<%BANNER%>

The Relative Value of Seagrass, Marsh Edge, and Oyster Habitats to the Brackish Grass Shrimp, Palaemonetes intermedius, ...

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 E20110113_AAAAEB INGEST_TIME 2011-01-13T21:43:48Z PACKAGE UFE0004787_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 4943 DFID F20110113_AACRCE ORIGIN DEPOSITOR PATH goodfriend_d_Page_28thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
843c58721ebcc5228be49878f164e717
SHA-1
96268a07cca42ff12bc17478d63f5bbe85073bdd
47099 F20110113_AACQVH goodfriend_d_Page_14.pro
c42553c980678a082bc221bd800ff178
34a7ff97eee595288a78b053a2c9f2aaad61bc9e
5580 F20110113_AACRCF goodfriend_d_Page_44thm.jpg
a9cb2b7c4317902688426a2353f2cc1d
4de68f83664d07d6b67390ca9b24bc4851ac915a
37234 F20110113_AACQQK goodfriend_d_Page_40.jpg
5de610b5e0c352236655450e85af12d5
f11e6e85f7a4bed0b3440cd9c96aeae25a20e6ac
37300 F20110113_AACQVI goodfriend_d_Page_15.pro
7efd25eb8d86e562783bae9474b14b50
2544887daf4c47d802568628957e7c5494c20b25
24273 F20110113_AACRCG goodfriend_d_Page_47.QC.jpg
b11f08a755e93d400bb8691afa0a6485
a41355513d825862b6364d9cba0e0507072be3bb
59967 F20110113_AACQQL goodfriend_d_Page_41.jpg
ea4f8a34afa3ec114d85fd1919791448
9e94891bddde10a5c93c29e6ccdabc70c67e1bc6
51654 F20110113_AACQVJ goodfriend_d_Page_16.pro
2e673f1b45bdd38c82f99a501246fed0
4568a310947524ba40c4226ab96b55aba52ef7e4
6615 F20110113_AACRCH goodfriend_d_Page_47thm.jpg
a9f57f86e580b542c040262d720f3457
7da99ac497844c733a762f9a7654129c7a45a910
66192 F20110113_AACQQM goodfriend_d_Page_42.jpg
3251a03d98949c2fea8ea36e39f6e7af
dcc922aaa85adab8c39cba68563bf4a8f37c6690
51122 F20110113_AACQVK goodfriend_d_Page_17.pro
f1f7290da4f111dc8481bacaff6d770a
4b68ad53a5199ff6b58bd9201301043596b94e69
7233 F20110113_AACRCI goodfriend_d_Page_56thm.jpg
e95740596b8269ad0d0a3f0f0f4d19ba
6062ecc23ed8778571ce07bef80f44438b1cb1a1
48894 F20110113_AACQVL goodfriend_d_Page_18.pro
cb13dcf5175a66f9d81ba741e50cf01a
c10c56e27c6d412d25641381ae241df0fa3f80ed
27746 F20110113_AACQQN goodfriend_d_Page_43.jpg
6427ee5727174e9cfe85dc7bec37f6c7
3dc263123b010abb809a4c72aaec1989c72f6a37
21952 F20110113_AACRCJ goodfriend_d_Page_57.QC.jpg
da9c32552f8eb44ade8b1a4cf9270952
de04f3aec6ec865afab47776fca871db5ca1a2aa
47468 F20110113_AACQVM goodfriend_d_Page_19.pro
a5c30aa38d16c3e525ee76f3cea7b6d8
f2a89cd8a8595deaea43dab8c8383f2830030fed
59434 F20110113_AACQQO goodfriend_d_Page_44.jpg
1cb6163b76898c922873006cba6f3918
ca00ba5db709cd627d7c6d8a64754d48acfef81c
48297 F20110113_AACQVN goodfriend_d_Page_20.pro
ad8899c08f045e854f567a97451f3508
caa4d7573b896f2ec17814bc26b658e6a1f7a293
68076 F20110113_AACQQP goodfriend_d_Page_45.jpg
c0f4d680add412564d6622a174136f74
1cf99a4d45f674ec96d43e93054c7843846def60
33286 F20110113_AACQVO goodfriend_d_Page_22.pro
6cec334bb67fb52528135db5504a89c8
a2fd17e6ad83d7607ad43fba6071944ac65f43eb
70689 F20110113_AACQQQ goodfriend_d_Page_46.jpg
25512c4f4f0e82cf05b5261ceef75e54
0e21a2c133addb1d99e4ba00cb5e2e1943539067
74497 F20110113_AACQQR goodfriend_d_Page_47.jpg
a4c81e1f870ae824768afb52cc2156b8
3d395ce0ec3cd72933393658ff372a3917cc2cba
41083 F20110113_AACQVP goodfriend_d_Page_23.pro
4e02054b3711a6ccde040c791497ae78
602cc3bf63acd9e8084a11ba96a11ed157ce92a4
74982 F20110113_AACQQS goodfriend_d_Page_48.jpg
418b339ec9e48d4beb788504e1808205
994b868ddbd01fd0f314bf01e670d6c4347916bb
49568 F20110113_AACQVQ goodfriend_d_Page_24.pro
be594973612f9e1ef7276a8e63ef6a61
5b0ba1081c89950d99aacad6eaed593ebbcc55e7
73256 F20110113_AACQQT goodfriend_d_Page_49.jpg
ff64b0d2449cf4e28a1e60128f3ae26d
dc91a36d773856879d01c24fb147a4bca8161ce2
51029 F20110113_AACQVR goodfriend_d_Page_25.pro
194e9496e1ac5ffa8ca9ef5c083ee118
ab4051da67a0e3c06592e7c2039bb306ea2227a3
70034 F20110113_AACQQU goodfriend_d_Page_50.jpg
ddb8bdc3f9c2af29eec6fab8ac0741e3
20e21bdaf96ef0a11c2f612d78abdf8ffe24d425
48776 F20110113_AACQVS goodfriend_d_Page_27.pro
8f8cb9a04261a76d69b27b9a44dc8937
d87fed85e3ec1ed212159283ef427717352d893b
70991 F20110113_AACQQV goodfriend_d_Page_51.jpg
4a012d88283898f86efe16de35964201
89feed938a19d69b9a4e97f3c3b0cf8093b3c6ac
33143 F20110113_AACQVT goodfriend_d_Page_28.pro
3af91c7b4b375a4c2efa23751c388b82
9a12927a77d2cb76e08beafb5e948a9cabfacaa6
71473 F20110113_AACQQW goodfriend_d_Page_52.jpg
a31b8aeeec52cf2225451ad6dc2bdb31
81cfae2f72a6b28bb326850d3c92ec4d1f9ac78a
19591 F20110113_AACQVU goodfriend_d_Page_29.pro
513f85583b698d4ea64000500db9d7c5
9f59901c839bc77f8ce45ac3a99a80ba8e8b3b71
69823 F20110113_AACQQX goodfriend_d_Page_53.jpg
164dc31c8a85b0eaadd59c7a33c477ac
cadd36845b1b7c6c2a10bfdff149c8168042a7db
11565 F20110113_AACQVV goodfriend_d_Page_30.pro
8db8fd033efc479b821c5240dad0c224
a0dec3148d3aa6f2e235b9bb5eaeeea445dd0bd9
18793 F20110113_AACQOA goodfriend_d_Page_06.pro
897c223dd3296f1761040da37db304e7
cd8a0c23c7059ae37bb3ad9e59dc62ad9023f548
86742 F20110113_AACQQY goodfriend_d_Page_54.jpg
bdec7f3ef24bcea7d8bdcc7f54e20a06
10dd02398549bbe43d35558db45c912a3289736c
5992 F20110113_AACQVW goodfriend_d_Page_31.pro
d18291995b88ccbe7735bd2e210e315d
7741f893328295e45eae481cf4232b5a77c5a55c
1053954 F20110113_AACQOB goodfriend_d_Page_39.tif
d0adbc4bebdd6e6539e42ccb813cd6ab
00800e797758ec277798be3332e1bff0919c957b
98247 F20110113_AACQQZ goodfriend_d_Page_55.jpg
570f5de90d47297010aac19aa18d1f7c
6fa296ba7b043c5e68f9bda723a3210497d64101
23453 F20110113_AACQVX goodfriend_d_Page_32.pro
7fa3f66aaa101ae5fabea4c55e4368c9
2919383dcc66431f6ccad86a593b9d6402b734e9
50529 F20110113_AACQOC goodfriend_d_Page_40.jp2
5139d20f2b596c9dd0b6f53914175505
53794b501306589001bcd139b88ba81fe33a00d6
97361 F20110113_AACQTA goodfriend_d_Page_58.jp2
5e0229cfd9dfa066a30daff61e2cd32b
12932b15d905e1f0a2d9c4b4333adb0c95a533f1
41053 F20110113_AACQVY goodfriend_d_Page_34.pro
8b7ad1de6d9869d69741017d3c3d5089
12ae4c1bd293e18d90e44fb9b6ae362eee9729c0
21600 F20110113_AACQOD goodfriend_d_Page_58.QC.jpg
eb3fb8de7cdb435f8289be369ee043d6
b4d75cb273406b819fd428610a6cd62543937897
F20110113_AACQTB goodfriend_d_Page_01.tif
26981f8a8fb2668c777eed25675203ff
74c21b40042279503179d7b4afcbcb4df747b8ea
37092 F20110113_AACQVZ goodfriend_d_Page_35.pro
1288742a372b13bb4ab5f9e35a9f184e
303c6fbb7e9168ebbeba9b20fcaa33d8f44ae7ec
32420 F20110113_AACQOE goodfriend_d_Page_06.jpg
58f64a6f8cc69db8f7a06339bb17257b
71c0f8af84e8f524bf073f0c5b32766744a74942
F20110113_AACQTC goodfriend_d_Page_02.tif
caa3c9053c39f697db39386281d73f4a
c23cc06872b9f57771de8682d3b624fe810fbbf5
1716 F20110113_AACQOF goodfriend_d_Page_02.txt
ed1bfd3f976cac9d4029ab9a70298f3f
d881fc83534babc133088104e95d56bd0c68623b
F20110113_AACQTD goodfriend_d_Page_03.tif
2adbcd6cb339b35a4efd2871412a679b
5fa299db231d57921ed0fe290a48d4e8194b1e22
5972 F20110113_AACQOG goodfriend_d_Page_41thm.jpg
40d011f730c8673c667f1cb7683860bf
9bf98fb7deec595f47eda5aaf0f2951d052a8034
1034 F20110113_AACQYA goodfriend_d_Page_40.txt
93e0e678f1418d6e5240ff0148f3b510
7d9b1707559ad59a8708669c087b16bf8b0f2a11
21711 F20110113_AACRAA goodfriend_d_Page_45.QC.jpg
d3d840e25930e45ec5d12dc95df4a48b
6839e85bf0578b12fb014d469413cd50215ce464
25271604 F20110113_AACQTE goodfriend_d_Page_04.tif
0c4bc01a728177eb198f0bc9780a6dbd
2d7ef6233e664e4f1e46f54912f3fd28092877d2
F20110113_AACQOH goodfriend_d_Page_13.tif
59c25465b271d804c36165afeee70386
4b1cc909075fa78cd495287ab2dcb0a865641af2
1966 F20110113_AACQYB goodfriend_d_Page_41.txt
04e19cc6f622615156841f16c104c8d5
349518ddb6fb991fcde897fee2e7c0b36ef783dd
23374 F20110113_AACRAB goodfriend_d_Page_52.QC.jpg
81ee6e9ff10581a53a8677701cebba95
dc828d3963177be38988719e51a83666a3f7c461
F20110113_AACQTF goodfriend_d_Page_05.tif
8e6a9af4adf7a976f3af3a3effb2d668
490182533d53a31c8dc57fd6e9f98552db50ac03
2436 F20110113_AACQYC goodfriend_d_Page_42.txt
c45b95783a082a0fb92453d51d0bd5ca
66bd3421c159d06945884e05d902dbf2758c1a5c
6568 F20110113_AACRAC goodfriend_d_Page_10thm.jpg
bdabb160452bd5d125064a7d7b1ab521
c27575a9d027b764ce8d595dba65d097708cf731
F20110113_AACQTG goodfriend_d_Page_06.tif
c5944734b50372d234ad1a941937d17e
97e4c8063d180580e7f9ef12d399e72b27da49cc
18961 F20110113_AACQOI goodfriend_d_Page_44.QC.jpg
bca23e3650728503b0e12da5ac568e47
e06010a67f1c8453593ce8a8928c5abb272c469f
798 F20110113_AACQYD goodfriend_d_Page_43.txt
0accde281d64a16aa3b7312fc473cff7
d83b3cefba31f5935cbe037fdfb0416aacdb1cb5
19026 F20110113_AACRAD goodfriend_d_Page_41.QC.jpg
09a094f1e1329eb70734650163a9e818
de3dffe65ca2d4a7d01643045ff9cab3f5e5ff0f
F20110113_AACQTH goodfriend_d_Page_08.tif
7994414d4381d6e708b5cf821e0a9cd0
cd099306570a63e3bea5ab59ca6efd27dfaef36a
1944 F20110113_AACQOJ goodfriend_d_Page_24.txt
5af0fc6156d764b9f397ce4b2377d91b
22f3186aa6dc228337031c38974a24055b01e667
1634 F20110113_AACQYE goodfriend_d_Page_44.txt
f4975f096abcd5632dfcefe1368ab25b
0d5053e86f161b674ae0d897258996e83d267329
5656 F20110113_AACRAE goodfriend_d_Page_02thm.jpg
2624e1d35ba081b153f1752a0fbb5b32
b4e817bb3594ca190507632b0da706ddbc88acbd
F20110113_AACQTI goodfriend_d_Page_10.tif
903d453c9599223536c854b8432064cd
f33f9fce5248d4db1f3dea01439deec96440df28
72308 F20110113_AACQOK goodfriend_d_Page_24.jpg
06938803c5157c5ee0122f90ea78701d
7ad62a83f5859a6dd502abe0155f3d53dc31b395
1846 F20110113_AACQYF goodfriend_d_Page_45.txt
de4b1f3bd36232701baa199d466b90f4
f3d58ddbd257b5eee4e1fcf9ac6ac19732925c2d
24597 F20110113_AACRAF goodfriend_d_Page_54.QC.jpg
076d45729c7c4e346e1e956ef9eb9ae1
0a05e6a8bc315e584f9f93c2b0385a7a6f987f3c
F20110113_AACQTJ goodfriend_d_Page_11.tif
2c8eff29d1a3ffb0111785979d4d64a2
45b3476de71f2b8be45646cf8be684f0b68783d6
F20110113_AACQOL goodfriend_d_Page_24.tif
2b385309b4cd19af0b20450db9df0593
e5cc213bdf5e2696ab2350a213f3a55a8f01f07e
1949 F20110113_AACQYG goodfriend_d_Page_46.txt
55d4fd96009bb150d27b485db8a8d1bc
88296459f50623ef5db417fa4b41be6830b82d42
6900 F20110113_AACRAG goodfriend_d_Page_54thm.jpg
e03609914f69fd4950415158765cd91f
94f450052e8b3bc5e703cea0564f2e745fcff976
6162 F20110113_AACQOM goodfriend_d_Page_14thm.jpg
243ae4592be8bd8ad2a5a782af07c73a
72f0d57372636d32c4ad037124ecba455c56936b
2045 F20110113_AACQYH goodfriend_d_Page_47.txt
5114109ff91887fd086e63c7baddad35
c4b69f6c2ec9eebab1e9457f145393024c71586f
2846 F20110113_AACRAH goodfriend_d_Page_30thm.jpg
f6ddf51c4d6ffba996120bc0de3986df
f80147322f7b77b8ab816f944522225f3315d09a
F20110113_AACQTK goodfriend_d_Page_14.tif
cd09d3c1c85fa94dd64c824f46eeeb09
f8d35f5f2cfdecd9242cb391db6949af910126a7
F20110113_AACQON goodfriend_d_Page_42.tif
ee7bb160a0bca8e26053658acda8d1c4
56de07cd6692e9178ffa2647b6957bfa585af51b
2068 F20110113_AACQYI goodfriend_d_Page_48.txt
842a98a0ab28e7f121df5567fa0629ed
aa6d6b8354ae8b3be71c578b4bbeee12e576cc0e
4344 F20110113_AACRAI goodfriend_d_Page_39thm.jpg
1b01f2594e8631bbf2b6a2b47abd580c
1f0a8852ad4884637c898d45d875e5f23d02470a
F20110113_AACQTL goodfriend_d_Page_15.tif
87c48f2e90c937750ceccaa9cbe87606
cbb1c3a65912969d4f5b30d523f6a8bfe2b60bdb
F20110113_AACQOO goodfriend_d_Page_09.tif
91d8bd81bec228e441ab6262cf875b03
7968e6cff9f4a5e39de05fe24994bb017f3c8e15
2002 F20110113_AACQYJ goodfriend_d_Page_49.txt
c87f9a432539d8efbb541bc8683be315
d125af48f1647baa776422f0d2ee25ab3886c115
6599 F20110113_AACRAJ goodfriend_d_Page_16thm.jpg
741b97023ec2e9aa66d5f3a18fa3e6fc
1681a810e9b986572c515de1325b9479d6319de3
F20110113_AACQTM goodfriend_d_Page_16.tif
649bac727cc5a0391c6cfc7294acb53a
a345b87d380d19968961d3d0ca10c822cf9f44a5
2544 F20110113_AACQOP goodfriend_d_Page_56.txt
72bf67d3faeddc650e12c024da5113a3
2b8a466927d222f06d3de472aeaf9e9312625dd1
1950 F20110113_AACQYK goodfriend_d_Page_50.txt
03cfe8f743800a31c1a2a4d0fdb825b7
24812086457d45b3d2522dbffc51d73d8552ed9e
20208 F20110113_AACRAK goodfriend_d_Page_05.QC.jpg
7465101022ac13dd2b1c12d853b904fc
a3f7a32094e0c8628ee5aa7f5d97e717e48af6f2
102488 F20110113_AACQOQ goodfriend_d_Page_14.jp2
4c988ffb4bd3c3ff13cd9c6c6a37933d
9f03b41942d2a699d2f8443e6c1caa2d91c9fcb2
2000 F20110113_AACQYL goodfriend_d_Page_51.txt
a590d9b73183af0901111cc45f0b2b51
588c622c05fa3309d3ae59613fb7f7fa6c06df8c
5396 F20110113_AACRAL goodfriend_d_Page_05thm.jpg
681112932db7aa603faf6e46eb8d8366
119d335bd9004779217ca41c7c723a9a3b908ecb
F20110113_AACQTN goodfriend_d_Page_18.tif
4f5308241f1d7ed4147c0c980715f8c1
af82bd3389caef4c1508b198bcfc613e165f7187
2898 F20110113_AACQOR goodfriend_d_Page_43thm.jpg
4359debcfb8e8fefbd70faca269d9cdb
a0ca1cfc5616c74134195d2d9cc4f7a38c3b43dd
1960 F20110113_AACQYM goodfriend_d_Page_52.txt
c7c04c38f32811527ecbd6c6706d4f30
0ec9f465a049d1670d8879aa510cf8e895951008
6719 F20110113_AACRAM goodfriend_d_Page_51thm.jpg
7a34e05828856e803aa28988a40f2890
bd5a33fad7f82b2f1c6ae7bdc495ede7c5a30878
F20110113_AACQTO goodfriend_d_Page_19.tif
e1825777d4c620a026daaf365eeaf382
694305041ddec0abcad22d6b559d04cf98f6befb
84515 F20110113_AACQOS goodfriend_d_Page_03.jp2
70381313fa6dae331c03021b0a6583df
2dc02faac14a8e546c240a0e644a0911ffc8f3f3
2519 F20110113_AACQYN goodfriend_d_Page_55.txt
38a70ce10b81ca3d913ece945b3d7130
cc45bf7b0bf0f81786f9d0f2308c2c2162312b95
22221 F20110113_AACRAN goodfriend_d_Page_14.QC.jpg
55bacd1195fbea26c122abdc1f429c50
ff3a2efeee22d8c774f93b29232842e32797db3e
F20110113_AACQTP goodfriend_d_Page_20.tif
61ed4ed11dc06f85688b439641151b53
1376b70ad9e6d72f6f47d885878e50fb8aa73d99
5300 F20110113_AACQOT goodfriend_d_Page_07thm.jpg
728e358a905daefb728173380c3e7c0b
eb955571f687d969376aa087d6ac90f5f941048d
2144 F20110113_AACQYO goodfriend_d_Page_57.txt
b8fa2a9f0053bbb684647c89fe9412ad
4cb51049c4eb0e0a2478042ed0a742593ebc4f1c
6287 F20110113_AACRAO goodfriend_d_Page_42thm.jpg
68c5e7b5aa94ab1aa0c7c33c150b94b9
ef85119edbd78ea7f7d9381082589656c07c07f2
F20110113_AACQTQ goodfriend_d_Page_21.tif
ce4e6011a2dd8b8ad74bf467199b0fa2
bcea4875d975c9f19ce0ba3605ef855164221e86
23477 F20110113_AACQOU goodfriend_d_Page_10.QC.jpg
ecc237625c4c5f504e227823e89b6473
ef848d014d5687d5671eeb6ed4fbb0057f7748b1
1808 F20110113_AACQYP goodfriend_d_Page_58.txt
3a20a935035804d91bd9e118e78cf954
8482af7c4ce244c80f8db8352857374486f5e3a7
22352 F20110113_AACRAP goodfriend_d_Page_19.QC.jpg
23f827a240fb858853db2d87e28aab34
ce676942364522c98b7797f631a933af6310007d
F20110113_AACQTR goodfriend_d_Page_22.tif
b48ec754666c55a03afbccc85048d344
c7b98f6968bc12fc8ab535addcfeecb59a1cdcd1
F20110113_AACQOV goodfriend_d_Page_17.tif
714f99327b7fb0be13f31927c982c1e8
c10c65ef6484b802d4b2320b3a7413226cd679cd
2507 F20110113_AACQYQ goodfriend_d_Page_01thm.jpg
1f0b92bf78e00a347e1212a05bf2ac1b
dcd0a35cd490eb17cada6a28600b13fdba7960e0
23173 F20110113_AACRAQ goodfriend_d_Page_18.QC.jpg
bf08d7920bd8e8deb2d290548dacf3f6
eb5afe7b97f5adb8db721a4846ba0b50fcf8d39e
F20110113_AACQTS goodfriend_d_Page_23.tif
6e0b8eb0eb43bed5d09904e09b66fba7
525bc8f3dd89990d10367fae5c24092ab0e58d5b
35667 F20110113_AACQOW goodfriend_d_Page_33.jp2
d3dc957300d50e896b8e3a4a7b316a4c
179a64d210008176128d2fc1a2871c14a56110e1
10945072 F20110113_AACQYR goodfriend_d.pdf
7945dc43dcf847e7e5230e529e578ee3
11a1463573ebb5510a7608781e352773d0e8d65d
2871 F20110113_AACRAR goodfriend_d_Page_06thm.jpg
ffed7247ba261259d0a7c314b026576f
e15ed1813c97fa14ab798ce668bac4fdde02a78d
F20110113_AACQTT goodfriend_d_Page_25.tif
7cf43b6ceff8212248a806c4f534bc43
49b1fe00a4f05cd57c8277a1ee7d630dabe1d600
70704 F20110113_AACQOX UFE0004787_00001.mets FULL
fdb4254d49219b6ebbc2fede23381dae
9ab77949b9e9113c0b9e10150379278847fa8a4f
3638 F20110113_AACRAS goodfriend_d_Page_34thm.jpg
97af60e60f0e0a0024ca3ebddff9b41d
a756c276a7985bfe8cc5a7903436a332c2b91e8a
F20110113_AACQTU goodfriend_d_Page_26.tif
d729769cb8bea82cce32e25085e0b940
f02051a33348e5e083c00257df133836941b6176
6869 F20110113_AACQYS goodfriend_d_Page_48thm.jpg
4b3d26cc2510480231cce055dbd04acc
6e75bc2e6dfa3794347791cb7dedecf0f04ee1fa
18647 F20110113_AACRAT goodfriend_d_Page_07.QC.jpg
4ac3e5e624c9df9066f6276c1be815b2
0101bac855c373c61498a6a5452db4f21b301243
F20110113_AACQTV goodfriend_d_Page_28.tif
448e420e87bce36e639badc2f65ae074
530330796ed22f857294edd757852e79c0c6d4d8
3167 F20110113_AACQYT goodfriend_d_Page_33thm.jpg
6537dc82a47bf9ddba4a54ff9f051962
478847958200384b50423de30f682e04d05b5d55
4428 F20110113_AACRAU goodfriend_d_Page_37thm.jpg
309b9f983d0cbf5136728aebb4d3e88c
753c8899bd427001446e1fef34a49ea7c19b2e78
8423998 F20110113_AACQTW goodfriend_d_Page_29.tif
b35c442ec18ad37263f65f6d83669523
e36c3699a0fcc44d6d979043dca1af7f72b82d62
6404 F20110113_AACQYU goodfriend_d_Page_52thm.jpg
6fbcd2c4abd3311d6cf2d572abe53571
f352083aabe85ce6d1e567d3c9f81d674b35f2ef
4336 F20110113_AACRAV goodfriend_d_Page_12thm.jpg
b1a4c08ffd96dcb411a854eca7895cf6
21921c895b3509195c409b857670644564ce1146
F20110113_AACQTX goodfriend_d_Page_30.tif
9f36a04e7b51d6bb7b6b52b4c80ad09f
5067be2ac669725d379df76c143b6adf23831c7d
6542 F20110113_AACQYV goodfriend_d_Page_17thm.jpg
0cb8c5d99d8abd0c4f7408807f6d216f
a6d53a8d4f8013c230034b0a44ca9c284f4ceb99
24254 F20110113_AACRAW goodfriend_d_Page_25.QC.jpg
4e6cb10179f60b4e7de733ea31d54137
4ea47e7927815bdf9f4c6f5e989d000e71d79a7d
102578 F20110113_AACQRA goodfriend_d_Page_56.jpg
5ab7e66e1fb203b5a8d1ede0c3a16bb1
8e31c1b33b5c2e953b936749d2f32cddc38bd5a9
F20110113_AACQTY goodfriend_d_Page_31.tif
4a2806037ce9e4e7cf1c41eeef5b132b
434baebefe11df8c8e0583d85cbbb1d36f17cb1c
20985 F20110113_AACQYW goodfriend_d_Page_13.QC.jpg
28026055d480ee5f0e54dbaa783960ba
f0ebde59a040bfa78732acc95bdc2b133c203b4e
23820 F20110113_AACRAX goodfriend_d_Page_16.QC.jpg
ccf0116660e3133ee8cccd5d6db47313
b02fdbb6de8d141c0456ce63c92c0101383d6242
77216 F20110113_AACQRB goodfriend_d_Page_57.jpg
66c566c50ac1369d408159f406e4ac26
429f31225887b0695aef0841b0fedbf6756f429b
F20110113_AACQTZ goodfriend_d_Page_32.tif
9c393d44afa15df75849cc06cb489d3d
49ddc452c6940f73b0d5f495b05ddf59a828fa32
6320 F20110113_AACQYX goodfriend_d_Page_11thm.jpg
03341736b57f9f93b5d8d5078f541e06
571d9d2a08a5288da3e73dbd1bafea4675ccbaa9
3434 F20110113_AACRAY goodfriend_d_Page_29thm.jpg
4cf348016fb17735f6bee40205eb059b
fe00f7e4649da680953c2b894835fbab646b45e3
65661 F20110113_AACQRC goodfriend_d_Page_58.jpg
d1fd2ef0e22ec3706e6abaffea15a8bf
f303a8a571cee4fbbbcdb065c0791d490d268c5e
30402 F20110113_AACQWA goodfriend_d_Page_36.pro
d6604d3e27de1c462a8683edd56c60c6
b723d11cf3d4cdab589e7e636cba48503c52692a
20993 F20110113_AACQYY goodfriend_d_Page_42.QC.jpg
93f627d1cd5b574dd187fb43fd4c4a4d
875d4ec3d864b8cac1143fab1cb433c0a51d2935
28598 F20110113_AACQRD goodfriend_d_Page_01.jp2
d46b5be985c09b331f7af01560585c47
71d282cb4382ea2a1480e60c68eb154447784453
29600 F20110113_AACQWB goodfriend_d_Page_37.pro
4786a99509052288c6ba03c248f7c87a
1487d467aaca44eb0f7f0438a9d4cc9adc78f2fc
6296 F20110113_AACQYZ goodfriend_d_Page_53thm.jpg
f86ce31c25ee34d6a5bf35351e6665d8
087d32f6b1f33fb96933ffb0953db1bc1172c224
6656 F20110113_AACRAZ goodfriend_d_Page_49thm.jpg
c71b1945731a425a9bbb594415f276ff
8bed139802aa9ef1d6fe360ab785657093c316db
91378 F20110113_AACQRE goodfriend_d_Page_02.jp2
283a71423cb3f3a49f4217e1fb8ced5f
4b065b9b4022d83ba9076f3c92a72d2a3e691985
30204 F20110113_AACQWC goodfriend_d_Page_38.pro
2d66552e89caa290e39c4cf2110e3354
292405a201d92d539ad3026fd903c19cc2e94311
1051952 F20110113_AACQRF goodfriend_d_Page_04.jp2
f392b0087a451e85f6cfb97ed259a602
d60882637a67da84d14c442b3c2f5fb6e7f5c000
27626 F20110113_AACQWD goodfriend_d_Page_39.pro
2f5456c0d378d9803b9b88c923eac858
88714269b00e38182622c70ca203f4488c015904
1051974 F20110113_AACQRG goodfriend_d_Page_05.jp2
b13a1dbed4179e6642d522d8009c48e3
9cf69d22b58b6d17f690e5f8b4c0f3e037f468ca
22409 F20110113_AACQWE goodfriend_d_Page_40.pro
ae1b8ca1bfe0b0ca501ed2be2d09ece7
73a723ecf79376c8528bded5447c545cf07bd55b
752441 F20110113_AACQRH goodfriend_d_Page_06.jp2
aa2fa8b6f06ac27c371a73de9494822a
1315fb21063113acc73cde97efb8c1d225b62875
39971 F20110113_AACQWF goodfriend_d_Page_41.pro
4441d76867ff4b2291c3881aefd7094d
825c75ecf77b0bcb3a6719eb0d7c72cc2afa2b29
85662 F20110113_AACQRI goodfriend_d_Page_07.jp2
596146ead7de5a85b09a2653c013c08a
abecd30e59f093c882f3a71f6ae8f4d35d49e7b5
86989 F20110113_AACQRJ goodfriend_d_Page_08.jp2
e2f0b1908cf109e4c79bc35315c79d65
a7218678484baa2b1eb0da785f6fd5ec9a4d1347
45722 F20110113_AACQWG goodfriend_d_Page_42.pro
31109552b67eb25f2eb462ce1f562af2
87e638cf0467b418482bf2c4be8c9e6bc4004078
97869 F20110113_AACQRK goodfriend_d_Page_09.jp2
7ad6e2c99c70ef47cdf9f19295099647
171b838bab17212bcb9599de600ba22693bc99c6
14267 F20110113_AACQWH goodfriend_d_Page_43.pro
73425fff605ea2512b65f989410b6f88
dac48f0d054d612301505f0544c4bb900982ce35
39082 F20110113_AACQWI goodfriend_d_Page_44.pro
b8c086abf4157adab45679c25ad2d101
b0095d0fd11c3f8da6f187de4ad8c19940b8fb36
112525 F20110113_AACQRL goodfriend_d_Page_10.jp2
f20de99b5891286ab243b9becefb3738
2a7016c31dc86fbec73d7beb608084d4cf38c989
46645 F20110113_AACQWJ goodfriend_d_Page_45.pro
04b4dcb1107b8f017067653716af3f9f
b53ba7fbfaaedcac1ecee0fa4c8acbd1bf51582b
66801 F20110113_AACQRM goodfriend_d_Page_12.jp2
934a3ad682cd155f24883139f1337393
f2a9059b091d6da1eacecefc1bc33f2715eaf41f
18728 F20110113_AACQMP goodfriend_d_Page_03.QC.jpg
45d207beae97575e22d835735a67a671
c1c3ebeeeab0524f830a13f30b0a7d72810161b7
49381 F20110113_AACQWK goodfriend_d_Page_46.pro
b56896bfe2d3f3c89c4cf6a45a314a11
9a2599d7062fc3503ddd8670c771dbbd3635ec27
98095 F20110113_AACQRN goodfriend_d_Page_13.jp2
94071524a68c626c4265d2f66d7bf3cb
a7b5cfec80eef27b108b79e7a400fa1b2660d624
18943 F20110113_AACQMQ goodfriend_d_Page_08.QC.jpg
7e2a3d9b6b6f9c2399c4877b0be078bd
4f8b8ec471e4515aaf077ae4e05e80c58b8627a0
51924 F20110113_AACQWL goodfriend_d_Page_47.pro
745afef5b3b2e440d71daf6fdce3d530
1194330dfc5a54093bfe35b6835601b819563d94
84954 F20110113_AACQRO goodfriend_d_Page_15.jp2
d0b5dcb7f05af5457929c2c1d4e41e8f
976b3436630e6b61ec795f695440b273d41e0ec9
F20110113_AACQMR goodfriend_d_Page_07.tif
070eeba2145a0d6eb0a526a50ef0fb02
48ede3a05c63272718ec4c748133678dfd12d0ff
52834 F20110113_AACQWM goodfriend_d_Page_48.pro
62a4fcc9058adf978d3fd9d030455300
5da2f111949d5fb8e966c64e1bf382113e805036
112227 F20110113_AACQRP goodfriend_d_Page_16.jp2
ec36d417783cb43dacffe94cd7419f67
a5eab9d8300217f0ebf2ec647b2fd42fe388113b
34950 F20110113_AACQMS goodfriend_d_Page_35.jpg
2361effd70e2ec2ff18adffae8cbfe7c
50fc4b553e36c069d1af104ccd098c193bfdcebb
50917 F20110113_AACQWN goodfriend_d_Page_49.pro
bf16ed7ce6650815ef40d308dc85b9a6
49bc0b556baf2a3bd0c13c6c4a38814c3e9e50ac
112347 F20110113_AACQRQ goodfriend_d_Page_17.jp2
8e8f4eac2437f5e6d08f1d92fc80c687
94f1fd8fee732917a4e4a16219c17a4bf32fe28a
104536 F20110113_AACQMT goodfriend_d_Page_20.jp2
e0d40f4e04434e400331fa4cf28f1e8c
d1010dd2d1abc930d0e8ef5f3fe995132ddbc79a
48706 F20110113_AACQWO goodfriend_d_Page_50.pro
5ab6824d39a2226324ab58fa25e67fb7
db082bfcc719b4ca75f2f06a4a38ed8a37cc9f39
108442 F20110113_AACQRR goodfriend_d_Page_18.jp2
cf639789d8aa87abef3abd2de7bb4629
37b045a517a20c964f61507688881ffddf20128a
1939 F20110113_AACQMU goodfriend_d_Page_11.txt
e04119e3f99cf27ed2b301f666de4565
42c46ffeb5f2634fc834521fa5a838b4ad9a0c35
49829 F20110113_AACQWP goodfriend_d_Page_51.pro
5829cdae4004a47471b7d46fc8b12449
c21240fbb827c94bc7d7fb71dac9e2977d513565
102227 F20110113_AACQRS goodfriend_d_Page_19.jp2
8bd7324c12ab3307f643c8a9c22bf7fa
ee7a6e36cdcf539ddefa9f0475fa1219f3056e6b
804 F20110113_AACQMV goodfriend_d_Page_06.txt
88e09694ec8b32ad72f98999e8b796c1
aaee815c24b316bb41ea87ed6f7aa37bae2e65dc
114078 F20110113_AACQRT goodfriend_d_Page_21.jp2
1e82f3907b1a334bb9f7e6bc159f4e12
30d4acdc44e0ff3bc42a2de8b83fffbbaab63ca4
15171 F20110113_AACQMW goodfriend_d_Page_12.QC.jpg
6852aebb03d25a3d5d7d1849d9e3b72c
dde00dbbeee5345a08172dd068ec25acc4f031dc
49444 F20110113_AACQWQ goodfriend_d_Page_52.pro
192541d43bbb14b9031f206df5c76b11
810f46b4cca465cbb82d1b36984f04bcefa5042f
75219 F20110113_AACQRU goodfriend_d_Page_22.jp2
ee092d68bc19746b0058f9967a9f4dab
00360261e492a1bbaab12100e29f1d5502cec88a
1925 F20110113_AACQMX goodfriend_d_Page_53.txt
87811a0124938604da37e10cce194b5c
0f5e3cb07ca87db9528f3273007050363b30ac0f
48794 F20110113_AACQWR goodfriend_d_Page_53.pro
eb1f9d7990226b05812132b971f7c910
8e983b4273421b630a269197f4621f6bfbc3a9ac
91487 F20110113_AACQRV goodfriend_d_Page_23.jp2
cb47b2c5c0424245d15a047af906dc20
cfc2fd147bd16608ead67bb4935e7abeaa5084fb
F20110113_AACQMY goodfriend_d_Page_27.tif
6a668157edc05ba35aa74e7bd6094f60
bd77d72e0334ac809a876f5ffed5954ab5219eef
65206 F20110113_AACQWS goodfriend_d_Page_55.pro
19d58da1d0fa326acb0c2fbdb0b112e0
9964fcaa7c2ce714c3e762064b2a3a9cf6809a57
108793 F20110113_AACQRW goodfriend_d_Page_24.jp2
3317c74ae7f1237096dc683f372fcb60
200fc7b060b58a2412e3cdc460ea498eef8cd8f8
4767 F20110113_AACQMZ goodfriend_d_Page_38thm.jpg
0553976aa60adcec8c02658e44bb3274
10d54b0a619ec75d93b4de764f437956b88b4ddc
65868 F20110113_AACQWT goodfriend_d_Page_56.pro
7c7fd382d4b14989277cf577fcb44306
6e607856fbffcd43d8dc93d9522979fd1fc3d6ac
112672 F20110113_AACQRX goodfriend_d_Page_25.jp2
05b536fa164e439356feec0047330b89
8e8e67f0ec1bd589eb729d6fef3419b5288eda08
44899 F20110113_AACQWU goodfriend_d_Page_58.pro
d652a1587a448cc75dc3a7650c382215
d665f1d11c228651eb6b33c705dbf8128cfb62d1
112761 F20110113_AACQRY goodfriend_d_Page_26.jp2
7846fc3e662c40eea337e6492b506e24
6766507c5e61f693abcb51a49b6816edd97ffc7f
531 F20110113_AACQWV goodfriend_d_Page_01.txt
af58c0b2ba6d8beaf98218ee2183e953
32383023af7006f33fb41b3817ce86dc46f5e4cc
27027 F20110113_AACQPA goodfriend_d_Page_01.jpg
5f19516739c5bc3689ff1fb1fe383f6e
7e03d6d8cc34f2a214da217e847ba5a553113168
110833 F20110113_AACQRZ goodfriend_d_Page_27.jp2
4c4bf45b58659a8dc2d3c5ad744a6881
3368e6926c324a41d2d7120e4c3c2199b8061077
1524 F20110113_AACQWW goodfriend_d_Page_03.txt
830603cd524119a860681766aad3809d
c09ca185eec91a63e15249c3f075ad6b4e048730
62076 F20110113_AACQPB goodfriend_d_Page_02.jpg
7155265c0664b968f3274a4d3867614b
bc2a6bf3694e2dfd426ce142af706975145a1f59
2800 F20110113_AACQWX goodfriend_d_Page_04.txt
58213202a163d8aed41702067786f7df
71b756f097349fe4cedc70d541c37230f7ba8996
57436 F20110113_AACQPC goodfriend_d_Page_03.jpg
7b9ee5f114a67811cba60f6f663410aa
5468e0307be472f6b407b5b081c894c0d3e06bb4
F20110113_AACQUA goodfriend_d_Page_33.tif
a2018599e90a5ec5cedb627b4345d5cc
f271431567049dd7b3609dfa041deaa0348877a8
1979 F20110113_AACQWY goodfriend_d_Page_05.txt
9b2e722f377619f0727830d5723d05ed
fc684a3768899fe130c4f9dfa94167e01c430a85
60959 F20110113_AACQPD goodfriend_d_Page_04.jpg
353f5339b7be5e8efe4edcebf390a5f9
317ba6df665732e36d5b38f7e34f4c24f6dacc01
1054428 F20110113_AACQUB goodfriend_d_Page_34.tif
46b2bc442c170e0d28473c12a53cd805
3cb421de74c19b9d6ab5ad455004a7308e810bea
1699 F20110113_AACQWZ goodfriend_d_Page_07.txt
5780fd9f57146ee2e122abc1c5bee394
9a1372d7ed6e06c683053ab9e801666a3719ee84
72820 F20110113_AACQPE goodfriend_d_Page_05.jpg
6723a4a827d3176653ef97d29adcb65b
c68ece06141eac4672a5883580e37d21b3de4ea0
F20110113_AACQUC goodfriend_d_Page_35.tif
fd657f16c3bf95e4665afad535d69217
bac64caae737527ef9b4c2c3e58c8b416eae96d4
59137 F20110113_AACQPF goodfriend_d_Page_07.jpg
a9201b0afcfe703e336f58605cbec491
4d35364f317e8ce003c1e839cc43b4d2b18fea1b
6011 F20110113_AACQZA goodfriend_d_Page_09thm.jpg
fc9afb582a1115dacb63966df904a387
d32cc178e7b15a81d5906b538117d224b999f0d0
3533 F20110113_AACRBA goodfriend_d_Page_35thm.jpg
2f1e6b16c0b2061850a190d8c9580564
656a279e8dddd5501a9f9a14b9268a78c01b4a70
F20110113_AACQUD goodfriend_d_Page_36.tif
1349d0cf65d26740fe514bdfa7fdf399
d059931d729ad194168a9a604292d0228cb21577
58349 F20110113_AACQPG goodfriend_d_Page_08.jpg
59f97ba12c789afa823c35d741ce71e0
77f94eead72550d82dce80b20297b1fd719c8891
6198 F20110113_AACQZB goodfriend_d_Page_45thm.jpg
2e201721b9e6add6a5cc5b4d9377198c
c4b57cb4e21f9fdafa6f7d662c29486a03aa5831
2115 F20110113_AACRBB goodfriend_d_Page_31thm.jpg
0b69a9e1b1404689221151f9f8fc3b47
00c3ad8ddb1667deb4babcb9e319a117629434e5
F20110113_AACQUE goodfriend_d_Page_37.tif
627a4c01316b27c6e4d9f2b4b8e42e3e
2208c53e1450a668809db088ff5edf15688b7f52
64914 F20110113_AACQPH goodfriend_d_Page_09.jpg
618715fb30c257657d34eabd05ad1aca
58ecf5d8d972910b6af4d2fedbc3779f2eda0688
8238 F20110113_AACRBC goodfriend_d_Page_01.QC.jpg
901b2d921860fb1b865e7090235944fb
ff4e9c9db2e958da9d4b59af1d65231effe178a6
F20110113_AACQUF goodfriend_d_Page_38.tif
e350eea466f3293da455b137fce4a723
3695166722061bf843fb8c2e044ee2bc2d6ae35b
73488 F20110113_AACQPI goodfriend_d_Page_10.jpg
26e521f5b23f62e2f245b18874d0efbb
263f07f7993de4c24b0fdc1b8e7ec516f2aa928f
5622 F20110113_AACQZC goodfriend_d_Page_23thm.jpg
2782d96b50978a53b0d5c306c9372139
3329cb1fe89ced839d5d8a229871a8dfcda12056
6374 F20110113_AACRBD goodfriend_d_Page_46thm.jpg
3d5cf9c35e12a4b450ca2c124157dcf5
90269f3a1f80455e14441d59084dcc4d498899a7
F20110113_AACQUG goodfriend_d_Page_40.tif
8ce0de5a7e01ca26752070db2ad3e60f
dd3cd1a245ac8102b509eb828e12bb459c8b721f
22424 F20110113_AACQZD goodfriend_d_Page_53.QC.jpg
53e336fabdc4ca10c2735b8df855bd4d
e9e3d46b9d7d87c912ff414d970d887f74b8226f
22760 F20110113_AACRBE goodfriend_d_Page_50.QC.jpg
4e0655a095b6424f56e753054f3942b1
8be594d977c0c8ba47e19e4fde5541be7482e261
F20110113_AACQUH goodfriend_d_Page_41.tif
66e7c0c6d20ffcb9eb095af2a020943d
36fbb555519194500dd59059df9eabc999f40c34
68836 F20110113_AACQPJ goodfriend_d_Page_11.jpg
146a15c3afb34560a94f895dda9a320b
fe361fcecb06ce781473f7b26b5987015ebefeec
12654 F20110113_AACQZE goodfriend_d_Page_34.QC.jpg
571ad5b615d67508272736d980306a42
c27182e7334ceb0ecf33e53311ec7461218a622d
24402 F20110113_AACRBF goodfriend_d_Page_48.QC.jpg
26adebc2f631b559cd41bb4ce75ae531
2cd200bd1684a34a110d6ea00861443b56da7b23
F20110113_AACQUI goodfriend_d_Page_43.tif
53ae23af9eb0f30d4cc944e41341b32a
0666f1b20c64cf598d506733579f38128e639fd5
46402 F20110113_AACQPK goodfriend_d_Page_12.jpg
01a886e4e4c80d7f11224c54327f1a9a
6885d69962273b1f221ccfe59b9517424990407c
14413 F20110113_AACQZF goodfriend_d_Page_39.QC.jpg
addcdcc1f3ff1cc8f4b664e5f2d366a0
aa27c7b8692296b846324fb6410785506f3b229a
9169 F20110113_AACRBG goodfriend_d_Page_30.QC.jpg
6ab4130b07b0b075d552a2c53123c821
520ce2335aefcc2fbbd64a1606a9d2e766dc715d
F20110113_AACQUJ goodfriend_d_Page_44.tif
b64c38cc26e615504bab4d1e88e61db6
3c9044384d2a363292f5a2981f959e78ad9d5bd9
64029 F20110113_AACQPL goodfriend_d_Page_13.jpg
89befbc3c6c327ed32d946203fe4db54
60ad0fa1d83ecddcbdf587728709132ae56f39ef
11046 F20110113_AACQZG goodfriend_d_Page_29.QC.jpg
0b6920d12917e9349a9727a1e5239d78
550f39bcd5a3962b681bdf2dcde52b83eb75f004
26501 F20110113_AACRBH goodfriend_d_Page_55.QC.jpg
7866a840a96ccd236556b4ea8e13abdb
d8f738ccb2c312ec550690728aa833f9af4c0ad3
F20110113_AACQUK goodfriend_d_Page_45.tif
76aa0156fd476962ca64a82d9a309967
ace332d669f2c8023b40bad1502e5f8ba69384be
68059 F20110113_AACQPM goodfriend_d_Page_14.jpg
707bd9823f6a092e2a3c4ea41679cf6a
9ccaffcc33b3a8ca2f31f62dbb268bc80acf4a31
4625 F20110113_AACQZH goodfriend_d_Page_36thm.jpg
7ba453aac6539de652208452dccdb4c1
05cc742beecb123cc2444dde0eccd319d471c615
17311 F20110113_AACRBI goodfriend_d_Page_22.QC.jpg
5577c9fdc3493b1fc0ed3eca86549215
dcf49ccc462a4477cd36ebd757751744270d3c99
F20110113_AACQUL goodfriend_d_Page_47.tif
4fbd18d07cec6b3932fa42a48359872f
f645ec764da31e3093562a1a26db8d6a211803d7
56916 F20110113_AACQPN goodfriend_d_Page_15.jpg
e4a82e1930d537397d889c4e738c2ddd
a078190b24717d4b8212566943c39cb8b56b6178
6440 F20110113_AACQZI goodfriend_d_Page_27thm.jpg
8087d79da487ca889f7bf1d512849bd2
bc3b948dfa36e4a44362b99f363970f97b08b5e8
F20110113_AACQUM goodfriend_d_Page_48.tif
d5bd44514e0379fa702b56b93df80c84
0a471a8a247e5d31a3db0ddfbd2dafff30408979
73225 F20110113_AACQPO goodfriend_d_Page_16.jpg
8ab2a7cf9e3ae77056d014dd9b8af21a
d86ff34d760fa91d5e960a44678a1d46bb986851
6573 F20110113_AACQZJ goodfriend_d_Page_24thm.jpg
66959fe2215c1a5db038f1dc9bd5f902
5a6a1761051e18ac989c17eedf72fa551d85960a
9667 F20110113_AACRBJ goodfriend_d_Page_32.QC.jpg
da7aeab01c2cf56f3e82855200f4e021
d9a5b3e53abdcaec674f97f44a2d8734a0ffe9c1
F20110113_AACQUN goodfriend_d_Page_49.tif
e4eba9720009e382c3c9c46b15e8d20e
7099b9d975e2049e5cb058778b419a8e3fe797e8
73693 F20110113_AACQPP goodfriend_d_Page_17.jpg
8368853980e698e96cf001db9408b504
c151ca315f50cdd314cd89b953ff0447d76a72d2
19860 F20110113_AACQZK goodfriend_d_Page_23.QC.jpg
812fe4babb0a9a2a650d2920381f0295
a03e9ba72c5a47522507ce52a36675dffeef660f
3673 F20110113_AACRBK goodfriend_d_Page_40thm.jpg
15b9944afaa209f200082ab6c211d2cd
a69e9373dfe44d58c391f2f7241f39108e844d9e
70319 F20110113_AACQPQ goodfriend_d_Page_18.jpg
7b2010be6d49ffbdf2f70b446a293484
578cb03dbc589d37d2f43d1939966386fddcee2e
23621 F20110113_AACQZL goodfriend_d_Page_51.QC.jpg
eb9b63209570b79c2192d042830148cb
eab77932d1dc8b89a1671dcd989c39de4d80da8f
16874 F20110113_AACRBL goodfriend_d_Page_28.QC.jpg
4a844ab8ee3abb042984008cf32a86e3
9c22c3d3ab139201a5c3f0558cf14df6ac35fb16
F20110113_AACQUO goodfriend_d_Page_50.tif
1980a6585cca20ba659e0bd0257f5fe7
080ed0bf2d99cceaacdf5bdad7cd0d853406bee6
69082 F20110113_AACQPR goodfriend_d_Page_19.jpg
dfeb16bf7c401f60397b9f424cce3a9b
f239137af87904a434731cbdcb7fbdf1d07c805b
23438 F20110113_AACQZM goodfriend_d_Page_26.QC.jpg
f5482c4b3bd28466c6d340ee3def08b4
e2ff2ce9b07f4b9bdb71106f4b29afd14ddd1ca7
6125 F20110113_AACRBM goodfriend_d_Page_57thm.jpg
245adb62ce277f7cc932ed01a9e46826
dfb26c84accf8d887e5cc8ef2f47a43ece048d48
F20110113_AACQUP goodfriend_d_Page_51.tif
a99ca1e0d75f41a1d46a85615b2d53c9
8fa99b421ddf3d9696e8b7b787f6c7ebb77311f8
69163 F20110113_AACQPS goodfriend_d_Page_20.jpg
3436373eb691aa73d635bad8bc4572f2
eb6fbffce108c0f147fea796da49191dbea1730b
20286 F20110113_AACQZN goodfriend_d_Page_02.QC.jpg
23dbeafd927ef8389a6cfa03855ad2a8
499ad36c46ca228195eacbdd21c42ba0b5dc61b8
6213 F20110113_AACRBN goodfriend_d_Page_31.QC.jpg
aed6cc6b15eb75b1429822a630704b34
4a33724862329bc7320671fb575b2f9c5fcf0bf4
F20110113_AACQUQ goodfriend_d_Page_52.tif
27c5f69af3546c5a46a68e06bde8bb9b
0780376b872436b4077d0c092dfcdc631de6b6b9
75273 F20110113_AACQPT goodfriend_d_Page_21.jpg
98be60edad4089dd62d13668e09f0677
04391327c80743ef8a5d81843cb5e5116b22c7ac
2761 F20110113_AACQZO goodfriend_d_Page_32thm.jpg
245f04c19e1dd87e29627cffe041edf3
8a08e1a1d67c93e9664e5ed830acd13ee4053cd1
9816 F20110113_AACRBO goodfriend_d_Page_06.QC.jpg
c1ba0a3dc14907359402ecc43a4ed26b
46d04d06e1ff67b0620cad8669f3822e9bf08044
F20110113_AACQUR goodfriend_d_Page_53.tif
a43de431214b7cdb1f5ac7b6211bee88
aa7e2be1b92a545f6bdccf650126a30ed7d29a57
51729 F20110113_AACQPU goodfriend_d_Page_22.jpg
da905c92eee2ca2f5f7284ee2040e1b1
1b9b55cc5d8edb65930438d285be6fa4aee500d6
6311 F20110113_AACQZP goodfriend_d_Page_19thm.jpg
0107bebdf1d53768f17d01c4292209d7
075eaf88c7c78f4df98f8ff3f73ee26c53431f07
5253 F20110113_AACRBP goodfriend_d_Page_08thm.jpg
cc2c8adde9632477294712a8b31f8d9b
df987c1a68ddbf863bdc4adfb0b566cfadd9ab19
F20110113_AACQUS goodfriend_d_Page_54.tif
3f68384fa7531ace0d13b4feb359682f
c118d79aaffb02bdee1cb4e112e35803e0ede056
61322 F20110113_AACQPV goodfriend_d_Page_23.jpg
ef963e6c85e1c804938c490cd9ca27fb
bedf7135c804d877d79782c91591f97f6ac15977
11949 F20110113_AACQZQ goodfriend_d_Page_40.QC.jpg
295f1679881db2553bb98fd15240a7b5
226659d41b74956ac8a4bf09249be286739a4909
23236 F20110113_AACRBQ goodfriend_d_Page_24.QC.jpg
19c1b4eb8e6a9115466908bef2dc001e
983fb3fd34fd1844884b1d47fcd5603a9ff61285
F20110113_AACQUT goodfriend_d_Page_55.tif
26341f0725aadb22ebc550e5d21a4a1b
700b0ff38190ea532d2b18291239099beaae7da2
74153 F20110113_AACQPW goodfriend_d_Page_25.jpg
d65ac7b3ff27c2ece28a82a3d48990e1
e33976275de1b89e69a9bbde59ac5d1a76295744
5267 F20110113_AACQZR goodfriend_d_Page_15thm.jpg
0ee99b91ec3e45b1f33bd1cf8366746d
9913d1d47728a8f1b33c2b7d008f152d11f84ea0
5434 F20110113_AACRBR goodfriend_d_Page_03thm.jpg
8951a46e854331868267b0e9b5f18894
d4b9869492eb8fc123a71a32afce2b0a89966b1d
F20110113_AACQUU goodfriend_d_Page_56.tif
aa3e2a9382d2cef3f3fd20648a286b18
9716a11a547b61fcad1422358970b9ead43880cc
72984 F20110113_AACQPX goodfriend_d_Page_26.jpg
23aeeb79c0a6f37edeb7bbe4a55fa0c0
ba3b2edf6ef6f29a69dca318d1c1b378fb8a02f2
6646 F20110113_AACQZS goodfriend_d_Page_25thm.jpg
0ef9c8ed6c737888e9216fa56e3aae46
0f3c642e47185b3cc72aa0e9328421274077c6c0
18954 F20110113_AACRBS goodfriend_d_Page_15.QC.jpg
d5012d48b50fa9179d6f1ec72fa136a2
3605d20603d3d6851d378b5a5f2bc2e149a3c900
F20110113_AACQUV goodfriend_d_Page_57.tif
91fb3fc53f882a0986b586a0a440e227
2dde0d6fb6c4c30bbcc47b8a1375c4ad741ad326
56978 F20110113_AACQNA goodfriend_d_Page_54.pro
9e9bd660316e7a178226f06a09a338c1
c909738383177884a996c97b782371827e2698e6
71736 F20110113_AACQPY goodfriend_d_Page_27.jpg
86a18570a6510977ffff9ada8fafa1f3
9d4bfa67f7dd9d5f102a50ce69013c717b03da2c
7174 F20110113_AACRBT goodfriend_d_Page_55thm.jpg
aaf51bf13e8e229a08c352345ce832d0
b4af8aeffa6dc9bde741d5f0cec93f1187aee1e9
F20110113_AACQUW goodfriend_d_Page_58.tif
82abc51e3848d837bf629818658709fb
53b1800134600ef4ce64cf6b18390b55ffbe6531
5896 F20110113_AACQNB goodfriend_d_Page_13thm.jpg
883d0b6e440e207a84050b29146049a6
b100cbbf57402426bd6f18124ddff4ad9f7f47f5
51144 F20110113_AACQPZ goodfriend_d_Page_28.jpg
0c171089733e45e32f7c5ad0e0b6fa6c
7d04da3a4a0c940af3660e2621216927d9835336
6492 F20110113_AACQZT goodfriend_d_Page_50thm.jpg
28dd54f9d08106b1545bdf57bd157bb4
a09aac70e2c0e1a77751329e7bf8a47f483eea6c
6595 F20110113_AACRBU goodfriend_d_Page_18thm.jpg
fc21ee29e69477a77e90d41dc07933b9
b54ca9d2a5efb565fa11081aaae9f5d9ffb487cf
10119 F20110113_AACQUX goodfriend_d_Page_01.pro
3c54c6014b06f12572c030d4186cc3dd
19f7d4f6d645e896daeac2549d8b5bef2eb152b8
55003 F20110113_AACQNC goodfriend_d_Page_57.pro
f852e45093f011f5deaebd10a8c5e444
9302477b15fc608e6b44a9fab6b4623f28374156
23040 F20110113_AACQZU goodfriend_d_Page_46.QC.jpg
ab938878610bf1195470c2581007b998
50c3f03a1f3caabf4d1bf8237a4660545946073c
6855 F20110113_AACRBV goodfriend_d_Page_21thm.jpg
2d873fd8c13c841522cbe0abf887583f
61b146e60cac6a7797a071f943ef7976c8728fcd
42311 F20110113_AACQUY goodfriend_d_Page_02.pro
9c7106c4243eda71e8155890a064a7d4
0ade15ae56cebedfdc3d7d99dec56c2cf7a7fe3c
106411 F20110113_AACQND goodfriend_d_Page_11.jp2
6c1d0a01d82d82ad660b7b45c97cb841
c5d9e6667ec321e74d04e8d8d8c5792e2bc147e0
24259 F20110113_AACQZV goodfriend_d_Page_21.QC.jpg
371395d341a210a9c00d1e1b3b59fbcd
c122a6961e76628e59d78f51f9091de4ae51c2c1
9130 F20110113_AACRBW goodfriend_d_Page_43.QC.jpg
6f82774ecdb4c2af9dad8052ed95309e
59158fbea6c5711cadbb4e4ccd133cdbcecb72e8
77717 F20110113_AACQSA goodfriend_d_Page_28.jp2
c392d2f558b873511b7c76d2e0953650
9646fac8395abf5a9b5f898001ba94045757d6ed
37855 F20110113_AACQUZ goodfriend_d_Page_03.pro
ea19942f793de6524db76996ad4dcb65
42dbd3d18c300cfe7b3eed06388e8a89ec30b21c
22488 F20110113_AACQNE goodfriend_d_Page_20.QC.jpg
5d6cbb72c76e347117016c170605d0a0
09379636f1271faeb2e672b880e663c436b1273e
15689 F20110113_AACQZW goodfriend_d_Page_36.QC.jpg
fb5a2c4481304fd612c1df3324ca362e
be573a8e5fac2d3f99e14d7c0700c082690f222b
15576 F20110113_AACRBX goodfriend_d_Page_38.QC.jpg
91f3ab8a929b3a27705e0e45c5fd5945
f90b174510e597933ffcb6f3f9a5b798b7d5cc84
464077 F20110113_AACQSB goodfriend_d_Page_29.jp2
39bed13b115be807f7fddb198c718396
9f4ce500b7ab110ed8e19176c8ec8630d549008a
F20110113_AACQNF goodfriend_d_Page_12.tif
b8959cdc6697e6647c231586b6dbe12f
07fe461afea1a633f26573a88202d8f3b4885ee8
15050 F20110113_AACQZX goodfriend_d_Page_37.QC.jpg
f4d256aae1760982d8065c6337405067
d231a865848f6007b6fbc38ddc68c125bac50803
91120 F20110113_AACRBY UFE0004787_00001.xml
80fc378db1cbeb51764759393d3a4990
29af670d3efd12462af031e8bbd4524c2abe5473
156703 F20110113_AACQSC goodfriend_d_Page_31.jp2
6c6a0add2b140aeacfe629cfedac1c68
b18de7a8ba7b7519a7c1c8d7fb7b6c8f33e7be08
354386 F20110113_AACQNG goodfriend_d_Page_30.jp2
485cc74bc84036b18ce5e49ba1dc6309
d7dfb1f997034deacabfc14756a3d31e3ff240b4
1534 F20110113_AACQXA goodfriend_d_Page_08.txt
988e49e82206340626213e337ed231d5
5f9d7093b73f38eba9d18f1bde3cac977ae552f8
F20110113_AACQZY goodfriend_d_Page_49.QC.jpg
94df48004479321599c4f3be6ad7a637
77a5a85d3dabdd65d77a029060346a81e5ea5207
16080 F20110113_AACRBZ goodfriend_d_Page_04.QC.jpg
3fbc14cb5856e498b92f833acc5ba6d1
c080ba15d25270f8ebb67097be184e43b50d8720
351489 F20110113_AACQSD goodfriend_d_Page_32.jp2
3e8693ca78a4a907a951fc3c18d0ed60
c660c40411db63773a07064def04802096630825
1831 F20110113_AACQXB goodfriend_d_Page_09.txt
e28e165df1bab5826870c776661a89fe
d2afda6178bdc2df41ea2ef1d0691af4afb719ac
4800 F20110113_AACQZZ goodfriend_d_Page_22thm.jpg
cc12418ccf9a50dbae06ae8e7a9a0bde
5bffc34392cedabb7761ba1afd3f0e5b88055fd0
90612 F20110113_AACQSE goodfriend_d_Page_34.jp2
70bd8de34427108ec6dfc3797d1b2dca
a76a14835e29775ea16741d1c79c06b7b1f45d7d
28043 F20110113_AACQNH goodfriend_d_Page_56.QC.jpg
2720dcb5ac5f0e255ee5529e19eb3935
96427d2e7b1135c71f028e3e62b50feaac81d3f8
2034 F20110113_AACQXC goodfriend_d_Page_10.txt
c5f6c1f320d4a192d6886a2f328ddca1
d5a55be30e46b26165caa523bfa53d03e239ba6e
82520 F20110113_AACQSF goodfriend_d_Page_35.jp2
dfd8b80c2295efb02c313fde48b1ec6f
8c3c5d225e607f3e1bd34549565cba9b09ce29d8
F20110113_AACQNI goodfriend_d_Page_16.txt
951b73d01a47fcaa62dd312177417422
7db0be955e06876a87ae27f798954858bec0d5e5
1184 F20110113_AACQXD goodfriend_d_Page_12.txt
03999e5023b6643aca878793df5a5735
57c7fbbff59448a881628be8d72f8d87116aa866
69735 F20110113_AACQSG goodfriend_d_Page_36.jp2
b7ee0ee07558a859a3d23a25f0265112
c3300108fa79c112fe4a0ba92e1965e2d29a1bdb
6322 F20110113_AACQNJ goodfriend_d_Page_20thm.jpg
e1330348819b1cf1a806a791fd5c1224
e0684f766d28fd3ee243378022ebe5f163371898
1786 F20110113_AACQXE goodfriend_d_Page_13.txt
90bf00df22ce41b3529779a165cc0dcb
f5da478ff96c2480f93591d6bdbac00cd226a3bc
67198 F20110113_AACQSH goodfriend_d_Page_37.jp2
6b16032c64fac3e3d06d821549b5a0d8
3a242961597e3e3a512c6ad968c5c642893f909c
51658 F20110113_AACQNK goodfriend_d_Page_10.pro
140f5813db3b5382fe7193316ce0cb21
c5ed58726e2657a4b7025326503e5f6e648c36fa
1866 F20110113_AACQXF goodfriend_d_Page_14.txt
f10ebd79885c2ff4e731d2739d4a062e
634be4a99ae66a918ff90ccfe87c78b0847ee86b
68756 F20110113_AACQSI goodfriend_d_Page_38.jp2
e96ed4e00a74573c7a998e101faede33
9134e52256826f2c518160b200b1746d88895810
F20110113_AACQNL goodfriend_d_Page_36.txt
150fc207d25358169db1385bf93d16a9
4312c7560b967acd2d219ee4c9756b45b6756add
1563 F20110113_AACQXG goodfriend_d_Page_15.txt
e506459f5ebc086cb3c0fa4aff9bd900
26cdcc7112c19e555c51a18d12a72a377fe6850a
64480 F20110113_AACQSJ goodfriend_d_Page_39.jp2
381dad4ec2d5bdc1296beb8c7a27e040
28b50f9a4c8505656f9eea41c34939627f4e29ca
2205 F20110113_AACQNM goodfriend_d_Page_54.txt
76792870446cfe260c68f2cecd1e0c9a
6b53ddf2ab0371321a410681c9d100c21b6bc044
2014 F20110113_AACQXH goodfriend_d_Page_17.txt
6a022325ca0ea524a11072a29c794020
091a3bbe84178f9ce942567f7e49cecfe1aacb6c
85739 F20110113_AACQSK goodfriend_d_Page_41.jp2
da1ea2977ac611bedc2a25f55dbf07cc
0d15e032e7df7cab53ff511aec986f3e529bfaeb
49197 F20110113_AACQNN goodfriend_d_Page_11.pro
e7cdd42e44ea654f648dc347bdef390c
93a8c59f681bace3422f839ac83bfa66f4d58722
1929 F20110113_AACQXI goodfriend_d_Page_18.txt
8a8547ea7ad43e1e50b576865c9db726
471e7a0b20bf16182856acbaab58765827f70196
98230 F20110113_AACQSL goodfriend_d_Page_42.jp2
ecbff46e4ea4ca50f23575c8f31dcb42
93fa840a31d59828bcaabc5cc23274fa5e217baf
15182 F20110113_AACQNO goodfriend_d_Page_33.pro
3feed5259805d0ad1fa19ac78042106c
9d50bdbf5f575aa8cf960b18047e86691d4435d4
1878 F20110113_AACQXJ goodfriend_d_Page_19.txt
845c48cecb52231b28e9fe7f40f4129e
20fe11c33e1788546ecc057cf32437b053323bd9
113352 F20110113_AACQNP goodfriend_d_Page_47.jp2
50fdca7cf1a528fc13a9de08444872d6
a22d710ce93b4fbf177914d5494338dc7bdddb3a
1921 F20110113_AACQXK goodfriend_d_Page_20.txt
bea1e4b9b359630295763aaeaf087cf1
8ac758b4b0ca5c0ef98c11c679dde059370694b3
34236 F20110113_AACQSM goodfriend_d_Page_43.jp2
17498c22bbb6fcac56ffb9e0ef84c9c4
58a990ed946466fdd0ba08be301f95a62cab8cbb
11478 F20110113_AACQNQ goodfriend_d_Page_35.QC.jpg
bc391294d152b3bb390f9f6e6e0100e1
ab236f69b342c4849ad7e2d2d824b07c08b58b09
2079 F20110113_AACQXL goodfriend_d_Page_21.txt
f0bd60132b8ff577c386ab1246ca4c0a
575e2a40106cc4051f2e3bab042f602ee0350c50
87215 F20110113_AACQSN goodfriend_d_Page_44.jp2
4e68c4ab1e2ff76efa7e785709b6e5e1
77ed2dd343bb5e149d2d567d9043a5d6b293259c
1327 F20110113_AACQXM goodfriend_d_Page_22.txt
bac0f8f56d21ac32e32561b44d7966c0
4199c45ec3ccdaa17c8ada1593054fbc9cab8b89
102332 F20110113_AACQSO goodfriend_d_Page_45.jp2
35a73507c5290f0b9b1bf51369644acc
e5889e3ccf2ec5da5e33962a1bf39cdb5e0531de
22756 F20110113_AACQNR goodfriend_d_Page_11.QC.jpg
c9ae032dfb6f119f14beaa2c83a66029
fc067660dbf5e9cf1c7ddd818dc4821b79773e26
1707 F20110113_AACQXN goodfriend_d_Page_23.txt
3285b6e0c4de166f25fd0c2d63ad91f8
6990aada0bfd7961b1fc3f247557b71321d5d573
107862 F20110113_AACQSP goodfriend_d_Page_46.jp2
391c2906cb210590e8eb88f730e6c757
ce7d4cf0f532a29c967a509667658832678e4b06
6068 F20110113_AACQNS goodfriend_d_Page_58thm.jpg
5aa9b7e3a20deae6b798ebf60b8a47ea
28dadb229b0b6f2ea91ca9e5486fc8366263b599
2015 F20110113_AACQXO goodfriend_d_Page_25.txt
ee1afe3a5541e0405d50744b93c3b4e9
2e1288d508afa49600a7c5a18874989ca4559616
116687 F20110113_AACQSQ goodfriend_d_Page_48.jp2
d5f8a784ea27132c76b865467600bacd
2406f20c48699e2533dcb6f1f39c4f99b8a9e979
9627 F20110113_AACQNT goodfriend_d_Page_33.QC.jpg
582f40d427c82d9c48596739da4692b5
5d63f83f8e8a45def78bcc7a93e7aa48fc1f3a68
1969 F20110113_AACQXP goodfriend_d_Page_26.txt
8dd114482b57240b08a4419003f930f4
64a94868e3070b12c88ac3f077bf281213088517
110552 F20110113_AACQSR goodfriend_d_Page_49.jp2
4cfff1e98075e7ef8f20cfe7dd9ce921
aca4a483e26b2cff795d3c1cb06766ae66092439
24048 F20110113_AACQNU goodfriend_d_Page_17.QC.jpg
4390298dbc8ed7af225fd1caf2930e35
2ad58a3b5a30478c754a74152753f1b500c642e2
1957 F20110113_AACQXQ goodfriend_d_Page_27.txt
6701915a78bd1b3ffd406b908a3e9296
7de030fa7c81ddf777e4c9b132ee5154de2583ac
108110 F20110113_AACQSS goodfriend_d_Page_50.jp2
45269ab06589764bda4c347d5d5caefb
c40fded7e92f116451f1ef43b571ec967190b714
F20110113_AACQNV goodfriend_d_Page_46.tif
75d584e00ea5a873f25a427eac2f27fb
2883047f573ce0e59fdb39415be9c18624fdb114
109856 F20110113_AACQST goodfriend_d_Page_51.jp2
cf0d334d6fd9554b9fd94b4fca312754
e1b02afd3bea8b87dab56405eb97b931798afa56
1466 F20110113_AACQNW goodfriend_d_Page_32.txt
12db1d7d9b88bf4c407127183ef3b899
3d9eaf197fd37cbf30b701a519ab8981a25602b0
1324 F20110113_AACQXR goodfriend_d_Page_28.txt
5d0492e1c89ee2ee204bae2ba0aaa4a9
b70b2d3256cc6a246aa7be3375f2e445bddecfd7
108905 F20110113_AACQSU goodfriend_d_Page_52.jp2
7f7109983812b4f744ca3ba5bf448dfa
2f09a751a80a1dc53a02ed6fb490f0b923c1c8dd
348 F20110113_AACQNX goodfriend_d_Page_31.txt
fbf4ee8af5eb7535ad5754ef9b3c61b8
20bb7ce354a1815b71d31fbfbae82cfc6b19f73d
932 F20110113_AACQXS goodfriend_d_Page_29.txt
120437bd79851c8cb558b7ed6a11d025
21650ea82a2054a2fd05b11d4c4245cea7ff1a04
105459 F20110113_AACQSV goodfriend_d_Page_53.jp2
ec3aac12220ac69fcef89a016a1e4b66
78f081d9e7bd07da023b75d3ffbb4949ba10b77e
52218 F20110113_AACQNY goodfriend_d_Page_21.pro
8cd92dc721f5cea6ce37bd97d174c64a
aa07892d35edf97d99b2917ea4bc647fe93e0fb7
595 F20110113_AACQXT goodfriend_d_Page_30.txt
62cc2e268c11e056bc635d0a85f1563c
41e752326be9319a50f120725130d62e42fe3616
1051979 F20110113_AACQSW goodfriend_d_Page_54.jp2
e0394eb42aa336f9db77868e1594f8db
b7d8bde9cedf89bba666681fd7bc50de222a475c
49880 F20110113_AACQNZ goodfriend_d_Page_26.pro
496945ac3bb14c49f8e5ea129e17cf9c
ecde455a9181bb2e75e0d94683787b845af65d7f
740 F20110113_AACQXU goodfriend_d_Page_33.txt
13fca576c5dfcea5150545cb5bc5afd8
8a2dbef6ce447d34ecd97e5adafcf3e6d3bcadb6
F20110113_AACQSX goodfriend_d_Page_55.jp2
0ca2745735c751c5917a1ac750b5afe1
170c6bf4bdac27b69224ee4b6874fbf6b0bda047
2093 F20110113_AACQXV goodfriend_d_Page_34.txt
6d6eaedaec4dd245e94148c092ec7955
6a180279ddda9a9d1d5030b5a9401c6839449ba9
37691 F20110113_AACQQA goodfriend_d_Page_29.jpg
822d874ebe1d77f324421bb4be40afad
37c4058c945f95b3a3db5a225d8fa917b53a3392
1051938 F20110113_AACQSY goodfriend_d_Page_56.jp2
a76f18187f2394e96af455c56f5c5af2
dc20d1f1cbf8ca1b3fa4f50d85714e751cc14708
1894 F20110113_AACQXW goodfriend_d_Page_35.txt
274cee57ba84610ace5a4d7d611b85d0
b2823654f49318760e18f2fdecc92b7c76486646
30999 F20110113_AACQQB goodfriend_d_Page_30.jpg
c70d839a36e566796255900acaf22a2a
dff6ababe36776b43f9e1737eed8a7f4664af51a
117672 F20110113_AACQSZ goodfriend_d_Page_57.jp2
cea9de6c67d8cc3a220bf800f900180a
e7ada796a8b76c34fabf4820a8f56ab4e5940c93
1380 F20110113_AACQXX goodfriend_d_Page_37.txt
2902ad86d9e7062556dd83b9d15cf931
0e4a7894b4407599c32f51f1134c75ce75811cc3
18307 F20110113_AACQQC goodfriend_d_Page_31.jpg
2fd71082141f5580eff02d46fd543903
eb8008feaa01c27c7847c364702c9b3487c7ae32
1416 F20110113_AACQXY goodfriend_d_Page_38.txt
dfec5263da622a882543717f6c79538d
94906ed9ad69fc2e9f19c91ea594f8cfb3b729bc
30757 F20110113_AACQQD goodfriend_d_Page_32.jpg
7cb3154aec33a476020c4962a73a4d5a
566ec9a666a5e7ed8748794b1cf4c1c72a244d8b
66834 F20110113_AACQVA goodfriend_d_Page_04.pro
0001b4e003f8317507bb3ad257b2382f
27335aec7cb8ce3e701bfa68ea766de90d74ecd5
1385 F20110113_AACQXZ goodfriend_d_Page_39.txt
3f564c6e4cc8b11aab4d5e4f863a2f91
7d7fe18874ecfc6b812e03e055b44c4c7f7ac73c
29633 F20110113_AACQQE goodfriend_d_Page_33.jpg
8f7f1cdafa9b19e1475a1934cd89a50c
e7e4a3a7bfeaf9f6db256519661f2a989dd73a43
49516 F20110113_AACQVB goodfriend_d_Page_05.pro
c84e9574d34933869478d457338e2127
e2c8a333ae5812aeb905a808f39aeb50fe954380
39511 F20110113_AACQQF goodfriend_d_Page_34.jpg
19657e68aa5ec78ee676bc9b83453657
3ddc3950be6606632baeb41046e8a131efc54783
38231 F20110113_AACQVC goodfriend_d_Page_07.pro
84db63ef144391fa54488c919f5e2339
3a15e2acf44656af465ea575fc2cf2c8f94dbaaa
50070 F20110113_AACQQG goodfriend_d_Page_36.jpg
969bd7c8865eb4e8e5e7a602b9cd8d98
ea48b4b8438e53407c24c859495dbd06c5ecf16f
4542 F20110113_AACRCA goodfriend_d_Page_04thm.jpg
5749e0936c53cc583ca0bdf9e5c93bb4
04028cd0d82ba839cd9a955915a76e2d9068507b
38504 F20110113_AACQVD goodfriend_d_Page_08.pro
7661cd50a6918a8336e88623c8145cea
81a000d734a532b3947dbd8e22f3af239ab7082a
21038 F20110113_AACRCB goodfriend_d_Page_09.QC.jpg
a7b79d92a2f25eed296c8b86d0598241
0d367d44412c709d956dcc2d64b8f90717f0ec90
48401 F20110113_AACQQH goodfriend_d_Page_37.jpg
f2ed1055c4d648193ed4ff2e408b2d02
74621b9f227b7d769334ffacb2832f088d6b6fa0
45426 F20110113_AACQVE goodfriend_d_Page_09.pro
833978abc3a5f53e55873dcc63eedc25
9251f5bf8b708ec3c6539bccf43843f256881546
6622 F20110113_AACRCC goodfriend_d_Page_26thm.jpg
353da325995bfcb3ab630396009d1f37
f140e248dbfb0b3b7c101e37f9a85c56295fcc8d
48038 F20110113_AACQQI goodfriend_d_Page_38.jpg
92567618d765235a46fe8be82ceecd00
2044a571178df7a42d9c952058996c1ac6592ccd
29631 F20110113_AACQVF goodfriend_d_Page_12.pro
f22b32714003793a82ee583d11da50ab
6d9acfdfa3b85cc8058b300600d44859de6f3ec6
23562 F20110113_AACRCD goodfriend_d_Page_27.QC.jpg
f8fe571f8d0755a8ba01d41c58608658
09fd696223193dbd9046b83676f466554f5beac5
45882 F20110113_AACQQJ goodfriend_d_Page_39.jpg
77a0c04680ffb1fa7c880dea66947d9e
c240378c08a67baab20ebf5c7b08a681f2994f9d
44135 F20110113_AACQVG goodfriend_d_Page_13.pro
c908252f178e7265307b23c884407842
98dd5286fadf707eb4ab442f84cfccf32df06e5c



PAGE 1

THE RELATIVE VALUE OF SEAGRASS, MARSH EDGE, AND OYSTER HABITATS TO THE BRACKISH GRASS SHRIMP, Palaemonetes intermedius ALONG THE GULF COAST OF FLORIDA By DANIEL SCOTT GOODFRIEND A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004

PAGE 2

ACKNOWLEDGMENTS There are probably many more people who deserve recognition for their help and encouragement than I have space for here, so let me begin by thanking everyone who has helped me in deciding on a career path, thinking more critically about the natural world, and enabling me to fulfill my dreams and goals. This applies particularly to the faculty of the Department of Wildlife at Humboldt State University, who first opened my eyes to the possibilities that present themselves when we apply the scientific method to better understand the natural world. I want to thank my advisor, Dr. Thomas Frazer, for his invaluable help. Dr. Frazer has proved to be a continuing source of support and encouragement, without which my graduate education would never have been as rewarding or challenging. His flexible and open style as an advisor provided me with the intellectual space I needed to learn to think more critically about ecological problems, and his high standards forced me to become a better ecologist. The other members of my committee, Dr. William Lindberg and Dr. Charles Jacoby, also deserve special thanks. Dr. Lindberg was particularly helpful in the early stages of this project, providing me with key sources of information on crustacean biology as well as on the philosophy of the scientific method in general. Dr. Jacoby was especially helpful nearer to the end of the project, substituting for Dr. Frazer as a proofreader of early drafts of this thesis. He provided me not only with suggestions, but potential solutions to the problems I encountered. ii

PAGE 3

There are many other people who proofread early drafts of this paper. For this I need to thank Rikki Grober-Dunsmore, Jason Hale, Chanda Jones, Kate Lazar, Darlene Saindon, and Deborah Schwartz. Dr. Kenneth Portier was also invaluable for his advice on statistical design and data management techniques. Many people lent me time and equipment which helped the field portion of this project to be successful. When this project was in its infancy, Thomas Glancy introduced me to what would become my study sites, provided important background data, and taught me to differentiate the species of Palaemonetes I would encounter. I also wish to thank the entire staff of the St. Martins Marsh Aquatic Preserve, in particular Seth Blitch and Chad Bedee, who graciously provided transportation to my shallow-water field sites on their airboat and assisted as field workers. Carla Beals, Jaime Greenwalt, Rikki Grober-Dunsmore, Jason Hale, Stephanie Keller, Kate Lazar, Benjamin Loughran, Sky Notestein, Troy Thompson, Chris Tilghman, and Duncan Vaughn also assisted with the field work, sometimes for long hours on very cold and rainy days. They deserve both thanks and credit for their work. I especially want to thank Deborah Schwartz for her sense of humor and encouragement which helped me to remain optimistic and to continue my research during those inevitable times in the course of a project like this when all hope seems to be lost. iii

PAGE 4

TABLE OF CONTENTS Page ACKNOWLEDGMENTS..................................................................................................ii LIST OF TABLES...............................................................................................................v LIST OF FIGURES...........................................................................................................vi ABSTRACT......................................................................................................................vii INTRODUCTION...............................................................................................................1 STUDY AREA....................................................................................................................5 METHODS..........................................................................................................................7 Field and Laboratory Methods......................................................................................7 Statistical Methods......................................................................................................13 RESULTS..........................................................................................................................15 Temporal Variability..................................................................................................15 Within Habitat Variability...................................................................................15 SMMAP Cross-Habitat Comparisons (Seagrass vs. Marsh Edges)....................17 Among Estuaries Comparisons...................................................................................19 DISCUSSION....................................................................................................................36 Determination of the Relative Quality of Habitat Types............................................36 Temporal Variability within the SMMAP..................................................................36 SMMAP Within Habitat Sampling.....................................................................36 SMMAP Cross-Habitat Comparisons (Seagrass vs. Marsh Edge)......................40 Among Estuaries Comparisons...................................................................................42 Overall Conclusions....................................................................................................43 LITERATURE CITED......................................................................................................46 BIOGRAPHICAL SKETCH.............................................................................................50 iv

PAGE 5

LIST OF TABLES Table page 1. Water chemistry at SMMAP by sampling period..........................................................25 2. Temporal changes in variables characterizing samples of P intermedius taken from seagrass at SMMAP.......................................................................................................26 3. Temporal changes in variables characterizing samples of P intermedius taken from marsh edges at SMMAP.................................................................................................27 4. Differences in variables characterizing samples of P intermedius taken in different habitats at SMMAP in the July/August 2002 and 2003 sampling periods....................28 5. Differences in variables characterizing samples of P intermedius taken in different habitats at SMMAP in the November 2002 sampling period........................................29 6. Differences in variables characterizing samples of P intermedius taken in different habitats at SMMAP in the May 2003 sampling period.................................................30 7. Differences in variables characterizing samples of P intermedius taken in different habitats at SMMAP in the February 2003 sampling period..........................................31 8. Differences in variables characterizing samples of P intermedius collected from different habitats in the Weeki Wachee estuary in August 2003. .............................32 9. Differences in variables characterizing samples of P intermedius in different habitats in different estuaries during August 2003.....................................................................33 10. Characteristics of seagrass vegetation in different estuaries in August 2003..............34 11. Water chemistry in the SMMAP, Weeki Wachee and Steinhatchee estuaries in August 2003.................................................................................................................35 v

PAGE 6

LIST OF FIGURES Figure page 1. Significant temporal changes in abundance and ISF variables in seagrass at SMMAP (error bars represent 5 th and 95 th percentiles in a and b and 95% confidence intervals in c, d, and e).............................................................................................21 2. Size distribution of shrimp populations in seagrass beds in different sampling periods..22 3. Temporal differences in median size of shrimp from marsh edges at SMMAP (error bars represent 5 th and 95 th percentiles).....................................................................23 4. Size distribution of shrimp populations in marsh edges in different sampling periods.24 vi

PAGE 7

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science THE RELATIVE VALUE OF SEAGRASS, MARSH EDGE, AND OYSTER HABITATS TO THE BRACKISH GRASS SHRIMP, Palaemonetes intermedius ALONG THE GULF COAST OF FLORIDA By Daniel Scott Goodfriend May 2004 Chair: Thomas Frazer Major Department: Fisheries and Aquatic Sciences Palaemonid shrimp are abundant and important components of estuarine faunas throughout much of the world, resulting in considerable interest in their ecology. Studies of the relative value of habitats to these animals have primarily used measures of relative abundance or habitat selection by individuals as indicators of habitat quality. However, these have produced varied and conflicting results. As such, new techniques are needed for investigating the relative quality of habitats for grass shrimp. This study examined the relative value of seagrass beds, marsh edge habitats, and oyster bars to the brackish grass shrimp, Palaemonetes inte rmedius using a variety of metrics that either directly measure, or provide surrogate measures for, fecundity, mortality, and growth rates. These metrics were compared with measures of relative abundance to determine whether abundance might be an effective indicator of habitat quality. Sampling was done during July, August, and November 2002, and February, vii

PAGE 8

May, July, and August 2003, in the St. Martins Marsh Aquatic Preserve (SMMAP) near Crystal River, Florida. This provided an assessment of temporal variability in the quality of these habitat types as well as their overall relative value. In August 2003, the study was broadened to include the Weeki Wachee and Steinhatchee estuaries in order to assess the degree of regional variability in the quality of these habitats for grass shrimp. In contrast with several other studies, P intermedius was never collected from oyster habitat. In the SMMAP, seagrass beds provided higher quality habitat for grass shrimp than marsh edges in the July/August, November, and May, when seagrass beds had greater abundances of shrimp, as well as higher proportions of gravid females than marsh edges. In February 2003, the two habitats were determined to be of similar value to grass shrimp. As in the SMMAP, seagrass beds in Weeki Wachee provided a higher quality habitat than marsh edges for grass shrimp. Within the Steinhatchee estuary, however, the quality of marsh edge habitats was similar to, or marginally higher than, that of seagrass beds. Differences in seagrass characteristics, species composition specifically, may account for the lower abundance of P intermedius in the sampled seagrass beds within the Steinhatchee estuary. In general, however, grass beds were determined to provide higher quality habitat than adjacent marsh edges in the broad study area. viii

PAGE 9

INTRODUCTION Grass shrimp ( Palaemonetes spp.) are abundant components of coastal benthic communities throughout much of the world, including along the Gulf and east coasts of the United States (Rozas and Minello 1998, Lewis and Foss 2000, Glancy et al. 2003). They are important consumers of organic detritus, and in turn serve as prey for a myriad of commercially and recreationally important fish. As such, there has been substantial interest in their ecology, and in particular, their use of habitats. A habitat should be defined as high quality if the animals inhabiting it have increased fitness relative to those in an alternative habitat, as indicated by measures of fecundity, mortality, and growth rates (Anderson and Gutzwiller 1996, Davenport et al. 2000, Franklin et al. 2000, Lin and Batzli 2001, Luck 2002, Walters et al. 2002, Ross 2003). Delineation of high quality habitats is, therefore, an important step toward the effective management and conservation of animals, including grass shrimp. There have been numerous attempts by ecologists to determine the relative quality of estuarine habitats for grass shrimp (Sheridan 1992, Knowlton et al. 1994, Khan et al. 1995, Eggleston et al. 1998, Rozas and Minello 1998, Bass et al. 2001). The vast majority of these studies have used relative abundance, density, or habitat selection by individuals as indicators of habitat quality. These studies have substantially increased our understanding of many aspects of the relationship between grass shrimp and their habitats. For example, it has been demonstrated that congeneric species may select different habitat types when given a choice in experimental and field settings (Knowlton 1

PAGE 10

2 et al. 1994, Khan et al. 1995, Sheridan 1992), and that paleomonid shrimp can occur in greater abundance in adjacent, but alternative habitat types within a given ecosystem (e.g. Eggleston et al. 1998, Rozas and Minello 1998, Glancy et al. 2003). Findings in some studies suggest, however, that grass shrimp are more abundant in seagrass beds (e.g. Glancy et al. 2003), whereas others suggest that grass shrimp are more abundant in marsh habitats (e.g. Rozas and Minello 1998). Others have shown seasonal shifts in the relative abundance of grass shrimp between habitat types (Eggleston et al. 1998). These inconsistencies imply that the processes determining habitat quality for, and relative abundance of, grass shrimp may be fairly complex and variable over both time and space and that new approaches are needed to investigate the relative quality of estuarine habitats for grass shrimp. Van Horne (1983) noted that there are several levels of increasing sophistication by which managers delineate habitat for species of concern, but that most often, especially for nongame species, relative density or abundance are used as indicators of habitat quality. As noted above, these indicators are often used when investigating habitat use by grass shrimp. However, as Van Horne (1983) pointed out, and as is apparent from other studies, e.g., Schantz (1981), density and abundance are not always correlated with habitat quality. In fact, density and habitat quality may be decoupled in six specific types of situations: (1) when habitat use changes seasonally, (2) when social dominance interactions result in large densities of subdominant individuals in suboptimal habitats, (3) when resources and other environmental conditions are temporally unpredictable within habitats, (4) when the species have high reproductive capacity, which can result in population sizes that poorly reflect the carrying capacity of the environment over the

PAGE 11

3 short term, (5) when there is spatial habitat patchiness and (6) when the species in question is a habitat generalist, which can cause the spatial distribution of habitats to have a greater influence over the distribution of populations than the relative quality of those habitats. Interestingly, none of the papers cited in Van Hornes review were of studies from estuarine or marine systems, or of invertebrates. Also, none of the studies of grass shrimp, cited above or otherwise encountered, examined any potential indicators of habitat quality other than relative abundance or habitat selection by individual shrimp. Nonetheless, many life history and ecological characteristics of grass shrimp comply with Van Hornes criteria. Seasonal shifts in habitat use by grass shrimp have been documented (Kneib 1987, Eggleston et al. 1998), grass shrimp live in estuarine environments that exhibit pronounced variability, and also invest significantly in reproduction (Vernberg and Piyatitivorakul 1998). Moreover, there is considerable patchiness in the distribution and species composition of seagrass beds (Hovel et al. 2002), one of the primary habitats in which grass shrimp are found (Glancy et al. 2003). Finally, grass shrimp are generalists, tolerating wide ranges in many physiochemical parameters (e.g. temperature and salinity) and utilizing multiple food sources (Vernberg and Piyativorakul 1998). In combination, these life history characteristics and ecological attributes suggest that the primary methods used by estuarine ecologists to investigate the relative value of habitats for grass shrimp, i.e.,,, preference and relative abundance, may be inadequate. Ideally, the relative quality of habitats for a species should be measured via comparison of population densities or relative abundance and demographic parameters such as

PAGE 12

4 fecundity, mortality, and growth rates. This was done by Chockley and St. Mary (2003), who identified and monitored individual banded coral shrimp ( Stenopus hispidus ), and showed that low density, inshore populations produced more eggs per area of habitat than high density, offshore populations, which implied that the inshore areas were higher quality habitats despite the fact that they supported lower shrimp abundances. In an attempt to extend this type of habitat assessment to grass shrimp, this study looks more closely at the relative values of key habitats to the brackish grass shrimp, Palaemonetes intermedius in terms of abundance and other measures that should indicate relative fitness of the shrimp within those habitats, termed individual shrimp fitness (ISF) variables. These ISF variables include measures of fecundity, growth, and several surrogate measures of mortality which were selected based on relevant literature. Abundance estimates were then compared to the ISF measures of habitat quality to determine whether or not abundance was an effective indicator of habitat quality for grass shrimp.

PAGE 13

STUDY AREA Sampling was conducted primarily at three sites within the St. Martins Marsh Aquatic Preserve (SMMAP), along Floridas northeast Gulf coast in Citrus county (28 53 N, 82 41 W). These three sites are the same as those previously described and sampled by Glancy et al. (2003), and the close proximity of oyster bar, seagrass, and marsh edge habitats at these sites allow a comparison of habitat quality in similar environments, as characterized by water temperature, salinity, and dissolved oxygen concentrations. In brief, this estuary is a complex mosaic of seagrass beds, salt marshes, oyster bars, tidal channels, bays, and sandy flats. Seagrass beds are primarily comprised of turtle grass ( Thalassia testudinum ), shoal grass ( Halodule wrightii ), and manatee grass ( Syringodium filiforme ). Intertidal oyster habitats are dominated by the eastern oyster ( Crassostrea virginica ), are low relief, and typically have a substrate of sand, mud, and shell fragments overlying a limestone base. Salt marshes are dominated by cordgrass ( Spartina alterniflora ) and black needle rush ( Juncus roemarianus ), but are interspersed with small stands of black ( Avicennia germinans ) and red ( Rhizophora mangle ) mangroves. As a whole, the estuary receives freshwater input from freshwater springs and the spring-fed Crystal River. In August 2003, additional sites were sampled in the Weeki Wachee and Steinhatchee estuaries to determine if patterns similar to those observed within the SMMAP occur at other Florida Gulf coast locations. The Weeki Wachee estuary is located at approximately 28 32.5 N, 82 39.5 W, and is characterized by a similar 5

PAGE 14

6 mosaic of habitats to SMMAP, although with a decreased prevalence of intertidal oyster bars. As a result, no oyster habitats were sampled in this estuary. Salt marsh and seagrass flora are typically the same as in the SMMAP (Frazer et al. 1998 and 2003), and the major input is from the spring-fed Weeki Wachee River. Three sites were sampled within this estuary, with selection based on similarities in size, water flow, and depth to the sites at SMMAP, as well as the close proximity of salt marsh and seagrass habitats. The Steinhatchee estuary is located at approximately 29.5 N, 83 25.5 W, and is also characterized by a similar mosaic of habitats to SMMAP, although with a decreased prevalence of intertidal oyster habitats in close proximity to seagrass beds. The seagrass beds at the sites in this estuary tended to be even more heavily dominated by turtle grass and manatee grass than the other two estuaries, and to have a decreased prevalence of shoal grass relative to the other estuaries. Salt marshes here also tended to have a decreased prevalence of both black and red mangrove trees, and to be found predominantly along the coasts rather than occurring as islands. The Steinhatchee estuary receives freshwater primarily from the Steinhatchee River, which is fed by more surface water runoff, and fewer spring inputs, than the Crystal or Weeki Wachee Rivers. Seven sites were sampled in this estuary. Three of these had a close association of seagrass and marsh edge habitats and were selected based on similarity in size, water flow, and depth to the sites sampled in other estuaries. Three of the sites in this estuary only contained seagrass, but were similar to the others in all other respects. The seventh site contained only intertidal oyster habitat.

PAGE 15

METHODS Field and Laboratory Methods Sampling within SMMAP was conducted during July, August, and November 2002, and February, May, July, and August 2003. As there were no statistically significant within year or among year differences in any of the variables measured during July and August sampling periods in 2002 and 2003, summer sample data were pooled and are hereafter referred to simply as the July/August sampling periods. After the May 2003 SMMAP sampling period, it was clear that the only sampling period during which all the ISF and abundance variables used in this study could be estimated in all habitats was July/August, which the data suggested was the primary breeding season, a conclusion that was expected based on relevant literature. Therefore, in August 2003, the sampling was broadened to include the Weeki Wachee and Steinhatchee estuaries to determine whether the patterns observed at SMMAP were representative of the Gulf coast region as a whole during the grass shrimp breeding season. Sampling of seagrass vegetation (see below) was also intensified during this period. Sampling, at all times and locations, was performed within 3 h of low tide to minimize the variability associated with changes in habitat use due to tidal fluctuations. This schedule facilitated comparisons to data collected and reported by Glancy et al. (2003) who also sampled during low tides. 7

PAGE 16

8 Seagrass beds, oyster bars, and marsh edges were sampled at each study site in each time period with ten 5-m sweeps with a standard D-frame sweep net 25 cm high by 35 cm wide. Sweep net sampling has been previously used in several estuarine habitat types to compare the relative abundances of fauna (e.g. Young et al. 1976, Young and Young 1978, Posey and Hines 1991, Townshend 1991, and Posey et al. 1999), and controlled tests for potential habitat bias using the sweep net collection method indicate that the method is appropriate for the habitats samples in this study (T.K. Frazer and M.H. Posey, pers comm.). Locations of sweeps in seagrass beds and in oyster habitats were chosen in an unbiased manner by taking a random number of paces (between 1 and 30) at a random bearing from the end of the previous sweep. Marsh edge sweeps were located similarly, but did not include a random bearing, as the edge of an island is measurable only in one dimension. The initial sweep in each habitat type was selected haphazardly by sighting a location in the habitat from the boat. All Palaemonetes spp. individuals caught in each sweep were put in a sealable plastic freezer bag, placed on ice, and transported to the lab for further analysis. In seagrass beds, after every third sweep, a 0.25 m 2 quadrat was placed on the benthos to estimate the percent areal cover of each vascular plant species present, as well as total grass coverage. This was primarily done looking down at the benthos from above the water line, but when water clarity was too poor for the bottom to be seen, or species to be easily identified, a mask and snorkel was used. In August 2003, the plant survey sample size was increased to 10; estimates of percent areal cover were made at the end of every sweep. In August 2003, a small ponar grab was used to take ten 625 cm 2 bottom samples from each habitat/site, to an approximate sediment depth of 8 cm. Samples were

PAGE 17

9 individually bagged in sealable plastic freezer bags and brought back to the lab for analysis. A YSI electronic meter model 650 was used to measure salinity, temperature, and dissolved oxygen (DO) concentrations in each habitat type within each site at a depth of approximately 0.25 m below the surface. All grass shrimp within each of the bagged sweep samples were enumerated to allow for estimates of relative abundance. Subsequent processing of shrimp collected in the sweep nets was carried out with a random subsample of 50 individuals. The shrimp for the subsample were selected randomly by counting all shrimp from the 10 sweeps associated with a given habitat/site/sampling period and assigning each a number. Then 50 random numbers between 1 and the total number of shrimp counted were selected and those shrimp were selected for the subsample. This subsample size was determined to be sufficient to detect differences between samples in all variables after measuring the variability associated with measurements from the first 200 shrimp. If less than 50 shrimp were caught from a specific habitat type within any given time period, all shrimp were processed as described below. Shrimp were identified as either Palaemonetes pugio or P intermedius according to Abele and Kim (1986). The relative proportion of P intermedius in each of the subsamples was multiplied by the total Palaemonetes counts from the respective sweep samples to estimate the abundances of P intermedius In the laboratory, all subsampled P intermedius were measured for total length. In addition, the sex, reproductive condition (gravid/not gravid), and clutch size of gravid females was recorded. The total length of each shrimp, as well as telson lengths of the first 200 shrimp, was measured using a WILD M3Z dissecting microsope fitted with a KR 221 120-increment optical micrometer. Measurements were recorded to the nearest

PAGE 18

10 0.15 mm. Shrimp were considered female if they were gravid, or if developing ovarian tissue could be observed through the translucent body of the individual. If these indicators were not observed, then the second pleopod of the shrimp was examined to determine if it bore an appendix masculina, a male characteristic (Berg and Sandifer 1984). Shrimp without this modification of the second pleopod were counted as female. Shrimp with a total length less than 12 mm were not included in estimates of sex ratio or fecundity values as shrimp this small may not have been sexually mature, in which case the appendix masculina may not have developed. The clutch size of gravid females was determined by teasing apart egg masses with a fine metal probe and counting individual eggs with the aid of a dissecting microscope. For the bagged vegetation samples from the ponar grabs, the above-ground vegetative portions were separated from the mud, detritus, rhizomes, and roots by hand. The below-ground portion, as well as nonliving material, was discarded. Seagrass shoots were then counted and total number of shoots for each seagrass species recorded. Above-ground biomass (wet weight) of each species of seagrass in the ponar samples was recorded and weighed with a Pesola 1000 g hanging scale and measurements recorded to the nearest 10 g. An instantaneous growth rate technique similar to that employed by Quetin and Ross (1991) was used to estimate shrimp molting rates and growth increments in SMMAP during November 2002, and February, May, and August 2003 in SMMAP only. P intermedius from each habitat/site in each sampling period were collected with sweep nets and/or a beach seine. The target sample size for each habitat/site combination was 100 shrimp. When shrimp densities were too low to collect 100 shrimp, the number of

PAGE 19

11 shrimp caught in approximately 1.5 h was used. Individual shrimp were placed in pre-labeled 265-ml glass jars with mosquito-netting tops secured by rubber bands. The jars were placed in plastic tubs and left immersed in a readily accessible nearshore area that had similar salinity, temperature, and DO values as the sites from which the shrimp were collected. These jars were monitored every 12 h for 3 d, and the shrimp that molted, as well as their exuviae, were preserved in a solution of 90% ETOH and 5% glycerin, and brought back to the lab for analysis. The number of shrimp that either escaped from the jars or died in any given experiment (< 3% in all experiments) was subtracted from the original number of total shrimp that were used. This number was further adjusted to estimate the total number of P intermedius that were used throughout the experiments by multiplying it by the proportion of P intermediu s in the subsample from the sweep net samples which were collected in the same location during the same sampling period. The number of P intermedius that were directly observed to molt during the course of the 3-d experiment was then divided by the estimated total number of P intermedius to calculate a molting frequency per 3 d. Intermolt period (IMP) was then calculated as the reciprocal of molting frequency, and expressed as days molt -1 In the laboratory, telson lengths of the preserved shrimp and their molts were measured. Species, sex, and reproductive condition of female shrimp were also recorded. The growth increment of molting shrimp (mm) was calculated as the difference between the estimated total lengths of the post-molt shrimp and their molts. Total lengths of the molts and post-molt shrimp were estimated using measured telson lengths and

PAGE 20

12 Equation 1, which was derived from a linear regression analysis between measured telson and total lengths for the first 200 shrimp from the sweep net subsamples (p < 0.0001, r 2 = 0.87). total length= 1.92 + (7.36 telson length) eq. (1) Growth rate was defined as the average growth (mm/day) for an individual P intermedius within a sample population, and was calculated by dividing the average growth increment (mm) by the IMP (days) for each habitat/site/sampling period combination. This provided an estimate of continuous growth which was a population-wide average, individual shrimp growth is incremental. The estimated values for IMP, growth increment, and growth rate were assumed to be indicative of the sample population as a whole for the time period in which the experiment took place. The combination of field, laboratory and mathematical techniques described above resulted in estimates for a variety of variables which provide information about P intermedius These fall into two general categories. (1) measures of abundance, and (2) those variables that provide information about the relative fitness of shrimp (individual shrimp fitness, or ISF, variables). The latter include measures of fecundity (proportion of gravid females and fecundity), and surrogate measures of mortality and nutritional stress (sex ratio, size distribution, and growth rates). These surrogate measures were necessary because the mortality of a high density, small-bodied, mobile species such as P intermedius is extremely difficult to measure directly in an open estuarine habitat. The validity of these surrogates is supported by the results of previous investigations. Females have been shown to respond to decreased food availability with increased mortality relative to males (Reinsel et al. 2001). Also,

PAGE 21

13 population size structure has been shown to be shifted towards smaller shrimp in ecosystems with increased predation on P pugio and with increased mortality due to nutritional stress (Cross and Stiven 1999, Bass et al. 2001, Reinsel et al. 2001). In addition, reduced food availability can result in decreased intermolt growth increments and increased intermolt periods in crustaceans in general (Hartnoll 2001). As such, population size and sex structure, as well as growth rates, were used as the surrogate measures of relative mortality and nutritional stress for shrimp. In addition, histograms of size distributions of the shrimp from each habitat/time period were created to further explore the relationships between life history, habitat quality, and shrimp size. Statistical Methods Within each sampling period and estuary, one way Analyses of Variance (ANOVAs) were performed with each of the numerical variables (abundance, size, clutch size, growth increment, total growth, as well as percent cover, biomass, and shoot density of seagrass and individual seagrass species) as dependent variables and habitat type as the independent variable (MINITAB Release 14). When necessary, variables were log 10 transformed to achieve normality and homoscedasticity and all tests were assumed significant at p < 0.05. Tukeys post-hoc test was used to determine where significant differences existed when ANOVAs detected significant effects (MINITAB Release 14). Students t-tests were used when only 2 habitats were compared within a particular season (MINITAB Release 14). When the assumptions of normality and homogeneity of variance could not be met, a Kruskal Wallis ANOVA was used as an alternative to ANOVA and Mann-Whitney U-tests were used as alternatives to T-tests (MINITAB Release 14). When Kruskal-Wallis ANOVA was used, post-hoc comparisons were made by comparing the intervals between the 5 th and 95 th percentiles of each distribution. Only

PAGE 22

14 three values for intermolt period, one from each site within the SMMAP, could be calculated for each habitat/sampling period with the method employed in this study. To increase statistical power, the molting frequency of P intermedius (molts day -1 ) was used as a proportional variable and analyzed via 2 tests (MINITAB Release 14) with the distribution of the molting shrimp assumed to be independent of habitat or time period. 2 tests were also used to compare the other proportional variables (sex ratio and proportion of gravid females), between habitat types, time periods, and estuaries (MINITAB Release 14). When there were only two categories being compared, i.e., marsh edge vs. seagrass, overall test significance at < 0.05 was sufficient to infer significant differences between mean values. When there were more than two categories being compared, such as time periods or estuaries, and the 2 test was significant at < 0.05, then statistically significant differences were inferred by comparing 95% confidence intervals. In all cases, the variables were assumed to be independent of habitat, time period, or estuary. Sampling periods were only compared within the SMMAP, and among estuary comparisons were only made with data from August 2003.

PAGE 23

RESULTS P intermedius were not collected from oyster bar habitats during the course of this investigation. As a consequence, subsequent analyses focus only on grass shrimp in marsh edge habitats and seagrass beds. Temporal Variability Within Habitat Variability As expected, because of the close proximity of sampling sites within the SMMAP, there were no statistically significant differences in salinity, dissolved oxygen concentrations, or water temperature between habitat types within any given sampling period. Thus, data from all habitats were pooled for all statistical comparisons. Dissolved oxygen concentrations and salinity were always well within the ranges reported for Palaemonetes (Vernberg and Piyatitivorakul 1998, Stickle et al. 1989, Bass et al. 2001, Table 1). Statistically significant temporal changes in water temperature were consistent with expected seasonal patterns (Frazer et al. 1998 and 2003, Table 1), but were within the range reported for grass shrimp (Knowlton and Schoen 1984). Water temperature was significantly lower during February 2003 (12.69C) than in any other time period (Table 1). Although water temperature was higher in November 2002 (18.80C) than in February 2003, the mean value was lower than in either July/August (29.57C) or May 2003 (29.73C) sampling periods, which were similar to one another (Table 1). Within seagrass beds, there were significant temporal differences in the relative abundance of P intermedius and several ISF variables, specifically total length, sex 15

PAGE 24

16 ratio, the proportion of gravid females, and IMP. Surprisingly, there were no significant differences observed for the percent cover of any individual seagrass species or cover of seagrass in general. The median abundance of P intermedius was significantly lower in February 2003 (1.5 shrimp/sweep) than in any other sampling period (Figure 1a, Table 2), coinciding with the period when mean water temperatures were lowest (Table 1). Gravid females were not collected from seagrass beds in February. (Figures 1d, Table 2). In May 2003, shrimp were more abundant than in February (14.5 vs. 1.5 shrimp/sweep), although smaller on average (Figures 1a,b, 2b,c, Table 2). Nineteen percent of females were gravid in May (Table 2). In the July/August sampling periods, the median abundance of shrimp was similar to that in May 2003 (18.0 shrimp/sweep), but shrimp were significantly larger (Figures 1a,b, 2a,d, Table 2). In May 2003, 77.8% of the shrimp were < 15 mm total length and probably not sexually mature (Figure 2d). In comparison, 70.1% of captured shrimp during the July/August sampling periods were > 15 mm total length, indicating that the majority of the sampled population was sexually mature (Figure 2a). Changes in ISF variables were consistent with previously documented life history information for P intermedius The proportion of gravid females in July/August was 47%, indicative of a peak reproductive period (Figure 1d, Table 2). Shrimp abundance was higher in November 2002 than in summer (46.5 vs. 18.0 shrimp/sweep), though size distribution was skewed towards smaller size classes (Figure 2b), reflecting a late summer/early fall recruitment period. In November 2002, 87.1% of the shrimp were < 15 mm total length (Figure 2b), and only 23% of the larger shrimp were gravid (Figure 1d, Table 2). In addition, the IMP for the sampled population

PAGE 25

17 was 8.4 d in November, which was significantly longer than in either July/August (5.5 days) or May (4.4 days) (Figure 1e, Table 2). Marsh edge habitats exhibited less temporal variability than seagrass beds. There were no statistically significant differences in the relative abundance of P intermedius within the marsh edge habitats sampled during the different time periods in SMMAP (Table 3). In fact, no variables exhibited significant or large temporal differences other than size and the proportion of gravid females (Table 3). Shrimp occupying marsh edges were significantly larger in July/August (median TL = 22.929 mm) than in May (median TL = 12.714 mm). In November and February, shrimp were intermediate in size (15.710 mm and 16.429 mm, respectively), but did not differ significantly from the summer sampling period (Figure 3, Table 3). A majority (94.4%) of P intermedius sampled in marsh edges during summer were > 15 mm and 27% of the females caught were gravid (Table 3, Figure 4). In fact, gravid females were only captured from marsh edges in summer. During the other sampling periods, there was a much greater proportion of individuals in the < 15 mm size classes (Figure 4). In summary, while the highest shrimp abundances in seagrass beds occurred in November 2002, the ISF variables measured indicated that conditions during July/August were the most appropriate for the cross-habitat comparisons central to this study. SMMAP Cross-Habitat Comparisons (Seagrass vs. Marsh Edges) Shrimp were significantly more abundant in seagrass than in marsh edges during every sampling period (Tables 4-7). Seagrasses yielded somewhere between twice as many shrimp in February and sixty times as many shrimp in November 2002 (Tables 4-7). During the July/August sampling period, the proportion of gravid females was also higher in seagrass (48%) than in marsh edge habitats (27%). Average shrimp

PAGE 26

18 size in July/August, however, was less in seagrass beds (19.143 mm) than in marsh edges (22.929 mm) (Table 4) as was clutch size (42.5 eggs/gravid female in seagrass beds and 89.0 eggs/gravid female in marsh edge habitats). Differences in shrimp size confounded a rigorous comparison of growth parameters (Table 4, Figures 2a, 4a) during the summer sampling periods. Estimates of fecundity at the population level were made by multiplying shrimp abundance by % females in the sample population (averaged between marsh edge and seagrass values due to lack of significant differences), by % gravid females, and by clutch size. These estimates showed that the average number of eggs/sweep was 209.3 in seagrass, and 0.7 in marsh edges, which strongly suggests that seagrass provides a higher quality habitat for P intermedius. Although the July/August sampling period was determined to be the most appropriate time to compare measures of relative abundance with other ISF variables, the estimates of abundance and ISF variables during other sampling periods provided an opportunity to make additional seagrass/marsh edge comparisons. In November 2002, for example, gravid females were only encountered in seagrass (23% gravid and 28.0 eggs/gravid female), but shrimp from marsh edges were larger (9.714 mm in seagrass and 15.710 mm in marsh edges) (Table 5). When fecundity was compared at the population level, the average number of eggs/sweep was 143.2 in seagrass and 0.0 in marsh edges. In May 2003, shrimp abundances and ISF variables both indicated that seagrass provided higher habitat quality than marsh edges. IMP was shorter in seagrass beds than in marsh edge habitats (4.4 d in seagrass and 8.1 d in marsh edge), while mean shrimp size and size distributions were similar, implying faster growth in seagrass beds than marsh edges for individual shrimp (Table 6).

PAGE 27

19 Interestingly, in February 2003, there was a higher proportion of females in marsh edge habitats than in seagrass beds (47% females in seagrass and 78% in marsh edges) (Table 7). However, the difference in abundance between seagrass and marsh edges was of lesser magnitude than during other sampling periods (1.5 shrimp/sweep in seagrass and 0.7 shrimp/sweep in marsh edges) (Table 7). When female abundance was estimated by multiplying relative abundance by sex ratio, the average number of females/sweep was 0.70 in seagrass and 0.55 in marsh edges, suggesting that these two habitats were capable of providing resources for similar numbers of female shrimp. Among Estuaries Comparisons Cross-habitat comparisons in the Weeki Wachee and Steinhatchee estuaries produced conflicting results. In the Weeki Wachee estuary, abundance was significantly greater in seagrass than marsh edges (11.5 shrimp/sweep in seagrass and 0.6 shrimp/sweep in marsh edges), although there were no significant differences between habitat types for any other ISF variables (Table 8). Within the Steinhatchee estuarine area, there were no significant differences between seagrass and marsh edge habitats in either abundance or any of the ISF variables measured. Among the three estuaries in this study, Steinhatchee had the lowest abundance of P intermedius in seagrass, and highest abundance of P intermedius in marsh edges (Table 9). There were only two ISF variables which showed significant differences among estuaries. i.e., proportion of gravid females in seagrass beds and size in marsh edge habitats. In Weeki Wachee seagrass beds, the proportion of gravid females (35%) was higher than in seagrass beds in the Steinhatchee (18%) or SMMAP (26%). Median total lengths were significantly less for shrimp occupying marsh edges in Steinhatchee

PAGE 28

20 (17.429 mm) than in the other two estuaries (22.857 mm in Weeki Wachee and 26.000 mm in SMMAP) (Table 9). There were also several differences detected in the species composition and biomass of the seagrass beds among estuaries (post-hoc comparisons made by comparison of intervals between the 5 th and 95 th percentiles), which may have significance for the shrimp. The percent areal cover and shoot density of Thalassia testudinum was different in each estuary, lowest in Weeki Wachee (0.0% cover, 9.6 shoots/m 2 ), higher in SMMAP (0.7% cover, 11.6 shoots/m 2 and highest in Steinhatchee (80.0% cover, 14.4 shoots/m 2 ) (Table 10). Percent areal cover of Halodule wrightii was different in each estuary, and followed the reverse spatial pattern, increasing from Steinhatchee (1.9% cover) to SMMAP (80.0% cover) to Weeki Wachee (100.0% cover) (Table 10). Above-ground biomass of H wrightii however, was higher in SMMAP (624.0 g/m 2 ) than in Weeki Wachee (280.0 g/m 2 ) (Table 10). While there were statistically significant differences in salinity and dissolved oxygen concentrations among the estuaries, these were well within the ranges reported for P intermedius There was not a significant difference in water temperature between the estuaries (Table 11).

PAGE 29

21 Fig 1a-e. Significant temporal changes in abundance and ISF variables in seagrass at SMMAP (error bars represent 5 th and 95 th percentiles in a and b and 95% confidence intervals in c, d, and e). a) abundance. b) total length. c) sex ratio. d) Proportion of gravid females e) IMP.

PAGE 30

22 Fig. 2a-d. Size distribution of shrimp populations in seagrass beds in different sampling periods. a) July/August 2002 and 2003. b) November 2002. c) February 2003. d) May 2003.

PAGE 31

23 Fig. 3. Temporal differences in median size of shrimp from marsh edges at SMMAP (error bars represent 5 th and 95 th percentiles).

PAGE 32

24 Fig. 4a-d. Size distribution of shrimp populations in marsh edges in different sampling periods. a) July/August 2002 and 2003. b) November 2002. c) February 2003. d) May 2003.

PAGE 33

25 Table 1. Water chemistry at SMMAP by sampling period. Variable Test Used p value Mean (95% Confidence Interval) July/August November 2002 February 2003 May 2003 Dissolved oxygen (mg/l) ANOVA < 0.001 8.80 (7.96-9.64) 13.36 (11.44-15.28) 10.07 (9.53-10.61) 5.41 (5.08-5.74) Salinity (ppt) ANOVA 0.476 16.33 (13.49-19.17) 19.54 (18.27-20.81) 14.06 (13.25-14.87) 15.11 (13.71-16.51) Water temperature (C) ANOVA < 0.001 29.57 (28.79-30.35) 18.80 (18.37-19.23) 12.69 (12.28-13.10) 29.73 (29.18-30.28)

PAGE 34

Table 2. Temporal changes in variables characterizing samples of P intermedius taken from seagrass at SMMAP. Variable Test Used p value Central Tendency* (95% confidence interval)** July/August November 2002 February 2003 May 2003 Abundance (shrimp/sweep) Kruskal-Wallis ANOVA < 0.0001 18.0 (3.0-21.1 ) 46.5 (22.0-352.8) 1.5 (0.2-2.6) 14.5 (4.0-18.2) Total length (mm) Kruskal-Wallis ANOVA < 0.0001 19.143 (15.233-21.371) 9.714 (5.571-12.824) 16.571 (15.979-30.214) 12.429 (8.877-15.979) Sex ratio (female/male) 2 0.006 0.55 (0.49-0.61) 0.49 (0.43-0.56) 0.47 (0.25-0.69) 0.61 (0.59-0.68) Proportion gravid females 2 < 0.0001 0.47 (0.39-0.62) 0.23 (0.17-0.30) 0.00 (0.00-0.00) 0.19 (0.06-0.32) Clutch size Kruskal-Wallis ANOVA 0.181 42.5 (16.6-102.5) 28.0 (18.0-47.0) *** *** 29.0 (11.8-71.2) Intermolt period 2 0.018 5.5 (4.6-6.5) 8.4 (7.0-16.4) 15.6 (6.0-25.2) 4.4 (3.8-5.2) Growth increment Kruskal-Wallis ANOVA 0.547 0.043 (-0.098-0.230) 0.079 (-0.139-0.127) 0.030 (-0.079-0.180) 0.049 (-0.089-0.243) Growth rate mm/day Kruskal-Wallis ANOVA 0.168 0.0009 (0.0003-0.0020) 0.001 (0.0006-0.0014) 0.0002 (0.0000-0.0007) 0.0017 (0.0010-0.0018) Mean when test is 2 median when test is Kruskal-Wallis ANOVA ** 5th and 95th percentiles when test is Kruskal-Wallis ANOVA *** Sample size is 0 or 1, therefore data were not included in statistical tests and confidence intervals were not calculated. 26

PAGE 35

Table 3. Temporal changes in variables characterizing samples of P intermedius taken from marsh edges at SMMAP. Variable Test Used p value Central Tendency* (95% confidence interval)** July/August November 2002 February 2003 May 2003 Abundance (shrimp/ sweep) Kruskal-Wallis ANOVA 0.331 0.6 (0.2-4.5) 0.8 (0.1-7.9) 0.7 (0.1-3.0) 0.6 (0.1-15.8) Total length (mm) Kruskal-Wallis ANOVA <0.001 22.929 (17.080-23.845) 15.710 (12.786-22.786) 16.429 (14.343-21.286) 12.714 (8.286-15.514) Sex ratio (female/male) 2 0.468 0.06 (0.00-0.33) 0.46 (0.00-1.00) 0.78 *** 0.53 (0.00-0.92) Proportion gravid females *** *** 0.27 (0.11-0.43) 0.00 *** 0.00 *** 0.00 (0.00-0.00) Clutch size *** *** 89.0 (16.0-116.6) *** *** *** *** *** *** Intermolt period 2 0.326 6.1 *** 4.8 (4.4-14.3) *** *** 8.1 (6.8-10.3) Growth increment Kruskal-Wallis ANOVA 0.233 0.006 (0.001-0.013) 0.010 (-0.0700.184) 0.008 (-0.056-0.182) 0.000 (-0.079-0.117) Growth rate mm/day Kruskal-Wallis ANOVA 0.194 0.0008 *** 0.0015 *** *** *** 0.0000 (-0.0001-0.0001) Mean when test is 2 median when test is Kruskal-Wallis ANOVA ** 5th and 95th percentiles when test is Kruskal-Wallis ANOVA *** Sample size is 0 or 1, therefore data were not included in statistical tests and confidence intervals were not calculated. 27

PAGE 36

28 Table 4. Differences in variables characterizing samples of P intermedius taken in different habitats at SMMAP in the July/August 2002 and 2003 sampling periods. Variable Test Used p value Central Tendency* (95% Confidence Interval)** Seagrass Marsh Edge Abundance (shrimp/sweep) Mann-Whitney U Test < 0.001 18.0 (3.0-30.3) 0.6 (0.2-4.5) Total Length (mm) Mann-Whitney U Test < 0.001 19.143 (15.233-28.371) 22.929 (17.080-23.845) Sex ratio (female/male) 2 0.773 0.55 (0.49-0.61) 0.06 (0.00-0.33) Proportion of females gravid 2 0.015 0.47 (0.39-0.62) 0.27 (0.11-0.43) Clutch Size Mann-Whitney U Test 0.0428 42.5 (16.6-102.5) 89.0 (16.0-116.6) Intermolt period *** 0.679 5.5 (4.6-6.5) 6.1 *** Growth increment Mann-Whitney U Test 0.9213 0.006 (-0.098-0.230) 0.006 (0.0012-0.0129) Growth rate *** *** 0.0009 (-0.0001-0.0020) 0.0008 *** Mean when test is 2 median when test is Mann Whitney U Test ** 5th and 95th percentiles when test is Mann-Whitney U Test *** Sample size is 0 or 1, therefore data were not included in statistical tests and confidence intervals were not calculated.

PAGE 37

29 Table 5. Differences in variables characterizing samples of P intermedius taken in different habitats at SMMAP in the November 2002 sampling period. Variable Test Used p value Central Tendency* (95% Confidence Interval)** Seagrass Marsh Edge Abundance (shrimp/sweep) Mann-Whitney U Test 0.0002 46.5 (22.0-352.8) 0.8 (0.1-7.9) Total length (mm) Mann-Whitney U Test < 0.0001 9.714 (5.571-12.824) 15.710 (12.786-22.786) Sex ratio (female/male) 2 0.798 0.49 (0.43-0.56) 0.46 (0.00-1.00) Proportion of females gravid 2 < 0.0001 0.23 (0.17-0.30) 0.00 *** Clutch size *** *** 28.0 (18.0-47.0) *** *** Intermolt period 2 0.259 8.4 (7.0-16.4) 4.8 (4.4-14.3) Growth increment Mann-Whitney U Test 0.259 0.011 (-0.139-0.127) 0.010 (-0.070-0.184) Growth rate Mann-Whitney U Test 0.083 0.0007 (0.0004-0.0010) 0.0015 *** Mean when test is 2 median when test is Mann Whitney U Test ** 5th and 95th percentiles when test is Mann-Whitney U Test *** Sample size is 0 or 1, therefore data were not included in statistical tests and confidence intervals were not calculated.

PAGE 38

30 Table 6. Differences in variables characterizing samples of P intermedius taken in different habitats at SMMAP in the May 2003 sampling period. Variable Test Used p value Central Tendency* (95% Confidence Interval)** Seagrass Marsh Edge Abundance (shrimp/sweep) Mann-Whitney U Test < 0.0001 14.5 (4.9-27.7) 0.6 (0.1-15.8) Total Length (mm) Mann-Whitney U Test 0.91 12.429 (8.877-15.979) 12.714 (8.286-15.514) Sex ratio (female/male) 2 0.337 0.61 (0.59-0.68) 0.53 (0.00-0.92) Proportion of females gravid 2 < 0.0001 0.19 (0.06-0.32) 0.00 (0.00-0.00) Clutch Size *** *** 29.0 (11.8-71.2) *** *** Intermolt period 2 0.009 4.4 (3.8-5.2) 8.1 (6.8-10.3) Growth increment Mann-Whitney U Test 0.7388 0.007 (-0.089-0.243) 0.000 (-0.079-0.117) Growth rate Mann-Whitney U Test 0.083 0.0017 (0.0010-0.0018) 0.0000 (-0.0001-0.0001) Mean when test is 2 median when test is Mann Whitney U Test ** 5th and 95th percentiles when test is Mann-Whitney U Test *** Sample size is 0 or 1, therefore data were not included in statistical tests and confidence intervals were not calculated.

PAGE 39

31 Table 7. Differences in variables characterizing samples of P. intermedius taken in different habitats at SMMAP in the February 2003 sampling period. Variable Test Used p value Central Tendency* (95% Confidence Interval)** Seagrass Marsh Edge Abundance (shrimp/sweep) Mann-Whitney U Test 0.0023 1.5 (0.2-2.6) 0.7 (0.1-3.0) Total Length (mm) Mann-Whitney U Test 0.8777 16.571 (15.979-30.214) 16.429 (14.343-21.286) Sex ratio (female/male) 2 0.027 0.47 (0.36-0.62) 0.78 *** Proportion of females gravid *** *** 0.00 (0.00-0.00) 0.00 *** Clutch Size *** *** *** *** *** *** Intermolt period 2 0.246 15.600 (6.0-25.2) *** *** Growth increment Mann-Whitney U Test 0.1622 0.004 (-0.079-0.180) 0.008 (-0.056-0.182) Growth rate *** *** 0.0017 (0.0012-0.0021) *** *** Mean when test is 2 median when test is Mann Whitney U Test ** 5th and 95th percentiles when test is Mann-Whitney U Test *** Sample size is 0 or 1, therefore data were not included in statistical tests and confidence intervals were not calculated.

PAGE 40

32 Table 8. Differences in variables characterizing samples of P intermedius collected from different habitats in the Weeki Wachee estuary in August 2003. Variable Test Used p value Central Tendency* (95% confidence intervals)** Seagrass Marsh Edge Abundance (shrimp/sweep) Mann-Whitney U Test < 0.0001 11.5 (4.4-27.0) 0.6 (0.1-0.7) Total Length (mm) Mann-Whitney U Test 0.635 20.571 (16.686-23.536) 22.857 (20.500-27.929) Sex ratio (female/male) 2 0.534 0.54 (0.34-0.93) 0.64 (0.34-0.93) Proportion of females gravid 2 0.718 0.35 (0.28-0.64) 0.29 (0.00-0.65) Clutch Size Mann-Whitney U Test 0.101 32.5 (22.0-58.1) 50.5 (42.0-59.1) Mean when test is 2 median when test is Mann Whitney U Test ** 5th and 95th percentiles when test is Mann-Whitney U Test

PAGE 41

33 Table 9. Differences in variables characterizing samples of P intermedius in different habitats in different estuaries during August 2003. Variable Test Used p value Central tendency* (95% Confidence Interval**) Seagrass SMMAP Weeki Wachee Steinhatchee Abundance KruskalWallis ANOVA 0.002 11.5 (7.4-26.2) 11.5 (4.4-27.0) 0.7 (0.1-3.2) Shrimp Total Length KruskalWallis ANOVA 0.133 21.000 (14.600-29.764) 20.571 (16.686-23.536) 18.288 (10.571-26.943) Sex Ratio 2 0.911 0.55 (0.04-0.58) 0.54 (0.34-0.93) 0.53 (0.22-0.84) Proportion of Females Gravid 2 0.012 0.26 (0.08-0.28) 0.35 (0.28-0.64) 0.18 (0.03-0.23) Clutch Size KruskalWallis ANOVA 0.354 27.0 (16.4-89.8) 32.5 (22.0-58.1) 39.0 (22.4-54.0) Marsh Edge Abundance KruskalWallis ANOVA 0.023 0.6 (0.1-0.7) 0.6 (0.1-0.7) 0.8 (0.7-1.5) Shrimp Total Length KruskalWallis ANOVA 0.004 26.000 (20.300-34.250) 22.857 (20.500-27.929) 17.429 (13.400-19.929) Sex Ratio 2 0.978 0.60 (0.36-0.84) 0.64 (0.34-0.93) 0.62 (0.36-1.60) Proportion of Females Gravid 2 0.763 0.25 (0.00-0.51) 0.29 (0.00-0.65) 0.17 (0.06-0.93) Clutch Size KruskalWallis ANOVA 0.165 89.0 (11.8-116.6) 50.5 (42.0-59.1) 32.5 (20.3-34.8) Mean when test is 2 median when test is Kruskal-Wallis ANOVA ** 5th and 95th percentiles when test is Kruskal-Wallis ANOVA

PAGE 42

34 Table 10. Characteristics of seagrass vegetation in different estuaries in August 2003. Variable Test Used p value Median (5th, 95th percentile) SMMAP Weeki Wachee Steinhatchee Halodule % Areal Cover KruskalWallis ANOVA < 0.001 80.0 (20.0-90.0) 100.0 (90.0-100.0) 1.9 (0.2-3.6) Halophila % Areal Cover KruskalWallis ANOVA 1 0.0 (0.0-0.0) 5.5 (1.0-10.0) 0.0 (0.0-0.0) Syringodium % Areal Cover KruskalWallis ANOVA 0.55 0.0 (0.0-0.0) 0.0 (0.0-0.0) 0.6 (0.1-12.8) Ruppia % Areal Cover KruskalWallis ANOVA 1 0.0 (0.0-0.0) 0.0 (0.0-0.0) 0.0 (0.0-0.0) Thallassia % Areal Cover KruskalWallis ANOVA < 0.001 0.7 (0.3-1.3) 0.0 (0.0-0.0) 80.0 (10.0-100.0) Halodule shoot density (shoots/m 2 ) KruskalWallis ANOVA < 0.001 792.0 (36.8-1651.2) 544.0 (80-1152.0) 14.4 (1.6-544.0) Halophila shoot density (shoots/m 2 ) KruskalWallis ANOVA 0.609 0.0 (0.0-0.0) 0.0 (0.0-0.0) 0.0 (0.0-0.0) Thalassia shoot density (shoots/m 2 ) KruskalWallis ANOVA < 0.001 11.6 (1.6-17.6) 9.6 (1.6-41.6) 14.4 (1.6-225.6) Syringodium shoot density (shoots/m 2 ) KruskalWallis ANOVA 0.114 0.0 (0.0-0.0) 0.0 (0.0-0.0) 11.2 (1.6-625.6) Halodule aboveground biomass (g/m 2 ) KruskalWallis ANOVA < 0.001 624.0 (36.8-1368.0) 280.0 (80.0-640.0) 14.4 (1.6-449.6) Halophila aboveground biomass (g/m 2 ) KruskalWallis ANOVA 0.976 0.0 (0.0-0.0) 0.0 (0.0-0.0) 0.0 (0.0-0.0) Thalassia aboveground biomass (g/m 2 ) KruskalWallis ANOVA 0.002 9.6 (1.6-614.4) 8.0 (1.6-43.2) 14.4 (1.6-142.7) Syringodium aboveground biomass (g/m 2 ) KruskalWallis ANOVA 0.114 0.0 (0.0-0.0) 0.0 (0.0-0.0) 11.2 (1.6-481.6)

PAGE 43

35 Table 11. Water chemistry in the SMMAP, Weeki Wachee and Steinhatchee estuaries in August 2003. Variable Test Used p value Mean (95% Confidence Interval) SMMAP Weeki Wachee Steinhatchee Dissolved oxygen (mg/l) ANOVA 0.013 7.60 (6.30-8.90) 11.86 (11.43-12.28) 4.78 (1.61-7.96) Salinity (ppt) ANOVA 0.007 9.99 (9.39-10.59) 17.68 (16.40-18.95) 10.52 (6.85-14.19) Water temperature (C) ANOVA 0.299 29.34 (28.76-29.91) 27.61 (26.66-28.56) 31.26 (29.91-32.61)

PAGE 44

DISCUSSION Determination of the Relative Quality of Habitat Types A habitat type was considered to provide higher habitat quality than another for P intermedius if one of the following three conditions were met: (1) if shrimp abundance and ISF variables both indicated higher habitat quality in the same habitat or estuary, (2) if shrimp abundance was similar between the two and the ISF variables indicated greater fitness in one over the other, (3) if the ISF variables were similar, but abundance was higher in one over the other. In the situations where abundance and ISF variables proved to be contradictory, then fitness related measures and abundances were used to estimate fecundity at the population level in an effort to provide additional insights into the relative qualities of the habitats. Temporal Variability within the SMMAP SMMAP Within Habitat Sampling Observed temporal changes in the demographics and physiology of grass shrimp within the seagrass and marsh edge habitats suggest seasonal changes consistent with those found in other studies (Kneib 1987, Knowlton et al. 1994, Grabe 2003). These temporal changes provide important background information that should be considered when comparing the relative habitat quality of seagrass beds and marsh edges for P intermedius The results suggest that summer months are the most appropriate time to compare and contrast estimates of abundance and fitness to infer relative habitat quality. 36

PAGE 45

37 This can best be illustrated through discussion of results from each of the sampling periods. In February 2003, the relative abundance of P intermedius was at its minimum in SMMAP, similar to reduced winter abundances documented by Knowlton et al. (1994) in North Carolina. The shift in sex ratio towards a more male-dominated population from November 2002 to February 2003 in this study suggested that females may have begun to suffer the effects of reduced habitat quality before males, which is consistent with Reinsel et al. (2001), who showed that limited food availability caused increased mortality of females relative to males. This increased susceptibility of females may be due to the large amount of energy invested in reproduction (Vernberg and Piyatitivorakul 1998). This large reproductive investment could, however, also support the alternative conclusion that female shrimp are simply succumbing to the metabolic demands of repeated reproduction during summer and late fall irrespective of changes in the quality of the habitat in which they are found. However, changes in other ISF variables in February 2003 also suggest declining habitat quality in winter. Apparently spawning did not occur in February 2002, beacuse gravid females were not collected, a pattern observed in other decapods during winter months (e.g. Grabe 2003). In February 2002, P intermedius size was at its maximum, suggesting that only the largest individuals were capable of surviving an autumn-winter mortality event. This result is consistent with that of Chockley and St. Mary (2003), who found that mortality of Stenopus hispidus decreased with increasing shrimp size. In addition, the long IMP during this sampling period suggested that conditions in the

PAGE 46

38 seagrass beds, or the estuarine waters in general, were negatively influencing growth rates in addition to precluding spawning. In May 2003, an increase in shrimp abundance, proportion of gravid females, and shorter IMPs suggested that conditions in the seagrass beds were improved relative to winter. In the July/August sampling periods, conditions within seagrass beds were most favorable for reproduction, as indicated by the high proportion of gravid females in the sample population. This result suggests that summer months are the primary breeding period for grass shrimp along the Gulf of Mexico coast, which is what would be expected for a decapod in a temperate or subtropical climate. Although the total number of shrimp increased in November 2002, the median size was reduced as a consequence of an influx of newly recruited shrimp. The increase in abundance is consistent with autumnal increases documented in other studies (Knowlton et al. 1994). Spawning continued during this time, but at a reduced rate compared to either spring or summer, as indicated by the lower proportion of gravid females. Along with decreased spawning, an increase in the intermolt period and a reduced proportion of females in the population suggested a decline in the quality of seagrass habitats or the environment as a whole during November and a critical period in the life history of P intermedius Shrimp abundance had declined markedly by February, indicating increased mortality during late fall and winter. Alternatively, shrimp may have emigrated to other areas. However, due to their small size, presumed lack of long distance mobility, apparent absence form offshore habitats and lack of increased abundance in any other habitats sampled in this study, emigration is an unlikely explanation for the decline in numbers between November and February.

PAGE 47

39 So what were the primary habitat characteristics or environmental factors driving these changes in shrimp demography? The lack of significant changes in the percent areal cover of seagrass as a whole, or of species composition of seagrasses, implies that seasonal changes in vegetation were not driving the changes observed in shrimp demographics. The most likely explanation for the changing demographics within the shrimp population is water temperature, which is an environmental, rather than a habitat, characteristic. Fall/winter mortalities of shrimp similar to those observed in this study have been documented in other locations (Knowlton et al. 1994), and Lemaire et al. (2002) suggested that reduced water temperatures could compromise the osmoregulatory abilities of juvenile and subadult Pennaeus stylirostris Vernberg and Piyatitivorakul (1998) demonstrated that temperature had significant effects on many grass shrimp metabolic processes. Changes in temperature and salinity have been shown to affect changes in caridean shrimp abundances in other locations as well (Walsh and Mitchell 1998). However, temperature and food production are also often correlated in marine environments, and productivity of seagrass beds is known to vary seasonally (e.g. Peterson and Fourquean 2001). Brockington and Clarke (2001), in an attempt to determine the relative influence of temperature and food availability for the sea urchin Sterechinus neum ayeri found that food availability may indeed be the key factor driving metabolic rates, as opposed to temperature. Further study is necessary to determine which factor is more important for grass shrimp, and P intermedius in particular. In the marsh edges, temporal variability was less pronounced than in the seagrass beds. There were no significant changes in abundance or any ISF variable other than total length. However, the quality of marsh edges seemed higher in the July/August sampling

PAGE 48

40 period than other sampling periods. Grass shrimp were significantly larger within marsh edge habitats in July/August, and this was the only period in which gravid females were encountered. Since the percent areal cover of plant species did not change significantly, and the DO, salinity, and water temperature values were similar to those encountered in seagrass, water temperature is the most likely cause of temporal changes in shrimp demographics in marsh edges as well. SMMAP Cross-Habitat Comparisons (Seagrass vs. Marsh Edge) In all sampling periods, greater abundances of shrimp and higher proportions of gravid females were found in seagrass beds than along marsh edges. Although this appears to contradict the findings of Rozas and Minello (1998), who documented higher abundances of grass shrimp in salt marshes than seagrass beds in a Texas estuary, it may simply reflect a difference in occupancy patterns between marsh edges and salt marshes that were not captured by either study. In May 2003, shrimp from seagrass had shorter IMPs, and gravid females were only caught in seagrass beds during this time period. Thus, it is concluded (based on criterion 1 above) that seagrass beds provided higher quality habitat than marsh edges in May. During the July/August sampling periods, shrimp size and clutch size indicated higher habitat quality in marsh edges. These variables were expected to covary, however, as a relationship between crustacean size and clutch size has been previously documented, notably by Chockley and St. Mary (2003), who documented this relationship for banded coral shrimp. Despite the differences in total length and clutch size, the much greater abundances and the higher proportion of gravid females in the seagrass beds were far more important in determining the net number of eggs produced per area of habitat than the larger clutch sizes in marsh edges, as indicated by the estimated number of eggs/sweep. The larger size of shrimp in

PAGE 49

41 marsh edges may suggest that only large shrimp are capable of surviving in this habitat, or that dominant individuals are monopolizing the habitat at low densities, implying that there is some degree of habitat segregation by size in these shrimp, as has been shown for the daggerblade grass shrimp ( P pugio ) when subjected to mummichog ( Fundulus heteroclitus ) predation (Davis et al. 2003), and also for tiger prawns ( Pennaeus esculentus and P semisulcatus ) in Australian seagrass beds (Loneragan et al. 1998). Based on the greater potential reproductive output of shrimp in seagrass beds, it was concluded that marsh edges provided an inferior habitat for P intermedius during summer time periods. In February 2003, the pattern was apparently somewhat different. Although shrimp abundance remained higher in seagrass beds than marsh edges, the difference was not as pronounced. However, the sex ratio of shrimp in the marsh was skewed towards females, in comparison with the nearly even sex ratio of the sampled shrimp population in seagrass beds. When the abundance values were multiplied by the sex ratios in each of the habitats, it was clear that there were approximately the same number of females per unit area in each of the habitats. The higher proportion of females in marsh edges may imply that habitat quality there was higher, and the higher abundance in seagrass may imply some degree of overcrowding of subdominant individuals into suboptimal habitat during this time period. This possibility is further supported by the observation that female grass shrimp tend to be larger than males, and so would presumably be dominant in intraspecific agonistic competition. This possibility merits further investigation. Since no other variables were significantly different between the habitats, and the sex ratio and abundance do not both point to the same habitat as having higher quality, it can not be

PAGE 50

42 conclusively stated that either of these habitats was higher quality than the other during February 2003. Among Estuaries Comparisons Weeki Wachee and SMMAP showed similar patterns in August 2003. Seagrass beds provided a higher quality habitat than the marsh edges for P intermedius during July and August, and the same was true in the Weeki Wachee estuary. Seagrass beds in the Weeki Wachee estuary were determined to provide higher quality habitat than adjacent marsh edges because shrimp were more abundant in seagrass beds than marsh edges and there were no significant differences in measured ISF variables (criterion 3 above). These results contrast with those from the Steinhatchee estuary, where shrimp abundances were similar between the two habitat types and no differences in ISF variables were observed. As a consequence it was not possible to designate either seagrass beds or marsh edges as the superior habitat in that estuary. Interestingly, the abundance of P intermedius in Steinhatchee seagrass beds was less than the abundances found in seagrass in SMMAP or Weeki Wachee. In fact, it was the marked reduction in numbers of P intermedius in seagrass that accounted for similarities in abundance between the two primary habitat types rather than an increase in marsh edge occupancy. Although shrimp abundance in the Steinhatchee marsh sites was greater than the other two estuaries, the cross-estuary differences in this habitat type were relatively small. The lower abundance of shrimp in seagrass beds in Steinhatchee relative to either SMMAP or Weeki Wachee coupled with the fact that the proportion of gravid females was also less suggests that seagrass habitat in Steinhatchee was of lower quality for P intermedius Many reasons may explain these findings, but data reported here indicate

PAGE 51

43 that the vegetative characteristics of the seagrass beds may be important. The percent areal cover of H wrightii was lowest in Steinhatchee and greatest in Weeki Wachee. The percent areal cover and shoot density of T testudinum on the other hand, was greater in the Steinhatchee estuary than in either SMMAP or Weeki Wachee. These patterns suggest that seagrass beds dominated by Halodule may provide a higher quality habitat for P intermedius than those dominated by Thalassia Although preference for a species of seagrass has not been demonstrated for P intermedius preference for specific seagrass communities has been documented for other shrimp species (Loneragan et al. 1998). The relative quality of the seagrass beds in these different estuaries may change temporally as they did in SMMAP. Broad generalizations based on these findings warrant caution. Shrimp abundance and ISF variables were more uniform within marsh edges across estuaries. However, marsh edge habitats in the Steinhatchee estuary had greater abundances of shrimp than marsh edges in the other two estuarine areas, which may indicate an increased use of otherwise suboptimal marsh habitats, perhaps due to the reduced quality of seagrass habitats in Steinhatchee. Overall Conclusions The lack of any shrimp near oyster bars in SMMAP contrasts with Eggleston et al. (1998), who documented grass shrimp using these habitats. Throughout all estuaries and sampling periods, marsh edges appeared to maintain consistent, fairly low-quality habitat for P intermedius This contrasts with seagrass beds, where habitat quality was generally much higher and more variable. The similarity in habitat quality between seagrass beds and marsh edges in February 2003 was likely due to a decrease in the quality of seagrass habitat. If seagrass beds in North Carolina show similarly variable

PAGE 52

44 habitat quality relative to alternative habitats such as marsh edge or oyster bars, this may provide an explanation for the findings of Eggleston et al. (1998), who found higher abundances in seagrasses than along oyster bars in spring but similar abundances between habitats in late fall. Temporal differences in the relative quality of seagrass and marsh edge habitats to grass shrimp in summer and winter sampling periods at SMMAP likely reflect broad-scale environmental shifts in temperature, that, in turn, may affect also food availability for this organism. The variation in quality among seagrass beds may also explain why the quality of marsh in Steinhatchee, equaled or surpassed that of seagrass. Seagrass in Steinhatchee was poorer quality habitat compared to seagrass in these other estuaries. Where seagrass quality was determined to be poor for grass shrimp, such as at Steinhatchee in August 2003, or SMMAP in February 2003, marsh edges were determined of equal or greater value to the shrimp. Poor quality seagrass habitat for P intermedius during summer was dominated by T testudinum and high quality habitat was dominated by H wrightii Relative abundance was a good indicator of habitat quality both between habitats and between estuaries during most time periods, which fails to support Van Horne (1983). Perhaps, Van Hornes (1983) ideas do not apply to estuarine invertebrates as well as they do to terrestrial quadrupeds. However, relative abundance and the ISF variables did appear to be decoupled at certain times of the year, such as when small shrimp appeared in November 2002. Perhaps the import of these findings can best be understood from a management perspective, if we consider P intermedius to be a model organism for others (such as penaeid shrimp or Macrobrachium spp.) that are commercially important and therefore

PAGE 53

45 more likely to require intensive management. If a manager wanted to set up a marine protected area for P intermedius abundance would likely be an acceptable measure for selecting the location with the best habitat quality and to protect. However, if a manager was attempting to use closed and open seasons as a management tool to generate productivity, then measures of fecundity, mortality, and growth rates would be necessary to determine the best time to restrict harvest. For example, during May, July, and August, when growth rates and fecundity were greatest, exploitation should be restricted to maximize productivity. Relative abundance (and likely biomass) was highest in November, when the period of mass mortality was beginning, and this would be the best time to utilize these shrimp as a resource. On this temporal scale, therefore, Van Hornes theory was supported because the greatest abundances did not always occur when ISF variables pointed to the highest quality of the habitat or environment. Future study of grass shrimp habitat ecology would be most informative if it focused on five key issues: (1) better resolution of the temporal variability within populations so as to determine if the seasonal patterns suggested by this study do, in fact exist; (2) better determination of what factors (i.e., food availability, water temperature or others) most influence the temporal variability in grass shrimp populations; (3) better determination of what factors (i.e., dominant grass type or spatial distribution of habitats) most influence variability in grass shrimp populations within seagrass; (4) better determination of what characteristics make marsh edges a less high-quality habitat relative to seagrass; and (5) why Van Hornes predictions apparently fail to hold in the context of this estuarine invertebrate.

PAGE 54

LITERATURE CITED Abele, L.G., and W. Kim. 1986. An Illustrated Guide to the Marine Decapod Crustaceans of Florida. State of Florida, Department of Environmental Protection, Tallahassee, Florida. 760pp. Anderson, S.H. and K.J. Gutzwiller. 1996. Habitat evaluation methods, p. 592-606. In T.A. Bookhout (ed), Research and Management Techniques for Wildlife and Habitats. The Wildlife Society. Bethesda. Bass, C.S., S. Bhan, G. M. Smith, and J. S. Weis. 2001. Some factors affecting size distribution and density of grass shrimp ( Palaemonetes pugio ) populations in two New Jersey estuaries. Hydrobiologia 450: 231-241 Berg, A.B.V. and P.A. Sandifer. 1984. Mating behavior of the grass shrimp Palaemonetes pugio Holthuis (Decapoda, Caridea). Journal of Crustacean Biology 4(3)417-424. Brockington, S. and A. Clarke. 2001. The relative influence of temperature and food on the metabolism of a marine invertebrate. Journal of Experimental Marine Biology and Ecology 258(1):87-99 Chockley, B.R. and C.M. St. Mary. 2003. Effects of body size on growth, survivorship, and reproduction in the banded coral shrimp, Stenopus hispidus Journal of Crustacean Biology 23(4): 836-848 Cross, R. E. and A. E. Stiven. 1999. Size-dependent interactions in salt marsh fish ( Fundulus heteroclitus Linnaeus) and shrimp ( Palaemonetes pugio Holthuis). Journal of Experimental Marine Biology and Ecology 242: 179-199 Davenport, D.E., Lancia, R.A., Walters, J.R. and P.D. Doerr. 2000. Red cockeaded woodpeckers: A relationship between reproductive fitness and habitat in the North Carolina sandhills. Wildlife Society Bulletin 28 (2): 426-434 Davis, J.L., W.J. Metcalfe, and A.H. Hines. 2003. Implications of a fluctuating fish predator guild on behavior, distribution, and abundance of a shared prey species: the grass shrimp Palaemonetes pugio Journal of Experimental Marine Biology and Ecology 293(1) 23-40. Eggleston, D. B., L. L. Etherington, and W. E. Elis. 1998. Organism response to habitat patchiness: species and habitat-dependent recruitment of decapod crustaceans. Journal of Experimental Marine Biology and Ecology 223: 111-132 46

PAGE 55

47 Franklin, A.B., Anderson, D.R., Guitierrez, R.J., and K.P. Burnham. 2000. Climate, habitat quality, and fitness in Northern Spotted Owl populations in northwestern California. Ecological Monographs 70 (4): 539-590. Frazer, T.K., Hoyer, M.V., Notestein, S.K., Canfield, D.E., Vose, F.E., Leavens, W.R., Blitch, S.B., and J. Conti. 1998. Nitrogen, phosphorus, and chlorophyll relations in selected rivers and nearshore coastal waters along the Big Bend region of Florida. Final Project Report. Suwannee River Water Management District (SRWMD Contract No. 96/97-156) and the Southwest Florida Water Management District (SWFWMD Contract No. 96/97/157R). 326 pp. Frazer, T.K. Notestein, S.K., Hale, J.A., Hoyer, M.V., Canfield, D.E., Blitch, S.B., and C. Bedee. 2003. Water quality characteristics of the nearshore gulf coast waters adjacent to Citru, Hernando, and Levy Counties Project COAST 2001. Annual Report. Southwest Florida Water Management District, Brooksville, Florida. 168 pp. Glancy, T. P., T.K. Frazer, C.E. Cichra, and W.J. Lindberg. 2003. Comparative patterns of occupancy by decapod crustaceans in seagrass, oyster, and marsh edge habitats in a northeastern Gulf of Mexico estuary. Estuaries 26(5):1291-1301 Grabe, S.A. 2003. Seasonal periodicity of decapod larvae and population dynamics of selected taxa in New Hampshire (USA) coastal waters. Journal of Plankton Research 25(4) 417-428. Hartnoll, R.G. 2001. Growth in crustaceatwenty years on. Hydrobiologia 449(1-3): 111-122 Hovel, K.A., Fonseca M.S., Myer, D.L., Kenwrthy, W.J., and P.E. Whitfield. 2002. Effects of seagrass landscape structure, structural complexity and hydrodynamic regime on macrofaunal densities in North Carolina seagrass beds. Marine Ecology Progress Series 243: 11-24. Khan, R. N., R. E. Knowlton, and H. C. Merchant. 1995. Distribution of two sympatric species of grass shrimp Palaemonetes pugio and Palaemonetes vulgaris in relation to homogeneous and heterogeneous aquarium substrates. The Journal of Elisha Mitchell Scientific Society 111(2): 83-95 Kneib, R.T. 1987. Seasonal abundance, distribution, and growth of postlarval and juvenile grass shrimp ( Palaemonetes pugio ) in a Georgia, USA, salt marsh. Marine Biology 96(2): 215-223 Knowlton, R. E., R. N. Khan, P. M. Arguin, T. A Aldaghlas, and R. Sivapathasundram. 1994. Factors determining distribution and abundance of delmarva grass shrimp. Virginia Journal of Science 45(4): 231-247

PAGE 56

48 Knowlton, R.E. and R.H. Schoen. 1984. Salinity tolerance and sodium balance in the prawn Palaemonetes vulgaris compared with Palaemonetes pugio Comparative Biochemistry and Physiology. 79(4): 519-524. Lemaire, P. E. Bernard, J.A. Martinez-Paz, and L. Chim. 2002. Combined effect of temperature and salinity on osmoregulation of juvenile and subadult Penaeus stylirostris Aquaculture 209(1-4):307-317. Lewis, M.A. and S.S. Foss. 2000. A caridean grass shrimp ( Palaemonetes pugio Holthuis) as an indicator of sediment quality in Florida coastal areas affected by point and nonpoint source contamination. Environmental Toxicology 15(3)234-242. Lin, Y.T.K. and G.O. Batzli. 2001. The influence of habitat quality on dispersal, demography, and population dynamics of voles. Ecological Monographs 71(2):245-275 Loneragan, N.R., R.A. Kenyon, D.J. Staples, I.R. Poiner, and C.A.Conacher. 1998. The influence of seagrass type on the distribution and abundance of postlarval and juvenile tiger prawns ( Penaeus esculentus and P semisulcatus ) in the western Gulf of Carpentaria, Austrailia. Journal of Experimental Marine Biology and Ecology 228(2): 175-195. Luck, G.W. 2002. Determining habitat quality for the cooperatively breeding Rufous Treecreeper, Climacteris rufa Austral Ecology 27 (2): 229-237 Peterson, B.J. and J.W. Fourqurean. 2001. Large scale patterns in seagrass ( Thalassia testudinum ) demographics in south Florida. Limnology and Oceanography 46(5):1077-1090. Posey, M.H. and A. H. Hines. 1991. Complex predator-prey interactions within an estuarine benthic community. Ecology 72: 2155-2169 Posey, M.H., Alphin, T.D., Powell, C.M., and E. Townshend. 1999. Use of oyster reefs as a habitat for epibenthic fish and decapods. Pp 229-238. In Luckenbach, M.W., Mann, R. and J.A. Wesson, (eds.). Oyster Reef Habitat Restoration: A Synopsis and Synthesis. Virginia Institute of Marine Science Press. Quetin, L.B. and R.M Ross. 1991. Behavioral and Physiological characteristics of the Antarctic krill, Euphausia superba. American Zoology 31:49-63 Reinsel, K.A., Glas, P.S., Rayburn, J.R., Pritchard, M.K., and W.S. Fisher. 2001. Effects of food availability on survival, growth, and reproduction of the grass shrimp Palaemonetes pugio : a laboratory study. Marine Ecology Progress Series 220:231-239 Ross, S.W. 2003. The relative value of different nursery areas in North Carolina for transient juvenile marine fishes. Fishery Bulletin Seattle 101 (2): 384-404.

PAGE 57

49 Rozas, L. P. and T. J. Minello. 1998. Nekton use of salt marsh, seagrass, and nonvegetated habitats in a south Texas (USA) Estuary. Bulletin of Marine Science 63(3):481-501 Schantz, T. von. 1981. Female cooperation, male competition, and dispersal in the red fox ( Vulpes vulpes ). Oikos 37:63-68 Sheridan, P. F. 1992. Comparative habitat utilization by estuarine macrofauna within the mangrove ecosystem of Rookery Bay, Florida. Bulletin of Marine Science 50(1)21-39 Stickle, W.B., Kapper, M.A., Liu, L.L., Gnaiger, E., and S.Y. Wang. 1989. Metabolic adaptations of several species of crustaceans and mollusks to hypoxia tolerance and microcalorimetric studies. Biological BulletinWoods Hole. 177 (2): 303-312. Townshend, E.C. 1991. Depth distribution of the grass shrimp Palaemonetes pugio in two contrasting tidal creeks in North Carolina and Maryland. M.S. Thesis, University of North Carolina at Wilmington. Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Journal of Wildlife Management 47:893-901 Vernberg, F. J., and S. Piyatiratitivorakul. 1998. Effects of salinity and temperature on the bioenergetics of adult stages of the grass shrimp ( Palaemonetes pugio Holthuis) from the North Inlet Estuary, South Carolina. Estuaries 21(1)176-193 Walsh, C.J. and B.D. Mitchell. 1998. Factors associated with variations in abundance of epifaunal caridean shrimps between and within estuarine seagrass meadows. Marine and Freshwater Research 49(8):769-777 Walters, J.R., S.J. Daniels, J.H. Carter, and P.D Doerr. 2002. Defining quality of red-cockaded woodpecker foraging habitat based on habitat use and fitness. Journal of Wildlife Management 66 (4): 1064-1082 Young, D.K, Buzas, M.A. and M.W. Young. 1976. Species density of macrobenthos associated with seagrasss: a field study of predation. Journal of Marine Resources 34: 577-592 Young, D.K. and M.W. Young. 1978. Regulation of species densities of seagrass-associated macrobenthos: evidence from field experiments in the Indian River estuary. Florida Journal of Marine Resources 36: 569-593

PAGE 58

BIOGRAPHICAL SKETCH Daniel Scott Goodfriend was born on March 15, 1976, in Suffern, New York. He lived in the suburbs of New York City until his graduation from Rockland Country Day School in May of 1994, where he spent a lot of time enjoying the outdoors in and around New Yorks Harriman State Park. After graduating high school, he spent two semesters as a philosophy of religion major at Grinnell College in Iowa, but found this subject matter too abstract. He left school in an attempt to find a path that was more personally fulfilling and spent the next three years traveling around the United States, living briefly in New Orleans, Louisiana and Lake Worth, Florida. In January 1996 he moved to Encinitas, California, and began attending classes at MiraCosta College in the nearby town of Oceanside. Drawing on his experiences in the woods of New York and his travels to various National Parks, he decided on a major in wildlife. To pursue this goal, he transferred to Humboldt State University in Arcata, California, in August 1997. After three and a half very enjoyable years, he graduated with a Bachelor of Science in wildlife in December of 2001. He then spent a year in the Americorps program monitoring a population of threatened Blandings turtles ( Emydoidea blandingii ) for the nonprofit group Hudsonia, based in Dutchess County, NY. After his year at Hudsonia, Daniel decided that his goals required him to continue his education, and in January 2002 he enrolled at the University of Florida in Gainesville. There he spent two and a half rewarding years studying coastal ecology with Dr. Thomas Frazer, and will receive a Master of Science degree in May 2000. 50


Permanent Link: http://ufdc.ufl.edu/UFE0004787/00001

Material Information

Title: The Relative Value of Seagrass, Marsh Edge, and Oyster Habitats to the Brackish Grass Shrimp, Palaemonetes intermedius, along the Gulf Coast of Florida
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0004787:00001

Permanent Link: http://ufdc.ufl.edu/UFE0004787/00001

Material Information

Title: The Relative Value of Seagrass, Marsh Edge, and Oyster Habitats to the Brackish Grass Shrimp, Palaemonetes intermedius, along the Gulf Coast of Florida
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0004787:00001


This item has the following downloads:


Full Text












THE RELATIVE VALUE OF SEAGRASS, MARSH EDGE, AND OYSTER
HABITATS TO THE BRACKISH GRASS SHRIMP, Palaemonetes intermedius,
ALONG THE GULF COAST OF FLORIDA
















By

DANIEL SCOTT GOODFRIEND


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2004















ACKNOWLEDGMENTS

There are probably many more people who deserve recognition for their help and

encouragement than I have space for here, so let me begin by thanking everyone who has

helped me in deciding on a career path, thinking more critically about the natural world,

and enabling me to fulfill my dreams and goals. This applies particularly to the faculty of

the Department of Wildlife at Humboldt State University, who first opened my eyes to

the possibilities that present themselves when we apply the scientific method to better

understand the natural world.

I want to thank my advisor, Dr. Thomas Frazer, for his invaluable help. Dr. Frazer

has proved to be a continuing source of support and encouragement, without which my

graduate education would never have been as rewarding or challenging. His flexible and

open style as an advisor provided me with the intellectual space I needed to learn to think

more critically about ecological problems, and his high standards forced me to become a

better ecologist.

The other members of my committee, Dr. William Lindberg and Dr. Charles

Jacoby, also deserve special thanks. Dr. Lindberg was particularly helpful in the early

stages of this project, providing me with key sources of information on crustacean

biology as well as on the philosophy of the scientific method in general. Dr. Jacoby was

especially helpful nearer to the end of the project, substituting for Dr. Frazer as a

proofreader of early drafts of this thesis. He provided me not only with suggestions, but

potential solutions to the problems I encountered.









There are many other people who proofread early drafts of this paper. For this I

need to thank Rikki Grober-Dunsmore, Jason Hale, Chanda Jones, Kate Lazar, Darlene

Saindon, and Deborah Schwartz. Dr. Kenneth Portier was also invaluable for his advice

on statistical design and data management techniques.

Many people lent me time and equipment which helped the field portion of this

project to be successful. When this project was in its infancy, Thomas Glancy introduced

me to what would become my study sites, provided important background data, and

taught me to differentiate the species of Palaemonetes I would encounter. I also wish to

thank the entire staff of the St. Martin's Marsh Aquatic Preserve, in particular Seth Blitch

and Chad Bedee, who graciously provided transportation to my shallow-water field sites

on their airboat and assisted as field workers. Carla Beals, Jaime Greenwalt, Rikki

Grober-Dunsmore, Jason Hale, Stephanie Keller, Kate Lazar, Benjamin Loughran, Sky

Notestein, Troy Thompson, Chris Tilghman, and Duncan Vaughn also assisted with the

field work, sometimes for long hours on very cold and rainy days. They deserve both

thanks and credit for their work.

I especially want to thank Deborah Schwartz for her sense of humor and

encouragement which helped me to remain optimistic and to continue my research during

those inevitable times in the course of a project like this when all hope seems to be lost.















TABLE OF CONTENTS
Page

A C K N O W L E D G M E N T S .................................................................................................. ii

LIST OF TABLES ............... .. ...... ........ .. ........ .......... .......... .... ....

L IST O F FIG U R E S .... .............................. ....................... ........ .. ............... vi

ABSTRACT .............. ............................................. vii

INTRODUCTION ............................... .................... .............

ST U D Y A R E A ....................................................... 5

M E T H O D S ............................................................................ . 7

Field and Laboratory M ethods........................................................... ............... 7
Statistical M eth od s........... ...... .......................................... .............. ......... ....... 13

R E S U L T S ................................................................................1 5

T em poral V ariability ........................ .... .......................... .. ...... .. ................15
W within H habitat V ariability .............................................................. .................. 15
SMMAP Cross-Habitat Comparisons (Seagrass vs. Marsh Edges)................ 17
A m ong E stuaries C om parisons........................................................ ..................... 19

D IS C U S S IO N ....................................................................................................3 6

Determination of the Relative Quality of Habitat Types................ ............... 36
Temporal Variability within the SM M AP............................................................... 36
SM M AP W within H habitat Sam pling ................................... ..............................36
SMMAP Cross-Habitat Comparisons (Seagrass vs. Marsh Edge)...................40
A m ong E stuaries C om parisons........................................................ .....................42
O overall C onclusions......... ............................................................ ... ... .... ..... 43

L IT E R A T U R E C IT E D ............................................................................ ....................46

B IO G R A PH IC A L SK E TCH ..................................................................... ..................50







iv












LIST OF TABLES


Tablege

1. Water chemistry at SMMAP by sampling period.........................................................25

2. Temporal changes in variables characterizing samples of P. intermedius taken from
seagrass at SMMAP..................... ..................................26

3. Temporal changes in variables characterizing samples of P. intermedius taken from
m arsh edges at SM M A P ................................................................ ........................... 27

4. Differences in variables characterizing samples of P. intermedius taken in different
habitats at SMMAP in the July/August 2002 and 2003 sampling periods ..................28

5. Differences in variables characterizing samples of P. intermedius taken in different
habitats at SMMAP in the November 2002 sampling period ......................................29

6. Differences in variables characterizing samples of P. intermedius taken in different
habitats at SMMAP in the May 2003 sampling period. ..............................................30

7. Differences in variables characterizing samples of P. intermedius taken in different
habitats at SMMAP in the February 2003 sampling period ......................................31

8. Differences in variables characterizing samples of P. intermedius collected from
different habitats in the Weeki Wachee estuary in August 2003. ............................32

9. Differences in variables characterizing samples of P. intermedius in different habitats
in different estuaries during A ugust 2003. ............................ .................................. 33

10. Characteristics of seagrass vegetation in different estuaries in August 2003 .............34

11. Water chemistry in the SMMAP, Weeki Wachee and Steinhatchee estuaries in
A ugust 2003 ................... .............................................. ......... 35


.35















LIST OF FIGURES


Figure pge

1. Significant temporal changes in abundance and ISF variables in seagrass at SMMAP
(error bars represent 5t and 95th percentiles in a and b and 95% confidence
intervals in c, d, and e). .......................... .................. ......................... 2 1

2. Size distribution of shrimp populations in seagrass beds in different sampling periods.. 22

3. Temporal differences in median size of shrimp from marsh edges at SMMAP (error
bars represent 5th and 95th percentiles). .............. ...................... ..................... 23

4. Size distribution of shrimp populations in marsh edges in different sampling periods. 24















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science


THE RELATIVE VALUE OF SEAGRASS, MARSH EDGE, AND OYSTER
HABITATS TO THE BRACKISH GRASS SHRIMP, Palaemonetes intermedius
ALONG THE GULF COAST OF FLORIDA

By

Daniel Scott Goodfriend

May 2004

Chair: Thomas Frazer
Major Department: Fisheries and Aquatic Sciences

Palaemonid shrimp are abundant and important components of estuarine faunas

throughout much of the world, resulting in considerable interest in their ecology. Studies

of the relative value of habitats to these animals have primarily used measures of relative

abundance or habitat selection by individuals as indicators of habitat quality. However,

these have produced varied and conflicting results. As such, new techniques are needed

for investigating the relative quality of habitats for grass shrimp.

This study examined the relative value of seagrass beds, marsh edge habitats, and

oyster bars to the brackish grass shrimp, Palaemonetes intermedius, using a variety of

metrics that either directly measure, or provide surrogate measures for, fecundity,

mortality, and growth rates. These metrics were compared with measures of relative

abundance to determine whether abundance might be an effective indicator of habitat

quality. Sampling was done during July, August, and November 2002, and February,









May, July, and August 2003, in the St. Martin's Marsh Aquatic Preserve (SMMAP) near

Crystal River, Florida. This provided an assessment of temporal variability in the quality

of these habitat types as well as their overall relative value. In August 2003, the study

was broadened to include the Weeki Wachee and Steinhatchee estuaries in order to assess

the degree of regional variability in the quality of these habitats for grass shrimp.

In contrast with several other studies, P. intermedius was never collected from

oyster habitat. In the SMMAP, seagrass beds provided higher quality habitat for grass

shrimp than marsh edges in the July/August, November, and May, when seagrass beds

had greater abundances of shrimp, as well as higher proportions of gravid females than

marsh edges. In February 2003, the two habitats were determined to be of similar value to

grass shrimp. As in the SMMAP, seagrass beds in Weeki Wachee provided a higher

quality habitat than marsh edges for grass shrimp. Within the Steinhatchee estuary,

however, the quality of marsh edge habitats was similar to, or marginally higher than,

that of seagrass beds. Differences in seagrass characteristics, species composition

specifically, may account for the lower abundance of P. intermedius in the sampled

seagrass beds within the Steinhatchee estuary. In general, however, grass beds were

determined to provide higher quality habitat than adjacent marsh edges in the broad study

area.















INTRODUCTION

Grass shrimp (Palaemonetes spp.) are abundant components of coastal benthic

communities throughout much of the world, including along the Gulf and east coasts of

the United States (Rozas and Minello 1998, Lewis and Foss 2000, Glancy et al. 2003).

They are important consumers of organic detritus, and in turn serve as prey for a myriad

of commercially and recreationally important fish. As such, there has been substantial

interest in their ecology, and in particular, their use of habitats. A habitat should be

defined as high quality if the animals inhabiting it have increased fitness relative to those

in an alternative habitat, as indicated by measures of fecundity, mortality, and growth

rates (Anderson and Gutzwiller 1996, Davenport et al. 2000, Franklin et al. 2000, Lin and

Batzli 2001, Luck 2002, Walters et al. 2002, Ross 2003). Delineation of high quality

habitats is, therefore, an important step toward the effective management and

conservation of animals, including grass shrimp.

There have been numerous attempts by ecologists to determine the relative quality

of estuarine habitats for grass shrimp (Sheridan 1992, Knowlton et al. 1994, Khan et al.

1995, Eggleston et al. 1998, Rozas and Minello 1998, Bass et al. 2001). The vast majority

of these studies have used relative abundance, density, or habitat selection by individuals

as indicators of habitat quality. These studies have substantially increased our

understanding of many aspects of the relationship between grass shrimp and their

habitats. For example, it has been demonstrated that congeneric species may select

different habitat types when given a choice in experimental and field settings (Knowlton









et al. 1994, Khan et al. 1995, Sheridan 1992), and that paleomonid shrimp can occur in

greater abundance in adjacent, but alternative habitat types within a given ecosystem (e.g.

Eggleston et al. 1998, Rozas and Minello 1998, Glancy et al. 2003).

Findings in some studies suggest, however, that grass shrimp are more abundant

in seagrass beds (e.g. Glancy et al. 2003), whereas others suggest that grass shrimp are

more abundant in marsh habitats (e.g. Rozas and Minello 1998). Others have shown

seasonal shifts in the relative abundance of grass shrimp between habitat types (Eggleston

et al. 1998). These inconsistencies imply that the processes determining habitat quality

for, and relative abundance of, grass shrimp may be fairly complex and variable over

both time and space and that new approaches are needed to investigate the relative quality

of estuarine habitats for grass shrimp.

Van Horne (1983) noted that there are several levels of increasing sophistication

by which managers delineate habitat for species of concern, but that most often,

especially for nongame species, relative density or abundance are used as indicators of

habitat quality. As noted above, these indicators are often used when investigating habitat

use by grass shrimp. However, as Van Horne (1983) pointed out, and as is apparent from

other studies, e.g., Schantz (1981), density and abundance are not always correlated with

habitat quality. In fact, density and habitat quality may be decoupled in six specific types

of situations: (1) when habitat use changes seasonally, (2) when social dominance

interactions result in large densities of subdominant individuals in suboptimal habitats,

(3) when resources and other environmental conditions are temporally unpredictable

within habitats, (4) when the species have high reproductive capacity, which can result in

population sizes that poorly reflect the carrying capacity of the environment over the









short term, (5) when there is spatial habitat patchiness and (6) when the species in

question is a habitat generalist, which can cause the spatial distribution of habitats to have

a greater influence over the distribution of populations than the relative quality of those

habitats.

Interestingly, none of the papers cited in Van Home's review were of studies from

estuarine or marine systems, or of invertebrates. Also, none of the studies of grass

shrimp, cited above or otherwise encountered, examined any potential indicators of

habitat quality other than relative abundance or habitat selection by individual shrimp.

Nonetheless, many life history and ecological characteristics of grass shrimp comply with

Van Home's criteria. Seasonal shifts in habitat use by grass shrimp have been

documented (Kneib 1987, Eggleston et al. 1998), grass shrimp live in estuarine

environments that exhibit pronounced variability, and also invest significantly in

reproduction (Vernberg and Piyatitivorakul 1998). Moreover, there is considerable

patchiness in the distribution and species composition of seagrass beds (Hovel et al. 2002),

one of the primary habitats in which grass shrimp are found (Glancy et al. 2003). Finally,

grass shrimp are generalists, tolerating wide ranges in many physiochemical parameters

(e.g. temperature and salinity) and utilizing multiple food sources (Vernberg and

Piyativorakul 1998).

In combination, these life history characteristics and ecological attributes suggest

that the primary methods used by estuarine ecologists to investigate the relative value of

habitats for grass shrimp, i.e.,,, preference and relative abundance, may be inadequate.

Ideally, the relative quality of habitats for a species should be measured via comparison

of population densities or relative abundance and demographic parameters such as









fecundity, mortality, and growth rates. This was done by Chockley and St. Mary (2003),

who identified and monitored individual banded coral shrimp (Stenopus hispidus), and

showed that low density, inshore populations produced more eggs per area of habitat than

high density, offshore populations, which implied that the inshore areas were higher

quality habitats despite the fact that they supported lower shrimp abundances.

In an attempt to extend this type of habitat assessment to grass shrimp, this study

looks more closely at the relative values of key habitats to the brackish grass shrimp,

Palaemonetes intermedius, in terms of abundance and other measures that should

indicate relative fitness of the shrimp within those habitats, termed individual shrimp

fitness (ISF) variables. These ISF variables include measures of fecundity, growth, and

several surrogate measures of mortality which were selected based on relevant literature.

Abundance estimates were then compared to the ISF measures of habitat quality to

determine whether or not abundance was an effective indicator of habitat quality for grass

shrimp.















STUDY AREA

Sampling was conducted primarily at three sites within the St. Martin's Marsh Aquatic

Preserve (SMMAP), along Florida's northeast Gulf coast in Citrus county (280 53' N,

820 41' W). These three sites are the same as those previously described and sampled by

Glancy et al. (2003), and the close proximity of oyster bar, seagrass, and marsh edge

habitats at these sites allow a comparison of habitat quality in similar environments, as

characterized by water temperature, salinity, and dissolved oxygen concentrations.

In brief, this estuary is a complex mosaic of seagrass beds, salt marshes, oyster

bars, tidal channels, bays, and sandy flats. Seagrass beds are primarily comprised of turtle

grass (Thalassia testudinum), shoal grass (Halodule wrightii), and manatee grass

(Syringodium filiforme). Intertidal oyster habitats are dominated by the eastern oyster

(Crassostrea virginica), are low relief, and typically have a substrate of sand, mud, and

shell fragments overlying a limestone base. Salt marshes are dominated by cordgrass

(Spartina alterniflora) and black needle rush (Juncus roemarianus), but are interspersed

with small stands of black (Avicennia germinans) and red (Rhizophora mangle)

mangroves. As a whole, the estuary receives freshwater input from freshwater springs

and the spring-fed Crystal River.

In August 2003, additional sites were sampled in the Weeki Wachee and

Steinhatchee estuaries to determine if patterns similar to those observed within the

SMMAP occur at other Florida Gulf coast locations. The Weeki Wachee estuary is

located at approximately 280 32.5' N, 820 39.5' W, and is characterized by a similar









mosaic of habitats to SMMAP, although with a decreased prevalence of intertidal oyster

bars. As a result, no oyster habitats were sampled in this estuary. Salt marsh and seagrass

flora are typically the same as in the SMMAP (Frazer et al. 1998 and 2003), and the

major input is from the spring-fed Weeki Wachee River. Three sites were sampled within

this estuary, with selection based on similarities in size, water flow, and depth to the sites

at SMMAP, as well as the close proximity of salt marsh and seagrass habitats.

The Steinhatchee estuary is located at approximately 2939.5' N, 830 25.5' W,

and is also characterized by a similar mosaic of habitats to SMMAP, although with a

decreased prevalence of intertidal oyster habitats in close proximity to seagrass beds. The

seagrass beds at the sites in this estuary tended to be even more heavily dominated by

turtle grass and manatee grass than the other two estuaries, and to have a decreased

prevalence of shoal grass relative to the other estuaries. Salt marshes here also tended to

have a decreased prevalence of both black and red mangrove trees, and to be found

predominantly along the coasts rather than occurring as islands. The Steinhatchee estuary

receives freshwater primarily from the Steinhatchee River, which is fed by more surface

water runoff, and fewer spring inputs, than the Crystal or Weeki Wachee Rivers. Seven

sites were sampled in this estuary. Three of these had a close association of seagrass and

marsh edge habitats and were selected based on similarity in size, water flow, and depth

to the sites sampled in other estuaries. Three of the sites in this estuary only contained

seagrass, but were similar to the others in all other respects. The seventh site contained

only intertidal oyster habitat.















METHODS

Field and Laboratory Methods

Sampling within SMMAP was conducted during July, August, and

November 2002, and February, May, July, and August 2003. As there were no statistically

significant within year or among year differences in any of the variables measured during

July and August sampling periods in 2002 and 2003, summer sample data were pooled

and are hereafter referred to simply as the July/August sampling periods.

After the May 2003 SMMAP sampling period, it was clear that the only sampling

period during which all the ISF and abundance variables used in this study could be

estimated in all habitats was July/August, which the data suggested was the primary

breeding season, a conclusion that was expected based on relevant literature. Therefore,

in August 2003, the sampling was broadened to include the Weeki Wachee and

Steinhatchee estuaries to determine whether the patterns observed at SMMAP were

representative of the Gulf coast region as a whole during the grass shrimp breeding

season. Sampling of seagrass vegetation (see below) was also intensified during this

period.

Sampling, at all times and locations, was performed within 3 h of low tide to

minimize the variability associated with changes in habitat use due to tidal fluctuations.

This schedule facilitated comparisons to data collected and reported by Glancy et al.

(2003) who also sampled during low tides.









Seagrass beds, oyster bars, and marsh edges were sampled at each study site in

each time period with ten 5-m sweeps with a standard D-frame sweep net 25 cm high by

35 cm wide. Sweep net sampling has been previously used in several estuarine habitat

types to compare the relative abundances of fauna (e.g. Young et al. 1976, Young and

Young 1978, Posey and Hines 1991, Townshend 1991, and Posey et al. 1999), and

controlled tests for potential habitat bias using the sweep net collection method indicate

that the method is appropriate for the habitats samples in this study (T.K. Frazer and

M.H. Posey, pers comm.). Locations of sweeps in seagrass beds and in oyster habitats

were chosen in an unbiased manner by taking a random number of paces (between 1 and

30) at a random bearing from the end of the previous sweep. Marsh edge sweeps were

located similarly, but did not include a random bearing, as the edge of an island is

measurable only in one dimension. The initial sweep in each habitat type was selected

haphazardly by sighting a location in the habitat from the boat. All Palaemonetes spp.

individuals caught in each sweep were put in a sealable plastic freezer bag, placed on ice,

and transported to the lab for further analysis.

In seagrass beds, after every third sweep, a 0.25 m2 quadrat was placed on the

benthos to estimate the percent areal cover of each vascular plant species present, as well

as total grass coverage. This was primarily done looking down at the benthos from above

the water line, but when water clarity was too poor for the bottom to be seen, or species

to be easily identified, a mask and snorkel was used. In August 2003, the plant survey

sample size was increased to 10; estimates of percent areal cover were made at the end of

every sweep. In August 2003, a small ponar grab was used to take ten 625 cm2 bottom

samples from each habitat/site, to an approximate sediment depth of 8 cm. Samples were









individually bagged in sealable plastic freezer bags and brought back to the lab for

analysis. A YSI electronic meter model 650 was used to measure salinity, temperature,

and dissolved oxygen (DO) concentrations in each habitat type within each site at a depth

of approximately 0.25 m below the surface.

All grass shrimp within each of the bagged sweep samples were enumerated to

allow for estimates of relative abundance. Subsequent processing of shrimp collected in

the sweep nets was carried out with a random subsample of 50 individuals. The shrimp

for the subsample were selected randomly by counting all shrimp from the 10 sweeps

associated with a given habitat/site/sampling period and assigning each a number. Then

50 random numbers between 1 and the total number of shrimp counted were selected and

those shrimp were selected for the subsample. This subsample size was determined to be

sufficient to detect differences between samples in all variables after measuring the

variability associated with measurements from the first 200 shrimp. If less than 50 shrimp

were caught from a specific habitat type within any given time period, all shrimp were

processed as described below. Shrimp were identified as either Palaemonetes pugio or

P. intermedius, according to Abele and Kim (1986). The relative proportion of

P. intermedius in each of the subsamples was multiplied by the total Palaemonetes counts

from the respective sweep samples to estimate the abundances of P. intermedius.

In the laboratory, all subsampled P. intermedius were measured for total length. In

addition, the sex, reproductive condition (gravid/not gravid), and clutch size of gravid

females was recorded. The total length of each shrimp, as well as telson lengths of the

first 200 shrimp, was measured using a WILD M3Z dissecting microsope fitted with a

KR 221 120-increment optical micrometer. Measurements were recorded to the nearest









0.15 mm. Shrimp were considered female if they were gravid, or if developing ovarian

tissue could be observed through the translucent body of the individual. If these

indicators were not observed, then the second pleopod of the shrimp was examined to

determine if it bore an appendix masculina, a male characteristic (Berg and

Sandifer 1984). Shrimp without this modification of the second pleopod were counted as

female. Shrimp with a total length less than 12 mm were not included in estimates of sex

ratio or fecundity values as shrimp this small may not have been sexually mature, in

which case the appendix masculina may not have developed. The clutch size of gravid

females was determined by teasing apart egg masses with a fine metal probe and counting

individual eggs with the aid of a dissecting microscope.

For the bagged vegetation samples from the ponar grabs, the above-ground

vegetative portions were separated from the mud, detritus, rhizomes, and roots by hand.

The below-ground portion, as well as nonliving material, was discarded. Seagrass shoots

were then counted and total number of shoots for each seagrass species recorded. Above-

ground biomass (wet weight) of each species of seagrass in the ponar samples was

recorded and weighed with a Pesola 1000 g hanging scale and measurements recorded to

the nearest 10 g.

An instantaneous growth rate technique similar to that employed by Quetin and

Ross (1991) was used to estimate shrimp molting rates and growth increments in

SMMAP during November 2002, and February, May, and August 2003 in SMMAP only.

P. intermedius from each habitat/site in each sampling period were collected with sweep

nets and/or a beach seine. The target sample size for each habitat/site combination was

100 shrimp. When shrimp densities were too low to collect 100 shrimp, the number of









shrimp caught in approximately 1.5 h was used. Individual shrimp were placed in pre-

labeled 265-ml glass jars with mosquito-netting tops secured by rubber bands. The jars

were placed in plastic tubs and left immersed in a readily accessible nearshore area that

had similar salinity, temperature, and DO values as the sites from which the shrimp were

collected. These jars were monitored every 12 h for 3 d, and the shrimp that molted, as

well as their exuviae, were preserved in a solution of 90% ETOH and 5% glycerin, and

brought back to the lab for analysis.

The number of shrimp that either escaped from the jars or died in any given

experiment (< 3% in all experiments) was subtracted from the original number of total

shrimp that were used. This number was further adjusted to estimate the total number of

P. intermedius that were used throughout the experiments by multiplying it by the

proportion of P. intermedius in the subsample from the sweep net samples which were

collected in the same location during the same sampling period. The number of

P. intermedius that were directly observed to molt during the course of the 3-d

experiment was then divided by the estimated total number of P. intermedius, to calculate

a molting frequency per 3 d. Intermolt period (IMP) was then calculated as the reciprocal

of molting frequency, and expressed as days molt1.

In the laboratory, telson lengths of the preserved shrimp and their molts were

measured. Species, sex, and reproductive condition of female shrimp were also recorded.

The growth increment of molting shrimp (mm) was calculated as the difference between

the estimated total lengths of the post-molt shrimp and their molts. Total lengths of the

molts and post-molt shrimp were estimated using measured telson lengths and









Equation 1, which was derived from a linear regression analysis between measured telson

and total lengths for the first 200 shrimp from the sweep net subsamples (p < 0.0001, r2 =

0.87).

total length= 1.92 + (7.36 telson length) eq. (1)

Growth rate was defined as the average growth (mm/day) for an individual P. intermedius

within a sample population, and was calculated by dividing the average growth

increment (mm) by the IMP (days) for each habitat/site/sampling period combination.

This provided an estimate of continuous growth which was a population-wide average,

individual shrimp growth is incremental. The estimated values for IMP, growth

increment, and growth rate were assumed to be indicative of the sample population as a

whole for the time period in which the experiment took place.

The combination of field, laboratory and mathematical techniques described

above resulted in estimates for a variety of variables which provide information about

P. intermedius. These fall into two general categories. (1) measures of abundance, and

(2) those variables that provide information about the relative fitness of shrimp

(individual shrimp fitness, or ISF, variables). The latter include measures of fecundity

(proportion of gravid females and fecundity), and surrogate measures of mortality and

nutritional stress (sex ratio, size distribution, and growth rates).

These surrogate measures were necessary because the mortality of a high density,

small-bodied, mobile species such as P. intermedius, is extremely difficult to measure

directly in an open estuarine habitat. The validity of these surrogates is supported by the

results of previous investigations. Females have been shown to respond to decreased food

availability with increased mortality relative to males (Reinsel et al. 2001). Also,









population size structure has been shown to be shifted towards smaller shrimp in

ecosystems with increased predation on P. pugio and with increased mortality due to

nutritional stress (Cross and Stiven 1999, Bass et al. 2001, Reinsel et al. 2001). In

addition, reduced food availability can result in decreased intermolt growth increments

and increased intermolt periods in crustaceans in general (Hartnoll 2001). As such,

population size and sex structure, as well as growth rates, were used as the surrogate

measures of relative mortality and nutritional stress for shrimp. In addition, histograms of

size distributions of the shrimp from each habitat/time period were created to further

explore the relationships between life history, habitat quality, and shrimp size.

Statistical Methods

Within each sampling period and estuary, one way Analyses of Variance

(ANOVAs) were performed with each of the numerical variables (abundance, size, clutch

size, growth increment, total growth, as well as percent cover, biomass, and shoot density

of seagrass and individual seagrass species) as dependent variables and habitat type as the

independent variable (MINITAB Release 14). When necessary, variables were logo

transformed to achieve normality and homoscedasticity and all tests were assumed

significant at p < 0.05. Tukey's post-hoc test was used to determine where significant

differences existed when ANOVAs detected significant effects (MINITAB Release 14).

Student's t-tests were used when only 2 habitats were compared within a particular

season (MINITAB Release 14). When the assumptions of normality and homogeneity of

variance could not be met, a Kruskal Wallis ANOVA was used as an alternative to

ANOVA and Mann-Whitney U-tests were used as alternatives to T-tests (MINITAB

Release 14). When Kruskal-Wallis ANOVA was used, post-hoc comparisons were made

by comparing the intervals between the 5th and 95th percentiles of each distribution. Only










three values for intermolt period, one from each site within the SMMAP, could be

calculated for each habitat/sampling period with the method employed in this study. To

increase statistical power, the molting frequency of P. intermedius (molts day-) was used

as a proportional variable and analyzed via X2 tests (MINITAB Release 14) with the

distribution of the molting shrimp assumed to be independent of habitat or time period.

X2 tests were also used to compare the other proportional variables (sex ratio and

proportion of gravid females), between habitat types, time periods, and estuaries

(MINITAB Release 14). When there were only two categories being compared, i.e.,

marsh edge vs. seagrass, overall test significance at a < 0.05 was sufficient to infer

significant differences between mean values. When there were more than two categories

being compared, such as time periods or estuaries, and the X2 test was significant at

a < 0.05, then statistically significant differences were inferred by comparing 95%

confidence intervals. In all cases, the variables were assumed to be independent of

habitat, time period, or estuary. Sampling periods were only compared within the

SMMAP, and among estuary comparisons were only made with data from August 2003.














RESULTS

P. intermedius were not collected from oyster bar habitats during the course of

this investigation. As a consequence, subsequent analyses focus only on grass shrimp in

marsh edge habitats and seagrass beds.

Temporal Variability

Within Habitat Variability

As expected, because of the close proximity of sampling sites within the

SMMAP, there were no statistically significant differences in salinity, dissolved oxygen

concentrations, or water temperature between habitat types within any given sampling

period. Thus, data from all habitats were pooled for all statistical comparisons. Dissolved

oxygen concentrations and salinity were always well within the ranges reported for

Palaemonetes (Vernberg and Piyatitivorakul 1998, Stickle et al. 1989, Bass et al. 2001,

Table 1). Statistically significant temporal changes in water temperature were consistent

with expected seasonal patterns (Frazer et al. 1998 and 2003, Table 1), but were within

the range reported for grass shrimp (Knowlton and Schoen 1984). Water temperature was

significantly lower during February 2003 (12.690C) than in any other time period

(Table 1). Although water temperature was higher in November 2002 (18.800C) than in

February 2003, the mean value was lower than in either July/August (29.570C) or

May 2003 (29.730C) sampling periods, which were similar to one another (Table 1).

Within seagrass beds, there were significant temporal differences in the relative

abundance of P. intermedius, and several ISF variables, specifically total length, sex









ratio, the proportion of gravid females, and IMP. Surprisingly, there were no significant

differences observed for the percent cover of any individual seagrass species or cover of

seagrass in general. The median abundance of P. intermedius was significantly lower in

February 2003 (1.5 shrimp/sweep) than in any other sampling period (Figure la,

Table 2), coinciding with the period when mean water temperatures were lowest (Table

1). Gravid females were not collected from seagrass beds in February. (Figures Id,

Table 2). In May 2003, shrimp were more abundant than in February (14.5 vs. 1.5

shrimp/sweep), although smaller on average (Figures la,b, 2b,c, Table 2). Nineteen

percent of females were gravid in May (Table 2). In the July/August sampling periods,

the median abundance of shrimp was similar to that in May 2003 (18.0 shrimp/sweep),

but shrimp were significantly larger (Figures la,b, 2a,d, Table 2). In May 2003, 77.8% of

the shrimp were < 15 mm total length and probably not sexually mature (Figure 2d). In

comparison, 70.1% of captured shrimp during the July/August sampling periods

were > 15 mm total length, indicating that the majority of the sampled population was

sexually mature (Figure 2a). Changes in ISF variables were consistent with previously

documented life history information for P. intermedius. The proportion of gravid females

in July/August was 47%, indicative of a peak reproductive period (Figure Id, Table 2).

Shrimp abundance was higher in November 2002 than in summer (46.5 vs. 18.0

shrimp/sweep), though size distribution was skewed towards smaller size classes

(Figure 2b), reflecting a late summer/early fall recruitment period. In November 2002,

87.1% of the shrimp were < 15 mm total length (Figure 2b), and only 23% of the larger

shrimp were gravid (Figure Id, Table 2). In addition, the IMP for the sampled population









was 8.4 d in November, which was significantly longer than in either July/August (5.5

days) or May (4.4 days) (Figure le, Table 2).

Marsh edge habitats exhibited less temporal variability than seagrass beds. There

were no statistically significant differences in the relative abundance of P. intermedius

within the marsh edge habitats sampled during the different time periods in SMMAP

(Table 3). In fact, no variables exhibited significant or large temporal differences other

than size and the proportion of gravid females (Table 3). Shrimp occupying marsh edges

were significantly larger in July/August (median TL = 22.929 mm) than in May

(median TL = 12.714 mm). In November and February, shrimp were intermediate in size

(15.710 mm and 16.429 mm, respectively), but did not differ significantly from the

summer sampling period (Figure 3, Table 3). A majority (94.4%) of P. intermedius

sampled in marsh edges during summer were > 15 mm and 27% of the females caught

were gravid (Table 3, Figure 4). In fact, gravid females were only captured from marsh

edges in summer. During the other sampling periods, there was a much greater

proportion of individuals in the < 15 mm size classes (Figure 4).

In summary, while the highest shrimp abundances in seagrass beds occurred in

November 2002, the ISF variables measured indicated that conditions during July/August

were the most appropriate for the cross-habitat comparisons central to this study.

SMMAP Cross-Habitat Comparisons (Seagrass vs. Marsh Edges)

Shrimp were significantly more abundant in seagrass than in marsh edges during

every sampling period (Tables 4-7). Seagrasses yielded somewhere between twice as

many shrimp in February and sixty times as many shrimp in November 2002

(Tables 4-7). During the July/August sampling period, the proportion of gravid females

was also higher in seagrass (48%) than in marsh edge habitats (27%). Average shrimp









size in July/August, however, was less in seagrass beds (19.143 mm) than in marsh edges

(22.929 mm) (Table 4) as was clutch size (42.5 eggs/gravid female in seagrass beds and

89.0 eggs/gravid female in marsh edge habitats). Differences in shrimp size confounded

a rigorous comparison of growth parameters (Table 4, Figures 2a, 4a) during the summer

sampling periods. Estimates of fecundity at the population level were made by

multiplying shrimp abundance by % females in the sample population (averaged between

marsh edge and seagrass values due to lack of significant differences), by % gravid

females, and by clutch size. These estimates showed that the average number of

eggs/sweep was 209.3 in seagrass, and 0.7 in marsh edges, which strongly suggests that

seagrass provides a higher quality habitat for P. intermedius.

Although the July/August sampling period was determined to be the most

appropriate time to compare measures of relative abundance with other ISF variables, the

estimates of abundance and ISF variables during other sampling periods provided an

opportunity to make additional seagrass/marsh edge comparisons. In November 2002, for

example, gravid females were only encountered in seagrass (23% gravid and 28.0

eggs/gravid female), but shrimp from marsh edges were larger (9.714 mm in seagrass and

15.710 mm in marsh edges) (Table 5). When fecundity was compared at the population

level, the average number of eggs/sweep was 143.2 in seagrass and 0.0 in marsh edges.

In May 2003, shrimp abundances and ISF variables both indicated that seagrass

provided higher habitat quality than marsh edges. IMP was shorter in seagrass beds than

in marsh edge habitats (4.4 d in seagrass and 8.1 d in marsh edge), while mean shrimp

size and size distributions were similar, implying faster growth in seagrass beds than

marsh edges for individual shrimp (Table 6).









Interestingly, in February 2003, there was a higher proportion of females in marsh

edge habitats than in seagrass beds (47% females in seagrass and 78% in marsh edges)

(Table 7). However, the difference in abundance between seagrass and marsh edges was

of lesser magnitude than during other sampling periods (1.5 shrimp/sweep in seagrass

and 0.7 shrimp/sweep in marsh edges) (Table 7). When female abundance was estimated

by multiplying relative abundance by sex ratio, the average number of females/sweep

was 0.70 in seagrass and 0.55 in marsh edges, suggesting that these two habitats were

capable of providing resources for similar numbers of female shrimp.

Among Estuaries Comparisons

Cross-habitat comparisons in the Weeki Wachee and Steinhatchee estuaries

produced conflicting results. In the Weeki Wachee estuary, abundance was significantly

greater in seagrass than marsh edges (11.5 shrimp/sweep in seagrass and 0.6

shrimp/sweep in marsh edges), although there were no significant differences between

habitat types for any other ISF variables (Table 8). Within the Steinhatchee estuarine

area, there were no significant differences between seagrass and marsh edge habitats in

either abundance or any of the ISF variables measured.

Among the three estuaries in this study, Steinhatchee had the lowest abundance of

P. intermedius in seagrass, and highest abundance of P. intermedius in marsh edges

(Table 9). There were only two ISF variables which showed significant differences

among estuaries. i.e., proportion of gravid females in seagrass beds and size in marsh

edge habitats. In Weeki Wachee seagrass beds, the proportion of gravid females (35%)

was higher than in seagrass beds in the Steinhatchee (18%) or SMMAP (26%). Median

total lengths were significantly less for shrimp occupying marsh edges in Steinhatchee









(17.429 mm) than in the other two estuaries (22.857 mm in Weeki Wachee and 26.000

mm in SMMAP) (Table 9).

There were also several differences detected in the species composition and

biomass of the seagrass beds among estuaries (post-hoc comparisons made by

comparison of intervals between the 5th and 95th percentiles), which may have

significance for the shrimp. The percent areal cover and shoot density of Thalassia

testudinum was different in each estuary, lowest in Weeki Wachee (0.0% cover, 9.6

shoots/m2), higher in SMMAP (0.7% cover, 11.6 shoots/m2, and highest in Steinhatchee

(80.0% cover, 14.4 shoots/m2) (Table 10). Percent areal cover of Halodule wrightii was

different in each estuary, and followed the reverse spatial pattern, increasing from

Steinhatchee (1.9% cover) to SMMAP (80.0% cover) to Weeki Wachee (100.0% cover)

(Table 10). Above-ground biomass of H. wrightii, however, was higher in SMMAP

(624.0 g/m2) than in Weeki Wachee (280.0 g/m2) (Table 10). While there were

statistically significant differences in salinity and dissolved oxygen concentrations among

the estuaries, these were well within the ranges reported for P. intermedius. There was

not a significant difference in water temperature between the estuaries (Table 11).












b)
b) 35-
30-
-25-
S20-
S15-
S 10-
5-


July/ Nov 02 Feb 03 May 03
August
02+03


d) 0.6-
0.5-
0.4-
0.3 -
0-u
0.2-
0.1 -


July/ Nov 02 Feb 03 May 03
August
02+03









n nm


July/ Nov 02
August
02 + 03


400 -
S 375 -
l350-
75-
50 -
25
0


July/
August
02+03


Nov02 Feb03 May 03


Nov02 Feb03 May 03


Feb 03 May 03


Fig la-e. Significant temporal changes in abundance and ISF variables in seagrass at
SMMAP (error bars represent 5th and 95th percentiles in a and b and 95%
confidence intervals in c, d, and e). a) abundance. b) total length. c) sex ratio. d)
Proportion of gravid females e) IMP.


July/
August
02+03


0.6
S 0.5
| 0.4
0o .3
FC


) 0

60 0
CTJ *
Ia


-Fri


v


r












70
60
S50
S40
0

8 20
10
0


Size class (mm)


d)
70 -

60 -
S50 -
S40 -
0
5 30 -
20 -
10 -
0


Size class (mm)


H


H -7


Size class (mm) Size class (mm)

Fig. 2a-d. Size distribution of shrimp populations in seagrass beds in different sampling
periods, a) July/August 2002 and 2003. b) November 2002. c) February 2003.
d) May 2003.


70
60
50
S40
4-
30
2 20
10


Y~"o~*,~,~".4", p














30 -


25 -



20 -


15 -
C

S 10 -


5



0
July/August Nov 02 Feb 03 May 03
02 + 03


Fig. 3. Temporal differences in median size of shrimp from marsh edges at SMMAP
(error bars represent 5th and 95th percentiles).















a) b)
70 70 -

60 60 -

2 50 2 50 -
40 40 -
0 0
S30 30 -

20 20 -
10 10 -
0 F|i- i I -r -t- 0


Size class (mm) Size class (mm)


c) d)
70 70 -

60 60 -

S 50- 2 50-
40 40 -

30 30 -

S20 20 -

10 10 -

0 0


Size class (mm) Size class (mm)


Fig. 4a-d. Size distribution of shrimp populations in marsh edges in different sampling
periods. a) July/August 2002 and 2003. b) November 2002. c) February 2003.
d) May 2003.












Table 1. Water chemistry at SMMAP by sampling period.


Variable Test Used p value


Mean
(95% Confidence Interval)


July/August


November
2002


May 2003


ANOVA < 0.001


8.80


13.36


10.07


5.41


(7.96-9.64) (11.44-15.28) (9.53-10.61) (5.08-5.74)


Salinity
(ppt)


Water
temperature
(C)


16.33
ANOVA 0.476
(13.49-19.17)


ANOVA < 0.001


29.57
(28.79-30.35)


19.54
(18.27-20.81)


18.80
(18.37-19.23)


14.06
(13.25-14.87)


12.69
(12.28-13.10)


15.11
(13.71-16.51)


29.73
(29.18-30.28)


Dissolved
oxygen
(mg/1)


February
2003












Table 2. Temporal changes in variables characterizing samples of P. intermedius taken from seagrass at SMMAP.


Variable



Abundance
(shrimp/sweep)

Total length (mm)

Sex ratio
(female/male)
Proportion gravid
females


Clutch size


Test Used



Kruskal-Wallis
ANOVA
Kruskal-Wallis
ANOVA


Kruskal-Wallis
ANOVA


Intermolt period


Growth increment


Growth rate mm/day


Kruskal-Wallis
ANOVA
Kruskal-Wallis
ANOVA


p value

July/August
18.0
< 0.0001
(3.0-21.1)
19.143
< 0.0001 19
(15.233-21.371)
0.55
0.006 0
(0.49-0.61)
0.47
< 0.0001
(0.39-0.62)
42.5
0.181 42
(16.6-102.5)
5.5
0.018
(4.6-6.5)

0.043
0.547043
(-0.098-0.230)
0.0009
0.1680009
(0.0003-0.0020)


Central Tendency*
(95% confidence interval)**
November 2002 February 2003
46.5 1.5
(22.0-352.8) (0.2-2.6)
9.714 16.571
(5.571-12.824) (15.979-30.214)
0.49 0.47
(0.43-0.56) (0.25-0.69)
0.23 0.00
(0.17-0.30) (0.00-0.00)
28.0 ***
(18.0-47.0) ***
8.4 15.6
(7.0-16.4) (6.0-25.2)
0.079 0.030
(-0.139-0.127) (-0.079-0.180)
0.001 0.0002
(0.0006-0.0014) (0.0000-0.0007)


May 2003
14.5
(4.0-18.2)
12.429
(8.877-15.979)
0.61
(0.59-0.68)
0.19
(0.06-0.32)
29.0
(11.8-71.2)
4.4
(3.8-5.2)
0.049
(-0.089-0.243)
0.0017
(0.0010-0.0018)


* Mean when test is X2, median when test is Kruskal-Wallis ANOVA
** 5th and 95th percentiles when test is Kruskal-Wallis ANOVA
*** Sample size is 0 or 1, therefore data were not included in statistical tests and confidence intervals were not calculated.












Table 3. Temporal changes in variables characterizing samples of P. intermedius taken from marsh edges at SMMAP.


Central Tendency*
(95% confidence interval)**


July/August


Abundance
(shrimp/ sweep)
Total length
(mm)
Sex ratio
(female/male)
Proportion
gravid females

Clutch size


Intermolt period


Kruskal-Wallis
ANOVA
Kruskal-Wallis
ANOVA


0.331


0.6
(0.2-4.5)


22.929
<0.001
(17.080-23.845)

0.06
0.468
(0.00-0.33)

0.27
(0.11-0.43)
89.0
(16.0-116.6)


0.326


November
2002


0.8
(0.1-7.9)
15.710
(12.786-22.786)
0.46
(0.00-1.00)
0.00
***


4.8
(4.4-14.3)


February 2003

0.7
(0.1-3.0)
16.429
(14.343-21.286)
0.78


0.00
***


May 2003


0.6
(0.1-15.8)
12.714
(8.286-15.514)
0.53
(0.00-0.92)
0.00
(0.00-0.00)


8.1
(6.8-10.3)


Growth Kruskal-Wallis 0.006 0.010 0.008 0.000
0.233
increment ANOVA (0.001-0.013) (-0.070- 0.184) (-0.056-0.182) (-0.079-0.117)
Growth rate Kruskal-Wallis 0.0008 0.0015 *** 0.0000
0.194
mm/day ANOVA *** *** *** (-0.0001-0.0001
* Mean when test is X2, median when test is Kruskal-Wallis ANOVA
** 5th and 95th percentiles when test is Kruskal-Wallis ANOVA
*** Sample size is 0 or 1, therefore data were not included in statistical tests and confidence intervals were not calculated.


Variable


Test Used


p value


)









Table 4. Differences in variables characterizing samples of P. intermedius taken in
different habitats at SMMAP in the July/August 2002 and 2003 sampling
periods.


Variable



Abundance
(shrimp/sweep)

Total Length
(mm)

Sex ratio
(female/male)

Proportion of
females gravid


Clutch Size


Intermolt period


Growth increment


Test Used



Mann-Whitney
U Test

Mann-Whitney
U Test


Mann-Whitney
U Test


Mann-Whitney
U Test


Central Tendency*
p value
(95% Confidence Interval)**

Seagrass Marsh Edge
18.0 0.6
< 0.001
(3.0-30.3) (0.2-4.5)
19.143 22.929
< 0.001
(15.233-28.371) (17.080-23.845


0.773


0.015


0.0428


0.679


0.9213


Growth rate


0.55
(0.49-0.61)
0.47
(0.39-0.62)
42.5
(16.6-102.5)
5.5
(4.6-6.5)
0.006
(-0.098-0.230)
0.0009
(-0.0001-0.0020)


0.06
(0.00-0.33)
0.27
(0.11-0.43)
89.0
(16.0-116.6)
6.1


0.006
(0.0012-0.0129)
0.0008
***


* Mean when test is X2, median when test is Mann Whitney U Test
** 5th and 95th percentiles when test is Mann-Whitney U Test
*** Sample size is 0 or 1, therefore data were not included in statistical tests and
confidence intervals were not calculated.


)









Table 5. Differences in variables characterizing samples of P. intermedius taken in
different habitats at SMMAP in the November 2002 sampling period.


Test Used


p value


Central Tendency*
(95% Confidence Interval)**


Abundance
(shrimp/sweep)
Total length
(mm)
Sex ratio
(female/male)
Proportion of
females gravid


Mann-Whitney
U Test
Mann-Whitney
U Test


0.0002


< 0.0001


0.798


< 0.0001


Clutch size


Intermolt period


Growth
increment

Growth rate


Mann-Whitney
U Test
Mann-Whitney
U Test


0.259


0.259


Seagrass
46.5
(22.0-352.8)
9.714
(5.571-12.824)
0.49
(0.43-0.56)
0.23
(0.17-0.30)
28.0
(18.0-47.0)
8.4
(7.0-16.4)
0.011
(-0.139-0.127)


0.0007
0.083000
(0.0004-0.0010)


Marsh Edge
0.8
(0.1-7.9)
15.710
(12.786-22.786)
0.46
(0.00-1.00)
0.00
***


4.8
(4.4-14.3)
0.010
(-0.070-0.184)
0.0015
***


* Mean when test is X2, median when test is Mann Whitney U Test
** 5th and 95th percentiles when test is Mann-Whitney U Test
*** Sample size is 0 or 1, therefore data were not included in statistical tests and
confidence intervals were not calculated.


Variable









Table 6. Differences in variables characterizing samples of P. intermedius taken in
different habitats at SMMAP in the May 2003 sampling period.


Test Used


p value


Central Tendency*
(95% Confidence Interval)**


Abundance
(shrimp/sweep)

Total Length
(mm)

Sex ratio
(female/male)

Proportion of
females gravid


Mann-Whitney
U Test

Mann-Whitney
U Test


< 0.0001


0.91


0.337


< 0.0001


Clutch Size


Intermolt period


Growth increment


Growth rate


0.009


Mann-Whitney
U Test

Mann-Whitney
U Test


Seagrass

14.5
(4.9-27.7)


12.429
(8.877-15.979)


0.61
(0.59-0.68)

0.19
(0.06-0.32)

29.0
(11.8-71.2)

4.4
(3.8-5.2)


0.007
0.7388
(-0.089-0.243)

0.0017
0.0830017
(0.0010-0.0018)


Marsh Edge

0.6
(0.1-15.8)

12.714
(8.286-15.514)


0.53
(0.00-0.92)

0.00
(0.00-0.00)


8.1
(6.8-10.3)


0.000
(-0.079-0.117)

0.0000
(-0.0001-0.0001)


* Mean when test is X2, median when test is Mann Whitney U Test
** 5th and 95th percentiles when test is Mann-Whitney U Test
*** Sample size is 0 or 1, therefore data were not included in statistical tests and
confidence intervals were not calculated.


Variable









Table 7. Differences in variables characterizing samples of P. intermedius taken in
different habitats at SMMAP in the February 2003 sampling period.


Test Used


Central Tendency*
(95% Confidence Interval)**


Abundance
(shrimp/sweep)

Total Length (mm)


Sex ratio (female/male)

Proportion of females
gravid


Mann-Whitney U
Test
Mann-Whitney U
Test


Seagrass

0.0023 1.5
(0.2-2.6)

0.8777 16.571
(15.979-30.214)

)

27 0.47
(0.36-0.62)
0.00
(0.00-0.00)


Marsh Edge
0.7
(0.1-3.0)
16.429
(14.343-21.286)
0.78


0.00
***


Clutch Size


Intermolt period


Growth increment


Mann-Whitney U
Test


Growth rate


0.246 15.600
(6.0-25.2)

0.1622 0004
(-0.079-0.180)
0.0017
(0.0012-0.0021)


0.008
(-0.056-0.182)


* Mean when test is X2, median when test is Mann Whitney U Test
** 5th and 95th percentiles when test is Mann-Whitney U Test
*** Sample size is 0 or 1, therefore data were not included in statistical tests and
confidence intervals were not calculated.


Variable


cal tests and
confidence intervals were not calculated.


Variable









Table 8. Differences in variables characterizing samples of P. intermedius collected from
different habitats in the Weeki Wachee estuary in August 2003.


Test Used


Central Tendency*
p valueconfidence intervals)
(95% confidence intervals)**


Abundanc


Abundance
(shrimp/sweep)


Total Length (mm)


Sex ratio (female/male)

Proportion of females
gravid


Clutch Size


Seagrass


Mann-Whitney
U Test

Mann-Whitney
U Test


Mann-Whitney
U Test


< 0.0001


0.635


0.534

0.718


0.101


11.5
(4.4-27.0)
20.571
(16.686-
23.536)
0.54
(0.34-0.93)
0.35
(0.28-0.64)

32.5
(22.0-58.1)


Marsh Edge


0.6
(0.1-0.7)
22.857
(20.500-
27.929)
0.64
(0.34-0.93)
0.29
(0.00-0.65)

50.5
(42.0-59.1)


* Mean when test is X2, median when test is Mann Whitney U Test
** 5th and 95th percentiles when test is Mann-Whitney U Test


Variable









Table 9. Differences in variables characterizing samples of P. intermedius in different
habitats in different estuaries during August 2003.


Variable Test Used p value


Seagrass


Abundance


Shrimp
Total
Length

Sex Ratio

Proportion
of Females
Gravid


Clutch Size


Marsh Edge


Abundance


Shrimp
Total
Length


Kruskal-
Wallis
ANOVA
Kruskal-
Wallis
ANOVA


0.002


Central tendency*
(95% Confidence Interval**)


SMMAP


11.5
(7.4-26.2)


133 21.000
0.133(14.600-29.764)
(14.600-29.764)


2 0.55
X2 0.911
(0.04-0.58)

20.26
X 0.0120.2
(0.08-0.28)


Kruskal-
Wallis
ANOVA


Kruskal-
Wallis
ANOVA
Kruskal-
Wallis
ANOVA


27.0
0.354 27
(16.4-89.8)


0.023


0.6
(0.1-0.7)


26.000
0.004 (20.300-34.250)
(20.300-34.250)


Weeki
Wachee

11.5
(4.4-27.0)


20.571
(16.686-23.536)

0.54
(0.34-0.93)

0.35
(0.28-0.64)


32.5
(22.0-58.1)


0.6
(0.1-0.7)


22.857


Steinhatchee


0.7
(0.1-3.2)


18.288
(10.571-26.943)

0.53
(0.22-0.84)

0.18
(0.03-0.23)


39.0
(22.4-54.0)


0.8
(0.7-1.5)


17.429


(20.500-27.929) (13.400-19.929)


Sex Ratio

Proportion
of Females
Gravid


Clutch Size


0.60
X2 0.97860
(0.36-0.84)

20.25
X 0.7630.
(0.00-0.51)


Kruskal-
Wallis
ANOVA


89.0
0.1658.
(11.8-116.6)


0.64
(0.34-0.93)

0.29
(0.00-0.65)


50.5
(42.0-59.1)


0.62
(0.36-1.60)

0.17
(0.06-0.93)


32.5
(20.3-34.8)


* Mean when test is X2, median when test is Kruskal-Wallis ANOVA
** 5th and 95th percentiles when test is Kruskal-Wallis ANOVA









Table 10. Characteristics of seagrass vegetation in different estuaries in August 2003.


p value


Median
(5th, 95th percentile)


Weeki
SMMAP Wa
Wachee


Halodule % Areal
Cover
Halophila % Areal
Cover
Syringodium %
Areal
Cover
Ruppia %
Areal
Cover
Thallassia % Areal
Cover
Halodule
shoot density
(shoots/m2)
Halophila
shoot density
(shoots/m2)
Thalassia
shoot density
(shoots/m2)
Syringodium shoot
density
(shoots/m2)
Halodule
aboveground
biomass (g/m2)
Halophila
aboveground
biomass (g/m2)
Thalassia
aboveground
biomass (g/m2)
Syringodium
aboveground
biomass (g/m2)


Kruskal- Wallis
ANOVA
Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA

Kruskal- Wallis
ANOVA


80.0
< 0.001
(20.0-90.0)

1 0.0
(0.0-0.0)


0.55


0.0
(0.0-0.0)


0.0
100
(0.0-0.0)


< 0.001


0.7
(0.3-1.3)


792.0
< 0.001
(36.8-1651.2)

0.0
0.609 0
(0.0-0.0)

11.6
< 0.001
(1.6-17.6)

0.0
0.114
(0.0-0.0)

624.0
< 0.001 4.
(36.8-1368.0)

0.0
0.976
(0.0-0.0)

9.6
0.002
(1.6-614.4)

0.0
0.114
(0.0-0.0)


100.0
(90.0-100.0)
5.5
(1.0-10.0)

0.0
(0.0-0.0)

0.0
(0.0-0.0)

0.0
(0.0-0.0)

544.0
(80-1152.0)

0.0
(0.0-0.0)

9.6
(1.6-41.6)

0.0
(0.0-0.0)

280.0
(80.0-640.0)

0.0
(0.0-0.0)

8.0
(1.6-43.2)

0.0
(0.0-0.0)


Steinhatchee

1.9
(0.2-3.6)
0.0
(0.0-0.0)

0.6
(0.1-12.8)

0.0
(0.0-0.0)

80.0
(10.0-100.0)

14.4
(1.6-544.0)

0.0
(0.0-0.0)

14.4
(1.6-225.6)

11.2
(1.6-625.6)

14.4
(1.6-449.6)

0.0
(0.0-0.0)

14.4
(1.6-142.7)

11.2
(1.6-481.6)


Variable


Test Used









Table 11. Water chemistry in the SMMAP, Weeki Wachee and Steinhatchee estuaries in
August 2003.


Variable


Test Used p value


SMMA


Dissolved
oxygen (mg/1)

Salinity (ppt)

Water
temperature (C)


7.60
ANOVA 0.013 6
(6.30-8.5
9.99
ANOVA 0.007 9.
(9.39-10.
29.34
ANOVA 0.299
(28.76-29


Mean
(95% Confidence Interval)
Weeki
P Stei
Wachee
11.86
)0) (11.43-12.28) (1.
17.68
59) (16.40-18.95) (6.8
27.61
.91) (26.66-28.56) (29.


inhatchee

4.78
61-7.96)
10.52
85-14.19)
31.26
91-32.61)















DISCUSSION

Determination of the Relative Quality of Habitat Types

A habitat type was considered to provide higher habitat quality than another for

P. intermedius if one of the following three conditions were met: (1) if shrimp abundance

and ISF variables both indicated higher habitat quality in the same habitat or estuary,

(2) if shrimp abundance was similar between the two and the ISF variables indicated

greater fitness in one over the other, (3) if the ISF variables were similar, but abundance

was higher in one over the other. In the situations where abundance and ISF variables

proved to be contradictory, then fitness related measures and abundances were used to

estimate fecundity at the population level in an effort to provide additional insights into

the relative qualities of the habitats.

Temporal Variability within the SMMAP

SMMAP Within Habitat Sampling

Observed temporal changes in the demographics and physiology of grass shrimp

within the seagrass and marsh edge habitats suggest seasonal changes consistent with

those found in other studies (Kneib 1987, Knowlton et al. 1994, Grabe 2003). These

temporal changes provide important background information that should be considered

when comparing the relative habitat quality of seagrass beds and marsh edges for

P. intermedius. The results suggest that summer months are the most appropriate time to

compare and contrast estimates of abundance and fitness to infer relative habitat quality.









This can best be illustrated through discussion of results from each of the sampling

periods.

In February 2003, the relative abundance of P. intermedius was at its minimum in

SMMAP, similar to reduced winter abundances documented by Knowlton et al. (1994) in

North Carolina. The shift in sex ratio towards a more male-dominated population from

November 2002 to February 2003 in this study suggested that females may have begun to

suffer the effects of reduced habitat quality before males, which is consistent with

Reinsel et al. (2001), who showed that limited food availability caused increased

mortality of females relative to males. This increased susceptibility of females may be

due to the large amount of energy invested in reproduction (Vernberg and

Piyatitivorakul 1998). This large reproductive investment could, however, also support

the alternative conclusion that female shrimp are simply succumbing to the metabolic

demands of repeated reproduction during summer and late fall irrespective of changes in

the quality of the habitat in which they are found.

However, changes in other ISF variables in February 2003 also suggest declining

habitat quality in winter. Apparently spawning did not occur in February 2002, beacuse

gravid females were not collected, a pattern observed in other decapods during winter

months (e.g. Grabe 2003). In February 2002, P. intermedius size was at its maximum,

suggesting that only the largest individuals were capable of surviving an autumn-winter

mortality event. This result is consistent with that of Chockley and St. Mary (2003), who

found that mortality of Stenopus hispidus decreased with increasing shrimp size. In

addition, the long IMP during this sampling period suggested that conditions in the









seagrass beds, or the estuarine waters in general, were negatively influencing growth

rates in addition to precluding spawning.

In May 2003, an increase in shrimp abundance, proportion of gravid females, and

shorter IMPs suggested that conditions in the seagrass beds were improved relative to

winter. In the July/August sampling periods, conditions within seagrass beds were most

favorable for reproduction, as indicated by the high proportion of gravid females in the

sample population. This result suggests that summer months are the primary breeding

period for grass shrimp along the Gulf of Mexico coast, which is what would be expected

for a decapod in a temperate or subtropical climate.

Although the total number of shrimp increased in November 2002, the median

size was reduced as a consequence of an influx of newly recruited shrimp. The increase

in abundance is consistent with autumnal increases documented in other studies

(Knowlton et al. 1994). Spawning continued during this time, but at a reduced rate

compared to either spring or summer, as indicated by the lower proportion of gravid

females. Along with decreased spawning, an increase in the intermolt period and a

reduced proportion of females in the population suggested a decline in the quality of

seagrass habitats or the environment as a whole during November and a critical period in

the life history of P. intermedius. Shrimp abundance had declined markedly by February,

indicating increased mortality during late fall and winter. Alternatively, shrimp may have

emigrated to other areas. However, due to their small size, presumed lack of long

distance mobility, apparent absence form offshore habitats and lack of increased

abundance in any other habitats sampled in this study, emigration is an unlikely

explanation for the decline in numbers between November and February.









So what were the primary habitat characteristics or environmental factors driving

these changes in shrimp demography? The lack of significant changes in the percent areal

cover of seagrass as a whole, or of species composition of seagrasses, implies that

seasonal changes in vegetation were not driving the changes observed in shrimp

demographics. The most likely explanation for the changing demographics within the

shrimp population is water temperature, which is an environmental, rather than a habitat,

characteristic. Fall/winter mortalities of shrimp similar to those observed in this study

have been documented in other locations (Knowlton et al. 1994), and Lemaire

et al. (2002) suggested that reduced water temperatures could compromise the

osmoregulatory abilities of juvenile and subadult Pennaeus stylirostris. Vernberg and

Piyatitivorakul (1998) demonstrated that temperature had significant effects on many

grass shrimp metabolic processes. Changes in temperature and salinity have been shown

to affect changes in caridean shrimp abundances in other locations as well (Walsh and

Mitchell 1998). However, temperature and food production are also often correlated in

marine environments, and productivity of seagrass beds is known to vary seasonally (e.g.

Peterson and Fourquean 2001). Brockington and Clarke (2001), in an attempt to

determine the relative influence of temperature and food availability for the sea urchin

Sterechinus neumayeri, found that food availability may indeed be the key factor driving

metabolic rates, as opposed to temperature. Further study is necessary to determine which

factor is more important for grass shrimp, and P. intermedius in particular.

In the marsh edges, temporal variability was less pronounced than in the seagrass

beds. There were no significant changes in abundance or any ISF variable other than total

length. However, the quality of marsh edges seemed higher in the July/August sampling









period than other sampling periods. Grass shrimp were significantly larger within marsh

edge habitats in July/August, and this was the only period in which gravid females were

encountered. Since the percent areal cover of plant species did not change significantly,

and the DO, salinity, and water temperature values were similar to those encountered in

seagrass, water temperature is the most likely cause of temporal changes in shrimp

demographics in marsh edges as well.

SMMAP Cross-Habitat Comparisons (Seagrass vs. Marsh Edge)

In all sampling periods, greater abundances of shrimp and higher proportions of

gravid females were found in seagrass beds than along marsh edges. Although this

appears to contradict the findings of Rozas and Minello (1998), who documented higher

abundances of grass shrimp in salt marshes than seagrass beds in a Texas estuary, it may

simply reflect a difference in occupancy patterns between marsh edges and salt marshes

that were not captured by either study. In May 2003, shrimp from seagrass had shorter

IMPs, and gravid females were only caught in seagrass beds during this time period.

Thus, it is concluded (based on criterion 1 above) that seagrass beds provided higher

quality habitat than marsh edges in May. During the July/August sampling periods,

shrimp size and clutch size indicated higher habitat quality in marsh edges. These

variables were expected to covary, however, as a relationship between crustacean size

and clutch size has been previously documented, notably by Chockley and

St. Mary (2003), who documented this relationship for banded coral shrimp. Despite the

differences in total length and clutch size, the much greater abundances and the higher

proportion of gravid females in the seagrass beds were far more important in determining

the net number of eggs produced per area of habitat than the larger clutch sizes in marsh

edges, as indicated by the estimated number of eggs/sweep. The larger size of shrimp in









marsh edges may suggest that only large shrimp are capable of surviving in this habitat,

or that dominant individuals are monopolizing the habitat at low densities, implying that

there is some degree of habitat segregation by size in these shrimp, as has been shown for

the daggerblade grass shrimp (P. pugio) when subjected to mummichog (Fundulus

heteroclitus) predation (Davis et al. 2003), and also for tiger prawns (Pennaeus esculentus

and P. semisulcatus) in Australian seagrass beds (Loneragan et al. 1998). Based on the

greater potential reproductive output of shrimp in seagrass beds, it was concluded that

marsh edges provided an inferior habitat for P. intermedius during summer time periods.

In February 2003, the pattern was apparently somewhat different. Although

shrimp abundance remained higher in seagrass beds than marsh edges, the difference was

not as pronounced. However, the sex ratio of shrimp in the marsh was skewed towards

females, in comparison with the nearly even sex ratio of the sampled shrimp population

in seagrass beds. When the abundance values were multiplied by the sex ratios in each of

the habitats, it was clear that there were approximately the same number of females per

unit area in each of the habitats. The higher proportion of females in marsh edges may

imply that habitat quality there was higher, and the higher abundance in seagrass may

imply some degree of overcrowding of subdominant individuals into suboptimal habitat

during this time period. This possibility is further supported by the observation that

female grass shrimp tend to be larger than males, and so would presumably be dominant

in intraspecific agonistic competition. This possibility merits further investigation. Since

no other variables were significantly different between the habitats, and the sex ratio and

abundance do not both point to the same habitat as having higher quality, it can not be









conclusively stated that either of these habitats was higher quality than the other during

February 2003.

Among Estuaries Comparisons

Weeki Wachee and SMMAP showed similar patterns in August 2003. Seagrass

beds provided a higher quality habitat than the marsh edges for P. intermedius during

July and August, and the same was true in the Weeki Wachee estuary. Seagrass beds in

the Weeki Wachee estuary were determined to provide higher quality habitat than

adjacent marsh edges because shrimp were more abundant in seagrass beds than marsh

edges and there were no significant differences in measured ISF variables (criterion 3

above).

These results contrast with those from the Steinhatchee estuary, where shrimp

abundances were similar between the two habitat types and no differences in ISF

variables were observed. As a consequence it was not possible to designate either

seagrass beds or marsh edges as the superior habitat in that estuary. Interestingly, the

abundance of P. intermedius in Steinhatchee seagrass beds was less than the abundances

found in seagrass in SMMAP or Weeki Wachee. In fact, it was the marked reduction in

numbers of P. intermedius in seagrass that accounted for similarities in abundance

between the two primary habitat types rather than an increase in marsh edge occupancy.

Although shrimp abundance in the Steinhatchee marsh sites was greater than the other

two estuaries, the cross-estuary differences in this habitat type were relatively small.

The lower abundance of shrimp in seagrass beds in Steinhatchee relative to either

SMMAP or Weeki Wachee coupled with the fact that the proportion of gravid females

was also less suggests that seagrass habitat in Steinhatchee was of lower quality for P.

intermedius. Many reasons may explain these findings, but data reported here indicate









that the vegetative characteristics of the seagrass beds may be important. The percent

areal cover of H. wrightii was lowest in Steinhatchee and greatest in Weeki Wachee. The

percent areal cover and shoot density of T. testudinum, on the other hand, was greater in

the Steinhatchee estuary than in either SMMAP or Weeki Wachee. These patterns

suggest that seagrass beds dominated by Halodule may provide a higher quality habitat

for P. intermedius than those dominated by Thalassia. Although preference for a species

of seagrass has not been demonstrated for P. intermedius, preference for specific seagrass

communities has been documented for other shrimp species (Loneragan et al. 1998).

The relative quality of the seagrass beds in these different estuaries may change

temporally as they did in SMMAP. Broad generalizations based on these findings warrant

caution.

Shrimp abundance and ISF variables were more uniform within marsh edges

across estuaries. However, marsh edge habitats in the Steinhatchee estuary had greater

abundances of shrimp than marsh edges in the other two estuarine areas, which may

indicate an increased use of otherwise suboptimal marsh habitats, perhaps due to the

reduced quality of seagrass habitats in Steinhatchee.

Overall Conclusions

The lack of any shrimp near oyster bars in SMMAP contrasts with Eggleston et

al. (1998), who documented grass shrimp using these habitats. Throughout all estuaries

and sampling periods, marsh edges appeared to maintain consistent, fairly low-quality

habitat for P. intermedius. This contrasts with seagrass beds, where habitat quality was

generally much higher and more variable. The similarity in habitat quality between

seagrass beds and marsh edges in February 2003 was likely due to a decrease in the

quality of seagrass habitat. If seagrass beds in North Carolina show similarly variable









habitat quality relative to alternative habitats such as marsh edge or oyster bars, this may

provide an explanation for the findings of Eggleston et al. (1998), who found higher

abundances in seagrasses than along oyster bars in spring but similar abundances between

habitats in late fall. Temporal differences in the relative quality of seagrass and marsh

edge habitats to grass shrimp in summer and winter sampling periods at SMMAP likely

reflect broad-scale environmental shifts in temperature, that, in turn, may affect also food

availability for this organism.

The variation in quality among seagrass beds may also explain why the quality of

marsh in Steinhatchee, equaled or surpassed that of seagrass. Seagrass in Steinhatchee

was poorer quality habitat compared to seagrass in these other estuaries. Where seagrass

quality was determined to be poor for grass shrimp, such as at Steinhatchee in August

2003, or SMMAP in February 2003, marsh edges were determined of equal or greater

value to the shrimp. Poor quality seagrass habitat for P. intermedius during summer was

dominated by T. testudinum, and high quality habitat was dominated by H. wrightii.

Relative abundance was a good indicator of habitat quality both between habitats

and between estuaries during most time periods, which fails to support Van Home

(1983). Perhaps, Van Home's (1983) ideas do not apply to estuarine invertebrates as well

as they do to terrestrial quadrupeds. However, relative abundance and the ISF variables

did appear to be decoupled at certain times of the year, such as when small shrimp

appeared in November 2002.

Perhaps the import of these findings can best be understood from a management

perspective, if we consider P. intermedius to be a model organism for others (such as

penaeid shrimp or Macrobrachium spp.) that are commercially important and therefore









more likely to require intensive management. If a manager wanted to set up a marine

protected area for P. intermedius, abundance would likely be an acceptable measure for

selecting the location with the best habitat quality and to protect. However, if a manager

was attempting to use closed and open seasons as a management tool to generate

productivity, then measures of fecundity, mortality, and growth rates would be necessary

to determine the best time to restrict harvest. For example, during May, July, and August,

when growth rates and fecundity were greatest, exploitation should be restricted to

maximize productivity. Relative abundance (and likely biomass) was highest in

November, when the period of mass mortality was beginning, and this would be the best

time to utilize these shrimp as a resource. On this temporal scale, therefore, Van Home's

theory was supported because the greatest abundances did not always occur when ISF

variables pointed to the highest quality of the habitat or environment.

Future study of grass shrimp habitat ecology would be most informative if it

focused on five key issues: (1) better resolution of the temporal variability within

populations so as to determine if the seasonal patterns suggested by this study do, in fact

exist; (2) better determination of what factors (i.e., food availability, water temperature or

others) most influence the temporal variability in grass shrimp populations; (3) better

determination of what factors (i.e., dominant grass type or spatial distribution of habitats)

most influence variability in grass shrimp populations within seagrass; (4) better

determination of what characteristics make marsh edges a less high-quality habitat

relative to seagrass; and (5) why Van Home's predictions apparently fail to hold in the

context of this estuarine invertebrate.
















LITERATURE CITED


Abele, L.G., and W. Kim. 1986. An Illustrated Guide to the Marine Decapod Crustaceans
of Florida. State of Florida, Department of Environmental Protection, Tallahassee,
Florida. 760pp.

Anderson, S.H. and K.J. Gutzwiller. 1996. Habitat evaluation methods, p. 592-606. In
T.A. Bookhout (ed), Research and Management Techniques for Wildlife and Habitats.
The Wildlife Society. Bethesda.

Bass, C.S., S. Bhan, G. M. Smith, and J. S. Weis. 2001. Some factors affecting size
distributi n and density of grass shrimp (Palaemonetes pugio) populations in two New
Jersey estuaries. Hydrobiologia. 450: 231-241

Berg, A.B.V. and P.A. Sandifer. 1984. Mating behavior of the grass shrimp Palaemonetes
pugio Holthuis (Decapoda, Caridea). Journal of Crustacean Biology. 4(3)417-424.

Brockington, S. and A. Clarke. 2001. The relative influence of temperature and food on
the metabolism of a marine invertebrate. Journal of Experimental Marine Biology and
Ecology. 258(1):87-99

Chockley, B.R. and C.M. St. Mary. 2003. Effects of body size on growth, survivorship,
and reproduction in the banded coral shrimp, Stenopus hispidus. Journal of Crustacean
Biology. 23(4): 836-848

Cross, R. E. and A. E. Stiven. 1999. Size-dependent interactions in salt marsh fish
(Fundulus heteroclitus Linnaeus) and shrimp (Palaemonetes pugio Holthuis). Journal of
Experimental Marine Biology and Ecology. 242: 179-199

Davenport, D.E., Lancia, R.A., Walters, J.R. and P.D. Doerr. 2000. Red cockeaded
woodpeckers: A relationship between reproductive fitness and habitat in the North
Carolina sandhills. Wildlife Society Bulletin. 28 (2): 426-434

Davis, J.L., W.J. Metcalfe, and A.H. Hines. 2003. Implications of a fluctuating fish
predator guild on behavior, distribution, and abundance of a shared prey species: the
grass shrimp Palaemonetes pugio. Journal of Experimental Marine Biology and Ecology.
293(1) 23-40.

Eggleston, D. B., L. L. Etherington, and W. E. Elis. 1998. Organism response to habitat
patchiness: species and habitat-dependent recruitment of decapod crustaceans. Journal of
Experimental Marine Biology and Ecology. 223: 111-132










Franklin, A.B., Anderson, D.R., Guitierrez, R.J., and K.P. Burnham. 2000. Climate,
habitat quality, and fitness in Northern Spotted Owl populations in northwestern
California. Ecological Monographs. 70 (4): 539-590.

Frazer, T.K., Hoyer, M.V., Notestein, S.K., Canfield, D.E., Vose, F.E., Leavens, W.R.,
Blitch, S.B., and J. Conti. 1998. Nitrogen, phosphorus, and chlorophyll relations in
selected rivers and nearshore coastal waters along the Big Bend region of Florida. Final
Project Report. Suwannee River Water Management District (SRWMD Contract No.
96/97-156) and the Southwest Florida Water Management District (SWFWMD Contract
No. 96/97/157R). 326 pp.

Frazer, T.K. Notestein, S.K., Hale, J.A., Hoyer, M.V., Canfield, D.E., Blitch, S.B., and C.
Bedee. 2003. Water quality characteristics of the nearshore gulf coast waters adjacent to
Citru, Hernando, and Levy Counties Project COAST 2001. Annual Report. Southwest
Florida Water Management District, Brooksville, Florida. 168 pp.

Glancy, T. P., T.K. Frazer, C.E. Cichra, and W.J. Lindberg. 2003. Comparative patterns
of occupancy by decapod crustaceans in seagrass, oyster, and marsh edge habitats in a
northeastern Gulf of Mexico estuary. Estuaries. 26(5): 1291-1301

Grabe, S.A. 2003. Seasonal periodicity of decapod larvae and population dynamics of
selected taxa in New Hampshire (USA) coastal waters. Journal of Plankton Research.
25(4) 417-428.

Hartnoll, R.G. 2001. Growth in crustacea- twenty years on. Hydrobiologia. 449(1-3):
111-122

Hovel, K.A., Fonseca M.S., Myer, D.L., Kenwrthy, W.J., and P.E. Whitfield. 2002.
Effects of seagrass landscape structure, structural complexity and hydrodynamic regime
on macrofaunal densities in North Carolina seagrass beds. Marine Ecology Progress
Series. 243: 11-24.

Khan, R. N., R. E. Knowlton, and H. C. Merchant. 1995. Distribution of two sympatric
species of grass shrimp Palaemonetes pugio and Palaemonetes vulgaris, in relation to
homogeneous and heterogeneous aquarium substrates. The Journal of Elisha Mitchell
Scientific Society. 111(2): 83-95

Kneib, R.T. 1987. Seasonal abundance, distribution, and growth of postlarval and
juvenile grass shrimp (Palaemonetes pugio) in a Georgia, USA, salt marsh. Marine
Biology. 96(2): 215-223

Knowlton, R. E., R. N. Khan, P. M. Arguin, T. A Aldaghlas, and R. Sivapathasundram.
1994. Factors determining distribution and abundance of delmarva grass shrimp.
Virginia Journal of Science. 45(4): 231-247









Knowlton, R.E. and R.H. Schoen. 1984. Salinity tolerance and sodium balance in the
prawn Palaemonetes vulgaris compared with Palaemonetes pugio. Comparative
Biochemistry and Physiology. 79(4): 519-524.

Lemaire, P. E. Bernard, J.A. Martinez-Paz, and L. Chim. 2002. Combined effect of
temperature and salinity on osmoregulation of juvenile and subadult Penaeus stylirostris.
Aquaculture. 209(1-4):307-317.

Lewis, M.A. and S.S. Foss. 2000. A caridean grass shrimp (Palaemonetes pugio
Holthuis) as an indicator of sediment quality in Florida coastal areas affected by point
and nonpoint source contamination. Environmental Toxicology. 15(3)234-242.

Lin, Y.T.K. and G.O. Batzli. 2001. The influence of habitat quality on dispersal,
demography, and population dynamics of voles. Ecological Monographs. 71(2):245-275

Loneragan, N.R., R.A. Kenyon, D.J. Staples, I.R. Poiner, and C.A.Conacher. 1998. The
influence of seagrass type on the distribution and abundance of postlarval and juvenile
tiger prawns (Penaeus esculentus and P. semisulcatus) in the western Gulf of Carpentaria,
Austrailia. Journal of Experimental Marine Biology and Ecology. 228(2): 175-195.

Luck, G.W. 2002. Determining habitat quality for the cooperatively breeding Rufous
Treecreeper, Climacteris rufa. Austral Ecology. 27 (2): 229-237

Peterson, B.J. and J.W. Fourqurean. 2001. Large scale patterns in seagrass (Thalassia
testudinum) demographics in south Florida. Limnology and Oceanography. 46(5):1077-
1090.

Posey, M.H. and A. H. Hines. 1991. Complex predator-prey interactions within an
estuarine benthic community. Ecology. 72: 2155-2169

Posey, M.H., Alphin, T.D., Powell, C.M., and E. Townshend. 1999. Use of oyster reefs
as a habitat for epibenthic fish and decapods. Pp 229-238. In Luckenbach, M.W., Mann,
R. and J.A. Wesson, (eds.). Oyster Reef Habitat Restoration: A Synopsis and Synthesis.
Virginia Institute of Marine Science Press.

Quetin, L.B. and R.M Ross. 1991. Behavioral and Physiological characteristics of the
Antarctic krill, Euphausia superba. American Zoology. 31:49-63

Reinsel, K.A., Glas, P.S., Rayburn, J.R., Pritchard, M.K., and W.S. Fisher. 2001. Effects
of food availability on survival, growth, and reproduction of the grass shrimp
Palaemonetes pugio: a laboratory study. Marine Ecology Progress Series. 220:231-239

Ross, S.W. 2003. The relative value of different nursery areas in North Carolina for
transient juvenile marine fishes. Fishery Bulletin Seattle. 101 (2): 384-404.









Rozas, L. P. and T. J. Minello. 1998. Nekton use of salt marsh, seagrass, and
nonvegetated habitats in a south Texas (USA) Estuary. Bulletin of Marine Science.
63(3):481-501

Schantz, T. von. 1981. Female cooperation, male competition, and dispersal in the red
fox (Vulpes vulpes). Oikos. 37:63-68

Sheridan, P. F. 1992. Comparative habitat utilization by estuarine macrofauna within the
mangrove ecosystem of Rookery Bay, Florida. Bulletin of Marine Science. 50(1)21-39

Stickle, W.B., Kapper, M.A., Liu, L.L., Gnaiger, E., and S.Y. Wang. 1989. Metabolic
adaptations of several species of crustaceans and mollusks to hypoxia tolerance and
microcalorimetric studies. Biological BulletinWoods Hole. 177 (2): 303-312.

Townshend, E.C. 1991. Depth distribution of the grass shrimp Palaemonetes pugio in two
contrasting tidal creeks in North Carolina and Maryland. M.S. Thesis, University of
North Carolina at Wilmington.

Van Home, B. 1983. Density as a misleading indicator of habitat quality. Journal of
Wildlife Management. 47:893-901

Vernberg, F. J., and S. Piyatiratitivorakul. 1998. Effects of salinity and temperature on
the bioenergetics of adult stages of the grass shrimp (Palaemonetes pugio Holthuis) from
the North Inlet Estuary, South Carolina. Estuaries. 21(1)176-193

Walsh, C.J. and B.D. Mitchell. 1998. Factors associated with variations in abundance of
epifaunal caridean shrimps between and within estuarine seagrass meadows. Marine and
Freshwater Research. 49(8):769-777

Walters, J.R., S.J. Daniels, J.H. Carter, and P.D Doerr. 2002. Defining quality of red-
cockaded woodpecker foraging habitat based on habitat use and fitness. Journal of
Wildlife Management. 66 (4): 1064-1082

Young, D.K, Buzas, M.A. and M.W. Young. 1976. Species density of macrobenthos
associated with seagrasss: a field study of predation. Journal of Marine Resources. 34:
577-592

Young, D.K. and M.W. Young. 1978. Regulation of species densities of seagrass-
associated macrobenthos: evidence from field experiments in the Indian River estuary.
Florida Journal of Marine Resources. 36: 569-593















BIOGRAPHICAL SKETCH

Daniel Scott Goodfriend was born on March 15, 1976, in Suffern, New York. He

lived in the suburbs of New York City until his graduation from Rockland Country Day

School in May of 1994, where he spent a lot of time enjoying the outdoors in and around

New York's Harriman State Park. After graduating high school, he spent two semesters

as a philosophy of religion major at Grinnell College in Iowa, but found this subject

matter too abstract. He left school in an attempt to find a path that was more personally

fulfilling and spent the next three years traveling around the United States, living briefly

in New Orleans, Louisiana and Lake Worth, Florida. In January 1996 he moved to

Encinitas, California, and began attending classes at MiraCosta College in the nearby

town of Oceanside. Drawing on his experiences in the woods of New York and his

travels to various National Parks, he decided on a major in wildlife. To pursue this goal,

he transferred to Humboldt State University in Arcata, California, in August 1997. After

three and a half very enjoyable years, he graduated with a Bachelor of Science in wildlife

in December of 2001. He then spent a year in the Americorps program monitoring a

population of threatened Blanding's turtles (Emydoidea blandingii) for the nonprofit

group Hudsonia, based in Dutchess County, NY.

After his year at Hudsonia, Daniel decided that his goals required him to continue

his education, and in January 2002 he enrolled at the University of Florida in Gainesville.

There he spent two and a half rewarding years studying coastal ecology with Dr. Thomas

Frazer, and will receive a Master of Science degree in May 2000.