae129d21eaa4a691657ab
b790d0ed82a301669ce0b4e6764a0ccbdb2d76f7
3578 F20101204_AAANQZ silvalugo_j_Page_115.txt
e34395eba101c1477617bb614fee4a03
007b1a354160b00a27ef8c45f3f59717a5e6f69f
36550 F20101204_AAAONT silvalugo_j_Page_245.QC.jpg
3fa022aecff352524b8a28b32b496e59
d7c3f1d3f53ac064227267f09eca0d78d95f2ab0
104556 F20101204_AAAMGW silvalugo_j_Page_085.jpg
d57c8b384785d9d926a89891eeca703f
e43ca0b9e66c118695d97e4edb455f38dfabe06a
F20101204_AAANDP silvalugo_j_Page_235.tif
fe515fbdf8972143116c21c83b030cc3
efbe4a3b6c3df422400a6ce6f28642d5904dd3eb
3396 F20101204_AAAOAI silvalugo_j_Page_221thm.jpg
70a24a60c09f87d8f08f3cf4ccdbef0d
a2d22919b37a22bb7d3cc99905bc66e3626cabb8
38574 F20101204_AAAONU silvalugo_j_Page_246.QC.jpg
e632ea2bed615bf81cd2ac7c9551af90
cb8387fab077fe25866364c84abd2d9f084fe062
F20101204_AAAMGX silvalugo_j_Page_086.jpg
bc5eff3d95c35128e39340c3f496cc45
af0b68be3e275c474e232acb5197cdb040394590
F20101204_AAAMZA silvalugo_j_Page_104.tif
a386efe79780de7990ecab76e5de14a8
49514627b8c00aa7a15c4d5190ddf9f8f299d203
F20101204_AAANDQ silvalugo_j_Page_236.tif
9e887f8e1b54d2f1e457b0f547a1e60d
a455763bc1748bdbafad1705aec36500ddfc3881
36290 F20101204_AAAOAJ silvalugo_j_Page_197.QC.jpg
9041c410e09fb85e27fa9be69e0a1e4a
0b609aee2256ce95af0fc88a495fc62050ea3f4d
32437 F20101204_AAAONV silvalugo_j_Page_248.QC.jpg
864be98568eef0d2b2e089f7b7bf8edf
512a39016eb822ab8d712b6c11746b80d5e8ddd5
109971 F20101204_AAAMGY silvalugo_j_Page_087.jpg
3625c2a9e3d36043c45868a56626acb3
5f1266479ba375e1c9654f7470a8e1289c9de0bc
F20101204_AAAMZB silvalugo_j_Page_105.tif
274e3a650651716c24e00b9b23544c42
deee2ef5d47b422a8434a9ef4bee7fc6a0663027
F20101204_AAANDR silvalugo_j_Page_237.tif
18dbec66d43486b61758991e18cbc6c0
0c99a80a36632232a8383be58787b715e90d0a2c
37465 F20101204_AAAOAK silvalugo_j_Page_161.QC.jpg
7f51e7d5f85cb8d3e2932ae85ad04e32
6180a805a8f302589ab769fb2525654bb8537b81
114174 F20101204_AAAMGZ silvalugo_j_Page_088.jpg
6f5f578a2bdc4bc8b864eaaf8275219c
ca19c75b19cc4f0ecabf56ffef9b44c8363d67b3
F20101204_AAAMZC silvalugo_j_Page_106.tif
562d96d06e947fe1f328a80f8f30ae9e
ba2ee4844ca966478c0b01714ee7f472b2ca5d54
F20101204_AAANDS silvalugo_j_Page_238.tif
c39fa35ea386e4a7fed9a4788ab22312
632524d1f375e86d2773fbaab27c90447a969898
12564 F20101204_AAAOAL silvalugo_j_Page_212.QC.jpg
6d285e5a3032b471d21f675babe1c1fd
55484c6569b97c1540d89297ceb80b6a79b9ce61
F20101204_AAAMZD silvalugo_j_Page_107.tif
0a17fe5c42abca44091ccee09d337775
7536a89e1ee78afdee4c856eeaf4a1b26c2ac4c5
F20101204_AAANDT silvalugo_j_Page_239.tif
c8dcd2c3ca50586969260f3de45f9392
e9f3d8349aab0470bbe0df7331d878bb8ca76d63
37384 F20101204_AAAOAM silvalugo_j_Page_240.QC.jpg
f60f325a545fe11614d61f17bdd6dd40
ecb992213edba28f06338fbeb9b9ee9689eee943
F20101204_AAAMZE silvalugo_j_Page_108.tif
8f033348742d0d1a226bdce123a30832
36cb5e0ea55984d19abae0ee9370186b6717fb4e
F20101204_AAANDU silvalugo_j_Page_240.tif
a4018df7d2a0da0c5044879fa30193f9
07da354dd52ddd58bf94c1e04252d89311e58048
31548 F20101204_AAAOAN silvalugo_j_Page_036.QC.jpg
783551dbdd2093f52e86073be84bd6b2
3696ef41a4bceabaf6c009261d480dc4f12a3e08
F20101204_AAAMZF silvalugo_j_Page_109.tif
cd88390c1936a31ef74bc70bfc410d8f
0169cf3b92f046f7bba414a8b97933179fe59ba2
F20101204_AAANDV silvalugo_j_Page_241.tif
63a98857f88490e303271a3f95f4e29b
817f0a68b6239d46597968b13051d95fe89b5062
35976 F20101204_AAAOAO silvalugo_j_Page_178.QC.jpg
13afea2b2fb8796620880f282292e8b2
7c8c70597a322b4e705d33f6dcdbe8493e087a1e
F20101204_AAAMZG silvalugo_j_Page_110.tif
7bb66dd8d00f8449a0490e8e7c175561
87fa250aba077429bc00406562c9b7125464b1e2
F20101204_AAANDW silvalugo_j_Page_242.tif
1153c57ea58ae145bdf415586db0f357
e4d39cd375c1dfd0b0a00a0844e7b3d5debdbb6c
4955 F20101204_AAAOAP silvalugo_j_Page_146thm.jpg
8310353e8969a65312cdddde9f00f4ad
1d5fe63e6686e013a60c535519de3b6f0132282d
F20101204_AAAMZH silvalugo_j_Page_111.tif
e205b90f182375831f2e824fca1a8e2d
ab19ffaa954a270de52263d779ff0627b10ee7c5
F20101204_AAANDX silvalugo_j_Page_243.tif
b12cfefb4b21a7deb4c2eb25f6ba728d
d3270e9473d1685b25c000f3129ec96da6b047b0
13642891 F20101204_AAANWA silvalugo_j.pdf
23d1897bea300a91ca1dfb1bbd686b20
6e5938b6d6cfecf0d296dcf2ef5699076df2999b
3261 F20101204_AAAOAQ silvalugo_j_Page_212thm.jpg
f916354f31aa600d0cf8b3a350ca2a10
34dec174c36b4d72af401b89c6b1370c5be2ae88
F20101204_AAAMZI silvalugo_j_Page_112.tif
cc7d867c2078531eda0204d3dde7ca6e
ea365e997abb0e7f688ff2dfca833706419208db
F20101204_AAANDY silvalugo_j_Page_244.tif
d8255a571b38ec12d83da3dc852ee37d
ca639db3dd2f130d0f8159848dc72eb578c288ea
36184 F20101204_AAANWB silvalugo_j_Page_065.QC.jpg
1357ade154ddf79916d518090cefe463
a6e3a3fe12ebddabca0ccfa0f347cf43e273adef
1301 F20101204_AAAOAR silvalugo_j_Page_002.QC.jpg
815280e09a1e30e7fd61e8455728c27d
9ca44a707f698d93ea51b898cfab21fbcb655436
F20101204_AAAMZJ silvalugo_j_Page_113.tif
cd822aca769a289b203978ac785ec72e
1db4deaad9af726af76e8dac13898f2d1b8b83b0
F20101204_AAANDZ silvalugo_j_Page_245.tif
200a8d982fa5fb129be5faf577320ca3
b4783fc2c8e1f5c04b91766e59a8048f697e34e6
3612 F20101204_AAANWC silvalugo_j_Page_217thm.jpg
18a82309450d4fe6b993f2531d0b0259
7c70827fc538638d7c75d4834432d0f184923e7e
37569 F20101204_AAAOAS silvalugo_j_Page_181.QC.jpg
273f3f234ddef51cb13dfb0012b65626
02b5319f5b81e05baf8da090abd2904dae91ec47
F20101204_AAAMZK silvalugo_j_Page_114.tif
8fa58f1dcc14f38cfa81208d246b66c1
76b8110479a7388e858c41f7d0acccc6f8eb1cc7
15430 F20101204_AAANWD silvalugo_j_Page_063.QC.jpg
41e091412282c6806c335bd6084b8afa
39cc8a8579cfb34871da801c91ff9941905b3827
33133 F20101204_AAAOAT silvalugo_j_Page_004.QC.jpg
fcb4a684559be6f56169c3336072f3c6
1ce59ed83b099564dec9305e3d3a55d11654bc1f
82302 F20101204_AAAMMA silvalugo_j_Page_233.jpg
719700dd3addaac6ece0c83cbd227d2d
53566c9911daa52ba75dfcdaad849bf855db47ee
F20101204_AAAMZL silvalugo_j_Page_116.tif
68675bf9e164e87ed69f76269b9d44ad
373b22cf766386878bfac22c3c17fddf576f422d
8332 F20101204_AAANWE silvalugo_j_Page_165thm.jpg
0a1cda8c61b4c1e9635c819e71fadb11
b610849f0eede3b308b56973219263be4d6689a0
574 F20101204_AAAOAU silvalugo_j_Page_002thm.jpg
441a97f38f7a1c2bcc3a53366af8ab80
05f94e5e0ae30d47feb8f1b6658992200bbbf704
112641 F20101204_AAAMMB silvalugo_j_Page_234.jpg
cd0d15233bd1b12e42ecd6949109defb
e0119eab665ac37b76a248df4a06e3051d589679
F20101204_AAAMZM silvalugo_j_Page_117.tif
78845200b2447f07c3695385e9791594
1cd289ef5180f093e980c22ad4a15b80e2ad92ac
2751 F20101204_AAANWF silvalugo_j_Page_247thm.jpg
4e1384f0bc379d353c8324d0dacfa7df
b5bf28212ee3d025d7747dbe15fcbbfca967cea0
35011 F20101204_AAAOAV silvalugo_j_Page_104.QC.jpg
82549d379e8657bdd8481ceb537cf6bc
c59b8c577822f16494b99641f5f47594031f97be
125178 F20101204_AAAMMC silvalugo_j_Page_235.jpg
6da1173bdfb0c003fd4cfb3397f8037b
aaae5c80820ce4e26212f0b241dcfb4e8ca5857b
F20101204_AAAMZN silvalugo_j_Page_118.tif
e0ff2cfdd007a42a987973e16dd49cbb
a79a1390de96d1d76d50fc67443fd53c86bb26f3
33602 F20101204_AAANWG silvalugo_j_Page_053.QC.jpg
d45d14e93d26c3583b372b0b56427bd0
cc03f9981f514d41798e758de47f619a91bb4212
40958 F20101204_AAAOAW silvalugo_j_Page_012.QC.jpg
eee5cfbc333d4d3aacd1cf91266b187e
08777cb2fa26849ea7d5bcf8fd4be65e17ab86da
123957 F20101204_AAAMMD silvalugo_j_Page_236.jpg
d87dfc635f9435eb668de8ba07bdbac4
247de602b72fe6030e2401465125f5252001a8b3
F20101204_AAAMZO silvalugo_j_Page_119.tif
95716264d708a5b3fa4e9cbd275c055a
501778ce0131c610c3045bb4e98028c3e0caaf29
8858 F20101204_AAANWH silvalugo_j_Page_087thm.jpg
684408df2408f33075c1a58102bfb080
6cbde4d448795ca6b6e031747ae49c907b016290
34769 F20101204_AAAOAX silvalugo_j_Page_095.QC.jpg
89d877e28ad620d8aae4cee54b2c31f6
2ae4d2bcd099a939c05302de355d35ac6c87fccb
123005 F20101204_AAAMME silvalugo_j_Page_237.jpg
ac011e954df367c6f2ab5e8bc2fab676
c0d0c97eda6eb006331623f409555d3362b9eb17
F20101204_AAAMZP silvalugo_j_Page_120.tif
d857f8b55d3684a80b77d9f6701e534c
11de12cd126cb8911953a52c9c1cb19a6b451f4e
9151 F20101204_AAANWI silvalugo_j_Page_229thm.jpg
76646f5f16f406e79444744072ae5575
ca07dae7c28dcc1fa361e74ec16a73bc29d17244
13922 F20101204_AAAOAY silvalugo_j_Page_201.QC.jpg
f0960265c79f6d884b1be1e4afd6c00f
7c0b0e64a96843fbd544a9ff51f5577eac4a6735
130830 F20101204_AAAMMF silvalugo_j_Page_238.jpg
0ee05d8e82bb9ec1b54241ed7986fe80
700f1d27d38e36cdfcb4f9dbf6d247f71f028e64
F20101204_AAAMZQ silvalugo_j_Page_121.tif
264f233f3a56a49794075f19112a6bce
9817144dfe56fa6a37aedadf9516ee36bea70907
36068 F20101204_AAANWJ silvalugo_j_Page_236.QC.jpg
2ce0b0709118aa45585476d2879c0f97
5cbe2b53b8931f122112c3b9ed15ac269bbfc947
7966 F20101204_AAAOAZ silvalugo_j_Page_077thm.jpg
9dc105e724466eab66cf128bc7dd17fe
cd7dd1eb5f4495ad128a13c599114c63193fe047
140202 F20101204_AAAMMG silvalugo_j_Page_239.jpg
dfaf4923a79f9ffbc1538e260cb2f3d6
c4dcf638beae70b0360988d579505bff926f0387
1054428 F20101204_AAAMZR silvalugo_j_Page_122.tif
b59eba8a620469e0c4ae1333d3a06ce5
69c24731f3f6281180058ab599e94ae24db0e8b6
8824 F20101204_AAANWK silvalugo_j_Page_228thm.jpg
729f63504f0214f9771321d0fff0eff2
064304d24ace948f90a7eaf537f91e1c92be9557
131628 F20101204_AAAMMH silvalugo_j_Page_240.jpg
a75f775c103912be8a57b84ee652a7bc
0ccf1a7ce9012f6045b9078092a91940ad957e0c
15236 F20101204_AAANJA silvalugo_j_Page_138.pro
784d9f48a60270b9f9cf2294e02a3f87
93e3559d2cb2ffabdd7b71d007ad66c984a35bf4
F20101204_AAAMZS silvalugo_j_Page_123.tif
b40dd02e31e33350de5202483dd95f5d
595f50a39e0618ac4454f1294720b8cd9a1a8ece
8924 F20101204_AAANWL silvalugo_j_Page_101thm.jpg
131b673e08aa8205c55345619a055220
38fdd9207c84d67b7df0f563380e568c1285a995
132633 F20101204_AAAMMI silvalugo_j_Page_241.jpg
1f094f724108295dbf4b566f4b90f37e
656a496e80261868ac826241f66ca1dd98524a31
15622 F20101204_AAANJB silvalugo_j_Page_139.pro
53803a8d0c1152155e342af28ada7413
483f4ffc5201fd12a0602c5542ff68c4b5f3c59d
F20101204_AAAMZT silvalugo_j_Page_125.tif
68e029d5d205b6beffe9c07d5b7865f6
57cace5d8bd46fc35d0651ee39b236f6f787c2ba
36407 F20101204_AAANWM silvalugo_j_Page_022.QC.jpg
f9acce66091d726641016ac7f9ec0e37
9129bcb1c0f985e9cfa43f901866fcae197e92e2
12668 F20101204_AAANJC silvalugo_j_Page_140.pro
2400e78fac7dd314fed4603571a0e202
728f395a584f3a2e4c33d8d8b3a3baacfb98c5c7
F20101204_AAAMZU silvalugo_j_Page_126.tif
ea720aed045d0f00c64a644efcaba2b0
8eae4afb4f7bb2ef7783bea2df0623bf77a783fb
36165 F20101204_AAANWN silvalugo_j_Page_237.QC.jpg
7c47b9241c446bd656f5cf8152e495e5
9c785c6880aec0c2ab01d809b1be0e7119c01786
137351 F20101204_AAAMMJ silvalugo_j_Page_242.jpg
79156752c49f65fcb7f2dbd371877ee5
34ebe2fae2afcc5606b458e02cf17bedad3c7633
16970 F20101204_AAANJD silvalugo_j_Page_141.pro
e405bc05d1bed109eb14a6e23037b686
19a1c94ad15697a7ea7a19d44783ad6ab49611cc
F20101204_AAAMZV silvalugo_j_Page_127.tif
ec3770327d9e3ba34c6ecb718cc256e8
7e860d6837a90613808dcddca653d30a51e9c04c
33659 F20101204_AAANWO silvalugo_j_Page_056.QC.jpg
d544c54bd46ae9fd3cfb4fee29bcd566
8e5d2ba5b70020144ffe1c99c281024353d33b16
125589 F20101204_AAAMMK silvalugo_j_Page_245.jpg
e2ff6b6696de60322c00b2c120c91247
ee3d63b88b14130032d64dd110858cc99b638acb
26865 F20101204_AAANJE silvalugo_j_Page_144.pro
df8b1f6ef591a1283efd319d4876db7c
64c48a6c28ce9a569ebd0afd46deef826116444a
8601 F20101204_AAANWP silvalugo_j_Page_052thm.jpg
fb29d44c54a996b94d7f6abbfdfd0ec0
40475f05c13782ad24f982372a7e3034eb0815e6
38851 F20101204_AAAMML silvalugo_j_Page_247.jpg
408bd526d0446c62babbd78e8bbe9350
50934382b1419126f6c89b6b9cee2036d0e87bc6
F20101204_AAAMZW silvalugo_j_Page_128.tif
e62b2b84aa1bd88829c3df8270f43bac
ebd771e854451299fbe389a57d8530ea469d9809
40848 F20101204_AAANWQ silvalugo_j_Page_111.QC.jpg
f16e449a4708862ef3f04916cbd5e195
da160514bf6f5513a351614dab24c66ece4a3bd9
100623 F20101204_AAAMMM silvalugo_j_Page_248.jpg
aa79e9dc4cc761a0cc7c1f8a1dffbb47
87f7f818c8f1c969552fc184b1bc97e2ef92dbd7
20897 F20101204_AAANJF silvalugo_j_Page_145.pro
5eea4c64c9acb7d2ce1f8abd6c6c285d
f416f81256ba9247864450d67b5c2494cfa5002f
25265604 F20101204_AAAMZX silvalugo_j_Page_129.tif
f130b6ad7eb21a07712bb2ae570afbe7
7ae78646cd015385a4ce7ab486e9e0aad3c46b14
9181 F20101204_AAANWR silvalugo_j_Page_161thm.jpg
52367babbd919252a8fc2c4403778e20
39ce5e34a028ce5172ff720622863a2a4147ffca
15871 F20101204_AAAMMN silvalugo_j_Page_249.jpg
d978f69222a6f10b8e5ef50551253628
87c85b083eb7c1128d6fd0e73e036890a32cad12
12964 F20101204_AAANJG silvalugo_j_Page_146.pro
31ec00e50612d951cbbb28a5f147d2da
a497517b0c7ea045f6869b65f3d4435766063ca1
F20101204_AAAMZY silvalugo_j_Page_130.tif
eec1e7697783c8110be0ad1d540b6035
33bf61e12a1abc39cbf2c8adaff591166e0484b8
5963 F20101204_AAAOGA silvalugo_j_Page_233thm.jpg
4384855896ac23a4039c00edf41f7bbb
b1917cee8fbbbd76b9ee56377024251ab1fe0c82
275280 F20101204_AAAMMO silvalugo_j_Page_001.jp2
1359ec9bacd95a05fc6a33ffecf9fe44
90240f3414e7605b139ce9de712921c6e4c08e26
29582 F20101204_AAANJH silvalugo_j_Page_147.pro
e149d9c46d23a8f47f35a499f0ea6872
58292be8c856e49f7369b40b329feeb168d1537f
F20101204_AAAMZZ silvalugo_j_Page_131.tif
295a1f0e7269b827b01d4935c257dc9a
197cfddc9a5d1399f701255da0aca23b6a05dee2
7410 F20101204_AAANWS silvalugo_j_Page_006thm.jpg
6ca00c47da6283eda5c93cdaebb28346
1bfe8489c6d9ceb4a9a21a07f79cccda51b986b1
100607 F20101204_AAAMMP silvalugo_j_Page_003.jp2
c8e101bf88e5c4b3a920c6484a5cd538
2c8333bd150b249adad16a736013b3ba21987612
10047 F20101204_AAANJI silvalugo_j_Page_148.pro
2ed203c5a2a08e4c23d1830725d7a19f
83f5bb7ec5bdfaa1b31d8f61c37e906374b926f3
14548 F20101204_AAAOGB silvalugo_j_Page_122.QC.jpg
8bb6bad299531120067ce375357691cf
51912f33f9ddbd6884fc334a27f08c2da10715f9
8925 F20101204_AAANWT silvalugo_j_Page_184thm.jpg
56996fc1edf61319879b56aa8f1d125b
adf03f07ce596b44a1e03d9c86af75f5da79d88f
F20101204_AAAMMQ silvalugo_j_Page_004.jp2
8cba027349de9db40c12d92a8e0c6dbf
39981dfb2f1d00f1e376e4f6bd4089d2693f6e39
12193 F20101204_AAANJJ silvalugo_j_Page_149.pro
d59af25a5fa799b6dd5f24bdfaff99e5
685852d23c90d356eafbd7e783ef48692496209e
4927 F20101204_AAAOGC silvalugo_j_Page_249.QC.jpg
2c7444a07dda08e5ea6b13c6f254c2f7
32faa207710d2d15ff765b0ae2dd1a78370a12c1
40040 F20101204_AAANWU silvalugo_j_Page_125.QC.jpg
5d65cee09ebda3db52353ae76ead5744
b4a9c733bfc39466c92adc2c70e7056b109c8128
1051978 F20101204_AAAMMR silvalugo_j_Page_005.jp2
a919190d27123e3e76bc2399dbcc0ebb
0a834f6bf760187c2c1a270ab569b506ffe0c09c
9757 F20101204_AAANJK silvalugo_j_Page_150.pro
d1dcc557ca83b87a66092af308c5a9c0
00d02af0115df48957da171e55191817912fcf9d
8632 F20101204_AAAOGD silvalugo_j_Page_168thm.jpg
f06623c213528c43f6d30b64ee8da3b9
286e57c5e35baab047eaea1559365fc758884bee
5338 F20101204_AAANWV silvalugo_j_Page_210thm.jpg
cc4ffd1dff3f81f53099d9fabdfb211d
7717e05d2c96d03f764f35230b3d94843e36d445
1051981 F20101204_AAAMMS silvalugo_j_Page_006.jp2
c348b7b7dd04679bd5925bfeebbdfa4f
d8c84a57933de47b3652a291d27abfef2d33e197
12371 F20101204_AAANJL silvalugo_j_Page_151.pro
8f45d7e02ee99536bc0b2a9af6d3fee9
c20893831e86153303347d3c2c0b0f480a00fbde
375461 F20101204_AAAOGE UFE0022089_00001.xml FULL
3fad6e8cd40255e90d8fb30c029bc740
85f27468894641731238fb0eeb09aaaf519c062a
2502 F20101204_AAANWW silvalugo_j_Page_130thm.jpg
cbeffeeb6dc6a71581fe37d82916aa6f
c73afb40821197bc5a32d7fe1c8b8d563f0595a9
F20101204_AAAMMT silvalugo_j_Page_007.jp2
af8b0fc8cddc9968a55bb752ebe76ff5
d13f9f87e956d046e90f2c1e8282ad336db8a3fc
8615 F20101204_AAANJM silvalugo_j_Page_152.pro
ab26b53e4d33f55e724e39b170a13641
17c583b77bd4b650473b266aebee73db079340ca
8950 F20101204_AAAOGF silvalugo_j_Page_001.QC.jpg
0d7ed993174561cd3466bd9138c8e5f3
9b04561a27d05344e484330f82b1d437f1fa70dc
3930 F20101204_AAANWX silvalugo_j_Page_219thm.jpg
d108a945957bb35e4b9a23cb4f9b157a
4cb54b8bab693afbbb23486d594b6707d22c26e4
F20101204_AAAMMU silvalugo_j_Page_008.jp2
ab497adebcaae386176e5064effe81ea
8ef9328200f6534e3cf5bb6d8c6367b057332e79
10892 F20101204_AAANJN silvalugo_j_Page_153.pro
a351b4bca0d20e147d03709751dfc94b
6ea7739842a1bcb8004e246993d2fd67788df5f9
3023 F20101204_AAAOGG silvalugo_j_Page_003.QC.jpg
fa305cf638f930a0fadec880d041b938
beb1c6621bf9a6a8680c8fb751e2351665ebc107
30253 F20101204_AAANWY silvalugo_j_Page_171.QC.jpg
46db0afb66284f0116d3a64d9bf34815
e6b8402cc1282dcc778ead209ab18d35b7e3c180
1051972 F20101204_AAAMMV silvalugo_j_Page_009.jp2
a327b7b425eee8cef66fba9e3305f526
e97095fb95f970de96644e0f6562fa84fc6a5638
18899 F20101204_AAANJO silvalugo_j_Page_155.pro
4179bd4b570eb3b3acb8088435fe7629
e58a1584cb8d3c3b84d50830006d1e046c20a7a3
935 F20101204_AAAOGH silvalugo_j_Page_003thm.jpg
798d446962721af01a0b0a6bc0372b9a
8ec0f0d385761d2a42466ad609ce1e2cc0193ac8
19092 F20101204_AAANWZ silvalugo_j_Page_138.QC.jpg
a0e84c04dfda2bd247fb9105c6390ab0
dfb7636074dd96628442eef98b7142ec6f76ce1f
200630 F20101204_AAAMMW silvalugo_j_Page_010.jp2
0415905fce7e7f8cae79d8ea6f67af6c
247537d15b4b28b8168b4e90e6b1a9c22c1abd1f
13289 F20101204_AAANJP silvalugo_j_Page_156.pro
4de4300f87fcbca917ae2cb69494c8c7
1e613bcbbfb5a903518ee782517828a275e0486e
8057 F20101204_AAAOGI silvalugo_j_Page_004thm.jpg
71d21a62a5a9ffa7312fc16c7c330242
b5292545a3d075e2c29849739385ca77e18e4317
52010 F20101204_AAANJQ silvalugo_j_Page_157.pro
27e677e0a5f7eb751f07bcae157272ce
72171416ff59a68565eb1ef3598faf3ae1e55d66
26588 F20101204_AAAOGJ silvalugo_j_Page_005.QC.jpg
1951ca5aaa938041085a83ac4e818a9d
2ae6bc1c18558cbb31f8f3bd27c7678df255b962
F20101204_AAAMMX silvalugo_j_Page_012.jp2
a3bbca1976b2a7fae1c17385f15ea775
592253b7c0d3c092c8d2c501c9473fa2ffee42b0
55963 F20101204_AAANJR silvalugo_j_Page_158.pro
1a613e576fc9ba33fcd811de08c27c42
3df132f796f32ba459de7198d344cfb89013c0b7
6648 F20101204_AAAOGK silvalugo_j_Page_005thm.jpg
f9fe798dadafaaa43a3bbe5d28f6d784
41587d499e4249bea6506967900f35b726e1b223
1051983 F20101204_AAAMMY silvalugo_j_Page_013.jp2
997be831b9df78f4ae2a33ed146a5924
13cb0659a2874f8705445d237f68b64b8b12d853
56639 F20101204_AAANJS silvalugo_j_Page_159.pro
6d8770081e98750590001c58ddc00439
a3df6aa468f3faab2100f7dba5edf6af6c34d3f4
44679 F20101204_AAAOGL silvalugo_j_Page_008.QC.jpg
10f985e1e6c2e1977e48aa1046619f1d
aba0da9c54b7ecd599fb855fcdc48d7e6bf79c29
1051966 F20101204_AAAMMZ silvalugo_j_Page_014.jp2
a6ba21a9deba0e280898e28812295c37
adce7d60b3e60e0c088914657d87d00874147ffb
55275 F20101204_AAANJT silvalugo_j_Page_160.pro
d2e3af087999e95abb1f71a9b534ff94
6a220b0c14017f82d6859c3f41d2319245567141
10361 F20101204_AAAOGM silvalugo_j_Page_008thm.jpg
dda519682388ab9130faaa2c90b82c5c
ae13d957939adf21a33c22ef6ca7d754bbaa52df
56865 F20101204_AAANJU silvalugo_j_Page_161.pro
caaa14ea0fb4e9158be447af81cc1429
a0b661934a7cf3ed54c73f2a3036864e2563bbd6
4437 F20101204_AAAOGN silvalugo_j_Page_010.QC.jpg
d19e96d4015479b2b359b68061032608
7af37a69cc60b58745138926adcb7190a5043702
55961 F20101204_AAANJV silvalugo_j_Page_162.pro
21e94e0912da24ed2e1e8b755ac1d042
851c994f043665caa9e70b7da756d7423ee86c25
9724 F20101204_AAAOGO silvalugo_j_Page_012thm.jpg
254afecb28deab132771f9467530e9e9
f3844952cde52dbf894b45114da576d5f3a1cee6
56831 F20101204_AAANJW silvalugo_j_Page_163.pro
f87ee1093a60ad73619d3fc68e753c45
7e800efd03482f7e25b6d41d95c7875f4baccd93
1579 F20101204_AAAOGP silvalugo_j_Page_015thm.jpg
442c7874843d22d637b2f0f9eab6064b
4d7569547723d22e2bde765c0844eeb82ee6cd48
55645 F20101204_AAANJX silvalugo_j_Page_164.pro
b909eb24ebf0d79628766c48e92761f6
19fc19207a894b90aa9a6d6f440ae1ab1afb46c7
7996 F20101204_AAAOGQ silvalugo_j_Page_016thm.jpg
df9bec1cb1335a50242549d244a34aa7
934cd0d76df0c067dcc284b7fec535448e57a122
50229 F20101204_AAANJY silvalugo_j_Page_165.pro
8220d8dc439aa01fc9bbaddd230b052b
0b68a6e5e4e8af538d6f3f2e3b22d61ea0815874
35135 F20101204_AAAOGR silvalugo_j_Page_017.QC.jpg
4273bab7a3a790586380af4e7388e401
150757be8331926654495b931bff749b50980a1a
54733 F20101204_AAANJZ silvalugo_j_Page_166.pro
b5ff2346f367789fe4af582ac546e73f
78a6d50a2e10f85a445c01c2577f205bd06909fe
36090 F20101204_AAAOGS silvalugo_j_Page_018.QC.jpg
8baa1e897973ba34fbaa43fda791aba7
275e726bb9bf297b4a48098cad208b5ca0974a85
35270 F20101204_AAAOGT silvalugo_j_Page_019.QC.jpg
4bdb791fa871ee4a2d39b78255c94433
243d3264a083a7d131d79e33c04ab5374d124bfa
8872 F20101204_AAAOGU silvalugo_j_Page_019thm.jpg
b30180daf52ceb6b890b7ff5e3b0fe48
5043f56a2016d64f0438324a0a5f527253372c97
337846 F20101204_AAAMSA silvalugo_j_Page_150.jp2
1db39d9f0fbb6817a4f9286a8e0802e8
58448c66715339e8524f32dc89e79ddf602cc903
F20101204_AAAOGV silvalugo_j_Page_022thm.jpg
ddf7256de14ea9dbcc5fc8cbf56da0c0
54b5a15521430e73b25a38895ef5df8688227405
331846 F20101204_AAAMSB silvalugo_j_Page_151.jp2
0d1c49239cd0f55cc14868252375f96b
4585add93ea3fd07f9ad3d949ea825da8efbfe3b
7804 F20101204_AAAOGW silvalugo_j_Page_024.QC.jpg
d944dccbf3a408a819fa626bc343554f
d7c20392a78a34340930699b07c9bc7cfe9fa1da
357330 F20101204_AAAMSC silvalugo_j_Page_152.jp2
4473ade6e95bc99bd578003127fa3995
b39200fbc61cfa280bfdf187bfef569048a81107
2197 F20101204_AAAOGX silvalugo_j_Page_024thm.jpg
3f9cd1f2c66196df1401ced686ff4ab1
dd609b50c7ffa1ca4323be29e2ac9ac0f5c6fd31
318192 F20101204_AAAMSD silvalugo_j_Page_153.jp2
f9e2391a398dd8b5ea01a0a1309c4a67
5415ff29ed57d427d71a1cc6eb20425d885f252c
19784 F20101204_AAAOGY silvalugo_j_Page_026.QC.jpg
cbcf7f20e276d198013c9b1bac0f09ce
462a646f11a9ec3a125ed52e93d6336602786ef6
343207 F20101204_AAAMSE silvalugo_j_Page_154.jp2
ab117940e2baa5a44fde41bab0edad58
409241ee59339424020737a3de0e1159a0d158c8
5808 F20101204_AAAOGZ silvalugo_j_Page_026thm.jpg
04c65414cbe7303b5483ad292a7d999d
e5ec34969e9e4dcb0da6f47362392ebbd4bf05cf
384720 F20101204_AAAMSF silvalugo_j_Page_156.jp2
a87847a6829435349605f32f80b86338
2dbfb14d3e19fe30d6c3c6ae3fe958671021255d
F20101204_AAAMSG silvalugo_j_Page_158.jp2
ce19ccbc838a39b3d74f955a95d3d78c
5b8026f11248181039da87653522abab6b1efeae
346 F20101204_AAANPA silvalugo_j_Page_057.txt
eccebe333270be8f72c84d03b755ae19
9e2cb58cb7f51e361c5d1fef5a5e181e9c6457af
F20101204_AAAMSH silvalugo_j_Page_159.jp2
1141f72c980754cb1216e6fd674a932a
6aa858500e0d73b3c86dce3ef6cea2cdd3ac66be
232 F20101204_AAANPB silvalugo_j_Page_058.txt
f7d53a1cf377a107e445bef6a3de5085
51162f84c032054db6104e07abac12b4c9230c6b
F20101204_AAAMSI silvalugo_j_Page_160.jp2
a01c1e5079e3c2998e2907c376b1c3c6
a369295c9613ba335d6f8c8119729dd7c8bfe1d5
319 F20101204_AAANPC silvalugo_j_Page_059.txt
88f695fb37f4def959d77caf1948b102
abee233f1aa19db5f65ad05e5fc302b90181c7a2
F20101204_AAAMSJ silvalugo_j_Page_161.jp2
803c458037a714701218b8882cd91a47
bfee7fcd233e9846e0761c608081c1def8425722
324 F20101204_AAANPD silvalugo_j_Page_060.txt
971abe9c89a04e0c474e1c01e5e6eb08
dfde553fcba47116f9a751b62f823ddfd5f6a2a6
1051967 F20101204_AAAMSK silvalugo_j_Page_162.jp2
5b0e241e6bb50f8a8c6c0a419b53604b
772abd7eabf976e69dbfd125ebc98fd499b5b129
301 F20101204_AAANPE silvalugo_j_Page_061.txt
92ed73f5c5e9605e9160a2b2876cff65
66b0fc6a0fe1fef3777d069e62744c11ee5b3b2a
72379 F20101204_AAAMFA silvalugo_j_Page_034.jpg
41450a65675f32e986e7d13c550feb94
5d92b2c36f6e61dec316e8230ade7729ae43ad27
F20101204_AAAMSL silvalugo_j_Page_163.jp2
abde7b3f172e1a40e8e8e24ca6c10ede
5cfa30397161f475f0da89f1600ff6bb8d6d4e9e
696 F20101204_AAANPF silvalugo_j_Page_062.txt
b053a1f6946b307ec7bda275f82bea3e
526afb0929aa141d9f341f25ba7aa90bf411e62c
109558 F20101204_AAAMFB silvalugo_j_Page_035.jpg
dc8e9a631bf59bad35bd14defed332ab
a609f1b120cc13ac28bdf049281aadd9bd0adc66
F20101204_AAAMSM silvalugo_j_Page_164.jp2
4bacfaef1083048510e2fdfb4b67629c
7b74886816fc13a475af1c109181e9461fed4294
579 F20101204_AAANPG silvalugo_j_Page_063.txt
4c8d1ce85a3d8b922fe2fff87d33dd5f
9f9e0f5ede297efac75cc637c06583f8a6f05e7d
F20101204_AAAMSN silvalugo_j_Page_165.jp2
ad5a568404d4326cdc18ec34496fbf22
320cf6bf2d885e9d07f9dd150f5a42b58ffa7cd1
9089 F20101204_AAAOMA silvalugo_j_Page_172thm.jpg
89176532977c11f79eaf78699ba099c2
a6274e483f2aa0d189929d0f17fa578b10a8ec4d
339 F20101204_AAANPH silvalugo_j_Page_064.txt
9bbd5a6f9981c4722a160b19dffc1859
baa01bb781c07707b24a596e032482d183cc5fed
112780 F20101204_AAAMFC silvalugo_j_Page_036.jpg
d8baee84f3aa7323c1f7f071351c2541
b092afe4fcfe61429080081c1b34d34c1b38f1e8
1051969 F20101204_AAAMSO silvalugo_j_Page_167.jp2
92fc3d0182bc9118db828c26c7f3ed8f
8361989ca5a046779bc462fa843f50c49b22d0d4
7711 F20101204_AAAOMB silvalugo_j_Page_173thm.jpg
5f0f68cf95d2678f920d4930ee0469b4
0081c969da27a7e108dfaf9d7b58cd4cb8638a6d
2155 F20101204_AAANPI silvalugo_j_Page_065.txt
5f63810d775bb1d20b1f80e8b9fa9a41
aa69c473e8a47934acbe2bbf6cb314a7e9a26a3e
88108 F20101204_AAAMFD silvalugo_j_Page_037.jpg
e8a81ddd830be16bb0e5b49dd7dce444
c66ae49211dcd2ae544b20fc7d27814be6776218
34938 F20101204_AAAOMC silvalugo_j_Page_174.QC.jpg
796749ed5ac07f52314bcc97e0129b32
dfc3e1fca9a6a7a7b737e3ff9f866369f5dfe4e9
2191 F20101204_AAANPJ silvalugo_j_Page_066.txt
105ae00b9976495f0ba08a3b4dbdf1d6
621685a423d35232c858cc70717683897e03eb4f
102877 F20101204_AAAMFE silvalugo_j_Page_038.jpg
e51daba805da2d36f6d2c61497dfbfb1
8ed5f45c3cbc2455f69a4f378398609a6e9a4fdb
1051973 F20101204_AAAMSP silvalugo_j_Page_168.jp2
204a7d90997adade1378b3041005914f
20448d6b560f1b445af6accfb56cfdcf2e1008a0
38862 F20101204_AAAOMD silvalugo_j_Page_177.QC.jpg
ce1048f88ad158191c3c305dc6536c6a
482f661a2cc02efa7a1a05ac614ce1fd4e7d6fd2
2076 F20101204_AAANPK silvalugo_j_Page_067.txt
da570b39a23687a67e3a758a70525930
f7df27da382d0beae80ccd51681521c00c1e0d11
94082 F20101204_AAAMFF silvalugo_j_Page_039.jpg
d5c5e68c79f9823da42c42639c5773cc
fbafb264817d6fd4de761d6b7682ed5dd984420d
1051909 F20101204_AAAMSQ silvalugo_j_Page_169.jp2
640d13fcf1ac1b99b37c10296b959e23
af3c73be91b56693db62f6b5f4ff0be9d303d22e
35910 F20101204_AAAOME silvalugo_j_Page_179.QC.jpg
4a8e581dcb8dd7b6d1bd478d879eddfd
801a09fa1dcb5d8fde35f55733a7916cd4f1ba10
109920 F20101204_AAAMFG silvalugo_j_Page_040.jpg
b3af88fcb66dc7361fadf10d102bf874
3bf14a167c18787f0d19a9b140e126b19dc2a47c
1051946 F20101204_AAAMSR silvalugo_j_Page_170.jp2
c5c8e5d4d5e02fc2cf0d168d38ce163f
508f3d8d5f7716c334b9f4434b5125092f6214e4
8520 F20101204_AAAOMF silvalugo_j_Page_182thm.jpg
57c795b6ad2d6dd4db18df70d3cff148
ae17a9dca97ed55331787302f113156c13e5ee7a
2149 F20101204_AAANPL silvalugo_j_Page_068.txt
7ba2a3f0e4356c1d86466908f257fed4
857d0d6b3e46fc8212891ea971c8110539376fea
111263 F20101204_AAAMFH silvalugo_j_Page_041.jpg
0621297d876743dce82f22c2c395bf24
b7aacbd22fd20c74e4fcdd0d90f9b7102f2971dd
F20101204_AAANCA silvalugo_j_Page_191.tif
c57c2ccf80229cdfd9e88fd6ffd00a87
265c3fa6ae131f2f072c7da5a8fc64c060ffc389
972952 F20101204_AAAMSS silvalugo_j_Page_171.jp2
6ad98ba61e8307dfcb5658d53e8c280f
5b9fb5c3244d323bbaa135bfa4cd01dbacfb6045
36833 F20101204_AAAOMG silvalugo_j_Page_185.QC.jpg
36096bdc5de53e991c26e8384be292ca
4be6324713834fd9589c077c12d6d1d2d3d24f85
2262 F20101204_AAANPM silvalugo_j_Page_069.txt
86d45b151bfbafe781f436da72ec11b5
19c20e2496b9cbf4124d30d761760ef7282f9dc2
96717 F20101204_AAAMFI silvalugo_j_Page_042.jpg
9f54faba0d9c78f148ae01ea1341727f
c76553224ba7cf41dc7b714965f6e15724f490f5
F20101204_AAANCB silvalugo_j_Page_192.tif
e86303603a514a1d55b2320325573f25
5e6d216b1c1f334de9b7c40f4fb7c9ca04f33509
F20101204_AAAMST silvalugo_j_Page_172.jp2
96df87d29dfa96dc6f3396e39828671c
5459f4d40a2001352e4b9ec34a10cf8e16c4494a
1900 F20101204_AAANPN silvalugo_j_Page_070.txt
4536e6fa62c5e798a1365218f25f2ecc
4ca9cd27f58b310c3bda9ac136c432cdc4559157
114568 F20101204_AAAMFJ silvalugo_j_Page_043.jpg
efe1210e32a6f48e1ebb6c36e7666fe3
a714d933d6773ea0e1a046cd2243d6e123a9e0cd
F20101204_AAANCC silvalugo_j_Page_193.tif
f078b16b1b5921993555e5bb2fa31375
ecff185d33a26005ff83984610578f4522cd6052
1034960 F20101204_AAAMSU silvalugo_j_Page_173.jp2
bfdc4879947008457bb8862efa3e51cf
ae7833546ef8f83e22e54907584c02321a52c811
8754 F20101204_AAAOMH silvalugo_j_Page_185thm.jpg
bcb1b3677cdc623a1e0c4164e760df4a
e989fb2a3de57b2eed8d41c194003f29b80704fc
2618 F20101204_AAANPO silvalugo_j_Page_071.txt
7e90d9f6894decd0a85b4239394c86ee
19c30d28af4ac13d34e75b683f00bc3899004666
107784 F20101204_AAAMFK silvalugo_j_Page_044.jpg
f8ede2697eefd967ccd360264f730347
ae50f52167ae2c59e8b9a3245d7b63c408a210e2
F20101204_AAANCD silvalugo_j_Page_194.tif
919bd1936596f89829d0e4a57eb03750
d88f46d511ec1d0513c93fce62a7a373dd146a05
F20101204_AAAMSV silvalugo_j_Page_174.jp2
271ca2f8a477a199d5da1a905bf82242
399fddbcba5f40243b5b8b34aaee8bae59ef40d6
37442 F20101204_AAAOMI silvalugo_j_Page_188.QC.jpg
2f320204bad90556a1c71a436c87bd3a
045b1b41cb11a6f10f7e9fe3d3802ab4db4190bf
2186 F20101204_AAANPP silvalugo_j_Page_072.txt
86ad85cab00c499e8dd7d04396634824
26b94e9cd44162a0207e4bfd01b8518323b24ff9
103194 F20101204_AAAMFL silvalugo_j_Page_045.jpg
a90dfac7b5f20c636ff896c08f785b7a
f9456632d40de366ab6184b8af69639de0288c5b
F20101204_AAANCE silvalugo_j_Page_195.tif
b84c76e9b8750d6800acd6fcf92e18a1
3b6caee917d295d65b3cc37a1e02cd12cb02874f
1051944 F20101204_AAAMSW silvalugo_j_Page_175.jp2
6bac8dd4b60188b8447910edff1ad6b1
a9b406e2eb5686af2c26124712b8f9d8939fd17c
37369 F20101204_AAAOMJ silvalugo_j_Page_189.QC.jpg
a7800359aebd1a6d4e2fcbba225bb0fc
71302c16ad2f1b244efc33818af040a4904df6f1
2103 F20101204_AAANPQ silvalugo_j_Page_073.txt
cfa05bf01fe6b192c6164fd830d46734
ad8ab6f6c9b026f2cf4b6676ba56a8580b5ef4a4
12855 F20101204_AAAMFM silvalugo_j_Page_046.jpg
622ce920bdec7a5526f7c96f9fc8d9a8
13dbe983824b9b685d2b02d630aacaaf076942f2
F20101204_AAANCF silvalugo_j_Page_196.tif
4e20784d15637a68373c6eccd37dca58
e148e7c9cc3074f52f90556e06cff2572dae393e
1051976 F20101204_AAAMSX silvalugo_j_Page_176.jp2
f299e8f2151a2dcfec3e99b1656d8485
c9daa6dad9a8e22e6109d0f7b6e3f32993601f75
35474 F20101204_AAAOMK silvalugo_j_Page_190.QC.jpg
5f5b958a08950424e4c2711be407c0a5
f9fb5182bdb5c03cf69855ae2433eb9814dd093e
2189 F20101204_AAANPR silvalugo_j_Page_074.txt
d1335343291b576152b366a718acb338
e30f0eabcc551d24c61de8d7ced6905dce0803a4
53861 F20101204_AAAMFN silvalugo_j_Page_047.jpg
7b78c8249026afa36d1deaa6d0f2e0ba
414113229833292fe195b91197a682cebdc46979
F20101204_AAANCG silvalugo_j_Page_197.tif
68ad5b7548513b17af529a332df7030d
64399d88d51d70aafb2167bcb0b5f5e0010554b1
1051901 F20101204_AAAMSY silvalugo_j_Page_177.jp2
b5bf70675a203876668853309d5785c0
13ff147e0339fa77f4786c958f727f944d733322
8837 F20101204_AAAOML silvalugo_j_Page_190thm.jpg
b4df7a50b1616eb4fb969bd4da505b03
4d560526d8122f89c03210505d85e4a0151e741f
F20101204_AAANPS silvalugo_j_Page_075.txt
83d239be46b0f225e95661407b0c47a0
e462eaf67dc8c78332f267d08d04c80a5442f7cd
44788 F20101204_AAAMFO silvalugo_j_Page_049.jpg
4224171af5fac2e3907c1c281dfa0b9d
2b92a58dcff812f5fc458568fab2001baf39924d
F20101204_AAANCH silvalugo_j_Page_198.tif
6e2f82b2a7cb63e452d1cc56f31ff4a0
349bec492a581a95385440e5d4ae449a1fd24c52
1051927 F20101204_AAAMSZ silvalugo_j_Page_178.jp2
e83917c26eebf6a11a22385fc287877d
177d654ec8a8fc4be53f3f7757336208b884920e
37391 F20101204_AAAOMM silvalugo_j_Page_194.QC.jpg
5edf65a610a41e0e6d163f0e9a0efd18
3e37ce8508e188aeb47ae5b615139467d1e095fc
1850 F20101204_AAANPT silvalugo_j_Page_076.txt
aa2340227c7814cf9772c971ea7935b9
ad95baee44545a7ec785396e6169d37802ed959a
74481 F20101204_AAAMFP silvalugo_j_Page_050.jpg
6ff4bb2ab477313b4d4822db7f1ef747
f51cba004687e573cf82181fb22b1d60e832dafa
F20101204_AAANCI silvalugo_j_Page_199.tif
fa56d3e2a7a5cf6cf61e04a6631cc098
fe653cc585a854bdbc241488ec9cbffb669ba6dc
9288 F20101204_AAAOMN silvalugo_j_Page_197thm.jpg
422066a466189f085e2ce0d0f4f9c23a
523cceedd25db39e41712a79a71a13a4d4be2138
2223 F20101204_AAANPU silvalugo_j_Page_078.txt
5a26eea7d3707787c4fee9548608ad76
48437f22d3b175dc96a702a8ef03839726ce069d
100222 F20101204_AAAMFQ silvalugo_j_Page_051.jpg
b4a705c6768e34f038da4699e844f564
04811bab25131670b27943b8359e107ba9879983
F20101204_AAANCJ silvalugo_j_Page_200.tif
1b3d91fdd71fdedc85341bdb72a86468
0c4bfdcfa3948fb154053a7b352e0373e23aeb3c
37254 F20101204_AAAOMO silvalugo_j_Page_199.QC.jpg
274d9c343b2a6086c8582386f64c190e
080d892398bd8e5a25e05ad60c3b123bfdb01108
2141 F20101204_AAANPV silvalugo_j_Page_079.txt
d82efc2a27d13e2fcb0771a26461763c
cd8821dc37e7c4ad18b3076e769e7fb1736d59fd
110970 F20101204_AAAMFR silvalugo_j_Page_052.jpg
39a68828b13413adaf855d541680539f
be9f4d1a28ade7713d3cdeb452bc1bc1d1d6d2fe
F20101204_AAANCK silvalugo_j_Page_201.tif
0890b6726c41865947a8983678ffc0de
2f3781bad4fa72be69553809431fa439822c7388
8886 F20101204_AAAOMP silvalugo_j_Page_200thm.jpg
8506d21931638114c92b87af634bd27b
ef8d6e1cbe4ac4d1e879085cc8f9a595b11f8fed
1941 F20101204_AAANPW silvalugo_j_Page_080.txt
afa516eae09af8fd0d1bcc52ec6cc3e8
d7aeba7d8ea09eb87f2dcbc890a5be824a04a74b
118493 F20101204_AAAMFS silvalugo_j_Page_053.jpg
15d427ae94f0b077a1c45c26531698ff
d837632fc060c4c04be08e121220149f8c16a146
F20101204_AAANCL silvalugo_j_Page_202.tif
6436fada1d3bbb235025d4dd0514fbb4
6fbbf7b7afe53d3eb7d7d1fa975726f511696916
1639 F20101204_AAAOMQ silvalugo_j_Page_202thm.jpg
06f3ca14123580862f56b3a1c7a92c2f
6f61a8e36621f1f34e81cffa77b9b6c40e57407a
F20101204_AAANPX silvalugo_j_Page_082.txt
a8cb8a2632dbb4e862b705b96d799990
28179b606dc954a08c2bf6bf2e9dc9f8ec0acc4d
54736 F20101204_AAAMFT silvalugo_j_Page_054.jpg
99d45549db20de5102cd415a249406af
122a09646b762b6f7d5404bc0917fdac01c6fe07
F20101204_AAANCM silvalugo_j_Page_203.tif
41d27c0ba8155fd321e986a8b1e9835a
70014056da5f9c9cce364068d45ead8b5087280f
19468 F20101204_AAAOMR silvalugo_j_Page_203.QC.jpg
1cb54257e1c47578e30e604addc7ec12
15af422c214c3fcbf16876ab82ab09ecefb54f5f
F20101204_AAANPY silvalugo_j_Page_084.txt
4c00db1b53d30be677ac9f083dd6b99d
d53f8018faaf6c646640a53ebbe1b052a3d0ec24
99954 F20101204_AAAMFU silvalugo_j_Page_055.jpg
0bb045b0dfa78235663235fc3e56784a
439b6be6189346c74f2b4974b76cca5254244b0b
F20101204_AAANCN silvalugo_j_Page_204.tif
abbcd2f3f3a26c7ab4faa55786be18b6
dceb8a82a5e058a9c2909ba5ca81ba940b0aa864
4724 F20101204_AAAOMS silvalugo_j_Page_203thm.jpg
c1b55799025218e294fc094d8d2d535c
b0df5ad0c7ae4862ce44fae8e978ddca6f90069c
126439 F20101204_AAAMFV silvalugo_j_Page_056.jpg
1c6e83eff912c00246f524aede9093e3
f724ccd862f2a22a8f850c77e8508cec13d3aee2
F20101204_AAANCO silvalugo_j_Page_205.tif
94a07156181c14b593b0cbd57ca9c354
2d27c5bfda442165867d04261cadc429a81026c8
2029 F20101204_AAANPZ silvalugo_j_Page_085.txt
8dab13a4181fc2036c04ca27049be17b
7b17c3a6cc25c5502c5f11a2d41de412f6804d15
27836 F20101204_AAAOMT silvalugo_j_Page_204.QC.jpg
4eef8122c1e7316e15c7c90c376299ea
2697d4915edb6bdbe790d0bc44c96d1bd2c3cbab
114648 F20101204_AAAMFW silvalugo_j_Page_057.jpg
85fbddf4ca635ce367f8e9eef46a40e8
b73c66a6d3575a5bd61a1b76465d242d5f847b44
F20101204_AAANCP silvalugo_j_Page_206.tif
f12fd69f563d6411afebf0318cd5a9e7
78af83adc99b15058a45d23fe02c2a994c7f5503
6454 F20101204_AAAOMU silvalugo_j_Page_204thm.jpg
5d1c57e823dbd11bd61c15dd38ff1c42
6043c09d3da0fdf67ac54cf372eed391c179c5fa
143925 F20101204_AAAMFX silvalugo_j_Page_058.jpg
903e88efc14abd3de561fb10bcde1345
0767b89e266140efe6e17e29429c41719a1cb1e3
F20101204_AAAMYA silvalugo_j_Page_076.tif
d1a6be627816ee2fd53e05e4f64236b0
2fe37324904d3de0a233338bfcfee34fe04512b4
F20101204_AAANCQ silvalugo_j_Page_208.tif
72e1315ee3029723441f470a12d4823f
4b6dc7bda723db97806a12c7f57274a93a0ac8ea
17688 F20101204_AAAOMV silvalugo_j_Page_205.QC.jpg
8c1d32b0687cc7a81a2a0f232172a980
bf89ae451e901f2b008da6991ddb66a60fd81d72
71081 F20101204_AAAMFY silvalugo_j_Page_059.jpg
103085a39d4aa7f73d7fbf6c823472db
907ecd2d4d54d742c75017cba6dd7c386fd14a6d
F20101204_AAAMYB silvalugo_j_Page_077.tif
8be06dd31fc5dec93b06ae204bbffb68
f35e8778b91159034902c324bcf3e0097758e11f
F20101204_AAANCR silvalugo_j_Page_209.tif
cb8fb9eaecc10e6800a73c77decb2e5f
29dd6e4b7cb0e49ad7f97e681462134624add461
4311 F20101204_AAAOMW silvalugo_j_Page_205thm.jpg
fb2913cad5e1080002588f85cc0b88b5
9a4e7810f16702ffe8368d8db8683b9d64285f41
129005 F20101204_AAAMFZ silvalugo_j_Page_060.jpg
8c3ddc85aec75bee0db94eb590adc7eb
b6cd051a5a78208ae6e93873f0e5aa4980da4271
F20101204_AAAMYC silvalugo_j_Page_079.tif
ab67176b044b7ac253d2c6c5934c6a2f
c34b0f8f8bab603c4a0aeb3ba7584d25dfca34ef
F20101204_AAANCS silvalugo_j_Page_210.tif
a6fb5cfe8401a3af36f228f7d959055e
816f86875d4e89cf1ec8eb573539126a2e2397b7
F20101204_AAAOMX silvalugo_j_Page_206.QC.jpg
9afa9812c64fa7fb7c9ad6516f2a6d02
cc89403d3a4bda030133af7f5949e6cc79aa109e
F20101204_AAAMYD silvalugo_j_Page_080.tif
0860e71ff1d092539dd552c51c138b06
070b6ea7ed73476093045a4be28ed3c7c7abdebf
F20101204_AAANCT silvalugo_j_Page_211.tif
783047f01301db9be55c8e9ae3f8c2cd
1644f9b98c3f0a611dd0eff78b6238556c90d430
7285 F20101204_AAAOMY silvalugo_j_Page_206thm.jpg
6c9af1d3070e83328338e188f91ed5ef
377ee0b182aae38eb0a215adcbb89c1adb6678cb
F20101204_AAAMYE silvalugo_j_Page_081.tif
6a17ebaa4f94f6be626ecb2e3fb9cb98
7e4f9f857235cf065116d39a453c07409479265b
F20101204_AAANCU silvalugo_j_Page_212.tif
3acf95656b6e251a82dca22547af8965
c654031246cc0437e1fa227b9225a54e9660c48d
2625 F20101204_AAAOMZ silvalugo_j_Page_207thm.jpg
a3b62e8e8309889de1e4d36975ae2691
327f77e53adff34eebc14c5d510483f4bc1af8a8
F20101204_AAAMYF silvalugo_j_Page_082.tif
63c7e9660a60e16b1a609ad3288f10a8
e4fbe5c9a0e9d2c8c9cf3295a1904063e4a6d5c6
F20101204_AAANCV silvalugo_j_Page_213.tif
de6ead4b9dcc6434b03f40f007b22e5f
b3701bc6e1271ac77807fdf711f897ebbdf65d95
F20101204_AAAMYG silvalugo_j_Page_083.tif
5c102b009ae63720ae4fd6bd69a41de0
ff15e37f64cb11b7ffca0b014ae2edd6db43b085
F20101204_AAANCW silvalugo_j_Page_214.tif
1830c60f6795a5e046d8c620121cfec2
7a90bdb71ee68c43971176c7a2091a6324588ba3
F20101204_AAAMYH silvalugo_j_Page_084.tif
3bf05ca17929dad412932856cee2e150
e0973858c6c02c8a9a0185f73af0490a68ec1b6e
F20101204_AAANCX silvalugo_j_Page_215.tif
a0bae83ec9d08bf381a99f6a57fe8044
7d7927f08674debf90e7dde935cf96a50f7b0b8a
820 F20101204_AAANVA silvalugo_j_Page_224.txt
d249aacac8a7dc84f63ebc02a24ead48
d72d3ae472eb6f17c06ca21409a48641283a4503
F20101204_AAAMYI silvalugo_j_Page_085.tif
0202525da42007a6a16e7cd8a2407e7f
3d60207b4938617bc586f319f2831dcafe403569
F20101204_AAANCY silvalugo_j_Page_216.tif
90a5837ea5eb9bcf0155260e6effce5b
bee18ae6d599263ec3916550a75352be101b376f
1963 F20101204_AAANVB silvalugo_j_Page_225.txt
18f95639103aefa3a20bb3257ca3d19e
08fb6092ec0ce404663bb6c8cdb03fab6e1028b1
F20101204_AAAMYJ silvalugo_j_Page_086.tif
402a4b388f352a1cd0a343124ed3de77
a604d13baf1cb9da52e714b431b1a5dfd3ff21fb
F20101204_AAANCZ silvalugo_j_Page_217.tif
521cfbe7af92cea2f100b36a491a5898
95350f7b4c5a3e348070f6e816563d9fa4192b05
2158 F20101204_AAANVC silvalugo_j_Page_226.txt
c3311e05a3e65b81a6ea6b49c588e575
2ef06b3c106a78c9e03fd306e40a942ccf2067d4
F20101204_AAAMYK silvalugo_j_Page_087.tif
d0ef498937426e51e7e6489d4dae2ebe
8b031c6c2dbe7a88753dc270f4977999124dc998
2173 F20101204_AAANVD silvalugo_j_Page_227.txt
33989b0c154c916b1e50ccf58d2c37ce
1fadbe897134add573d9eec8efc594b7e813940f
34826 F20101204_AAAMLA silvalugo_j_Page_207.jpg
9858d64368afae3a46a2fba5c730d820
c094e37823637ea8617beeb162dc55f0c0967a7d
F20101204_AAAMYL silvalugo_j_Page_088.tif
6c248e01343606b90b1f9b42cba09765
bcc5d1828f532369f65b0eae18af6583bbaabe97
2127 F20101204_AAANVE silvalugo_j_Page_228.txt
c4256fb8d3df02a8760ec7a73e371356
f9dd5422bb00cb56166d4432d0aa827faf82a367
60895 F20101204_AAAMLB silvalugo_j_Page_208.jpg
aec517f9c0d70258ba6b8f4b926a2c6f
c477bfb906ef130cb60e884b65857616c86da7de
F20101204_AAAMYM silvalugo_j_Page_089.tif
aa33e21e4ce76a8f1433ee811bb7a37d
d53de00bf5457876052432b58c42d389decdbca6
2269 F20101204_AAANVF silvalugo_j_Page_229.txt
c70543f3cadf599c78a0dbad4986a972
b59b799a4f92b6034a93a91ba9d9526557106485
72390 F20101204_AAAMLC silvalugo_j_Page_209.jpg
cfb5ef848332e4fec20090369199292e
1ccb31b9299dc38f27c953ab3a6f63cf653e9e16
F20101204_AAAMYN silvalugo_j_Page_090.tif
3008668a4e93ebd7224c58970af1f27d
b58f6c25df78775ba6375a1ac170e11c9ae9ed66
2168 F20101204_AAANVG silvalugo_j_Page_230.txt
cd22b10ed9feda06424c4fe380c78a19
491e93dbe284f9cd42a233b374266787fe37a260
85926 F20101204_AAAMLD silvalugo_j_Page_210.jpg
92e87707019e8071aec076a47a75226c
1b7cc501b37d5c22f6e8d46bc49eb13abf088a9d
F20101204_AAAMYO silvalugo_j_Page_091.tif
69aa0b5a804920f3085c0d0f06ff77d5
89b16e391c1040062f431e7559e79d1aecfb2814
2094 F20101204_AAANVH silvalugo_j_Page_231.txt
aa7d7407c477500ed647db693c7b2185
4bbbf06fecd182fcdf28c6d25c786a55840da0f7
120209 F20101204_AAAMLE silvalugo_j_Page_211.jpg
1d9ce5f44fb391c1aaaf64365e3d2964
7abcee4e703d5263ef3071e8dcbbd5e756315d60
F20101204_AAAMYP silvalugo_j_Page_092.tif
37a18c53203aa9e79b63697dd01f438c
b496ad8aa38374ce55a1d1568782c4bdef50d6b4
1236 F20101204_AAANVI silvalugo_j_Page_232.txt
41b421030fb82821c63a5579171ae1b2
48982822534f5934eb30d38875943e3bb18792fe
41949 F20101204_AAAMLF silvalugo_j_Page_212.jpg
b98940e31eef6e9259a94b01d18aa557
9d705ff03f4cec2868c7c5f044dc831dbf875c40
F20101204_AAAMYQ silvalugo_j_Page_093.tif
6a6d99bcb1c2f2476e00c78c0dc082b5
3f48eb6a8e1584a580415b626d3937fba19e0919
2512 F20101204_AAANVJ silvalugo_j_Page_233.txt
3c838eefeac702833515cb8f1aab84c3
2c618bc2ad1749579c66f731b3cd8420112a76eb
66030 F20101204_AAAMLG silvalugo_j_Page_213.jpg
b5de3628dbaec469a73af3be064ef83c
3267ba8ad26c13a9875382d262d50ba7b582df47
F20101204_AAAMYR silvalugo_j_Page_094.tif
ec786592a6b16d78d5d082d5a6096f48
cfc1f5d237956c9b7d40d16aed32b3244e977a99
2391 F20101204_AAANVK silvalugo_j_Page_234.txt
613708371295c63b77b73b6f4de82d8f
04d2bf3b2f6ec9a90e01c073ed839ac0f008316d
15564 F20101204_AAAMLH silvalugo_j_Page_214.jpg
3848f1ce948e83e33c6aad19fa39cc9a
528e21deec4868afe44f4a8828bd391058d4a0e9
78692 F20101204_AAANIA silvalugo_j_Page_110.pro
add417cd06f2f575054ea28ec7f280ed
1835afacfe6d2b98a9a6c4b915555cbec3545649
F20101204_AAAMYS silvalugo_j_Page_096.tif
9fd5d6abec4e964ca4e851f924b8e3f5
faacf01be627c105f7c013328d1f7f58ed710280
2583 F20101204_AAANVL silvalugo_j_Page_235.txt
b9e2616f6afc80babb8354086b3a0563
ea46b19214aa8c268633146357ab1e935b00558c
87791 F20101204_AAANIB silvalugo_j_Page_111.pro
50481d24684061607d908ee76237c0e9
7773e01444a36e10f8793ade19a2d6f0e9f8777f
F20101204_AAAMYT silvalugo_j_Page_097.tif
8a90e050a9f14b23e96f8250f0f17e5e
174c78db50328e4bd3b398674a633a5b40700ba8
2651 F20101204_AAANVM silvalugo_j_Page_236.txt
eec397061816b95eab70051b6d2647cc
5b43a1f7059e397610ca73d085cd2e6f12e1e511
41445 F20101204_AAAMLI silvalugo_j_Page_215.jpg
e713c1ea86ba45f429fd13b6219fe160
2eceabd75892e4e6206f23046829c0749c920f5d
72878 F20101204_AAANIC silvalugo_j_Page_112.pro
c52694bc1747f1854fec0c7cfb08f594
f2f803479e37292287ba9ba894a13f198f19edc4
F20101204_AAAMYU silvalugo_j_Page_098.tif
49becaeeed74df5b5ca7d45efabdc6f1
75f146665ab7ea468fddf68ee519f1db054e0b54
2616 F20101204_AAANVN silvalugo_j_Page_237.txt
343708e7175fdbf07a85125f61461c82
0d47beaa2cb3f05495e9ec14c794d11306e8bd6b
40406 F20101204_AAAMLJ silvalugo_j_Page_216.jpg
6a67f349700786256c4128d014b22ff6
08bc6de39466bfccfe0a1056a5a1b04dbbb00381
68785 F20101204_AAANID silvalugo_j_Page_113.pro
b80308d8715a4f37aaf80f996e5fb1e6
81fa37fd264eff5c46ce70bcddd26dad1f7ed2eb
2906 F20101204_AAANVO silvalugo_j_Page_239.txt
3b4910a14028bdfcf5f332cdc5d8f10c
ee98151ea8b122c7e3981f75fa3a50c029ec770b
35279 F20101204_AAAMLK silvalugo_j_Page_217.jpg
1c50f5e822d43fa6bb8608026e56f6f5
fd20b77855f63d14a4851c65ca709a98da800103
F20101204_AAAMYV silvalugo_j_Page_099.tif
935cd93bf9418c687530d37637ac23a5
bbc75fee27ee685011bf41d610d6a7857675c8dd
2742 F20101204_AAANVP silvalugo_j_Page_240.txt
78f6ca1eeeda11f486b0d3f23bf73ca1
84e94406e0c3d74fa9420938046fbefd38bb9d1d
32156 F20101204_AAAMLL silvalugo_j_Page_218.jpg
be0aad3278288e5e8cb5bc897e76e007
a503e45577f132bcdfc7bf48452f6469b3e5bb0f
74080 F20101204_AAANIE silvalugo_j_Page_114.pro
f8558c50cb95d0c84d845ede0d87173e
a6ca4318d2d4f837178b4394d4cfac8d2b692fe2
F20101204_AAAMYW silvalugo_j_Page_100.tif
36b1daf94c5f12ef3563e0613af3aaa8
7c99b24aa90b1a0c14f64621a9158f716edae68c
2685 F20101204_AAANVQ silvalugo_j_Page_241.txt
98abad3705c7b0d46641a8b7153f3bca
1844eca3b1dc64b650462da7b71c8e6ad62dc62f
34203 F20101204_AAAMLM silvalugo_j_Page_219.jpg
a14adb20cffa9663f06571d70b7211da
4f21d93621308261a2307aca49237b371ffd8630
88006 F20101204_AAANIF silvalugo_j_Page_115.pro
e225394278bc2e1c0a8a7f97ac3f8720
563783fa55ba5eb19814f3c909d958181bca149c
F20101204_AAAMYX silvalugo_j_Page_101.tif
d11dc2b09f85b196e0a2f63fb617c5e4
c234881a856aebbc954735e5baaaf2902b1ad992
28121 F20101204_AAAMLN silvalugo_j_Page_220.jpg
ec4d196dcea17de00d1cf1ddc9b1b126
034a64d62f428edd00b33f0387889aff802f4847
57938 F20101204_AAANIG silvalugo_j_Page_116.pro
8c2036c9f1d1f53d6087efaf40d26891
fc06e8ca37cfc5ebab90b2b67d88a8a9974ddc4d
F20101204_AAAMYY silvalugo_j_Page_102.tif
085c6a71fae73883c067e9527138ccc3
2566de92d8af02df1aba9852decf51a78e998016
2719 F20101204_AAANVR silvalugo_j_Page_242.txt
96942216d228d3cabade16ee63c4c51c
434c8b3c4d5640ca3703356adb0f81881144d001
30979 F20101204_AAAMLO silvalugo_j_Page_221.jpg
b164dc3681b35a9689e4b2ee85d3f615
0aeb46e9a21d364dc5a04704c85102d7b320067e
58034 F20101204_AAANIH silvalugo_j_Page_117.pro
60b075f71fd61e0864cf34385bf42db8
1828b3c1fee3cfa3180527b920b3ba1237aad803
F20101204_AAAMYZ silvalugo_j_Page_103.tif
a98ebb7de0cadf2ee635a2fe0a2fe3c0
26e11df208d34123f1aedc27f24c95c9579d7ee0
8944 F20101204_AAAOFA silvalugo_j_Page_187thm.jpg
debe4fdf2b4e3321ab5d383273110328
c814055e74d8dc4f1457421199ec6ea2eb1454fa
F20101204_AAANVS silvalugo_j_Page_243.txt
f816acf8e6045a6d80c774260a807d69
e7901d60eb0847fb79a36dcb55f8f788579972bc
34315 F20101204_AAAMLP silvalugo_j_Page_222.jpg
62d9a0bb31d5c57b3af3536434eea9e4
49de731e0faf73d92668ec684c6fd0664037219c
22145 F20101204_AAANII silvalugo_j_Page_118.pro
d859c47305e444e2ab838d4a089c4cc5
317b5ac39f72cfce2981a814308bbcb5a8595aa5
16586 F20101204_AAAOFB silvalugo_j_Page_136.QC.jpg
53b53f28140f84b2258f3f869721c658
4be68548c6c47cf9cf3a7dbfc5d1cc07444e0382
2631 F20101204_AAANVT silvalugo_j_Page_244.txt
b37caa1a505c1d3fd3ddb49f886f473b
fdc1f5bb5922b283f69e040b22a1035992f14ab7
32792 F20101204_AAAMLQ silvalugo_j_Page_223.jpg
0b1ddeb3e38e3f495c7a42c40c55d1c0
09a395ca343a4cf01db33271ec0437311768966f
23134 F20101204_AAANIJ silvalugo_j_Page_119.pro
bada3c4db468de6c1bbe4eeb79a2c856
552644bb0aca34c09c70e6889fed3484c9c89490
31069 F20101204_AAAOFC silvalugo_j_Page_006.QC.jpg
cdbffacda385f2eaf9a0b17bb12aca14
be0a44ceb17c5877bbd342c5da7f7f93d0081c34
2646 F20101204_AAANVU silvalugo_j_Page_245.txt
ff85155103d7380a3d7e77943dd17aa0
55153ab650f7cfa4309764555c4fb6d358e42962
36995 F20101204_AAAMLR silvalugo_j_Page_224.jpg
31c467312fbcc0a1b7aa15a98cf8519d
094aba58d034e8e5f2e1c426a128a4821f32661e
27042 F20101204_AAANIK silvalugo_j_Page_121.pro
6b5c25d9e6d2e9d5d9b4532e0c08e2c2
6b996fda1a22531402c5402c2b530b9159c597ad
10947 F20101204_AAAOFD silvalugo_j_Page_215.QC.jpg
bb4be7f78767fd1afeb710e383df7db5
f651910377292c92479389aaf68954e980b58281
2771 F20101204_AAANVV silvalugo_j_Page_246.txt
b76732bb4b79caf070a3796f2422cbbc
43e9c74b368419a193c40633a7f6e4cda122e93d
98358 F20101204_AAAMLS silvalugo_j_Page_225.jpg
6606ac6e8425100125fc57b933df0a12
e123895bcde81efc315a18cfda688e62d6d5c1c6
42096 F20101204_AAANIL silvalugo_j_Page_122.pro
8d5be2f3abf197b157624706f464951f
e51a9a819fb015ccea59c98f1c30c4b0f58173d4
9158 F20101204_AAAOFE silvalugo_j_Page_245thm.jpg
653aab24a3b30ec203b60193834900b9
0ae537e815a1848b03b9df3f6f7f5a2aaf6c418f
672 F20101204_AAANVW silvalugo_j_Page_247.txt
21e07d2c8ee619ece58b08f7993c6f12
779a11a2de35343cbb5df67bbc5b8616e96518fd
111073 F20101204_AAAMLT silvalugo_j_Page_226.jpg
cb5db1ef1a9d7243146a511f74e9409c
0af9b470557b11ceebf3a1580f70b257bcb63673
3846 F20101204_AAANIM silvalugo_j_Page_123.pro
d62ba52d6f8b2458ee5cdf5dc3b94649
890a317e02a9d0f605a015846fe0f6d37710f631
8050 F20101204_AAAOFF silvalugo_j_Page_080thm.jpg
999d62f417334316fa4091a492a6776d
fe7c57a5ae2b53c527feb613334e2e455bb81746
1933 F20101204_AAANVX silvalugo_j_Page_248.txt
adc2ae74bf0eb67714c80921dc8f7e78
d23117794488ea205f292f690f599783a62fdafc
112212 F20101204_AAAMLU silvalugo_j_Page_227.jpg
33327f3474873e7db2ce86c85f963b29
d71dd2ff59483aea49f2023f8438d975dbcb8040
4126 F20101204_AAANIN silvalugo_j_Page_124.pro
cf03e7c38bff827c2f9686b02a9255d5
4c1ef299bd410e49801d3ed6bb4fed7a79e392bb
35167 F20101204_AAAOFG silvalugo_j_Page_023.QC.jpg
cc16ca9b6c0879b6757d17325f4d4073
f3a82777e8c4ab2f4bdb04f402383cfbe44c5ac2
251 F20101204_AAANVY silvalugo_j_Page_249.txt
3104e7254217a107a6619272ab72dd6f
8b68d29f7cb506e95a1efea53ff0794055f27733
108193 F20101204_AAAMLV silvalugo_j_Page_228.jpg
7007e5c983584af7b5541cf74c53958d
ef39181fee9f33d6a0a4d2f96fa6730e95e28f1c
4098 F20101204_AAANIO silvalugo_j_Page_125.pro
434e108db9184566fe25f0fb57e1b9ca
1ee188a6098eb89505f3b413b510962645c30e33
13832 F20101204_AAAOFH silvalugo_j_Page_120.QC.jpg
dc1d8568486e1f60a18af3c43c45ce8a
c12d6e0fb852a271df4cfcd399f31fe9494b44ca
2139 F20101204_AAANVZ silvalugo_j_Page_001thm.jpg
94d736ae2f7359ecb12e040a622897d8
dd771dca3f2d757ee735569d351f7433997774b8
116080 F20101204_AAAMLW silvalugo_j_Page_229.jpg
c68be628d73c9d220a0145bd7335c51d
cc98110e1e900f26ec3670f6f8beca56aafb8cd4
3915 F20101204_AAANIP silvalugo_j_Page_126.pro
b669227dfd6e3384650833241c19aabe
1b89be790d34ca319658e2c1a0c9d25ae1f7fa7c
9973 F20101204_AAAOFI silvalugo_j_Page_013thm.jpg
94b98045e943fd090777d6c5e6f1eedb
e9934fd360e33f5df4461abb90b6910e48990d94
110514 F20101204_AAAMLX silvalugo_j_Page_230.jpg
bb4c01d7e7535863fb54eb6801fa7004
876fd744e47583b2f254904becf910e7666ccc40
4769 F20101204_AAANIQ silvalugo_j_Page_127.pro
716a634d3a359396e8e60207b1d4e037
04587f3f9fe334b9ea8f7b33e6ae0c9dee3a6561
9126 F20101204_AAAOFJ silvalugo_j_Page_244thm.jpg
d54a44d7591e849d614f6e41f76fd13a
18c9a1c8b2b0fe03c416a9a49304fe5216c8e9cf
109098 F20101204_AAAMLY silvalugo_j_Page_231.jpg
29ace8e31d8c158dc1b5ab126559e548
99718c83ae00b89feddedf300fac4ac614ab47b8
16129 F20101204_AAANIR silvalugo_j_Page_128.pro
acc5ee55ea98c208c415d691756300fb
82ad1c5b3cb60e482bfc723997ebad2917104d94
24844 F20101204_AAAOFK silvalugo_j_Page_233.QC.jpg
cd2dee8e9c99e20be07bc9423a677d4a
af8f196989c87f112c4b51aacd67aa59d23467a7
65138 F20101204_AAAMLZ silvalugo_j_Page_232.jpg
134eb51572a5d033522698b80a219f59
1429a1c85c5202f5d64797ca6294c96d5340101a
15676 F20101204_AAANIS silvalugo_j_Page_129.pro
6b63469f93e3c0f1a6fa1f9a395818ca
a37be3254a8075b29f5c35d99cdd9dfdb933c76c
8276 F20101204_AAAOFL silvalugo_j_Page_038thm.jpg
22423bd9c465e3cdc64c89cb6acd90e9
44c49f5a0a2dc34f8fc4f2849a11229d6289dafe
11374 F20101204_AAANIT silvalugo_j_Page_130.pro
015382653336ee8c51616e3bf137e5a8
98fe4f0d06095a56791ebc5d3ed3c4d145406cb1
37893 F20101204_AAAOFM silvalugo_j_Page_098.QC.jpg
a86987a1c10a39175646e2aec522171b
db2cb7cb70847045b80385056daca3059c2a3113
15101 F20101204_AAANIU silvalugo_j_Page_131.pro
e9583abb4b758cfc71a73fba2ff7ae71
c029c95fb8cb0c69583dd6706f65ae82020a6837
12386 F20101204_AAAOFN silvalugo_j_Page_151.QC.jpg
5067f1dad84dfef7e25da87bac9ccbd8
04f7dcaf42be1920fe39291c72289aab8f8c4a1e
10611 F20101204_AAANIV silvalugo_j_Page_132.pro
f1028028c6da2caeb239a2c3416e2511
ef9a29953f7137d4bba2451d195015e2400a4660
8946 F20101204_AAAOFO silvalugo_j_Page_056thm.jpg
1b9807bb457ebb5791b6f364a580edd6
bfb9a94a09268d9864af0b87c4ed58118fd1fa92
17449 F20101204_AAANIW silvalugo_j_Page_133.pro
9c518a9597d1e741a4f2cb634eddc253
e5905c636ae3e4d50015af57b5fe485e20c68096
9180 F20101204_AAAOFP silvalugo_j_Page_234thm.jpg
184e9087f5f10f4d051186512359c6e4
391ba751584253bac2962c58bb8eb14954e1e682
11623 F20101204_AAANIX silvalugo_j_Page_134.pro
4da0c668668147de407f083e9b09439d
35fec12e1912c0674d34bdcab3f4d82fd53de0af
9043 F20101204_AAAOFQ silvalugo_j_Page_163thm.jpg
6666cbbf4c61f667b18fa30fff597085
648a75b31d0bca5d17043aa0192ca1a93455ecba
8758 F20101204_AAANIY silvalugo_j_Page_135.pro
8fee0355c82bbee33ab74215579651be
e97e0f56665d7634b87b0c81c96b459caa93a2f9
35552 F20101204_AAAOFR silvalugo_j_Page_230.QC.jpg
0c49a485e5794f1348fde4d50ac08dfc
4261ef0b73a02126e484353252f716079201e04a
12159 F20101204_AAANIZ silvalugo_j_Page_136.pro
ac662391c4cf7bbfa5edab57a1742f95
575925e3dc3459fa571b916b9b9aaac7a76721df
33868 F20101204_AAAOFS silvalugo_j_Page_038.QC.jpg
804dcb8e54f67951f0fd6ccef48b224d
81f5436a997e9b6e38290b086342ab13248ce999
32854 F20101204_AAAOFT silvalugo_j_Page_225.QC.jpg
0a762b86965c77a91b030e6532c69197
3718a593777ff172fa9a95879a27832e5631a083
21283 F20101204_AAAOFU silvalugo_j_Page_209.QC.jpg
454dfd00c6a7e9cb7b7fdfef541c3db6
c55915b16119dae596ba1906f57753d686f479c4
587993 F20101204_AAAMRA silvalugo_j_Page_121.jp2
de6342600ff5e14dbfccd5df80179255
1cdb4995e1c4c1ad9c2b683f58a52f8e00616427
3602 F20101204_AAAOFV silvalugo_j_Page_107thm.jpg
fb4399b7a217741842d050c8a98ab05e
6edbab3d3706ce4e072b56a340a8a424cc044250
F20101204_AAAMRB silvalugo_j_Page_123.jp2
7bae798244e1b429dadf37e5152a01bd
2b993112d35733271fbc0af9dff972709d4c4f4f
7371 F20101204_AAAOFW silvalugo_j_Page_113thm.jpg
926facc89ae93d2cd598c8f1085794f8
aff7589eb8091033b946b01664874c78f903d8f3
F20101204_AAAMRC silvalugo_j_Page_125.jp2
31803c78644bccd98305de3e72a814c7
2e2f110ef85f0045befbd56bf366279fa66dc915
9031 F20101204_AAAOFX silvalugo_j_Page_111thm.jpg
c309718cbabdf9628d2d3e275914bd16
5d9fa87aacb9c003190a98602cc2e522e8d53a91
1051938 F20101204_AAAMRD silvalugo_j_Page_126.jp2
a7861f7ac083865dcf9638751cb09c8a
29843e18aa637ff3c42bf8b23a415f85b960b970
8909 F20101204_AAAOFY silvalugo_j_Page_230thm.jpg
8a029b93c4dede3057e981d077bab88c
d810bba8010e6da038b8d74ee830f2729a3a3408
1051947 F20101204_AAAMRE silvalugo_j_Page_127.jp2
7a1f119be46fd86d1cf6abfa7fa1a8b4
17c32b84f619ebab9a01f4451ec099ca57ab4723
9974 F20101204_AAAOFZ silvalugo_j_Page_207.QC.jpg
632046cbc3482316186d7a1bba14f01f
9d6bd6abb0ed606cbc8f19d8fe80f320dff25b8e
441265 F20101204_AAAMRF silvalugo_j_Page_128.jp2
dba8244c0dcd678d0b8e34f138cc57ee
26babe65f7f9a85f80a7b560a593dcc7986beea9
439845 F20101204_AAAMRG silvalugo_j_Page_129.jp2
1548555e7a55d7a6b55a8ccbe4e65115
d2be4a6e01761a5ff6c4d67c63c94ef62e9b9f8f
327 F20101204_AAANOA silvalugo_j_Page_029.txt
0e5e40ac5c10096a6211f0060809c683
7e89bba4f4f95ae22be58a0d377241c97d93941e
320449 F20101204_AAAMRH silvalugo_j_Page_130.jp2
36b605a625635df331d3777467a6de57
5f32934583eba7f1c47fabb86417712bf098ad68
314 F20101204_AAANOB silvalugo_j_Page_030.txt
74b73742472363528631782a4d02a217
a2ce7d009cbdad51c3e59a72c991c62f7cf6e1dc
452090 F20101204_AAAMRI silvalugo_j_Page_131.jp2
3acfa2f8909e3421200fc8423d407b7f
a8dfe43735d5c73358a7370d1d102fd0400d31d1
437 F20101204_AAANOC silvalugo_j_Page_031.txt
d0b2727293ac604d4a45c27b7b45761d
6ea5c4357d3d0c211efc81ffa63e6c2b992ddc82
329277 F20101204_AAAMRJ silvalugo_j_Page_132.jp2
6c64b9f56d621b78853f718bf7a82c4a
8d5821d01f1fe1aef398748b55b6750633d14e29
333 F20101204_AAANOD silvalugo_j_Page_032.txt
8d9bb0f54c372186768856ace9e50597
04587af40fafc9bc98e9c4ed6bfcd5cb50433258
405491 F20101204_AAAMRK silvalugo_j_Page_133.jp2
b5f98a82bca4b2cf599ac087083e67fb
98ce322655f2ee2bc130a445fe00dafae33f5050
306 F20101204_AAANOE silvalugo_j_Page_033.txt
491ccc93c2641bd88f0c777675c3bde4
4a1762d84f2047c3a6fce504b111e8621f494fab
172945 F20101204_AAAMEA silvalugo_j_Page_008.jpg
9dd88dbd5d85080a7f5026c47fe3b9ac
ea44565d0a737ca345f7c2f732c269f613ae27a4
316174 F20101204_AAAMRL silvalugo_j_Page_134.jp2
e17211e8154f7310f76bc3482bf1062e
5b79c62d24abe4f1ef54dc247219492d6f2ab479
310 F20101204_AAANOF silvalugo_j_Page_034.txt
c5648bfab227345c6af37be9f9a75813
78ad3d7e37719e34ea1bab95d6e6c8fd65118d3c
206603 F20101204_AAAMRM silvalugo_j_Page_135.jp2
6114a6d18fdcff0f0f0a85d4a21e90dd
a64572d40579c6dd96da213b5f0c8773db1b48bd
352 F20101204_AAANOG silvalugo_j_Page_035.txt
c875176e3afaf5f06a3286b996bed70e
623e978384f830d75b4ba26473439a54874f7dd4
184261 F20101204_AAAMEB silvalugo_j_Page_009.jpg
a0bb68c085ce246a3bfbb3708011d22b
78cf99d086f5ae748ca873e0c185d35f6480d391
552628 F20101204_AAAMRN silvalugo_j_Page_136.jp2
c2d515fce20da8f7c737bbad10de96be
894e130487e4fce7dbe60e324d9aba0327bc4dc0
381 F20101204_AAANOH silvalugo_j_Page_036.txt
abc6884133479cf97cc911c54febfe1e
a035404e6ef84ca42b4f9cf52245a6c2b072d207
5071 F20101204_AAAOLA silvalugo_j_Page_136thm.jpg
782b42cabef46a6be9b1c991a1fbef8a
d320d1d7644be434e993907d29e59aee791eaf21
13876 F20101204_AAAMEC silvalugo_j_Page_010.jpg
16a324b950069747d14da5af1ad65d64
2e731aeaf51d2d2981dd1f58f04c4c9621738c1b
1830 F20101204_AAANOI silvalugo_j_Page_037.txt
1e1e72d8c57a04f7f3e2128f49bf98a7
33f4c45aa01c7fa9db94a48b435eca31f19a575c
9953 F20101204_AAAOLB silvalugo_j_Page_137.QC.jpg
e222ee4d2b24875539bb407a1b621479
f22acb7b1a6ecccebe032bff63cb1dc9c9b32947
118491 F20101204_AAAMED silvalugo_j_Page_011.jpg
82b255bbda37e957a67b628921bbdb05
8baef5cc673dd41b937100b916c2a99a127f1222
289610 F20101204_AAAMRO silvalugo_j_Page_137.jp2
002d341f356377dee12c8a9845353dfe
6ee1ad48071a68d95f9cea63177953308edf33c0
1814 F20101204_AAANOJ silvalugo_j_Page_039.txt
b5b6134f2f28faef0f7bab8e2a317eb3
cfd7385a75cd17a113512d72884884ef5f99ed97
3303 F20101204_AAAOLC silvalugo_j_Page_137thm.jpg
bcf5c8c57cd59655faff914c2051fcf1
c81d6c7332b0c4b1a092ef39c311dcb0e68dc387
145834 F20101204_AAAMEE silvalugo_j_Page_012.jpg
fe0127bea22e30db29877917d4d1761c
7373f4ea8400eb409cd078301b402d45ddc37698
757584 F20101204_AAAMRP silvalugo_j_Page_138.jp2
685e51859668d3a28ef5d26c7af1168c
35cd494679664cba1fd423f24ce54c4bb268772c
18578 F20101204_AAAOLD silvalugo_j_Page_139.QC.jpg
d423756a6af01816fed1c088f19b5d88
fbe568a80a6490e95f85cef3b4678f6db4cefca0
155113 F20101204_AAAMEF silvalugo_j_Page_013.jpg
d86197986c6f09fdbb245dd9d7e394a4
9bdbf1f8529346183a3fc2b3ad29ef9c9d6cc836
768876 F20101204_AAAMRQ silvalugo_j_Page_139.jp2
c601db316ff1efa2d89a95ba7818e31d
c646ecd8bbf756c2c6f1164e1079540252013fa2
2134 F20101204_AAANOK silvalugo_j_Page_040.txt
1d844956a979e5eea20c613e2dbfb814
1d36622470434bb16138f11d4cd17eaa5e499ec5
16546 F20101204_AAAOLE silvalugo_j_Page_140.QC.jpg
2195b0bba5b3a3539dbeadd89ab6c304
713c731249e41bffff91b00b874e85afe6217556
159654 F20101204_AAAMEG silvalugo_j_Page_014.jpg
0392a0d0a28814473cf016f5a8e83c71
79261b358ec5280e86f398c170cf5278a1ca8b13
639216 F20101204_AAAMRR silvalugo_j_Page_140.jp2
412cb379696d560408b1e284a6fbf3b8
788d423f6cc7841eaeab813f6ae3da195b2f5351
2164 F20101204_AAANOL silvalugo_j_Page_041.txt
ca0dcc9ca9046620e9b04aa7afa21902
3bac41a307b550736d1ab265cf51abceef442728
5384 F20101204_AAAOLF silvalugo_j_Page_140thm.jpg
73ea8db43a23cf087af417a7a3748f2e
ff8169571b03caffb3a31bf94a189fa2b4269cb3
20822 F20101204_AAAMEH silvalugo_j_Page_015.jpg
4711637603141b7ab345022cd12b4dd4
a7604f6472dac2663cabecedccac4e4692781a5e
F20101204_AAANBA silvalugo_j_Page_160.tif
085ae83ea1d9456e1ec9d1ce3b880cf0
ebb9599b8d3b05c1ab67efdc5abdfe903d6d4fa1
836929 F20101204_AAAMRS silvalugo_j_Page_141.jp2
318a1bc130f9b305094b80fcf5ac684b
55ac831c9d8de15086c91a9b42095e1c753ce0f4
1844 F20101204_AAANOM silvalugo_j_Page_042.txt
2eb9f5b6c8dd68fad724445bc889be42
71fb3ea669d52663a580fb4d8b58fb08025aa876
100799 F20101204_AAAMEI silvalugo_j_Page_016.jpg
b554f68644ecd8838bbfc9bb83902de4
56d7538994394767627973cfc9c67a91bd4b1ecf
F20101204_AAANBB silvalugo_j_Page_161.tif
388897355932c857a9bedafdac521f1b
18a69e6b51e4c0c135b18f21b1d501ff4529c455
909514 F20101204_AAAMRT silvalugo_j_Page_142.jp2
9ad8b01cb45e10a4432e45ab9c2dd449
d3bfa3b838c4fb9de19d7293cf8080c947af74d6
2203 F20101204_AAANON silvalugo_j_Page_043.txt
0fe98308ddbbc15d1ad2a3863da9191c
0684d60e4adde03176b671ac066a42fa60f530f0
5831 F20101204_AAAOLG silvalugo_j_Page_141thm.jpg
f43eb21ea4c5b97fec91423c52ed6f21
14e3520f0e7750bdfb39326f3c531165f3d7dd38
108548 F20101204_AAAMEJ silvalugo_j_Page_017.jpg
5bfa7623b5da4ecb8f814f7fcba7978d
2236c2745a86ad47cd477b6b33df6e9d79292c83
F20101204_AAANBC silvalugo_j_Page_163.tif
accc677c8425f3fce714f5b46d55a6f3
614d66f516b7c627d483b75a116f07b4dc008989
850359 F20101204_AAAMRU silvalugo_j_Page_144.jp2
318f93cef3f6f15e2434283fd1e76245
f3ff35a57463c4cb07383459b18008481ccc5041
2071 F20101204_AAANOO silvalugo_j_Page_044.txt
c306d8c9c93125f998170f9972bc840f
fb594c630f3271bf152fdeccb1aaf065cee4815c
6069 F20101204_AAAOLH silvalugo_j_Page_142thm.jpg
30423f346bc31297c3e94f8e21c19561
73eb1e0177cc29911cddee2456d187861da63256
110590 F20101204_AAAMEK silvalugo_j_Page_018.jpg
8e15826fa47a8d13b0e1fdff350e2882
4ac53b8bceb834d37ed7e1b8469e9a83915544b0
F20101204_AAANBD silvalugo_j_Page_164.tif
8ea8a2241aa6add3267246643d6903ae
7338b2efdbbde566fe29eb2f5ce0937b7f255464
993954 F20101204_AAAMRV silvalugo_j_Page_145.jp2
1a6a316fd4a246435f710809a6702557
6e7048f7f54289df5ee02f8eedcf8205f132b674
1979 F20101204_AAANOP silvalugo_j_Page_045.txt
8eb1714eb446d44ed3105c959ceecb4e
2f5430e11954a80b84c0b1dd3b6e46baac4c10d8
20342 F20101204_AAAOLI silvalugo_j_Page_143.QC.jpg
8d1df1566104a8d949749e140281fdc3
3b8a433a4e40d572141bec05d74523e94c826083
109357 F20101204_AAAMEL silvalugo_j_Page_019.jpg
869aea9b1ea2c28a5c6a8e9723d50397
9da09137659c43c1c2787d3d978fd67726c42e1c
F20101204_AAANBE silvalugo_j_Page_165.tif
33f8a0febbcb072b8570765675650dfe
90188cebdc11b33c555ca338bd42e121f4a6e6e1
688229 F20101204_AAAMRW silvalugo_j_Page_146.jp2
61e66df0865404e8c8fe24ab7e77c8d7
2effb17058d4f3cff3a3798c77c782b334152324
194 F20101204_AAANOQ silvalugo_j_Page_046.txt
fd11f0b09981b0541f5a305b5a9d710d
1c4269d777b8aec97778ca32fa1305ed65f07668
6251 F20101204_AAAOLJ silvalugo_j_Page_143thm.jpg
94dae7fa8d25e31260941ccdb8694781
50d613837884490dade5afdef1a9b87406ea246d
109944 F20101204_AAAMEM silvalugo_j_Page_020.jpg
0830641fc980cc7e022568098b77a6e2
2a7c5785918e8cd5eaf5e7bba4285dea352655e2
F20101204_AAANBF silvalugo_j_Page_166.tif
95e5fe633c0c5bd6fb5249334bb50a26
cf14e4e50223498a9aad0e7f9d992468d3241732
803400 F20101204_AAAMRX silvalugo_j_Page_147.jp2
626210d02c1c6d333c33c71083699d9c
9d041e4fec2056f2f5b94164196aec05e6d3bd8e
1244 F20101204_AAANOR silvalugo_j_Page_047.txt
79797e619965e3944ac4f53852bf6d76
5b3bbe6e8cb9d6f7dc90595d92e427044e7717ca
6813 F20101204_AAAOLK silvalugo_j_Page_145thm.jpg
e655b8130d668bc43ddcdd945e376b68
19eca7756408e751f1dc7b8652585827d368e412
110017 F20101204_AAAMEN silvalugo_j_Page_021.jpg
c63cd2a72927cd4bae0079ba77708b94
2805095d599b33c58436266296f0f6c1c2c67e4c
F20101204_AAANBG silvalugo_j_Page_167.tif
930fbd76489bc3bde93a165c9396af96
eb426ff56dce18c8dbf8d4f8889e8e8cd5365a3c
251227 F20101204_AAAMRY silvalugo_j_Page_148.jp2
42c34121da427ae50728443a4278876c
8023e3bc345d3e5ee7526f733ba9a999a212391c
2632 F20101204_AAANOS silvalugo_j_Page_048.txt
424aa5f9e13f9f106283c7104d524e0e
3a71cf99705ad5434fc7477d720a082de8ca7067
4710 F20101204_AAAOLL silvalugo_j_Page_147thm.jpg
27353ee41bde80f6128f8f4fc82f404c
42a5043a2b8d8b41cffbb3dee933c14f84701eb5
112942 F20101204_AAAMEO silvalugo_j_Page_022.jpg
1815fafbf4da14e948429d9df1ed6619
7fd782f844778d21c16fbd28fcf92f209c218872
F20101204_AAANBH silvalugo_j_Page_168.tif
a42f6fe0b1b507c20b51e9af5c2865be
cd36c6ebb6cb98159cc6564e6e1aa3a9d7daa877
566527 F20101204_AAAMRZ silvalugo_j_Page_149.jp2
8598d9823b5b7c2b1dc78b0df94c9c8c
2eb7dd2928f74ae9f396247729b45a932924dd22
990 F20101204_AAANOT silvalugo_j_Page_049.txt
0f88d4ad0ac4e2589f46e762d26483e6
41ac2032d5a959aaba581450e061dc7f58daf15e
4967 F20101204_AAAOLM silvalugo_j_Page_149thm.jpg
f3493425c686440a3b4c8325e2671ab9
9766fdfc3e19553c1c79a8facee0de3480a5535c
108961 F20101204_AAAMEP silvalugo_j_Page_023.jpg
286a1c3409d4e5146363c71fb93d554e
faf929db1956f79acae04ef061bc6d5f0d8a36cd
F20101204_AAANBI silvalugo_j_Page_170.tif
2e5c6002508b518c2faf5d7bbcba846b
90a4cd809eb22982d6d8bcde2de0fb4e566dfe8f
433 F20101204_AAANOU silvalugo_j_Page_050.txt
d82affcabeb659113e443f08b6a6dec6
e76fbee89148de2f92b1056eec73647e5189b534
9601 F20101204_AAAOLN silvalugo_j_Page_150.QC.jpg
19b547f85f2a7a1b4f638829657f67eb
d856bd4fdcce42c498c5bc030d1beb23e02a6625
24263 F20101204_AAAMEQ silvalugo_j_Page_024.jpg
beddd7ed7af2ad0260be5dd9b2088d4b
d5bfb001f10904f55ebdd5c8d7dd3969148f0d88
F20101204_AAANBJ silvalugo_j_Page_171.tif
c2db3ebeeebeb15c7750a78cbfebc16e
4a91ddcc59d8e77c60c0f7e30994a8c2578b3f08
357 F20101204_AAANOV silvalugo_j_Page_051.txt
0dbdf4aa2ecaa08d5f6aa12e27459a56
ffa579118566d581a3750c43a6b011b1ecd82933
11316 F20101204_AAAOLO silvalugo_j_Page_153.QC.jpg
69e6f90df4878f5156f9f035cdb156a9
9d1fb06985aa9dc059dcb64ed4facd746244bcc8
96108 F20101204_AAAMER silvalugo_j_Page_025.jpg
8ac69881e840faa2efdf1a53553b368c
489dbb91908e9b27bb88cc6a3163cccd147e8039
F20101204_AAANBK silvalugo_j_Page_172.tif
2edd1fd2b73d4fe2c2b33eb8720b2089
1b70f7f0f46c21b2116c0c875bd16e4d573c78c7
328 F20101204_AAANOW silvalugo_j_Page_052.txt
bcebaa24785175b29195216a3f3ad86e
42af35c9495b616cc90cd805ddf370045691e0ca
3646 F20101204_AAAOLP silvalugo_j_Page_153thm.jpg
4cca388a799f578d9110b39dd3a6e8c2
1a6b529e3ef7a4dc3a1c655ea8c5fa0d4d22849a
54501 F20101204_AAAMES silvalugo_j_Page_026.jpg
cb8c9efb2e4b9d9f45b29ac168a5a57e
3e8ee6540d0e34667fe0c7e919e47c8b34af8ccf
F20101204_AAANBL silvalugo_j_Page_173.tif
a0107fafbbb74c7da08e420c62c76872
aaae4ef3727d079e75a6170b638ac7e9e0588353
192 F20101204_AAANOX silvalugo_j_Page_053.txt
725dd56079ceb743b30accb86b1d1748
11e4f3652c57dee29e1fec0ebace60dcaf22626b
3516 F20101204_AAAOLQ silvalugo_j_Page_156thm.jpg
692b7c2138afcad0a5ac7049d7f0e56f
31a3c2787aa841ec4158b2c8969cf4ef3c29fc2d
66790 F20101204_AAAMET silvalugo_j_Page_027.jpg
4cbda99c5337866f7f289f43cf77eccc
4ed4abcb755691c2573889ae5e750dc0782842b5
F20101204_AAANBM silvalugo_j_Page_174.tif
b0b0e88a8c3456270d2b5e9d5745aaf9
66acf66aa223038e5d88d042bf2eb1bc83fe8a81
184 F20101204_AAANOY silvalugo_j_Page_054.txt
1a52f145963acb8c127ff062138adb43
90f064a1554b516256826997a75ce905ec1998a8
36764 F20101204_AAAOLR silvalugo_j_Page_158.QC.jpg
cdb7649bd1c465ac46a4c599b91aadd2
adca479cbf56e43e2e087b945ad76878e0a1eb08
76968 F20101204_AAAMEU silvalugo_j_Page_028.jpg
a8bd70482c20c590649a4abb1fc01132
d4c72ca86dc0a59f68161872284f89b4b705e2d5
F20101204_AAANBN silvalugo_j_Page_175.tif
57d7de934a6dee6fb94e48c980a2685d
6e9433781f185b2304665eb5ab01b7d3de02f085
431 F20101204_AAANOZ silvalugo_j_Page_055.txt
7a0ec065fb058fd05787f0c0683fde51
59babf5b34e94c939e11bcf1b94f73eecdfc9c18
37438 F20101204_AAAOLS silvalugo_j_Page_159.QC.jpg
66c7623c63a29bcbdd566b992dc7c36c
e55b8b613b84ac06a90de7de833932eeaf772476
72773 F20101204_AAAMEV silvalugo_j_Page_029.jpg
f04b44258f685d0ecf295a2eb6c6a842
0499c418a1d226b3944c5734b3ca780cdcc60854
F20101204_AAANBO silvalugo_j_Page_176.tif
3c5d0f2ce8723eb0289295905fbf1130
34d8a9a18fd8231a5e644c52049f04dcd892ac00
36283 F20101204_AAAOLT silvalugo_j_Page_160.QC.jpg
aaace7e2e5f1c05fe90779d6415f4b43
2ee89077e00e591a8f377d666a05372c8f75a822
142757 F20101204_AAAMEW silvalugo_j_Page_030.jpg
949bc1ead49e063996fde3021e28c9b8
148245b34bf4539ad1bb6b630b8101cabea3c9b4
F20101204_AAANBP silvalugo_j_Page_177.tif
836f6ff147190ee0b04d49c7098c6c10
427d6e7c857854cb32b3d7b5941a7593e12a1c46
35869 F20101204_AAAOLU silvalugo_j_Page_164.QC.jpg
3654f63e001a8aa36d923f1c3983b500
b5770ddfafe92b881a3cfeeb9c0b0e61f8249eb0
77157 F20101204_AAAMEX silvalugo_j_Page_031.jpg
28d719b40a19c17e11c74ffcdb49b0ab
564d079b24487a4c7c300b454889e9dc2c831626
F20101204_AAAMXA silvalugo_j_Page_048.tif
f4c625bf6b84a24e3554627b515c56d1
58ca441442f774efe26828ab95bedf0bf49c1a10
F20101204_AAANBQ silvalugo_j_Page_178.tif
c706a2ca702fa7ac84ae9872aed50c3d
efe4225e50eef64486e46bd2427ea691399a2263
35057 F20101204_AAAOLV silvalugo_j_Page_166.QC.jpg
4f291154768819ffe7d1bf33ffadfaff
320850a2d4e1c1b91fe06cb41c26962fcc618511
101922 F20101204_AAAMEY silvalugo_j_Page_032.jpg
c9511897c9dad984775057f1d7fe07d7
b2c1a0daab1d63058d13c605ebb7ef664ebe6eaf
F20101204_AAAMXB silvalugo_j_Page_049.tif
49857f2fe399ff4b9a8962c7a13c00b1
5455d90d97257f0c8228cca0a9ad44ba5f0dd95b
F20101204_AAANBR silvalugo_j_Page_179.tif
28e0747c2c44e3dbb155f755190276bc
5a8e323b72d8fc45794e2168f9c1fbe469b7f1c3
F20101204_AAAOLW silvalugo_j_Page_166thm.jpg
0d4837572974e13414371b790b461fe2
3ec548d92e2da0aa1d4623c16c1badf8e353d65f
131972 F20101204_AAAMEZ silvalugo_j_Page_033.jpg
93076cc21e0b76389b71b8489dfba970
e90ee3553dfec65336d4c4f7e5eaba0e0e3540d7
F20101204_AAAMXC silvalugo_j_Page_050.tif
24d8bb7f87defcf49953a268e5fb0e32
f36dded6f794c4ab60de6fea8b6395438244088c
F20101204_AAANBS silvalugo_j_Page_180.tif
0a1ea27578b5723e3a707224cdd4ea89
4ed0c70863eaa14d09b75a5684685160fb87754c
8938 F20101204_AAAOLX silvalugo_j_Page_167thm.jpg
ec94ef6f5b72fae9c48fb8b44f2f05f8
a9ffb85e4d8ed7de5f0327270e14eadf32607ed5
F20101204_AAAMXD silvalugo_j_Page_051.tif
bd80abfe3af3dfa9e3b6d3579006508e
94a86f0c443e7aacfdff4e935184a72b4779cb00
F20101204_AAANBT silvalugo_j_Page_181.tif
d2ddfdf01756e6a6a1f486f2bb23701c
998715fb2afd180e8989f2ad167685f92331e618
9044 F20101204_AAAOLY silvalugo_j_Page_169thm.jpg
6436e62db2e2503d51a6d5470320f057
46b2b5640cbe4f8a12e1af70b8cef114d44548bd
F20101204_AAAMXE silvalugo_j_Page_052.tif
022286eece522418e3bae54589f0a86c
7bd449e9a36144f7fd2de087ff22837c77749518
F20101204_AAANBU silvalugo_j_Page_183.tif
dfe16b7749dfb6d1b051dffe1ab26dc5
0268bc364e5e1d00c69e18fa3651e55e6262a86f
8891 F20101204_AAAOLZ silvalugo_j_Page_170thm.jpg
27404e66bb32b31377ca2ac59f0b55c3
477eea8fd75ab6e6bb4d0ca5f32aca88261c8a1a
F20101204_AAAMXF silvalugo_j_Page_053.tif
7fc00d0337b074a2d6bed516012f7ec3
01e7f52b86a21caf7893eecd82c0da2aa6e67821
F20101204_AAANBV silvalugo_j_Page_184.tif
55b7f540b87e61db39e98827e8728cd7
d4642749a3afe13f9a913617093bd34ad5b4acfc
F20101204_AAAMXG silvalugo_j_Page_054.tif
ae3d9c0f420a1ab21d50d692ab264904
570df1f7702a881d1ed5600b45008de0e899d066
F20101204_AAANBW silvalugo_j_Page_185.tif
d0817bc7893b78177548c9da1cf46c14
18a4473089713d694a5d43f819316ffce2f2a44c
F20101204_AAAMXH silvalugo_j_Page_055.tif
2f64f746aa40571edaed2bb796203a8d
505ccd74ad50d495cfbbffe7e5ec19ac69c4e9d1
F20101204_AAANBX silvalugo_j_Page_186.tif
72e14e01d1d3e4b39f275c06df90fb38
0ed221ceadcaa653a92612564121c1b6d21277b5
2234 F20101204_AAANUA silvalugo_j_Page_198.txt
956fc0d6785855c27c6093a517745c3a
99a0aab367261acd0db8c815b7cef77f67ee3f40
F20101204_AAAMXI silvalugo_j_Page_056.tif
583266c59329c7b1d6095e9d25124299
86a615146d86eb5583ca220f1475366cb8b8eb3d
F20101204_AAANBY silvalugo_j_Page_187.tif
bc91dfb91e0e31b68dda7d45d383f8ce
3c932b2693868914a31ff504e774f2dd93976a6d
2219 F20101204_AAANUB silvalugo_j_Page_199.txt
aae28064a25981c0e6239b9f435c50bc
060886da8b6ae0c98b444a08c50f57d44a7b930e
F20101204_AAAMXJ silvalugo_j_Page_057.tif
31576c1707495d06259bde13ab7d1414
c831737d798928cedf4ee0e92734729fb131ef64
F20101204_AAANBZ silvalugo_j_Page_189.tif
9e6631d73ea56088c2f42cbfb2484a92
192167d144d6cce3610cf2bb2a83eb5e4ddd1a02
F20101204_AAANUC silvalugo_j_Page_200.txt
df96ab7161d6ee41e2c8f153db5a9199
1af0cbe717a684c0c015f90d0127e5ee457cd1d4
F20101204_AAAMXK silvalugo_j_Page_058.tif
d4cca5d02ab3021188b13702edeef35c
76b072a7d27298a88cf00effd51dcc9f943241f6
2531 F20101204_AAANUD silvalugo_j_Page_201.txt
74aa5e8a2d580d8a9c7298ebc2613aeb
7897a97dc80564fbcbf9acee46aa7d4338824ef0
111043 F20101204_AAAMKA silvalugo_j_Page_178.jpg
f5d3680cd0b7bbacc5eb6989fbd72b8e
765668398ae6f3b3252ad62f1e97f9864848273a
F20101204_AAAMXL silvalugo_j_Page_059.tif
fc212b46cda1c46c15d0ac18cb531fb7
5df800d964421ec853fa60f1d0894c41b0d80cb1
629 F20101204_AAANUE silvalugo_j_Page_202.txt
008c3cffe2fd57f95eda4b1c835c3f3d
608be14a1c8741a9b056428b8694e6d5a982f284
110617 F20101204_AAAMKB silvalugo_j_Page_179.jpg
bbb43702a5d79e76fb4e2c478f249d23
6c090704bc7925a0f78ace96404c6634b26e6e00
F20101204_AAAMXM silvalugo_j_Page_060.tif
14dc93567dd1a41f19942a7bcfff6b4c
4f6fe67aef401eda999a033c6460df6143590541
1591 F20101204_AAANUF silvalugo_j_Page_203.txt
c082fa700bfa5fd5ca8166ee06117976
20551d8927cb297eff3ab1f8df03cfd0d9d135ee
118094 F20101204_AAAMKC silvalugo_j_Page_181.jpg
50deb9ee4512713cbfe5fd1df2ee0a67
baf98e7e6b9a5e7688d0f4b053a01e8bf3c413b6
F20101204_AAAMXN silvalugo_j_Page_061.tif
9266ef982a7e505009982a63cb8999f0
baad5e57d39ff8ed91b907a70e7ada8687cf2a67
2368 F20101204_AAANUG silvalugo_j_Page_204.txt
7726354563a55638870d272eb136d8e4
2b0bafea0627dc6a37afe7df78396e6a5c9b916e
102609 F20101204_AAAMKD silvalugo_j_Page_182.jpg
c0f08e597ab04fd506089ac449ae584f
912c93ab9163c3db1853709c3957d44b4027e271
F20101204_AAAMXO silvalugo_j_Page_062.tif
c06b42b29723cfc486f6f3c96109a467
31cf1c04b3e06ce2a0353f5327bb50dbb8e2b50a
1343 F20101204_AAANUH silvalugo_j_Page_205.txt
eaeb88fa479a14980c5b51bff1d0daa8
59ec2fee574b51aaa150e4f165632a4cd0b18798
106248 F20101204_AAAMKE silvalugo_j_Page_183.jpg
c005b6c6c4946bb1d39b5154a5f71146
8d73e5443f19c0120d0ba3963e3c7ce73fe72efc
F20101204_AAAMXP silvalugo_j_Page_063.tif
9d43f5bcc9543458d91f9b37ab5619da
3201594e812af6c8be8b13f37b3402aa95bfa1cc
F20101204_AAANUI silvalugo_j_Page_206.txt
a02cbbf4283cf377ca85263b6d7a6219
bd43216c2f856f624996377f028e24e5b6c67a63
112033 F20101204_AAAMKF silvalugo_j_Page_184.jpg
8fbe217087332cc338ade94c83bb0c56
4a896a8b52a7441086b2dad5d9fb5ba0133d60a6
F20101204_AAAMXQ silvalugo_j_Page_064.tif
5f4ab3511969f0dd56953fe524ccb1e5
5737060a4090273d54f707bdceeccf7ccb9e9d06
766 F20101204_AAANUJ silvalugo_j_Page_207.txt
e1fd4013baa6988a589c47e4a759199e
2b2a2e98ae930ca28bbadaf6f346bb769ab27aea
110132 F20101204_AAAMKG silvalugo_j_Page_185.jpg
9bfda66433f808b9b8908880a54ab02f
3c47db2a293800e91f668e4e69f05d57410e9f34
F20101204_AAAMXR silvalugo_j_Page_066.tif
1c437430be726eb234dab14865735aef
4c2423441f9c37534e6a1983d2d7b20cfcfbe3f7
1229 F20101204_AAANUK silvalugo_j_Page_208.txt
fc5e1b2b330ef7c30c4b190a80622993
67da9141e0be75978e5cb99718ef17c3dbe2e50b
52835 F20101204_AAANHA silvalugo_j_Page_081.pro
c6778f691e8fe5471d87758ffc9375a8
d9a0c5e3e066aca63bc7c0a60cead8f5b3f260d5
F20101204_AAAMXS silvalugo_j_Page_067.tif
95c8288f7594aca2dff0f49aa72630b7
4bb960682e69a497efcc172ed36cb0f44d2d140d
1648 F20101204_AAANUL silvalugo_j_Page_209.txt
a21c86d42c24399cec354191f52a0450
ae1972306615383b59c50cd4bdd58483947efcc9
111514 F20101204_AAAMKH silvalugo_j_Page_186.jpg
a0a002d88b211d31a1010e337065d098
4952fd92e6062da542d56e79a4c322d49de862eb
53095 F20101204_AAANHB silvalugo_j_Page_082.pro
49445d6eb16940989bb33986bb23ddd2
a57022c9d2a60e76121d4ff3f7e5269b7a2ec067
F20101204_AAAMXT silvalugo_j_Page_069.tif
99a08466b5ddfebf62fca7ebd14be434
9e9c6ca35a8551bf65d0e4c5bdd6531437e57f3e
1834 F20101204_AAANUM silvalugo_j_Page_210.txt
b160f2869d620854bf045363188da3ea
1cbb0fe54abd6b7569e0abae2d72d7b50b8d43ab
111065 F20101204_AAAMKI silvalugo_j_Page_187.jpg
cf100f8595d49739f33cf4346df105cf
9988276e0b0c45a6ad4174ca7034e5bc4b7a84cc
54524 F20101204_AAANHC silvalugo_j_Page_083.pro
ab6790d687df032161555af55527a07c
8f107c3ad41715cd562b731b8c08a2c9024a2f2e
2358 F20101204_AAANUN silvalugo_j_Page_211.txt
ce85538a2dafdd84cda6d8f3bcbe8a2c
8e6651a2b8aa3dfa5aba2cabb09fe11678c40d57
116384 F20101204_AAAMKJ silvalugo_j_Page_188.jpg
884f540839df440f6667f31ff00c14b7
562bc3cb96ec30f6779fad7ded34a97b9459a111
F20101204_AAAMXU silvalugo_j_Page_070.tif
cd6d725473416546c5900d99501e106a
2402ce551fa5e5ff1ac3fc9ac8f4a6a48a42b001
1053 F20101204_AAANUO silvalugo_j_Page_212.txt
2ce38a1b802e8f0942f9a08c0f1f2d1f
3dc0fc3b99f9f9bd44f5479e63b48ba1aa52c63e
113065 F20101204_AAAMKK silvalugo_j_Page_189.jpg
8ffc266b19c840f80724458ddcb1218a
045e96c26593847243bfd9685349f340837e950e
54782 F20101204_AAANHD silvalugo_j_Page_084.pro
57ba04782d4100c8250232119792fb44
571d50afcaca0bd3acb8775cf82c15c3f6f32b6e
F20101204_AAAMXV silvalugo_j_Page_071.tif
3a9afc4fabca7732cebd12473879e9ed
1798443eb9b696f8c0975ce76c8d846fcdd43d70
1203 F20101204_AAANUP silvalugo_j_Page_213.txt
eeb3b5d07700bc654fc827e88b969053
92739a34352b2c8941cfe0d7398314fc05080be3
109601 F20101204_AAAMKL silvalugo_j_Page_190.jpg
210b8da2c3a841eaa95c67a8e885810c
71db80b325080bfb05570dadd46c4f059901e436
51663 F20101204_AAANHE silvalugo_j_Page_085.pro
1108cb800c97b34d79ddf8ab0cef4ae4
4cdfeb657357436449df8f30dd7291dbe974bed1
F20101204_AAAMXW silvalugo_j_Page_072.tif
848e597dfeeb58e560c2f58c5b187c2e
402bd8da0c9f9eeed94dd50bd15a567ecb52a997
111558 F20101204_AAAMKM silvalugo_j_Page_191.jpg
b22ab0539000205a228e30d8323e9d2b
98bca5726251b8b3257d9920451f97598add0a1e
57925 F20101204_AAANHF silvalugo_j_Page_086.pro
aee707fed160370cd3cd1bc3983956b6
9b749df91562e9f367fc07b94bfc880e219a41f9
F20101204_AAAMXX silvalugo_j_Page_073.tif
c6347bf2c020dc1a8b72a646e795a748
983344a33f2b3dbe80066a2b2bffdcc375f8ef3c
445 F20101204_AAANUQ silvalugo_j_Page_214.txt
4ba3a2cb5ff91ab3bfa646d08663fcaa
36291ce9ba440a18b4060e374d0d235aedd9b9f8
118081 F20101204_AAAMKN silvalugo_j_Page_192.jpg
07d0f28ae8d4dffedba70a96fd0134fe
a5b6bb6d5d5795aa70d22ea2d4ddf0c9c09815c8
54587 F20101204_AAANHG silvalugo_j_Page_087.pro
fc5d18b066d3400bd026a988d4657041
9dfa20f6b7f4cfba28c57be6d16c2975797d5518
F20101204_AAAMXY silvalugo_j_Page_074.tif
21ca49f01ca2fcc647aeea2fa0554b5d
58e3d20198777d69449e7edafaa4b2755a3fc47a
1256 F20101204_AAANUR silvalugo_j_Page_215.txt
1af70830f837e7957a01c7e0fc4b8830
c170b4f617c7cb5eb57f3b7495636dcd655678e8
116154 F20101204_AAAMKO silvalugo_j_Page_193.jpg
6b074d8b05f2882903373cc9468c0633
080d266e73842b3419d4e613ad95e8789b8dd481
52445 F20101204_AAANHH silvalugo_j_Page_089.pro
c75dfba7226b651494b97dc4aa1c9e74
b5edd6a601534db9cd84fe808e4b47f5ae1135eb
F20101204_AAAMXZ silvalugo_j_Page_075.tif
ce719d70f136204def146b4d7e8715c6
9dd4a953a275442721f7f45a97cf0af541b5a41c
9214 F20101204_AAAOEA silvalugo_j_Page_096thm.jpg
9b818f04ecaef6b12c2e8dfd78555990
edd388ac0c658cd59f43514f2cd5f0a7201603ed
500 F20101204_AAANUS silvalugo_j_Page_216.txt
50416a9ead592aa22ad58d4b343a4553
4c82b6145b389c21a87d2523a376004958b14eca
116614 F20101204_AAAMKP silvalugo_j_Page_194.jpg
3778f1600c59978708a320e9c4d196d9
c61ef3a22acbcc238c0bf8697ad7dbd3d0452415
56283 F20101204_AAANHI silvalugo_j_Page_090.pro
51bf9100e6f666c5a94941224c50a04e
913cde2f5bfced7090f96e0ce299a90431f41a8f
4032 F20101204_AAAOEB silvalugo_j_Page_151thm.jpg
1bbe940bfd43867af55f30844f06b901
4b0af81e16bb6b652cf575768c66c88f2d89d2dd
640 F20101204_AAANUT silvalugo_j_Page_217.txt
43c6c55586febc255adce3245067d0eb
972abc25c6e4e059f1abe23e100209a03579ec51
116670 F20101204_AAAMKQ silvalugo_j_Page_195.jpg
56a4a004746932386c21ddc1a27c4ca5
e6fa8cc79f32e213395fba5484384124d8c13fc9
52199 F20101204_AAANHJ silvalugo_j_Page_092.pro
f6168d54cae85b4982b8040e3702e647
a4f7a59be5cd131402bd570429239d1326275749
15608 F20101204_AAAOEC silvalugo_j_Page_146.QC.jpg
b68e0962f67ec2df62115385af71f779
b75e1b94b37e8d82f2258f3d7b00d4ed515ec1a5
654 F20101204_AAANUU silvalugo_j_Page_218.txt
b7205e915c0514262229f1d7b4e22217
c4248f746c6e92be17e9bd9cfecf24a3bd9aabc3
117379 F20101204_AAAMKR silvalugo_j_Page_196.jpg
619d9d564f8327419812e9cb85357a19
7aa447e03e284634f9dc8f16ca468b79585745f2
51870 F20101204_AAANHK silvalugo_j_Page_093.pro
8a9f6fed3d072d15d860d9bd9f943372
3d03e8390b177dd3e04b92cb502fe9c6fba7cbfa
9231 F20101204_AAAOED silvalugo_j_Page_196thm.jpg
c4ac30d4181b06652d947630902751b2
07b5eeebe179a28654daee409db4f2071b623835
368 F20101204_AAANUV silvalugo_j_Page_219.txt
f1ee1149fad187fc40bee62379ac31d8
f642ed47cf3d8fdc52d41aeab9e63ce8bd91a4aa
114129 F20101204_AAAMKS silvalugo_j_Page_197.jpg
885b6018fb8043930c32ed1e648290b6
0adba52c0e6a9f554f4ea3e620147d7bf2f8ea79
54288 F20101204_AAANHL silvalugo_j_Page_094.pro
60546a9ce1634936f2b2450e66506e03
c06034ede28fe1c6117fa86857d21765e880dec0
8808 F20101204_AAAOEE silvalugo_j_Page_237thm.jpg
a71ef9534e3ff90e0a4b84c8e2f35af3
18d179d1f923d428eecb4f5f53877eafdccfadcf
458 F20101204_AAANUW silvalugo_j_Page_220.txt
b789110e1294687b220be2d85b2f21a0
f625035e3208ba771aa3206418689d8d0935412e
116564 F20101204_AAAMKT silvalugo_j_Page_198.jpg
d111cbf38c2cd6a638876a1cc6a9ea6d
07cbe18c7faf2b8f63e2a1f15f0aa447c29094f1
52340 F20101204_AAANHM silvalugo_j_Page_095.pro
79e08e3571665543645e4f144f374832
5ad7c9eb50f2e4be09982efd4a416cceb0aa69aa
27603 F20101204_AAAOEF silvalugo_j_Page_055.QC.jpg
efa118e8808bc95821ff13bf08521d23
a3cda2b95510e8b4fdd94d21c2b6a9e561a7ed33
407 F20101204_AAANUX silvalugo_j_Page_221.txt
ad3a05033ca2abc274cdd39cfe65c7e2
acb341a7207855a7d7e30de8d045a00dac2cd4ee
116958 F20101204_AAAMKU silvalugo_j_Page_199.jpg
1cb77dea89d5991682693f5ab5bd6dbf
595f5a951f58b8d8c478ba46422a2bcb1ef1fe72
55393 F20101204_AAANHN silvalugo_j_Page_096.pro
f7cb2820dab129891f2aca9ca6a58901
98f90d340aad377b10e51c2c03d093400a68e746
2790 F20101204_AAAOEG silvalugo_j_Page_134thm.jpg
f3759783d7af2182fce0e63e174f8cc5
511600580fcbc33171687c35525df4dc9fba602f
518 F20101204_AAANUY silvalugo_j_Page_222.txt
f3c5be699b67fbe79ad08f8afd5aa3fa
8f870925fdb0ef83f8b9e6f8dbe8b95cf889a30b
110719 F20101204_AAAMKV silvalugo_j_Page_200.jpg
2575d0273131b3edfeaef3772a864223
b2565b1a2f0786a6d245d7126ee44efbfa94f5f6
52882 F20101204_AAANHO silvalugo_j_Page_097.pro
c65fda2b1f3383770dcd09a000238a7b
6152237997a03d40586f8842d85e79ad75142fd6
9077 F20101204_AAAOEH silvalugo_j_Page_094thm.jpg
5ce9d65bb21e3089069618806405775f
d162966bba2313e8b831997f0d139e18394075fb
550 F20101204_AAANUZ silvalugo_j_Page_223.txt
380f6b2558ab39e398de5cb37f855fd6
bae62e47b2adc76eeddc286daafc01a4b53fd6bf
44124 F20101204_AAAMKW silvalugo_j_Page_201.jpg
98ee1568a5caee38d4df2b6ea1d36801
b42738e62630860d88ea595b4e625124287f207b
55254 F20101204_AAANHP silvalugo_j_Page_098.pro
c0630307a4384d97dd6693355a2d465f
af897343f1c0af2c949f5a5b09919020bd3c25f3
5656 F20101204_AAAOEI silvalugo_j_Page_128thm.jpg
a2c45b0be48af33b9e5c31576052c2d7
c8cc454925911d5c1aafde8266303aa80301e233
69982 F20101204_AAAMKX silvalugo_j_Page_203.jpg
28da8d049b122f2b40252ed12958797a
1a46f57ba3d1432636016085da4aadd7f91f4a32
54679 F20101204_AAANHQ silvalugo_j_Page_099.pro
e5d496ab6bc0d81f02d34452f043a1c5
aa84a0109c4a8d7edabbc03c415e2cf46a7d444f
7467 F20101204_AAAOEJ silvalugo_j_Page_117thm.jpg
f7c8305b1790c0e6b4d50fb8ec4e4b2f
3c3bfc4bf28cab6602d1818074a899e0c7914ea2
66420 F20101204_AAAMKY silvalugo_j_Page_205.jpg
fba19e29cecac2ba3fd0a093d27f53ff
9b417d2f2ac69163faa9bcfafd5dc99d04af8a2e
54876 F20101204_AAANHR silvalugo_j_Page_100.pro
d7a844fb1e29c909e963e138153240fd
0a4a7e1271eca9aba57e5287b7501c5151557f8f
32542 F20101204_AAAOEK silvalugo_j_Page_042.QC.jpg
7848d348c288fd9fbd6e59cc375d1397
ad4693d178b81038e8e5de588d999b264d547367
112201 F20101204_AAAMKZ silvalugo_j_Page_206.jpg
e51536dee7f8d4b22548105248f595d5
2b1b91a510eb7f8d82a98655b01e146a95d8583b
53825 F20101204_AAANHS silvalugo_j_Page_101.pro
a3216e825c57d37f67ef7d8768239740
3f51ad010f8ff84ff1fc1a0e7544a5bc3098174c
5706 F20101204_AAAOEL silvalugo_j_Page_059thm.jpg
1bf5d37335d7aa8f4363fecc09cc2f22
b78ca9451826461c1ee6bf2d15fa2d600897ada2
56401 F20101204_AAANHT silvalugo_j_Page_102.pro
c6012dde962c196b619891e19fbb49a7
4ca9ecb5b8a43d212112959b32eddeeb0327e209
37641 F20101204_AAAOEM silvalugo_j_Page_175.QC.jpg
b3234ce9caf239edc811d0f47ff6d546
db3117cf5f618c1dab95a63a6d9803e5a3e5c210
5987 F20101204_AAANHU silvalugo_j_Page_103.pro
2a44d09ee6024e695851494940dd424b
b50ae373d8ffce88b10deb7ea48fd07979fab80f
33348 F20101204_AAAOEN silvalugo_j_Page_176.QC.jpg
a78202a5d02e0ffad620d3e24203fa8c
206a01b663e474388d693c5325369b8359897092
70726 F20101204_AAANHV silvalugo_j_Page_104.pro
5a35c7d015887ac7c690ce49ad6fbf87
48becc001c5a625a7a4e1b5e2b43ff87139268ef
8750 F20101204_AAAOEO silvalugo_j_Page_098thm.jpg
1c330f6f1b61cba505eb4712f67c36e8
c3b30eecd9ebe82d95992ff79d55c5474f3d6570
11012 F20101204_AAANHW silvalugo_j_Page_105.pro
ca5b0ce3359ef1ee8bc27149587d15eb
43d82fd7465fd54d63d256fcd138cfc67dfc6ec4
8805 F20101204_AAAOEP silvalugo_j_Page_238thm.jpg
f5c63b4468de6481255aa06b4e1956b0
e383e8441aad05aecfac0461ad4694ea2d544363
57498 F20101204_AAANHX silvalugo_j_Page_106.pro
3200d3302e7c7ba79ca7ba2dda6a2937
a19fad45914b03eca76a8396a59ed3723199233a
31232 F20101204_AAAOEQ silvalugo_j_Page_035.QC.jpg
598b143f3f0b5c85fec9616f15534b29
4f9ba1cda05d66db15873d19c908b78a8ae38948
27731 F20101204_AAANHY silvalugo_j_Page_107.pro
0e02623384e231d25e7531dc3e97d10f
cab3d38d939115756e70a63f6539a2788c664d67
3794 F20101204_AAAOER silvalugo_j_Page_154thm.jpg
df733a7168ffc266edbc93bff65eb595
afe04c9f5f4799210c9b305b638e01d70a68b2fe
68028 F20101204_AAANHZ silvalugo_j_Page_109.pro
1027d2ec261ce45a35a85cb28cffd678
0c2caf02661c15c0d586b92a05b18ada76633959
11216 F20101204_AAAOES silvalugo_j_Page_222.QC.jpg
248610432ef443fedb8f8cf961b27283
b972c25183bcb7a563b29b6e464dfe5bc15fdec4
14972 F20101204_AAAOET silvalugo_j_Page_216.QC.jpg
252195a87b08867199c7465649a6f9b3
4f7523b43ac3046307981c408963aaacc907a222
9177 F20101204_AAAOEU silvalugo_j_Page_099thm.jpg
c8ac488dcc0463f58651237b5b9d6447
bda60b7e10066c0b2bede4454c8098050f691769
F20101204_AAAMQA silvalugo_j_Page_095.jp2
4e5b5e556202f9288bd22a4777e59c9f
59da86d20df47b9baacd8319f1e3b79eea63f2f1
31133 F20101204_AAAOEV silvalugo_j_Page_117.QC.jpg
a4d5f732573a8bfd26686615a627f984
8e73dff418a675260d4608a7168e5e7ae2ef7886
F20101204_AAAMQB silvalugo_j_Page_096.jp2
9be26d74eb537b9963f281a4040405f7
a61da29477ffb4d9ce1d850d583bee0de323e9eb
4962 F20101204_AAAOEW silvalugo_j_Page_209thm.jpg
3bb65f9bd03c4119b3797d2b7dd7623a
d5a7e5a531e696e838cb9f344c96c3af299d6f7e
F20101204_AAAMQC silvalugo_j_Page_097.jp2
6fcd4a19a827b8acfac15172ea0177f9
d87a1027c0f5797ac6c6610abbcd98e2febcce8b
12556 F20101204_AAAOEX silvalugo_j_Page_217.QC.jpg
1cb06521aee3db536eeffbe01d1bd9c5
86e0694f98d5c51e14898035b485df8c8447c221
1051864 F20101204_AAAMQD silvalugo_j_Page_098.jp2
4509dfbd132ecd7a5e05788953f72fd5
6ca3d0b1c26af7776d2ac24e63f3367b3f18c527
9064 F20101204_AAAOEY silvalugo_j_Page_071thm.jpg
5a96a2b2a3c51238668d00e3c02ab4ce
b20a4559d9b221b8b2834f2deaf23ba598abbbe1
F20101204_AAAMQE silvalugo_j_Page_099.jp2
2988e8caf65d5bb81515e00e03b9f17e
c8f76735eb63275528eaa7ab0ec4b3f8f9dd12eb
3983 F20101204_AAAOEZ silvalugo_j_Page_121thm.jpg
10b2bcdc1e486289ea56590849c4e5fa
651af4a83aad75107ec8740e0cf64aa41ac27d5b
1051928 F20101204_AAAMQF silvalugo_j_Page_100.jp2
ac22d4accb9dc3d0f0074957f913c6d4
9739af0c09d99d298048ab2bef77fc8efd28ec83
F20101204_AAAMQG silvalugo_j_Page_101.jp2
17a0fc317826b93400e9ae95d5c49054
57bf48b8d843355d11f12ee4e501c8606c848fa2
6065 F20101204_AAANNA silvalugo_j_Page_249.pro
a234414ed59ba653a39d6c67288ceeb0
47962c67f34163e47a0175510e95ca88ecf5b635
F20101204_AAAMQH silvalugo_j_Page_102.jp2
db8cd49eb8a6266ad7f2544e921861da
190fe1b474e41ea91e459593e423e58c9c6ca8d1
496 F20101204_AAANNB silvalugo_j_Page_001.txt
01672c9b9fe154fa35a83a1a6b354ce4
0b1be7fc367645be055192d67d713488d0db6cc5
139926 F20101204_AAAMQI silvalugo_j_Page_103.jp2
67936a700717d5ff3c76b9023be43e2f
f60b46ff35b67e921adf1d58f8fe5bf9b473c6a3
99 F20101204_AAANNC silvalugo_j_Page_002.txt
bf85e032be0a243102b3fe3fdd7cefe4
8072faecdfce4a6e6c0c5581e990366097f62aa0
1051960 F20101204_AAAMQJ silvalugo_j_Page_104.jp2
07a358166ee7b4c42583813688712fa7
3131abcdc6a53a6885672609b5a47af75adb8a6d
238 F20101204_AAANND silvalugo_j_Page_003.txt
0bd626ca4b7197cd9d56d30ce431889c
a166b6ec6209225f1fb2d88f865b402ae5dda721
236259 F20101204_AAAMQK silvalugo_j_Page_105.jp2
d93d4ba424333a39ee2fbeb0f8bab83e
bec2c7bb5cb32950fbe8a75f0e00f684e983bcc4
3730 F20101204_AAANNE silvalugo_j_Page_005.txt
55d55c68b72ce141cf553a332bb3fd40
c66eefa76be186f930ecf8fd68d5f9956ed15db2
1051922 F20101204_AAAMQL silvalugo_j_Page_106.jp2
8b7505504bb31a60de21ae3f04f53b4a
0aa0342ba48220597e18d7885cc919f7a585ea47
3699 F20101204_AAANNF silvalugo_j_Page_006.txt
e9080c8d685ca63bd5f0915024d3270a
172e073291fd5778bb95fbbf5e233b15cafe682e
1064 F20101204_AAAMDA silvalugo_j_Page_145.txt
a33c1107f89d96e8a78dcd0199171900
cdb6ec4037f85f4e899bfc1aeeca146951b21533
582995 F20101204_AAAMQM silvalugo_j_Page_107.jp2
bb542826534ae07f391bc7aeffa54a38
5161359fc9bb852af1b6d5b7ec237c0d648509d6
3304 F20101204_AAANNG silvalugo_j_Page_007.txt
62e7fe81747ea6d89144899878a99f9d
9ca5a7ee98c3c71812241f4d537ba47faa4c9fe8
122268 F20101204_AAAMDB silvalugo_j_Page_114.jpg
f27abd67a06b61e3941c4a01056673db
5caa7e0546fccc228a91df69017aa06471a576cc
3568 F20101204_AAANNH silvalugo_j_Page_008.txt
b90473f92abf2715ed74eee2c4978c52
7c7194daf58abf579f26fa614bce5d8e4ba981bc
8394 F20101204_AAAOKA silvalugo_j_Page_104thm.jpg
cb708010252ae7eb35327d55f67e9f99
56ebaff64718879d224ba5cc415e8b2e2c2a326e
1051958 F20101204_AAAMDC silvalugo_j_Page_204.jp2
4fea9707e67028dcd1935e9c1433b46b
72fa75f98e39fa98307dece1fbe2ca80e829942e
F20101204_AAAMQN silvalugo_j_Page_108.jp2
21a27d0910a84cb105c8583d6cc69a89
cbc45632c89cd199ba8b5b7f2a537085dad786dc
3764 F20101204_AAANNI silvalugo_j_Page_009.txt
8a23341adc11879964c7e443857fc9f9
525a4f5b6bdf8a78f697510c6f89fe932d2ebe68
7474 F20101204_AAAOKB silvalugo_j_Page_106thm.jpg
ed5b3387bec9e9c57c51d7fdc21dba16
e54a2fa7e826140470824c00ac79988a652649d6
19358 F20101204_AAAMDD silvalugo_j_Page_202.jpg
afd0af145a843b7bd0c05942f0eb4817
6f33639763b40c91cf73b951d97daf2ca1928bd4
914824 F20101204_AAAMQO silvalugo_j_Page_109.jp2
47cbbc8277e1455324ca374da28cb889
c7734b90637f317cc55b976b05234cc3f108329d
15586 F20101204_AAAOKC silvalugo_j_Page_107.QC.jpg
b4785b0f847abd957ef5c80681d672f4
d70fd102e677d8b2e788f677c53989beb446d31d
F20101204_AAAMDE silvalugo_j_Page_139.tif
850e919906d8b42e0a5175489ba78b07
05da3d289de024ec70952a88336e362648822dfb
1051943 F20101204_AAAMQP silvalugo_j_Page_110.jp2
e96c3212bfd6026d7ff4d7dd488c3208
e4b4b6cbe4acfe8c1078c57c9f11e80349271809
227 F20101204_AAANNJ silvalugo_j_Page_010.txt
37a84207b227aeee6adc5b6b8819cfff
e66f828a42261d45ceb86f2b425b3c7b8e1bfff4
27656 F20101204_AAAOKD silvalugo_j_Page_109.QC.jpg
734ab1e34fb38052815889adffbdb273
dbaa268e84152d70a727f0d4cf212735c7514d9e
611 F20101204_AAAMDF silvalugo_j_Page_056.txt
19d15b0606af3eb9df93e6cfbf5735cb
b58e84e620d4f32975cd530dc1e46c1592b27811
F20101204_AAAMQQ silvalugo_j_Page_111.jp2
8289e3bb6742e77ada64b6955edd01ff
9ff6c13fbc72890b5ec63e00b76fa6ea57438338
2762 F20101204_AAANNK silvalugo_j_Page_011.txt
3e3bb3d32e13a0c1b069a86c1860b941
e3c2f0edbfbeff91c4807940a1d2180fc9445f77
35378 F20101204_AAAOKE silvalugo_j_Page_110.QC.jpg
3d1ce1900343001f36edf8e77a5b00a9
b4e94453173410e1337bab45dd63b3d648d6ee1b
55231 F20101204_AAAMDG silvalugo_j_Page_107.jpg
f058f6cbe82421ddb977cfecb1723866
efe675ef868a6e0fe01041abb751ccf842206782
1051982 F20101204_AAAMQR silvalugo_j_Page_112.jp2
b9bea59b1d0dbdfc7022768c0706a11a
fd6508540b836fd295f629cae9fab924669989ef
3137 F20101204_AAANNL silvalugo_j_Page_012.txt
e7b69e71f04a251f7a83e5e4aa590530
3167dd0b54c34f0d604d8bdb232215f6663524b0
1051935 F20101204_AAAMDH silvalugo_j_Page_036.jp2
be1c7124eb85bc883d44e9c407d31cd4
e6cefc18277a444d81a5726fed376d7cc2df1709
F20101204_AAANAA silvalugo_j_Page_133.tif
6df04b289d748fe8c5d1efa335c9447a
f678192604bd04480da62e1caa0d532ffb3a6fcb
923012 F20101204_AAAMQS silvalugo_j_Page_113.jp2
3ae29af53004f702187dc3c49014abd0
5057c1cf861058528d6aa79284de7cbc588dadce
3401 F20101204_AAANNM silvalugo_j_Page_013.txt
b1f4ac6cbb7b3466612fc75c826e8d65
3bdf2a7ae33974e1e58b580b8001ae3fffa3e04c
8311 F20101204_AAAOKF silvalugo_j_Page_110thm.jpg
5b3fa654ac8f4341c21f41f730989b94
61a46c2b2fe4e575a5bbd3b1c9e937be37b16445
16372 F20101204_AAAMDI silvalugo_j_Page_121.QC.jpg
d38e053783cd59256e96f0fd03484d2e
64a9c38e5b88b77fa734fb26714586fe6eda1ca0
F20101204_AAANAB silvalugo_j_Page_134.tif
9b4b1a41ab8d95aef680ee27bcbd5661
94f7b11cacf1a984bfac4f657a9be52514ea4e2c
F20101204_AAAMQT silvalugo_j_Page_114.jp2
8b0be094be7e9d620277119bb76ab3f9
c45e583de830e6bc1d08576e17bc75425c950ca2
3388 F20101204_AAANNN silvalugo_j_Page_014.txt
9a678eb1d91f0080c65f2ecb09bd1925
0dda77009371078a075ab1f82da0e39f22e153ec
35592 F20101204_AAAOKG silvalugo_j_Page_112.QC.jpg
d390971d32746efe2f6e6832f1cb2d86
f0e3fa9072f00269ac822bc4a6e2a495e8cc17e2
106160 F20101204_AAAMDJ silvalugo_j_Page_081.jpg
a0e0bd9e929a5ee824d83a34816f96ef
feaef1ec1f451aa0af7e054070b82ccc74b1bbb7
F20101204_AAANAC silvalugo_j_Page_135.tif
1de7a688716efd81e067a923a396bb0e
87711e21800620863f10cbcbd015a3ba8e7faaff
1051954 F20101204_AAAMQU silvalugo_j_Page_115.jp2
090ac7b9316d3746fcf67101b85df1ee
14d8ff3bc1c922b16eaa6860d595f2a05d31fd50
291 F20101204_AAANNO silvalugo_j_Page_015.txt
619b183d88f3a28c8186d5fa46ab80e6
c520e319ed162a64e4d784b65d052104cab24902
7991 F20101204_AAAOKH silvalugo_j_Page_112thm.jpg
27036c1d92d5f252ab264539238bc2b1
8e6b339486425a0e4a678352b9ab69f9bb7c23b3
F20101204_AAAMDK silvalugo_j_Page_095.tif
67fc9dc71d50c6775268db8378e4a9b3
c17116e57086ea883a091e30e454731000fab8d1
F20101204_AAANAD silvalugo_j_Page_136.tif
41631004c05bc4ef9414797c05abd80c
1753b07a24b668b871aa049c18cdc475ab10c03a
F20101204_AAAMQV silvalugo_j_Page_116.jp2
b15d16fe0f4e6c1260c2f80d8fbee387
30dc86ac5abe377c2b802931e57ca14cfa9d61e2
2062 F20101204_AAANNP silvalugo_j_Page_016.txt
4b449cdaa7ec9a796b49b45507e52d4b
59d948e3a5e21be340f3f6fc923cb379aa1ca113
26752 F20101204_AAAOKI silvalugo_j_Page_113.QC.jpg
df8d009cdb40b8f293190f7280f16be2
ea709c38f4d0a7be290acab3db88f3f2909654c5
19534 F20101204_AAAMDL silvalugo_j_Page_144.QC.jpg
f3291dfd538a777bced1ac7ef741f38a
33160ac4a5b50f69c80de0ea429affeb444dcb2b
F20101204_AAANAE silvalugo_j_Page_137.tif
d9238c3884dd7d5eafd26659bc8973aa
c3be4a6cb9fd4bcdd85c6b37719145f53cc2c498
F20101204_AAAMQW silvalugo_j_Page_117.jp2
ce63da8354e5d037773dab10316f74a5
ac86fe405f05aa7bd8bdc60eb2e059cbd530c50d
2086 F20101204_AAANNQ silvalugo_j_Page_017.txt
d1ae1d74bb08316beaf3a66f3c0023e3
cbf27fb6745c352972505e849e72e26043d06ca4
7321 F20101204_AAAOKJ silvalugo_j_Page_116thm.jpg
e0d21d294ff52c1bd460705b5f656139
6fbbab630d3d97ecad46524fe020a9333fa99138
1907 F20101204_AAAMDM silvalugo_j_Page_173.txt
44c936b5b9776b354ae4184eb24449b3
ab6ca5993b1e742e3a018f58ec18bc6e1e022116
F20101204_AAANAF silvalugo_j_Page_138.tif
b5129c3ce6eb5ca755c6fb246d44afed
5ac40d3a097ce6755c8e91283dd6bd75c1a8c4d8
454512 F20101204_AAAMQX silvalugo_j_Page_118.jp2
13333ced1425bb04929565fc5635648d
acb7428397ffdad405173540fd095caf4ff6a887
F20101204_AAANNR silvalugo_j_Page_018.txt
02c16aa3b16a80b1b4ba440fd578d409
eb48cf9544205a6600058208bd83e6a839f00312
15633 F20101204_AAAOKK silvalugo_j_Page_118.QC.jpg
1196be986f2da286397b5ad42443aca0
552a2a204c59f1e2b062606f94ad68b9474cf8cb
F20101204_AAAMDN silvalugo_j_Page_207.tif
845d16f3bc77dfca35a58bee12a5b93d
1ed33d7b4a1aff972871cfbfb6a6804901fe00c6
F20101204_AAANAG silvalugo_j_Page_140.tif
5d86f4c82165e1ccdc0033d89f454717
2a21b972aecf1755efb07fe0c0462af2025c9f91
569259 F20101204_AAAMQY silvalugo_j_Page_119.jp2
ca2a570f587793ae1fd0af4402347139
dd905838bd7779114408984cacd7ed531c11442b
2104 F20101204_AAANNS silvalugo_j_Page_019.txt
5940973033fa85226bf5e556f71606f4
7a8b8c8629fab70d68baab6af5bc3ca969dc0b5a
3276 F20101204_AAAOKL silvalugo_j_Page_120thm.jpg
60e6e3dc86283502686de08903a68220
75e40c9c3dc9109a864852e7acd7238d86eced22
2079 F20101204_AAAMDO silvalugo_j_Page_081.txt
1da97e693886590227ba76cc55fd46a0
5b017813acfb01ebd7946c3166caf38e3f1005d6
F20101204_AAANAH silvalugo_j_Page_141.tif
9cd9a544700ff5fe057a77b27f848ca2
662437548762e0d2697dae33df3792c90092fbd9
484824 F20101204_AAAMQZ silvalugo_j_Page_120.jp2
e5d2b088f8c39316cab1af394360d722
a5154f21b2e8635210f6401b504c34d6d976c5a3
2090 F20101204_AAANNT silvalugo_j_Page_020.txt
3d1aa8b9824bdeb7a816963f7bc67147
32722adec04ac7e5f6b704ce96324d98f27cedcc
38630 F20101204_AAAOKM silvalugo_j_Page_123.QC.jpg
dae6aa687e9eaff68514f14d88318b41
e0f318363bcc10dce4ea35b3cbac4c1c776311cc
5906 F20101204_AAAMDP silvalugo_j_Page_144thm.jpg
59f3b11fc95cc0bbc6f4de482c482c9e
176a10b68b843bfbf3b415144ea806c4f7b5ab50
F20101204_AAANAI silvalugo_j_Page_142.tif
8ac2ce9a9b731a8fa344bdc58170f959
621602d9c638d36f138e04b4ffefc302cf5e1928
2080 F20101204_AAANNU silvalugo_j_Page_021.txt
d8cb7d62460d6a2d40a8b7789ed04bb8
9eeb961042ec591108d0e8c66c0e2638cd78d3d9
9451 F20101204_AAAOKN silvalugo_j_Page_123thm.jpg
7a3182652e536360070f6d54c58985c9
37fb9535a4e67a2c3e4063de0abbd15ea2479de6
1051980 F20101204_AAALZA silvalugo_j_Page_011.jp2
802519ad780810f425481e8d56da1364
8067dad13421984e7068b774c8c6027dab8f4d3f
289540 F20101204_AAAMDQ UFE0022089_00001.mets
da04d1e3c6a85c730515c3a29e20deb3
be98941439bc143642980492a43f6fc4a0d1a468
F20101204_AAANAJ silvalugo_j_Page_143.tif
3a0cf7493b64d531161bf577ec003f8f
b73f2dce68ff982f4f8e4f4bc8670f67f421c074
2169 F20101204_AAANNV silvalugo_j_Page_022.txt
a02c410f6811ea7bf8f9df96f921f2c1
4a3d520757feb5c55413213b9f9b9d70cb1587ea
39387 F20101204_AAAOKO silvalugo_j_Page_126.QC.jpg
051fcb536500032a44aa1f56b6a636e1
255946fd33da97a20338ab24e50f72163c779d7f
F20101204_AAALZB silvalugo_j_Page_024.tif
b9c84ecab7ed619a04ec0c985141bb1d
a99aaea78a437775d2097a17b082c818896d6cb5
F20101204_AAANAK silvalugo_j_Page_144.tif
40d122dde9d160a970c7a7cc8bfa5529
da5ed55195cd043977621dbed150518f30ec1712
2107 F20101204_AAANNW silvalugo_j_Page_023.txt
1ba31cc6ef7b429714c4a091b3486402
4d34112764e7c9b2a22929976d3f1a8f170922b1
10223 F20101204_AAAOKP silvalugo_j_Page_126thm.jpg
871c94fb8346f8d86552ff8abfb78dba
ca5b3018f7bc73f60de318d9c849dddbd0c49525
85355 F20101204_AAALZC silvalugo_j_Page_109.jpg
7c8535fbd5b58950368faeaa5c511e02
d4ab719209548aa5d41bed63ce76cc08a7cedca7
F20101204_AAANAL silvalugo_j_Page_145.tif
9cad9e6daa9dc010e54c299e247b04f9
d763975952c273f5a210b9dab1eb80609bd14524
403 F20101204_AAANNX silvalugo_j_Page_024.txt
dfc9a13519f966ee2f021a6024d7b7b0
5fe8041cfa862f896c219ade2e5c331735f77f1e
9136 F20101204_AAAOKQ silvalugo_j_Page_127thm.jpg
11dbb4db89292deb5ad101fd1bffe173
0cfa877714046b8ffe8812fee78dd13673ecddf3
35812 F20101204_AAALZD silvalugo_j_Page_169.QC.jpg
710033b690f8af536e8d4e81340c3343
a5b5e7e50946e05aea1c0fa92e6b61fec1ec630a
27980 F20101204_AAAMDT silvalugo_j_Page_001.jpg
11189a410fbc7f7197be08da545dbba2
48be435a707dfdc8b1b1e4c5ac933f2737b78065
F20101204_AAANAM silvalugo_j_Page_146.tif
e159e78628882199428d98ef5dc93bc6
8212f960cae205147489f0c014b461959f219194
1157 F20101204_AAANNY silvalugo_j_Page_026.txt
30132afb394fc82f8ab2934156969bc0
418dd8e607c61c80a6b94840d7ffb6cd7f5b7e16
18295 F20101204_AAAOKR silvalugo_j_Page_128.QC.jpg
4c2dd8861311e150abd6bc0c2b5b7d9a
406635037b0649841f7741691c18991f0a64ef31
58085 F20101204_AAALZE silvalugo_j_Page_192.pro
1e746f719814dd8821c81f05b79e366a
d7766a467b2057876185e2c77b69f8371b7bd17f
4462 F20101204_AAAMDU silvalugo_j_Page_002.jpg
1e7399d5ec8d23cd975f1e53a3917337
970fd19d823a3fb028068b97c0881e8ae2747b56
F20101204_AAANAN silvalugo_j_Page_147.tif
e69cec96c4a2eb60aa0c77370aa5a208
a7c6e3d0fa41aa941facd0e0f5f307700b5eb3bb
303 F20101204_AAANNZ silvalugo_j_Page_028.txt
b96bfc36d21bc8f817b05e02bf3e5117
6d240a1e44e5b026952490dd208d6337b85f5b02
2865 F20101204_AAAOKS silvalugo_j_Page_129thm.jpg
d022abeb9e4498f864b739bbbe7ce302
cf58350ac8ed2da8a0ab78c7518ed7ed9e6504b8
F20101204_AAALZF silvalugo_j_Page_124.tif
754dd524e4eb9cc917267808f92d4a59
c28e6e961af1505c5686284b2f8a7342d22941fd
12273 F20101204_AAAMDV silvalugo_j_Page_003.jpg
59424623ba51bb2b1d5379127bd672d1
c5012e60776cc59685564b3ab86d0fb927fed653
F20101204_AAANAO silvalugo_j_Page_148.tif
0da402891128dd93d3e0951b9d872a4c
7864adea43329ec24913c36e55d6a22fe3b01884
8407 F20101204_AAAOKT silvalugo_j_Page_130.QC.jpg
64ee2d0561d2aac7fadd4ff078b5e4ef
3f8c5a4e467f3ad7a75573e517580ad1952132d5
F20101204_AAALZG silvalugo_j_Page_236.jp2
9fa0aac7b587d6ba677c54cd06309ec1
1b6dd936b9e66c2f9033cdf390b0e292b2fce5bf
102224 F20101204_AAAMDW silvalugo_j_Page_004.jpg
844afe2813e8370718f65791d88f76c1
2cf8dc408009f6b85f552e1f81d161a3e18e8ba5
F20101204_AAANAP silvalugo_j_Page_149.tif
1232349ad20002cb0643918101a29b6b
04eebeb3ca8ebff7d31625925c6905983f21f577
12325 F20101204_AAAOKU silvalugo_j_Page_131.QC.jpg
27ac3a7a59f57e37a825aaf45580c7d7
faae8a41ba7d8d5f5a7dd2ab6f38c08cb095ca3c
F20101204_AAALZH silvalugo_j_Page_184.jp2
915bb0c3167cceae0539914fba738373
7071f003713e2e592f4bb94df66feae9b67825fc
115692 F20101204_AAAMDX silvalugo_j_Page_005.jpg
d387ab6532225cbf1e6fbacbf973557a
603633c6ad343687bcc054dba09f226e90f86962
F20101204_AAAMWA silvalugo_j_Page_013.tif
1651a7e87db294856aa9256c2d23ff15
ecdd66a19c5c4d646a9fe634be7f4ec442999e04
F20101204_AAANAQ silvalugo_j_Page_150.tif
7f4f3e94b45934963cec8ebf6cab1126
d8f0a18e451d803cbc3e9da6e570b2181f1cddbb
8067 F20101204_AAAOKV silvalugo_j_Page_132.QC.jpg
72bacba345d3748f0c5b919e6b7f0bcd
f8d8054868de1377a2e31475ae44eb20258bb0f5
5910 F20101204_AAALZI silvalugo_j_Page_015.QC.jpg
9a4105d2dad6c363bfabf7612098c131
ea0b188b9ddf4c9a207be9456a940965b25dde11
133173 F20101204_AAAMDY silvalugo_j_Page_006.jpg
7f6ff531c8674eace84aa91b5a9a6496
4bedf2f23ccc5e1b5bb5426e03bc31f114ba86f3
F20101204_AAAMWB silvalugo_j_Page_014.tif
7edf5594924583eff2da9830e68e3893
da9b002d3d2303187cde78cb5d3ede9cf46ee4c3
F20101204_AAANAR silvalugo_j_Page_151.tif
c1a42740df912024055e3027d31b7b9a
a2ecf4e462ab79080060f7025ad96014291a3657
2585 F20101204_AAAOKW silvalugo_j_Page_132thm.jpg
ba8673606343fc0fbe2e50d7f924234a
c1ab25250373bab3e6bb09d0d8e12d6240f3cd32
F20101204_AAALZJ silvalugo_j_Page_027.txt
25191c136452d9954e3f8abac0b736c9
ac6dcfbc04a90051ed07301adf4ff4e0f50d48ae
154819 F20101204_AAAMDZ silvalugo_j_Page_007.jpg
38acbd7221a5738251b1f29208bfd362
ba521a02355ae8dfc1223b493f1dfb717f34491f
F20101204_AAAMWC silvalugo_j_Page_016.tif
9521e963c18a94b87b90036b839b626e
d3fb04bb5032a3397d1d653d5c80fb51962ba164
F20101204_AAANAS silvalugo_j_Page_152.tif
422711c1cd169d11cc1c129f9cfc1322
0c7bb769b2e62eb9b72bfc999ab4546f5f0ba403
10045 F20101204_AAAOKX silvalugo_j_Page_133.QC.jpg
b785a6e009b011e2b094de51c715fd8e
093f2cda5bdf4a5e4fe7165c9321feccdbe1670c
F20101204_AAALZK silvalugo_j_Page_132.tif
a3493ed0d1920f2069892bba7ea55be9
1759775d6f7095db890b2d423fb732a52b7844e0
F20101204_AAAMWD silvalugo_j_Page_017.tif
a9c9204cc0281d36f55ace059c9115aa
0c876a3097fa48eb827311f3c5bc83360e8d9059
F20101204_AAANAT silvalugo_j_Page_153.tif
83d75b25823a21bff8cba283bcfc84ae
e1ae0f180adf7f6aef6674a2505caee8e4e97fcd
2916 F20101204_AAAOKY silvalugo_j_Page_133thm.jpg
fb399f05a19d3312c903406b7b517a62
9769455d002889cf4b2b0cb591533b1977cb2075
21272 F20101204_AAALZL silvalugo_j_Page_120.pro
5f55d617e204e48b05d2c4fb87f3283f
5c6e16501ba5ce865025be9b3f0a1deab373d07b
F20101204_AAAMWE silvalugo_j_Page_022.tif
6640bd34d9a2e7f0483ea41e63d693fb
9b33b691a604e12baf841658867d15045ace2dc6
F20101204_AAANAU silvalugo_j_Page_154.tif
d32fa1f0a028486b310794ff9150e69d
0e42f58c6a47358ad16c86c50fc7a16cb0f8a4a7
8919 F20101204_AAAOKZ silvalugo_j_Page_134.QC.jpg
7bcf97ee429133397cb582636c2304a1
b81e11c66d0ef485db9eb1b1d43ad76a94ea938b
25500 F20101204_AAALZM silvalugo_j_Page_130.jpg
97ad9cce560b73bda41aacf20b40b44e
207da606f30f7ce5b6fe6373d7fad960b993525b
F20101204_AAAMWF silvalugo_j_Page_023.tif
4d9c003af0b9ee543d624edf9b82c4f0
b20cc0617b2487271517385c3b44a0c54ae1d90e
F20101204_AAANAV silvalugo_j_Page_155.tif
f794d860f0dd5ea9c082e38d7a95cc15
44ceaac6414cb84d8c68a8ec00b799c92d551886
35096 F20101204_AAALZN silvalugo_j_Page_186.QC.jpg
37703edf3150d275bbf29601fbcd6910
98f171965933d4325c70345c3f72b535a3c32ad5
F20101204_AAAMWG silvalugo_j_Page_025.tif
543af796424dd95f8bb98a4517dac094
53fe73b90d065b3e1bc2eb0669e77f3fd038ba12
F20101204_AAANAW silvalugo_j_Page_156.tif
42e664800209794f900119aab8378f12
cd0f040da27b62e8ebb2467b61a6e1ca9278ec1b
131861 F20101204_AAALZO silvalugo_j_Page_244.jpg
514db839ea6dba37a74bf750f1e37849
c2660c0bad1dc090d9f4f5f3c25d92d920bf8bed
F20101204_AAAMWH silvalugo_j_Page_026.tif
5cce92c57cb25a145a2cbf82caff1613
88227c88b173b96cbf4af343d14dad20e31a2fe8
F20101204_AAANAX silvalugo_j_Page_157.tif
18afcf82b1a026704bc45cc23a0c3bd6
94da7bf5ee4e00cb06b6426a58afe0dcf0e107ab
1829 F20101204_AAANTA silvalugo_j_Page_171.txt
9dcca1909251fcd2f754619dc7d5fdb0
1f009a64c5a23815edff056b62c60086af574c32
73208 F20101204_AAALZP silvalugo_j_Page_108.pro
18c93d3695c060d724f115b9ad9a8de3
0b0fcae62afa01d55b8b7b5b2ade7bc59906524f
F20101204_AAAMWI silvalugo_j_Page_027.tif
55024eb9b4c7912cf2083c96d671364c
deb9a7f25402a14b6d3e5366ba2aa0975e02fe2f
F20101204_AAANAY silvalugo_j_Page_158.tif
0c29c545032e011c939185cee5f6156f
e4c6500c067605fb263c6ead23fd7f1101a64377
F20101204_AAANTB silvalugo_j_Page_172.txt
607608e10071a33f34e824b431cb4397
894cb3abd5deba32461508985705031d16bcdc3a
F20101204_AAALZQ silvalugo_j_Page_188.tif
515d702e156c1fbcdc6d63b387aa9ffd
b31ddd740e2d093ef373daa31c8f05409b4fc8e0
F20101204_AAAMWJ silvalugo_j_Page_028.tif
5a68bf1d765cf313848e75d99c979af5
be2435cefeeafa8a74d7ef6d90d558ffe3aa39cd
F20101204_AAANAZ silvalugo_j_Page_159.tif
d593c76e306ac3f1391c1794e277a8a2
167ae8a6079e6027f9dc1d7fcedbd64450ca067e
2132 F20101204_AAANTC silvalugo_j_Page_174.txt
43237e997b8f5f29e643d83e3f693126
b47f412465c0ce0c18c85b2c5710d9a0071bab0e
62006 F20101204_AAALZR silvalugo_j_Page_141.jpg
078efc131343d3635a59647875021f14
99c112e831503e74311e2abdff620719f8e913c3
F20101204_AAAMWK silvalugo_j_Page_029.tif
beff520690fd82a59292789ca614a8f3
13ef474d7317f9ae9cb9bec14db9484dc68e153f
2271 F20101204_AAANTD silvalugo_j_Page_175.txt
3538864972633e427a448c2cd7afbc3e
2754174b30f25e9af0e9ead85ef45a40e0a823d9
45144 F20101204_AAAMJA silvalugo_j_Page_149.jpg
87a36542e61ebcd766ac5b3c00576348
ce9c2524dcc2ab0009abffae31c9156e880be3a7
9207 F20101204_AAALZS silvalugo_j_Page_021thm.jpg
38e378d774e69761a904a25398a1a0d7
79e2b7336b78cf5bef0bfbe40466e0909a2b74a2
F20101204_AAAMWL silvalugo_j_Page_030.tif
5f3860e8070a02047eed13444d985c50
159a23d2c2ad36c6e9b8bc145341c12446256a92
1995 F20101204_AAANTE silvalugo_j_Page_176.txt
8b80604f31378a6effcb564c48048ee2
c96d8be0029e87f9ea7ee0792db773f292672dbe
29171 F20101204_AAAMJB silvalugo_j_Page_150.jpg
c4b8bc9fef9305af0c854f3311deab54
061656833ac6c15177a93a4013790e51d1695155
F20101204_AAALZT silvalugo_j_Page_068.tif
7cfe2abe114557ba135012b8577df8df
ec49eee0beff50ae762df8b49b87aba12377b97b
F20101204_AAAMWM silvalugo_j_Page_031.tif
f405f55a4b82dcc2640c19fb86b87d28
9696d0e3dece075a85852ddae2086629785f2f31
2335 F20101204_AAANTF silvalugo_j_Page_177.txt
0b8397c38b385a919075d0284c0adb9d
ea746179caaf2b08a99cc1168ff30e8a18452a11
35232 F20101204_AAAMJC silvalugo_j_Page_151.jpg
e992c5fbd7b6e8dc472628a9b3942370
1c5f5ae22210ccf7876c7e548d68aee51dbc03c2
9175 F20101204_AAALZU silvalugo_j_Page_243thm.jpg
38548c373171b1c3e20397868c3ee4bf
82ec74017145976cf7420284cb01fde6e4cc5c32
F20101204_AAAMWN silvalugo_j_Page_033.tif
ddfb94046763fd6953a5c742f4be49b3
30439f1eb2678f0465dcbeb00dcdc8b2c1e76905
2153 F20101204_AAANTG silvalugo_j_Page_178.txt
ff26858db5fbec8f80a687aa3a6eaabb
76fb9292d7d5cf9201ea3d76afcc13094b02093c
36956 F20101204_AAAMJD silvalugo_j_Page_152.jpg
4ac8327709034f1b30351af815e50e34
5e669218a9bb91a70facaa6fca98a1a290ff46fc
8433 F20101204_AAALZV silvalugo_j_Page_011thm.jpg
a339209204c368dd3a07dbd673801fbf
8049ba64b16fefcf866b54076fd28aaec56c5e52
F20101204_AAAMWO silvalugo_j_Page_034.tif
6ea838faf094c3a808b718237dc005e8
d32995662ccc1af7fa73a56ab189d3678da59a7d
2194 F20101204_AAANTH silvalugo_j_Page_179.txt
33392c8e21ef45b856d5a1f4c1641bff
75936eb922a3ea550578f4b3344109389babf218
34991 F20101204_AAAMJE silvalugo_j_Page_154.jpg
74548afdad885e9570f88daedd03b953
96d0270c84217e449d278599fad68ca5063222db
19039 F20101204_AAALZW silvalugo_j_Page_141.QC.jpg
20cb862d413988a94c2c2d4977742d6d
e9ce8a3b50fe47312de56da6a1fe008ca6bda601
F20101204_AAAMWP silvalugo_j_Page_035.tif
e812e2a62cacfe56ce423e8cbc3eb5f7
5444c7afb8bfe98e47062696162cd6aa13688321
2184 F20101204_AAANTI silvalugo_j_Page_180.txt
8aae8500d23267c14ac171eca2e7d218
0555884b69a75ff55686d202e5bdd253c71ce5e3
39301 F20101204_AAAMJF silvalugo_j_Page_155.jpg
25875dc746602604cbebdc7d105f1b2f
6cd8f45052a14518e5e40d109c153c4990e2a174
F20101204_AAAMWQ silvalugo_j_Page_036.tif
3a82536769f26c268047762ae394d7f0
ff0de2c72976bb7821aab4840aa1243d4cee15ef
2272 F20101204_AAANTJ silvalugo_j_Page_181.txt
e2f513dfea153549b4744ade682ec5b9
3be2f73b77ec270f12495ad7570eb19223a02fe6
F20101204_AAALZX silvalugo_j_Page_199.jp2
90e1409961dad37c69202fe4ce124c33
2bdb2186aba15db63ea9049cb09d8b9693a99530
F20101204_AAAMWR silvalugo_j_Page_037.tif
07c970db6d9bcd2c0c159918f9ecb697
4ef183a39e9d8468a99d82442cce24aa3ae3fe73
F20101204_AAANTK silvalugo_j_Page_182.txt
9d1a2263fe445035057f12c64400c34d
8ec8c780a2c3cf0182e3d8c60baf741eedb274f3
36528 F20101204_AAAMJG silvalugo_j_Page_156.jpg
92a7880197f4ad8c768fc5519d6cc510
3b6f83c9d21650ea1d2068cca4c8ebb6fd12781b
F20101204_AAALZY silvalugo_j_Page_021.tif
a577376b4ea454604389017dbfc76c82
0578d57baab6d1a117e9fb9ca474e759153f42c3
2840 F20101204_AAANGA silvalugo_j_Page_053.pro
53ae6a11c1dbf79fe2775e3ef62c579f
400589b00a0ca4a838eb71db2bebb8f73309b2cc
F20101204_AAAMWS silvalugo_j_Page_038.tif
91764b012e10608e051439a6bbb44f7d
c72daaeeae3b224f3be9b798fd6b30b914905131
2057 F20101204_AAANTL silvalugo_j_Page_183.txt
56ebc49537936740684372d87655833f
764fc904a9d19bb4cf019c5bcd18e3e0a7add9a7
107861 F20101204_AAAMJH silvalugo_j_Page_157.jpg
674d8492f3bd0ec2b8e1c841d97dc6ec
132ff28bf799c297b72c533a3f8a99e3f6a8351d
F20101204_AAALZZ silvalugo_j_Page_115.tif
00d70eb0fec22fb287d81f29cb7d3cb7
5836e800ec3427c363f0efa6494cb5479acf75be
3138 F20101204_AAANGB silvalugo_j_Page_054.pro
8ca1084750e35df814cc1d63d6ef0d07
b9e8b669ee75613ebb10f769b22e935ae21dd5cd
F20101204_AAANTM silvalugo_j_Page_184.txt
ff32f85c63a4bfecbe4ae4b34b2f9007
5e69418d781245a78661e205b1f7b72b9b4686e9
115441 F20101204_AAAMJI silvalugo_j_Page_158.jpg
841e9a1cd181777ae80963e149954949
8647b26938db0b08e020bf6ceab27d051e045852
F20101204_AAAMWT silvalugo_j_Page_039.tif
59e800632891e276115139adf809a368
57602cd9d7260493cbfcb4f897796f2bda4e1a5b
2161 F20101204_AAANTN silvalugo_j_Page_185.txt
ee8861bcacba49b25619804dc03a466b
bdd90f5ea0b74e7a8ff44b1763038687ff0fa8e4
113807 F20101204_AAAMJJ silvalugo_j_Page_160.jpg
87def353ee23018f7e99d33b8331a4bb
2e41b103e16b13d1aff117c1c5172bd8c7172100
5655 F20101204_AAANGC silvalugo_j_Page_055.pro
cec974c94e22ddde7f32389989b90399
4997ba25c2e42ccf51dd4ffc5bc9ddcc874615f1
F20101204_AAAMWU silvalugo_j_Page_041.tif
02636acfbfa602da113065c486ac8f7b
bcc5c2e5a12ba91ed80fa082c94bf24c914aff06
2125 F20101204_AAANTO silvalugo_j_Page_186.txt
4c23c225f247607a3d763a55f6d02097
a827c0ceca91fbfe4582d895fc0ffc83955c4e93
116959 F20101204_AAAMJK silvalugo_j_Page_161.jpg
4efc003f8e55960f69cc7bbeb84e6fb4
e6f2ae810a3dd50bb39b281939446dcaabf47fd4
8325 F20101204_AAANGD silvalugo_j_Page_056.pro
1deb2b42438b68c212b488e3d0bc6e92
974745136effea811429372a11b04273cae50387
F20101204_AAAMWV silvalugo_j_Page_043.tif
819ebf6fa5d2d4ea48ffa04aa8f5aec6
db9359a69855a6e65cb139196727c0415021f0f6
115091 F20101204_AAAMJL silvalugo_j_Page_162.jpg
4cb6a719d596b00f2c03ebc4d9cae06f
868347a609a626731ad53e0b0dc4040a8aae8cf4
4490 F20101204_AAANGE silvalugo_j_Page_057.pro
9adcbb95a83dd7b797dbb7fdd97c7cab
1db8c117408d58ac0a6b05ce026389c5a9ab16be
F20101204_AAAMWW silvalugo_j_Page_044.tif
c13e3adce5c61dd3dbe95ee3afec10ae
5fa697a8146f624394b9cea8f2ae4e8f71c703fa
2230 F20101204_AAANTP silvalugo_j_Page_187.txt
672102d0a314bcabf600e29394d83779
d9f8eecd3b8f202aa9925746b691760c1ddc0915
115880 F20101204_AAAMJM silvalugo_j_Page_163.jpg
a2a8270e6687c7fa513f6d8378558789
21ce474cbf01e7eaba4270410dd1f2bcc3972ede
4678 F20101204_AAANGF silvalugo_j_Page_058.pro
b61eebe198cbdd6403be422354c187b2
c5b01c99e96616e6d0d032392fd01ca6680f3e39
F20101204_AAAMWX silvalugo_j_Page_045.tif
0f6ab945e203f6e970b933f572f304e6
9e1c9cf17cfa1e77223dc4aeafe16d1227a8647d
2250 F20101204_AAANTQ silvalugo_j_Page_188.txt
8e8350ccc3e3ec4f70858d1fa017f107
a0315f1afcf82c6bed6857c6834e206238571b9f
113564 F20101204_AAAMJN silvalugo_j_Page_164.jpg
1bbe5cfd42bf3da2941b3bf6d7ee0a61
cda109ab3ac73180b91cf0098ba2ad66538ca8fc
4518 F20101204_AAANGG silvalugo_j_Page_059.pro
2235554a6db6ca94c4ec28929f0f94ec
c1530ee841155b0cc600e00896da76eac1d50b6e
F20101204_AAAMWY silvalugo_j_Page_046.tif
a06feaa055a1463ce8961b5878d2a3ef
22e90c49bf68d4a8005c47926a9a34e5bd4ce9f1
2227 F20101204_AAANTR silvalugo_j_Page_189.txt
68a43991f52e64e7e614dae9b5afc859
6b83f3f2b2fc73921b600dc3bd57c93174acb7ee
102880 F20101204_AAAMJO silvalugo_j_Page_165.jpg
aa941f17b82f1af7350746141967e77c
d1c5c54eff38af56690c9a120152b3dce30c5610
6070 F20101204_AAANGH silvalugo_j_Page_060.pro
3635278d97f4a4b0a6808104517a18a5
bc4b0117ede83918d154f6a3bf24ff45a21a0b9a
F20101204_AAAMWZ silvalugo_j_Page_047.tif
4fa6db83d95defed98a303bd9590695d
60b78f16ebd974488c3c618982b95b6cc2224db0
8579 F20101204_AAAODA silvalugo_j_Page_044thm.jpg
bc742ae24fdf5adcfe692c319f91a141
971d02d1ec2a106aa2ed635ed432f1d9044c8f20
F20101204_AAANTS silvalugo_j_Page_190.txt
25de416bad28a3ad0382d4089f9f3d16
9ecb9b9cb8aef6dd9b83ff2949f95a551dbcae7f
111044 F20101204_AAAMJP silvalugo_j_Page_167.jpg
9b928041c8dfaabdb8779e8417cc74dc
bd6413f0acfd2960ddeaf6a89c59312257632a10
4080 F20101204_AAANGI silvalugo_j_Page_061.pro
f81f189a0d9ac34b32a539347acf53f3
42522307d0b0bee0f749227f7efffb52db035037
5670 F20101204_AAAODB silvalugo_j_Page_202.QC.jpg
cbfc4c779e5637768555c763b5ae9f4f
93a7980be5e133321279000560c2a85f8ef52b8d
2166 F20101204_AAANTT silvalugo_j_Page_191.txt
004187d81f7f2cb491befbdefa3ef41d
e4785d5db51e9b8d26503c05860b7bc10f2ec8c1
101549 F20101204_AAAMJQ silvalugo_j_Page_168.jpg
1fa7a2746179614d387e5ecccdda48c8
ed291f237c2d7f2a18aeeed47ce6e1b046c1b08f
11940 F20101204_AAANGJ silvalugo_j_Page_063.pro
c9340e5e9bbdce59c33231a030c3a10c
25e4c7d185a22b6819eb4b397afe979815f4047b
8420 F20101204_AAAODC silvalugo_j_Page_082thm.jpg
95cddd85f0e433244294b6aff9b85778
2adab8034f3dc160e1b9ce9bf863b2403de77afb
2266 F20101204_AAANTU silvalugo_j_Page_192.txt
c925c16706edb9db2c444549b5923472
f67a441c199691aee5fc6edbeeb88083a2af2f6d
112549 F20101204_AAAMJR silvalugo_j_Page_169.jpg
ed1e54e9f4f568d8069b4b703622f6b1
a8d1dc48b6207ef62bf82053eb4ce478eda03f84
F20101204_AAANGK silvalugo_j_Page_064.pro
55f5797059fc68080198800c77416d3c
dd4def5bc9d17a4e81f3f38056ca354ec5d623f5
9059 F20101204_AAAODD silvalugo_j_Page_191thm.jpg
a357015b188bccc74beb3a2b8f60f68b
1cb1aeee5a6f33b6cbfaf1cff69a24a3fd0a595a
F20101204_AAANTV silvalugo_j_Page_193.txt
40618b04a235a49f6ffc67da3eb886ed
336c09595555a80c063b9063482e0b6fb60a4437
106564 F20101204_AAAMJS silvalugo_j_Page_170.jpg
333650d521d920b56f716d1bbda54eed
936e7c63e28ec64eda0b03d163813c8d2bfab4cc
52592 F20101204_AAANGL silvalugo_j_Page_065.pro
4c0c4aed8b2a5eaa96224dca10c77970
6f51fc54916675b6b5552d745dcedc7f30739bfe
9522 F20101204_AAAODE silvalugo_j_Page_240thm.jpg
2c8fbc8d5a7e7a128a47e69bdba3a1eb
73996a355eab0ac79285a3b6ab4c1dc36bfbfbfa
2237 F20101204_AAANTW silvalugo_j_Page_194.txt
349ef789dade581c8dc9607800068ce4
6da47b5fc59046f31a074dd21aa21a0eec6db024
91988 F20101204_AAAMJT silvalugo_j_Page_171.jpg
b812c670d61e72fee77355d22225a19d
fc1c51aed8747a3a23b239c735b0cc52b5d7fffc
55969 F20101204_AAANGM silvalugo_j_Page_066.pro
2cf776ad125d01c2f113d0aa53c1c848
6d84c3a577cfe513de1634bcf05b60f12fcdbbd8
35968 F20101204_AAAODF silvalugo_j_Page_084.QC.jpg
0cea82ebde8a90753f49156a41f6ba18
711cd4dcb27325c1c8be7b51664df3e8bb8f44b2
2198 F20101204_AAANTX silvalugo_j_Page_195.txt
2ab8e715eea9432dac98c8383fc4ff13
39eb4a51aa2d75e341987ac6e1d4641cbe696bdc
113157 F20101204_AAAMJU silvalugo_j_Page_172.jpg
a4c6a9851364b92ce25496942ff73fe0
6e07e73bf7ab16c5f24021284cd57923bbe00df7
52876 F20101204_AAANGN silvalugo_j_Page_067.pro
38ed46343f5c9c609172762f7f1c681d
858869fb28c2eccd239401319c023cb7f44d5e9a
10554 F20101204_AAAODG silvalugo_j_Page_009thm.jpg
c927859a21288a2b50b93428ba8ac92e
a24690bd681fc79a11e00d9e5774c2fa389e02b7
2297 F20101204_AAANTY silvalugo_j_Page_196.txt
1884bb2f6ede50b1472d046ae373783f
46f5cda8e9d7d3fd33c87333151a85757883f2b3
96205 F20101204_AAAMJV silvalugo_j_Page_173.jpg
9172b6d42343f4eb8c89447d8fa3d6c4
7217cb735c6951a2f73c07c06c25eaddc7bd12fb
54749 F20101204_AAANGO silvalugo_j_Page_068.pro
bdfe141983b7c9947326c0219319910c
87027383bdc5207f1f79077b4840df20254bbf20
9150 F20101204_AAAODH silvalugo_j_Page_091thm.jpg
e4f0f9a2971ee54a421730bc4ebd4b08
76cad2d9040308e0a3f05cdafa56354f9aecd75c
2200 F20101204_AAANTZ silvalugo_j_Page_197.txt
1a99a423f77f35b0b0df5d3045bb3fe4
10a98399d0eb1efe58b932d1dabeb6c64db29b11
107837 F20101204_AAAMJW silvalugo_j_Page_174.jpg
5a293c707393e6fb435b21a9d7296ea5
791b3b9d00caf0c3c14a5a85059a82576e97d379
57853 F20101204_AAANGP silvalugo_j_Page_069.pro
9878d825dc27719f380ee0c08e1e36b6
131f132d4d54f597e5334a294718f439d7e94233
22817 F20101204_AAAODI silvalugo_j_Page_029.QC.jpg
9f8bd2c901c6673c01a3e268d8c509aa
ec93c553391803d78b650bd8f0f9d945300ba9cc
116230 F20101204_AAAMJX silvalugo_j_Page_175.jpg
cb19beec234c516b28cb490be0b026fc
ea63507ac1b5dcabd17d4f93f452e295265d5dc5
47852 F20101204_AAANGQ silvalugo_j_Page_070.pro
2ff1740f32759eb3c3f004d1b49f8e92
3af89be2dbfe74a781177ab96c341ce7ecaac157
36799 F20101204_AAAODJ silvalugo_j_Page_162.QC.jpg
bc40c92f7c88db4e635eaa3a1b057935
8f2e44a785a71e0f318b9c061e817ab5cde8263c
103500 F20101204_AAAMJY silvalugo_j_Page_176.jpg
795ec5d8e74ddb74902270a1f740ba31
50d9c4a19a3534ffc20c0bfdda6870ed8bb4d05e
63665 F20101204_AAANGR silvalugo_j_Page_071.pro
f09af0e7358cea37a5cdef7d5e90884a
20a713cbc59798e8f99c0fbeb2d16fb1f22ca78e
8092 F20101204_AAAODK silvalugo_j_Page_248thm.jpg
c6486a031b2eda5b1ceef91a14268545
8bf60766cad909d84881007b279b569db047c51f
118706 F20101204_AAAMJZ silvalugo_j_Page_177.jpg
5b665716ab61353f16853678df500358
ca09ead316703f11e354a175a69985ff7b756d84
55843 F20101204_AAANGS silvalugo_j_Page_072.pro
f1713b0a420258e8d32a24e1adeedd70
701e4487c6392cfae6ccedaff3e5babf003306d1
36568 F20101204_AAAODL silvalugo_j_Page_227.QC.jpg
22cf882bc54bca6edfcb70b429805b86
cdcc159691cd7e829ffed59ce7b2b5b6e7a4aeb0
53580 F20101204_AAANGT silvalugo_j_Page_073.pro
ceda9d39674d5bc590ad4ac6470864f4
fefe66b1a3cc998a693ca88621e557ac8f1f7f8b
9917 F20101204_AAAODM silvalugo_j_Page_007thm.jpg
7120f066656c1cadeba411939affa38e
f55e036db5425b8c055c9ca1b6d4875b976f36e7
55913 F20101204_AAANGU silvalugo_j_Page_074.pro
29cbe379148ca1a151b537587ab6ece9
f06d26bd3019a63daef5a4c029a81d4dc7dc7562
37355 F20101204_AAAODN silvalugo_j_Page_066.QC.jpg
43b94eb843dfdcb8699c8afef8c627e9
5688b60f64ad6c61ff45d87c1ebf5b3ac8c0adfe
46195 F20101204_AAANGV silvalugo_j_Page_076.pro
7d66c6be867fbcf8f0e294293819bbc2
bb1b2668e53c7d29a7808015b2aa012ff3f93ded
1270 F20101204_AAAODO silvalugo_j_Page_010thm.jpg
5242437aa65496569ba4ab514f0980b2
11dd3825fc200ae70f784f886fb3c3e9e1d9e578
48308 F20101204_AAANGW silvalugo_j_Page_077.pro
ea497d54325e7d2f79d8b3e9e7539f05
7f1516a7e377aa7b4283ea45eaa6baa2f2f59048
34552 F20101204_AAAODP silvalugo_j_Page_060.QC.jpg
875103f4960866cc0a33a94e14c3c993
878645c7b8792e388f92f58180ea6ad0b82003fe
56821 F20101204_AAANGX silvalugo_j_Page_078.pro
706f05ddd389e51c18d6a0765a647a58
f2329d72842269bcca924ea0079e371c39610888
32637 F20101204_AAANZA silvalugo_j_Page_032.QC.jpg
6e74c03486963ae22245d8911d51fb9a
fc1b06a6ac73944d2ed3ac07cbf3ad516ab6c286
34767 F20101204_AAAODQ silvalugo_j_Page_114.QC.jpg
7fba225f2ad171313f53342057ab6add
cbcddc27c6d0087104deb1177f112ddf22abb490
54676 F20101204_AAANGY silvalugo_j_Page_079.pro
b7850598bbb0e7d5d7f43abddcb26cf1
531ec0fce72d589225f5c5ec88bd30f842b43bec
4999 F20101204_AAANZB silvalugo_j_Page_050thm.jpg
1bdadc38a77b514ec7d9135e8f91c826
97f23dc2a506acf3db10bf71af96a4b0eaad8366
7831 F20101204_AAAODR silvalugo_j_Page_042thm.jpg
95e04f6e2d83e366bec51b1f554d013b
eb924fa638d0fc66d26cd026d79af00f990d276b
47994 F20101204_AAANGZ silvalugo_j_Page_080.pro
b1dda4ae5d113d49f5e207eac8cbb778
0891e94c11ec987976b9bb89a285e02476191ab5
34135 F20101204_AAANZC silvalugo_j_Page_011.QC.jpg
05500ea03f39ee8ff7c248c6df520d81
888dbe9b4b6d7b4216dd0c7c7a0a0c2ba67d09ce
40465 F20101204_AAAODS silvalugo_j_Page_013.QC.jpg
3c45179e24e95d830a163eefcb301354
37010512c999f9b47ae0d560e4b338bce9a43748
10378 F20101204_AAANZD silvalugo_j_Page_125thm.jpg
eeb81991fd174827eb59ffb72331026d
fe83f606705ca9583cfce09e940e1e3bfc40a001
35997 F20101204_AAAODT silvalugo_j_Page_172.QC.jpg
0fbb1471085b41f29f8816b2ef71c499
811ab19035b8c0c881271d34d76bd0138a800cf5
24403 F20101204_AAANZE silvalugo_j_Page_028.QC.jpg
d0cd3a8350665d3721b72d213495c918
1f165ff6ce8b8a485c94be31092fed8569cd322d
36071 F20101204_AAAODU silvalugo_j_Page_200.QC.jpg
e19de97422a53f717ca698ccbe1e3ab7
ea8717ec42d38ac356105cfab88c84d25d08100f
F20101204_AAAMPA silvalugo_j_Page_069.jp2
c0488c07f7f6a47833ae8a529a4962b7
8d88afc3f60c50733504cc1a2c1ff820783ab5f7
36802 F20101204_AAANZF silvalugo_j_Page_187.QC.jpg
85c3f58ef6e50ab3b413e8fb70d98e8f
1a902ba7458d053b171e95cc12b097eaef66667d
37818 F20101204_AAAODV silvalugo_j_Page_180.QC.jpg
3da59cdbe088f5989c56bb18fce8d57d
11ae716ad35e0efc1c7c5de43ec115da48add5d9
1051797 F20101204_AAAMPB silvalugo_j_Page_070.jp2
23d09954f23b169fae228e69cd04e38c
19ff67a4bcb461211d278c5db4256a9531120ecd
3213 F20101204_AAANZG silvalugo_j_Page_220thm.jpg
fa90323c81e5043c8b6d2c700e4a4ba3
a532f14f154e9ce8bb8c02e637ee0405371abc34
19880 F20101204_AAAODW silvalugo_j_Page_059.QC.jpg
11147417eb597a93e485b78476cba8b9
131726c9941eaec05adec67180151c2f90ba6671
1051961 F20101204_AAAMPC silvalugo_j_Page_071.jp2
18e5ef34be9912d973f529bf22408daa
a8a8848537782b78a5d230a2ecdd10c7a1ba9933
37734 F20101204_AAANZH silvalugo_j_Page_096.QC.jpg
e06d6a05b72be6b083919555200d82ae
b386e7cb9c7aa28140015a36513d783459055309
F20101204_AAAODX silvalugo_j_Page_138thm.jpg
312f39035120ce3a5a8425c92109defa
cb9640546185d5773105bb00274ca6b0e1bf5173
F20101204_AAAMPD silvalugo_j_Page_072.jp2
a207eb1442f1264918333aec80cd340e
2e8749e33e33ac26c8485c1cea60de79d2c79146
9005 F20101204_AAANZI silvalugo_j_Page_072thm.jpg
874bb956ab0cb53566894125302b7995
8c500f00b96319ae3928d7ed2a107f02b9702fbc
9225 F20101204_AAAODY silvalugo_j_Page_069thm.jpg
b03ebb49be7f32dfb38fc67642cd668b
105551d0debcd30f9fab5ef10edb28809735a721
1051951 F20101204_AAAMPE silvalugo_j_Page_073.jp2
572174a26b195b50ff94aba4ef28c873
073b594f55e1270b202451854c9ee34a83a6c9fd
38649 F20101204_AAANZJ silvalugo_j_Page_196.QC.jpg
263df24e81f040e510b5782127f8ec9f
a1d136d61254d961339952d491a459caf3d90f93
8803 F20101204_AAAODZ silvalugo_j_Page_053thm.jpg
bb1877c51065b5466099d399a6e9179d
536d042cb9264c3e5b7e5bdabbe3c7a25a658d63
F20101204_AAAMPF silvalugo_j_Page_074.jp2
70fda53a1f2662ab0c75fd66d0ab505a
a6abdb0d7148292343486899373f1a34601aae7b
9282 F20101204_AAANZK silvalugo_j_Page_241thm.jpg
0452210a4603a2d96139969160dbe09d
d5cebe4b8befd3f3d5561e95fd434a8c0f8da99e
F20101204_AAAMPG silvalugo_j_Page_075.jp2
33f3a1caa85c1c2257351dadc2eb26f5
8bc41ed49cb1f6c9a3acb4e27baa9e9f105dba29
11184 F20101204_AAANMA silvalugo_j_Page_223.pro
c2947ab0961a79b4b7db48c11b7e6413
e04b5567622b851afcf8025c56b1a42f8df9a713
4059 F20101204_AAANZL silvalugo_j_Page_118thm.jpg
9efab8c32e36e37e3115b8bad25fa0e2
5bc425bc5a7701bff9e8f23b26ddc08735a0f953
1034782 F20101204_AAAMPH silvalugo_j_Page_076.jp2
98f2ff99692ce2cd107b978988d0c01c
72105cb77e222730f3d3eaa9d1cb84c9ca4781c5
15719 F20101204_AAANMB silvalugo_j_Page_224.pro
11b603e74966569fc1a5cbe61902c6ef
176a54dd238f34c1fd676382beaf7910d2d55f75
10653 F20101204_AAANZM silvalugo_j_Page_220.QC.jpg
0113e7047a298e8af4b68513e7308d71
1c34b7b7cdee3295a1a59f2dd3d6dddc35ffa5db
F20101204_AAAMPI silvalugo_j_Page_077.jp2
eb0eabb1581e3394b9cf1351b50ad872
e49d98f6cf17bcae41ecfaadfe117d759062b581
47092 F20101204_AAANMC silvalugo_j_Page_225.pro
1fc9bc369d5c131dd0eb2070ec8a8d06
6d190df04a575cd299c9f109c9f609d4251e344e
37593 F20101204_AAANZN silvalugo_j_Page_244.QC.jpg
3834db788a18b2bd51a2ba0145540531
1b48cb4f49f39e744860330f544cd60a7e07b5aa
F20101204_AAAMPJ silvalugo_j_Page_078.jp2
20755256d6af7581968d26d65a199510
f5e1992b8195aaa3eaffca321a0f8b08202cb2d1
54814 F20101204_AAANMD silvalugo_j_Page_226.pro
4d7a674e141e31170eceb30ca3f5c7ff
c7728b6de90bac6293ddf3a1feeb7a3aeae53a14
7663 F20101204_AAANZO silvalugo_j_Page_105.QC.jpg
6f93d415a1baf59cee8f0bfe736e31e7
670dda1f4b10851b47a27ddde956ca3c6e2f169d
F20101204_AAAMPK silvalugo_j_Page_079.jp2
f5c463c69dae36b153017396b07ee138
99a4e6088b12fe37569b6f435e68205b53348ea3
55371 F20101204_AAANME silvalugo_j_Page_227.pro
2b1069636afe3c22500b886769781a92
be107e8ba3e7390cc56505070411b93c9d18a6e1
5038 F20101204_AAANZP silvalugo_j_Page_119thm.jpg
0fd64c16b6a5f54a6522ef8aef66a866
645f88da86076b5ff58b16c7caac02742581021b
F20101204_AAAMPL silvalugo_j_Page_080.jp2
a9b7066028c7b2914321af0c4e70f958
7bef5809b8211406ebd5b8f70b2dec208a28cba4
52912 F20101204_AAANMF silvalugo_j_Page_228.pro
0dd6e97c2133d3b7c24c79b9a7fd308f
05774e1dff0afa4d2d0bc385480d9436d0e9c159
83852 F20101204_AAAMCA silvalugo_j_Page_122.jp2
d5d24fc12e9ed20d277d54d323d1cba4
2aaf7fb29ecc0efa9024488ab9ba51752296e9a6
1885 F20101204_AAANZQ silvalugo_j_Page_105thm.jpg
dab5b500fd2d1bbed85fe23bb941f3b6
83105835a931db9f8e1c0f6c36759591b4a085b2
57874 F20101204_AAANMG silvalugo_j_Page_229.pro
05cb3f82abe348ad789daf3ae898bbd7
d67c67986ddf81337126d3c8467a785b1f85850e
730 F20101204_AAAMCB silvalugo_j_Page_130.txt
4c968f306932835d787860a77e35f10b
34bdc467beb834ba9518b5d2a1edf5c87adaecfa
3980 F20101204_AAANZR silvalugo_j_Page_155thm.jpg
a9f78d2ea019377c8158ef0dba2fa54b
260b375bb94152fc02b4c3aeb675c84f23f0cc64
F20101204_AAAMPM silvalugo_j_Page_081.jp2
8a10fe092a9a4a3918091eb6eb617a56
544476eb72eeaae6af50f3dc7fb4ff747b19fbb2
55014 F20101204_AAANMH silvalugo_j_Page_230.pro
98295e9760ebd736de372406a07a414f
73e3035f819f17acd8ba05d8930ed67cac5ae5e9
34180 F20101204_AAAOJA silvalugo_j_Page_081.QC.jpg
be2222f491890459f0fe779de6bfb317
b2deb7d05953cede9a52ff9adab781441d940217
F20101204_AAAMCC silvalugo_j_Page_015.tif
5829bd73fe24d9ec055d5299ff717adc
de9e4e341e8b1a02b9071ac11e8aff21b0715641
6645 F20101204_AAANZS silvalugo_j_Page_135.QC.jpg
043d9827eb0bced3e49db7989aa88628
be92485578ca9364ad60e73525757004ba03e33a
F20101204_AAAMPN silvalugo_j_Page_082.jp2
43b27af8256f46503a53d58c523da080
286449ed0464c12977ffa33ca71af16a76caa1aa
8505 F20101204_AAAOJB silvalugo_j_Page_081thm.jpg
8318d06e2d0d071cc9c05841cd87b99f
b2467aa63d18d09315120f20c15294f304cdcc46
F20101204_AAAMCD silvalugo_j_Page_065.tif
42c880a2f3774cd483aa1523af652daf
3d6dce3d3399ee41b805f34b10680264e0b5ed4f
20917 F20101204_AAANZT silvalugo_j_Page_232.QC.jpg
6b83acd1d251a1b5a5ccc8f4ece38da9
0c4534b2e745339895df9cfa1d6069bf130f147e
1051926 F20101204_AAAMPO silvalugo_j_Page_083.jp2
8b71f7101856256df590040d3174cdf2
69771925c8115c6cc5e1d8e49f39f734838c9eb6
53197 F20101204_AAANMI silvalugo_j_Page_231.pro
68650e62c10b52055d44fa83aa4a95c6
0380b35cc7c62ae0e8bd525421626d54d43d4202
9093 F20101204_AAAOJC silvalugo_j_Page_083thm.jpg
67239c7fd9d2aac9f5f8a7610dcaf810
857f4bef363b20b883dad55741cb34e09061a95c
F20101204_AAAMCE silvalugo_j_Page_169.tif
078dc3609b9b8ccfe8b3afa50492e6f2
9e40d42494564782673388431d4baed275f7e8e9
32182 F20101204_AAANZU silvalugo_j_Page_077.QC.jpg
0919a7accefc8c0e0c3f2d9ab36cafbc
8c902caef00578ce08d80fb50d6247a715d8b5f6
F20101204_AAAMPP silvalugo_j_Page_084.jp2
a7f54c9772427f08c1cf07f3a8293d61
39ced5b67c570ddc7b7325b5f5e54ec364a9f721
31009 F20101204_AAANMJ silvalugo_j_Page_232.pro
f49614b7f11dbb44bfc35ded291e356e
233d988c2062be4187eefe323f1b4400c3615e26
8958 F20101204_AAAOJD silvalugo_j_Page_084thm.jpg
79135c39ce7a7a8115a95d631d88d2d1
2bd4d46d47d4b2cfa875e33d8b74c9320bd48061
8139 F20101204_AAAMCF silvalugo_j_Page_064thm.jpg
57732d7e2192e364faee0ed20be7a4d3
8df11d8021e011d07701279fca3b9dfbf81de315
1051949 F20101204_AAAMPQ silvalugo_j_Page_085.jp2
ab6bdba82defb116ef8be98122ae94ab
a1dc347a9353aa819a7e801921c0045dd7cf4913
41683 F20101204_AAANMK silvalugo_j_Page_233.pro
e16a995ca7dc0bfef418cda3b5c78409
85d25db5d3f7fe5de215285325bffeca7bc62645
F20101204_AAAMCG silvalugo_j_Page_190.tif
03d8ae18bdabc21d5a5e80731dda6ae3
4672977bd8607ff346e69d3c662ec296f6ee26d1
8916 F20101204_AAANZV silvalugo_j_Page_023thm.jpg
26f17658dd3ad5a534697a591739e0f5
da6fb254ec5c6a43f728a9625e4060876a06f82c
F20101204_AAAMPR silvalugo_j_Page_086.jp2
88ec6660a424c50bfe5bd48f9fc284c5
578faf92fa4b9058aac92577348e734c0e689ce7
56498 F20101204_AAANML silvalugo_j_Page_234.pro
7400b55e0067eea12466792febbb0106
ac6a38701530153956d952245ac355ae6c816c92
33665 F20101204_AAAOJE silvalugo_j_Page_085.QC.jpg
24587726544c57a92db09ea57c60eb81
2ace60e501039ee0635d50a946c2175165bbe3b4
F20101204_AAAMCH silvalugo_j_Page_040.tif
dcaa6cd08f1f7c7a1b1a5f5f9f41e217
27b65829ad0a24075fbf38acc12bce93d0871889
10151 F20101204_AAANZW silvalugo_j_Page_129.QC.jpg
8c7c4abf03604bcd26c591babe2945f7
d4bd707eea377c6b02b9d7a65d139935a86dcfb1
F20101204_AAAMPS silvalugo_j_Page_087.jp2
b4a4af0f4873e70d4b296670ab7316e9
3addf554051565ee74aeffb2f4b0f93fe980cd7f
62375 F20101204_AAANMM silvalugo_j_Page_235.pro
bc1bc322b03d8310535a88d13060df1c
7859ff0841a448bfedef3fc2b73de879aa955158
8393 F20101204_AAAOJF silvalugo_j_Page_085thm.jpg
c7d5ba6b2e349b20ae8c39c8d3d33f34
fc0c5873f9b6e0253872715543dab3ee393633e9
136672 F20101204_AAAMCI silvalugo_j_Page_246.jpg
aaec7d138aef1d17ce0cf2a007585b57
07a16c67a6b119992930997363a8f43fdc260c62
33666 F20101204_AAANZX silvalugo_j_Page_182.QC.jpg
6447cc5c7920f07ce95ba7478eabb56b
e92db4de7420cc5651c4598062b21a04743a8305
F20101204_AAAMPT silvalugo_j_Page_088.jp2
cfdfcb4ef80dda238f2035810f39ebf5
f53009ade04d468fea0022420edcabd604d96446
63882 F20101204_AAANMN silvalugo_j_Page_236.pro
f44547bc3847dd3651393a5aaaf76af5
711cbd717fd832f497b013c8c0ccb6b2d1b0fd10
38751 F20101204_AAAOJG silvalugo_j_Page_086.QC.jpg
aa98563f327063505f1aae8ab25cafac
65e2d81fdd93244762d06202a2e4f980d274230b
2040 F20101204_AAAMCJ silvalugo_j_Page_165.txt
150805215aaf9eb5460310edf01e5a47
6b04f0810a28227289e12dcae96652fd4725a205
8817 F20101204_AAANZY silvalugo_j_Page_051thm.jpg
af810cdc1d4f362812b05c8335193141
be02d673bdb0bb58454cd40e3b996069eda7bed4
F20101204_AAAMPU silvalugo_j_Page_089.jp2
b98d364c40c50e78f152776a09b5643d
eb969bd266947887d7a997427c55edf4b2924f68
63090 F20101204_AAANMO silvalugo_j_Page_237.pro
b60e0b9eafa47e5b1f261c029c6eddee
15df44990fc2039d220eeaca1d0103dbed98b471
9308 F20101204_AAAOJH silvalugo_j_Page_086thm.jpg
3be306e25d78f25abecd479888329ad4
7d26215d85dcd3252775342163bc31b89029994e
12762 F20101204_AAAMCK silvalugo_j_Page_154.pro
6850c9d30bf4db88000beb22c90a52cb
2ceda7a19508cdd54fa383b36b03da4f172d0e70
6782 F20101204_AAANZZ silvalugo_j_Page_211thm.jpg
323db35cf0ed8539c262e598e3883a6f
393e268a58977d756bf971b413642bbacad35109
1051919 F20101204_AAAMPV silvalugo_j_Page_090.jp2
1169776bf0acb56c859ba875dda392d2
4e0b0c51067e9f9bf02ed1a7402863c8d466e2b1
64702 F20101204_AAANMP silvalugo_j_Page_238.pro
19f4f564a3b44425e54bc73d57475ca9
bc57d7cc6e4d68abcf68318d33c4e481f2c918d4
36130 F20101204_AAAOJI silvalugo_j_Page_087.QC.jpg
b318764457afa3230fc3373fc27a3496
b96aa75e9d14a9afbe7b4ff86f1baecae4a0fbab
F20101204_AAAMCL silvalugo_j_Page_032.tif
8a1435d8e53e197f2f9fe010d4c00271
31694c0aa988011300b3b6e1e3dd4c14593c72ff
F20101204_AAAMPW silvalugo_j_Page_091.jp2
74b71a952b93e99359cb1059c5850c06
403feaf1264adb0f1c451a0bf2353a12820ce4f7
70935 F20101204_AAANMQ silvalugo_j_Page_239.pro
ed84d565c2aa09443e0e12cc90eadcc9
37cb8a84a17eebc8b328b729926f8847f38eae6c
37193 F20101204_AAAOJJ silvalugo_j_Page_088.QC.jpg
f1f9661ec5070b43e999e11efb7f4f33
83cdc3aa58b6dbd663dca7f66185151d370238b6
1051942 F20101204_AAAMCM silvalugo_j_Page_240.jp2
6006c8e38fe91a97ee72d331e8055a51
36cf85ea5e3bcf52d03c42274b74f26ea62225eb
F20101204_AAAMPX silvalugo_j_Page_092.jp2
a2582c87f3f44d692e47665c08241e88
fe10e2d8a78edba5bf5441a70478bb0d655a43c8
66149 F20101204_AAANMR silvalugo_j_Page_240.pro
28ac8a78f731c67dc379dd8f71f7ce91
ad70367238e74d7dcdb39bd400dac758c2162e1a
F20101204_AAAOJK silvalugo_j_Page_088thm.jpg
ce0ecc62b5695f6ac40d597e9be4cf70
3f83b8b540edc32edb3c9003231a501cfb93d4d5
8509 F20101204_AAAMCN silvalugo_j_Page_017thm.jpg
045cb4014baf5b6a4542a2d66ff47e3c
2e18203cf7806b8e61fd90b1eaa85e5728684927
F20101204_AAAMPY silvalugo_j_Page_093.jp2
562ffe065566d117f702582bc874d756
568019c9f6c28becc95e906ce1281d52df216aa8
65804 F20101204_AAANMS silvalugo_j_Page_241.pro
0c7fb3629c41cbd28d00f8cf698a4292
41127929f480c363b1d5a4475de776d007c31c15
35034 F20101204_AAAOJL silvalugo_j_Page_089.QC.jpg
abc1201a0878fa306f49a2eeecdf1826
8a3656f96d77fa438d73ec7622f612095fbcb237
2215 F20101204_AAAMCO silvalugo_j_Page_135thm.jpg
9f35878bc8964d6513b7613d2c96728f
ca29f8ed542d4d85c1d6e466b57789cf9fb84529
F20101204_AAAMPZ silvalugo_j_Page_094.jp2
2881b742aa12d263e39dfc351326743a
15c03325cdebcd7e4ba861fbaebb40636fc55703
66910 F20101204_AAANMT silvalugo_j_Page_242.pro
520df6cca289968de2c98909d82f0429
ff214f14996c97632e1e76f0055d26f0d6753fa6
35903 F20101204_AAAOJM silvalugo_j_Page_090.QC.jpg
82b6214fbb3b8d92d9859bca23eb5638
3b039758fcb3bcd5eda6c32ae768962ead192579
23459 F20101204_AAAMCP silvalugo_j_Page_047.pro
ca90f66b210f1ddd5aef3dc8cea0f076
9ee849a1ac8f7dc38f544a52f14e126d3590670e
64663 F20101204_AAANMU silvalugo_j_Page_243.pro
6cd360929d6e3d72e6e352dc7c85d9f7
87d36a875ca5d81d5f3cee31d86c27c39e400034
8904 F20101204_AAAOJN silvalugo_j_Page_090thm.jpg
11e4c5e3c9c62c54c663590eafe9378b
f7d1345ea8947e7fd20bc6b2c5707b77a900f084
56412 F20101204_AAAMCQ silvalugo_j_Page_088.pro
3450da8c73520a6f03a9801c79485c0f
b0e3cca8db4925ef43a1fb31b99c6a6f63918e90
64221 F20101204_AAANMV silvalugo_j_Page_244.pro
117ae3c8ce70e704e942599f8d972eef
d7bf7fcfda989ac84f9d970b9d035f05f331a998
38081 F20101204_AAAOJO silvalugo_j_Page_091.QC.jpg
f41169ce01c2c3e93f176fb9c85f8f31
f4b7a11474066fec72a0022908547820047552ee
8572 F20101204_AAAMCR silvalugo_j_Page_089thm.jpg
7217930e9a3a323d219564231f4f86a4
d85a30ff2740e9bb6d5ebfb74350ee8899020ce3
64140 F20101204_AAANMW silvalugo_j_Page_245.pro
dc7a1b46efb5e5e4c646f16150c96c91
fe69ad487ed151f11b500e7ab07bcf21359d3252
34679 F20101204_AAAOJP silvalugo_j_Page_092.QC.jpg
987dc9421f6bed5cb916ffd7797f5f4e
3f5eb5ba83eb7e28d79a9514d72ca1f91fbb6a72
24275 F20101204_AAAMCS silvalugo_j_Page_132.jpg
e1d5b66c1d5fb4267d96f1e7ff4a1129
177e409894bdcf415c99932952a59309e65e62da
68438 F20101204_AAANMX silvalugo_j_Page_246.pro
91190f67c6e84aa77eb8256d2c2e705b
f99a5f5e57109a432e3e9c639399b2249bbcd600
8684 F20101204_AAAOJQ silvalugo_j_Page_092thm.jpg
e59f3d54e53442eb21ccfc0db525616b
009c3116dba238654dffd783886b11c049cea764
9238 F20101204_AAAMCT silvalugo_j_Page_194thm.jpg
b6431c205f053a788c8ecbad6ed9b0ad
ecc5a7b3243f6475aec400b49ddfe93dd8513ca0
16358 F20101204_AAANMY silvalugo_j_Page_247.pro
c69a32289056278987fd28b633928bc0
174ebcc9e9342d2178dc399330d7c876e8f25b4f
8738 F20101204_AAAOJR silvalugo_j_Page_093thm.jpg
ad9c069832c9b1572ba11da45e2c13bc
59b63b60725875da1bf8db949f101c10a18f1afe
100928 F20101204_AAAMCU silvalugo_j_Page_204.jpg
a18745caedb07cbeae80f4287047be58
dd7e38eb298b586f1d451b98f07a3a342585da3c
47923 F20101204_AAANMZ silvalugo_j_Page_248.pro
4e659a7b6c98025894376a6889f3f506
d12b29da1047a1a1023cc3f5174ddedf9a375c2b
36734 F20101204_AAAOJS silvalugo_j_Page_094.QC.jpg
2f9bd10a768a506700d34c66dd562350
307723ff170ababd6ecf10ccb3594ed99c04e493
58428 F20101204_AAAMCV silvalugo_j_Page_119.jpg
390675982c5985d05d546b67c7d9f108
408d3ce8ed077e1167fa55b8aa8481ad0102f1a4
35550 F20101204_AAAOJT silvalugo_j_Page_097.QC.jpg
303b0b993783353c6200b03bc136e28c
4b7374c97c998b66eb350b9bcbdf23d23720b620
2052 F20101204_AAAMCW silvalugo_j_Page_092.txt
8fb1c3448da01b91a9ec57accc3e0dee
abbb37b0e2c5ed79e2d90650f11f3d83a894ce38
8918 F20101204_AAAOJU silvalugo_j_Page_097thm.jpg
951675e75bd0278996d091adcd93b43d
10c605e6182e2a53b802223d7c1add44d62df803
21268 F20101204_AAAMCX silvalugo_j_Page_142.QC.jpg
fcb7f666038a0899149973715a4d740f
65e205c13071b30fa596503d0c75262e0a1492ac
F20101204_AAAMVA silvalugo_j_Page_234.jp2
24015eb9902f596869f0b0a375c43fcc
031e405bdc53b11e5fb46989677519e45f721611
36593 F20101204_AAAOJV silvalugo_j_Page_099.QC.jpg
1d8f909d0f35295d69ea017e5f3f500a
356083d65ab99352c80e41fd9b5241c6c3a24094
71967 F20101204_AAAMCY silvalugo_j_Page_048.jpg
7009af31f33e9706682689811244255b
7beac0a7574fccd7afcc496dc03d1426d5c5f5a3
F20101204_AAAMVB silvalugo_j_Page_235.jp2
7a330ff8dbaf1ec48788c8729ccd51e5
dcab1adaa224166adc7d0c059ab639324c766810
36021 F20101204_AAAOJW silvalugo_j_Page_100.QC.jpg
eb337d0d0d613264fc44af3835daed9d
3bb2b5ae33b7a7ab5bf5d9ad69af593e8bdd4286
15898 F20101204_AAAMCZ silvalugo_j_Page_054.QC.jpg
01b63ca313d160e28f97249e49110ddd
fbf681c911b71ae16d92a03ba6337e86ac283915
F20101204_AAAMVC silvalugo_j_Page_237.jp2
5f90772aea201ea3e1c593dd00502e1a
278749d91a06c4f707c3baa8b4d207f33ad07b94
9105 F20101204_AAAOJX silvalugo_j_Page_100thm.jpg
54983a753b5fbe33bfaafb3f09ab7535
969f199ad723e3fd9c5b7d5d56aaedcf1c6f076f
F20101204_AAAMVD silvalugo_j_Page_238.jp2
1cb1c3f43f0153fb8b6a92ed36d3f51c
7a0052622b878cf7f734fd4ec13104b198e77df9
36055 F20101204_AAAOJY silvalugo_j_Page_101.QC.jpg
a849c585d949e69a3489ef8f5c83798e
3d369aaeb0d3112d5c34a0561a74b128abb678f8
F20101204_AAAMVE silvalugo_j_Page_239.jp2
8fe9785c3ea6b57afa3e5b557df97df3
6e8dcc659b0525aa0c33aa3a4666cbb5e983b749
38384 F20101204_AAAOJZ silvalugo_j_Page_102.QC.jpg
a03fb4b9458fd15d31581a4e77f93007
e1acf7961077679e7ad1d28576a05b8f0f07c43a
1051925 F20101204_AAAMVF silvalugo_j_Page_241.jp2
40edfa587802ddb7300832bb1e598b73
9cf918c97abaa7d52e52f7adcb49a0dbd8af9435
F20101204_AAAMVG silvalugo_j_Page_242.jp2
af633ced9ee6806655ef45a0cfc393c4
a2e351a5dee276e6b80cb5f98ca00d6550d8a9d7
F20101204_AAAMVH silvalugo_j_Page_243.jp2
d40f0a91bbdfb59a474abfb6acd913ef
dd44b4f75a22d62edba780a01bd3ed023bfd5db0
1432 F20101204_AAANSA silvalugo_j_Page_143.txt
894396eff33050c0298d196e56a81c67
c8772297c0369e4d3c1475b85b749e80fe4f6cdb
F20101204_AAAMVI silvalugo_j_Page_244.jp2
a81049544be2c43111396d3cfb2d514e
a039ff1892561250d18df3a859bf759f5569dcfe
1770 F20101204_AAANSB silvalugo_j_Page_144.txt
68857404249324fd30570a7b72d70716
349d01aa9c3fcba39fa3078791f9986917b58169
F20101204_AAALYQ silvalugo_j_Page_003.pro
73b9d9137db67506d074a3b055100b69
ecae5658431f4b94608ea66f2bcc206ca3b46a4a
F20101204_AAAMVJ silvalugo_j_Page_245.jp2
e1eca83de6604ba8e1c5fc8c89cbba96
6d5567cb4ecfef3c3c24e804a09267b28c8eb6d3
651 F20101204_AAANSC silvalugo_j_Page_146.txt
9ae0053a067a62e758eb077c364a2a90
584c24be0bd464fb34cbb6837a3f866ec74c0830
4431 F20101204_AAALYR silvalugo_j_Page_213thm.jpg
cce15509c0660a01f762534a319ceccc
748eeb56832805b304dd93f4e23a4c290e3ecf43
F20101204_AAAMVK silvalugo_j_Page_246.jp2
f89323a8a2886440ffba2baead92b34a
4e1fc1e3e08e3f2684a2e8fd3798dfb2a28f2269
2152 F20101204_AAANSD silvalugo_j_Page_147.txt
ff1bcd0ed9a55617979f180cab9f2546
1595a932669d0fbc7aae44db0300c1ab8df93784
48671 F20101204_AAAMIA silvalugo_j_Page_120.jpg
49d7894be867ecc74d94501956d43eb3
e1491f976fe476860e2d2a1c3fbf14c84764b868
2666 F20101204_AAALYS silvalugo_j_Page_238.txt
45181288e69637483a7f1b435ed1cb05
38f39f056759807390f60698be1b08af0a9e8d65
376989 F20101204_AAAMVL silvalugo_j_Page_247.jp2
8e783135bb03da2b07e8ef9e96e058b2
ee5925994b0d7f1ce2a8dfbfee5a97a2c6a5c9ed
658 F20101204_AAANSE silvalugo_j_Page_148.txt
8615b2899293bb2a8f4de1d39de0cc55
6a960a2f2eaac0a96ffff0bee99424702a1e7625
60795 F20101204_AAAMIB silvalugo_j_Page_121.jpg
fde774cbbe17ed243d3d43dde6183950
0b98924f198ff93ddc04b2428c956a950d31c055
F20101204_AAALYT silvalugo_j_Page_232.tif
1957076a790a0ba43e490d68827708db
a660449829167f691d9a68a2eb3b203cbf481977
F20101204_AAAMVM silvalugo_j_Page_248.jp2
dc77de9e2851b5ee47eca0d52a24f6f4
9d1d7fefcf7d4432c7dd2ca19c003a40cd9d0186
603 F20101204_AAANSF silvalugo_j_Page_149.txt
c1ff9a9100106aa78c55059b9c68836a
e9f2b17ccb76a67caa7c65656ffe5750616bdaa9
53398 F20101204_AAAMIC silvalugo_j_Page_122.jpg
d6e063e78f56ab8914645e56ba838fde
643ff044ad4601b95704d188ffb48e19c6fd96f8
53401 F20101204_AAALYU silvalugo_j_Page_019.pro
befba64d7064b336aa2bb26253d106cd
2c0b3a8c840e66553e7f5826ac9229ad39c511d1
141283 F20101204_AAAMVN silvalugo_j_Page_249.jp2
125d21e69faf685f3f660082871d6e43
3a3ae9caf50ad0c5c58bdeb2db4083e8c2f234ec
665 F20101204_AAANSG silvalugo_j_Page_150.txt
e942b6e097332f4ec17eb94c38adc2e3
9a4d9cb019fe1795df3e8babe54624e47501e658
143836 F20101204_AAAMID silvalugo_j_Page_123.jpg
b318e3fb5b925b9418ad4ea0df958eb3
beb830be7aaf0a394234be01f2883baf0b1e89a1
6548 F20101204_AAALYV silvalugo_j_Page_137.pro
88aa9a2bef0bb8be288cec768c938da7
8ed7af231807ba6a9c96e2576e492ddb492c3e5d
F20101204_AAAMVO silvalugo_j_Page_001.tif
29314953920444a0889c9ae55a6708ce
b38b3e322ef160de48cbf125a3421a2b1ee2d5ca
1027 F20101204_AAANSH silvalugo_j_Page_151.txt
42ccceb59075c3dea11ed755630656c2
1d2b58b6a83762f8d6dabb5b3ff4567343243450
129093 F20101204_AAAMIE silvalugo_j_Page_124.jpg
2993a6d96074528b383aba8cd359672b
a014356d831e8bc465a5b9e412e729853e750784
F20101204_AAAMVP silvalugo_j_Page_002.tif
b6c90c28a150ff23c1f750d4443f4272
65eebc9c5d13741e0a5ec47240de263166db180d
639 F20101204_AAANSI silvalugo_j_Page_152.txt
c74fc99f066178a7b3bca6884124bff3
f4a9d6cab8f8a49b84826c1dd666340b8856c917
881907 F20101204_AAALYW silvalugo_j_Page_143.jp2
bb422d841d6572cd9810ea5f37a7429c
89d774b754f18ea18a26360d75c6b2ebe81506c4
F20101204_AAAMVQ silvalugo_j_Page_003.tif
8bcd480d245f150f11433a81b567dac4
987fe36e480061ec206082d811e75a1a4f69b1da
646 F20101204_AAANSJ silvalugo_j_Page_153.txt
e391fe95d983a44c07b5682bd5d4067d
5adb03d7ecba69aca9def257afc53546d2253886
135655 F20101204_AAAMIF silvalugo_j_Page_125.jpg
5ad55973cfd01ed2d32b897007490068
f1201896fc2d968e9d76c8ff4b7a4991f8294e05
88419 F20101204_AAALYX silvalugo_j_Page_008.pro
5a5d5e8c0e827294e680ea85a01850cd
63a3970e18bcf587835ca8a1f65a701695cbba75
F20101204_AAAMVR silvalugo_j_Page_004.tif
4cef049f4b4ce630af5449b29f0e72a3
603e659d18f5c2c2bf25ef492c40ceaa64dcd5ad
1091 F20101204_AAANSK silvalugo_j_Page_154.txt
4cd6018db9aee57290d3d32e7d358183
c5f562f2f2a9b6786920cf2722409762abe40525
134728 F20101204_AAAMIG silvalugo_j_Page_126.jpg
74dd317490d5f5b873ee2f2c2a48bfd0
1fa1b37c0da20a58d5ec2228d99885c34e550cc3
9090 F20101204_AAALYY silvalugo_j_Page_164thm.jpg
2fa5458932c873399800a87669ffbb18
a7452472f703a3d3ba01ebb53305be48f7772391
13663 F20101204_AAANFA silvalugo_j_Page_026.pro
462ce5dd6b49e0093428154f5e46b3e6
82845bef5c6ee5bf6ffa149cdfdf8a71c423529d
997 F20101204_AAANSL silvalugo_j_Page_155.txt
d978c84154df140fb34b0b7b49995a21
d2dbf125dfdfbfc2d72fb269696eaddca23b02e3
142165 F20101204_AAAMIH silvalugo_j_Page_127.jpg
c5fa3eae25912401317b9e49a04966cc
2c27d4753c890c67b975f61f5c3746c6e1393cbe
33378 F20101204_AAALYZ silvalugo_j_Page_153.jpg
b3606f1e6718816863f8b014e56e97e0
7e2db7d440599ac8b4b2a6c50f20e6be739113fe
F20101204_AAAMVS silvalugo_j_Page_005.tif
8189646d4f0eeca530ecc7d521daeca5
baff9d4828bc14074d0d5c32d4c39f9f42e0b2a7
636 F20101204_AAANSM silvalugo_j_Page_156.txt
85b59e2338b54bc4eedd6fd56c1b6ba6
2a45e8b5e1dfe22e6a7bc3f58916cb102bc2f3fc
52204 F20101204_AAAMII silvalugo_j_Page_128.jpg
e03c67c478e59684f7a71f6cf02bc180
7f0a3598134ca5728fc43869556bea9810b2e3f0
15293 F20101204_AAANFB silvalugo_j_Page_027.pro
73b6ca72f107b54ba1766188a2b87d6e
c70fdc8c3be55fc2a9fedbb08b233df0c78347af
F20101204_AAAMVT silvalugo_j_Page_006.tif
35a0b39ddc36057417cdf91e2a4cc5c5
9b70993b2bbb4381ec3d2a9c2d7a6b672459c75f
2133 F20101204_AAANSN silvalugo_j_Page_157.txt
100791bb1aeb5417a23a17f364e83bad
3f48c2d057140f2bca436b20e81fd9adc1f018be
34547 F20101204_AAAMIJ silvalugo_j_Page_129.jpg
00f8050a47d2209bc582530791f9ee9e
a2fc9a05f710473215636a8ba27cd94253883335
5307 F20101204_AAANFC silvalugo_j_Page_028.pro
ab298991ce2d22b7ef90a9489234e1f0
d9135c515f34808c063035daa43a72645057501a
F20101204_AAAMVU silvalugo_j_Page_007.tif
769e70eb80b41841ba5425ea959f9329
7109abf0ed38de36042acc90dd933e6564e12134
39127 F20101204_AAAMIK silvalugo_j_Page_131.jpg
fddb2c90e2a16c28d9db790d45a61bd4
0c4e4d1e2e3c41d504777d4534e8655766e1982c
5235 F20101204_AAANFD silvalugo_j_Page_029.pro
5d6dc4cba53bdf8e582b36c99705b674
a4280fda0df1520b18ab12f5a9d4cc6cd9978d20
F20101204_AAAMVV silvalugo_j_Page_008.tif
9bbdb63db1c7d2e0ad8c910227d62c1c
bbbf6f0a89577016cbe26eebab15ed9a54e63761
2188 F20101204_AAANSO silvalugo_j_Page_158.txt
58d9106ee81bbaeb95087f6dec9803ce
4bfe04a4fb6dfdf8cfaf92761e02b20b01297745
30741 F20101204_AAAMIL silvalugo_j_Page_133.jpg
cd3520a4b1a38f6f22bceb1e0c1ef6f3
20df14f2419054403b85c6e43a5d4ef33f663164
3874 F20101204_AAANFE silvalugo_j_Page_030.pro
df160eb3f7282a67815fba34f0ba35b4
503e77f9a8c6a7564173fc7dcd7fcaafc17ad719
F20101204_AAAMVW silvalugo_j_Page_009.tif
1814c6df16b3eb7e7a2cfd7e9b6415dd
dbcbf0845be1a7258d0b38ef7735fdd06c1dca45
2214 F20101204_AAANSP silvalugo_j_Page_159.txt
fcb5fc3868eab4e4d89a4566d154d39d
2cd038f82b2f6c2900a4bb0b80f9bbb1b6fa5278
25838 F20101204_AAAMIM silvalugo_j_Page_134.jpg
7d8734bfb0f50af9ce6007cd731689f7
3fa15fef8643b3082642f3af3ae1a5bbeb413296
5584 F20101204_AAANFF silvalugo_j_Page_031.pro
a639e886dc7a2d9af2715d2058f27f54
6ab6d4e44ab75e83a79ef113b42ce5ff6fda1e38
F20101204_AAAMVX silvalugo_j_Page_010.tif
1c99f41eabcaf6e9bbe630037e592389
d3c4f87448bee6db723bad621bfa240f1536c187
2167 F20101204_AAANSQ silvalugo_j_Page_160.txt
31379c94cbcae748c9e67c58713272f0
1e8e323d717e50910a07b50944d398504b228129
20404 F20101204_AAAMIN silvalugo_j_Page_135.jpg
6e4d3629693c8950b5e71833154cabe0
5da93c5edebbbdc9b4a8384056a86bce0256c728
F20101204_AAANFG silvalugo_j_Page_032.pro
d82872363f3d7f3e86ed744717c1df22
716788bfb7ba8e34628c7bbafbe206469f524648
F20101204_AAAMVY silvalugo_j_Page_011.tif
37c8474b046dde551d79ad61603854db
f3991fa218034ea39dd2029ca86b7c6826f957ac
2220 F20101204_AAANSR silvalugo_j_Page_161.txt
a6ac49166701a8fc396c8b93e5fae785
ff1e8199ca007a510be804d4415d4a255c942be8
46011 F20101204_AAAMIO silvalugo_j_Page_136.jpg
9ce65220e1db80d16d986eed574bea6f
42633958f672bfedd5e61926686d25f5683e6e6d
F20101204_AAANFH silvalugo_j_Page_033.pro
31efe87ac56742f41fd7a04e29e45ec6
b7cde5fdc2720e07e7957edfa5a943c4473fea86
F20101204_AAAMVZ silvalugo_j_Page_012.tif
ace7e00da5b23e8446e05af6287a829d
4d695882c848a02e29bf58d51b915db49e26cc02
9463 F20101204_AAAOCA silvalugo_j_Page_236thm.jpg
2d0e66fa12cb0f0b008b8c3d72f22c27
48cbe6769de4cf6ecc720e4532442e913c6724b2
2195 F20101204_AAANSS silvalugo_j_Page_162.txt
248ac8ade113df974e647dbf48b85a23
600263639a61c5c1b2a11e98fa5464e312e60459
28160 F20101204_AAAMIP silvalugo_j_Page_137.jpg
9a2f347ba7d6ec310e2d77e29fe53d65
9bf5adccb78827ad2c15c2def41d2b9adfe13caa
F20101204_AAANFI silvalugo_j_Page_034.pro
118fdda04955f4173031f7ba0d1be0ee
0bb9092725a04c66acffb68b326770b1b6831055
F20101204_AAAOCB silvalugo_j_Page_181thm.jpg
3fd0fc19ac1ba39ce05efe9dfad9db87
8c2097b9104410296e10a58c83fe19e55eb60f01
F20101204_AAANST silvalugo_j_Page_163.txt
ba50ebaca1c9321b2bcfdcf20197ee85
e53f220aabd16e0ca148f5b86e246235e0fd07ac
54816 F20101204_AAAMIQ silvalugo_j_Page_138.jpg
f90988875847f2af489e652d210b4c88
dafc24dec31edcfb3f8283dd12a127d5430f8a2c
4574 F20101204_AAANFJ silvalugo_j_Page_035.pro
3ecb56c345a45e435e968aa7ad949012
10df2e9c46b9127d43346ae768335bae7de50fa9
34245 F20101204_AAAOCC silvalugo_j_Page_106.QC.jpg
97ffc283515c9c11dab5cd9db98a831b
0fdb79cff19eefe3240ac6c80a0e770e8012e096
2178 F20101204_AAANSU silvalugo_j_Page_164.txt
ee8d63796bfb742e535f6daca65549ea
cb6a38d90aa1758b8258e41e94a8526781a08d72
57145 F20101204_AAAMIR silvalugo_j_Page_139.jpg
569909e069e22af830181fbce7a6bb24
fc8b16283c84f86176bd2a73d35447c7c03d6636
4668 F20101204_AAANFK silvalugo_j_Page_036.pro
d685d7b33798633770c0babb5ea130a1
60c214c303868305a422455b557d19a966adca7c
8551 F20101204_AAAOCD silvalugo_j_Page_174thm.jpg
cef7a4ae2e84277b6d9947722964ffd0
9c856a4e25cd28bd4cf5350b43b1d0518a5d176c
F20101204_AAANSV silvalugo_j_Page_166.txt
4189341dcf739a4a0223f7a6be4950bc
102664f69c7fa02c9a63d0268c35437550d7d0d5
47548 F20101204_AAAMIS silvalugo_j_Page_140.jpg
781a91a35feb24760def288f52328d7a
20c3144929417320ac6c8ed90243b1e175724ce4
42414 F20101204_AAANFL silvalugo_j_Page_037.pro
d5b6b12e91e1b1bc8dd2db731b3e8e35
5c4f4b3fad3a3bba36633465f2a6383e4cafceb8
30837 F20101204_AAAOCE silvalugo_j_Page_116.QC.jpg
cc73df4495873bb0d8382f9bd5f60b0c
947eea9a201ffb7142964147cd889a00de6debb8
2196 F20101204_AAANSW silvalugo_j_Page_167.txt
ee80b06de15a5d5a887ca0845b910135
efaf7067ac4e5e92572307cfcd527fe92e624099
65192 F20101204_AAAMIT silvalugo_j_Page_142.jpg
ee202d47ffd1f7a845cad114e7999726
bce3884e738f143761df1fe9adfc1970d08c9c73
50374 F20101204_AAANFM silvalugo_j_Page_038.pro
7192c30c4ff27f617e3866d917d9652e
ad3dff9d9be2a91f3467cb576d6809b9e5b9991b
36775 F20101204_AAAOCF silvalugo_j_Page_163.QC.jpg
49e9c054e67ad843379cd0f324f339f0
7b603f0f436deb6bbc94979d5273f9c9dd92e954
2025 F20101204_AAANSX silvalugo_j_Page_168.txt
409a1dab6c516d5102f20ef3cd123048
d84c13222455796f9694cbe69dc279a519966e15
61871 F20101204_AAAMIU silvalugo_j_Page_143.jpg
d1c8ae8717b8f949d92f7748fa986cd1
cb113ffa8f350a80c444812ef3b56b16f843884c
45100 F20101204_AAANFN silvalugo_j_Page_039.pro
080c2f4fb15aa735877ebbb50234f1da
1a91bc7e44e7b40dc1cddf1faaca57ac51975f6a
3310 F20101204_AAAOCG silvalugo_j_Page_148thm.jpg
426103ce80e8c5105fba32e59a5a187b
ca147af14fdee6c64bdeedeca82f47a64f34de76
F20101204_AAANSY silvalugo_j_Page_169.txt
dcae0cae16ae179dadbb83485467c6a5
bc1706d04b6a5e3401ec643a1f73e985b6eb0dfb
58915 F20101204_AAAMIV silvalugo_j_Page_144.jpg
75c61ea379cf2d01ceaa13e65d61fa02
ac4bced937a2b68c31850960af5c9b1975dc87eb
54364 F20101204_AAANFO silvalugo_j_Page_040.pro
d3069f97a3f554afa75a15471631a78b
18cd97f92aa7a9e33b46bd4cf50691057b4f295c
9306 F20101204_AAAOCH silvalugo_j_Page_159thm.jpg
2552ef990fffb3de92e2fb114af68447
bd7317132317e63d093140edce79dc465906b467
2021 F20101204_AAANSZ silvalugo_j_Page_170.txt
89a83b6b39e032c37c131f98890223f3
12b76c5ba67c1493b36e4a982e6b8a72d8743eab
70614 F20101204_AAAMIW silvalugo_j_Page_145.jpg
a646ed239ecdeff6932aebdf7406050d
20bfe2f92fb4cd2c573448ca4e388c77d2bd7ecf
55033 F20101204_AAANFP silvalugo_j_Page_041.pro
d5315fe212fe6d3d181ac877a3e27709
5b2222cfe00656e05993bd64001edae5b15cd89e
9563 F20101204_AAAOCI silvalugo_j_Page_061thm.jpg
e6a9082837bc4f0efc382fdc89758468
5d53c2bc2fe3406b69649283684a8a0d7e09d2ff
46638 F20101204_AAAMIX silvalugo_j_Page_146.jpg
b84d0447e429f7265e490d18a4b98a6b
66b72aa0e079d330c012400d0e76cdb58b7f43ed
46552 F20101204_AAANFQ silvalugo_j_Page_042.pro
d8f9bce29599d6eabfa264c7761a1595
da4665aaab3528c8ff9c46f7584a2ba858cf417c
8564 F20101204_AAAOCJ silvalugo_j_Page_183thm.jpg
1c89fb5cdbe4ca11bdbb3b706bbeb2d8
bebd1a111499845671b39feb7cb8e0e406368cad
F20101204_AAAMIY silvalugo_j_Page_147.jpg
8657b507b7ee757e7dc3e96f8acf36e6
1f884f06c4762f7a48d6e716027c2895cde0b7ae
55048 F20101204_AAANFR silvalugo_j_Page_043.pro
7b805853453435b483ea62bd645b8736
b52f6a34832838c662ee2117747b185875e40b71
14939 F20101204_AAAOCK silvalugo_j_Page_049.QC.jpg
71a4d7ce175ec1ff212e82d7281feb9c
7f2ebc4edbb33515eb65ff6ee41f3b19b05b2f65
27357 F20101204_AAAMIZ silvalugo_j_Page_148.jpg
42dc17bb96d65998759360d6309c4ccc
8b9f61564fae3e3f8d02363ed6182b4a0fb867d3
52578 F20101204_AAANFS silvalugo_j_Page_044.pro
b0b86e146019e81434cb8b69ba98076b
79f9e4dd5829d9046f4a42c0ee7d7e708bbc237a
4227 F20101204_AAAOCL silvalugo_j_Page_208thm.jpg
e280155de962be2db4cececb534045c0
91f9c75c29469f2bc2bd7343a3529aa12220dcce
50210 F20101204_AAANFT silvalugo_j_Page_045.pro
a6ec97bbaea72607a8b145f555040515
5200facd4defebcbf04a06585976e5470e625b92
43219 F20101204_AAAOCM silvalugo_j_Page_014.QC.jpg
a14e9b10662aa3313030fffbb74928bf
5a1b84b000381dfa0fb407c33477340168cae869
4828 F20101204_AAANFU silvalugo_j_Page_046.pro
918463583afda94d34bd471a2529e629
29d263b886913a7e04e6a9c8555bb7d51029b8a7
8529 F20101204_AAAOCN silvalugo_j_Page_157thm.jpg
8e5d02419793bf86f363205e9ac6c39c
1728d23eb05976dda1d3e35dec6f881cde80c7ba
40878 F20101204_AAANFV silvalugo_j_Page_048.pro
34373178fa857fc6fea736feddd0b503
6895965e9ed4a12f86b7640fe41dd83ba5a8d01e
36151 F20101204_AAAOCO silvalugo_j_Page_082.QC.jpg
2a809a2b3916c43e126b8a4d1233783c
586eb79f418d33a583e486c899b2b769ff398236
18522 F20101204_AAANFW silvalugo_j_Page_049.pro
ebc251a62d36a837ce07d46bb6b88ad9
83f57e0f0be02ce875123efdc5feae684a022537
9065 F20101204_AAAOCP silvalugo_j_Page_162thm.jpg
01dc6c5aaa2bbe74a04db14b74717e7a
7b12f395106ce8c79323b1809567c092f8c38ec1
8888 F20101204_AAANFX silvalugo_j_Page_050.pro
0ca51ea2cdb0e33365efce282fd5bed9
5491550bb4d1428b34d2abe3971881f78f1f77cb
31268 F20101204_AAANYA silvalugo_j_Page_173.QC.jpg
4357092795c7b6a3806a467dea22960d
c0b7abf0a363a78840af66775f087f824ca87020
11630 F20101204_AAAOCQ silvalugo_j_Page_154.QC.jpg
3b28712aa3f0a33b3d5e5bd40206dd04
21ea7a20f3d1e4ee52c9cec4fa0a361fd3462a94
5702 F20101204_AAANFY silvalugo_j_Page_051.pro
9b2df914fad2350442d0eb4929b089fb
b15f56463eb9343f051112c5a0de39b29ac01b05
8596 F20101204_AAANYB silvalugo_j_Page_114thm.jpg
d77073d370013c86a055bfe18726209f
c05152bb21634c037f30e84cba39a688d5ca5c2c
34779 F20101204_AAAOCR silvalugo_j_Page_108.QC.jpg
bfcb812be1324b4c17d53be2ec59d5be
88eb4f61447c20113d2f5901734d0e5a9c252546
3820 F20101204_AAANFZ silvalugo_j_Page_052.pro
ff9c0d068d77b498298f7bc7b79429b5
d75a842e132ea1383c098077c4d8a3fc8ac21c51
23969 F20101204_AAANYC silvalugo_j_Page_210.QC.jpg
6a1a264abc564d1fa768601fffbe84f0
884000ad493442fb9133eef116b7151f3c9cb3f6
6260 F20101204_AAAOCS silvalugo_j_Page_048thm.jpg
b849d2b06ad67202d01e1d84686ea8a7
4ca9b3ae71d9d19922c02c92c38da97d965f28ed
38827 F20101204_AAANYD silvalugo_j_Page_033.QC.jpg
367d5b814c60c6679fbb13c726893b74
017cd1f206b0da879f1137978d36a7be3a53dd2c
8843 F20101204_AAAOCT silvalugo_j_Page_115thm.jpg
d3b47549d7b179b3ca43e5b14c545158
b6101e8653bc5821905e9e5318c793e13cd44152
8708 F20101204_AAANYE silvalugo_j_Page_018thm.jpg
b8affb49181bd4357ff57c19ebe97c40
2cad0f73143dde820c37a985b5d910829181f724
5533 F20101204_AAAOCU silvalugo_j_Page_103.QC.jpg
a718f9b3c937790f1ea8e2fbc6e2a5cd
0b2e7120055f8934426a10cc009aa2dd4f9bbf5d
F20101204_AAAMOA silvalugo_j_Page_043.jp2
df3d8d3bc0c82408d17428105e30dc8f
1dd9564bd0036a381f66735f85526c9b92421351
8880 F20101204_AAANYF silvalugo_j_Page_160thm.jpg
10f997d4854fc1af838f2fe3db229875
86fe8257feb6667cb2c09ef42c7b40761a74abfc
9321 F20101204_AAAOCV silvalugo_j_Page_198thm.jpg
4264ecf04643c143d9618cf48c678049
7f23e493e1df932dbc1fe622dd52f802f20614d3
1051964 F20101204_AAAMOB silvalugo_j_Page_044.jp2
9ba8401e31672497c98d16e110b57691
ae6f629928899565460219e1c6b52eb11ac92561
8523 F20101204_AAANYG silvalugo_j_Page_045thm.jpg
23161e8307599cb2a9a1c1aaee4181fd
38d97c24fd6cc25d5a93a9dd8dccce7808b9efc6
8752 F20101204_AAAOCW silvalugo_j_Page_020thm.jpg
f892125b630e2fc8dd8c98ec3f87ff19
2f23d5a0f2080cf6b8bfc8b38de07336350bdfad
F20101204_AAAMOC silvalugo_j_Page_045.jp2
b1d166182c0f28456ca8cb87e21cbd70
ced22869fc2ab415990427797ad1a5b724ca5831
9311 F20101204_AAANYH silvalugo_j_Page_032thm.jpg
be5f0ba1ada8137ff945a1fbcbf7437d
8fc33aed11b71205cc02882fe70f7b8a13ff487d
8049 F20101204_AAAOCX silvalugo_j_Page_025thm.jpg
bbd5ca471882b8d6a7c18c2df797d382
c679ad3c8624ef89638dceffe8a7056f45ccb985
114310 F20101204_AAAMOD silvalugo_j_Page_046.jp2
a72ba1d58ae65608537d5e98724f06f1
37635db559b3a647d169b800d49d3a2f37474818
5629 F20101204_AAANYI silvalugo_j_Page_139thm.jpg
d41684a69a99f3b1ca0617335578a992
b106c1637c2656e41d8543ba8dd9fe19c461ccba
9219 F20101204_AAAOCY silvalugo_j_Page_066thm.jpg
fc38439428db97a5e7feb0764a3d8121
301ae93e9841f373049f8f3a9afbed6153fa61d2
536997 F20101204_AAAMOE silvalugo_j_Page_047.jp2
86e132ce0ca0dbf39d2194d314972c46
ca4f002391145b5966344caa5b8947edd2331b0d
35312 F20101204_AAANYJ silvalugo_j_Page_020.QC.jpg
3d3b39d5ce44045dbd283c175d5d9f4d
3cf797315b9ae904ef8e2d18822fb47e1fcacf03
9128 F20101204_AAAOCZ silvalugo_j_Page_180thm.jpg
15aa428366f72836c50249dbe5ecdf14
921deaf99fc264bf5959ba57ab982c933bfd91dd
702525 F20101204_AAAMOF silvalugo_j_Page_048.jp2
943c359c0ad61760acef07234aa323a3
5c1c28b92884aa8e79ec7bd11aca76b8da008f4b
7829 F20101204_AAANYK silvalugo_j_Page_171thm.jpg
ebbc61487f61ae70d6cd0b026d46fe3d
d9a5444052149220a3d0edd8cbcdb3c21f95d148
425585 F20101204_AAAMOG silvalugo_j_Page_049.jp2
99af7ec880aadc9efc4dad206c300e59
c16b3832ab8baa20fbbcba248ccc45ba7dbb260e
57401 F20101204_AAANLA silvalugo_j_Page_196.pro
0872b2310071d6fc395b7b1e081f6e51
b3999a7d9af6e562a4de7de7d5915d773151af24
9296 F20101204_AAANYL silvalugo_j_Page_043thm.jpg
8fd98e05ea6b2957091a1731991dbdee
acc75bc2430309aaa8715e65124676b16a07d234
1051814 F20101204_AAAMOH silvalugo_j_Page_050.jp2
8afdc99e6c71e56cca958683c8f39a2c
d88faa71487951d263ae44f8ce2e32075d5b592b
56201 F20101204_AAANLB silvalugo_j_Page_197.pro
74796484a66b845ab1b4dc649f77d6cc
4c7a58127a1dc0b4aa3136e9eac641c2f44105bd
9258 F20101204_AAANYM silvalugo_j_Page_199thm.jpg
9d1706622f7371ac2b7ef2dff72837dd
96e8d090a407d6bfbef830f5e4e6d13f93d25033
F20101204_AAAMOI silvalugo_j_Page_051.jp2
d272d67c2edf37b5611bd0a687d847d7
e9a834977f09f9ec95ce235e398eb6f59f71b195
57017 F20101204_AAANLC silvalugo_j_Page_198.pro
7249b11abaadf3b3c454fa705da25958
9bb5aca469aec79577f219c95e7c91f95ba6e5df
9038 F20101204_AAANYN silvalugo_j_Page_074thm.jpg
dc5d5c0f14e949909cd2a811bdf55e36
e49cf826e1b74371ac1b324370024208dd796b02
F20101204_AAAMOJ silvalugo_j_Page_052.jp2
85901032f6083d82211a2ffaeefa1824
be3570d5e38ed4e1fcfd4bee9a5025f5f77d848c
56679 F20101204_AAANLD silvalugo_j_Page_199.pro
667228969d9153abf3f3e7a333781572
3522070103f306e7c7f483acfb73fee215d78ef9
10936 F20101204_AAANYO silvalugo_j_Page_247.QC.jpg
2c3ca224c02119b8ce948e57d0c1aa77
2b9733c1c243e0870df1d6d3c67bd6f54042e6f5
1051956 F20101204_AAAMOK silvalugo_j_Page_053.jp2
f85aa400b899ce08b823552463cfbe03
ac7aad7d72a191b9c65e25508668fb2450fb49b0
54329 F20101204_AAANLE silvalugo_j_Page_200.pro
eebc3e48b0aad83a20b5469d1942901e
cac96c753f19eb2181f8c57380b09c4e37495bfe
8683 F20101204_AAANYP silvalugo_j_Page_189thm.jpg
299e766b924a3b50ce88fe91b6ab788f
0e7800e5857ede42c051235283f0036b6d1db6dd
43946 F20101204_AAANLF silvalugo_j_Page_201.pro
579cbba66d9980b6771013b33fcc41e4
2c4031de1040b610b51c691e30c819f05a5f8baa
F20101204_AAAMBA silvalugo_j_Page_017.jp2
e3f666d6f2fb3055b879a94d831a6c67
3e9e18faa56d15fc804dc7d97df3a68f7af521e6
33407 F20101204_AAANYQ silvalugo_j_Page_045.QC.jpg
4050b68fde0957c3adab676546936d8e
d6b90743dfcf9853b59ee1adfb1711c6e790b1c0
1051852 F20101204_AAAMOL silvalugo_j_Page_054.jp2
bbe154371b7c11c2ffc12f593ac781cd
a8576424421040663eef1822ab3c4f4372f625ab
13842 F20101204_AAANLG silvalugo_j_Page_202.pro
29a817cd481057645c36438766c4f3bf
2c42edf384459f9fa9a22989dad8f89a50b0b817
F20101204_AAAMBB silvalugo_j_Page_162.tif
3847bb7905e7ff2ad8310b0339e82e92
6835f09f18669eb436d0978feeeb4064a173db1e
F20101204_AAANYR silvalugo_j_Page_186thm.jpg
133dc4a451ce2598f9e18bd85f0dea1c
c115f9d835091da1091c72f634efc296f41d9e4a
F20101204_AAAMOM silvalugo_j_Page_055.jp2
b261687062c71e276139f228fa736df0
f13f3ff9cde6323385717c04c814854eaf932808
31313 F20101204_AAAOIA silvalugo_j_Page_057.QC.jpg
12275f38472758bdd76d20f46810f238
bda8b1bf7ba61f376e4d92b23a728da9acb35c12
2143 F20101204_AAAMBC silvalugo_j_Page_083.txt
91940c8cdae5a6b2cce2dbfd6eb84ecb
e5625353a98f279ac1338be1fcde844b4771faaf
8326 F20101204_AAANYS silvalugo_j_Page_176thm.jpg
657d431f5e0b53edb9c0243e27891ae4
2250585257b81bbc63adbe4947b82d080aca6108
1051858 F20101204_AAAMON silvalugo_j_Page_056.jp2
c0c828887405b5ac394f53410747ce64
1fe059088fe0d481d3b050cc4936161f0cafa4ea
33724 F20101204_AAANLH silvalugo_j_Page_203.pro
26581469f142e85e2f331fc43ec4f781
ec8ba3aac4e38e64512dfaf69c36035bb1785b62
8588 F20101204_AAAOIB silvalugo_j_Page_057thm.jpg
f8b6f4a082e41885ac6d21c9da115a7d
a0546e688300a3239ff4733b0109106de30fca4a
27784 F20101204_AAAMBD silvalugo_j_Page_143.pro
4dd29caac81e663b4da30085a9661352
44ce60c11da5d80ab6d1badf6e55657e69baafc7
32503 F20101204_AAANYT silvalugo_j_Page_016.QC.jpg
dbe3483f181d68e66318e0c0881e01f7
925637d50369e548e3c89653d57a3fee3ca35355
F20101204_AAAMOO silvalugo_j_Page_057.jp2
6f4498cf8bcb08f83b684e9529fc7bcd
a26bd3de2f1592e8033ad4311c912213fff3a9a2
29607 F20101204_AAANLI silvalugo_j_Page_205.pro
7934942f377790c156a63941e4a5fa1e
68787342d94fa40e61629f9f16b9c20f5264a008
37332 F20101204_AAAOIC silvalugo_j_Page_058.QC.jpg
788eddb991186a596748aeb683b0af93
6400b989e495cdd625d56760cd40cd3942e12284
131456 F20101204_AAAMBE silvalugo_j_Page_243.jpg
d8f64668356f71036cf4e31c83ebf1f9
ffd308ace494896bcc42229594f9c72a68301b6c
F20101204_AAAMOP silvalugo_j_Page_058.jp2
9586f56719481a485c5101b404f21773
04859f94e784f3dd8397e8e505ba5e840e29cd89
F20101204_AAANLJ silvalugo_j_Page_206.pro
3d66ad4d124ab09203145468675c9a06
272432c96b56ddbdab5cee1136502bf95371097d
52227 F20101204_AAAMBF silvalugo_j_Page_183.pro
ae03ccb76bee328e2c9b2d906b6414e7
66fcafc2a80a36024207b234055fd476183ecf6e
3542 F20101204_AAANYU silvalugo_j_Page_223thm.jpg
89b8798b99685261a669054afb482cc6
b255c07c15efadd6aad5aaf07d77d787d221d792
1051952 F20101204_AAAMOQ silvalugo_j_Page_059.jp2
62bc311925cb5ddbed792a5cf51474b4
214c39e9eb0aecbb347155e0ee9da46f8ad6852f
14889 F20101204_AAANLK silvalugo_j_Page_207.pro
a3ffde402a016c8ca2d4f119645f357d
f4abf68b5c1aafe67003b61c381923c6fc728be2
9287 F20101204_AAAOID silvalugo_j_Page_058thm.jpg
ba584378efd2669f83019bbacbfced89
63236780b27c147e88357e842812ed21d1cb50cb
114468 F20101204_AAAMBG silvalugo_j_Page_180.jpg
273fbf19f3e19850f5ed031f6c5b5b30
ed5bf4908e3319f4ec2ccd0d95e175599d803a8e
8417 F20101204_AAANYV silvalugo_j_Page_075thm.jpg
d33cb716f8bcae69cfebda7484012aeb
5634b3668c09b51fb2eab6958225ed09ea382fec
1051740 F20101204_AAAMOR silvalugo_j_Page_060.jp2
0748146dc049c038fef18e809b3db07d
0574e4f203c4c330e8c1bfef46d01f10afb77093
28610 F20101204_AAANLL silvalugo_j_Page_208.pro
dd864c02dd94482180fa5a4e4252ec2e
9d226ba9acab4bdd21029c10c8b0db97652815ac
36254 F20101204_AAAOIE silvalugo_j_Page_061.QC.jpg
edee65cdf01ee4760b1481b4064813ad
7b42adbe1bdc5707ceb271686eec932000a7b691
40724 F20101204_AAAMBH silvalugo_j_Page_007.QC.jpg
7dfcc9940023ab238e82a1c5a7208ad6
e03fe23a7cd4b4be3e249460d16d23c0378c5f8e
9216 F20101204_AAANYW silvalugo_j_Page_193thm.jpg
b11cb1cd9f2ebb30aba88892be4162ac
bbec0e800dc1c00963047fe64f082f7b3df43f0d
F20101204_AAAMOS silvalugo_j_Page_061.jp2
437ac5a18af2f803de9ee3a505c93594
d2e7f8855a9e3b6037e0d1ee8d3e2147b88e352e
33967 F20101204_AAANLM silvalugo_j_Page_209.pro
b3cf7f71209cd02926605ebda3b251bb
706a2eb3c850e8f885e09b4ff5cf9686b9b491cb
15793 F20101204_AAAOIF silvalugo_j_Page_062.QC.jpg
9ff802393529468aa642502944102fe7
327459840ff441c438c133b4afe2dc12bc75108b
12906 F20101204_AAAMBI silvalugo_j_Page_224.QC.jpg
d2875a75062612e5380d3aee740046b0
2ee5a62d48db0baa1d225ddc076afad6e0afeafc
12276 F20101204_AAANYX silvalugo_j_Page_152.QC.jpg
b80b6da83fd64ea62934546268217f3f
d0981bfc7031a295837b4b98308a407c95b38f5a
448481 F20101204_AAAMOT silvalugo_j_Page_062.jp2
6ef398f7936cbea334b2b13f12dcf076
1d26bb152cec4e40f0fc6a2831203e34549606c7
39385 F20101204_AAANLN silvalugo_j_Page_210.pro
df2eb2d444f2d2d8d21b929b1132bfc4
45efff7394fb5063a0ebed4d868da78f681caaf1
F20101204_AAAOIG silvalugo_j_Page_062thm.jpg
6cf2ad8baf94c82f58a0a35107867bdf
244e1be7763fb72a277b259ed195963b8993bf4a
F20101204_AAAMBJ silvalugo_j_Page_020.tif
c037aa00bd49c9b18071c26b9750834b
81a94b775cb63db12817e912b25a2bfecf52b7cf
3111 F20101204_AAANYY silvalugo_j_Page_131thm.jpg
9e8b76a85951444a7924957cb9a15905
9e05cf6ac31e631686780287c6c50546e7eebb2b
428926 F20101204_AAAMOU silvalugo_j_Page_063.jp2
3849e2a4bddd76471544a0af43839a1a
c7a72740e7a8871aad6769ccef5a48ba5c2b475d
F20101204_AAANLO silvalugo_j_Page_211.pro
1574fd84505db4727373a763426af8cd
a2312f674338a18a9ce99cb61df9931f8f74861f
4720 F20101204_AAAOIH silvalugo_j_Page_063thm.jpg
36fe6a203be3718b857a98bfc81437f1
c0f1c586dd5791be9f21a49d156c5bd77b3210fd
51344 F20101204_AAAMBK silvalugo_j_Page_075.pro
013eb68752c1f89c6cc61008e3f1bad4
ee8a57a9cb0aeb29fcd76b938e6eda95ebb2f540
11423 F20101204_AAANYZ silvalugo_j_Page_218.QC.jpg
b8b67f1da67c742e50b8dc5a91ae7799
f3abe63ef7e56644cc5f6696bc12185a27609edd
F20101204_AAAMOV silvalugo_j_Page_064.jp2
7636b2685a2d3ec8ffd5ad64230f1f7b
cc92039eef9c849b19495d052b2ca1fb22c1fc46
18750 F20101204_AAANLP silvalugo_j_Page_212.pro
055634da7b77ba32efd9d9771914eeb8
3d7e5439a0316ddc5dd6201cf4c68bab55314aed
8582 F20101204_AAAOII silvalugo_j_Page_065thm.jpg
1b291226082b776a834292097aefdfda
2eb06d3c317c2532d64541cdbf0c1a443526826c
96769 F20101204_AAAMBL silvalugo_j_Page_080.jpg
abe86fdd6adc64edf2ee98439abe13a1
7a28b3806e0491878b3a1c613ca96d39f627bb11
F20101204_AAAMOW silvalugo_j_Page_065.jp2
577ec6e19f32d96a8adc2b8c442ca471
777219dac17a68e6aa668b66fe820b97b5fc064d
28038 F20101204_AAANLQ silvalugo_j_Page_213.pro
f1fe0ec7710b821032b008931a3f648e
ad02043a856036678c01a6c307034aa27f833597
35119 F20101204_AAAOIJ silvalugo_j_Page_067.QC.jpg
89daf68ce58d271a27123da9a60d5a5e
fd6eaee5e34ee29347f0dc0146bded1c8b371e97
9293 F20101204_AAAMBM silvalugo_j_Page_242thm.jpg
3d38021543025ab4a7358f36f9cef3bb
6098f87b5d529474c4d1ba45054709c16e9f26f5
1051934 F20101204_AAAMOX silvalugo_j_Page_066.jp2
c33a2197c98ee6a36cb5701517f19f69
5ee82e4037a3c0a014a66fea85d2335f1bfc5ac5
7090 F20101204_AAANLR silvalugo_j_Page_214.pro
0cd56544eac07834cf3d751589e55686
c9e8a858b0fe26231219c5b29d347ecadafdb103
8954 F20101204_AAAOIK silvalugo_j_Page_067thm.jpg
9fcbc830130e1a555377891c5769666f
416afb8c3ffd1cc25c9e738395528a995c4a58ad
F20101204_AAAMBN silvalugo_j_Page_166.jp2
993d44f8ddd2519babc252d6d1293609
219ff2100d15714166719730c4e7f6711cda9e00
F20101204_AAAMOY silvalugo_j_Page_067.jp2
aa97f72fb1d51e9258c758ebf3b47266
ac86a4387e766becf4b8703b1565bad6a6428f4c
30436 F20101204_AAANLS silvalugo_j_Page_215.pro
fec3761705ed72fccff18cb3f2bdcace
752bc0bf962d78752205f337d233a1a5772a523c
9030 F20101204_AAAOIL silvalugo_j_Page_068thm.jpg
b55d18f082d97bfe5fcd9518cae99efd
0e7f1680ae6450e556caab2343726a20c85ef1cf
19809 F20101204_AAAMBO silvalugo_j_Page_142.pro
d36326b3d870ee6d308e6f685866db11
c170ef7a2c5e8a938e2493354e1944d2c4226b4d
F20101204_AAAMOZ silvalugo_j_Page_068.jp2
7daf0454a42d32ea9aa4780a27131618
5b1d9187c804729a98094960fc46b2872e780d35
11565 F20101204_AAANLT silvalugo_j_Page_216.pro
8d9377a9a65fee0002812ed1ccd0333f
7e4c3f2bd40ac6a53d275fca6f049415eee0857a
37761 F20101204_AAAOIM silvalugo_j_Page_069.QC.jpg
4c4930c442007844d1f1fc041ee61aec
d956c98a85b826075a9ffdb3af97d7b606cd18d5
F20101204_AAAMBP silvalugo_j_Page_042.tif
003fbb42096c2737af336cab7a0b73b6
49a4297bef33373c14abca6a174d8a6c4f2611bc
10254 F20101204_AAANLU silvalugo_j_Page_217.pro
95894df258086bbb85dddd31b8a5d61b
df5233369b94f66c0de78ce6e4bac52dd8d1a69b
30420 F20101204_AAAOIN silvalugo_j_Page_070.QC.jpg
c2eb9f9abd3a468fc46978b0557e3b08
46235423547aff3887d1ee95a333b7be653966fa
F20101204_AAAMBQ silvalugo_j_Page_019.tif
b1e7413a4e46f2f5a2cf4c318546e2ce
d5cd12d2d4cabc92788fe86ec1613d5f2b27cfb0
10922 F20101204_AAANLV silvalugo_j_Page_218.pro
9d67d09c01c9c43412f426343f310c84
180aedca27ca9bf1b0a4393a41613bffdfc3f920
7840 F20101204_AAAOIO silvalugo_j_Page_070thm.jpg
d41a2e073ea659288dab4332e8abe7a1
bccc74db6845be43d8c4b8c80bc0859c0763230a
1938 F20101204_AAAMBR silvalugo_j_Page_004.txt
683f95fadbb015827bd50249a6d33d3f
006fccecf6bb189878596218ce806c6d68e8578f
7942 F20101204_AAANLW silvalugo_j_Page_219.pro
a00ec82b2e1b6a2aa5d7744ebf0ffa0d
d3331b344923d221ba54a1195451caebbd4c3282
36048 F20101204_AAAOIP silvalugo_j_Page_073.QC.jpg
9430fed07a971378789e2041dcf09fe2
7feb7e71324e5d176639d85862d75ae17c4e5dd3
13858 F20101204_AAAMBS silvalugo_j_Page_062.pro
596c86e9f83f245b02ecdc86fb470327
b67aa78e841f059a28c1573cbbda5c53b42858c2
8472 F20101204_AAANLX silvalugo_j_Page_220.pro
f64c93e554235bc886ab0931fde1d4a9
ffd4f48ffd6030f26cef5a956b6668a1e36eaa35
8855 F20101204_AAAOIQ silvalugo_j_Page_073thm.jpg
6eb9316e0c4e5de50c0ed3cb3245fc85
e70442cbaa5ff013387c2c0286aa9339ff8d320b
F20101204_AAAMBT silvalugo_j_Page_219.tif
3f13a459f29a48c88b707977aa88f023
d18a9273a949079c84f563704b20d7a48f7449a8
8868 F20101204_AAANLY silvalugo_j_Page_221.pro
7af68f6587301a6dc9eb97231a87223d
09b44a30704b84b5a4e1f7a667b131ad1d979198
36955 F20101204_AAAOIR silvalugo_j_Page_074.QC.jpg
52b137cbddbce7689e1d7e33b65010a4
08649f179ba8999718f2b9d9a243a8475f72e319
F20101204_AAAMBU silvalugo_j_Page_018.tif
7a28160e66931e9b050c1d9d7d15a0cf
d47b03faad35b1e1474ccf98a4ac818cc010585f
11081 F20101204_AAANLZ silvalugo_j_Page_222.pro
f0000114e477b331718bf0d09f06a2fa
44f3c26ced69c8e2bf72f34cfbcbd43cb154178d
34743 F20101204_AAAOIS silvalugo_j_Page_075.QC.jpg
198e607ac9228fc5e38dd0d70557fd19
7a2efcd1009ef7a23cb21dd58b94c12092bfea9b
F20101204_AAAMBV silvalugo_j_Page_157.jp2
d08a567940e4222e2c84180341f1e2b6
7de29aa423f2b707cb9ba934fdf86f5df3590932
30045 F20101204_AAAOIT silvalugo_j_Page_076.QC.jpg
6f8460d7fc29e1c4ee50545ec542c279
ce8cf437c57443360923c7fdf3093b43473193cb
2130 F20101204_AAAMBW silvalugo_j_Page_094.txt
4b7d772992ecb79122f97efbc472891c
d80f43c1a9f0e319cc9e555dfe566ceca1cce975
7500 F20101204_AAAOIU silvalugo_j_Page_076thm.jpg
10a3e4eab214135492281c98328a17f6
c1dfeee1829094f66452d1f45c37a53123093e0c
1915 F20101204_AAAMBX silvalugo_j_Page_077.txt
9ccbb03d1337556d1c054e2a5e8f6918
68a883ae9fa6a676da1953e5d38d775063b28599
679148 F20101204_AAAMUA silvalugo_j_Page_208.jp2
a28744fcd8d3ebeeb790d41b7a404c1d
6396e6caf465c24a78eb0f2e310638527c466eeb
37174 F20101204_AAAOIV silvalugo_j_Page_078.QC.jpg
3e6975f75535db60576571b46c42c3cd
f8fedf9d53439881513a4384cfd652b4e8da377f
8184 F20101204_AAAMBY silvalugo_j_Page_036thm.jpg
94f018304a3e4f0b074eda261b94bdcf
1943e4e50d3267c33395cbd83775469b8e245091
750267 F20101204_AAAMUB silvalugo_j_Page_209.jp2
c86151b39381c72809ca99fe05a9544a
911f6c1c1a02d3d86ccb4b8d9298d5b116890f9a
9066 F20101204_AAAOIW silvalugo_j_Page_078thm.jpg
5eb93b59874d00a76a51851d4f90e8f6
62a860a0bb7d9265a05b0e772594150fcf3d5252
2146 F20101204_AAAMBZ silvalugo_j_Page_099.txt
2e6f23061b06ca2a6d984b3e42d20d52
0cac239796ced1af14f1dcb40e6322fd0f775fb1
879817 F20101204_AAAMUC silvalugo_j_Page_210.jp2
b6332766c1776423b1744b40ef1769c6
fbd3ef78ad160f3729b0d1a6da9b059810acd480
36352 F20101204_AAAOIX silvalugo_j_Page_079.QC.jpg
83f00e6db60db8ddd9e1028f832bceba
2b6d36953e784ef11f053567cc976c1bdf0ed6ec
1051940 F20101204_AAAMUD silvalugo_j_Page_211.jp2
fdcc6fb21c12b728932b996b4b2b3605
20f993aff9212e054bdc8ad2e141e580f1d55b98
9002 F20101204_AAAOIY silvalugo_j_Page_079thm.jpg
43f384db64c7d425f4553cfe08645ece
dbb16f8b1144d2d8fb0e297f50b3e1569d176431
402036 F20101204_AAAMUE silvalugo_j_Page_212.jp2
bf05abb3aa53c9ebf09e2b074d65e0fd
9203d5d2089d5e01f3803fee51b953d65ad931e2
32136 F20101204_AAAOIZ silvalugo_j_Page_080.QC.jpg
e6bfe5cb998fc69d871bfffe304a45e9
18948756f38126d7fc9ebca888b49800da3c4ddd
666145 F20101204_AAAMUF silvalugo_j_Page_213.jp2
2c81f62d5c916ed000234f018c284964
d80c9a6429ea460cde24167ac54a25675e85f7d8
140268 F20101204_AAAMUG silvalugo_j_Page_214.jp2
ff611074850df025c514ff98cab1f10b
d7ff368b30bb5ca946d4cd2a2ec5c68617382d66
2776 F20101204_AAANRA silvalugo_j_Page_116.txt
deae4f232535b85f8f8d881fdf7412f7
cd42cba022e92b6cdfc77fcf45686c47fa9f8a33
69289 F20101204_AAAMUH silvalugo_j_Page_215.jp2
0eb9133f841a4b838378718dde7dbe0f
23313409bfbf8b95b12c0224694db78c597ac84b
2802 F20101204_AAANRB silvalugo_j_Page_117.txt
114826202dd02453ddd88ad014341758
4d458ed053e10a7febae7276df629ba1c04576d0
406120 F20101204_AAAMUI silvalugo_j_Page_216.jp2
b962008a29441cde20460021a27b6419
8c767632ca7ee876bddd7489e1b2ad0b47eda8f2
343534 F20101204_AAAMUJ silvalugo_j_Page_217.jp2
2e9ec037e4a5ab0ccbea5658808fd7a0
fa1974332d5336294a37a2b7776ce28175b8755a
1278 F20101204_AAANRC silvalugo_j_Page_118.txt
78b0fd1ef53f3c7cb8cb9e5370f6f3fa
eb43fa0de2ca7d87af653205b3d3905e116de9ef
309241 F20101204_AAAMUK silvalugo_j_Page_218.jp2
16fe0deec8eb84634c2f64f25afa03f4
8dc038fbf0717276395b795fb59a02b364c18400
1284 F20101204_AAANRD silvalugo_j_Page_119.txt
60e40df4e68165c952e1012a19c1454b
033d7a323fcb8d264867a45567e46624611eb70c
107823 F20101204_AAAMHA silvalugo_j_Page_089.jpg
9bfbc4e36203ac6313f8935dfba26e74
f756a01c2ef9cf64ba1b0587d7d6bfe43228a108
341925 F20101204_AAAMUL silvalugo_j_Page_219.jp2
cebd1ec7c414a1baf56955a6fa0a4ab2
ae24a78a694eb2ee3e5f79e2cb645137ce1c426b
934 F20101204_AAANRE silvalugo_j_Page_120.txt
da489ae9b16f63e9da202a24cce0e8ba
2583e658e025a28788e4ab506796c16bd8d5a5c3
111598 F20101204_AAAMHB silvalugo_j_Page_090.jpg
10dd6611350f49171ca74a566b68dca9
26c2fd4b8b6519759ce38aad569781c11ac485b5
260195 F20101204_AAAMUM silvalugo_j_Page_220.jp2
62d8164879cc7e4ae469cfbaa837c619
9941b7c5c945abd548ee828d9fa47413de9b7d17
1097 F20101204_AAANRF silvalugo_j_Page_121.txt
b4574d886eb37422ded265c38b9d519f
7a9db358fead16d082494303f393e982b4acfc39
114805 F20101204_AAAMHC silvalugo_j_Page_091.jpg
b8c3486e379c43401d20f9bb662c1129
06ec0e4c9e31855df3f7155448fd304e027d386d
280029 F20101204_AAAMUN silvalugo_j_Page_221.jp2
af8fdf53fa453dd0157c4d7ae9548938
a85ea121a15a83bf7441fce1a19ba2f9ecc79e9f
1949 F20101204_AAANRG silvalugo_j_Page_122.txt
bb410878fc189dcf85908d2fa4caa561
692e95bae32682b4f709a5ed7e756ef6c0b3d27a
108719 F20101204_AAAMHD silvalugo_j_Page_092.jpg
6bc497d7743d0c94c9ed7484f8b5b1b4
4789119915cf44659842e2852244255110aaea02
345533 F20101204_AAAMUO silvalugo_j_Page_222.jp2
56f8581ec74caa732588f99a642bc6fc
ab892456de4f0ce82cc026cfb70e813f27a4efc4
317 F20101204_AAANRH silvalugo_j_Page_123.txt
416247d448e2247f2dc220336851c41a
e6994d15daaf0d8ab5cb8031750100b9db754b28
282121 F20101204_AAAMUP silvalugo_j_Page_223.jp2
91f4605b6bec32e82439e7430e5a9596
a8387b3d474ac520a1105b3e3744279f162efb8e
318 F20101204_AAANRI silvalugo_j_Page_124.txt
2fa306c112d766de3f26eb224fb0f7ad
3ae36470f631eabeb62216d67928703c46c01150
106475 F20101204_AAAMHE silvalugo_j_Page_093.jpg
4e044781368f90b7045c31040c505940
b9d3a60428544baf121a524da227122518e48559
364387 F20101204_AAAMUQ silvalugo_j_Page_224.jp2
bce23038061172ee05526f5c664c87a4
d35cc511921de98bc7d6da282bae5c2aca9bc0e5
F20101204_AAANRJ silvalugo_j_Page_125.txt
4311877daa882e00e3ffbc3f929d35f4
723c84e0f0027a98119f9307c680b9b24328ad3f
111151 F20101204_AAAMHF silvalugo_j_Page_094.jpg
8a9670993dfecb8cd01cbcce9e3fe732
d7e72d44a3019eb973dcc37e1516821feeb6b887
379 F20101204_AAANRK silvalugo_j_Page_126.txt
1b18831eb99fbc5aee35e59d3d1eb62e
14eeb9138068b4d94a60c62c007e7ddea1b1a42e
106121 F20101204_AAAMHG silvalugo_j_Page_095.jpg
b911954699a2c959396fdc1ce9fdeca8
d78f4289c11c429ce25a3ebb9a7eda1851cdcc19
F20101204_AAAMUR silvalugo_j_Page_225.jp2
5c81ef81e714a6c305a31d0af0542a4c
6df7bd07a1bd8a3e84d6cabf88d77c845f0c2376
F20101204_AAANRL silvalugo_j_Page_127.txt
246dd517161b3f2f83de5181737ef178
f10163b4f81212bf5a75bdf41934e458a49f961b
115642 F20101204_AAAMHH silvalugo_j_Page_096.jpg
a12ab89567128b679286596658216294
84b27771a301ce78d50dbf75931ecd0c0487b964
F20101204_AAANEA silvalugo_j_Page_246.tif
d9b88261d092f8bf3c569c0d73674691
8cb4da87a98cec8a858c338f6f2570930ef8a1de
F20101204_AAAMUS silvalugo_j_Page_226.jp2
7af522839f3f83ebec14cd5ec12e3987
81916aa152a8460ab0c0ac588af9400099abda55
838 F20101204_AAANRM silvalugo_j_Page_128.txt
2243e7267ba78b5e0675c25b64690e94
60b05fb7164613cb9d04ed2c2af5a3700da88c6d
108144 F20101204_AAAMHI silvalugo_j_Page_097.jpg
34bf05b9d1cf5e285e1703ab7c95fe84
3c4b6cb476538d94523cfbe2ffa28cbb0a8da8eb
F20101204_AAANEB silvalugo_j_Page_247.tif
157fc37e1aa68f28582108b50cc84831
89a3e8240fd08f1ac1c0166abe61b0dff6727f20
1051913 F20101204_AAAMUT silvalugo_j_Page_227.jp2
514680321104aaf01a1e279b9ae127dd
5ee77a114d25ed2892cbfef3a32da512a710de6d
112892 F20101204_AAAMHJ silvalugo_j_Page_098.jpg
2fa70dcdbd5817eb55e104cabb08514b
b725583b14cc9c461099c467dd465818262f773d
F20101204_AAANEC silvalugo_j_Page_248.tif
c4c1b613d64b06f1de3e9423f8020121
2e5900c727259771ed5352e1076c1809daa3b1f6
F20101204_AAAMUU silvalugo_j_Page_228.jp2
c8cefc2a2e46abd1a38e5893e161904e
f9481c6150d354fd3f93c89d8e655ffa9a5a88bb
770 F20101204_AAANRN silvalugo_j_Page_129.txt
c32351f1a5bdbe69ac1e949497dc3be3
762f081e469b19a18ab5573c489d5d156f657bc2
110224 F20101204_AAAMHK silvalugo_j_Page_099.jpg
3138f2c8023e0f78aee9d0ef7d235cce
4516720cbbded02a5f3eefc2d769e332cc6a3ec5
F20101204_AAANED silvalugo_j_Page_249.tif
67356ee3fccaf8e0d0b58a9dadbc5d6d
37b84ec5d3a59ec8f1db009fcf5236459eb21e63
F20101204_AAAMUV silvalugo_j_Page_229.jp2
60f8a6951c38fe3801567ee41239a218
b8ab5352c1d0e01aef48e6e43b0f620a7b4b531d
842 F20101204_AAANRO silvalugo_j_Page_131.txt
11172c410c6e1301df44651f73fb1596
bf2d5130b4bf05a4212fcc5c8b2c36bb8dbb699a
112338 F20101204_AAAMHL silvalugo_j_Page_100.jpg
903b35f62f87789bfa806cac8188c3bb
a8bb4559c0e2477d649ed3109d080bbe74cbd7c7
9003 F20101204_AAANEE silvalugo_j_Page_001.pro
2ecb48e282abbe1b5eda484075633794
d916a568815093502ac4315ef078685267f6214c
F20101204_AAAMUW silvalugo_j_Page_230.jp2
2551c8bf1ca10ad43aec9dfcc63ad214
11c7b7ed0d50ef0d188faab010aea9b38b1b03f3
679 F20101204_AAANRP silvalugo_j_Page_132.txt
70a599d454e953b293876fe2283ee5dd
5053243d6597b3541c8064a4a2a7f38450e1789d
111075 F20101204_AAAMHM silvalugo_j_Page_101.jpg
c43ced9d2ffcc0293a8ce2fd3ab7b4af
27b550acb768f60ae04f247039505f984c3712f5
1071 F20101204_AAANEF silvalugo_j_Page_002.pro
f99dbaf2fe1be76b3f6e53b45359afa7
8a451e9a93a1b0db27e5079a91aeff969ab6d987
F20101204_AAAMUX silvalugo_j_Page_231.jp2
13864ee5d758399fce56d29622536b09
8d46ae11ef3459bfad582982fca1c67d11d1ecf3
809 F20101204_AAANRQ silvalugo_j_Page_133.txt
e653ce8e9089dd64f5179ffc023ce9cb
55f1f5cf4ff6b389f164b7340aaf1ca8e4bf134d
115164 F20101204_AAAMHN silvalugo_j_Page_102.jpg
5f3a966fe15ee10b80df0da2ec589959
a5e1bc912a6f4ba2c86f5fb12d80c018c1f0b353
48178 F20101204_AAANEG silvalugo_j_Page_004.pro
eff07f2e89b201f748c8c015f66c827d
16e4b1d13064323406a717046cf1873935b63628
708844 F20101204_AAAMUY silvalugo_j_Page_232.jp2
cfed93c0ee73757193e717d914f6f178
fcfe38103e24b00c590b35fe7f0a0fc9c11bcfc5
617 F20101204_AAANRR silvalugo_j_Page_134.txt
bb649799be53f98d2db3309c1226ff97
9be599ba5ff3a25d2aa743ba47a37e31c2085a9f
128033 F20101204_AAAMHO silvalugo_j_Page_104.jpg
4ae63cf83916a6ca57a90b787c9a5dee
5023041d8f79230a697d285e9c1c855ffcf8a279
92325 F20101204_AAANEH silvalugo_j_Page_005.pro
5e8796877c9be06020643cb35e1aaaa5
165ffe96c50ea15f34f812f8ef96ba24d512fd75
819992 F20101204_AAAMUZ silvalugo_j_Page_233.jp2
78f297f642aa82f6a18536d2a3f5239c
682c5af0ac53a0db59a03daa29c389c7ae5fbb42
4012 F20101204_AAAOBA silvalugo_j_Page_152thm.jpg
955461fe76c20cd0e0cbd90e262558dd
a0f3f67bce2b0e487f2134a2b4bca2af91c313d3
405 F20101204_AAANRS silvalugo_j_Page_135.txt
53561dc80c33357d59fa241f66ace273
babca5c56bda6ec9420a86a46c701462ac95ad28
25887 F20101204_AAAMHP silvalugo_j_Page_105.jpg
613620b269a4f500c3c603d3c0e0a490
252e3a0e3e0cde49bae974cebde515b8b238cd6a
88482 F20101204_AAANEI silvalugo_j_Page_006.pro
72a5e55f3ee51f8fb7d8d3dc48ffecd8
323fe7317e3b9524881ed0cc12a23a68a87e0ab6
23620 F20101204_AAAOBB silvalugo_j_Page_145.QC.jpg
797e67727fc5f5c7298872e2d386e8fb
328f77094a7436eafdbf06097a623c93fe29872c
697 F20101204_AAANRT silvalugo_j_Page_136.txt
0f69bc57204b63c5e22f72c23d151994
c63016bac4775e0a4d1223e1b21ab8ad654213b5
118555 F20101204_AAAMHQ silvalugo_j_Page_106.jpg
528b61cd75aa5814f5716d1fc2aefe6f
854b2939a17baa180a92ec5c11897422c512b40a
80763 F20101204_AAANEJ silvalugo_j_Page_007.pro
54f4208f35eca929f05a3911bfa4ada4
b4fdc82ad0ae8b92f45ca076c1ba2a9afff462f9
35109 F20101204_AAAOBC silvalugo_j_Page_083.QC.jpg
cb17d4e177588646677eb76afd87d823
5eac25678365537a4d82c2aaa71cab05aa68cb5b
423 F20101204_AAANRU silvalugo_j_Page_137.txt
69124f6f350e1cbe2b1151158af0e1aa
6dc018867f5128a47bc3ac1b55ae8f67e678cdc9
122945 F20101204_AAAMHR silvalugo_j_Page_108.jpg
511bbc5b12d2bc5758342af8d4cb4529
9325cdbfd26549b3b20ace317a76d98092c29ea0
92751 F20101204_AAANEK silvalugo_j_Page_009.pro
521eb1e101ed4aad66ec250cafeaa60e
7a9259f2e740d52506849912febcc83ea986f37c
35458 F20101204_AAAOBD silvalugo_j_Page_157.QC.jpg
1c81e77f387c44c39b2ebd4b1c945d37
74968d8fe356701ef4fd4db4e189539368031c48
784 F20101204_AAANRV silvalugo_j_Page_138.txt
b7edd8f224a80525c83579b559f7180c
75a391f20f0aaceff33c3b788c7685b5ba7b6c50
123887 F20101204_AAAMHS silvalugo_j_Page_110.jpg
441adecf4410664ec8ce047f3bfb38c9
d65eb6376aada6c7f7ef3bfbe460e575f0ba7446
5510 F20101204_AAANEL silvalugo_j_Page_010.pro
2267c3fda52e2248ea866d0402d3ec62
1e2779c3569627b12fdbd7e4451d766f6bc3ac33
10303 F20101204_AAAOBE silvalugo_j_Page_124thm.jpg
fae66d84cf3dab9ca44726beffe99139
0a4f0d0d8e8c611abb144c01bc881710559b8e6a
836 F20101204_AAANRW silvalugo_j_Page_139.txt
d5a551cd77558d211f3978e95cff14dd
e4ec2e6479d38cf671c6d1c97fe555d6d559a470
154088 F20101204_AAAMHT silvalugo_j_Page_111.jpg
a5e7b382cb832f47a34a1f3d9788933d
b650d6e0d1013d38b9f8a17304dc27eab28904ac
68797 F20101204_AAANEM silvalugo_j_Page_011.pro
d7d1d7617514dae77b42a2424a458162
b6e1d34130c3425c89bb2d80dce9bf62746f827a
8910 F20101204_AAAOBF silvalugo_j_Page_235thm.jpg
2dd284cf0397841686d4279310c44279
72c929715d376bfb36d994dc4aff19f475129124
662 F20101204_AAANRX silvalugo_j_Page_140.txt
a5293e20b6bf6cddabb371dcb51a5dee
34d66a5b65d220572d3e1dcf9cf1a55bbd4c2670
124022 F20101204_AAAMHU silvalugo_j_Page_112.jpg
4a58a484692586d0e6394d91a631898a
d88c6af0ed8d6660e00339281168520bcb1c0727
78556 F20101204_AAANEN silvalugo_j_Page_012.pro
767154c489e695d304b874cbfb213f6c
79651498ce9d06506776301e266e3e6b6eec5813
3563 F20101204_AAAOBG silvalugo_j_Page_201thm.jpg
27beb7a09d07a1c50034e507d0ad0149
b3ce20bc4552a5d40184d7857607b38430dde952
867 F20101204_AAANRY silvalugo_j_Page_141.txt
976687c87899c4cd29d3a3e25dd5a2a0
2b8b881de8ccc2dd8148b6e39d24846fbfd1a9dd
87408 F20101204_AAAMHV silvalugo_j_Page_113.jpg
1b65b677f9bd970ffb7a1e7a97c92f7b
e8d5b12cf663ed594184a55e34dfba57b14ea336
84401 F20101204_AAANEO silvalugo_j_Page_013.pro
af1f71014323e2fcefae8848c881d9e3
8c10f6da988592a5c1f58a90a5b52d771b0ae4f2
29489 F20101204_AAAOBH silvalugo_j_Page_064.QC.jpg
7b70cd785cd44737721566ba6c0bd9d3
1be4805e000447683683c6ad8c7328cceb896259
944 F20101204_AAANRZ silvalugo_j_Page_142.txt
dd3a90bb5e82b05990ac22b1fcbb96d0
35de0d827c4487961926b690cc815ed3d94e55b1
155378 F20101204_AAAMHW silvalugo_j_Page_115.jpg
811a1a277bb925a67b0a022942a4bd7e
674c36110738185fcffcdc38fbe63c52c4ba1408
84616 F20101204_AAANEP silvalugo_j_Page_014.pro
f04cf7a1cc7b8fed349cf8ec43a92128
236c1b386f9ecfff5275c681e4260f5d181e1888
38002 F20101204_AAAOBI silvalugo_j_Page_242.QC.jpg
63f13fdd18fa42b4c4cc36e66d45b11b
e1f4c94f13ec07462a4a9d6f71591b7248367b41
107779 F20101204_AAAMHX silvalugo_j_Page_116.jpg
4a15699b6f7c94e3f8d73d1eb1db7453
64aa6b616e0e25e982c1d9b2d73bc53f0384c34e
6977 F20101204_AAANEQ silvalugo_j_Page_015.pro
288b0b8c54da2ae255039062d8bb4834
5daf67730ef1fd43242190e49e6b7974e867bfd1
3720 F20101204_AAAOBJ silvalugo_j_Page_222thm.jpg
f684cb09f889267e60df24cddc7688d4
463e385d3229954cc5cb411ed91c724c055c8cbe
108585 F20101204_AAAMHY silvalugo_j_Page_117.jpg
055c0f51494b0df79cb98df7d66771e9
2a4fb14ddae35b705ce2b5d687b109c714328a7d
47848 F20101204_AAANER silvalugo_j_Page_016.pro
5277181b31d4eeb97dbfe3f1c04d69b8
a4eeb05117df6ca963ef13036aa3e72ad6c4b799
36795 F20101204_AAAOBK silvalugo_j_Page_072.QC.jpg
94ff44e1f5f2c99dda556ff73a7ff006
a5bf030f438a3fb5949a8f14bc2d78e2b8111b55
48655 F20101204_AAAMHZ silvalugo_j_Page_118.jpg
55600579660baae185bdfdda75464f4f
b66ec44e3be82d39a4524a74d572091db9079954
53082 F20101204_AAANES silvalugo_j_Page_017.pro
7a2649a70c3b58a9baee1ec7c845def8
c7f23cba0ca97f5f0d480015a6f0363f285b6d8b
7709 F20101204_AAAOBL silvalugo_j_Page_109thm.jpg
01d41e8355895e67f809d499006d533d
c7e5a1b164fcf2908eca8d06ac4dfcc7458867a3
54036 F20101204_AAANET silvalugo_j_Page_018.pro
444ebc47bc4bebff857fa398b4ecd398
d189987cab113d763113a11bc12cbe84cbe1ed47
F20101204_AAAOBM silvalugo_j_Page_231thm.jpg
71bd09774f3ebccbab6b5232a59276ed
16c143928372d049285b4f55b3f6a677dd6ec983
53240 F20101204_AAANEU silvalugo_j_Page_020.pro
97cb2bcb739bdc3d9c02bf71cdb639af
d63ed18ceec7b8d113f6d2d23cdd2e6949247e44
4722 F20101204_AAAOBN silvalugo_j_Page_216thm.jpg
b78340b90a18398e3e0dc499ffbac788
4686c81e565caa59e31329bc5549b506af095fd5
52772 F20101204_AAANEV silvalugo_j_Page_021.pro
5c894370626de0f4df2d95f58640116d
693ad4131341bbde0b9a7a6f428275eeccabdc2e
40436 F20101204_AAAOBO silvalugo_j_Page_115.QC.jpg
aa7d7b58319ffe5b73966db6be8ed87d
4082f05abe9b1e92634828b2595d1421df690695
55193 F20101204_AAANEW silvalugo_j_Page_022.pro
56c71c704118714d35d2b80bd039ec17
9b068e997f70abdb57b74ed3997fb9b34596d871
37814 F20101204_AAAOBP silvalugo_j_Page_127.QC.jpg
876311f8eec9a2f49a3efa5cc5b95a18
ba7f13933d23a6d605b2ee1c5da3ae4216e9fc78
53692 F20101204_AAANEX silvalugo_j_Page_023.pro
c1f05c3e7fec5300d2765d706f467aa9
c755c240bf6c8864c5992dae4c953d7022456545
34946 F20101204_AAANXA silvalugo_j_Page_231.QC.jpg
711da32fbf3b832d72ff8658aef48013
6860167345868dd6ede31bcbb9f2626327043e7a
9171 F20101204_AAAOBQ silvalugo_j_Page_175thm.jpg
6fe658a2253cd8602683a5a515eb5c9e
46ca4effd8a9f2e967a797ab7e6f11b1017fc794
10136 F20101204_AAANEY silvalugo_j_Page_024.pro
45465df14c1d4191a73acfd59e978e1a
34507c764711fc676800891c9b884c04c44c3f10
19356 F20101204_AAANXB silvalugo_j_Page_050.QC.jpg
32c7bb2663306671377cd3510bc19534
3793333b03cd7d2789a68a7752081f9dd40578fd
37608 F20101204_AAAOBR silvalugo_j_Page_071.QC.jpg
32a834e0ae17f27f1c1ce078b1b44dd2
3c47b89c01fb7ed23e897bdd468714c35d06a415
2924 F20101204_AAANEZ silvalugo_j_Page_025.pro
3d0bde77f9cbc7b36345e5dae2020d56
d0dce19c5aad2ca4354a0f9fdb9ee3f0200a7d17
36694 F20101204_AAANXC silvalugo_j_Page_068.QC.jpg
4c753ddb58414931063b99a77aa12eb3
10ac6ee07cf63eaa9440435d971f7d34a5042b19
16417 F20101204_AAAOBS silvalugo_j_Page_147.QC.jpg
dc5ba2627374957890a20eb60678082c
7e4ee381a8e01013b97d9100bbf92ca6aaddaea8
9149 F20101204_AAANXD silvalugo_j_Page_148.QC.jpg
255c52311190ff88664f82924a63c906
ee49d72339899e1099e5da4fc673b032841b4061
36822 F20101204_AAAOBT silvalugo_j_Page_193.QC.jpg
2e3f3952838496a7c893714de83d35b5
f36fd042b35296ed3fae66e9b542b897bda5cead
332535 F20101204_AAAMNA silvalugo_j_Page_015.jp2
c088bafd63add7fafd6353a25201fd3e
8a307046d0b23aa2aae1db4d005ee6ff85212492
9127 F20101204_AAANXE silvalugo_j_Page_102thm.jpg
7897a57bf28961d62204040af58a07ac
0cfc5a35c94b7f1f91efef4175eecdf883e4e47b
8650 F20101204_AAAOBU silvalugo_j_Page_108thm.jpg
0c65733086288ebc4db6171e082a58ee
708b66c532c186acb100d4db6e54884a6a68b46c
1051953 F20101204_AAAMNB silvalugo_j_Page_016.jp2
20d22750166d05bd355ead80f2fe1ff3
171c95729e1455e1d5043bdf9ac4f6b86bebc461
12630 F20101204_AAANXF silvalugo_j_Page_219.QC.jpg
74571f978b647aef6b402183beea3ead
0bcace8e8d2cae8c0023752adcabddfb13ccd55b
39334 F20101204_AAAOBV silvalugo_j_Page_124.QC.jpg
8bfdb760693b25a3c11887d38eed73c2
84b45d4fa13147bb016c1eabf437fb855a7a2059
1051931 F20101204_AAAMNC silvalugo_j_Page_018.jp2
56d759c70b301c6989dd0293f03b4180
5d5e8bbca7eac5dc976d013a25dfd5aad3b04731
28407 F20101204_AAANXG silvalugo_j_Page_025.QC.jpg
ba129e0a133c19f0e16baf7cd65bc2cf
ca5681174f918ef8bcc95f32f17b1eaa94db81b3
36420 F20101204_AAAOBW silvalugo_j_Page_184.QC.jpg
903cc9fc8f77589912d522be1a27581d
ba464131d532e4aecbe867269d0782a84e21a91a
F20101204_AAAMND silvalugo_j_Page_019.jp2
ebdd082aba34abff26ec9752ffdcaa54
7b5ba3f3d9267adb96177231806480888a8052d8
9165 F20101204_AAANXH silvalugo_j_Page_227thm.jpg
6e3156d045af289d3a71d69e70b7dd99
d7f8ccaae0652f92b8cd15d138bc84792f1bf088
18260 F20101204_AAAOBX silvalugo_j_Page_119.QC.jpg
fc6b7f35f465e217e1d6894b512bb973
7e702d6fa110711adcec060679f5b507d06ead44
F20101204_AAAMNE silvalugo_j_Page_020.jp2
df6f5dc9e43217fcfa3f2fff89e672d6
a0332a4ccb6b01d1ef40bfa208f6431f3af18901
33267 F20101204_AAANXI silvalugo_j_Page_168.QC.jpg
14cc595c33bf5af8147d65f807ef0264
e505eb6d272be831e9067a4c28506f6e17cca455
F20101204_AAAOBY silvalugo_j_Page_195thm.jpg
961cd7ff4dbb6068292e1c20240879af
0beb9c06ac746f2dc990c042e6db9834f58e64d9
F20101204_AAAMNF silvalugo_j_Page_021.jp2
2f5bab14d8186de083bdbaca2a74b552
d038025794a4efaaf456c2817eaec66314b59f56
9261 F20101204_AAANXJ silvalugo_j_Page_188thm.jpg
772b5217e8d3d8c395f1d7a90e459ca7
f9231b0eed20afea7cf06df49ac00fd8fe13973e
3192 F20101204_AAAOBZ silvalugo_j_Page_150thm.jpg
f75eb5874c2e0de45fb9e95beac3e014
e7bfb8ae5f217362667efb392d047773c5819078
F20101204_AAAMNG silvalugo_j_Page_022.jp2
3215ffab510cf5bc50676b8d2db70a61
70c97289c4b9413f65a37cd350dee92d640a9cf4
12791 F20101204_AAANXK silvalugo_j_Page_155.QC.jpg
45e95dd983769d6196ff12a8196034dc
fa3955b7c79164cb17e4af287e76340fcfc2dc84
F20101204_AAAMNH silvalugo_j_Page_023.jp2
8d9f50e0816f0aa685d4b74208416a3a
db61f953bf7fe5f85d548e012915b7ee50facfa8
54699 F20101204_AAANKA silvalugo_j_Page_167.pro
04d768cafec3ad990f1193449ba3f141
2771b2b196bff2ce67f0d3b2a6a0ea44406efc98
36331 F20101204_AAANXL silvalugo_j_Page_167.QC.jpg
07f67ef6fc3549ab27e2d607f5c29246
2dba52b43f9e16379513f828311e881bc15cc695
238576 F20101204_AAAMNI silvalugo_j_Page_024.jp2
ebda638bb9d1d93ce4390a54356546a7
0a617f14d00074478dbdbbd9d12dbcf03270e8e7
50785 F20101204_AAANKB silvalugo_j_Page_168.pro
4090f8b66975d85a3276ae2549727f70
26ad43d7561d72cffe771b56953804dccb6b3545
12064 F20101204_AAANXM silvalugo_j_Page_223.QC.jpg
05ee04664d9625f7791966c75f49dd1b
d05b3344dd96dd3122f015c291a5810885a88f94
1051903 F20101204_AAAMNJ silvalugo_j_Page_025.jp2
265e3dd31d24fcaca086776dbbe234d5
6df07d4affb7e0c70f2790750fc23c5e47aa650b
54653 F20101204_AAANKC silvalugo_j_Page_169.pro
bdfb000c08e1b31215f19af1b647fcf2
0595c5c92e8822d4ad736e8353ad88c947e195b5
36954 F20101204_AAANXN silvalugo_j_Page_041.QC.jpg
5c2324123841533a6696ed4b7d7722f0
962f04cf614c428b7e3f55c7391734f5bdfed4e0
51293 F20101204_AAANKD silvalugo_j_Page_170.pro
7e93d612438032d2ec5bcb1fdf52c8bf
a58509f75e00ea150e1c4b4631375b2b81f6c663
1422 F20101204_AAANXO silvalugo_j_Page_249thm.jpg
5f1f25135e2b9c824322cab43b468166
dd93b08c3c8a207e1ef1bcb6b571d4a08a845686
448752 F20101204_AAAMNK silvalugo_j_Page_026.jp2
ce1ce2dc523215f9e8009df38ae2dc79
935c971cde0c67e864b30393f8b7f0e0d56c920a
44197 F20101204_AAANKE silvalugo_j_Page_171.pro
600a986f2eb782c191c6383c1a7047fa
16f78d464f0ac5949ad93ae05a2588c0778637f1
8451 F20101204_AAANXP silvalugo_j_Page_060thm.jpg
cd2c7452c60e9dd3f03ba1f3f9222f40
a9ca086e27280b9dff7bfe28a97e1cdd571233d7
54715 F20101204_AAANKF silvalugo_j_Page_172.pro
6f09b82b36bf69ca8ef13f1d59f662cd
c5198fae3c1fa8e17e5e4219b8d721c13faf0d5d
F20101204_AAAMAA silvalugo_j_Page_038.txt
74bfe1ef0bc27efe92f5576da7ac3070
f9f1a3f0a529fc336af6fd7e21176c24866b5de7
37122 F20101204_AAANXQ silvalugo_j_Page_243.QC.jpg
d7cb60c6a4754b012fac74753618230b
06044db467692eb054ffa131040475832e88fea6
686528 F20101204_AAAMNL silvalugo_j_Page_027.jp2
ace6e62dfd6b4853b01295c3da33366a
32b004ef553e28f888ef2ba9c4970d379048eb1a
35734 F20101204_AAAMAB silvalugo_j_Page_040.QC.jpg
991520d7695957b3623b3cc0bebec4c1
8f038867b2cb8d1190cd8979d5bbb76847361e88
38083 F20101204_AAANXR silvalugo_j_Page_192.QC.jpg
bb98f8397ca7b66844e7e74d1de4933b
85b537658963a89d40ffce1d0170412dbc0516bc
1051923 F20101204_AAAMNM silvalugo_j_Page_028.jp2
2b2292fdced63c55d21d92aca62d5778
242d53b46909dc5c585eee8bfc93fb8f95d90e65
46954 F20101204_AAANKG silvalugo_j_Page_173.pro
b84500271ae8567e9539e432b569e029
0778940379a8e9dbdc74235ac92db70560fc396d
20977 F20101204_AAAOHA silvalugo_j_Page_027.QC.jpg
98c5de413d4d8be62f808fc6116335c0
67ee3784c86e86b5b0b264802b0f285830c07912
197 F20101204_AAAMAC silvalugo_j_Page_025.txt
8e35592316030c671633e9a7f364f5e6
9f259a92891ac2e7f8af12199139959e280e60b2
37329 F20101204_AAANXS silvalugo_j_Page_198.QC.jpg
9702b0a3251907ed2a1b0444c1737e81
80b361e726a6f19c8880906c0d8a0f9a93322ee6
F20101204_AAAMNN silvalugo_j_Page_029.jp2
ae8f6495a786083b6421d4e6ffe0c0ed
428aff2a5f26cbad53f69dcc5da23c15e9b80f65
53046 F20101204_AAANKH silvalugo_j_Page_174.pro
fd541d662f11b964382336902c45a9cb
904dc020ec9748f11cbd17224fc68d72ac1beb03
6165 F20101204_AAAOHB silvalugo_j_Page_027thm.jpg
b1269d6a14bd36075ce9fb31632c78d2
064157cbc9a3c7801322f16263b7b59be54a3ac0
34737 F20101204_AAAMAD silvalugo_j_Page_093.QC.jpg
7a7b41e2b0e7e2d63d6498af41e49852
439bd354a7f134f9aa84928c70a06ec258246055
F20101204_AAAMNO silvalugo_j_Page_030.jp2
5cb573448665573c11fa9d697ad4d520
2b027825216bf2987cefd47177673c4d092a8ca1
56670 F20101204_AAANKI silvalugo_j_Page_175.pro
56c202b259bce0adf2ea60bc8c8a1b01
895cf44a7f9410d143d5c6908ec96e3a923f371f
110607 F20101204_AAAMAE silvalugo_j_Page_166.jpg
c13e2f20d10f0e253f47b11b3d335c0a
fd80e6ef61572697ab5dbf115e326275aed431c8
9606 F20101204_AAANXT silvalugo_j_Page_246thm.jpg
315ba2cb8c2e55101034d6126c68eac9
880e7abf36796853b3ff59cda955556ef81f4625
F20101204_AAAMNP silvalugo_j_Page_031.jp2
acac3739a49af2897258b790932093ac
09ca45a15a2533eaeb12041a4b5754c011f76b31
50628 F20101204_AAANKJ silvalugo_j_Page_176.pro
1ff577f65de133b68d5b4124e7bae241
555716961a057d91c917010f8594a69edd4604d9
6220 F20101204_AAAOHC silvalugo_j_Page_029thm.jpg
15d71c92a06ee477fb4945e373bf7bff
74fb6aeb8dc88bf07692fe34ee59a0383c929a09
1051857 F20101204_AAAMAF silvalugo_j_Page_124.jp2
6701fb7931e89c9d130cbad4a68a63f0
03700c3da04e7cd4264b2378ea3a8bc8b45a6484
9407 F20101204_AAANXU silvalugo_j_Page_192thm.jpg
418817b26034c867bb199149d1aa480b
dce79796c57328335c7913f543c06445b2a5d32a
F20101204_AAAMNQ silvalugo_j_Page_032.jp2
ce69962ad3e679837d366d8b041fe908
09b0db38e93df357c25f1607765a5934b3aa87f0







RESPONSES OF PLANT AND SMALL MAMMAL COMMUNITIES TO PRESCRIBED
BURNING IN CEDAR KEY SCRUB STATE RESERVE

















By

JOSE LORENZO SILVA-LUGO


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008


































2008 Jose Lorenzo Silva-Lugo



































To my mother, my wife, and my three children for all their love, support, and sacrifice.
To my adviser, Dr. George Tanner, for all his help, support, and advice.









ACKNOWLEDGMENTS

The Department of Wildlife Ecology and Conservation and the Department of

Environmental Protection provided the financial aid and logistic needed for carrying out the field

work. Florida Fish and Wildlife Conservation Commission provided a GPS unit and participated

during prescribed burning. Special acknowledgments go to my adviser, Dr. George Tanner, for

his support, help, and advice when I needed it. Dr. Tanner always replies to all my requests, and

he was always pleasant, cordial, and calm. I have fulfilled a very important goal in my life thanks

to him. I also thank my committee members, Dr. Dick Franz, Dr. Deborah Miller, Dr. Hardin

Waddle, and Dr. Alan Long for their help during the correction process.

M.S. James Collee and Jeff DiMaggio reviewed chapter drafts in advance and provided

helpful comments. Dr. Waddle, Dr. Arpat Ozul, and Dr. Gary White helped me out with the

MARK program. James Collee assisted me with the multivariate statistical analysis and spent

time studying and analyzing the complexity of the results and their interpretations. Jeff

DiMaggio and David Hoyt helped me to cut the dense scrub vegetation for making four trapping

grids during three months and provided logistic help during field work. When I did not have

transportation to go to Cedar Key, Jason Hall shared the field vehicle assigned to him with me

and Jeff DiMaggio provided transportation in the reserve. The staff members in the department,

Sam Jones, Monica Lindberg, McRae Caprice and Delores Tillman were always helpful in all

my requests. Dr. Susan Carr and Zachariah Welch lent PC-ORD and books to me. All these

years have been difficult, and I always had my mother, my wife, and my three children

supporting and giving me their love and the incentive that I needed to complete my education.









TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

L IST O F T A B L E S ..................................................................................................... . 7

LIST O F FIG U R E S .................................... .. .... .............. .................. ............. 11

A B S T R A C T ......... ....................... ............................................................ 16

CHAPTER

1 IN TR OD U CTION ......................................................... ................. .. ........ 18

2 PRE SCR IBED B U RN IN G PLA N .............................................................. .....................37

In tro du ctio n ................... ...................3...................7..........
O bje ctiv e s ................... ...................3...................7..........
M e th o d s ...........................................................................3 8
B before Burning ....................... .......................... 38
M easurem ents During Prescribed Burning ........................................ ............... 39
Measurements After Prescribed Burning ...........................................41
Predicting Fire B behavior ......................................................................................... ........42
R results and D iscu ssion .................................................................................................... 43
Rate of Spread, Flame Length, and Fire Intensity ....................................................43
Fuel M odel Predictions............................................. 44

3 RESPONSES OF LON-UNBURNED SCRUBBY FLATWOODS TO BURNING ...........65

In tro d u c tio n ............. ...... ......... ....................................................................................... 6 5
O objective ......... ......... .................... ....................... ............... 75
M methodology ......... ........... ......................... ...........................75
R e su lts .............. ...... .. ............................................................................... .................. 8 0
Species List and Recovery Modes........................... ...............80
A Site Analysis ........................................................ 80
Postburn Recovery and Survival ........................................ ............ ...............82
Structural and Compositional Changes in Response to Prescribed Burning ................... 86
Flowering and Fruiting after Prescribe Burning ..............................................................89
Discussion ..................... .......................... ..................89
True Control Sites...................................... ..........................89
Fire Survival and R recovery M odes .......................................................................... ..91
Speed of R recovery Process ....................... ............................. ............... ....92
Community Shift in Response to Prescribed Burning................................................98
Age at First Flowering after Prescribed Fire ........................................................... 102









4 SMALL MAMMALS RESPONSEs TO PRESCRIBED FIRE............... .................157

Introdu action ............... ............ ......... ......................... 157
Objectives and Research Hypotheses ........... ..... ......... ........................ 165
M methodology ......... ..................................... ...........................165
Trapping M methods ........................... .......................... .... .... ......... ......... 165
Data Analysis.................... ....................................167
Assumptions of the Cormack-Jolly-Seber model............................168
The goodness of fit (G OF) test................. .. ......................................................168
Model comparison, model selection, and hypothesis testing ..............................170
R results ................... ................... ......................................................... .. 174
Number of Captured Individuals in Treatment and Control Sites..............................174
Number of Captured Individuals in Scrubs and Wetlands ................. ... ............. 175
Flood and Fire Effects on Podomysfloridanus ............................................................177
Flood and Fire Effects on Sigmodon hispidus...........................................................179
T rapping P predators ........................ .. ........................ .... .... .................182
D iscu ssio n ................... ...................1.............................2
Low Capture Success .......................... ......... .. .. ..... .. ............ 82
Flooding E effect .................................................................................. 185
Population Responses during and after Prescribed Burning ......................................187
Burrow s as refugia during prescribed burning ...................................................... 187
P described burning effect ..................... ........................................ ..................... 189
Population increases/decreases after prescribed burning .................... ........... 190
Habitat selection: immigration, emigration, and returning to burned areas...........196

5 CONCLUSIONS AND RECOMENDATIONS..................... ... ....................... 225

C o n c lu sio n s ........................................................................................................................... 2 2 5
Prescribed Burning Plan ......... ............. ................................... ............... 225
Plant Species Responses to Prescribed Burning ............ .... ....................225
Small M ammal Responses to Prescribed Burning ................................ ............... 227
R eco m m en d atio n s.................................................................................................... .. 2 2 8

APPENDIX

A TWO TAILS T TEST COMPARISON AMONG ENVIRONMENTAL VARIABLES
AND FIRE BEHAVIOR CHARACTERISTICS BETWEEN SITES 5C AND 2M .........233

L IST O F R E F E R E N C E S ..................................................................................... ..................234

B IO G R A PH IC A L SK E T C H ......................................................................... ... .....................248









6









LIST OF TABLES


Table page

2-1 Comparison of the prescribed burning plan between sites 5C and 2M in Cedar Key
Scrub State R reserve. .................. .................................. ........... ............... 47

2-2 Environmental variables during prescribed burning in sites 5C and 2M in Cedar Key
Scrub State Reserve. ..................................... ... .......... ...............48

2-3 Observed fire behavior characteristics in sites 5C and 2M in Cedar Key Scrub State
R e se rv e ................... ............. ........................................... ................ 4 9

3-1 List of plant species recorded in quadrats in treatment and control sites in Cedar Key
Scrub State Reserve. .................................. ......... ........... ....... ...... 104

3-2 Species richness, Simpson's index, Shannon-Wiener's index, and Shannon-
Wiener's evenness for preburn conditions in control (5A & 5D) and treatment (5C
& 2M) sites in Cedar Key Scrub State Reserve.......................................................105

3-3 Multiple mean comparison (Duncan's test) between clusters created by using
Euclidean distances and Ward's minimum variance linkage fusion. Abundance and
mean percent cover were standardized. Significant level = 0.05..................................106

3-4 T test, ANOVA test, and Kruskal-Wallis test for comparing means and medians
among treatment and control sites under preburn conditions in Cedar Key Scrub
State Reserve. Data for T test and ANOVA were standardized. Test for ANOVA and
Kruskal-Wallis is F test and Chi-squared test, respectively. Significant level = 0.05 .....107

3-5 Pre- and postburn absolute densities for all species in 5C in Cedar Key Scrub State
Reserve. Absolute densities for control sites 5A and 5D are also shown. d = days. M
= m months ................................................................................108

3-6 Pre- and postburn absolute frequencies for all species in 5C in Cedar Key Scrub
State Reserve. Absolute frequencies for control sites 5A and 5D are also shown. d =
day s. M = m month s.................................................................... 109

3-7 Pre- and postburn absolute mean % cover of herb and woody species in 5C in Cedar
Key Scrub State Reserve. Absolute mean percent cover for control sites 5A and 5D
are also shown. d = days. M = M months. .............. ... .... ........................... 110

3-8 Pre- and postburn absolute importance values of herb and woody species in 5C in
Cedar Key Scrub State Reserve. Absolute importance values for control sites 5A and
5D are also displayed. d = days. M = Months. ..............................................111

3-9 Pre- and postburn absolute densities of herb and woody species in 2M in Cedar Key
Scrub State Reserve. Absolute densities for control sites 5A and 5D are also shown.
d = days. M = m months. ........ .. ........................ .......... .. .. ....... .... 112









3-10 Pre- and postburn absolute frequencies of herb and woody species in 2M in Cedar
Key Scrub State Reserve. Absolute frequencies for control sites 5A and 5D are also
presented. d = days. M = M onths......... ................. ............................... ............... 113

3-11 Pre- and postburn absolute mean percent cover of herb and woody species in 2M in
Cedar Key Scrub State Reserve. Absolute mean percent cover for control sites 5A
and 5D are also shown. d = days. M = Months. ........................ ................114

3-12 Pre- and postburn absolute importance values of herb and woody species in 2M in
Cedar Key Scrub State Reserve. Absolute importance values for control sites 5A and
5D are also displayed. d = days. M = months.......................... ............... ... ............ 115

3-13 Density of ramets in 5C after prescribed burning in Cedar Key Scrub State Reserve. ... 116

3-14 Density of ramets in 2M after prescribed burning in Cedar Key Scrub State Reserve. ..117

3-15 Tree mortality in quadrats and in grids 5C and 2M in Cedar Key Scrub State
R eserv e ........................................................................ ............... 1 18

3-16 Coefficient of determination (r2) resulting from Detrended Correspondence Analysis
of the pre- and postburn sites of a long-unburned scrubby flatwoods in Cedar Key
S crub State R eserv e. .......................................................................... .. .............. 119

3-17 Summary statistics of the Multi-response Permutation Procedure for woody absolute
densities and mean percent cover between control and treatment sites at preburn and
12 months postburn in Cedar Key Scrub State Reserve. Results are given for
Euclidean and Sorensen distances. ...............................................................................120

3-18 Multiple comparison for absolute densities and mean percent cover between control
and treatment sites at 12 months postburn and between preburn and 12 months
postburn values of treatment sites in Cedar Key Scrub State Reserve. ...........................121

3-19 Comparison among several studies and Cedar Key Scrub State Reserve (CKSRR)
regarding common variables measured in each research. The data presented for
bareground and for plant species is mean percent cover. Data for Schmalzer and
Hinkle's study average all strata. Data for CKSSR is the average of the two treatment
site s ......................................................... ....................................12 2

4-1 Number of captured individuals in the scrubby flatwoods and in the vegetation
surrounding wetlands per trapping session in treatment and control sites in Cedar
K ey Scrub State R reserve ..... .......................................... ...................... ............... 201

4-2 The Goodness of Fit test for the general model phi(g*t) p(g*t) for Podomys
floridanus and Sigmodon hispidus ................................. ............... ............... 202

4-3 Result browser of the fitted candidate model set of 16 models for Podomys
floridanus after adjusting to c-hat = 1.1387 in treatment and control sites in Cedar
K ey Scrub State R reserve ................................................................................ ...... ....... 203









4-4 Set of 25 models after adding phi(Flood + Fire) in combination with p constant (.),
time dependent (t), and group dependent (g) in the analysis Flood and Fire effect on
survival probabilities of Podomysfloridanus in treatment and control sites in Cedar
K ey Scrub State R reserve ..... .......................................... ...................... ............... 204

4-5 Estimated survival (phi) and recapture (p) parameters by using model phi(Flood+
Fire) p(.) in the analysis Flood and Fired effect on survival probabilities of Podomys
floridanus in treatment and control sites in Cedar Key Scrub State Reserve.
C confidence interval = 95% .................................................................. .....................205

4-6 Set of 28 models after adding phi(Flood + Fire) in combination with p(Flood + Fire),
p(Flood), and p(Fire) in the analysis Flood and Fire effect on survival probabilities of
Podomysfloridanus in treatment and control sites in Cedar Key Scrub State Reserve. .206

4-7 Estimated P parameters from models phi(Flood+ Fire) p(.) in the analysis Flood and
Fired effect on the survival probabilities of Podomysfloridanus in treatment and
control sites in Cedar Key Scrub State Reserve. Confidence interval = 95%. ................207

4-8 Estimated survival parameters (phi) by using model averaging for the set of 28
models in the analysis Flood and Fire effect on the survival probabilities of the
Podomysfloridanus in treatment and control sites in Cedar Key Scrub State Reserve.
C-hat = 1.1387; 95% confidence interval. ............................................ ............... 208

4-9 Result browser of the candidate model set of 16 models for Sigmodon hispidus fitted
and adjusted to c-hat = 1.5694 in treatment and control sites in Cedar Key Scrub
State R eserv e................................. ........................................................ ............... 2 09

4-10 Set of 22 models after adding phi(Flood + Fire) in combination with p (., g) to the
candidate model set in the analysis Flood and Fire effect on survival probabilities of
Sigmodon hispidus in treatment and control sites in Cedar Key Scrub State Reserve. ...210

4-11 Estimated survival (phi) and recapture (p) parameters by using model phi(Flood) p(.)
and phi (Flood + Fire) p(.) in the analysis Flood and Fired effect on survival
probabilities of Sigmodon hispidus in treatment and control sites in Cedar Key Scrub
State Reserve. Confidence interval = 95% .................................. ............. ................. 211

4-12 Estimated P parameters from models phi(Flood) p(.) and phi(Flood + Fire) p(.) in the
analysis Flood and Fire effect on the survival probabilities of Sigmodon hispidus in
treatment and control sites in Cedar Key Scrub State Reserve. Confidence interval =
95% ........................................... ......... .................2 12

4-13 Estimated survival parameters (phi) by using model averaging for the set of 25
models in the analysis Flood and Fire effect on the survival probabilities of
Sigmodon hispidus in treatment and control sites in Cedar Key Scrub State Reserve.
C-hat = 1.5694; 95% confidence interval. ............................................ .....................213

4-14 Predators captured before the 8th trapping session in treatment and control sites in
Cedar Key Scrub State Reserve. ..... ........................... ....................................... 214









4-15 Trapping effort carried out by several studies conducted on small mammals in
Florida. C ode: TN = trapping nights..................................................................... .... 215









LIST OF FIGURES


Figure page

1-1 Cedar K ey Scrub State Reserve....................................................................... 25

1-2 Thirty years of weather data for Cedar Key (Accuweather.com) ............................... 26

1-3 Weather data from local station in Cedar Key Scrub State Reserve...............................27

1-4 Three species of oaks in Cedar Key Scrub State Reserve. .............................................28

1-5 Ericaceous shrubs in Cedar Key Scrub State Reserve ............................................. 29

1-6 Palm species in Cedar Key Scrub State Reserve. .................................. ............... 30

1-7 Herbaceous species in Cedar Key Scrub State Reserve. ................................................31

1-8 Two rodent species found in Cedar Key Scrub State Reserve........................................32

1-9 Two cotton rodents found in Cedar Key Scrub State Reserve. .......................................33

1-10 Location of treatment (5C and 2M) and control sites (5A and 5D) in Cedar Key
Scrub State R eserve.. ................. .................................. ........... .. .............34

1-11 Treatment sites in Cedar Key Scrub State Reserve. .................................. ............... 35

1-12 Control sites in Cedar Key Scrub State Reserve ............ ........................36

2-1 Improvement of the firebreak located at the north side of site 2M in Cedar Key Scrub
State R eserv e ................................................... ..... ............... 50

2-2 Sensor with bands of temperature-sensitive painting. ....................................................51

2-3 Sensors experim ented tem perature> 1094 C ..................................................................52

2-4 Prescribe burning maps in Cedar Key Scrub State Reserve. ...........................................53

2-5 Fire line created by the firefighter at the border of the stand in Cedar Key Scrub State
R e se rv e ................... ............. ............................................ ................ 5 4

2-6 Examples of fire front in Cedar Key Scrub State Reserve.................................... ............55

2-7 Applying prescribed burning in Cedar Key Scrub State Reserve................................56

2-8 Taking pictures with a reference person of know height in Cedar Key Scrub State
R e se rv e ................... ............. ............................................ ................ 5 7









2-9 Effect of prescribed burning in the scrubby flatwoods in site 5C in Cedar Key Scrub
State R eserv e ................................. ........................................................... ............... 5 8

2-10 Effect of prescribed burning in the scrubby flatwoods in site 2M in Cedar Key Scrub
State R eserv e ................................. ........................................................... ............... 59

2-11 After prescribed burning in Cedar Key Scrub State Reserve. ........................................60

2-12 Burned animals in Cedar Key Scrub State Reserve ......................................................61

2-13 Rate of spread comparison between Cedar Key and fuel models 4, sh5, and sh8
according to B ehavePlus 3.0.2 .............................................. .... .......................... 62

2-14 Flame length comparison between Cedar Key and fuel models 4, sh5, and sh8
according to B ehavePlus 3.0.2 .............................................. .... .......................... 63

2-15 Examples of flame length in Cedar Key Scrub State Reserve...................................64

3-1 Quadrats in Cedar Key Scrub State Reserve.............. ........................... ...... ............ 123

3-2 Ramets of Quercus myrtifolia in Cedar Key Scrub State Reserve. .............................. 124

3-3. Ramets of Quercus chapmanii in Cedar Key Scrub State Reserve.............................125

3-4 Ramets of Lyoniaferruginea in Cedar Key Scrub State Reserve. .................................126

3-5 Ramets of Gaylussacia nana in Cedar Key Scrub State Reserve..................................127

3-6 Similarity coefficients among the four study sites in Cedar Key Scrub State Reserve...128

3-7 Median linkage dendrogram for herb and woody species in treatment and control
sites under preburn conditions in Cedar Key Scrub State Reserve...............................129

3-8 Scatterplot of the first two canonical axes corresponding to the cluster analysis with
Median linkage fusion method for herb and woody species in treatment and control
sites under preburn conditions in Cedar Key Scrub State Reserve...............................130

3-9 Average linkage dendrogram for herb and woody species in treatment and control
sites under preburn conditions in Cedar Key Scrub State Reserve ..............................131

3-10 Scatterplot of the first two canonical axes corresponding to the cluster analysis with
Average linkage fusion method for herb and woody species in treatment and control
sites under preburn conditions in Cedar Key Scrub State Reserve.............................132

3-11 Ward's minimum-variance linkage dendrogram for herb and woody species in
treatment and control sites under preburn conditions in Cedar Key Scrub State
R eserv e ................... ........................................................... .............. 13 3









3-12 Scatterplot of the first two canonical axes corresponding to the cluster analysis with
Ward's minimum-variance linkage fusion method for herb and woody species in
treatment and control sites under prebum conditions in Cedar Key Scrub State
R e serve e ................... ......................................................................... 13 4

3-13 Duncan's multiple comparisons for the median abundances of Serenoa repens and
Vaccinium myrsinites among treatment and control sites in Cedar Key Scrub State
R eserv e ................... ......................................................................... 13 5

3-14 Prebum and postburn mean percent cover ofbareground, litter, and debris in Cedar
K ey Scrub State R reserve ......... .............................. .. ........................... ............... 136

3-15 Prebum and postburn vegetation height in 5C and 2M in Cedar Key Scrub State
R eserv e. .. ................ .... ......... .... ......................................................137

3-16 Prebum and postburn absolute densities of the most abundance herb species in Cedar
K ey Scrub State R reserve .. .... ............................ ............................. ............... 138

3-17 Prebum and postburn absolute frequencies of the most abundance herb species in
Cedar Key Scrub State Reserve. ...... ........................... ....................................... 139

3-18 Prebum and postburn absolute mean percent cover of the most abundance herb
species in Cedar Key Scrub State Reserve. ........................................ ............... 140

3-19 Prebum and postburn absolute importance values of the most abundance herb species
in Cedar K ey Scrub State Reserve ................................. ............... ............... 141

3-20 Prebum and postbum absolute densities of the most abundance woody species in
Cedar Key Scrub State Reserve. ...... ........................... ....................................... 142

3-21 Prebum and postburn absolute frequencies of the most abundance woody species in
C edar K ey Scrub State R reserve. .......................................................... .....................143

3-22 Prebum and postburn absolute mean percent cover of the most abundance woody
species in Cedar Key Scrub State Reserve. ........................................ ............... 144

3-23 Prebum and postburn absolute importance values of the most abundance woody
species in Cedar Key Scrub State Reserve. ........................................ ............... 145

3-24 Prebum and postburn absolute ramet density of the most abundance herb species in
Cedar K ey Scrub State R eserve...................................................................... 146

3-25 Prebum and postburn absolute ramet density of the most abundance woody species
in Cedar K ey Scrub State Reserve .................................. ............... ............... 147

3-26 Prebum and postburn species richness in treatment sites in Cedar Key Scrub State
R eserve........ ................................ ................................................14 8









3-27 Prebum and postburn species diversity in treatment sites in Cedar Key Scrub State
R reserve .................. ................. ...............................................149

3-28 Prebum and postburn evenness in treatment sites in Cedar Key Scrub State Reserve....150

3-29 Detrended correspondence analysis (DCA) sample ordination for densities in 5C in
Cedar Key Scrub State Reserve. ..... ........................... ....................................... 151

3-30 Detrended correspondence analysis (DCA) sample ordination for mean % cover in
5C in Cedar Key Scrub State Reserve. ........................................ ........................ 152

3-31 Detrended correspondence analysis (DCA) sample ordination for densities in 2M in
Cedar Key Scrub State Reserve. ..... ........................... ....................................... 153

3-32 Detrended correspondence analysis (DCA) sample ordination for mean % cover in
2M in Cedar Key Scrub State Reserve. ........................................ ....................... 154

3-33 Stand and species ordination of oak-saw palmetto scrub based on preburn absolute
mean percent cover in Kennedy Space Center. .................................. ................ 155

3-34 Site and species ordination of scrubby flatwoods based on preburn absolute mean
percent cover in Cedar Key Scrub State Reserve. ................................... ..................... 156

4-1 Trapping sessions (red blocks) carried out in Cedar Key Scrub State Reserve...............216

4-2 Number of captured individuals per species per trapping session in treatment sites 5C
and 2M in Cedar Key Scrub State Reserve................................. ........................ 217

4-3 Number of captured individuals per species per trapping session in control sites 5A
and 5D in Cedar Key Scrub State Reserve. ........................................ ............... 218

4-4 Number of captured individuals in scrubs and wetlands per trapping session in
treatment sites 5C and 2M in Cedar Key Scrub State Reserve................. ........... 219

4-5 Number of captured individuals in scrubs and wetlands per trapping session in
control sites 5A and 5D in Cedar Key Scrub State Reserve........................................220

4-6 Survival probabilities of Podomysfloridanus estimated by the model phi(Flood +
Fire ) p(.) in the set of 28 models in treatment and control sites in Cedar Key Scrub
State R reserve (chat= 1.1387). ........................................ .............................................22 1

4-7 Podomysfloridanus's survival probabilities calculated by the model phi(Flood + Fire
) p(.) and by model averaging (28 models) in treatment and control sites in Cedar
K ey Scrub State Reserve (chat=1.1387) ......................................................................... 222

4-8 Survival probabilities of Sigmodon hispidus quantified by the model phi(Flood +
Fire) p(.) in the set of 22 models in treatment and control sites in Cedar Key Scrub
State R reserve (chat= 1.5694). ........................................ .............................................223









4-9 Sigmodon hispidus' survival probabilities estimated by the model phi(Flood + Fire)
p(.) and by model averaging in the set of 22 models in treatment and control sites in
Cedar Key Scrub State Reserve (chat=1.5694).................... ...............224









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

RESPONSES OF PLANTS AND SMALL MAMMAL COMMUNITIES TO PRESCRIBED
BURNING IN CEDAR KEY SCRUB STATE RESERVE

By

Jose Lorenzo Silva-Lugo

May 2008

Chair: George Tanner
Major: Wildlife Ecology and Conservation

Although prescribed burning is an important management tool for ecosystem restoration in

Cedar Key Scrub State Reserve, this is the first study that analyzes the effect of prescribed

burning on plants and small mammals. In addition, this is the first research carried out on plant

community response to prescribed fire in coastal scrub on the west side of Florida, and the 12th

study about the effects of prescribed burning on small mammals in Florida. The main objectives

were to determine: (a) if there were structural and compositional changes in the plant community

after prescribed burning, (b) if small mammals used wetlands as temporal refugia after

prescribed fire; and (c) if prescribed burning had a negative effect on the survival of the small

mammal species.

The experimental design consisted of two treatment and two control sites that were

sampled before and after burning from December 2003 to August 2006. Preburn vegetation

samples were conducted one time in all sites, and postburn vegetation samples were carried out

every three months for a 12 month period. Fifty quadrats (4 m2 each) per site were assessed in

each sampling. Resprouting was the main way of surviving and recovering from fire by the

majority of the species, and almost all of the dominant species reached preburn levels during the

12 months period. This fast recovery of the vegetation after burning has been reported in the









literature but not in one year. The Detrended Correspondence Analysis showed that woody

species had structural and compositional changes during the first three months postburn, but

there were more compositional than structural changes after that. According to the Multi-

response Permutation Procedure, the structural changes were significant; therefore, there were

significant changes in absolute densities in treatment sites between pre- and 12 months postbum

and between control values and 12 months postbum as a consequence of prescribed burning.

A total of 29,340 trapping nights were completed in treatment and control sites. Each site

had a grid (100 traps) and a wetland next to it with two transects (10 traps each). Mice were

marked to monitor movements between scrub and the vegetation surrounding wetlands during

four trapping sessions before and after prescribed burning. A total of 184 individuals of

Sigmodon hispidus (cotton rat), Podomys floridanus (Florida mouse), Peromyscus gossypinus

(cotton mouse), and Ochrotomys nuttalli (golden mouse) were monitored during this study. In

treatment sites, mice were captured mainly in the scrub (75%) before burning, they used the

vegetation surrounding wetlands as temporal refugia for 11 months after burning, and they

returned to the scrub after that. In control sites, mice were captured mainly in the scrub (91%)

during the study. MARK analysis was only carried out on S. hispidus and P. floridanus because

of the small sample size obtained for the other two species. MARK indicated that fire did not

have a negative effect on the survival of S. hispidus. I cannot state the same for P. floridanus

because the p parameter was not estimable. However, the data indicated that mice moved to

wetlands and survived for 11 months. These results will provide guidance to managers in

prescribed burning plans to establish a fire return interval according to the recuperation of the

vegetation and to maintain viable populations of small mammals.









CHAPTER 1
INTRODUCTION

Fire is a natural disturbance and an important ecological factor for ecosystem management.

Fire has occurred across the landscape of United States for at least 2 millions of years (Franz and

Quitmyer 2005). This natural disturbance alters landscape structure, functions, and maintains

biodiversity (Pyne et al. 1996). Therefore, fire is an ecological process that greatly influences

composition, structure, and dynamics of many ecosystems. The ecological role of fire for

ecosystem management has been appreciated because managers rely on fire history to document

land management planning and silvicultural prescription, to study the effects of past fires and

past fire exclusion, to simulate natural fire intervals, to perpetuate communities, and to schedule

prescribed fire (Pyne et al. 1996). Particularly, prescribed fire started to be used in the southern

United States before most other regions.

The prescribed burning era began after a long period of fire suppression in the southeastern

United States. Fire suppression started in 1890, declined in 1930, and continued through the

1940s (Williams 2002). Prescribed burning began in the 1930s after several scientific

publications supported the idea of burning wild lands. Some of these publication were "The Use

and Abuse of Fire on Southern Quail Preserve," published in 1931; "Use of Prescribe Fire in

Southeastern Upland Game Management," published in the Journal of Forestry in 1935; and

"Relation of Burning to Timber and Wildlife" in the North American Wildlife Conference in

1935 (Kennard 2007). These publications created the political atmosphere to re-introduce

prescribed burning. However, prescribed burning was discontinued in Okefenokee Swamp in

1930 (reintroduced in 1970) and in Welaka Reserve in 1935 (reintroduced in 1980). The first

official prescribed burning carried out on federal land took place in Osceola National Forest in

1943 (Stanturf et al. 2002). Therefore, the process of re-introducing prescribed burning into the









southeastern United States was not a one-time event. But, it started to be more frequently used in

1945 and extensively used after 1980.

During the 1980s, an estimated 16 million ha of forest land and 1.6 million ha of range and

agricultural lands were treated with prescribed fire each year in the southern United States (Wade

and Lunsford 1989). The majority of this treated area was for wildfire hazard reduction, wildlife

habitat improvement, and range management. Nearly 0.81 million ha of rangeland were burned

annually in Florida alone (Brown and Smith 2000). The main reason for carrying out such

extensive prescribed burning in the Southeast, specifically in Florida, was because several

natural ecosystems depend on fire. One of these ecosystems is the Florida scrub.

Florida scrub is a distinctive and threatened ecosystem (Myers 1990, Whelan 1995,

Menges 1999, Brown and Smith 2000, Schmalzer 2003). It is distinct because it supports a high

number of threatened and endangered plants and animals (Myers 1990, Stout and Marion 1993,

Stout 2001). It is threatened because of natural fragmentation, human perturbations, and fire

exclusion (Myers 1990). In addition, natural fire no longer occurs with the same intensity and

frequency because the scrub habitat has been fragmented and reduced. Conservation of this

unique ecosystem relies on management and research.

Management and research of the scrub habitat is essential for the survival of many plant

and animal species. Prescribed burning has been the primary management technique for

maintaining scrub communities because it is a fire-maintained system (Myers 1990, Whelan

1995, Menges 1999). Understanding the structural and compositional changes of the scrub

communities after prescribed fire is important for making management decisions to maintain

appropriate conditions for plants and animals relying on these communities (Schmalzer and

Hinkle 1992a). In addition, understanding prescribed burning effects on wildlife is critical in









order to provide a more comprehensive management based on the knowledge of plant and animal

responses to prescribed burning. Even though the combination of management and research is

needed for conservation and restoration purposes, research about plant and animal responses to

prescribed fire is strongly needed in several public lands in Florida. One of these public lands is

Cedar Key Scrub State Reserve (CKSSR).

Prescribed burning is the main management tool in CKSSR, and it has been intensively

used since 1985. Prescribed fire is considered the most potent and critical natural resource

management tool at the reserve (DEP 1998). The main objectives of the program are to restore

the natural fire regimen within the reserve, create a mosaic of different successional stages, and

maximize ecological diversity (DEP 1998). To achieve these objectives, CKSSR was divided

into burn zones, and burn programs were assigned to each zone. In addition, these objectives

were established because natural communities, and the associated plant and animal communities

adapted to them, have been negatively impacted by extended periods of fire suppression. For this

reason, the program targets restoring the habitat for the endangered Aphelocoma coerulescens

(Florida scrub jay) and other species of interest such as Podomysfloridanus (Florida mouse) and

Gopheruspolyphemus (gopher tortoise). These species have specific habitat requirements that

are not satisfied by long-unburned scrubs. Proper management of the scrub habitat in the reserve

will be critical for the long-term survival of these species of interest. Although prescribed

burning has an important role in the reserve, no study has evaluated the effect of prescribed

burning on plants or wildlife within this scrub community. Research on this topic is essential for

proper management. This study is the first research conducted with the purpose of evaluating

plants and small mammal responses to prescribed burning in scrubby flatwoods in CKSSR.









Study Area. CKSSR is located in Levy County, Florida, approximately 4 km east of the

town of Cedar Key (Figure 1-1). It consists of 1973 ha, and it was acquired in 1978 under the

Environmentally Endangered Land program (DEP 1998). Cedar Key has a warm and humid

climate. Based on 30 years of weather records (Weather.com), annual temperature and

precipitation average 20.8 C and 126.3 mm (Figure 1-2), respectively, but year-to-year

variability is high. For instance, based on March 2004 February 2005 data from

Accuweather.com and March 2005 August 2006 data from a local station in CKSSR, annual

temperature and precipitation average 21.0 C and 110.4 mm, respectively (Figure 1-3). The

heaviest rainfall typically takes place from June to September with some precipitation in all

months of the year. Thunderstorms are frequent during the summer, and lightning strikes are

common.

Hurricanes Charley (9-14 August), Frances (25 August 8 September), and Jeanne (13-28

September) hit the Florida peninsula in 2004. Charley did not hit Cedar Key directly, but its

winds brought some rainfall to the area. Frances and Jeanne hit Cedar Key directly and brought a

high precipitation into the area. A total of 372.5 mm fell in Cedar Key during September (Figure

1-3). This is 204.5 mm over the monthly average precipitation (Figure 1-2). This amount of

precipitation caused water from nearby wetlands to inundate the scrubby flatwoods and the sand

pine scrub, and they remained partially flooded for several weeks.

Different types of soils support different plant communities. CKSSR has ancient dunes of

aeolian origin (White 1970). Sand deposits in the reserve are considered to be part of the Silver

Bluff Terrace (DEP 1998). The reserve has eight types of soils that range from well-drained

sandy soils in the upland to poorly drained, frequently flooded, and mucky soils in tidal marsh

(Slabaugh et al. 1996). Numerous wetlands are integrated with other communities in the









landscape. CKSSR is a mosaic of wetlands (basin swamps, basin marshes, depression marsh,

tidal marshes, hydric hammock, and estuarine), mesic flatwoods, scrubby flatwoods, sand pine

scrub, and sandhill. This mosaic of habitats makes prescribed burning difficult because each

habitat has a different set of optimal burning conditions. Particularly, scrubby flatwoods are

surrounded by mesic flatwoods and wetlands and may occassionaly be adjacent to sand pine

scrubs. Occasionally, the prescription for an upland vegetative community includes surrounding

wetlands. This is one aspect of interest land managers because the vegetation surrounding

wetlands might play a role as refugia after prescribed fire in the scrubby flatwoods. Scrubby

flatwoods occur on sites of well-drained sandy and acid soils and low in nutrients. They

represent a great percentage of the total land area in the reserve, and several of them are over-

mature because the last wildfire in the area occurred in 1955 (DEP 1998).

Scrubby flatwoods are represented by several oaks, ericaceous, palm, and herb species.

The most common oaks are Quercus myrtifolia (myrtle oak), Q. geminata (sand live oak) andQ.

chapmanii (Chapman oak) (Figure 1-4). Ericaceous species are Lyoniaferruginea (rusty lyonia),

L. lucida fetterbushh), and L. fruticosa staggerbushh) (Figure 1-5). Palms are Serenoa repens

(saw-palmetto) and Sabalpalmetto saball palm) (Figure 1-6). Herb species richness is low

because of the overgrown condition of the scrubby flatwoods. Some of them are the following:

Galactia elliottii (Elliot's milkpea), G. mollis (soft milkpea), Solidago odora (Chapman's golden

rod), Crotalaria rotundifolia (rabbit bells), Woodwardia virginica (Virginia chain fern) and

several Panicum spp (Figure 1-7). Oaks and ericaceous species dominate the scrubby flatwoods,

and this is the habitat for several animal species of interest.

CKSSR has species of concern (G. polyphemus and P. floridanus) and threatened species

(Drymarchon corais couperi (eastern indigo snake) and A. coerulescens) according to Florida









Fish and Wildlife Conservation Commission. This study deals with P. floridanus and other small

mammals found in scrubby flatwoods such as Ochrotomys nuttalli (golden mouse), Peromyscus

gossypinus (cotton mouse), and Sigmodon hispidus (cotton rat) (Figure 1-8 and 1-9). Even

though CKSSR has the potential to be a study area for scientific research, it has not received the

attention that it deserves from the scientific community.

Little research has been conducted in CKSSR. Amoroso (1993) carried out a floristic

study, in which several carnivorous orchids species were reported. Morgan (1998) studied the

association of P. floridanus with G. polyphemus's burrows and vegetation characteristics.

Several surveys for P. floridanus and A. coerulescens have been carried out. Dr. James Layne

monitored the population of P. floridanus in one stand from 1957 to 1995. Later, Florida Fish

and Wildlife Conservation Commission and the Department of Park and Recreation continued

trapping P. floridanus in the reserve in several locations from 1995 to 1997. A. coerulescens has

been surveyed annually since 1980. Since no study has evaluated the effects of prescribed

burning on plants and small mammals, an experimental design was planned to carry out this

research.

Four sites were selected to study the effects of prescribed burning. The experimental

design considered two treatment sites (5C and 2M) and two control sites (5A and 5D) not

selected at random (Figure 1-10 through 1-12). The park manager already had planned to bum

long-unburned scrubby flatwoods in the reserve, and we visited them to do the selection. I chose

four sites with the same characteristics regarding fire history, plant species composition, a

wetland next to them, and no mechanical treatment (cutting or roller chopping). The other sites

fulfilled the first three criteria, but they received partial mechanical treatment. The experimental

design included vegetation sampling and trapping in these sites and in the vegetation surrounding









wetlands next to trapping grids before and after prescribed burning. The vegetation surrounding

wetlands was an ectone between the scrubby flatwoods and the proper vegetation of wetlands.

This dissertation was interested in determining the potential role of the vegetation surrounding

wetlands as refugia for small mammals when adjacent scrubby flatwoods are prescribed burned.




















































Figure 1-1. Cedar Key Scrub State Reserve. A) Location in Florida. B) Boundary of the reserve.












26 A


W 20

O
2420

15 14 14









I ~ 1 20 -----------11 I I----------
15
I-

S10


5


0)
Jan Feb Mar Apr May Jml Jul Aug Sep Oct Nov Dec

Months


300
B
249
250
217

E 200
171 168
0
150
115 120

0 100
75 67
I-
50
50


Jan Feb Mar Apr May Jim Jul Aug Sep Oct Nov Dec

Months

Figure 1-2. Thirty years of weather data for Cedar Key (Accuweather.com). A) Average monthly
temperature. B) Average monthly precipitation.










30 27282726 262826 262727 A
S25
325 -
21 21
O 20
16
15 __1414 15
S12511












400 373 B
350 323

1 50 ------
E 300

0 199
gD 1160 162 153


100 37 B7 7fu
350 2
r k4I 116 153




Months


Figure 1-3. Weather data from local station in Cedar Key Scrub State Reserve. A) Monthly
average temperature. B) Average monthly precipitation.



















































Figure 1-4. Three species of oaks in Cedar Key Scrub State Reserve. A) Quercus myrtifolia
(myrtle oak). B) Quercus geminata (sand live oak). C) Quercus chapmanii
(Chapman's oak).
























B
kziX Kk


C













Figure 1-5. Ericaceous shrubs in Cedar Key Scrub State Reserve. A) Lyoniaferruginea (Rusty
lyonia). B) Lyonia lucida fetterbushh). C) Lyoniafruticosa staggerbushh).






























B




















Figure 1-6. Palm species in Cedar Key Scrub State Reserve. A) Serenoa repens (scrub palmetto).
B) Sabalpalmetto saball palm).























B


























Figure 1-7. Herbaceous species in Cedar Key Scrub State Reserve. A) Galactia elliottii (Elliot's
milkpea). B) Solidago odora (Chapman's golden rod). C) Galactia mollis (soft
milkpea).

A






























B




















Figure 1-8. Two rodent species found in Cedar Key Scrub State Reserve. A) Podomys
floridanus (Florida mouse). B) Ochrotomys nuttalli (golden mouse).
























B














Figure 1-9. Two cotton rodents found in Cedar Key Scrub State Reserve. A) Peromyscus
gossypinus (Cotton mouse). B) Sigmodon hispidus (cotton rat).































Figure 1-10. Location of treatment (5C and 2M) and control sites (5A and 5D) in Cedar Key
Scrub State Reserve. Yellow blocks are trapping grids installed in each site. The
actual size of each site is bigger than the grid.




































B



















Figure 1-11. Treatment sites in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. There
were two transects of traps between grid and the wetland.





































B


















ir#


Figure 1-12. Control sites in Cedar Key Scrub State Reserve. A) Site 5A. There were two
transects of traps between grid and the wetland. B) Site 5D.











CHAPTER 2
PRESCRIBED BURNING PLAN

Introduction

The prescribed burn plan for sites 5C and 2M was elaborated by the Park Manager Jeff

DiMaggio. The prescribed bum plan considered the most relevant environmental variables that

affect fire behavior and other factors of interest such as area to be burned, fire history, plant

communities, smoke screening test, smoke sensitive areas, fire break/site preparation, special

precaution for specific areas, firing procedure, contacted agencies, bum zone map, weather data,

prebum checklist, safety procedures, and required personal and equipment. This plan required

inspection of the sites to be burned in order to observe the condition of the vegetation (not

previously burned or roller chopped) and to monitor environmental variables and the Keetch

Byram Drought Index (KBDI) index (Keeth and Byram 1968) to assure that prescribed burning

would be conducted under preferred environmental conditions. In addition, prescribed burning

should be the same across treatment sites.

Fire behavior characteristics should not vary between treatments. This is a very strong

constraint from the experimental point of view. Even though this was a difficult challenge to

achieve, the park manager developed two plans with the same objectives and outcomes expected

from them.

Objectives

The prescribed burning plan for sites 5C and 2M had the following objectives: (a) reduce

vegetative biomass on the ground by a minimum of 60%, (b) top kill woody vegetation by a

minimum of 75%, and (c) reduce the vegetative mass for habitat improvement for listed species.









This dissertation also had the following objectives: (a) to quantify rate of spread and flame

length, (b) to indirectly estimate fire intensity through recording temperature during prescribed

burning, (c) to determine if fire behavior characteristics were the same in treatment sites, and (d)

to compare the results from Cedar Key fires with the predicted fire behavior from three model

fuels in BehavePlus 3.0.2.

Methods

Before Burning

Mowing vegetation at the border of stands was the first site preparation (Figure 2-1). This

job was carried out by using a Gyrotrac machine with a drum mounted on the front with many

cutting blades. The border was mowed up to 3 m width around the stand and at least 2 months in

advance of burning. In addition, a Brown tree cutter machine cut the remaining scrub to mineral

soil to improve the firebreak.

Placing markers, taking vegetation samples, and installing sensors were needed to quantify

rate of spread, moisture content, and temperature during prescribed burning, respectively.

Existing pine trees, posts, and flags were used to mark specific places for quantifying distance

and time from the starting ignition point. These distances and times were used to calculate rate of

spread. A stratified random sampling was used to take vegetation samples in 20 points the day

before burning between 11:00 am and 4:00 pm. Even though sampling points were assigned

randomly, samples were selected in relatively undisturbed places and representative of the fuel

complex and species composition. In each sampling point, a total of 20 samples of dry fuel were

collected for each of the following classes: 1-h timelag (< 14 inch diameter), 10-h timelag (1/4 to

1 inch), 100-h timelag (1 to 3 inches). In addition, 20 samples of live herb and woody vegetation

were collected including only stem, branches, and leaves. Samples were put in brown paper bags,









labeled, and weighed in situ. Later, samples were oven-dried for 10 days at 60 OC and reweighed.

Fuel moisture content was calculated by using Equation 2-1:

Percent moisture content =- x 100 (2-1)



At the same sampling point, two temperature sensors (Figure 2-2) were installed to measure

temperature during burning. Mica tags (12.7 x 7.6 cm; material resistant to high temperature)

were painted with five Tempilaq temperature-sensitive paints that melted at the following

temperatures: 204 C, 427 C, 621 C, 816 C, and 1093 C (see Figure 2-3). One tag was

inserted at the ground level and the other was attached with wire to a tree at 1.5 m above the

ground.

Details about the prescribed burning plan in 5C and 2M are displayed in Table 2-1.

According to this table, the plan was the same for both sites. The differences between 5C and 2M

prescribed burning plans were the area to be burned, the estimated flame length, and personnel

needed.

Measurements During Prescribed Burning

The prescribed burning map for site 5C is presented in Figure 2-4. This map illustrates the

ignition point (NE comer of the site) and where the fire lines ended (SW comer of the site). Fire

line is the fire spread by the firefighter at intervals of approximately 15-20 m at the border of the

site in order to control the spread of the fire (Figure 2-5). The test fire was completed at 10:38

am, and two fire lines were immediately started at the border of the site. The first line started on

the north side of the site, moved west and later moved south on the west side of the stand. The

second line began at the same point where the first line was started and moved south on the east









side of the site. The two lines met at the SW corner of the stand. This burning design was

planned to burn the site in an effective way.

Burning with a combination of back and head fire made the plan successful. The park

manager knew in advance from previous days that wind was blowing mainly in the SE-SW

direction and planned to start burning at the SW corner of the site. However, wind direction

shifted to NW-W the evening before burning and in the morning of the burning day. Therefore,

the park manager decided to start burning at the NE corner due to the shift in wind direction and

because 5C should start burning with back fire. The two fire lines moved simultaneously, but the

first line burned faster than the second one. Then, the park manager coordinated the movement

of fire lines on both the west and east sides of the stand in such a way that they did not move

farther than half site 5C going south. The idea was to bum half 5C at the north side first by using

head fire from the west side fire line (moving south on the west side of 5C) and back fire from

the east side fire line (Figure 2-6). In this way, we created a "firebreak" for the head fire

originated by the time the fire lines were near the SW corner of the stand. This plan worked even

though wind direction changed 14 times between 10:38 am and 2:09 pm. By the last time, 5C

was completely burned.

Prescribed burning in 2M used the same technique as in 5C. Wind direction was SE-SW

on previous days, so the park manager decided to start at point A of 2M (Figure 2-4) in order to

burn the treatment site first (where the trapping grid was located). Fire test was carried out at

10:54 am (Figure 2-7) and immediately after it, a first fire line started going NE at the border of

2M (Figure 2-4 and 2-7). From 10:54 am to 1:25 pm, the front fire (back fire) moved slowly and

reached transect C of the trapping grid (approximately half the distance between the first and

third yellow arrow at the border of 2M on Figure 2-4), and from 1:25 pm to 2:20 pm, it moved









fast (head fire) because the wind changed direction to NE. This change in wind direction not

only increased rate of spread, but also increased fire intensity. Burning the side of 2M where the

trapping grid was located finished at 2:20 pm. Starting points B and C began at 11:20 am and

4:00 pm, respectively, and the corresponding fire lines burned the rest of 2M.

Environmental variables were quantified during prescribed burning every 30 minutes. A

Dwyer hand-held wind meter was used to record wind speed, and a Sling-Psychrometer was used

to measure air temperature and relative humidity. In addition, cloud type, state of weather, and

fire conditions were monitored.

Photograph documentation and monitoring time of back and head fires were carried out

during prescribed burning. A digital camera was used to take pictures and mini videos (up to 3

minutes) during the entire process. Since these pictures would be used to estimate flame length,

an object or a firefighter was used as a reference (Figure 2-8). Out of 92 and 79 pictures taken in

5C and 2M, respectively, 30 pictures were selected in each site to measure flame length. Rate of

spread of the fire front was measured from the time the firefighter started to make the fire line.

Special attention was focused on changes in wind direction in order to quantify back or head fire

rate of spread. Back and head fires were only quantified up to 20-30 m from the border of the site

because the visibility was limited due to smoke. Standing at the top of a truck helped me to

watch and to record the advance of the fire front until it reached marked pine trees, posts, or

flags.

Measurements After Prescribed Burning

Sites 5C and 2M were monitored right after prescribed burning. Firefighters reviewed each

site after burning, particularly 8-10 m from the border of each site in order to look for spots still

burning and to stop these fires by adding water. In addition, firefighters also sprayed water on

pine trees still burning. The burned areas were re-checked during the night and the next day for









smoke and flare-ups. I visited sites 5C and 2M the next day after prescribed burning with the

purpose of collecting temperature sensors, estimating flame length with the pictures in situ,

taking pictures, and looking for wildlife killed by fire.

Predicting Fire Behavior

The program BehavePlus 3.0.2 (Andrews and Bevins 2005) was used to predict fire

behavior by using the default worksheet. This worksheet contains fuel models, fuel moisture,

surface wind speed, and slope steepness. Fuel model 4 (Chaparral), fuel model sh5 (high load,

dry climate shrub), and fuel model sh8 (high load, humid climate shrub) were run with the

moisture contents measured for the three types of dry fuel, live herbs, and live woody vegetation

collected before burning and the wind speed recorded during burning. Slope steepness was input

to zero. Fuel model 4 was selected because this is the model for the shrub group characterized by

the California mixed chaparral. Fuel model sh5 was chosen because both sites had low

precipitation during the last 2 weeks before burning (91.1 mm in April and 93.6 mm in May).

However, model sh8 was also selected because the KBDI index during prescribed burning (5C =

220 and 2M = 234) suggested soils had wet conditions. The minimum, maximum, and average

wind speed registered during prescribed burning and the average fuel moisture content for each

dry and live fuel type were input into each model. As a consequence, each model produced three

rates of spread and flame lengths. These results were compared with the minimum, maximum,

and average rates of spread and flame length observed during prescribed burning and calculated

from the distance/time quantified in situ for head fire and from the pictures taken during

prescribed burning.









Results and Discussion

Rate of Spread, Flame Length, and Fire Intensity

Table 2-2 shows duration of prescribed burning, wind direction change, surface wind

speed, air temperature, air relative humidity, fuel moisture content, and KBDI index during

prescribed burning in 5C and 2M. The duration of the burning was approximately the same in

both sites. However, the time recorded for 2M corresponded only to the burning time for the

portion of 2M where the trapping grid was installed. Wind changed direction 14 times in 5C and

one time in 2M. This un-controlled variable affected fire behavior in 5C and the difference

between the two sites regarding how burning occurred in both sites. Wind speeds were not

significantly different between the two sites, though, but air temperature and air relative

humidity were significantly higher in 2M than in 5C (Appendix A). Average fuel moisture

content for each fuel type was slightly higher in 2M than in 5C with the exception of live

herbaceous, but they were not significantly different (Appendix A). Probably, this was due to the

higher air temperature and relative humidity found in 2M. The KBDI index suggested wet soils

in both sites by the time of the burning.

Observed fire behavior characteristics such as flame length, rate of spread, and fire

intensity are presented on Table 2-3. Flame length and rate of spread were slightly higher in 5C

than in 2M, but they were not significantly different (Appendix A). Average fire intensity was

significantly lower in 5C than in 2M. The change in wind direction probably was one of the

factors that varied and lowered fire intensity in 5C (Appendix A). In this site, out of 60 censors,

the temperature registered by 28 censors varied between 204 C and 816 C, and the rest of the

censors experienced temperatures equal or higher than 1093 C. I state that the temperature was

higher than 1093 C because there was no paint remaining on the mica sheets. In 2M, out of 60

censors, five censors recorded 204 C, five censors registered 427 C, one censor recorded 816









C, and the rest experienced temperatures equal to or higher than 1093 C. So, fire intensity was

higher an almost homogeneous in 2M in comparison with 5C. Therefore, fire behavior

characteristics were not exactly the same in treatment sites, but at least flame length and rate of

spread were similar. Even though fire intensity was not the same in treatment sites at the heights

where temperature was recorded, it was high enough to reduce almost 100% of the vegetation

and top kill almost 100% of the above-ground woody vegetation in both sites with very little

damage to wildlife.

The objectives of the prescribed burning plans were achieved. Figures 2-9 through 2-11

illustrate how sites 5C and 2M appeared the day after burning. Almost all trees were burned in

5C and all trees were burned in 2M. Since fire intensity varied in 5C, flames did not completely

consume all tree foliage. Approximately, 7% of the trees did not bum completely, but the stems

fell down after several weeks. Also, I only found two Terrapene carolina bauri (eastern box

turtle) in 5C, one in 2M, and several insects burned after prescribed burning (Figures 2-12).

Fuel Model Predictions

The comparison among models and the results found in Cedar Key are shown in Figures 2-

13 and 2-14. A clarification about this comparison is needed because models predicted values

according to mathematical models, but this is not the case for the relationship between observed

rate of spread / flame length and surface wind speed in Cedar Key. The minimum, average, and

maximum observed values of wind speed were matched with the minimum, average, and

maximum observed values for rate of spread and flame length. This match was a valid

assumption because it was expected that minimum, average, and maximum rates of spread or

flame length would occur when wind speed was also at the minimum, average, and maximum

value. Based on this assumption, the comparison was made. Another assumption was that surface









wind speed measured during prescribed burning was equal to midflame wind speed used in

BehavePlus.

BehavePlus models a nearly linear relationship between rate of spread or flame length with

midflame wind speed. In both sites, rate of spread or flame length increased when wind speed

increased. In addition, rate of spread or flame length decreased when relative humidity increased.

However, more data are needed to confirm the linear relationship found in this study. Figure 2-

13 illustrates that rates of spreads found in Cedar Key fell between models sh8 and sh5 or

between humid and dry climate shrubs in both sites. It is important to highlight that the values

from Cedar Key were close to the predicted values from model sh8 in 2M. Model 4 and sh5

over-estimated rate of spread and flame length for Cedar Key. Figure 2-14 presents a different

scenario. The average flame length recorded in Cedar Key was between model sh5 and model 4,

but the minimum and maximum values of flame length fell beyond the predicted values from the

models in both sites. Now, the reality was that minimum and maximum flame lengths were

obtained from pictures taken during prescribed burning in both sites. This is a fact illustrated on

Figure 2-15. Therefore, models 4, sh5, and sh8 over-estimated flame length at the minimum

wind speed and under-estimated flame length at the maximum wind speed recorded during

prescribed burning in both sites.

Prescribed burning studies have not reported fire behavior characteristics in the reviewed

literature. Few studies have described the environmental variables during prescribed burning

(Abrahamson and Abrahamson 1996a, Weekly and Menges 2003, and Greenberg 2003);

however, fire behavior characteristics were not reported. This is a critical aspect in experimental

designs that focus in fire effects. Although it is difficult to achieve the same fire behavior









characteristics among sites considered under treatment, an attempt must be made in order to

know how similar/dissimilar stands are regarding the application of the treatment.











Table 2-1. Comparison of the prescribed burning plan between sites 5C and 2M in Cedar Key
Scrub State Reserve.


Category 5C


Area to burn
Starting time
Last burn/years fire suppression
Fire procedure
Wind direction
Surface wind speed (Min/Max)
Transport wind speed (Min /
Max)
Minimum mixing height
Dispersion Index
Air temperature (Min / Max)
Air relative humidity maximum
(Min/ Max)
Fine fuel moisture content
Drought index
Rate of spread
Flame length
Personnel required
Passed smoke screening system


12 ha
10:00 AM
1955 / 50 years
Baking, flanking, strip head
SE-SW
11 / 23 kph


2M
33 ha
10:00 AM
1955 / 50 years
Baking, flanking, strip head
SE-SW
11 / 23 kph


15 / 32 kph
610m
Day 65 max
4/32 C

35-50 %
8-12%
< 550
2-5 m/min
2-7 m
6-8
Yes


15 / 32 kph
610m
Day 65 max
4/ 32 C

35-50%
8-12%
< 550
2-5 m/min
2-5 m
6-10
Yes









Table 2-2. Environmental variables during prescribed burning in sites 5C and 2M in Cedar Key
Scrub State Reserve.
Category Category 5C 2M
Date 04/21/05 05/18/05
Starting time 10:38 10:54
Ending time 14:09 14:20
Duration 3:31 h 3:26 h
Wind direction change 14 times 1 time
Minimum 2.41 1.61
Surface wind speed (kph) Maximum 4.83 4.83
Average 3.22 3.62
Minimum 23.89 26.67
Air temperature (C) Maximum 27.22 32.22
Average 26.60 29.31
Minimum 40.00 55.00
Air relative humidity (%) Maximum 54.00 75.00
Average 49.00 62.00
Minimum 3.81 2.43
FMC lh timelag (%) Maximum 10.10 11.96
Average 6.51 7.42
Minimum 1.69 3.45
FMC 10h timelag (%) Maximum 19.27 22.20
Average 7.52 8.31
Minimum 6.08 4.41
FMC 100h timelag (%) Maximum 36.14 38.85
Average 13.95 14.85
Minimum 27.36 19.66
FMC live herb (%) Maximum 86.92 82.93
Average 59.20 55.93
Minimum 42.86 39.48
FMC live woody (%) Maximum 86.05 74.94
Average 60.56 62.13
KBDI index 220 234
Codes: FMC = fuel moisture content. KBDI = Keetch Byram Drought Index.












Table 2-3. Observed fire behavior characteristics in sites 5C and 2M in Cedar Key
Reserve.


Category


Flame length (m)



Rate of spread-back fire (m/min)



Rate of spread-head fire (m/min)


Fire intensity-Temperature at the
ground level (oC)


Minimum
Maximum
Average
Minimum
Maximum
Average
Minimum
Maximum
Average
Minimum
Maximum
Average


5C
1.00
6.50
3.99
0.78
2.29
1.66
3.71
8.67
6.21
204.0
1093.0
858.0


Scrub State


2M
1.00
5.50
3.44
0.73
2.22
1.47
2.22
6.67
4.87
204.0
1093.0
1019.0


Minimum 204.0 204.0
Fire intensity-Temperature at 1.5 m Maxim 1 0
Maximum 1093.0 1093.0
above the ground level (oC)
Average 600.0 899.0

































Figure 2-1. Improvement of the firebreak located at the north side of site 2M in Cedar Key
Scrub State Reserve. The road is the fire break and it can be seen at the left side (6
cm from left to right) of the picture. The firebreak was improved by increasing its
width (the remaining 6 cm from middle to the right of the picture).































S.- B




















Figure 2-2. Sensor with bands of temperature-sensitive painting. A) Sensor not exposed to fire.
B) First band was melted indicating that temperature reached 204 C.


































B






















Figure 2-3. Sensors experimented temperature > 1094 C. A) Sensor at 1.5 m above the ground.
B) Sensor at the ground level.



















































Figure 2-4. Prescribe burning maps in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M
































Figure 2-5. Fire line created by the firefighter at the border of the stand in Cedar Key Scrub
State Reserve.


































B

















Figure 2-6. Examples of fire front in Cedar Key Scrub State Reserve. A) Back fire B) Head fire.




Figure 2-6. Examples of fire front in Cedar Key Scrub State Reserve. A) Back fire. B) Head fire.


















































i ) ,- ..: _
.1 sB


















__ .. -. ?s.. .
,* "- .... '.'*" '

Figure 2-7. Applying prescribed burning in Cedar Key Scrub State Reserve. A) Fire test in site
2M. B) Fire line at the border of the site 2M.


































B





















Figure 2-8. Taking pictures with a reference person of know height in Cedar Key Scrub State
Reserve. A) Jeff DiMaggio in 5C. B) David Romano in 2M.































IjU- B


Figure 2-9. Effect of prescribed burning in the scrubby flatwoods in site 5C in Cedar Key Scrub
State Reserve. A) Before burning. B) After burning.

































B





















Figure 2-10. Effect of prescribed burning in the scrubby flatwoods in site 2M in Cedar Key
Scrub State Reserve. A) Before burning. B) After burning.









A


LA B



















Figure 2-11. After prescribed burning in Cedar Key Scrub State Reserve. A) Site 5C showing a
burned palm, palmettos, and pine trees. B) Site 2M with burned oaks and palmettos.



































." .B






















Figure 2-12. Burned animals in Cedar Key Scrub State Reserve. A) Terrapene carolina bauri
(eastern box turtle). B) Cockroaches.














20
18
S16
E
14
E 14
12

a 8

0 6

CU 4
2
0


Midflame Wind Speed (Km/h)


- Model 4

-eModrel H5

-Model SHB


1.61 3.62 4.83


MIdflame Wind Speed (Kmlh)



Figure 2-13. Rate of spread comparison between Cedar Key and fuel models 4, sh5, and sh8
according to BehavePlus 3.0.2. A) Site 5C. B) Site 2M.


--Model 4
SModel SH5
--Model SH8
- Cedar Key


3.22


4.83


20.0
18.0
S16.0
S14.0
12.0
10.0
S8.0
6.0
4.0
x 2.0
0.0


i -


-
-




















--Model 4
Model SH5
--Model SH8
- Cedar Key


2.41 3.22 4.83

Midflame Wind Speed (Km/h)


3.62


-- Model 4
- Model SH5
- Model SH8
- Cedar Key


4.83


M idflame Wind Speed (Km/h)



Figure 2-14. Flame length comparison between Cedar Key and fuel models 4, sh5, and sh8
according to BehavePlus 3.0.2. A) Site 5C. B) Site 2M.


6.0

5.0

t 4.0

3.0

- 2.0
a,


CI
S4.0


* 3.0
-J

E 2.0
u-


17


I


-































B



















Figure 2-15. Examples of flame length in Cedar Key Scrub State Reserve. A) Flame length up to
one m height. B) Flame length of at least 6 m height.









CHAPTER 3
RESPONSES OF LONG-UNBURNED SCRUBBY FLATWOODS TO BURNING

Introduction

Florida has had colossal changes that affected the size and the distribution of the scrub

ecosystem. The scrub habitat probably appeared during the early Tertiary Period originating in

the southern Rocky Mountains and northern Mexico with eventual spread along the Gulf coast to

Florida (Axelrod 1958). In Florida, the scrub habitat formed during the early Miocene (20

million years before the present time (mybp)), and it is one of the most ancestral habitats of

Florida in conjunction with the mesic forest (Webb 1990). During the Pliocene epoch, the

Florida peninsula started a process of contraction and expansion of land area because of sea level

rise during glacial and interglacial periods. At one time, Florida was about two times the current

size because of lower sea level and xeric conditions. During the late-Pleistocene, the scrub

vegetation was probably wide-spread across the peninsula. However, during the later part of this

epoch, sea level rose and reduced the size of the Florida Peninsula, and mesic conditions

extended into the landscape (Myers 1990). During the last million years (still the Pleistocene),

broad areas of xeric habitats persisted in Florida, but they were replaced by wet subtropical

mesicc) habitat because of increased precipitation and increased water tables. In this process, the

original widespread and almost continuous scrub habitat was reduced and became fragmented

(Webb 1974, Clark et al. 1999). The fragmentation process increased in the last 5000 7000

years because the climate became more humid, water levels rose, and electrical storms and

lightning fires developed. In addition, natural fragmentation has been coupled with human

perturbations in the last 200 years.

The current distribution of the Florida scrub is a consequence of the historical

biogeography of the Florida peninsula plus the effect of human perturbations. During the post-









European period, the natural fragmented scrub ecosystem was gradually reduced by conversion

to housing developments, citrus groves, and golf courses (Myers 1990). From 1940 to 1981,

approximately 64% of the xeric upland habitat was destroyed and an additional 10% was

disturbed (Peroni and Abrahamson 1985). The scrubs have lost more than 60% of the original

extent, and 85% of the scrubs in Lake Wales Ridge were converted to agriculture, commercial,

or residential development (Peroni and Abrahamson 1985, Christman 1988). The last five

decades of development have reduced considerably the extent of the scrubs (McCoy and

Mushinsky 1994). These perturbations have decreased the number and size of scrub patches and

have increased their isolation. As a result, the original widespread scrub ecosystem has been

fragmented and reduced to patches in the interior and along the coast of the peninsula.

Scrubs in Florida have different ages. Because of the glacial and interglacial periods, at

least six ancient shorelines were created during the 25 mybp, and the scrub habitats along the

central portion of Florida are the oldest (approximately 9 mybp) among all scrub habitats (Myers

1990). So, the current Florida landscape can be described as old scrub habitat on the north -

south axis in the center of the state, and young scrub habitats towards the coastlines

(approximately 0.5-2 mybp). Also, the scrub habitat can be classified as inland or coastal scrub.

Inland scrubs are the largest block of scrubs and occur along a complex of sand ridges running

north-south from Clay and Putnam Counties to Highland County. These sand ridges, dated from

the Miocene to early Pleistocene, form the Florida Central Ridge, and they include: Ocala

National Forest, Lake Wales Ridge (conformed by Arbuckle, Carter Creek, and Archbold) and

Avon Park Air Force Range (Myers 1990, Clark et al. 1999). These ridges have been separated

by several kilometers, and each ridge has an assemblage of scrub patches separated by mesic

habitats and by human development that together act as a matrix for scrub species. Coastal









scrubs are the smallest scrubs found on both the Atlantic and Gulf coasts. The northernmost

examples of these scrubs are located in the Panhandle and are restricted to a narrow strip along

the Gulf coast. They extend from west Ochlockonee River in Franklin County, Florida, to Gulf

Bay State Park in Baldwin County, Alabama. In north-central Florida, coastal scrubs occur on

the east coast in St. John's County near Durbin and on the west coast in Levy County near Cedar

Key. The southernmost scrubs are found on the west coast at Marco Island in Collier County

(probably already extirpated) and on the east coast in Merritt Island National Wildlife Refuge,

Cape Canaveral barrier island complex (Kennedy Space Center), and Jonathan Dickinson State

Park. A high number of endemic species, which are habitat specialists, characterize inland and

coastal scrubs.

Scrub is the most unique and restricted natural ecosystem in Florida. The scrub in Florida

is a shrubland ecosystem located on contemporary or relict beach dune substrates maintained by

recurrent disturbances (Myers 1990, Gibson and Menges 1995, Menges 1999). Scrub

communities are dominated by a well-developed layer of evergreen oaks (shrubs), with or

without a sand pine overstory, sparse ground cover with few herbaceous plants, and many

patches of bare ground occupying well-drained, infertile, sandy soils (Layne 1963, 1990; Myers

1990). Oaks species typically include Quercus geminata, Q. myrtifolia, Q. inopina (inopine oak)

and Q. chapmanii. In addition, other shrubs such as Lyoniaferruginea and Ceratiola ericoides

(false rosemary) are common. Ninety percent of the shrub layer consists of the same six species

in approximately the same order of abundance: Q. myrtifolia, Q. inopina, Serenoa repens, L.

ferruginea, and C. ericoides (Myers 1990). The ground cover is always sparse and includes

species such as Cladonia evansii (deer moss), Licania michauxii (gopher apple), Galactia spp.

(milk peas), and other herbs.









Several types of scrub have been named depending on dominant species, location,

elevation, soil type, fire history, and other factors (Myers 1990, Menge 1999). These scrubs are

rosemary scrub, oak scrub, oak-saw palmetto scrub, sand pine scrub, slash pine scrub, and

scrubby flatwoods. Rosemary scrub is the most common, and it is characterized by the common

species C. ericoides and by gaps supporting an herbaceous flora that includes terrestrial lichens

and many rare species. Oak scrub and oak-saw palmetto scrub are dominated by oaks and the

association oak-saw palmetto, respectively. Sand pine and slash pine scrubs have sand pine and

slash pine, respectively, as representative species in conjunction with other shrub species. The

name scrubby flatwoods is applied to scrubs that either lack a pine overstory or have slash pine

in place of sand pine. Scrub is one of the most endangered communities in Florida not only for

the number, size, and distribution of the patches, but also because of the number of endemic

plant and animal species.

Scrubs have several endemic plant and animal species. Currently, 22 plant species are

federally-listed as endangered or threatened (U.S. Fish and Wildlife Service 1999). Examples of

these are Ilex opaca (scrub holly), Persea humilis (silk bay), Garberia heterophylla (garberia),

Palafoxiafeayi (palafoxia), and Osmanthus megacarpa (wild olive). However, the truly rare

endemic species are restricted to the Lake Wales Ridges such as Hypericum cumulicola (scrub

hypericum), Dicerandrafrutescens (scrub balm), Eryngium cuneifolium (wedge-leave

snakeroot), Lupinus aridorum (Beckner's lupine), and Warea carter (Carter's warea). This

concentration of endemism is probably due to the age of the scrub and its isolation. A very rare

shrub is Ziziphus celata (Garret's ziziphus), collected only twice, not seen since 1955, but

rediscovered in 1987 (Myers 1990). Vertebrate endemic species that occur in the scrub are the

following: Podomysfloridanus (Florida mouse), Aphelocoma caerulecens (Florida scrub jay),









Sceloropus woodi (Florida scrub lizard), Neoseps reynoldsi (sand skink), and a mole skink

species with three subspecies, Eumeces egrerius egrerius (brown red-tailed skink), E. e. lividus

(blue-tailed skink), and E. e. insularis. The structure and stage of the vegetation is very important

for these species and others. For example, if the height of the scrub reaches a critical level, and a

pine canopy develops, then P.floridanus, A. caerulecens, and many bird species leave the patch

of scrub (Myers 1990). In addition, the development of a mature scrub creates habitat for other

species such as Glaucomys volans (flying squirrel), Sciurus carolenensis (gray squirrel),

Ochrotomys nuttalli (golden mouse), Peromyscus gossypinus (cotton mouse), and many species

of birds. Therefore, scrub species need a very specific structure and stage of the vegetation for

habitat, and the only way that scrub communities maintain this status is through fire periodicity.

Scrub is a pyrogenic ecosystem that requires catastrophic fire for self-maintenance. Scrub

fires are devastating, resulting in extensive consumption of the above ground vegetation. The

natural frequency of fires is one every 10-100 years according to Myers (1990) or one every 20-

50 years according to Layne (1990). In the past, the scrub ecosystem had a natural fire frequency

that allowed its maintenance and persistence. In the absence of a natural frequency of fire, tree

and shrub layer density increase and scrub transforms into xeric hardwood forest. However,

when frequency of fire becomes more frequent, sand pine disappears, and the association

becomes oak scrub or changes to high pine (Myers 1990). This was what actually happened

during the pre- and post-European periods. Humans altered the natural frequency and intensity

of fires. As a result, the scrub ecosystem become more fragmented or changed to another type of

vegetation. Currently, prescribed burning is extensively used in Florida, and it has helped to

maintain the scrub ecosystem. Under prescribed fire and natural conditions, fire maintains sand

pine scrub and scrubby flatwoods (Layne 1990, Myers 1990) as stable and non-successional









associations. How plant species respond to prescribed burning has been analyzed by several

studies.

There are two general approaches to describe the effects of prescribed fire on flora. The

first one takes into consideration the concept of fire regimen in order to understand fire effect at

the community level. The second one focuses on the response at the species level. Let us starts

with the definition of fire regimen.

The concept of fire regimen encompasses several components. Kilgore (1987) defines fire

regimen as a set of several factors such as fire frequency (time between fires), season of burn,

fire periodicity, fire intensity, size of fire, pattern on the landscape, and depth of burn. Brown

and Smith's (2000) definition includes pattern of fire occurrence, size, uniformity, and severity.

Whelan (1995) considers fire regimen as a global concept to summarize fire frequency, season of

burning, type of fire (only organic layer soil, only above ground, or crown fire) and extent of the

fire (continuous vs patchy). This dissertation synthesizes the three previous concepts and defines

fire regimen as a global concept to summarize fire frequency, intensity, severity, type of fire,

size, season, and extent. Fire regimen has been used in order to categorize plant community

responses to fire.

The most recent classification of plant community responses to fire uses fire severity as the

main criterion. Brown and Smith (2000) used a fire regimen classification relying only on fire

severity. According to those authors, the use of fire severity as the key component to describe

fire regimen is interesting because it connects directly to the effects of disturbance, especially on

survival and structure of the dominant vegetation. Brown and Smith's (2000) classification is as

follows:









Understory fire regime (applies to forest and woodland vegetation). Approximately 80%
or more of the aboveground dominant vegetation survive fire. Fire is not lethal for
dominant vegetation and does not change its structure.

Stand-replacement regime (applies to forests, woodlands, shrublands, and grasslands).
Approximately 80% or more of the aboveground dominant vegetation is either consumed
or killed by fires. Since fire consumed or killed the aboveground parts of the dominant
vegetation, they dramatically change the structure of it as well, particularly in shrublands
and forests.

Mixed severity regime (applies to forests and woodlands). Fire selectively kills species of
the dominant vegetation depending on the species' susceptibility to fire. This type of fire
varies between understory and stand-replacement.

Nonfire regime. Little or no occurrence of natural fire. All type of forest can be classified
according to the categories above that correspond to low, medium, and high fire severity
types.

The second general approach about how plant species respond to prescribed burning takes

into consideration how species survive fire. Whelan (1995) has named four categories as follows:

fire ephemerals, obligate seeders, sprouters, and facultative sprouters. The first category

describes plants that do not survive the fire. The second category refers to plants that germinate

after fire through a seed bank stored in the soil or in the canopy. The third category presents

plant species that survive fire through protected buds in the stems or roots. The fourth category

introduces species in which the ability to sprout would depend on the characteristics of the

prescribed fire. However, recovery mode of individual species in Florida scrubs may be

correlated with habitat characteristics and fire regime.

The scrub ecosystem in Florida falls under the category of a stand-replacement regime and

the majority of the species are resprouters. After a long fire-free period of fuel accumulation, a

high intensity fire takes place. If sand pine or slash pines are present, they might be killed. The

above-ground shrub layer is totally consumed. According to Menges and Kohfeldt (1995),

species recovery varies, and they classified 95 species of the scrubby flatwoods and rosemary

scrub into seven guilds of recovery mechanism: 24 species were resprouters (many woody

71









shrubs), 24 were resprouters and seeders (small-statured shrubs, palmettos, and herbaceous

perennials), 14 were resprouters and clonal spreaders (the majority of the dominant shrub genera

in scrubby flatwoods such as Quercus, Lyonia, and Vaccinium), five were resprouters, clonal

spreaders, and seeders (herbs), 26 were obligate seeders (C. ericoides and many herbs), one

aerial seeder (Pinus clausa), and one seeder and survivor (Pinus elliottii). Shrub species sprout

from previously suppressed underground buds on buried roots. Few shrub species, for instance

C. ericoides, regenerate from seeds stored in the soil. P. clausa regenerates from fire-induced

seed release from individuals with serotinous closed cones. P. elliottii is the only pine tree that

might survive moderate or high-intensity fire and also recovers by seeds. Both species might

reseed from other pine trees in adjacent stands. Post-fire species composition is usually an

assemblage of many of the species previously growing on the site. However, there are some

variations in the way Florida scrubs recover after fire.

Florida inland scrubs, such as scrubby flatwoods and rosemary scrubs, have different

recovery strategies (Menge and Kohfeldt 1995). Scrubby flatwoods specialists usually depend on

vegetative recovery modes (61%; resprouting and clonal spread) and less often on mixed modes

(23%) or obligate seeding (16%). In general, specialist species and dominant shrubs spread

clonally and resprout in the scrubby flatwoods. In contrast, half (50%) of the of rosemary scrub

specialists are obligate seeders, 17% were mixed, and 33% were vegetative. Also, the dominant

species of C. ericoides are mainly obligate seeders. Species found in both habitats are

intermediate in recovery modes (32% vegetative, 36% mixed, 32% obligate seeders). Therefore,

in general, scrubby flatwoods appear to be more favorable for post-fire resprouting and clonal

growth and rosemary scrub more favorable for post-fire seedling recovery. This difference

occurs despite the overlap in species composition (Abrahamson et al. 1984) and close









concurrence of these communities in the landscape. Coexistence of resprouters, seeders, and

species with mixed recovery modes in Florida inland scrubs suggests that fire-return intervals

may be quite variable.

Post-fire recovery of Florida inland scrub ecosystem varies with dominant shrub species

and fire history. Most scrubby flatwoods species recover by resprouting and/or clonal spread

oaks, recovery is rapid, and there is little change in species composition at a scale of four to 10

years (Abrahamson 1984a, 1984b, Johnson and Abrahamson 1990, Abrahamson and

Abrahamson 1996b). In contrast, recovery of rosemary scrub takes more time because C.

ericoides and P. clausa (the dominant species) recover through post-fire seedling establishment,

and C. ericoides takes a decade to reach sexual maturity (Johnson 1982, Johnson et al. 1986).

Fire return intervals for scrubby flatwoods vary from 5 to 20 years (Menges & Kohfeldt 1995)

and to 20 to 80 years for rosemary scrubs (Myers 1990, Menges 1999).

Post-fire recovery of Florida coastal scrub also varies with dominant species and fire

history. The majority of plant studies conducted in coastal scrub have been carried out on Merritt

Island National Wildlife Refuge and Cape Canaveral (Simon 1986, Breininger and Schmalzer

1990, Schmalzer and Hinkle 1991, 1992a, 1992b, Schmalzer and Boyle 1998, Schmalzer and

Adrian 2001, Schmalzer 2003, Schmalzer et al. 2003). Myers (1990) classified this scrub as a

coastal scrub; Weekly and Menges (2003) referred to it as a coastal oak-palmetto scrub. However,

according to Schmalzer et al. (2003), the most common types of plant communities in Merritt

Island and Cape Canaveral are oak-saw palmetto scrub, scrubby flatwoods, and coastal scrub.

Schmalzer (2003) classified the scrubby flatwoods without slash pine overstory as oak-saw

palmetto scrub. Dominant species in these two communities were: myrtle oak, sand live oak,

Chapman oak, saw palmetto, and ericaceous shrubs such as L.ferruginea. Schmalzer et al.









(2003) considered coastal scrub a different type of community because it was dominated by

Quercus virginiana (Live Oak), Serenoa repens, and ericaceous shrub species were absent. The

type of soils was also different. Oak-saw palmetto scrub was on soils that varied from neutral to

acid, but the majority was acid soils. Coastal scrub soils were alkaline. In oak-saw palmetto

scrub, recovery of dominant oaks and ericaceous species after fire is primarily through

resprouting and clonal spread (Schmalzer and Hinkle 1992a, 1992b). Resprouting allowed a

rapid reestablishment of the dominant shrubs. Saw-palmetto reestablished cover faster than

woody shrubs. Saw palmetto cover equaled preburn values between one and 1.5 year postburn

and changed little after that. Q. myrtifolia, Q. geminata, and Q. chapmanii recovered rapidly

after burning but at different rates. Cover in these three species equaled preburn values between

4 and 5 years postburn and changed little by 10 years postburn. Few changes occur through 10

years post-fire except for continued height growth. In coastal scrub that received

cutting/prescribed burning treatments, cover of saw palmetto was reduced by mechanical

treatment. Recovery of Q. virginiana was also through resprouting, which reestablished cover

within 5 years postburn (Schmalzer et al. 2003). Growth rate of Q. virginiana was higher than

shrubs in oak-saw palmetto scrub under the same treatment. The rapid growth rate of these types

of scrubs has suggested that prescribed burning would need to be more frequent than often

applied in these communities in order to maintain the desired structure of the vegetation. Fire

return interval for oak-saw palmetto scrub has been estimated between 5 and 20 years

(Schmalzer 2003). However, the length of the restoration period needs to be determined

(Schmalzer et al 2003). These types of studies are also needed on the Gulf coastal scrub.

Even though some research has been conducted on the Atlantic coastal scrub, only one

study has been carried out in the Gulf coastal scrub, particularly in the Panhandle. Ruth et al.









(2007) studied the effect of reintroduction of fire in long-unburned coastal scrub in Naval Live

Oaks areas of the Gulf Islands National Seashore. This study addressed the effect of

environmental variables and fire on plant distribution and abundance, and it found that elevation

and time since fire were the most important environmental variables that affected species

distribution and abundance. No study has been carried out on the west coast of Florida's

peninsula. This is particularly important in CKSSR because prescribed burning has been

practiced since 1985. Understanding how the scrub ecosystem in CKSSR responds to prescribed

burning is critical to know the direction and rates of changes in composition and structure of

scrub communities after fire and to make effective management decisions. This is the first study

carried out to determine the responses of a long-unburned (since 1955) scrubby flatwoods to

prescribe burning in the west coastal scrub of Florida.

Objective

Prescribed burning was applied to two long-unbumed scrubby flatwoods sites in order to

study the post-bum dynamics of this community. The objective was to document recovery modes

and structural and compositional changes in the post-burn community. To achieve this objective,

a site analysis was needed to determine if treatment and control sites were ecologically similar

before burning.

Methodology

Vegetation sampling was carried out in control sites one time before prescribed burning.

Sampling in treatment sites was conducted as follows: pre-bum and post-burn at 11 days, 3, 6, 9,

and 12 months. I sampled the density of vegetation by placing a quadrat (4 m2) (Figure 3-1) on

specific points of the trapping grid selected by a stratified random sampling. Cover was

quantified along a 2 m line that intercepted the center point of two opposite sides of each

quadrat. This sampling measured the following predictor variables:

75









1. Ground cover of bare ground, litter, hard woody debris, and herbaceous species measured in
cm.

2. Shrub cover (< 1.5 m tall) per species was quantified in cm.

3. Number of palmetto plants in each quadrat.

4. Number of individuals per species of trees, saplings, and seedlings, which were classified
with the following criteria: oak and pine trees with dbh > 7.6 cm, saplings with dbh < 7.6 cm
and height > 1.0 m, and seedlings with height < 1.0 m.

5. Maximum vegetation height in each quadrat.

Flowering and fruiting were not systematically surveyed, but they were recorded as

encountered.

Before burning, the center of each quadrat was mapped with a Global Positioning System

GPSmap 76S (Garmin) fitted with an external antenna (1-2 m accuracy). In addition, I marked

the center each quadrat and the two of its north facing corners with a 20 cm wire. These wires

helped to locate the quadrat after burning, and they were replaced with flags after the burn

(Figure 3-1).

Counting of individual plants was done per stratum. However, there were species that

could be classified as a seedling or sprout, but they could not be classified as sapling or tree

because they were herbs, a lichen, a palmetto, a cactus or woody species of low height. Examples

of these woody species were as follows: Vaccinium myrsinites, C. ericoides, Osmanthus

americanus, Gaylussacia dumosa, Gaylussacia nana, Licania michauxii, and Rhus copallinum.

In this case, the abundance per species was recorded without establishing an association with a

particular stratum.

Counting ramets started in sites 5C and 2M at 11 and 12 days after burning, respectively

(Figures 3-2, 3-3, and 3-4). Since species identification is difficult at this stage, groups of ramets









were labeled and recorded with pictures. Later, these species were identified when ramets

matured.

Counting of ramets was done carefully to minimize human error. Counting was carried out

twice to make sure that the recorded number was accurate. If the numbers did not match, I

counted slowly the third time. The two numbers that matched were selected. All ramets were

counted even though they belonged to the same stem (Figure 3-5).

Recovery mode was determined by excavation of ramets. In each quadrat, I excavated 10

ramets per species in order to find seedlings or resprouting individuals at 11 days and 3 months

after burning. Ramets were carefully excavated using hand tools and fingers, with care taken to

preserve root systems and rhizome connections. I followed the technique used by Menge and

Kohfeldt (1995) to classify seedling, resprout, or clonal ramet. Seedlings were independent

plants with small root systems and no sign of pre-fire biomass (e.g. no charred stem bases).

Resprouts were classified as ramets resprouting within 20 cm of pre-fire stems. Ramets more

than 50 cm away from pre-fire ramets were named clonal ramets.

Univariate and multivariate data analyses were performed using SAS 9.1.3 (SAS Institute

Inc. 2002-2003) and PC-ORD v.5.0 (McCune and Mefford 1999), respectively. The significance

level chosen was 0.05. The data set was summarized by computing absolute and relative

abundance, density, frequency (number of occupied quadrats), and mean percent cover (total

distance intercepted above, below, or touching by species divided by 2 m and multiplied by 100)

per species. Importance values were quantified by summing relative values for density,

frequency, and mean percent cover for each species with the exception of pine trees because they

did not have mean percent cover records.









A site analysis was carried out to determine if treatment and control sites at preburn

conditions did not differ regarding plant species structure and composition. This analysis was

critical for the experimental design because otherwise treatment effect could not be verified. This

analysis started by quantifying species richness (as number of species), diversity (Simpson's

index 1/D and Shannon-Wiener's index H'), and evenness (as J = H'/lnS, Pielou 1969) to

measure structural and compositional differences among sites. Jaccard's and Sorensen's

coefficients of similarity were also calculated to analyze the species composition among sites.

After that, a cluster analysis was performed to study whether there were differences or not

among sites by taking into consideration absolute abundance and mean percent cover for all herb

and woody species. The dataset was standardized and outliers were deleted (McGarigal et al.

2000). In the cluster analysis, the Euclidean distance was used for the resembling matrix and

Median linkage, Average linkage, and Ward minimum-variance linkage were used as fusion

methods. In addition, to decide the number of significant clusters to retain, a F-ratio test in

combination with Duncan's test were carried out to assess the null hypothesis that the mean for

each variable was not different between multispecies clusters. Finally, a mean or median

comparison was done to determine significant differences among sites for the variables found

with significant differences in the F-ratio test. The idea was to determine which site(s) was

(were) significantly different from other sites regarding that particular variable. Since plant

species variables did not have a normal distribution according to the Shapiro-Wilk test, the data

set for abundance and mean percent cover was transformed using the Arcsin function. Then, an

ANOVA test and multiple comparisons (Duncan's and Bonferroni's procedures) were carried out

to determine if at least two means were significantly different and which means were

significantly different, respectively. If the Arcsin transformed variable did not have a normal









distribution, medians were compared by using Kruskal-Wallis test and Duncan's multiple

comparison procedure.

Species richness, evenness, and diversity (Simpson's and Shannon-Wiener's index) were

quantified to measure and document structural and compositional changes between the pre- and

the postburn community. Detrended Correspondence Analysis (DCA) was used to visualize the

multivariate changes in woody species densities and mean percent cover in the preburn and

postburn samples over time. Ordinations were carried out on absolute (to highlight structural

changes) and relativized (to emphasize compositional changes) values of densities and mean

percent cover data by using PC-ORD. Absolute values were relativized using standardization by

the norm (Greig-Smith 1983). The quality of the ordinations was evaluated with the coefficient

of determination that measured the proportion of the variance represented by the ordination axes.

Scatter-plots between the first two axes were done to visualize structural and compositional

changes. Only sampling time scores, and not species scores, were plotted for this reason. Finally,

a Multi-response Permutation Procedure (MRPP) and multiple comparison was conducted using

PC-ORD. MRPP is the non-parametric test analogue of MANOVA. Unfortunately, MRPP could

only be performed on treatment sites pre- and 12 months postburn and control sites. MRPP can

only be carried out on independent samples. Therefore, it is not the right test for a sequence of

sampling through time on the same quadrats. Multi-response Block Procedure (MRBP) is a

variant of MRPP, and it can be used for dependent samples. However, it requires a second matrix

in PC-ORD and a balanced design (McCune and Grace 2002). The second matrix in PC-ORD is

used with environmental variables that were not measured in this study, and sites had different

sample sizes. Thus, MRPP was carried out to test the hypothesis of no treatment effect or no

significant structural differences between treatments at 12 months postburn and treatments at









preburn levels and control sites. Only the absolute density and mean percent cover of woody

species were used for this analysis because the herb data set did not have a large enough sample

size. Euclidean and Sorensen distances were used to calculate the distance matrices.

Results

Species List and Recovery Modes

Table 3-1 illustrates the list of species recorded in quadrats in treatment and control sites.

A total of 10 herb species, 26 woody species, a lichen, and a cactus was recorded during the

study. All species resprouted after burning. Even though digging was carried out on 10

individuals per species per quadrat, I did not find evidence of recovery by seeds. The only

exception was Pinus clausa with seedlings at six months postburn in 5C. Also, the criterion of

ramets more than 50 cm away from pre-fire ramets to name clonal ramets did not work.

Therefore, I did not consider this recovery mode. Pre-fire ramets were burned completely in

almost all quadrats.

A Site Analysis

A comparison of the structure and composition of the four sites is displayed by Table 3-2.

As can be seen, control (5A and 5D) and treatment (5C and 2M) sites differed in species

richness, species diversity, and evenness. In general, control sites had higher species richness,

species diversity, and evenness than treatment sites. The only exception is 5D and 2M sites that

had the same Shannon-Wiener's index and a similar evenness. Therefore, control and treatment

sites were not ecologically similar by using these criteria.

Figure 3-6 reveals Jaccard's and Sorensen's similarity coefficients among the four study

sites. All sites had a similarity higher than 54% and 63% according to Jaccard's and Sorensen's

coefficients, respectively. The only exception was Jaccard's coefficient between 5C-preburn and









5D (46%). Consequently, these communities shared some structural similarities according to the

coefficients.

A more powerful criterion was needed to determine if control and treatment sites were

ecologically similar. Species richness, species diversity, evenness, and similarity coefficients

drew different results. These criteria used the proportion of individuals and the number of

species as data in a single dimension. A multivariate approach such as cluster analysis gives

more insight about the actual similarity among sites.

Figures 3-7 through 3-12 show the results of the cluster analysis. Median linkage/Average

linkage and Ward's minimum-variance linkage clearly displayed one cluster (Figures 3-7

through 3-10) and two clusters (Figures 3-11 and 3-12), respectively. However, looking at

Ward's dendrogram, clusters were composed of a mix of sample units corresponding to

treatment and control sites. As a result, I could not state that one cluster corresponds to control

sites and the other to treatment sites. In order to decide the number of significant clusters to

retain, Table 3-3 presents the results of the F-ratio test and Duncan's test. Out of 52 variables

tested, 23 (44%) did not show results because of the small sample sizes; the means of 21 (41%)

variables were not significantly different between the two clusters, and the means of eight (15%)

variables were significantly different. Of these eight variables, only the mean abundances of S.

repens and V. myrsinites had a significant result when means were compared among treatment

and control sites (Table 3-4). The multiple comparison procedure revealed that only the median

of the abundance of V. myrsinites in 2M preburn was significantly different from 5D and 5C

preburn according to the Duncan's test. Therefore, there was enough evidence to suggest that

treatment and control sites were ecologically similar, and prescribed burning effect could be

determined by comparing treatment and control sites.









Postburn Recovery and Survival

Figure 3-14 shows absolute mean percent cover ofbareground, litter, and debris in 5C and

2M. Bareground had postburn values higher in 2M than in 5C, but these values were not higher

than 13 %. Even though the preburn value for litter was lower in 5C (83.9%) than in 2M (94.5%)

and the postburn values were higher in 5C than in 2M for 3, 6, and 9 months, litter had almost

exactly the same mean percent cover in both sites (5C 68.3 %; 2M 67.5%) at 12 months

postburn. However, these values were lower than preburn mean percent cover in 5C, 2M, and

control sites. Debris had a similar curve pattern in both sites with very low values after burning.

Preburn and postburn vegetation height in 5C and 2M is presented in Figure 3-15. As

revealed by the graph, preburn height in 5C was higher than in 2M and postburn heights were

similar. The vegetation took 6 months to reach one m tall, and the height remained constant until

12 months.

Tables 3-5 through 3-12 provide absolute densities, frequencies, mean percent cover, and

importance values of 10 herb species in 5C and 2M. In both sites, the Family Poaceae had the

highest importance value, and it was represented by several grass species. Galactia elliottii,

Solidago odora, and Galactia mollis had the next three highest importance values, and

Crotalaria rotundifolia and Woodwardia virginica were only recorded in 5C and 2M,

respectively. The species above were counted during the preburn and/or postburn sampling

periods, and the rest of the species in Tables 3-5 through 3-12 were only found during the

preburn sampling period.

The postburn recovery of the most common herb species in 5C and 2M is illustrated in

Figures 3-16 through 3-19. Poaceae had an upward trend for density, frequency, and importance

value in 5C. Cover also had an increasing trend, but declined at nine months and increased again

at 12 months. In 2M and for all variables, Poaceae leveled off from 3 to 9 months and increased

82









one more time at 12 months. In general, Poaceae was the only taxon with preburn values and had

higher values in 5C than in 2M. After burning, the behavior of the curves for G. elliottii was

alike for all variables in both sites. G. elliottii increased at 3 months, decline at 6 or 9 months,

and then increased again at 12 months. In general, G. elliottii had higher densities, frequencies,

cover, and importance values in 2M than in 5C, and it surpassed density control value (5D) in 5C

and both control values in 2M. S. odora, G. mollis, C. rotundifolia, and W. virginica had the

same pattern for all variables in both sites. These species had low values for all variables until

the 9 months and increased at 12 months, with this increase higher in 5C than in 2M. Besides G.

elliottii, G. mollis was the only herb species that reached density control value (5D) at 12 months

postburn in 5C.

Tables 3-5 through 3-12 display absolute densities, frequencies, mean percent cover, and

importance values of 26 woody species in 5C and 2M. These tables reveal that the seven species

with high values for all variables in both sites were the following: Quercus myrtifolia, Serenoa

repens, Quercus geminata, Lyoniaferruginea, Lyonia lucida, Quercus chapmanii, and

Vaccinium myrsinites. Ilex glabra and Gaylussacia nana were also common in 5C and 2M,

respectively. Rare species were as follows: Quercus nigra, Quercus sp., Ceratolia ericoides,

Osmanthus americanus, Salix caroliniana, Smilax sp, and Opuntia humifusa. Pinus clausa, P.

elliottii, and P. palustris were present in 5C and the last two species in 2M, but few trees were

recorded in quadrats. The rest of the species had moderate values in density, frequency, mean

percent cover, and importance value in both sites.

The postburn absolute density recovery of the most common woody species in 5C and 2M

is shown by Figure 3-20. Q. myrtifolia had the fastest recovery in both sites, with densities in 5C

higher than in 2M. The rest of the species, S. repens, Q. geminata, L.ferruginea, L. lucida, Q.









chapmanii, and V. myrsinites had similar recovery patterns in both sites, with densities slightly

higher in 2M than in 5C. However, these species did not have densities higher than 15

individuals/m2 in both sites. Q. myrtifolia recuperated over preburn and control sites values. The

other species achieved preburn values at 3 months, with the exception of V. myrsinites, and they

also had equal or higher absolute densities than control sites.

Figure 3-21 reveals that the most common species had similar patterns of absolute

frequency after burning in 5C and 2M. All species increased frequency at 3 months and then

leveled off until 12 months, with the exception of V. myrsinites that continued increasing after 3

months. The only difference was that values were higher in 2M than in 5C. All species had

frequencies equal or higher than preburn values at 3 months with the exception of L. ferruginea

and L. lucida in 5C, and these two species and Q. myrtifolia and V. myrsinites in 2M. However,

these species achieved at least a control site frequency at 3 months or later during the next 9

months.

Absolute mean percent cover for the most common species in 5C and 2M is presented in

Figure 3-22. As can be seen, all species had a very similar pattern in both sites. Q. myrtifolia and

S. repens had the highest cover registered in both sites. Q. myrtifolia increased cover at 3 months

in both sites, leveled off until 9 months and grew until 12 months in 5C, and decreased at 6

months and increased until 12 months in 2M. S. repens had a sharp rise until 6 months, and then

it tended to level out at 9 and 12 months in both sites. The other five species had an increased no

higher than 5% at 3 months and remained relatively constant until 12 months, with the exception

of Q. geminata in 5C and this species and L.ferruginea in 2M. These last two species in their

respective sites increased frequency at 12 months with values higher than 5%. Q. geminata, Q.

chapmanii, L. ferruginea, and V. myrsinites, and G. nana recovered preburn and/or control cover









at 12 months in 2M, and Q. myrtifolia, Q. geminata, and V. myrsinites in 5C (Tables 3-11 and 3-

7).

Figure 3-23 provides the postburn importance values of the most common species in 5C

and 2M. In general, Q. myrtifolia had importance values higher in 5C than in 2M. It increased in

its importance value at 3 months and later leveled off until 12 months in 5C. It had its highest

value at 3 and 6 months, and then decreased until 12 months in 2M. S. repens had similar

postburn recovery in both sites. It had the highest importance value at 11 and 12 days in 5C and

2M, respectively, due to their high resprout and frequency. Then, it decreased at 3 months,

increased until 9 months, and declined one more time at 12 months. Q. geminata also had a

similar recovery pattern in both sites. The importance value increased at 3 months, and then

remained almost constant until 12 months. L. ferruginea was relatively constant in 5C, and it

increased at 6 months and leveled out after that in 2M. L. lucida kept almost the same

importance value after burning in both sites, being a little bit higher in 5C. Q. chapmanii

resprouted at a higher level in 2M than in 5C, and the importance value at 12 days was

considerably higher in 2M than in 5C for this reason. However, Q. chapmanii's importance

values were very similar in both sites after that with the tendency of declining. The importance

value of V. myrsinites was relatively constant in 5C after burning, and remained steadily

increasing in 2M, but at low importance values. Q. myrtifolia, S. repens, Q. geminata, and Q.

chapmanii had importance values equal or higher than preburn and control sites values at 3

months in both sites. L.ferruginea achieved control site values during the lapse of 12 months,

but not the preburn value in 5C. However, it had a higher importance value at 6 months than

preburn/control sites in 2M. L. lucida did not reach preburn or control importance values in 5C









and 2M. V. myrsinites recovered its preburn value at 6 months, but not control site values in 5C.

It obtained control site values between 6 and 12 months, but not the preburn value in 2M.

The postburn recovery pattern described above belonged to all individuals recorded in

quadrats. This included ramets, new saplings, and trees that survived fire. If I consider the

calculation of the variables for ramets only, absolute density for ramets were lower than absolute

density for individuals in both sites. Tables 3-13 and 3-14 show ramet absolute densities for herb

and woody species in 5C and 2M and Figures 3-24 and 3-25 illustrate recovery patterns for the

most common herb and woody species in 5C and 2M, respectively. As revealed by the graphs,

recovery patterns for herb and woody species followed exactly the same pattern as for absolute

density for individuals in both sites.

Tree mortality in both quadrats and grids is presented by Table 3-15. Tree mortality was

low in quadrats at both sites (7.5%) because of the adaptation of Quercus spp. and Lyonia

ferruginea to fire. These trees had 80-100% burned stem, but roots were alive. Therefore, all

roots that remained alive resprouted after prescribed burning. Census on pine trees carried out in

the grids before and after prescribed burning suggested that fire intensity was high enough to kill

91.6% of pine trees, including the fire adapted P. clausa and the fire resistant P. palustris.

Structural and Compositional Changes in Response to Prescribed Burning

Figure 3-26 provides preburn and postburn species richness in treatment sites in

comparison with control sites. Preburn species richness was lower in 5C than in 2M. Then,

species richness decreased immediately following burning in 5C and was stable in 2M. After

that, species richness in 5C increased (surpassing its preburn level) and reached species richness

in 5D, and obtained the highest species richness in 2M at 3 months. Later, species richness in

both sites fluctuated exactly with the same pattern, being higher in 5C than in 2M. In general,

species richness in 5C was higher than its preburn value and species richness in 2M after 3

86









months. In addition, species richness in 5C reached 5D control site value after burning, but

species richness in 2M did not attain control site values.

Figures 3-27 and 3-28 indicate that species diversity and evenness, respectively, were

almost constant in both sites after prescribed burning. Simpson's indexes pre- and postburn

values were higher in 2M than in 5C. At both sites, Simpson's index originally decreased just

after burning; then it was almost constant from several days to 9 months after burning, but

increased at 12 months. Preburn species diversity was achieved in both sites at 12 months, and

only 5D control value was surpassed in 2M. Shannon-Wiener's indices had the same pattern as

Simpson's indices in both sites. Preburn values were reached at three months in 5C and at 12

months in 2M. Again, only 5D control index was surpassed in 2M. Evenness followed exactly

the same pattern as species diversity with moderate values indicating modest predominance of

the common species. Figures 3-27 and 3-28 suggest small structural and compositional changes

in treatment sites until 12 months postburn.

The results of the Detrended Correspondence Analysis are shown in Table 3-16. In

general, the r2 coefficient of determination was high in both treatment sites. This indicated that

the analysis provided ordinations of good quality, and a high proportion of the variance was

explained by the axes. However, absolute densities had higher r2 than relativized densities, and

relativized cover had higher r2 than absolute cover in both sites. In addition, site 5C had lower r2

than 2M for densities and higher r2 than 2M for cover. Except for absolute cover in 2M, the first

axis had higher r2 than the second and third axis in both sites. The scatter-plots illustrate DCA,

but it is important to explain what distances represent in this ordination space.

Understanding the distance among sampling times in the ordination space is key for the

interpretation of structural and compositional change of the postburn community. Each point in









the species/sampling time ordination space represents the site's position on the first two axes of

the ordination at a given time prior to or after fire. According to Schmalzer and Hinkle (1992a),

the distance between sampling time scores (points) is an index of similarity. Sampling times that

occur together are similar. Also, the distance between pre- and postburn sampling times reveal

vegetation change after fire and recovery. In other words, the lengths of vectors between pre- and

postbum times in ordination space indicate the vegetation change during the recovery process.

Hence, the distance between sampling times (vector lengths) is an index of change and recovery

(Schmalzer and Hinkle 1992a). These concepts are important to understand structural and

compositional change in the scrubby flatwoods in CKSSR.

Figures 3-29 through 3-32 display absolute and relative densities and mean percent cover

scores of plant species during preburn and postburn sampling times. Treatment sites 5C and 2M

had structural and compositional changes through time, but they had in common the following

changes: (a) a structural and compositional change after prescribed burning between preburn and

three months postburn in both absolute and relativized density and cover, and (b) a structural and

compositional changes between three and six months shown by the absolute and relativized

cover in 5C and by absolute cover and relativized density in 2M. Treatments sites 5C and 2M

differed in changes between six and 12 months. Site 5C had moderate structural and

compositional changes in both absolute and relativized density and cover, but 2M had more

compositional than structural changes in both variables. Now, are structural changes significant?

The results of the MRPP are shown by Table 3-17. As can be seen in this table, the test

statistic T was significant at the 0.05 level for both absolute densities and mean percent cover. T

described the separation among groups. The more negative is T, the stronger the separation.

Therefore, at least two sites were significantly different regarding density or cover, and the









multiple comparison revealed them. Absolute densities in treatment sites (5C preburn, 5C-12

months postburn, 2M prebum, and 2M-12 months postbum) and control sites (5A and 5D) were

significantly different by using both Euclidean and Sorensen distances (Table 3-18). Therefore,

prescribed fire did have a significant effect on absolute densities on treatment sites by changing

the structure of the community between pre- and 12 months postburn. Absolute cover was only

significant between control site 5D and treatment sites 5C and 2M at 12 months postburn by

using Euclidean and Sorensen distances. Hence, prescribed fire did not have a significant effect

on changing absolute mean percent cover between 5A and 5C/2M 12 months postburn and

between preburn and 12 months postburn mean percent cover of treatment sites. This is due to

the fact that prebum and control mean percent cover values were achieved by 12 months in the

majority of the species (see Table 3-7 and 3-11).

Flowering and Fruiting after Prescribe Burning

Flowering and fruiting season were not synchronous in CKSSR within and among species.

V. myrsinites started to have flowers in March and fruits in April 2006 in some sites. G. nana and

S. repens started to have flowers in April, but G. nana had fruits in May and S. repens in June

2006. Other species such as L. ferruginea, L. lucida, Ilex glabra, and Brevaria racemosa started

to flower in April-May and fruits in June-July 2006.

Discussion

True Control Sites

This is the first site analysis carried out to determine the veracity of control sites. In the

reviewed literature about the effects of prescribed fire, control sites have been taken very loosely

and without scientific rigor. In general, control sites have been assigned at random or not at

random as an assumption and without scientific validation. The assumption has been made and

accepted by the majority of the scientific community of fire ecology including editors of

89









prestigious scientific journals. This study suggests that a site analysis should be the starting point

for future research.

Cluster analysis in combination with ordination and univariate techniques can be used to

determine if treatment and control sites are ecologically similar. Species richness, species

diversity, evenness, and similarity coefficients produced contradictory results. The cluster

analysis (by using three fusion techniques) in combination with discriminant analysis (to plot the

first two pairs of canonical variates) and the F-ratio test (with Duncan's and Bonferroni's

multiple comparison procedures) helped to conclude that treatment and control sites were

ecologically similar. The reason for using three fusion methods relied on the purpose of

conducting a site analysis: to determine if treatment and control sites were ecologically similar

by carrying out a cluster analysis that really represented the structure of the data. Two space-

conservative methods (Median and Average linkage) and one space-distorting method (Ward's

minimum-variance linkage) were applied to the data for this reason. According to McGarigal et

al. (2000), space-conservative methods are the best choice when the objective is to reveal the

true structure of the data, which is usually the case in most ecological research. Space-distorting

methods do not truly represent the spatial relationship of the data because they contract or dilate

the space in the immediate vicinity of the groups. A comparison between space-conservative and

space-distorting methods was done to be more objective in the decision. Since, Median/Average

linkage and Ward's minimum variance linkage found one and two clusters, respectively, the F-

ratio test with the multiple comparison definitively helped to determine if the structure of the

dataset fit to one cluster or not. Having true control sites is critical in fire ecology research.

A site analysis must be done a priori and not a posteriori. Researchers in fire ecology need

to be careful selecting control sites and test them before applying treatments to plots. Before and









during this process, park managers' advice and involvement are desirable due to their

experience. The success of selecting good control sites will rely on park managers' and

researchers' judgment, preliminary sampling, and the statistical analysis conducted a priori.

Fire Survival and Recovery Modes

Resprouting was the main mechanism of survival and recovery with fire by the majority of

the species in CKSSR. All species that occurred before burning in CKSSR exhibit resprouting

life histories, except P. clausa, P. elliottii, P. palustris, and C. ericoides. Post-fire recovery

through sprouting has been documented in scrubby flatwoods and sand pine scrub in Archbold

Biological Station (Abrahamson 1984a, 1984b, Abrahamson and Abrahamson 1996a, 1996b), in

oak-saw palmetto scrub in Kennedy Space Center (Schmalzer 2003, Schmalzer and Hinkle

1992a, 1992b, Schmalzer et al. 2003) and in sand pine scrub in National Seashore in the

Panhandle (Ruth et al 2007). However, there are some species that recovery through seeding and

clonal spread. For instance, S. repens has been reported as resprouter and seeder, Quercus spp.

Lyonia spp., and Smilax auriculata have been cited as resprouters and clonal spreaders (Menges

and Kohfeldt 1995, Menges and Hawkes 1998 see Table 3-1).

Resprouting from dormant buds, rhizome tips, root crowns, and protected meristems can

account for a substantial proportion of postfire recruitment (Lyon and Stickney 1976). It is clear

that almost all species (except C. ericoides) tolerated fire in CKSSR, and soil was a good

insulator that protected underground roots and meristematic tissues as well. Tissues deeper than

5 cm rarely experience significant increase in temperature (Whelan 1995). But, did soil protect

the seed bank already established during the spring 2005?

Seeds were probably not buried deep enough in the soil by the time of prescribed burning

in CKSSR. Prescribed burning took place at the end of the reproductive and growing season.

This was the right moment biologically and environmentally because the majority of the species

91









already produced seeds and weather characteristics were appropriate to assure a catastrophic fire.

However, fire intensity was high enough to top-kill all aboveground vegetation in treatment sites

and the seed bank. No seedlings were found until 6 months postburn, and these seedlings

belonged to P. clausa. Seeds experienced temperatures higher than 1000 C in treatment sites,

and most likely they were lying on the leaf litter and thus they were consumed. Seeds must be

protected from direct heat to survive fire.

Speed of Recovery Process

Postfire changes in CKSSR were described at a very short time interval in comparison with

the literature. The results obtained in CKSSR correspond to only one year postfire. Research

carried out in oak-saw palmetto scrub (stands 2, 4, 8, and 24 years since burning) at Kennedy

Space Center/Merritt Island National Wildlife Refuge (KSC; Schmalzer and Hinkle 1992a,

1992b, Schmalzer 2003), in long-unburned scrubby flatwoods (>35 years) at Archbold

Biological Station (ABS; Abrahamson and Abrahamson 1996a), and long-unburned sand pine

scrub (57 years) at Ocala National Forest (ONF; Greenberg 2003) have lasted at least 7 years.

Therefore, the comparison between CKSSR and these studies was made by mainly taking into

consideration the first year postburn (see Table 3-19). Even though the results obtained from

CKSSR were short term, they are important because this is the first study conducted on coastal

scrub in the west coast of the Florida Peninsula.

Another factor to take into consideration is how variables were measured among studies.

Mean height, species richness, and mean percent cover of bareground and species presented in

Table 3-19 were the only variables and the most common species shared among these three

studies. Mean percent cover in CKSSR was quantified for herb species and for shrub species in

the stratum < 1.5 m tall and the results are comparable with ABS and ONF because these studies









did not establish any stratum. At KSC, Schmalzer (2003) measured cover at two stratum: <0.5 m

and >0.5 m. Therefore, the data from these two strata were summed to make them comparable

with the other three studies. Unfortunately, other studies carried out in ABS, KSC, and Naval

Live Oaks/Gulf Island National Seashore (NLO; Ruth et al. 2007) could not be included in the

comparison because they did not quantify the same variables through time. However, important

results from these studies are cited to highlight particular points in this discussion.

Bareground recovery is slow through time in long-unburned scrubs. Since, long-unburned

scrubs have zero or very low mean percent ofbareground cover, fire considerably increases

bareground cover during the first 6 to 12 months postburn, then it tends to decrease as Quercus

spp., Lyonia spp., and S. repens increase density and cover through time. In KSC, bareground

increased significantly after burning until 6 months (22.9%), but declined rapidly to 0.7% at 36

months. In ONF, bareground increased significantly after burning, reaching its maximum at 16

months postburn (25%), and by 101 months postburn (8%) still was not near preburn level

(0.00%). Even though the study in CKSSR was at short term, the recovery of the vegetation was

so fast that bareground had 3.8% and 6.8% at 6 and 12 months postburn, respectively. Probably,

bareground in CKSSR will recuperate to the preburn level faster than in the other two study

areas.

The recovery process for litter and debris has not been well documented in the literature.

Greenberg (2003) reported these values in a sand pine stand in ONF. Litter was recorded as

depth of litter layer. It decreased from preburn level of 6.5 cm to 2.0 cm and 2.5 cm at 5 and 16

months postburn, respectively, and then slowly decreased to 1% at 101 months postburn. Debris

had a preburn value of 0.2%, which was constant until 16 months, then it increased to 8.4% at









101 months. In CKSSR, litter and debris recovered faster than in ONF, achieving litter more than

70% and debris more than 30% of the preburn level in treatment sites in 12 months.

Vegetation height was one of the variables with the slowest recovery process. In KSC,

mean pre-burn height was 1.08 m and reached 0.32 m and 0.50 m at 6 and 12 months,

respectively. Height growth continued throughout the postbum period reaching prebum value at

85 months. In ONF, the mean height of the vegetation was determined by measuring the height

of Q. myrtifolia and Sabal etonia. Table 3-19 only shows the mean height of S. etonia because it

was the highest data recorded at 5 and 16 months. By 5 months postbum, S. etonia reached its

preburn value (1 m), and Q. myrtifolia needed almost 64 months to achieve its prebum mean

value of 1.25 m. In CKSSR, the mean height of the vegetation recovered slower than in the other

two study areas. It was 31% of the prebum value at 12 months, while in KSC was 46% of the

prebum value.

The recovery process in KSC is described below. Aristida strict was the most common

herb species and recovered preburn value in six months. Other herb species also found in

CKSSR such as Carphephorus spp. and Galactia elliottii were not recorded during the preburn

sample, and they were censused with low cover during the postburn period. Regarding woody

species in general, Q. myrtifolia, S. repens, and L. lucida were the dominant species (Table 3-

19), with Q. geminata and Q. chapmanii also relatively common. At the <0.5 m stratum, Q.

myrtifolia and L. lucida increased cover more than 5 times the preburn levels after burning.

Then, cover decreased at four years for Q. myrtifolia and two years for L. lucida, and reached

prebum level after seven and five years, respectively. At the >0.5 m stratum, Q. myrtifolia and L.

lucida significantly decreased cover after one year postbum and needed five years to recover pre-

burn values. S. repens did not have a high increase in cover in the <0.5 m stratum after burning,









and it had a low cover relatively constant through time. S. repens reestablishes cover faster than

woody shrubs at >0.5 m stratum. It increased cover right after prescribed burning and reached

the preburn level between one and one and a half year. Q. geminata and Q. chapmanii had a low

increase in cover after burning at the <0.05 m stratum. Then, they gradually reduced cover after

three years. Both species decreased cover after burning at the >0.5 m stratum and obtained

prebum levels after 5 years postburn. L. lucida recovered cover between 4 and 5 years.

Schmalzer and Hinkle reported shifts in dominance after fire due to differences in recovery rates

in shrub species and S. repens. V. myrsinites had a low increase after burning that persisted

relatively constant through time at the <0.05 m stratum, and it started to appear after three years

at the >0.05 m stratum, and it kept low cover after that.

In ABS, Abrahamson and Abrahamson did not show data of cover for herb species during

five years of postburn period. They only presented the average percent cover for five 200-m

transects of seven species not found in CKSSR. However, they reported that long-unbumed

scrubby flatwoods contained fewer herb species during the postburn period than recently burned

scrubby flatwoods. A possible explanation suggested by the authors was that fire had the effect

of creating or enlarging gaps making the persistence of herbs (gap specialists) sensitive to time

since fire. Long fire-free periods may allow shrubs to clonally spread into gaps and make the

gap-specialist herbs disappear through time. S. repens, Q. chapmanii, and Q. geminata were

dominant species. Cover of Q. chapmanii, Q. geminata, Q. minima, Lyoniafruticosa, L. lucida,

V. myrsinites and M. cerifera returned to or exceeded preburn levels within three years following

fire. S. repens reached preburn level in two years, did not maintain preburn dominance, but cover

was higher than 15% through time.









In ONF, Rynchospora megalocarpa was the dominant herb species and recuperated its

preburn value (1.84%) in 16 months. Species common to CKSSR were: Clitoria mariana (0.01-

0.80%), G. elliottii (0.02%), and Zamiapumila (0.02-0.53%), and they occurred with low mean

percent cover. Of 28 herb species, 19 were absent from transects prior to the bum and occurred

on transects during postbum samples with low cover (0.01-1.53%). In general, scrub woody

species composition and cover were similar to preburn values after 16 months. Q. myrtifolia was

the most dominant shrub before fire, recovered rapidly, and attained 67% of prebum cover level

at 16 months postburn. Q. geminata regained 84% of its preburn level in 16 months. Mean

percent cover of Q. chapmanii was 0.98% prebum, and it almost achieved this value at 16

months (0.93%). S. repens recovered 75% of its prebum level and L.ferruginea surpassed its

preburn level in 16 months. In general, recovery rates among species did not result in long-term

shifts in species dominance because Sabal etonia recovered its prebum level faster than Q.

myrtifolia, but Q. myrtifolia continued to be the dominant species.

Comparing the recovery pattern in CKSSR with KSC, ABS, and ONF, we found a

different story for herb and woody species. Galactia elliottii was the most common herb species

in CKSSR, but with limited comparison because it was only recorded by few sampling periods in

KSC and ONF. Carphephorus corymbosus was only recorded in control site 5D at CKSSR, and

it had low cover in almost all samples in KSC. Clitoria mariana and Zamiapumila were sampled

only in 5A and 5D, respectively, in CKSSR, and they also had low cover in almost all samples in

ONF. At CKSSR, G. elliottii and G. moilis were the only herb species sampled on control sites

and during the postbum period, while Crotalaria rotundifolia, Solidago odora and Woodwardia

virginica were registered only during the postbum period. Therefore, few species are present

with low cover before and after burning, and other species colonize gaps available after burning









for a temporal use (with low cover) and until they are replaced by the dominant and growing

woody vegetation.

In general, the scrub woody vegetation in KSC, ABS, ONF, and CKSSR returned to

preburn conditions rapidly after a high-intensity prescribed bum. This aspect has been reported

by other studies in scrub vegetation in Florida (Abrahamson 1984a, 1984b; Menges et. al. 1993;

Schmalzer et al. 2003; Ruth et al. 2007). As can be seen in Table 3-19, CKSSR and ONF had the

fastest recovery. However, I did not investigate cover after 12 months in CKSSR. The recovery

time for the other studies suggest that there is variation depending on the type of scrub.

Regarding the dominant species, Q. myrtifolia and S. repens were the most common species in at

least three study areas. There was a shift in dominance after fire for shrubs and palmetto in KSC,

for S. repens in ABS, and for Q. chapmanii and L.ferruginea in CKSSR (see Figure 3-22).

Although P. clausa is a species adapted to fire, its seedlings take some time to appear after

fire. In ABS, P. clausa appearance is delayed three years postbum (Abrahamson 1984b). In

ONF, P. clausa seedlings were established after 5 months postburn (Greenberg 2003). In NLO

(Ruth et al. 2007), P. clausa seedlings were absent eight months to two years after burning from

sand pine scrub. In CKSSR, seedlings appeared six months postburn. Hence, P. clausa seedlings

appear after several months to several years in the scrub and very little is known about seedlings

survival and establishment.

Even though several vegetation variables returned to prebum conditions in CKSSR during

an interval of 12 months, this time was still too short to predict that treatment sites would be

restored to scrubby flatwoods without a history of fire suppression. According to Abrahamson

and Abrahamson (1996a, 1996b) and Baker (1992, 1994), prescribed burning does not

necessarily reestablish the prebum conditions in all landscapes after several years of fire









suppression. This is an aspect that needs to be determined in Cedar Key with continued

monitoring of the study sites. Schmalzer and Boyle (1998), Schmalzer and Adrian (2001), and

Schmalzer et al. (2003) recommended a combination of mechanical treatment and prescribed

burning in restoring long-unburned scrub vegetation. Mechanical cutting should be used only

one time to reestablish shrub vegetation structure, which can be maintained with periodic

prescribed burning after that. All mechanical treatments result in some loss of S. repens cover,

and this loss persists (Schmalzer et al. 2003). So, mechanical treatment must be applied

carefully.

Whether the reintroduction of fire to long-unburned scrub might produce or not the desired

effect of returning the association to a state similar to scrubs without fire suppression will rely on

fire intensity and the season of burning. The reintroduction of frequent and low-intensity fires to

sand pine scrub may shift scrub to sandhill (Myers 1985). In contrast, a single, high-intensity fire

in sand pine scrub may make possible the perseverance of the scrub stand (Myers 1985, Menges

et al. 1993, Menges and Hawkes 1998). Instead, a single-low intensity fire might facilitate the

shift of scrub toward xeric hammock. This situation would occur if fire increases the abundance

of the sprouter species, and consequently repress the regeneration of obligate seeders such as

sand pine and herb species. According to Abrahamson & Abrahamson (1996b), a single fire may

not be effective at restoring long-unburned scrubby flatwoods to states characteristic of more

recently burned stands. Several fires may be needed before scrubby flatwoods are returned to

communities similar to those without fire suppression.

Community Shift in Response to Prescribed Burning

The scrubby flatwoods community in CKSSR had structural and community changes after

prescribed burning, and they were reported for an interval of 12 months. As previously

mentioned, studies conducted in KSC, ABS, and ONF have datasets for at least 7 years, which

98









constraint the comparison with CKSSR. This study presents a comparison between CKSSR and

these three studies regarding species richness, species diversity (Shannon-Wiener's index),

evenness (H'/ In S), and the results of the Detrended Correspondence Analysis mainly restricted

to 12 months postburn.

Mean species richness in KSC increased in the <0.5 m stratum, but declined in the >0.5 m

stratum, during the first 12 months. Considering both strata together, there was an increase in the

species richness 12 months postburn. Species diversity and evenness were not measured in this

study. The DCA carried out only with mean percent cover indicated that the composition of the

scrub varied along a gradient closely related to the depth of the water table with oak dominating

drier sites and S. repens and I. glabra in wetter places. The species ordination located oak

species to the left and S. repens, I. glabra, and M cerifera to the right of the axis. This pattern

was found in all oak-dominated transects and saw palmetto-dominated transects. The other

transects with a mixed oak-saw palmetto composition before burning located sample units in the

middle of the ordination axis. Both, oak-dominated and saw palmetto-dominated transects in the

ordination space, returned to preburn locations after three years. Vector lengths were greater for

the mixed oak-saw palmetto transects than those dominated by S. repens, indicating a greater

degree of change in the mixed oak-saw palmetto transects.

In one long-unburned (>35 yr) and two recently burned (<20 yr) scrubby flatwoods in

ABS, species richness was constant in long-unburned and decreased in recently burned during

the first 12 months. Later, species richness increased in all stands relative to preburn levels until

36 months, and increases were most pronounced at recently burned scrubby flatwoods. Also,

species richness of long-unburned scrubby flatwoods was reduced relative to preburn levels after

36 months. Species diversity increased and reached the average preburn index in the long-









unburned stand at 12 months postburn. However, species diversity increased just after burning

and then decreased in recently burned stands in the first 12 months. But the index at 12 months

was higher than preburn index. After the first year, species diversity increased and remained high

for the recently burned stands, which had a higher index through time than the long-unburned

stand. Species diversity in the long-unburned stand decreased at 24 months and 48 months

postburn and then linearly increased through time after that. Overall, evenness was reduced after

fire. Long-unburned stands achieved preburn evenness at approximately 72 months postburn;

however, recently burned stands did not. The DCA was carried out to visualize the multivariate

changes in species dominance (percentage of cover based on crown intercepts) in the preburn

and postfire samples. The analyses, based on absolute and relativized dominance, showed that

scrubby flatwoods were very stable following fire. Changes related to composition were more

noticeable than structural changes. In general, the effect of fire on these stands caused slight

changes in stand structure and some degree of shifts in stand composition regardless of time

since fire. These results are similar to the conclusions found from other studies in Florida

(Abrahamson 1984a, 1984b; Abrahamson et al. 1984; Abrahamson & Hartnett 1990; Schmalzer

&Hinkle 1992a).

In a sand pine scrub stand in ONF, herbaceous species richness increased within 5 months

postburn, peaked at 16 months, and declined by 40 months postburn. Woody species richness

decreased immediately after fire (five weeks), then increased until 28 months and remained

constant after that. In general, species richness increased during the first 12 months. The increase

followed by the gradual decline of herbaceous species richness appeared to be related with

gradual increases in shrub cover and decreases in bare ground availability. Carrington and

Keeley (1999) suggested that the low postburn seedling recruitment in sand pine scrub is









probably due to the elimination of suitable microsites by resprouting shrubs. Greenberg (2003)

did not quantify species diversity, evenness, and did not carry out a DCA.

The results found in CKSSR are similar to the results found in the other studies. KSC,

ONF, and CKSSR had an increased of species richness, but ABS did not during the first 12

months. This increase in species richness is expected due to the disturbance caused by fire, the

amount of species resprouting and clonal spread, and new herb species that are temporarily

colonizing gaps. Species diversity in the long-unburned scrubby flatwoods in ABS and in

CKSSR achieved preburn levels in 12 months postburn. Evenness regained preburn levels in

ABS and CKSSR, but was faster in CKSSR. The information obtained from species richness,

species diversity, and evenness suggests little structural and compositional changes at short term

in CKSSR and at long term in ABS. In contrast, DCA was able to reveal results not detected by

these indices.

The DCA revealed both structural and compositional changes at short term in CKSSR and

more compositional than structural changes at long term in ABS. In CKSSR, there were both

structural and compositional changes with similar magnitude during the first 3 months according

to the lengths of the vectors (Figure 3-29 through 3-32). These changes were expected because

many species were resprouting simultaneously. Between 3 and 12 months, there were also

structural and compositional changes because almost all species already resprouted (little

variation in species richness and diversity), their frequencies were almost constant thorough time

(Figure 3-21), and mainly the density and cover of Q. myrtifolia and the cover of S. repens, Q.

geminata, and L. ferruginea increased through time (Figures 3-20 and 3-22). In a low-intensity

fire on long-unburned sand pine scrub in ABS, Abrahamson and Abrahamson (1996b) reported

that the largest amount of structural and compositional change occurred immediately after









prescribed fire, just between the preburn and 1-year postburn censuses. In contrast, changes

between the 1-year and 2-year censuses after fire were primarily compositional; little structural

change was measured.

The DCA carried out in CKSSR on mean percent cover also showed that the composition

of the scrub varied along a gradient associated with the water table. Figures 3-33 and 3-34

presents the ordination found in KSC and in CKSSR with preburn samples of mean percent

cover, respectively. Comparing both figures, we can see that oaks are in the left side of the

ordination (drier places) and S. repens and I. glabra are in the right side of the ordination (wetter

places). In Cedar Key, M. cerifera, L. lucida, and Q. geminata are almost in the middle of the

gradient, but not in KSC. However, the position of L. lucida is not exactly to the left and the

location ofM. cerifera is not exactly to the right in the ordination space in KSC.

Age at First Flowering after Prescribed Fire

Age at first flowering might vary among species within a single community, between

populations of a single species, and within one population. In addition, year of first flowering for

several species might vary among sites in the same study area (Whelan 1995). In CKSSR,

flowering was not synchronous within and among species and among stands without burning.

Flowering started in March 2006 after the prescribed burnings in April and May 2005. Probably,

sprouting species delayed flowering after prescribed fire because the energy produced through

photosynthesis was devoted to vegetative growth. In sand pine scrub in ABS (Abrahamson and

Abrahamson 1996b), S. etonia did not flower until spring the year following a February burn. In

contrast, in ONF (Greenberg 2003), S. etonia flowered within 5 weeks and fruited within 5

months. The factors that affect post-fire sprouting are likely to affect post-fire flowering when

the number of flowers or inflorescences per plant are determined by the number of active shoots.

Each shoot sprouting after fire can produce a terminal inflorescence (Whelan 1995). Therefore,

102









the season of bum, fire intensity, or both may be important factors that affect flowering in

sprouting species. Unfortunately, the effects of these factors on flowering of sprouting species

have rarely been quantified.










Table 3-1. List of plant species recorded in quadrats in treatment and control sites in Cedar Key


Scrub State Reserve.
Common Name


Family


Recovery Mode


Herbaceous
Agalinis ilifolia
Asclepias sp
Carphephorus
corymbosus
Clitoria mariana
Crotalaria rotundifolia
Galactia elliottii
Galactia mollis
Solidago odora
Woodwardia virginica
Zamia pumila

Woody
Pinus clausa
Pinus elliottii
Pinus palustris
Quercus geminata
Quercus myrtifolia
Quercus chapmanii
Quercus minina
Quercus nigra
Quercus sp
Brevaria racemosa
Ceratolia ericoides
Ilex glabra
Lyoniaferruginea
Lyoniafruticosa
Lyonia lucida
Myrica cerifera
Gaylussacia dumosa
Gaylussacia nana
Licania michauxii
Osmanthus americanus
Rhus copallinum
Salix caroliniana
Serenoa repens
Smilax auriculata
Smilax sp
Vaccinium myrsinites


Seminole False Foxglove
Milkweed

Florida Paintbrush
Atlantic Pigeonwings
Rabbitbells
Elliottis Milkpea
Soft Milkpea
Chapman's Golden Rod
Virginia Chain Fern
Florida Arrowroot


Sand Pine
Slash Pine
Long-leaf Pine
Sand Live Oak
Myrtle Oak
Chapman Oak
Runner Oak
Water Oak

Tar Flower
False Rosemary
Gallberry
Rusty Lyonia
Stagger-bush
Fetterbush
Wax Myrtle
Dwarfhuckleberry
Dangleberry
Gopher Apple
Wild Olive
Winged Sumae
Carolina Willow
Saw Palmetto
Catbrier

Blueberry


Orobanchaceae
Apocynaceae

Asteraceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Asteraceae
Blechnaceae
Zamiaceae


Pinaceae
Pinaceae
Pinaceae
Fagaceae
Fagaceae
Fagaceae
Fagaceae
Fagaceae
Fagaceae
Ericaceae
Empetraceae
Aquifoliaceae
Ericaceae
Ericaceae
Ericaceae
Myricaceae
Ericaceae
Ericaceae
Chrysobalanaceae
Oleaceae
Anacardiaceae
Salicaceae
Arecaceae
Smilacaceae
Smilacaceae
Ericaceae


Resprouter (1)
Resprouter (2)

Resprouter (2)
Resprouter (3)
Resprouter
Resprouter (2)
Resprouter
Resprouter (*) (2)
Resprouter (1)
Resprouter (3)


Obligate seeder (2)
Obligate seeder (*) (2)
Obligate seeder (*)
Resprouter (4) (2)
Resprouter (4) (2)
Resprouter (4) (2)
Resprouter (4) (2)
Resprouter
Resprouter
Resprouter (2)
Obligate seeder (2)
Resprouter
Resprouter (2)
Resprouter (4) (2)
Resprouter (4) (2)
Resprouter (4) (2)
Resprouter (4) (2)
Resprouter (1)
Resprouter (4) (2)
Resprouter
Resprouter (2)
Resprouter
Resprouter-seeder
Resprouter (*) (2)
Resprouter
Resprouter (4) (2)


Cladonia evansii
Opuntia humifusa


Deer Moss
Devil's tongue


Cladoniaceae
Cactaceae


Species


Seeder (2)
Resprouter


Codes: (*) Resprouter, clonal spreader, and seeder. (4) Resprouter and clonal spreaders. (*)
Obligate seeder and survivor. (1) United States Department of Agriculture website. (2) Menge &
Kohfeldt (1995). (3) Greenberg (2003).


i i









Table 3-2. Species richness, Simpson's index, Shannon-Wiener's index, and Shannon-
Wiener's evenness for preburn conditions in control (5A & 5D) and treatment (5C
& 2M) sites in Cedar Key Scrub State Reserve.
Site Richness Simpson Shannon-Wiener Evenness
5A 26 10.64 2.54 0.6981
5D 24 7.06 2.28 0.6271
5C 17 5.03 1.99 0.5483
2M 20 8.41 2.28 0.6265









Table 3-3. Multiple mean comparison (Duncan's test) between clusters created by using
Euclidean distances and Ward's minimum variance linkage fusion. Abundance and
mean percent cover were standardized. Significant level = 0.05.


Variable
Galactia elliottii (abundance)
Galactia elliottii (cover)
Quercus geminata (abundance)
Quercus geminata (cover)
Quercus myrtifolia (abundance)
Quercus myrtifolia (cover)
Quercus chapmanii (abundance)
Quercus chapmanii (cover)
Quercus minima (abundance)
Quercus minima (cover)
Quercus sp (abundance)
Brevaria racemosa (abundance)
Ceratolia ericoides (cover)
Ilex glabra (abundance)
Lyoniaferruginea (abundance)
Lyoniaferruginea (cover)
Lyonia fruticosa (abundance)
Lyonia lucida (abundance)
Lyonia lucida (cover)
Myrica cerifera (abundance)
Myrica cerifera (cover)
Gaylussacia dumosa (abundance)
Gaylussacia nana (abundance)
Licania michauxii (abundance)
Serenoa repens (abundance)
Serenoa repens (cover)
Smilax auriculata (abundance)
Vaccinium myrsinites (abundance)
Vaccinium mvrsinites (cover)


Cluster 1
n Mean
34 -0.2873
34 -0.2375
34 -0.1107
34 -0.1750
34 -0.6000
34 -0.5288
34 -0.2150
34 -0.1741
34 -0.0354
34 -0.2316
34 -0.1589
34 -0.0374
34 -0.2030
34 -0.0432
34 -0.2414
34 -0.4303
34 -0.1792
34 0.1317
34 0.2238
34 -0.2175
34 -0.1522
34 -0.1296
34 -0.1939
34 -0.0984
34 0.6573
34 1.0759
34 -0.1325
34 -0.3902
34 -0.2928


Cluster 2
n Mean
47 -0.2635
47 -0.2089
47 -0.1272
47 -0.2104
47 0.6634
47 0.7118
47 -0.1768
47 -0.2252
47 -0.1105
47 -0.1840
47 -0.1404
47 -0.1302
47 -0.1770
47 -0.2026
47 -0.0951
47 -0.1164
47 -0.2084
47 -0.1755
47 -0.3472
47 -0.1783
47 -0.1760
47 -0.1108
47 -0.2177
47 -0.0978
47 -0.3788
47 -0.4110
47 -0.1809
47 -0.1371
47 -0.1172


F value
0.08
0.13
0.01
0.14
41.47
38.54
0.08
0.42
0.25
0.72
0.03
2.86
0.72
5.19
1.13
6.85
1.45
3.74
15.7
0.11
0.26
0.07
0.18
0.00
35.23
72.16
1.39
5.21
2.52


P-value
0.7742
0.7234
0.9163
0.7130
<.0001
<.0001
0.7792
0.5173
0.6219
0.3984
0.8578
0.0945
0.3984
0.0254
0.2919
0.0106
0.2315
0.0567
0.0002
0.7398
0.6135
0.7939
0.6720
0.9909
<.0001
<.0001
0.2421
0.0251
0.1165


I \ j









Table 3-4. T test, ANOVA test, and Kruskal-Wallis test for comparing means and medians
among treatment and control sites under preburn conditions in Cedar Key Scrub State
Reserve. Data for T test and ANOVA were standardized. Test for ANOVA and
Kruskal-Wallis is F test and Chi-squared test, respectively. Significant level = 0.05.
Test Species 5A 5D 5C pre 2M pre Test DF P-value
H Ilex glabra (*) -0.2907 0.1642 -1.1570 16 0.2642
Quercus myrtifolia (*) 0.0721 0.0655 -0.1537 -0.1519 0.7000 3 0.5523
< Quercus myrtifolia (%) -0.0683 0.1724 0.0835 0.0867 0.3300 3 0.8014
0 Lyonia ferruginea (%) 0.2617 0.2827 0.4946 0.3457 0.1700 3 0.9169


SLyonia lucida (%) 0.0167 0.1522 -0.0267 -0.6480 1.7500 3 0.1654
Serenoa repens (%) -0.2689 0.1440 0.0508 0.0926 0.8600 3 0.4657
SSerenoa repens (*) 4.0000 3.0000 4.0000 3.0000 7.9563 3 0.0469
4 Vaccinium myrsinites (*) 5.0000 3.0000 2.0000 10.0000 18.7388 3 0.0003


Codes: (*) mean abundance. (%) mean percent cover.










Table 3-5. Pre- and postburn absolute densities for all species in 5C in Cedar Key Scrub State
Reserve. Absolute densities for control sites 5A and 5D are also shown. d = days. M
= months.


Species
Herbaceous
Poaceae
Solidago odora
Galactia elliottii
Galactia mollis
Crotalaria rotundifolia
Agalinis ilifolia
Asclepias sp
Carphephorus
corymbosus
Clitoria mariana
Woodwardia virginica
Zamia pumila

Woody
Quercus myrtifolia
Lyoniaferruginea
Quercus geminata
Lyonia lucida
Quercus chapmanii
Serenoa repens
Ilex glabra
Licania michauxii
Lyoniafruticosa
Vaccinium myrsinites
Gaylussacia dumosa
Gaylussacia nana
Brevaria racemosa
Myrica cerifera
Rhus copallinum
Pinus clausa
Smilax auriculata
Quercus minina
Pinus palustris
Pinus elliottii
Quercus nigra
Quercus sp
Ceratolia ericoides
Osmanthus americanus
Salix caroliniana
Smilax spp

Cladonia evansii
Opuntia humifusa


Control
5A 5D Pre


Time after burning
l d 3M 6M 9M 12 M


0.37 0.14 0.03 0.09 0.61 1.15 0.98 1.90
0.00 0.00 0.00 0.00 0.12 0.11 0.06 1.82
2.46 1.09 0.00 0.00 2.66 1.26 0.00 1.35
0.00 0.18 0.00 0.00 0.04 0.05 0.00 1.22
0.00 0.00 0.00 0.00 0.08 0.07 0.07 0.03
0.00 0.15 0.00 0.00 0.00 0.00 0.00 0.00
0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.19 0.00 0.00 0.00 0.00 0.00 0.00
0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00


10.91 12.17 12.53 1.46 45.58 43.46 47.76 42.11
2.09 5.06 3.87 0.58 8.32 10.39 11.22 11.60
2.90 2.24 2.20 0.38 13.20 12.68 13.24 11.58
7.03 10.70 4.49 0.55 8.23 8.83 8.87 8.88
3.01 2.01 2.79 0.29 9.99 8.58 7.33 6.43
1.14 1.01 1.41 1.80 3.31 4.32 4.34 4.94
2.98 0.55 0.13 0.06 4.63 4.91 4.74 3.55
0.76 0.20 0.00 0.00 2.65 2.43 0.90 2.34
1.88 0.00 0.25 0.15 1.24 1.85 2.22 1.71
5.08 2.04 0.51 0.00 0.97 1.43 1.44 1.69
5.07 1.37 0.00 0.00 1.69 1.41 1.46 1.40
0.31 5.30 0.03 0.00 1.32 1.19 0.28 1.13
0.57 0.05 0.00 0.03 1.92 2.00 1.65 1.06
1.04 0.47 0.85 0.01 0.43 0.77 0.69 0.50
0.00 0.00 0.00 0.00 0.25 0.28 0.00 0.31
0.00 0.00 0.02 0.01 0.01 0.03 0.08 0.25
0.12 0.08 0.00 0.00 0.03 0.11 0.17 0.12
2.50 1.70 1.21 0.00 0.30 0.38 0.35 0.09
0.05 0.00 0.00 0.00 0.00 0.00 0.00 0.08
0.04 0.00 0.04 0.00 0.00 0.01 0.02 0.05
0.00 0.00 0.08 0.00 0.00 0.00 0.00 0.00
0.07 0.17 0.11 0.00 0.02 0.00 0.00 0.00
0.03 0.02 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.03 0.02 0.00 0.00
0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00

3.51 1.80 2.26 0.00 0.00 0.00 0.00 0.00
0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00










Table 3-6. Pre- and postburn absolute frequencies for all species in 5C in Cedar Key Scrub State
Reserve. Absolute frequencies for control sites 5A and 5D are also shown. d = days.
M = months.
Control Time after burning
Species 5A 5D Pre 1 d 3 M 6M 9M 12 M
Herbaceous
Poaceae 12 6 4 3 15 17 18 25
Galactia elliottii 31 29 0 0 35 28 1 20
Galactia mollis 0 1 0 0 1 2 0 11
Solidago odora 0 0 0 0 5 5 4 8
Crotalaria rotundifolia 0 0 0 0 1 1 1 1
Agalinisfilifolia 0 1 0 0 0 0 0 0
Asclepias sp 1 0 0 0 0 0 0 0
Carphephorus
corymbosus 0 1 0 0 0 0 0 0
Clitoria mariana 2 0 0 0 0 0 0 0
Woodwardia virginica 0 0 0 0 0 0 0 0
Zamiapumila 0 1 0 0 0 0 0 0

Woody
Quercus myrtifolia 34 44 35 29 36 35 35 35
Quercus geminata 29 34 30 14 34 33 34 34
Serenoa repens 27 29 33 31 35 35 35 34
Quercus chapmanii 32 35 27 12 29 30 31 31
Lyoniaferruginea 19 34 27 18 20 24 24 26
Lyonia lucida 28 38 31 13 28 25 24 24
Vaccinium myrsinites 34 27 14 0 14 20 19 23
Quercus minina 26 40 15 0 9 11 18 15
Pinus clausa 0 0 2 1 1 2 5 14
Myrica cerifera 14 12 15 2 8 14 13 13
Licania michauxii 5 6 0 0 13 15 5 13
Ilex glabra 13 4 3 1 6 11 9 9
Rhus copallinum 0 0 0 0 11 11 0 9
Smilax auriculata 5 4 0 0 1 7 8 7
Gaylussacia nana 1 29 2 0 8 7 5 6
Pinus elliottii 2 0 3 0 0 1 2 4
Lyoniafruticosa 13 0 4 3 3 3 3 4
Gaylussacia dumosa 24 22 0 0 6 3 3 4
Brevaria racemosa 8 2 0 2 2 3 3 3
Pinus palustris 3 0 0 0 0 0 0 2
Quercus nigra 0 0 1 0 0 0 0 0
Quercus sp 3 12 3 0 1 0 0 0
Ceratolia ericoides 3 6 0 0 0 0 0 0
Osmanthus americanus 0 1 0 0 0 0 0 0
Salix caroliniana 0 0 0 0 1 1 0 0
Smilax spp 1 0 0 0 0 0 0 0

Cladonia evansii 8 12 14 0 0 0 0 0
Opuntia humifusa 1 0 0 0 0 0 0 0










Table 3-7. Pre- and postburn absolute mean % cover of herb and woody species in 5C in Cedar
Key Scrub State Reserve. Absolute mean percent cover for control sites 5A and 5D
are also shown. d = days. M = Months.
Control Time after burning
Species 5A 5D Pre 1 d 3 M 6M 9M 12 M
Herbaceous
Poaceae 0.47 0.00 0.14 0.04 0.46 2.41 1.92 2.07
Galactia elliottii 2.06 1.76 0.00 0.00 1.83 0.13 0.00 0.51
Solidago odora 0.00 0.00 0.00 0.00 0.00 0.50 0.00 0.47
Galactia mollis 0.00 0.00 0.00 0.00 0.04 0.00 0.00 0.11
Agalinisfilifolia 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Asclepias sp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Carphephorus corymbosus 0.00 0.26 0.00 0.00 0.00 0.00 0.00 0.00
Clitoria mariana 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Crotalaria rotundifolia 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Woodwardia virginica 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Zamiapumila 0.00 0.09 0.00 0.00 0.00 0.00 0.00 0.00

Woody
Quercus myrtifolia 20.07 27.36 28.43 0.00 15.21 16.83 16.07 24.95
Serenoa repens 23.95 24.39 28.66 0.00 13.26 19.32 22.48 23.00
Quercus geminata 6.75 0.64 2.71 0.00 4.26 4.51 3.76 7.64
Lyonia lucida 8.94 18.00 4.35 0.00 2.05 2.35 2.82 3.58
Lyoniaferruginea 4.22 5.70 5.32 0.00 1.44 1.84 3.28 3.10
Quercus chapmanii 2.89 5.97 2.75 0.00 2.12 2.47 1.75 2.61
Ilex glabra 2.28 0.00 0.28 0.00 0.57 2.18 0.77 1.27
Myrica cerifera 0.39 1.32 1.87 0.00 0.42 0.43 0.95 0.90
Lyoniafruticosa 1.48 0.00 0.07 0.00 0.00 0.55 0.42 0.73
Vaccinium myrsinites 3.55 2.80 0.37 0.00 0.05 0.34 0.24 0.59
Rhus copallinum 0.00 0.00 0.00 0.00 0.16 0.31 0.00 0.26
Licania michauxii 0.04 0.00 0.00 0.00 0.25 0.08 0.00 0.14
Gaylussacia dumosa 0.43 0.06 0.00 0.00 0.09 0.07 0.00 0.10
Gaylussacia nana 0.28 1.60 0.00 0.00 0.15 0.00 0.00 0.07
Pinus clausa 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Pinus elliottii 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Pinus palustris 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Quercus minina 0.59 0.27 0.23 0.00 0.00 0.00 0.00 0.00
Quercus nigra 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Quercus sp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Brevaria racemosa 1.44 0.72 0.00 0.00 0.00 0.00 0.00 0.00
Ceratolia ericoides 2.71 7.49 0.00 0.00 0.00 0.00 0.00 0.00
Osmanthus americanus 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Salix caroliniana 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Smilax auriculata 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Smilax spp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Cladonia evansii 1.05 0.00 0.33 0.00 0.00 0.00 0.00 0.00
Opuntia humifusa 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00










Table 3-8. Pre- and postburn absolute importance values of herb and woody species in 5C in
Cedar Key Scrub State Reserve. Absolute importance values for control sites 5A and
5D are also displayed. d = days. M = Months.


Control
5A 5D


Species
Herbaceous
Poaceae
Galactia elliottii
Solidago odora
Galactia mollis
Crotalaria rotundifolia
Agalinis filifolia
Asclepias sp
Carphephorus corymbosus
Clitoria mariana
Woodwardia virginica
Zamia pumila

Woody
Quercus myrtifolia
Serenoa repens
Quercus geminata
Lyoniaferruginea
Lyonia lucida
Quercus chapmanii
Vaccinium myrsinites
Ilex glabra
Licania michauxii
Myrica cerifera
Quercus minina
Pinus clausa
Lyoniafruticosa
Rhus copallinum
Gaylussacia nana
Gaylussacia dumosa
Smilax auriculata
Brevaria racemosa
Pinus elliottii
Pinus palustris
Quercus nigra
Quercus sp
Ceratolia ericoides
Osmanthus americanus
Salix caroliniana
Smilax spp

Cladonia evansii
Opuntia humifusa


Time after burning
Pre 1 d 3M 6M 9M 12 M


0.0418 0.0164
0.1457 0.1055
0.0000 0.0000
0.0000 0.0059
0.0000 0.0000
0.0000 0.0053
0.0026 0.0000
0.0000 0.0088
0.0056 0.0000
0.0000 0.0000
0.0000 0.0033


0.5225 0.6254
0.3733 0.3338
0.2049 0.1288
0.1353 0.2377
0.3038 0.4868
0.1680 0.1805
0.2186 0.1309
0.1136 0.0202
0.0267 0.0176
0.0581 0.0500
0.1165 0.1275
0.0000 0.0000
0.0839 0.0000
0.0000 0.0000
0.0115 0.1897
0.1566 0.0781
0.0157 0.0106
0.0473 0.0128
0.0056 0.0000
0.0083 0.0000
0.0000 0.0000
0.0086 0.0305
0.0402 0.0900
0.0000 0.0025
0.0000 0.0000
0.0032 0.0000

0.0962 0.0638
0.0028 0.0000


0.0173 0.0769
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000


0.8850 0.4947
0.5421 0.5730
0.2116 0.1788
0.2861 0.2467
0.3067 0.2024
0.2191 0.1466
0.0712 0.0000
0.0185 0.0188
0.0000 0.0000
0.1050 0.0174
0.0942 0.0000
0.0079 0.0096
0.0230 0.0510
0.0000 0.0000
0.0082 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0210
0.0121 0.0000
0.0000 0.0000
0.0061 0.0000
0.0142 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000

0.1239 0.0000
0.0000 0.0000


0.0619
0.1736
0.0162
0.0043
0.0038
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000


0.8913
0.4496
0.3258
0.1717
0.2095
0.2304
0.0524
0.0746
0.0697
0.0382
0.0300
0.0031
0.0206
0.0394
0.0400
0.0360
0.0033
0.0239
0.0000
0.0000
0.0000
0.0032
0.0000
0.0000
0.0033
0.0000


0.1031
0.0932
0.0243
0.0061
0.0035
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000


0.8120
0.4946
0.2940
0.1981
0.1958
0.2098
0.0760
0.1169
0.0663
0.0547
0.0346
0.0059
0.0358
0.0394
0.0308
0.0228
0.0208
0.0270
0.0029
0.0000
0.0000
0.0000
0.0000
0.0000
0.0030
0.0000


0.0000 0.0000 0.0000 0.0000
0.0000 0.0000 0.0000 0.0000


0.1027 0.1120
0.0032 0.0722
0.0135 0.0446
0.0000 0.0419
0.0039 0.0029
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000


0.8514 0.8342
0.5668 0.4545
0.3021 0.3040
0.2422 0.2204
0.2120 0.1961
0.2008 0.1779
0.0794 0.0843
0.0873 0.0747
0.0246 0.0580
0.0660 0.0512
0.0617 0.0401
0.0170 0.0390
0.0380 0.0368
0.0000 0.0301
0.0188 0.0273
0.0233 0.0250
0.0275 0.0195
0.0250 0.0178
0.0067 0.0109
0.0000 0.0060
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000










Table 3-9. Pre- and postburn absolute densities of herb and woody species in 2M in Cedar Key
Scrub State Reserve. Absolute densities for control sites 5A and 5D are also shown. d
= days. M = months.
Control Time after burning
Species 5A 5D Pre 12 d 3 M 6M 9M 12 M


Herbaceous
Galactia elliottii
Poaceae
Solidago odora
Woodwardia virginica
Galactia mollis
Agalinis ilifolia
Asclepias sp
Carphephorus corymbosus
Clitoria mariana
Crotalaria rotundifolia
Zamia pumila

Woody
Quercus myrtifolia
Lyoniaferruginea
Gaylussacia nana
Quercus geminata
Lyonia lucida
Vaccinium myrsinites
Quercus chapmanii
Lyoniafruticosa
Serenoa repens
Ilex glabra
Licania michauxii
Brevaria racemosa
Myrica cerifera
Smilax auriculata
Gaylussacia dumosa
Quercus minina
Rhus copallinum
Pinus clausa
Pinus elliottii
Pinus palustris
Quercus nigra
Quercus sp
Ceratolia ericoides
Osmanthus americanus
Salix caroliniana
Smilax spp


Cladonia evansii
Opuntia humifusa


2.46 1.09 0.00
0.37 0.14 0.17
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.18 0.00
0.00 0.15 0.00
0.01 0.00 0.00
0.00 0.19 0.00
0.04 0.00 0.00
0.00 0.00 0.00
0.00 0.01 0.00


10.91 12.17 12.13
2.09 5.06 5.40
0.31 5.30 2.78
2.90 2.24 6.62
7.03 10.70 5.27
5.08 2.04 7.02
3.01 2.01 2.02
1.88 0.00 0.55
1.14 1.01 1.21
2.98 0.55 0.74
0.76 0.20 0.03
0.57 0.05 0.04
1.04 0.47 1.01
0.12 0.08 0.02
5.07 1.37 0.00
2.50 1.70 1.25
0.00 0.00 0.00
0.00 0.00 0.00
0.04 0.00 0.00
0.05 0.00 0.01
0.00 0.00 0.15
0.07 0.17 0.95
0.03 0.02 0.03
0.00 0.01 0.00
0.00 0.00 0.00
0.04 0.00 0.00

3.51 1.80 7.44
0.02 0.00 0.00


2.48 3.37
0.02 0.55
0.02 0.14
0.00 0.30
0.12 0.13
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00


9.31 29.55
2.16 8.84
5.23 16.54
1.82 13.52
4.10 7.12
0.29 2.69
4.80 8.42
0.77 3.33
1.64 2.83
0.07 3.28
0.14 1.37
0.02 1.06
0.03 0.77
0.00 0.13
0.18 0.29
0.00 0.03
0.00 0.04
0.00 0.00
0.00 0.00
0.01 0.01
0.02 0.00
0.03 0.00
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00


0.59 0.40
0.51 0.50
0.12 0.02
0.04 0.00
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00


2.50
0.77
0.35
0.21
0.07
0.00
0.00
0.00
0.00
0.00
0.00


31.68 31.27 21.84
13.66 14.11 13.12
15.82 10.19 11.54
13.71 13.18 11.50
8.44 8.10 10.95
4.29 8.06 9.67
9.97 9.00 6.90
4.45 5.04 6.20
3.20 3.25 3.17
3.33 3.29 2.38
1.49 0.34 1.69
1.09 1.43 0.85
1.05 1.06 0.76
0.16 0.11 0.20
0.00 0.00 0.17
0.33 0.88 0.11
0.01 0.00 0.05
0.00 0.00 0.00
0.00 0.00 0.00
0.02 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00


0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00


0.00
0.00










Table 3-10. Pre- and postburn absolute frequencies of herb and woody species in 2M in Cedar
Key Scrub State Reserve. Absolute frequencies for control sites 5A and 5D are also
presented. d = days. M = Months.
Control Time after burning
Species 5A 5D Pre 12 d 3 M 6M 9M 12 M
Herbaceous
Galactia elliottii 31 29 0 30 40 28 3 36
Poaceae 12 6 9 8 15 14 14 17
Solidago odora 0 0 0 2 6 6 3 8
Galactia mollis 0 1 0 1 2 0 1 3
Woodwardia virginica 0 0 0 0 2 1 0 2
Agalinis filifolia 0 1 0 0 0 0 0 0
Asclepias sp 1 0 0 0 0 0 0 0
Carphephorus corymbosus 0 1 0 0 0 0 0 0
Clitoria mariana 2 0 0 0 0 0 0 0
Crotalaria rotundifolia 0 0 0 0 0 0 0 0
Zamiapumila 0 1 0 0 0 0 0 0

Woody
Quercus myrtifolia 34 44 48 44 44 45 46 44
Quercus geminata 29 34 43 23 44 43 43 44
Gaylussacia nana 1 29 9 30 38 38 33 39
Vaccinium myrsinites 34 27 38 6 32 37 40 39
Serenoa repens 27 29 35 36 36 36 36 37
Quercus chapmanii 32 35 34 28 33 37 36 34
Myrica cerifera 14 12 19 1 15 18 17 19
Lyoniaferruginea 19 34 20 17 18 19 18 18
Lyonia lucida 28 38 17 12 16 16 16 17
Quercus minina 26 40 23 0 3 15 23 12
Licania michauxii 5 6 1 3 11 11 5 10
Lyoniafruticosa 13 0 4 3 5 6 6 6
Brevaria racemosa 8 2 2 2 5 5 5 5
Ilex glabra 13 4 4 2 5 5 5 5
Smilax auriculata 5 4 1 0 4 4 4 4
Gaylussacia dumosa 24 22 0 2 4 2 0 3
Rhus copallinum 0 0 0 0 3 1 0 3
Pinus palustris 3 0 1 1 1 1 0 1
Pinus clausa 0 0 0 0 0 0 0 0
Pinus elliottii 2 0 0 0 0 0 0 0
Quercus nigra 0 0 6 1 0 0 0 0
Quercus sp 3 12 21 1 0 0 0 0
Ceratolia ericoides 3 6 4 0 0 1 1 0
Osmanthus americanus 0 1 0 0 0 0 0 0
alix caroliniana 0 0 0 0 0 0 0 0
Smilax spp 1 0 0 0 0 0 0 0

Cladonia evansii 8 12 21 0 0 0 0 0
Opuntia humifusa 1 0 0 0 0 0 0 0










Table 3-11. Pre- and postburn absolute mean percent cover of herb and woody species in 2M in
Cedar Key Scrub State Reserve. Absolute mean percent cover for control sites 5A
and 5D are also shown. d = days. M = Months.
Control Time after burning
Species 5A 5D Pre 12 d 3 M 6M 9M 12 M


Herbaceous
Galactia elliottii
Poaceae
Solidago odora
Galactia mollis
Agalinis ilifolia
Asclepias sp
Carphephorus corymbosus
Clitoria mariana
Crotalaria rotundifolia
Woodwardia virginica
Zamia pumila

Woody
Serenoa repens
Quercus myrtifolia
Lyoniaferruginea
Quercus geminata
Quercus chapmanii
Vaccinium myrsinites
Lyonia lucida
Gaylussacia nana
Lyoniafruticosa
Ilex glabra
Licania michauxii
Myrica cerifera
Pinus clausa
Pinus elliottii
Pinus palustris
Quercus minina
Quercus nigra
Quercus sp
Brevaria racemosa
Ceratolia ericoides
Gaylussacia dumosa
Osmanthus americanus
Rhus copallinum
Salix caroliniana
Smilax auriculata
Smilax spp

Cladonia evansii
Opuntia humifusa


2.06 1.76
0.47 0.00
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.26
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.09


23.95 24.39
20.07 27.36
4.22 5.70
6.75 0.64
2.89 5.97
3.55 2.80
8.94 18.00
0.28 1.60
1.48 0.00
2.28 0.00
0.04 0.00
0.39 1.32
0.00 0.00
0.00 0.00
0.00 0.00
0.59 0.27
0.00 0.00
0.00 0.00
1.44 0.72
2.71 7.49
0.43 0.06
0.00 0.00
0.00 0.00
0.00 0.00
0.10 0.00
0.00 0.00

1.05 0.00
0.00 0.00


0.00
0.15
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00


24.40
22.21
6.27
4.96
2.61
2.80
7.30
0.63
0.02
0.44
0.00
1.36
0.00
0.00
0.00
0.00
0.03
0.00
0.20
3.53
0.00
0.00
0.00
0.00
0.00
0.00


0.04 1.61 0.08 0.13 1.23
0.00 0.16 0.39 0.33 0.55
0.00 0.00 0.05 0.00 0.30
0.00 0.06 0.00 0.00 0.11
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.05 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00


1.00 13.34 20.07 21.57 21.64
0.73 14.17 12.47 14.48 18.68
0.00 3.24 5.98 5.26 11.18
0.15 3.98 5.13 4.63 8.58
0.62 2.09 2.35 2.97 4.07
0.00 1.06 1.20 1.43 2.76
0.00 1.81 3.18 3.11 2.62
0.15 1.70 1.84 0.61 1.75
0.00 0.10 0.30 0.66 0.77
0.00 0.90 0.54 0.46 0.74
0.00 0.20 0.07 0.00 0.36
0.00 0.06 0.45 0.19 0.32
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.03 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.09 0.14 0.00
0.00 0.00 0.22 0.14 0.00
0.00 0.02 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00


0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00 0.00


''










Table 3-12. Pre- and postburn absolute importance values of herb and woody species in 2M in
Cedar Key Scrub State Reserve. Absolute importance values for control sites 5A


and 5D are also displayed. d = days. M


months.


Control Time after burning
Species 5A 5D Pre 12 d 3 M 6M 9M 12 M


Herbaceous
Galactia elliottii
Poaceae
Solidago odora
Galactia mollis
Woodwardia virginica
Agalinisfilifolia
Asclepias sp
Carphephorus corymbosus
Clitoria mariana
Crotalaria rotundifolia
Zamia pumila

Woody
Quercus myrtifolia
Serenoa repens
Quercus geminata
Lyoniaferruginea
Gaylussacia nana
Vaccinium myrsinites
Quercus chapmanii
Lyonia lucida
Lyoniafruticosa
Myrica cerifera
Licania michauxii
Ilex glabra
Quercus minina
Brevaria racemosa
Smilax auriculata
Gaylussacia dumosa
Rhus copallinum
Pinus palustris
Pinus clausa
Pinus elliottii
Quercus nigra
Quercus sp
Ceratolia ericoides
Osmanthus americanus
Salix caroliniana
Smilax spp

Cladonia evansii
Opuntia humifusa


0.1457
0.0418
0.0000
0.0000
0.0000
0.0000
0.0026
0.0000
0.0056
0.0000
0.0000


0.5225
0.3733
0.2049
0.1353
0.0115
0.2186
0.1680
0.3038
0.0839
0.0581
0.0267
0.1136
0.1165
0.0473
0.0157
0.1566
0.0000
0.0083
0.0000
0.0056
0.0000
0.0086
0.0402
0.0000
0.0000
0.0032


0.1055
0.0164
0.0000
0.0059
0.0000
0.0053
0.0000
0.0088
0.0000
0.0000
0.0033


0.6254
0.3338
0.1288
0.2377
0.1897
0.1309
0.1805
0.4868
0.0000
0.0500
0.0176
0.0202
0.1275
0.0128
0.0106
0.0781
0.0000
0.0000
0.0000
0.0000
0.0000
0.0305
0.0900
0.0025
0.0000
0.0000


0.0000
0.0280
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000


0.6288
0.4283
0.2928
0.2293
0.0809
0.2592
0.1568
0.2327
0.0203
0.0843
0.0031
0.0292
0.0811
0.0084
0.0029
0.0000
0.0000
0.0027
0.0000
0.0000
0.0183
0.0706
0.0566
0.0000
0.0000
0.0000


0.1984 0.1693
0.0302 0.0466
0.0080 0.0165
0.0072 0.0076
0.0000 0.0090
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000


0.7094 0.7121
0.5528 0.4175
0.1955 0.3298
0.1267 0.2028
0.3220 0.2923
0.0308 0.1303
0.4753 0.2108
0.1654 0.1492
0.0338 0.0468
0.0046 0.0466
0.0152 0.0453
0.0095 0.0642
0.0000 0.0079
0.0080 0.0228
0.0000 0.0113
0.0127 0.0133
0.0000 0.0080
0.0040 0.0026
0.0000 0.0000
0.0000 0.0000
0.0043 0.0000
0.0046 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000
0.0000 0.0000


0.0962 0.0638 0.1869 0.0000 0.0000
0.0028 0.0000 0.0000 0.0000 0.0000


0.0780 0.0141 0.1276
0.0475 0.0486 0.0559
0.0172 0.0084 0.0268
0.0000 0.0027 0.0094
0.0029 0.0000 0.0069
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000


0.6230 0.6674 0.5620
0.4904 0.5130 0.4062
0.3247 0.3195 0.3300
0.2787 0.2708 0.3166
0.2696 0.1933 0.2280
0.1541 0.2078 0.2234
0.2253 0.2328 0.2023
0.1735 0.1726 0.1803
0.0599 0.0737 0.0838
0.0634 0.0594 0.0577
0.0424 0.0167 0.0451
0.0519 0.0516 0.0446
0.0412 0.0714 0.0302
0.0239 0.0291 0.0203
0.0116 0.0119 0.0116
0.0051 0.0000 0.0089
0.0026 0.0000 0.0078
0.0027 0.0000 0.0024
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000
0.0000 0.0000 0.0000

0.0000 0.0000 0.0000
0.0000 0.0000 0.0000









Table 3-13. Density of ramets in 5C after prescribed burning in Cedar Key Scrub State Reserve.
Time after burning
11 3 6 9 12
Species days Months Months Months Months
Herbaceous
Poaceae 0.09 0.61 1.15 0.9800 1.9000
Solidago odora 0.00 0.07 0.05 0.0500 1.8000
Galactia elliottii 0.00 2.66 1.26 0.0000 1.3500
Gallactia moilis 0.00 0.04 0.05 0.0000 1.2200
Crotalaria rotundifolia 0.00 0.08 0.07 0.0700 0.0300
Agalinisfilifolia 0.00 0.00 0.00 0.0000 0.0000
Asclepias sp 0.00 0.00 0.00 0.0000 0.0000
Carphephorus
corymbosus 0.00 0.00 0.00 0.0000 0.0000
Clitoria mariana 0.00 0.00 0.00 0.0000 0.0000
Woodwardia virginica 0.00 0.00 0.00 0.0000 0.0000
Zamiapumila 0.00 0.00 0.00 0.0000 0.0000

Woody
Quercus myrtifolia 0.00 44.13 41.98 46.2500 40.5300
Quercus geminata 0.00 12.82 12.28 12.8600 11.1900
Lyoniaferruginea 0.00 7.74 9.81 10.6400 11.0100
Lyonia lucida 0.00 7.68 8.28 8.3200 8.3300
Quercus chapmanii 0.00 9.70 8.28 7.0400 6.1100
Serenoa repens 1.80 3.31 4.32 4.3400 4.9400
Ilex glabra 0.00 4.57 4.84 4.6700 3.4700
Licania michauxii 0.00 2.65 2.43 0.9000 2.3400
Vaccinum myrsinites 0.00 0.97 1.43 1.4400 1.6900
Lyoniafruticosa 0.00 1.09 1.70 2.0600 1.5500
Gaylussacia dumosa 0.00 1.69 1.41 1.4600 1.4000
Gaylussacia nana 0.00 1.32 1.19 0.2800 1.1300
Brevaria racemosa 0.00 1.89 1.97 1.6200 1.0200
Myrica cerifera 0.00 0.42 0.76 0.6800 0.4900
Rhus copallinum 0.00 0.25 0.28 0.0000 0.3100
Pinus clausa 0.00 0.00 0.03 0.0800 0.2500
Smilax auriculata 0.00 0.03 0.11 0.1700 0.1200
Quercus minina 0.00 0.30 0.38 0.3500 0.0900
Pinuspalustris 0.00 0.00 0.00 0.0000 0.0800
Pinus elliottii 0.00 0.00 0.01 0.0200 0.0500
Quercus nigra 0.00 0.00 0.00 0.0000 0.0000
Quercus sp 0.00 0.02 0.00 0.0000 0.0000
Ceratolia ericoides 0.00 0.00 0.00 0.0000 0.0000
Osmanthus americanus 0.00 0.00 0.00 0.0000 0.0000
Salix caroliniana 0.00 0.03 0.02 0.0000 0.0000
Smilax spp 0.00 0.00 0.00 0.0000 0.0000









Table 3-14. Density of ramets in 2M after prescribed burning in Cedar Key Scrub State Reserve.
Time after burning
12 3 6 9 12
Species days Months Months Months Months
Herbaceous
Galactia elliottii 2.48 3.37 0.59 0.4000 2.5000
Poaceae 0.02 0.55 0.51 0.5000 0.7700
Solidago odora 0.02 0.08 0.07 0.0200 0.3500
Woodwardia virginica 0.00 0.30 0.04 0.0000 0.2100
Gallactia moilis 0.12 0.13 0.00 0.0000 0.0700
Agalinis filifolia 0.00 0.00 0.00 0.0000 0.0000
Asclepias sp 0.00 0.00 0.00 0.0000 0.0000
Carphephorus
corymbosus 0.00 0.00 0.00 0.0000 0.0000
Clitoria mariana 0.00 0.00 0.00 0.0000 0.0000
Crotalaria rotundifolia 0.00 0.00 0.00 0.0000 0.0000
Zamiapumila 0.00 0.00 0.00 0.0000 0.0000

Woody
Quercus myrtifolia 7.85 28.08 30.18 29.7700 20.1700
Lyoniaferruginea 1.00 7.68 12.50 12.9500 11.7900
Gaylussacia nana 5.23 16.54 15.82 10.1900 11.5400
Quercus geminata 1.50 13.20 13.39 12.8600 11.0600
Lyonia lucida 3.42 6.47 7.79 7.4500 10.3000
Vaccinum myrsinites 0.29 2.69 4.29 8.0600 9.6700
Quercus chapmanii 4.52 8.14 9.67 8.7000 6.5500
Lyoniafruticosa 0.74 3.30 4.42 5.0100 6.1700
Serenoa repens 1.64 2.83 3.20 3.2500 3.1700
Ilex glabra 0.00 3.21 3.26 3.2200 2.3100
Licania michauxii 0.14 1.37 1.49 0.3400 1.6900
Brevaria racemosa 0.00 1.04 1.07 1.4100 0.8200
Myrica cerifera 0.03 0.77 1.05 1.0600 0.7600
Smilax auriculata 0.00 0.13 0.16 0.1100 0.2000
Gaylussacia dumosa 0.18 0.29 0.00 0.0000 0.1700
Quercus minina 0.00 0.03 0.33 0.8800 0.1100
Rhus copallinum 0.00 0.04 0.01 0.0000 0.0500
Pinus clausa 0.00 0.00 0.00 0.0000 0.0000
Pinus elliottii 0.00 0.00 0.00 0.0000 0.0000
Pinuspalustris 0.00 0.00 0.02 0.0000 0.0000
Quercus nigra 0.02 0.00 0.00 0.0000 0.0000
Quercus sp 0.03 0.00 0.00 0.0000 0.0000
Ceratolia ericoides 0.00 0.00 0.00 0.0000 0.0000
Osmanthus americanus 0.00 0.00 0.00 0.0000 0.0000
Salix caroliniana 0.00 0.00 0.00 0.0000 0.0000
Smilax spp 0.00 0.00 0.00 0.0000 0.0000









Table 3-15. Tree mortality in quadrats and in grids 5C and 2M in
Reserve.


Cedar Key Scrub State


Pinus clausa
Pinus elliottii
_ Pinuspalustris
Quercus myrtifolia
z Quercus chapmanii
Quercus geminata
Lyoniaferruginea
Ceratolia ericoides
Subtotal
Total


Trees
Before Burning
5C 2M
1 0
3 0
0 1
43 21
13 4
15 9
35 53
0 1
110 89


Dead Trees
After Burning
5C 2M
0 0
3 0
0 1
0 3
3 0
0 0
4 0
0 1
10 5


Mortality
Percentage
5C 2M
0.0 0.0
100.0 0.0
0.0 100.0
0.0 14.3
23.1 0.0
0.0 0.0
11.4 0.0
0.0 100.0
9.1 5.6


SPinus clausa 18 0 14 0 77.8 0.0
Pinus elliottii 108 1 96 1 88.9 0.0
Pinus palustris 47 53 44 53 93.6 100.0
Subtotal 173 54 154 54 89.0 100.0
Total 227 208 91.6











Coefficient of determination (r2) resulting from Detrended Correspondence Analysis
of the pre- and postburn sites of a long-unburned scrubby flatwoods in Cedar Key
Scrub State Reserve.
r squared


Data
Absolute Density



Absolute Mean % Cover



Relativized Density



Relativized Mean % Cover



Absolute Density



Absolute Mean % Cover


Axis
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3


2M Woody


Relativized Density


Relativized Mean % Cover


Site


Table 3-16.


5C Woody


Increment
0.8440
0.0760
0.0300
0.6260
0.3600
-0.0520
0.7260
0.2320
0.0270
0.8790
0.0980
-0.0040
0.9290
0.0430
0.0050
0.2570
0.6590
0.0380
0.9090
0.0560
0.0230
0.5480
0.3830
-0.0090


Cumulative
0.8440
0.9200
0.9500
0.6260
0.9860
0.9340
0.7260
0.9590
0.9860
0.8790
0.9770
0.9720
0.9290
0.9720
0.9770
0.2570
0.9160
0.9540
0.9090
0.9650
0.9880
0.5480
0.9310
0.9220










Table 3-17. Summary statistics of the Multi-response Permutation Procedure for woody absolute
densities and mean percent cover between control and treatment sites at preburn and
12 months postburn in Cedar Key Scrub State Reserve. Results are given for
Euclidean and Sorensen distances.
Variable Distance Observed d Expected Variance Skewness T p
Absolute Euclidean 46.9029 50.8919 0.0159 -0.7659 -31.6566 < 0.001
Density Sorensen 0.7039 0.7569 0.0000 -0.4809 -35.9750 < 0.001
Absolute Euclidean 73.2009 73.6672 0.0611 -0.8182 -1.8872 0.0469
Cover Sorensen 0.7608 0.7703 0.0000 -0.7717 -4.0362 0.0014
Code: Observed = Observed delta. Expected = Expected delta. Delta is the weighted average
distance within-group distance. T is the value of the T test statistics.










Multiple comparison for absolute densities and mean percent cover between control
and treatment sites at 12 months postburn and between preburn and 12 months
postburn values of treatment sites in Cedar Key Scrub State Reserve.


Absolute Density


Sites
5A vs 5C post
5A vs 2M post
5D vs 5C post
5D vs 2M post
5C pre vs 5C post
2M pre vs 2M post
5A vs 5C post
5A vs 2M post
5D vs 5C post
5D vs 2M post
5C pre vs 5C post
2M pre vs 2M post


T
-17.5538
-12.9969
-22.5284
-24.4070
-20.0190
-12.0040
-17.9852
-13.8534
-26.5362
-29.5114
-15.4409
-10.0212


p-value
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001


Absolute Cover
T p-value
-0.358058 0.26219348
-0.449258 0.24910358
-3.657711 0.00845920
-4.644292 0.00223107
0.151180 0.42555303
-0.018410 0.38794972
-1.145694 0.11978385
-1.447951 0.08799726
-4.079046 0.00532423
-6.554787 0.00020113
0.224953 0.45691273
-0.622968 0.21616540


T is the value of the T test statistics.


Table 3-18.










Table 3-19. Comparison among several studies and Cedar Key Scrub State Reserve (CKSRR) regarding common variables measured
in each research. The data presented for bareground and for plant species is mean percent cover. Data for Schmalzer and
Hinkle's study average all strata. Data for CKSSR is the average of the two treatment sites.
Schmalzer & Hinkle 1992 Abrahamson 1996 Greenberg 2003 Silva-Lugo


Oak saw palmetto scrub


Scrubby flatwoods


Sand pine scrub


Scrubby flatwoods


Bareground
Mean height (m)
Species richness
G. elliottii
Q. myrtifolia
Q. geminata
Q. chapmanii
L. ferruginea
L. lucida
S. repens
V myrsinites


Pre
0.00
1.08
8.50
0.00
36.00
15.30
6.70

17.60
31.80
1.50


6M
22.90
0.32
10.50
1.70
17.70
11.70
4.30

10.20
18.00
2.30


12 M
14.60
0.50
10.40
0.00
18.40
10.60
4.20

13.80
30.40
2.40


Pre 12 M Rec Pre
0.00
1.00
18.60 13.50 10.00
0.00
58.73
4.80 4.80 3.0 2.67
15.20 15.00 3.0 0.98
0.84


2.75
21.70
0.52


2.90
18.00
1.00


5M 16M Rec Pre
15.00 25.00 1.00
1.00 0.90 0.4 3.53
12.30 14.80 18.50
0.00 0.00 0.00
22.67 39.59 1.3 25.32
1.14 2.23 1.6 3.83
0.33 0.93 1.3 2.68
0.20 1.10 1.0 5.79
5.82
0.33 0.30 1.8 26.53
1.59


The citation for Abrahamson 1996 is Abrahamson & Abrahamson 1996. Codes: Pre = preburn.
(years). Empty cells mean data not available.


M = months. Rec.


recovery time


6M
3.80
0.97
22.00
0.10
14.65
4.82
2.41
3.91
2.76
19.70
0.77


12 M
6.80
1.08
22.50
0.87
21.82
8.11
3.34
7.14
3.10
22.32
1.67


Rec






1.0
1.0
1.0
1.0



1.0


































B





















Figure 3-1. Quadrats in Cedar Key Scrub State Reserve. A) Quadrat placement in the scrub. B)
Location of flags in a quadrat.
































B



















Figure 3-2. Ramets of Quercus myrtifolia in Cedar Key Scrub State Reserve. A) Twelve days
after burning. B) Twenty nine days postburn.

































B




















Figure 3-3. Ramets of Quercus chapmanii in Cedar Key Scrub State Reserve. A) Twelve days
after burning. B) Twenty nine days postburn.











A























B





















Figure 3-4. Ramets of Lyoniaferruginea in Cedar Key Scrub State Reserve. A) Twelve days
after burning. B) One year postburn.







































B
























Figure 3-5. Ramets of Gaylussacia nana in Cedar Key Scrub State Reserve. A) Apparently,
there are three individuals. B) These are sprouts from the same root.













5C preburn 2M preburn

5C preburn 5A

5C preburn 5D

2M preburn 5A S

2M preburn 5D

5A-5D 5

0.00


I I


-I-I-I-I,


--____


0.10 0.20 0.30 0.40

Jaccard's Simlarity


0.50 0.60 0.70 0.80

Coefficient


5C preburn 2M preburn

5C preburn 5A

5C preburn 5D

2M preburn 5A

2M preburn 5D

5A-5D

0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90

Sorensen's Similarity Coefficient


Figure 3-6. Similarity coefficients among the four study sites in Cedar Key Scrub State Reserve.
A) Jaccard's similarity coefficient. B) Sorensen's similarity coefficient.



















1.25


1.00






e
d
0.75

n

D


t

n 0.50
e






0.25








0.00
5255555555555 5555525 555525 5555252 5252525 555522 5525555 555555 52555555555555 2 2 5 5 22 25
AMACACCCCCDDCACCCCMADCCCMCCDCCMCM CMCMAMDDDDDMMCDMDDDCCCDDCDDMDDDDDAAACCCCMMDDMMMD
3132131412521 1234311 1222145179 1 442647231 2232238441451 342451 1213311 2342444422321 32
7260 827 0659057 652708808 6448 9 26 362978 4276043673 4003 9698524371 1941 322

sample_no




Figure 3-7. Median linkage dendrogram for herb and woody species in treatment and control sites under preburn conditions in Cedar

Key Scrub State Reserve.




















+ -I-


+ +
++
+ -F-
+ ++
,+ ++

+ Jr


++
+4
+


-3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13

Can1
CLUSTER + + + 1 + + 2




Figure 3-8. Scatterplot of the first two canonical axes corresponding to the cluster analysis with Median linkage fusion method for
herb and woody species in treatment and control sites under preburn conditions in Cedar Key Scrub State Reserve.


















1 .25


A

e
r 1.00







a 0.75-





e
t


e 0.50
n

C
u
U

t
e
r 0.25-








0.00

AMACCCCCDAMCCCACCDACCDDDMCMN DCCCMCDDCCMAMDCMCDCCMCDCMCMMDMDDDDADDDMMMDACDDCCCDDDM
3 1 1 3 1 25 1 1 234323144413224411222 1 5244572 3 1 264179 1 42 1 4 23441 22323 1 2213321 21344523
70 827 5 9 0 5 5 4 7 9 2 3 7 6 9 4 9 1 1 6 5 2 7 0 8 9 3 8 0 2 6 8 9 8 6 4 0 6 4 7 8 4 2 4 3 6 2 8 30 0 3 1 2 6 7 743 6 6 0 22

sampleno




Figure 3-9. Average linkage dendrogram for herb and woody species in treatment and control sites under preburn conditions in Cedar

Key Scrub State Reserve.































+ 4 -





+ ++
+ A

+ + ++
+


-3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Can1
CLUSTER + + + 1 + + 4 2




Figure 3-10. Scatterplot of the first two canonical axes corresponding to the cluster analysis with Average linkage fusion method for

herb and woody species in treatment and control sites under preburn conditions in Cedar Key Scrub State Reserve.
















Ward Linkage Dendrogram for All Species Data Set


0.20










S 0.15
e
i

P
a
r
t
I





S
q

a
r
e
d
0.05









0.00


525555555255555522525 5255522555555552222555555555555555255555 5 2 22525555555555255
AMACDADCCMCCCCDDMMAMCCMDDCMMDDDDDCCCMMMMDDDDCCCACCCCCDAMAACCCCDCMMAMDDDCCCDDDDMDC
313221122151242421234571426422311479142314521341231415113423443443221122342331218
726265278 60993 32680 48 2362484 647876074360 827 5 52473711128969090549 003

sampleno


sample_no


Figure 3-11. Ward's minimum-variance linkage dendrogram for herb and woody species in treatment and control sites under preburn

conditions in Cedar Key Scrub State Reserve.


I




































+ ++
+ 4+


4-1
*
4
+

+


-14-
4- 4

4- 4- -I4


+ +


+

+


Can1
CLUSTER .1 + + + 2


sampleno


Figure 3-12.


Scatterplot of the first two canonical axes corresponding to the cluster analysis with Ward's minimum-variance linkage

fusion method for herb and woody species in treatment and control sites under preburn conditions in Cedar Key Scrub

State Reserve.


+
4-I
4 4-


4- 4-












Serenoa repens Sites


Medians 4


Vacinium myrsinites Sites


2M pre 5A
ans 10 5


Figure 3-13. Duncan's multiple comparisons for the median abundances of Serenoa repens and
Vaccinium myrsinites among treatment and control sites in Cedar Key Scrub State
Reserve.


5A 5C pre 2M pre


5C pre
2













120.00


--Bareground
--Litter
Debris


100.00


80.00

U
60.00
f1@

2 40.00


20.00


0.00










120.00


100.00


80.00


S60.00


: 40.00


20.00


0.00


Figure 3-14. Preburn and postburn mean percent cover of bareground, litter, and debris in Cedar
Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and 5D are
included.


- -- U13F-------







N; s I
^ &6 ^ g^^ *


--Bareground
-m-Litter
Debris












4.50


4.00

3.50



Z 2.50

S200 --

I 1.50

1.00

0.50

0.00


.9 .t OC .t



Figure 3-15. Preburn and postburn vegetation height in 5C and 2M in Cedar Key Scrub State
Reserve. Control values for 5A and 5D are included.













3.00



r 2.00

E 1.50

1.00 -

o 0.50 -

000 -

q$ N> 49 qe,^<^ ^
'b t Q


350 -

3.00

2 50

g 2.00

1.50

1.00

o 0.50

0.00 -


[]- / /^ '-


--Poaceae
-- Galactia elliottii
Solidagoodora
Galactiamollis
--- Grotalaria rotundifo lia


-*- Poaceae
--- Galactia elliottii
Solidagoodora
Galactia mollis
--Woo dward ia virginica


Figure 3-16. Preburn and postburn absolute densities of the most abundance herb species in
Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and
5D are included















40

35



I 25 ---------f ----- ^
2 30



25
'S 20
d

15
" 10
Q-

. 5 -_. _____

0 /

,"\ac* N
ni AV


35

S30 --

U 25

' 20

15

0
0


5

0 -- -- -"


--- Poaceae
-m- Galactia elliottii
Solidago odora
Galactia mollis
-- Grotalaria rotun difolia


--- Poaceae
- -- Galactia elliottii
Solidago odora
Galactia mollis
---Woodward iavirginica


Figure 3-17. Preburn and postburn absolute frequencies of the most abundance herb species in
Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and
5D are included.





















2.00 "



S 1.50
U

I 1.00



0.50



0.00 -


F ^ / '/


2.50



2.00



S1.50
U

I 1.00



0.50



0.00 -


*U___T_


-*- Poaceae
--- Galactia elliottii
Solidagoodora
Galactia mollis


-*-Poaceae
- Galactia elliottii
Solidagoodora
Galactia mollis


Figure 3-18. Preburn and postburn absolute mean percent cover of the most abundance herb
species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values
for 5A and 5D are included.













0.2000

0.1800

0.1600

0.1400

0.1200

S0.1000

0 0.0800

0.0600

0.0400

0.0200


.000 --- -* *

0 0000 y


0.2000

0.1800

0 1600

0.1400

0.1200

0 1000

0 0.0800
E 0.0600

0.0400

0.0200

0.0000


U



U


--Poaceae
- Galactiaelliottii
Solidagoodora
Galatia mollis
- Crotalaria rotundifolia


---Poaceae
--- Galactiaelliotti
Solidagoodora
Galactia mollis
-- Woodwardiavirginica


Figure 3-19. Preburn and postburn absolute importance values of the most abundance herb
species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values
for 5A and 5D are included.


I A


-

-












50.00

45.00

40.00

35.00

30.00

25.00

20.00

15.00

10.00

5.00

0.00








50.00

45.00

40.00

35.00

30.00

25.00

20.00

15.00

10.00

5.00

0.00


g T i- ^ -- 7 -' '
OP^t ei;
S ^ 40


-- Quercus myrtifolia
---- Serenoa repens
Querous geminata
Lyoniaferruginae
Lyonia lucida
---Quercus chapmanii
--eVaccinium myrsinites


't C: 4;p

J; R5


Figure 3-20. Preburn and postburn absolute densities of the most abundance woody species in
Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and
5D are included.


-*-Querous myrtifolia
---Serenoa repens
Querous geminata
Lyoniaferruginae
x Lyonialucida
---Querous chapmanii
-i-Vaccinium myrsinites













50

45

4 40
S-4-- Quercus myrtifolia
S-- Serenoarepens
t 30 ---M Quercus geminata

25 Lyoniaferruginae

S20 --- x Lyonialucida

c 15 --- Querus chapmanii
S10 __ ----- Vaccinium myrsinites

5










q=
5 Lyoni.ferruginae
50 Bonia l







15 0--- Burcuschapmanil
45





u 3- 0 Pr n ad pstb n aQuercbus geminata
0 -
25 i Lyoniaferruginae

S Ce20 Key Scrb Lyonialucida
5D are iuerc chncludedapmani
2 10 \.i -i--Vaccinium myi~r,,,.i.

5
0r ----I-----------------------------

0 i i







Figure 3-21. Preburn and postburn absolute frequencies of the most abundance woody species in
Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and
5D are included.















30.00


25.00
250 -*-Quercus myrtifolia

20.00 -B-- Serenoarepens
Sx \ Quercus geminata
15.00 Lyoniaferruginae

M x Lyonialucida
10.00 -*--Quercus chapmanli

5.0 ----Vaccinium myrsinites
5.00












30.00 B


25.00
250 --Quercus myrtifolia

20.00 --- Serenoarepens
SQuercus geminata
15.00 Lyoniaferruginae

S' Lyonia lujcida
E10.00 a -- Q-uercus chapmanii

5.00 ----Vaccinium myrsinites
5.00 ----I-


0.00








Figure 3-22. Preburn and postburn absolute mean percent cover of the most abundance woody
species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values
for 5A and 5D are included.


0'~













0.9000

0.8000

0.7000

0.6000

0.5000

0.4000

0.3000

0.2000

0.1000

0.0000


4

-.., .-->'- --+

-s r s o s4


^ ^ ^ ^ ^


e .-



,,.-
B -






i i i i i


!? cC c c


Figure 3-23. Preburn and postburn absolute importance values of the most abundance woody
species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values
for 5A and 5D are included.


---Quercus myrtifolia
-- Serenoa repens

Querous geminata

Lyoniaferruginae

X Lyonialucida

---Quercus chapmanii

---Vaccinium myrsinites


0.9000

0.8000

0.7000

0.6000

0.5000

0.4000

0.3000

0.2000

0.1000

0.0000


--*-Quercus myrtifolia

-w- Serenoa rep ens

Quercus geminata

Lyoniaferruginae

Lyonialucida

-a--Querus chapmanii

--Vaccinium myrsinites



























~ b ~R-" ~U



S i- ~s
-C, o F2 i


/


- Poaceae
-*- Galactiaelliottii
Solidago odora
Gallactia moilis
-- Crotalaria rotundifolia


--- Poaceae
- -- Galactia elliottii
Solidagoodora
Gallactiamoilis
--- Woodwardiavirginica


/ 4
Rb^


Figure 3-24. Preburn and postburn absolute ramet density of the most abundance herb species in
Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M.


=I
2-50

a- 200

E 1-50

1.00

0 0.50

0.00


3 50


0 00


Kid~


K















50.00

45.00

S40.00
35.00

S30.00o --*-Quercus myrtifolia
--Serenoarepens
25.00
Quercus geminata
20.00 Lyoniaferruginae

15.00 Lyonialucida
C ---Quercus chapmanii
c 10.00 ------
o 0 ----Vaccinum myrsinites
5.00

0.00 ----- -- --










5000 B

45.00

3 4000

3500
--Quercus myrtifolia
30.00 --- --Serenoarepens

S25 00 Quercus geminata

E 20.00 / Lyoniaferruginae
Lyonialucida
S15.00 / .
-a --Quercus chapmanii
S10.00 -- Vaccinum myrsinites

5.00 -- Gaylussacia nana

0.00 ----









Figure 3-25. Preburn and postburn absolute ramet density of the most abundance woody species
in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M.
















25


20

4 15
5 5C
10 --- 2 M






5A 5D Preburn 11days 3 6 9 12
Months Months Months Months
Pre- and postburn time




Figure 3-26. Preburn and postburn species richness in treatment sites in Cedar Key Scrub State
Reserve. Species richness for control sites 5A and 5D are also displayed.





























2.00


0.00











3.00


2.50

-I
pt 2.00


S1.50
c
a
C
I 1.00
.-
W'


-*-5C
----2M


Preburn 11 days 3 Months 6 Months 9 Months 12


U


Preburn 11days 3 Months 6 Months 9 Months 12 Months

5A 51 Pre-and Post-burn Time


Figure 3-27.


Preburn and postburn species diversity in treatment sites in Cedar Key Scrub State
Reserve. A) Simpson's index. B) Shannon-Wiener's index. Species diversity for
control sites 5A and 5D are also shown.


















0.8000

0.7000 *

0.6000 U

0.5000

------

0.3000 5-

02000

0.1000

0.0000
Prebum 11 days 3 Months 6 Months 9 Months 12 Months

5A 5D Pre-and Post-burn Time



Figure 3-28. Preburn and postburn evenness in treatment sites in Cedar Key Scrub State
Reserve. Evenness for control sites 5A and 5D are also displayed.



















































































Figure 3-29.


mogd B












Prourn
3 9 MontUlhs





months


Months
40 80
Axis I




Detrended correspondence analysis (DCA) sample ordination for densities in 5C in
Cedar Key Scrub State Reserve. A) DCA carried out with absolute densities. B)
DCA done with relativized densities.


Preburn
O


80-






( 0 12 Months

O 9 Months

40 /



O 6 Months






0 O 3 Months
0 40 80
Axis 1


12 M
0























































3 months B









80
riobburn

40
9 Months



00-




0 40 80
Axig 1



Figure 3-30. Detrended correspondence analysis (DCA) sample ordination for mean % cover in
5C in Cedar Key Scrub State Reserve. A) DCA carried out with absolute mean %
cover. B) DCA done with relativized mean % cover.














O 12 Months








O 9 Months



6 Month/ Preburn







O 3 Months
Axis1


80








S40 /



20
LI 172 NMnlI


0 3 2men lth
0 40 80
Axis 1


Figure 3-31. Detrended correspondence analysis (DCA) sample ordination for densities in 2M in
Cedar Key Scrub State Reserve. A) DCA carried out with absolute density. B) DCA
done with relativized density.


















































B


D
80









10







S Bnths
Praburn 6 MonthsO r_ n12 Monlhs 3onts

0 40 80
Axis 1

B
Figure 3-32. Detrended correspondence analysis (DCA) sample ordination for mean % cover in
2M in Cedar Key Scrub State Reserve. A) DCA carried out with absolute mean %
cover. B) DCA done with relativized mean % cover.


3 Months
0


80








Preburn
40 0






12 Months
S D6s Months

0) 9I S Months
0 40 80
Axis 1







































0 50 100 150
AXIS 1


200 250 300 350


Figure 3-33.


Stand and species ordination of oak-saw palmetto scrub based on preburn absolute
mean percent cover in Kennedy Space Center. Codes: 11 = Lyonia lucida, as =
Aristida strict, qm = Quercus myrtifolia, qch = Quercus chapmanii, qg = Quercus
geminata, Ifr = Lyoniafruticosa, br = Brevaria racemosa, mc = Myrica cerifera, sr
= Serenoa repens, ig = Ilex glabra, pb = Persea borbonia.


250


200

150

100

50

0

-50

-100


-150

-200


-250
-5


Ifr br mc
*qg


sr ign
*qch
mqm


*as

SII


pb"
i i I I I I I


I I I I I I I


0


0


































qch
qm Ig
I sr
*
qmi


vm







_________* Q_______


Axis 1


Figure 3-34.


Site and species ordination of scrubby flatwoods based on preburn absolute mean
percent cover in Cedar Key Scrub State Reserve. Codes: 11 = Lyonia lucida, If =
Lyoniaferruginea, Ifr = Lyoniafruticosa, qm = Quercus myrtifolia, qch = Quercus
chapmanii, qmi = Quercus minima, qg = Quercus geminata, vm = Vacinium
myrsinites, mc = Myrica cerifera, sr = Serenoa repens, ig = Ilex glabra,









CHAPTER 4
SMALL MAMMALS RESPONSES TO PRESCRIBED FIRE

Introduction

Even though prescribed fire is the primary method of fuel reduction in the United States, the

effects of controlled burns on fauna are not well understood (Pilliod et al. 2003). Of the five

groups of vertebrates, mammals have received more attention. The effects of prescribed fire on

amphibians have been summarized by Russell et al. (1999), Bury et al. (2002), and Pilliod et al.

(2003). The effects of prescribed burning on birds were reviewed by Smith (2000). General

effects of prescribed fire on mammals were summarized by Bendell (1974), Lyon et al. (1978),

Wright and Bailey (1982), Peek (1986), and Landers (1987), and general effects of prescribed

fire on small mammals were reviewed by Ream (1981) and Smith (2000).

Ream (1981) presented an annotated bibliography of 237 papers, and a very brief summary

about prescribed fire effect on Sorex spp. (shrews), Sylvilagus spp. (rabbits), Lepus americanus

(snowshoe hare), Castor canadiensis (beaver), Eutamias spp. (chipmunks), Spermophilus spp.

(ground squirrels), Tamiasciurus hudsonicus (red squirrel), Glaucomys spp. (flying squirrel),

Thomomys spp. (pocket gophers), Peromyscus maniculatus (P. maniculatus), and Cledhi it,,uiuy,

spp. and Microtus spp (voles).

Smith (2000) presented direct and indirect effects of fire, both wildfire and prescribed

burning together in a single discussion, although the majority of the references were prescribed

burning studies. He stated that the direct effect of fire (injury or mortality) is lower than the

indirect effect through habitat modification. Fires generally do not kill or kill a very small

proportion of small mammals because during a fire they use underground refugia, where

adequate ventilation is essential for animal survival (Bendell 1974). Immediately after fire, some

species leave their habitats and emigrate because of lack of food and cover in the burned area.










Other species immigrate to take advantage of the altered habitat. The length of time before

species return depends on how much fire altered the habitat structure and food supply. Thus,

each species is likely to respond differently to fire and subsequent habitat changes. Actually,

most of the literature about the relationship between fire and small mammals is about how

vegetation changes affect their populations. These changes have been mainly reported in terms of

abundance and densities following fire. But, little is known about other demographic factors that

may be essential for understanding population responses. These are the most general statements

about small mammals' responses to fire made by Smith (2000). Another point of view about the

effects of prescribed burning is present below.

Even though the literature presents a variety of responses of small mammals to prescribe fire,

there are some aspects that can be synthesized. The majority of the studies cited were of short

duration (<30 months) and direct mortality was rarely documented (Tevis 1956, Chew et al.

1959, Taylor 1981, Ver Steeg et al. 1983, Singer and Schullery 1989, Harty et al. 1991); some

studies indicated no change in abundance after prescribe burning (Arata 1959, Cook 1959, Writz

1977, Kaufman et al. 1983, Jones 1990, Ford et al. 1999, Vreeland and Tietje 1998); the

majority of the studies, however, showed a positive response (population increase) to prescribed

fire in almost all habitats (Tevis 1956, Shadowen 1963, Hatchell 1964, Ahlgren 1966, Lawrence

1966, Beck and Vogl 1972, Stout et al. 1971, Kreftin and Ahlgren 1974, Layne 1974, Wirtz

1977, Hon 1981, Kaufman et al. 1982, McGee 1982, Bock and Bock 1978, 1983, Gunther et al.

1983, and Forde et al. 1984, Kaufman et al. 1988a, Wirtz et al. 1988, Jones 1990, Blanchard

1991, Sullivan 1995b, 1995b, Greenberg et al. 2006) with population increases attributed to an

increase of abundance in seeds and/or insects (Tevis 1956, Ahlgren 1966, Hooven 1973, Layne

1974, 1990, McGee 1982, Halford 1981, and Gunther et al. 1983). Some studies reported small










mammal use of refugia during and after a patchy prescribed burning (MacGee 1982, Schwilk and

Keeley 1998); dispersal to unburned areas were described (Tevis 1956, Arata 1959, Lee 1963,

Komarek 1965, 1969, Odum et al. 1974, Wirtz 1977, Blankenship 1982, Forde et al. 1984, Wirtz

et al. 1988), and dispersal to burned areas was documented in several studies (Cook 1959,

Gashwiler 1959, Hatchell 1964, Ahlgren 1966, Sims and Buckner 1973, Krefting and Ahlgren

1974, Layne 1974, Schramm 1983, Martell 1984, Vacanti and Geluso 1985, Kaufman et al.

1988b, 1990, Monroe and Converse 2006). Differences in fire regime and management practices

(fire in combination with logging, chopping, clearcutting, and mowing), habitat types,

topography, and climate among studies make generalizations difficult to draw. However, there

are two aspects about the effects of prescribed burning that are relevant because of their direct

connotations for researchers and land managers. These aspects are the use of surrounding un-

burned habitats as refugia and the influence of the re-growth of the vegetation on re-colonization.

Few studies have indicated that adjacent unburned habitats might be used as refugia and

population recovery is due to plant species regrowth several months/years after prescribed fire.

Goatcher (1990) and Blanchard (1991) carried out research of the possible use of stream-terrace

hardwood forest as refuge for Peromyscus gossypinus in Lee Memorial Forest, Baton Rouge,

Louisiana. They used live-trap capture, radio telemetry, and fluorescent tracking pigments to

monitor movements before, during, and after prescribed burning. No movements across the fire-

break were detected by these methods. They concluded that P. gossypinus apparently does not

use stream-terrace hardwood forest as refuge after prescribed fire in adjacent pine forests.

However, more research is needed because several studies have documented that small mammals

temporarily leave burned sites. If they temporarily abandon burned sites, most likely they use

other habitats as temporary refugia. Also, the following studies have indicated that small










mammals recolonize burned sites after the regrowth of the vegetation. P. maniculatus and

Spermophilus armatus (Uinta ground squirrel) populations approached control numbers after

three years following a spring burning, when total cover of the understory was near control levels

in Burro Hill, Bridger-Teton National Park, Wyoming (McGee 1982). The impact of fire on

small mammal communities in the central Appalachians in Pennsylvania was transitory, and the

differences in small mammal abundance between unburned and burned sites disappeared within

eight months after fire. This rapid recovery of small mammal populations was explained by the

fast regrowth of ground cover within the study area (Kirkland et al. 1996). This link between

small mammals and regrowth of the vegetation was also used as explanation for population

recovery in the study conducted by Ahlgren (1966) in Minnesota and Sullivan and Boateng

(1996) in British Columbia. Both studies found an increase in P. maniculatus on burn sites

following fire and a decrease in Clethi/ iiiuniny gapperi (southern red-backed vole) numbers 2-3

years following fire until recovery of the ground cover vegetation occurred. Neotoma mexicana

(Mexican woodrats) benefits from thinning and/or prescribed burning that encourages shrub

densities in the long-term by reducing canopy cover in Coconino National Forest, Arizona

(Converse et al. 2006). Research on recovery of small mammal populations relative to

development rate of vegetation structure following fire is highly needed (Taylor 1981). In

Florida, unfortunately, few studies have been conducted to provide evidence of the impact of

prescribed fire on small mammals, and no study has ever evaluated the importance of adjacent

habitats to burned sites as refugia. In addition, little documentation has been reported on how

closely population recovery is linked with the regrowth of the vegetation.

Only 11 studies have evaluated the effects of fire (nine prescribed burning and two

wildfires studies) on small mammals in Florida. Arata (1959) found no change in the










composition of the populations of P. floridanus, Peromyscus polionotus (old-field mouse), and

Sigmodon hispidus before and after burning in longleaf/turkey oak habitat in north-central

Florida. Also, he reported that S. hispidus moved from burned to unburned areas, while P.

polionotus and P. floridanus stayed in the burned areas. Therefore, prescribed fire did not have a

detrimental effect on these species. Even though Vogl (1973) trapped only during five nights and

obtained a low trapping success, he stated that the densities of Peromyscus gossypinus, S.

hispidus, and Blarina brevicauda (short tailed shrew) appeared to be similar in the burned and

unburned hardwood forest in north Florida (Gannet Pond, Leon County). Gates and Tanner

(1988) found no apparent effect of prescribed burning on Geomyspinetis (pocket gopher) at the

successional stages of sandhill communities in Ordway-Swisher Preserve (OSP, Putnam County)

in north-central Florida. Jones (1989, 1990) found that three populations of Podomysfloridanus

had little or no mortality due to prescribed fire, and populations were higher on burned areas that

on unburned sites in longleaf/turkey oak habitat in OSP. In addition, she found that all

individuals except one did not move out of the burned area after prescribed burning. Layne

(1974) reported a population increase in P. gossypinus and S. hispidus after a wildfire in

slash/longleaf pine habitat in north-central Florida and suggested that burned areas could act as

"dispersal sinks." Layne further stated that the reappearance of Reitl, i,,,r1inuiy, humulis

(Eastern harvest mouse) and S. hispidus on the burned area appeared to be correlated with

redevelopment of the ground cover. Layne (1990) also conducted a long-term monitoring of P.

floridanus population in sand pine scrub at Cedar Key Scrub State Reserve (CKSSR), Levy

County. He found that the species survived a heavy wildfire in 1955 and was still present in

1986; however, absolute density and relative abundance declined 10 years after fire. Layne

(1990) also reported that P. floridanus was still present at low numbers in sandhill and scrub










sites that were burned in 1927 in Archbold Biological Station, Highlands County. In comparison,

populations were higher and more stable in similar nearby habitats that were burned periodically.

According to Layne (1992), P. floridanus populations are higher in early successional stages of

scrub and sandhill vegetation following fire. Later, in the absence of fire, populations decline as

habitat structure becomes denser, shadier, and microclimatic conditions more mesic. Even

though Layne's papers (1974, 1990) reported on the effect of wildfires, these papers are included

in this introduction because of the low number of studies on the subject in Florida. Fitzgerald

(1990) stated no significant treatment effects were detected on S. hispidus and P. gossypinus in

dry prairie of Myakka River State Park in southwestern Florida. Jones (1992) reviewed 38 papers

on the effects of fire on Peromyscus and Podomys, and she found that the majority of the papers

described responses ofP. maniculatus, whose numbers increased on burned areas in forest and

grasslands. Other species differed in their response to fire according to the type of habitats. P.

floridanus appeared to have little or no short term effects following prescribed burning, and

abundance equaled or exceeded pre-fire levels after two or three months. Depue (2005) found

that P. floridanus in central Florida increased or recovered to pre-burn levels within six months

following prescribed burning in Bullfrog Creek Mitigation Park; it dropped in numbers following

prescribed fire, but started to increase when the study ended in Split Oak Mitigation Park; and

the decrease in animal numbers remained unaffected by prescribed fire in Chuluota Wilderness

Area. No study has ever determined the response of Ochrotomys nuttalli to prescribed fire in

Florida.

After reviewing the literature on the effects of prescribed burning on small mammals in the

United States, I found that the majority of the studies had methodological problems that did not

permit comparison among them and the possibility of making generalizations. A high proportion










of these studies were short-term (from weeks to one year) and limited in geographic area, fire

behavior characteristics were not measured, they had small sample sizes, and lacked pre-fire

data, experimental control, replication, and randomization. Therefore, these studies are neither

true experimental designs nor quasi-experimental ones that could provide strong causal inference

and domain (James and MacCulloch 1995). According to Pyne et al. (1996), two methods have

been frequently used in fire studies. The most common one is to compare burned with unbumed

(controls) areas, but two important assumptions are not validated. First, treatment and control

must be ecologically similar. Hence, soils, slopes, species composition, vegetation structure, and

fire history need to be similar. Second, fire behavior characteristics must not vary between

treatments. The second common method of study is pre-burn versus post-bum comparisons. In

this type of study, only the second assumption needs to be made and validated. However, this

approach also needs control, replications, and randomization when it is possible. Whelan (1995)

and Russel et al. (1999) recommended that future prescribed fire studies should have more

rigorous experimental designs, including larger sample sizes, pre-fire baseline data, more

carefully selected controls, and better replications. Another aspect of importance among

prescribed fire studies is the way data analysis has been conducted.

The majority of the studies on prescribed burning effects on small mammals focus at the

population level, providing information about changes in abundance indices and densities after

fire. The problem with evaluating fire effects by using abundance indices, such as minimum

number alive or catch per unit effort, is that these indices tend to be biased in their estimates of

the true abundance. They are also biased because they do not account for differences in detection

probabilities that are likely in small mammal studies. These include differences in detection

probabilities between individual animals over time, or in response to an experimental treatment










(Monroe and Converse 2006). In contrast, few studies (Lee and Tietje 2005, Converse et al.

2006, Monroe and Converse 2006) have used other demographic factors such as survival and

recapture probabilities to analyze prescribed fire effects. This approach might be critical for a

better understanding of prescribed burning effects.

Since prescribed fire has been extensively used in Florida as a management tool for

restoring fire-adapted ecosystems (i.g., longleaf pine-sandhill, sand pine scrub, and scrubby

flatwoods), prescribed burning effects on small mammals become very important because of

their roles in ecosystem functions. Small mammals constitute the prey for many forest predators

(Zeilinski et al 1983; Williams et al. 1992). They influence the structure of vegetative

communities through seed predation and dispersal (Vander Wall, 1993; Hollander and Vander

Wall 2004; Schnurr et al. 2004). They play an essential role as dispersers of ectomycorrhizal

fungi (Pyare and Longland 2001). Therefore, more studies on prescribed burning effects are

needed because several small mammal species have restricted geographical ranges, occurs in

only localized habitats that may be vulnerable to management practices, or may be listed under

the Endangered Species Act. In addition, no study has ever assessed the role of adjacent habitats

as refugia, and no study has been conducted to evaluate the immediate (days after burning) and

short-term effects (one month after initiation of plant species growth, 6 months, and one year

after burning) of prescribed fire on small mammal species in Florida. Furthermore, no

information is available on survival rate when small mammals leave the burned area. These

aspects may be critical for population survival and have been overlooked in fire studies. These

topics of research are relevant for fire managers and those responsible for assessing the potential

effects of prescribed burning on rare, sensitive, and endangered small mammal species in

Florida. Managers, researchers, and non-game species and their habitats will all benefit from a










more comprehensive understanding of how small mammal species respond to prescribed

burning.

Objectives and Research Hypotheses

The objectives of this study were to document small mammal responses to prescribed fire, to

evaluate the importance of vegetation surrounding wetlands next to burned sites as refugia, and

to determine whether or not population recovery is linked with the re-growth of the vegetation.

The research hypotheses were the following:

1. P. floridanus, S. hispidus, P. gossypinus, and 0. nuttalli in Cedar Key Scrub State Reserve
use the vegetation surrounding wetlands next to burned sites as temporary refugia after
prescribed burning.
2. Prescribed burning does not have a negative effect on the survival probability of P.
floridanus and S. hispidus in Cedar Key Scrub State Reserve.
Methodology

Trapping Methods

Four 10x10 grids with 10 trap lines were used for capturing, marking, and recapturing mice.

Grids were installed in the scrubby flatwoods of treatment and control sites. Each grid had 100

standard-sized Sherman Live Traps (7.6 cm x 8.9 cm x 22.9 cm) arranged in ten lines with 10

trapping stations each and 15 m between trapping stations. Each trapping station was identified

by a flag. Also, two trap lines with 10 traps each (15 m between traps) were placed between each

grid and the wetland next to it. Trap lines were in the vegetation near the border of each wetland.

The idea was to detect movements between the grids and trap lines. In addition, two traps were

located at the entrance of each burrow found in grids and trap lines, and 4-6 traps were placed

around the small wetlands found inside treatment and control sites. Each trap was baited with a

50-50 mix of crimped oats and sunflower seeds, and polyester was used as nesting material

during periods of cool weather. Palmetto fronds were used to shade and insulate traps. Traps









were checked at sunrise and mid afternoon because of the diurnal activity of S. hispidus, and

traps were left set at all times during the session.

Four trapping sessions were conducted before and after prescribed burning (Figure 4-1).

During the eight trapping sessions, trapping was performed up to a 5-day sequence to avoid

possible loss in body mass associated with capture (Slade 1991). The 3rd trapping session was

carried out right after three hurricanes hit Cedar Key in August-September 2004, and the 5th

trapping session was done after prescribed burning. In treatment and control sites, pre-treatment

data were collected from 03/02/04 to 04/20/05, and post-treatment data were collected from

04/26/05 to 07/19/06. Collection of post-treatment data started 5 days after prescribed burning

and continued at intervals of 3 months after that. In 2006, the 8th trapping session in February,

just 9 months after burning, was cancelled because of the low temperature and rescheduled for

April 2006.

Trapping effort was increased at three opportunities. The first one took place right after the

3rd trapping session, in which five trap lines (10 traps each; 15 m between traps) were installed

on places of higher elevation in/near each treatment and control sites during five nights. The

purpose of this trapping was to recapture 79 individuals marked during the 1st and 2nd trapping

sessions and not recaptured during the 3rd session. The second effort occurred after prescribed

burning, and it was included during the 5th trapping session and in treatment sites only. A trap

line (10 traps) was placed at the east and west side of the grid in 5C and at the east and south side

of the grid in 2M. Also, 18 traps were added to the two trap lines located in the vegetation

surrounding the wetland in 5C and 2M. The idea was to detect movements of marked individuals

from the scrubby flatwoods to wetlands after burning. The third effort was carried out only in

control sites 5D and 5A in June and July 2005, respectively, because these sites were mowed









with the exception of the grid and 150-200 m buffer zone around the grids. Trapping occurred in

half a grid for five nights to determine if mice left the site because of the mechanic activities near

the grid in each site.

Collected data included species, weight, sex, reproductive condition, trap location, and tag

number. Reproductive conditions were as follows: juveniles with varying stages of gray pelage,

subadult (non-breeding) individuals with adult pelage but not evidence of current sexual activity,

and breeding animals with adult pelage and evidence of sexual activity. Sexual activity was

determined in males by testicles in scrotal position and for females by pregnant conditions and

nipple size. Each mouse was identified with two unique ear-tags (National Band and Tag Co.,

Covington, KY) and released at point of capture.

Trapping for predators was carried out to remove them from the grids before the eighth

trapping session in each site. Procyon lotor (raccoon), Urocyon cinereoargenteus (grey fox),

Didelphis marsupialis (opossum), and Mustelafrenatapeninsulae (Florida long-tailed weasel)

were the most problematic predators found in treatment and control sites. From the lst to the 7th

trapping session, predators disturbed traps for not more than 3 days. The strategy was to close

traps and wait 3-5 days for predators to move to other places. However, before the eighth

trapping session, predators were abundant (particularly P. lotor) in treatment sites. Therefore,

trapping predators took place for 7-10 days in each site before trapping small mammals.

Predators were released at a minimum distance of 3.0 km from the grid of capture.

Data Analysis

Due to the small sample size for the capture-recapture dataset, data analysis was only

carried out for P. floridanus and S. hispidus. In addition, the data for treatment sites were

combined for each species, and the same procedure was done for control sites. Data collected

during the extra trapping effort done only in control sites in June-July 2005 were not considered

167









for the analysis. I evaluated the effect of prescribed burning on the survival probability of P.

floridanus and S. hispidus by using information-theoretic model selection and inference

framework (Burnham and Anderson 2002). Also, I used the program MARK 4.3 (White and

Burnham 1999) for testing lack of fit and for estimating the survival and recapture probabilities

by using the Cormack-Jolly-Seber model (Cormack 1964, Jolly 1965, Seber 1965).

Assumptions of the Cormack-Jolly-Seber model

The Cormack-Jolly-Seber (CJS) model is used alone to estimate survival and recapture

probabilities. This model requires information only on the recapture of the marked animals, and

that these individuals are representative of the populations. Several assumptions have to be made

to be able to estimate the parameters associated with this model. These assumptions are as

follows (Williams et al. 2002):

1. Every marked individual at time (i) has the same probability of recapture (pi).

2. Every marked individual immediately after time (i) has the same probability of survival
to time (i+1).

3. Marks are not lost or missed.

4. All samples are taken in a short period, and captured animals are released immediately.

5. All emigration from the sampled area is permanent.

6. The fate of each animal regarding capture and survival probabilities is independent of the
fate of any other animal.

The goodness of fit (GOF) test

The GOF test was used to test the lack of fit of the data to the underlying assumptions of the

CJS model. The survival probability phi was considered constant (.), time dependent (t), group

dependent (g; treatment control), and dependent of the interaction group and time (g*t). The

recapture probability was also considered under the same conditions, and the combination of all

these possibilities added up 16 models. This set of models was used as the candidate model set. I










fitted these models to the data by using the "pre-defined model" option in MARK and carried out

the GOF test for the full time-dependent model phi(g*t) p(g*t). This model is the general model

because it contains the largest number of parameters. The idea was to assess if this model

adequately fit the data, which means verifying whether or not the arrangement of the data meet

the expectation determined by the assumptions underlying the CJS model.

The GOF test was done by using Bootstrap and Release methods. Both methods estimate the

variance inflation factor (c-hat), which quantify the lack of fit of the model to the data or the

amount of under or over dispersion that we have in the dataset. The Bootstrap method provides

two ways to estimate c-hat: (a) the deviance method which divides the observed deviance

(obtained from the summary statistic of the general model) by the mean deviance from the

bootstrap summary statistics, and (b) the c-hat method that divides the observed c-hat (obtained

from the summary statistic of the general model) by the mean c-hat from bootstrap summary

statistic. The Release method presents three tests of which Test 2 and Test 3 check if the 1st and

2nd assumptions are met. I assumed that the 3rd, 4th, 5th, and 6th assumptions were met. C-hat was

estimated by dividing the overall chi-square from Test 2 and Test 3 by the overall degree of

freedom.

I followed Cooch and White's (2006) recommendations regarding which c-hat to choose

between Bootstrap and Release. These authors suggest to choose the largest c-hat value in the

interval 1 < c-hat < 3 in order to make the model selection more conservative and minimizing the

chances of Type II error. A c-hat value in the interval 1 < c-hat < 3 means overdispersion of the

dataset, and an adjustment for lack of fit is needed with MARK by using that particular c-hat

value. So, MARK displays quasi-likelihood Akaike's Information Criterion values (QAIC) after

adjustment. A c-hat > 3 means that the general model does not fit the data. A c-hat = 1 means a









perfect fit and the data do not need any adjustment. A c-hat < 1 means underdispersion of the

data, and Cooch and White suggest to treat a c-hat < 1 as a c-hat = 1. These criteria allowed

choosing a c-hat value for the general model and made the corresponding correction to it and the

candidate model set.

Model comparison, model selection, and hypothesis testing

To compare and select models, the following steps were carried out. First, selecting the most

parsimonious model in the candidate model set by using the Akaike's Information Criterion

(AIC). The most parsimonious model is the one that best explains the variation in the data with

the lowest number of parameters, and it is better supported by the data. AIC is a good and well-

justified criterion for selecting the most parsimonious model and it is considered a robust way of

model selection (Burnham and Anderson 2002). The criteria for model selection were the

following: AIC, Delta AIC, the normalized Akaike weights, model likelihood, the number of

parameters, and model deviance.

The AIC index measured how much the model explained the variation in the data. The

AIC = -21n( (q/data) ) +2K = -2 x model log likelihood of parameters (q) given the data + 2

Nro. Parameters. The model with the lowest AIC value was better supported by the data and

more parsimonious than other models. Since the dataset is small, MARK calculated the corrected

AIC or AICc. Since the candidate model set was adjusted by using the c-hat obtained from the

GOF test, MARK displayed the quasi AICc or QAICc values.

Delta AIC (AAIC): the difference between each model and the one with the lowest AIC

value (the most parsimonious). MARK calculated AAIC, because of the small sample size, and

AQAIC, after adjustment. The following rules of thumb were applied to determine what models

were different or not: (a) If AQAIC, < 2, both models have equally weight in the data. No real










difference between the two models, (b) if 2 < AQAIC, < 7, there is considerable support for a

real difference between the two models, and (c) if AQAIC, > 7, there is a strong evidence to

support to the conclusion of difference between the two models.

The rule of thumb above was used in combination with the normalized Akaike weights.

The weight (wi) for each model was calculated with Equation 4-1:

Cp-AQAIC,
exp ---c
wi= exp- (4-1)
ex< f-AQAIC,


So, wi is the proportion of the data that support a particular model in comparison with all

models. The model with the highest wi would be the best model because it had more support than

any other model. But, to know how much better it was than the next model, the wi of the best

model was divided by the wi of the next best model. This quotient stated how much the best

model was supported by the data than the next best model.

Model likelihood (index of relative plausibility): the ACI weight of the model of interest

divided by the ACI weight of the best model. This quotient indicated how likely a particular

model was in comparison with the best model. It is important to highlight that the AIC approach

helps to select the best model; however, there is an uncertainty of which model is the best model.

The normalized Akaike weights and the likelihood of the model measure this uncertainty.

The number of parameters and model deviance were the last two criteria. The model

deviance is the difference in -21n (q/data) of the current model and -21n (q/data) of the

saturated model. The saturated model is the model with the number of parameters equal to the

sample size.









Second, P. floridanus, the most parsimonious model was phi(t) p(.), and the second and

third best models were phi(t) p(g) and phi(t) p(t), respectively. In S. hispidus, the most

parsimonious model was phi(t) p(.) and the second one was phi(t) p(g) The survival probability

(phi) was only time dependent, and the probability of recapture (p) was constant (.), time

dependent (t), and group dependent (g). Therefore, I considered building other models to test the

prescribed burning and other effects based on these preliminary results.

Third, adding models phi(Flood + Fire) p(., t, g) for P. floridanus and phi (Flood + Fire) p(.,

g) for S. hispidus. Flood and prescribed burning (Fire) are time dependent variables and their

additive effects were modeled with the probability of recapture constant (.), time dependent (t),

and group dependent (g). This combination resulted in nine models for P. floridanus and six

models for S. hispidus that were added to the candidate model set of 16 models for comparison

purposes. Then, the most parsimonious model was selected out of 25 and 22 models for P.

floridanus and S. hispidus, respectively. The covariate Flood was included because three

hurricanes hit Cedar Key and partially flooded treatment and control sites. No previously marked

mice were recaptured after flooding. Since, this covariate had a strong influence in the survival

probability in both species; its additive effect with prescribed burning was modeled by adding

linear constraints to MARK. The basic sequence of steps of building design matrices followed

Cooch and White (2006).

Fourth, checking for the number or real parameters and adjusting them. Even though MARK

estimates the number of parameters, the model structure determines the number of parameters

that are theoretically estimable. However, if the sample size is small, not all theoretically

estimable parameter can be estimated. In addition, when survival and recapture are time

dependent, the terminal parameter is not individually identifiable (Lebreton et al. 1992). Since












the sample size for the mark-recapture data is small in this study, I manually checked the number

of estimable parameters indicated by MARK matched the number of 3 parameters theoretically

estimable for a particular model. If they did not match (one or more 3 parameters were not

estimated), I manually adjusted the number of parameters to the theoretical number.

Fifth, the model selection carried out in the previous steps allowed the testing of the flood

and prescribed burning effect on the survival probabilities in both species and to estimate

survival and recapture parameters. The most parsimonious model provided the most precise and

less biased survival estimates. However, since there was an uncertainty about which model was

the best model; there was also an uncertainty with survival values. Reporting survival estimates

from a single model in the candidate model set, even if it was the most parsimonious model,

ignored model uncertainty. For this reason, survival estimates were reported by using modeling

averaging, which allowed estimating the average of each parameter of interest from the model

set. Modeling averaging takes the estimates of various models, and weights them by using the

normalized AIC weights. The following equation was used (Equation 4-2):

A R A
avg (0) = wi fi
1=1 (4-2)


The average value for the parameter 0 was calculated by multiplying the AIC weight of model i

and the parameter value estimated by that model and adding up all these products in the model

set. Finally, survival estimates from both the most parsimonious model and model averaging

were plotted. The purpose of this comparison was to show similarities or dissimilarities between

estimated parameter from both approaches.










Results

Number of Captured Individuals in Treatment and Control Sites

A total of 182 individuals were marked and recaptured 426 times in 29,340 trapping

nights. Figure 4-2 illustrates the number of captured individuals per species per trapping session

in treatment sites. All species were captured with low numbers (1-9 individuals) at the beginning

of the study. Of 39 individuals marked in the first 2 trapping sessions, I did not recapture any of

them after the hurricanes during the 3rd trapping session and during the additional trapping effort

carried out in upper grounds. Most likely, they died. After the 3rd trapping session, the number of

individuals of S. hispidus and P. floridanus increased through time from 13 to 34 individuals and

from 13 to 21 individuals, respectively, even after prescribed burning. Also, after the 3rd trapping

session, the number of captured individuals ofP. gossypinus was stable (4-7 individuals) even

after prescribed burning. The number of captured individuals of 0. nuttalli did not increase after

prescribed fire, and only one individual was recaptured through time.

Figure 4-3 shows the number of captured individuals per species per trapping session in

control sites. Again, all species were captured in low numbers (1-11 individuals) at the beginning

of the study. Of 39 individuals marked during the first 2 trapping sessions, I did not recapture

any of them after the hurricanes during the 3rd trapping session and during the trapping done in

upper grounds. Probably, they also died. I captured 9 and one new individuals of P. floridanus

and S. hispidus, respectively, during the 3rd trapping session. After the 3rd trapping session, the

number of individuals of S. hispidus increased from 2 to 13 individuals, and the number of

captured individuals of P. floridanus (6-8 individuals) and P. gossypinus (3-6 individuals)

remained relatively stable through time. Only one individual of 0. nuttalli was recaptured after

the 3rd trapping session.









Figures 4-2 and 4-3 present two different patterns. The number of individuals of S.

hispidus and P. floridanus increased in treatment sites only, and this event took place after

flooding and prescribed burning. In contrast, a similar pattern was found for P. gossypinus and

0. nuttalli in treatment and control sites. Obviously, flooding, prescribe burning, or both affected

demography parameters of P. floridanus and S. hispidus.

The extra trapping effort carried on for control sites in June-July 2005 indicated that mice

did not move because of mowing. Mowing the vegetation took 2-3 weeks in each site, and the

mechanical activity was very intense. Even though the mechanical activity produced a loud noise

that could be heard from 500 m, marked mice were not affected by this type of perturbation

because they remained in the grids.

Number of Captured Individuals in Scrubs and Wetlands

Table 4-1 presents the number of captured individuals in the scrubby flatwoods (scrubs

from now on) and in the vegetation surrounding wetlands (wetlands from now on) per trapping

session in treatment and control sites. In treatment sites before prescribed burning (1st- 4th

trapping sessions), 54 (74%) out of 73 individuals were captured in scrubs (see also Figure 4-4).

During the 4th trapping session after the hurricanes, 26 (77%) out of 34 new marked individuals

were captured in scrubs, and 22 (85%) out of the 26 mice were recaptured in wetlands during the

5th trapping session right after prescribed burning. During the 5th trapping session in wetlands,

five mice previously trapped in wetlands were recaptured and 18 new individuals were captured.

Therefore, marked mice moved to or stayed in wetlands after prescribed burning and new

individuals preferred wetlands rather than scrubs. During the 6th and 7th trapping sessions, 90

individuals were captured in wetlands and six in scrubs. The 90 individuals included mainly

previously marked individuals (81) rather than new ones (9). The six individuals found in scrubs

corresponded to one P. floridanus marked in the 6th and recaptured in the 7th trapping session,

175











two new P. floridanus marked in the 7th session, and one P. gossypinus recaptured in the 6th and

7th trapping sessions. Of 46 mice captured in wetlands during the 7th trapping session, 31(67%)

were recaptured in scrubs during the 8th trapping session. During this last trapping session, one

year after prescribed burning, 53 (90%) out of 59 captured individuals were trapped in scrubs.

The 53 individuals included 31 recaptured mice from the 7th trapping session and 22 new

individuals. Also, six new individuals were captured in wetlands. During all trapping sessions in

control sites, 153 individuals were captured, from which 139 (91%) mice were

captured/recaptured only in scrubs (Table 4-1 and Figure 4-5). Therefore, mice returned to the

scrubs in treatment sites after at least 11 months (May 2005-April 2006) following prescribed

burning. These results clearly supported the research hypothesis and indicated that the vegetation

surrounding wetlands provided refuge to the small mammals for at least 11 months following

prescribed burning.

Mice returned to scrubs after plant species offered both cover and food, and this event took

place at least 11 months after prescribed burning. The 7th trapping session occurred in November

2005, and the majority of the mice were still in wetlands. I could not trap at nine months after

burning (January 2006) because of the low temperature. Trapping started in April 2006 at 12

months after burning and mice were recaptured in scrubs. Therefore, mice may have moved to

scrubs in March 2006 or after it. Mice probably did not move to scrubs in January-February

because insect activity was assumed to be low due to the low temperatures and plant species did

not start to produce flowers and fruits until April-May 2006. Most likely, mice found both cover

and food in scrubs after 11 months following prescribed burning and returned to scrubs for this

reason.









Flood and Fire Effects on Podomysfloridanus

The GOF test for P. floridanus is presented in Table 4-2. The Bootstrap c-hat method did not

give any results, which may be due to the small sample size, and the Bootstrap deviance method

provided a c-hat larger than the c-hat obtained from the Release method. Hence, the general

model and the candidate model set were adjusted to the c-hat = 1.1387 in order to be

conservative. This c-hat indicated that there was a little of overdispersion in the dataset.

The additive effect of Flood and Fire had an effect on the survival probability of P.

floridanus. Table 4-3 presents the candidate model set of 16 models adjusted to c-hat = 1.1387.

Model phi(t) p(.) was the most parsimonious model with 73.16% support in the data. But,

because of the first three models had 99.88% support in the data, phi in these models was time

dependent, and p was constant (.), group dependent (g), and time dependent (t), I modeled phi

with the covariates Flood and Fire in combination with p (., g, t). Table 4-4 displays the 25

models after including nine models from the combination phi(Flood + Fire) p(., g, t) and

correcting for the number of parameters. As shown in this table, the top model phi (Flood + Fire)

p(.) has only 29.92% support in the data, and there is no enough evidence to indicate that this

model is different from the 2nd to the 5th model because delta QAIC, < 2.0. However, the additive

effect of Flood + Fire and the covariate Flood had a strong influence on the survival probability

of P. floridanus because this set of models is supported by 61.67% of the data, and phi is

dependent of Flood + Fire and Flood in the first four models. Even though the covariate Fire by

itself does not have support in the data, the effect of fire can be seen by comparing model phi

(Flood + Fire) p(.) with phi (Flood) p(.). Delta QAIC, = 2.83, and this is the Fire effect. The

effect of Flood can be noted by comparing phi (Flood + Fire) p(.) with phi (Fire) p(.). Delta

QAIC, = 30.11. This is the effect of Flood. Model phi(Flood + Fire) p(.) estimated one survival

and one recapture parameters because this was a particular case of time-dependence where

177










trapping sessions with the same Flood and Fire conditions shared the same survival rate (Table

4-5). MARK does not count parameters with standard errors equal to zero in the parameter total.

So, the survival rate for non-flood and non- prescribed fire times was 0.7871 in treatment and

control sites, and 0.00 and 1.00 for flooding and after prescribed fire times, respectively.

Therefore, the additive effect of Flood + Fire had an influence on the survival probability of P.

floridanus.

The time-dependent covariate Flood and Fire did not have an important influence in the

recapture probability. Because models phi(Flood + Fire) p(t) and phi(Flood) p(t) had 31.76%

support in the data and p(t) was present in these two models, I added models phi(Flood + Fire)

p(Flood+Fire), phi(Flood + Fire) p(Flood), and phi(Flood + Fire) p(Fire) to analyze the effect of

the time-dependent covariates Flood and Fire in the recapture probability. Table 4-6 presents

these set of models, and each of them had very little support in the data (< 8.7 %). Consequently,

model phi(Flood + Fire) p(.) still was the most parsimonious model.

Prescribed burning did not have a negative effect and flooding probably negatively

influenced the survival probability of P. floridanus. The results of the current analysis indicated

that phi(Flood + Fire) p(.) was the most parsimonious model among 28 models. The survival

parameters estimated for treatment and control sites from this model revealed that both curves

were similar, and the only difference was for phi4, where prescribed burning apparently

increased survival (Figure 4-6). But, I cannot conclude if this increase was significant or not

because the p parameter was not estimable (Table 4-7). Nevertheless, I can state that prescribed

fire did not have a negative influence on the survival probability of P. floridanus because any of

the models in which Fire was involved alone had support in the data. Furthermore, in the real

scenario, prescribed burning indirectly increased survival because of the role of the vegetation









surrounding wetlands as refugia. Thus, these results support the research hypotheses. In contrast,

flooding probably had a negative effect. The 0 parameter in the model phi(Fire + Flood) p(.) was

not estimable. Statistically, I cannot make any conclusion. But, practically, it is likely that Flood

decreased the survival probability ofP. floridanus to zero before the 3rd trapping session. I did

not recapture 31 P. floridanus marked between the 1st and 2nd trapping sessions after the

hurricanes.

Table 4-8 summarizes the estimated survival parameters for model averaging, and Figure 4-7

displays the estimated survival parameters for model phi(Flood + Fire) p(.) and model averaging

in treatment and control sites. The fact that four curves were quite similar indicated that model

phi(Flood + Fires) p(.) was the best model and provided pretty good estimates of the survival

parameters. The main difference was on phi4, in which the two estimates in treatment sites

differed by 0.032212. However, model averaging provided a better estimate on phi4 because of

13 P. floridanus marked in the 4th trapping session, 12 were recaptured in the 5th trapping

session. So, phi4 should not be equal tol.00.

Flood and Fire Effects on Sigmodon hispidus

Table 4-2 shows the GOF test for S. hispidus. The Bootstrap deviance method gave a c-hat

larger than the c-hat obtained from the Release method, and the Bootstrap c-hat method did not

provide any result probably because of the small sample size. The general model and the

candidate model set were adjusted to the c-hat given by the Bootstrap deviance method (c-hat =

1.5694). This c-hat revealed overdispersion in the dataset.

The covariate Flood and the additive effect of Flood and Fire had an influence on the

survival probability ofS. hispidus. The candidate model set of 16 models adjusted to a c-hat =

1.5694 is shown in Table 4-9. The most parsimonious model was phi(t) p(.) with 69.78% support

in the data, and the first two models had 95.59% support in the data. The survival probability phi

179









in these models was time dependent, and the recapture probability p was constant (.) and group

dependent (g). Consequently, I decided to model phi with the time-dependent variables Flood

and Fire in combination with p (., g). The combination of these models produced six models that

were added to the candidate model set. Table 4-10 shows the 22 models fitted, adjusted to c-hat

= 1.5694, and corrected for the number of parameters. As can be seen in this table, the top

model, phi (Flood) p(.) had 32.51% support in the data, but it was no different from the 2nd to the

4th model because delta QAIC, < 2.0. Of the remaining 18 models, phi(Flood + Fire) p(g) and

phi(t) p(g) had 9.96% and 5.15% support in the data, respectively, but the rest of the models had

a support < 0.37% in the data. In addition, there were considerable evidences for a real difference

between phi(Flood) p(.) and the 5th and 6th models because AQAICc > 2.0. The first four models

had 83.75% supports in the data and it is conformed mainly by the covariates Flood and Flood +

Fire. Thus, only models with Flood and Flood + Fire on the apparent survival rate had a

substantial support in the data. Therefore, these covariates had an effect of the survival

probability of S. hispidus.

The covariate Flood and the additive effect of Flood and Fire apparently reduced the

survival probability of S. hispidus, but Fire did not have any negative effect on the survival rate.

The most parsimonious model phi(Flood) p(.) and the second most parsimonious one phi(Flood

+ Fire) p(.) had 32.51% and 23.77% support in the data, respectively (Table 4-10). In both

models, Flood apparently decreased survival from values such as 0.9082 and 0.9276 to 0.0000

between the 2nd and 3rd trapping sessions (see Table 4-11, Figure 4-8). Also, Fire in model

phi(Flood + Fire) p(.) apparently decreased survival from 0.9276 to 0.7736 between the 4th and

the 5th trapping sessions. The word 'apparently' is used because the P parameters corresponding

to Flood in the two most parsimonious models were not estimable, and Fire in model phi(Flood +









Fire) p(.) was estimable but not significant (Table 4-12). Practically, Flood probably killed all

marked S. hispidus in the study during hurricanes even though it could not be demonstrated

statistically. Six S. hispidus marked during the first two trapping sessions were not recaptured

after the hurricanes. In contrast, the covariate Fire had < 0.186% support in the data in the three

models that stood alone, and it was not significant (Table 4-12). Therefore, prescribed burning

did not have a significant negative effect on the survival probability of S. hispidus. Out of 13

marked rats during the 4th trapping session, 10 were recaptured in the vegetation surrounding

wetlands after prescribed burning. Hence, these results support the research hypothesis.

The estimated survival parameters for model phi(Flood + Fire) p(.) and model averaging are

presented in Table 4-11 and Table 4-13, respectively, and Figure 4-9 displays the comparison.

Even though phi(Flood + Fire) p(.) was not the most parsimonious model, it was compared with

model averaging because this model and model phi(Flood) p(.) had equal weight in the data, and

phi(Flood) p(.) did not contain Fire and model averaging did. The survival parameters estimated

by model phi(Flood + Fire) p(.) were the same for treatment and control sites, with the exception

of phi4, because this was a case of time-dependence where trapping sessions with the same flood

conditions would share the same survival rate. These estimated parameters were similar to the

parameters estimated by modeling averaging with the exception of phi3 and phi4. For phi3, the

value estimated by model phi (Flood + Fire) p(.) in treatment and control sites (0.9275804) was

higher than the value estimated by modeling averaging in treatment (0.8513742) and control

(0.8512919) sites. Nonetheless, these last two values were so similar that are shown as one point

in Figure 4-9. For phi4, model phi(Flood + Fire) p(.) estimated a value for treatment (0.7736407)

lower than for control sites (0.9275804). Model averaging also produced a value for treatment

(0.8413499) lower than for control (0.9037845) sites. But, the difference between treatment and









control sites for model phi(Flood + Fire) p(.) was higher (0.1539397) than the difference for

model averaging (0.0624346). Therefore, model phi(Flood + Fire) p(.) provided good estimates

of the survival parameters with the exception of phi3 and phi4. Parameter phi2 = 0.00 because six

rats marked during the first two trapping sessions were not recaptured after the hurricanes during

the 3rd trapping session.

Trapping Predators

An increased in the number of predators was noticed in all sites before the 8th trapping

session, but mainly in treatment sites. Before prescribed burning, P. lotor, U. cinereoargenteus,

D. marsupialis, and M. frenata sprung traps for no more than three days. The first three species

were identified by their tracks, and the Florida long-tailed weasel was seen on 04/06/05 at 5:37

pm hunting in the grid in 5A. These predators moved to other places after finding traps closed for

3-5 days. However, before the 8th trapping session and coincidently with the return of mice to the

scrubs, predators were trapped in all sites (Table 4-14), and they did not move to other places

after closing traps for 3 days. This problem was solved by trapping predators before small

mammals for 7-10 days and translocating them to other places at least 3.0 km away from the

grids.

Discussion

Low Capture Success

Morgan's study (1998) and the current study shared sites. Morgan trapped on transects and

at burrows in three places named CKA, CKB, and CKC. CKA and CKC are in the 5A site in this

study; but, transects installed in CKA and CKC and the grid installed by this study were in

different places. CKB is the same 5C site in the present study, and trapping occurred at the same

place. Even though trapping did not occur in the exactly same place in 5A site, trapping took

place in scrubby flatwoods in the same study area. Sites 5A and 5C were not burned since 1957

182











(Morgan 1998), and 2M and 5D did not have records of the last burn, but apparently both sites

burned in a wildfire in 1955 (Jeff DiMaggio, per. com.). Morgan trapped in 1997-98, at least 7

years prior to my trapping period and she obtained a high trapping success.

This study found a low capture success even though a major trapping effort was carried

out. Table 4-15 shows the trapping effort of several studies conducted for small mammals in

Florida. Jones (1990) had the highest with 33,000 trapnights. The current study had the next

highest with 29,340 trapnights. Nonetheless, Jones trapped during 60 months, and I trapped

during 31 months. Morgan trapped 2,970 trapnights, and she reported a higher number of

individuals than the present study. Morgan trapped 430 P. floridanus, 275 S. hispiduss, 221 S.

hispidus, and 33 0. nuttalli. I expected to obtain a high capture success due to Morgan's capture

success in the same study area. In addition, Layne (1990) reported peaks in populations in late

winter and early spring, exactly when I did the first trapping session. There was a lapse of at least

6 years between Morgan's and the current study. During this time, vegetation density increased

and structure changed with all the consequences for small mammals' habits and behaviors

associated with these changes.

The age of the vegetation is an important factor because all sites were not burned for at

least 47 years. When fire suppression leads to habitat conversion, P. floridanus populations tend

to be reduced or eliminated (Layne 1990, 1992, Jones 1992). Populations of P. floridanus in

scrub habitat have shown to decline with increased vegetation density (Layne 1990, 1992).

However, this explanation is not valid for S. hispidus and P. gossypinus since they are

opportunistic and able to use other types of habitats (Layne 1974, Eisenberg 1983). But,

vegetation succession without fire with the corresponding increase in vegetation density might










explain a high capture of 0. nuttalli in comparison with Morgan's study. The vegetation in all

sites may be more suitable for 0. nuttalli due to its arboreal habit.

Fire suppression also decreases habitat suitability for Gopheruspolyphemus (Cox et al.

1987), and this aspect influences P. floridanus presence in all sites. Even though P. floridanus

has been cited as a facultative user of gopher tortoise's burrows in scrubby flatwoods (Morgan

1998), the high capture success found in burrows by Morgan was an indication of the close

association between P. floridanus and the gopher tortoise's burrows in this type of habitat. Of 92

burrows found in four sites in this study, only 13 were gopher tortoise's burrows. It is very likely

that 79 burrows were Dasypus novemcinctus (armadillo) burrows because of the circular shape

of the entrances, and traps placed near entrances were not removed.

Small mammals have natural population fluctuations through time. Particularly, population

cycles have been reported for small mammals (Batzli 1992). It is possible that trapping occurred

when the population size for these species was low. A low mast production could be responsible

for a low population level, but there were no data to support this statement. However, I observed

that acorn production was low during 2004-2006 in all sites.

Vegetation age, fire suppression, population cycles, and acorn production or a combination

of these factors probably influence capture success. A combination of these factors maybe

worked together. Other factors such as trapping design and type of bait were not the causes of the

low capture success. The 10x10 grid with traps every 15 m has been used as the standard grid

design in the majority of the small mammal studies in the United States (John Eisenberg, per.

com.). In general, scientists with experience trapping small mammals in Florida consider

trapping success low on grids and between 1% and 5% (John Eisenberg and James Layne, per.

com). However, if we compare Morgan's trapping design (three U-shaped loops of 210 m long









with 14 trapping station separated by 15 m and 45 m between loops) with the current study,

loops cover a smaller area (1.42 ha) than grids (2.25 ha), so I would expect to have a higher

capture success on grids, and Morgan's results were evidence of the opposite. Morgan used

crimped oat as bait with an excellent result. I used a 50-50 mixture of crimped oat and sun flower

seeds, which probably is much better bait than crimped oat by itself. I found that P. floridanus, S.

hispidus, P. gossypinus, and 0. nuttalli ate all the sunflower seeds and not all the crimped oats in

the traps.

Flooding Effect

Hurricanes Charley (9-14 August), Frances (25 August 8 September), and Jeanne (13-28

September) hit the Florida peninsula in 2004. Charley did not hit Cedar Key directly, but its

winds brought rainfall to the area. Frances and Jeanne hit Cedar Key directly and brought a high

precipitation into the area. A total of 372.5 mm fell in Cedar Key during September (data from

http://www.AccuWeather.com). This is 217.6 mm over the monthly average precipitation (154.9

mm). This amount of precipitation inundated wetlands, scrubby flatwoods, and sand pine scrub.

All sites were partially flooded. The approximate percentage of the grids covered by water was

as follows: 5C = 50%, 2M = 60%, 5A = 50%, and 5D = 30%. All sites stayed flooded for at least

2 weeks. Could small mammals survive this amount of precipitation? Did they move to upper

grounds?

Mice only had two options, move to upper ground or die. All sites had upper grounds

inside the grids; therefore, the 3rd trapping session already sampled upper grounds in all of them.

The extra effort involved the upper ground in/near the grids. However, no mice were captured

during this time. The soil in scrubby flatwoods and sand pine scrubs could not absorb the amount

of precipitation that fell in Cedar Key during the impact of the hurricanes. Thus, a high portion

of the preserve was flooded, and this condition strongly impacted the small mammal community.

185










Most likely, 79 terrestrial and arboreal mice marked during the first two trapping sessions in

treatment and control sites died. I do not think that mice had time to move to other upper ground

areas not covered by traps.

Flooding negatively affected the small mammal community independently of their life

form. Even though the effect of flooding on the survival probabilities ofP. floridanus and S.

hispidus was not conclusive, the effect of flooding upon terrestrial species such as P. floridanus,

S. hispidus, and P. gossypinus was expected to be detrimental to their populations. Arboreal

species such as 0. nuttalli should have a better possibility to tolerate this type of disturbance, but

it did not because flooding lasted at least 2 weeks in all sites. No previous survival analysis of

the effect of flooding on small mammals was found, but some references support the negative

effect of a long period of flooding. No detrimental effect upon the population of P. gossypinus

and 0. nuttalli in Texas was recorded when flooding occurred up to 8 days. Flooding for a 3-

week period caused a marked decrease in the populations. This probably happened because

individuals tended to remain within established home range even during long periods of flooding

(McCarley 1959). Peromyscus leucopus (white-footed mice) completely disappeared from

floodplain plots after severe flooding (Blair 1939, Turner 1966). P. leucopus, Microtus montanus

(mountain vole), and Dipodomys ordii (kangaroo rat) generally experience habitat inundation as

catastrophic (Andersen et al. 2000).

Owls hunted snakes during day light hours after the impact of the hurricanes. I observed

three owls hunting near wetlands between 3 pm and 4 pm in 5C and 5D. Probably, the species

was Bubo virginianus (great hored owls) because of the big size and ear tufts. Two of them

already caught a snake. This species is mainly nocturnal and eats rodents. Even though these

observations are not enough to drive a conclusive statement, maybe owls were hunting snakes









during light hours because of the reduction of small mammal populations due to the impact of

the hurricanes.

Population Responses during and after Prescribed Burning

Small mammals hide in burrows to survive and respond to change in cover and food

caused by prescribed burning. Each species, independent of the life form, must retreat to burrows

to survive the passage of the fire front, and it is likely to respond differently to prescribe burning

and subsequent habitat changes. Some species increase in population and others decrease in

populations after prescribed burning. Other species just disappear from burned areas. Some

species avoid recent burns until habitat requirements removed by prescribed burning are

restored. These aspects are discussed in more detail in the following subtitles.

Burrows as refugia during prescribed burning

Probably, the stand-replacing prescribed fire did not cause mortality of the small mammal

community in CKSSR because mice hid in burrows. Fire intensity should be equal in treatment

sites from the experimental point of view. However, this aspect was not as important as expected

because fire intensity was high enough in both sites to remove all above ground vegetation. No

evidence of mice mortality caused by prescribed burning was found after a careful search of both

sites, and this result might be a consequence of the behavior of the species during prescribed

burning. P. floridanus and P. gossypinus are nocturnal species that live in burrows and have to

stay in them to survive the passage of fire. S. hispidus is a diurnal-nocturnal species, and if

individuals are active during the day, they have to move and to hide in burrows or to flee if their

home range is burned. 0. nuttalli is arboreal and nocturnal species that live in above ground

nests and would have to seek refuge in burrows to survive. There were a total of 28 burrows in

each treatment site. Out of 33 individuals of P. floridanus, S. hispidus, and P. gossypinus marked

during the 4th trapping session, 26 individuals were recaptured in wetlands after prescribed

187









burning during the 5th trapping session. Only one individual of 0. nuttalli was marked during the

4th trapping session, and it was recaptured in wetlands after prescribed burning in the 5th trapping

session. Probably this individual also hid in a burrow because otherwise it would not survive.

Burrows increase survivorship because of the insulating characteristic of the soil. Some studies

have shown that temperatures higher than 100 C at the surface of the ground decline in the first

2.5 cm of soil depth to temperatures around 20-30 C in longleaf pine in south-eastern USA

(Heyward 1938), in Australia eucalypt forest (Beadle 1940), in California Chaparral (DeBano et

al. 1998), and in heavy slash fuels after logging a forest (Neal et al. 1965; cited by Whelan

1995). One reason for poor penetration of heat is that convective heat is transferred upward.

Burrows as refugia for small mammals during prescribed fire have been documented in the

literature. Small mammals can survive fires by remaining in their burrows (Tester 1965, Beck

and Vogl 1972; Hendlund and Rickard 1981; Smith 2000). Most species look for refugia

underground, where ventilation inside burrows is vital for animal survival (Bendell 1974). In this

regard, burrows with more than one entrance might be better ventilated than those with one

entrance (Geluso et al. 1986). No mortality ofP. floridanus was found in two sites in Ordway-

Swisher Preserve possibly because of the uneven distribution of litter and bare patches of sands

caused a mosaic effect of varying intensities, and mice were protected in the tortoise burrows

(Jones 1990). However, Jones also reported 87% return rate in one site, and probably, some

mortality occurred after prescribed fire in this site. Burrowing rodents, such as Dipodomys,

survived in substantial numbers after a stand-replacing fire in California chaparral because their

burrows protected them from heat (Quinn 1979). Populations of Townsend's ground squirrels

living in burrows were unaffected by stand-replacing fire in sagebrush-grass community in

southeastern Washington (Hendlund and Rickard 1981). Regarding arboreal species, woodrats









usually suffered relatively high mortality because their nests were above ground (Simons 1991).

However, populations of woodrats were "unexpectedly high in burned areas because burns left

patches of "lightly burned "vegetation in California chaparral and coastal sage scrub, which may

have provided refugia for woodrat populations (Schwilk and Keeley 1998). In summary, during

fire, the majority of the species seek refugia underground or in sheltered places above the

ground.

Prescribed burning effect

The lack of a negative prescribed burning effect found in this study has also been cited in

the literature. Even though three studies have reported survival analysis by using information-

theoretic model selection and inference framework through the program MARK, two of them

indicated that prescribed burning did not have any effect on the species involved. On all plots,

independent of shrub density or burn treatment, the abundance of Neotomafuscipes (dusky-

footed woodrats) increased from 1993 to a peak in 1997, and decreased from fall 1997 to fall

2001 after prescribed fire in California oak woodlands. Apparently, juvenile survival appeared to

be the cause of the population fluctuation in this species. Prescribe fire by itself did not have any

support in the data (Lee and Tietje 2005). In a single prescribed fire in old growth mixed conifer

forest in Sequoia National Park, California, where fire had been suppressed for over a century,

year effects had greater influences on P. maniculatus densities, P. maniculatus age ratios,

Neotomias speciosus (lodgepole chipmunk) densities, and total small mammal biomass than did

prescribed fire effects. Fire by itself had less than 0.01% support in the data (Monroe and

Converse 2006). In ponderosa pine in Coconino National Forest, Arizona, forest thinning

increased densities of P. maniculatus, Tamias cinereicollis (gray-collared chipmunks),

Spermophilus lateralis (golden-mantled ground squirrels), and Neotoma mexicana (Mexican

woodrats), but the combination of thinning and frequent prescribed fire might have reduced

189









small mammal densities (Converse et al. 2006). Prescribed burning was not applied as a

treatment by itself.

Population increases/decreases after prescribed burning

One important aspect to take into consideration for the experimental design is the

homogeneity of treatment and control sites regarding soils, slopes, plant and small mammal

species composition, vegetation structure, and fire history. Particularly, the normal seasonal

variation in numbers of individuals before prescribed burning should be similar in treatment and

control sites in order to detect prescribed burning effects. A high proportion of the studies have

this problem. The current study attempted to determine this aspect by trapping four times before

and after prescribed burning. The interval of time before and after was 13 and 15 months,

respectively. However, the flooding effect caused by the hurricanes did not allow determining

the normal seasonal variation in experimental sites before burning. This natural disturbance was

recorded in conjunction with prescribed burning, which makes the current study unique in this

sense. The effect of flooding previously presented was different from the effect of prescribed

burning.

There was an increase in the number of individuals of P. floridanus and S. hispidus after

prescribed burning. The increase in the number of individuals of P. floridanus and S. hispidus

was clearly identified when I compared treatment and control sites (Figures 4-2 and 4-3). There

was no doubt that prescribed burning forced these individuals to move to the vegetation

surrounding wetlands. However, P. gossypinus and the only individual of 0. nuttalli did the

same, but the number of individuals did not increase through time. So, prescribed burning was

the stimulus to move, but the increase in number of individuals in the vegetation surrounding

wetlands had to deal with other factors such as immigration, food/space availability, and

competition. This dissertation only has data for the first factor. The increase in the number of

190










individuals was due to new adult individuals. No juvenile was captured between the 5th and the

7th trapping session. So, apparently, the burned area attracted these mice but they had to seek

refuge in the vegetation surrounding wetlands because of the lack of cover and food in the

burned area.

The lack of increase in the number of individuals of P. gossypinus was surprising, but the

drastic decline in the number of individuals of 0. nuttalli was expected. The majority of the

studies have shown a positive response of the genus Peromyscus to prescribed burning (Jones

1992). It was astonishing to see that the number of individuals of P. gossypinus that maintained

relatively stable low population levels in treatment and control sites (Figures 4-2 and 4-3).

Maybe, competition for food and space did not allow new S. hispidus to establish in the

vegetation surrounding wetlands. However, the decline in the number of individuals of O.

nuttalli was predictable because of its arboreal life form and the combined effect of flooding and

prescribed burning. Both curves in Figures 4-2 and 4-3 look alike. 0. nuttalli appears to be

susceptible to prescribed burning even though it might hide in burrows to survive the path of fire.

The increase, decrease, or no change in the number of individuals after prescribed burning

has been reported by other studies conducted in Florida. P. floridanus and P. polionotus

increased population after prescribed fire while S. hispidus declined. However, no change in the

composition of the populations of P. floridanus and S. hispidus was recorded before and after

burning in longleaf/turkey oak habitat in north-central Florida (Arata 1959). Densities of P.

gossypinus and S. hispidus appeared to be similar in the burned and unburned hardwood forest in

north Florida (Vogl 1973). Three populations of P. floridanus had little or no mortality due to

prescribed fire, and populations were higher on burned areas than on unburned sites in

longleaf/turkey oak habitat in Ordway-Swisher Preserve. This means that P. floridanus did not










move from the burned area after prescribed burning. Numbers of mice were more constant in the

burn site than in the unburned one. Significantly more mice were caught in burned areas

immediately following prescribed fire. Also, significantly more mice were caught on burned sites

than on unburned burrows (Jones 1989, 1990). P. floridanus populations appeared to have little

or no short term effects following prescribed burning, and abundance equaled or exceeded pre-

fire levels after two or three months (Jones 1992). P. gossypinus and S. hispidus had an increase

in the number of individuals after a wildfire in slash/longleaf pine habitat in north-central Florida

(Layne 1974). Layne also suggested that burned areas could act as "dispersal sinks." This

statement might be true in some ecosystems, but not in others. In the current study in CKSSR,

four P. floridanus and two P. gossypinus were recaptured only in the scrubs after prescribed

burning. Layne (1990) also carried out a long-term monitoring of P. floridanus population in

CKSSR. He reported that the species survived a wildfire in 1955, population declined 10 years

after fire, and the species was still present in 1986. Another study conducted by Layne (1990) at

Archbold Biological Station revealed that P. floridanus was present at low numbers in scrub sites

that were burned in 1927. In contrast, populations were higher and more stable in similar nearby

habitats that were burned periodically. According to Layne (1992), P. floridanus populations are

higher in early successional stages of scrub vegetation following fire. Particularly, high

population numbers can be found in 2 years old scrub (Layne, personal communication). Later in

the absence of fire, populations decline as habitat structure becomes denser and microclimatic

conditions more mesic. However, Morgan (1998) reported a high capture success between 1997

and 1998 for P. floridanus, S. hispidus, P. gossypinus, and 0. nuttalli in CKSSR, and the current

study found a low capture success between 2004 and 2006. This change in population size can be

explained by natural population fluctuations, and probably the high population abundance found










by Morgan after 42 years might be related with mast production. Depue (2005), in central

Florida, found that P. floridanus increased or recovered pre-burn levels within 6 months

following prescribed burning in Bullfrog Creek Mitigation Park; it dropped in numbers following

prescribed fire, but started to increase when the study ended in Split Oak Mitigation Park; and

the decrease in animal numbers remained unaffected by prescribed fire in Chuluota Wilderness

Area. Apparently, there is not a cut-clear pattern by a single species in Florida, and some

variation in the response to prescribed burning might occur within the same region. Therefore,

the same variation might be expected in other studies conducted in other states.

Some species increase population and others decrease populations after prescribed fire in

other studies conducted in the United States. P. maniculatus had a post-fire increase in

population size in California chaparral (Cook 1959) and in northeastern Minnesota (Ahlgren

1966), and these increases were probably related to the increase of seeds of annual grasses

stimulated by fire. Prescribed burning reduced the population of small mammal species with the

exception of P. maniculatus in north-central Pennsylvania. The P. maniculatus established in the

burned area one month following prescribed burning (Fala 1975). Populations ofP. maniculatus

generally increase after fire (Ream 1981). P. maniculatus were more abundant on 1-2-year-old

burns in tallgrass prairie than in unburned areas in eastern Kansas. However, Reilnl ,,/1,mily\

megalotis (western harvest mouse) was more abundant in unburned areas (Kaufman et al. 1982).

Also, Kaufman et al. (1983) reported that R. megalotis densities in the burn site increased the

following spring and summer because the population in the un-burned sites served as a source of

dispersing individuals. P. maniculatus dramatically increased population size one year following

prescribed burning; however, this effect disappeared and reversed itself during the second year in

Ponderosa Pine, South Dakota (Bock and Bock 1983). The total number of small mammals was










lower in burned sites than in unburned ones in shrub-steppe habitat in Idaho. Nonetheless, P.

maniculatus were more abundant in the bum site (Groves and Steenhof 1988). A dramatic

increase in P. maniculatus population on burn sites following prescribed fire in British Columbia

was explained by the rodents' ability to forage for seeds and insects that were greatly increased.

In contrast, southern red-backed vole numbers were decreased for 2-3 years following burning

(Sullivan and Boateng 1996). The studies cited above found stronger support for positive

prescribed fire impacts on P. maniculatus abundances than reported by Monroe and Converse

(2006) in Sierra Nevada mixed conifer forest, California. These authors said that the tendency for

P. maniculatus densities to be greater on burned areas does not necessarily indicate that burned

habitat is optimal for P. maniculatus, and may reflect dispersal to marginal sink habitat. In

California oak woodland, the population of N. fuscipes increased from 1993 to 1997, and then

decreased steadily after prescribed burning (Lee and Tietje 2005). P. leucopus increased

population size in treatment sites (thinning, prescribed fire, and thinning + prescribed burning),

but this increase was higher when the two types of treatments were combined in the southern

Appalachian hardwood forest in North Carolina (Greenberg et al. 2006). A comparison between

three sites (brush, prairie, and savanna) burned frequently during 15 years and three forested sites

unburned during 35 years in western Wisconsin were carried out to study small mammal's

abundance in these sites. P. leucopus and Clethi/ iiu,,uin gapperi (red-backed vole) were more

common in the unburned forest, and P. maniculatus and Spermophilus tridecemlineatus (13-

lined ground squirrel) were more abundant in the prairie created and maintained by fire. Burning

the forest did not significantly reduce the number of mice present (Beck and Vogl 1972). In the

first year after burning in the California grassland and chaparral, populations of Chaetodipus

californicus (California pocket mouse), Peromyscus californicus (California mouse), and









Dipodomys agilis (agile kangaroo rat) were either unchanged or greater on burned than in

unburned sites. However, populations ofR. humulis, Peromyscus boylii (brush mouse), and

Neotoma spp. (woodrat) decreased or disappeared in the burned sites (Wirtz 1977). Populations

of S. hispidus in Arizona grassland were greatly reduced by a summer fire, while populations of

Perognathus hispidus (seed-eating pocket mice) and Dipodomys merriami (kangaroo rats)

increased. This difference was explained by the food habits of the species. S. hispidus fed on

green vegetation that decreased after fire. The heteromyid rodents fed on seeds, which increased

after fire due to the invasion of weedy forbs (Bock and Bock 1978). Even though there was no

evidence that prescribed fire killed any small mammal, it negatively impacted Mus musculus

(house mice), P. maniculatus, and Microtuspennsylvanicus (meadow voles) in Oxford, Ohio. M.

musculus and M. pennsylvanicus disappeared from the burned site, while the three species

prevailed in the control site (Crowned and Barret 1979). Populations of Spermophilus spp.

(ground squirrels) and Thonomys spp. (pocket gophers) generally increase after fire (Ream

1981). Only Dipodomy agilis out of five species of rodents increased in abundance after

prescribed burning in a coastal sage scrub in southern California (Prise and Waser 1984). Six

small mammal species were not eliminated, and they did not increase in numbers in the months

following prescribed burning in the California Chaparral (Lawrence 1966). A year after a

prescribed burning in conifer woodland with shrubby understory in California, the abundance of

small mammals was almost three times greater on unburned than burned sites, even though

species composition did not vary significantly between burned and unburned sites (Blankenship

1982). The populations of the small mammal community (11 species) in the unburned sagebrush

in Burro Hill, Wyoming, varied little before burning, the populations were at low level following

burning, and populations approached control values three years after burning (McGee 1982). The









impact of fire on small mammal communities in the central Appalachians on Pennsylvania was

transitory, and the differences in small mammal abundance between unburned and burned sites

disappeared within eight months after fire (Kirkland et al. 1996). Also, there were not significant

differences in small mammals' mean total captured efficiency between treatment and control

sites in southern Appalachian, North Carolina (Ford et al. 1999). Reduction of shrubs and woody

debris by thinning with overly frequent prescribed fire may reduce small mammal densities

(Converse et al. 2006). In summary, small mammal species' responses to prescribed burning

vary greatly within the same geographic area and among states in the United States.

Habitat selection: immigration, emigration, and returning to burned areas

In general, prescribed burning affects small mammals mainly through the way it affects

their habitats. Direct effects such as injury, mortality, and movement (immigration and

emigration) might be the short-term population responses. Indirect effects through habitat

alteration could influence long-term responses such as feeding, movement, reproduction, and

availability of refugia (Smith 2000). In both circumstances, immigration and emigration play an

important role in population demography, food availability, reproduction, and re-colonization of

the burned areas. Immigration might occur because burned areas attract small mammals;

however, emigration could also take place because there is insufficient food and cover in the

burned area. Characteristics of an animal species such as mobility and particular food and cover

requirements will determine its ability to re-invade a burned site (Whelan 1995). The length of

time before these species return to burned sites depends on how much fire altered the habitat

structure and food supply (Smith 2000). The last three sentences summarize and explain what P.

floridanus and S. hispidus experienced at CKSSR.

Although I did not trap new individuals in burned sites following prescribed burning, there

were data that showed possible immigration to the burned area. A total of 27 new adult

196










individuals were trapped between the 5th and the 7th trapping sessions in wetlands corresponding

to 3 P. floridanus, 20 S. hispidus, 3 P. gossypinus, and one 0. nuttalli. These individuals were

probably attracted by the burned area and moved to wetlands looking for refugia. Immigration to

burned areas immediately after prescribed burning has been cited in the literature. Odors from

burned areas might stimulate immigration of P. maniculatus from suboptimal habitats (Kaufman

et al. 1988b). The high reproductive potential of P. maniculatus and R. megalotis populations in

Kansas tallgrass prairie enables them to increase rapidly in favorable environments and disperse

readily into recently burned areas (Kaufman et al. 1988b). The number of resident individuals of

P. floridanus in burned areas of Ordway-Swisher Preserve was higher than in the unburned ones.

P. maniculatus invaded a burned area after a heavy rain in California (Tevis 1956). Also, P.

maniculatus invaded burned areas immediately after prescribed fire in jack pine in northeastern

Minnesota (Ahlgreen 1966).

The lack of prescribed burning effect on P. floridanus and S. hispidus was probably

because they used the vegetation surrounding wetlands as refugia. Statistically, I could not draw

any conclusion, but practically mice may have moved to wetlands looking for cover and food. If

mice did not have the wetlands near the treatment sites, probably they have had to move farther

until finding another wetland or unburned site. P. floridanus in the current study moved out of

burned areas. In contrast, Jones (1990) reported that all individuals of P. floridanus except one

did not leave burned areas after prescribed burning in Ordway-Swisher Preserve. This is

probably due a patchy burn in the Ordway sandhills in comparison with a more continuous burn

in the CKSSR scrub. A similar result was found by Arata (1959) near Gainesville, Florida. The

number of captured individuals of P. floridanus and P. polionotus remained at pre-burn levels in

burned and unburned sites, whereas S. hispidus moved from the burned to the un-burned site









following prescribed fire. However, Ream (1981) reported that red squirrel, voles, northern

flying squirrel, and showshoe hare emigrated from recent burned areas. Of 25 species common in

chaparral brushlands, Townsend's chipmunk and dusly-footed woodrat were not abundant in

recently burned areas (Biswell 1989). This suggest that species such as P. floridanus found in the

same geographic area might respond differently to prescribed burning depending on the habitat

type and how much fire altered it. Also, other species always emigrate from burned sites, and

most of the time they return to the same burned sites they moved away from months ago.

P. floridanus and S. hispidus returned to the burned sites in CKSSR after at least 11

months. This amount of time appears to be a long time for a mouse to live and survive. Out of 50

P. floridanus (20 individuals) and S. hispiduss (30 individuals) marked between the 5th and the

7th trapping sessions, 33 (66%) individuals (nine P. floridanus and 24 S. hispiduss) were

recaptured again in scrub in the 8th trapping session. The survival curve for both species was high

(Figures 4-6 and 4-8), and the amount of time involved between the 5th and the 8th trapping

session was 373 days for 5C and 403 days for 2M. Are P. floridanus and S. hispidus long lived

species? Jones (1990), at Smith Lake sandhill in Ordway Swisher Preserve, found that 8.6% of

all marked mice (225 individuals) were present for 360 days or more. Of these mice, half were

females first marked as juveniles, and most of the males were first marked as subadults and

adults. The longevity records were 649 days for females and 920 days for males. Layne (1974)

reported two S. hispidus females originally trapped as subadults were recorded on the study area

during the entire 14-month period. A juvenile female and an adult and a juvenile male were first

captured in October 1960 and they were recaptured 10 months later in July 1961.

P. floridanus and S. hispidus in CKSSR returned to the burned sites after they found cover

and food to live there. It is surprising that both species returned approximately at least 11 months










later. But, it is not as surprising when we consider the phenology of the plant species. Accoms of

Myrtle oak, Chapman oak, and sand live oak developed surrounding wetlands in September, and

production ended in December 2005. Thus, P. floridanus and S. hispidus had food to stay in the

vegetation surrounding wetlands. In the burned sites 5C and 2M, the percentage of shrub cover at

11 months after prescribed burning was not quantified, but at 12 months was 72 % and 105%,

respectively. This percent cover was mainly composed of Quercus myrtifolia, Quercus geminata,

Lyoniaferruginea, Lyonia lucida, and Quercus chapmanii (see Tables 3-7 and 3-11). Neither

flowering season nor fruiting season was synchronous in CKSSR within and among species, but

Vaccinium myrsinites developed flowers in March and fruits in April 2006. Gaylussacia nana

and Serenoa repens developed flowers in April, but G. nana had fruits in May and Serenoa

repens in June 2006. Other species such as L. ferruginea, L. lucida, Ilex glabra, and Brevaria

racemosa flowered in April-May and fruited in June-July 2006. Therefore, P. floridanus and S.

hispidus returned to scrubs after plant species offered cover and food.

The relationship between the amount of cover and mice returning to burned areas has been

reported in the literature. Arata (1959) indicated that S. hispidus did not return to longleaf/turkey

oak habitat in north-central Florida at 5 months after a bum. However, he stated that the burned

area was recolonized by S. hispidus within 6 months following the bum. Layne (1974) was the

first study in Florida to report that the return ofS. hispidus to a burned area appeared to be

correlated with redevelopment of the ground cover in slash/longleaf pine habitat in north-central

Florida. Ahlgren (1966) showed that the southern red-backed vole numbers decreased for 2-3

years in Minnesota following prescribed burning until recovery in the ground story vegetation

occurred. McGee (1982) reported that the populations of the small mammal community (11

species) in the unburned sagebrush in Burro Hill, Wyoming, varied little before burning, the










populations were at low level following burning, and populations approached control values

three years after burning when total cover of the understory was near unburned levels. West

(1982) indicated that northern red-backed voles avoided a burned area in black spruce for one

year and established a resident population in post fire year 4, which it was the first year of berry

production in central Alaska. Geluso et al. (1986) found that voles survived a prescribed burning

in Nebraska grassland and left the burned areas until a new litter layer had accumulated about

two growing seasons later. Possible reasons for emigration included decreased protection from

predators and decreased food availability. Kirkland et al. (1996) showed that the impact of fire

on small mammal communities in the central Appalachians on Pennsylvania was transitory, and

the differences in small mammal abundance between unburned and burned sites disappeared

within eight months after fire. This rapid recovery of small mammal populations was explained

by the fast re-growth of ground cover within the study area, particularly of blueberry. Sullivan

and Boateng (1996) reported that southern red-backed vole numbers decreased for 2-3 years in

British Columbia following prescribed burning until recovery in the ground story vegetation took

place. Schwilk and Keeley (1998) carried out a patchy bum in a California chaparral and coastal

sage scrub. These refugia allow small mammals colonize severely burned sites during the first

six months after prescribed fire. Ford et al. (1999) also used the link between small mammals

and re-growth of the vegetation as the explanation of the population recovery in the study

conducted in Southern Appalachian, North Carolina.

In summary, immediately after fire, some species are attracted to burned areas and

immigrate into them. Other species emigrate due to insufficient food and cover in the burned area.

The length of time before species return depends on how much fire altered the habitat structure

and food supply.

















Table 4-1. Number of captured individuals in the scrubby flatwoods and in the vegetation surrounding wetlands per trapping session
in treatment and control sites in Cedar Key Scrub State Reserve.
Trapping Podomys Peromyscus Ochrotomys Sigmodon Total
Session S W T S W T S W T S W T S W T
1 4 3 7 2 1 3 7 1 8 1 0 1 14 5 19
2 3 3 6 3 0 3 7 2 9 1 1 2 14 6 20
3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
E 4 9 4 13 5 2 7 1 0 1 11 2 13 26 8 34
U 5 0 16 16 0 4 4 0 1 1 0 24 24 0 45 45
S 6 1 15 16 1 6 7 0 0 0 0 23 23 2 44 46
7 3 13 16 1 6 7 0 1 1 0 26 26 4 46 50
8 21 0 21 2 2 4 0 0 0 30 4 34 53 6 59
Total 95 Total 35 Total 20 Total 123 113 160 273
1 7 3 10 0 0 0 11 0 11 1 0 1 19 3 22
2 4 4 8 0 0 0 8 0 8 2 0 2 14 4 18
3 8 1 9 0 0 0 0 0 0 1 0 1 9 1 10
o 4 8 0 8 5 1 6 0 0 0 2 0 2 15 1 16
o 5 6 0 6 6 0 6 1 0 1 9 0 9 22 0 22
6 6 0 6 4 1 5 1 0 1 9 0 9 20 1 21
7 7 0 7 3 0 3 0 0 0 11 0 11 21 0 21
8 6 0 6 4 0 4 0 0 0 9 4 13 19 4 23
Total 60 Total 24 Total 21 Total 48 139 14 153
Codes: S = scrubs. W = wetlands. T = Total.

















Table 4-2. The Goodness of Fit test for the general model phi(g*t) p(g*t) for Podomysfloridanus and Sigmodon hispidus.
Species Bootstrap GOF Method Observe Mean c-hat GOF Method Chi-square df c-hat
Pod s Deviance method 12.5835 11.0510 1.1387 R e 0 7 4
Podomys Release 0.9167 4 0.2292
c-hat method 0.0000 0.0000 0.0000
Sigmodon Deviance method 24.4433 15.5750 1.5694 Release 5.6026 9 0.6225
c-hat method 0.0000 0.0000 0.0000










Table 4-3. Result browser of the fitted candidate model set of 16 models for Podomysfloridanus
after adjusting to c-hat = 1.1387 in treatment and control sites in Cedar Key Scrub
State Reserve.


Models
{Phi(t) p(.) PIM}
{Phi(t) p(g) PIM}
{Phi(t) p(t) PIM}
{Phi(g*t) p(.) PIM}
{Phi(g*t) p(g) PIM}
{Phi(g*t) p(t) PIM}
{Phi(t) p(g*t) PIM}
{Phi(.) p(t) PIM}
{Phi(g) p(t) PIM}
{Phi(.) p(.) PIM}
{Phi(g) p(.) PIM}
{Phi(.) p(g) PIM}
{Phi(g) p(g) PIM}
{Phi(g) p(g*t) PIM}
{Phi(.) p(g*t) PIM}
{Phi(g*t) p(g*t) PIM}


QAICc
123.822
125.949
131.653
137.206
139.728
149.739
150.755
153.549
153.681
158.311
158.629
160.259
160.716
165.704
166.047
171.137


Delta
QAICc
0.00
2.13
7.83
13.38
15.91
25.92
26.93
29.73
29.86
34.49
34.81
36.44
36.89
41.88
42.23
47.31


QAICc
weight
0.73160
0.25265
0.01458
0.00091
0.00026
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000


Model
likelihood
1.0000
0.3453
0.0199
0.0012
0.0004
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000


#Par
8
9
13
15
16
21
21
8
9
2
3
3
4
16
15
28


QDeviance
18.930
18.741
14.778
15.239
15.145
11.340
12.356
48.657
46.474
66.533
64.753
66.383
64.708
41.121
44.079
11.051










Table 4-4. Set of 25 models after adding phi(Flood + Fire) in combination with p constant (.),
time dependent (t), and group dependent (g) in the analysis Flood and Fire effect on
survival probabilities of Podomysfloridanus in treatment and control sites in Cedar
Key Scrub State Reserve.


Models
{Phi(Flood+Fire) p(.)}
{Phi(Flood+Fire) p(t)}
{Phi(Flood) p(t)}
{Phi(Flood+Fire) p(g)}
{Phi(t) p(.) PIM}
{Phi(Flood) p(.)}
{Phi(t) p(g) PIM}
{Phi(Flood) p(g)}
{Phi(t) p(t) PIM}
{Phi(g*t) p(.) PIM}
{Phi(g*t) p(g) PIM}
{Phi(Fire) p(t)}
{Phi(.) p(g*t) PIM}
{Phi(g*t) p(t) PIM}
{Phi(t) p(g*t) PIM}
{Phi(Fire) p(.)}
{Phi(.) p(t) PIM}
{Phi(g) p(t) PIM}
{Phi(Fire) p(g)}
{Phi(.) p(.) PIM}
{Phi(g) p(.) PIM}
{Phi(.) p(g) PIM}
{Phi(g) p(g) PIM}
{Phi(g) p(g*t) PIM}
{Phi(g*t) p(g*t) PIM}


QAICc
121.892
122.950
123.393
123.668
123.822
124.720
125.949
126.384
131.653
137.206
139.728
148.336
148.972
149.739
150.755
151.999
153.549
153.681
153.973
158.311
158.629
160.259
160.716
165.704
171.137


Delta
QAICc
0.00
1.06
1.50
1.78
1.93
2.83
4.06
4.49
9.76
15.31
17.84
26.44
27.08
27.85
28.86
30.11
31.66
31.79
32.08
36.42
36.74
38.37
38.82
43.81
49.24


QAICc
weight
0.29916
0.17630
0.14128
0.12308
0.11397
0.07274
0.03936
0.03166
0.00227
0.00014
0.00004
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000


Model
likelihood
1.0000
0.5893
0.4723
0.4114
0.3810
0.2432
0.1316
0.1058
0.0076
0.0005
0.0001
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000


#Par
4
10
9
5
8
3
9
4
13
15
16
9
8
21
21
3
8
9
4
2
3
3
4
16
28


QDeviance
25.885
13.387
16.185
25.494
18.930
30.845
18.741
30.377
14.778
15.239
15.145
41.129
44.079
11.340
12.356
58.123
48.657
46.474
57.966
66.533
64.753
66.383
64.708
41.121
11.051


\ \_ I_\_ I I













Table 4-5. Estimated survival (phi) and recapture (p) parameters by using model phi(Flood+
Fire) p(.) in the analysis Flood and Fired effect on survival probabilities of Podomys
floridanus in treatment and control sites in Cedar Key Scrub State Reserve.


Confidence interval


= 95%.


Estimate
0.7870664
0.0000000
0.7870664
Treatment 1.0000000
0.7870664
0.7870664
0.7870664


Control


0.7870664
0.0000000
0.7870664
0.7870664
0.7870664
0.7870664
0.7870664


15:p Recapture 0.9430430


Phi
1:Phi
2:Phi
3:Phi
4:Phi
5:Phi
6:Phi
7:Phi

8:Phi
9:Phi
10:Phi
1 :Phi
12:Phi
13:Phi
14:Phi


Standard
error
0.0507500
0.0000000
0.0507500
0.0000000
0.0507500
0.0507500
0.0507500

0.0507500
0.0000000
0.0507500
0.0507500
0.0507500
0.0507500
0.0507500

0.0398357


Lower CI
0.6712425
0.0000000
0.6712425
1.0000000
0.6712425
0.6712425
0.6712425

0.6712425
0.0000000
0.6712425
0.6712425
0.6712425
0.6712425
0.6712425

0.7946523


Upper CI
0.8699882
0.0000000
0.8699882
1.0000000
0.8699882
0.8699882
0.8699882

0.8699882
0.0000000
0.8699882
0.8699882
0.8699882
0.8699882
0.8699882

0.9860803













Table 4-6. Set of 28 models after adding phi(Flood + Fire) in combination with p(Flood + Fire),
p(Flood), and p(Fire) in the analysis Flood and Fire effect on survival probabilities of
Podomysfloridanus in treatment and control sites in Cedar Key Scrub State Reserve.


Models
{Phi(Flood+Fire) p(.)}
{Phi(Flood+Fire) p(t)}
{Phi(Flood) p(t)}
{Phi(Flood+Fire) p(g)}
{Phi(t) p(.) PIM}
{Phi(Flood + Fire) p(Fire)}
{Phi(Flood + Fire) p(Flood)}
{Phi(Flood) p(.)}
{Phi(t) p(g) PIM}
{Phi(Flood + Fire) p(Flood +
Fire)}
{Phi(Flood) p(g)}
{Phi(t) p(t) PIM}
{Phi(g*t) p(.) PIM}
{Phi(g*t) p(g) PIM}
{Phi(Fire) p(t)}
{Phi(.) p(g*t) PIM}
{Phi(g*t) p(t) PIM}
{Phi(t) p(g*t) PIM}
{Phi(Fire) p(.)}
{Phi(.) p(t) PIM}
{Phi(g) p(t) PIM}
{Phi(Fire) p(g)}
{Phi(.) p(.) PIM}
{Phi(g) p(.) PIM}
{Phi(.) p(g) PIM}
{Phi(g) p(g) PIM}
{Phi(g) p(g*t) PIM}
jPhi(g*t) p(a*t) PIM}


QAICc
121.892
122.950
123.393
123.668
123.822
123.935
124.059
124.720
125.949

126.138
126.384
131.653
137.206
139.728
148.336
148.972
149.739
150.755
151.999
153.549
153.681
153.973
158.311
158.629
160.259
160.716
165.704
171.137


Delta
QAICc
0.00
1.06
1.50
1.78
1.93
2.04
2.17
2.83
4.06

4.25
4.49
9.76
15.31
17.84
26.44
27.08
27.85
28.86
30.11
31.66
31.79
32.08
36.42
36.74
38.37
38.82
43.81
49.24


QAICc
weight
0.24033
0.14164
0.11350
0.09888
0.09156
0.08652
0.08135
0.05843
0.03162

0.02877
0.02543
0.00182
0.00011
0.00003
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000


Model
likelihood
1.0000
0.5894
0.4723
0.4114
0.3810
0.3600
0.3385
0.2431
0.1316

0.1197
0.1058
0.0076
0.0005
0.0001
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000


#Par
4
10
9
5
8
5
5
3
9

6
4
13
15
16
9
8
21
21
3
8
9
4
2
3
3
4
16
28


QDeviance
25.885
13.387
16.185
25.494
18.930
25.761
25.885
30.845
18.741

25.761
30.377
14.778
15.239
15.145
41.129
44.079
11.340
12.356
58.123
48.657
46.474
57.966
66.533
64.753
66.383
64.708
41.121
11.051










Table 4-7. Estimated 3 parameters from models phi(Flood+ Fire) p(.) in the analysis Flood and
Fired effect on the survival probabilities of Podomysfloridanus in treatment and
control sites in Cedar Key Scrub State Reserve. Confidence interval = 95%.
Standard
Model Parameter 3 error Lower CI Upper CI
Interception 1.3073323 0.3028174 0.7138103 1.9008543
phi(Flood+Fire) p(.) Flood -22.5146010 0.0000000 -22.5146010 -22.5146010
Fire 30.5618730 0.0000000 30.5618730 30.5618730
p 2.8068159 0.7416407 1.3532002 4.2604316










Table 4-8. Estimated survival parameters (phi) by using model averaging for the set of 28
models in the analysis Flood and Fire effect on the survival probabilities of the
Podomysfloridanus in treatment and control sites in Cedar Key Scrub State Reserve.
C-hat = 1.1387; 95% confidence interval.


Estimate
0.7847947
0.0000000
0.8269939
0.9677876
0.8251911
0.8247459
0.7939472

0.7847475
0.0000000
0.8270833
0.8353632
0.8251870
0.8247389
0.7939708


Standard
error
16.2037846
0.0000055
20.3813127
0.0613967
0.0687758
0.0694662
49.0490509

16.2037852
0.0000055
0.0763053
0.0781662
0.0687882
0.0694835
49.0490508


Lower CI Upper CI
0.0000000 1.0000000
-0.0000108 0.0000108
0.0000000 1.0000000
0.3875942 0.9992993
0.6496347 0.9231836
0.6472239 0.9234956
0.0000000 1.0000000


Phi
1:phi
2:phi
3:phi
4:phi
5:phi
6:phi
7:phi

1:phi
2:phi
3:phi
4:phi
5:phi
6:phi
7:phi


1.0000000
0.0000108
0.9315558
0.9392373
0.9231923
0.9235066
1.0000000


Average
Treatment


Average
Control


0.0000000
-0.0000108
0.6269967
0.6248433
0.6495939
0.6471662
0.0000000









Table 4-9. Result browser of the candidate model set of 16 models for Sigmodon hispidus fitted
and adjusted to c-hat = 1.5694 in treatment and control sites in Cedar Key Scrub State
Reserve.


Models
{Phi(t) p(.) PIM}
{Phi(t) p(g) PIM}
{Phi(.) p(.) PIM}
{Phi(g) p(.) PIM}
{Phi(.) p(g) PIM}
{Phi(g) p(g) PIM}
{Phi(t) p(t) PIM}
{Phi(.) p(t) PIM}
{Phi(g) p(t) PIM}
{Phi(g*t) p(.) PIM}
{Phi(g*t) p(g) PIM}
{Phi(.) p(g*t) PIM}
{Phi(g*t) p(t) PIM}
{Phi(t) p(g*t) PIM}
{Phi(g) p(g*t) PIM}
{Phi(g*t) p(g*t) PIM}


QAICc
86.170
88.159
93.408
94.814
95.071
96.668
97.014
99.994
101.753
102.937
105.261
116.520
118.334
118.380
118.632


Delta
QAICc
0.00
1.99
7.24
8.64
8.90
10.50
10.84
13.82
15.58
16.77
19.09
30.35
32.16
32.21
32.46


QAICc
weight
0.69776
0.25817
0.01871
0.00926
0.00815
0.00367
0.00308
0.00070
0.00029
0.00016
0.00005
0.00000
0.00000
0.00000
0.00000


140.013 53.84 0.00000


Model
likelihood
1.0000
0.3700
0.0268
0.0133
0.0117
0.0053
0.0044
0.0010
0.0004
0.0002
0.0001
0.0000
0.0000
0.0000
0.0000
0.0000


#Par QDeviance
8 17.575
9 17.237
2 37.966
3 37.271
3 37.528
4 36.989
13 16.362
8 31.399
9 30.831
15 17.150
16 16.834
15 30.733
21 15.932
21 15.978
16 30.205
28 15.575


\ \_ I_\_ I I











Table 4-10. Set of 22 models after adding phi(Flood + Fire) in combination with p (., g) to the
candidate model set in the analysis Flood and Fire effect on survival probabilities of
Sigmodon hispidus in treatment and control sites in Cedar Key Scrub State Reserve.


Models
{Phi(Flood) p(.)}
{Phi(Flood + Fire) p(.)}
{Phi(t) p(.) PIM}
{Phi(Flood) p(g)}
{Phi(Flood + Fire) p(g)}
{Phi(t) p(g) PIM}
{Phi(.) p(.) PIM}
{Phi(Fire) p(.)}
{Phi(g) p(.) PIM}
{Phi(.) p(g) PIM}
{Phi(Fire) p(g)}
{Phi(g) p(g) PIM}
{Phi(t) p(t) PIM}
{Phi(.) p(t) PIM}
{Phi(g) p(t) PIM}
{Phi(g*t) p(.) PIM}
{Phi(g*t) p(g) PIM}
{Phi(.) p(g*t) PIM}
{Phi(g*t) p(t) PIM}
{Phi(t) p(g*t) PIM}
{Phi(g) p(g*t) PIM}
{Phi(g*t) p(g*t) PIM}


QAICc
84.472
85.098
86.170
86.220
86.838
88.159
93.408
94.797
94.814
95.071
96.463
96.668
97.014
99.994
101.753
102.937
105.261
116.520
118.334
118.380
118.632
140.013


Delta
QAICc
0.00
0.63
1.70
1.75
2.37
3.69
8.94
10.33
10.34
10.60
11.99
12.20
12.54
15.52
17.28
18.46
20.79
32.05
33.86
33.91
34.16
55.54


QAICc
weight
0.32506
0.23770
0.13907
0.13565
0.09962
0.05145
0.00373
0.00186
0.00185
0.00162
0.00081
0.00073
0.00061
0.00014
0.00006
0.00003
0.00001
0.00000
0.00000
0.00000
0.00000
0.00000


#Par
3
4
8
4
5
9
2
3
3
3
4
4
13
8
9
15
16
15
21
21
16
28


Qdev.
26.930
25.419
17.575
26.541
24.986
17.237
37.966
37.254
37.271
37.528
36.784
36.989
16.362
31.399
30.831
17.150
16.834
30.733
15.932
15.978
30.205
15.575










Table 4-11.





Model


phi(Flood) p(.)


phi(Flood+Fire)
p(.)


Estimated survival (phi) and recapture (p) parameters by using model phi(Flood)
p(.) and phi (Flood + Fire) p(.) in the analysis Flood and Fired effect on survival
probabilities of Sigmodon hispidus in treatment and control sites in Cedar Key


Scrub State Reserve. Confidence interval =


Parameter
l:phi
2:phi
3:phi
4:phi
5:phi
6:phi
7:phi
8:phi
9:phi
10:phi
1 :phi
12:phi
13:phi
14:phi
15:p
l:phi
2:phi
3:phi
4:phi
5:phi
6:phi

7:phi
8:phi
9:phi
10:phi
1 :phi
12:phi
13:phi
14:phi
15:p


Estimate
0.9081981
0.0000000
0.9081981
Treatment 0.9081981
0.9081981
0.9081981
0.9081981
0.9081981
0.0000000
0.9081981
Control 0.9081981
0.9081981
0.9081981
0.9081981
Recapture 0.9295392
0.9275804
0.0000000
0.9275804
Treatment 0.7736407
0.9275804
0.9275804

0.9275804
0.9275804
0.0000000
0.9275804
Control 0.9275804
0.9275804
0.9275804
0.9275804
Recapture 0.9268919


95%.
Standard
error
0.0355371
0.0000523
0.0355371
0.0355371
0.0355371
0.0355371
0.0355371
0.0355371
0.0000523
0.0355371
0.0355371
0.0355371
0.0355371
0.0355371
0.0352384
0.0353713
0.0000438
0.0353713
0.1472561
0.0353713
0.0353713

0.0353713
0.0353713
0.0000438
0.0353713
0.0353713
0.0353713
0.0353713
0.0353713
0.0357070


Lower CI
0.8109827
0.0000000
0.8109827
0.8109827
0.8109827
0.8109827
0.8109827
0.8109827
0.0000000
0.8109827
0.8109827
0.8109827
0.8109827
0.8109827
0.8212892
0.8202518
0.0000000
0.8202518
0.3967210
0.8202518
0.8202518

0.8202518
0.8202518
0.0000000
0.8202518
0.8202518
0.8202518
0.8202518
0.8202518
0.8186314


Upper CI
0.9580029
0.7602520
0.9580029
0.9580029
0.9580029
0.9580029
0.9580029
0.9580029
0.7602520
0.9580029
0.9580029
0.9580029
0.9580029
0.9580029
0.9742732
0.9729370
0.7246399
0.9729370
0.9467035
0.9729370
0.9729370

0.9729370
0.9729370
0.7246399
0.9729370
0.9729370
0.9729370
0.9729370
0.9729370
0.9726868










Table 4-12. Estimated 3 parameters from models phi(Flood) p(.) and phi(Flood + Fire) p(.) in
the analysis Flood and Fire effect on the survival probabilities of Sigmodon hispidus
in treatment and control sites in Cedar Key Scrub State Reserve. Confidence interval
= 95%.
Standard
Model Parameter 3 error Lower CI Upper CI
Intersection 2.29 0.4262 1.4564 3.1272
phi(Flood) p(.) Flood -21.14 7999.2294 -15699.6274 15657.3518
p 2.58 0.5380 1.5251 3.6342

Intersection 2.55 0.5266 1.5181 3.5822
Flood -21.58 8080.7068 -15859.7679 -15816.7679
phi(Flood+Fire) p(.) Fire -1.32 0.9970 -3.2752 0.6329
p 2.54 0.5269 1.5071 3.5727











Table 4-13. Estimated survival parameters (phi) by using model averaging for the set of 25
models in the analysis Flood and Fire effect on the survival probabilities of
Sigmodon hispidus in treatment and control sites in Cedar Key Scrub State Reserve.
C-hat = 1.5694; 95% confidence interval.


Estimate
0.9225670
0.0075199
0.8513742
0.8413499
0.9160235
0.9225670
0.9162478

0.9225064
0.0074594
0.8512919
0.9037845
0.9159611
0.9225064
0.9161860


Standard
error
0.0406832
0.0148244
0.1855291
0.1178554
0.0393425
0.0406832
0.0761627

0.0407450
0.0147077
0.1301067
0.0523791
0.0394113
0.0407450
0.0760357


Lower CI
0.7960088
0.0215359
0.2444639
0.4844399
0.8001250
0.7960088
0.6099526

0.7957258
0.0213677
0.4330015
0.7425754
0.7998231
0.7957258
0.6108419


Upper CI
0.9732462
0.0365758
0.9902356
0.9676693
0.9674515
0.9732462
0.9871024

0.9732474
0.0362864
0.9772273
0.9683423
0.9674598
0.9732474
0.9870342


Site


Treatment


Control


1:phi
2:phi
3:phi
4:phi
5:phi
6:phi
7:phi

1:phi
2:phi
3:phi
4:phi
5:phi
6:phi
7:phi













Table 4-14. Predators captured before the 8th trapping session in treatment and control sites in
Cedar Key Scrub State Reserve.
Predator 5C 2M 5A 5D
Raccoon 18 15 6 5
Grey Fox 2 3 1 1
Opossum 7 9 2 3
















Table 4-15. Trapping effort carried out by several studies conducted on small mammals in Florida. Code: TN = trapping nights.


Author
Jones (1990)
Current study
Frank (1996)
Fitzgerald (1990)
Layne (1974)
Franz et al. (1998)
Newman (1997)
Depeu (2005)
S Morgan (1998)
Arata (1959)
Humphrey et al. (1985)
Sasso and Gaines (2002)
Vogl (1973)


Study Area
Ordway/Swisher Preserve
Cedar Key Scrub Preserve
Anastasia Island
Myakka River State Park
North Central Florida
Avon Park
Ordway/Swisher Preserve
3 study areas, central FL
Cedar Key Scrub Preserve
North Central Florida
Ordway/Swisher Preserve
Key Largo
North Florida


Species
Podomysfloridanus
Several
Peromyscus polionotus
Several
Several
Podomysfloridanus
Podomysfloridanus
Podomysfloridanus
Several
Several
Podomysfloridanus
Several
Several


Vegetation Type
Sandhill
Scrubby Flatwoods
Dune
Dry Prairie
Scrubby Flatwoods


Scrub
Sandhill
Scrub, scrubby flatwoods, others
Scrubby Flatwoods
Longleaf Pine/Turkey Oak
Sandhill
Hardwood Hammocks
Hardwood forest


TN
33,000
29,340
23,200
15,608
11,370
8,576
6,084
4,458
2,970
1,950
1,114
66
5








Flooding


2004 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec



-
2005 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Prescribed Burning


2006 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Figure 4-1. Trapping sessions (red blocks) carried out in Cedar Key Scrub State Reserve. Three
hurricanes hit Cedar Key before the 3rd trapping session, and prescribed burning
occurred before the 5th trapping session.




















S25 -- Podomys
o -2- Peromyscus
C5 _______5 -- Ochrotomys



5
z -oSigmodon



0 ---
1st 2nd 3rd 4rd 5th 6th 7th 8th
Trapping Sessions


Figure 4-2. Number of captured individuals per species per trapping session in treatment sites
5C and 2M in Cedar Key Scrub State Reserve.














40 n--------------------------------------|_
40
35

i" 30 Hurricanes
S--- Podomys
25
S--- Peromyscus
20
--- Ochrotomys
4- 15 -
10 .-a-- Sigmodon

1 5



1st 2nd 3rd 4rd 5th 6th 7th 8th
Trapping Sessions


Figure 4-3. Number of captured individuals per species per trapping session in control sites 5A
and 5D in Cedar Key Scrub State Reserve.














Prescribed
burning


1 year after
burning

It


60


50


040




- 20

-o
10


-- Scrubs
--Wetlands


Trapping sessions


Number of captured individuals in scrubs and wetlands per trapping session in
treatment sites 5C and 2M in Cedar Key Scrub State Reserve.


Hurricanes


Figure 4-4.












60


50


U 40
3






-0
> 30
-o

t 20
6
S10


0


-- Scrubs
--Wetlands


Number of captured individuals in scrubs and wetlands per trapping session in
control sites 5A and 5D in Cedar Key Scrub State Reserve.


Hurricanes










------ r---_----__----



1 2 3 4 5 6 7 8
Trapping sessions


Figure 4-5.











1.20


1.00


0.80


0.60


0.40


0.20


0.00


1 2 3 4 5 6 7
Parameter No.


Survival probabilities of Podomysfloridanus estimated by the model phi(Flood +
Fire ) p(.) in the set of 28 models in treatment and control sites in Cedar Key Scrub
State Reserve (chat=1.1387).


--Treatment
- Control


Figure 4-6.















1.00


0.80
Co

0 0.60


; 0.40


0.20


0.00


1 2 3 4 5
Parameter No.


6 7


Podomysfloridanus's survival probabilities calculated by the model phi(Flood +
Fire ) p(.) and by model averaging (28 models) in treatment and control sites in
Cedar Key Scrub State Reserve (chat=1.1387).


Figure 4-7.


---Treat.
phi(Flood+Fire)p(.)
-- Cont.
phi(Flood+Fire)p(.)
-- Treat. Average
Cont. Average











1.20


1.00

4-
=0.80
,0
20.60


.50.40

0)
0.20


0.00


1 2


m m m




4- 1


3 4 5 6 7
Parameter Nro.


Survival probabilities of Sigmodon hispidus quantified by the model phi(Flood +
Fire) p(.) in the set of 22 models in treatment and control sites in Cedar Key Scrub
State Reserve (chat=1.5694).


Figure 4-8.


--- Treatment


-- Control












1.20


1.00


0.80 -Treatment
Sphi(Flood+Fire)p(.)
.m- Control
0 0.60 phi(Flood+Fire)p(.)
-a-- Treatment Average
o Control Average
.>0.40
I-

0.20


0.00 -
1 2 3 4 5 6 7
Parameter Nro.


Figure 4-9. Sigmodon hispidus' survival probabilities estimated by the model phi(Flood + Fire)
p(.) and by model averaging in the set of 22 models in treatment and control sites in
Cedar Key Scrub State Reserve (chat=1.5694).









CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS

Conclusions

Prescribed Burning Plan

The prescribed burning plan carried out in treatment sites 5C and 2M achieved the

management and the research objectives. Burning reduced the vegetative mass on the ground

more than 60%, and top-killed more than 75% of the woody vegetation. Whether this

management action in 5C and 2M will improve the habitat for scrub jays, needs to be evaluated

following this study. Even though the prescribed burning plan was almost the same for each

treatment site, fire behavior characteristics were not exactly the same. Rate of spread and flame

lengths were similar, but fire intensity in 5C was lower and varied more than in 2M, which was

more homogeneous across the site. The change in wind speed during the burn of 5C might be

one factor responsible for this difference. Even though 5C and 2M sites did not have exactly the

same fire behavior characteristics, the outcome of reducing and top killing the above ground

vegetation was the same in both sites. This result made possible the comparison of plant

responses between treatment and control sites.

Plant Species Responses to Prescribed Burning

A cluster analysis in combination with an ordination technique and a F-ratio test (with the

respective multiple comparison test) was used to carry out a site analysis. Statistically, treatment

and control sites in CKSSR were ecologically similar, and they were compared to determine

prescribed burning effects.

Resprouting was the main way of surviving fire and recovery by the majority of the species

in CKSSR. During the first 12 months in treatment sites, the recovery of the vegetation was so

fast that bareground did not have values higher than 13%, and litter achieved more than 69% of









the preburn level. Debris and vegetation height were the variables with the slowest recovery,

having debris 32% and height 31% of the preburn value at 12 months.

The recovery process of herb and woody species was different. The Family Poaceae was

dominant over other herb species and was the only taxon with preburn value. Only Galactia

elliottii and G. mollis recuperated control values at 12 months postburn. Solidago odora,

Crotalaria rotundifolia, and Woodwardia virginica were only recorded with low densities and

cover during the postburn period and the other five species were sampled only in control sites.

Prescribed burning created gaps that were temporarily used by S. odora, C. rotundifolia, and W.

virginica and eventually filled by the growing woody vegetation.

The seven woody species with high densities, frequencies, cover, and importance values in

treatment and control sites were: Quercus myrtifolia, Serenoa repens, Q. geminata, Lyonia

ferruginea, Lyonia lucida, Q. chapmanii, and Vaccinium myrsinites. Q. myrtifolia and S. repens

were the most dominant species. Rare species were: Quercus nigra, Quercus sp., Ceratolia

ericoides, Osmanthus americanus, Salix caroliniana, Smilax sp, and Opuntia humifusa. In

general, the seven dominant species recovered preburn values (except for height) and/or

achieved at least a control value at 3 months with the exception of V. myrsinites. Also, there was

a shift in dominance for Q. chapmanii and L. ferruginea through the 12 month period. Tree

mortality was low in quadrats (7.5%) because oaks and ericaceous shrubs are adapted to fire and

pine trees had very low densities, but pine tree mortality was high in grids (91.6%). The

Detrended Correspondence Analysis indicated that woody species had structural and

compositional changes during the first 3 months postburn, but there were more compositional

than structural changes after that as one of the effects of prescribed burning. According to the

Multi-response Permutation Procedure, the structural changes were significant, which means that









prescribed burning had a significant effect on absolute densities on treatment sites between pre-

and 12 months postburn. V. myrsinites was the first species to have flowers in March and fruits

in April, which was an important factor for small mammals returning to burn sites.

Small Mammal Responses to Prescribed Burning

The combination of Flood and Fire effects affected the survival probability of the Florida

mouse. Model phi(Flood + Fire) p(.) was the most parsimonious model. In this model, even

though I cannot statistically conclude that Flood decreased survival from 0.8732 to 0.0000 in all

sites because the p parameter was not estimable, practically and most likely, marked mice died

during or after flooding. Also, although I cannot statistically state that prescribed fire increased

survival (from 0.7871 to 1.0000) in treatment sites because the 0 parameter was not estimable,

most likely prescribed fire forced the Florida mouse to emigrate from the scrubs to the vegetation

surrounding wetlands, and mice were able to survive. Models in which Fire was involved alone

did not have support in the data. Therefore, prescribed burning did not negatively affect the

survival probability of the Florida mouse because the vegetation surrounding wetlands were used

as refugia after prescribed burning for at least 11 months.

The covariate Flood and the additive effect of Flood and Fire influenced the survival

probability of the cotton rat. Models phi(Flood) p(.) and phi(Flood + Fire) p(.) were the most

parsimonious ones. Statistically, I cannot conclude that Flood in both models decreased survival

from values such as 0.9082 and 0.9276 to 0.0000 between the 2nd and 3rd trapping sessions

because the 3 parameters were not estimable. However, marked cotton rats probably died after

flooding before the 3rd trapping session. Fire in model phi(Flood + Fire) p(.) did not significantly

decrease survival from 0.9276 to 0.7736 between the 4th and the 5th trapping sessions because

the P parameter was not significant. Hence, prescribed burning did not have a significant









negative effect on the survival probability of the cotton rat because the majority of these rats

emigrated from scrubs to the vegetation surrounding wetlands and used it as refugia for at least

11 months.

The number of individuals of Florida mice and cotton rats increased after prescribed

burning because of immigration of new adult individuals in the vegetation surrounding wetlands.

The cotton mouse maintained relatively low, but stable numbers of individuals, in treatment and

control sites. Probably, competition for food and space did not allow immigrant cotton mice to

establish in the vegetation surrounding wetlands. The golden mouse drastically declined in the

number of individuals because of its arboreal life form and the combined effect of flooding and

prescribed burning. These species have been reported with a similar or different reaction to

prescribed fire at other localities in Florida and the United States. Therefore, these species do not

have a specific pattern in their responses to prescribed burning.

Emigration to refugia occurred because there was no cover and food in the burned sites.

Immigration of new adult individuals possibly took place because burned areas attracted rodents,

but they had to move to the vegetation surrounding wetlands because of the lack of food and

cover. The Florida mouse and the cotton rat were the only species that established in the

wetlands, where they had acorns from September to December. Returning to the scrubs

happened in or after March 2006, where the amount of cover was at least 70%, and only V.

myrsinites were flowering. Fruit production started with V. myrsinites in April 2006. So, the re-

colonization of the burned sites can be explained by the re-growth of the shrub cover and the

production of fruits by V. myrsinites.

Recommendations

Prescribed burning studies in Florida are strongly needed. Regarding the amount of

knowledge obtained from prescribed burning studies, Florida is behind in comparison with









studies conducted in California, Arizona, Montana, and the Appalachian region. Arata (1959)

wrote: "Considering the ecological importance and frequency of fire in the southeast, the lack of

data on its effects on small mammals is surprising." Jones (1990) also stated: "Finally, I wish to

emphasize how little is known about the effects of fires on small mammals....It is rather

surprising; however, that with the prevalence of both prescribed burns and lightning strikes in the

southeastern United States, there is still so little research on effects on non-game animals of the

region." It is even more surprising that Arata's and Jone's statements are still valid in 2008.

The experimental design constraint of having the same fire behavior characteristics in

treatment sites is extremely difficult to satisfy. However, an attempt must be made in order to try

to achieve this important goal. In this regard, this dissertation recommends burning treatment

sites in the same day when possible or on different dates not separated by more than one month.

The season of burning is critical for both plant species and wildlife. A prescribed fire too

early in the season might kill flowering buds or developing flowers. This mortality could

significantly reduce a full year of potential seeds. Sprouting shrubs may take some years to

recover sufficiently to support flowering again (Whelan 1995). Also, it is critical to avoid

prescribed burning during periods of reproduction in small mammals in order to reduce juvenile

mortality and to increase the possibility of re-nesting in bird species. Spring fires may impact

small mammal populations more than fires in other seasons because of limited mobility of

young. The species with the most vulnerable young are small mammals, most of which also have

high reproductive rates. If postfire habitat provides food and shelter for them, their populations

recover rapidly. Following these ideas, land managers should limit the size of prescribed fires

during peak of reproductive periods, which occur during February-March in south Florida and

May-June in north Florida. Hence, according to the current information available for CKSSR,









prescribed burning should be applied in April to May because the majority of the plant species

have produced seeds and small mammals have reproduced.

Do not burn wetlands when they are the only refugia available, and burn them when other

habitats might work as refugia. I suggest that the prescribed burning plan should consider refugia

for wildlife. If the area to be burned does not have surrounded habitats not included in the

prescribed burning plan, wetlands next to the area to be burned should not be included in the

prescribed burning plan. These wetlands will be the only refugia available for wildlife.

Otherwise, they should be burned because the vegetation will change to another type of habitat.

Another possibility is to burn wetlands from one to 2 years out of the sequence with the scrub,

after giving the scrub enough time to establish cover and food production.

The prescribed burning plan should not include large areas to be burned in the same

season. It would not be wise to include large extension of land because the prescribed fire plan

might fail. Large extension of land could not be controlled, and the damage to both plants and

wildlife may be irreversible. Land managers always make the assumption that wildlife,

especially small mammals, will recover soon from adjacent habitats. Particularly, if we are

talking about burning small plots, there is no doubt about it. But, if large areas are considered,

the prescribed burning might drastically cause decline or even eliminate rare species. In addition,

there is no guarantee that rare species would recover because in general prescribed burning plans

do not have a monitoring program working at the same time to measure fire effects on vegetation

and wildlife. In contrast, burning small plots makes the job easier, allows wildlife to recolonize

burned sites faster, and makes the restoration goals feasible.

Mechanical treatments have already started to be used in CKSSR, but cautio should be

considered. The rapid recovery to previous conditions occurs in scrub only when the sprouting









ability of the dominant shrubs remains intact. Mechanical disturbances that remove roots and

rhizomes of oaks, palmettos, and ericaceous shrubs cause long-lasting changes in structure and

composition (Breininger and Schmalzer 1990, Schmalzer et al. 2003). Therefore, mechanical

treatment should be inspected in situ to assure no damage of root and rhizome systems.

Fire return interval should be established based on the knowledge of plant species response

to prescribed burning. This aspect is critical in any public land where restoration through

prescribed burning is needed. Prescribed burning should not be applied in short intervals because

it might occur at the beginning of seed production of obligate seeders such as P. clausa or C.

ericoides. Also, it should not be applied in the long intervals because obligate-seeder species like

herb species are lost because oaks and ericaceous shrubs storage large amounts of underground

resources. For this reason, it is very important to monitor structural and compositional change

after fire in order to determine when the community returns to conditions similar to stands

without fire suppression. This criterion might be used in order to establish a fire return interval

for scrubby flatwoods in CKSSR.

Currently, CKSSR is applying prescribed burning every 5 years. Fire-return interval for

scrubby flatwoods is one every 5-60 yr (Abrahamson and Abrahamson 1984a, Abrahamson and

Hartnett 1990). The research program carried out in Kennedy Space Center indicates that 5 years

is the average time for communities without fire suppression to return to preburn conditions.

Therefore, CKSSR is applying the fire return interval for the reserve according to the literature.

However, CKSSR needs a researcher in charge of monitoring prescribed burning effects on

plants and wildlife and to determine if 5 years is the right time for the Reserve. The recovery

process in CKSSR is so fast (for the majority of the vegetation variables) that debris cover and









vegetation height might be taken as criteria because they have the slowest recovery among all

variables.

Future research should address microsite post-fire conditions, patchiness within bums, and

seasonality of fire effects for specific ecosystems. Few studies about microclimatic conditions

before and after prescribed burning have been conducted in Florida. A more comprehensive

database is needed in order to have a better understanding of the changes in environmental

variables after prescribed burning and their effects on small mammals. The importance of food,

cover, and predation as important factors regulating post-fire small mammal populations require

investigations. The responses of small mammal populations to post-fire vegetation changes,

especially in relation to fire patchiness, are important aspects to take into consideration because

they might have a strong impact on both the responses of plant population and of small mammals

to prescribed burning. To improve long-term management for sustaining ecosystems,

information is needed about the effects of fire on small mammals, at different seasons and under

different conditions and over several decades.










APPENDIX A
TWO TAILS T TEST COMPARISON AMONG ENVIRONMENTAL VARIABLES AND
FIRE BEHAVIOR CHARACTERISTICS BETWEEN SITES 5C AND 2M. a = 0.05

Variable Mean Variance N T Stat P-value
Wind Speed 5C (Km/h) 3.2 0.81 9
-0.81 0.4336
Wind Speed 2M (Km/h) 3.6 1.30 8
Air Temperature 5C (OC) 26.6 1.27 8
-3.10 0.0112
Air Temperature 2M (C) 29.3 4.82 8
Relative Humidity 5C (%) 48.5 23.43 8 -4.07 0.0016
Relative Humdity 2M (%) 61.8 61.36 8
lh timelag 5C 6.5 2.29 20
-1.46 0.1553
lhtimelag 2M 7.4 6.29 20
10htimelag 5C 7.5 18.17 20
-0.60 0.5550
10 h timelag 2M 8.3 17.91 20
100htimelag 5C 13.9 85.36 20
-0.29 0.7768
100htimelag 2M 14.8 97.57 20
Live Herb 5C 59.2 279.03 20
0.63 0.5298
Live Herb 5C 55.9 218.57 20
Live Woody 5C 60.6 109.44 20
-0.52 0.6038
Live Woody 2M 62.1 69.79 20
Flame Length 5C 4.0 1.73 30
1.59 0.1171
Flame Length 2M 3.4 1.95 30
Back fire 5C 1.7 0.24 30
1.09 0.3009
Back fire 2M 1.5 0.15 30
Head fire 5C 6.2 2.82 30
2.11 0.0641
Head fire 2M 4.9 1.50 30
Temperature ground 5C (C) 857.9 141510.46 30
-2.00 0.0506
Temperature ground 2M (C) 1019.0 52172.59 30
Temperature above ground 5C (C) 600.4 190308.66 30
Temperatureaboveground2M-2C) 898.6 11849.620048
Temperature above ground 2M (C) 898.6 118459.62 30









LIST OF REFERENCES

1. Abrahamson W.G. 1984a. Post-fire recovery of Florida Lake Wales Ridge vegetation.
American Journal of Botany 71:9-21.

2. Abrahamson W.G. 1984b. Species responses to fire on the Florida Lake Wales Ridge.
American Journal of Botany 71:35-43.

3. Abrahamson, W.G., A.F. Johnson, J.N. Layne, and P.A. Peroni. 1984. Vegetation of the
Archbold Biological Station, Florida: An Example of the Southern Lake Wales Ridge.
Florida Scientist. 47: 209-251.

4. Abrahamson W.G., and J.R. Abrahamson 1996a. Effects of low-intensity winter fire on
long-unbumed Florida sand pine scrub. Natural Areas Journal 16: 171-183.

5. Abrahamson W.G., and J.R. Abrahamson 1996b. Effects of fire on long-unbumed Florida
uplands. Journal of Vegetation Science 7: 565-574.

6. Abrahamson. W.G. and D.C. Hartnett. 1990. Pine flatwoods and dry prairies. Pp. 103-
149 in R.L. Myers, and J.J. Ewel, J.J., eds., Ecosystems of Florida. University of Presses
Florida, Gainesville.

7. Ahlgreen, C. E. 1966. Small mammals and reforestation following prescribed burning.
Journal of Forestry 64: 614-618.

8. Amoroso, J.L. 1993. A floristic study of the Cedar Key Scrub State Reserve, Levy
County, Florida. M.S. Thesis, University of Florida, Gainesville.

9. Andersen, D. C., K. R. Wilson, M. S. Miller, and M. Falck. 2000. Movement patterns of
riparian small mammals during predictable floodplain inundation. Journal of Mammalogy
81: 1087-1099.

10. Andrews, P.L., and C.D. Bevins. 2005. BehavePlus 3.0.2: Fire behavior prediction and
fuel modeling system. Available online .

11. Arata, A. A. 1959. Effects of burning on vegetation and rodent populations in a longleaf
pine turkey oak association in north central Florida. Quarter Journal of Florida Academic
Science 22: 94-104.

12. Axelrod, D.I. 1958. Evolution of the Macro-Tertiary geoflora. Botanical Review 24: 432-
509.

13. Baker, W.L. 1992. Effects of settlement and fire suppression on landscape structure.
Ecology 73: 1879-1887.

14. Baker, W.L. 1994. Restoration and landscape structure altered by fire suppression.
Conservation Biology 8: 763-769.









15. Batzli, G. O. 1992. Dynamics of small mammal populations: a review. Pp 831-850 in D.
R. McCullough, and R. H. Barrett, eds., Wildlife 2001. Elsevier Applied Science,
London. 1163 pp.

16. Beadle, N.C.W. 1940. Soil temperatures during forest fires and their effect on the
survival of vegetation. Journal of Ecology 28:180-192.

17. Beck, A.M., and R. J. Vogl. 1972. The effect of spring burning on rodent populations in a
brush prairie savanna. Journal of Mammalogy 53: 336-346.

18. Bendell, J.F. 1974. Effects of fire on birds and mammals. Pp. 73-138 in T. T. Kozlowski,
and C. E. Ahlgreen eds., Fire and Ecosystems. Academic Press, Orlando. 250 pp.

19. Biswell, H.H. 1989. Prescribed burning in California wildlands vegetation management.
University of California Press, Berkeley, C.A. 255 pp.

20. Blair, W. F. 1939. Some observed effects of stream-valley flooding on mammalian
populations in eastern Oklahoma. Journal of Mammalogy 20: 304-306.

21. Blanchard, V.J. 1991. Stream-terrace hardwood forest as refuge for small mammals when
adjacent pine forest is burned. M.S. Thesis, Louisiana State University, Baton Rouge,
LA, 89 pp.

22. Blankenship, D. J. 1982. Influence of prescribed burning on small mammals in
Cuyamaca Rancho State Park, California. Pp. 587 in C.E. Conrad, and W.C. Oechel, eds.,
Proceedings of the Symposium of Dynamics and Management of Mediterranean Type
Ecosystems; 1871 Jun 22-26; Sand Diego, CA. GTR-PSW-58, U.S. Department of
Agriculture, Forest Service, Berkeley, CA.

23. Bock, C.E. and J.H. Bock. 1978. Response of birds, small mammals, and vegetation to
burning sacaton grasslands in southeastern Arizona. Journal Range Management 31:296-
300.

24. Bock, C. E. and J. H. Bock. 1983. Responses of birds and deer mice to prescribed
burning in ponderosa pine. Journal of Wildlife Management 47:836-840.

25. Breininger, D.R. and P.A. Schmalzer. 1990. Effects of fire and disturbance on plants and
birds in a Florida oak/palmetto scrub community. American Midland Naturalist 123: 64-
74.

26. Brown, J.K., and J.K. Smith. 2000. Wildland fire in ecosystems: effects of fire on flora.
GTR-RMRS-42 U.S. Department of Agriculture, Forest Service, Ogden, UT.

27. Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference: A
Practical Information-Theoretic Approach. Springer, New York.









28. Bury, R. B., D.J. Major, D.S. Pilliod. 2002. Responses of amphibians to fire disturbances
in Pacific Northwest forest: a review. Pp. 34-42 in W.M. Ford, K.R. Russell, and C.E.
Moorman, eds., The Role of Fire in Nongame Wildlife Management and Community
Restoration: Traditional uses and New Directions. GTR-NE-288. U.S. Department of
Agriculture, Forest Service, Washington, D.C.

29. Carrington, M.E., and J.E. Keeley. 1999. Comparison of post-fire seedling establishment
between scrub communities in Mediterranean and non-Mediterranean climate
ecosystems. Journal of Ecology 87: 1025-1036.

30. Chew, R. M., B.B. Butterworth, and R. Greghman. 1959. The effects of fire on the small
mammal populations of chaparral. Journal of Mammalogy 40: 253.

31. Christman, S.P. 1988. Endemism and Florida's interior sand pine scrub. Final Proj. Rep.
GFC-84-101, Florida Fish and Wildlife Conservation Commission, Tallahassee.

32. Clark, A.M., B.W. Bowen, and L.C. Branch. 1999. Effects of natural habitat
fragmentation on an endemic scrub lizard (Sceloporus woodi): an historical perspective
based on mitochondrial DNA gene genealogy. Molecular Ecology 8: 1093-1104.

33. Converse, S. J., Block, W. M., and White, G. C. 2006. Small mammal population and
habitat responses to forest thinning and prescribed fire. Forest Ecology and Management
228: 263-273.

34. Cooch, E., and G. White. (2006). Program MARK: A Gentle Introduction. Available
online: < http://www.phidot.org/software/mark/docs/book/>.

35. Cook, S.R. 1959. The effects of fire on a population of small rodents. Ecology 40:102-
108.

36. Cormack, R. M. 1964. Estimates of survival from the sightings of marked animals.
Biometrika 51:429-438.

37. Cox, J., K. Inkley, and R. Kautz. 1987. Ecology and habitat protection needs of gopher
tortoise (Gopheruspolyphemus) populations found on lands slated for large-scale
development in Florida. Florida Game and Fresh Water Fish Commission, Nongame
Wildlife Program, Technical Report 4. 75 pp.

38. Crowner, A.W. and G.W. Barrett. 1979. Effects of fire on the small mammal component
of an experimental grassland community. Journal of Mammalogy 60:803-813.

39. DeBano, L. F., D.G. Neary, P.F. Ffolliott. 1998. Fire's effects on ecosystems. New
York,NY: John Wiley and Sons. 333 pp.

40. DEP. 1998. Cedar Key Scrub State Reserve: Unit Management Plan. Division of
Recreation and Parks, Tallahassee, FL.









41. Depue, J. R. 2005. Responses of the Florida mouse (Podomysfloridanus) to habitat
management. M.S. thesis. University of Central Florida, Orlando, FL.

42. Eisenberg, J.F. 1983. The gopher tortoise as a keystone species. Proceedings of the
Annual Management Gopher tortoise council 4:1-4.

43. Fala, R. A. 1975. Effects of prescribed burning of small mammal populations in a mixed-
oak clearcut. Journal of Forest 73:586-587.

44. Fitzgerald, S. M. 1990. Small mammal, bird, and plant responses to the rehabilitation of a
dry prairie grassland association using fire and roller chopping applied in two seasons.
M.S. Thesis, University of Florida, Gainesville, FL. 104 pp.

45. Ford W.M., J. Laerm, T.S. McCay, M.A. Menzel, and D.W. McGill. 1999. Effects of a
community restoration fire on small mammals and herpetofauna in the southern
Appalachians. Forest Ecology and Management 114:233-243.

46. Forde, J. D., N.F. Sloan, and D.A. Shown. 1984. Grassland habitat management using
prescribed burning in Wind Cave National Park, South Dakota. Prairie Naturalist 16:97-
110.

47. Frank, P. A. 1996. Ecology and conservation of the Anastasia island beach mouse
(Peromyscus polionotus phasma). Ph.D. Dissertation, University of Florida, Gainesville,
FL. 111 pp.

48. Franz, R., and I. R. Quitmyer. 2005. A fossil and zooarcheological history of the gopher
tortoise (Gopheruspolyphemus) in the southeastern United States. Bulletin of Florida
Museum of Natural History 45:179-199.

49. Franz, R. D. Meahr, A. Kinlaw, C. O'Brien, and R. D. Owen. 1998. Distribution and
abundance of sensitive wildlife species at Avon Park Air Force Range. Final Report,
Project RWO-169. Florida Museum of Natural History, University of Florida,
Gainesville, FL. 90 pp.

50. Gashwiler, J. S. 1959. Small mammal study in west-central Oregon. Journal of
Mammalogy. 40: 128-139.

51. Gates, C.A. and G.W. Tanner. 1988. Effects of prescribed burning on herbaceous
vegetation and gophers (Geomyspinetis) in a sandhill community. Florida Scientist
51:129-139.

52. Geluso, K.N., G. D. Schroder, and T.B. Bragg. 1986. Fire-avoidance behavior of meadow
voles (Microtuspennsylvanicus). The American Midland Naturalist 116:202-205.

53. Gibson, D.J., and E. S. Menges. 1994. Population structure and spatial pattern in the
dioeciously shrub Ceratiola ericoides. Journal Vegetation Science 5: 337-346.









54. Goatcher, B.L. 1990. A preliminary investigation of the importance of stream-terrace
hardwood forests as refuge for small mammals to escape fires. M.S. Thesis, Louisiana
State University, Baton Rouge, LA, p. 53.

55. Greenberg, C. H. 2003. Vegetation recovery and stand structure following a prescribed
stand-replacement burn in sand pine scrub. Natural Areas Journal 23: 141-151.

56. Greenberg, Cathryn, David L. Otis, and Thomas A. Waldrop. 2006. Response of white-
footed mice (Peromyscus leucopus) to fire and fire surrogate fuel reduction treatments in
a southern Appalachian hardwood forest. Forest Ecology and Management 234:355-362.

57. Greig-Smith, P. 1983. Quantitative plant ecology. Blackwell, Oxford. 198 pp.

58. Groves, F.R., K. Steenhof 1988. Responses of small mammals and vegetation to wildfire
in shadscale communities of southwestern Idaho. Northwest Science. 62:205-210.

59. Gunther, P.M., B.S. Horn, and G.D. Babb. 1983. Small mammal populations and food
selection in relation to timber harvest practices in the western Cascade Mountains.
Northwest Science 57:32-44.

60. Halford, D. K. 1981. Repopulation and food habits ofPeromyscus maniculatus on a
burned sagebrush desert in southeastern Idaho. Northwest Science 55:44-49.

61. Harty, F. M., J. M. Ver Steeg, R. R. Heidorn, and L. Harty. 1991. Direct mortality and
reappearance of small mammals in an Illinois grassland after a prescribed burn. Natural
Areas Journal 11: 114-118.

62. Hatchell, G. E. 1964. Small mammals species and populations in the loblolly-shortleaf
pine forest type of Lousiana. Research Paper SO-10. U.S. Department of Agriculture.
Forest Service. Washington, D.C. 12 pp.

63. Hedlund, J. D. and W.H. Richard. 1981. Wildfire and the short-term response of small
mammals inhabiting a sagebrush-bunchgrass community. Murrelet 62: 10-14.

64. Heyward, F. 1938. Soil temperatures during forest fires in the Longleaf Pine region.
Journal of Forestry 36:478-491.

65. Hollander, J.L., S.B. Vander Wall. 2004. Effectiveness of six species of rodents as
dispersers of singleleaf pinon pine (Pinus monophylla). Oecologia 138:57-65.

66. Hon, T. 1981. Effects of prescribed fire on furbearers in the South. Pp. 121-128 in Wood,
Gene W., ed. Prescribed Fire and Wildlife in Southern Forests: Proceedings of a
Symposium, 1981 April 6-8; Myrtle Beach, SC. Georgetown, SC: Clemson University,
Belle W. Baruch Forest Science Institute. 548 pp.









67. Hooven, E. F. 1973. Effects of vegetation changes on small forest mammals. Pp. 75-97 in
R.K. Hermann, and D. P. Lavender, eds., Even-age Management: Proceedings of a
Symposium; 1972 August 1; [Location of conference unknown]. Paper 848. Oregon State
University, School of Forestry, Corvallis, OR.

68. Humphrey, S. R., J.F. Eisenberg, and R. Franz. 1985. Possibilities for restoring wildlife
of a longleaf pine savanna in an abandoned citrus grove. Wildlife Society Bulletin 13:
487-496.

69. James, F.C., and C.E. McCulloch. 1995. The strength of inferences about causes of trends
in populations. Pp. 40-51 in T.E. Martin, and D. M. Finch eds., Ecology and Management
of Neotropical Migratory Birds. Oxford University Press, New York. 489 pp.

70. Johnson, A.F. 1982. Some demographic characteristics of the Florida rosemary,
Ceratiola ericoides Michx. American Midland Naturalist 108:170-174.

71. Johnson, A.F. and W.G. Abrahamson. 1990. A note on the fire responses of species in
rosemary scrubs on the southern Lake Wales Ridge. Florida Scientist 53: 138-143.

72. Johnson, A.F., W.G. Abrahamson, and K.D. McCrea. 1986. Comparison ofbiomass
recovery after fire of a seeder (Ceratiola ericoides) and a sprouter (Quercus inopina)
species from south-central Florida. American Midland Naturalist 116: 423-428.

73. Jolly, G. M. 1965. Explicit estimates from capture-recapture data with both death and
immigration: stochastic model. Biometrika 52:225-247.

74. Jones, C. A. 1989. Fire and the Florida mouse (Podomysfloridanus). Florida Scientist
52:19.

75. Jones, C. A. 1990. Microhabitat use by Podomysfloridanus in the high pine lands of
Putnam County, Florida. Ph.D. dissertation, Univ. Florida, Gainesville, Fla.

76. Jones, C. A. 1992. Review of the effects of fire on Peromyscus and Podomys. Florida
Scientist 55:75-84.

77. Kaufman, G.A., D. W. Kaufman, and E.J. Finck. 1988a. Influence of fire and topography
on habitat selection by Peromyscus maniculatus and Reitil ,, i ungrazed tallgrass prairie. Journal of Mammalogy 69:342-352.

78. Kaufman, D.W., S.K. Gurtz, and G.A. Kaufman. 1988b. Movements of the deer mouse in
response to prairie fire. Prairie Naturalist 20: 225-229.

79. Kaufman, D. W, G. A. Kaufman, and E. J. Finck. 1983. Effects of fire on rodents in tall
grass prairie of the Flint Hills region of eastern Kansas. Prairie Naturalist 15: 49-56.

80. Kaufman, G. A., D.W. Kaufman, E. J. Elmer. 1982. The effect of fire frequency on
populations of the deer mouse (Peromyscus maniculatus) and the western harvest mouse
(Reitiln t,ini,,lwnuy megalotis). Bulletin of the Ecological Society of America 63: 66.









81. Keeth, J.J., and G.M. Bryam. 1968. A drought index for forest fire control. Research
Paper SE-38, U.S. Department of Agriculture, Forest Service, Southeast Forest
Experimental Station, Asheville, NC.

82. Kennard, D. 2007. A history of people and fire in the south. In: Forest Encyclopedia
Network, [Online}. USDA Forest Service Southern Research Station and Southern
Regional Extension Forestry. Available online < http://www.forestryencyclopedia.net>.

83. Kilgore, B.M. 1987. The role of fire in wilderness: a state-of-knowledge review. Pp.71-
103 in C. L. Rober, ed., Proceedings National Wilderness Research Conference: issues,
state-of-knowledge, future directions. GTR-INT-220. U.S. Department of Agriculture,
Forest Service, Ogden, UT.

84. Kirkland Jr., G.L., H.W. Snoddy, and R.L. Amsler. 1996. Impact of fire on small
mammals and amphibians in a central Appalachian deciduous forest. American Midland
Naturalist 135:253-260.

85. Komarek, E.V. 1965. Fire Ecology grassland and man. Proceedings of the Annual Tall
Timbers Fire Ecology Conference 4: 169-220.

86. Komarek, E.V. 1969. Fire and animal behavior. Proceedings of the Annual Tall Timbers
Fire Ecology Conference 9: 160-207.

87. Krefting, L.W. and C.E. Ahlgren. 1974. Small mammals and vegetation changes after fire
in a mixed conifer-hardwood forest. Ecology 55: 1391-1398.

88. Landers, J.L. 1987. Prescribed burning for managing wildlife in southeastern pine forest.
Proc. 1986 Society of American Foresters National Convention pp. 151-159.

89. Layne, J.L. 1963. A study of the parasites of the Florida mouse, Peromyscusfloridanus,
in relation to host and environmental factors. Tulane Studies in Zoology 11:1-27.

90. Layne, J.N. 1974. Ecology of small mammals in a flatwoods habitat in north-central
Florida, with emphasis on the cotton rat (Sigmodon hispidus). American Museum
Novitates. 2544: 1-48.

91. Layne, J.N. 1990. The Florida mouse. Pp 1-21 in C. K. Dodd, R. E. Ashton. Jr., R. Franz,
and E. Wester, eds., Burrow associates of the gopher tortoise. Eight Ann. Mtg. Gopher
Tortoise Council. Florida Museum of Natural History, University of Florida, Gainesville.
134 pp.

92. Layne, J. N. 1992. The Florida mouse. Pp 250-264 in S.R. Humphrey ed., Rare and
Endangered Biota of Florida. Volume I: Mammals. University Press of Florida,
Gainesville, Fla. 392 pp.

93. Lawrence, G.E. 1966. Ecology of vertebrate animals in relation to chaparral fire in the
Sierra Nevada foothills. Ecology 47: 278-291.









94. Lebreton, J.D., K.P. Burham, J. Clobert, and D.R. Anderson. 1992. Modeling survival
and testing biological hypotheses using marked animals: a unified approach with case
studies. Ecological Monographs 62:67-118.

95. Lee, A.K. 1963. The adaptations to arid environments in woodrats of the genus Neotoma.
University of California Publications in Zoology 64: 57-96.

96. Lee, D.E., and W.D. Tietje. 2005. Dusky-footed woodrat demography and prescribed fire
in a California oak woodland. Journal of Wildlife Management 69: 1211-1220.

97. Lyon, J.L., and P.F. Stickney. 1976. Early vegetation succession following large
Northern Rocky Mountain wildfires. Tall Timbers Fire Ecology Conference Proceedings
14: 335-375.

98. Lyon, L.J., H.S. Crawford, E. Czuhai, R.L. Frediksen, F. Harlow, L.J. Metz, and H.A.
Pearson. 1978. Effects of fire on fauna: a state-of-knowledge review, GTR-WO-6. U.S.
Department of .Agriculture, Forest Service.

99. Martell, A.M. 1984. Changes in small mammal communities after fire in north central
Ontario. Canadian Field Naturalist 98:223-226.

100. McCarley, W.H. 1959. Fluctuations and structure of Peromyscus gossypinus in eastern
Texas. Journal of Mammalogy 35:526-532.

101. Mccoy, E. D. and H. R. Mushinsky. 1994. Effects of fragmentation on richness of
vertebrates in the Florida scrub habitat. Ecology 75 (2): 446-457.

102. McCune, B., and J. B. Grace. 2002. Analysis of ecological communities. MjM Software
Design, Gleneden Beach. 284 pp.

103. McCune, B., and M.J. Mefford. 1999. PC-ORD 5.0. Multivariate Analysis of Ecological
Data. MjM Software, Gleneden Beach, Oregon.

104. McGarigal, K., S. Cushman, and S. Stafford. 2000. Multivariate Statistics for wildlife and
Ecology research. Springer, New York. 283 pp.

105. McGee, J. M. 1982. Small mammal populations in an unburned and early fire
successional sagebrush community. Journal Range Management 35: 177-180.

106. Menges, E.S. 1999. Ecology and Conservation of Florida Scrub. Pp. 7-22 in R.C.
Anderson, J.S. Fralish, and J.M. Baskin, eds., Savannas, Barrens and Rock Outcrop Plant
Communities of North America. Cambridge University Press, New York. 470 pp.

107. Menges, E.S., and N. Kohfeldt. 1995. Life history strategies on Florida scrub plants in
relation to fire. Bulleting of the Torrey Botanical Club 122: 282-293.

108. Menges, E.S., and C.V. Hawkes. 1998. Interactive effects of fire and microhabitat on
plants of Florida scrub. Ecological Applications 8: 935-946.









109. Menges, E.S., W.G. Abrahamson, K.T. Givens, N.P. Gallo, and J.N. Layne. 1993.
Twenty years of vegetation change in five long-unburned Florida plant communities.
Journal of Vegetation Science 4:375-386.

110. Monroe, M.E. and S.J. Converse. 2006. The effects of early season and late season
prescribed fires on small mammals in a Sierra Nevada mixed conifer forest. Forest
Ecology and Management 236: 229-240.

111. Morgan, G.L. 1998. Florida mouse (Podomysfloridanus) association with gopher tortoise
(Gopheruspolyphemus) burrows and vegetation characteristics in the Cedar Key Scrub
State Reserve, Levy County, Florida. M.S. Thesis, University of Florida, Gainesville.

112. Myers, R. L. 1985. Fire and the dynamic relationship between sandhill and sand pine
scrub vegetation. Bulletin Torrey Botanical Club 112:241-252.

113. Myers, R.L. 1990. Scrub and high pine. Pp. 150-193 in R.L. Myers and J.J. Ewel, eds.,
Ecosystems of Florida. University of Presses Florida, Gainesville. 765 pp.

114. Newman, Ch. 1997. The Florida Mouse (Podomysforidanus): long term dynamics of a
population in the high pine sandhill of Putnam County, Florida. Master Thesis,
University of Florida. Florida, Gainesville. 102 pp.

115. Odum, E.P., S.E. Pomeroy, J.C. Dickinson III, and K. Hutcheson. 1974. The effects of
late winter litter burn on the composition, productivity and diversity of a 4-year old
fallow-field in Georgia. Pp. 399-419 in Proceedings, Annual Tall Timbers Fire Ecology
Conference; 1973 March 22-23; Tallahassee, Fla. No. 13. Tallahassee, Fla.

116. Peek, J.M. 1986. A review of wildlife management. Prentice-Hall, Englewood Cliffs, NJ.
486 pp.

117. Peroni, P. A., and W.G. Abrahamson. 1985. Post-settlement vegetation loss on the
southern Lake Wales Ridge, Florida. Palmetto 5: 6-7.

118. Pielou, E. C. 1969. An introduction to mathematical ecology. John Wiley and Sons, New
York. 365 pp.

119. Pilliod D.S., P.S. Corn, R.B. Bury, E.J. Hyde, and C.A. Pearl. 2003. Fire and amphibians
in North America. Forest Ecology and Management 178: 163-181.

120. Price, M.V. and N.M. Waser. 1984. On the relative abundance of species: postfire
changes in a coastal sage scrub rodent community. Ecology 65:1161-1169.

121. Pyare, S. and W.S. Longland. 2001. Patterns of ectomycorrhizal-fungi consumption by
small mammals in remnant old-growth forest of the Sierra Nevada. Journal of
Mammalogy 85:681-689.

122. Pyne, S.J., P.L. Andrews, and R.D. Laven. 1996. Introduction to Wildland Fire. John
Wiley and Sons, Inc., New York.









123. Quinn, Ronald D. 1979. Effects of fire on small mammals in the chaparral. Cal-Neva
Wildlife Transactions. 1979:125-133.

124. Ream, C. H. 1981. The effects of fire and other disturbances on small mammals and their
predators: an annotated bibliography. GTR-INT-106. U.S. Department of Agriculture,
Forest Service, Ogden UT. 55 p.

125. Ruth, A.D., D.L. Miller, J.Shibu, and A. Long. 2007. Effects of reintroduction of fire into
fire suppressed coastal scrub and longleaf pine communities along the lower gulf coastal
plain. Natural Areal Journal 27: 332-344.

126. Russell, K.R., D.H.V. Lear, D.C. Guynn Jr. 1999. Prescribed fire effects on herpetofauna:
Review and management implications. Wildlife Society Bulletin. 27: 374-384.

127. Sasso, C. R., and M. Gaines. 2002. Population ecology of three species of small
mammals on Key Largo, Florida. Florida Scientist 65: 115-125.

128. Schmalzer, P.A. 2003. Growth and recovery of oak-saw palmetto scrub through ten years
after fire. Natural Areas Journal 23: 5-13.

129. Schmalzwer, P.A., and FW. Adrian. 2001. Scrub restoration on Kennedy Space
Center/Merrit Island National Wildlife Refuge, 1992-2000. Pp. 17-20 in D. Zarrau, ed.,
Proceedings of the Florida Scrub Symposium 2001. U. S. Fish and Wildlife Service.
Jacksonville, Fla.

130. Schmalzer, P.A. and S.R. Boyle. 1998. Restoring long-unburned oak-saw palmetto scrub
requires mechanical cutting and prescribed burning (Florida). Restoration and
Management Notes. 16: 96-97.

131. Schmalzer, P.A., and G.R. Hinkle. 1991. Dynamics of vegetation and soils of oak/saw
plametto scrub after fire: observations from permanent transects. NASA Tech. Mem.
103817. 149 pp.

132. Schmalzer, P.A., and C. R. Hinkle. 1992a. Recovery of oak-saw palmetto scrub after fire.
Castanea 57: 158-173.

133. Schmalzer, P.A., and C. R. Hinkle. 1992b. Species composition and structure of oak-saw
palmetto scrub vegetation. Castanea 57:220-251.

134. Schmalzer, P.A., T.E. Foster, and F.W. Adrian. 2003. Responses of long-unburned scrub
on the Merrit Island/Cape Canaveral Barrier Island Complex to cutting and burning.
Proceedings of the Second International Wildland Fire Ecology and Fire Management
Congress. 16-20 November 2003. Orlando, Fla.

135. Schnurr, J.L., C.D. Canham, R.S. Otsfield, and R.S. Inouye. 2004. Neighborhood
analysis of small mammal dynamics: impacts on seed predation and seedling
establishment. Ecology 85: 741-755.









136. Schramm, P., and B.J. Willcutts. 1983. Habitat selection of small mammals in burned and
unburned tallgrass prairie. Pp. 49-55 in R. Brewer ed., Proceedings, 8th North American
Prairie Conference; 1982 August 1-4; Kalamazoo, MI. Western Michigan University,
Department of Biology, Kalamazoo, MI.

137. Schwilk, D.W., and J.E. Keeley. 1998. Rodent populations after a large wildfire in
California chaparral and coastal sage scrub. Southwestern Naturalist 43:480-483.

138. Seber, G.A. 1965. A note on the multiple recapture census. Biometrika 52: 249-259.

139. Shadowen, H.E. 1963. A live-trap study of small mammals in Louisiana. Journal of
Mammalogy 44:103-108.

140. Sims, H. P. and C. H. Buckner. 1973. The effect of clear cutting and burning of Pinus
banksiana forest on the populations of small mammals in southern Manitoba. American
Middland. Naturalist 90:228-231.

141. Simon, D.M. 1986. Fire effects in coastal habitats of East Central Florida. National Park
Service Cooperative Park Studies Unit Technical Report 27. Institute of Ecology,
University of Georgia, Athens.

142. Simons, L. H. 1991. Rodent dynamics in relation to fire in the Sonoran Desert. Journal of
Mammalogy 72(3): 518-524.

143. Singer, F.J., and P. Schullery. 1989. Yellowstone wildlife: populations in process.
Western Wildlands 15: 18-22.

144. Slabaugh, J.D., A.O. Jones, W.E. Puckett, and J. N. Schuster. 1996. Soil Survey of Levy
County, Florida. U.S. Department of Agriculture, Natural Resource Conservation
Service, Tallahassee, Fla.

145. Slade, N.A. 1991. Loss of body mass associated with capture of Sigmodon and Microtus
from Northeastern Kansas. Journal of Mammalogy 72: 171-176.

146. Smith, J.K. 2000. Wildland fire in ecosystems: effects of fire on fauna. GTR-RMRS-
GRT-42-I U.S. Department of Agriculture, Forest Service. Ogden, UT. 83 pp.

147. Stanturf, J.A., D.D. Wade, T.A. Waldrop, D.K. Kennard, and G.L. Achtemeier. 2002.
Background paper: fire in southern forest landscapes. Pp. 607-622 in: D.N.Wear and J.G.
Greis, eds., Southern forest resource assessment. U.S. Department of Agriculture, Forest
Service, Southern Research Station, Asheville, N.C.

148. Stout, I.J. 2001. Rare plants of the Florida scrub, USA. Natural Areas Journal 21: 50-60.

149. Stout, I.J., A. L. Farris, and V.L. Wright. 1971. Small mammal populations of an area in
northern Idaho severely burned in 1967. Northwest Science 45:219-226.









150. Stout, I.J., and W.R. Marion. 1993. Pine flatwoods and xeric pine forests of the southern
(lower) coastal plain. Pp.373-446 in W.H. Martin, S.G. Boyce, and A.C Echteracht,
eds., Biodiversity of the Southeastern United States Lowland Terrestrial Communities.
John Wiley and Sons, New York. 502 pp.

151. Sullivan, J. 1995 a. Sigmodon hispidus. In: Fire Effects Information System, [Online].
U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available online .

152. Sullivan, J. 1995 b. Peromyscus maniculatus in Fire Effects Information System. U.S.
Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available online .

153. Sullivan, T.P., and J.O. Boateng. 1996. Comparison of small mammal community
responses to broadcast burning and herbicide application in cutover forest habitats.
Canadian Journal of Forest Research 26: 462-473.

154. Taylor, D.L. 1981. Effects of prescribed fire on small mammals in the southeastern
United States. Pp. 109-120, in, G. W. Wood ed., Prescribed Fire and Wildlife in
Southern Forest. Belle W. Baruch Forest Sci. Inst. Clemson Univ., Georgetown, SC. 170
pp.

155. Tester, J. R. 1965. Effects of a controlled burn on small mammals in a Minnesota oak-
savanna. American Midland Naturalist 74:240-243.

156. Tevis, L. Jr. 1956. Effect of a slash bur on forest mice. Journal of Wildland
Management 20: 405-409.

157. Turner, R. W. 1966. Effects of flooding on the mouse Peromyscus leucopus. Trans. Ill.
State Academic Science. 59: 390-391.

158. U.S. Fish and Wildlife Service. 1999. South Florida multi-species recovery plan. U.S.
Fish and Wildlife Service, Atlanta, GA.

159. Vacanti, P.L. and K.N. Geluso. 1985. Recolonization of a burned prairie by meadow
voles (Microtuspennsylvanicus). Prairie Naturalist 17:15-22.

160. Vander Wall, S.B. 1993. Cache site selection by chipmunks (Tamias spp) and its
influence on the effectiveness of seed dispersal in Jeffrey pine (Pinusjefferyi). Oecologia
96:246-252.

161. Ver Steeg, J. M.; F. M. Harty; and L. Harty. 1983. Prescribed fire kills meadow voles
(Illinois). Restoration and Management Notes 1: 21.

162. Vogl, R. J. 1973. Effects of fire on the plants and animals of a Florida wetland. American
Midland Naturalist 89: 334-347.









163. Vreeland, J. K. and W.D. Tietje. 1998. Initial response of woodrats to prescribed burning
in oak woodland. Transactions of the Western Section of the Wildlife Society 34:21-31.

164. Wade, D.D. and J.D. Lunsford. 1989. A guide for prescribed fire in the southern United
States. R8-TP 11, U.S. Department of Agriculture, Forest Service, Atlanta, GA.

165. Webb, S.D. 1974. Pleistocene mammals of Florida. University Presses of Florida,
Gainesville. 270 pp.

166. Webb, S.D. 1990. Historical biogeography. Pp 70-100 in R.L. Myers and J.J. Ewel, eds.,
Ecosystems of Florida. University of Presses Florida, Gainesville. 765 pp.

167. Weekly, C.W., and E.S. Menges. 2003. Species and vegetation responses to prescribed
fire in a long-unbumed, endemic rich Lake Wales Ridge scrub. Journal of the Torrey
Botanical Society 130: 265-282.

168. West, S. D. 1982. Dynamics of colonization and abundance in central Alaskan
populations of the northern red-backed vole Clethrionomys rutilus. Journal of
Mammalogy 63:128-143.

169. Whelan, R.J. 1995. The Ecology of Fire. Cambridge University Press, Cambridge. 346
pp.

170. White, G. and K.P. Burnham. 1999. Program MARK. Available online:
.

171. White, W. 1970. The Geomorphology of Florida peninsula. Geological Bulletin No. 51.
State of Florida, Department of Natural Resources, Bureau of Geology, Division of
Resource Management, Florida Department of Natural Resources, Tallahassee, Fl.

172. Williams, B.K., J.D. Nichols, and M.J. Conroy. 2002. Analysis and Management of
Animal Populations. Academic Press, San Diego. 817 pp.

173. Williams, D.F., J. Verner, G.F. Sakai, and J.R. Waers. 1992. General biology of major
prey species of the California spotted owl. Pp. 207-221 in J. Verner, K.S. Mckelvey, B.R.
Noon, R. F. Gutierrez, G.I. Goud Jr., and T.W Beck, eds., The California Spotted Owl: A
Technical Assessment of its Current Status. GTR-PSW-133, U.S. Forest Service,
California.

174. Williams, G.W. 2002. Aboriginal use of fire: are there any "natural" communities?
Washington, DC: U.S. Department of Agriculture, Forest Service, Washington D.C.

175. Wirtz, W.O., D. Hoekman: J.R. Muhm; and S.L. Souza. 1988. Postfire rodent succession
following prescribed fire in southern California chaparral. Pp. 333-339 in Szaro, P.C.,
K.E. Severson, and D.R. Patton, eds., Proccedings of Symposium about Management of
Amphibians, Reptiles, and Small Mammals in North America. GTR-RM-166. U.S.
Department of Agriculture, Forest Service. Ft. Collins, CO. 458 pp.









176. Wirtz,W.O. 1977. Vertebrate post-fire succession. Pp. 46-57 in H.A. Mooney, and C.E.
Conrad, eds., Proceedings of the Symposium on Environmental Consequences of Fire
and Fuel Management in Mediterranean Ecosystems, 1977 Aug 1-5; Palo Alto, CA.
GTR-WO-3. U.S. Department of Agriculture, Forest Service, Washington, D.C.

177. Wright, H.A. and A.W. Bailey. 1982. Fire ecology: United States and Southern Canada.
John Wiley & Sons, New York. 501 pp.

178. Zeilinski, W.J., W.D. Spencer, and R.H. Barret. 1983. Relationship between food habits
and activity patterns of pine martens. Journal of Mammalogy 64:387-396.









BIOGRAPHICAL SKETCH

I was raised in Caracas, Venezuela, where I received the degree "Licenciado en Biologia"

at Central University of Venezuela in 1981. From 1981 to 1985, I completed research on

population ecology, community ecology, and the behavior of snakes and small mammals. In

addition, I carried out the first evaluation of human impacts in a national park in Venezuela.

From 1985 to 1990, I worked for three projects sponsored by New York Zoological Society-The

Wildlife Conservation Society. These included (a) behavioral ecology and conservation of the

Family Cracidae (birds) in Venezuela, (b) human impacts on cracids populations in Venezuela,

and (c) uses, preferences, and hunting impact on wildlife in Venezuela. I was responsible for the

last two projects. Based on my results in relation to hunting pressure in national parks, an

educational program was recommended to the Venezuelan government.

From 1990 to 1996, I coordinated an educational program for hunters in four national

parks. I obtained a LASPAU-Fulbright scholarship in 1996 and began my master's program in

the University of Florida in 1997. I graduated in December 1998 and started the PhD program in

January 1999. The course work was completed in 2001 and presented the qualifying exams in

February 2002. Unfortunately, I did not obtain funding until October 2003 after changing the

original project. From October 2003 to August 2006, I carried out field work for my dissertation

in Cedar Key Scrub State Reserve in Florida.

I have coordinated the CIRCA Operations Training Program in the University of Florida

since fall 2000. I have trained 298 technology consultants for assisting professors, students, and

staff in five computers labs. In addition, I trained all supervisors and training specialists during

this time.










I have published a book, a section in another book, 16 scientific publications, 20 technical

reports, six publications for nonscientific audience, and I have given 38 speeches in Venezuela

and in international congresses.





PAGE 1

RESPONSES OF PLANT AND SMALL MAMMAL COMMUNITIES TO PRESCRIBED BURNING IN CEDAR KEY SCRUB STATE RESERVE By JOSE LORENZO SILVA-LUGO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Jose Lorenzo Silva-Lugo 2

PAGE 3

To my mother, my wife, and my three childre n for all their love, support, and sacrifice. To my adviser, Dr. George Tanner, for all his help, support, and advice. 3

PAGE 4

ACKNOWLEDGMENTS The Department of Wildlif e Ecology and Conservation and the Department of Environmental Protection pr ovided the financial aid and logistic needed for carrying out the field work. Florida Fish and Wildlife Conservation Co mmission provided a GPS unit and participated during prescribed burning. Special acknowledgments go to my adviser, Dr. George Tanner, for his support, help, and advice when I needed it. Dr Tanner always replies to all my requests, and he was always pleasant, cordial, and calm. I have fulfilled a very important goal in my life thanks to him. I also thank my committee members, Dr Dick Franz, Dr. Deborah Miller, Dr. Hardin Waddle, and Dr. Alan Long for their he lp during the correction process. M.S. James Collee and Jeff DiMaggio reviewed chapter drafts in advance and provided helpful comments. Dr. Waddle, Dr Arpat Ozul, and Dr. Gary White helped me out with the MARK program. James Collee assi sted me with the multivariate statistical analysis and spent time studying and analyzing the complexity of the results and their interpretations. Jeff DiMaggio and David Hoyt helped me to cut the dense scrub vegetation for making four trapping grids during three months and provided logistic help during field work. When I did not have transportation to go to Cedar Key, Jason Hall shar ed the field vehicle assigned to him with me and Jeff DiMaggio provided transp ortation in the reserve. The st aff members in the department, Sam Jones, Monica Lindberg, McRae Caprice and Delores Tillman were always helpful in all my requests. Dr. Susan Carr and Zachariah We lch lent PC-ORD and books to me. All these years have been difficult, and I always had my mother, my wife, and my three children supporting and giving me their love and the incentive that I needed to complete my education. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................7LIST OF FIGURES .......................................................................................................................11ABSTRACT ...................................................................................................................... .............16 CHAPTER 1 INTRODUCTION ................................................................................................................ ..182 PRESCRIBED BURNING PLAN .........................................................................................37Introduction .................................................................................................................. ...........37Objectives .................................................................................................................... ...........37Methods ..................................................................................................................................38Before Burning ................................................................................................................38Measurements During Prescribed Burning .....................................................................39Measurements After Prescribed Burning ........................................................................41Predicting Fire Behavior ..................................................................................................42Results and Discussion ........................................................................................................ ...43Rate of Spread, Flame Length, and Fire Intensity ...........................................................43Fuel Model Predictions ....................................................................................................443 RESPONSES OF LON-UNBURNED SC RUBBY FLATWOODS TO BURNING ............65Introduction .................................................................................................................. ...........65Objective ..................................................................................................................... ............75Methodology ................................................................................................................... ........75Results .....................................................................................................................................80Species List and Recovery Modes ...................................................................................80A Site Analysis ................................................................................................................80Postburn Recovery and Survival .....................................................................................82Structural and Compositional Changes in Response to Prescribed Burning ...................86Flowering and Fruiting after Prescribe Burning ..............................................................89Discussion .................................................................................................................... ...........89True Control Sites ............................................................................................................89Fire Survival and Recovery Modes .................................................................................91Speed of Recovery Process .............................................................................................92Community Shift in Response to Prescribed Burning .....................................................98Age at First Flowering af ter Prescribed Fire .................................................................102 5

PAGE 6

4 SMALL MAMMALS RESPONSEs TO PRESCRIBED FIRE ...........................................157Introduction .................................................................................................................. .........157Objectives and Research Hypotheses ...................................................................................165Methodology ................................................................................................................... ......165Trapping Methods .........................................................................................................165Data Analysis .................................................................................................................167Assumptions of the Cormack-Jolly-Seber model ...................................................168The goodness of fit (GOF) test ...............................................................................168Model comparison, model selection, and hypothesis testing .................................170Results ...................................................................................................................................174Number of Captured Individuals in Treatment and Control Sites .................................174Number of Captured Individua ls in Scrubs and Wetlands ............................................175Flood and Fire Effects on Podomys floridanus .............................................................177Flood and Fire Effects on Sigmodon hispidus ...............................................................179Trapping Predators ........................................................................................................182Discussion .................................................................................................................... .........182Low Capture Success ....................................................................................................182Flooding Effect ..............................................................................................................185Population Responses during and after Prescribed Burning .........................................187Burrows as refugia durin g prescribed burning .......................................................187Prescribed burning effect .......................................................................................189Population increases/decreases after prescribed burning .......................................190Habitat selection: immigration, emigra tion, and returning to burned areas ...........1965 CONCLUSIONS AND RECOMENDATIONS ...................................................................225Conclusions ...........................................................................................................................225Prescribed Burning Plan ................................................................................................225Plant Species Responses to Prescribed Burning ............................................................225Small Mammal Responses to Prescribed Burning ........................................................227Recommendations ............................................................................................................... ..228 APPENDIX A TWO TAILS T TEST COMPARISON AM ONG ENVIRONMENTAL VARIABLES AND FIRE BEHAVIOR CHARACTERISTI CS BETWEEN SITES 5C AND 2M ...........233LIST OF REFERENCES .............................................................................................................234BIOGRAPHICAL SKETCH .......................................................................................................248 6

PAGE 7

LIST OF TABLES Table page 2-1 Comparison of the prescribed burning pl an between sites 5C and 2M in Cedar Key Scrub State Reserve. .......................................................................................................... 472-2 Environmental variables dur ing prescribed burning in si tes 5C and 2M in Cedar Key Scrub State Reserve. .......................................................................................................... 482-3 Observed fire behavior characteristics in sites 5C and 2M in Cedar Key Scrub State Reserve. ...................................................................................................................... ........493-1 List of plant species recorded in quadrat s in treatment and control sites in Cedar Key Scrub State Reserve. ........................................................................................................10 43-2Species richness, Simpsons index Shannon-Wieners index and ShannonWieners evenness for preburn conditions in control (5A & 5D) and treatment (5C & 2M) sites in Cedar Key Scrub State Reserve. ..............................................................1053-3Multiple mean comparison (Duncans test ) between clusters created by using Euclidean distances and Wards minimu m variance linkage fusion. Abundance and mean percent cover were standa rdized. Significant level = 0.05. ....................................1063-4T test, ANOVA test, and Kruskal-Wallis test for comparing means and medians among treatment and control sites under preburn conditions in Cedar Key Scrub State Reserve. Data for T test and ANOVA were standardized. Test for ANOVA and Kruskal-Wallis is F test and Chi-squared test, respectively. Significant level = 0.05 .....1073-5 Preand postburn absolute densities for all species in 5C in Cedar Key Scrub State Reserve. Absolute densities for control si tes 5A and 5D are also shown. d = days. M = months. ..................................................................................................................... ....1083-6 Preand postburn absolute frequencies for all species in 5C in Cedar Key Scrub State Reserve. Absolute frequencies for c ontrol sites 5A and 5D are also shown. d = days. M = months. ............................................................................................................1093-7 Preand postburn absolute mean % cover of herb and woody species in 5C in Cedar Key Scrub State Reserve. Absolute mean pe rcent cover for control sites 5A and 5D are also shown. d = days. M = Months. ...........................................................................1103-8 Preand postburn absolute importance va lues of herb and woody species in 5C in Cedar Key Scrub State Reserve. Absolute im portance values for control sites 5A and 5D are also displayed. d = days. M = Months. ................................................................1113-9 Preand postburn absolute densities of herb and woody species in 2M in Cedar Key Scrub State Reserve. Absolute densities fo r control sites 5A and 5D are also shown. d = days. M = months. .....................................................................................................112 7

PAGE 8

3-10 Preand postburn absolute frequencies of herb and woody species in 2M in Cedar Key Scrub State Reserve. Absolute frequencies for control sites 5A and 5D are also presented. d = days. M = Months. ....................................................................................1133-11 Preand postburn absolute mean percent cover of herb and woody species in 2M in Cedar Key Scrub State Reserve. Absolute mean percent cover fo r control sites 5A and 5D are also shown. d = days. M = Months. ..............................................................1143-12 Preand postburn absolute importance va lues of herb and woody species in 2M in Cedar Key Scrub State Reserve. Absolute im portance values for control sites 5A and 5D are also displayed. d = days. M = months. .................................................................1153-13 Density of ramets in 5C after prescribed burning in Cedar Key Scrub State Reserve. ...1163-14 Density of ramets in 2M after prescribed burning in Cedar Key Scrub State Reserve. ..1173-15 Tree mortality in quadrats and in grid s 5C and 2M in Cedar Key Scrub State Reserve. ...................................................................................................................... ......1183-16 Coefficient of determination (r2) resulting from Detrended Correspondence Analysis of the preand postburn sites of a lo ng-unburned scrubby flatwoods in Cedar Key Scrub State Reserve. ........................................................................................................11 93-17 Summary statistics of the Multi-response Permutation Procedure for woody absolute densities and mean percent c over between control and treatment sites at preburn and 12 months postburn in Cedar Key Scrub State Reserve. Results are given for Euclidean and Sorensen distances. ..................................................................................1203-18 Multiple comparison for absolute densities and mean percent cover between control and treatment sites at 12 months post burn and between preburn and 12 months postburn values of treatment sites in Cedar Key Scrub State Reserve. ...........................1213-19 Comparison among several studies and Cedar Key Scrub State Reserve (CKSRR) regarding common variables measured in each research. The data presented for bareground and for plant species is mean percent cover. Data for Schmalzer and Hinkles study average all strata. Data for CKSSR is the average of the two treatment sites. ........................................................................................................................ .........1224-1 Number of captured i ndividuals in the scrubby flat woods and in the vegetation surrounding wetlands per trapping session in treatment and control sites in Cedar Key Scrub State Reserve. .................................................................................................2014-2 The Goodness of Fit test for the general model phi(g*t) p(g*t) for Podomys floridanus and Sigmodon hispidus ..................................................................................2024-3 Result browser of the fitted candidate model set of 16 models for Podomys floridanus after adjusting to c-ha t = 1.1387 in treatment and control sites in Cedar Key Scrub State Reserve. .................................................................................................203 8

PAGE 9

4-4 Set of 25 models after a dding phi(Flood + Fire) in combin ation with p constant (.), time dependent (t), and group dependent (g) in the analysis Flood and Fire effect on survival probabilities of Podomys floridanus in treatment and c ontrol sites in Cedar Key Scrub State Reserve. .................................................................................................2044-5 Estimated survival (phi) and recapture (p) parameters by using model phi(Flood+ Fire) p(.) in the analysis Flood and Fire d effect on survival probabilities of Podomys floridanus in treatment and control sites in Cedar Key Scrub State Reserve. Confidence interval = 95%. .............................................................................................2054-6 Set of 28 models after a dding phi(Flood + Fire) in combination with p(Flood + Fire), p(Flood), and p(Fire) in the analysis Flood and Fire effect on survival probabilities of Podomys floridanus in treatment and control sites in Cedar Key Scrub State Reserve. .2064-7 Estimated parameters from models phi(Flood+ Fire) p(.) in the analysis Flood and Fired effect on the survival probabilities of Podomys floridanus in treatment and control sites in Cedar Key Scrub State Reserve. Confidence interval = 95%. ................2074-8 Estimated survival parameters (phi) by using model averaging for the set of 28 models in the analysis Flood and Fire eff ect on the survival probabilities of the Podomys floridanus in treatment and control sites in Cedar Key Scrub State Reserve. C-hat = 1.1387; 95% confidence interval. .......................................................................2084-9 Result browser of the candida te model set of 16 models for Sigmodon hispidus fitted and adjusted to c-hat = 1.5694 in treatment and control sites in Cedar Key Scrub State Reserve. ...................................................................................................................2094-10 Set of 22 models after a dding phi(Flood + Fire) in combin ation with p (., g) to the candidate model set in the an alysis Flood and Fire effect on survival probabilities of Sigmodon hispidus in treatment and control sites in Cedar Key Scrub State Reserve. ...2104-11 Estimated survival (phi) and recapture (p) parameters by using model phi(Flood) p(.) and phi (Flood + Fire) p(.) in the anal ysis Flood and Fired effect on survival probabilities of Sigmodon hispidus in treatment and control sites in Cedar Key Scrub State Reserve. Confid ence interval = 95%. .....................................................................2114-12 Estimated parameters from models phi(Flood) p( .) and phi(Flood + Fire) p(.) in the analysis Flood and Fire effect on the survival probabilities of Sigmodon hispidus in treatment and control sites in Cedar Key Sc rub State Reserve. Confidence interval = 95%. .......................................................................................................................... .......212 4-13 Estimated survival parameters (phi) by using model averaging for the set of 25 models in the analysis Flood and Fire e ffect on the survival probabilities of Sigmodon hispidus in treatment and control sites in Cedar Key Scrub State Reserve. C-hat = 1.5694; 95% confidence interval. .......................................................................2134-14 Predators captured before the 8th trapping session in treatment and control sites in Cedar Key Scrub State Reserve. ......................................................................................214 9

PAGE 10

4-15 Trapping effort carried out by several studies conducted on small mammals in Florida. Code: TN = trapping nights. ...............................................................................215 10

PAGE 11

LIST OF FIGURES Figure page 1-1 Cedar Key Scrub State Reserve.. .......................................................................................251-2 Thirty years of weather data for Cedar Key (Accuweather.com). .....................................261-3 Weather data from local station in Cedar Key Scrub State Reserve. .................................271-4 Three species of oaks in Ce dar Key Scrub State Reserve. ...............................................281-5 Ericaceous shrubs in Ceda r Key Scrub State Reserve. ......................................................291-6 Palm species in Cedar Ke y Scrub State Reserve. .............................................................301-7 Herbaceous species in Cedar Key Scrub State Reserve. ...................................................311-8 Two rodent species found in Cedar Key Scrub State Reserve. ..........................................321-9 Two cotton rodents found in Cedar Key Scrub State Reserve. .........................................331-10 Location of treatment (5C and 2M) and control sites (5A and 5D) in Cedar Key Scrub State Reserve.. ......................................................................................................... 341-11 Treatment sites in Cedar Key Scrub State Reserve. ..........................................................351-12 Control sites in Cedar Key Scrub State Reserve.. ..............................................................362-1 Improvement of the firebreak located at th e north side of site 2M in Cedar Key Scrub State Reserve. .....................................................................................................................502-2 Sensor with bands of temp erature-sensitive painting. .......................................................512-3 Sensors experimented temperature 1094 C. ..................................................................522-4 Prescribe burning maps in Ce dar Key Scrub State Reserve. .............................................532-5 Fire line created by the fi refighter at the border of the stand in Cedar Key Scrub State Reserve. ...................................................................................................................... ........542-6 Examples of fire front in Cedar Key Scrub State Reserve.. ...............................................552-7 Applying prescribed burning in Cedar Key Scrub State Reserve. .....................................562-8 Taking pictures with a reference person of know height in Cedar Key Scrub State Reserve. ...................................................................................................................... ........57 11

PAGE 12

2-9 Effect of prescribed burning in the scrubby flatwoods in site 5C in Cedar Key Scrub State Reserve.. ....................................................................................................................582-10 Effect of prescribed burning in the scrubby flatwoods in site 2M in Cedar Key Scrub State Reserve. .....................................................................................................................592-11 After prescribed burning in Ce dar Key Scrub State Reserve. ...........................................602-12 Burned animals in Cedar Key Scrub State Reserve. ..........................................................612-13 Rate of spread comparison between Ce dar Key and fuel models 4, sh5, and sh8 according to BehavePlus 3.0.2. ..........................................................................................622-14 Flame length comparison between Cedar Key and fuel models 4, sh5, and sh8 according to BehavePlus 3.0.2. ..........................................................................................632-15 Examples of flame length in Cedar Key Scrub State Reserve. ..........................................643-1 Quadrats in Cedar Ke y Scrub State Reserve....................................................................1233-2 Ramets of Quercus myrtifolia in Cedar Key Scrub State Reserve. ................................1243-3.Ramets of Quercus chapmanii in Cedar Key Scrub State Reserve. ................................1253-4 Ramets of Lyonia ferruginea in Cedar Key Scrub State Reserve. ..................................1263-5Ramets of Gaylussacia nana in Cedar Key Scrub State Reserve.. ..................................1273-6 Similarity coefficients among the four study sites in Cedar Key Scrub State Reserve. ..1283-7 Median linkage dendrogram for herb and woody species in treatment and control sites under preburn conditions in Cedar Key Scrub State Reserve. .................................1293-8 Scatterplot of the first tw o canonical axes corresponding to the cluster analysis with Median linkage fusion method for herb a nd woody species in treatment and control sites under preburn conditions in Cedar Key Scrub State Reserve. .................................1303-9 Average linkage dendrogram for herb a nd woody species in tr eatment and control sites under preburn conditions in Cedar Key Scrub State Reserve. .................................1313-10 Scatterplot of the first tw o canonical axes corresponding to the cluster analysis with Average linkage fusion method for herb a nd woody species in treatment and control sites under preburn conditions in Cedar Key Scrub State Reserve. ................................1323-11 Wards minimum-variance linkage de ndrogram for herb and woody species in treatment and control sites under preburn conditions in Cedar Key Scrub State Reserve. ...................................................................................................................... ......133 12

PAGE 13

3-12 Scatterplot of the first tw o canonical axes corresponding to the cluster analysis with Wards minimum-variance linkage fusion method for herb and woody species in treatment and control sites under prebur n conditions in Cedar Key Scrub State Reserve. ...................................................................................................................... ......1343-13 Duncans multiple comparisons for the median abundances of Serenoa repens and Vaccinium myrsinites among treatment and control si tes in Cedar Key Scrub State Reserve. ...................................................................................................................... ......1353-14 Preburn and postburn mean percent cover of bareground, litter, and debris in Cedar Key Scrub State Reserve. .................................................................................................1363-15 Preburn and postburn vege tation height in 5C and 2M in Cedar Key Scrub State Reserve. ...................................................................................................................... ......1373-16 Preburn and postburn absolute densities of the most abundance herb species in Cedar Key Scrub State Reserve. .................................................................................................1383-17 Preburn and postburn absolute frequencie s of the most abundan ce herb species in Cedar Key Scrub State Reserve. ......................................................................................1393-18 Preburn and postburn absolute mean pe rcent cover of the most abundance herb species in Cedar Key Scrub State Reserve. .....................................................................1403-19 Preburn and postburn absolute importance values of the most abundance herb species in Cedar Key Scrub State Reserve. ..................................................................................1413-20 Preburn and postburn abso lute densities of the most abundance woody species in Cedar Key Scrub State Reserve. ......................................................................................1423-21 Preburn and postburn absolute frequenc ies of the most abundance woody species in Cedar Key Scrub State Reserve. .....................................................................................1433-22 Preburn and postburn absolute mean percent cover of th e most abundance woody species in Cedar Key Scrub State Reserve. .....................................................................1443-23 Preburn and postburn absolute importa nce values of the most abundance woody species in Cedar Key Scrub State Reserve. .....................................................................1453-24 Preburn and postburn absolute ramet dens ity of the most abundance herb species in Cedar Key Scrub State Reserve.. .....................................................................................1463-25 Preburn and postburn absolute ramet de nsity of the most abundance woody species in Cedar Key Scrub State Reserve.. .................................................................................1473-26 Preburn and postburn species richness in treatment sites in Cedar Key Scrub State Reserve.. ..................................................................................................................... ......148 13

PAGE 14

3-27 Preburn and postburn species diversity in treatment sites in Cedar Key Scrub State Reserve. ...................................................................................................................... ......1493-28 Preburn and postburn evenness in treatment sites in Cedar Key Scrub State Reserve. ...1503-29 Detrended correspondence analysis (DCA) sa mple ordination for densities in 5C in Cedar Key Scrub State Reserve. ......................................................................................1513-30 Detrended correspondence analysis (DCA) sample ordination for mean % cover in 5C in Cedar Key Scrub State Reserve. ............................................................................1523-31 Detrended correspondence analysis (DCA) sa mple ordination for densities in 2M in Cedar Key Scrub State Reserve. ......................................................................................1533-32 Detrended correspondence analysis (DCA) sample ordination for mean % cover in 2M in Cedar Key Scrub State Reserve. ...........................................................................1543-33 Stand and species ordinati on of oak-saw palmetto scrub based on preburn absolute mean percent cover in Kennedy Space Center. ...............................................................1553-34 Site and species ordination of scrubby flatwoods based on preburn absolute mean percent cover in Cedar Key Scrub State Reserve. ...........................................................1564-1 Trapping sessions (red blocks) carried out in Cedar Key Scrub State Reserve. ..............2164-2 Number of captured indivi duals per species per trapping session in treatment sites 5C and 2M in Cedar Key Scrub State Reserve. .....................................................................2174-3 Number of captured indivi duals per species per trapping session in control sites 5A and 5D in Cedar Key Scrub State Reserve. .....................................................................2184-4 Number of captured indi viduals in scrubs and wetla nds per trapping session in treatment sites 5C and 2M in Ce dar Key Scrub State Reserve. .......................................2194-5 Number of captured indi viduals in scrubs and wetla nds per trapping session in control sites 5A and 5D in Ce dar Key Scrub State Reserve. ...........................................2204-6 Survival probabilities of Podomys floridanus estimated by the model phi(Flood + Fire ) p(.) in the set of 28 models in tr eatment and control sites in Cedar Key Scrub State Reserve (chat=1.1387). ...........................................................................................2214-7 Podomys floridanuss survival probabilitie s calculated by the model phi(Flood + Fire ) p(.) and by model averaging (28 models) in treatment and control sites in Cedar Key Scrub State Reserve (chat=1.1387). .........................................................................2224-8 Survival probabilities of Sigmodon hispidus quantified by the model phi(Flood + Fire) p(.) in the set of 22 models in treatment and control sites in Cedar Key Scrub State Reserve (chat=1.5694). ...........................................................................................223 14

PAGE 15

4-9 Sigmodon hispidus s survival probabilities estima ted by the model phi(Flood + Fire) p(.) and by model averaging in the set of 22 models in treatment and control sites in Cedar Key Scrub State Reserve (chat=1.5694)................................................................224 15

PAGE 16

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy RESPONSES OF PLANTS AND SMALL MAMMAL COMMUNITIES TO PRESCRIBED BURNING IN CEDAR KEY SCRUB STATE RESERVE By Jose Lorenzo Silva-Lugo May 2008 Chair: George Tanner Major: Wildlife Ecology and Conservation Although prescribed burning is an important management tool for ecosystem restoration in Cedar Key Scrub State Reserve, this is the first study that analyzes the effect of prescribed burning on plants and small mammals. In addition, this is the first research carried out on plant community response to prescribed fire in coastal scrub on the west side of Florida, and the 12th study about the effects of prescr ibed burning on small mammals in Florida. The main objectives were to determine: (a) if there were structural and compositional changes in the plant community after prescribed burning, (b) if small mammals used wetlands as temporal refugia after prescribed fire; and (c) if pres cribed burning had a negative eff ect on the survival of the small mammal species. The experimental design consisted of two treatment and two control sites that were sampled before and after burning from D ecember 2003 to August 2006. Preburn vegetation samples were conducted one time in all sites, an d postburn vegetation samples were carried out every three months for a 12 m onth period. Fifty quadrats (4 m2 each) per site were assessed in each sampling. Resprouting was the main way of surviving and recovering from fire by the majority of the species, and almost all of the dominant species reached preburn levels during the 12 months period. This fast recovery of the ve getation after burning has been reported in the 16

PAGE 17

17 literature but not in one year. The Detrended Correspondence Analysis showed that woody species had structural and compositional change s during the first three months postburn, but there were more compositional than structural changes after that. According to the Multiresponse Permutation Procedure, the structural changes were si gnificant; therefore, there were significant changes in absolute densities in trea tment sites between preand 12 months postburn and between control values and 12 months postbur n as a consequence of prescribed burning. A total of 29,340 trapping nights were completed in treatment and contro l sites. Each site had a grid (100 traps) and a wetla nd next to it with two transect s (10 traps each). Mice were marked to monitor movements between scr ub and the vegetation surrounding wetlands during four trapping sessions before and after pres cribed burning. A total of 184 individuals of Sigmodon hispidus (cotton rat), Podomys floridanus (Florida mouse), Peromyscus gossypinus (cotton mouse), and Ochrotomys nuttalli (golden mouse) were monitored during this study. In treatment sites, mice were captured mainly in the scrub (75%) before burning, they used the vegetation surrounding wetlands as temporal re fugia for 11 months after burning, and they returned to the scrub after that. In control site s, mice were captured mainly in the scrub (91%) during the study. MARK analys is was only carried out on S. hispidus and P. floridanus because of the small sample size obtained for the other two species. MARK indicated that fire did not have a negative effect on the survival of S. hispidus I cannot state the same for P. floridanus because the parameter was not estimable. However, th e data indicated that mice moved to wetlands and survived for 11 months. These results will provide guidance to managers in prescribed burning plans to establish a fire return interval according to the recuperation of the vegetation and to maintain viable populations of small mammals.

PAGE 18

CHAPTER 1 INTRODUCTION Fire is a natural disturbance and an important ecological factor for ecosystem management. Fire has occurred across the landscape of United Stat es for at least 2 millions of years (Franz and Quitmyer 2005). This natural disturbance alters landscape structure, functions, and maintains biodiversity (Pyne et al. 1996). Ther efore, fire is an ecological pr ocess that greatly influences composition, structure, and dynamics of many ecosystems. The ecological role of fire for ecosystem management has been appreciated becau se managers rely on fire history to document land management planning and silvicultural prescr iption, to study the effects of past fires and past fire exclusion, to simulate natural fire intervals, to perpet uate communities, and to schedule prescribed fire (Pyne et al. 1996). Particularly, prescribed fire star ted to be used in the southern United States before most other regions. The prescribed burning era began after a long pe riod of fire suppression in the southeastern United States. Fire suppression started in 1890, declined in 1930, and continued through the 1940s (Williams 2002). Prescribed burning bega n in the 1930s after several scientific publications supported the idea of burning wild lands. Some of th ese publication were "The Use and Abuse of Fire on Southern Quail Preserve, published in 1931; Use of Prescribe Fire in Southeastern Upland Game Management, published in the Journal of Fo restry in 1935; and "Relation of Burning to Timber and Wildlife" in the North American Wildlife Conference in 1935 (Kennard 2007). These publications created the political atmosphere to re-introduce prescribed burning. However, prescribed bur ning was discontinued in Okefenokee Swamp in 1930 (reintroduced in 1970) and in Welaka Rese rve in 1935 (reintroduce d in 1980). The first official prescribed burning carried out on fede ral land took place in Osceola National Forest in 1943 (Stanturf et al. 2002). Therefore, the process of re -introducing prescribed burning into the 18

PAGE 19

southeastern United States was not a one-time event. But, it started to be more frequently used in 1945 and extensively used after 1980. During the 1980s, an estimated 16 million ha of forest land and 1.6 million ha of range and agricultural lands were treated with prescribed fire each year in the southern United States (Wade and Lunsford 1989). The majority of this treated area was for wildfire hazard reduction, wildlife habitat improvement, and range management. Nearly 0.81 million ha of rangeland were burned annually in Florida alone (Brown and Smith 2000 ) The main reason for carrying out such extensive prescribed burning in the Southeast, specifically in Florida, was because several natural ecosystems depend on fire. One of th ese ecosystems is the Florida scrub. Florida scrub is a distin ctive and threatened ecosystem (Myers 1990, Whelan 1995, Menges 1999, Brown and Smith 2000, Schmalzer 2003). It is distinct because it supports a high number of threatened and endangered plants and animals (Myers 1990, Stout and Marion 1993, Stout 2001). It is threatened because of natu ral fragmentation, human perturbations, and fire exclusion (Myers 1990). In addi tion, natural fire no longer occurs with the same intensity and frequency because the scrub habitat has been fragmented and reduced. Conservation of this unique ecosystem relies on management and research. Management and research of th e scrub habitat is essential fo r the survival of many plant and animal species. Prescribed burning has been the primary management technique for maintaining scrub communities because it is a fire-maintained system (Myers 1990, Whelan 1995, Menges 1999). Understanding the structural and compositional changes of the scrub communities after prescribed fire is important for making management decisions to maintain appropriate conditions for plants and animal s relying on these communities (Schmalzer and Hinkle 1992a). In addition, understanding prescribed burning effects on w ildlife is critical in 19

PAGE 20

order to provide a more comprehensive manageme nt based on the knowledge of plant and animal responses to prescribed burning. Even though the combination of management and research is needed for conservation and rest oration purposes, research about plant and animal responses to prescribed fire is strongly needed in several publ ic lands in Florida. One of these public lands is Cedar Key Scrub State Reserve (CKSSR). Prescribed burning is the main management tool in CKSSR, and it has been intensively used since 1985. Prescribed fire is considered the most potent and cr itical natural resource management tool at the reserve (DEP 1998). The main objectives of the program are to restore the natural fire regimen within the reserve, crea te a mosaic of different successional stages, and maximize ecological diversity (DEP 1998). To achieve these objectives CKSSR was divided into burn zones, and burn programs were assigned to each zone. In addition, these objectives were established because natural communities, a nd the associated plant and animal communities adapted to them, have been negatively impacted by extended periods of fire suppression. For this reason, the program targets restorin g the habitat for the endangered Aphelocoma coerulescens (Florida scrub jay) and other species of interest such as Podomys floridanus (Florida mouse) and Gopherus polyphemus (gopher tortoise). These species have specific habitat requirements that are not satisfied by long-unburned scrubs. Proper management of th e scrub habitat in the reserve will be critical for the long-term survival of these species of interest. Although prescribed burning has an important role in the reserve, no study has evaluated the effect of prescribed burning on plants or wildlife within this scrub community. Research on this topic is essential for proper management. This study is the first research conducted with the purpose of evaluating plants and small mammal responses to pres cribed burning in scrubby flatwoods in CKSSR. 20

PAGE 21

Study Area CKSSR is located in Levy County, Flor ida, approximately 4 km east of the town of Cedar Key (Figure 1-1). It consists of 1973 ha, and it was acquired in 1978 under the Environmentally Endangered Land program (DEP 1998). Cedar Key has a warm and humid climate. Based on 30 years of weather reco rds (Weather.com), annual temperature and precipitation average 20.8 C and 126.3 mm (Figure 1-2), re spectively, but year-to-year variability is high. For in stance, based on March 2004 February 2005 data from Accuweather.com and March 2005 August 2006 data from a local station in CKSSR, annual temperature and prec ipitation average 21.0 C and 110.4 mm, respectively (Figure 1-3). The heaviest rainfall typically takes place from June to September with some precipitation in all months of the year. Thunderstorms are freque nt during the summer, and lightning strikes are common. Hurricanes Charley (9-14 Augus t), Frances (25 August 8 September), and Jeanne (13-28 September) hit the Florida peninsula in 2004. Char ley did not hit Cedar Key directly, but its winds brought some rainfall to the area. Frances and Jeanne hit Cedar Key directly and brought a high precipitation into the area. A total of 372.5 mm fell in Ceda r Key during September (Figure 1-3). This is 204.5 mm over the monthly averag e precipitation (Figure 12). This amount of precipitation caused water from nearby wetlands to inundate the scrubby flatwoods and the sand pine scrub, and they remained pa rtially flooded for several weeks. Different types of soils support different pl ant communities. CKSSR has ancient dunes of aeolian origin (White 1970). Sand de posits in the reserve are considered to be part of the Silver Bluff Terrace (DEP 1998). The reserve has eight t ypes of soils that range from well-drained sandy soils in the upland to poorly drained, freq uently flooded, and mucky soils in tidal marsh (Slabaugh et al. 1996). Numerous wetlands are integrated with other communities in the 21

PAGE 22

landscape. CKSSR is a mosaic of wetlands (bas in swamps, basin marshes, depression marsh, tidal marshes, hydric hammock, and estuarine), mesic flatwoods, scrubby flatwoods, sand pine scrub, and sandhill. This mosaic of habitats makes prescribed burning difficult because each habitat has a different set of optimal burning conditions. Particularly, scrubby flatwoods are surrounded by mesic flatwoods and wetlands and may occassionaly be adjacent to sand pine scrubs. Occasionally, the prescription for an upland vegetative comm unity includes surrounding wetlands. This is one aspect of interest la nd managers because the vegetation surrounding wetlands might play a role as refugia after pr escribed fire in the scrubby flatwoods. Scrubby flatwoods occur on sites of well-drained sandy and acid soils and low in nutrients. They represent a great percentage of the total land area in the reserve, and several of them are overmature because the last wildfire in the area occurred in 1955 (DEP 1998). Scrubby flatwoods are represented by several oaks, ericaceous, palm, and herb species. The most common oaks are Quercus myrtifolia (myrtle oak), Q. geminata (sand live oak) and Q chapmanii (Chapman oak) (Figure 1-4). Ericaceous species are Lyonia ferruginea (rusty lyonia), L. lucida (fetterbush), and L. fruticosa (staggerbush) (Figure 1-5). Palms are Serenoa repens (saw-palmetto) and Sabal palmetto (sabal palm) (Figure 1-6). Herb species richness is low because of the overgrown condition of the scrub by flatwoods. Some of them are the following: Galactia elliottii (Elliots milkpea), G. mollis (soft milkpea), Solidago odora (Chapmans golden rod), Crotalaria rotundifolia (rabbit bells), Woodwardia virginica (Virginia chain fern) and several Panicum spp (Figure 1-7). Oaks and ericaceous species dominate the scrubby flatwoods, and this is the habitat for several animal species of interest. CKSSR has species of concern ( G. polyphemus and P. floridanus ) and threatened species ( Drymarchon corais couperi (eastern indigo snake) and A. coerulescens) according to Florida 22

PAGE 23

Fish and Wildlife Conservation Commission. This study deals with P. floridanus and other small mammals found in scrubby flatwoods such as Ochrotomys nuttalli (golden mouse), Peromyscus gossypinus (cotton mouse), and Sigmodon hispidus (cotton rat) (Figure 1-8 and 1-9). Even though CKSSR has the potential to be a study area for scientific res earch, it has not received the attention that it deserves from the scientific community. Little research has been conducted in CKSS R. Amoroso (1993) carried out a floristic study, in which several carnivorou s orchids species were reported. Morgan (1998) studied the association of P. floridanus with G polyphemuss burrows and vegetation characteristics. Several surveys for P. floridanus and A. coerulescens have been carried out. Dr. James Layne monitored the population of P. floridanus in one stand from 1957 to 1995. Later, Florida Fish and Wildlife Conservation Comm ission and the Department of Park and Recreation continued trapping P. floridanus in the reserve in several locations from 1995 to 1997. A. coerulescens has been surveyed annually since 1980. Since no stu dy has evaluated the effects of prescribed burning on plants and small mammals, an experime ntal design was planned to carry out this research. Four sites were selected to study the eff ects of prescribed burning. The experimental design considered two treatment sites (5C a nd 2M) and two control sites (5A and 5D) not selected at random (Figure 110 through 1-12). The park manager already had planned to burn long-unburned scrubby flatwoods in th e reserve, and we visited them to do the selection. I chose four sites with the same characteristics rega rding fire history, plant species composition, a wetland next to them, and no mechanical treatmen t (cutting or roller c hopping). The other sites fulfilled the first three criteria, but they received partial mechanical treatment. The experimental design included vegetation sampling and trapping in these sites and in the vegetation surrounding 23

PAGE 24

wetlands next to trapping grids before and after prescribed burning. The vegetation surrounding wetlands was an ectone between the scrubby flatwoods and the proper vegetation of wetlands. This dissertation was interested in determining the potential ro le of the vegetation surrounding wetlands as refugia for small mammals when adj acent scrubby flatwoods are prescribed burned. 24

PAGE 25

A B Figure 1-1. Cedar Key Scrub State Re serve. A) Location in Florida. B) Boundary of the reserve. 25

PAGE 26

A B Figure 1-2. Thirty years of weathe r data for Cedar Key (Accuweathe r.com). A) Average monthly temperature. B) Average monthly precipitation. 26

PAGE 27

A B Figure 1-3. Weather data from local station in Cedar Key Scr ub State Reserve. A) Monthly average temperature. B) Average monthly precipitation. 27

PAGE 28

A B C Figure 1-4. Three species of oaks in Cedar Key Scrub State Reserve. A) Quercus myrtifolia (myrtle oak). B) Quercus geminata (sand live oak). C) Quercus chapmanii (Chapmans oak). 28

PAGE 29

A B C Figure 1-5. Ericaceous shrubs in Cedar Key Scrub State Reserve. A) Lyonia ferruginea (Rusty lyonia). B) Lyonia lucida (fetterbush). C) Lyonia fruticosa (staggerbush). 29

PAGE 30

A B Figure 1-6. Palm species in Ceda r Key Scrub State Reserve. A) Serenoa repens (scrub palmetto). B) Sabal palmetto (sabal palm). 30

PAGE 31

A B C Figure 1-7. Herbaceous species in Ce dar Key Scrub State Reserve. A) Galactia elliottii (Elliots milkpea). B) Solidago odora (Chapmans golden rod). C) Galactia mollis (soft milkpea). A 31

PAGE 32

A B Figure 1-8. Two rodent species found in Cedar Key Scrub State Reserve. A) Podomys floridanus (Florida mouse). B) Ochrotomys nuttalli (golden mouse). 32

PAGE 33

A B Figure 1-9. Two cotton rodents found in Cedar Key Scrub State Reserve. A) Peromyscus gossypinus (Cotton mouse). B) Sigmodon hispidus (cotton rat). 33

PAGE 34

Figure 1-10. Location of treatment (5C and 2M ) and control sites (5A and 5D) in Cedar Key Scrub State Reserve. Yellow blocks are tr apping grids installed in each site. The actual size of each site is bigger than the grid. 34

PAGE 35

A B Figure 1-11. Treatment sites in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. There were two transects of traps between grid and the wetland. 35

PAGE 36

A B Figure 1-12. Control sites in Cedar Key Scrub State Reserve. A) Site 5A. There were two transects of traps between grid and the wetland. B) Site 5D. 36

PAGE 37

CHAPTER 2 PRESCRIBED BURNING PLAN Introduction The prescribed burn plan for sites 5C and 2M was elaborated by the Park Manager Jeff DiMaggio. The prescribed burn plan considered th e most relevant environmental variables that affect fire behavior and other f actors of interest such as area to be burned, fire history, plant communities, smoke screening test, smoke sensitive areas, fire break/site preparation, special precaution for specific areas, firing procedure, co ntacted agencies, burn zone map, weather data, preburn checklist, safety procedures, and required personel and equipmen t. This plan required inspection of the sites to be burned in order to observe the condition of the vegetation (not previously burned or roller chopped) and to monitor environmental variables and the Keetch Byram Drought Index (KBDI) index (Keeth and Byra m 1968) to assure that prescribed burning would be conducted under preferred environmen tal conditions. In addition, prescribed burning should be the same across treatment sites. Fire behavior characteristics should not vary between treatments. This is a very strong constraint from the experimental point of view. Even though this was a difficult challenge to achieve, the park manager devel oped two plans with the same obj ectives and outcomes expected from them. Objectives The prescribed burning plan for sites 5C a nd 2M had the following objectives: (a) reduce vegetative biomass on the ground by a minimum of 60%, (b) top kill woody vegetation by a minimum of 75%, and (c) reduce the vegetative mass for habitat improvement for listed species. 37

PAGE 38

This dissertation also had the following objectives : (a) to quantify rate of spread and flame length, (b) to indirectly estimat e fire intensity through recordi ng temperature during prescribed burning, (c) to determine if fire be havior characteristics were the sa me in treatment sites, and (d) to compare the results from Cedar Key fires with the predicted fire behavior from three model fuels in BehavePlus 3.0.2. Methods Before Burning Mowing vegetation at the border of stands was th e first site preparation (Figure 2-1). This job was carried out by using a Gyrotrac machine with a drum mounted on the front with many cutting blades. The border was mowed up to 3 m wi dth around the stand and at least 2 months in advance of burning. In addition, a Brown tree cutter machine cut the remaining scrub to mineral soil to improve the firebreak. Placing markers, taking vegetation samples, and installing sensors were needed to quantify rate of spread, moisture content, and temp erature during prescribed burning, respectively. Existing pine trees, posts, and flags were used to mark specific places for quantifying distance and time from the starting ignition po int. These distances and times were used to calculate rate of spread. A stratified random sampling was used to take vegetation samples in 20 points the day before burning between 11:00 am and 4:00 pm. Even though sampling points were assigned randomly, samples were selected in relatively un disturbed places and representative of the fuel complex and species composition. In each sampling poi nt, a total of 20 samples of dry fuel were collected for each of the following classes: 1-h timelag (< inch diameter), 10-h timelag (1/4 to 1 inch), 100-h timelag (1 to 3 inches). In a ddition, 20 samples of live herb and woody vegetation were collected including only stem, branches, and leaves. Samples were put in brown paper bags, 38

PAGE 39

labeled, and weighed in situ Later, samples were oven-dried fo r 10 days at 60 C and reweighed. Fuel moisture content was calcu lated by using Equation 2-1: Percent moisture content = x 100 (2-1) At the same sampling point, two temperature sensors (Figure 2-2) were installed to measure temperature during burning. Mica ta gs (12.7 x 7.6 cm; material re sistant to high temperature) were painted with five Tempilaq temperature-sensitive paints that melted at the following temperatures: 204 C, 427 C, 621 C, 816 C, a nd 1093 C (see Figure 2-3). One tag was inserted at the ground level and the other was att ached with wire to a tree at 1.5 m above the ground. Details about the prescribed burning plan in 5C and 2M are displayed in Table 2-1. According to this table, the plan was the same for both sites. The differe nces between 5C and 2M prescribed burning plans were the area to be burned, the estimated flame length, and personnel needed. Measurements During Prescribed Burning The prescribed burning map for site 5C is pres ented in Figure 2-4. This map illustrates the ignition point (NE corner of the si te) and where the fire lines ended (SW corner of the site). Fire line is the fire spread by the fire fighter at intervals of approximate ly 15-20 m at the border of the site in order to control the spr ead of the fire (Figure 2-5). The test fire was completed at 10:38 am, and two fire lines were immediately started at the border of the site. The first line started on the north side of the site, moved west and later moved south on the west side of the stand. The second line began at the same poi nt where the first line was started and moved south on the east 39

PAGE 40

side of the site. The two lines met at the SW corner of the stand. This burning design was planned to burn the site in an effective way. Burning with a combination of back and head fire made the plan successful. The park manager knew in advance from previous days that wind was blowing mainly in the SE-SW direction and planned to start burning at the SW corner of the site. However, wind direction shifted to NW-W the evening before burning and in the morning of the burning day. Therefore, the park manager decided to start burning at th e NE corner due to the shift in wind direction and because 5C should start burning wi th back fire. The two fire lines moved simultaneously, but the first line burned faster than the second one. Th en, the park manager coordinated the movement of fire lines on both the west a nd east sides of the stand in such a way that they did not move farther than half site 5C going s outh. The idea was to burn half 5C at the north side first by using head fire from the west side fi re line (moving south on the west si de of 5C) and back fire from the east side fire line (Figure 26). In this way, we created a firebreak for the head fire originated by the time the fire lines were near th e SW corner of the stand. This plan worked even though wind direction changed 14 times between 10: 38 am and 2:09 pm. By the last time, 5C was completely burned. Prescribed burning in 2M used the same t echnique as in 5C. Wind direction was SE-SW on previous days, so the park manager decided to st art at point A of 2M (Figure 2-4) in order to burn the treatment site first (where the trapping gr id was located). Fire te st was carried out at 10:54 am (Figure 2-7) and immediat ely after it, a first fire line started going NE at the border of 2M (Figure 2-4 and 2-7). From 10:54 am to 1:25 pm, the front fire (back fire) moved slowly and reached transect C of the trapping grid (approxi mately half the distance between the first and third yellow arrow at the border of 2M on Figur e 2-4), and from 1:25 pm to 2:20 pm, it moved 40

PAGE 41

fast (head fire) because the wind changed direct ion to NE. This change in wind direction not only increased rate of spread, but also increased fire intensity. Bu rning the side of 2M where the trapping grid was located finished at 2:20 pm. Starting points B and C began at 11:20 am and 4:00 pm, respectively, and the corresponding fire lines burned the rest of 2M. Environmental variables were quantified during prescribed burning every 30 minutes. A Dwyer hand-held wind meter was used to record wind speed, and a Sling-Psychrometer was used to measure air temperature and relative humidit y. In addition, cloud type, state of weather, and fire conditions were monitored. Photograph documentation and monitoring time of back and head fires were carried out during prescribed burning. A digita l camera was used to take pictur es and mini videos (up to 3 minutes) during the entire process. Since these pi ctures would be used to estimate flame length, an object or a firefighter was used as a reference (Figure 2-8) Out of 92 and 79 pictures taken in 5C and 2M, respectively, 30 pictures were selected in each site to measure flame length. Rate of spread of the fire front was measured from the tim e the firefighter started to make the fire line. Special attention was focused on changes in wind di rection in order to quantify back or head fire rate of spread. Back and head fires were only qu antified up to 20-30 m from the border of the site because the visibility was limite d due to smoke. Standing at the top of a truck helped me to watch and to record the advance of the fire front until it reached marked pine trees, posts, or flags. Measurements After Prescribed Burning Sites 5C and 2M were monitored right after pr escribed burning. Firefi ghters reviewed each site after burning, particularly 8-10 m from the borde r of each site in order to look for spots still burning and to stop these fires by adding water. In addition, firefighters also sprayed water on pine trees still burning. The burned areas were re-checked during the night and the next day for 41

PAGE 42

smoke and flare-ups. I visited sites 5C and 2M the next day after pres cribed burning with the purpose of collecting temperature sensors, estimating flame length with the pictures in situ taking pictures, and looking for wildlife killed by fire. Predicting Fire Behavior The program BehavePlus 3.0.2 (Andrews and Be vins 2005) was used to predict fire behavior by using the default worksheet. This wo rksheet contains fuel models, fuel moisture, surface wind speed, and slope steepness. Fuel m odel 4 (Chaparral), fuel model sh5 (high load, dry climate shrub), and fuel model sh8 (high lo ad, humid climate shrub) were run with the moisture contents measured for the three types of dry fuel, live herbs, and live woody vegetation collected before burning and the wind speed re corded during burning. Slop e steepness was input to zero. Fuel model 4 was selected because this is the model for the sh rub group characterized by the California mixed chaparral. Fuel model sh5 was chosen because both sites had low precipitation during the last 2 weeks before burning (91.1 mm in April and 93.6 mm in May). However, model sh8 was also selected because the KBDI index during prescribed burning (5C = 220 and 2M = 234) suggested soils had wet condi tions. The minimum, maximum, and average wind speed registered during prescribed burning a nd the average fuel moisture content for each dry and live fuel type were input into each m odel. As a consequence, each model produced three rates of spread and flame lengths. These results were compared with the minimum, maximum, and average rates of spread a nd flame length observed during pr escribed burning and calculated from the distance/time quantified in situ for head fire and fr om the pictures taken during prescribed burning. 42

PAGE 43

Results and Discussion Rate of Spread, Flame Length, and Fire Intensity Table 2-2 shows duration of prescribed burning, wind dire ction change, surface wind speed, air temperature, air rela tive humidity, fuel moisture content, and KBDI index during prescribed burning in 5C and 2M. The duration of the burning was approximately the same in both sites. However, the time recorded for 2M corresponded only to the burning time for the portion of 2M where the trapping grid was instal led. Wind changed direction 14 times in 5C and one time in 2M. This un-controlled variable affected fire behavior in 5C and the difference between the two sites regarding how burning oc curred in both sites. Wind speeds were not significantly different between the two sites, though, but air temperature a nd air relative humidity were significantly hi gher in 2M than in 5C (Appendix A). Average fuel moisture content for each fuel type was slightly higher in 2M than in 5C with the exception of live herbaceous, but they were not significantly differe nt (Appendix A). Probably, this was due to the higher air temperature and relativ e humidity found in 2M. The KBDI index suggested wet soils in both sites by the time of the burning Observed fire behavior characteristics such as flame length, rate of spread, and fire intensity are presented on Table 2-3. Flame length a nd rate of spread were slightly higher in 5C than in 2M, but they were not significantly di fferent (Appendix A). Average fire intensity was significantly lower in 5C than in 2M. The cha nge in wind direction probably was one of the factors that varied and lowered fire intensity in 5C (Appendix A). In this site, out of 60 censors, the temperature registered by 28 censors varied between 204 C and 816 C, and the rest of the censors experienced temperatures equal or higher than 1093 C. I state that the temperature was higher than 1093 C because there was no paint re maining on the mica sheets. In 2M, out of 60 censors, five censors recorded 204 C, five cen sors registered 427 C, one censor recorded 816 43

PAGE 44

C, and the rest experienced temperatures equal to or higher than 1093 C. So, fire intensity was higher an almost homogeneous in 2M in comp arison with 5C. Therefore, fire behavior characteristics were not exactly the same in treatme nt sites, but at least flame length and rate of spread were similar. Even though fire intensity wa s not the same in treatm ent sites at the heights where tempetature was recorded, it was high en ough to reduce almost 100% of the vegetation and top kill almost 100% of the above-ground woody vegetation in both site s with very little damage to wildlife. The objectives of the prescrib ed burning plans were achie ved. Figures 2-9 through 2-11 illustrate how sites 5C and 2M appeared the day after burning. Almost all trees were burned in 5C and all trees were burned in 2M. Since fire in tensity varied in 5C, flames did not completely consume all tree foliage. Approximately, 7% of the trees did not burn completely, but the stems fell down after several weeks. Also, I only found two Terrapene carolina bauri (eastern box turtle) in 5C, one in 2M, and several insects bu rned after prescribed burning (Figures 2-12). Fuel Model Predictions The comparison among models and the results found in Cedar Key are shown in Figures 213 and 2-14. A clarification about this comparis on is needed because models predicted values according to mathematical models, but this is not the case for the relationship between observed rate of spread / flame length and surface wind sp eed in Cedar Key. The minimum, average, and maximum observed values of wind speed were matched with the minimum, average, and maximum observed values for rate of spr ead and flame length. This match was a valid assumption because it was expected that minimum, average, and maximum rates of spread or flame length would occur when wind speed was al so at the minimum, average, and maximum value. Based on this assumption, the comparison was made. Another assumption was that surface 44

PAGE 45

wind speed measured during prescribed burning was equal to midflame wind speed used in BehavePlus. BehavePlus models a nearly lin ear relationship between rate of spread or flame length with midflame wind speed. In both sites, rate of sp read or flame length increased when wind speed increased. In addition, rate of sp read or flame length decreased when relative humidity increased. However, more data are needed to confirm th e linear relationship found in this study. Figure 213 illustrates that rates of spreads found in Ce dar Key fell between models sh8 and sh5 or between humid and dry climate shrubs in both sites. It is important to highlight that the values from Cedar Key were close to the predicted va lues from model sh8 in 2M. Model 4 and sh5 over-estimated rate of spread and flame length fo r Cedar Key. Figure 2-14 presents a different scenario. The average flame length recorded in Cedar Key was between model sh5 and model 4, but the minimum and maximum values of flame le ngth fell beyond the predicted values from the models in both sites. Now, the reality was that minimum and maximum flame lengths were obtained from pictures taken during prescribed burn ing in both sites. This is a fact illustrated on Figure 2-15. Therefore, models 4, sh5, and sh8 over-estimated flame length at the minimum wind speed and under-estimated flame length at the maximum wind speed recorded during prescribed burning in both sites. Prescribed burning studies have not reported fire be havior characteristics in the reviewed literature. Few studies have described the envi ronmental variables during prescribed burning (Abrahamson and Abrahamson 1996a, Week ly and Menges 2003, and Greenberg 2003); however, fire behavior characterist ics were not reported. This is a cr itical aspect in experimental designs that focus in fire eff ects. Although it is difficult to ach ieve the same fire behavior 45

PAGE 46

46 characteristics among sites considered under treatme nt, an attempt must be made in order to know how similar/dissimilar stands are regarding the application of the treatment.

PAGE 47

Table 2-1. Comparison of the prescribed burning plan between sites 5C and 2M in Cedar Key Scrub State Reserve. Category 5C 2M Area to burn 12 ha 33 ha Starting time 10:00 AM 10:00 AM Last burn/years fire suppressi on 1955 / 50 years 1955 / 50 years Fire procedure Baking, flanking, st rip head Baking, flanking, strip head Wind direction SE-SW SE-SW Surface wind speed (Min/Max) 11 / 23 kph 11 / 23 kph Transport wind speed (Min / Max) 15 / 32 kph 15 / 32 kph Minimum mixing height 610 m 610 m Dispersion Index Day 65 max Day 65 max Air temperature (Min / Max) 4 / 32 C 4 / 32 C Air relative humidity maximum (Min / Max) 35-50 % 35-50 % Fine fuel moisture content 8-12% 8-12% Drought index < 550 < 550 Rate of spread 2-5 m/min 2-5 m/min Flame length 2-7 m 2-5 m Personnel required 6-8 6-10 Passed smoke screening system Yes Yes 47

PAGE 48

Table 2-2. Environmental variables during prescribed burning in sites 5C and 2M in Cedar Key Scrub State Reserve. Category Category 5C 2M Date 04/21/05 05/18/05 Starting time 10:38 10:54 Ending time 14:09 14:20 Duration 3:31 h 3:26 h Wind direction change 14 times 1 time Surface wind speed (kph) Minimum 2.41 1.61 Maximum 4.83 4.83 Average 3.22 3.62 Air temperature (C) Minimum 23.89 26.67 Maximum 27.22 32.22 Average 26.60 29.31 Air relative humidity (%) Minimum 40.00 55.00 Maximum 54.00 75.00 Average 49.00 62.00 FMC 1h timelag (%) Minimum 3.81 2.43 Maximum 10.10 11.96 Average 6.51 7.42 FMC 10h timelag (%) Minimum 1.69 3.45 Maximum 19.27 22.20 Average 7.52 8.31 FMC 100h timelag (%) Minimum 6.08 4.41 Maximum 36.14 38.85 Average 13.95 14.85 FMC live herb (%) Minimum 27.36 19.66 Maximum 86.92 82.93 Average 59.20 55.93 FMC live woody (%) Minimum 42.86 39.48 Maximum 86.05 74.94 Average 60.56 62.13 KBDI index 220 234 Codes: FMC = fuel moisture content. KBDI = Keetch Byram Drought Index. 48

PAGE 49

Table 2-3. Observed fire behavi or characteristics in sites 5C and 2M in Cedar Key Scrub State Reserve. Category 5C 2M Flame length (m) Minimum 1.00 1.00 Maximum 6.50 5.50 Average 3.99 3.44 Rate of spread-back fire (m/min) Minimum 0.78 0.73 Maximum 2.29 2.22 Average 1.66 1.47 Rate of spread-head fire (m/min) Minimum 3.71 2.22 Maximum 8.67 6.67 Average 6.21 4.87 Fire intensity-Temperature at the ground level (C) Minimum 204.0 204.0 Maximum 1093.0 1093.0 Average 858.0 1019.0 Fire intensity-Temperature at 1.5 m above the ground level (C) Minimum 204.0 204.0 Maximum 1093.0 1093.0 Average 600.0 899.0 49

PAGE 50

Figure 2-1. Improvement of the firebreak located at the north side of site 2M in Cedar Key Scrub State Reserve. The road is the fire br eak and it can be seen at the left side (6 cm from left to right) of the picture. The firebreak was improved by increasing its width (the remaining 6 cm from midd le to the right of the picture). 50

PAGE 51

A B Figure 2-2. Sensor with bands of temperature-sensitive painting. A) Sensor not exposed to fire. B) First band was melted indicati ng that temperature reached 204 C. 51

PAGE 52

A B Figure 2-3. Sensors expe rimented temperature 1094 C. A) Sensor at 1.5 m above the ground. B) Sensor at the ground level. 52

PAGE 53

A B Figure 2-4. Prescribe burning maps in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M 53

PAGE 54

Figure 2-5. Fire line created by the firefighter at the border of the stand in Cedar Key Scrub State Reserve. 54

PAGE 55

A B Figure 2-6. Examples of fire front in Cedar Key Scrub State Reserve. A) Back fire. B) Head fire. 55

PAGE 56

A B Figure 2-7. Applying prescribed bur ning in Cedar Key Scrub State Rese rve. A) Fire test in site 2M. B) Fire line at the border of the site 2M. 56

PAGE 57

A B Figure 2-8. Taking pictures with a reference person of know hei ght in Cedar Key Scrub State Reserve. A) Jeff DiMaggio in 5C. B) David Romano in 2M. 57

PAGE 58

A B Figure 2-9. Effect of prescribed burning in the scrubby flatwoods in site 5C in Cedar Key Scrub State Reserve. A) Before burning. B) After burning. 58

PAGE 59

A B Figure 2-10. Effect of prescribed burning in the scrubby flatwoods in site 2M in Cedar Key Scrub State Reserve. A) Befo re burning. B) After burning. 59

PAGE 60

A B Figure 2-11. After prescribed burning in Cedar Key Scrub State Reserve. A) Site 5C showing a burned palm, palmettos, and pine trees. B) Site 2M with burned oaks and palmettos. 60

PAGE 61

A B Figure 2-12. Burned animals in Cedar Key Scrub State Reserve. A) Terrapene carolina bauri (eastern box turtle). B) Cockroaches. 61

PAGE 62

A B Figure 2-13. Rate of spread comparison betw een Cedar Key and fuel models 4, sh5, and sh8 according to BehavePlus 3.0.2. A) Site 5C. B) Site 2M. 62

PAGE 63

A B Figure 2-14. Flame length comparison between Cedar Key and fuel models 4, sh5, and sh8 according to BehavePlus 3.0.2. A) Site 5C. B) Site 2M. 63

PAGE 64

A B Figure 2-15. Examples of flame length in Cedar Key Scrub State Reserve. A) Flame length up to one m height. B) Flame length of at least 6 m height. 64

PAGE 65

CHAPTER 3 RESPONSES OF LONG-UNBURNED SC RUBBY FLATWOODS TO BURNING Introduction Florida has had colossal changes that affected the size and the dist ribution of the scrub ecosystem. The scrub habitat probably appeared dur ing the early Tertiary Period originating in the southern Rocky Mountains and northern Mexico with eventual spread along the Gulf coast to Florida (Axelrod 1958). In Flor ida, the scrub habitat formed during the early Miocene (20 million years before the present time (mybp)), and it is one of the most ancestral habitats of Florida in conjunction with the mesic fore st (Webb 1990). During the Pliocene epoch, the Florida peninsula started a process of contracti on and expansion of land area because of sea level rise during glacial and in terglacial periods. At one time, Flor ida was about two times the current size because of lower sea level and xeric conditions. During th e late-Pleistocene, the scrub vegetation was probably wide-spread across the peninsula. However, during the later part of this epoch, sea level rose and reduced the size of the Florida Peni nsula, and mesic conditions extended into the landscape (Myers 1990). During the last million years (s till the Pleistocene), broad areas of xeric habitats persisted in Florida, but they were replaced by wet subtropical (mesic) habitat because of increas ed precipitation and increased wate r tables. In this process, the original widespread and almost continuous scrub habitat was reduced and became fragmented (Webb 1974, Clark et al. 1999). The fragmentation process in creased in the last 5000 7000 years because the climate became more humid, wa ter levels rose, and electrical storms and lightning fires developed. In addition, natural fragmentation has been coupled with human perturbations in th e last 200 years. The current distribution of the Florida scrub is a cons equence of the historical biogeography of the Florida peninsula plus the effect of human pertur bations. During the post65

PAGE 66

European period, the natural fragmented scrub ecosystem was gradually reduced by conversion to housing developments, citrus groves, a nd golf courses (Myers 1990). From 1940 to 1981, approximately 64% of the xeric upland habita t was destroyed and an additional 10% was disturbed (Peroni and Abrahamson 1985). The scrubs have lost more than 60% of the original extent, and 85% of the scrubs in Lake Wales Ridge were converted to agriculture, commercial, or residential development (Peroni and Abrahamson 1985, Christman 1988). The last five decades of development have reduced considerably the extent of the scrubs (McCoy and Mushinsky 1994). These perturbations have decreased the number and size of scrub patches and have increased their isolation. As a result, the original widespread scrub ecosystem has been fragmented and reduced to patches in the in terior and along the coas t of the peninsula. Scrubs in Florida have different ages. Because of the glacial and interglacial periods, at least six ancient shorelines we re created during the 25 mybp, a nd the scrub habitats along the central portion of Florida are the oldest (approximately 9 mybp) among all scrub habitats (Myers 1990). So, the current Florida landscape can be described as old scrub habitat on the north south axis in the center of the state, and young scrub habita ts towards the coastlines (approximately 0.5-2 mybp). Also, the scrub habitat can be classified as inland or coastal scrub. Inland scrubs are the largest block of scrubs and occur along a complex of sand ridges running north-south from Clay and Putnam Counties to Highland County. These sa nd ridges, dated from the Miocene to early Pleistocene, form the Florida Central Ridge, and they include: Ocala National Forest, Lake Wales Ridge (conformed by Arbuckle, Cart er Creek, and Archbold) and Avon Park Air Force Range (Myers 1990, Clark et al. 1999). These ridges have been separated by several kilometers, and each ridge has an asse mblage of scrub patches separated by mesic habitats and by human developmen t that together act as a matr ix for scrub species. Coastal 66

PAGE 67

scrubs are the smallest scrubs found on both th e Atlantic and Gulf co asts. The northernmost examples of these scrubs are lo cated in the Panhandle and are re stricted to a narrow strip along the Gulf coast. They extend from west Ochlocko nee River in Franklin Co unty, Florida, to Gulf Bay State Park in Baldwin County, Alabama. In north-central Florida, coastal scrubs occur on the east coast in St. Johns County near Durbin and on the west coast in Levy County near Cedar Key. The southernmost scrubs are found on the we st coast at Marco Island in Collier County (probably already extirpated) a nd on the east coast in Merritt Island National Wildlife Refuge, Cape Canaveral barrier island complex (Kennedy Space Center), and Jonathan Dickinson State Park. A high number of endemic species, which ar e habitat specialists, characterize inland and coastal scrubs. Scrub is the most unique and re stricted natural ecosystem in Florida. The scrub in Florida is a shrubland ecosystem located on contemporary or relict beach dune substrates maintained by recurrent disturbances (Myers 1990, Gi bson and Menges 1995, Menges 1999). Scrub communities are dominated by a well-developed la yer of evergreen oaks (shrubs), with or without a sand pine overstory, sparse ground cover with few herbaceous plants, and many patches of bare ground occupying well-drained, in fertile, sandy soils (Layne 1963, 1990; Myers 1990). Oaks species typically include Quercus geminata, Q. myrtifolia Q inopina (inopine oak) and Q chapmanii In addition, other shrubs such as Lyonia ferruginea and Ceratiola ericoides (false rosemary) are common. Ninety percent of th e shrub layer consists of the same six species in approximately the sa me order of abundance: Q myrtifolia Q inopina, Serenoa repens L ferruginea and C ericoides (Myers 1990). The ground cover is always sparse and includes species such as Cladonia evansii (deer moss), Licania michauxii (gopher apple), Galactia spp. (milk peas), and other herbs. 67

PAGE 68

Several types of scrub have been named depending on dominant species, location, elevation, soil type, fire histor y, and other factors (Myers 1990, Menge 1999). These scrubs are rosemary scrub, oak scrub, oak-saw palmetto sc rub, sand pine scrub, slash pine scrub, and scrubby flatwoods. Rosemary scrub is the most common, and it is characterized by the common species C. ericoides and by gaps supporting an herbaceous fl ora that includes terrestrial lichens and many rare species. Oak scrub and oak-saw palmetto scrub are dominated by oaks and the association oak-saw palmetto, resp ectively. Sand pine and slash pine scrubs have sand pine and slash pine, respectively, as representative speci es in conjunction with other shrub species. The name scrubby flatwoods is applied to scrubs that either lack a pine overstory or have slash pine in place of sand pine. Scrub is one of the most endangered communities in Florida not only for the number, size, and distribution of the patches, but also because of the number of endemic plant and animal species. Scrubs have several endemic plant and anim al species. Currently, 22 plant species are federally-listed as endangered or threatened (U .S. Fish and Wildlife Service 1999). Examples of these are Ilex opaca (scrub holly), Persea humilis (silk bay), Garberia heterophylla (garberia), Palafoxia feayi (palafoxia), and Osmanthus megacarpa (wild olive). However, the truly rare endemic species are restricted to the Lake Wales Ridges such as Hypericum cumulicola (scrub hypericum), Dicerandra frutescens (scrub balm), Eryngium cuneifolium (wedge-leave snakeroot), Lupinus aridorum (Beckners lupine), and Warea carteri (Carters warea). This concentration of endemism is probably due to the age of the scrub and its isolation. A very rare shrub is Ziziphus celata (Garrets ziziphus), collected onl y twice, not seen since 1955, but rediscovered in 1987 (Myers 1990). Vertebrate endemic species that occur in the scrub are the following: Podomys floridanus (Florida mouse), Aphelocoma caerulecens (Florida scrub jay), 68

PAGE 69

Sceloropus woodi (Florida scrub lizard), Neoseps reynoldsi (sand skink), and a mole skink species with three subspecies, Eumeces egrerius egrerius (brown red-tailed skink), E e lividus (blue-tailed skink), and E e insularis. The structure and stage of the vegetation is very important for these species and others. For example, if the he ight of the scrub reaches a critical level, and a pine canopy develops, then P floridanus A caerulecens and many bird species leave the patch of scrub (Myers 1990). In addition, the developmen t of a mature scrub creates habitat for other species such as Glaucomys volans (flying squirrel), Sciurus carolenensis (gray squirrel), Ochrotomys nuttalli (golden mouse), Peromyscus gossypinus (cotton mouse), and many species of birds. Therefore, scrub species need a very specific structure and stage of the vegetation for habitat, and the only way that sc rub communities maintain this stat us is through fire periodicity. Scrub is a pyrogenic ecosystem that requires ca tastrophic fire for self-maintenance. Scrub fires are devastating, resulting in extensive consumption of the above ground vegetation. The natural frequency of fires is one every 10-100 ye ars according to Myers (1990) or one every 2050 years according to Layne (1990). In the past, the scrub ecosystem had a natural fire frequency that allowed its maintenance and persistence. In the absence of a natural frequency of fire, tree and shrub layer density increase and scrub tran sforms into xeric hardwood forest. However, when frequency of fire becomes more frequent, sand pine disappears, and the association becomes oak scrub or changes to high pine (Myers 1990). This was what actually happened during the preand post-European periods. Huma ns altered the natural frequency and intensity of fires. As a result, the scrub ecosystem become more fragmented or changed to another type of vegetation. Currently, prescribed bu rning is extensively used in Florida, and it has helped to maintain the scrub ecosystem. Under prescribed fire and natural conditions, fire maintains sand pine scrub and scrubby flatwoods (Layne 1990, Myers 1990) as stable and non-successional 69

PAGE 70

associations. How plant species respond to pr escribed burning has been analyzed by several studies. There are two general approaches to describe the effects of prescribed fire on flora. The first one takes into consideration the concept of fire regimen in or der to understand fire effect at the community level. The second one focuses on the response at the species level. Let us starts with the definition of fire regimen. The concept of fire regimen encompasses seve ral components. Kilgore (1987) defines fire regimen as a set of several factors such as fire frequency (time between fires), season of burn, fire periodicity, fire intensity, size of fire, pattern on the land scape, and depth of burn. Brown and Smiths (2000) definition includes pattern of fire occurrence, size, uniformity, and severity. Whelan (1995) considers fire regimen as a global concept to summarize fire frequency, season of burning, type of fire (only organi c layer soil, only above ground, or crown fire) and extent of the fire (continuous vs patchy). This dissertation sy nthesizes the three previous concepts and defines fire regimen as a global concept to summarize fire frequency, intensity, severity, type of fire, size, season, and extent. Fire regimen has been used in order to cat egorize plant community responses to fire. The most recent classification of plant community responses to fire uses fire severity as the main criterion. Brown and Smith (2000) used a fire regimen classification relying only on fire severity. According to those author s, the use of fire severity as the key component to describe fire regimen is interesting because it connects directly to the effect s of disturbance, especially on survival and structure of the dominant vegetation. Brown and Smiths (2000) classification is as follows: 70

PAGE 71

Understory fire regime (applies to forest and woodland vegetation). Approximately 80% or more of the aboveground dominant vegetation survive fire. Fire is not lethal for dominant vegetation and does not change it s structure. Stand-replacement regime (applies to forest s, woodlands, shrublands, and grasslands). Approximately 80% or more of the abovegroun d dominant vegetation is either consumed or killed by fires. Since fire consumed or killed the aboveground parts of the dominant vegetation, they dramatically change the structur e of it as well, particularly in shrublands and forests. Mixed severity regime (applies to forests and woodlands). Fire selectively kills species of the dominant vegetation depending on the species susceptibility to fire. This type of fire varies between understory and stand-replacement. Nonfire regime. Little or no occurrence of natural fire. All type of forest can be classified according to the categories above that correspond to low, medium, and high fire severity types. The second general approach about how plant species respond to prescribed burning takes into consideration how species survive fire. Whel an (1995) has named four categories as follows: fire ephemerals, obligate seeders, sprouters, and facultative sprouter s. The first category describes plants that do not survive the fire. The second category refers to plants that germinate after fire through a seed bank stored in the so il or in the canopy. The third category presents plant species that survive fire through protected buds in the stem s or roots. The fourth category introduces species in which the ability to sprout would depend on the ch aracteristics of the prescribed fire. However, recovery mode of individual species in Florida scrubs may be correlated with habitat characteristics and fire regime. The scrub ecosystem in Florida falls under th e category of a stand-replacement regime and the majority of the species are resprouters. Af ter a long fire-free period of fuel accumulation, a high intensity fire takes place. If sand pine or slash pines are pres ent, they might be killed. The above-ground shrub layer is totally consumed. According to Menges and Kohfeldt (1995), species recovery varies, and th ey classified 95 species of th e scrubby flatwoods and rosemary scrub into seven guilds of recovery mechan ism: 24 species were resprouters (many woody 71

PAGE 72

shrubs), 24 were resprouters and seeders (small-statured shrubs, palmettos, and herbaceous perennials), 14 were resprouters a nd clonal spreaders (the majority of the dominant shrub genera in scrubby flatwoods such as Quercus, Lyonia and Vaccinium ), five were resprouters, clonal spreaders, and seeders (herbs), 26 were obligate seeders ( C. ericoides and many herbs), one aerial seeder ( Pinus clausa ), and one seeder and survivor ( Pinus elliottii ). Shrub species sprout from previously suppressed underground buds on bur ied roots. Few shrub species, for instance C. ericoides, regenerate from seeds stored in the soil. P. clausa regenerates from fire-induced seed release from individuals with serotinous closed cones. P. elliottii is the only pine tree that might survive moderate or high-intensity fire an d also recovers by seeds. Both species might reseed from other pine trees in adjacent sta nds. Post-fire species composition is usually an assemblage of many of the speci es previously growing on the site. However, there are some variations in the way Florida scrubs recover after fire. Florida inland scrubs, such as scrubby flat woods and rosemary scrubs, have different recovery strategies (Menge a nd Kohfeldt 1995). Scrubby flatwoods specialists usually depend on vegetative recovery modes (61% ; resprouting and cl onal spread) and less often on mixed modes (23%) or obligate seeding (16%). In general, specialist speci es and dominant shrubs spread clonally and resprout in th e scrubby flatwoods. In contrast, half (50%) of the of rosemary scrub specialists are obligate seeders, 17% were mixe d, and 33% were vegetative. Also, the dominant species of C ericoides are mainly obligate seeders. Sp ecies found in both habitats are intermediate in recovery modes (32% vegetativ e, 36% mixed, 32% obligate seeders). Therefore, in general, scrubby flatwoods appear to be more favorable for post-fire resprouting and clonal growth and rosemary scrub more favorable fo r post-fire seedling recovery. This difference occurs despite the overlap in species com position (Abrahamson et al. 1984) and close 72

PAGE 73

concurrence of these communities in the landscape. Coexistence of resprouters, seeders, and species with mixed recovery mode s in Florida inland scrubs sugge sts that fire-return intervals may be quite variable. Post-fire recovery of Florid a inland scrub ecosystem varies with dominant shrub species and fire history. Most scrubby fl atwoods species recover by resp routing and/or clonal spread oaks, recovery is rapid, and there is little change in species composition at a scale of four to 10 years (Abrahamson 1984a, 1984b, Johnson and Abrahamson 1990, Abrahamson and Abrahamson 1996b). In contrast, recovery of rosemary scrub takes more time because C ericoides and P clausa (the dominant species) recover thr ough post-fire seedling establishment, and C ericoides takes a decade to reach sexual matu rity (Johnson 1982, Johnson et al. 1986). Fire return intervals for scrubby flatwoods vary from 5 to 20 years (Menges & Kohfeldt 1995) and to 20 to 80 years for rosema ry scrubs (Myers 1990, Menges 1999). Post-fire recovery of Florida coastal scrub also varies with dominant species and fire history. The majority of plant st udies conducted in coastal scrub ha ve been carried out on Merritt Island National Wildlife Refuge and Cape Canaveral (Simon 1986, Breininger and Schmalzer 1990, Schmalzer and Hinkle 1991, 1992a, 1992b, Schm alzer and Boyle 1998, Schmalzer and Adrian 2001, Schmalzer 2003, Schmal zer et al. 2003). Myers (1990) classified this scrub as a coastal scrub; Weekly and Menges (2003) refered to it as a coasta l oak-palmetto scrub. However, according to Schmalzer et al. (2003), the most common types of plant communities in Merritt Island and Cape Canaveral are oak-saw palmetto scrub, scrubby flatwoods, and coastal scrub. Schmalzer (2003) classified th e scrubby flatwoods without slas h pine overstory as oak-saw palmetto scrub Dominant species in these two comm unities were: myrtle oak, sand live oak, Chapman oak, saw palmetto, and ericaceous shrubs such as L ferruginea Schmalzer et al. 73

PAGE 74

(2003) considered coastal scrub a di fferent type of community because it was dominated by Quercus virginiana (Live Oak), Serenoa repens and ericaceous shrub species were absent. The type of soils was also different. Oak-saw palmetto scrub was on soils that varied from neutral to acid, but the majority was acid soils. Coastal scru b soils were alkaline. In oak-saw palmetto scrub, recovery of dominant oaks and ericaceous species after fire is primarily through resprouting and clonal spread (Schmalzer and Hinkle 1992a, 1992b). Resprouting allowed a rapid reestablishment of the dominant shrubs. Saw-palmetto reestablished cover faster than woody shrubs. Saw palmetto cover equaled preburn va lues between one and 1.5 year postburn and changed little after that. Q myrtifolia Q geminata, and Q chapmanii recovered rapidly after burning but at different ra tes. Cover in these three specie s equaled preburn values between 4 and 5 years postburn and changed little by 10 years postburn. Few ch anges occur through 10 years post-fire except for continued height growth. In coastal scrub that received cutting/prescribed burning tr eatments, cover of saw palmetto was reduced by mechanical treatment. Recovery of Q. virginiana was also through resprouting, which reestablished cover within 5 years postburn (Schmalzer et al. 2003). Growth rate of Q. virginiana was higher than shrubs in oak-saw palmetto sc rub under the same treatment. The rapid growth rate of these types of scrubs has suggested that prescribed burning would need to be more frequent than often applied in these communities in order to mainta in the desired structure of the vegetation. Fire return interval for oak-saw palmetto scr ub has been estimated between 5 and 20 years (Schmalzer 2003). However, the length of the restoration period need s to be determined (Schmalzer et al 2003). These types of studies are also needed on the Gulf coastal scrub. Even though some research has been conducte d on the Atlantic coastal scrub, only one study has been carried out in the Gulf coastal sc rub, particularly in the Panhandle. Ruth et al. 74

PAGE 75

(2007) studied the effect of rein troduction of fire in long-unburne d coastal scrub in Naval Live Oaks areas of the Gulf Islands National S eashore. This study addressed the effect of environmental variables and fire on plant distri bution and abundance, a nd it found that elevation and time since fire were the most important e nvironmental variables that affected species distribution and abundance. No study has been carried out on the west coast of Floridas peninsula. This is particularly important in CKSSR because prescribed burning has been practiced since 1985. Understandi ng how the scrub ecosystem in CKSSR responds to prescribed burning is critical to know the di rection and rates of changes in composition and structure of scrub communities after fire and to make effective management de cisions. This is the first study carried out to determine the responses of a long-unburned (since 1955) scrubby flatwoods to prescribe burning in the west coastal scrub of Florida. Objective Prescribed burning was applied to two long-unbu rned scrubby flatwoods sites in order to study the post-burn dynamics of this community. The objective was to document recovery modes and structural and compositional changes in the post-burn community. To achieve this objective, a site analysis was needed to determine if treatment and control sites were ecologically similar before burning. Methodology Vegetation sampling was carried out in control sites one time before prescribed burning. Sampling in treatment sites was conducted as fo llows: pre-burn and post-burn at 11 days, 3, 6, 9, and 12 months. I sampled the density of vegetation by placing a quadrat (4 m2) (Figure 3-1) on specific points of the trapping grid select ed by a stratified random sampling. Cover was quantified along a 2 m line that intercepted the center point of two opposite sides of each quadrat. This sampling measured th e following predictor variables: 75

PAGE 76

1. Ground cover of bare ground, litter, hard woody de bris, and herbaceous species measured in cm. 2. Shrub cover ( 1.5 m tall) per species was quantified in cm. 3. Number of palmetto plan ts in each quadrat. 4. Number of individuals per species of trees, sa plings, and seedlings, which were classified with the following criteria: oak and pine trees with dbh 7.6 cm, saplings with dbh < 7.6 cm and height 1.0 m, and seedlings with height < 1.0 m. 5. Maximum vegetation height in each quadrat. Flowering and fruiting were not systematical ly surveyed, but they were recorded as encountered. Before burning, the center of each quadrat was mapped with a Global Positioning System GPSmap 76S (Garmin) fitted with an external antenna (1-2 m accuracy). In addition, I marked the center each quadrat and the two of its north facing corners with a 20 cm wire. These wires helped to locate the quadrat af ter burning, and they were repl aced with flags after the burn (Figure 3-1). Counting of individual plants was done per stratum. However, there were species that could be classified as a seedling or sprout, but they could not be classified as sapling or tree because they were herbs, a lichen, a palmetto, a cactus or woody species of low height. Examples of these woody species were as follows: Vaccinium myrsinites, C ericoides Osmanthus americanus Gaylussacia dumosa Gaylussacia nana, Licania michauxii and Rhus copallinum In this case, the abundance per species was recorded without esta blishing an association with a particular stratum. Counting ramets started in sites 5C and 2M at 11 and 12 days afte r burning, respectively (Figures 3-2, 3-3, and 3-4). Since species identification is difficult at this stage, groups of ramets 76

PAGE 77

were labeled and recorded with pictures. Later, these species were identified when ramets matured. Counting of ramets was done carefully to minimize human error. Counting was carried out twice to make sure that the recorded number was accurate. If the numbers did not match, I counted slowly the third time. The two numbers that matched were selected. All ramets were counted even though they belonged to the same stem (Figure 3-5). Recovery mode was determined by excavation of ramets. In each quadrat, I excavated 10 ramets per species in order to find seedlings or resprouting individuals at 11 days and 3 months after burning. Ramets were carefully excavated us ing hand tools and fingers, with care taken to preserve root systems and rhizome connections I followed the technique used by Menge and Kohfeldt (1995) to classify seedling, resprout or clonal ramet. Seed lings were independent plants with small root systems and no sign of pre-fire biomass (e.g. no charred stem bases). Resprouts were classified as ramets resprouting within 20 cm of pre-fire stems. Ramets more than 50 cm away from pre-fire ramets were named clonal ramets. Univariate and multivariate da ta analyses were performed using SAS 9.1.3 (SAS Institute Inc. 2002-2003) and PC-ORD v.5.0 (McCune and Mefford 1999), respectively. The significance level chosen was 0.05. The data set was summa rized by computing absolute and relative abundance, density, frequency (number of occupi ed quadrats), and mean percent cover (total distance intercepted above, below, or touching by species divided by 2 m and multiplied by 100) per species. Importance values were quantif ied by summing relative values for density, frequency, and mean percent cover for each species with the exception of pine trees because they did not have mean percent cover records. 77

PAGE 78

A site analysis was carried out to determin e if treatment and control sites at preburn conditions did not differ regarding plant specie s structure and composition. This analysis was critical for the experimental design because otherwise treatment effect could not be verified. This analysis started by quantifying species richness (as number of species), diversity (Simpsons index 1/D and Shannon-Wieners index H), and evenness (as J = H/lnS, Pielou 1969) to measure structural and compositional differen ces among sites. Jaccards and Sorensens coefficients of similarity were also calculated to analyze th e species composition among sites. After that, a cluster analysis was performed to study whether there were differences or not among sites by taking into consideration absolute abundance and mean percen t cover for all herb and woody species. The dataset was standardized and outliers were deleted (McGarigal et al. 2000). In the cluster analysis, th e Euclidean distance was used for the resembling matrix and Median linkage, Average linkage, and Ward mi nimum-variance linkage were used as fusion methods. In addition, to decide th e number of significant clusters to retain, a F-ratio test in combination with Duncans test were carried out to assess the null hypothe sis that the mean for each variable was not different between multispe cies clusters. Finally, a mean or median comparison was done to determine significant di fferences among sites for the variables found with significant differences in the F-ratio te st. The idea was to determine which site(s) was (were) significantly different from other sites re garding that particular variable. Since plant species variables did not have a normal distributi on according to the Shapiro-Wilk test, the data set for abundance and mean percent cover was tr ansformed using the Arcsin function. Then, an ANOVA test and multiple comparisons (Duncan's a nd Bonferroni's procedures) were carried out to determine if at least two means were significantly different and which means were significantly different, respectivel y. If the Arcsin transformed vari able did not have a normal 78

PAGE 79

distribution, medians were compared by using Kruskal-Wallis test and Duncans multiple comparison procedure. Species richness, evenness, and diversity (S impsons and Shannon-Wieners index) were quantified to measure and documen t structural and compositional changes between the preand the postburn community. Detrended Correspondence Analysis (DCA) was used to visualize the multivariate changes in woody sp ecies densities and mean per cent cover in the preburn and postburn samples over time. Ordinations were carri ed out on absolute (to highlight structural changes) and relativized (to emphasize compositi onal changes) values of densities and mean percent cover data by using PC-ORD. Absolute values were relativized using standardization by the norm (Greig-Smith 1983). The quality of the ordinations was evaluate d with the coefficient of determination that measured the proportion of the variance represented by the ordination axes. Scatter-plots between the first two axes were done to visualize struct ural and compositional changes. Only sampling time scores, and not specie s scores, were plotted for this reason. Finally, a Multi-response Permutation Procedure (MRPP) and multiple comparison was conducted using PC-ORD. MRPP is the non-parametric test anal ogue of MANOVA. Unfo rtunately, MRPP could only be performed on treatment sites preand 12 months postburn and co ntrol sites. MRPP can only be carried out on independent samples. Therefore, it is not the right test for a sequence of sampling through time on the same quadrats. Mu lti-response Block Procedure (MRBP) is a variant of MRPP, and it can be used for dependent samples. However, it requires a second matrix in PC-ORD and a balanced design (McCune and Grace 2002). The second matrix in PC-ORD is used with environmental variable s that were not measured in th is study, and sites had different sample sizes. Thus, MRPP was carried out to te st the hypothesis of no treatment effect or no significant structural di fferences between treatments at 12 months postburn and treatments at 79

PAGE 80

preburn levels and control sites. Only the ab solute density and mean percent cover of woody species were used for this analysis because the herb data set did not have a large enough sample size. Euclidean and Sorensen distances were used to calculate the distance matrices. Results Species List and Recovery Modes Table 3-1 illustrates the list of species recorded in quadrats in treatment and control sites. A total of 10 herb species, 26 woody species, a lichen, and a c actus was recorded during the study. All species resp routed after burning. Even t hough digging was carried out on 10 individuals per species per quadr at, I did not find evidence of recovery by seeds. The only exception was Pinus clausa with seedlings at six months post burn in 5C. Also, the criterion of ramets more than 50 cm away from pre-fire ramets to name clonal ramets did not work. Therefore, I did not consider th is recovery mode. Pre-fire rame ts were burned completely in almost all quadrats. A Site Analysis A comparison of the structure and composition of the four sites is displayed by Table 3-2. As can be seen, control (5A and 5D) and treatment (5C and 2M) sites differed in species richness, species diversity, and evenness. In gene ral, control sites had hi gher species richness, species diversity, and evenness than treatment site s. The only exception is 5D and 2M sites that had the same Shannon-Wieners index and a simila r evenness. Therefore, control and treatment sites were not ecologically similar by using these criteria. Figure 3-6 reveals Jaccards and Sorensens si milarity coefficients among the four study sites. All sites had a similarity higher than 54% and 63% according to Jaccards and Sorensens coefficients, respectively. The only exception wa s Jaccards coefficient between 5C-preburn and 80

PAGE 81

5D (46%). Consequently, these communities shared some structural similarities according to the coefficients. A more powerful criterion was needed to determine if control and treatment sites were ecologically similar. Species richness, species di versity, evenness, and similarity coefficients drew different results. These cr iteria used the proportion of individuals and the number of species as data in a single dimension. A multivar iate approach such as cluster analysis gives more insight about the actual similarity among sites. Figures 3-7 through 3-12 show the results of the cluster analysis. Me dian linkage/Average linkage and Wards minimum-va riance linkage clearly displaye d one cluster (Figures 3-7 through 3-10) and two clusters (Figures 3-11 and 3-12), respectively. However, looking at Wards dendrogram, clusters were composed of a mix of sample units corresponding to treatment and control sites. As a result, I could not st ate that one cluster corresponds to control sites and the other to treatment sites. In order to decide the number of significant clusters to retain, Table 3-3 presents the re sults of the F-ratio test and D uncans test. Out of 52 variables tested, 23 (44%) did not show results because of the small sample sizes; the means of 21 (41%) variables were not significantly different between the two clusters, and the means of eight (15%) variables were signifi cantly different. Of these eight vari ables, only the mean abundances of S. repens and V myrsinites had a significant resu lt when means were compared among treatment and control sites (Table 3-4). The multiple compar ison procedure revealed that only the median of the abundance of V myrsinites in 2M preburn was significantly different from 5D and 5C preburn according to the Duncans test. Theref ore, there was enough evidence to suggest that treatment and control sites were ecologically sim ilar, and prescribed bu rning effect could be determined by comparing treatment and control sites. 81

PAGE 82

Postburn Recovery and Survival Figure 3-14 shows absolute mean percent cover of bareground, litter, an d debris in 5C and 2M. Bareground had postburn values higher in 2M than in 5C, but these values were not higher than 13 %. Even though the preburn value for litt er was lower in 5C (83.9%) than in 2M (94.5%) and the postburn values were highe r in 5C than in 2M for 3, 6, and 9 months, litter had almost exactly the same mean percent cover in both sites (5C 68.3 %; 2M 67.5%) at 12 months postburn. However, these values were lower than preburn mean percent cover in 5C, 2M, and control sites. Debris had a simila r curve pattern in both sites with very low values after burning. Preburn and postburn vegetation height in 5C and 2M is presented in Figure 3-15. As revealed by the graph, preburn height in 5C wa s higher than in 2M a nd postburn heights were similar. The vegetation took 6 months to reach one m tall, and the height remained constant until 12 months. Tables 3-5 through 3-12 provide absolute densities, frequencies, mean percent cover, and importance values of 10 herb species in 5C a nd 2M. In both sites, the Family Poaceae had the highest importance value, and it was represented by several grass species. Galactia elliottii, Solidago odora and Galactia mollis had the next three highest importance values, and Crotalaria rotundifolia and Woodwardia virginica were only recorded in 5C and 2M, respectively. The species above were counted during the preburn and/or postburn sampling periods, and the rest of the species in Tabl es 3-5 through 3-12 we re only found during the preburn sampling period. The postburn recovery of the most common herb species in 5C and 2M is illustrated in Figures 3-16 through 3-19. Poaceae had an upward trend for density, frequency, and importance value in 5C. Cover also had an increasing trend, but declined at nine months and increased again at 12 months. In 2M and for all variables, Poaceae leveled off from 3 to 9 months and increased 82

PAGE 83

one more time at 12 months. In general, Poaceae was the only taxon with preburn values and had higher values in 5C than in 2M. After burning, the behavior of the curves for G elliottii was alike for all variables in both sites. G elliottii increased at 3 months, d ecline at 6 or 9 months, and then increased again at 12 months. In general, G elliottii had higher densities, frequencies, cover, and importance values in 2M than in 5C, and it surpassed density c ontrol value (5D) in 5C and both control values in 2M. S. odora G mollis C rotundifolia and W virginica had the same pattern for all variables in both sites. Th ese species had low values for all variables until the 9 months and increased at 12 months, with th is increase higher in 5C than in 2M. Besides G elliottii, G mollis was the only herb species that reached density control value (5D) at 12 months postburn in 5C. Tables 3-5 through 3-12 display absolute densities, frequencies, mean percent cover, and importance values of 26 woody species in 5C and 2M These tables reveal that the seven species with high values for all variables in both sites were the following: Quercus myrtifolia Serenoa repens Quercus geminata Lyonia ferruginea, Lyonia lucida Quercus chapmanii and Vaccinium myrsinites. Ilex glabra and Gaylussacia nana were also common in 5C and 2M, respectively. Rare species were as follows: Quercus nigra Quercus sp., Ceratolia ericoides Osmanthus americanus Salix caroliniana Smilax sp, and Opuntia humifusa. Pinus clausa, P elliottii, and P palustris were present in 5C and the last two species in 2M, but few trees were recorded in quadrats. The rest of the species had moderate values in density, frequency, mean percent cover, and importance value in both sites. The postburn absolute density recovery of the most common woody species in 5C and 2M is shown by Figure 3-20. Q myrtifolia had the fastest recovery in both sites, with densities in 5C higher than in 2M. The rest of the species, S. repens Q geminata L ferruginea L lucida Q 83

PAGE 84

chapmanii and V myrsinites had similar recovery patterns in both sites, with densities slightly higher in 2M than in 5C. However, these sp ecies did not have de nsities higher than 15 individuals/m2 in both sites. Q myrtifolia recuperated over preburn and control sites values. The other species achieved preburn values at 3 months, with the exception of V myrsinites and they also had equal or higher absolute densities than control sites. Figure 3-21 reveals that the most common species had sim ilar patterns of absolute frequency after burning in 5C and 2M. All specie s increased frequency at 3 months and then leveled off until 12 months, with the exception of V myrsinites that continued increasing after 3 months. The only difference was that values were higher in 2M than in 5C. All species had frequencies equal or higher than preburn values at 3 months with the exception of L. ferruginea and L lucida in 5C, and these two species and Q myrtifolia and V myrsinites in 2M. However, these species achieved at least a control site fr equency at 3 months or later during the next 9 months. Absolute mean percent cover for the most co mmon species in 5C and 2M is presented in Figure 3-22. As can be seen, all species ha d a very similar pattern in both sites. Q myrtifolia and S. repens had the highest cover regi stered in both sites. Q myrtifolia increased cover at 3 months in both sites, leveled off until 9 months and gr ew until 12 months in 5C, and decreased at 6 months and increased until 12 months in 2M. S. repens had a sharp rise until 6 months, and then it tended to level out at 9 and 12 months in both sites. The other five species had an increased no higher than 5% at 3 months and remained relati vely constant until 12 months, with the exception of Q geminata in 5C and this species and L ferruginea in 2M. These last two species in their respective sites increased fr equency at 12 months with values higher than 5%. Q geminata Q chapmanii L ferruginea and V myrsinites, and G nana recovered preburn and/or control cover 84

PAGE 85

at 12 months in 2M, and Q myrtifolia Q geminata and V myrsinites in 5C (Tables 3-11 and 37). Figure 3-23 provides the postburn importance values of the most common species in 5C and 2M. In general, Q myrtifolia had importance values higher in 5C than in 2M. It increased in its importance value at 3 months and later levele d off until 12 months in 5C. It had its highest value at 3 and 6 months, and then decreased until 12 months in 2M. S. repens had similar postburn recovery in both sites. It had the highe st importance value at 11 and 12 days in 5C and 2M, respectively, due to their high resprout and frequency. Then, it decreased at 3 months, increased until 9 months, and declined one more time at 12 months. Q geminata also had a similar recovery pattern in both sites. The importance value in creased at 3 months, and then remained almost constant until 12 months. L ferruginea was relatively constant in 5C, and it increased at 6 months and le veled out after that in 2M. L lucida kept almost the same importance value after burning in both sites, being a littl e bit higher in 5C. Q chapmanii resprouted at a higher level in 2M than in 5C, and the importance value at 12 days was considerably higher in 2M than in 5C for this reason. However, Q chapmanii s importance values were very similar in both sites after that with the tendency of declining. The importance value of V myrsinites was relatively constant in 5C after burning, and remained steadily increasing in 2M, but at low importance values. Q myrtifolia S. repens Q geminata and Q chapmanii had importance values equal or higher than preburn and control sites values at 3 months in both sites. L ferruginea achieved control site values during the lapse of 12 months, but not the preburn value in 5C. However, it ha d a higher importance value at 6 months than preburn/control sites in 2M. L lucida did not reach preburn or cont rol importance values in 5C 85

PAGE 86

and 2M. V myrsinites recovered its preburn value at 6 months but not control site values in 5C. It obtained control site values between 6 a nd 12 months, but not the preburn value in 2M. The postburn recovery pattern described above belonged to all individuals recorded in quadrats. This included ramets, new saplings, and trees that survived fire. If I consider the calculation of the variables for ra mets only, absolute density for ramets were lower than absolute density for individuals in both si tes. Tables 3-13 and 3-14 show ra met absolute densities for herb and woody species in 5C and 2M and Figures 3-24 and 3-25 illustrate recovery patterns for the most common herb and woody species in 5C and 2M, respectively. As revealed by the graphs, recovery patterns for herb and woody species followed exactly the same pattern as for absolute density for individuals in both sites. Tree mortality in both quadrats and grids is presented by Table 3-15. Tree mortality was low in quadrats at both sites (7.5 %) because of the adaptation of Quercus spp. and Lyonia ferruginea to fire. These trees had 80-100% burned st em, but roots were alive. Therefore, all roots that remained alive resprouted after prescr ibed burning. Census on pine trees carried out in the grids before and after prescribed burning sugg ested that fire intensity was high enough to kill 91.6% of pine trees, including the fire adapted P clausa and the fire resistant P palustris Structural and Compositional Changes in Response to Prescribed Burning Figure 3-26 provides preburn and postburn species richness in treatment sites in comparison with control sites. Preburn species richness was lower in 5C than in 2M. Then, species richness decreased immediately following burning in 5C and was stable in 2M. After that, species richness in 5C increased (surpassin g its preburn level) and reached species richness in 5D, and obtained the highest sp ecies richness in 2M at 3 mont hs. Later, species richness in both sites fluctuated exactly with the same pattern, being higher in 5C than in 2M. In general, species richness in 5C was higher than its preb urn value and species ri chness in 2M after 3 86

PAGE 87

months. In addition, species richness in 5C r eached 5D control site value after burning, but species richness in 2M did not attain control site values. Figures 3-27 and 3-28 indicate that species diversity and evenness, respectively, were almost constant in both site s after prescribed bu rning. Simpsons indexe s preand postburn values were higher in 2M than in 5C. At both sites, Simpsons index originally decreased just after burning; then it was almost constant from several days to 9 months after burning, but increased at 12 months. Preburn species diversity was achieved in both sites at 12 months, and only 5D control value was surpassed in 2M. Shannon-Wieners indices had the same pattern as Simpsons indices in both sites. Preburn values were reached at three months in 5C and at 12 months in 2M. Again, only 5D control index wa s surpassed in 2M. Evenness followed exactly the same pattern as species diversity with mode rate values indicating modest predominance of the common species. Figures 3-27 and 3-28 sugges t small structural and compositional changes in treatment sites unti l 12 months postburn. The results of the Detrended Corresponden ce Analysis are shown in Table 3-16. In general, the r2 coefficient of determination was high in both treatment sites. This indicated that the analysis provided ordinations of good quality, and a high proportion of the variance was explained by the axes. However, absolute densities had higher r2 than relativized densities, and relativized cover had higher r2 than absolute cover in both sites. In addition, site 5C had lower r2 than 2M for densities and higher r2 than 2M for cover. Except for absolute cover in 2M, the first axis had higher r2 than the second and third axis in both sites. The scatter-plots illustrate DCA, but it is important to explain what dist ances represent in this ordination space. Understanding the distance among sampling times in the ordination space is key for the interpretation of structural and compositional change of the postburn community. Each point in 87

PAGE 88

the species/sampling time ordination space represents the sites position on the first two axes of the ordination at a given time prior to or afte r fire. According to Schmalzer and Hinkle (1992a), the distance between sampling time scores (points) is an index of similarity. Sampling times that occur together are similar. Also, the distance between preand postburn sampling times reveal vegetation change after fire and recovery. In othe r words, the lengths of vectors between preand postburn times in ordination space indicate the ve getation change during th e recovery process. Hence, the distance between sampling times (vector lengths) is an index of change and recovery (Schmalzer and Hinkle 1992a). These concepts are important to understand structural and compositional change in the scrubby flatwoods in CKSSR. Figures 3-29 through 3-32 display absolute and relative dens ities and mean percent cover scores of plant species during preburn and postb urn sampling times. Treatment sites 5C and 2M had structural and compositional changes through time, but they had in common the following changes: (a) a structural and co mpositional change after prescribed burning between preburn and three months postburn in both absolute and relativized density and cover, and (b) a structural and compositional changes between three and six mo nths shown by the absolute and relativized cover in 5C and by absolute cove r and relativized density in 2M Treatments sites 5C and 2M differed in changes between six and 12 months Site 5C had moderate structural and compositional changes in both absolute and rela tivized density and cover, but 2M had more compositional than structural changes in both vari ables. Now, are structural changes significant? The results of the MRPP are shown by Table 3-17. As can be seen in this table, the test statistic T was significant at the 0.05 level for both absolute densities and mean percent cover. T described the separation among groups. The more negative is T, the stronger the separation. Therefore, at least two sites were significantl y different regarding dens ity or cover, and the 88

PAGE 89

multiple comparison revealed them. Absolute dens ities in treatment sites (5C preburn, 5C-12 months postburn, 2M preburn, and 2M-12 months postburn) and control s ites (5A and 5D) were significantly different by using bot h Euclidean and Sorensen distan ces (Table 3-18). Therefore, prescribed fire did have a signi ficant effect on absolute densiti es on treatment sites by changing the structure of the community between preand 12 months postburn. Absolute cover was only significant between control site 5D and treatment sites 5C an d 2M at 12 months postburn by using Euclidean and Sorensen distances. Hence, pr escribed fire did not ha ve a significant effect on changing absolute mean percent cover be tween 5A and 5C/2M 12 months postburn and between preburn and 12 months postburn mean percen t cover of treatment sites. This is due to the fact that preburn and contro l mean percent cover values we re achieved by 12 months in the majority of the species (see Table 3-7 and 3-11). Flowering and Fruiting after Prescribe Burning Flowering and fruiting season we re not synchronous in CKSSR within and among species. V myrsinites started to have flowers in March an d fruits in April 2006 in some sites. G nana and S. repens started to have flowers in April, but G nana had fruits in May and S. repens in June 2006. Other species such as L ferruginea L lucida Ilex glabra and Brevaria racemosa started to flower in April-May a nd fruits in June-July 2006. Discussion True Control Sites This is the first site analysis carried out to determine the veracity of control sites. In the reviewed literature about the effect s of prescribed fire, control sites have been taken very loosely and without scientific rigor. In general, control sites have b een assigned at random or not at random as an assumption and without scientific validation. The assumption has been made and accepted by the majority of the scientific comm unity of fire ecology including editors of 89

PAGE 90

prestigious scientific journals. This study suggests th at a site analysis shou ld be the starting point for future research. Cluster analysis in combination with ordina tion and univariate techniques can be used to determine if treatment and control sites are eco logically similar. Species richness, species diversity, evenness, and similar ity coefficients produced contradictory results. The cluster analysis (by using three fusion tec hniques) in combination with disc riminant analysis (to plot the first two pairs of canonical vari ates) and the F-ratio test (wit h Duncans and Bonferronis multiple comparison procedures) helped to conclude that treatment and control sites were ecologically similar. The reason for using th ree fusion methods relied on the purpose of conducting a site analysis: to dete rmine if treatment and control sites were ecologically similar by carrying out a cluster analysis that really represented the structur e of the data. Two spaceconservative methods (Median and Average linka ge) and one space-disto rting method (Wards minimum-variance linkage) were ap plied to the data for this r eason. According to McGarigal et al. (2000), space-conservative methods are the best choice when the objective is to reveal the true structure of the data, whic h is usually the case in most eco logical research. Space-distorting methods do not truly represent the sp atial relationship of the data because they contract or dilate the space in the immediate vicinity of the groups. A comparison between space-conservative and space-distorting methods was done to be more objective in the decision. Since, Median/Average linkage and Wards minimum va riance linkage found one and two clusters, respectively, the Fratio test with the multiple comparison definitivel y helped to determine if the structure of the dataset fit to one cluste r or not. Having true control sites is critical in fire ecology research. A site analysis must be done a priori and not a posteriori. Researcher s in fire ecology need to be careful selecting control sites and test them before applying treatments to plots. Before and 90

PAGE 91

during this process, park managers advice and involvement are desirable due to their experience. The success of selecting good cont rol sites will rely on park managers and researchers judgment, preliminary sampling, and the statistical analysis conducted a priori. Fire Survival and Recovery Modes Resprouting was the main mechanism of survival and recovery with fire by the majority of the species in CKSSR. All spec ies that occurred before burning in CKSSR exhibit resprouting life histories, except P. clausa, P. elliottii, P. palustris and C. ericoides. Post-fire recovery through sprouting has been documented in scrub by flatwoods and sand pi ne scrub in Archbold Biological Station (Abrahamson 1984a, 1984b, Abrahamson and Abrahamson 1996a, 1996b), in oak-saw palmetto scrub in Kennedy Space Ce nter (Schmalzer 2003, Schmalzer and Hinkle 1992a, 1992b, Schmalzer et al. 2003) and in sand pine scrub in National Seashore in the Panhandle (Ruth et al 2007). However, there are some species that recovery through seeding and clonal spread. For instance, S. repens has been reported as resprouter and seeder, Quercus spp. Lyonia spp., and Smilax auriculata have been cited as resprouter s and clonal spreaders (Menges and Kohfeldt 1995, Menges and Hawkes 1998 see Table 3-1). Resprouting from dormant buds, rhizome tips, root crowns, and protected meristems can account for a substantial proportion of postfire recruitment (Lyon a nd Stickney 1976). It is clear that almost all species (except C ericoides ) tolerated fire in CK SSR, and soil was a good insulator that protected underground roots and meristematic tissues as well. Tissues deeper than 5 cm rarely experience significan t increase in temper ature (Whelan 1995). But, did soil protect the seed bank already established during the spring 2005? Seeds were probably not buried deep enough in the soil by the time of prescribed burning in CKSSR. Prescribed burning took place at th e end of the reproductive and growing season. This was the right moment biologically and envi ronmentally because the ma jority of the species 91

PAGE 92

already produced seeds and weathe r characteristics were appropriat e to assure a catastrophic fire. However, fire intensity was high enough to top-kill all aboveground vegetation in treatment sites and the seed bank. No seedlings were found until 6 months postburn, and these seedlings belonged to P clausa Seeds experienced temperatures higher than 1000 C in treatment sites, and most likely they were lying on the leaf litter and thus they were consumed. Seeds must be protected from direct he at to survive fire. Speed of Recovery Process Postfire changes in CKSSR were described at a very short time interval in comparison with the literature. The results obtai ned in CKSSR correspond to only one year postfire. Research carried out in oak-saw palmetto scrub (stands 2, 4, 8, and 24 years si nce burning) at Kennedy Space Center/Merritt Island Na tional Wildlife Refuge (KSC ; Schmalzer and Hinkle 1992a, 1992b, Schmalzer 2003), in long-unburned scru bby flatwoods (>35 years) at Archbold Biological Station (ABS; Abrahamson and Abrahamson 1996a), and long-unburned sand pine scrub (57 years) at Ocala Nationa l Forest (ONF; Greenberg 2003) ha ve lasted at least 7 years. Therefore, the comparison between CKSSR and th ese studies was made by mainly taking into consideration the first year postburn (see Tabl e 3-19). Even though the results obtained from CKSSR were short term, they are important beca use this is the first study conducted on coastal scrub in the west coast of the Florida Peninsula. Another factor to take into consideration is how variables were measured among studies. Mean height, species richness, and mean percen t cover of bareground and species presented in Table 3-19 were the only variables and the most common species shared among these three studies. Mean percent cover in CKSSR was quantified for herb sp ecies and for shrub species in the stratum 1.5 m tall and the results are comparable with ABS and ONF because these studies 92

PAGE 93

did not establish any stratum. At KSC, Schmalzer (2003) measured cover at two stratum: <0.5 m and >0.5 m. Therefore, the data from these two strata were summed to make them comparable with the other three studies. Unfortunately, other studies carried out in ABS, KSC, and Naval Live Oaks/Gulf Island National Seashore (NLO; Ruth et al. 2007) could not be included in the comparison because they did not quantify the sa me variables through time. However, important results from these studies are cited to highlight particular points in this discussion. Bareground recovery is slow through time in long-unburned scrubs. Since, long-unburned scrubs have zero or very low mean percent of bareground cover, fire considerably increases bareground cover during the first 6 to 12 months postburn, then it tends to decrease as Quercus spp., Lyonia spp., and S. repens increase density and cover through time. In KSC, bareground increased significantly after burning until 6 months (22.9%), but declined rapidly to 0.7% at 36 months. In ONF, bareground increased significantly after burning, reachi ng its maximum at 16 months postburn (25%), and by 101 months postburn (8%) still was not near preburn level (0.00%). Even though the study in CKSSR was at s hort term, the recovery of the vegetation was so fast that bareground had 3.8% and 6.8% at 6 and 12 months postbur n, respectively. Probably, bareground in CKSSR will recuperate to the preb urn level faster than in the other two study areas. The recovery process for litter and debris has not been well documented in the literature. Greenberg (2003) reported these values in a sand pine stand in ONF. Litter was recorded as depth of litter layer. It decreas ed from preburn level of 6.5 cm to 2.0 cm and 2.5 cm at 5 and 16 months postburn, respectively, and then slowly d ecreased to 1% at 101 months postburn. Debris had a preburn value of 0.2%, which was constant until 16 months, then it increased to 8.4% at 93

PAGE 94

101 months. In CKSSR, litter and debr is recovered faster than in ONF, achieving litter more than 70% and debris more than 30% of the prebur n level in treatment sites in 12 months. Vegetation height was one of the variables w ith the slowest recovery process. In KSC, mean pre-burn height was 1.08 m and reach ed 0.32 m and 0.50 m at 6 and 12 months, respectively. Height growth continued throughout the postburn period reach ing preburn value at 85 months. In ONF, the mean height of the vege tation was determined by measuring the height of Q myrtifolia and Sabal etonia Table 3-19 only shows the mean height of S. etonia because it was the highest data recorded at 5 and 16 months. By 5 months postburn, S. etonia reached its preburn value (1 m), and Q myrtifolia needed almost 64 months to achieve its preburn mean value of 1.25 m. In CKSSR, the mean height of the vegetation recovered slow er than in the other two study areas. It was 31% of the preburn valu e at 12 months, while in KSC was 46% of the preburn value. The recovery process in KSC is described below. Aristida stricta was the most common herb species and recovered preburn value in si x months. Other herb sp ecies also found in CKSSR such as Carphephorus spp. and Galactia elliottii were not recorded during the preburn sample, and they were censused with low c over during the postburn period. Regarding woody species in general, Q myrtifolia S. repens and L lucida were the dominant species (Table 319), with Q geminata and Q chapmanii also relatively common. At the <0.5 m stratum, Q myrtifolia and L lucida increased cover more than 5 times the preburn levels after burning. Then, cover decreased at four years for Q myrtifolia and two years for L lucida, and reached preburn level after seven and five years, respectively. At the >0.5 m stratum, Q myrtifolia and L lucida significantly decreased cover after one year pos tburn and needed five years to recover preburn values. S. repens did not have a high increase in cove r in the <0.5 m stratum after burning, 94

PAGE 95

and it had a low cover relati vely constant through time. S. repens reestablishes cover faster than woody shrubs at >0.5 m stratum. It increased cover right after prescribed burning and reached the preburn level between one and one and a half year. Q geminata and Q chapmanii had a low increase in cover after burning at the <0.05 m stratum. Then, they gradually reduced cover after three years. Both species decr eased cover after burning at the >0.5 m stratum and obtained preburn levels after 5 years postburn. L. lucida recovered cover between 4 and 5 years. Schmalzer and Hinkle reported shifts in dominance after fire due to differences in recovery rates in shrub species and S. repens. V. myrsinites had a low increase afte r burning that persisted relatively constant through time at the <0.05 m stratum, and it star ted to appear after three years at the >0.05 m stratum, and it kept low cover after that. In ABS, Abrahamson and Abrahamson did not show data of cover for herb species during five years of postburn period. They only presented the average percen t cover for five 200-m transects of seven species not found in CKSSR. However, they reported that long-unburned scrubby flatwoods contained fewer herb species during the postburn period than recently burned scrubby flatwoods. A possibl e explanation suggested by the author s was that fire had the effect of creating or enlarging gaps making the persisten ce of herbs (gap specialists) sensitive to time since fire. Long fire-free periods may allow shrubs to clonally spread into gaps and make the gap-specialist herbs disappear through time. S. repens Q chapmanii and Q geminata were dominant species. Cover of Q chapmanii Q geminata Q minima Lyonia fruticosa L lucida, V myrsinites and M cerifera returned to or exceeded preburn levels within three years following fire. S. repens reached preburn level in two years, did not maintain preburn dominance, but cover was higher than 15% through time. 95

PAGE 96

In ONF, Rynchospora megalocarpa was the dominant herb sp ecies and recuperated its preburn value (1.84%) in 16 months. Species common to CKSSR were: Clitoria mariana (0.010.80%), G elliottii (0.02%), and Zamia pumila (0.02-0.53%), and they occurred with low mean percent cover. Of 28 herb species, 19 were absent from transects prior to the burn and occurred on transects during postburn samples with low cover (0.01-1.53%). In general, scrub woody species composition and cover were simila r to preburn values after 16 months. Q myrtifolia was the most dominant shrub before fire, recovered ra pidly, and attained 67% of preburn cover level at 16 months postburn. Q geminata regained 84% of its prebur n level in 16 months. Mean percent cover of Q chapmanii was 0.98% preburn, and it almost achieved this value at 16 months (0.93%). S. repens recovered 75% of its preburn level and L ferruginea surpassed its preburn level in 16 months. In general, recovery rates among species did not result in long-term shifts in species dominance because Sabal etonia recovered its preburn level faster than Q myrtifolia but Q myrtifolia continued to be the dominant species. Comparing the recovery pattern in CKSSR with KSC, ABS, and ONF, we found a different story for herb and woody species. Galactia elliottii was the most common herb species in CKSSR, but with limited comparison because it was only recorded by few sampling periods in KSC and ONF. Carphephorus corymbosus was only recorded in control site 5D at CKSSR, and it had low cover in almost all samples in KSC. Clitoria mariana and Zamia pumila were sampled only in 5A and 5D, respectively, in CKSSR, and they also had low cover in almost all samples in ONF. At CKSSR, G elliottii and G moilis were the only herb species sampled on control sites and during the postburn period, while Crotalaria rotundifolia Solidago odora and Woodwardia virginica were registered only during the postburn period. Therefore, few species are present with low cover before and after burning, and othe r species colonize gaps available after burning 96

PAGE 97

for a temporal use (with low cover) and until th ey are replaced by the dominant and growing woody vegetation. In general, the scrub woody vegetation in KSC, ABS, ONF, and CKSSR returned to preburn conditions rapidly after a high-intensity prescribed burn. This aspect has been reported by other studies in scrub vegetation in Flor ida (Abrahamson 1984a, 1984b; Menges et. al. 1993; Schmalzer et al. 2003; Ruth et al. 2007). As can be seen in Table 3-19, CKSSR and ONF had the fastest recovery. However, I did not investigate c over after 12 months in CKSSR. The recovery time for the other studies suggest that there is variation depending on the type of scrub. Regarding the dominant species, Q myrtifolia and S. repens were the most common species in at least three study areas. There was a shift in dominance after fire for shrubs and palmetto in KSC, for S. repens in ABS, and for Q chapmanii and L ferruginea in CKSSR (see Figure 3-22). Although P clausa is a species adapted to fire, its seed lings take some time to appear after fire. In ABS, P clausa appearance is delayed three y ears postburn (Abrahamson 1984b). In ONF, P. clausa seedlings were established af ter 5 months postburn (Greenberg 2003). In NLO (Ruth et al. 2007), P. clausa seedlings were absent eight months to two years after burning from sand pine scrub. In CKSSR, seedlings appeared six months postburn. Hence, P. clausa seedlings appear after several months to se veral years in the scrub and very little is known about seedlings survival and establishment. Even though several vegetation variables retu rned to preburn conditions in CKSSR during an interval of 12 months, this time was still t oo short to predict that treatment sites would be restored to scrubby flatwoods without a histor y of fire suppression. According to Abrahamson and Abrahamson (1996a, 1996b) and Baker (1 992, 1994), prescribed burning does not necessarily reestablish the pre burn conditions in all landscapes after several years of fire 97

PAGE 98

suppression. This is an aspect that needs to be determined in Cedar Key with continued monitoring of the study sites. Schmalzer and B oyle (1998), Schmalzer and Adrian (2001), and Schmalzer et al. (2003) recommended a combina tion of mechanical treatment and prescribed burning in restoring long-unburne d scrub vegetation. Mechanical cutting should be used only one time to reestablish shrub ve getation structure, which can be maintained with periodic prescribed burning after that. All mechanical treatments result in some loss of S. repens cover, and this loss persists (Schma lzer et al. 2003). So, mechanical treatment must be applied carefully. Whether the reintroduction of fire to long-unbur ned scrub might produce or not the desired effect of returning the associati on to a state similar to scrubs w ithout fire suppression will rely on fire intensity and the season of burning. The reintr oduction of frequent and low-intensity fires to sand pine scrub may shift scrub to sandhill (Myers 1985). In contrast, a single, high-intensity fire in sand pine scrub may make possible the pers everance of the scrub st and (Myers 1985, Menges et al. 1993, Menges and Hawkes 1998). Instead, a si ngle-low intensity fire might facilitate the shift of scrub toward xeric hammock. This situ ation would occur if fire increases the abundance of the sprouter species, and consequently repress the regeneration of ob ligate seeders such as sand pine and herb species. According to Abrahamson & Abrahamson (1996b), a single fire may not be effective at restoring l ong-unburned scrubby flatwoods to st ates characteristic of more recently burned stands. Several fires may be needed before scrubby flatwoods are returned to communities similar to those without fire suppression. Community Shift in Response to Prescribed Burning The scrubby flatwoods community in CKSSR ha d structural and commu nity changes after prescribed burning, and they were reported fo r an interval of 12 months. As previously mentioned, studies conducted in KSC, ABS, and ONF have datasets for at least 7 years, which 98

PAGE 99

constraint the comparison with CKSSR. This study presents a comparison between CKSSR and these three studies regarding species richness, species dive rsity (Shannon-Wieners index), evenness ( H' / ln S ), and the results of the Detrended Correspondence Analysis mainly restricted to 12 months postburn. Mean species richness in KSC in creased in the <0.5 m stratum, but declined in the >0.5 m stratum, during the first 12 months. Considering bo th strata together, there was an increase in the species richness 12 months postbur n. Species diversity and evenne ss were not measured in this study. The DCA carried out only with mean percen t cover indicated that the composition of the scrub varied along a gradient clos ely related to the depth of the water table with oak dominating drier sites and S. repens and I. glabra in wetter places. The spec ies ordination located oak species to the left and S. repens, I. glabra and M. cerifera to the right of the axis. This pattern was found in all oak-dominated transects and sa w palmetto-dominated transects. The other transects with a mixed oak-saw palmetto compos ition before burning located sample units in the middle of the ordination axis. Both, oak-dominated and saw palmetto-dominated transects in the ordination space, returned to preburn locations afte r three years. Vector lengths were greater for the mixed oak-saw palmetto tran sects than those dominated by S. repens indicating a greater degree of change in the mixed oak-saw palmetto transects. In one long-unburned (>35 yr) and two recentl y burned (<20 yr) scrubby flatwoods in ABS, species richness was constant in long-unburn ed and decreased in recently burned during the first 12 months. Later, species richness increase d in all stands relative to preburn levels until 36 months, and increases were most pronounced at recently burned scrubby flatwoods. Also, species richness of long-unburned sc rubby flatwoods was reduced relative to preburn levels after 36 months. Species diversity increased and re ached the average preburn index in the long99

PAGE 100

unburned stand at 12 months postburn. However, species diversity increa sed just after burning and then decreased in recently burned stands in the first 12 months. But the index at 12 months was higher than preburn index. After the first year species diversity increased and remained high for the recently burned stands, which had a hi gher index through time th an the long-unburned stand. Species diversity in the long-unburned stand decreased at 24 months and 48 months postburn and then linearly increased through time after that. Overall, ev enness was reduced after fire. Long-unburned stands achieved preburn ev enness at approximately 72 months postburn; however, recently burned stands did not. The DCA was carried out to visualize the multivariate changes in species dominance (percentage of co ver based on crown interc epts) in the preburn and postfire samples. The analyses, based on abso lute and relativized dominance, showed that scrubby flatwoods were very stable following fire. Changes related to composition were more noticeable than structural changes. In general, the effect of fire on these stands caused slight changes in stand structure and some degree of shifts in stand compos ition regardless of time since fire. These results are similar to the conclusions found from other studies in Florida (Abrahamson 1984a, 1984b; Abrahamson et al. 1984; Abrahamson & Hartnett 1990; Schmalzer & Hinkle 1992a). In a sand pine scrub stand in ONF, herbaceous species richness increased within 5 months postburn, peaked at 16 months, and declined by 40 months postburn. Woody species richness decreased immediately after fire (five weeks), then increased until 28 months and remained constant after that. In general, species richness increased during the first 12 months. The increase followed by the gradual decline of herbaceous sp ecies richness appeared to be related with gradual increases in shrub c over and decreases in bare gr ound availability. Carrington and Keeley (1999) suggested that the low postburn seedling recruitment in sand pine scrub is 100

PAGE 101

probably due to the elimination of suitable mi crosites by resprouting sh rubs. Greenberg (2003) did not quantify species di versity, evenness, and di d not carry out a DCA. The results found in CKSSR are similar to th e results found in the other studies. KSC, ONF, and CKSSR had an increased of species richness, but ABS did not during the first 12 months. This increase in species richness is expected due to the disturbance caused by fire, the amount of species resprouting a nd clonal spread, and new herb species that are temporarily colonizing gaps. Species divers ity in the long-unburned scru bby flatwoods in ABS and in CKSSR achieved preburn levels in 12 months postburn. Evenness regained preburn levels in ABS and CKSSR, but was faster in CKSSR. Th e information obtained from species richness, species diversity, and evenness sugg ests little structural and com positional changes at short term in CKSSR and at long term in ABS. In contrast, DCA was able to reveal results not detected by these indices. The DCA revealed both structural and compos itional changes at short term in CKSSR and more compositional than structural changes at long term in ABS. In CKSSR, there were both structural and compositional changes with similar magnitude during the first 3 months according to the lengths of the vectors (Figure 3-29 thr ough 3-32). These changes were expected because many species were resprouting simultaneously. Between 3 and 12 months, there were also structural and compositional changes because a lmost all species already resprouted (little variation in species richness and diversity), their frequencies we re almost constant thorough time (Figure 3-21), and mainly the density and cover of Q. myrtifolia and the cover of S. repens, Q. geminata and L. ferruginea increased through time (Figures 3-20 and 3-22). In a low-intensity fire on long-unburned sand pine scrub in ABS, Abrahamson and Abrahamson (1996b) reported that the largest amount of stru ctural and compositional change occurred immediately after 101

PAGE 102

prescribed fire, just between the preburn and 1-year postburn censuses. In contrast, changes between the 1-year and 2-year censuses after fire were primarily compositional; little structural change was measured. The DCA carried out in CKSSR on mean percen t cover also showed that the composition of the scrub varied along a gr adient associated with the water table. Figures 3-33 and 3-34 presents the ordination found in KSC and in CK SSR with preburn samples of mean percent cover, respectively. Comparing both figures, we can see that oaks are in the left side of the ordination (drier places) and S. repens and I. glabra are in the right side of the ordination (wetter places). In Cedar Key, M cerifera L lucida, and Q geminata are almost in the middle of the gradient, but not in KSC. However, the position of L lucida is not exactly to the left and the location of M cerifera is not exactly to the right in the ordination space in KSC. Age at First Flowering after Prescribed Fire Age at first flowering might vary among sp ecies within a single community, between populations of a single species, and within one population. In additi on, year of first flowering for several species might vary among sites in the same study area (Whelan 1995). In CKSSR, flowering was not synchronous within and am ong species and among st ands without burning. Flowering started in March 2006 after the prescribed burnings in April and May 2005. Probably, sprouting species delayed flower ing after prescribed fire be cause the energy produced through photosynthesis was devoted to vegetative growth. In sand pine scrub in ABS (Abrahamson and Abrahamson 1996b), S. etonia did not flower until spring the y ear following a February burn. In contrast, in ONF (Greenberg 2003), S. etonia flowered within 5 weeks and fruited within 5 months. The factors that affect pos t-fire sprouting are likely to affect post-fire flowering when the number of flowers or inflorescences per plant are determined by the number of active shoots. Each shoot sprouting after fire can produce a terminal inflorescence (Whelan 1995). Therefore, 102

PAGE 103

the season of burn, fire intensity, or both may be important factors that affect flowering in sprouting species. Unfortunately, the effects of th ese factors on flowering of sprouting species have rarely been quantified. 103

PAGE 104

Table 3-1. List of plant species recorded in qua drats in treatment and c ontrol sites in Cedar Key Scrub State Reserve. Species Common Name Family Recovery Mode Herbaceous Agalinis filifolia Seminole False Foxglove Orobanchaceae Resprouter (1) Asclepias sp Milkweed Apocynaceae Resprouter (2) Carphephorus corymbosus Florida Paintbrush Asteraceae Resprouter (2) Clitoria mariana Atlantic Pigeonwings Fabaceae Resprouter (3) Crotalaria rotundifolia Rabbitbells Fabaceae Resprouter Galactia elliottii Elliottis Milkpea Fabaceae Resprouter (2) Galactia mollis Soft Milkpea Fabaceae Resprouter Solidago odora Chapman's Golden Rod Asteraceae Resprouter (*) (2) Woodwardia virginica Virginia Chain Fern Blechnaceae Resprouter (1) Zamia pumila Florida Arrowroot Zamiaceae Resprouter (3) Woody Pinus clausa Sand Pine Pinaceae Obligate seeder (2) Pinus elliottii Slash Pine Pinaceae Obligate seeder ( ) (2) Pinus palustris Long-leaf Pine Pinaceae Obligate seeder ( ) Quercus geminata Sand Live Oak Fagaceae Resprouter ( ) (2) Quercus myrtifolia Myrtle Oak Fagaceae Resprouter ( ) (2) Quercus chapmanii Chapman Oak Fagaceae Resprouter ( ) (2) Quercus minina Runner Oak Fagaceae Resprouter ( ) (2) Quercus nigra Water Oak Fagaceae Resprouter Quercus sp Fagaceae Resprouter Brevaria racemosa Tar Flower Ericaceae Resprouter (2) Ceratolia ericoides False Rosemary Empetraceae Obligate seeder (2) Ilex glabra Gallberry Aquifoliaceae Resprouter Lyonia ferruginea Rusty Lyonia Ericaceae Resprouter (2) Lyonia fruticosa Stagger-bush Ericaceae Resprouter ( ) (2) Lyonia lucida Fetterbush Ericaceae Resprouter ( ) (2) Myrica cerifera Wax Myrtle Myricaceae Resprouter ( ) (2) Gaylussacia dumosa Dwarfhuckleberry Ericaceae Resprouter ( ) (2) Gaylussacia nana Dangleberry Ericaceae Resprouter (1) Licania michauxii Gopher Apple Chrysobalanaceae Resprouter ( ) (2) Osmanthus americanus Wild Olive Oleaceae Resprouter Rhus copallinum Winged Sumae Anacardiaceae Resprouter (2) Salix caroliniana Carolina Willow Salicaceae Resprouter Serenoa repens Saw Palmetto Arecaceae Resprouter-seeder Smilax auriculata Catbrier Smilacaceae Resprouter (*) (2) Smilax sp Smilacaceae Resprouter Vaccinium myrsinites Blueberry Ericaceae Resprouter ( ) (2) Cladonia evansii Deer Moss Cladoniaceae Seeder (2) Opuntia humifusa Devil's tongue Cactaceae Resprouter Codes: (*) Resprouter, clonal spreader, and seeder. ( ) Resprouter and clonal spreaders. () Obligate seeder and survivor. (1) United States Department of Agriculture website. (2) Menge & Kohfeldt (1995). (3) Greenberg (2003). 104

PAGE 105

Table 3-2. Species richness, Simpsons i ndex Shannon-Wieners index and ShannonWieners evenness for preburn conditions in control (5A & 5D) and treatment (5C & 2M) sites in Cedar Ke y Scrub State Reserve. Site Richness Simpson Shannon-Wiener Evenness 5A 26 10.64 2.54 0.6981 5D 24 7.06 2.28 0.6271 5C 17 5.03 1.99 0.5483 2M 20 8.41 2.28 0.6265 105

PAGE 106

Table 3-3. Multiple mean comparison (Duncans test) between clusters created by using Euclidean distances and Wards minimu m variance linkage fusion. Abundance and mean percent cover were standa rdized. Significan t level = 0.05. Cluster 1 Cluster 2 Variable n Mean n Mean F value P-value Galactia elliottii (abundance) 34 -0.2873 47 -0.2635 0.08 0.7742 Galactia elliottii (cover) 34 -0.2375 47 -0.2089 0.13 0.7234 Quercus geminata (abundance) 34 -0.1107 47 -0.1272 0.01 0.9163 Quercus geminata (cover) 34 -0.1750 47 -0.2104 0.14 0.7130 Quercus myrtifolia (abundance) 34 -0.6000 47 0.6634 41.47 <.0001 Quercus myrtifolia (cover) 34 -0.5288 47 0.7118 38.54 <.0001 Quercus chapmanii (abundance) 34 -0.2150 47 -0.1768 0.08 0.7792 Quercus chapmanii (cover) 34 -0.1741 47 -0.2252 0.42 0.5173 Quercus minima (abundance) 34 -0.0354 47 -0.1105 0.25 0.6219 Quercus minima (cover) 34 -0.2316 47 -0.1840 0.72 0.3984 Quercus sp (abundance) 34 -0.1589 47 -0.1404 0.03 0.8578 Brevaria racemosa (abundance) 34 -0.0374 47 -0.1302 2.86 0.0945 Ceratolia ericoides (cover) 34 -0.2030 47 -0.1770 0.72 0.3984 Ilex glabra (abundance) 34 -0.0432 47 -0.2026 5.19 0.0254 Lyonia ferruginea (abundance) 34 -0.2414 47 -0.0951 1.13 0.2919 Lyonia ferruginea (cover) 34 -0.4303 47 -0.1164 6.85 0.0106 Lyonia fruticosa (abundance) 34 -0.1792 47 -0.2084 1.45 0.2315 Lyonia lucida (abundance) 34 0.1317 47 -0.1755 3.74 0.0567 Lyonia lucida (cover) 34 0.2238 47 -0.3472 15.7 0.0002 Myrica cerifera (abundance) 34 -0.2175 47 -0.1783 0.11 0.7398 Myrica cerifera (cover) 34 -0.1522 47 -0.1760 0.26 0.6135 Gaylussacia dumosa (abundance) 34 -0.1296 47 -0.1108 0.07 0.7939 Gaylussacia nana (abundance) 34 -0.1939 47 -0.2177 0.18 0.6720 Licania michauxii (abundance) 34 -0.0984 47 -0.0978 0.00 0.9909 Serenoa repens (abundance) 34 0.6573 47 -0.3788 35.23 <.0001 Serenoa repens (cover) 34 1.0759 47 -0.4110 72.16 <.0001 Smilax auriculata (abundance) 34 -0.1325 47 -0.1809 1.39 0.2421 Vaccinium myrsinites (abundance) 34 -0.3902 47 -0.1371 5.21 0.0251 Vaccinium myrsinites (cover) 34 -0.2928 47 -0.1172 2.52 0.1165 106

PAGE 107

Table 3-4. T test, ANOVA test, and Kruskal-Wall is test for comparing means and medians among treatment and control sites under preb urn conditions in Cedar Key Scrub State Reserve. Data for T test and ANOVA we re standardized. Test for ANOVA and Kruskal-Wallis is F test and Chi-squared test, respectively. Significant level = 0.05. Test Species 5A 5D 5C pre 2M pre Test DF P-value T Ilex glabra (*) -0.2907 0.1642 -1.1570 16 0.2642 ANOVA Quercus myrtifolia (*) 0.0721 0.0655 -0.1537 -0.1519 0.7000 3 0.5523 Quercus myrtifolia (%) -0.0683 0.1724 0.0835 0.0867 0.3300 3 0.8014 Lyonia ferruginea (%) 0.2617 0.2827 0.4946 0.3457 0.1700 3 0.9169 Lyonia lucida (%) 0.0167 0.1522 -0.0267 -0.6480 1.7500 3 0.1654 Serenoa repens (%) -0.2689 0.1440 0.0508 0.0926 0.8600 3 0.4657K-W Serenoa repens (*) 4.0000 3.0000 4.0000 3.0000 7.9563 3 0.0469 Vaccinium myrsinites (*) 5.0000 3.0000 2.0000 10.0000 18.7388 3 0.0003 Codes: (*) mean abundance. (%) mean percent cover. 107

PAGE 108

Table 3-5. Preand postburn abso lute densities for all species in 5C in Cedar Key Scrub State Reserve. Absolute densities for control si tes 5A and 5D are also shown. d = days. M = months. Control Time after burning Species 5A 5D Pre 11 d 3 M 6 M 9 M 12 M Herbaceous Poaceae 0.37 0.14 0.03 0.09 0.61 1.15 0.98 1.90 Solidago odora 0.00 0.00 0.00 0.00 0.12 0.11 0.06 1.82 Galactia elliottii 2.46 1.09 0.00 0.00 2.66 1.26 0.00 1.35 Galactia mollis 0.00 0.18 0.00 0.00 0.04 0.05 0.00 1.22 Crotalaria rotundifolia 0.00 0.00 0.00 0.00 0.08 0.07 0.07 0.03 Agalinis filifolia 0.00 0.15 0.00 0.00 0.00 0.00 0.00 0.00 Asclepias sp 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Carphephorus corymbosus 0.00 0.19 0.00 0.00 0.00 0.00 0.00 0.00 Clitoria mariana 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Woodwardia virginica 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Zamia pumila 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 Woody Quercus myrtifolia 10.91 12.17 12.53 1.46 45.58 43.46 47.76 42.11 Lyonia ferruginea 2.09 5.06 3.87 0.58 8.32 10.39 11.22 11.60 Quercus geminata 2.90 2.24 2.20 0.38 13.20 12.68 13.24 11.58 Lyonia lucida 7.03 10.70 4.49 0.55 8.23 8.83 8.87 8.88 Quercus chapmanii 3.01 2.01 2.79 0.29 9.99 8.58 7.33 6.43 Serenoa repens 1.14 1.01 1.41 1.80 3.31 4.32 4.34 4.94 Ilex glabra 2.98 0.55 0.13 0.06 4.63 4.91 4.74 3.55 Licania michauxii 0.76 0.20 0.00 0.00 2.65 2.43 0.90 2.34 Lyonia fruticosa 1.88 0.00 0.25 0.15 1.24 1.85 2.22 1.71 Vaccinium myrsinites 5.08 2.04 0.51 0.00 0.97 1.43 1.44 1.69 Gaylussacia dumosa 5.07 1.37 0.00 0.00 1.69 1.41 1.46 1.40 Gaylussacia nana 0.31 5.30 0.03 0.00 1.32 1.19 0.28 1.13 Brevaria racemosa 0.57 0.05 0.00 0.03 1.92 2.00 1.65 1.06 Myrica cerifera 1.04 0.47 0.85 0.01 0.43 0.77 0.69 0.50 Rhus copallinum 0.00 0.00 0.00 0.00 0.25 0.28 0.00 0.31 Pinus clausa 0.00 0.00 0.02 0.01 0.01 0.03 0.08 0.25 Smilax auriculata 0.12 0.08 0.00 0.00 0.03 0.11 0.17 0.12 Quercus minina 2.50 1.70 1.21 0.00 0.30 0.38 0.35 0.09 Pinus palustris 0.05 0.00 0.00 0.00 0.00 0.00 0.00 0.08 Pinus elliottii 0.04 0.00 0.04 0.00 0.00 0.01 0.02 0.05 Quercus nigra 0.00 0.00 0.08 0.00 0.00 0.00 0.00 0.00 Quercus sp 0.07 0.17 0.11 0.00 0.02 0.00 0.00 0.00 Ceratolia ericoides 0.03 0.02 0.00 0.00 0.00 0.00 0.00 0.00 Osmanthus americanus 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 Salix caroliniana 0.00 0.00 0.00 0.00 0.03 0.02 0.00 0.00 Smilax spp 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Cladonia evansii 3.51 1.80 2.26 0.00 0.00 0.00 0.00 0.00 Opuntia humifusa 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 108

PAGE 109

Table 3-6. Preand postburn abso lute frequencies for all species in 5C in Cedar Key Scrub State Reserve. Absolute frequencies for control sites 5A and 5D are also shown. d = days. M = months. Control Time after burning Species 5A 5D Pre 11 d 3 M 6 M 9 M 12 M Herbaceous Poaceae 12 6 4 3 15 17 18 25 Galactia elliottii 31 29 0 0 35 28 1 20 Galactia mollis 0 1 0 0 1 2 0 11 Solidago odora 0 0 0 0 5 5 4 8 Crotalaria rotundifolia 0 0 0 0 1 1 1 1 Agalinis filifolia 0 1 0 0 0 0 0 0 Asclepias sp 1 0 0 0 0 0 0 0 Carphephorus corymbosus 0 1 0 0 0 0 0 0 Clitoria mariana 2 0 0 0 0 0 0 0 Woodwardia virginica 0 0 0 0 0 0 0 0 Zamia pumila 0 1 0 0 0 0 0 0 Woody Quercus myrtifolia 34 44 35 29 36 35 35 35 Quercus geminata 29 34 30 14 34 33 34 34 Serenoa repens 27 29 33 31 35 35 35 34 Quercus chapmanii 32 35 27 12 29 30 31 31 Lyonia ferruginea 19 34 27 18 20 24 24 26 Lyonia lucida 28 38 31 13 28 25 24 24 Vaccinium myrsinites 34 27 14 0 14 20 19 23 Quercus minina 26 40 15 0 9 11 18 15 Pinus clausa 0 0 2 1 1 2 5 14 Myrica cerifera 14 12 15 2 8 14 13 13 Licania michauxii 5 6 0 0 13 15 5 13 Ilex glabra 13 4 3 1 6 11 9 9 Rhus copallinum 0 0 0 0 11 11 0 9 Smilax auriculata 5 4 0 0 1 7 8 7 Gaylussacia nana 1 29 2 0 8 7 5 6 Pinus elliottii 2 0 3 0 0 1 2 4 Lyonia fruticosa 13 0 4 3 3 3 3 4 Gaylussacia dumosa 24 22 0 0 6 3 3 4 Brevaria racemosa 8 2 0 2 2 3 3 3 Pinus palustris 3 0 0 0 0 0 0 2 Quercus nigra 0 0 1 0 0 0 0 0 Quercus sp 3 12 3 0 1 0 0 0 Ceratolia ericoides 3 6 0 0 0 0 0 0 Osmanthus americanus 0 1 0 0 0 0 0 0 Salix caroliniana 0 0 0 0 1 1 0 0 Smilax spp 1 0 0 0 0 0 0 0 Cladonia evansii 8 12 14 0 0 0 0 0 Opuntia humifusa 1 0 0 0 0 0 0 0 109

PAGE 110

Table 3-7. Preand postburn abso lute mean % cover of herb a nd woody species in 5C in Cedar Key Scrub State Reserve. Absolute mean pe rcent cover for control sites 5A and 5D are also shown. d = days. M = Months. Control Time after burning Species 5A 5D Pre 11 d 3 M 6 M 9 M 12 M Herbaceous Poaceae 0.47 0.00 0.14 0.04 0.46 2.41 1.92 2.07 Galactia elliottii 2.06 1.76 0.00 0.00 1.83 0.13 0.00 0.51 Solidago odora 0.00 0.00 0.00 0.00 0.00 0.50 0.00 0.47 Galactia mollis 0.00 0.00 0.00 0.00 0.04 0.00 0.00 0.11 Agalinis filifolia 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Asclepias sp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Carphephorus corymbosus 0.00 0.26 0.00 0.00 0.00 0.00 0.00 0.00 Clitoria mariana 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Crotalaria rotundifolia 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Woodwardia virginica 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Zamia pumila 0.00 0.09 0.00 0.00 0.00 0.00 0.00 0.00 Woody Quercus myrtifolia 20.07 27.36 28.43 0.00 15.21 16.83 16.07 24.95 Serenoa repens 23.95 24.39 28.66 0.00 13.26 19.32 22.48 23.00 Quercus geminata 6.75 0.64 2.71 0.00 4.26 4.51 3.76 7.64 Lyonia lucida 8.94 18.00 4.35 0.00 2.05 2.35 2.82 3.58 Lyonia ferruginea 4.22 5.70 5.32 0.00 1.44 1.84 3.28 3.10 Quercus chapmanii 2.89 5.97 2.75 0.00 2.12 2.47 1.75 2.61 Ilex glabra 2.28 0.00 0.28 0.00 0.57 2.18 0.77 1.27 Myrica cerifera 0.39 1.32 1.87 0.00 0.42 0.43 0.95 0.90 Lyonia fruticosa 1.48 0.00 0.07 0.00 0.00 0.55 0.42 0.73 Vaccinium myrsinites 3.55 2.80 0.37 0.00 0.05 0.34 0.24 0.59 Rhus copallinum 0.00 0.00 0.00 0.00 0.16 0.31 0.00 0.26 Licania michauxii 0.04 0.00 0.00 0.00 0.25 0.08 0.00 0.14 Gaylussacia dumosa 0.43 0.06 0.00 0.00 0.09 0.07 0.00 0.10 Gaylussacia nana 0.28 1.60 0.00 0.00 0.15 0.00 0.00 0.07 Pinus clausa 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Pinus elliottii 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Pinus palustris 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Quercus minina 0.59 0.27 0.23 0.00 0.00 0.00 0.00 0.00 Quercus nigra 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Quercus sp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Brevaria racemosa 1.44 0.72 0.00 0.00 0.00 0.00 0.00 0.00 Ceratolia ericoides 2.71 7.49 0.00 0.00 0.00 0.00 0.00 0.00 Osmanthus americanus 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Salix caroliniana 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Smilax auriculata 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Smilax spp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Cladonia evansii 1.05 0.00 0.33 0.00 0.00 0.00 0.00 0.00 Opuntia humifusa 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 110

PAGE 111

Table 3-8. Preand postburn ab solute importance values of herb and woody species in 5C in Cedar Key Scrub State Reserve. Absolute im portance values for control sites 5A and 5D are also displayed. d = days. M = Months. Control Time after burning Species 5A 5D Pre 11 d 3 M 6 M 9 M 12 M Herbaceous Poaceae 0.0418 0.0164 0.0173 0.07 69 0.0619 0.1031 0.1027 0.1120 Galactia elliottii 0.1457 0.1055 0.0000 0.0000 0.1736 0.0932 0.0032 0.0722 Solidago odora 0.0000 0.0000 0.0000 0.0000 0.0162 0.0243 0.0135 0.0446 Galactia mollis 0.0000 0.0059 0.0000 0.0000 0.0043 0.0061 0.0000 0.0419 Crotalaria rotundifolia 0.0000 0.0000 0.0000 0.0000 0.0038 0.0035 0.0039 0.0029 Agalinis filifolia 0.0000 0.0053 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Asclepias sp 0.0026 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Carphephorus corymbosus 0.0000 0.0088 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Clitoria mariana 0.0056 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Woodwardia virginica 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Zamia pumila 0.0000 0.0033 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Woody Quercus myrtifolia 0.5225 0.6254 0.8850 0.4947 0.8913 0.8120 0.8514 0.8342 Serenoa repens 0.3733 0.3338 0.5421 0.5730 0.4496 0.4946 0.5668 0.4545 Quercus geminata 0.2049 0.1288 0.2116 0.1788 0.3258 0.2940 0.3021 0.3040 Lyonia ferruginea 0.1353 0.2377 0.2861 0.2467 0.1717 0.1981 0.2422 0.2204 Lyonia lucida 0.3038 0.4868 0.3067 0.2024 0.2095 0.1958 0.2120 0.1961 Quercus chapmanii 0.1680 0.1805 0.2191 0.1466 0.2304 0.2098 0.2008 0.1779 Vaccinium myrsinites 0.2186 0.1309 0.0712 0.0000 0.0524 0.0760 0.0794 0.0843 Ilex glabra 0.1136 0.0202 0.0185 0.0188 0.0746 0.1169 0.0873 0.0747 Licania michauxii 0.0267 0.0176 0.0000 0.0000 0.0697 0.0663 0.0246 0.0580 Myrica cerifera 0.0581 0.0500 0.1050 0.0174 0.0382 0.0547 0.0660 0.0512 Quercus minina 0.1165 0.1275 0.0942 0.0000 0.0300 0.0346 0.0617 0.0401 Pinus clausa 0.0000 0.0000 0.0079 0.0096 0.0031 0.0059 0.0170 0.0390 Lyonia fruticosa 0.0839 0.0000 0.0230 0.0510 0.0206 0.0358 0.0380 0.0368 Rhus copallinum 0.0000 0.0000 0.0000 0.0000 0.0394 0.0394 0.0000 0.0301 Gaylussacia nana 0.0115 0.1897 0.0082 0.0000 0.0400 0.0308 0.0188 0.0273 Gaylussacia dumosa 0.1566 0.0781 0.0000 0.0000 0.0360 0.0228 0.0233 0.0250 Smilax auriculata 0.0157 0.0106 0.0000 0.0000 0.0033 0.0208 0.0275 0.0195 Brevaria racemosa 0.0473 0.0128 0.0000 0.0210 0.0239 0.0270 0.0250 0.0178 Pinus elliottii 0.0056 0.0000 0.0121 0.0000 0.0000 0.0029 0.0067 0.0109 Pinus palustris 0.0083 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0060 Quercus nigra 0.0000 0.0000 0.0061 0.0000 0.0000 0.0000 0.0000 0.0000 Quercus sp 0.0086 0.0305 0.0142 0.0000 0.0032 0.0000 0.0000 0.0000 Ceratolia ericoides 0.0402 0.0900 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Osmanthus americanus 0.0000 0.0025 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Salix caroliniana 0.0000 0.0000 0.0000 0.0000 0.0033 0.0030 0.0000 0.0000 Smilax spp 0.0032 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Cladonia evansii 0.0962 0.0638 0.1239 0.0000 0.0000 0.0000 0.0000 0.0000 Opuntia humifusa 0.0028 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 111

PAGE 112

Table 3-9. Preand postburn abso lute densities of herb and w oody species in 2M in Cedar Key Scrub State Reserve. Absolute densities for control sites 5A and 5D are also shown. d = days. M = months. Control Time after burning Species 5A 5D Pre 12 d 3 M 6 M 9 M 12 M Herbaceous Galactia elliottii 2.46 1.09 0.00 2.48 3.37 0.59 0.40 2.50 Poaceae 0.37 0.14 0.17 0.02 0.55 0.51 0.50 0.77 Solidago odora 0.00 0.00 0.00 0.02 0.14 0.12 0.02 0.35 Woodwardia virginica 0.00 0.00 0.00 0.00 0.30 0.04 0.00 0.21 Galactia mollis 0.00 0.18 0.00 0.12 0.13 0.00 0.00 0.07 Agalinis filifolia 0.00 0.15 0.00 0.00 0.00 0.00 0.00 0.00 Asclepias sp 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Carphephorus corymbosus 0.00 0.19 0.00 0.00 0.00 0.00 0.00 0.00 Clitoria mariana 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Crotalaria rotundifolia 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Zamia pumila 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 Woody Quercus myrtifolia 10.91 12.17 12.13 9.31 29.55 31.68 31.27 21.84 Lyonia ferruginea 2.09 5.06 5.40 2.16 8.84 13.66 14.11 13.12 Gaylussacia nana 0.31 5.30 2.78 5.23 16.54 15.82 10.19 11.54 Quercus geminata 2.90 2.24 6.62 1.82 13.52 13.71 13.18 11.50 Lyonia lucida 7.03 10.70 5.27 4.10 7.12 8.44 8.10 10.95 Vaccinium myrsinites 5.08 2.04 7.02 0.29 2.69 4.29 8.06 9.67 Quercus chapmanii 3.01 2.01 2.02 4.80 8.42 9.97 9.00 6.90 Lyonia fruticosa 1.88 0.00 0.55 0.77 3.33 4.45 5.04 6.20 Serenoa repens 1.14 1.01 1.21 1.64 2.83 3.20 3.25 3.17 Ilex glabra 2.98 0.55 0.74 0.07 3.28 3.33 3.29 2.38 Licania michauxii 0.76 0.20 0.03 0.14 1.37 1.49 0.34 1.69 Brevaria racemosa 0.57 0.05 0.04 0.02 1.06 1.09 1.43 0.85 Myrica cerifera 1.04 0.47 1.01 0.03 0.77 1.05 1.06 0.76 Smilax auriculata 0.12 0.08 0.02 0.00 0.13 0.16 0.11 0.20 Gaylussacia dumosa 5.07 1.37 0.00 0.18 0.29 0.00 0.00 0.17 Quercus minina 2.50 1.70 1.25 0.00 0.03 0.33 0.88 0.11 Rhus copallinum 0.00 0.00 0.00 0.00 0.04 0.01 0.00 0.05 Pinus clausa 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Pinus elliottii 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Pinus palustris 0.05 0.00 0.01 0.01 0.01 0.02 0.00 0.00 Quercus nigra 0.00 0.00 0.15 0.02 0.00 0.00 0.00 0.00 Quercus sp 0.07 0.17 0.95 0.03 0.00 0.00 0.00 0.00 Ceratolia ericoides 0.03 0.02 0.03 0.00 0.00 0.00 0.00 0.00 Osmanthus americanus 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 Salix caroliniana 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Smilax spp 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Cladonia evansii 3.51 1.80 7.44 0.00 0.00 0.00 0.00 0.00 Opuntia humifusa 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 112

PAGE 113

Table 3-10. Preand postburn ab solute frequencies of herb a nd woody species in 2M in Cedar Key Scrub State Reserve. Absolute frequencies for control sites 5A and 5D are also presented. d = days. M = Months. Control Time after burning Species 5A 5D Pre 12 d 3 M 6 M 9 M 12 M Herbaceous Galactia elliottii 31 29 0 30 40 28 3 36 Poaceae 12 6 9 8 15 14 14 17 Solidago odora 0 0 0266 3 8 Galactia mollis 0 1 0120 1 3 Woodwardia virginica 0 0 0021 0 2 Agalinis filifolia 0 1 0000 0 0 Asclepias sp 1 0 0000 0 0 Carphephorus corymbosus 0 1 0000 0 0 Clitoria mariana 2 0 0000 0 0 Crotalaria rotundifolia 0 0 0000 0 0 Zamia pumila 0 1 0000 0 0 Woody Quercus myrtifolia 34 44 48 44 44 45 46 44 Quercus geminata 29 34 43 23 44 43 43 44 Gaylussacia nana 1 29 9 30 38 38 33 39 Vaccinium myrsinites 34 27 38 6 32 37 40 39 Serenoa repens 27 29 35 36 36 36 36 37 Quercus chapmanii 32 35 34 28 33 37 36 34 Myrica cerifera 14 12 19 1 15 18 17 19 Lyonia ferruginea 19 34 20 17 18 19 18 18 Lyonia lucida 28 38 17 12 16 16 16 17 Quercus minina 26 40 23 0 3 15 23 12 Licania michauxii 5 6 1 3 11 11 5 10 Lyonia fruticosa 13 0 4356 6 6 Brevaria racemosa 8 2 2255 5 5 Ilex glabra 13 4 4255 5 5 Smilax auriculata 5 4 1044 4 4 Gaylussacia dumosa 24 22 0242 0 3 Rhus copallinum 0 0 0031 0 3 Pinus palustris 3 0 1111 0 1 Pinus clausa 0 0 00 00 0 0 Pinus elliottii 2 0 0000 0 0 Quercus nigra 0 0 6100 0 0 Quercus sp 3 12 21 100 0 0 Ceratolia ericoides 3 6 4001 1 0 Osmanthus americanus 0 1 0000 0 0 alix caroliniana 0 0 0000 0 0 Smilax spp 1 0 0000 0 0 Cladonia evansii 8 12 21 000 0 0 Opuntia humifusa 1 0 0000 0 0 113

PAGE 114

Table 3-11. Preand postburn absolute mean per cent cover of herb and woody species in 2M in Cedar Key Scrub State Reserve. Absolute mean percent cover fo r control sites 5A and 5D are also shown. d = days. M = Months. Control Time after burning Species 5A 5D Pre 12 d 3 M 6 M 9 M 12 M Herbaceous Galactia elliottii 2.06 1.76 0.00 0.04 1.61 0.08 0.13 1.23 Poaceae 0.47 0.00 0.15 0.00 0.16 0.39 0.33 0.55 Solidago odora 0.00 0.00 0.00 0.00 0.00 0.05 0.00 0.30 Galactia mollis 0.00 0.00 0.00 0.00 0.06 0.00 0.00 0.11 Agalinis filifolia 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Asclepias sp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Carphephorus corymbosus 0.00 0.26 0.00 0.00 0.00 0.00 0.00 0.00 Clitoria mariana 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Crotalaria rotundifolia 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Woodwardia virginica 0.00 0.00 0.00 0.00 0.05 0.00 0.00 0.00 Zamia pumila 0.00 0.09 0.00 0.00 0.00 0.00 0.00 0.00 Woody Serenoa repens 23.95 24.39 24.40 1.00 13.34 20.07 21.57 21.64 Quercus myrtifolia 20.07 27.36 22.21 0.73 14.17 12.47 14.48 18.68 Lyonia ferruginea 4.22 5.70 6.27 0.00 3.24 5.98 5.26 11.18 Quercus geminata 6.75 0.64 4.96 0.15 3.98 5.13 4.63 8.58 Quercus chapmanii 2.89 5.97 2.61 0.62 2.09 2.35 2.97 4.07 Vaccinium myrsinites 3.55 2.80 2.80 0.00 1.06 1.20 1.43 2.76 Lyonia lucida 8.94 18.00 7.30 0.00 1.81 3.18 3.11 2.62 Gaylussacia nana 0.28 1.60 0.63 0.15 1.70 1.84 0.61 1.75 Lyonia fruticosa 1.48 0.00 0.02 0.00 0.10 0.30 0.66 0.77 Ilex glabra 2.28 0.00 0.44 0.00 0.90 0.54 0.46 0.74 Licania michauxii 0.04 0.00 0.00 0.00 0.20 0.07 0.00 0.36 Myrica cerifera 0.39 1.32 1.36 0.00 0.06 0.45 0.19 0.32 Pinus clausa 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Pinus elliottii 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Pinus palustris 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Quercus minina 0.59 0.27 0.00 0.00 0.00 0.00 0.03 0.00 Quercus nigra 0.00 0.00 0.03 0.00 0.00 0.00 0.00 0.00 Quercus sp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Brevaria racemosa 1.44 0.72 0.20 0.00 0.00 0.09 0.14 0.00 Ceratolia ericoides 2.71 7.49 3.53 0.00 0.00 0.22 0.14 0.00 Gaylussacia dumosa 0.43 0.06 0.00 0.00 0.02 0.00 0.00 0.00 Osmanthus americanus 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Rhus copallinum 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Salix caroliniana 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Smilax auriculata 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Smilax spp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Cladonia evansii 1.05 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Opuntia humifusa 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 114

PAGE 115

Table 3-12. Preand postburn ab solute importance values of herb and woody species in 2M in Cedar Key Scrub State Reserve. Absolute importance values for control sites 5A and 5D are also displayed. d = days. M = months. Control Time after burning Species 5A 5D Pre 12 d 3 M 6 M 9 M 12 M Herbaceous Galactia elliottii 0.1457 0.1055 0.0000 0.1984 0.1693 0.0780 0.0141 0.1276 Poaceae 0.0418 0.0164 0.0280 0.03 02 0.0466 0.0475 0.0486 0.0559 Solidago odora 0.0000 0.0000 0.0000 0.0080 0.0165 0.0172 0.0084 0.0268 Galactia mollis 0.0000 0.0059 0.0000 0.0072 0.0076 0.0000 0.0027 0.0094 Woodwardia virginica 0.0000 0.0000 0.0000 0.0000 0.0090 0.0029 0.0000 0.0069 Agalinis filifolia 0.0000 0.0053 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Asclepias sp 0.0026 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Carphephorus corymbosus 0.0000 0.0088 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Clitoria mariana 0.0056 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Crotalaria rotundifolia 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Zamia pumila 0.0000 0.0033 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Woody Quercus myrtifolia 0.5225 0.6254 0.6288 0.7094 0.7121 0.6230 0.6674 0.5620 Serenoa repens 0.3733 0.3338 0.4283 0.5528 0.4175 0.4904 0.5130 0.4062 Quercus geminata 0.2049 0.1288 0.2928 0.1955 0.3298 0.3247 0.3195 0.3300 Lyonia ferruginea 0.1353 0.2377 0.2293 0.1267 0.2028 0.2787 0.2708 0.3166 Gaylussacia nana 0.0115 0.1897 0.0809 0.3220 0.2923 0.2696 0.1933 0.2280 Vaccinium myrsinites 0.2186 0.1309 0.2592 0.0308 0.1303 0.1541 0.2078 0.2234 Quercus chapmanii 0.1680 0.1805 0.1568 0.4753 0.2108 0.2253 0.2328 0.2023 Lyonia lucida 0.3038 0.4868 0.2327 0.1654 0.1492 0.1735 0.1726 0.1803 Lyonia fruticosa 0.0839 0.0000 0.0203 0.0338 0.0468 0.0599 0.0737 0.0838 Myrica cerifera 0.0581 0.0500 0.0843 0.0046 0.0466 0.0634 0.0594 0.0577 Licania michauxii 0.0267 0.0176 0.0031 0.0152 0.0453 0.0424 0.0167 0.0451 Ilex glabra 0.1136 0.0202 0.0292 0.0095 0.0642 0.0519 0.0516 0.0446 Quercus minina 0.1165 0.1275 0.0811 0.0000 0.0079 0.0412 0.0714 0.0302 Brevaria racemosa 0.0473 0.0128 0.0084 0.0080 0.0228 0.0239 0.0291 0.0203 Smilax auriculata 0.0157 0.0106 0.0029 0.0000 0.0113 0.0116 0.0119 0.0116 Gaylussacia dumosa 0.1566 0.0781 0.0000 0.0127 0.0133 0.0051 0.0000 0.0089 Rhus copallinum 0.0000 0.0000 0.0000 0.0000 0.0080 0.0026 0.0000 0.0078 Pinus palustris 0.0083 0.0000 0.0027 0.0040 0.0026 0.0027 0.0000 0.0024 Pinus clausa 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Pinus elliottii 0.0056 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Quercus nigra 0.0000 0.0000 0.0183 0.0043 0.0000 0.0000 0.0000 0.0000 Quercus sp 0.0086 0.0305 0.0706 0.0046 0.0000 0.0000 0.0000 0.0000 Ceratolia ericoides 0.0402 0.0900 0.0566 0.0000 0.0000 0.0000 0.0000 0.0000 Osmanthus americanus 0.0000 0.0025 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Salix caroliniana 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Smilax spp 0.0032 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Cladonia evansii 0.0962 0.0638 0.1869 0.0000 0.0000 0.0000 0.0000 0.0000 Opuntia humifusa 0.0028 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 115

PAGE 116

Table 3-13. Density of ramets in 5C after prescr ibed burning in Cedar Key Scrub State Reserve. Time after burning Species 11 days 3 Months 6 Months 9 Months 12 Months Herbaceous Poaceae 0.09 0.61 1.15 0.9800 1.9000 Solidago odora 0.00 0.07 0.05 0.0500 1.8000 Galactia elliottii 0.00 2.66 1.26 0.0000 1.3500 Gallactia moilis 0.00 0.04 0.05 0.0000 1.2200 Crotalaria rotundifolia 0.00 0.08 0.07 0.0700 0.0300 Agalinis filifolia 0.00 0.00 0.00 0.0000 0.0000 Asclepias sp 0.00 0.00 0.00 0.0000 0.0000 Carphephorus corymbosus 0.00 0.00 0.00 0.0000 0.0000 Clitoria mariana 0.00 0.00 0.00 0.0000 0.0000 Woodwardia virginica 0.00 0.00 0.00 0.0000 0.0000 Zamia pumila 0.00 0.00 0.00 0.0000 0.0000 Woody Quercus myrtifolia 0.00 44.13 41.98 46.2500 40.5300 Quercus geminata 0.00 12.82 12.28 12.8600 11.1900 Lyonia ferruginea 0.00 7.74 9.81 10.6400 11.0100 Lyonia lucida 0.00 7.68 8.28 8.3200 8.3300 Quercus chapmanii 0.00 9.70 8.28 7.0400 6.1100 Serenoa repens 1.80 3.31 4.32 4.3400 4.9400 Ilex glabra 0.00 4.57 4.84 4.6700 3.4700 Licania michauxii 0.00 2.65 2.43 0.9000 2.3400 Vaccinum myrsinites 0.00 0.97 1.43 1.4400 1.6900 Lyonia fruticosa 0.00 1.09 1.70 2.0600 1.5500 Gaylussacia dumosa 0.00 1.69 1.41 1.4600 1.4000 Gaylussacia nana 0.00 1.32 1.19 0.2800 1.1300 Brevaria racemosa 0.00 1.89 1.97 1.6200 1.0200 Myrica cerifera 0.00 0.42 0.76 0.6800 0.4900 Rhus copallinum 0.00 0.25 0.28 0.0000 0.3100 Pinus clausa 0.00 0.00 0.03 0.0800 0.2500 Smilax auriculata 0.00 0.03 0.11 0.1700 0.1200 Quercus minina 0.00 0.30 0.38 0.3500 0.0900 Pinus palustris 0.00 0.00 0.00 0.0000 0.0800 Pinus elliottii 0.00 0.00 0.01 0.0200 0.0500 Quercus nigra 0.00 0.00 0.00 0.0000 0.0000 Quercus sp 0.00 0.02 0.00 0.0000 0.0000 Ceratolia ericoides 0.00 0.00 0.00 0.0000 0.0000 Osmanthus americanus 0.00 0.00 0.00 0.0000 0.0000 Salix caroliniana 0.00 0.03 0.02 0.0000 0.0000 Smilax spp 0.00 0.00 0.00 0.0000 0.0000 116

PAGE 117

Table 3-14. Density of ramets in 2M after prescr ibed burning in Cedar Key Scrub State Reserve. Time after burning Species 12 days 3 Months 6 Months 9 Months 12 Months Herbaceous Galactia elliottii 2.48 3.37 0.59 0.4000 2.5000 Poaceae 0.02 0.55 0.51 0.5000 0.7700 Solidago odora 0.02 0.08 0.07 0.0200 0.3500 Woodwardia virginica 0.00 0.30 0.04 0.0000 0.2100 Gallactia moilis 0.12 0.13 0.00 0.0000 0.0700 Agalinis filifolia 0.00 0.00 0.00 0.0000 0.0000 Asclepias sp 0.00 0.00 0.00 0.0000 0.0000 Carphephorus corymbosus 0.00 0.00 0.00 0.0000 0.0000 Clitoria mariana 0.00 0.00 0.00 0.0000 0.0000 Crotalaria rotundifolia 0.00 0.00 0.00 0.0000 0.0000 Zamia pumila 0.00 0.00 0.00 0.0000 0.0000 Woody Quercus myrtifolia 7.85 28.08 30.18 29.7700 20.1700 Lyonia ferruginea 1.00 7.68 12.50 12.9500 11.7900 Gaylussacia nana 5.23 16.54 15.82 10.1900 11.5400 Quercus geminata 1.50 13.20 13.39 12.8600 11.0600 Lyonia lucida 3.42 6.47 7.79 7.4500 10.3000 Vaccinum myrsinites 0.29 2.69 4.29 8.0600 9.6700 Quercus chapmanii 4.52 8.14 9.67 8.7000 6.5500 Lyonia fruticosa 0.74 3.30 4.42 5.0100 6.1700 Serenoa repens 1.64 2.83 3.20 3.2500 3.1700 Ilex glabra 0.00 3.21 3.26 3.2200 2.3100 Licania michauxii 0.14 1.37 1.49 0.3400 1.6900 Brevaria racemosa 0.00 1.04 1.07 1.4100 0.8200 Myrica cerifera 0.03 0.77 1.05 1.0600 0.7600 Smilax auriculata 0.00 0.13 0.16 0.1100 0.2000 Gaylussacia dumosa 0.18 0.29 0.00 0.0000 0.1700 Quercus minina 0.00 0.03 0.33 0.8800 0.1100 Rhus copallinum 0.00 0.04 0.01 0.0000 0.0500 Pinus clausa 0.00 0.00 0.00 0.0000 0.0000 Pinus elliottii 0.00 0.00 0.00 0.0000 0.0000 Pinus palustris 0.00 0.00 0.02 0.0000 0.0000 Quercus nigra 0.02 0.00 0.00 0.0000 0.0000 Quercus sp 0.03 0.00 0.00 0.0000 0.0000 Ceratolia ericoides 0.00 0.00 0.00 0.0000 0.0000 Osmanthus americanus 0.00 0.00 0.00 0.0000 0.0000 Salix caroliniana 0.00 0.00 0.00 0.0000 0.0000 Smilax spp 0.00 0.00 0.00 0.0000 0.0000 117

PAGE 118

Table 3-15. Tree mortality in quadrats and in grids 5C and 2M in Cedar Key Scrub State Reserve. Trees Dead Trees Mortality Before Burning After Burning Percentage 5C 2M 5C 2M 5C 2M Quadrats Pinus clausa 1 0 0 0 0.0 0.0 Pinus elliottii 3 0 3 0 100.0 0.0 Pinus palustris 0 1 0 1 0.0 100.0 Quercus myrtifolia 43 21 0 3 0.0 14.3 Quercus chapmanii 13 4 3 0 23.1 0.0 Quercus geminata 15 9 0 0 0.0 0.0 Lyonia ferruginea 35 53 4 0 11.4 0.0 Ceratolia ericoides 0 1 0 1 0.0 100.0 Subtotal 110 89 10 5 9.1 5.6 Total 199 15 7.5 Grids Pinus clausa 18 0 14 0 77.8 0.0 Pinus elliottii 108 1 96 1 88.9 0.0 Pinus palustris 47 53 44 53 93.6 100.0 Subtotal 173 54 154 54 89.0 100.0 Total 227 208 91.6 118

PAGE 119

Table 3-16. Coefficient of determination (r2) resulting from Detrended Correspondence Analysis of the preand postburn sites of a lo ng-unburned scrubby flatwoods in Cedar Key Scrub State Reserve. r squared Site Data Axis Increment Cumulative Absolute Density 1 0.8440 0.8440 2 0.0760 0.9200 3 0.0300 0.9500 Absolute Mean % Cover 1 0.6260 0.6260 2 0.3600 0.9860 5C Woody 3 -0.0520 0.9340 Relativized Density 1 0.7260 0.7260 2 0.2320 0.9590 3 0.0270 0.9860 Relativized Mean % Cover 1 0.8790 0.8790 2 0.0980 0.9770 3 -0.0040 0.9720 Absolute Density 1 0.9290 0.9290 2 0.0430 0.9720 3 0.0050 0.9770 Absolute Mean % Cover 1 0.2570 0.2570 2 0.6590 0.9160 2M Woody 3 0.0380 0.9540 Relativized Density 1 0.9090 0.9090 2 0.0560 0.9650 3 0.0230 0.9880 Relativized Mean % Cover 1 0.5480 0.5480 2 0.3830 0.9310 3 -0.0090 0.9220 119

PAGE 120

Table 3-17. Summary statistics of the Multi-response Permutati on Procedure for woody absolute densities and mean percent c over between control and treatment sites at preburn and 12 months postburn in Cedar Key Scrub State Reserve. Results are given for Euclidean and Sorensen distances. Variable Distance Observed d Expected Variance Skewness T p Absolute Density Euclidean 46.9029 50.8919 0.0159 -0.7659 -31.6566 < 0.001 Sorensen 0.7039 0.7569 0.0000 -0.4809 -35.9750 < 0.001 Absolute Cover Euclidean 73.2009 73.6672 0.0611 -0.8182 -1.8872 0.0469 Sorensen 0.7608 0.7703 0.0000 -0.7717 -4.0362 0.0014 Code: Observed = Observed delta. Expected = E xpected delta. Delta is the weighted average distance within-group distance. T is th e value of the T test statistics. 120

PAGE 121

121 Table 3-18. Multiple comparison for absolute dens ities and mean percent cover between control and treatment sites at 12 months post burn and between preburn and 12 months postburn values of treatment sites in Cedar Key Scrub State Reserve. Absolute Density Absolute Cover Sites T p-value T p-value Euclidean 5A vs 5C post -17.5538 <0.0001 -0.358058 0.26219348 5A vs 2M post -12.9969 <0.0001 -0.449258 0.24910358 5D vs 5C post -22.5284 <0.0001 -3.657711 0.00845920 5D vs 2M post -24.4070 <0.0001 -4.644292 0.00223107 5C pre vs 5C post -20.0190 <0.0001 0.151180 0.42555303 2M pre vs 2M post -12.0040 <0.0001 -0.018410 0.38794972Sorensen 5A vs 5C post -17.9852 <0.0001 -1.145694 0.11978385 5A vs 2M post -13.8534 <0.0001 -1.447951 0.08799726 5D vs 5C post -26.5362 <0.0001 -4.079046 0.00532423 5D vs 2M post -29.5114 <0.0001 -6.554787 0.00020113 5C pre vs 5C post -15.4409 <0.0001 0.224953 0.45691273 2M pre vs 2M post -10.0212 <0.0001 -0.622968 0.21616540 T is the value of the T test statistics.

PAGE 122

Table 3-19. Comparison among several studies and Cedar Key Scrub State Reserve (CKS RR) regarding common variables measured in each research. The data presented for bareground and for pl ant species is mean percent cover. Data for Schmalzer and Hinkles study average all strata. Data for CKSSR is the average of the two treatment sites. Schmalzer & Hinkle 1992 Abrahamson 1996 Greenberg 2003 Silva-Lugo Oak saw palmetto scrub Scrubby flatw oods Sand pine scrub Scrubby flatwoods Pre 6 M 12 M Rec Pre 12 M Rec Pre 5 M 16 M Rec Pre 6 M 12 M Rec Bareground 0.00 22.90 14.60 0.00 15.00 25.00 1.00 3.80 6.80 Mean height (m) 1.08 0.32 0.50 7.1 1.00 1.00 0.90 0.4 3.53 0.97 1.08 Species richness 8.50 10.50 10.40 18.60 13.50 10.00 12.30 14.80 18.50 22.00 22.50 G. elliottii 0.00 1.70 0.00 0.00 0.00 0.00 0.00 0.10 0.87 Q. myrtifolia 36.00 17.70 18.40 5.0 58.73 22.67 39.59 1.3 25.32 14.65 21.82 1.0 Q. geminata 15.30 11.70 10.60 5.0 4.80 4.80 3.0 2.67 1.14 2.23 1.6 3.83 4.82 8.11 1.0 Q. chapmanii 6.70 4.30 4.20 5.0 15.20 15.00 3.0 0.98 0.33 0.93 1.3 2.68 2.41 3.34 1.0 L. ferruginea 0.84 0.20 1.10 1.0 5.79 3.91 7.14 1.0 L. lucida 17.60 10.20 13.80 4.5 2.75 2.90 3.0 5.82 2.76 3.10 S. repens 31.80 18.00 30.40 1.3 21.70 18.00 2.0 0.40 0.33 0.30 1.8 26.53 19.70 22.32 V. myrsinites 1.50 2.30 2.40 0.5 0.52 1.00 3.0 1.59 0.77 1.67 1.0 122 The citation for Abrahamson 1996 is Abrahamson & Abrahamson 1996. Codes: Pre = preburn. M = months. Rec. = recovery time (years). Empty cells mean data not available.

PAGE 123

A B Figure 3-1. Quadrats in Cedar Ke y Scrub State Reserve. A) Quadrat placement in the scrub. B) Location of flags in a quadrat. 123

PAGE 124

A B Figure 3-2. Ramets of Quercus myrtifolia in Cedar Key Scrub State Reserve. A) Twelve days after burning. B) Twenty nine days postburn. 124

PAGE 125

A B Figure 3-3. Ramets of Quercus chapmanii in Cedar Key Scrub State Reserve. A) Twelve days after burning. B) Twenty nine days postburn. 125

PAGE 126

A B Figure 3-4. Ramets of Lyonia ferruginea in Cedar Key Scrub State Reserve. A) Twelve days after burning. B) One year postburn. 126

PAGE 127

A B Figure 3-5. Ramets of Gaylussacia nana in Cedar Key Scrub State Reserve. A) Apparently, there are three individuals. B) Thes e are sprouts from the same root. 127

PAGE 128

128 A B Figure 3-6. Similarity coefficien ts among the four study sites in Cedar Key Scrub State Reserve. A) Jaccards similarity coefficient. B) Sorensens similarity coefficient.

PAGE 129

129 Figure 3-7. Median linkage dendr ogram for herb and woody species in treatment and control sites under preburn conditions in Ce dar Key Scrub State Reserve.

PAGE 130

130 Figure 3-8. Scatterplot of the first two ca nonical axes corresponding to the cluster analysis with Median linkage fusion method for herb and woody species in treatment and control sites under prebur n conditions in Cedar Key Scrub State Reserve.

PAGE 131

131 Figure 3-9. Average linkage dendrogram for he rb and woody species in treatment and c ontrol sites under preburn conditions in Ce dar Key Scrub State Reserve.

PAGE 132

132 Figure 3-10. Scatterplot of the first two canonical axes corresponding to the cluster analysis with Average linkage fusion met hod for herb and woody species in treatment and control sites under prebur n conditions in Cedar Key Scrub State Reserve.

PAGE 133

133 Figure 3-11. Wards minimum-variance linkage dendrogram for herb and woody species in treatment and control sites under prebur n conditions in Cedar Key Scrub State Reserve.

PAGE 134

134 Figure 3-12. Scatterplot of the first two canonical axes corresponding to the cluster analysis with Wards minimum-variance li nkage fusion method for herb and woody species in treatment and control sites under preburn conditions in Cedar Key Scrub State Reserve.

PAGE 135

Serenoa repens Sites 5A 5C pre 2M pre 5D Medians 4 4 3 3 Vacinium myrsinites Sites 2M pre 5A 5D 5C pre Medians 10 5 3 2 Figure 3-13. Duncans multiple comparisons for the median abundances of Serenoa repens and Vaccinium myrsinites among treatment and control si tes in Cedar Key Scrub State Reserve. 135

PAGE 136

A B Figure 3-14. Preburn and postburn mean percent cover of bareground, litter, and debris in Cedar Key Scrub State Reserve. A) Site 5C. B) S ite 2M. Control values for 5A and 5D are included. 136

PAGE 137

Figure 3-15. Preburn and postburn vegetation hei ght in 5C and 2M in Cedar Key Scrub State Reserve. Control values for 5A and 5D are included. 137

PAGE 138

A B Figure 3-16. Preburn and postburn absolute dens ities of the most abunda nce herb species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and 5D are included 138

PAGE 139

A B Figure 3-17. Preburn and postburn absolute freq uencies of the most abundance herb species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and 5D are included. 139

PAGE 140

A B Figure 3-18. Preburn and postburn absolute m ean percent cover of the most abundance herb species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and 5D are included. 140

PAGE 141

A B Figure 3-19. Preburn and postburn absolute im portance values of the most abundance herb species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and 5D are included. 141

PAGE 142

A B Figure 3-20. Preburn and postburn absolute de nsities of the most a bundance woody species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and 5D are included. 142

PAGE 143

A B Figure 3-21. Preburn and postburn absolute freq uencies of the most abundance woody species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and 5D are included. 143

PAGE 144

A B Figure 3-22. Preburn and postburn absolute m ean percent cover of the most abundance woody species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and 5D are included. 144

PAGE 145

A B Figure 3-23. Preburn and postburn absolute im portance values of the most abundance woody species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. Control values for 5A and 5D are included. 145

PAGE 146

A B Figure 3-24. Preburn and postburn absolute ramet density of the most abundance herb species in Cedar Key Scrub State Reserve. A) Site 5C. B) Site 2M. 146

PAGE 147

A B Figure 3-25. Preburn and postburn absolute ra met density of the most abundance woody species in Cedar Key Scrub State Reserv e. A) Site 5C. B) Site 2M. 147

PAGE 148

Figure 3-26. Preburn and postburn species richness in treatment sites in Cedar Key Scrub State Reserve. Species richness for control sites 5A and 5D are also displayed. 148

PAGE 149

A B Figure 3-27. Preburn and postburn species diversity in treatment sites in Cedar Key Scrub State Reserve. A) Simpsons index. B) ShannonWieners index. Species diversity for control sites 5A and 5D are also shown. 149

PAGE 150

Figure 3-28. Preburn and postburn evenness in treatment sites in Cedar Key Scrub State Reserve. Evenness for control site s 5A and 5D are also displayed. 150

PAGE 151

A B Figure 3-29. Detrended correspondence analysis (DCA) sample ordina tion for densities in 5C in Cedar Key Scrub State Reserve. A) DCA ca rried out with absolute densities. B) DCA done with relati vized densities. 151

PAGE 152

A B Figure 3-30. Detrended correspond ence analysis (DCA) sample ordi nation for mean % cover in 5C in Cedar Key Scrub State Reserve. A) DCA carried out with absolute mean % cover. B) DCA done with re lativized mean % cover. 152

PAGE 153

A B Figure 3-31. Detrended correspondence analysis (DCA) sample ordina tion for densities in 2M in Cedar Key Scrub State Reserve. A) DCA ca rried out with absolute density. B) DCA done with relativized density. 153

PAGE 154

A B B Figure 3-32. Detrended correspond ence analysis (DCA) sample ordi nation for mean % cover in 2M in Cedar Key Scrub State Reserve. A) DCA carried out with absolute mean % cover. B) DCA done with re lativized mean % cover. 154

PAGE 155

Figure 3-33. Stand and species ordination of oak-saw palmetto scrub based on preburn absolute mean percent cover in Kennedy Space Center. Codes: ll = Lyonia lucida, as = Aristida stricta qm = Quercus myrtifolia qch = Quercus chapmanii qg = Quercus geminata lfr = Lyonia fruticosa, br = Brevaria racemosa mc = Myrica cerifera sr = Serenoa repens ig = Ilex glabra pb = Persea borbonia 155

PAGE 156

Figure 3-34. Site and species ordination of sc rubby flatwoods based on preburn absolute mean percent cover in Cedar Key Scr ub State Reserve. Codes: ll = Lyonia lucida lf = Lyonia ferruginea lfr = Lyonia fruticosa qm = Quercus myrtifolia qch = Quercus chapmanii qmi = Quercus minima qg = Quercus geminata, vm = Vacinium myrsinites, mc = Myrica cerifera sr = Serenoa repens, ig = Ilex glabra 156

PAGE 157

CHAPTER 4 SMALL MAMMALS RESPONSES TO PRESCRIBED FIRE Introduction Even though prescribed fire is the primary meth od of fuel reduction in the United States, the effects of controlled burns on fauna are not well understood (Pilliod et al. 2003). Of the five groups of vertebrates, mammals have received more attention. The effects of prescribed fire on amphibians have been summarized by Russell et al (1999), Bury et al. ( 2002), and Pilliod et al. (2003). The effects of prescribed burning on bi rds were reviewed by Smith (2000). General effects of prescribed fire on mammals were summarized by Bendell (1974), Lyon et al. (1978), Wright and Bailey (1982), Peek (1986), and Land ers (1987), and general effects of prescribed fire on small mammals were reviewed by Ream (1981) and Smith (2000). Ream (1981) presented an annotated bibliogra phy of 237 papers, and a very brief summary about prescribed fire effect on Sorex spp. (shrews), Sylvilagus spp. (rabbits), Lepus americanus (snowshoe hare), Castor canadiensis (beaver), Eutamias spp. (chipmunks), Spermophilus spp. (ground squirrels), Tamiasciurus hudsonicus (red squirrel), Glaucomys spp. (flying squirrel), Thomomys spp. (pocket gophers), Peromyscus maniculatus ( P. maniculatus), and Clethrionomys spp. and Microtus spp (voles). Smith (2000) presented direct a nd indirect effects of fire, both wildfire and prescribed burning together in a single discussion, although th e majority of the references were prescribed burning studies. He stated that the direct effect of fire (injury or mortality) is lower than the indirect effect through habitat modification. Fire s generally do not kill or kill a very small proportion of small mammals because during a fire they use underground refugia, where adequate ventilation is essential for animal su rvival (Bendell 1974). Immedi ately after fire, some species leave their habitats and emigrate because of lack of food and cover in the burned area. 157

PAGE 158

Other species immigrate to take advantage of the altered habita t. The length of time before species return depends on how much fire alte red the habitat structur e and food supply. Thus, each species is likely to respond differently to fi re and subsequent habitat changes. Actually, most of the literature about the relationship between fire a nd small mammals is about how vegetation changes affect their populations. These ch anges have been mainly reported in terms of abundance and densities following fire. But, little is known about othe r demographic factors that may be essential for understandi ng population responses. These are the most general statements about small mammals responses to fire made by Smith (2000). Another point of view about the effects of prescribed bur ning is presente below. Even though the literature presen ts a variety of responses of sm all mammals to prescribe fire, there are some aspects that can be synthesized. The majority of the studies cited were of short duration (<30 months) and direct mortality wa s rarely documented (Tevis 1956, Chew et al. 1959, Taylor 1981, Ver Steeg et al 1983, Singer and Schullery 1989, Harty et al. 1991); some studies indicated no change in abundance after prescribe burn ing (Arata 1959, Cook 1959, Writz 1977, Kaufman et al. 1983, Jones 1990, Ford et al. 1999, Vreeland and Tietje 1998); the majority of the studies, however, showed a posi tive response (population increase) to prescribed fire in almost all habitats (Tevis 1956 Shadowen 1963, Hatchell 1964, Ahlgren 1966, Lawrence 1966, Beck and Vogl 1972, Stout et al. 1971, Kreftin and Ahlgren 1974, Layne 1974, Wirtz 1977, Hon 1981, Kaufman et al. 1982, McGee 1982, Bock and Bock 1978, 1983, Gunther et al. 1983, and Forde et al. 1984, Kaufman et al. 1988a, Wirtz et al. 1988, Jones 1990, Blanchard 1991, Sullivan 1995b, 1995b, Greenberg et al. 2006) with population increases attributed to an increase of abundance in seeds and/or in sects (Tevis 1956, Ahlgren 1966, Hooven 1973, Layne 1974, 1990, McGee 1982, Halford 1981, and Gunther et al 1983). Some studies reported small 158

PAGE 159

mammal use of refugia during and after a patc hy prescribed burning (MacGee 1982, Schwilk and Keeley 1998); dispersal to unburned areas were described (Tevis 1956, Arata 1959, Lee 1963, Komarek 1965, 1969, Odum et al. 1974, Wirtz 1 977, Blankenship 1982, Forde et al. 1984, Wirtz et al. 1988), and dispersal to burned areas was documented in se veral studies (Cook 1959, Gashwiler 1959, Hatchell 1964, Ahlgren 1966, Sims and Buckner 1973, Krefting and Ahlgren 1974, Layne 1974, Schramm 1983, Martell 1984, V acanti and Geluso 1985, Kaufman et al. 1988b, 1990, Monroe and Converse 2006). Differences in fire regime and management practices (fire in combination with logging, chopping, clearcutting, and mowing), habitat types, topography, and climate among studies make genera lizations difficult to draw. However, there are two aspects about the effects of prescribed bur ning that are relevant be cause of their direct connotations for researchers a nd land managers. These aspect s are the use of surrounding unburned habitats as refugia and the influence of the re-growth of the vege tation on re-colonization. Few studies have indicated that adjacent unburned habitats mi ght be used as refugia and population recovery is due to plan t species regrowth several mont hs/years after prescribed fire. Goatcher (1990) and Blanchard (19 91) carried out research of th e possible use of stream-terrace hardwood forest as refuge for Peromyscus gossypinus in Lee Memorial Forest, Baton Rouge, Louisiana. They used live-trap capture, radio telemetry, and fluorescent tracking pigments to monitor movements before, during, and after pres cribed burning. No move ments across the firebreak were detected by these methods. They concluded that P. gossypinus apparently does not use stream-terrace hardwood forest as refuge after prescribed fire in adjacent pine forests. However, more research is needed because seve ral studies have documented that small mammals temporarily leave burned sites. If they temporarily abandon burned sites, most likely they use other habitats as temporary refugia. Also, the following studies have indicated that small 159

PAGE 160

mammals recolonize burned sites afte r the regrowth of the vegetation. P. maniculatus and Spermophilus armatus (Uinta ground squirrel) populations approached control numbers after three years following a spring burning, when total c over of the understory wa s near control levels in Burro Hill, Bridger-Teton National Park, Wyoming (McGee 1982). The impact of fire on small mammal communities in th e central Appalachians in Penns ylvania was transitory, and the differences in small mammal abundance between unburned and burned sites disappeared within eight months after fire. This rapid recovery of small mammal populations was explained by the fast regrowth of ground cover within the study area (Kirkland et al. 1996 ). This link between small mammals and regrowth of the vegetation was also used as explanation for population recovery in the study conducted by Ahlgren (196 6) in Minnesota and Sullivan and Boateng (1996) in British Columbia. Both studies found an increase in P. maniculatus on burn sites following fire and a decrease in Clethrionomys gapperi (southern red-backed vole) numbers 2-3 years following fire until recovery of the ground cover vegetation occurred. Neotoma mexicana (Mexican woodrats) benefits from thinning an d/or prescribed burning that encourages shrub densities in the long-term by reducing canopy cover in Coconino National Forest, Arizona (Converse et al. 2006). Research on recovery of small mammal popul ations relative to development rate of vegetation structure followi ng fire is highly needed (Taylor 1981). In Florida, unfortunately, few studies have been conducted to provide evidence of the impact of prescribed fire on small mammals, and no study ha s ever evaluated the importance of adjacent habitats to burned sites as refugia. In addi tion, little documentation has been reported on how closely population recovery is linked with the regrow th of the vegetation. Only 11 studies have evaluated the effects of fire (nine prescribed burning and two wildfires studies) on small mammals in Fl orida. Arata (1959) found no change in the 160

PAGE 161

composition of the populations of P. floridanus Peromyscus polionotus (old-field mouse), and Sigmodon hispidus before and after burning in longleaf /turkey oak habitat in north-central Florida. Also, he reported that S. hispidus moved from burned to unburned areas, while P. polionotus and P. floridanus stayed in the burned areas. Therefore, prescribed fire did not have a detrimental effect on these species. Even though V ogl (1973) trapped only during five nights and obtained a low trapping success, he stated that the densities of Peromyscus gossypinus S. hispidus, and Blarina brevicauda (short tailed shrew) appeared to be similar in the burned and unburned hardwood forest in north Florida (Gannet Pond, Leon County). Gates and Tanner (1988) found no apparent effect of prescribed burning on Geomys pinetis (pocket gopher ) at the successional stages of sandhill communities in Or dway-Swisher Preserve (OSP, Putnam County) in north-central Florida. Jones (1989, 1990) found that three populations of Podomys floridanus had little or no mortality due to prescribed fire, and populations we re higher on burned areas that on unburned sites in longleaf/turkey oak habita t in OSP. In addition, she found that all individuals except one did not move out of th e burned area after prescribed burning. Layne (1974) reported a popul ation increase in P. gossypinus and S. hispidus after a wildfire in slash/longleaf pine habitat in north-central Florida and suggested that burned areas could act as dispersal sinks. Layne further stated that the reappearance of Reithrodontomys humulis (Eastern harvest mouse) and S. hispidus on the burned area appeared to be correlated with redevelopment of the ground cover. Layne (1990 ) also conducted a long-term monitoring of P. floridanus population in sand pine scrub at Ceda r Key Scrub State Reserve (CKSSR), Levy County. He found that the species survived a heavy wildfire in 1955 and was still present in 1986; however, absolute density and relative abundance declined 10 years after fire. Layne (1990) also reported that P. floridanus was still present at low nu mbers in sandhill and scrub 161

PAGE 162

sites that were burned in 1927 in Archbold Biological Station, Highlands County. In comparison, populations were higher and more stable in simila r nearby habitats that were burned periodically. According to Layne (1992), P. floridanus populations are higher in ea rly successional stages of scrub and sandhill vegetation following fire. Later, in the absence of fire, populations decline as habitat structure becomes denser, shadier, a nd microclimatic conditions more mesic. Even though Laynes papers (1974, 1990) reported on the eff ect of wildfires, these papers are included in this introduction becau se of the low number of studies on the subject in Florida. Fitzgerald (1990) stated no significant treatme nt effects were detected on S. hispidus and P. gossypinus in dry prairie of Myakka River State Park in southw estern Florida. Jones ( 1992) reviewed 38 papers on the effects of fire on Peromyscus and Podomys, and she found that the majority of the papers described responses of P. maniculatus, whose numbers increased on burned areas in forest and grasslands. Other species differed in their response to fire accor ding to the type of habitats. P. floridanus appeared to have little or no short te rm effects following prescribed burning, and abundance equaled or exceeded pre-fire levels after two or three months. Depue (2005) found that P. floridanus in central Florida increased or recovere d to pre-burn levels within six months following prescribed burning in Bullfrog Creek Mitigation Park; it dropped in numbers following prescribed fire, but started to increase when the study ended in Split Oak Mitigation Park; and the decrease in animal numbers remained unaffect ed by prescribed fire in Chuluota Wilderness Area. No study has ever determined the response of Ochrotomys nuttalli to prescribed fire in Florida. After reviewing the literature on the effects of prescribed bu rning on small mammals in the United States, I found that the majority of the studies had methodological problems that did not permit comparison among them and the possibility of making generalizations. A high proportion 162

PAGE 163

of these studies were short-term (from weeks to one year) and limited in geographic area, fire behavior characteristics were not measured, they had small sample sizes, and lacked pre-fire data, experimental control, re plication, and randomization. Theref ore, these studies are neither true experimental designs nor quasi-experimental ones that could provide strong causal inference and domain (James and MacCulloch 1995). Accordi ng to Pyne et al. (1996), two methods have been frequently used in fire studies. The mo st common one is to compare burned with unburned (controls) areas, but two important assumptions are not validated. First, treatment and control must be ecologically similar. Hence, soils, slope s, species composition, ve getation structure, and fire history need to be similar. Second, fire behavior characteristics must not vary between treatments. The second common method of study is pre-burn versus post-burn comparisons. In this type of study, only the second assumption need s to be made and validated. However, this approach also needs control, replications, and randomization when it is possible. Whelan (1995) and Russel et al. (1999) recommended that future prescribed fire studies should have more rigorous experimental designs, in cluding larger sample sizes, pre-fire baseline data, more carefully selected controls, a nd better replications. Another aspect of importance among prescribed fire studies is the way data analysis has been conducted. The majority of the studies on prescribed bur ning effects on small mammals focus at the population level, providing information about cha nges in abundance indices and densities after fire. The problem with evaluating fire effects by using abundance indices, such as minimum number alive or catch per unit effort, is that thes e indices tend to be biased in their estimates of the true abundance. They are also biased because they do not account for differences in detection probabilities that are likely in small mammal st udies. These include diffe rences in detection probabilities between individual animals over time, or in response to an experimental treatment 163

PAGE 164

(Monroe and Converse 2006). In contrast, few studies (Lee a nd Tietje 2005, Converse et al. 2006, Monroe and Converse 2006) have used other demographic factors such as survival and recapture probabilities to analyze prescribed fire effects. This approach might be critical for a better understanding of prescr ibed burning effects. Since prescribed fire has been extensively used in Florida as a management tool for restoring fire-adapted ecosystems (i.g., longleaf pine-sandhill, sand pine scrub, and scrubby flatwoods), prescribed burning effects on sma ll mammals become very important because of their roles in ecosystem functions. Small mammals constitute the prey for many forest predators (Zeilinski et al 1983; Williams et al. 1992). They influence the structure of vegetative communities through seed predation and disper sal (Vander Wall, 1993; Hollander and Vander Wall 2004; Schnurr et al. 2004). They play an essent ial role as dispersers of ectomycorrhizal fungi (Pyare and Longland 2001). Therefore, more studies on prescrib ed burning effects are needed because several small mammal species ha ve restricted geographical ranges, occurs in only localized habitats that may be vulnerable to management practices, or may be listed under the Endangered Species Act. In addition, no study ha s ever assessed the role of adjacent habitats as refugia, and no study has been conducted to ev aluate the immediate (days after burning) and short-term effects (one month af ter initiation of plant species growth, 6 months, and one year after burning) of prescribed fire on small ma mmal species in Florida. Furthermore, no information is available on survival rate when small mammals leave the burned area. These aspects may be critical for population survival an d have been overlooked in fire studies. These topics of research are relevant for fire managers and those responsible for assessing the potential effects of prescribed burning on rare, sensit ive, and endangered small mammal species in Florida. Managers, researchers, and non-game speci es and their habitats will all benefit from a 164

PAGE 165

more comprehensive understanding of how sm all mammal species respond to prescribed burning. Objectives and Research Hypotheses The objectives of this study were to document small mammal responses to prescribed fire, to evaluate the importance of vegetation surrounding we tlands next to burned sites as refugia, and to determine whether or not population recovery is linked with the re-growth of the vegetation. The research hypotheses were the following: 1. P. floridanus S. hispidus P. gossypinus and O. nuttalli in Cedar Key Scrub State Reserve use the vegetation surrounding wetlands next to burned sites as temporary refugia after prescribed burning. 2. Prescribed burning does not have a negativ e effect on the survival probability of P. floridanus and S. hispidus in Cedar Key Scrub State Reserve. Methodology Trapping Methods Four 10x10 grids with 10 trap lines were used for capturing, marking, and recapturing mice. Grids were installed in the scrubby flatwoods of treatment and control sites. Each grid had 100 standard-sized Sherman Live Traps (7.6 cm x 8.9 cm x 22.9 cm) arranged in ten lines with 10 trapping stations each and 15 m between trapping stations. Each trapping station was identified by a flag. Also, two trap lines with 10 traps each (15 m between traps) were placed between each grid and the wetland next to it. Trap lines were in the vegetati on near the border of each wetland. The idea was to detect movements between the gr ids and trap lines. In addition, two traps were located at the entrance of each burrow found in grids and trap lines, and 4-6 traps were placed around the small wetlands found inside treatment and control sites. Each trap was baited with a 50-50 mix of crimped oats and sunflower seeds, and polyester was used as nesting material during periods of cool weather. Palmetto fronds were used to shade and insulate traps. Traps 165

PAGE 166

were checked at sunrise and mid after noon because of the diurnal activity of S. hispidus and traps were left set at all times during the session. Four trapping sessions were conducted before and after prescribed burning (Figure 4-1). During the eight trapping sessions, trapping was performed up to a 5-day sequence to avoid possible loss in body mass associated with capture (Slade 1991). The 3rd trapping session was carried out right after three hurricanes hit Cedar Key in August-September 2004, and the 5th trapping session was done after pr escribed burning. In treatment and control sites, pre-treatment data were collected from 03/02/04 to 04/20/05, and post-treatmen t data were collected from 04/26/05 to 07/19/06. Collection of post-treatment data started 5 days after prescribed burning and continued at intervals of 3 months after that. In 2006, the 8th trapping session in February, just 9 months after burning, was cancelled because of the low temperature and rescheduled for April 2006. Trapping effort was increased at three opportunities. The first one took place right after the 3rd trapping session, in which five trap lines (10 traps each; 15 m between traps) were installed on places of higher elevation in/n ear each treatment and control sites during five nights. The purpose of this trapping was to recapt ure 79 individuals marked during the 1st and 2nd trapping sessions and not reca ptured during the 3rd session. The second effort occurred after prescribed burning, and it was included during the 5th trapping session and in treat ment sites only. A trap line (10 traps) was placed at the eas t and west side of the grid in 5C and at the east and south side of the grid in 2M. Also, 18 traps were added to the two trap lines located in the vegetation surrounding the wetland in 5C and 2M. The idea was to detect movements of marked individuals from the scrubby flatwoods to wetlands after bur ning. The third effort was carried out only in control sites 5D and 5A in June and July 2005, respectively, because these sites were mowed 166

PAGE 167

with the exception of the grid and 150-200 m buffer zone around the grids. Trapping occurred in half a grid for five nights to determine if mice left the site because of the mechanic activities near the grid in each site. Collected data included species, weight, se x, reproductive condition, trap location, and tag number. Reproductive conditions were as follows: juveniles with varying st ages of gray pelage, subadult (non-breeding) individual s with adult pelage but not evid ence of current sexual activity, and breeding animals with adult pelage and evidence of sexua l activity. Sexual activity was determined in males by testicles in scrotal pos ition and for females by pregnant conditions and nipple size. Each mouse was identified with two unique ear-tags (Na tional Band and Tag Co., Covington, KY) and released at point of capture. Trapping for predators was carried out to rem ove them from the grids before the eighth trapping session in each site. Procyon lotor (raccoon), Urocyon cinereoargenteus (grey fox), Didelphis marsupialis (opossum), and Mustela frenata peninsulae (Florida long-tailed weasel) were the most problematic pr edators found in trea tment and control sites. From the 1st to the 7th trapping session, predators distur bed traps for not more than 3 days. The strategy was to close traps and wait 3-5 days for predators to move to other places. However, before the eighth trapping session, predators were abundant (particularly P. lotor) in treatment sites. Therefore, trapping predators took place for 7-10 days in each site before trapping small mammals. Predators were released at a minimum dist ance of 3.0 km from the grid of capture. Data Analysis Due to the small sample size for the capture -recapture dataset, data analysis was only carried out for P. floridanus and S. hispidus In addition, the data for treatment sites were combined for each species, and the same procedure was done for control sites. Data collected during the extra trapping effort done only in control sites in June -July 2005 were not considered 167

PAGE 168

for the analysis. I evaluated the effect of pr escribed burning on the survival probability of P. floridanus and S. hispidus by using information-theoretic model selection and inference framework (Burnham and Anderson 2002). Also, I used the program MARK 4.3 (White and Burnham 1999) for testing lack of fit and for estimating the survival and recapture probabilities by using the Cormack-Jolly-Seber mode l (Cormack 1964, Jolly 1965, Seber 1965). Assumptions of the Cormack-Jolly-Seber model The Cormack-Jolly-Seber (CJS) model is used alone to estimate survival and recapture probabilities. This model requires information onl y on the recapture of the marked animals, and that these individuals are represen tative of the populations. Severa l assumptions have to be made to be able to estimate the parameters associated with this model. These assumptions are as follows (Williams et al. 2002): 1. Every marked individual at time (i) has the same probability of recapture (pi). 2. Every marked individual immediately after time (i) has the same proba bility of survival to time (i+1). 3. Marks are not lost or missed. 4. All samples are taken in a short period, a nd captured animals are released immediately. 5. All emigration from the sampled area is permanent. 6. The fate of each animal regarding capture and survival probabilities is independent of the fate of any other animal. The goodness of fit (GOF) test The GOF test was used to test the lack of fit of the data to the underlying assumptions of the CJS model. The survival probability phi was considered constant ( .), time dependent ( t ), group dependent ( g; treatment control), and dependent of the interaction group and time ( g*t ). The recapture probability was also considered under the same conditions, and the combination of all these possibilities added up 16 models. This set of models was used as the candidate model set. I 168

PAGE 169

fitted these models to the data by using the pre -defined model option in MARK and carried out the GOF test for the full time-dependent model phi(g*t) p(g*t). This model is the general model because it contains the largest number of parame ters. The idea was to assess if this model adequately fit the data, which m eans verifying whether or not the arrangement of the data meet the expectation determined by the assumptions underlying the CJS model. The GOF test was done by using Bootstrap and Release methods. Both methods estimate the variance inflation factor (c-hat), which quantify the lack of fit of the model to the data or the amount of under or over dispersion that we have in the dataset. The Bootstrap method provides two ways to estimate c-hat: (a) the deviance method which divides the observed deviance (obtained from the summary statistic of the general model) by the mean deviance from the bootstrap summary statistics, and (b) the c-hat me thod that divides the ob served c-hat (obtained from the summary statistic of the general m odel) by the mean c-hat from bootstrap summary statistic. The Release method pres ents three tests of which Test 2 and Test 3 check if the 1st and 2nd assumptions are met. I assumed that the 3rd, 4th, 5th, and 6th assumptions were met. C-hat was estimated by dividing the overall chi-square fr om Test 2 and Test 3 by the overall degree of freedom. I followed Cooch and Whites (2006) recommen dations regarding which c-hat to choose between Bootstrap and Release. These authors s uggest to choose the larg est c-hat value in the interval 1 < c-hat < 3 in order to make the mode l selection more conservative and minimizing the chances of Type II error. A c-ha t value in the interval 1 < c-ha t < 3 means overdispersion of the dataset, and an adjustment for lack of fit is needed with MARK by using that particular c-hat value. So, MARK displays quasi-likelihood Akai kes Information Criterion values (QAIC) after adjustment. A c-hat > 3 means that the general model does not fit the data. A c-hat = 1 means a 169

PAGE 170

perfect fit and the data do not need any adjust ment. A c-hat < 1 means underdispersion of the data, and Cooch and White suggest to treat a c-hat < 1 as a c-hat = 1. These criteria allowed choosing a c-hat value for the general model and made the corresponding co rrection to it and the candidate model set. Model comparison, model select ion, and hypothesis testing To compare and select models, the following steps were carried out. First, selecting the most parsimonious model in the candidate model set by using the Akaikes Information Criterion (AIC). The most parsimonious model is the one that best explains the vari ation in the data with the lowest number of parameters, and it is better supported by the da ta. AIC is a good and welljustified criterion for selecting the most parsimon ious model and it is considered a robust way of model selection (Burnham and Anderson 2002). Th e criteria for model selection were the following: AIC, Delta AIC, the normalized Akai ke weights, model likelihood, the number of parameters, and model deviance. The AIC index measured how much the model explained the variation in the data. The AIC = -2ln( ) +2K = -2 x model log likelihood of parameters (q) given the data + 2 Nro. Parameters. The model with the lowest AIC value was better supported by the data and more parsimonious than other models. Since the da taset is small, MARK calculated the corrected AIC or AICc. Since the candidate model set was adjusted by using the c-hat obtained from the GOF test, MARK displayed the quasi AICc or QAICc values. q/dataDelta AIC ( AIC): the difference between each model and the one with the lowest AIC value (the most parsimonious). MARK calculated AICc because of the small sample size, and Q AICc after adjustment. The following rules of thum b were applied to determine what models were different or not: (a) If Q AICc < 2, both models have equally weight in the data. No real 170

PAGE 171

difference between the two models, (b) if 2 < Q AICc < 7, there is consid erable support for a real difference between the two models, and (c) if Q AICc > 7, there is a strong evidence to support to the conclusion of diffe rence between the two models. The rule of thumb above was used in combination with the normalized Akaike weights. The weight ( wi ) for each model was calculated with Equation 4-1: c cAIC exp 2 AIC exp 2 Q wi Q (4-1) So, wi is the proportion of the data that suppor t a particular model in comparison with all models. The model with the highest wi would be the best model because it had more support than any other model. But, to know how much better it was than the next model, the wi of the best model was divided by the wi of the next best model. This quotient stated how much the best model was supported by the data than the next best model. Model likelihood (index of relative plausibility): the ACI weight of the model of interest divided by the ACI weight of the best model. This quotient indicated how likely a particular model was in comparison with the best model. It is important to highlight that the AIC approach helps to select the best model; however, there is an uncertainty of which model is the best model. The normalized Akaike weights and the likelihood of the mode l measure this uncertainty. The number of parameters and model devian ce were the last two criteria. The model deviance is the difference in -2ln of the current model and -2ln of the saturated model. The saturated model is the mode l with the number of parameters equal to the sample size. q/dataq/data 171

PAGE 172

Second, P. floridanus the most parsimonious model was phi(t) p(.), and the second and third best models were phi(t) p(g) and phi(t) p(t), respectively. In S. hispidus the most parsimonious model was phi(t) p(.) and the second one was phi(t) p(g) The survival probability (phi ) was only time dependent, and the probability of recapture (p ) was constant ( .), time dependent (t), and group dependent (g). Therefore, I considered bu ilding other models to test the prescribed burning and other effect s based on these preliminary results. Third, adding models phi(Fl ood + Fire) p(., t, g) for P. floridanus and phi (Flood + Fire) p(., g) for S. hispidus Flood and prescribed burning (Fire) ar e time dependent variables and their additive effects were modeled with th e probability of recapture constant (. ), time dependent (t), and group dependent (g). This combin ation resulted in nine models for P. floridanus and six models for S. hispidus that were added to the candidate model set of 16 models for comparison purposes. Then, the most parsimonious model was selected out of 25 and 22 models for P. floridanus and S. hispidus respectively. The covariate Flood was included because three hurricanes hit Cedar Key and partially flooded treatm ent and control sites. No previously marked mice were recaptured after flooding. Since, this c ovariate had a strong influence in the survival probability in both species; its additive effect with prescribed burning was modeled by adding linear constraints to MA RK. The basic sequence of steps of building design matrices followed Cooch and White (2006). Fourth, checking for the number or real para meters and adjusting them. Even though MARK estimates the number of parameters, the model st ructure determines the number of parameters that are theoretically estimable. However, if the sample size is small, not all theoretically estimable parameter can be estimated. In addition, when survival and recapture are time dependent, the terminal parameter is not individually identifiable (Lebreton et al. 1992). Since 172

PAGE 173

the sample size for the mark-recapture data is sm all in this study, I manually checked the number of estimable parameters indicated by MARK matched the number of parameters theoretically estimable for a particular model. If they did not match (one or more parameters were not estimated), I manually adjusted the number of parameters to the theoretical number. Fifth, the model selection carried out in the previous steps allowed the testing of the flood and prescribed burning effect on the survival pr obabilities in both species and to estimate survival and recapture parameters. The most parsimonious model provided the most precise and less biased survival estimates. However, since there was an uncertainty about which model was the best model; there was also an uncertainty wi th survival values. Reporting survival estimates from a single model in the candidate model set, even if it was the most parsimonious model, ignored model uncertainty. For this reason, survival estimates were re ported by using modeling averaging, which allowed estimating the average of each parameter of interest from the model set. Modeling averaging takes the estimates of various models, and weights them by using the normalized AIC weights. The following equation was used (Equation 4-2): R i = 1avg () = wi i (4-2) The average value for the parameter was calculated by multiplying the AIC weight of model i and the parameter value estimated by that model and adding up a ll these products in the model set. Finally, survival estimate s from both the most parsimonious model and model averaging were plotted. The purpose of this comparison was to show similarities or dissimilarities between estimated parameter from both approaches. 173

PAGE 174

Results Number of Captured Individuals in Treatment and Control Sites A total of 182 individuals were marked and recaptured 426 times in 29,340 trapping nights. Figure 4-2 illustrates the number of captured individuals per species per trapping session in treatment sites. All species were captured wi th low numbers (1-9 individuals) at the beginning of the study. Of 39 individuals marked in the fi rst 2 trapping sessions, I di d not recapture any of them after the hurricanes during the 3rd trapping session and during th e additional trapping effort carried out in upper grounds. Most likely, they died. After the 3rd trapping session, the number of individuals of S. hispidus and P. floridanus increased through time from 13 to 34 individuals and from 13 to 21 individuals, respectively, even after prescribed bur ning. Also, after the 3rd trapping session, the number of captured individuals of P. gossypinus was stable (4-7 individuals) even after prescribed burning. The num ber of captured individuals of O. nuttalli did not increase after prescribed fire, and only one indivi dual was recapture d through time. Figure 4-3 shows the number of captured indivi duals per species per trapping session in control sites. Again, all species were captured in low numbers (1-11 individuals) at the beginning of the study. Of 39 individuals marked during th e first 2 trapping sessions, I did not recapture any of them after the hurricanes during the 3rd trapping session and durin g the trapping done in upper grounds. Probably, they also died. I captured 9 and one new individuals of P. floridanus and S. hispidus respectively, during the 3rd trapping session. After th e 3rd trapping session, the number of individuals of S. hispidus increased from 2 to 13 individuals, and the number of captured individuals of P. floridanus (6-8 individuals) and P. gossypinus (3-6 individuals) remained relatively stable thr ough time. Only one individual of O. nuttalli was recaptured after the 3rd trapping session. 174

PAGE 175

Figures 4-2 and 4-3 present two different patterns. The number of individuals of S. hispidus and P. floridanus increased in treatment sites onl y, and this event took place after flooding and prescribed burni ng. In contrast, a similar pattern was found for P. gossypinus and O. nuttalli in treatment and control sites. Obviousl y, flooding, prescribe burning, or both affected demography parameters of P. floridanus and S. hispidus The extra trapping effort carried on for contro l sites in June-July 2005 indicated that mice did not move because of mowing. Mowing the vegetation took 2-3 weeks in each site, and the mechanical activity was very inte nse. Even though the mechanical activity produced a loud noise that could be heard from 500 m, marked mice we re not affected by this type of perturbation because they remained in the grids. Number of Captured Individuals in Scrubs and Wetlands Table 4-1 presents the number of captured individuals in the sc rubby flatwoods (scrubs from now on) and in the vegetation surrounding wetlands (wetlands from now on) per trapping session in treatment and contro l sites. In treatment sites before prescribed burning (1st 4th trapping sessions), 54 (74%) out of 73 individuals were captured in scrubs (see also Figure 4-4). During the 4th trapping session after the hurricanes, 26 ( 77%) out of 34 new marked individuals were captured in scrubs, and 22 (85%) out of th e 26 mice were recaptured in wetlands during the 5th trapping session right after prescribed burning. During the 5th trapping session in wetlands, five mice previously trapped in wetlands were recaptured and 18 new i ndividuals were captured. Therefore, marked mice moved to or stayed in wetlands after pres cribed burning and new individuals preferred wetlands ra ther than scrubs. During the 6t h and 7th trapping sessions, 90 individuals were captured in we tlands and six in scrubs. The 90 individuals included mainly previously marked individuals (81) rather than new ones (9). The six individuals found in scrubs corresponded to one P. floridanus marked in the 6th and recaptured in the 7th trapping session, 175

PAGE 176

two new P. floridanus marked in the 7th session, and one P. gossypinus recaptured in the 6th and 7th trapping sessions. Of 46 mice cap tured in wetlands during the 7th trapping session, 31(67%) were recaptured in scrubs during the 8th trapping session. During this last trapping session, one year after prescribed burning, 53 (90%) out of 59 captured individuals were trapped in scrubs. The 53 individuals included 31 recaptured mice from the 7th trapping session and 22 new individuals. Also, six new individuals were captured in wetl ands. During all trapping sessions in control sites, 153 individuals were ca ptured, from which 139 (91%) mice were captured/recaptured only in scrubs (Table 4-1 and Figure 4-5). Therefore, mice returned to the scrubs in treatment sites after at least 11 months (May 2005-April 2006) following prescribed burning. These results clearly suppo rted the research hypothesis and indicated that the vegetation surrounding wetlands provided refuge to the sma ll mammals for at least 11 months following prescribed burning. Mice returned to scrubs after plant species offered both cover and food, and this event took place at least 11 months after prescribed burning. The 7th trapping session occurred in November 2005, and the majority of the mice were still in we tlands. I could not trap at nine months after burning (January 2006) because of the low temper ature. Trapping started in April 2006 at 12 months after burning and mice were recaptured in scrubs. Therefore, mice may have moved to scrubs in March 2006 or after it Mice probably did not move to scrubs in January-February because insect activity was assumed to be low due to the low temperatures and plant species did not start to produce flowers and fruits until April-May 2006. Most like ly, mice found both cover and food in scrubs after 11 months following prescribed burning and returned to scrubs for this reason. 176

PAGE 177

Flood and Fire Effects on Podomys floridanus The GOF test for P. floridanus is presented in Table 4-2. Th e Bootstrap c-hat method did not give any results, which may be due to the sma ll sample size, and the Bootstrap deviance method provided a c-hat larger than the c-hat obtained from the Release method. Hence, the general model and the candidate model set were adjusted to the c-hat = 1.1387 in order to be conservative. This c-hat indicated that there was a little of overdispersion in the dataset. The additive effect of Flood and Fire had an effect on the survival probability of P. floridanus. Table 4-3 presents the candidate model set of 16 models adju sted to c-hat = 1.1387. Model phi(t) p(.) was the most parsimonious model with 73.16% support in the data. But, because of the first three models had 99.88% sup port in the data, phi in these models was time dependent, and p was constant (.), group dependent (g), and time dependent (t), I modeled phi with the covariates Flood and Fire in combination with p (., g, t). Table 4-4 displays the 25 models after including nine models from th e combination phi(Flood + Fire) p(., g, t) and correcting for the number of parameters. As shown in this table, the top model phi (Flood + Fire) p(.) has only 29.92% support in the data, and there is no enough evidence to indicate that this model is different from the 2nd to the 5th model because delta QAICc < 2.0. However, the additive effect of Flood + Fire and the covariate Flood ha d a strong influence on the survival probability of P. floridanus because this set of mode ls is supported by 61.67% of the data, and phi is dependent of Flood + Fire and Flood in the first four models. Even though the covariate Fire by itself does not have support in the data, the eff ect of fire can be seen by comparing model phi (Flood + Fire) p(.) with phi (Flood) p(.). Delta QAICc = 2.83, and this is the Fire effect. The effect of Flood can be noted by comparing phi (F lood + Fire) p(.) with phi (Fire) p(.). Delta QAICc = 30.11. This is the effect of Flood. Model phi (Flood + Fire) p(.) estimated one survival and one recapture parameters because this wa s a particular case of time-dependence where 177

PAGE 178

trapping sessions with the same Flood and Fire c onditions shared the same survival rate (Table 4-5). MARK does not count parameters with standard errors equal to zero in the parameter total. So, the survival rate for non-flood and nonpres cribed fire times was 0.7871 in treatment and control sites, and 0.00 and 1.00 for flooding and af ter prescribed fire times, respectively. Therefore, the additive effect of Flood + Fire had an influence on the survival probability of P. floridanus. The time-dependent covariate Fl ood and Fire did not have an important influence in the recapture probability. Because models phi(Flood + Fire) p(t) and phi(Flood) p(t) had 31.76% support in the data and p(t) was present in thes e two models, I added models phi(Flood + Fire) p(Flood+Fire), phi(Flood + Fire) p(Flood), and phi(Flood + Fire) p(Fire) to analyze the effect of the time-dependent covariates Flood and Fire in the recapture probability. Table 4-6 presents these set of models, and each of th em had very little support in the data (< 8.7 %). Consequently, model phi(Flood + Fire) p(.) still was the most parsimonious model. Prescribed burning did not have a negativ e effect and flooding probably negatively influenced the survival probability of P. floridanus The results of the current analysis indicated that phi(Flood + Fire) p(.) was the most pa rsimonious model among 28 models. The survival parameters estimated for treatment and control sites from this model revealed that both curves were similar, and the only difference was for phi4, where prescribed burning apparently increased survival (Figure 4-6). But, I cannot conclude if this increase was significant or not because the parameter was not estimable (Table 4-7). Nevertheless, I can st ate that prescribed fire did not have a negative influe nce on the survival probability of P. floridanus because any of the models in which Fire was involved alone had support in the data. Furthermore, in the real scenario, prescribed burning indirectly increased survival because of the role of the vegetation 178

PAGE 179

surrounding wetlands as refugia. Thus, these results support the research hypotheses. In contrast, flooding probably had a negative effect. The parameter in the model phi(Fire + Flood) p(.) was not estimable. Statistically, I cannot make any co nclusion. But, practically it is likely that Flood decreased the survival probability of P. floridanus to zero before the 3rd trapping session. I did not recapture 31 P. floridanus marked between the 1st and 2nd trapping sessions after the hurricanes. Table 4-8 summarizes the estimated survival parameters for model averaging, and Figure 4-7 displays the estimated survival parameters for model phi(Flood + Fire) p(.) and model averaging in treatment and control sites. Th e fact that four curves were quite similar indicated that model phi(Flood + Fires) p(.) was the best model and provided pretty good estimates of the survival parameters. The main difference was on phi4, in which the two estimates in treatment sites differed by 0.032212. However, model averaging provided a better estimate on phi4 because of 13 P. floridanus marked in the 4th trapping session, 12 were recaptured in the 5th trapping session. So, phi4 should not be equal to1.00. Flood and Fire Effects on Sigmodon hispidus Table 4-2 shows the GOF test for S. hispidus The Bootstrap deviance method gave a c-hat larger than the c-hat obtained from the Release method, and the Bootstrap c-hat method did not provide any result probably because of the sm all sample size. The general model and the candidate model set were adjusted to the c-ha t given by the Bootstrap deviance method (c-hat = 1.5694). This c-hat revealed overd ispersion in the dataset. The covariate Flood and the a dditive effect of Flood and Fi re had an influence on the survival probability of S. hispidus The candidate model set of 16 models adjusted to a c-hat = 1.5694 is shown in Table 4-9. The most parsimonious model was phi(t) p(.) with 69.78% support in the data, and the first two models had 95.59% s upport in the data. The survival probability phi 179

PAGE 180

in these models was time dependent, and the r ecapture probability p was constant (.) and group dependent (g). Consequently, I decided to m odel phi with the time-de pendent variables Flood and Fire in combination with p (., g). The combination of these models produced six models that were added to the candidate model set. Table 4-10 shows the 22 models fitted, adjusted to c-hat = 1.5694, and corrected for the number of parameters. As can be seen in this table, the top model, phi (Flood) p(.) had 32.51% support in the data, but it was no different from the 2nd to the 4th model because delta QAICc < 2.0. Of the remaining 18 models, phi(Flood + Fire) p(g) and phi(t) p(g) had 9.96% and 5.15% suppor t in the data, respectively, bu t the rest of the models had a support 0.37% in the data. In addition, there were c onsiderable evidences for a real difference between phi(Flood) p(.) and the 5th and 6th models because QAICc > 2.0. The first four models had 83.75% supports in the data a nd it is conformed mainly by the covariates Flood and Flood + Fire. Thus, only models with Flood and Flood + Fire on the apparent survival rate had a substantial support in the data. Therefore, these covariates ha d an effect of the survival probability of S. hispidus The covariate Flood and the a dditive effect of Flood and Fi re apparently reduced the survival probability of S. hispidus but Fire did not have any nega tive effect on the survival rate. The most parsimonious model phi(Flood) p(.) an d the second most parsimonious one phi(Flood + Fire) p(.) had 32.51% and 23.77% support in the data, respec tively (Table 4-10). In both models, Flood apparently decreased survival from values such as 0.9082 and 0.9276 to 0.0000 between the 2nd and 3rd trapping sessions (see Table 4-11, Figure 4-8). Also, Fire in model phi(Flood + Fire) p(.) apparently decreased su rvival from 0.9276 to 0.7736 between the 4th and the 5th trapping sessions. The word apparently is used because the parameters corresponding to Flood in the two most parsimonious models were not estimable, and Fire in model phi(Flood + 180

PAGE 181

Fire) p(.) was estimable but not significant (Tab le 4-12). Practically, Fl ood probably killed all marked S. hispidus in the study during hurricanes even though it could not be demonstrated statistically. Six S. hispidus marked during the first two trapping sessions were not recaptured after the hurricanes. In cont rast, the covariate Fire had 0.186% support in the data in the three models that stood alone, and it was not significant (Table 4-12). Therefore, prescribed burning did not have a significant negative ef fect on the survival probability of S. hispidus Out of 13 marked rats during the 4th trapping session, 10 were recaptured in the vegetation surrounding wetlands after prescribed burning. Hence, thes e results support the research hypothesis. The estimated survival parameters for model phi(Flood + Fire) p(.) a nd model averaging are presented in Table 4-11 and Table 4-13, respec tively, and Figure 4-9 displays the comparison. Even though phi(Flood + Fire) p(.) was not the mo st parsimonious model, it was compared with model averaging because this model and model phi(F lood) p(.) had equal weight in the data, and phi(Flood) p(.) did not contain Fi re and model averaging did. The survival parameters estimated by model phi(Flood + Fire) p(.) were the same for treatment and control site s, with the exception of phi4, because this was a case of time-dependence where trapping sessions with the same flood conditions would share the same survival rate. Th ese estimated parameters were similar to the parameters estimated by modeling averaging with th e exception of phi3 and phi4. For phi3, the value estimated by model phi (Flood + Fire) p(.) in treatment and cont rol sites (0.9275804) was higher than the value estimated by modeling averaging in treatmen t (0.8513742) and control (0.8512919) sites. Nonetheless, these last two values were so similar that are shown as one point in Figure 4-9. For phi4, model phi(Flood + Fire) p(.) estim ated a value for treatment (0.7736407) lower than for control sites (0.9275804). Model av eraging also produced a value for treatment (0.8413499) lower than for control (0.9037845) sites. But, the difference between treatment and 181

PAGE 182

control sites for model phi(Flood + Fire) p(.) was higher (0.1539397) than the difference for model averaging (0.0624346). Therefore, model phi (Flood + Fire) p(.) provided good estimates of the survival parameters with the exception of phi3 and phi4. Parameter phi2 = 0.00 because six rats marked during the first two trapping sessions were not recaptured af ter the hurricanes during the 3rd trapping session. Trapping Predators An increased in the number of predators was noticed in all sites before the 8th trapping session, but mainly in treatment s ites. Before prescribed burning, P. lotor, U. cinereoargenteus D. marsupialis, and M. frenata sprung traps for no more than three days. The first three species were identified by their tracks, and the Florida long-tailed weas el was seen on 04/06/05 at 5:37 pm hunting in the grid in 5A. These predators moved to other places after finding traps closed for 3-5 days. However, before the 8th trapping session and coincidently with the return of mice to the scrubs, predators were trapped in all sites (Table 4-14), and they did not move to other places after closing traps for 3 days. This problem was solved by trapping predators before small mammals for 7-10 days and transl ocating them to other places at least 3.0 km away from the grids. Discussion Low Capture Success Morgans study (1998) and the current study shar ed sites. Morgan trapped on transects and at burrows in three places named CKA, CKB, and CKC. CKA and CKC are in the 5A site in this study; but, transects installed in CKA and CKC and the grid installed by this study were in different places. CKB is the same 5C site in the present study, and trapping occurred at the same place. Even though trapping did not occur in the exactly same place in 5A site, trapping took place in scrubby flatwoods in the same study area. Sites 5A and 5C were not burned since 1957 182

PAGE 183

(Morgan 1998), and 2M and 5D did not have reco rds of the last burn, but apparently both sites burned in a wildfire in 1955 (Jeff DiMaggio, per. com.). Morgan trapped in 1997-98, at least 7 years prior to my trapping period an d she obtained a high trapping success. This study found a low capture success even though a major trapping effort was carried out. Table 4-15 shows the trapping effort of se veral studies conducted for small mammals in Florida. Jones (1990) had the highest with 33,000 trapnights. The curre nt study had the next highest with 29,340 trapnights. N onetheless, Jones trapped du ring 60 months, and I trapped during 31 months. Morgan trapped 2,970 trapni ghts, and she reported a higher number of individuals than the pres ent study. Morgan trapped 430 P. floridanus 275 S. hispidus s, 221 S. hispidus, and 33 O. nuttalli. I expected to obtain a high capture success due to Morgans capture success in the same study area. In addition, Layne (1990) reported peaks in populations in late winter and early spring, exactly wh en I did the first trapping session. There was a lapse of at least 6 years between Morgans and th e current study. During this time vegetation density increased and structure changed with all the consequen ces for small mammals habits and behaviors associated with these changes. The age of the vegetation is an important fact or because all sites were not burned for at least 47 years. When fire suppres sion leads to ha bitat conversion, P. floridanus populations tend to be reduced or eliminated (La yne 1990, 1992, Jones 1992). Populations of P. floridanus in scrub habitat have shown to decline with increased vegetation density (Layne 1990, 1992). However, this explanation is not valid for S. hispidus and P. gossypinus since they are opportunistic and able to use other types of habitats (La yne 1974, Eisenberg 1983). But, vegetation succession without fire with the co rresponding increase in vegetation density might 183

PAGE 184

explain a high capture of O. nuttalli in comparison with Morgans study. The vegetation in all sites may be more suitable for O. nuttalli due to its arboreal habit. Fire suppression also decreas es habitat suitability for Gopherus polyphemus (Cox et al. 1987), and this aspect influences P. floridanus presence in all sites. Even though P. floridanus has been cited as a facultative user of gopher to rtoises burrows in sc rubby flatwoods (Morgan 1998), the high capture success found in burrows by Morgan was an indication of the close association between P. floridanus and the gopher tortoises burrows in this type of habitat. Of 92 burrows found in four sites in this study, only 13 we re gopher tortoises burrows It is very likely that 79 burrows were Dasypus novemcinctus (armadillo) burrows because of the circular shape of the entrances, and traps placed near entrances were not removed. Small mammals have natural population fluctua tions through time. Particularly, population cycles have been reported for small mammals (Batzli 1992). It is possible that trapping occurred when the population size for these species was lo w. A low mast production could be responsible for a low population level, but ther e were no data to support this statement. However, I observed that acorn production was low during 2004-2006 in all sites. Vegetation age, fire suppression, population cy cles, and acorn production or a combination of these factors probably influence capture su ccess. A combination of these factors maybe worked together. Other factors su ch as trapping design and type of bait were not the causes of the low capture success. The 10x10 grid with traps ever y 15 m has been used as the standard grid design in the majority of the small mammal studi es in the United States (John Eisenberg, per. com.). In general, scientists with experience trapping small mammals in Florida consider trapping success low on grids and between 1% a nd 5% (John Eisenberg and James Layne, per. com). However, if we compare Morgans tra pping design (three U-shaped loops of 210 m long 184

PAGE 185

with 14 trapping station separa ted by 15 m and 45 m between loops) with the current study, loops cover a smaller area (1.42 ha) than grids (2 .25 ha), so I would expect to have a higher capture success on grids, and Morgans results were eviden ce of the opposite. Morgan used crimped oat as bait with an excellent result. I used a 50-50 mixture of crimped oat and sun flower seeds, which probably is much better bait than crimped oat by itself. I found that P. floridanus S. hispidus, P. gossypinus and O. nuttalli ate all the sunflower seeds a nd not all the crimped oats in the traps. Flooding Effect Hurricanes Charley (9-14 Augus t), Frances (25 August 8 September), and Jeanne (13-28 September) hit the Florida peninsula in 2004. Char ley did not hit Cedar Key directly, but its winds brought rainfall to the area. Frances and Je anne hit Cedar Key directly and brought a high precipitation into the area. A total of 372.5 mm fell in Cedar Ke y during September (data from http://www.AccuWeather.com). This is 217.6 mm over the monthly averag e precipitation (154.9 mm). This amount of precipitation inundated wetla nds, scrubby flatwoods, and sand pine scrub. All sites were partially flooded. The approximate percentage of the grids covered by water was as follows: 5C = 50%, 2M = 60%, 5A = 50%, and 5D = 30%. All sites stayed flooded for at least 2 weeks. Could small mammals survive this amo unt of precipitation? Did they move to upper grounds? Mice only had two options, move to upper gr ound or die. All sites had upper grounds inside the grids; therefore, the 3rd trapping session already sample d upper grounds in all of them. The extra effort involved the upper ground in/near the grids. However, no mice were captured during this time. The soil in scrubby flatwoods an d sand pine scrubs could not absorb the amount of precipitation that fell in Cedar Key during th e impact of the hurricanes. Thus, a high portion of the preserve was flooded, and this condition strongly impacted the small mammal community. 185

PAGE 186

Most likely, 79 terrestrial and arboreal mice marked during the fi rst two trapping sessions in treatment and control sites died. I do not think that mice had time to mo ve to other upper ground areas not covered by traps. Flooding negatively affected the small mammal community independently of their life form. Even though the effect of flooding on the survival probabilities of P. floridanus and S. hispidus was not conclusive, the effect of flooding upon terres trial species such as P. floridanus S. hispidus and P. gossypinus was expected to be detrimenta l to their populations. Arboreal species such as O. nuttalli should have a better possibility to to lerate this type of disturbance, but it did not because flooding lasted at least 2 weeks in all sites. No previous survival analysis of the effect of flooding on small mammals was fo und, but some references support the negative effect of a long period of flooding. No detrimental effect upon the population of P. gossypinus and O. nuttalli in Texas was recorded when flooding o ccurred up to 8 days. Flooding for a 3week period caused a marked decrease in th e populations. This probably happened because individuals tended to remain w ithin established home range ev en during long periods of flooding (McCarley 1959). Peromyscus leucopus (white-footed mice) completely disappeared from floodplain plots after severe fl ooding (Blair 1939, Turner 1966). P. leucopus, Microtus montanus (mountain vole), and Dipodomys ordii (kangaroo rat) generally e xperience habita t inundation as catastrophic (Andersen et al. 2000). Owls hunted snakes during day light hours afte r the impact of the hurricanes. I observed three owls hunting near wetlands between 3 pm and 4 pm in 5C and 5D. Probably, the species was Bubo virginianus (great horned owls) because of the big size and ear tufts. Two of them already caught a snake. This species is mainly nocturnal and eats rodents. Even though these observations are not enough to dr ive a conclusive statement, ma ybe owls were hunting snakes 186

PAGE 187

during light hours because of the reduction of small mammal populations due to the impact of the hurricanes. Population Responses during and after Prescribed Burning Small mammals hide in burrows to survive and respond to change in cover and food caused by prescribed burning. Each species, independent of the life form, must retreat to burrows to survive the passage of the fire front, and it is likely to respond differe ntly to prescribe burning and subsequent habitat changes. Some species increase in population and others decrease in populations after prescribed burning. Other spec ies just disappear from burned areas. Some species avoid recent burns until habitat requi rements removed by prescribed burning are restored. These aspects are discussed in more detail in the following subtitles. Burrows as refugia during prescribed burning Probably, the stand-replacing prescribed fire did not cause mortality of the small mammal community in CKSSR because mice hid in burrows. Fire intensity should be equal in treatment sites from the experimental point of view. However, this aspect was not as important as expected because fire intensity was high enough in both sites to remove all above ground vegetation. No evidence of mice mortality caused by prescribed burning was found after a careful search of both sites, and this result might be a consequence of the behavior of the species during prescribed burning. P. floridanus and P. gossypinus are nocturnal species that live in burrows and have to stay in them to survive the passage of fire. S. hispidus is a diurnal-noctur nal species, and if individuals are active during the day, they have to move and to hide in burrows or to flee if their home range is burned. O. nuttalli is arboreal and nocturnal species that live in above ground nests and would have to seek refuge in burrows to survive. There were a total of 28 burrows in each treatment site. Out of 33 individuals of P. floridanus S. hispidus and P. gossypinus marked during the 4th trapping session, 26 individuals were r ecaptured in wetlands after prescribed 187

PAGE 188

burning during the 5th trapping session. Only one individual of O. nuttalli was marked during the 4th trapping session, and it was recaptured in we tlands after prescribed burning in the 5th trapping session. Probably this individual also hid in a burrow because otherwise it would not survive. Burrows increase survivorship because of the insu lating characteristic of the soil. Some studies have shown that temperatures higher than 100 C at the surface of the ground decline in the first 2.5 cm of soil depth to temperatures around 20-30 C in longleaf pine in south-eastern USA (Heyward 1938), in Australia eucalypt forest (B eadle 1940), in California Chaparral (DeBano et al. 1998), and in heavy slash fuels after loggi ng a forest (Neal et al 1965; cited by Whelan 1995). One reason for poor penetration of heat is that convective heat is transferred upward. Burrows as refugia for small mammals during pres cribed fire have been documented in the literature. Small mammals can survive fires by remaining in their burrows (Tester 1965, Beck and Vogl 1972; Hendlund and Rickard 1981; Sm ith 2000). Most species look for refugia underground, where ventilation inside burrows is vital for animal su rvival (Bendell 1974). In this regard, burrows with more than one entrance might be better ventilated than those with one entrance (Geluso et al. 1986). No mortality of P. floridanus was found in two sites in OrdwaySwisher Preserve possibly because of the uneven distribution of litter and bare patches of sands caused a mosaic effect of varying intensities, and mice were protected in the tortoise burrows (Jones 1990). However, Jones also reported 87% re turn rate in one site, and probably, some mortality occurred after prescr ibed fire in this site. Burrowing rodents, such as Dipodomys survived in substantial numbers after a stand-re placing fire in California chaparral because their burrows protected them from heat (Quinn 1979) Populations of Townsends ground squirrels living in burrows were unaffected by stand-repl acing fire in sagebrus h-grass community in southeastern Washingt on (Hendlund and Rickard 1981). Regard ing arboreal species, woodrats 188

PAGE 189

usually suffered relatively high mortality becau se their nests were above ground (Simons 1991). However, populations of woodrats were unexpected ly high in burned areas because burns left patches of lightly burned vegetation in California chaparral and coastal sage scrub, which may have provided refugia for woodrat populations (Schwilk and Keeley 1998). In summary, during fire, the majority of the species seek ref ugia underground or in sheltered places above the ground. Prescribed burning effect The lack of a negative prescribed burning eff ect found in this study ha s also been cited in the literature. Even thoug h three studies have reported surv ival analysis by using informationtheoretic model selection and inference fram ework through the program MARK, two of them indicated that prescribed burning did not have a ny effect on the species involved. On all plots, independent of shrub density or burn treatment, the abundance of Neotoma fuscipes (duskyfooted woodrats) increased from 1993 to a peak in 1997, and decreased from fall 1997 to fall 2001 after prescribed fire in California oak woodla nds. Apparently, juvenile survival appeared to be the cause of the population fluctu ation in this species. Prescribe fire by itself did not have any support in the data (Lee and Tietje 2005). In a single prescribed fi re in old growth mixed conifer forest in Sequoia National Park, California, wh ere fire had been suppressed for over a century, year effects had greater influences on P. maniculatus densities, P. maniculatus age ratios, Neotomias speciosus (lodgepole chipmunk) densities, and total small mammal biomass than did prescribed fire effects. Fire by itself had less than 0.01% support in the data (Monroe and Converse 2006). In ponderosa pine in Coconi no National Forest, Ariz ona, forest thinning increased densities of P. maniculatus Tamias cinereicollis (gray-collared chipmunks), Spermophilus lateralis (golden-mantled gr ound squirrels), and Neotoma mexicana (Mexican woodrats), but the combination of thinning and fr equent prescribed fire might have reduced 189

PAGE 190

small mammal densities (Converse et al. 2006). Prescribed burning was not applied as a treatment by itself. Population increases/decreases after prescribed burning One important aspect to take into consider ation for the experimental design is the homogeneity of treatment and control sites re garding soils, slopes, plant and small mammal species composition, vegetation st ructure, and fire history. Pa rticularly, the normal seasonal variation in numbers of individuals before prescribed burning should be similar in treatment and control sites in order to detect prescribed burning effects. A hi gh proportion of the studies have this problem. The current study attempted to determine this aspect by trapping four times before and after prescribed burning. The interval of time before and after was 13 and 15 months, respectively. However, the flooding effect cause d by the hurricanes did not allow determining the normal seasonal variation in experimental site s before burning. This natural disturbance was recorded in conjunction with pr escribed burning, which makes the current study unique in this sense. The effect of flooding previously presente d was different from the effect of prescribed burning. There was an increase in the number of individuals of P. floridanus and S. hispidus after prescribed burning. The increase in the number of individuals of P. floridanus and S. hispidus was clearly identified when I compared treatment and control sites (Figur es 4-2 and 4-3). There was no doubt that prescribed burning forced th ese individuals to m ove to the vegetation surrounding wetlands. However, P. gossypinus and the only individual of O. nuttalli did the same, but the number of individu als did not increase through time. So, prescribed burning was the stimulus to move, but the increase in numbe r of individuals in the vegetation surrounding wetlands had to deal with other factors such as immigration, food/ space availability, and competition. This dissertation only has data for the first factor. The increase in the number of 190

PAGE 191

individuals was due to new adult individuals No juvenile was captured between the 5th and the 7th trapping session. So, apparently, the burned area attracted thes e mice but they had to seek refuge in the vegetation surrounding wetlands becau se of the lack of cover and food in the burned area. The lack of increase in th e number of individuals of P. gossypinus was surprising, but the drastic decline in the number of individuals of O. nuttalli was expected. The majority of the studies have shown a positive response of the genus Peromyscus to prescribed burning (Jones 1992). It was astonishing to see th at the number of individuals of P. gossypinus that maintained relatively stable low population levels in trea tment and control sites (Figures 4-2 and 4-3). Maybe, competition for food and space did not allow new S. hispidus to establish in the vegetation surrounding wetlands. However, the decline in the number of individuals of O. nuttalli was predictable because of it s arboreal life form and the combined effect of flooding and prescribed burning. Both curves in Figures 4-2 and 4-3 look alike. O. nuttalli appears to be susceptible to prescribed burning even though it might hide in burrows to survive the path of fire. The increase, decrease, or no change in the number of individuals af ter prescribed burning has been reported by other studies conducted in Florida. P. floridanus and P. polionotus increased population after prescribed fire while S. hispidus declined. However, no change in the composition of the populations of P. floridanus and S. hispidus was recorded before and after burning in longleaf/turkey oak ha bitat in north-central Florid a (Arata 1959). Densities of P. gossypinus and S. hispidus appeared to be similar in the burned and unburned hardwood forest in north Florida (Vogl 1973). Three populations of P. floridanus had little or no mortality due to prescribed fire, and populations were highe r on burned areas than on unburned sites in longleaf/turkey oak habitat in Ordway-S wisher Preserve. This means that P. floridanus did not 191

PAGE 192

move from the burned area after prescribed burni ng. Numbers of mice were more constant in the burn site than in the unburned one. Significan tly more mice were caught in burned areas immediately following prescribed fire. Also, sign ificantly more mice were caught on burned sites than on unburned burrows (Jones 1989, 1990). P. floridanus populations appeared to have little or no short term effects following prescribed burning, and abundance e qualed or exceeded prefire levels after two or three months (Jones 1992). P. gossypinus and S. hispidus had an increase in the number of individuals after a wildfire in slash/longleaf pine habitat in north-central Florida (Layne 1974). Layne also suggested that burned areas could act as d ispersal sinks. This statement might be true in some ecosystems, but not in others. In the current study in CKSSR, four P. floridanus and two P. gossypinus were recaptured only in the scrubs after prescribed burning. Layne (1990) also carried out a long-term monitoring of P. floridanus population in CKSSR. He reported that the species survived a wildfire in 1955, popula tion declined 10 years after fire, and the species was still present in 1986. Another study conducted by Layne (1990) at Archbold Biological Station revealed that P. floridanus was present at low numbers in scrub sites that were burned in 1927. In contrast, populations were higher and more stable in similar nearby habitats that were burned periodi cally. According to Layne (1992), P. floridanus populations are higher in early successional stages of scrub vegetation following fire. Particularly, high population numbers can be found in 2 years old scrub (Layne, pe rsonal communicatio n). Later in the absence of fire, populations decline as habi tat structure becomes denser and microclimatic conditions more mesic. However, Morgan (19 98) reported a high capture success between 1997 and 1998 for P. floridanus S. hispidus P. gossypinus and O. nuttalli in CKSSR, and the current study found a low capture success between 2004 and 2006. This change in population size can be explained by natural population fluctuations and probably the high population abundance found 192

PAGE 193

by Morgan after 42 years might be related with mast production. Depue (2005), in central Florida, found that P. floridanus increased or recovered pre-burn levels within 6 months following prescribed burning in Bullfrog Creek Mitigation Park; it dropped in numbers following prescribed fire, but started to increase when the study ended in Split Oak Mitigation Park; and the decrease in animal numbers remained unaffect ed by prescribed fire in Chuluota Wilderness Area. Apparently, there is not a cut-clear patt ern by a single species in Florida, and some variation in the response to prescribed burning might occur within the same region. Therefore, the same variation might be expected in other studies conducte d in other states. Some species increase population and others decrease populations after prescribed fire in other studies conducted in the United States. P. maniculatus had a post-fire increase in population size in California chap arral (Cook 1959) and in north eastern Minnesota (Ahlgren 1966), and these incrases were probably related to the increase of seeds of annual grasses stimulated by fire. Prescribed burning reduced the population of small mammal species with the exception of P. maniculatus in north-central Pennsylvania. The P. maniculatus established in the burned area one month following prescribed burning (Fala 1975). Populations of P. maniculatus generally increase afte r fire (Ream 1981). P. maniculatus were more abundant on 1-2-year-old burns in tallgrass prairie than in unburned areas in eastern Kansas. However, Reithrodontomys megalotis (western harvest mouse) was more abundant in unburned areas (Kaufman et al. 1982). Also, Kaufman et al. (1983) reported that R. megalotis densities in the burn site increased the following spring and summer because the population in the un-burned sites served as a source of dispersing individuals. P. maniculatus dramatically increased populat ion size one year following prescribed burning; however, this effect disappear ed and reversed itself during the second year in Ponderosa Pine, South Dakota (Bock and Bock 1983). The total number of small mammals was 193

PAGE 194

lower in burned sites than in unburned ones in shrub-steppe habitat in Idaho. Nonetheless, P. maniculatus were more abundant in the burn site (Groves and Steenhof 1988). A dramatic increase in P. maniculatus population on burn sites following pres cribed fire in British Columbia was explained by the rodents ability to forage fo r seeds and insects that were greatly increased. In contrast, southern red-back ed vole numbers were decreased for 2-3 years following burning (Sullivan and Boateng 1996). The studies cited above found stronger support for positive prescribed fire impacts on P. maniculatus abundances than reported by Monroe and Converse (2006) in Sierra Nevada mixed conifer forest, California. These authors said that the tendency for P. maniculatus densities to be greater on burned areas does not necessarily indicate that burned habitat is optimal for P. maniculatus and may reflect dispersal to marginal sink habitat. In California oak woodland, the population of N. fuscipes increased from 1993 to 1997, and then decreased steadily after prescribed burning (Lee and Tietje 2005). P. leucopus increased population size in treatment sites (thinning, prescribed fire, and thinning + prescribed burning), but this increase was higher when the two types of treatments were combined in the southern Appalachian hardwood forest in North Carolina (Greenberg et al. 2006). A comparison between three sites (brush, prairie, and savanna) burned frequently during 15 years and three forested sites unburned during 35 years in western Wisconsin we re carried out to study small mammals abundance in these sites. P. leucopus and Clethrionomys gapperi (red-backed vole ) were more common in the unburned forest, and P. maniculatus and Spermophilus tridecemlineatus (13lined ground squirrel ) were more abundant in the prairie cr eated and maintained by fire. Burning the forest did not significantly reduce the number of mice presen t (Beck and Vogl 1972). In the first year after burning in the California grassland and chapar ral, populations of Chaetodipus californicus (California pocket mouse), Peromyscus californicus (California mouse), and 194

PAGE 195

Dipodomys agilis (agile kangaroo rat) we re either unchanged or greater on burned than in unburned sites. However, populations of R. humulis Peromyscus boylii (brush mouse), and Neotoma spp. (woodrat) decreased or disappeared in the burned sites (Wirtz 1977). Populations of S. hispidus in Arizona grassland were greatly redu ced by a summer fire, while populations of Perognathus hispidus (seed-eating pocket mice) and Dipodomys merriami (kangaroo rats) increased. This difference was explained by the food habits of the species. S. hispidus fed on green vegetation that decreased af ter fire. The heteromyid rodents fed on seeds, which increased after fire due to the invasi on of weedy forbs (Bock and Bock 1978). Even though there was no evidence that prescribed fire killed any small mammal, it negatively impacted Mus musculus (house mice), P. maniculatus, and Microtus pennsylvanicus (meadow voles) in Oxford, Ohio. M. musculus and M. pennsylvanicus disappeared from the burned site, while the three species prevailed in the contro l site (Crowned and Ba rret 1979). Populations of Spermophilus spp. (ground squirrels) and Thonomys spp. (pocket gophers) generally increase after fire (Ream 1981). Only Dipodomy agilis out of five species of rode nts increased in abundance after prescribed burning in a coastal sage scrub in southern California (Prise and Waser 1984). Six small mammal species were not eliminated, and th ey did not increase in numbers in the months following prescribed burning in the California Chaparral (Lawrence 1966). A year after a prescribed burning in conifer woodland with shrubby understory in California, the abundance of small mammals was almost three times greater on unburned than burned sites, even though species composition did not vary significantly between burned and unburned sites (Blankenship 1982). The populations of the small mammal comm unity (11 species) in the unburned sagebrush in Burro Hill, Wyoming, varied little before burning, the populations were at low level following burning, and populations approached control valu es three years after burning (McGee 1982). The 195

PAGE 196

impact of fire on small mammal communities in the central Appalachians on Pennsylvania was transitory, and the differences in small ma mmal abundance between unburned and burned sites disappeared within eight months af ter fire (Kirkland et al. 1996). Also, there were not significant differences in small mammals mean total captu red efficiency between treatment and control sites in southern Appalachian, North Carolina (F ord et al. 1999). Reduction of shrubs and woody debris by thinning with overly frequent pres cribed fire may reduce small mammal densities (Converse et al. 2006). In summary, small mammal species responses to prescribed burning vary greatly within the same geographic area and among states in the United States. Habitat selection: immigration, emig ration, and returning to burned areas In general, prescribed burning affects sm all mammals mainly through the way it affects their habitats. Direct effect s such as injury, mortality, and movement (immigration and emigration) might be the short-term populati on responses. Indirect effects through habitat alteration could influence long-term responses such as feeding, move ment, reproduction, and availability of refugia (Smith 2000). In both ci rcumstances, immigration and emigration play an important role in population demography, food ava ilability, reproduction, a nd re-colonization of the burned areas. Immigration might occur be cause burned areas attract small mammals; however, emigration could also take place because there is insufficient food and cover in the burned area. Characteristics of an animal species such as mobility and particular food and cover requirements will determine its ability to re-i nvade a burned site (Whelan 1995). The length of time before these species return to burned site s depends on how much fi re altered the habitat structure and food supply (Smith 2000). The last three sentences summarize and explain what P. floridanus and S. hispidus experienced at CKSSR. Although I did not trap new individuals in burned sites following prescribed burning, there were data that showed possi ble immigration to the burned area. A total of 27 new adult 196

PAGE 197

individuals were tr apped between the 5th and the 7th trapping sessions in wetlands corresponding to 3 P. floridanus 20 S. hispidus 3 P. gossypinus and one O. nuttalli These individuals were probably attracted by the burned area and moved to wetlands looking for ref ugia. Immigration to burned areas immediately after prescribed burning has been cited in the literature. Odors from burned areas might stim ulate immigration of P. maniculatus from suboptimal habitats (Kaufman et al. 1988b). The high reproductive potential of P. maniculatus and R. megalotis populations in Kansas tallgrass prairie enables them to increase rapidly in favorable envi ronments and disperse readily into recently burned areas (Kaufman et al. 1988b). The number of resident individuals of P. floridanus in burned areas of Ordway-Swisher Preser ve was higher than in the unburned ones. P. maniculatus invaded a burned area after a heavy ra in in California (Tevis 1956). Also, P. maniculatus invaded burned areas immediately after prescr ibed fire in jack pine in northeastern Minnesota (Ahlgreen 1966). The lack of prescribed burning effect on P. floridanus and S. hispidus was probably because they used the vegetation surrounding wetland s as refugia. Statistically, I could not draw any conclusion, but practically mice may have move d to wetlands looking for cover and food. If mice did not have the wetlands near the treatment sites, probably they have had to move farther until finding another wetland or unburned site. P. floridanus in the current study moved out of burned areas. In contrast, Jones (1990) reported that all individuals of P. floridanus except one did not leave burned areas after prescribed burning in Ordway-Swisher Preserve. This is probably due a patchy burn in the Ordway sandh ills in comparison with a more continuous burn in the CKSSR scrub. A similar result was found by Arata (1959) near Gainesville, Florida. The number of captured individuals of P. floridanus and P. polionotus remained at pre-burn levels in burned and unburned sites, whereas S. hispidus moved from the burned to the un-burned site 197

PAGE 198

following prescribed fire. However, Ream (1981) reported that red squirrel, voles, northern flying squirrel, and showshoe ha re emigrated from recent burned areas. Of 25 species common in chaparral brushlands, Townsends chipmunk and dusly-footed woodrat were not abundant in recently burned areas (Biswell 1989). This suggest that species such as P. floridanus found in the same geographic area might respond differently to prescribed burning depending on the habitat type and how much fire altered it. Also, other species always em igrate from burned sites, and most of the time they return to the same burned sites they moved away from months ago. P. floridanus and S. hispidus returned to the burned sites in CKSSR after at least 11 months. This amount of time appears to be a long time for a mouse to live and survive. Out of 50 P. floridanus (20 individuals) and S. hispidus s (30 individuals) ma rked between the 5th and the 7th trapping sessions, 33 ( 66%) individuals (nine P. floridanus and 24 S. hispidus s) were recaptured again in scrub in the 8th trapping session. The survival curve for both species was high (Figures 4-6 and 4-8), and the amount of time involved between the 5th and the 8th trapping session was 373 days for 5C and 403 days for 2M. Are P. floridanus and S. hispidus long lived species? Jones (1990), at Smith Lake sandhill in Ordway Swisher Preser ve, found that 8.6% of all marked mice (225 individuals) were present fo r 360 days or more. Of these mice, half were females first marked as juveniles, and most of the males were first marked as subadults and adults. The longevity records were 649 days fo r females and 920 days for males. Layne (1974) reported two S. hispidus females originally trapped as subadul ts were recorded on the study area during the entire 14-month period. A juvenile female and an adult and a juvenile male were first captured in October 1960 and they were re captured 10 months later in July 1961. P. floridanus and S. hispidus in CKSSR returned to the burn ed sites after they found cover and food to live there. It is surprising that both species returned approximately at least 11 months 198

PAGE 199

later. But, it is not as surpri sing when we consider the phenology of the plant species. Accorns of Myrtle oak, Chapman oak, and sand live oak deve loped surrounding wetlands in September, and production ended in December 2005. Thus, P. floridanus and S. hispidus had food to stay in the vegetation surrounding wetlands. In the burned sites 5C and 2M, the percentage of shrub cover at 11 months after prescribed burning was not qua ntified, but at 12 months was 72 % and 105%, respectively. This percent c over was mainly composed of Quercus myrtifolia, Quercus geminata, Lyonia ferruginea, Lyonia lucida, and Quercus chapmanii (see Tables 3-7 and 3-11). Neither flowering season nor fruiting se ason was synchronous in CKSSR within and among species, but Vaccinium myrsinites developed flowers in March and fruits in April 2006. Gaylussacia nana and Serenoa repens developed flowers in April, but G. nana had fruits in May and Serenoa repens in June 2006. Other species such as L. ferruginea, L. lucida, Ilex glabra and Brevaria racemosa flowered in April-May and fru ited in June-July 2006. Therefore, P. floridanus and S. hispidus returned to scrubs after plan t species offered cover and food. The relationship between the amount of cover and mice returning to burned areas has been reported in the literature. Arata (1959) indicated that S. hispidus did not return to longleaf/turkey oak habitat in north-central Flor ida at 5 months after a burn. Howe ver, he stated that the burned area was recolonized by S. hispidus within 6 months following the burn. Layne (1974) was the first study in Florida to re port that the return of S. hispidus to a burned area appeared to be correlated with redevelopment of the ground cover in slash/longleaf pine ha bitat in north-central Florida. Ahlgren (1966) showed that the southern red-backed vole numbers decreased for 2-3 years in Minnesota following prescribed burning until recovery in the ground story vegetation occurred. McGee (1982) reported that the popula tions of the small mammal community (11 species) in the unburned sagebrush in Burro H ill, Wyoming, varied little before burning, the 199

PAGE 200

200 populations were at low level following burning, and populations approached control values three years after burning when to tal cover of the unde rstory was near unburned levels. West (1982) indicated that northern red-backed voles avoided a burned area in black spruce for one year and established a resident population in post fire year 4, which it was the first year of berry production in central Alaska. Geluso et al. (1986) found that voles survived a prescribed burning in Nebraska grassland and left the burned areas until a new litter layer had accumulated about two growing seasons later. Possible reasons for emigration included decreased protection from predators and decreased food availa bility. Kirkland et al. (1996) showed that the impact of fire on small mammal communities in the central Appa lachians on Pennsylvani a was transitory, and the differences in small mammal abundance be tween unburned and burned sites disappeared within eight months after fire This rapid recovery of small mammal populations was explained by the fast re-growth of ground cover within the study area, particularly of blueberry. Sullivan and Boateng (1996) reported that southern red-backed vole numb ers decreased for 2-3 years in British Columbia following prescribed burning until recovery in the ground story vegetation took place. Schwilk and Keeley (1998) carried out a pa tchy burn in a California chaparral and coastal sage scrub. These refugia allow small mammals colonize severely burned sites during the first six months after prescribed fire. Ford et al. (1999) also used the link between small mammals and re-growth of the vegetation as the explan ation of the population recovery in the study conducted in Southern Appal achian, North Carolina. In summary, immediately after fire, some species are attracted to burned areas and immigrate into them. Other species emigrate due to insuficient food and cover in the burned area. The length of time before species return depends on how much fire altered the habitat structure and food supply.

PAGE 201

Table 4-1. Number of captur ed individuals in the scrubby fl atwoods and in the vegetation su rrounding wetlands per trapping se ssion in treatment and control sites in Cedar Key Scrub State Reserve. Trapping Podomys Peromyscus Ochrotomys Sigmodon Total Session S W T S W T S W T S W T S W T Treatment 1 4 3 7 2 1 3 7 1 8 1 0 1 14 5 19 2 3 3 6 3 0 3 7 2 9 1 1 2 14 6 20 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 9 4 13 5 2 7 1 0 1 11 2 13 26 8 34 5 0 16 16 0 4 4 0 1 1 0 24 24 0 45 45 6 1 15 16 1 6 7 0 0 0 0 23 23 2 44 46 7 3 13 16 1 6 7 0 1 1 0 26 26 4 46 50 8 21 0 21 2 2 4 0 0 0 30 4 34 53 6 59 Total 95 Total 35 Total 20 Total 123 113 160 273 Control 1 7 3 10 0 0 0 11 0 11 1 0 1 19 3 22 2 4 4 8 0 0 0 8 0 8 2 0 2 14 4 18 3 8 1 9 0 0 0 0 0 0 1 0 1 9 1 10 4 8 0 8 5 1 6 0 0 0 2 0 2 15 1 16 5 6 0 6 6 0 6 1 0 1 9 0 9 22 0 22 6 6 0 6 4 1 5 1 0 1 9 0 9 20 1 21 7 7 0 7 3 0 3 0 0 0 11 0 11 21 0 21 8 6 0 6 4 0 4 0 0 0 9 4 13 19 4 23 Total 60 Total 24 Total 21 Total 48 139 14 153 201 Codes: S = scrubs. W = wetlands. T = Total.

PAGE 202

Table 4-2. The Goodness of Fit test for the general model phi(g*t) p(g*t) for Podomys floridanus and Sigmodon hispidus Species Bootstrap GOF Method Observe Mean c-hat GOF Method Chi-square df c-hat Podomys Deviance method 12.5835 11.0510 1.1387 Release 0.9167 4 0.2292 c-hat method 0.0000 0.0000 0.0000 Sigmodon Deviance method 24.4433 15.5750 1.5694 Release 5.6026 9 0.6225 c-hat method 0.0000 0.0000 0.0000 202

PAGE 203

Table 4-3. Result browser of the fitted candidate model set of 16 models for Podomys floridanus after adjusting to c-hat = 1.1387 in treatment and control sites in Cedar Key Scrub State Reserve. Delta QAICc Model Models QAICc QAICc weight likelihood #Par QDeviance {Phi(t) p(.) PIM} 123.822 0.00 0.73160 1.0000 8 18.930 {Phi(t) p(g) PIM} 125.949 2.13 0.25265 0.3453 9 18.741 {Phi(t) p(t) PIM} 131.653 7.83 0.01458 0.0199 13 14.778 {Phi(g*t) p(.) PIM} 137.206 13.38 0.00091 0.0012 15 15.239 {Phi(g*t) p(g) PIM} 139.728 15.91 0.00026 0.0004 16 15.145 {Phi(g*t) p(t) PIM} 149.739 25.92 0.00000 0.0000 21 11.340 {Phi(t) p(g*t) PIM} 150.755 26.93 0.00000 0.0000 21 12.356 {Phi(.) p(t) PIM} 153.549 29.73 0.00000 0.0000 8 48.657 {Phi(g) p(t) PIM} 153.681 29.86 0.00000 0.0000 9 46.474 {Phi(.) p(.) PIM} 158.311 34.49 0.00000 0.0000 2 66.533 {Phi(g) p(.) PIM} 158.629 34.81 0.00000 0.0000 3 64.753 {Phi(.) p(g) PIM} 160.259 36.44 0.00000 0.0000 3 66.383 {Phi(g) p(g) PIM} 160.716 36.89 0.00000 0.0000 4 64.708 {Phi(g) p(g*t) PIM} 165.704 41.88 0.00000 0.0000 16 41.121 {Phi(.) p(g*t) PIM} 166.047 42.23 0.00000 0.0000 15 44.079 {Phi(g*t) p(g*t) PIM} 171.137 47.31 0.00000 0.0000 28 11.051 203

PAGE 204

Table 4-4. Set of 25 models afte r adding phi(Flood + Fire) in comb ination with p constant (.), time dependent (t), and group dependent (g) in the analysis Flood and Fire effect on survival probabilities of Podomys floridanus in treatment and c ontrol sites in Cedar Key Scrub State Reserve. Delta QAICc Model Models QAICc QAICc weight likelihood #Par QDeviance {Phi(Flood+Fire) p(.)} 121.892 0.00 0.29916 1.0000 4 25.885 {Phi(Flood+Fire) p(t)} 122.950 1.06 0.17630 0.5893 10 13.387 {Phi(Flood) p(t)} 123.393 1.50 0.14128 0.4723 9 16.185 {Phi(Flood+Fire) p(g)} 123.668 1.78 0.12308 0.4114 5 25.494 {Phi(t) p(.) PIM} 123.822 1.93 0.11397 0.3810 8 18.930 {Phi(Flood) p(.)} 124.720 2.83 0.07274 0.2432 3 30.845 {Phi(t) p(g) PIM} 125.949 4.06 0.03936 0.1316 9 18.741 {Phi(Flood) p(g)} 126.384 4.49 0.03166 0.1058 4 30.377 {Phi(t) p(t) PIM} 131.653 9.76 0.00227 0.0076 13 14.778 {Phi(g*t) p(.) PIM} 137.206 15.31 0.00014 0.0005 15 15.239 {Phi(g*t) p(g) PIM} 139.728 17.84 0.00004 0.0001 16 15.145 {Phi(Fire) p(t)} 148.336 26.44 0.00000 0.0000 9 41.129 {Phi(.) p(g*t) PIM} 148.972 27.08 0.00000 0.0000 8 44.079 {Phi(g*t) p(t) PIM} 149.739 27.85 0.00000 0.0000 21 11.340 {Phi(t) p(g*t) PIM} 150.755 28.86 0.00000 0.0000 21 12.356 {Phi(Fire) p(.)} 151.999 30.11 0.00000 0.0000 3 58.123 {Phi(.) p(t) PIM} 153.549 31.66 0.00000 0.0000 8 48.657 {Phi(g) p(t) PIM} 153.681 31.79 0.00000 0.0000 9 46.474 {Phi(Fire) p(g)} 153.973 32.08 0.00000 0.0000 4 57.966 {Phi(.) p(.) PIM} 158.311 36.42 0.00000 0.0000 2 66.533 {Phi(g) p(.) PIM} 158.629 36.74 0.00000 0.0000 3 64.753 {Phi(.) p(g) PIM} 160.259 38.37 0.00000 0.0000 3 66.383 {Phi(g) p(g) PIM} 160.716 38.82 0.00000 0.0000 4 64.708 {Phi(g) p(g*t) PIM} 165.704 43.81 0.00000 0.0000 16 41.121 {Phi(g*t) p(g*t) PIM} 171.137 49.24 0.00000 0.0000 28 11.051 204

PAGE 205

Table 4-5. Estimated survival (phi) and recapture (p) parameters by using model phi(Flood+ Fire) p(.) in the analysis Flood and Fire d effect on survival probabilities of Podomys floridanus in treatment and control sites in Cedar Key Scrub State Reserve. Confidence interval = 95%. Phi Estimate Standard error Lower CI Upper CI 1:Phi 0.7870664 0.0507500 0.6712425 0.8699882 2:Phi 0.0000000 0.0000000 0.0000000 0.0000000 3:Phi 0.7870664 0.0507500 0.6712425 0.8699882 4:Phi Treatment 1.0000000 0.0000000 1.0000000 1.0000000 5:Phi 0.7870664 0.0507500 0.6712425 0.8699882 6:Phi 0.7870664 0.0507500 0.6712425 0.8699882 7:Phi 0.7870664 0.0507500 0.6712425 0.8699882 8:Phi 0.7870664 0.0507500 0.6712425 0.8699882 9:Phi 0.0000000 0.0000000 0.0000000 0.0000000 10:Phi 0.7870664 0.0507500 0.6712425 0.8699882 11:Phi Control 0.7870664 0.0507500 0.6712425 0.8699882 12:Phi 0.7870664 0.0507500 0.6712425 0.8699882 13:Phi 0.7870664 0.0507500 0.6712425 0.8699882 14:Phi 0.7870664 0.0507500 0.6712425 0.8699882 15:p Recapture 0.9430430 0.0398357 0.7946523 0.9860803 205

PAGE 206

Table 4-6. Set of 28 models af ter adding phi(Flood + Fire) in co mbination with p(Flood + Fire), p(Flood), and p(Fire) in the analysis Flood and Fire effect on survival probabilities of Podomys floridanus in treatment and control sites in Cedar Key Scrub State Reserve. Delta QAICc Model Models QAICc QAICc weight likelihood #Par QDeviance {Phi(Flood+Fire) p(.)} 121.892 0.00 0.24033 1.0000 4 25.885 {Phi(Flood+Fire) p(t)} 122.950 1.06 0.14164 0.5894 10 13.387 {Phi(Flood) p(t)} 123.393 1.50 0.11350 0.4723 9 16.185 {Phi(Flood+Fire) p(g)} 123.668 1.78 0.09888 0.4114 5 25.494 {Phi(t) p(.) PIM} 123.822 1.93 0.09156 0.3810 8 18.930 {Phi(Flood + Fire) p(Fire)} 123.935 2.04 0.08652 0.3600 5 25.761 {Phi(Flood + Fire) p(Flood)} 124.059 2.17 0.08135 0.3385 5 25.885 {Phi(Flood) p(.)} 124.720 2.83 0.05843 0.2431 3 30.845 {Phi(t) p(g) PIM} 125.949 4.06 0.03162 0.1316 9 18.741 {Phi(Flood + Fire) p(Flood + Fire)} 126.138 4.25 0.02877 0.1197 6 25.761 {Phi(Flood) p(g)} 126.384 4.49 0.02543 0.1058 4 30.377 {Phi(t) p(t) PIM} 131.653 9.76 0.00182 0.0076 13 14.778 {Phi(g*t) p(.) PIM} 137.206 15.31 0.00011 0.0005 15 15.239 {Phi(g*t) p(g) PIM} 139.728 17.84 0.00003 0.0001 16 15.145 {Phi(Fire) p(t)} 148.336 26.44 0.00000 0.0000 9 41.129 {Phi(.) p(g*t) PIM} 148.972 27.08 0.00000 0.0000 8 44.079 {Phi(g*t) p(t) PIM} 149.739 27.85 0.00000 0.0000 21 11.340 {Phi(t) p(g*t) PIM} 150.755 28.86 0.00000 0.0000 21 12.356 {Phi(Fire) p(.)} 151.999 30.11 0.00000 0.0000 3 58.123 {Phi(.) p(t) PIM} 153.549 31.66 0.00000 0.0000 8 48.657 {Phi(g) p(t) PIM} 153.681 31.79 0.00000 0.0000 9 46.474 {Phi(Fire) p(g)} 153.973 32.08 0.00000 0.0000 4 57.966 {Phi(.) p(.) PIM} 158.311 36.42 0.00000 0.0000 2 66.533 {Phi(g) p(.) PIM} 158.629 36.74 0.00000 0.0000 3 64.753 {Phi(.) p(g) PIM} 160.259 38.37 0.00000 0.0000 3 66.383 {Phi(g) p(g) PIM} 160.716 38.82 0.00000 0.0000 4 64.708 {Phi(g) p(g*t) PIM} 165.704 43.81 0.00000 0.0000 16 41.121 {Phi(g*t) p(g*t) PIM} 171.137 49.24 0.00000 0.0000 28 11.051 206

PAGE 207

Table 4-7. Estimated parameters from models phi(Flood+ Fire) p(.) in the analysis Flood and Fired effect on the survival probabilities of Podomys floridanus in treatment and control sites in Cedar Ke y Scrub State Reserve. Confidence interval = 95%. Model Parameter Standard error Lower CI Upper CI Interception 1.3073323 0.3028174 0.7138103 1.9008543 phi(Flood+Fire) p(.) Flood -22.5146010 0.0000000 -22.5146010 -22.5146010 Fire 30.5618730 0.0000000 30.5618730 30.5618730 p 2.8068159 0.7416407 1.3532002 4.2604316 207

PAGE 208

Table 4-8. Estimated survival parameters ( phi) by using model averaging for the set of 28 models in the analysis Flood and Fire eff ect on the survival probabilities of the Podomys floridanus in treatment and control sites in Cedar Key Scrub State Reserve. C-hat = 1.1387; 95% confidence interval. Phi Estimate Standard error Lower CI Upper CI 1:phi 0.7847947 16.2037846 0.0000000 1.0000000 2:phi 0.0000000 0.0000055 -0.0000108 0.0000108 Average 3:phi 0.8269939 20.3813127 0.0000000 1.0000000 Treatment 4:phi 0.9677876 0.0613967 0.3875942 0.9992993 5:phi 0.8251911 0.0687758 0.6496347 0.9231836 6:phi 0.8247459 0.0694662 0.6472239 0.9234956 7:phi 0.7939472 49.0490509 0.0000000 1.0000000 1:phi 0.7847475 16.2037852 0.0000000 1.0000000 2:phi 0.0000000 0.0000055 -0.0000108 0.0000108 Average 3:phi 0.8270833 0.0763053 0.6269967 0.9315558 Control 4:phi 0.8353632 0.0781662 0.6248433 0.9