<%BANNER%>

Community-Based Timber Management in Acre, Brazil and Its Implications for Sustainable Forest Management

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 E20110330_AAAALD INGEST_TIME 2011-03-30T13:12:27Z PACKAGE UFE0012840_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 29912 DFID F20110330_AAAYGP ORIGIN DEPOSITOR PATH rockwell_c_Page_090.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
eb1b6fdaea4529acfb8819abd876fcdf
SHA-1
efdcc134c6f1017fd5926c38b0ec2f77dc12d02a
4422 F20110330_AAAYHD rockwell_c_Page_008thm.jpg
ddc76eb53f0690e34667c53b3d8ffca3
6059520965844708c82fb2f460c84ed23ad0d5c0
33274 F20110330_AAAYGQ rockwell_c_Page_091.QC.jpg
a4f7e7aa9ebccbacfe887b065e5dfd58
844e9e9e71f9ba91c2f2f5ff297ef75c31c88d96
4522 F20110330_AAAYHE rockwell_c_Page_010thm.jpg
e6f8a45c49b954af349e3b20e61479cf
d01f41e9c2c9e6d9df156e0366bfefe60852f5cd
9020 F20110330_AAAYGR rockwell_c_Page_092.QC.jpg
5d6922a59eb58c0266fa64a461b270c6
a9192c966fa4e04b073a69d134bb85fa7d6cf7b0
6729 F20110330_AAAYHF rockwell_c_Page_011thm.jpg
62a0c4ded36fb7380379b988214067fb
e5877ab05b33ea3a452d3f38bd27c28f5d9a954c
30541 F20110330_AAAYGS rockwell_c_Page_093.QC.jpg
efca00e6cbe081809466e95bd2077278
f311662d2bad1b1d1169b96af64e9ebc25b28daa
6953 F20110330_AAAYHG rockwell_c_Page_012thm.jpg
152ecf90e2c2faebee2ab3baf4625b7e
13f8ffa43473c78de6674354b244ac45bd43533c
7948 F20110330_AAAYHH rockwell_c_Page_017thm.jpg
2d9cc93ece05d237f377b9baf62d6fa0
4b7a0fa825cfc7b923c6d35d3edd1c3134d03690
35965 F20110330_AAAYGT rockwell_c_Page_095.QC.jpg
111b5a9a9b791668a45fff081c8dd88d
a1907afc5b3e59f20b5bcee3c190b2809db52774
2380 F20110330_AAAYHI rockwell_c_Page_019thm.jpg
09ef7a4be6cff62e0995a786f5799cc5
a3b6a5ad70e9d8392e16f2dec6660601c16f286b
34135 F20110330_AAAYGU rockwell_c_Page_096.QC.jpg
e20483ecd9a924f247a5ab0d643a870e
564c379b30943b247cc3f07c8747705e7df4bed6
8353 F20110330_AAAYHJ rockwell_c_Page_021thm.jpg
f145779b94e0cbab06ad620bc945784a
219e27c9fbec9aa652dd3006d77b413344cfcbcb
36778 F20110330_AAAYGV rockwell_c_Page_097.QC.jpg
3facd7c2dba19cb8833bf837160b627a
78142131cebdaa22f34933dd86b397b8e685e3e7
8611 F20110330_AAAYHK rockwell_c_Page_023thm.jpg
eb72365a1768b7cf0f716b2ab104503d
230d2a768dab4855634d587a8f2cf022a131a5c6
33387 F20110330_AAAYGW rockwell_c_Page_098.QC.jpg
a65cb12588b8ef98ea8bcc6cd7fcca02
3c3df9aee49edcbc00c0ac0f232fb25c64229a41
8180 F20110330_AAAYHL rockwell_c_Page_024thm.jpg
52e5f63502f22ebe5423d64f89a62a3c
d37a5b90c3a4c3e9dd70d0c11b13116b84b596b0
35126 F20110330_AAAYGX rockwell_c_Page_100.QC.jpg
e3db0351ca8b58c208857318f11d6610
e318d6554b88cef8fb57b7fb34334e8903143696
8668 F20110330_AAAYIA rockwell_c_Page_044thm.jpg
f98036db9495b52d14131388d562d920
9adefb50e39a4116e29bd778ab37bdb9b2a622a6
8278 F20110330_AAAYHM rockwell_c_Page_025thm.jpg
aa2ea9e379edf8a711c3d88b0aede329
16123de32f94a4489f19f70ad482007c67da9114
22992 F20110330_AAAYGY rockwell_c_Page_106.QC.jpg
3e94f4aba003e02bc2d082acd38f2946
be0ffb5a1bcb1a96c2dcc9a3acb900e2b1ab2083
8222 F20110330_AAAYIB rockwell_c_Page_046thm.jpg
ff746dcf76173618df329a9a348afd21
77c8e98c52b77cf1f20a19dac0bf5a2512a05c30
8469 F20110330_AAAYHN rockwell_c_Page_027thm.jpg
ebe9f65c1a34f48e063889aa2aeb1de6
224623310e144575f3140dc02b0af7817ee06275
622 F20110330_AAAYGZ rockwell_c_Page_002thm.jpg
a027d10ef23ff19cdb25e88f7aba7607
7860a8d182db83e27c33c8b15730f3d671b4cd48
8132 F20110330_AAAYIC rockwell_c_Page_048thm.jpg
7b19a66b55e76e52db3fd42171dbf9f1
1f06e04c4493c3bf095f181cd6054074b46a8562
8591 F20110330_AAAYHO rockwell_c_Page_028thm.jpg
7153e90d20635d79dfbbe8211115585b
7b6620db4710c8a21103bcba820adbdf3770c17c
8494 F20110330_AAAYID rockwell_c_Page_049thm.jpg
9b4c42c1c1d5b9d55ed02eb1383df36d
eaaa6711ed6940c51d84db28e8aa9885018d784b
8377 F20110330_AAAYHP rockwell_c_Page_029thm.jpg
a1841304a9b7e431c3e8b5d426105b1b
0063322117fe53d0592ef73b6cb4909079fa2018
8473 F20110330_AAAYIE rockwell_c_Page_050thm.jpg
e1f6dfc457b34c4a5e61d634792140ba
07297d91a595eccc8edb2cf0e8d686f8a050b7d0
8454 F20110330_AAAYHQ rockwell_c_Page_030thm.jpg
dc8674dd8e332155526736b148b2a505
260aaf10cab261cf2a8d0c3a1f0a84464a0c0a6b
8114 F20110330_AAAYIF rockwell_c_Page_051thm.jpg
ab177bad9078ad131c86d4e574b229b3
a1abb1c7e1938343b202a9cd0112b407560e2114
8185 F20110330_AAAYHR rockwell_c_Page_032thm.jpg
56c20e3ade391f186380dbeb530e942f
f5e8c55de28cb55f5a79ba80279ab7b0d22130bd
8307 F20110330_AAAYIG rockwell_c_Page_052thm.jpg
2b9bacba9e57bdb59131b12a500c80af
3697e5f248b83a241a878513cf417e0213e4ce42
8390 F20110330_AAAYHS rockwell_c_Page_033thm.jpg
6cec7ecb2019f7eb9f5fb397116a5b1d
29f8bd15e758f20c4e96eed8ae9ae5c4e10ea7d1
3071 F20110330_AAAYIH rockwell_c_Page_053thm.jpg
3362b1a1ab590dab1d85076d738e70d9
f9fbb539f98b0361e283ef01f7a655c23a5fc9c1
8378 F20110330_AAAYHT rockwell_c_Page_035thm.jpg
af421dcb30ef07adb21f4e167596a149
b3ef4ddabd9a5452eb39fc17262f819414035e4e
3846 F20110330_AAAYII rockwell_c_Page_054thm.jpg
78db567d0aa5b338d65b698365059bc4
097c0c864a791ca4e519693843e6f2c2a03ab126
4158 F20110330_AAAYIJ rockwell_c_Page_055thm.jpg
9432dba1137b0f1c0a22a25078fafcaf
1f72c918f790d34ff40ab78e9376ceb5ca9b277d
8169 F20110330_AAAYHU rockwell_c_Page_036thm.jpg
0b2625510459fcdac9f757879704a7d7
ef2c76c857b7425163a909cdd558fb1a25416fb7
7089 F20110330_AAAYIK rockwell_c_Page_056thm.jpg
85f070bb158ab43b39ae17f612f83b74
ab2ed765a6fa9378cbf7a925e50537cd186723cf
8530 F20110330_AAAYHV rockwell_c_Page_037thm.jpg
6f35bec77b275119f18cd9ad8309e9a9
8d3556e7ccfdb3e7d0ef7f1b1e7a24c7544a5dbe
8133 F20110330_AAAYIL rockwell_c_Page_057thm.jpg
76f5a76a4b195553e2187c2137a8b791
546e5944b6af1ff864eb0bb8906a7f30641d2eb5
8490 F20110330_AAAYHW rockwell_c_Page_038thm.jpg
abdeb55f679e42cbda88d3d33cf1173c
dbe4dee67620b89837ce53e99a4905bf7ff9e460
8330 F20110330_AAAYJA rockwell_c_Page_075thm.jpg
4de510d6e7b76ab3f400925434c52ed2
d247bad33321183664464a154f27df89338a420f
8386 F20110330_AAAYIM rockwell_c_Page_058thm.jpg
9f2d41c52b0add692e5f7336cd7d861f
05a3de9dadd21e224fcec09e4a1c95edaf905116
8214 F20110330_AAAYHX rockwell_c_Page_039thm.jpg
c7f62cc1bacebd294a61be8d55f0d731
1b66450c4581633bd39f8caa6b12be1fee51173c
9062 F20110330_AAAYJB rockwell_c_Page_077thm.jpg
1854f9d88a5807cb4ad2492523e5344a
bd65a95724ef18d0afd03307faed7ca98a17a7b9
8174 F20110330_AAAYIN rockwell_c_Page_059thm.jpg
fc364eb5b91bd9acebd1f385169ef78d
4e9c53103be3162b555dc1eeef6cc2392f3cc686
7974 F20110330_AAAYHY rockwell_c_Page_040thm.jpg
bc84cd6c73716367aeaae44a653022c3
39f5c7a47d1bd4436827e1165041980e400b6443
1496 F20110330_AAAYJC rockwell_c_Page_078thm.jpg
62bb9cb524eb7cd44f3dd914a2799d7d
d9104465991ed86dc07b332ef31779ae19186b45
8423 F20110330_AAAYIO rockwell_c_Page_060thm.jpg
9dca0d979f72699a4484b570ad9afe89
db5d36a5c4965bf1cc39f9d8e6261cbdf9cd8eda
8356 F20110330_AAAYHZ rockwell_c_Page_041thm.jpg
4104cb05c144815308d5fc4c44e03c2b
9ea1395980bf09afd858db82c94069f38eb81ddd
2344 F20110330_AAAYJD rockwell_c_Page_081thm.jpg
78e64741594b9a661cf7a23439b90be5
82f82e4b01f0ea6fa50dd144a3e306f2dbeda786
8622 F20110330_AAAYIP rockwell_c_Page_062thm.jpg
28411f7764d804fc2531e6584bbf53d5
3e2b3c431ba457380b577d2f7b521eab6c005ff8
5172 F20110330_AAAYJE rockwell_c_Page_084thm.jpg
fd9a3a583c13ede5fcd9127d6be6e091
807a794a5284385d3ee693c26a3b14c151ad5857
7979 F20110330_AAAYIQ rockwell_c_Page_063thm.jpg
379ab1891e3c058d8d2e7b17f1b33d43
178bcd43dea4bc660aaac91ff0e59b2cdbf58bbd
3578 F20110330_AAAYJF rockwell_c_Page_085thm.jpg
548b23e687d5c096dc1bfa466251f592
8041331cb78a4119c96fe42ef14268b012d2d50a
7962 F20110330_AAAYIR rockwell_c_Page_064thm.jpg
3819e27a57b4960ccf8de1a017bf3241
9377b465b773108a4018839597fb840e517a8443
3400 F20110330_AAAYJG rockwell_c_Page_086thm.jpg
ed19aa72ddd2448de289935d8b7d0fd1
4fb7de152e7527453f3917cfa3459ce80d4f6584
7287 F20110330_AAAYIS rockwell_c_Page_065thm.jpg
22c8e8e713974de4734848ca40324dac
4778383af70873dc65927c296655e16504314d4f
3209 F20110330_AAAYJH rockwell_c_Page_087thm.jpg
5c21b6b1407625ff426c4fb0844f7393
838322bd4bc12a4cae4211b3ae2a49b2433036f1
7424 F20110330_AAAYIT rockwell_c_Page_066thm.jpg
35499780d30f720ade6ffc95f10c79ec
8eb78c41d3dc5fa41ef187f1e240750b22c2c98a
3257 F20110330_AAAYJI rockwell_c_Page_089thm.jpg
bf1f6ead59b67e3fd223e73e7353bab4
caef5842362d9a37b6f3fd5367d6d19eb8dfe915
7727 F20110330_AAAYIU rockwell_c_Page_067thm.jpg
7cf0a49427434bf4e8566f91f90ba26e
876a1529bb1cbe4f00756bef285802f26c3c32e7
8419 F20110330_AAAYJJ rockwell_c_Page_091thm.jpg
99d3f009d01f235f99f934b945aa7589
415181efd70731c4f3ad329af829cfbd5351b350
2260 F20110330_AAAYJK rockwell_c_Page_092thm.jpg
176b6770a7d3ce72ed33135a8f8c6db7
cc628dc4494e98ad1424e2687b2d101be24f711f
8104 F20110330_AAAYIV rockwell_c_Page_068thm.jpg
b2f06d3dcae65cfaafb665ffd9da9a0f
a56600eb4fb1177b7c9d07c54f78ef98a115fe9c
7596 F20110330_AAAYJL rockwell_c_Page_093thm.jpg
875e2ea768ec87225dd2e9caea64d9b6
d58460cf3175ed2646de36f35d910876c4c576fe
8067 F20110330_AAAYIW rockwell_c_Page_069thm.jpg
cb4c43f574d30e3c7eb3a6ca6c7e529b
d27fbc527e31bbc3fba185d51b06a3eba59c673d
8513 F20110330_AAAYJM rockwell_c_Page_094thm.jpg
667ba3e291df6ee9500697b53d49f29d
ee8c5ec2cc0dd9614518890b2653fbbdc44a42b2
8531 F20110330_AAAYIX rockwell_c_Page_071thm.jpg
b099a8d77c6d098540c59fb191c0cd3a
21a59e711b2f3aac9988a7071326d6a7255f9bae
9112 F20110330_AAAYJN rockwell_c_Page_095thm.jpg
e5d17ef5c61da01de0b6ba1076adbcb6
4113d203857a49b0c592397bb9544472b233b008
7989 F20110330_AAAYIY rockwell_c_Page_073thm.jpg
bd2b023bd25a6ca0090b715ef8af3573
ef561b725cd5f30f3e2127349e1a3135a93a9a63
8447 F20110330_AAAYJO rockwell_c_Page_096thm.jpg
db0578807893ca0f810f9bed7282bbf3
d9839500cb0aebaa643105a39058a98945db0552
8674 F20110330_AAAYIZ rockwell_c_Page_074thm.jpg
955795dfe12921988778bbf17ce48192
427378a999b2d45998eacd2121dd2ab0f386a545
8934 F20110330_AAAYJP rockwell_c_Page_097thm.jpg
d213dac9fe6dcd128a16cc247cd526ab
7737fcaaf12507dfca8ef93e3cccd0a3092dd7c7
8272 F20110330_AAAYJQ rockwell_c_Page_098thm.jpg
56f1ab95231b17d943b1cef205fc6356
2914a0998dbfe0c7846c22d4e9c37425b4d1544b
F20110330_AAAYJR rockwell_c_Page_099thm.jpg
292b359eff1711904379e97343bf138b
1243526e7f592d42bbf156b60d11c54cedd6fff8
8623 F20110330_AAAYJS rockwell_c_Page_100thm.jpg
daafd13b6af21cce95d62b5d22a5666b
d246c82bc945da2273c9fee084e68117d6d5ae19
8960 F20110330_AAAYJT rockwell_c_Page_102thm.jpg
5982dd58df0dffede6e0ef67e982f883
77a91cbe54d67015e6cbf246b89cccc9eb14f08b
32871 F20110330_AAAXHG rockwell_c_Page_042.QC.jpg
c3d5b2e05659d2015704c8ea80a9cca0
fa0dc65a0310d37e45589df4dfdfe3ca3fcc6f7c
8717 F20110330_AAAYJU rockwell_c_Page_104thm.jpg
a6d034881e76f785e3145e2eda8f65b5
4d7c0aa280a36ee7736f063878fec5757b06cc75
5154 F20110330_AAAXHH rockwell_c_Page_003.pro
e9108e90ce77c1666ff125a6d71f4fb5
e2ec81df5abbdbe66c422cd155fe7249367f150b
6761 F20110330_AAAYJV rockwell_c_Page_105thm.jpg
4aebd2ee00df97573b2593e240eee426
8585097f978a105d9ac1b3552abd8f338fad077c
8020 F20110330_AAAXHI rockwell_c_Page_022thm.jpg
41d0ef7c93605da607e2bf8f2aa8e2e9
62f8e2cf066975cbb63710d1b31298f05adc7bbe
1053954 F20110330_AAAXHJ rockwell_c_Page_104.tif
239c175d63771d2316034074ff59271b
613f4fbea60c155fdf075b521c943e7b1616389b
5742 F20110330_AAAYJW rockwell_c_Page_106thm.jpg
27eb140f964d03d359828d4ed1bfd07a
24461db9ef3be9e6f9fe9aba4f743f142d6a7ae4
1051953 F20110330_AAAXHK rockwell_c_Page_099.jp2
e01e2134f03a81c058c8ec90f1473808
9b93c3530d4728a6bda2dd775049af0a18a4abe3
808346 F20110330_AAAYJX rockwell_c.pdf
b513fba125837d9d6c01b7c3190ac592
14b2c88f9144047b7befcf2dc459f3e4a9142e58
11506 F20110330_AAAXHL rockwell_c_Page_087.QC.jpg
1c0d053f624d4e19677456073ff92ce7
e8046da9bc1415d29686de1d2021d4c0dd5823cc
25271604 F20110330_AAAXIA rockwell_c_Page_010.tif
c56dc215b0b5174dbcdbf4fc9bf4d350
4373077743dc4ca00d4b65084f0238ca0bb22663
43786 F20110330_AAAXHM rockwell_c_Page_004.pro
0af415e6936050ba1082c9e1efc01035
4157127c17dc3e085c78d0eb1d18e44db3c645fa
F20110330_AAAXIB rockwell_c_Page_011.tif
556ab96f1be0b5c8371c9133292a4a35
466da77ef89bf9f0db00a0efb94c5ba8e6221b8f
42517 F20110330_AAAXHN rockwell_c_Page_086.jpg
c409868ca79d5a149337db6552b9ae9e
fef657781446d60682c1e98135942d19f78aa94b
F20110330_AAAXIC rockwell_c_Page_012.tif
82b52ebce25b874cd9166b6d3eadcdb2
2b2ea29ec668c81417b4865d991de3eacdc2df76
11390 F20110330_AAAXHO rockwell_c_Page_085.QC.jpg
ad36417b21fe97728c68809267198b58
a48074af140b198d579574956507e58308e4f1b1
F20110330_AAAXID rockwell_c_Page_013.tif
2fe421c229d3c551f545245cd9ae9014
c89d67156e4b558429415c0ad9649b3fb6648223
126213 F20110330_AAAXHP UFE0012840_00001.mets FULL
c2b3d88cd125e2419d78b8290edfa9d5
6d00a34b25c32d6db5944ff53fc3973838e9bbde
BROKEN_LINK
rockwell_c_Page_001.tif
F20110330_AAAXIE rockwell_c_Page_014.tif
78c7e363f5f1fc2d4f4de53272e00c25
758eb949cdba0f303a6ef0670f92488def0e2e0f
F20110330_AAAXIF rockwell_c_Page_015.tif
e1c86a66766b51624af6f7b266c14a55
fc8897fa26fc6c89c44f749e7924209deff95a56
F20110330_AAAXIG rockwell_c_Page_016.tif
a352de71ea355c740332c880feccce1e
6013102d49752baab9d9497e9775519c0fd974ad
F20110330_AAAXHS rockwell_c_Page_002.tif
bbd164132a789795a11666387c4adbb9
bb1403d37d283f8c3b9999e2dfc6be89a359659c
F20110330_AAAXIH rockwell_c_Page_017.tif
aca1ddbd77872674555d89836e0fd731
03e632fb8e4f84d485e361871b6388f91eb9ead7
F20110330_AAAXHT rockwell_c_Page_003.tif
c537cc813c5efa7d6f2224322b5cb761
211a61b40dc304d6f82bf42908ac129b7e96804d
F20110330_AAAXII rockwell_c_Page_018.tif
6b8b3f9a42d16f46b5d281253930470e
5550559f502a2af533ff23f9b39af15ea6bfad4b
F20110330_AAAXHU rockwell_c_Page_004.tif
13f8e70fdf631af99b957a0dc07474b7
71dd291aa79bd7f308566cedc5421b1005742224
F20110330_AAAXIJ rockwell_c_Page_019.tif
5c6b4bbfd3ea0fd2367cacced3f42ca9
728e5938ad97dc67a3a94355b86c0266fbf34f92
F20110330_AAAXIK rockwell_c_Page_020.tif
2d741094be4435521b25a75331a1c3ad
1ae9b5d8fbb6bc4be2a2efb99457cb22e7f0d0d8
F20110330_AAAXHV rockwell_c_Page_005.tif
f6161494ae729a794473913aa84b39ed
89a6bea403e1c7af1cec3eacc0fd83bcdc8743c4
F20110330_AAAXIL rockwell_c_Page_021.tif
99dd0338c3182dc7b9eba7ba6e250988
b767dcc153209cddd231e953e122fab61faf12e2
F20110330_AAAXHW rockwell_c_Page_006.tif
aad49ffb37d9c821b35b6f03df69fbe0
3daeb018fe3cb082e939bf8e1128c96a03b304ea
F20110330_AAAXIM rockwell_c_Page_022.tif
7c9276a33b0f0958928c99b2be9cfde2
6447563ed84392abfecdc19aca116f7ed4ae9077
F20110330_AAAXHX rockwell_c_Page_007.tif
7b8e16093ca0ba6da75d688b3412628d
13e5e3a6de7f934dd0d154f2d81c5ed503f47fa2
F20110330_AAAXJA rockwell_c_Page_036.tif
725d81a88be339d62811b359306a348c
79e0b2048f3e761267da60646eb1eea92de0b91d
F20110330_AAAXIN rockwell_c_Page_023.tif
7cb926f130f9dab1682d21a14c0009af
1adf5b183b32418b44f00e621cfffd2e3e9f0009
F20110330_AAAXHY rockwell_c_Page_008.tif
2ec6aa3694351120090cfca3f7ad144e
c3596505464ef5f7feab0a58e15720c0248898ec
F20110330_AAAXJB rockwell_c_Page_037.tif
a90444997f1291bd9732bbb7b7be9cca
3afcd57f8c355558fe875cdac21de24bb0199142
F20110330_AAAXIO rockwell_c_Page_024.tif
7407a64eeca4d686f3deb3b6ee743980
04594a0d080d2e67d086eccc0382d50f8966dd38
F20110330_AAAXHZ rockwell_c_Page_009.tif
cca40ad80d88d30e21cb94e41d96101e
8b9a3a8828aead66d7f367786c434db9efb31be2
F20110330_AAAXJC rockwell_c_Page_038.tif
2e143b8e5cf55d5e9b7b8c789835bda4
7e1b88a4026882439bc62b540aefbcda33fb5094
F20110330_AAAXIP rockwell_c_Page_025.tif
c5b9ded86ae7953cb6dbf7f7d4641691
57489fe7883e199617977b7f161855567216a474
F20110330_AAAXJD rockwell_c_Page_039.tif
115ba1d913c0a7cac9bce7869000c4db
8d77c740e5cd68f5f967bd0587624f930f99c678
F20110330_AAAXIQ rockwell_c_Page_026.tif
3351f3942a3b67146c4ef713a98d0c1d
0a0b7168abc2b9304f56f75ebdd304aa21b201ef
F20110330_AAAXJE rockwell_c_Page_040.tif
34fd441d0ae23e190998e9b32d212d34
4ad95c5de838a9121f5f50e368484de70577e6d6
F20110330_AAAXIR rockwell_c_Page_027.tif
3b1d44c7b65c8a655af1f40a93b6db23
2cd590dd46284da5c748bf790be17371efe50d67
F20110330_AAAXJF rockwell_c_Page_041.tif
47559d709c3091c7db117329725722f8
0bb376c9025539848099339144e08743cb9156b8
F20110330_AAAXIS rockwell_c_Page_028.tif
737b234f7ad34ab8c73994a910ebf97f
975a155f79e34a6e02c6237fb8366b7c9d352d02
F20110330_AAAXJG rockwell_c_Page_042.tif
5354e84206e416a6925e19a0b057cab4
14db394955afd33c298a07ecaeb303cd7edddf9b
F20110330_AAAXIT rockwell_c_Page_029.tif
d6a52fcc8552c24f27f751aa8c81bed4
cab1e6c5cc8963e28d5a3a4b88ed411a589b26e0
F20110330_AAAXJH rockwell_c_Page_043.tif
4e14c0a4de221c1b3d43281e3b06dff1
a8be72c3faf75d9d419f241972645f1e87a38116
F20110330_AAAXIU rockwell_c_Page_030.tif
829a81e9bdc1ed0527b3549b8cae2660
15eb7459c519e960eaeba6a972d820f3be0aef4b
F20110330_AAAXJI rockwell_c_Page_044.tif
37d92888612d8724b26fb757bca06abb
e38747a8b0595bae1c5e40fd7f9b88a2fc98067f
F20110330_AAAXIV rockwell_c_Page_031.tif
db61e2f2e522247bd4e1a884099a8ab1
444325931e126ad21e779d53d01ed8ede5661cc6
F20110330_AAAXJJ rockwell_c_Page_045.tif
05915afbab957713e5424cd6b28e7a3c
31ce02fa2346bc317c763ff1ec1c9105e23766c7
F20110330_AAAXJK rockwell_c_Page_046.tif
d18be4bc3a6a9a45efd5ec59acd50c3d
d0f7293321e43dc1713fc5222df741604a6f03a6
F20110330_AAAXIW rockwell_c_Page_032.tif
adc36e476e29226403726c02f7d06935
f861aa88088115c4ce4acd221d4ff1daba5d6e09
F20110330_AAAXJL rockwell_c_Page_047.tif
06a77d152df031acee878eca185aec0b
e571652be916f690e443dce4c7e6a422b442ef68
F20110330_AAAXIX rockwell_c_Page_033.tif
7d88fe4de244067ba5094566aa891082
7149f89e8dfdca9069a037a6c61712779cdafa2f
F20110330_AAAXKA rockwell_c_Page_062.tif
581750d3aef1926f80689cd452de309b
99ff11eb7df7bb230058c184790be18341f49505
F20110330_AAAXJM rockwell_c_Page_048.tif
53a34fd74dd527633aa4fad06b1dbd07
93f6d997a12def01516f5a781093391df54ad7b1
F20110330_AAAXIY rockwell_c_Page_034.tif
7c9e6d02948b8947180bdd42e1171523
ada408b6d2a264909a631496f93ba43c6697201a
F20110330_AAAXKB rockwell_c_Page_063.tif
267310e8d05fb213fbfd80f17bcc11c1
3777cda298c1efce965902102ba3acbeb2cf1b07
F20110330_AAAXJN rockwell_c_Page_049.tif
ff38f3bd433530f68ee6beb2a8bc0c7b
d6ee01b0835d0cd097ce5de11d80627ba320168c
F20110330_AAAXIZ rockwell_c_Page_035.tif
a7c60c649121b5c24a5f2345f5e6cd40
52bc621454d8a9097f7d8c92be21a56805425b06
F20110330_AAAXKC rockwell_c_Page_064.tif
327eadeb5364dc4dc50f9f4702701bfb
9d7b304a3c6ccdb3500b38b0de34787dad136ef4
F20110330_AAAXJO rockwell_c_Page_050.tif
4ea5aef79c8ea926ab7b204dccc32120
02ad9080dd0b504a7395f253b96776c1a9f1d2dc
F20110330_AAAXKD rockwell_c_Page_065.tif
fcf7c4513836deec343306f540891ab4
bfdd9c306db1a485784b1661d64311326998e6a2
F20110330_AAAXJP rockwell_c_Page_051.tif
815f4ccada3038eb8760df4c6098ffe1
014c003c18e1149f5ec212c1b7dd3999a76f6503
F20110330_AAAXKE rockwell_c_Page_066.tif
32f87749a4b12f4ebfba10cc3fa96129
76fb3b2e02fcf447b1cd6e0c2062b54465a7b023
F20110330_AAAXJQ rockwell_c_Page_052.tif
3332959b1de9c05e30f0a641b0cc4c11
d23a15d9f598f62a7dce5f4f399360c0797e8de9
F20110330_AAAXKF rockwell_c_Page_067.tif
89d8eeae6da38c80a0192c373fa37957
e669f7b5b5379d710554c7a1e153c3c33127ab09
F20110330_AAAXJR rockwell_c_Page_053.tif
235e6379a17e36af5ae30ff53f75d1f7
bf2cf235a66838554d91b1430d30948f132af509
F20110330_AAAXKG rockwell_c_Page_068.tif
611717606529019720cfd88e79f3cd08
ac911b52b93e395c6200785a0c98fa1c8500b6e8
1054428 F20110330_AAAXJS rockwell_c_Page_054.tif
2b072ba0a68628b9e44702aeed31d8ee
88f325d877d064aa67eddf5943a5c29411e0b661
F20110330_AAAXKH rockwell_c_Page_069.tif
fcbeee5b630f54ccd7a8bd9cb5374cf6
8589bf385f34a295b4f8223687c9c602266d542d
F20110330_AAAXJT rockwell_c_Page_055.tif
d030766b8da723fcee2d7d8810a458b9
d4cf4a3d404178d510c1eab6b0fb067552808919
F20110330_AAAXKI rockwell_c_Page_070.tif
344e6bbd8eec5615588e468fd134f7e6
37f1493b79335ad08b2351dece883b2ee866ca8d
F20110330_AAAXJU rockwell_c_Page_056.tif
771a0f5833a2220865a75be98bdd4005
180e7bb7e25f13719ccc5b751a10221d9ee37aeb
F20110330_AAAXKJ rockwell_c_Page_071.tif
b7cb94a2a88f99f25348dd2b99b39b9b
3b0651611de58dd6e5f7c585c39361567add0899
F20110330_AAAXJV rockwell_c_Page_057.tif
c5463ca0ab663c337d927b5f1b54f129
ac815bd3cc0985f778b4171d238b0b21bbee8205
F20110330_AAAXKK rockwell_c_Page_072.tif
e656f129809ee04e7870ae77a4557703
172f8d326ad5db71549dd0750277c2ed07511969
F20110330_AAAXJW rockwell_c_Page_058.tif
8927810e8a7d9091daee356cd5e0bbae
e398b5b582f52bf11d7e1db2a17f07bd50b57cf7
F20110330_AAAXKL rockwell_c_Page_073.tif
36b71b41af1e36c43ac5634f9c849288
e466f4d99566489ca1c4266469ddc5c63df49d5b
25265604 F20110330_AAAXLA rockwell_c_Page_088.tif
1700145896ba97e8c71b1aea3154c0ff
fd1b338e38f9d90f7b01a9c09cd75f6ba3285a93
F20110330_AAAXKM rockwell_c_Page_074.tif
4003f87069950d42138996bf467338e6
151e3acf4e83e99ef67e134e4af062fbe2e120a1
F20110330_AAAXJX rockwell_c_Page_059.tif
1dfb83e32af2c37f1adb6aa2a4d16b02
6862b5bac7a7d912d37cf665e9354b10c110b90c
F20110330_AAAXKN rockwell_c_Page_075.tif
e0417af8665dc44108edf3e8d653a3b8
ce2408f4c4396de3703e08840db5fa5e4149eded
F20110330_AAAXJY rockwell_c_Page_060.tif
7d5110f81ec96abefab62ebd4c2a044c
b6b4e1ceb69da835ba8c4cbaebb9384b733344ac
8423998 F20110330_AAAXLB rockwell_c_Page_089.tif
f3f6282828ac21fba801f8dd43984650
0795ef340808b24c3ad58c4a4c9f398db99f5415
F20110330_AAAXKO rockwell_c_Page_076.tif
af232f77da40c2c6892502492fe67b94
6d8ee24897550e5b43d08627841fb4a8408700ab
F20110330_AAAXJZ rockwell_c_Page_061.tif
0ea3425315eeb8797b9bea6088e586b5
2621ce54f7ddb5b6b1dea2d68fbdf315b59b262a
F20110330_AAAXLC rockwell_c_Page_090.tif
2e873d58a9fb47667bdfdeaffac3fbcb
e4c8958f385ab581b9e4a116c0e6ba4f76627a31
F20110330_AAAXKP rockwell_c_Page_077.tif
44e5d59f3421625788997836e299bcb5
0543df49f7942ac597824650dce874521daa2744
F20110330_AAAXLD rockwell_c_Page_091.tif
b694954aef7d752576595c8c9d26a76f
a3adf9bd146109e8087dbcb2f4fee1d0ae4e6498
F20110330_AAAXKQ rockwell_c_Page_078.tif
acd367277604d6d840a7f7c69ad3f393
b809a22a0b63d32bab1c8dffb0d8054bf7d164e1
F20110330_AAAXLE rockwell_c_Page_092.tif
eb2198f41c0ea6ce6638ae699151d0ec
3395ca9d46d10ff7feb486ddee1c73906cedecda
F20110330_AAAXKR rockwell_c_Page_079.tif
087758cf728c54266b0d0bb32663c4d1
2f8fbe62b9964c359f6d7aa83f8569fdaaeac2ac
F20110330_AAAXLF rockwell_c_Page_093.tif
36af1fc308b891343f0d24747470b14f
ed712c09c44dbbba4d73694ec40b9bba14ba08b6
F20110330_AAAXKS rockwell_c_Page_080.tif
e91c953b359fcf37186f2ea0c2746f7f
425adbf7b02b291acf5471fbc14015e051f23af0
F20110330_AAAXLG rockwell_c_Page_094.tif
955f7b15d0f0b2eb313d59a13f3f19e8
673093d6451b3db6094b01e799b4f21bc6cf1595
F20110330_AAAXKT rockwell_c_Page_081.tif
e4d2131d5e9bfb30b5d3b83729b7dfb0
210dd78ef53256a8fb71f5788c66e5429020cbda
F20110330_AAAXLH rockwell_c_Page_095.tif
6aa18a9546fede6b5013e89b0ed1bbf3
9a29ec8a52cddbe350b3b57ca7b9d8ffea7e2bb8
F20110330_AAAXKU rockwell_c_Page_082.tif
65216c1033a21487e000d4a7b37178b4
1963b524f5b0894bc0abe061dcd568019512df5e
F20110330_AAAXLI rockwell_c_Page_096.tif
ff3dcfee366ff371a29df8531a2e7cbc
d408f461c3ed57c294a4f87df5afeba7b73f3d11
F20110330_AAAXKV rockwell_c_Page_083.tif
e57ead2707dc14d004d9c2f91a9ece10
3621fec9df69ac7b48b4c1e70b6039733d6447e6
F20110330_AAAXLJ rockwell_c_Page_097.tif
d19d61ace7bee59e12a39dfbc4b71429
4988593dc638633e6836c3a5b09086cce92ec45e
F20110330_AAAXKW rockwell_c_Page_084.tif
71dc38ff4bffcd9d32a9f4983a0c7f88
06bc2524cb8e1185a50521138f899a66ba3ed1f3
F20110330_AAAXLK rockwell_c_Page_098.tif
9d9c02d41c814a34a46b20e0b8b2df57
ed2f0d4372541c9e2c172f95a5f875cc8f4d5e7c
F20110330_AAAXKX rockwell_c_Page_085.tif
e7212ff2c8d02f29b77e3d2025686f7d
17e0f3b9e19e68e834759a11ca6eb97cabfd68b2
F20110330_AAAXLL rockwell_c_Page_099.tif
636215f7edfbfb462c7f57fb990337a4
429de81ba059cc8aa09442fd4393b7d993915acf
1439 F20110330_AAAXMA rockwell_c_Page_009.txt
079d6e006a38e27f45e24f16bfa3f0db
cb76bacbf8a823e24a698c86a64078f67c57e7b8
F20110330_AAAXLM rockwell_c_Page_100.tif
169a0d278d81923b66f48eb242ab1c19
40c6a76b10fed315e759d59da199f2e8f28e3af7
F20110330_AAAXKY rockwell_c_Page_086.tif
807abb23911ae845cbb8d343f30d3b4a
d8d888a30fb5128bb3bcc248b2b6a9510f90d714
1280 F20110330_AAAXMB rockwell_c_Page_010.txt
97e5e05458bd48b48162b778d429cda8
859494e9cf83989909d94510c22a86c5b72cffb7
F20110330_AAAXLN rockwell_c_Page_101.tif
aaef8eceb9c270e6a63beaee7829f8d1
c1193982125f7f504d7db8b6ccc8cda5a9cc01bc
F20110330_AAAXKZ rockwell_c_Page_087.tif
5dbbc635db5aca4530cf8963828d2251
e6a620eb5d1b40e24266e0047eaeb35738d4df36
1772 F20110330_AAAXMC rockwell_c_Page_011.txt
598225499010a1db9bbf6baf66e6c561
0d3b5c78bcb54b4d0ac945b8d9bf66df2edfbac8
F20110330_AAAXLO rockwell_c_Page_102.tif
73841db2715850d862eabc08eec955b9
bae2ee52efd583409c4399a4043eeb0f58823a08
1619 F20110330_AAAXMD rockwell_c_Page_012.txt
6fb249f413e6809754b80247418711a1
20a9366f8980548b263e90c0b1eeca578d9b0451
F20110330_AAAXLP rockwell_c_Page_103.tif
7cd64886a86ff07964ecac362a551d9d
145059d70c8f646d6bcc8622539b2838952f155d
1867 F20110330_AAAXME rockwell_c_Page_013.txt
e0c2ea4c6575f52fb94d1b5f7a89b6d3
b760be59144a08a3f316f90ac96fe4f3a126c550
F20110330_AAAXLQ rockwell_c_Page_105.tif
891163472f5d676574e64131b9213318
dbb4e7cd63eeb37b39de4b2b279b46ec20c19253
2004 F20110330_AAAXMF rockwell_c_Page_014.txt
b0f3d32920b345bf9dd101ef0818bb8b
4f493a804eb83becdcdc8ab7254b24a97179e881
F20110330_AAAXLR rockwell_c_Page_106.tif
3839458fdff5678f8efe964026f45c8c
25199bc665e60c50dcd8726851275e28e19c114e
1791 F20110330_AAAXMG rockwell_c_Page_015.txt
97512625a7898050fde6ca9f5e030e0e
525f6300018fa4a8f8fcea33306fefcd7d619601
474 F20110330_AAAXLS rockwell_c_Page_001.txt
c63206792ff0db88192aaba87973d06b
ffdcd2095bd356995bef5812ce8a9bf4e8da1521
1984 F20110330_AAAXMH rockwell_c_Page_016.txt
977c80a39f2de548b770fe38c9c5cf59
deed371180a356f535711792385a1b69ed4db302
111 F20110330_AAAXLT rockwell_c_Page_002.txt
0e74b16e88bb36942d441b97c3a106c9
e7415ef8ab8cab630c830a730d0123a50a4b13da
1946 F20110330_AAAXMI rockwell_c_Page_017.txt
933fc4ec0907559d3dde51daf6814ad5
fc3902fe70b1b26e2e500c82f24a30a12f86fec2
278 F20110330_AAAXLU rockwell_c_Page_003.txt
850fed3dc449796974f9363fcfff5c31
d9125c0010dea1ce7569f1cccfda5bac8d0abf34
1903 F20110330_AAAXMJ rockwell_c_Page_018.txt
2eede77d8d65a1ce2251fd4559fcf48b
4ae935d3240e3bf3ecce28976cd94372f5d5310b
1770 F20110330_AAAXLV rockwell_c_Page_004.txt
df201d38fc23ce105d48ed3383d7e0e2
9a0a29be22441c34cd2ccb107212d96420e03c8a
531 F20110330_AAAXMK rockwell_c_Page_019.txt
21013659c6cb198de5a8165031cabeb5
10d80f081a7c7d1103e56ad3581aefd02f76630a
2006 F20110330_AAAXLW rockwell_c_Page_005.txt
5c6bae45bb52c96925396874b754cbf4
bae2cec0d3e3f020c95f4775e6b86ddd45d0d0e0
1776 F20110330_AAAXML rockwell_c_Page_020.txt
4fa9ba8a9d19647dca3ffe18e4b36d06
c39c78262d21e27b3c0e64ff0d437ebaa662abe5
1124 F20110330_AAAXLX rockwell_c_Page_006.txt
08489881df864a61e96a81a66ae18b5e
fff6359152c15a4c2e5efb3ccd36df1f2a00ce16
2124 F20110330_AAAXNA rockwell_c_Page_035.txt
eebb76bf1fd7075a1bc61027612c2e66
647135f16d4fec096c350d2aef00f19a52646a12
2045 F20110330_AAAXMM rockwell_c_Page_021.txt
ae7000357c63158f5fbe5379ec787fce
99b746c1a1147413e82129bbda769736b597bd8e
2799 F20110330_AAAXLY rockwell_c_Page_007.txt
f6ab53dbcb135010bb04fc6e39feae03
388c8fedb8b568758c67daa213c2460f6301a4a9
1982 F20110330_AAAXNB rockwell_c_Page_036.txt
81cea349bf44a6437f0861096d35ffd7
955d95c9a7876e6c031aed260276e0e9e984b3c6
1898 F20110330_AAAXMN rockwell_c_Page_022.txt
1f00154712bfeab3fff0cf8a1f2de51d
9c734ba334060950bdd4ec18312db46343c2abf6
2053 F20110330_AAAXNC rockwell_c_Page_037.txt
83f0667abab7f2054cf6491267e18044
bfd2cc58a6a50767dd5039885c3b8085eb301159
2085 F20110330_AAAXMO rockwell_c_Page_023.txt
cb0368708478c618cf8eee2ac700fdb0
692f036fdbb3ce4f8222beac6d3a506612508fc9
2841 F20110330_AAAXLZ rockwell_c_Page_008.txt
8d7485d20e7b7e8088571b6681aae090
22449a6c12782a3f199e322af2a6a52c03c758d5
1976 F20110330_AAAXND rockwell_c_Page_038.txt
060a2adc0def5a3852764fa215a67e7a
423bb4811acad5a5f604b82f5071b00788ce42ad
1997 F20110330_AAAXMP rockwell_c_Page_024.txt
63bb39879c4e92fb0d27bfee25815563
a00f3a7e19bd7d15f7538a795c2c0e0302c33c9b
1987 F20110330_AAAXNE rockwell_c_Page_039.txt
9c6ba811cf6e172344e33451abe1c0e5
d05ce2f2cb296d81408e824973c926565da4fcbb
2038 F20110330_AAAXMQ rockwell_c_Page_025.txt
0382798d118c81d8036bb7632ac31688
b5b766547e802b98cbb9401242d64be07e635199
1949 F20110330_AAAXNF rockwell_c_Page_040.txt
a078850f5a8e23c70fe05f6f718ffcc4
5b51853bd5a8dcefcb913752168f9754efc52926
F20110330_AAAXMR rockwell_c_Page_026.txt
af2c6d2331ec1a2f35551661e51c53d9
0d064af77c80f54298a4f10b4b969e264d539edf
1966 F20110330_AAAXNG rockwell_c_Page_041.txt
9d581c33b349204c70e087f2c9176453
995cbfdf06303389c03dd724ad0c2be7997542b9
2050 F20110330_AAAXMS rockwell_c_Page_027.txt
a72ac682914be4af60415f67fe39eb5c
d2f8ef140feb1f4fd76b89843180a39da3b6218f
1960 F20110330_AAAXNH rockwell_c_Page_042.txt
2a6159456330b7b200d66a57db2e7db2
958c556230d346242401586e8c96e90072a9ca88
2081 F20110330_AAAXMT rockwell_c_Page_028.txt
6eca749197686f7d5f50e0f28d097125
16adbd94334efc1cdb605b7c50d333d41fe91e98
2023 F20110330_AAAXNI rockwell_c_Page_043.txt
45f47b802d5985b79d6f7b13191d680f
8b7f3698598bfc5a2dc98e60fdcd8e1b698080e3
2001 F20110330_AAAXMU rockwell_c_Page_029.txt
c76eaf0b979cbaaad041c23436668057
1ebff3a5f0bfe175b9c238057843d971da40a0bd
2103 F20110330_AAAXNJ rockwell_c_Page_044.txt
bd5a4fc5a29295897e6931ff21bc995a
59c0f7ce9716a2c3c9335ab2f79e3e14a8c2bfb2
2013 F20110330_AAAXMV rockwell_c_Page_030.txt
8b281172389e04024c7b386fed1b5fd3
fe5d9eedfd108d9a23c72c61952dde3ffb0a4930
F20110330_AAAXNK rockwell_c_Page_045.txt
55fea67d1a662bf4789cc4f0151915a1
4c7b0a6f579fee3f1cb4e8eeb9dab48860351be3
2061 F20110330_AAAXMW rockwell_c_Page_031.txt
9d459640813b3f4ca8e643caf0ec8af3
7a6932817a0899c02b2db0398b59654c4df864da
2059 F20110330_AAAXNL rockwell_c_Page_046.txt
a7a2744ca5cdbd9f56c78258e8017a0e
1f9be11f8a599ca99cdeddbe6ebd436bcd784562
1930 F20110330_AAAXMX rockwell_c_Page_032.txt
78d44966face6db4ae0c08ccca18adb1
74c2faaad0f1648d2ad5d168905e325c96b0568f
2048 F20110330_AAAXNM rockwell_c_Page_047.txt
3c908039f03a2ca2856d070b0ecceae7
89bbb7172d8734ead86e92d4a58d338827f69dad
1956 F20110330_AAAXMY rockwell_c_Page_033.txt
def7c0f32940a32331716a22503ff271
4eb8d77afca9d268b20a135d53571d6be473e363
2033 F20110330_AAAXOA rockwell_c_Page_061.txt
81f16ce9e9f38e220979f4b089f83eba
2e8cfaf3ed407d22ed6c1ddae2a282e3d56ac25c
2036 F20110330_AAAXNN rockwell_c_Page_048.txt
cbe3a6cafefe77506e6752f7afbec67c
772d8c0704c6b3682a6eaa41293c05ce86868712
1853 F20110330_AAAXMZ rockwell_c_Page_034.txt
222db81f053dec3262aeaa98d277ce42
f530130ffb340c0720706a92aee70d953445106f
2044 F20110330_AAAXOB rockwell_c_Page_062.txt
3d0c2c90f97514583f561f773427d077
165b0eb5b851afe9af8eca3f511b4ee085980237
2120 F20110330_AAAXNO rockwell_c_Page_049.txt
30d8fba4a5a2afb3b4beb28494133a6c
c768021c6461affe13d114d743f9e72cfe54c926
2003 F20110330_AAAXOC rockwell_c_Page_063.txt
b5c177bfe94dec2ad5cf1d27a557ae30
af42c4a7c30475a77d9d22a13e8ec5fd897c73b8
1962 F20110330_AAAXNP rockwell_c_Page_050.txt
8c72b6c6bad4910c59825cc59754cf84
cc7902cf9de64f5f5da52fce47d8c0aa3018a4b9
1971 F20110330_AAAXOD rockwell_c_Page_064.txt
e2cb6755c34c15c43b042a69d1b2ad8e
1bdbde8ea732b4fdee78d1768b64ecc0bddd7a48
1979 F20110330_AAAXNQ rockwell_c_Page_051.txt
43accb21cf6ed225894517975528d012
a1517368afca714a9cda63fd5d571343fc1cb4b4
1657 F20110330_AAAXOE rockwell_c_Page_065.txt
2c809c6b251e18f0afde510081caa2d9
27258e78125b6480d063f990a03a4bb02ce7af0c
2005 F20110330_AAAXNR rockwell_c_Page_052.txt
4d7a2574a4ec3ad486af4cbde7969656
ff888f1403369d7ef912c6bce551901f3e591c1c
1803 F20110330_AAAXOF rockwell_c_Page_066.txt
8ac0952d4c4fb404111f691382cc19f4
1107e7d8c59314dba1a0882e838ac9087150de37
706 F20110330_AAAXNS rockwell_c_Page_053.txt
338f46a50faac2b3a25932a221d01275
00ed7d048ace80413e42b65c1984524efc56a3f1
1873 F20110330_AAAXOG rockwell_c_Page_067.txt
3c6f013d8a06c50a221362eee1de2f5c
ae28210d09d40ce6387eef71502d927008b417e3
2557 F20110330_AAAXNT rockwell_c_Page_054.txt
3c45dd95857cd7d3da7234627478f164
6b2a8b68be2b1f53cbe5b39c6d32380ebd646076
2037 F20110330_AAAXOH rockwell_c_Page_068.txt
d3baf79748705c74c8a53d867302b708
5f637bc030f292f7252a80c5de27f9ba07ef46bc
1567 F20110330_AAAXNU rockwell_c_Page_055.txt
492464d94afd3c914d480c97c565afcd
9aac5d5677bb4db9b673cf299876d786413e1f41
1924 F20110330_AAAXOI rockwell_c_Page_069.txt
ad58a8cd2c25d08a050543e8a2c8164f
38745fefb32beef5c6719b61357927e75cbff574
1794 F20110330_AAAXNV rockwell_c_Page_056.txt
bc7aa40e806f1cb90480b0e2a92f2528
aae1eb589d01e963fe01d638e7c885d7ed41fda5
1876 F20110330_AAAXOJ rockwell_c_Page_070.txt
00a61bd651bd52ebf5af6c6ed78eca13
caffe8788ec3ccaecb400546d3a627d9b258f834
1916 F20110330_AAAXNW rockwell_c_Page_057.txt
474bfc00f537f113bf96ded52447284f
24bec5c9505830ea1637c4161ce379f85d0fbe8d
2086 F20110330_AAAXOK rockwell_c_Page_071.txt
cc859b75374da6270e4f1385f901720d
2e765a5df04c1a14ad676118718bd0d8a79d9730
1973 F20110330_AAAXNX rockwell_c_Page_058.txt
372009ce2f144ef2bd350e11d7e9d56d
12fd44b612bce2bea835f8dc74b9a62931250aed
2024 F20110330_AAAXOL rockwell_c_Page_072.txt
a265d9f1caebec3bc72e14173e0e0428
1b0ca347bb1f8b8f5a5f5cfbb5347f2d62285421
F20110330_AAAXNY rockwell_c_Page_059.txt
ab3736b1c9606813d9744b337c4ea4b3
11823165f18885801df835eb4ccbabe20460df9c
238 F20110330_AAAXPA rockwell_c_Page_087.txt
8ce1e2d668b98d2fd34549a1325172ba
e4ad7dbc6d06a986ef6942f3f9b019770906ac35
1941 F20110330_AAAXOM rockwell_c_Page_073.txt
4c81dabcdb1a90507960cd2626f469e7
4b67551872d887179394c5b7bc9285a2b44ffc4a
2020 F20110330_AAAXNZ rockwell_c_Page_060.txt
1ee3ffeeba47bc20765843a2a83c51c3
a5782872728b8da94e9e44665dd25ebb9b1e8548
F20110330_AAAXPB rockwell_c_Page_088.txt
06e48add54a6589fa644b0586e9bf00a
92b70c19a9afdb0c78f2cb971c6d4e6d1d02f03f
2149 F20110330_AAAXON rockwell_c_Page_074.txt
9d37438b7db728afa6042ef53f9192d6
062bdcdda9b79d6c6d446da2cf3a2bd1085e993e
594 F20110330_AAAXPC rockwell_c_Page_089.txt
3dd55e6646fd72bf53e694df076c62ab
96c01579d88cc95bfd0446023d24ea7ed480ab5b
2022 F20110330_AAAXOO rockwell_c_Page_075.txt
f6abc142a8b88c975d500c97e16dac28
31ab69ef782a0c38fd8a81ea4488dda2d3cf3a19
1852 F20110330_AAAXPD rockwell_c_Page_090.txt
870f347f951dc76f9a946263c2f395c6
b0a35fb55908e32dfb13f88fdf99427bd78aee69
2065 F20110330_AAAXOP rockwell_c_Page_076.txt
fd684a9a1cdd1cb9bffe93aa848d6d98
3642bc4ff95c1b5cae57a58a3f94bccbf8866e60
2029 F20110330_AAAXPE rockwell_c_Page_091.txt
b3c0ce869f033653097eccbbff18997a
59e62cabe6a850ccd94137ba7014de59e062e9b6
2064 F20110330_AAAXOQ rockwell_c_Page_077.txt
7470e2e85eaf84d8ce5b30e61f966c79
5fb193539a23eb85da7a87b74740289d507fadac
522 F20110330_AAAXPF rockwell_c_Page_092.txt
7ee6a1e64bdfd556613eefb38d2a3500
09df518233d35061b13a9adcc73613bfc4b7b709
268 F20110330_AAAXOR rockwell_c_Page_078.txt
6457bee9409d93c62ddd5691d71e9df4
a7d0c0494c2401abc32e1249bcb2ced1e047d19b
2105 F20110330_AAAXPG rockwell_c_Page_093.txt
aba223eddba5a0d3eee5b395692c2094
54946de8e22f23e6836d8b8d390690b876b2bf25
2247 F20110330_AAAXOS rockwell_c_Page_079.txt
77d86041cff1320a47c2758048b866b8
d8e275a53a8ad6f7e4fdcbca9e04fbd64677be79
2629 F20110330_AAAXPH rockwell_c_Page_094.txt
91349b54f03dd210a41df0f025bbd3db
abbd2d575bfa788bae62f247a67dbed5a15cbe9e
1599 F20110330_AAAXOT rockwell_c_Page_080.txt
1e9042b75435a9de579fc9eb5761b9b7
73c7fda4e2ee3ac4c6132eb06a778e6f5562f377
2580 F20110330_AAAXPI rockwell_c_Page_095.txt
8546f74d081db7613c38ed9166a1c22e
c858945ec52c7b156aa746560d279352ddd1699d
674 F20110330_AAAXOU rockwell_c_Page_081.txt
12cbc9579cf8cafd66ec2b2830685918
3553f115548b533bdfbe08ae2c255f388c4cba78
2481 F20110330_AAAXPJ rockwell_c_Page_096.txt
256bdade5d6d5dd047b4c7ca19fb4bc6
c11db627055e382f95bf802b04d239bc606ccb6f
1992 F20110330_AAAXOV rockwell_c_Page_082.txt
0ef4077c5e3a65c4d8597b7271091784
f0b1fb5e29fb46e9e23dd7085442a22ccc63858c
2725 F20110330_AAAXPK rockwell_c_Page_097.txt
3474ebf8e4ece023d48e54399e39c3be
07956f8f441cc9ccc24faf6b6b4044adf2ab42cb
1804 F20110330_AAAXOW rockwell_c_Page_083.txt
6e5f8b0645e4a2bae1d0a493f0e0d7bf
c3887486071061f91aa08f946f1fde8f567eaff2
2403 F20110330_AAAXPL rockwell_c_Page_098.txt
29322094d935d9a3dae9ad83516e5e6b
ef4a462988b28d60136493f72c45ce46234874b8
208 F20110330_AAAXOX rockwell_c_Page_084.txt
d295df83dfbf7e4811b7be6285a2208d
4cc39bd99059068f64bc2a9a9eb0c588ed387d88
2513 F20110330_AAAXPM rockwell_c_Page_099.txt
9c9d8ffe50f1dd6376362ae577a18e0a
adb77c040e2015ff761015b642005f42ee490b22
419 F20110330_AAAXOY rockwell_c_Page_085.txt
6e26a756032b62d7a300dcd38aed0be6
abf010a763014bc672f1ff5666726f7bf417409a
32568 F20110330_AAAXQA rockwell_c_Page_009.pro
50a2a8961fd0dcacdcd0bec4602c8af6
0b20e87aa5fee0ae8d8f9726471f38b30d562dbe
2521 F20110330_AAAXPN rockwell_c_Page_100.txt
e75b3fe4da9633639c0ab43b0217b4de
fa67d7160e6c7db246d3a5c244c1f4b3320ce7a0
240 F20110330_AAAXOZ rockwell_c_Page_086.txt
50d79e1001aa5f854ba3d6ea00844843
9f167331266f74540322953aabeaa6fad24dd4c3
30457 F20110330_AAAXQB rockwell_c_Page_010.pro
ab325624adc1d4a022e24dfec18f24a7
23801b05fb79d8354e06d8bfaf89643abdcc0ca9
2310 F20110330_AAAXPO rockwell_c_Page_101.txt
709ce7218579e0142103f3ba0441e9df
f10a8f952224861fc17cba7f0246c4cc4083c400
39641 F20110330_AAAXQC rockwell_c_Page_011.pro
eb9c673fd6a5f6e72e712e5fc521937e
328b85299109c82dbf33069c23db12623f15936a
2733 F20110330_AAAXPP rockwell_c_Page_102.txt
717269e89766008d12d41f45f7b7b5bd
983dd0b056e9e908a48edb9b2e2fdcdb55796435
40651 F20110330_AAAXQD rockwell_c_Page_012.pro
4e69bad2f9761677811027476aa68501
d281fdc32581d9309c2804c8289c75c7f6153139
2589 F20110330_AAAXPQ rockwell_c_Page_103.txt
45b7a0898c8cddd4d0332f7032f31d21
3b91251751741aeff057f4ded9d71f61644285f2
45775 F20110330_AAAXQE rockwell_c_Page_013.pro
38df82c25fb498348c971351ba820edd
7957b46904f462408d5a5477a2ee0ed9fc2e0a23
2583 F20110330_AAAXPR rockwell_c_Page_104.txt
41bc4975dcf956672267a4ae946a2651
2640f65e5636032d95f702a0e010934554d2c5ee
51091 F20110330_AAAXQF rockwell_c_Page_014.pro
a02207f9b25d2d29ca8e4c7da365ab06
d50f72fc5e1132668fe8da232337afa9a51c8a26
2017 F20110330_AAAXPS rockwell_c_Page_105.txt
d18385e0501dda526724f1dbc93ade56
92be13328148b585fe9f195ebbbc40b85d5b154a
45268 F20110330_AAAXQG rockwell_c_Page_015.pro
210b29457e86c1417035664118b6b1b4
add6cb01eb8372ae722d4eca16e089e44859184a
1349 F20110330_AAAXPT rockwell_c_Page_106.txt
963de04e10bc986f4ba22e284e735498
6574ef5e9976304806dc009c34c829bca325bb71
50378 F20110330_AAAXQH rockwell_c_Page_016.pro
3631d8624501f52acee3e1a23edec754
75e7dd59525d97bf21192a6bc91f722bb227cf11
8631 F20110330_AAAXPU rockwell_c_Page_001.pro
79abf4add5a8026da9e589933719f35f
a7840ec7263ac6e662c736a77614fd7e7a57de30
49446 F20110330_AAAXQI rockwell_c_Page_017.pro
f7cd9f1513024be33cf37c9886287252
4e00f4a578792fe1863607a485e0f5e45d83f0f6
1202 F20110330_AAAXPV rockwell_c_Page_002.pro
c19a16359d724913c5c12ab52782ec94
0c1964ef73e537018e3ef650267aa6ba90183a3b
48018 F20110330_AAAXQJ rockwell_c_Page_018.pro
e9fcd46c3b7f75e28108ed839ccc5141
ba10f29fd28a0282e58d1c3da383ef1fd0f517fc
51127 F20110330_AAAXPW rockwell_c_Page_005.pro
eb8bd6a1be0f3276b90114b04fda5d72
86ea84744dc70dc8ba0f14f0fc46386fe663c25b
12062 F20110330_AAAXQK rockwell_c_Page_019.pro
491b52f98b00519148e1b3670ad26b52
5527b59c2417d456b56426eea9fdafebd3b311bb
27784 F20110330_AAAXPX rockwell_c_Page_006.pro
45a46fc56cbcad70a1baa61a90610bd5
1bebe55c635425b1fbd1bd88cb0ae4698d429385
42521 F20110330_AAAXQL rockwell_c_Page_020.pro
bf619a2edafb516c9fc704429ea576bc
0e7c943693bf48e9f61774a87b12618bd7a2921a
69336 F20110330_AAAXPY rockwell_c_Page_007.pro
f74b220490ea1c47590f03f39b2f4090
64dc80d98c4e5398564b9538058a19ae319cf6c3
54080 F20110330_AAAXRA rockwell_c_Page_035.pro
6f7963ebf198c541934fafcf75bddb57
e7349716ab1e4dd721db6f50ddf4f111213599e2
51926 F20110330_AAAXQM rockwell_c_Page_021.pro
8fdd4beee0aaf9847d15b4c04563ba71
accf430000054d2cd1a620680fe02e433eb25b09
67031 F20110330_AAAXPZ rockwell_c_Page_008.pro
9468c2d9d2b2a14246dd54647a0687f2
f354f3d63d143cf79f3bf20d6a9ecf7b60e26f35
50343 F20110330_AAAXRB rockwell_c_Page_036.pro
15755f72724c499562c3f15091834fc9
7c50b677009b6813b13c5b3873449340357f2679
48195 F20110330_AAAXQN rockwell_c_Page_022.pro
422f311ffb1a722df26a6644d2d2223b
bc4dcdc872430e6e11cb042c7ac45c3b8425b843
52325 F20110330_AAAXRC rockwell_c_Page_037.pro
e3cac2bc07606b9b399a64943eac58c1
c96b61757c12f2e6d3b3ecffc74c8e3f2f6c9e44
53128 F20110330_AAAXQO rockwell_c_Page_023.pro
70badbc4c04d9a9e47605d22631ceda3
2c2e301d34eead29fcad34f82c2d01bc4a782d3a
50196 F20110330_AAAXRD rockwell_c_Page_038.pro
f8134fd67b17e361a28853d66ffc64db
57ca401023c5fbc53ffd3174000949b5a29e9609
50752 F20110330_AAAXQP rockwell_c_Page_024.pro
763b03cad7d0dc15433b0b1e679cf488
7d0f5a060ca889f3996821465711561ed424d7e2
50601 F20110330_AAAXRE rockwell_c_Page_039.pro
b1d9917d9de4624212c37fd56d8f6d44
e6395b8ca4761e7f1be0bc11c231f6b45b060be7
51330 F20110330_AAAXQQ rockwell_c_Page_025.pro
7172c3fdc3584bdb026d7bd9c5f163e6
41b7597bd5307fe37d6819206d16df91052c5db1
49557 F20110330_AAAXRF rockwell_c_Page_040.pro
fcc56d2e96e121535fe85d0d0df54467
ae77826d0cec2fa7484cf2a5a74fbd8c505d0b0e
52114 F20110330_AAAXQR rockwell_c_Page_026.pro
1f9ec8b3e0f637fbfcd1b911cfd2726d
0fe261ff4c3f5d5642e2aa1a786246516129edfd
49917 F20110330_AAAXRG rockwell_c_Page_041.pro
876802c0924a5a171952967c12e92ef8
dea61a47764137d770a2252e4f5571071b7f3b04
52428 F20110330_AAAXQS rockwell_c_Page_027.pro
e5c8d57a21770f9088772b2b235dc87a
77275f03456326ccded57929b47d584619849241
49901 F20110330_AAAXRH rockwell_c_Page_042.pro
034b35ff0595197a9efa6c8e4a02105a
b193e20751a73310b6e36e9112d78bc6fddf1dc4
53239 F20110330_AAAXQT rockwell_c_Page_028.pro
218f1718dec7cd0cbbb1fdbd3ba63241
9fed8b8ee304f3c43efe45f0cd6fa34d0d7cae33
51465 F20110330_AAAXRI rockwell_c_Page_043.pro
af4e5f35bffeb2b30a07985069ebeb86
8b5be2e5a961baf02a7e4b8a0da5c7272f05cc5e
51138 F20110330_AAAXQU rockwell_c_Page_029.pro
64eae630812e90f97a26a43dc1ecffe3
92cb65e2bb0cfd518ca764c23fe885c15bb05bb8
53750 F20110330_AAAXRJ rockwell_c_Page_044.pro
8907017a5d518104d94c235828205e3a
c8e9725ebe287d4b1244b866d186d3624a279416
51454 F20110330_AAAXQV rockwell_c_Page_030.pro
fd0f6ad7c32a9dd7b3dc74d28fef7c88
f4eb20664ec1e5465a3fac575deccaaebdfc48ce
51470 F20110330_AAAXRK rockwell_c_Page_045.pro
05d32ea30f4d4a6653dbc2abc57d7fa1
88c5bc8725017d63632371d7d2a0117234dd4c67
51539 F20110330_AAAXQW rockwell_c_Page_031.pro
7a82aec022dd959bad06ec2fd5fbcfe7
80bb01c0748b12f301a2e98e62a574ccf1e52d7f
52261 F20110330_AAAXRL rockwell_c_Page_046.pro
8d2c23a30559de884e751d985bcb25ef
1c57b108bc205d0c49ca80fa09cb9af580af8208
49048 F20110330_AAAXQX rockwell_c_Page_032.pro
3b6d2448a72b7d7130ae2dc24b10cbfd
1b8944d5d4450bf1699f8146f768a87c3520ff3d
51162 F20110330_AAAXSA rockwell_c_Page_061.pro
da90ae8e9c723b9c049b3a61ae060a8c
5d47536e5ba947e3feff2f971d5ef91bb0b87a55
52205 F20110330_AAAXRM rockwell_c_Page_047.pro
7249e32dd3bbf6fe6f656ca368e6d705
3850ad897ce965a5a78a37d6ba00618851e6b77d
49674 F20110330_AAAXQY rockwell_c_Page_033.pro
cfbc9c42b084710f218decc6cce0006a
516d3439841a11bfffc239d184a2447abdccf9c5
52140 F20110330_AAAXSB rockwell_c_Page_062.pro
382f890149dd1adbac792e6a51c1de74
c6b5e41e40f07db6acb08c5e3c20c58b6f4e8f7d
50788 F20110330_AAAXRN rockwell_c_Page_048.pro
2a7b31947ceafb7bc3071d3f3c3ccc46
fd486dbd4e5b874601ba473a703f3cb92226e12b
46970 F20110330_AAAXQZ rockwell_c_Page_034.pro
3afd27ada945d766fad0521fc03f8846
e8697ca4b9b2cdb01d9503f8692fb1e92391ee54
49932 F20110330_AAAXSC rockwell_c_Page_063.pro
c2289aae2a74d54d7600b3a0461bb08b
fba824b167bd9eba632231d58c9479f40eefe10f
54146 F20110330_AAAXRO rockwell_c_Page_049.pro
94a84352817e2ec8a3abff2bfe4f966f
6991bc0dd3b8d4519c14e5f7d1a51e80a042588a
50100 F20110330_AAAXSD rockwell_c_Page_064.pro
b75aea68a7e42055e68c2f798908ff0b
60143d4f918f4b68375d00e2aeab0badbf3e70f6
49997 F20110330_AAAXRP rockwell_c_Page_050.pro
85f1941926f0fc78ee0d933197d7a1bf
9dd6d8f0521d6ed86d200db925dbe6b2d6e123c9
41417 F20110330_AAAXSE rockwell_c_Page_065.pro
fcd317226f283a9631801da4d23542ae
4e52b7dc2089e6f2e0fd3642ae6075b8c1d32700
50396 F20110330_AAAXRQ rockwell_c_Page_051.pro
afd28aeebfd0b7356b51b09e36f5e3e2
c53c83d76ddfdf948a0f247aae491d94d4b2d286
45338 F20110330_AAAXSF rockwell_c_Page_066.pro
cf8c087e09dade86fc76e627bff603a0
681e2bee16f3695c09bf993dcbbe5c5ef9efe1e4
50971 F20110330_AAAXRR rockwell_c_Page_052.pro
a570225247c52384c719af756f06517c
2a66f77c38bc02ba3d373dfc0c81e373c09a272e
46279 F20110330_AAAXSG rockwell_c_Page_067.pro
7e84ee2db33b0ca7aa341f54291c8300
868c7bbf7446194e7c51c55d8aec37a9daeb1df2
16729 F20110330_AAAXRS rockwell_c_Page_053.pro
f3c59e9a2d76a5a537e90e1cea50409f
8d251f43bb1a98c06686057ae3404821f0813abf
51845 F20110330_AAAXSH rockwell_c_Page_068.pro
feb7b9c393c6e1b30cf6e9e554ebdd4a
58653af80c791c0f1910b0b72e8f6fca99a7eb75
45941 F20110330_AAAXRT rockwell_c_Page_054.pro
87b2dc23e0848722062a26bc351fa325
df34f4e221675c9dfdf48adf080a915179b4feea
47766 F20110330_AAAXSI rockwell_c_Page_069.pro
b80e0d1e991b30b28f9d528135ce054e
fa5b882fe3012a6324992e6552b9276c7ebc62e5
31999 F20110330_AAAXRU rockwell_c_Page_055.pro
9cc82c363b56716d17e4a8c349ee7bba
a1d2feeabc5471931db597668188e82370b8b561
47554 F20110330_AAAXSJ rockwell_c_Page_070.pro
407c9889d37946f96297332a59cbe42d
8152b96bc9667b9d8effa8ce5ee3ab56231a380d
42510 F20110330_AAAXRV rockwell_c_Page_056.pro
e673f9b3affa81f3aef908acc3c3eae5
fbf9dafb32ecb27c7634b325c557e2c47e55a3f0
53136 F20110330_AAAXSK rockwell_c_Page_071.pro
0af4ae574507d676d892887f77dc163f
9e32243279b4e79ffb2e7e17a33be27ab3e0a70f
48599 F20110330_AAAXRW rockwell_c_Page_057.pro
e7a8b0ce237dc7ee606ee1bd1f308363
d8799f024f9ae734157eef541c89e82fe0f15dc0
51586 F20110330_AAAXSL rockwell_c_Page_072.pro
3b6286b8726a35e9b35e78f66e613c61
7b9c9042cd8b7a98ccbe1338884b94d777bb14d1
50112 F20110330_AAAXRX rockwell_c_Page_058.pro
74bd5c2f8ee0c5e341babbffe6686f4c
1206fa7f6a5aeb1d70e9b9633174798725ad66ef
4246 F20110330_AAAXTA rockwell_c_Page_087.pro
a912506d4f63597b5a60935f7710231f
057105216a3a9bf045551389de8af94136e47858
49508 F20110330_AAAXSM rockwell_c_Page_073.pro
9cb0a25b7de4878f4d61c9cbcde93010
eb9b57b2b4a647ba1a8ff829277d958578495cc2
49251 F20110330_AAAXRY rockwell_c_Page_059.pro
63aa7fb11998c91e3b6eee48a0c78222
d126b36a02379bad8b9e5e08839cc9059faf269f
54978 F20110330_AAAXSN rockwell_c_Page_074.pro
c4441d4a6f71c7962947c499250927c8
cd70d07c89aed4e6bf1c8c137a88262b3d18b07c
51466 F20110330_AAAXRZ rockwell_c_Page_060.pro
4e4c0f7bc4b8bff9f4e8c02ca60894c5
2f1c44ae18771d1a82445146103d692280f2f382
4228 F20110330_AAAXTB rockwell_c_Page_088.pro
bde82660f290eaa9e9270a9fe234177f
c884e60fa47001a8c715ece02cfd739d2639432a
51471 F20110330_AAAXSO rockwell_c_Page_075.pro
ab0c8c21c91843b7f78e1d81f2fee732
e6e9a96e6403a91980478466d6f9001a9ffb03c0
8305 F20110330_AAAXTC rockwell_c_Page_089.pro
3d5af87119cec77a3604efe974e99b6d
2dfd2c137b9d4255b3da528ce9e491727e838b9c
51648 F20110330_AAAXSP rockwell_c_Page_076.pro
010bf2da1b15f4283f95157ec75e23ca
3257803a7562908e7def9da542e055d6ea113c82
45063 F20110330_AAAXTD rockwell_c_Page_090.pro
06df97e8089c816417254d6b7478282d
7c2315993242379b56f13b7b9175159050229e53
52622 F20110330_AAAXSQ rockwell_c_Page_077.pro
75329eed6ed854a2b1e51f0db3980fd8
62adf6634d114eab703dcc1221efab44a208b3c0
51559 F20110330_AAAXTE rockwell_c_Page_091.pro
2f33a005a219a3ccd20e03b18920e436
44e9b1d05a6328347cc18ee16b6f4836938666cc
5682 F20110330_AAAXSR rockwell_c_Page_078.pro
67c394465202b81be08ac8d34509c1f7
56aa3f2441d54c18cd432860c21ed6ba94ca58a6
12044 F20110330_AAAXTF rockwell_c_Page_092.pro
7568e3223649e5343f1093320c26ef54
8be2e2380cab7580bd035387e4cd8a61103e8c54
47404 F20110330_AAAXSS rockwell_c_Page_079.pro
f6128f8d003c5f1217d492c963a48b02
df4623ccc45b55b119ec1c6c433103b42f31fb29
51625 F20110330_AAAXTG rockwell_c_Page_093.pro
03077bed3f91fe3169e65af1fe59fc77
2bbc18bfa4119e3f8b0ba816357f27e4701ad1bb
30284 F20110330_AAAXST rockwell_c_Page_080.pro
3592e874a36efc81effd57d09f64f216
120bddfe72949590ce1eee6c45a289aaf6911c8f
64529 F20110330_AAAXTH rockwell_c_Page_094.pro
6990dcfacc5c423fc58a4d4a068790ce
3a7763564ddbb36f7c0434987f42bdd1a297dd5f
11416 F20110330_AAAXSU rockwell_c_Page_081.pro
20eb3306b1837ed731911ba459ff60fb
b976100622f66573e700c1b6f0e19ae6dfc3f233
63967 F20110330_AAAXTI rockwell_c_Page_095.pro
cb3618e056926dcbc0149108a712c77e
f6b8a6182914030bc1327392c2d3eec2244d02c0
36870 F20110330_AAAXSV rockwell_c_Page_082.pro
501fa85e74b91fd64dc6e046fcaf3a75
fcf42c0020d7a109085546aa910939848e75faf1
61050 F20110330_AAAXTJ rockwell_c_Page_096.pro
867285be7b3d1b65322174d3c9f96eed
e834da048e164adf1b814c093a34bddff79e7c78
31925 F20110330_AAAXSW rockwell_c_Page_083.pro
89033c7ee5fb17b3816012065d17096b
07b2d10d5f88240a605b597edec2fa1dd38a3765
67291 F20110330_AAAXTK rockwell_c_Page_097.pro
00f0cdb867ffa7a6d496b2eb0b95b585
9754d92b776c7399fbdf1c048f5d7ba71ce4a598
105195 F20110330_AAAXUA rockwell_c_Page_007.jpg
0daa3ca9db7ab66428a97e3d35db2d85
1109964931cea6d6bc1cb28c9bbdb8f61761cf69
58857 F20110330_AAAXTL rockwell_c_Page_098.pro
b708cc280338425e6abb348e302c7af0
9de63e47ee3972c583d25c21a549078af1160205
3579 F20110330_AAAXSX rockwell_c_Page_084.pro
7767f678049a8740ea775d2f7025a202
7baf0655b6f5920c665c777111ca4b15b834fa23
83370 F20110330_AAAXUB rockwell_c_Page_008.jpg
72ef41e0472d711327f4ef23c116f6d4
40d4a5b1d81f9bf20f94a5d4e4025a8a6e88b057
62037 F20110330_AAAXTM rockwell_c_Page_099.pro
70fd3320ef50d8540e525edc7dfcbcb6
3a755e1689c3a869e2de798e42db0f695fa86f53
9292 F20110330_AAAXSY rockwell_c_Page_085.pro
915a7c7338676d43bd3edd7dc3f056be
75115389d45c6d5589f2940858272a658ae861f9
61863 F20110330_AAAXTN rockwell_c_Page_100.pro
e41b486fd5411dcec47ac530ee2fd025
a7f53fd9e9230919b1d599710f657a2d4bca9411
4068 F20110330_AAAXSZ rockwell_c_Page_086.pro
04ce600f063710fe08406525be2523dc
274219eb3dbfeca16ca8eaf5cf1fac0ae0ae4074
70404 F20110330_AAAXUC rockwell_c_Page_009.jpg
f8efd8ba7a5d06eae41ea42ca0df59a3
7c23cf808d311c98b130d82a0cc1099972ecbc68
56527 F20110330_AAAXTO rockwell_c_Page_101.pro
cf41a6e310fa2867c1706ab0694a2fbd
055e3a011c1fda6370824b96b16d014d7a0a4a77
60084 F20110330_AAAXUD rockwell_c_Page_010.jpg
a47847cb99236447bdbd6c9577e077cc
618ca94dcec5e4a612bf1a9b279cae0454e2e7f3
67518 F20110330_AAAXTP rockwell_c_Page_102.pro
eb670abc3afc1c9cbfb43b897f2634d1
5d7a04424a6de95dfd200c4002287fdead35ba75
84678 F20110330_AAAXUE rockwell_c_Page_011.jpg
96b8976cb1630b485ffd219f5c7a8f97
e20159e737df577cc3b1dc1673374e3c018ed54e
64279 F20110330_AAAXTQ rockwell_c_Page_103.pro
1f420035af7694b0d704ecfc629678c1
664a3e96a9c56056444dd1e284cc52c95d150cb1
84642 F20110330_AAAXUF rockwell_c_Page_012.jpg
a5d7f3c878dab8ea9fb5afd0b68ceb40
be6dd8fcd6e3ea3e489099cdb2238d619ca34d41
63573 F20110330_AAAXTR rockwell_c_Page_104.pro
236bd9d97ab3c1b7c3cbf595bbd84386
973ab4300b68076c5c30a51c1f225bdd8a830933
112197 F20110330_AAAYAA rockwell_c_Page_058.jp2
9254b98689b5e4aadab092fffac0498c
20474b97b01eb6622192325e45e04373503179d3
93714 F20110330_AAAXUG rockwell_c_Page_013.jpg
c469e91c338e4b0213a5fe7a17ec23e0
bc2e6ffe59c09c8e2e86d37e6dc0953cf40ee225
49054 F20110330_AAAXTS rockwell_c_Page_105.pro
1e0d71f0592e8e695400b9ec721f6758
9fad0dcb6f0abe76324def36c52946e92ceba7c1
108854 F20110330_AAAYAB rockwell_c_Page_059.jp2
5667a4c47fb2417a584064157e473bdc
2518aa79a188e2c875570f6fd4c8e70e6b642bcb
102910 F20110330_AAAXUH rockwell_c_Page_014.jpg
75cc533f5f89ad6e1224402579c95280
4515b3f24f142a0ff6ed2e1d0aa56d6824f9b4b3
32780 F20110330_AAAXTT rockwell_c_Page_106.pro
38f7005ff2d4b2a5865d43aa3a2fc539
1f164b7c85f74a182d923b5bc7a6dfd735593f4a
114237 F20110330_AAAYAC rockwell_c_Page_060.jp2
4f173f758e6b2fdb12a46c3d463c3650
63c87f2debd6709c8ed24117d6a24be28179806a
93445 F20110330_AAAXUI rockwell_c_Page_015.jpg
a71f72760e553a5f2fc61f35a060e1ec
d4543677d92e160943fed29ab72d2b137fe2f9c0
27691 F20110330_AAAXTU rockwell_c_Page_001.jpg
42a8b0485324016083eea6737a25355b
e57645a585435c67d536050de89a6a419205ffd9
112838 F20110330_AAAYAD rockwell_c_Page_061.jp2
d7cd855b7176ab68b2b279b76b7034c9
8eb2d86a297b053d656a8c8b6d2ed740880012fe
103611 F20110330_AAAXUJ rockwell_c_Page_016.jpg
748bc1d8109a213be10c130534c5b6f8
0aaf431d2b4cb3f4de93f3dfad8d3d80cb1ae89a
4782 F20110330_AAAXTV rockwell_c_Page_002.jpg
1489a8da9b5aded5aaf51331fce25d9e
34425a4c00384ceb3a0354655e7484a10f3be87e
111597 F20110330_AAAYAE rockwell_c_Page_062.jp2
67004d748c6d1b3012a6cc90c1754e79
b3733c03c5365fc64fde10807c674e28278e35a5
100876 F20110330_AAAXUK rockwell_c_Page_017.jpg
ea2967a48f51559392d9884997f22860
1635642d568a08a3c1ad344e517415bf86809699
12501 F20110330_AAAXTW rockwell_c_Page_003.jpg
3670f7955089307f716ab95b3bcba296
691757750e26c2845b316a1420472df3f79f375a
108770 F20110330_AAAYAF rockwell_c_Page_063.jp2
b22d9dd205038b7a74e0704af59463ff
e6a027625c72bcc0e42bd33bd308b0c8e7d0a134
99104 F20110330_AAAXUL rockwell_c_Page_018.jpg
edd4045a48ef106453e7105d59be9686
bfacad3b5669aa8bb5acf6963c4b4465c0a57adb
91708 F20110330_AAAXTX rockwell_c_Page_004.jpg
c5f475d7877e06de7c7211af02ecb3b1
0118ad7d2a4f8dd9bc497c0805e60df8df986a20
109776 F20110330_AAAYAG rockwell_c_Page_064.jp2
005f164d60713998932b9b6d7a305080
47afdd976768ca5c9757918d8748db206d1220d3
99495 F20110330_AAAXVA rockwell_c_Page_033.jpg
5e13bda54bd9e8cd4b7e7657d6046078
fd671ba78b71fca21fc3da03c837c3036ccd6e6f
26370 F20110330_AAAXUM rockwell_c_Page_019.jpg
53c23c7ece593588e002e644e467ad31
777fb59bd4705cade0c0fcec274a463722c87da8
105380 F20110330_AAAXTY rockwell_c_Page_005.jpg
bf527890ee101adf3cb3d4efdfdb1310
8c2ec8cc19e0fa22b8b473db159a7f29d3629c09
91324 F20110330_AAAYAH rockwell_c_Page_065.jp2
995277f8dbaf29cbd6ad67c93f89a922
5d6a8df9ad033bf46c1640f57b33d5965ce7afe1
96825 F20110330_AAAXVB rockwell_c_Page_034.jpg
764ee0e58235587668faad6571d8de8e
e5118206835c0964f3c5de9e81fb59e48e609900
88955 F20110330_AAAXUN rockwell_c_Page_020.jpg
a09f06d22fd4756600d2a368c34eafcc
28dd171e798cc1be3edd070e1242c3099df73d26
59467 F20110330_AAAXTZ rockwell_c_Page_006.jpg
e0b7c39452d18362cccfe5ac3d0f8b56
bfb2397bf228de103b005f36581b1ca470dbba1f
100026 F20110330_AAAYAI rockwell_c_Page_066.jp2
9595e5a2a4d93786fc9476be28e181ee
e8606df1755c0bf33857dc0a16a67eedbaee9834
108058 F20110330_AAAXVC rockwell_c_Page_035.jpg
0277ae65046f053d3533f4978a0feaa6
0fef6347c68f3ad5ffa51a32d652d412cf71796c
104994 F20110330_AAAXUO rockwell_c_Page_021.jpg
c9fa1ba5e40d3d02aae59653abb5a8dd
5ff4ef1fd30c0190097bc079f278317d2a65aa1a
101265 F20110330_AAAYAJ rockwell_c_Page_067.jp2
5ea59cca676969c32b70bcd4e8195ec4
e7ba11e11633f270989bfcf81d9af0b82637d439
99257 F20110330_AAAXUP rockwell_c_Page_022.jpg
53f94af1c0e87d9a84e6d46fe74a857b
d5fcfee5d74dccc635799b7adbfc9591931e15d3
112617 F20110330_AAAYAK rockwell_c_Page_068.jp2
2b0dc3fb0df601af356a78eb48cad369
e06487af8ff6884b7b771be3ae9dc8bfd680c8a7
100848 F20110330_AAAXVD rockwell_c_Page_036.jpg
4ddcbd2cc4365539e84c05bb67ded69e
4dc0a9d174c41cb5e1b6aecc2aa4f472ffe46217
109413 F20110330_AAAXUQ rockwell_c_Page_023.jpg
5e81fc65d5e96df6564acf543f138b78
563f3c05660362fc3d4cb53ea8385e8c263dd4be
104910 F20110330_AAAYAL rockwell_c_Page_069.jp2
06d215755a9bf026cf96cedb33ef4669
81ef1c9c8044e7467d7e490ad1a24ffd0c3e90fe
105419 F20110330_AAAXVE rockwell_c_Page_037.jpg
4bf6248a71dce4f33113a68c481dbc97
6fbd6d7dd4621645b409b846bb056fd993d0f269
103052 F20110330_AAAXUR rockwell_c_Page_024.jpg
67da36224461032c696ee30aad70c82f
65019fb79d3b83d0777fdefff96e6ec2e326a5f9
611288 F20110330_AAAYBA rockwell_c_Page_084.jp2
bab7d0ae03c2fb475a3e3652c7e266e5
35aa9ae4943a753db2007fe35084375cc0d38075
104639 F20110330_AAAYAM rockwell_c_Page_070.jp2
406ee25716c2bebff884822d7b4fc2d8
e44727c55e9d742546df8ad128af4036d96c3750
102161 F20110330_AAAXVF rockwell_c_Page_038.jpg
4e277ed75536201aabbaf30b445d32d4
8e1d5d88a8e641960f981f6c293e67d0ca346a39
105786 F20110330_AAAXUS rockwell_c_Page_025.jpg
7271d07aaa39c1e4237696158cf8b90a
77b8743a958e909a71ad3b1502e1110c9d7ce9b8
669108 F20110330_AAAYBB rockwell_c_Page_085.jp2
c2460e2127f12f730cc689b542999eb9
731d65be809620b276bf2487690aef1cad4f2051
102232 F20110330_AAAXVG rockwell_c_Page_039.jpg
55ebab85018c3280f8c2c81d4bfe3191
73399c4da97824166d56525fca970a1d61d243f0
107469 F20110330_AAAXUT rockwell_c_Page_026.jpg
7c87e763672da6ee77ca45d8bda3e8d5
9b766f19dea8ef2d009088be1e6bdf10b075474d
822539 F20110330_AAAYBC rockwell_c_Page_086.jp2
0eeab93a1d4fb0c9b051274d7afcf3ed
82e4cd0c39b2a8e0430937e958cf7b26ff3d9c79
116564 F20110330_AAAYAN rockwell_c_Page_071.jp2
159f31d8072ead7907324ec9ffa28ffa
0572628540d9df0c590fd0dad539b65d35348d08
99932 F20110330_AAAXVH rockwell_c_Page_040.jpg
73e4a10e764414ac9f9b06165929cf57
ebf8e485b4d4a9f49a8f82357ed237bc3cf69e0e
105891 F20110330_AAAXUU rockwell_c_Page_027.jpg
8dda33c2bff3f1ccd9faeb3b60817123
6aabc4a7f5131362fbe9484f32fdda21a81eeec9
779352 F20110330_AAAYBD rockwell_c_Page_087.jp2
0c9593aba7114a55385fcb9f8554e6dc
b83ceab46748996a1c481bcdc4c1dabeed938efd
112633 F20110330_AAAYAO rockwell_c_Page_072.jp2
5106b79409521f743fe974e7f065ce82
cb9fce11fba37dcc6f1d914c61a0d3c5e08ecc56
100141 F20110330_AAAXVI rockwell_c_Page_041.jpg
24f3e5cffc004e960df5e514298c7d82
1ffe63e8ca2c4fdcb5a6a1ff5d31e5525dcebad7
107840 F20110330_AAAXUV rockwell_c_Page_028.jpg
4d29720d7ecfa13b34e312c65d7eeb7f
18e1f18ef8cc0e682a9e0e32be02cc153a7b6cf3
720511 F20110330_AAAYBE rockwell_c_Page_088.jp2
fc5954684bf316e56dcddd2cf25df620
0c884b9bb10d70df7b9d7efbb6627a051d218d24
107359 F20110330_AAAYAP rockwell_c_Page_073.jp2
6fe5a707a07b8fb4a86e280d5f96c6b5
0066184c41a1ec3a511dcf7ab1208b66df5b48b8
99859 F20110330_AAAXVJ rockwell_c_Page_042.jpg
87c6e873587ef231a3882d8505b7848e
0b2a9e4d8e16d6010b537b36fd19f071318bc12c
104840 F20110330_AAAXUW rockwell_c_Page_029.jpg
eaa3217eb1d0053bf83721e089bc21a1
97b6f799506deced7d49767871c00a9d8e9ac4d3
256858 F20110330_AAAYBF rockwell_c_Page_089.jp2
7d16257565c51041f0625127c03fa350
4039863085b86416f33702d314fd9c357a508f4a
117935 F20110330_AAAYAQ rockwell_c_Page_074.jp2
0e475a28e9204a49c7eaa766524845b9
0c51902692bf5b365c01e32d7e8e8d9737253373
102292 F20110330_AAAXVK rockwell_c_Page_043.jpg
16e498ff234279c2c1b7d43c213ff95a
804760251e0a8be68339e5a996cee6d2ea3258f1
104899 F20110330_AAAXUX rockwell_c_Page_030.jpg
c5b5eac6b68599a0f1fbecd5cf6f47e9
b700fce86930352b9d34f5edb4b54b4adbe1c725
99676 F20110330_AAAYBG rockwell_c_Page_090.jp2
e3ae7a9f7ad104536c1349f564d53514
e4989cf7ee7a79cfa56c84d120b60924bd3af1a0
102661 F20110330_AAAXWA rockwell_c_Page_059.jpg
e6428810259f08273544063ffd43b6c0
06cd806249660ad08781dab004cdf5e58a6c0608
110424 F20110330_AAAYAR rockwell_c_Page_075.jp2
186b8b651b417ea340d6e486bf3aa31f
e827628c7c85c0354e6ed884dce7026d5b23d9e2
108790 F20110330_AAAXVL rockwell_c_Page_044.jpg
bbaf26a77757047170e43efdb95aa513
1ee9113b25cda95cdbce7977099642e97ba305f4
105602 F20110330_AAAXUY rockwell_c_Page_031.jpg
57ae2a278c193863d1e54d12fa58aa97
0e9ce1949b4ed847bf364e6de14868389f94ce2c
111202 F20110330_AAAYBH rockwell_c_Page_091.jp2
e820a75aa6be906aec4587d0e6525e3f
ddc0309913fa0b3cf18f8b82fd56caf70524487a
104610 F20110330_AAAXWB rockwell_c_Page_060.jpg
0b25c794c3b941580d018dbf0f019189
4cf09e2ae6e993de8c1201089fb3df8333e04da8
112581 F20110330_AAAYAS rockwell_c_Page_076.jp2
2410eb9a473e396fc6df33b5cd047743
b39a097ef7a647b8cc6574e3892ba4150896d291
104353 F20110330_AAAXVM rockwell_c_Page_045.jpg
ebaf13a5a3787365a7f35e33ad3a78cd
4011b576f5917191c190d342b1cf7b501e41fb45
100673 F20110330_AAAXUZ rockwell_c_Page_032.jpg
2f3bb1e85b53c46c9c43f39c1f634978
98780ff9e6275403645c6c6701b42e8bab60d946
29213 F20110330_AAAYBI rockwell_c_Page_092.jp2
f5dbde133d50475c2be58aafca2e498d
c4fc539120356490fc40fc18202b3d54dfc72ff1
106442 F20110330_AAAXWC rockwell_c_Page_061.jpg
b971b6fa0c082fa9e6af9cce392107ca
d71f4755cadb4ac9a94f17b4c871b6b7cfa3c449
114247 F20110330_AAAYAT rockwell_c_Page_077.jp2
2ba085356483690165d5ae3737062c32
38d8a775c14073459b9523c6989a9d494548c2eb
104730 F20110330_AAAXVN rockwell_c_Page_046.jpg
993e8bd5dd39afb283767cd8c11c0c2b
563bb8992afb38eefc412b6f614f748c4d4ee494
1051984 F20110330_AAAYBJ rockwell_c_Page_093.jp2
f7e464967f779751008b5f2c732ecfb2
b61d829c95da4aca8871eaf40d5212b95cbbe599
104475 F20110330_AAAXWD rockwell_c_Page_062.jpg
80517433d1e133d4e36c5bc14c929027
5512f81c3e0e94d1a0f386227a1e5cb987f757fa
16057 F20110330_AAAYAU rockwell_c_Page_078.jp2
b803446d47d13388986149d69f6a1664
412b9c3b20203d92bac51d9da799c31ab37593a9
104436 F20110330_AAAXVO rockwell_c_Page_047.jpg
4767fffb70db6c628f470e39365dc88f
21903d55a084428ae21786042ead24312818e6ac
134009 F20110330_AAAYBK rockwell_c_Page_094.jp2
e9dd6ae4caf9ed260f5b41f6f9f59d5c
69303ba159e626df58452f63e71013188a9b2c0f
109803 F20110330_AAAYAV rockwell_c_Page_079.jp2
7d5036a7e0bc48eb7800559d1489b941
9ad55d4e86b554886d987c50592691b6ad6fbde9
101713 F20110330_AAAXVP rockwell_c_Page_048.jpg
8b32627c946feb8c79d22baacc362769
24f3817f89646eefcb3b0caa93975a9b4752793c
131498 F20110330_AAAYBL rockwell_c_Page_095.jp2
de86446388932fd20ace9f19aabe4992
4b8cd7bf022ad16cd04ce6388540d444525e5ccd
101782 F20110330_AAAXWE rockwell_c_Page_063.jpg
460bdc27e04e2e1cb7aba2f52f926a82
4e4cf1a16991fdf73eea2fe8b08ea773cfcfa36e
66227 F20110330_AAAYAW rockwell_c_Page_080.jp2
c9b1ff9bf6fe9712585de52122696bd6
f50c71934e7a44c78654319263b14f803b9fcc3a
109585 F20110330_AAAXVQ rockwell_c_Page_049.jpg
6907794d92590b06acec60df03399307
5c368fbf3cfc3f2ff29cbc471bb840c927abf497
127356 F20110330_AAAYBM rockwell_c_Page_096.jp2
4f351729089398dc10728390accacb65
8d7fe55fe25b3d1cfa878eaac6db1cf16176de1b
102531 F20110330_AAAXWF rockwell_c_Page_064.jpg
5969f3ad5a5afcf989cfd1e184a0c7e5
ee01e6cfadb56103f1104b15dbfb85cb7e8972cc
27602 F20110330_AAAYAX rockwell_c_Page_081.jp2
b46f00e77e9a74942c2f1e78725d56b2
cc2794fd4037561999673ef2d0b5031710de3df3
101642 F20110330_AAAXVR rockwell_c_Page_050.jpg
0df8e0c367ec3c3e11df0bfe63059db9
ec275d69769ff8aadc1b1f9af8c339d7f19d365f
31713 F20110330_AAAYCA rockwell_c_Page_018.QC.jpg
d031618f5fb714f8ce1668da29f9adb7
81d6a78358885d28e50bc2b99366ef79340eb548
138272 F20110330_AAAYBN rockwell_c_Page_097.jp2
f75bb7c0864af4cf954b1880888d3a99
60ee02d83503c30ca9ea1a6f89fcf929f996384a
84734 F20110330_AAAXWG rockwell_c_Page_065.jpg
2a45f28d0cd9320804557cc08d23bb2a
df993f746cb4a5c652c20cdd2949421308ccf43e
61280 F20110330_AAAYAY rockwell_c_Page_082.jp2
1c2c372c77899c3fa0e373dfc766b89c
9a723096ab38842bba76bbb5ebd310253c437cd7
F20110330_AAAXVS rockwell_c_Page_051.jpg
adbf76edca1f766bc7e13becc0299a24
0a0837376e16f3b1f2c15b235c2151c11e97c66d
31681 F20110330_AAAYCB rockwell_c_Page_101.QC.jpg
0f9b12add8e725f3a27b04f632ee396f
a76ae2d0bdbbfd042301138ffbd33c2a2ee261f8
93862 F20110330_AAAXWH rockwell_c_Page_066.jpg
3fda2348581278859f75e1764b100a47
6a61e059058ee19fd61be7ad7d5f7ff1001a80c0
62760 F20110330_AAAYAZ rockwell_c_Page_083.jp2
f61ccb9b737b81b31a88fededf2f4a65
5f86a5e5cf671078abb72d2dd19e48ef392fde31
104104 F20110330_AAAXVT rockwell_c_Page_052.jpg
067f027dd3b06b758a3ab74858d3155a
8113bdf47767d1344190090a94c50f7ab6ee4449
8110 F20110330_AAAYCC rockwell_c_Page_101thm.jpg
8618d921d2fce8c98649783c5b26bf64
e67064d20c8e541031297f41905f409ec48ffdfa
122791 F20110330_AAAYBO rockwell_c_Page_098.jp2
37a911d25172c6c1c1a43d96bc1a9f58
9be0740b2ab66d26dfa4b2addb5bcd20e7aaabc1
94349 F20110330_AAAXWI rockwell_c_Page_067.jpg
3e9d654c0e58d7453251292b2c784880
4bd5e26e23f453bb63223064d55d18144391f753
36241 F20110330_AAAXVU rockwell_c_Page_053.jpg
2a58eaebe4ad13c91ddd0dabe1e1e9dc
06ed60d0299b98d1831a250d3ce5c09c5ed9850b
33201 F20110330_AAAYCD rockwell_c_Page_024.QC.jpg
33c165ed260322164fc980301796f9d3
77ad0fdc9de401220e50dd397d03263d65f4d774
126434 F20110330_AAAYBP rockwell_c_Page_100.jp2
b3b686d0de980d95621e83ec97565a86
6b22a7883405ab8d017e2ccb381cfac8dc56c12a
105231 F20110330_AAAXWJ rockwell_c_Page_068.jpg
c7cc9be9587b178fafa2a28b86ed9ac5
6a627704e41232b2a7cf3844a8e299db4c65f7a9
52986 F20110330_AAAXVV rockwell_c_Page_054.jpg
d3b4d4c4ef6ce94d0d49772046782f7f
2ef902fc3d9126aa724e5ff6000c2d2de616e1ee
33461 F20110330_AAAYCE rockwell_c_Page_063.QC.jpg
926c1e5fbc785aa6e6ffad32739fab19
fb21db881586acddfd80e3c7453809488fa342f2
116575 F20110330_AAAYBQ rockwell_c_Page_101.jp2
300182491beea22b7ca113e2d1f05ed5
81d0a065dc5efc327959ca637e86ba3fc1455048
98401 F20110330_AAAXWK rockwell_c_Page_069.jpg
4f64102127d0815570380b07dfc4900d
c19d8a0cd6c46d6247f2c6b1d5a7ffaa8d2c01b0
64247 F20110330_AAAXVW rockwell_c_Page_055.jpg
6d6a7292c3675bc8c7e4fa695f4c5b47
dc492fc8bcefa504764c63501ef373858e485be8
34733 F20110330_AAAYCF rockwell_c_Page_061.QC.jpg
3f8ba33fa4585f32ddce98965105bc68
f637274df2767400fc88286cb7f900894b7983c7
34925 F20110330_AAAXXA rockwell_c_Page_085.jpg
fe736651bb4ee0a51a85811df291362c
e5f1ee174c0e6653ae2d3e3d921f999c003969e2
138743 F20110330_AAAYBR rockwell_c_Page_102.jp2
c07b9d7b3e6be1dbab6eda59508da03f
c7f8313aca53cab032c44dd6486d7e026c0f2be3
96920 F20110330_AAAXWL rockwell_c_Page_070.jpg
c793cc9bbd2e78b36e742c6f9b5c8a04
3bfd3d515bac64821a96dce9c8be5f0ff681670b
91187 F20110330_AAAXVX rockwell_c_Page_056.jpg
39a6d513b2443b142780cd099d002800
47eeff6fe07b188f3385773b2842fcc845696a71
32719 F20110330_AAAYCG rockwell_c_Page_064.QC.jpg
00cf127fdfde77ca20858c9f7439e4da
af0e522316661a57c64c4d06d57ca609f41757c3
38909 F20110330_AAAXXB rockwell_c_Page_087.jpg
58294aeec86be56e6505b9d4de498fc5
bd9fe2fefe7272e5199016ce09eaf1401d78fef6
131942 F20110330_AAAYBS rockwell_c_Page_103.jp2
19ba713c6a5e706de89d5b3935f880bc
a42e20f13e356a6b3de302b2e823f1d53d584401
108210 F20110330_AAAXWM rockwell_c_Page_071.jpg
c563555c45bdda2cc939b4dae3889172
bbde6cdc8a848c549faa0be4b3417b736582638b
99504 F20110330_AAAXVY rockwell_c_Page_057.jpg
e114ce2e73d9b00e5cf27be898cb3fa2
2ec4ca043e8eaba8fef76f6f8f8d99b1dbad132c
7232 F20110330_AAAYCH rockwell_c_Page_079thm.jpg
59deb29ab0997986f49bba514c4d6581
8fca3c9774d8279c4dfe77664c8dbbf035d310fd
35771 F20110330_AAAXXC rockwell_c_Page_088.jpg
cdb5716ae997dc772d962b9385221ec0
cf1d30de50f0efca82a3df111a5916137f3103a1
129236 F20110330_AAAYBT rockwell_c_Page_104.jp2
db6feb842e9455cc024ce93527f28537
73b545a2b2d84ed4197f4f542a4f2dce56de14bd
104276 F20110330_AAAXWN rockwell_c_Page_072.jpg
c4012962ccd90dcf760626bed645c1d7
516720fbb8427909b5a077a2c6e4ed2c2bfd76bc
102782 F20110330_AAAXVZ rockwell_c_Page_058.jpg
a44b8c510bf4fd4ff44db9f37b91d554
dedf892fe0557959d9c4d4dc108fb12dbc943d1c
8522 F20110330_AAAYCI rockwell_c_Page_014thm.jpg
0b98f864c9b99a8dabfa50d336fbc06e
85004a345361872653c64abbb898af9bfd64c8a1
26826 F20110330_AAAXXD rockwell_c_Page_089.jpg
c753d2cb47a236877537bbae77add8de
7d1efccee81af2eaabe74777db00655030c12492
104738 F20110330_AAAYBU rockwell_c_Page_105.jp2
0db97736713fc9c170cd82313d5f8f53
50ceaa495f72324e00139a005eed4e8b8c1b72d2
100936 F20110330_AAAXWO rockwell_c_Page_073.jpg
4618793cee94cb6a429e1ace0be73667
c5632862d68815cc970246fcea64503caca7339b
8313 F20110330_AAAYCJ rockwell_c_Page_061thm.jpg
39d8d6735f45dc9b39a579e96576b8ba
867ad23e0a44acf6d6c917bebe1af177d0e25dd2
92047 F20110330_AAAXXE rockwell_c_Page_090.jpg
9fd63f2454621be72695190bdc37e5d7
78fa580a7fe5882a91e27ce2218e47fd3288c486
74322 F20110330_AAAYBV rockwell_c_Page_106.jp2
45954e01bcbf3658bf8313a1ad51f9ef
a37bb25b1f1d6f0ce816208be1cc6ddd60f97a31
110707 F20110330_AAAXWP rockwell_c_Page_074.jpg
0727c737318d0b4c8fa1d1529a00dced
d452c241632eb92d381ce054e72a5a53baae7de9
34634 F20110330_AAAYCK rockwell_c_Page_045.QC.jpg
c2b0cef0c7fe3215509075965f3826b7
b39a497d70ee063d7ffb6cf78bfe318316409b11
29892 F20110330_AAAYBW rockwell_c_Page_067.QC.jpg
7b89a48f4c7ae65e7903c86ed1db8580
a29c6107de5a9832107f06b8efc8539f12b0203f
103063 F20110330_AAAXWQ rockwell_c_Page_075.jpg
32e688d514551a648f1a621adb1b3bc5
c5db580712e100b7b3aaf86db14d2d54a38a0be6
3780 F20110330_AAAYCL rockwell_c_Page_003.QC.jpg
2c09051e4a05e8ac5ce559968da9bdec
2727b187a478fb3b1efea795402c55a748b4412b
103144 F20110330_AAAXXF rockwell_c_Page_091.jpg
40f65bed9da1389e21a98d9558c07cea
0abf9b0680441c24f0c9e9fd564b5fb8634b5229
4401 F20110330_AAAYBX rockwell_c_Page_080thm.jpg
01965edb2e53ec77ef52a1508965d261
70255c7622ac5b360f5f65416da9782b032c8bf0
104972 F20110330_AAAXWR rockwell_c_Page_076.jpg
48d557ddbb54639fb94cf1c3bd021d72
d26bc312ed4fc02529ec033f3cc76ab9298386cb
3316 F20110330_AAAYDA rockwell_c_Page_088thm.jpg
3e306ed7d7508522b94166e58f6a62ed
093e4272fa9ef1ecf7d8c3bd845c5f27aa04c37b
8723 F20110330_AAAYCM rockwell_c_Page_103thm.jpg
fa7c2a8f75aa9d794d6db96e73b91cbc
2236b66dbabc3c71262979a3b991ee801e8e1ea4
26711 F20110330_AAAXXG rockwell_c_Page_092.jpg
a802f63b219d441b168837c85ed2cd18
cba7226a0d71a55a758b452b158cb7cd2112dde0
8592 F20110330_AAAYBY rockwell_c_Page_026thm.jpg
3cefae1f5657ace7655ddfafd8018a2c
01099dbe9b9a2a894ce83886a3dcd3e0f532cca1
107538 F20110330_AAAXWS rockwell_c_Page_077.jpg
cad91a9bb27fbb02db4d988373f5fa71
4332d7a7943eee4f0babc7fccf81b854aa88df14
35736 F20110330_AAAYDB rockwell_c_Page_049.QC.jpg
9d8e3e4386d70789c8718e32ded1a31d
6eca2ce313114ebb0cae5d6202772bdc48121039
7430 F20110330_AAAYCN rockwell_c_Page_015thm.jpg
1a0ff993dc80137c65e02c8cac0a7249
c47058027b078be9507ca84d7fe4112ddc74323d
108493 F20110330_AAAXXH rockwell_c_Page_093.jpg
af0c44a7b571b81092ddd0a23b00e61e
c0e24b0bb2406a0c1488b512279467aeadc675b0
33682 F20110330_AAAYBZ rockwell_c_Page_051.QC.jpg
2a40a9818252625adb50360cb7ee3d65
1563b83fdfa43f39773f28c35f31f2d0551a5f18
14809 F20110330_AAAXWT rockwell_c_Page_078.jpg
e62fe00d466f631e4b78ecd7b516eb6c
e6a2ef5751176752eb02909f7647d7e0dc156dd8
36114 F20110330_AAAYDC rockwell_c_Page_099.QC.jpg
368c1a03101a6c60f09c8c06e2dd863c
a91885968129c428b8ea3ea7f12417fba76da8f1
7800 F20110330_AAAYCO rockwell_c_Page_070thm.jpg
9e232c104190717078421b723c8f11b6
648e305f7121f11c3a04fbe66633296648d5fc86
127006 F20110330_AAAXXI rockwell_c_Page_094.jpg
529fd223043fa6af93981eeb12dc3775
3307d97d9e1192b302b82744cc04511ccbb4d821
95321 F20110330_AAAXWU rockwell_c_Page_079.jpg
b7816b95ccadad8d3920062d7c1867b4
75fdf3687d9c405311c197fdf473a93f5d65bfe4
31716 F20110330_AAAYDD rockwell_c_Page_070.QC.jpg
6b728a6c443a2c895cace5b45454d643
3cf55dc7faefa3da8274d97506344648d7732700
120673 F20110330_AAAXXJ rockwell_c_Page_095.jpg
dfe157d49958e67a38729f631f09ce06
8bcbf29db952409e1401bcdabb8a904071e98bd0
57965 F20110330_AAAXWV rockwell_c_Page_080.jpg
11e5169d081595eaf459ac6aa763a91f
14afedbefbc8c23c1f585c5b5390169990c051d8
7737 F20110330_AAAYDE rockwell_c_Page_034thm.jpg
cfb78d877546be9957ef3e6a61ae689f
930422febba21b410558b99c206403152b5ea849
32434 F20110330_AAAYCP rockwell_c_Page_069.QC.jpg
e2687612fc8f48e24c7bfedab235aab1
aeff1ab1bbcd9cfb07b8d25a368bdd5c8fc07795
117423 F20110330_AAAXXK rockwell_c_Page_096.jpg
dfd263e5fc6896c48850c321b248aeb6
5a64edc8d2330ba0ac23a6cb7460e32e8e0aa45e
26272 F20110330_AAAXWW rockwell_c_Page_081.jpg
a4a8f1a493e58cc7c35aa1440644202b
0a253c31856c6fe817b0ee92f89cf05449d3c94f
27649 F20110330_AAAYDF rockwell_c_Page_105.QC.jpg
d6fdfc3d8cbaaebc19cb15b0ce6e96f7
52e71d56a9260b7cb85b24a10cc98ffec0445cc9
7180 F20110330_AAAYCQ rockwell_c_Page_020thm.jpg
c866268cbd044fab04690c8478cd88ab
86130bcad62439a1499e412340623cc17fb6f42e
130592 F20110330_AAAXXL rockwell_c_Page_097.jpg
84016903ef9cf6663488801f20198a3f
b0ea21542a3c9fb891fe572031e93a05a6491145
56803 F20110330_AAAXWX rockwell_c_Page_082.jpg
c2d3bc2b013cb9dbefca77d3fa6d264c
d494ea74588f8efedc5685100adba518c5a66586
25275 F20110330_AAAYDG rockwell_c_Page_007.QC.jpg
eed05f3ccff817cb3d0a35922b7b95b5
0e867861c637fa3c6f4369580081b887ab79c342
61991 F20110330_AAAXYA rockwell_c_Page_006.jp2
20cad8b8e76544015e4c875ff51ddcb7
5b3fd4bf5a500402e0d3c9dd20ed4e8fe45ca4ea
36604 F20110330_AAAYCR rockwell_c_Page_102.QC.jpg
03aea77f1bdc144048f70267ddfb23b2
eaf7a50fd0681cb9c5acfbf550ee693c43ec90ba
118300 F20110330_AAAXXM rockwell_c_Page_098.jpg
88fd5d219a31b1f331b8ef0320252c0b
3016a64a70ce0c34ca1daf943ec042773e695cef
62549 F20110330_AAAXWY rockwell_c_Page_083.jpg
8feb1551ef0f91227bbda7ef883e0b2e
81a77f8fe8228a04223c5cbb4bd00f8d331ab34e
33974 F20110330_AAAYDH rockwell_c_Page_059.QC.jpg
f4f878bc45be4ea91b9eceacd81eb614
719ad190681ef0b36fc3851612bcaf61d1b0fc91
F20110330_AAAXYB rockwell_c_Page_007.jp2
6593dd8488aad89fa249b6d8193b5ca7
a1cde12d598f38a19dcf0dbf2f349d386b8ba463
4597 F20110330_AAAYCS rockwell_c_Page_009thm.jpg
d4da7425e54c4b699a4b61c65e7665e6
4d87e5219d074ba8374ed54102a75877027cd66f
130218 F20110330_AAAXXN rockwell_c_Page_099.jpg
6228d0e10f2be8700b41e55ed6883200
209e5d2bb3c4cddc00ba31396cbaef37ea279541
49232 F20110330_AAAXWZ rockwell_c_Page_084.jpg
a18335ffe8720a8e3b46d749597be72f
685309eba9525d91dde47cb34458ad1c1c45a2f6
8458 F20110330_AAAYDI rockwell_c_Page_076thm.jpg
165b893d29e6c4ee25563e13db7a5ee3
b7818c1aefcf2dcf7b704b42722a1a849e4ce90f
1051982 F20110330_AAAXYC rockwell_c_Page_008.jp2
2f4ddd25d22144154d7a95128fedbee6
83b3177e13f39621257d24eb41ba52dd95b9043a
33006 F20110330_AAAYCT rockwell_c_Page_048.QC.jpg
818237936a4e08ea8defa8d45c4cb756
bf9e8269e37e4a672d9b4dbbd7bada3939fe73fe
123235 F20110330_AAAXXO rockwell_c_Page_100.jpg
e4a28bf25cfab0009d72093bc8663d22
cb32b4371e26cfc93b4ca6494d45d863de4416f0
34399 F20110330_AAAYDJ rockwell_c_Page_060.QC.jpg
132b43d7973e97e723567f59131680b6
cf711c6a6f24411bce53bdf74dc8bd7a1acb30ef
1051959 F20110330_AAAXYD rockwell_c_Page_009.jp2
00e3fee31af19cfe45a3cc33c987e997
c7d8f531f3dd3b2f54fed5b25e1d0e9567a3b316
8179 F20110330_AAAYCU rockwell_c_Page_043thm.jpg
18d3293e2012d117cc946e2a8d6ff1d6
24a5094ba4657de8885b76ce34990ff404565bdd
109727 F20110330_AAAXXP rockwell_c_Page_101.jpg
8ea714b066173190b0523ea6b9307cf3
bdfed0b16b3b6f6caba1f9aa487effd774ae723b
8243 F20110330_AAAYDK rockwell_c_Page_045thm.jpg
88aa70878c34c3d915b8ed4c6da4f220
1d262019aa44efc34bf439d2e5aa85e594ea9278
1051962 F20110330_AAAXYE rockwell_c_Page_010.jp2
99403890da670e36cfad28cd0aa3e7cc
f9f4b8c1a30a83f1e589d9817c63fea8b6beda8c
4301 F20110330_AAAYCV rockwell_c_Page_083thm.jpg
8c4ec937e819074ada56583ad7aa1f6b
408d092ffd517630b243fb2f83915846ca8e1963
131306 F20110330_AAAXXQ rockwell_c_Page_102.jpg
c848128978ead2f02e5297940f224add
8a2c17387eef1bb354bc9fa4a0ef71f9cf383659
8662 F20110330_AAAYDL rockwell_c_Page_005thm.jpg
46f0bdfedbcb687ce6b8ae2212fcc25a
7e7189f7a13abcda8818178d8592d45071720c89
88763 F20110330_AAAXYF rockwell_c_Page_011.jp2
3f7a60930b7299d3162e6a57aa5fb55c
7169669de4893c34ebc00915db4745763e9832c9
33750 F20110330_AAAYCW rockwell_c_Page_029.QC.jpg
2e6694c35796099e84706dbfc5d7e5e5
7d96d6515d69603f6eba52d10d5e7d9c4e0aa079
130292 F20110330_AAAXXR rockwell_c_Page_103.jpg
f120bb2d5a8c784bf48f84173f78361e
2107d8b708bf61d319d9e3fb58b81bbc8aed36b4
7627 F20110330_AAAYEA rockwell_c_Page_013thm.jpg
9ae6c6f0e5f99d207c605fcb4e58cf4c
1eebe2e2a2cd084cfef1b8060858be1bcafd60e5
33726 F20110330_AAAYDM rockwell_c_Page_047.QC.jpg
60cf93abe10f9f9bc999a641e4658a96
9f4427175de178875c55d30059a3cecf63a55071
4653 F20110330_AAAYCX rockwell_c_Page_082thm.jpg
b7838df486f7aee5ae2d03b8ab8a5f63
f9114ba91f5505f87ab2394de17f44599c99ec0a
123185 F20110330_AAAXXS rockwell_c_Page_104.jpg
7042df78d99420c9f6710dcf327e2054
e16a10a1c65608727078dc69b4fea661abe63cda
8338 F20110330_AAAYEB rockwell_c_Page_072thm.jpg
0f6701c01ddfd3c622171cfe5a779dd6
6bec9248ec8205b8cf7398363f2b51cc15298e5c
32424 F20110330_AAAYDN rockwell_c_Page_017.QC.jpg
88ba53bab622fbdf0040e6d185d66051
fd16c22d9967e0f96b42dabb0e0740808760cd09
91373 F20110330_AAAXYG rockwell_c_Page_012.jp2
271b5d2319d9e2f773096b9ebbeb880b
983f3f5ee815606d742e8f4359fd18605fafa553
7455 F20110330_AAAYCY rockwell_c_Page_004thm.jpg
cc57251ab9a55cc136e5a23ce1430230
f5de55195c5a8dd1d6039af7096eb3abd4653f4a
98143 F20110330_AAAXXT rockwell_c_Page_105.jpg
65aff0eef8a97fd507e42bac8e4a2528
6a5ff9f34459587fac54422ea7fb5ae8e7ff0935
173262 F20110330_AAAYEC UFE0012840_00001.xml
2a2fadc91d843e3b7ba23a7d7714a688
6c5822d6b69a78c37b2916620e8c2d91496147d5
rockwell_c_Page_001.tif
2405 F20110330_AAAYDO rockwell_c_Page_001thm.jpg
cfba2b8960031ad8fc3ecce4da642273
381f7f191b13166aa748c4d47573c92f912b92f3
100577 F20110330_AAAXYH rockwell_c_Page_013.jp2
40cf668a5840822b09ca5ae3ce7e3fb2
7000f1080583594341e67bad62f0a1607052dd12
34682 F20110330_AAAYCZ rockwell_c_Page_094.QC.jpg
f952a41e422d01bc3c83d7b4c05fc364
e458b738fbc392f5ec59d3f76bed71cd35358e72
8698 F20110330_AAAYED rockwell_c_Page_001.QC.jpg
32b9eeb4c02d481a95801436f8e81a8d
7a00a214cd88ec0a7818330c1282157a2cb2fdf5
7394 F20110330_AAAYDP rockwell_c_Page_090thm.jpg
678fe5690c21b7654577fbad0ad64f63
dd21bb15325ae5517be20aa50a757a5ec737efe3
110521 F20110330_AAAXYI rockwell_c_Page_014.jp2
91f15ed02cc7f694f170c7b5cd4e6ef6
5127f4645aefa783c3c29d1acb6c908b4081d0fc
70890 F20110330_AAAXXU rockwell_c_Page_106.jpg
85b705f21bb9858ee7f9fc7a5bbfebf6
45c53118b62b93790aea23d28f71624796e1d269
1590 F20110330_AAAYEE rockwell_c_Page_002.QC.jpg
f7874fe70294586d20932b4b16e2967e
7f4992f215e7d60b44d1b025c5f2a092263ce602
101322 F20110330_AAAXYJ rockwell_c_Page_015.jp2
e08e187a9b7fad21b1b799cc1b4db0da
6c306eeb7aedc6d4b21f959ee27ae8782d291ff9
25581 F20110330_AAAXXV rockwell_c_Page_001.jp2
5934215dd87a164150501fa11f42c5ef
d6a54b07f746557807b554850062fcca0f2fdd8f
30194 F20110330_AAAYEF rockwell_c_Page_004.QC.jpg
922f184be0bda085edad8ba43274382d
d98e3180198173e823c298aaac3f449b633430fe
32991 F20110330_AAAYDQ rockwell_c_Page_016.QC.jpg
7e61e545c6b3597dfbba5d906c4a05a9
c0c9429e82d95e7bed681c62f4a4e237aecce311
110126 F20110330_AAAXYK rockwell_c_Page_016.jp2
38d6aa77188534ee30151954b491a06a
a84294659fd0548ae4d36870158051ed6f60d5ba
5670 F20110330_AAAXXW rockwell_c_Page_002.jp2
ea4a38215683093c8355250fbd1ab14d
be132484f8c1f51d1e44d4a664847e57d51f9326
34213 F20110330_AAAYEG rockwell_c_Page_005.QC.jpg
825e4fe86d0e28a5fe9947b473df8888
e5de19b9b4ac514f9b411ea35ae777f102957281
107570 F20110330_AAAXZA rockwell_c_Page_032.jp2
4858a4f8a680efbfcabe44777a03ba1b
6ac88717837d22e2460f83b70b363f2ff1f9284d
8405 F20110330_AAAYDR rockwell_c_Page_031thm.jpg
05865e69f1d39668332186587e92a519
20beee95cc7cbc582d4d3014e9acbcb44a2753c4
106866 F20110330_AAAXYL rockwell_c_Page_017.jp2
5fd7bd2e8d8bf31bdbb03fa6f5242ef2
86cacbee6d2d86005a64818bf8d786845aa444d7
13881 F20110330_AAAXXX rockwell_c_Page_003.jp2
91ef9c936a1003d8578dc874a93f7214
e723d26a110ab55cd00516dd2ae551e9c29d17a4
19103 F20110330_AAAYEH rockwell_c_Page_006.QC.jpg
994bdba9f404e660ae6d50bd24f1eaa0
3e0418f855af0373066ca582387676f08cacd3e7
107901 F20110330_AAAXZB rockwell_c_Page_033.jp2
cac1857e7043073d5f55f86eb5bb2afb
42eff8214735f7f9cb9fa87c7eceea61772ce07b
8053 F20110330_AAAYDS rockwell_c_Page_042thm.jpg
a0d38e97aa35a399f3ccb38649e312fe
e601aab5285127bbbebe90087362c291b2d81791
104652 F20110330_AAAXYM rockwell_c_Page_018.jp2
4c6ff75ab99f2d31d9e57557c0e230c7
4cda369a3311dbe6576954215c00c7ecd5e11cb6
97210 F20110330_AAAXXY rockwell_c_Page_004.jp2
cad83d07ff698b1f24e5a20558df3655
00d00fb8c583d9ffb372e82394cb2d3e74491697
18544 F20110330_AAAYEI rockwell_c_Page_008.QC.jpg
010dcc87005d9c430d33825d3c2a7e31
fd6018d4907de65945e663acbea54a8a2a2cbabd
104633 F20110330_AAAXZC rockwell_c_Page_034.jp2
fcc946082257654b46cb195c1a2241ee
fd3bc04d46049d80dbc16d0bd675ee22b696fb99
8514 F20110330_AAAYDT rockwell_c_Page_047thm.jpg
961ae995d9cb2bf13eac9920b8ec5ac6
3af45229694959485bc6c8cec98059b0ed63eaee
28989 F20110330_AAAXYN rockwell_c_Page_019.jp2
3fa7641b2e69080e6631bee62dcff984
4e7e7a36a52003fe0607d28c00ef9923e4a6b7ed
111970 F20110330_AAAXXZ rockwell_c_Page_005.jp2
efc37c28b8ad8f08af3176f92a1a113e
dd7e297b3f142238177fd56771c67f9e562df49d
18457 F20110330_AAAYEJ rockwell_c_Page_009.QC.jpg
91851fe18c78af90f43de3d32f95b6c9
b7f018885f63a24f63b9a52b4270e1240d3369ae
116050 F20110330_AAAXZD rockwell_c_Page_035.jp2
8e97f459e4d032fedfd671ab2aa36d12
2fb4e6dc30e06928e4509fe6e828e13adaefe85c
35016 F20110330_AAAYDU rockwell_c_Page_104.QC.jpg
1c016b20c1692144aec9287b0f233ad2
16eca60c7900e349a33d16e6c3b5eae4cc621e6b
94532 F20110330_AAAXYO rockwell_c_Page_020.jp2
05afef5baadaea59e0fb67e8c7d2dd06
722249adf8b6ecd36e5ce162ada55fa75ad04485
16591 F20110330_AAAYEK rockwell_c_Page_010.QC.jpg
6f6c20f0bb48582b701c16d5983bed58
4e1aa286f63a45fdf02630bc7a2cbaa79ca945aa
109580 F20110330_AAAXZE rockwell_c_Page_036.jp2
9c21e028a49f487f2a9ff49d33e6db13
4acc9efa4afe5fa1dd2e922c5781e297a84a7773
36469 F20110330_AAAYDV rockwell_c_Page_103.QC.jpg
d89b82996d738aaa4018257f809dec9f
440b716e4f10a935d800fb9d4d89b7264887ae2b
113345 F20110330_AAAXYP rockwell_c_Page_021.jp2
92708b33f9e39413f479f38db1c43914
c5782f26eb7f351928272ae66d5fa210a0c150c0
25849 F20110330_AAAYEL rockwell_c_Page_011.QC.jpg
1941a3936fc9e995369ced0ec3594cfd
26cbbf1edbdd2c16d6127ee494d52fbb4893e097
113278 F20110330_AAAXZF rockwell_c_Page_037.jp2
e51c6ba8942a239b65c38a22efb0a10a
fe4e5ea45ff10aadf76ee21f839791895e805415
8373 F20110330_AAAYDW rockwell_c_Page_016thm.jpg
1438f7d04f47c07d71e0c70d70abc8be
d2c2128c136b4e8e1d2d25dc4afc0bda8fa2b558
103778 F20110330_AAAXYQ rockwell_c_Page_022.jp2
4b1616037187e29fb02ec1b7f54e41a6
2d460184702bf6b9d3e964a158a6e916145e711e
27173 F20110330_AAAYEM rockwell_c_Page_012.QC.jpg
18378780ca8669ba2611166257b68cf5
24e8096a7566d906239dab6424a65f6cd74ec344
109376 F20110330_AAAXZG rockwell_c_Page_038.jp2
d372111a69905387dbf9be918efdab53
4c08cbcacdf4ad4b8b9eaea9b820e29836a5047c
17264 F20110330_AAAYDX rockwell_c_Page_055.QC.jpg
ddfc15d654ab530c2726dd769db1095c
fe1862fa6e3ee1aaac1374bc10da7456dc3e538a
116619 F20110330_AAAXYR rockwell_c_Page_023.jp2
286d5ce70e00a9254b77bf17c35b0240
dabd2fdce779e0837e6964e405fda1bdae64bd94
32042 F20110330_AAAYFA rockwell_c_Page_032.QC.jpg
f167246eaef96b75ef6a6f9672add6c3
b425d036be1d82438f0425aeb775bf5a5f7c30f4
30343 F20110330_AAAYEN rockwell_c_Page_013.QC.jpg
39c96329c3c5527b78e95979c54430e5
fc39eb8f44e8bb37f0c5dfd65749774d1aa5ef0c
34566 F20110330_AAAYDY rockwell_c_Page_025.QC.jpg
6b5b58f9f6f1a13325c36053df4fba06
3d77f88d398b0a0b6027b74cb7cffec8bcf076cc
110406 F20110330_AAAXYS rockwell_c_Page_024.jp2
f54c3dc9b384244079e8879516a7d90b
77699bfb428b8f90b184b303481e586ab07395c3
32336 F20110330_AAAYFB rockwell_c_Page_033.QC.jpg
bc51a1488138b75269f493410e0a3b0b
a99907022bea5ba4290811c1470a9c7f63fddcb8
33060 F20110330_AAAYEO rockwell_c_Page_014.QC.jpg
17ba787d8822b0ada993aea31741cff1
a094d6d2413223b88fbe3ff1fa9fbb0f618e415d
109934 F20110330_AAAXZH rockwell_c_Page_039.jp2
cad093538364d848775b216f11734a60
7b2856fec4230bd3c15c7e7a4bf4b41bc95de135
7953 F20110330_AAAYDZ rockwell_c_Page_018thm.jpg
ee015425b6233b8c7f4803ec396a7b7c
809ac2dcdad4e24f4c137d453773fdf715ac2e35
113001 F20110330_AAAXYT rockwell_c_Page_025.jp2
7bf46944fdecc9139804613d179e4587
a565c62d793132dee935290bd34a99864158de26
31690 F20110330_AAAYFC rockwell_c_Page_034.QC.jpg
bc3964050b6efe1ee1ea4bbd0a090ee5
80804b041bcd3946b38806ee211418a94ed10ea6
30328 F20110330_AAAYEP rockwell_c_Page_015.QC.jpg
ae77143a01272b6f68352e16d644549b
86b2b6813f694b646a937794fe37b399ec563a31
107489 F20110330_AAAXZI rockwell_c_Page_040.jp2
bc677eb414c2b169c0e95d3dc8ddda57
38e8efe4c72a35d245a0e54259588247361f5c6e
114724 F20110330_AAAXYU rockwell_c_Page_026.jp2
ef0aec1c7f3b772a01cbab158528aa64
8f1d958fa756400a2c87ad9c745f149be5a9b9af
35285 F20110330_AAAYFD rockwell_c_Page_035.QC.jpg
1715678a7f22bccaa6c4698078ef4d4e
bda22a5d9981ee3ca37d9bef9c50e8bb2be89205
8846 F20110330_AAAYEQ rockwell_c_Page_019.QC.jpg
7d2477c13ba4b719ab39e4a92bb9dd75
6c1ba3fcad929c8a763288734238c5e45999f11b
109088 F20110330_AAAXZJ rockwell_c_Page_041.jp2
a248e86561a2d04b2846a13f2725adef
03d6d7db68660900b735cc555bb5ae7d031dffa2
113016 F20110330_AAAXYV rockwell_c_Page_027.jp2
a9955df240e2e96db0b14c6489c772eb
519094c7777dc71871ca1ab37969efc2d08b8578
32401 F20110330_AAAYFE rockwell_c_Page_036.QC.jpg
4721d8a5368fef279ffb710ea5cd9bda
f579f76c8026f2421e773d264a88f55f678d2804
107403 F20110330_AAAXZK rockwell_c_Page_042.jp2
8f1c1376cf34b8128363c9fc2c755925
5ea39555a79e38bff3e8d43748f72df67e40e6e8
115477 F20110330_AAAXYW rockwell_c_Page_028.jp2
f76802698b3f21bb11b307917b1ca4a7
b9e66c0dfb120678654cc0e079c91fd2373f55d2
33766 F20110330_AAAYFF rockwell_c_Page_037.QC.jpg
42cf3b86fddc6b6f899d44186478802d
e21aee0b912dbd0f2b85487a338c8bcd2090479d
109241 F20110330_AAAXZL rockwell_c_Page_043.jp2
5d52c08873ea73e766cff1644a8732df
0fe7b416a0a6be4caa5fdd099e06afe6f6f18b92
111849 F20110330_AAAXYX rockwell_c_Page_029.jp2
43a29a6226d8bde7ba729b03714483d4
21704d45669440f1ebd6a7c69f3821edcf28158d
32863 F20110330_AAAYFG rockwell_c_Page_038.QC.jpg
c2d13831730005a985fd7451b4935369
d0375fca5fddd44a6ec1d892ed99d09ede114a4b
28770 F20110330_AAAYER rockwell_c_Page_020.QC.jpg
769bb561c2d6e60143b7d56edc44ac50
962bc6e9c50e7021aec9d5f2e1afd55eb0bcc51f
115871 F20110330_AAAXZM rockwell_c_Page_044.jp2
8dab608ac7c021347fed5dfd08da86ad
5e1e129be7928cdc72c4f776926616e6998b22a6
112240 F20110330_AAAXYY rockwell_c_Page_030.jp2
03c7364fc22167f8a166cebe51aaf62c
78b6552f3b7b01671e624a741384eb1f254938e5
33154 F20110330_AAAYFH rockwell_c_Page_039.QC.jpg
4bb90b7e48aebaa6fb82d9b5fde88663
859f4f552bc80a3822766f8abdd4b9d9b05f1211
34019 F20110330_AAAYES rockwell_c_Page_021.QC.jpg
ad2ef24ac836d77b91c804fd08798133
cfec630fd194e3240b511e2f6d945c863b33aa3b
111556 F20110330_AAAXZN rockwell_c_Page_045.jp2
bfa0a2bf43559e077f7e41746cebffd4
9a91430a286c8ea94f6746c4b2f8e1cf5ec7f2b4
111578 F20110330_AAAXYZ rockwell_c_Page_031.jp2
d73267606f2621da9d097e43294416d5
3f029f403ce0e40952d89f70f7a39346c864d492
31976 F20110330_AAAYFI rockwell_c_Page_040.QC.jpg
bd85d643e345444fc8441016ac567f83
7b9c6a39c8c3c88c103addd4ce3aefe38668fcc4
31528 F20110330_AAAYET rockwell_c_Page_022.QC.jpg
ce74f523fcbac3c7abdfbf57d8c7e19a
16f284660fa915994440b96dbaf8de51f0f3baca
112267 F20110330_AAAXZO rockwell_c_Page_046.jp2
24314828b841183b243132f377f9c163
18ed12fddf363a7b842da76317baa2c1bbce595d
32175 F20110330_AAAYFJ rockwell_c_Page_041.QC.jpg
6f00cf7979cc272c56aa5f0532c39695
37361d885e8277ae958ea4162158006de6c1fc5a
35538 F20110330_AAAYEU rockwell_c_Page_023.QC.jpg
2fba441e32876f06215effcb05319ef4
00824d7427b4cc07d2c9db2d9324485d1edd437d
112814 F20110330_AAAXZP rockwell_c_Page_047.jp2
79d3ae2f067b072878404fe535d99771
aaf2077e82e54169a6682a96d6b6c7bab8eebf2a
32852 F20110330_AAAYFK rockwell_c_Page_043.QC.jpg
95d0317ae9c4b5254d0d408090612c63
141ff7d17834464f494bfcb224c3f5571d68bd63
34575 F20110330_AAAYEV rockwell_c_Page_026.QC.jpg
d9afcc81ce8d5d57bd8d60441fe303f7
4695c6f2096de4e34ba32eaf8c2da474b65687dd
108345 F20110330_AAAXZQ rockwell_c_Page_048.jp2
ca46f7af76ef10e3c3173bff35421049
1d1dc5934ea35d5455b6c76196161759be96cab9
35464 F20110330_AAAYFL rockwell_c_Page_044.QC.jpg
1a714d16dfa3119492d435e4b3454ddb
0f2c8eea991706441f0390536920509040f91853
34143 F20110330_AAAYEW rockwell_c_Page_027.QC.jpg
5ab22a7d3a148beb056bfd1e9e39acc4
2578d52da4edab731f74e392b53ce532a1e38601
115632 F20110330_AAAXZR rockwell_c_Page_049.jp2
aee92e8c012cae313eb414d55c920d2e
313e9d270ff3b84e7bdd1a644c8b1764424bb762
33232 F20110330_AAAYGA rockwell_c_Page_073.QC.jpg
1c8273fae7eb82dd10b20df18b100e9e
a6380df851cb4775b25b709f11b114c4a0370c4b
34835 F20110330_AAAYFM rockwell_c_Page_046.QC.jpg
2c5899ac7192c25b9f0faa9b55fe369a
6566d4de8391ff3b301575c305145dd6128cacd7
34613 F20110330_AAAYEX rockwell_c_Page_028.QC.jpg
66b17570323dfb5d214f6c5e14b3cbb8
95736f0f5fea1fe2e915d4bdd2da47bbf8bb781c
108997 F20110330_AAAXZS rockwell_c_Page_050.jp2
e98574e5cf459f472c3a8deb8cd4de91
e2cc51b1fdf407cfc2c2481662c937c0b77b1faf
35932 F20110330_AAAYGB rockwell_c_Page_074.QC.jpg
7ea61a8ebabab553b71955dc62b483e7
bf7a4936d3ba30ffca65a9809b94e740d3e6d877
32311 F20110330_AAAYFN rockwell_c_Page_050.QC.jpg
4fe397179971e26b3116778010bfbf07
15355640f638bfab236a6f58860d87e6a33bcc74
33907 F20110330_AAAYEY rockwell_c_Page_030.QC.jpg
8dfb4f64134697fc24a42e5d7439b3e6
5b6de79858326cd4e06fb6d36852ef8bba23c0ae
110756 F20110330_AAAXZT rockwell_c_Page_051.jp2
484b3c65c4c1e90e4bacc53b6f3c6daa
792a0c7014ed7fd3433279eadb583f0f4351413f
34320 F20110330_AAAYGC rockwell_c_Page_075.QC.jpg
9e5a30f3cce73ee310a9ba358d7d29e7
53acde87f4a5a6d6efd94649b03d0705dbe05672
34365 F20110330_AAAYFO rockwell_c_Page_052.QC.jpg
fdf42a7f0f81c8ab9bfbaf2f07c4d2ad
881977b78683c266a17f46ea1003078c0a3a3b0f
34288 F20110330_AAAYEZ rockwell_c_Page_031.QC.jpg
2439fe6de63fc1c4081275894a814c7a
e1ec8f6bc8b4efc9e91eafa4112c780ce905c41c
110851 F20110330_AAAXZU rockwell_c_Page_052.jp2
f163f75d2b46355fbd553adafb571c99
14f537ffc431795355f916f60774c1f14585ba18
34826 F20110330_AAAYGD rockwell_c_Page_076.QC.jpg
94bb4cdf63127590753eb0cbb215c93f
b88c72fd9b5bb8c61529ddff2645bd3cb4e76099
12572 F20110330_AAAYFP rockwell_c_Page_053.QC.jpg
80e8a6e5d0a6bb0221f806ad823bd6b6
f629e99746686678d145d5a0e8df047c8a97268c
39749 F20110330_AAAXZV rockwell_c_Page_053.jp2
48d7e7308e13cbc5825b9d865e6ecf43
f5c5332e124e39674c43882ac721a9cef10d6195
F20110330_AAAYGE rockwell_c_Page_077.QC.jpg
4d9a68d72871c4d72978697210eb45d6
c5b8956940be49e784db05a8857c0c246229e50d
15140 F20110330_AAAYFQ rockwell_c_Page_054.QC.jpg
1a7f87a71d5ac90612f54221cd6a3048
6ffc74bada3ab976669e40e6665ccca3298d0547
85078 F20110330_AAAXZW rockwell_c_Page_054.jp2
1f93eb6a9edd1836d20ee6ee477e826e
fe37d080240e694e503276020f6aae823826577f
5496 F20110330_AAAYGF rockwell_c_Page_078.QC.jpg
9a92f34612dffc360ddabdd3d0edb97e
4dbf4471c8bec9aa82b079d6ddf31a4a12d5d867
29496 F20110330_AAAYFR rockwell_c_Page_056.QC.jpg
9fbf003c3ad711b58e9d127ff6f150d0
8b0adf551315dbc7483826c05d66f21ccb624e09
67600 F20110330_AAAXZX rockwell_c_Page_055.jp2
e9b23fb9c87b9186f86c0b494f6fe06e
1a63df2e661588ca4d053da5f099119eef3347c2
27017 F20110330_AAAYGG rockwell_c_Page_079.QC.jpg
75c37497f9308bc9f558df5ab20a6d1a
815400886a52ce1dc43c87755d3f13d73598cb59
95218 F20110330_AAAXZY rockwell_c_Page_056.jp2
6964a6518b63770bb4632defdb292024
1c00df493db6fe8dd5e3e785545fedefeba65f18
16724 F20110330_AAAYGH rockwell_c_Page_080.QC.jpg
f8ffd4de317c2189477cc834ed8e5649
3b85f806e3d621cdbad313bb143d442c9c97ef8b
32010 F20110330_AAAYFS rockwell_c_Page_057.QC.jpg
ed9a3ebcb8056a55627cf7674fcd4279
17ce436c212a62ede49d03d6f616d19244b1c572
107038 F20110330_AAAXZZ rockwell_c_Page_057.jp2
b446312108a1abd92e6ab3f4b70ee4b5
cdf364aaf2acc4b6c7fe89bb8afab7ae3ed0fe33
7882 F20110330_AAAYGI rockwell_c_Page_081.QC.jpg
75109c88acac88973ec8b6a0f5b29336
10e8da1da0d538bb146b112f6cf5dbfe82fb2ab6
32802 F20110330_AAAYFT rockwell_c_Page_058.QC.jpg
b499ca651dc21d3684d2a608efdf2f48
e3048a26dd92a322c8c3b231fb0cbe691a9fb287
17358 F20110330_AAAYGJ rockwell_c_Page_082.QC.jpg
df7883a1e57cc0cacd3bc0e313dae6c7
fd6eb2239a18c0fffe456cde083a9688de53a174
33706 F20110330_AAAYFU rockwell_c_Page_062.QC.jpg
621981810c0503faad2cd7b422cf368f
c1b93900b4748477c4e3853b4c64abbda78f7e6e
17909 F20110330_AAAYGK rockwell_c_Page_083.QC.jpg
0b42409527c3167a9a437b97dd926fd8
2e62088abb5756484c615f2d36dc8c4763221985
28365 F20110330_AAAYFV rockwell_c_Page_065.QC.jpg
60d14ef4e6c96ce495f0887a26d1c8f7
4bcf370ddade863d075b7eeb103f0ac259f6fca5
16608 F20110330_AAAYGL rockwell_c_Page_084.QC.jpg
b1403fdb1ec29adcc76aaaefd3d3987a
dc6e0a1a602a88bb183891b2d5845d10e15592d0
30236 F20110330_AAAYFW rockwell_c_Page_066.QC.jpg
36a4229019e2e9e7fab9c76ddb60a9c3
5863fa2faa777c94babc976a6bb0aef4c5679d2e
12322 F20110330_AAAYGM rockwell_c_Page_086.QC.jpg
49467e9bd76ebdb96ab9f33b6c477fd8
1561cc8926ad68e848bc479ebb812f85933cfbd8
34438 F20110330_AAAYFX rockwell_c_Page_068.QC.jpg
6e5570deb1fae14197bc26a86160f90b
3de3b63310af975b5757909ec39411feb6fb0f5a
1286 F20110330_AAAYHA rockwell_c_Page_003thm.jpg
cd5b162ad30377ca85a071f871e7f3f9
74af88ea0f35dc7526be03149146991421105daa
11050 F20110330_AAAYGN rockwell_c_Page_088.QC.jpg
de47614e95ccf2b4ec5da5280b9d3d3b
0285dd9864fcfffd294d16fabe544479e2481d97
34838 F20110330_AAAYFY rockwell_c_Page_071.QC.jpg
17c3f69af25f2ba09bacfb91ff5bb372
9f1ada985a025f13d024f1be721442ccf4392c31
4909 F20110330_AAAYHB rockwell_c_Page_006thm.jpg
f5de6ffff85ccdb25454e823c91191e5
e8deaf7f4f039a833654cafb6eec47bcd362468e
10645 F20110330_AAAYGO rockwell_c_Page_089.QC.jpg
b49daba7488cc2490565b36873cad328
51dd77c5353a9847ce488e7b1bdacbe1a76243a0
33594 F20110330_AAAYFZ rockwell_c_Page_072.QC.jpg
7a1a114f657742cee2e2e0c7068029e2
27a0c2b208680e9f878936f719302a4029c1e27b
6270 F20110330_AAAYHC rockwell_c_Page_007thm.jpg
f6c4f2dc42d9f4ac0f98670c87c3085f
9b4245a069102ce233074d6512557f04abad9e46



PAGE 1

COMMUNITY-BASED TIMBER MANAGEMENT IN ACRE, BRAZIL, AND ITS IMPLICATIONS FOR SUSTAINABLE FOREST MANAGEMENT By CARA A. ROCKWELL A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Cara A. Rockwell

PAGE 3

This thesis is dedicated to my friends and colleagues in PAE Porto Dias, who welcomed me into their homes, and to my husband, Chris Baraloto, for coming this far in the journey together. iii

PAGE 4

ACKNOWLEDGMENTS I would like to express my sincerest gratitude to my advisor and committee chair, Karen Kainer, for all of her unfailing support. The past two and a half years have been a truly enjoyable experience. My committee members, Marianne Schmink and Jack Putz, have been excellent sources of encouragement and knowledge. I extend my full appreciation to their involvement in the development of this thesis. Thanks go to the School of Forest Resources and Conservation, the Tropical Conservation and Development Program and Sigma Delta Epsilon/Graduate Women in Science for the funding and support of this research. The execution and completion of this project could not have been accomplished without the participation of many key partners in Brazil. First, I must thank the families of PAE Porto Dias, the Associao Seringueira Porto Dias, and the Projeto Manejo Florestal Comunitario de Mltiplo Uso for their patience as well as their kindness during my stay in their community. I must particularly thank the following families for welcoming us inside their homes; such hospitality will never be forgotten: the family of Francisco Correia da Cunha and Raimunda Feitosa do Nascimento, the family of Adilton Ferreira de Souza and Maria Alves Barboza, the family of Juscelino da Silva Correia and Sebastiana do Santo Venncio, and the family of Lzaro da Silva Salgueiro and Marinalva Souza Amora Salgueiro. Also, a special obrigada goes to the following community members for their participation in the study: Lzia Amora Salgueiro, Lziani Amora Salgueiro, Juciclei Venncio Correia, and Juciane Venncio Correia, and thanks iv

PAGE 5

go also to Luciana Roncoletta, who was conducting her own research in Porto Dias, but still found time to assist with my project. Undoubtedly, the staff of the Centro dos Trabalhadores da Amaznia (CTA) was there for me from beginning to end. They are exemplary models of how collaborative partnerships can work. I especially would like to thank my friends and colleagues, Nivea Marcondes, Magna Cunha dos Santos, Evandro Arajo, Patricia Roth, and Pedro Bruzzi. I received valuable information and support from the staff of the Federal University of Acre herbarium as well as faculty members Marcos Silveira and Evandro Ferreira. Lucia Wadt from the Empresa Brasileira de Pesquisa Agropecuria (EMBRAPA) provided support both in terms of research interest and in friendship. And across the border in Bolivia, Marielos Pea-Claros of the Instituto Boliviano de Investigacin Forestal (IBIF) gave me great insight and suggestions pertaining to the implementation of this study. I never would have gotten through the crazy moments without the support from so many fellow students, both in Gainesville and in Acre. If I miss any names, it is only because there have been so many in the last couple of years. My warmest regards and thanks go out to the girls of the Kainer lab: Diana Alvira, Jamie Cotta, Rosa Cossio Solano, Amy Duchelle, Shoana Humphries, Christie Klimas, and Joanna Tucker. I thank fellow RPCV Paraguayers who have seemed to find their way to University of Florida, for all of their kind moments, advice, and plentiful terere: Mandy Baily, Maria DiGiano, Tom Henshaw, and Katie Painter. The hospitality of Rich Wallace and Ana Puentes in Acre will never be forgotten. I thank Samantha Stone for her introduction to Porto Dias and CTA. She has set a high standard for all to follow. Other fellow SFRC students who have given great support along the way and have always been willing listeners are: Julie v

PAGE 6

Clingerman, Robin Collins, Trina Hofreiter, Louise Loudermilk, George McCaskill, and Miriam Wyman. Other faculty and staff at the University of Florida have often provided much-needed counsel and direction. Their efforts are duly appreciated: Cherie Arias, Hannah Covert, Jon Dain, Eric Jokela, Ramon Littell, Christine Staudhammer, and Daniel Zarin. I must not neglect to thank those who assisted both in terms of hospitality and technical assistance during the writing process in French Guiana: Sabrina Coste, Paul Arrive, Eric and Darouny Marcon, Franois Mornau, and Jean-Yves Goret. I never could have gotten to this stage without the support and love of my parents and my family. And there are no words sufficient to thank my husband, Chris Baraloto. My appreciation will probably never make up for all of the cuts and stings suffered in the tabocal forest of Acre while he acted as my botanist and trail cutter, but it has nonetheless been an incredible journey. His love of the forest has been one of the greatest educations in this process. vi

PAGE 7

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES ............................................................................................................x ABSTRACT.......................................................................................................................xi CHAPTER 1 GENERAL INTRODUCTION....................................................................................1 2 ECOLOGICAL LIMITATIONS OF REDUCED IMPACT LOGGING AT COMMUNITY AND SMALLHOLDER SCALES.....................................................8 Introduction...................................................................................................................8 Sustainable Forest Management (SFM) and Reduced-Impact Logging (RIL) Models....................................................................................................................11 Community/Smallholder Scale vs Industrial Operations..........................................13 To What Extent Does the RIL Model Ensure SFM at the Smallholder/Community Level?............................................................................19 Pre-harvest Forest Inventories, Botanical Identification and Selection of Seed Trees.................................................................................................................23 Minimum Diameter Cuts....................................................................................27 Low Harvest Intensity.........................................................................................30 Skid Trail Planning.............................................................................................32 Rotation Cycles...................................................................................................33 Liana Cutting......................................................................................................34 Conclusions.................................................................................................................36 Logging Block Size and Implications for Pre-harvest Inventories.....................37 Monitoring..........................................................................................................39 Moving beyond RIL............................................................................................40 3 EVALUATING LOGGING DAMAGE TO FUTURE CROP TREES IN ACRE, BRAZIL: IMPLICATIONS FOR MANAGEMENT OF A CERTIFIED COMMUNITY FOREST...........................................................................................44 Introduction.................................................................................................................44 vii

PAGE 8

Site Description..........................................................................................................47 Forest Management and Harvesting Operations.........................................................49 Methods......................................................................................................................51 Future Crop Tree (FCT) Selection......................................................................51 Experimental Design and Pre-logging Inventory................................................52 Area Disturbed.....................................................................................................52 FCT Location in Bamboo (Guadua)-dominated Forest......................................53 Liana Presence.....................................................................................................53 Harvest Intensity..................................................................................................53 Marking Treatment..............................................................................................54 Assessment of FCT Damage...............................................................................54 Statistical Analyses..............................................................................................55 Results.........................................................................................................................55 Overview of the Logging Operation....................................................................55 FCT Damage.......................................................................................................56 Area Disturbed.....................................................................................................57 Discussion...................................................................................................................57 FCT Damage in Guadua-dominated Forest........................................................59 Impact of Lianas on FCT Damage......................................................................61 Impact of Harvest Intensity on FCT Damage......................................................62 Impact of Marking Treatment on FCT Damage..................................................63 Conclusion..................................................................................................................64 4 CONCLUSION...........................................................................................................78 REFERENCES..................................................................................................................81 BIOGRAPHICAL SKETCH.............................................................................................94 viii

PAGE 9

LIST OF TABLES Table page 2-1 Contrasting tree population parameters from inventories at two different spatial scales commonly used by smallholders...................................................................42 2-2 An example of spatial aggregation among neotropical tree species harvested by communities in Acre, Brazil.....................................................................................43 3-1 Commercial timber species inventoried during 2004 logging operations in the PAE Porto Dias, Acre, Brazil For those species harvested in 2004-2005, total volume (m3 ha-1) is given.........................................................................................67 3-2 Classification of damage sustained by FCTs (future crop trees) during logging operations (modified from Jackson et al 2002 and Krueger 2004).........................69 3-3 Summary of harvesting impacts in 2004 for the three logging blocks at PAE Porto Dias, Acre, Brazil...........................................................................................70 3-4 Results of a logistic regression for the probability of damage to future crop trees (FCTs) at Porto Dias, Acre, Brazil during 2004 logging operations........................71 ix

PAGE 10

LIST OF FIGURES Figure page 3-1 Photograph taken in PAE Porto Dias, Acre, Brazil, demonstrating the re-curved thorn on a branch of a Guadua culm........................................................................72 3-2 PAE Porto Dias, with the four landholdings inventoried in the study, including the three (in blue) harvested in 2004........................................................................73 3-3 Logging block of Barrinha 1, showing skid trails, harvested trees, and location of future crop trees (FCTs) in the zones of impact...................................................74 3-4 Logging block of Barrinha 3, showing skid trails, harvested trees, as well as location of future crop trees (FCTs) in the zones of impact.....................................75 3-5 Logging block of Palestina, showing skid trails, harvested trees, as well as location of future crop trees (FCTs) in the zones of impact.....................................76 3-6 Relationship between size of felled trees and the surface area of the corresponding logging gap created during felling....................................................77 x

PAGE 11

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 COMMUNITY-BASED TIMBER MANAGEMENT IN ACRE, BRAZIL, AND ITS IMPLICATIONS FOR SUSTAINABLE FOREST MANAGEMENT By Cara A. Rockwell December 2005 Chair: Karen Kainer Major Department: Forest Resources and Conservation In recent years, there has been much debate about whether conservation goals can be achieved by logging old-growth tropical forests. Even so, the global demand for tropical hardwood is not likely to subside anytime soon, making good management of forests outside of parks and other protected areas essential. Some researchers have suggested that rural communities may be the most ideal candidates to engage in sustainable timber management, since they generally harvest at low intensities. Yet, it has also been noted that the impact of natural resource exploitation at this scale is still not well known, and that sustainable forest management (SFM) guidelines will necessarily vary, depending on the location in which they are being implemented. Field studies were conducted in Acre, Brazil to assess the impacts of the Porto Dias community timber management project. The main objectives of this investigation were (1) to determine the extent of area disturbed (treefall gap and skid trail construction) in a small-scale certified logging operation, (2) to determine a difference in damage incidence xi

PAGE 12

to future crop trees (FCTs) 20cm dbh between forest with or without bamboo (Guadua spp.), (3) to determine a difference in damage incidence to FCTs 20cm dbh between those associated with lianas and those that are not, (4) to determine to what extent harvesting can be conducted more intensely (m3ha-1), without incurring greater damage to FCTs 20cm dbh, and (5) to determine to what extent marking of FCTs diminishes damage during logging. Related to this experimental study, an examination of key reduced-impact logging (RIL) harvesting techniques and their potential influences on population dynamics of tropical tree species in small scale timber operations was also initiated, with the goal of addressing potential future management strategies for the smallholder system. The results from the study indicate that although the effects of marking, liana presence and location in Guadua-dominated forest were not significantly associated with FCT damage, the high values reported for area damaged merit attention. Previous studies have demonstrated that low intensity harvests are often associated with increased damage levels per tree due to skid trails, suggesting that the gap mosaic of the forest will often change, even when low-intensity RIL guidelines are followed. Given the series of poorly understood parameters outlined in this thesis, it is recommended that systematic post-harvest monitoring accompany logging operations, so that the benefits of recommended practices may be evaluated by managers at the appropriate scale. xii

PAGE 13

CHAPTER 1 GENERAL INTRODUCTION Deforestation is one of the greatest environmental challenges of the past few decades (Food and Agricultural Organization (FAO) 2005). Brazil, which claims 63% of the Amazon basin (Lentini et al. 2003), has long been at the center of the tropical conservation debate. The Brazilian Amazon constitutes approximately one-third of the worlds extant tropical forests (Uhl et al. 1997). Since the 1960s, however, the country has been plagued by destructive development policies that have opened up the region to industrial logging, mining, ranching, and large-scale colonization projects, contributing to a 14% forest cover loss to date (Nepstad et al. 2002). Compounding the problem of forest cover loss, Brazil is simultaneously the worlds largest producer and consumer of tropical timber, most of which is generated through land clearing or unplanned selective logging (Verissimo and Barreto 2004). Selective logging is the harvesting method in which individual trees (rather than an entire group) are removed. When conducted on a rotational basis, selective logging is generally regarded to have a low impact on the forest (Uhl and Buschbacher 1988, but see Sist et al. 2003 a,b, Asner et al. 2005). Unplanned selective logging, however, typically excludes the use of a systematic timber inventory or management plan and often harvests the same area over short, repeated intervals, frequently removing high volumes of timber. The primary goal in this case is to extract the largest and most valuable commercial species with little regard to surface area damaged or future harvests (Schulze 2003). In many instances, this has led to acute (up to 50% or more) canopy reduction, 1

PAGE 14

2 severely disturbed mineral soils and damage to the residual stand (Uhl and Vieira 1989, Verissimo et al. 1992, Johns et al. 1996). In response, scientists and practitioners alike have proposed implementing sustainable forest management (SFM) practices that could limit canopy cover loss and protect ecosystem function and biodiversity while maintaining the economic value of the forest (Putz 1994, 1996, Pearce et al. 2003). Definitions of SFM differ, but the term generally implies the sustenance of many forest products and services over long periods of time (Pearce et al. 2003). Yet, despite the recent conservation interest in SFM, there are many who would argue that to link conservation with sustainable forest management is erroneous, and that the only way to curb deforestation is outright protection (Rice et al. 1997, Bowles et al. 1998, Terborgh 1999). Recently, there has been a massive undertaking in Brazil to reduce the negative environmental impacts of regional silvicultural practices, much of which is generated by large scale industrial operations. Forest management research in the region was first initiated by FAO in the 1950s (Oliveira 2000), but it was not until the early 1990s that the Brazilian federal government began advocating reduced-impact logging (RIL) techniques as a means to combat rampant predatory logging in the Amazon basin (Holmes et al. 2002, Schulze 2003). RIL is an essential component of SFM if timber is to be harvested from the forest (Pearce et al. 2003), typically employing a pre-harvest forest inventory, road and skid trail planning, pre-harvest vine cutting, and directional felling (Dykstra and Heinrich 1996). Indeed, in the absence of secure prices for non-timber forest products (NTFPs) and infrastructural support for ecotourism efforts, selective logging under the RIL model at low intensities may be one of the few assured income-generating activities

PAGE 15

3 that can preserve the structure of the forest without compromising the entire ecosystem (Oliveira 2000). The SFM model has generated considerable interest for community managers and smallholders in the region, with the last few years witnessing the creation of a number of innovative community-based conservation experiments founded on SFM principles (Kainer et al. 2003, Stone 2003). Extractive reserves, or tracts of forest under community control with secure usufruct rights, figure prominently among these community development efforts. The remarkable and often violent history that preceded the emergence of such initiatives as the extractive reserve (RESEX) program immediately gave these forest communities widespread international support, but their conservation goals have since become the focus of intense scrutiny (Nepstad et al. 1992, Redford and Stearman 1993, Terborgh 1999, 2000). Many researchers, although generally not contesting the social justice value that such endeavors provide to a historically disenfranchised population, have questioned the wisdom in holding up the extractive reserves as models for biodiversity conservation (Redford and Stearman 1993, Terborgh 2000). This issue has become especially pertinent now that many community forests have shifted from local economies based on non-timber forest products (NTFPs) to that of timber extraction (Kainer et al. 2003, Schmink 2004). The general consensus among some researchers is that external economic and cultural pressures, in concert with a tendency to abuse the resource commons (see Hardin 1968), will hinder the ability of community forest managers to protect biodiversity (Redford and Stearman 1993, Terborgh 2000).

PAGE 16

4 In Brazil, though, voluntary forest certification has provided a strong incentive for both community forest managers and commercial logging companies to adopt SFM practices (May 2002, Kainer et al. 2003). As of April 2004, Brazil ranked fourth in the world in number of certified forests and eighth in area (slightly under 1.6 million ha, 529,079 ha of which are considered natural forest) (May 2002). The failure of government policies to curb illegal logging gave birth in the 1990s to the certification instrument, which endorses forest products originating from sustainable forestry operations, a process often executed by a third-party organization (Molnar 2003). Forest certifications role as a community-based conservation mechanism has received widespread support (Molnar 2003). This has been particularly true of the Forest Stewardship Council (FSC), an international organization that has had a strong influence in the development of certification standards in the Brazilian Amazon (Humphries 2005). Ultimately, the certification process could provide access not only to the global certified forest product market but also to silvicultural evaluation and assistance, thereby creating motivation to conserve the resource base (Putz 1996, Putz and Viana 1996, Putz and Romero 2001). Despite their positive associations with the conservation movement, a number of researchers have criticized forest certification bodies and biologists for producing SFM guidelines and criteria that fail to recognize the problems that such a system might create for smallholders (Pretty and Hine 1999, Molnar 2003). Even though community timber management operators may have the knowledge, human nature and economics can offset the desire to sustainably manage the forest (Salafsky et al. 1998). Smallholders and community forest managers in the rural tropics have limited financial and technical

PAGE 17

5 resources, so implementing additional silvicultural treatments and monitoring may be unfeasible for them, unless the proposed activities are seen as adding to their productivity. Furthermore, given intense competition from less sustainable commercial industrial timber enterprises and illegal logging, the long-term impact that certification will have on the industry and deforestation in general is questionable, particularly in Latin America (Sierra 2001). In Brazil, for example, certified wood is currently meeting only 2% of the annual demand for logs (Lentini et al. 2003). Community and smallholder forest enterprises are generally thought to be more environmentally benign than their industrial counterparts (Salafsky et al. 1998), making them seemingly ideal candidates to implement the SFM and RIL models. There are, however, a number of inherent ecological limitations associated with small-scale timber operations that are rarely addressed in the management literature, with few ecological investigations regarding the impact of timber extraction having been conducted at the level of the community or smallholder level (Oliveira 2000, Walters 2005). Although RIL is, in general, ecologically beneficial to the residual stand, there is evidence that the magnitude of those benefits varies, depending on the forest being exploited (Fredericksen and Putz 2003, Schulze 2003) and the specific RIL techniques being used. RIL was initially developed for large scale industrial operations where harvest intensities and investment potential are high (Oliveira et al. 1998). Nevertheless, community forest managers have eagerly adopted RIL standards, even though the ecological impacts of RIL at this scale are not well-known. Given that these impacts are poorly understood, there is a need to explore both the limitations and the benefits of RIL for these

PAGE 18

6 community and smallholder-based systems rather than embracing what is held to be the ideal by researchers working with larger operations. In this thesis I examine both RIL guidelines as generally implemented at the community and smallholder levels as well as a specific case study in PAE Porto Dias, Acre, Brazil, testing the harvesting impacts generated by the activities of the Multiple-Use Community Forest Management Project on the residual stand. This approach potentially addresses some of the challenges of community timber operations in Acre as well as examining future crop tree (FCT) damage as sustained during RIL operations. In Chapter Two, I examine some of the ecological constraints associated with timber operations at the community and smallholder scales. To do so, I evaluate several assumptions of selected RIL recommendations and their implications for smallholder management. After a brief introduction to SFM and RIL, particularly as they relate to the Brazilian Amazon, I discuss eleven key differences between industrial and community/smallholder operations. I then ask to what extent the RIL model contributes to SFM at the smallholder/community level, with an examination of RIL harvesting techniques and their potential influences on the population dynamics of tropical tree species in smallholder systems, with a focus on the Brazilian Amazon. I conclude the paper with a discussion of potential future management strategies for smallholders. In Chapter Three, I present research investigating the impacts that forest type (bamboo/non-bamboo), liana presence, harvesting intensity, and marking have on FCT damage during logging operations in PAE Porto Dias. The study focuses on the adult and sub-adult size classes ( 20 cm dbh) of commercial timber species in the hopes of

PAGE 19

7 capturing information about the availability of timber at the time of the next harvest (30 years). This thesis has been organized such that the second and third chapters are two individually and fully structured papers for publication. Each one of these chapters has its own conclusions, while Chapter 4 summarizes the main findings of the entire study and provides a discussion of the future directions for small-scale timber management research.

PAGE 20

CHAPTER 2 ECOLOGICAL LIMITATIONS OF REDUCED IMPACT LOGGING AT COMMUNITY AND SMALLHOLDER SCALES Introduction Researchers and practitioners alike have proposed that large tracts of tropical forests will not be conserved without the involvement of local forest residents (Scherr et al. 2002, Shanley and Gaia 2002). Already, approximately one quarter of the worlds forests are community-owned or managed, a proportion that is expected to double in the next 15 years (Molnar 2003). Some have advised that these forests will stand a greater chance in surviving when placed under the jurisdiction of communities, assuming they will have a desire to conserve their forest resource base (Camino and Alfaros 2000). In concert with a growing smallholder interest in formal timber production and a need for sustainable forest management (SFM) practices outside of protected areas (Putz et al. 2000), it is likely that community and smallholder forests will play an increasingly important role in global conservation. For various reasons, including an international market demand for sustainably harvested timber products, many of these community and smallholder operations are becoming third-party certified (Molnar 2003). This voluntary process emphasizes rigorous social and environmental standards for production forests, potentially providing a strong economic incentive for local forest managers to value the resource base (Putz 1996, Putz and Viana 1996). The reduced impact logging (RIL) management models favored by the certification instrument, however, have failed to develop situation8

PAGE 21

9 appropriate strategies that accommodate the ecological and social complexities of small-scale commercial timber extraction. Instead, the tendency has been to impose management practices of large commercial operations onto smallholder systems (Oliveira 2000), causing some forest operations to take on corporate features usually relegated to industrial timber enterprises (Putz 2004). It has been suggested that further research and operational training are needed to evaluate how variations in forest type and size of logging operations affect the performance of RIL systems (Holmes et al. 2002). Without much empirical information suggesting otherwise, it is often assumed that because many smallholders harvest at low intensities and utilize animal traction (Oliveira 2000) or small machines, they are more environmentally benign than industrial logging enterprises (Salafsky et al. 1998). Nevertheless, there are a number of inherent ecological limitations associated with small-scale timber operations and low intensity harvests, which have been shown to have substantial impacts on forest structure, composition and regeneration (Smiet 1992, Gullison and Hardner 1993, Boot and Gullison 1995, Whitman et al. 1997, Awasthi et al. 2003, Walters 2005). Community and smallholder-owned forests generally encompass smaller areas compared to their industrial counterparts ( 500 ha) (Lentini et al. 2003). When managing for timber in small, non-contiguous management units, ecological constraints to regeneration, such as insufficient pollen transfer and/or inbreeding (Ghazoul et al. 1998), dispersal limitation (Dalling et al. 2002), and predator satiation (Schupp 1992) can become exacerbated, as population sizes of all species are correspondingly small (Putz et al. 2000). RILs emphasis on environmental sustainability is often assumed to be a natural fit for rural forest dwellers (Braz and Oliveira 1996). However, it is becoming increasingly

PAGE 22

10 clear that the shift from traditional management systems to that of third-party certified commercial timber extraction is challenging. Smallholders and communities are typically required to implement RIL techniques to meet certification standards, compelling them to invest in both equipment and human capital not normally required for their daily subsistence needs (Stone 2003). Although the RIL model has been demonstrated to have many ecological benefits in tropical forests, there is a need to explore both the limitations and the benefits of RIL specifically for smallholder systems. In this paper, I evaluate key assumptions of several RIL recommendations and their implications for smallholder management. I use the terms smallholder and community interchangeably throughout, though there is a distinction (albeit broad) between smallholders living on individual lots in colonization projects and traditional or indigenous forest-dwelling communities, such as those living in or near forests under some form of protection. As well, when referring to the terms community and smallholder management in this paper, I am specifically addressing locally-run operations, not to be confused with commercial logging concessions on smallholder properties (e.g. Nepstad et al. 2004). After a brief introduction to SFM and RIL, I discuss eleven key differences between industrial and community/smallholder operations. I then ask to what extent the RIL model ensures SFM at the smallholder/community level, with an examination of RIL harvesting techniques and their potential influences on the population dynamics of tropical tree species in small scale systems, with a focus on the Brazilian Amazon. I conclude the paper with a discussion of potential future management strategies for the smallholder system.

PAGE 23

11 Sustainable Forest Management (SFM) and Reduced-Impact Logging (RIL) Models Selective logging when applied on a long rotational basis and at low harvest intensities is generally regarded to have a low impact on the residual stand (Uhl and Buschbacher 1988, but see Sist et al. 2003a,b, Asner et al. 2005). Selective logging, however, can be one of the first steps in forest conversion, providing a subsidy for pasture formation as in the Brazilian Amazon case (Oliveira 2000). Additionally, very few commercial timber operations in this region employ sustainable selective logging techniques, with an estimated 95% of all timber extracted through unplanned conventional logging (Verissimo et al. 2002). The extent to which selective logging causes permanent damage can be controlled in some ways by planned management activities, as has been suggested by the SFM model. SFM can be defined based on the following assumptions: (1) management can be exercised in a manner compatible with the maintenance of biodiversity, (2) management of tropical forests is economically viable, and (3) management can result in sustained timber yield over the long term (Bawa and Seidler 1998). Translating these concepts into practice, Bruenig (1996) maintains that when discussing timber extraction, SFM should encompass the following: canopy openings should be kept within the limits of natural gap formation; stand and soil damage must be minimized; felling cycles must be sufficiently long and tree marking so designed that a selection forestry canopy structure and a self regulating stand table are maintained without, or with very little, silvicultural manipulation; (and) production of timber should aim for high quality and versatility. Bawa and Seidler (1998) conclude, however, that too often, SFM is promoted without rigorous examination of the interrelatedness between biodiversity maintenance, economic viability, and long-term yields, causing some researchers to doubt the validity of SFM

PAGE 24

12 and calling for the outright protection of forested regions instead (see Rice et al. 1997, Bowles et al. 1998, Terborgh 1999). As well, some researchers have questioned whether timber harvests in at least some forests can be sustained without silvicultural intervention (Daube et al. 2005). Nevertheless, the SFM philosophy has been embraced by multiple forest actors. In Brazil, for example, the federal government, state institutions, NGOs and research organizations have actively promoted SFM, hoping that it will quell unchecked land conversion and predatory logging activities (Schwartzman et al. 2000, Nepstad et al. 2002, Fearnside 2003, Kainer et al. 2003). RIL is considered a necessary component to achieve the goals of SFM, though its implementation alone cannot maintain biodiversity and ecosystem functions nor even secure sustained timber yield (referring to the objective of maintaining timber production for future harvests) (Webb 1997, Putz et al. 2000, Sist et al. 2003b). RIL implies a series of preand post-logging guidelines (see Pinard et al. 1995, Dykstra and Heinrich 1996) which are designed to protect advanced regeneration (i.e., seedlings, saplings and larger future crop trees), minimize soil damage, limit damage to non-target populations (i.e.wildlife and non-timber plants species), protect water quality, and preserve ecosystem processes (i.e. carbon sequestration) (Putz et al. 2000). Operations under these guidelines typically employ long-term cutting cycles, pre-harvest forest inventories, road and skid trail planning, pre-harvest vine cutting, and directional felling (Dykstra and Heinrich 1996). RIL has been shown to reduce damage to the residual stand by 30-50% and has also demonstrated an increased efficiency of logging operations (see Johns et al. 1996, Pinard and Putz 1996, Bertault and Sist 1997, Barreto et al. 1998, Sist et al. 1998, Holmes et al. 2002).

PAGE 25

13 Under pressure from both domestic and international concerns to improve the ecological sustainability of the tropical hardwood industry, Brazil began to adopt RIL techniques in the early 1990s, following the model that was originally developed in Southeast Asia and Suriname a decade earlier. Since that time, many Brazilian timber companies have agreed to abide by RIL regulations at the behest of federal government regulations, and in some cases, third-party forest certification standards, hoping to tap a high-end niche market (Schulze 2003). Community/Smallholder Scale vs. Industrial Operations The SFM model (and implicitly, RIL) has generated considerable interest for community forest managers and smallholders (Braz and Oliveria 1996). Researchers often point to local peoples long-term commitment to place and their intimate knowledge of ecosystems (Schmink 2004) as evidence of their ability to both manage and (when necessary) restrict access to forest resources (Agrawal and Gibson 1999). Yet, it has also been noted that the implementation of a systematic management regime can often be overwhelming for communities and smallholders, requiring significantly different conditions for long-term success than large-scale commercial enterprises (Schmink 2004). Indeed, to mitigate the challenges faced by these operations, some suggest that community and smallholder forest management systems must be considered within case-specific ecological, socio-economic, political, and cultural settings (Berkes and Folke 2001, Schmink 2004). This viewpoint recognizes the unique character of individual community and smallholder systems and also further emphasizes the need to distinguish these enterprises from their industrial counterparts. Several general operational characteristics distinguish community and smallholder timber enterprises from the large-scale commercial timber industry, building on a list of

PAGE 26

14 five proposed by Salafsky et al. (1998): (1) Ownership: Industrial operations usually buy logging rights via a concession, or the timber company may own the land. More than likely, the proprietors live offsite. In contrast, community and smallholder timber harvesting rights are owned or, at the very least, held by local inhabitants (Salafsky et al. 1998). For example, in the Brazilian Amazon, one third of the land area is inhabited by smallholder and indigenous communities with the rights to manage their own forests (Shanley and Gaia 2002). Due to a strong sense of ownership and multigenerational responsibilities, community forest managers may implement more sustainable management practices than their industrial counterparts (Walters et al. 2005). Indeed, in those community cases where secure land tenure is lacking, there appear to be fewer incentives to manage the resource sustainably (Banana and Bombya-Ssembajjwe 2000, Gibson and Becker 2000). (2) Timber as integrated livelihood component: Community and smallholder forest managers rely on the forest for a suite of goods and services, viewing timber management as one component of an integrated system rather than a sole income-generating activity (Salafsky et al. 1998). Since timber harvesting is a seasonal activity (usually during the driest months), local managers pursue other livelihood activities throughout the remainder of the year (Oliveira 2000). Even so, timber management and other livelihood activities are not always compatible, especially when harvesting is conducted by those from outside the community. In one Eastern Amazonian case, community members from the Tapajs-Arapiuns Extractive Reserve believed that commercial logging negatively impacted NTFP collection (due to incidental damage) and hunting (through easier game access by logging roads) (Menton 2003). Researchers also have reported a lack of documented systems explicitly integrating timber and NTFPs in

PAGE 27

15 management plans (Bawa and Seidler 1998, but see Salick et al. 1995, Romero 1999). Local forest managers who wish to protect a multitude of forest services (NTFPs, game species, watershed) may be more inclined to adopt RIL guidelines geared to protect components of the forest beyond future timber harvests. (3) Scale: Community and smallholder enterprises generally operate on a smaller scale than their industrial counterparts, the latter tending to rely on capital intensive techniques and heavy machinery rather than local labor (Salafsky et al. 1998). Many techniques developed specifically for communities have been initiated in secondary forests, where the use of animal traction is ideal for the extraction of small diameter logs (Pinedo-Vasquez et al. 2001). Smallholder logging in primary forests has become more common in recent years, particularly in the Brazilian state of Acre (see Kainer et al. 2003, Stone 2003), but the need for heavy machinery in these systems and the subsequent competition with larger enterprises has constrained community participation in this type of extraction (Oliveira 2000). However, smaller scale operations may facilitate more intensive management for a suite of services and products, something that might be impractical in a larger management unit. Also, in Brazilian community timber efforts, harvest intensities tend to be low, usually no more than 10 m3 (or two trees) ha-1 (Oliveira et al. 1998), which (in theory) should be compatible with RIL guidelines. In comparison, most Brazilian conventional forest management systems harvest anywhere from 30 to 40 m3 ha-1 yr-1 (Oliveira 2000). (4) Added-value: Industrial logging operations harvest roundwood logs that are then transported to a centralized offsite-processing mill. Conversely, community and smallholder operations often seek to add value to the lumber onsite, transforming the raw material into finished or semi-finished products (Salafsky et al. 1998). This

PAGE 28

16 potentially provides greater income to local residents, such as the case of at least one certified community operations in Acre, Brazil that is producing artisan objects from locally harvested wood (pers. obs.). (5) Local reinvestment: In community-based systems, capital is generally reinvested locally and there is a greater long-term incentive for sustainability (Salafsky et al. 1998). In comparison, industrial timber enterprises tend to move profits to other localities or sectors of the economy, thereby reducing any potential social benefits from forestry activities (Salafsky et al. 1998). For example, in Oaxaca, Mexico, Bray et al. (2003) note community use of logging profits to encourage more benign forest enterprises (i.e., water bottling, ecotourism, and resin-tapping) and build more community assets (i.e., potable-water networks, schools, and clinics). And even though timber extraction has subsidized cattle pasture expansion in the Brazilian Amazon (Oliveira 2000), Summers et al. (2004) found that smallholder timber managers in Rondnia tended to clear less forest and had smaller cattle herds than their neighbors. (6) Adaptive management/traditional governance: Communities and smallholders sometimes develop adaptive management systems that permit local users to identify internal and external threats to the forest (Margoluis and Salafsky 1998, Salafsky et al. 1998). Some researchers have recognized this as a benefit of traditional governance, or locally-defined institutions and regulations that guide decision making processes (Banana and Gombya-Ssembajjwe 2000, Becker and Len 2000, Bray et al. 2003). For instance, Becker and Len (2000) found that the Yuracar of the Bolivian Amazon have clear, locally-defined rules regarding forest resources, upon which they can control both access and use of the forest at multiple scales. (7) Value of forest to future generations: Intergenerational values or investment in the forest for future generations also distinguish

PAGE 29

17 community timber enterprises from their industrial counterparts, a factor often overlooked (see Bowles et al. 1998) when discussing excessive production costs of community forests (Bray et al. 2003). In several studies conducted in the Brazilian Amazon, rural respondents identified silvicultural activities as positively impacting the future of their children and grandchildren (Walters et al. 2005). (8) Traditional ecological knowledge: Rural peoples who have often lived in the same geographic region for generations have accumulated considerable knowledge about the natural resource base through daily observations in their subsistence activities, usually demonstrating a comprehensive understanding of ecosystem linkages (Folke et al.1998, Berkes 1999, Calheiros et al. 2000, Berkes and Folke 2002). For example, it has been noted that despite the highly technical silvicultural expertise available in Finland, the main factor explaining the excellent forest health in that country is that conservation practices have been based on TEK passed from one generation of forest managers to another (Banuri and Apffel-Marglin 1993). Accordingly, TEK can be applied to a systematic forest resource management plan, providing details on timber quantity and location, regeneration requirements, and other requisites for successful implementation of RIL guidelines. (9) Commitment to forest conservation: Mexican communities in Oaxaca and Quintana Roo consistently logged below the authorized volume (some by as much as 37%), suggesting a willingness to reduce the volume of extraction when inventories indicate external threats to the resource base. Some of these same communities also prohibit or limit hunting in local forests, displaying an increasing commitment to biodiversity conservation (Bray et al. 2003) or perhaps, a clear understanding of sustainable game harvest limits. Other forest managers in Acre, Brazil

PAGE 30

18 have developed local indicators for advanced regeneration as determinants for harvest. By identifying a sufficient number of what they term as daughters and granddaughters of the tree(s) to be harvested, they acknowledge the importance of leaving several individuals of differing size classes of a particular species in the management unit, not just reproductive adults (P. Roth, pers. comm.). (10) Participatory monitoring: Local participants, especially those termed parabiologists or parataxonomists, have been instrumental in monitoring environmental impacts and planning formal conservation initiatives, and, in some cases, have actually improved the data quality of the academically-trained researcher (Janzen 2004, Sheil and Lawrence 2004). In one Brazilian case, local participants monitored the populations and phenology of selected fruit tree species in collaboration with professional biologists, eventually determining that the environmental impact of timber extraction was too great a risk to the long-term NTFP value of a particular fruit species (Shanley and Gaia 2000). These examples suggest that local forest dwellers are very aware of forest dynamics and capable of executing environmental monitoring (especially those post-logging surveys encouraged by the RIL paradigm) when provided with some capacity building and equipment (Sheil and Lawrence 2004). (11) Influence/funding from outside organizations: Particularly since the early 1990s, a network of community forests in the Brazilian Amazon have received financial and infrastructural support of a number of NGOs and governmental institutions (Stone 2003). This support has led to the voluntary third-party certification of at least 7 community operations in the region, fortifying the competitiveness of community forest management in a demanding market dominated by industrial giants. Indeed, two of the principal barriers faced by community and

PAGE 31

19 smallholder managers are the small scale production and consequent marketing difficulties, limitations that certification may be able to improve, primarily through access to a niche market (May 2002). Yet, Stone (2003) also warns of the increasing reliance of certified community timber projects on outside funding and technical assistance, mostly because of the complex requirements inherent with RIL adoption, such as investment in expensive equipment and the writing of management plans. To What Extent Does the RIL Model Ensure SFM at the Smallholder/Community Level? Prior to the recent emphasis on the RIL model, traditional smallholder forest exploitation in the Brazilian Amazon was based on low technology methods that produced minimal financial returns, but also resulted in low environmental impacts (Oliveira 2000). Indigenous inhabitants have been managing forests in the region since pre-Columbian times (see Heckenberger et al. 2003) for a suite of subsistence products and services (Summers et al. 2004), a pattern that has continued to the present. Rural Amazonian inhabitants typically integrate wood production into their existing swidden-fallow systems, resulting in a highly patchy landscape and allowing smallholders to maintain favorable environments for several different species (Pinedo-Vasquez et al. 2001). It was not until the early 1990s, amidst a deepening international concern for global deforestation and the creation of the Program for the Conservation of the Brazilian Rain Forest (PPG7) in 1992, that the formal establishment of SFM projects for community and smallholder forests emerged (Stone 2003), eventually leading to the first RIL project for privately-owned smallholder forests in the state of Par in 1993 (Amaral and Neto 2000). By the mid 1990s, the Brazilian federal government also began to take an interest in community-based SFM models and in 1998, the Brazilian Institute of the

PAGE 32

20 Environment and Renewable Resources (IBAMA) (the federal agency directly responsible for the execution of federal forest policy) implemented the PMFSimples (Simple Sustainable Forest Management Plan) protocol, allowing smallholders to manage up to 500 ha for timber (Stone 2003). It also enabled community forest initiatives to submit one management plan for all participating properties, rather than do this individually. Since 1995, at least 7 smallholder and community forest projects have been established in the region, under a wide range of property regimes (extractive reserves, national and state forests, indigenous reserves, and agricultural colonization projects) (Stone 2003). Community and smallholder interest in timber has been motivated in great part by a need to increase and diversify livelihoods, fueled in particular by low net returns for NTFPs (Kainer et al. 2003). However, as the sophistication and intensity of community and smallholder forest management systems increases, it has also become evident that silvicultural prescriptions appropriate to local conditions are needed, especially those that take into account a shortage of investment and human labor (Oliveira 2000). The Brazilian RIL model was initially developed for large scale industrial operations where investment potential is high (Oliveira et al. 1998). For example, since 1995, training in RIL techniques for Brazilian operations has been a primary focus of the Tropical Forest Foundation (TFF) and its Brazilian subsidiary, Fundao Floresta Tropical (FFT) (Holmes et al. 2002). This effort has resulted in the development of several model systems throughout the Amazon basin, but most of these examples are considered large-scale operations. Not surprisingly, these types of models have been eagerly adopted by community forest managers, even though

PAGE 33

21 the RIL ecological impacts on the overall forest structure at this scale are not well-known (Oliveira et al. 1998). In its original conception, one of the basic tenets of RIL mandated that logging should mimic the so-called gap-phase dynamics of the neotropical rainforest (see Brokaw 1982) by creating small-scale disturbances around the extracted crop tree (Hartshorn 1989). Natural small-scale disturbances (usually involving the death and subsequent fall of a neighboring dominant tree) permit access to limiting light (and other) resources. By attempting to mimic this type of disturbance during RIL operations, it is hoped that residual stand damage will decrease and that growth rates of surviving trees will increase. Natural disturbances, however, are qualitatively and quantitatively different from those logging-induced disturbances (Boot and Gullison 1995, Vitt et al. 1998, Dickinson et al. 2000). Boot and Gullison (1995) note that mortality rates of trees in forests with only natural disturbances are on the order of 1-2% (see Swaine et al. 1987), while conventional selective timber harvests can damage as much as 55% of the residual stems. Additionally, they point to other sources of residual stand degradation, such as soil compaction, erosion, and changes in drainage, that are not part of a natural small-scale disturbance. Although RIL techniques are designed to minimize this type of aggravated impact, some species of commercially-valuable trees require larger disturbances for reproduction than is mandated by minimal impact logging techniques, which usually emphasize low harvest intensities (Fredericksen and Putz 2003). Careful planning may maintain timber volume but the removal of scarce, high value logs could result in replacement by less important species, ultimately reducing forest biodiversity and dooming the second cutting

PAGE 34

22 cycle to timber not yet recognized by the market (Schulze 2003). In this case, the gap-phase management paradigm could be in direct opposition to the ecological role of catastrophic natural and anthropogenic disturbances in forests, conditions to which particular species have adapted over time (Deneven 1992, Bush and Colinvaux 1994, Fredericksen and Putz 2003). Fredericksen and Putz (2003) point to examples of neotropical trees that may not benefit from a low-impact management regime, such as Swietenia macrophylla, Bertholletia excelsa and Cedrela spp., species which appear to establish in large disturbed areas. Another of the underlying assumptions of many RIL guidelines is that advanced regeneration of valuable commercial timber species is abundant and exists uniformly throughout the stand (Schulze 2003). However, given the suite of differing growth rates, wood densities, population dynamics, and modes of seed dispersal, both between and within commercial species in the Amazon basin (Schulze 2003), it may be that knowledge of tree autecology is one of the greatest limiting factors in implementing SFM (Hammond et al. 1996, Guariguata and Pinard 1998). Currently, there is a lack of sufficient ecological information to help guide forest management decisions in the Brazilian Amazon, which is one of the reasons forestry methods have not advanced much beyond RIL (Schulze 2003). Yet, it is prudent to evaluate small-scale, RIL-based logging operations to date, and discuss how some of the basic tenets of RIL (pre-harvest forest inventories, selection of harvest and seed trees, minimum diameter cuts, low harvest intensity, skid trail planning, rotation cycles, and liana cutting) have worked at the smallholder level and how they can be modified.

PAGE 35

23 Pre-harvest Forest Inventories, Botanical Identification, and Selection of Seed Trees A full pre-harvest inventory provides managers with detailed information on distribution, quantity, and size of potential harvest trees and future crop trees (Schulze 2003). Unfortunately, for many small operations (as well as some large ones), 100% inventories are cost prohibitive, especially when the entire management unit (as is the case of those community management projects in the extractive settlement projects and extractive reserves in Amazonia) is as large as 300 ha. In at least one study in Quintana Roo, Mexico, local rural people identified time and finances as two of the biggest constraints for the inventory process (Lawrence and Roman 1996). In response, it has been common to inventory only the logging block for the following year, as the inventories are generally carried out one year prior to harvesting activities. Although this may eliminate the need to pay for a costly comprehensive inventory, it limits the managers ability to harvest with respect to species population biology (see Table 2-1). For example, in a 10 ha logging block chosen from a 300 ha landholding, several individuals of a particularly valuable species are identified during the 10 ha inventory. For management purposes, these individuals are then the sole representatives of that species within the 300 ha landholding, since a full inventory of adult trees has not been carried out. Many commercial tropical tree species are characterized by highly clumped spatial distributions (Hubbell 1979; Table 2-2) and are self-incompatible (Bawa et al. 1985), so harvesting even half of the inventoried individuals could potentially lead to inbreeding and genetic drift (Jennings et al. 2001). Ghazoul et al. (1998) demonstrated that the reproductive success of Shorea siamensis was limited by compatible pollen transfer at a reduced population density after logging. This Allee effect (see Allee 1949) can increase the probability of extinction,

PAGE 36

24 particularly in small populations. For those species occurring in widely spaced clumps, a reduced density in aggregations may result in a decrease in pollen exchange between clumps, severely limiting genetic variability within aggregations and thus, their ability to adapt to changing conditions (Ghazoul et al. 1998). When asked about this type of problem, local forest managers often respond that they know their forest well enough to have a general understanding of species distribution patterns (pers. obs.). This might be true in many cases, but at least one study conducted in Par, Brazil revealed that during Rapid Rural Appraisal interviews, local residents grossly overestimated the densities and distributions of game-attracting fruit species, presumably because they were favoring forest groves where the tree species occurred in high densities (Shanley and Gaia 2002). Even if the landowner or community cannot afford to pay for a 100% inventory, other forms of surveys (like rapid-assessment type inventories) should be tested to give the forest manager a more accurate idea of the availability of species and their distributions across the landholding. For example, in the ejido system of Quintana Roo, Mexico, Lawrence and Roman (1996) report that local inventory specialists prefer small circular plots as opposed to a 100% inventory design. Although these circular plots may lack some of the statistical accuracy of a full-scale inventory, they satisfy local needs for ease of implementation and available financial resources. Botanical identification during the inventory is crucial for making the most basic management decisions and should be considered one of the first steps in SFM. In Brazilian community forest settings, much of the identification is carried out by a forest manager or tree identifier (mateiro) from the community. Usually in these cases, initial

PAGE 37

25 identifications are recorded as common names. Thanks to the substantial botanical knowledge base in the community, most mateiros are able to identify species to common name with few difficulties, using a series of morphological features (i.e., bark, slash, and odor). The greater problem arises when the inventory sheet is sent back to the office of a local NGO or state agency and entered into a database. At this point, the common names are converted to scientific epithets based on lists (often created by a state or federal agency distant from the management region) that have been circulated between agencies. Too often, these lists do not correspond correctly to all common names, as the master list has been generated by a different set of mateiros. The assumption that these common names will then correspond with each other is highly optimistic, as even mateiros in the same community will often use different names for the same species. In Lacerda and Griffiths (2005) study from the Tapajs forest in Par, for example, more than 90% of common names were technically correct, but once the inventory sheet arrived in the office and was converted to corresponding scientific names, botanical accuracy was reduced to less than 40% correct. This type of error can have serious repercussions for both the selection of harvest trees and post-harvest population dynamics. The pre-harvest inventory and subsequent botanical identification have strong implications for the selection of seed trees. Some RIL guidelines require that mother trees, or seed trees, are maintained in the management unit at the appropriate spacing and density, based on means of dispersal, individual size, and tree diameter at first reproduction, as suggested by Guariguata and Pinard (1998). The seed tree concept does not elude forest managers on the ground. One landowner in a community timber management project in Acre, Brazil expressed his reluctance at cutting down large

PAGE 38

26 crowned individuals of Dipteryx spp. (even though such trees would bring an excellent price on the market), believing them to be good seed sources (as well as being concerned about the subsequent large gap size created when felled). He would instead take a smaller conspecific, or what is locally known as a daughter tree (F. Correa da Cunha, pers. comm.). Target mother tree densities are often around 3-4 trees ha-1, or (as in accordance with federal Brazilian law) 10% of the harvest compartment, but given that reproductively mature conspecific densities in many neotropical forests are much lower (Lieberman and Lieberman 1994, Guariguata and Pinard 1998), these target goals may be inadequate. As well, if individuals cannot be properly identified to species, their selection as seed trees remains dubious. For example, current community timber operations in Acre identify Dipteryx spp. to one common name, cumaru ferro, even though at least 7 species of this genus exist in the region (C. Baraloto, unpublished data). Consequently, in a hypothetical situation where 30 mature cumaru ferro in a 10 ha logging block have been left as seed trees, it is highly probable that this group entails several species that may be isolated from other conspecific reproductive adults. In general, the dilemmas of implementing a pre-harvest inventory, identifying scientific species, and selecting seed trees can occur at both the community and industrial scales. Brazilian federal law requires all companies to survey their logging concessions prior to harvest, regardless if they are certified or not. The difference for community and smallholder operations is that a pre-harvest botanical inventory will require long-term budgetary commitments not normally required on the part of a community-based organization or forest-dwelling individuals (Janzen 2004). Local parataxonomists, during

PAGE 39

27 the time of the inventories, are taken away from other responsibilities with families and subsistence activities, potentially compromising their situations at home. As such, they may need to be compensated financially by the forest management group, which can cause resentment among untrained community members who do not benefit directly from the botanical survey. Minimum Diameter Cuts The most common harvesting method in the tropics is selective logging, in which only the largest, most valuable individuals are removed from the forest (Webb 1997). One of the justifications used to support the practice of employing a minimum diameter cut is that smaller individuals of the same species will grow to replace the harvested trees. However, application of minimum-diameter cuts (often termed high grading, or taking the biggest and the best) is riddled with undesirable outcomes, and has thus been criticized as a short-term gain for long-term loss (Nyland 2002). As minimum diameter cuts are often defined by market demands rather than tree biology, Sist et al. (2003a) suggest that diameter limits be based instead on the structure, density and diameter at reproduction of target species. Minimum diameter cuts are not necessarily an inherent part of RIL, but their implementation is linked to the polycyclic cutting cycle, which is a common RIL recommendation. High grading typically dooms the next harvest cycle to small, poorly-formed trees of lesser commercial value or, at the very least, leaves inferior genetic stock to produce the next generation (Smith et al. 1997, Fredericksen and Putz 2003). For example, Vester and Navarro-Martnez (2005) point out that selective logging of the largest trees may deplete a population of the most favorable genotypes, eventually leading to genetic erosion. The extent to which this selection will affect populations depends on a number

PAGE 40

28 of factors, including harvest intensity, the contribution of the residual trees to regeneration, and the degree to which undesirable traits are genotypic rather than phenotypic (Jennings et al. 2001). As well, strong correlations have been drawn between tree size and fecundity (Chapman et al. 1992), so the removal of a stands largest individuals often eliminates the most prolific seed producers, important for both wildlife as well as regeneration (Thiollay 1992). Most tropical rain forests have more individuals below the minimum diameter cutting limit than above. When the reverse is true, the population is often assumed to be in decline (Jennings et al. 2001). One forest type representing the latter category is the arborescent bamboo (Guadua spp.)-dominated forest of southwestern Amazonia, which has been overlooked by the literature for management considerations (see Chapter 3). Griscom (2003) hypothesizes that if Guadua achieves dominance in managed forests, either the majority of tree stems will become competitively excluded, or a substantial proportion of the remaining stems will become deformed as a result of mass loading (referring to the added physical mass of the bamboo, which can displace aboveground tree structure (Griscom 2003)). He notes that it is not necessarily the inability of juvenile trees to germinate and establish in Guadua stands that distinguishes the reduced above-ground biomass in these forests, but rather the failure of seedlings and saplings to recruit into larger size classes. These factors present a few challenges to the minimum diameter cut, such as lack of advanced regeneration (one of the requisites for RIL) and removal of a substantial portion of what little canopy exists in the form of large trees (pers. obs.). As such, the stand dynamics of this particular type of forest might lead to an increased

PAGE 41

29 probability of high grading. Yet many smallholders, because they are limited to their landholdings, may be forced to harvest in these areas. Another major drawback of the minimum diameter cut affecting both large and small-scale operations is that knowledge of tree size at first reproduction (which is essential for maintaining adequate regeneration levels) is very limited in the neotropics (Guariguata and Pinard 1998). In the Yucatn forests of Mexico, for example, an original 60 cm dbh diameter limit for Swietenia macrophylla (later reduced to 55 cm) was based not on ecological constraints, but rather, market demand (Snook 1998). Even with the new 55 cm limit, it is assumed that it will take 75 years for the next generation of S. macrophylla to reach that minimum, an assumption that has been met with much skepticism (Vester and Navarro-Martnez 2005). Brazilian forest law requires that all harvested timber trees must be 45 cm diameter at breast height (dbh) (Schulze 2003). This is considered to be a fairly conservative lower limit by many forest managers, and some impose their own higher diameter limits on certain species, based on what they think is the ideal sapwood-heartwood ratio of the tree (F. Correa da Cunha, pers. comm.). Whereas a few species are characterized by little sapwood (e.g. Peltogyne spp.) and thus are considered worthwhile to harvest at diameters as small as 45 cm dbh, many species are not harvested unless they are much larger. In one community operation in Acre, Brazil, Tabebuia spp. and Hymenaea spp. must reach 70 cm dbh before they are cut, whereas other taxa, such as Dipteryx spp., must attain 80 or 90 cm dbh before they can be considered for harvest. This management strategy is also employed in operations that manage vast forest concessions, and in either case, such a minimum diameter limit could remove many of

PAGE 42

30 the largest trees on the site. Schulze (2003) suggests that by increasing the minimum diameter cut to 60 cm dbh, one runs the risk of removing the most vigorous individuals on site, not necessarily the oldest. In this scenario, it would be best to employ a proportional cutting across adult size classes rather than concentrate efforts on a minimum harvestable size, thereby ensuring the maintenance of fast-growing individuals for the next cutting cycle. Low Harvest Intensity Low harvest intensities (few trees harvested per hectare) are common in smallholder systems, and indeed, low intensity selective harvesting is attractive to those conservationists promoting SFM (Whitman et al. 1997). In small community management units in Acre, Brazil, the desire to maintain NTFP resources coupled with local investment in timber certification initiatives have encouraged communities to try and create as little damage as possible to the residual forest during logging operations (pers. obs.). This aspiration is also apparently reflected in Oaxaca and Quintana Roo, Mexico, where community forest managers generally log below the authorized volume (Bray et al. 2003). However, low harvest intensities associated with these smallholder timber operations may actually aggravate logging disturbance. Several studies have documented that as intensity of harvest increased, fewer skid trails and logging roads were required, resulting in a declining per-tree damage rate (Gullison and Hardner 1993, Whitman et al. 1997, Panfil and Gullison 1998). Panfil and Gullison, in their 1998 study of short term impacts of experimental timber harvest intensities in the Chimanes Forest of Bolivia, cautiously argue for an intensive (albeit at a fixed volume), localized harvest rather than an extensive extraction over a larger area. They suggest that the primary benefit of doing

PAGE 43

31 so is the reduction in mortality of smaller size classes due to skid trail construction and use. The tradeoffs involved with this scenario include the following: first, by logging one area of a large management unit more intensely, there is an increased chance of reducing local population levels. Second, even if the area damaged due to skid trail and road construction is decreased, there will be more concentrated disturbance in the treefall area in the form of large gaps, which are more susceptible to small-scale disturbances, such as windthrow and fire (Webb 1997). Third, the threshold values for the release of pioneer and vine species after such harvests is unknown for many forests, necessitating monitoring plots for this potentially site-specific information (Panfil and Gullison 1998). Fourth, the impacts of logging damage should ideally consider biodiversity maintenance in the event of a more intensive and localized extraction (Panfil and Gullison 1998). Tradeoffs associated with low harvest intensities will necessarily vary, depending on site and operational constraints. One of the primary factors that should be taken into consideration is the location and abundance of commercial species in a given logging block and to compare these values with other management units (if they exist) in the project. If considerable skid trail construction is required to reach one individual of a desired species, the destruction and cost necessary to harvest it may outweigh the benefit of acquiring the log. In this case, it may be possible to take another individual of the same species in a management unit better suited to its harvest (i.e. greater abundance of the targeted species or closer proximity), if there are multiple landowners harvesting simultaneously in one year as a community effort. It should not be assumed that just because an operation is extracting at a low intensity that it is having minimal impact on

PAGE 44

32 the forest. It is possible to have few commercial-sized trees injured during logging while simultaneously incurring substantial ground and smaller stem damage (see Chapter 3). Skid Trail Planning One way to minimize extraction damage is to reuse old skid trails, but this method can prove challenging for smallholders, depending on the management system being implemented. For example, in the Multiple-Use Community Forest Management Project in the extractive settlement of Porto Dias in Acre, Brazil, harvests are conducted in 10 ha (200 m x 500 m) logging blocks within individual landholdings (approximately 300 ha), with at least five landholdings participating in a given year. The majority of the first logging blocks (from 2000) in the participating landholdings were located near existing roads, which greatly facilitated the removal of stems from the logged areas and the associated construction of log decks. Logging blocks for the subsequent years were then positioned next to these initial logged sites so that old skid trails from previous years were reused. This is beneficial to the forest managers in that they do not need to construct an extra length of primary skid trail (although it does require them to open up these trails again every year), and it is favorable to the forest because the amount of superfluous skid trails is minimized. Using this method, though, means that future tree regeneration will never become permanently established along the length of these reused skid trails. It also requires the constant perturbation of the previous logging blocks. In comparison, the forest management project of So Luis de Remanso (another extractive settlement project in the state of Acre) utilizes 50 ha logging blocks, slightly reducing the need to reenter the management area in sequential years (depending on the number of members that remain in the project and the yearly execution of management plans). Instead of several different landholdings simultaneously participating in the

PAGE 45

33 harvesting (as is the case in Porto Dias), only one landholding (in general), or 50 ha, is logged per year. In comparison to the Porto Dias system, this setup may be more economically viable, as the forest managers usually harvest in just one landholding per year, so ground efforts are not dispersed. Conversely, in Porto Dias, if five landholdings are logged within a given year, the logging crew is compelled to maintain roads and construct/reopen skid trails in several sites (N. Marcondes, pers. comm.), adding extra cost and effort to the operation. Rotation Cycles One prerequisite for RIL is the implementation of growth and yield data for different management regimes to guide the choice of rotation cycle (Silva et al. 1995). In the Brazilian Amazon, the rotation cycle generally ranges from 30-35 years, based on an average diameter growth rate of 0.5 cm year -1 (Silva et al. 1995). Silva et al. (1995), however, caution that ongoing measurement and analysis of permanent plots over the next 30 years are needed before forest managers embrace an ironclad management prescription. They point to the variable diameter increments in any given class of trees, based in part on environmental conditions, growth habits, and genotypes. Another drawback of the Brazilian harvest rotation is the application of silvicultural recommendations generated in one region to another. By 1989, approximately 105 ha of permanent plots had been established in the Brazilian Amazon, but most of them are located in the state of Par (Silva et al. 1995). Meyer and Helfman (1993) point to the need for sustainable resource management plans based on an understanding of specific local conditions. Many of the Brazilian community timber operations are emerging in the western state of Acre (see Kainer et al. 2003), yet these management systems tend to base their diameter and volume increment projections on eastern Amazonian data. One

PAGE 46

34 exception is the permanent plot study that has been established in the Pedro Peixoto Settlement Project by the Brazilian Agricultural Research Corporation (EMBRAPA), which has based the smallholder management system on locally-generated growth and yield data. Drawing from the results generated by the project, Oliveira et al. (1998) propose that extraction of fewer trees per hectare at shorter intervals, in concert with silvicultural treatments, will maintain pre-intervention forest structure and biodiversity levels by creating a gap mosaic of different ages. In other words, a short rotation cycle (in this case, ten years whereby four hectares are harvested annually on a 40-ha stand) will substitute for the larger impact caused by more intensive interventions. Yet, even if site-specific growth and yield models state otherwise, a short rotation could severely degrade the quality of the resource base. Webb (1997) warns that although increased light levels due to removal of harvested individuals will generally promote the rapid growth of residual trees, repeated harvests over short rotation periods will potentially result in low levels of hardwood regeneration as well as a highly discontinuous canopy. Oliveira (2000), though, counters that damage levels from short interval cycles can be mitigated to a certain extent by reusing old logging roads and skid trails. As well, he makes the observation that the Brazilian SFM system is difficult to apply to the smallholder setting given that the required rotation cycles (20-30 years) make the clearing of forest for agriculture and pasture a more enticing option since the profits from those activities are much more immediate. Liana Cutting Proponents of RIL regard the pre-harvest cutting of lianas on crop trees essential if canopy and future crop tree damage are to be minimized and lianas are abundant (Fox 1968, Appanah and Putz 1984, Vidal et al. 1997, Gerwing and Vidal 2002, Pereira Jr. et

PAGE 47

35 al. 2002). On average, lianas connect each canopy tree to anywhere from three to nine others in Amazonian terra firme forest (Vidal et al. 1997). During tree felling, this liana connectivity inadvertently can bring down several other trees, creating a far larger gap than would occur if lianas associated with harvest trees were pre-cut (Appanah and Putz 1984, Vidal et al. 1997). This suggests that directional felling (a basic tenet of most RIL guidelines whereby harvest trees are felled in a specified direction to minimize residual stand damage), is less effective when interconnected vines are not eliminated (Fox 1968). Furthermore, as these uncut lianas fall with the crop tree, they typically resprout vigorously, thus competing with new tree recruits and reducing establishment success in gaps (Schnitzer et al. 2000, Gerwing 2001). Yet, there is little doubt that lianas are essential components of tropical forests, providing an important food source for wildlife species as well as supplying a canopy pathway for some arboreal fauna (Putz et al. 2001). Lianas also contribute to canopy closure after treefall and stabilize the microclimate of the subcanopy layer (Schnitzer and Bongers 2002). When asked about the disadvantages of cutting lianas, local forest managers in Acre, Brazil articulated concern for game species because many liana fruits are available when tree fruits are not (pers. obs.), an observation that is corroborated by several scientific studies (Heideman 1989, Morellato and Leitao-Filho 1996). These same forest managers have also observed that lianas constitute pathways between individual tree crowns for many animals (especially primates), something that has also been substantiated in the literature (Schwarzkopf and Rylands 1989, Bobadilla and Ferrari 2000). Consequentially, local logging crews from some Brazilian community forests have expressed reluctance at cutting lianas prior to the felling cycle (pers. obs.).

PAGE 48

36 Implementing this particular RIL recommendation might present a few quandaries for the community or smallholder forest manager who is looking to maintain more than just the next cycles timber yield, a consideration that may not manifest itself with an industrial commercial logging operation. Yet, not cutting lianas presents an obvious tradeoff, as they negatively influence growth, leaf production and sexual reproduction of host trees (Parren 2003). This is an important consideration, particularly in a landholding that is limited in size, and, consequently, in future crop tree abundance. Perhaps more importantly, though, is the shear risk that the logging crew takes when felling a tree that has not been freed of climbing vines (Amaral et al. 1998), a factor that is valid at both the smallholder and industrial scales. Parren and Bongers (2001) propose that the benefits of cutting or not cutting lianas prior to logging is heavily site dependent, so it may be that local forest managers must experiment with this facet of the RIL paradigm before making a final decision as to pre-harvest silvicultural prescriptions. Conclusions In this review, I have attempted to highlight some of the ecological constraints of RIL as practiced at the community and smallholder scales, mostly drawing upon examples from the Brazilian Amazon region. Many of the limitations outlined here can also be applied to the industrial commercial setting, such as the importance of a pre-harvest botanical inventory. However, a key distinction with smallholders is that these forest managers often extract timber from undersized management units, necessarily reducing the size and often the density of the populations from which they are harvesting. Ghazoul et al. (1998) point out that SFM (and presumably RIL) requires the maintenance of viable tree populations, but admit that minimum effective population sizes and

PAGE 49

37 densities are rarely known. Logging, even under the RIL model, can reduce population viability when conspecifics are isolated from each other (Ghazoul et al. 1998). Although RIL is, in general, ecologically beneficial to the residual stand, there is evidence that the magnitude of benefits varies, depending on which forest system is being exploited (Fredericksen and Putz 2003, Schulze 2003). Some logging guidelines outlined in this review are unlikely to change in the near future, simply because they are part of Brazilian federal law, such as the requisite rotation cycle and the minimum diameter cut. Nor perhaps should they, given the alternative options, such as reentering previously harvested stands after a short time or taking undersized trees. Given that many of these small scale timber operations must continue to function within the confines of their realities, I see the following points as potential recommendations for smallholder timber management following RIL guidelines. Logging Block Size and Implications for Pre-harvest Inventories A diversity of smallholder systems has been described in this review, illuminating the need to consider details of each one (such as logging block size) when planning timber harvests. In the state of Acre alone, several logging block sizes are used, from the four-ha harvest compartment in Pedro Peixoto, to the 10and 50-ha systems of Porto Dias and So Luis de Remanso, respectively. As logging operations are currently executed, I believe that the 50-ha system is best suited to maintain timber species population viability simply because managers implement a pre-harvest inventory in a larger unit, providing a more accurate assessment of species distributions than with a 10-ha block (see Table 2-1). Also at this scale, logging crews will not need to reenter the 50-ha area as frequently as is required in the Porto Dias system, in which skidders often return annually to previously harvested logging blocks, reopening primary skid trails. Also, by harvesting

PAGE 50

38 just one 50-ha unit per year, the logging crew is arguably exerting less effort in person hours and equipment usage, as compared to harvesting several non-contiguous 10-ha blocks in a given year. Even so, political, cultural and socio-economic reasons exist for choosing one system over another. As such, although harvesting a 10-ha block in one year may not be the most ecologically viable option, it is what certain communities have to deal with, along with all of the associated environmental disadvantages. The 10-ha logging block system is also likely to grow in usage, as it is reportedly being favored by new communities wishing to manage for certified timber in Acre (N. Marcondes, pers. comm.). This reality suggests the need for full scale inventories of the entire landholding projected for timber management to minimize potential dangers of inbreeding and inadequate seed dispersal. The measurement of genetic variation is impractical for forest managers, but the potential for genetic drift can be reduced by basic knowledge of species populations generated by a landscape scale inventory (Jennings et al. 2001). Failing implementation of a 100% survey for those trees above at least the minimum diameter cut, rapid assessment inventories should be considered. Botanical identification is a critical component of logging inventories, for which local parataxonomists are key partners. Before inventories take place, however, common names should be agreed upon, so that the same nomenclature is applied throughout the survey process. To ensure that these common names then correspond with their correct scientific counterparts, care must be taken to have samples identified with known herbarium examples, if possible. At the very least, care should be taken to not solely rely on databases generated by other organizations, whose own tree identifiers may also be

PAGE 51

39 using common names that differ from those being used by a local parataxonomist from the timber project. Monitoring Recognizing that smallholder and community forests must establish their own management strategies while building on RIL guidelines, I suggest that systematic post-harvesting monitoring become integrated into logging operations, so that the benefits of recommended practices (such as pre-harvest liana cutting) may be tested at the appropriate scale. I recognize that smallholders in the rural tropics have limited financial and technical resources, and that monitoring may be unfeasible for them, unless they perceive that proposed activities add to their livelihood productivity. Accordingly, Ghazoul (2001) concedes that given associated high costs, an adaptive management approach, entailing collaboration between biologists and local forest managers, would allow for sustainable management to be based on the technical skills, ecological knowledge base, equipment, and funds available to the managers. This collaborative approach to forestry requires considerable communication between local managers and professional forestry technicians and an assessment of existing local ecological knowledge and traditional management systems (Carter 1996). To this end, collaboration between smallholders, the agencies that support them and researchers who may be able to contribute technical expertise to the process should be fully encouraged. I recognize that working outside of traditional field sites in community settings obliges researchers and practitioners to approach forest management and monitoring differently, requiring them to consider social, political, and economic factors that affect the way that local peoples interact with the natural resource base. Beyond this consideration, we need to also think about the dynamics of the forest managed for a suite

PAGE 52

40 of products and services at the scale discussed in this review, providing communities and smallholders with valuable feedback on their management systems and further providing a link between biodiversity conservation and SFM. Moving beyond RIL It is indisputable that RIL methods are a vast improvement over the much more destructive and haphazard conventional selective logging practices that are, unfortunately, still the norm (Schulze 2003). Yet, forest management under RIL guidelines alone could be at odds with the SFM model, which demands forest management for a suite of goods and services as well as the maintenance of ecological functions (Fredericksen and Putz 2003). Arguably, it is this broader suite of outputs that many communities target in their forest management schemes. Rural communities depend on wildlife, NTFPs, watersheds, and soil properties for their livelihoods, underscoring the importance of looking beyond sustained timber production as the most important forest management objective. Even if seed sources and dispersal mechanisms remain intact, thereby (in theory) ensuring the viability of future timber regeneration (Baraloto and Forget 2004), the long term logging impacts on these other livelihood and conservation elements must not be overlooked. It seems particularly appropriate then to move beyond an emphasis on RIL guidelines for smallholders engaged in timber management, while still incorporating its basic tenets into practical application. It is highly probable that the so-called sustainability of intensive timber management may not be known until at least three cycles of harvest have been completed (Poore et al. 1989). This observation bolsters Simberloffs (1999) suggestion that any proposed silvicultural system designed to maintain biodiversity while producing timber should be treated as a hypothesis, due to limited empirical evidence supporting its

PAGE 53

41 viability. Until then, though, we must continue to focus on forest management research that is not just rooted in tree reproduction biology (Guariguata and Pinard 1998), but also consider the application of su ch investigations at the sma llholder and community levels, necessarily incorporating the local ecological knowledge that is often overlooked by academic researchers yet when it is acknowledged. In this way, we might be able to avoid the pitfalls of relying on RIL as proxy for SFM, especially in terms of the smallholder and community settings, which necessarily re quire considerations beyond the biological context.

PAGE 54

42 Table 2-1. Contrasting tree population parameters from inventories at two different spatial scales commonly used by smallholders. Shown are the results of a 50 ha inventory of commercial species stems 35 cm DBH from PAE So Luis de Remanso, Acre, Brazil, along with the range of values (5 and 95%) obtained by 500 simulations of randomly-placed 10 ha plots within the larger area. Species were included only if they were represented by at least 10 trees in the plot, and their botanical identification had been verified. Bold cases indicate that the distribution of values from the 10 ha simulations exceeded the actual value at 50 ha, suggesting that management decisions would change depending on the scale at which inventories are conducted. Species Common Name Total Density (ind ha-1) Harvestable Density1 (ind ha-1) 50 ha 10 ha 50 ha 10 ha Aspidosperma vargasii Amarelo 0.92 0.5 1.3 0.36 0.1 0.5 Myroxylon balsamum Blsamo 0.31 0.2 0.6 0.15 0.1 0.4 Tetragastris panamensis Breu-vermelho 4.62 2.6 8.6 0.82 0.2 2.2 Cedrela odorata Cedro-vermelho 0.87 0.3 1.7 0.67 0.3 1.3 Amburana cearensis Cerejeira 0.56 0.1 0.9 0.41 0.1 0.9 Apuleia leiocarpa Cumaru-cetim 0.97 0.7 1.7 0.79 0.5 1.3 Barnabydendron riedelii Guaribeiro 0.82 0.3 1.3 0.67 0.2 1.0 Tabebuia serratifolia Ip-amarelo 0.23 0.1 0.4 0.08 0 0.2 Hymenaea parvifolia Juta 1.21 0.7 2.1 0.64 0.2 1.2 1 refers to minimum size harvested, which varies among species from 45 60 cm DBH.

PAGE 55

43 Table 2-2. An example of spatial aggregation among neotropical tree species harvested by communities in Acre, Brazil. Shown are the minimum distance, if any, for each species at which individuals (stems 35 cm DBH) are significantly clustered (Ripleys L-statistic exceeding 95% confidence interval, based on 1000 bootstrapped values at 5 m distance intervals using Besags edge correction conducted in Ripley software package; Marcon 2005). Data are from a 50 ha inventory of commercial species from PAE So Luis de Remanso, Acre (Centro dos Trabalhadores da Amazonia, unpublished data). Species were included only if they were represented by at least 10 trees in the plot, and if their botanical identification had been verified. Species Common Name Spatial Distribution Clumped? Distance Aspidosperma vargasii Amarelo No Myroxylon balsamum Blsamo No Tetragastris panamensis Breu-vermelho Yes 6 m Cedrela odorata Cedro-vermelho Yes 8 m Amburana cearensis Cerejeira No Apuleia leiocarpa Cumaru-cetim No Barnabydendron riedelii Guaribeiro Yes 15 m Tabebuia serratifolia Ip-amarelo No Hymenaea parvifolia Juta No

PAGE 56

CHAPTER 3 EVALUATING LOGGING DAMAGE TO FUTURE CROP TREES IN ACRE, BRAZIL: IMPLICATIONS FOR MANAGEMENT OF A CERTIFIED COMMUNITY FOREST Introduction The guiding principle of sustainability dictates that a commercial forest should be managed in a way that limits as much undesirable damage as possible to the ecosystem. This is essential for not only ecological reasons, but economic ones as well (Mller 1998). For example, it is well known that unplanned (conventional) logging results in unnecessary damage to the residual forest and reduces the chances for future timber production (Pinard et al. 1995). Johnson and Cabarle (1993) estimate that anywhere from 26 to 75% of a given tropical stand of future crop trees (FCTs) are injured during conventional logging operations. Damage to residual stands is often correlated with logging intensity, which varies greatly, depending on spatial and temporal factors, access, equipment, terrain, climate, supplemental treatments, and market acceptance of lesser-known species (Metzger and Schultz 1984, Sist et al. 1998, Putz et al. 2000). The most common harvesting method in the tropics is selective logging, whereby only scattered trees of sufficient economic value are extracted from the forest. In unplanned operations, selective logging has been observed to cause extensive canopy cover removal (Johns et al. 1996, Webb 1997, Whitman et al. 1997, Pereira et al. 2002), soil compaction (Whitman et al. 1997), long-term changes in tree species composition (Thiollay 1992), and reduced faunal diversity (Johns 1991, Thiollay 1992). Accordingly, 44

PAGE 57

45 harvesting techniques and intensities need to be controlled to reduce residual stand damage (Webb 1997). Controlled selective logging, or reduced-impact logging (RIL), utilizes methods that minimize damage to the residual stand, such as pre-harvest inventory, mapping, directional felling, vine cutting, and planning of skid trails, log decks, and roads (Uhl et al. 1997), potentially avoiding some of the post-logging damage noted by other authors (Pinard and Putz 1996, Johns et al. 1996). RIL techniques have been shown to reduce damage to residual stands in mixed dipterocarp forests in Southeast Asia by as much as 30-50% (Pinard and Putz 1996, Bertault and Sist 1997, Sist et al. 1998). Furthermore, RIL methods have demonstrated an increased efficiency of logging operations and a reduction in the amount of timber wasted during a typical harvest (Barreto et al. 1998, Holmes et al. 2002). Even with implementation of RIL methods, sustainability of harvesting will ultimately depend upon advanced regeneration to replace the larger size classes during the next cutting cycle (Putz et al. 2000). This assumption is challenged by the damage incurred by the residual stand during logging operations. For example, Jackson et al. (2002) report that RIL operations in a FSC-certified timber concession in Bolivia killed or severely damaged an average of 22 trees ( 10 cm dbh) for every one extracted. In the same forest, Krueger (2004) found that flagging reduced damage to FCTs by 20% in felling gaps and by 10% along skid trails receiving 2-10 skidder passes, demonstrating that marking, or flagging, of FCTs can reduce damage to the residual stand by improving visibility during on-the-ground operations (Dykstra and Heinrich 1996). Currently, general Brazilian forest management practices do not employ this method.

PAGE 58

46 Mechanisms underlying these residual stand impacts need to be understood in the context of specific local conditions (Sheil and Van Heist 2000). Commercial species distribution and human exploitation varies greatly between neotropical regions, supporting the argument that silvicultural treatments appropriate in one location will not necessarily be applicable in another (Putz 1996). For example, Webb (1997) points out that difficulties encountered in Carapa nicaraguensis forest swamp in Costa Rica required management techniques not employed in other logging operations. Given the need to develop and evaluate site-specific management strategies that will mitigate the negative effects of timber harvesting, it is essential to look at several variables across locales. One forest type on which little has been written about in the management literature is the bamboo-dominated forest of southwestern Amazonia, which covers almost 180,000 km2 (Griscom and Ashton 2003). This forest is characterized by a mix of structurally heterogeneous stands without bamboo and patches of trees scattered within a canopy dominated by dense stands of Guadua species (with some species of Guadua reaching heights of 30 m) (Griscom 2003). Similar to other mast seeding bamboo genera (Janzen 1976), Guadua spp. undergo a single synchronized reproduction event followed by synchronized mortality, at 25-30 year intervals (Nelson et al. 2001). At least two species, G. sarcocarpa and G. weberbaueri have been linked with reduced forest basal area as well as tree species richness (Silveira 2001, Griscom 2003). This phenomena is strongly linked to the species ecology: branches of G. sarcocarpa and G. weberbaueri culms are endowed with curved barbs which act as grappling hooks (see Fig. 3-1), enabling the bamboo to be partially or fully supported by trees for vertical growth (Griscom 2003),

PAGE 59

47 weighing them down, and increasing the likelihood of blow-downs (Griscom and Ashton 2003). In concert with the common tree blow-downs in this region, Griscom and Ashton (2003) propose that low recruitment of trees into the larger size classes explain the persistence of Guadua domination, rather than the catastrophic disturbances of fire and agricultural intervention cited by other authors. If Guadua achieves dominance in managed forests, commercial trees may become competitively excluded and a substantial proportion of the remaining stems may become deformed, due to mass loading (Griscom 2003). The low visibility in some areas of the Guadua dominated forest also presents an impediment to sawyers trained in directional felling (F. Correa da Cunha, pers. comm.). Combined, these factors present some unique challenges for silvicultural management and timber extraction in Guadua-dominated forest. This study evaluated logging damage to FCTs in a certified community forest in Acre, Brazil. The main objectives of this investigation were (1) to determine the extent of area disturbed (canopy gaps and skid trails) during a small-scale certified logging operations, (2) to determine a difference in damage incidence to FCTs between forest with and without bamboo (Guadua), (3) to determine if likelihood of FCT damage is increased when lianas are present, (4) to determine to what extent harvesting can be conducted more intensely (m3ha-1) without incurring greater residual stand damage in general and damage to FCTs 20 cm dbh in particular, and (5) to determine to what extent marking of FCTs diminishes damage and/or mortality of FCTs (20 cm dbh) during logging. Site Description The Porto Dias Extractive Settlement Project (PAE) (S 1000, 9, W 6646,4) is situated in the northeastern corner of the Brazilian state of Acre, in the

PAGE 60

48 municipality of Acrelndia. It is bordered by various rubber estates, cattle ranches, and colonization projects (see Fig. 3-2), and is separated from Bolivia by the Abun River (Stone 2003). The site is defined by red-yellow latisols of low fertility and relatively flat topography, with an annual precipitation of 1890 mm yr-1, most of which falls between November and March. As of 2001, approximately 97% of the reserve was still forested (CTA, unpublished data), but it is one of the last large contiguous pieces of forest in the municipality (N. Marcondes, pers. comm.). As such, it is under constant pressure from the influences of agricultural conversion within the PAE as well as illegal clearing on its borders. Older residents recall Swietenia macrophylla (mahogany) being logged in the region and transported via the Abun River (Stone 2003), but mahogany is currently extremely rare in the PAE. Three tropical humid terra firme forest types are generally recognized in the reserve: open canopy forest; open canopy forest mixed with arborescent bamboo (Guadua spp.); and closed canopy forest. Three species of Guadua are found in the study site: G. weberbaueri, G. sarcocarpa, and one as-yet unnamed species, which occurs in riverine forest along the Abun River (M. Silveira, pers. comm.). At the end of 2004, when logging operations commenced, G. sarcocarpa underwent a monocarpic dieoff that affected much of the study area. Both G. weberbaueri, and G. sarcocarpa are thought to be synchronized on a 30 year masting cycle across large areas (100-100,000 km2) (Nelson et al. 2001). The two species exhibit similar autecological characteristics, with an apparent lack of seed dormancy, germination occurring in both gaps and non-gaps, and partial shade tolerance among seedlings (Griscom 2003). In at least one study in Madre de Dios, Peru, G. weberbaueri demonstrated a lower mean stem diameter (cm) in comparison with G.

PAGE 61

49 sarcocarpa (4.04 vs. 6.67, respectively), but exhibited higher stem density (stems/100 m2) (34.2 vs. 23.8) (Griscom 2003). The reserve encompasses approximately 22,145 ha, and is divided among colocaes, or individual family landholdings constructed around pre-existing rubber trails (approximately 300 ha each). There are around 100 registered families who live in the PAE, in addition to an estimated 20 families who are squatting illegally in the reserve (Santos 2000). Traditional family livelihoods in PAE Porto Dias include forest extraction, hunting, subsistence agriculture, and small-scale animal husbandry (Stone 2003). Cash was customarily earned from rubber and Brazil nut harvests, but in recent years rubber tappers have been turning to logging as an alternative income source. The Center for Amazonian Workers (CTA), a non-governmental organization based in Acre, has been instrumental in developing the management framework of the Multiple-Use Community Forest Management Project. The project was initiated in 1995, but the first timber harvest was not harvested until 2000. SmartWood certified the logging operation using Forest Stewardship Council (FSC) criteria in 2002, distinguishing it as the second FSC-certified community timber management project in Brazil. FSC certification requires that the project implement a special monitoring program in the bamboo-dominated forest, as there is concern that logging damage may be greater there than in forest lacking bamboo (N. Marcondes, pers. comm.). Additionally, Guadua-dominated stands have been identified as important habitat for many valuable wildlife species, making them focal areas of interest for conservation efforts (Griscom 2003). Forest Management and Harvesting Operations Landholdings (approximately 300 ha each) of reserve residents serve as the timber management areas, and therefore do not favor the compartmentalization typical in

PAGE 62

50 Brazilian industrial timber operations (Braz and Oliveira 1996). In compliance with federal regulation, 10% of the landholdings may be cleared for agricultural activities and 5% must be set aside for preservation (no harvesting of trees). Thus, most residents have at least 255 ha available for timber management, result in a 25-year logging rotation (255 ha/10 ha yr-1= 25.5 years). The original ten members of the project all harvested 10 ha per year, but within a few years after the first timber harvest, it was decided that only five landholdings per year would be harvested. In the latter case, project members began harvesting every two years in their respective landholdings. However, given that some families have since dropped out of the timber project, other members have still harvested in sequential years (i.e. more than one or two years in a row), further confounding the question of the cycle length. Inventories and mapping of all tree species 35 cm dbh are carried out one year prior to harvesting in 10 ha blocks within each landholding to be logged; as of 2005, though, a 100% inventory of the entire property had not been executed in any of the participating landholdings. After selecting the crop trees, the project must submit its logging plan to the Brazilian Institute of the Environment and Renewable Resources (IBAMA), which will then approve or disapprove the plan, based on an acceptable basal area removed, desired residual stand species distribution and crop tree proximity to water sources. Only trees 45 cm dbh may be legally harvested. Logging is generally carried out in the driest part of the year, from July to November. Twelve major timber species are presently exploited (see Table 3-1), with a low intensity of cutting (1-3 trees ha-1 or approximately 10 m3 ha-1). Since the timber project members began logging biannually, the total annual harvest for the project is around 500 m3, or 100 m3 landholding-1. Sawn

PAGE 63

51 board and artisan products from the community sawmill are primarily sold to the domestic certified market in Southern Brazil. The project does not market logs. Members of the timber project (as of August 2005 there were seven participating families) conduct most of the logging activities. These techniques include inventories, selection of crop trees, assessment of defects in crop trees, directional felling, and skid trail planning, RIL skills for which they have been extensively trained over the last ten years by the Technological Foundation of Acre (FUNTAC) and the Tropical Forest Foundation (FFT) in Par, and for which they now train other forest managers. The only activity that they do not directly participate in is skidding logs to the sawmill, a process contracted to heavy equipment owners and operators from outside the community. During logging operations, the landowner typically assists the construction of skid trails by flagging the path that the skidder or tractor operator will use. Methods Future Crop Tree (FCT) Selection Fifty-two tree species were selected for the study, based on their potential importance as commercial timber sources (see Table 3-1; Ribeiro et al. 1999, Lorenzi 2000). Due to external market demands, however, only about 10-12 species are exploited on a regular basis in Porto Dias, most commonly Amburana cearensis, Cedrela spp., Peltogyne spp., Dipteryx spp., Apuleia leiocarpa, Tabebuia spp., Aspidosperma vargasii, and Hymenaea intermedia. For purposes of this study, the definition of FCTs was broadened to account for all species of commercial value, both current and future (see Table 3-1). When proper field identification was questionable, species verifications were conducted at the Federal University of Acre herbarium in Rio Branco, Acre, Brazil, but some trees were only identified to genus (see Table 3-1).

PAGE 64

52 Experimental Design and Pre-logging Inventory The greatest amount of damage to FCTs during logging occurs in areas closest to felled crop trees and skid trails (Johns et al. 1996). Therefore, observations of logging impacts were concentrated in a 50 m radius (hereafter referred to as zone of impact) around each designated harvest tree in each of the 10 ha (200 x 500 m) logging blocks in four landholdings scheduled to be harvested in 2004-2005 (Barrinha 1, Barrinha 3, Palestina, and So Pedro). This number was eventually reduced to three, as the rainy season began before So Pedro could be logged. A full inventory of all commercial FCTs 20 cm dbh was conducted in the zones of impact. Sample sizes in the logging blocks (i.e. the number of zones of impact) depended on the abundance of crop trees, with a total of 18 zones of impact and 161 FCTs in Barrinha I, 20 zones of impact and 216 FCTs in Barrinha III, and 12 zones of impact and 109 FCTs in Palestina, including some overlap between zones. In 2004, before logging operations commenced, the following data were noted for FCTs: species, location (x, y coordinates), dbh (1.3 m or above the buttresses); total estimated height; qualitative trunk and crown quality (good, tolerable, or inferior); crown position (dominant, intermediate, suppressed); presence in bamboo-dominated stand (yes/no), and presence of at least one liana to the canopy (yes/no). Area Disturbed All primary and secondary skid trails and treefall gaps were mapped using a compass and meter tape. The size of treefall gaps was estimated using the center-point system (modified from Runkle 1992). From a point in the approximate center of each treefall gap, compass angles in eight different directions were measured to the edge of intact canopy (Johns et al. 1996). This information was then entered into ArcGIS 9.0

PAGE 65

53 (ESRI, Redlands, CA), to calculate area of the resulting polygon. The ground area transformed by the total length of skid trails was determined using modified methods from Johns et al. (1996), with skid trail width being measured at the beginning and end of each straight segment of trail. Trail width measurements were also taken at each bend to account for potential bulldozer maneuvering. This information was then combined with total skid trail length (estimated using pre-harvest inventory maps) in ArcGIS 9.0. FCT Location in Bamboo (Guadua)-dominated Forest Zones of impact (both in areas of high and low harvest intensity) naturally fell in locations where bamboo was present and in areas where it was absent. Location in Guadua-dominated stands was defined as at least one culm of Guadua spp. within 2 m of the FCT trunk. Liana Presence To test the effect of presence of lianas on FCT damage probability, liana occurrence on a FCT was denoted as at least one vine connecting to the canopy. Harvest Intensity To detect the impact that a range of timber harvest intensities could have on FCTs in Porto Dias, zones of impact were established in locations scheduled to receive a locally high intensity of cutting (3 trees ha-1), and also in areas of locally low intensity cutting (1-2 trees ha-1). This design allowed for establishing a harvest intensity gradient in each logging block, but did not necessarily allocate the treatments evenly across the logging blocks, as the abundance and distribution of crop trees varied greatly between landholdings.

PAGE 66

54 Marking Treatment To test the effect of marking, about half of all inventoried FCTs in each of the four logging blocks were marked. All FCTs within the zone of impact of a tree to be harvested were either marked or left unmarked, stratifying the treatment between regions of low and high crop tree density (see Figs. 3-3-5). Marking entailed a single band of orange paint 20 cm wide at approximately 1.7 m above the ground (adapted from Krueger 2004). Initially, the random assignment of the marking treatment was considered when designing this experiment. However, it was decided that the highly visible presence of even one marked FCT in the vicinity of unmarked FCTs could induce the logging crew to alter their activities in some way as to favor neighboring FCTs. This potential source of bias suggests that statistical independence of the FCT treatments would have been difficult to achieve even with this type of experimental design. Assessment of FCT Damage Inventory data were used to construct a spatial database in ArcGIS 9.0. For each FCT, distance to nearest felled tree, nearest gap edge and nearest skid trail were measured. Harvest intensity (trees ha-1) was estimated for each FCT as the volume exploited in the surrounding 1 ha (circumscribed by a radius of 56.4 m). Local FCT density was estimated as all FCTs 20 cm dbh within a 25 m radius of a given FCT. This measurement was included to control for variation explained by local density when testing for other factors such as marking. Seven months after logging activities were completed, in June and July, 2005, all previously inventoried FCTs were revisited and classified according to levels of damage or mortality (see Table 3-2).

PAGE 67

55 Statistical Analyses Quantifying FCT damage was considered for this study. A logistic regression analysis was used to test the binary response variable (damaged/not damaged) by treating each FCT as an independent statistical unit to which a series of independent variables was assigned: local harvest intensity (continuous); distance to nearest felled crop tree (continuous); distance to nearest skid trail, (continuous); distance to nearest gap (continuous); neighborhood FCT density (continuous); presence of bamboo (binary); marking (binary); and presence of lianas (binary). A regression analysis was performed to determine the influence of crop tree dbh on felling gap area. Differences were considered significant at P 0.05. Results Overview of the Logging Operation The mean tree ( 35 cm dbh) basal area was determined to have the following values for the three landholdings: 7.7 m2 ha-1 in Barrinha I, 8.3 m2 ha-1 in Barrinha III, and 8.8 m2 ha-1 in Palestina. All harvesting occurred between October and November 2004, and a bulldozer (Valmet 128) was used to extract logs. Two community-based sawyers participated in the operations: one felled trees only in Palestina and the other cut in both Barrinha I and Barrinha III. No new roads were created to facilitate logging operations, and pre-existing skid trails in logging blocks from the projects previous harvesting activities were reopened to access new areas. All three logging blocks were located close to existing roads, so boles were eventually skidded to log decks adjacent to these routes. A total of 22.3 ha of 30 ha were sampled in the three logging blocks. Forty five trees were extracted from all of the landholdings, with 38 actually cut and 7 that fell

PAGE 68

56 under natural circumstances at least one year prior to harvesting, for a total harvested volume of 237.7 m3, or approximately 7.9 m3 ha-1. Felled tree (n=38) dbh (cm) ranged from 60-114, with a mean of 71.7 (s.d. 14.73). Average harvested tree density was approximately equal between the three logging blocks (see Table 3-3), suggesting that any differences in damage incurred by FCTs was not due to differences in logging intensity (Johns et al. 1996). Some of the trees originally selected were not cut for various reasons (usually they were found to be hollow prior to harvesting), so loggers harvested several substitute trees. As a result, some of the trees designated as experimental FCTs the previous year were not affected by logging operations, as they were far from any skid trails or felling gaps ( 60 m from felling gap, cut stump, or skid trail). These trees (a total of 18) were not considered for the analysis. At the same time, damaged trees that were near the newly substituted crop tree sites were not tallied after harvesting, because no pre-logging data were gathered for them. FCT Damage A total of 70 FCTs out of 468 were damaged during logging operations, or approximately 15% of the experimental population. The most common damage types were slightly damaged crowns, followed by scraped boles and roots, classified as minor damage (see Table 3-3). Tree felling operations damaged a greater proportion of trees than skid trail construction and use, but the results were not statistically significant. Marking did not significantly reduce overall FCT damage in the Porto Dias operations (see Table 3-4). Furthermore, the incidence of FCT damage levels was not higher in Guadua-dominated forest, nor was there a significant interaction between a FCTs presence in a Guadua-dominated stand and the marking treatment (interaction term in Table 3-4). Fifty-six percent of the inventoried FCTs carried lianas, but damage

PAGE 69

57 probability due to liana presence on the FCT was not highly significant (P = 0.10), but many damaged FCTs had at least one large liana attached to its crown (67%). The closer a FCT was to a skid trail or logging gap, the greater the likelihood of it being damaged (see Table 3-4). In contrast, distance to cut stump and local FCT neighborhood density had no significant effect. Harvest intensity (see Table 3-4) also did not influence the probability of FCT damage, implying that in this study, the probability of FCT damage did not increase with the number of trees harvested in the surrounding hectare. Area Disturbed The percentage of total sampled area affected by harvesting was 28.9%, with the majority in felling gaps (16.8%). Average area per felling gap amongst all three logging blocks (mean s.d.) was estimated to be 393 m2 181. Regression analysis revealed a significant positive effect of crop tree dbh in single tree gaps area (see Fig. 3-6). Skidding damaged 11.8% of the total surface area. Discussion Timber harvesting, as conducted by the Multiple-Use Community Forest Management Project in PAE Porto Dias, had minimal impacts on residual stands, at least on commercial individuals 20 cm dbh. Additionally, potential damage was avoided by not having to construct new log landings and log roads. Even so, the high values reported for area damaged, both skid trails and felling gaps, merit analysis (see Table 3-3). The values for felling gap area per tree cut were in some cases almost twice that of other reported neotropical studies (see Johns et al. 1996). In many instances, increased gap sizes can be explained by the size of the trees that were felled. For example, one of the most valuable taxa in Porto Dias, Dipteryx spp., comprised some of the largest

PAGE 70

58 individuals amongst harvested trees (up to 114 cm dbh), contributing to the strong positive correlation between logging gap size (m2) and dbh of crop trees (see Figure 6). In the case of Palestina, more than 60% of the harvested trees were 70 cm dbh, explaining in part why this particular logging block had the greatest mean gap size (433 m2) amongst the three landholdings. When considering exceptionally large trees for harvest, forest managers might do well to consider the tradeoffs between creating a very large gap (thereby potentially increasing fire and windthrow risk) and the substantial profit for the sake of the wood. It is also worth noting that only those species valued for their merchantable timber were inventoried for the experiment. For example, locally important NTFP species such as Bertholletia excelsa and Hevea brasiliensis were not tallied during the original inventories, but during the post-logging survey, damage to several trees of these species was observed. For future studies, it would be prudent to address some of the issues of impact of timber extraction on other livelihood activities, such as NTFP collection, particularly in such regions as the Brazilian Amazon where many communities are attempting to integrate a suite of income-generating activities. Commercial logging, with potentially high financial yields, is unlikely to disappear in community forests. Nevertheless, interest persists in the harvesting of non-timber products such as Brazil nuts, rubber, and Copaifera and Carapa oils, species that may be adversely affected by timber harvests. When considering these factors, can timber harvesting and other livelihood strategies be made more compatible and if so, what are the tradeoffs (see Salick et al. 1995, Romero 1999, Menton 2003)?

PAGE 71

59 FCT Damage in Guadua-dominated Forest Given the low visibility in Guadua-dominated forest, I expected that FCT damage rates would be higher in these areas, but results proved otherwise. There could be a number of explanations for this outcome: even though poor visibility in Guadua groves has been cited as an impediment in the felling of large trees because neighboring FCTs are not always easily seen, FCT damage might be minimal simply because there are fewer FCTs to damage, as demonstrated by the low basal area estimates cited by other authors (Silveira 2001), or, more specifically, by the low representation of smaller stems in those same estimates. Guadua frequently outcompetes tree seedlings and saplings such that they fail to recruit into larger size classes (Griscom 2003). Therefore, one could conclude that there are comparatively more FCTs from the lower diameter classes to damage in forests without bamboo, or at the very least, there are more trees that operators must avoid when skidding or felling. It is also uncertain as to how the synchronized mortality of G. sarcocarpa may have directly or indirectly affected felling operations, although visibility may have improved in these stands because mature bamboo culms were drying out and falling to the ground in some areas. Even though I did not find that selective logging increased the probability of FCT damage in Guadua-dominated stands, the potential for other types of long-term residual stand damage warrants future investigations. We observed dramatically increased light levels in post-logging bamboo-dominated forest, possibly attributable to the loss of the biggest canopy trees in the block, of which there were few to begin with. There are a few potential problems associated with this. First, by opening large areas during harvesting, the chances of repeated small-scale disturbances, such as windthrow and fire, are increased (Webb 1997). Given that Guadua benefits from intermediate and large canopy

PAGE 72

60 disturbances (Griscom 2003), once a gap is opened, bamboo may proliferate and competitively exclude the majority of tree stems. One example of this phenomenon is Whitmores (1984) observation of the reduction in value of Malayan forests that were invaded by arborescent bamboo after unplanned logging operations. The high light intensity characteristic of large gaps might also select for a suite of different species as compared to small gaps (Denslow 1980), a dilemma if the regeneration of small gap-preferring species is desired by the forest managers. In general, Guadua-dominated stands are already characterized by a discontinuous canopy, which may amplify the effects of openings created by logging activities (i.e. increased light levels). As such, for future operations, forest managers may want to select crop trees based on the presence of large neighboring trees, ensuring that some sort of basic canopy structure is kept intact around the logging gap if smaller openings are desired. Another concern in this forest type is the potential for high grading, or the type of harvesting that selects for the largest and most valuable individuals. One of the underlying assumptions of RIL is that advanced regeneration of valuable commercial timber species exists uniformly throughout the stand so that future harvests are not compromised by the present one (Schulze 2003). Yet in Guadua-dominated stands, the lack of advanced regeneration potentially increases the chances of having small, poorly-formed trees of lesser commercial value dominate the next harvest. For the operation described in this paper, FSC certification standards, and indeed, Brazilian federal law, require that forest managers leave at least 10% of the harvested species populations as seed trees. This perhaps ensures harvests for generations to come, but it does not guarantee a sustainable timber yield for the immediate second cycle. As well, even

PAGE 73

61 though Brazilian federal law imposes a minimum diameter cut of 45 cm dbh, many species in this operation are not harvested until they are much larger (see Chapter 2), since taxa such as Dipteryx spp. are characterized by high levels of sapwood and thus attain commercial value at a greater girth. As a result, there is often an understandable temptation to take the biggest and the best individuals from a site, although at least one landowner in Porto Dias admitted hesitation in taking large Dipteryx spp. individuals, as they are often the most prolific seed sources (F. Correa da Cunha, pers. comm.). The complexities associated with timber harvesting in Guadua-dominated forest may require a much different approach to silvicultural techniques than developed in other parts of Amazonia, and underscore the need for future monitoring of harvesting activities and residual stand dynamics as well as extreme caution in the removal of the largest trees on site. Impact of Lianas on FCT Damage Lianas connecting the crowns of trees will often cause a domino-like effect during felling operations (Vidal et al. 1997), although few adjacent trees were completely brought down by felled trees during this operation. Other neotropical studies (see Fox 1968, Appanah and Putz 1984, Johns et al. 1996, Vidal et al. 1997, Pereira Jr. et al. 2002, Gerwing and Uhl 2003), have demonstrated increased damage during logging operations when vines were present. Although I expected vines to increase FCT damage, I did not observe any statistically significant effects at the level, though given results in other studies, it is probable that the magnitude of the outcome is consequential from a management standpoint (Steidl and Thomas 2001). It was evident from post-logging observations in the field that parts of crowns were pulled down by bulldozers at they

PAGE 74

62 moved through the forest. Undoubtedly, if lianas associated with FCTs had been cut at least one year prior to harvesting, some of this damage could have been avoided. Still, Porto Dias forest managers expressed reluctance to cut lianas prior to the felling cycle (pers. obs.), citing their importance as a food source for wildlife species. Yet, it has been shown that lianas negatively influence the growth, leaf production and sexual reproduction of host trees (Parren 2003), and when not cut prior to harvesting operations, increase the potential danger for logging crews (Amaral et al. 1998). Parren and Bongers (2001) propose that the benefits of cutting or not cutting lianas prior to logging is site dependent, so it may be that local forest managers must experiment with this particular RIL guideline before making a final decision as to pre-harvest silvicultural prescriptions. Impact of Harvest Intensity on FCT Damage Local harvest intensities in the logging operation described here removed 1-4 trees ha-1, seemingly with no increased levels of residual stand damage in the 20 cm dbh class as the maximum ceiling of harvest intensity (4 trees ha-1) was reached. This type of low-intensity selective logging is attractive for conservationists espousing the implementation of sustainable forest management as a way to limit canopy cover loss and protect ecosystem function and biodiversity (Whitman et al. 1997). Even so, Gullison and Hardner (1993) have pointed out that as the number of trees removed increases, fewer new skid trails are required (although the damaged area per tree due to logging gaps remains constant). In the case of the operation described here, even though there was no evidence supporting increased levels of FCT damage as harvest intensity increased, the average area (m2) damaged per tree extracted due to skid trail construction was quite high (see Table 3-3) as compared to other neotropical studies, relative to the number of trees extracted per hectare. For instance, Johns et al. (1996) found that unplanned logging

PAGE 75

63 operations using a bulldozer to skid trees damaged 119 m2 per bole removed in relation to skid trails (with an average harvest intensity of 5.6 trees ha-1). The high levels of area damaged per tree extracted in this study suggest that there might be a tradeoff associated with trying to harvest isolated trees across even a relatively small area of 10 ha, both in terms of cost and as a function of area and trees damaged. One way to offset this effect of low intensity harvests would be to increase localized harvest intensity at a fixed volume. The primary benefits of this method would be reduction in mortality of smaller size classes as well as fewer skid trails, not necessarily reduction in mortality or damage to FCTs (Panfil and Gullison 1998). Another consideration is that large gaps allow for more solar radiation to reach the forest floor, drying remnant organic matter such as leaves and increasing the amounts of fuel in post-logging sites (Johns et al. 1996). Furthermore, the release thresholds of pioneer species, lianas and bamboo in the large gaps resulting from this scenario are unknown (although conditions might be ideal for the regeneration of such species as S. macrophylla) (Panfil and Gullison 1998). As such, the implementation of an increased localized harvest intensity is risky, especially for a landowner limited to a management unit of 300 ha. Impact of Marking Treatment on FCT Damage Despite positive results of marking FCTs in at least one other study (see Krueger 2004), marking did not appear to reduce FCT damage in the Porto Dias timber operations. That FCT marking had no effect on damage levels along skid trails was surprising, given that machine operators were from outside the community and therefore not as familiar with the forest. Even so, the owners of the landholdings were responsible for flagging the path of the skid trails prior to extraction, so as to avoid any damage

PAGE 76

64 associated with skidding. This precaution could explain why marking had no effect on FCT damage along skid trails in this particular study, as bulldozer operators were already able to move without hesitation, especially when creating secondary and tertiary trails. As well, most logs were bucked to about 5 m prior to skidding, perhaps avoiding some of the damage that occurs when tree-length logs are skidded. Two local sawyers were also in charge of felling operations, perhaps adding to the effectiveness of the operation, as they are both experienced harvesters in Porto Dias, in addition to having lived for many years in the reserve. Conclusion As more traditional forest dwellers begin to incorporate formalized timber management into their livelihood strategies (which in the recent past have been based on the exploitation of NTFPs), it will become increasingly important to monitor and document these systems. Historically, most of the neotropical research and forest legislation has focused on large-scale operations (Oliveira 2000). Smallholders will necessarily differ from their industrial counterparts, given their low harvest intensities and management regimes that favor a suite of forest products and services beyond timber. Even though many general sustainable logging guidelines exist, one of the greatest barriers to good silvicultural management is developing site-specific management objectives and monitoring principles that recognize the heterogeneity inherent in tropical forest systems (Putz 1996). As was shown in this study, the gap mosaic of the forest will change dramatically in comparison to its pre-logging state, even when low intensity, RIL guidelines are followed (Webb 1997). Accordingly, community forest managers must establish their own management paradigm while building on the RIL model. Given the series of poorly understood parameters outlined in this study (pre-harvest liana cutting,

PAGE 77

65 post-logging stand dynamics in bamboo-dominated forest, etc.), it is suggested that systematic post-harvesting monitoring accompany logging operations (already one of the requirements for FSC certification), so that the benefits of recommended practices may be evaluated by the managers at the appropriate scale. Pretty and Hine (1999) argue that monitoring methods must be sensitive to local needs in order to work in the long term, primarily due to limited financial and technical resources. It stands to reason then that for monitoring purposes, local peoples efforts may be better placed in protecting the larger size classes of FCTs, particularly if the concessions are on a relatively short (30 years in this case) cutting cycle. This particular study focuses on the adult and sub-adult size classes ( 20 cm dbh) in the hopes of capturing information about the next harvest generation. Many previous logging experiments have concentrated on seedling regeneration, but these seedlings are unlikely to be recruited to harvestable size classes within at least two subsequent harvest cycles (Clark and Clark 1992, Connell and Green 2000). There may be more of a solid commitment to participatory monitoring if the individuals in question could easily project a young trees value across thirty years (which would be in their childrens lifetime), as opposed to a seedling that may take up to 100 years to get to a harvestable size, if it survives at all. Reducing damage is one of the most important components of a sustainable forest management program (Johns et al. 1996). Direct damage to FCTs was shown to be slight for this study, and indeed, it was considerably less than in other neotropical investigations. Even so, damage could be reduced even further. The Porto Dias timber management project is harvesting lightly (7.92 m3 ha-1), but the implementation of liana cutting, cautious selection of both crop and seed trees and avoidance of stands heavily dominated

PAGE 78

66 by Guadua could all help approach sustainable forest management, a concept which may have the best chances of being executed by those who directly benefit from the natural resource base on a daily basis.

PAGE 79

67 Table 3-1. Commercial timber species inventoried during 2004 logging operations in the PAE Porto Dias, Acre, Brazil. For those species harvested in 2004-2005, total volume (m3 ha-1) is given. Scientific Name Common Name Family Total volume harvested (m3) 2004-2005 Amburana cearensis. cerejeira Fabaceae 56.6 Andira sp. angelim branca Fabaceae Apuleia leiocarpa cumaru cetim Caesalpiniaceae 23.4 Aspidosperma megalocarpon carapanaba preta Apocynaceae Aspidosperma sp. carapanaba amarela Apocynaceae Aspidosperma spp. pereiro Apocynaceae Aspidosperma vargasii amarelo Apocynaceae 9.7 Astronium lecointeii aroeira Anacardiaceae 1.8 Astronium sp. 1 aroeira preta Anacardiaceae Brosimium uleanum manit Moracae 9.4 Carapa guianensis andiroba Meliaceae Cariniana sp. 1 tauari churu Lecythidaceae Cedrela cf .fissilis cedro branco Meliaceae 1.7 Cedrela odorata. cedro rosa Meliaceae 4.4 Clarisia racemosa guariba amarela Moraceae Copaifera cf. langsdorfii copaiba preta Caesalpiniaceae Copaifera multijuga copaiba branca Caesalpiniaceae Couratari spp. tauari Lecythidaceae Diplotropis spp. sucupira preta Fabaceae Dipteryx sp. 1 cumaruzinho Fabaceae 6.1 Dipteryx spp. cumaru ferro Fabaceae 65.0 Enterolobium shomburgkii fava orelhinha Mimosaceae 7.7 Geissospermum sp. quariquara amarelo Apocynanceae Hymenaea intermedia jatob Fabaceae Hymenaea oblongifolia juta Fabaceae 19.5 Hymenaea sp. juta da folha grande Fabaceae Hymenolobium cf. excelsum angelim preto Fabaceae 14.1* Hymenolobium sp. angelim da mata Fabaceae Hymenolobium sp. angelim pedra Fabaceae Hymenolobium sp. favela preta Fabaceae Jacaranda copaia marup Bignoniaceae Manilkara spp. maaranduba Sapotaceae 6.5

PAGE 80

68 Continued Table 3-1. Scientific Name Common Name Family Total volume harvested (m3) 2004-2005 Martiodendrom elatum pororoca Caesalpiniaceae Mezilaurus itauba itaba Lauraceae Minquartia guianensis? quariquara branca Olacaceae Myroxylon balsamum blsamo Fabaceae 2.5 Parkia pendula angico vermelho Mimosaceae Peltogyne paniculatum roxinho da folha grande Caesalpiniaceae Peltogyne sp. roxinho Caesalpiniaceae Pouteria cf. reticulata abiurana preta Sapotaceae Pouteria sp. 1 abiurana roxo Sapotaceae Pouteria sp. 2 abiurana abiu Sapotaceae Qualea cf. tesmanni catuaba Vochysiaceae 3.5 Simarouba amara marup preto Simaroubaceae Tabebuia cf. impetigenosa ip roxo Bignoniaceae Tabebuia spp. ip amarelo Bignoniaceae 2.91 Tetragastris altissima breu vermelho Burseraceae Vatairea spp. sucupira amarela Fabaceae Vochysia sp. 1 cedrinho Vochysiaceae Vochysia spp. guaruba Vochysiaceae volume recorded under one common name, but likely reflects more than one species of Hymenolobium.

PAGE 81

69 Table 3-2. Classification of damage sustained by FCTs (future crop trees) during logging operations (modified from Jackson et al. 2002 and Krueger 2004). Damage type Bole Root Crown Severe Snapped at base, bent or leaning Uprooted Loss 66% crown Moderate Exposed cambium Exposed cambium tissue Loss 33-66% crown Minor bark scrape root scrape Loss 33% of crown

PAGE 82

70 Table 3-3. Summary of harvesting impacts in 2004 for the three logging blocks at PAE Porto Dias, Acre, Brazil. Impact variable Barrinha I Barrinha III Palestina Area sampled (ha) 8.0 8.4 5.4 Harvest density Number of harvested trees ha-1 1.5 1.6 1.4 Percent of harvested trees 70 cm dbh 27 50 64 Volume harvested (m3 ha-1) 7.6 8.5 7.7 Disturbance dimensions Number of logging gaps 14 13 11 Felling gap area (excluding bole) (m2 ha-1) 545.6 433.4 476.6 Proportion of area damaged due to felling gap (excluding bole) (%) 5 4 5 Mean gap size SD, excluding bole (m2) 389.7 119 361.2 249 433.3 167 Skid trail area (m2 ha-1) 519.4 318.6 347.0 Proportion of area damaged due to skid trail (%) 5 3 4 Total area damaged (ha) 1.15 0.83 0.89 FCT damage rate Total number of damaged FCTs 18 37 15 Basal area of damaged FCTs (m2) 0.34 0.63 0.48 Proportion of FCTs damaged (%) 12 17 15 Frequency of damage categories (%) Minor or moderate crown, bole, or root damage 78 70 80 Severe crown damage 6 11 0 Severe bole damage 22 24 7 Severe root damage 6 3 27

PAGE 83

71 Table 3-4. Results of a logistic regression for the probability of damage to future crop trees (FCTs) at Porto Dias, Acre, Brazil during 2004 logging operations. Shown are the estimates for five continuous covariates, including distance to nearest stump, felling gap, and skid trail, neighborhood FCT density (number of stems within a 25 m radius) and local felling intensity (number of trees felled in the surrounding hectare); and for three categorical independent variables, including liana presence, location in bamboo-dominated forest, and marking treatment. Statistically significant variables at P < 0.05 are shown in bold. Variable Estimate Wald p Log distance nearest stump -0.07 0.01 0.91 Log distance nearest felling gap -1.48 22.96 0.001 Log distance nearest skid trail -2.25 28.37 0.001 Log neighborhood FCT density -0.26 0.24 0.62 Local felling intensity 0.16 0.88 0.35 Liana presence (L) 0.31 2.79 0.09 Location in bamboo (G) -0.02 0.01 0.90 Marking treatment (M) -0.14 0.57 0.45 L*G 0.14 0.52 0.47 L*M 0.24 1.55 0.21 G*M -0.10 0.29 0.59 L*G*M 0.11 0.32 0.57

PAGE 84

72 Figure 3-1. Photograph taken in PAE Porto Dias, Acre, Brazil, demonstrating the re-curved thorn on a branch of a Guadua culm.

PAGE 85

05102.5KilometersLEGEND Primary Road Secondary Road Creek & River Barrinha 1 Barrinha 3 Palestina So PedroPAE Porto Dias, Acrelandia, Acre, Brasil 73 Figure 3-2. PAE Porto Dias, with the four landholdings inventoried in the study, including the three (in blue) harvested in 2004.

PAGE 86

! .! .! .! .! .! .! .! .! .! .! .! .! .! ######################### #################5755044338334315249241150138135119107 #######################################################################################################################!245 030609012015Meters4Barrinha I Logging Block LEGEND B1 Zone of Impact! .B1 Cut & Skidded Trees!B1 Fallen & Skidded Trees B1 Skid Trails B1 Felling Gaps B1 Fallen TrunksB1 Future Crop TreesMARCADA#0#1 74 Figure 3-3. Logging block of Barrinha 1, showing skid trails, harvested trees, and location of future crop trees (FCTs) in the zones of impact.

PAGE 87

! .! .! .! .! .! .! .! .! .! .! .! .! ### #############90888380392722327360306213163162 ####################### #################################################################################################################################################################################!!!61334278 030609012015Meters4Barrinha 3 Logging Block LEGEND B3 Zone of Impact! .B3 Cut & Skidded Trees!B3 Fallen & Skidded Trees B3 Skid Trails B3 Gaps B3 Fallen TrunksB3 Future Crop TreesMARCADA#0#1 75 Figure 3-4. Logging block of Barrinha 3, showing skid trails, harvested trees, as well as location of future crop trees (FCTs) in the zones of impact.

PAGE 88

76 Figure 3-5. Logging block of Palestina, showing skid trails, harvested trees, as well as location of future crop trees (FCTs) in the zones of impact. .! .! .! .! .! .! .! .! .! .! ###################### ###51341323321311275271252216215183 ################################### #################################################&&&317192179 025507510012.5Meters4Palestina Logging Block LEGEND PAL Zone of Impact! .PAL Cut & Skidded Trees&PAL Fallen & Skidded Trees PAL Skid Trails PAL Felling Gaps PAL Fallen TrunksPAL Future Crop TreesMARCADA#0#1

PAGE 89

77 Figure 3-6. Relationship between size of felled trees and the surface area of the corresponding logging gap created during felling.

PAGE 90

CHAPTER 4 CONCLUSION Redford and Stearman (1993) propose that despite claims that low-impact extraction is compatible with biodiversity conservation, if a full range of genetic, species and ecosystem diversity is to be maintained, then any significant activity by humans must not be allowed. Advocates of SFM have countered this argument by stating that large-scale preserves have a limited chance of succeeding in the face of restricted financial resources, human population growth, and economic pressures (Putz et al. 2000, Pearce et al. 2003). Who is correct? Or is there no one right answer? Instead, should areas managed for timber resources be but one component of an interconnected model of complementary nature, production and indigenous reserves, allowing for various degrees of protection (Peres and Terborgh 1995)? The forest is going to be exploited, whether so-called preservationists like it or not. The extent to which that exploitation may incur permanent damage can be controlled in some ways by planned management activities. Even though a strong link between biodiversity conservation and SFM may never be fully satisfied, the global demand for tropical hardwood is not likely to subside anytime soon. This is especially true of Brazil, which is simultaneously the worlds largest producer and consumer of tropical hardwood (May 2002). Questions about economic and environmental sustainability still beleaguer community-based timber projects quest for staying power, though many see the timber projects and forest managers as key assets in protecting forest ecosystems from encroaching mechanized agriculture and cattle production, two of the greatest threats to 78

PAGE 91

79 Amazon biodiversity (Nepstad et al. 2002, Kainer et al. 2003). With future studies, it will be challenging to detect changes in the resource base given that many plant and animal species are harvested simultaneously in tropical forests; ecological tradeoffs associated with these multispecies systems are so complex as to warrant long-term trial-and-error experiments (Becker and Ostrom 1995). Thus, complex community level changes may be difficult to asses (Hall and Bawa 1993). Also, a critical factor to consider in this debate is such ambiguous vocabulary as sustainable, biodiversity and ecosystem management (Ghazoul 2001). The inability of biologists to agree upon this lexicon makes the practical application of sustainability guidelines by community forest managers even more challenging. Consequently, there is a greater risk in resource degradation if specific ecological problems are not readily identifiable. The conclusions reached in this thesis suggest that forest dwellers may be able to sustainably use common resources of the forest, but whether or not they do so will depend on a wider setting, that of a shifting national and international politico-economic climate (Cardoso 2002). Fortunately, particularly in the state of Acre, the government has been extremely supportive of forest dweller communities, in addition to providing a strong facilitative role. With the continued presence of external agencies, these communities will stand a better chance at conserving forest resources. One issue that is frequently mentioned in the literature but still lacks substantial investigation is the idea that economic sustainability is not always consistent with ecological sustainability (Hall and Bawa 1993). The links between sustainable development and resource management must be studied further, as the rural poor do indeed have an invested interest in sustaining the resource base, but at the same time they

PAGE 92

80 must also increase and diversify their overall livelihood (Cardoso 2002). Ultimately, the success of SFM will be contingent upon addressing the drivers of forest degradation and deforestation. Conservation will not occur through SFM and forest certification alone; local, regional, national, and international policy makers must also be engaged in this process. Otherwise, SFM as a tool for biodiversity conservation has little chance of success.

PAGE 93

REFERENCES Agrawal, A. A. and C. C. Gibson. 1999. Enchantment and disenchantment: The role of community in natural resource conservation. World Development 27: 629-649. Alder, D. and J. N. M. Silva. 2000. An empirical cohort model for management of terra firme forests in the Brazilian Amazon. Forest Ecology and Management 130: 141-157. Allee, W. C. 1949. Group survival value for Philodina roseola, a rotifer. Ecology 30: 395-397. Amaral, P. and M. A. Neto. 2000. Manejo florestal comunitrio na Amaznia Brasileira: Situao atual, desafios e perspectivas. Brasilia: Instituto Internacional de Educao do Brasil (IIEB). Electronic document, http://www.tech-inform.de/taller_regional/bibliografia/bibliografia.htm accessed November 30, 2005. Amaral, P. A. Verssimo, P. Barreto, and E. Vidal. 1998. Floresta para sempre: Um manual para a produo de madeira na Amaznia. World Wildlife Fund/IMAZON/USAID, Belm, Par, Brazil. Appanah, S. and F. E. Putz. 1984. Climber abundance in virgin Dipterocarp forest and the effect of pre-felling climber cutting on logging damage. The Malaysian Forester 47: 335-342. Asner, G. P., D. E. Knapp, E. N. Broadbent, P. J. C. Oliveira, M. Keller, and J. N. Silva. 2005. Selective logging in the Brazilian Amazon. Science 21: 480-482. Awasthi, A., S. K. Uniyal, G. S. Rawat, and A. Rajvanshi. 2003. Forest resource availability and its use by migratory villages of Uttatkashi Garwhal Himalaya (India). Forest Ecology and Management 174: 13-24. Banana, A. Y. and W. Gombya-Ssembajjwe. 2000. Successful forest management: The importance of security of tenure and rule enforcement in Ugandan forests. Pages 87-98 in C. C. Gibson, M. A. McKean and E. Ostrom, eds. People and forests: Communities, institutions, and governance. MIT Press, Cambridge, MA, USA. Banuri, T. and F. Apffel-Marglin. 1993. Who will save the forest? Knowledge, power and environmental destruction. Zed Books, London, UK. 81

PAGE 94

82 Baraloto, C. and P.-M. Forget. 2004. Effects of post-logging microhabitats on regeneration from seed: Implications for sustainable forest management. Pages 191-203 in S. Gourlet-Fleury, J.-M. Guehl, and O. Laroussinie, eds. Ecology and management of a neotropical rainforest: Lessons drawn from Paracou, a long-term experimental research site in French Guiana. Elsevier, Paris, France. Barreto, P., P. Amaral, E. Vidal, and C. Uhl. 1998. Costs and benefits of forest management for timber production in eastern Amazonia. Forest Ecology and Management 108: 9-26. Barreto, P., C. Uhl and G. Yared. 1993. O potencial de produo sustentavel de Madeira em Paragominas, Par na Amaznia Oriental: Consideraes eclogicas e econmicas. 7 Congresso Florestal Brasileiro, Anais da Sociedade Brasileiro de Engenheiros Florestais, pp. 387-392. Bawa, K. S., D. R. Perry and J. H. Beach. 1985. Reproductive biology of tropical lowland rainforest trees. I. Sexual systems and incompatibility mechanisms. American Journal of Botany 72: 331-345. Bawa, K. S. and R. Seidler. 1998. Natural forest management and conservation of biodiversity in tropical forests. Conservation Biology 12: 46-55. Becker, C. D. and R. Len. 2000. Indigenous forest management in the Bolivian Amazon: Lessons from the Yuracar people. Pages 163-191 in C. C. Gibson, M. A. McKean and E. Ostrom, editors. People and forests: Communities, institutions, and governance. MIT Press, Cambridge, MA, USA. Becker, C. D. and E. Ostrom. 1995. Human ecology and resource sustainability: The importance of institutional diversity. Annual Review of Ecology and Systematics 26: 113-133. Berkes, F. 1999. Sacred ecology: Traditional ecological knowledge and resource management. Taylor and Francis, Philadelphia, Pennsylvania, USA. Berkes, F. and C. Folke. 2002. Back to the future: Ecosystem dynamics and local knowledge. Pages 121-146 in L.H. Gunderson and C.S. Holling, editors. Panarchy: Understanding transformations in human and natural systems. Island Press, Washington, USA. Bertault, J.-G. and P. Sist. 1997. An experimental comparison of different harvesting intensities with reduced-impact and conventional logging in East Kalimantan, Indonesia. Forest Ecology and Management 94: 209-218. Bobadilla, U. L. and S. F. Ferrari. 2000. Habitat use by Chiroptes satanas utahicki and Syntopic platyrrhines in eastern Amazonia. American Journal of Primatology 50: 215-224.

PAGE 95

83 Boot, R. G. A. and R. E. Gullison. 1995. Approaches to developing sustainable extraction systems for tropical forest products. Ecological Applications 5: 896-903. Bowles, I., R. Rice, R. Mittermeier, and A. daFonseca. 1998. Logging and tropical conservation. Science 280: 1899-1900. Bray, D. B., L. Merino-Perez, P. Negreros-Castillo, G. Segura-Warnholtz, J. M. Torres-Rojo, and H. F. M. Vester. 2003. Mexico's community-managed forests as a global model for sustainable landscapes. Conservation Biology 17: 672-673. Braz, E. M. and M. V. N. dOliveira. 1996. Planning to reduce damage. Tropical Forest Update 6: 13-18. Brokaw, N. V. I. 1982. The definition of treefall gap and its effects on measures of forest dynamics. Biotropica 11: 158-160. Bruenig, F. E. 1996. Conservation and management of tropical rainforests: An integrated approach to sustainability. CAB International, New York, USA. Bush, M. B. and P. A. Colinvaux. 1994. A paleoecological perspective of tropical forest disturbance: Records from Darien, Panama. Ecology 75: 1761-1768. Calheiros, D. F., A. F. Seidl and C. J. A. Ferreira. Participatory methods in environmental science: Local and scientific knowledge of a limnological phenomenon in the Pantanal wetland of Brazil. Journal of Applied Ecology 37: 684-696. Camino, R. and M. Alfaros. 2000. Certification in Latin America: Experience to date. Forest, Trees and People Newsletter 43: 25-27. Cardoso, C. A. S. 2002. Extractive reserves in Brazilian Amazonia: Local resource management and the global political economy. Ashgate Studies in Environmental Policy and Practice, Hampshire, UK. Carter, J. 1996. Recent approaches to participatory forest resource management. Rural Development Forestry Network, Overseas Development Institute, London, UK. Chapman, C. A., L. J. Chapman, R. Wangham, K. Hunt, D. Gebo, and L. Gardner. 1992. Estimators of fruit abundance of tropical trees. Biotropica 24: 527-531. Clark, D. A. and D. B. Clark. 1992. Life history diversity of canopy and emergent trees in a neotropical rain forest. Ecological Monographs 62: 315-344. Connell, J. H. and P. T. Green. 2000. Seedling dynamics over thirty-two years in a tropical rain forest tree. Ecology 81: 568-584. Dalling, J. W., H. C. Muller-Landau, S. J. Wright, and S. P. Hubbell. 2002. Role of dispersal in the recruitment limitation of neotropical pioneer species. Journal of Ecology 90: 714-727.

PAGE 96

84 Daube, E., T. Fredericksen and M. Pea. 2005. Sustainability of timber harvesting in Bolivian tropical forests. Forest Ecology and Management 214: 294-304. Deneven, W. M. 1992. The pristine myth: The landscape of the Americas in 1492. Annals of the Association of American Geographers 82: 369-385. Denslow, J. S. 1980. Gap partitioning among tropical rainforest trees. Biotropica 12: 47-55. Dickinson, M. B., D. F. Whigham and S. M. Hermann. 2000. Tree regeneration in felling and natural treefall disturbances in a semideciduous tropical forest in Mexico. Forest Ecology and Management 134: 137-151. Dykstra, D. P., and R. Heinrich. 1996. FAO model code of forest harvesting practice. Food and Agricultural Organization of the United Nations, Rome, Italy. Food and Agricultural Organization (FAO). 2005. State of the worlds forests. Food and Agricultural Organization of the United Nations, Rome, Italy. Fearnside, P. M. 2003. Conservation policy in Brazilian Amazonia: Understanding the dilemmas. World Development 31: 757-779. Folke, C., F. Berkes and J. Colding 1998. Ecological practices and social mechanisms for building resilience and sustainability. Pages 414-436 in F. Berkes and C. Folke, editors. Linking social and ecological systems. Cambridge University Press, London, UK. Fox, J. E. D. 1968. Logging damage and the influence of climber cutting prior to logging in the lowland Dipterocarp forest of Sabah. The Malayan Forester 31: 326-347. Fredericksen, T. S. and F. E. Putz. 2003. Silvicultural intensification for tropical forest conservation. Biodiversity and Conservation 12: 1445-1453. Gerwing, J. J. 2001. Testing liana cutting and controlled burning as silvicultural treatments for a logged forest in the eastern Amazon. Journal of Applied Ecology 38: 1264-1278. Gerwing, J. J. and E. Vidal. 2002. Changes in liana abundance and species diversity eight years after liana cutting and logging in an eastern Amazonian forest. Conservation Biology 16: 544-548. Ghazoul, J. 2001. Barriers to biodiversity conservation in forest conservation. Conservation Biology 15: 315-317. Ghazoul, J., K. A. Liston and T. J. B. Boyle. 1998. Disturbance-induced density-dependent seed set in Shorea siamensis (Dipterocarpaceae), a tropical forest tree. Journal of Ecology 86: 462-473.

PAGE 97

85 Gibson, C. C. and C. D. Becker. 2000. A lack of institutional demand: Why a strong local community in western Ecuador fails to protect its forest. Pages 135-161 in C. C. Gibson, M. A. McKean and E. Ostrom, editors. People and forests: Communities, institutions, and governance. MIT Press, Cambridge, MA, USA. Griscom, B. W. 2003. The influence of bamboo (Guadua sarcocarpa and Guadua weberbaueri) on stand dynamics in lowland terra firme forests of southeastern Peru. PhD dissertation. Yale University, New Haven, CT, USA. Griscom, B. W. and P. M. S. Ashton. 2003. Bamboo control of forest succession: Guadua sarcocarpa in southeastern Peru. Forest Ecology and Management 175: 445-454. Guariguata, M. R. and M. A. Pinard. 1998. Ecological knowledge of regeneration from seed in neotropical forest trees: Implications for natural forest management. Forest Ecology and Management 112: 87-99. Gullison, R. E. and Hardner, J. J. 1993. The effects of road design and harvest intensity on forest damage caused by selective logging: empirical results and a simulation model for the Bosque Chimanes, Bolivia. Forest Ecology and Management 59: 1-14. Hall, P. and K. Bawa. 1993. Methods to assess the impact of extraction of non-timber tropical forest products on plant populations. Economic Botany 47: 234-247. Hammond, D. S., S. Gourlet-Fleury, P. van der Hout, H. ter Steege, and V. K. Brown. 1996. A compilation of known Guianan timber trees and the significance of their dispersal mode, seed size and taxonomic affinity to tropical rainforest management. Forest Ecology and Management 83: 99-116. Hardin, G. 1968. The tragedy of the commons. Science 162: 1243-1248. Hartshorn, G. S. 1989. Application of gap theory to tropical forest management: Natural regeneration in strip clearcuts in the Peruvian Amazon. Ecology 70: 567-569. Heckenberger, M. J., A. Kuikuro, U. T. Kuikuro, J. C. Russell, M. Schmidt, C. Fausto, and B. Franchetto. Amazonia 1492: Pristine forest or cultural parkland? Science 301: 1710-1714. Heideman, P. D. 1989. Temporal and spatial variation in the phenology of flowering and fruiting in a tropical rainforest. Journal of Ecology 77: 1059-1079. Holmes, T. P., G. M. Blate, J. C. Zweede, R. Pereira, P. Barreto, F. Boltz, and R. Bauch. 2002. Financial costs and benefits of reduced-impact logging relative to conventional logging in the eastern Amazon. Forest Ecology and Management 163: 93-110. Hubbell, S. P. 1979. Tree dispersion, abundance, and diversity in a tropical dry forest. Science 203: 1299-1309.

PAGE 98

86 Humphries, S. 2005. Forest certification for community-based forest enterprises in Brazils Western Amazon: Local stakeholders perceptions of negative and positive aspects and the certification process. MSc Thesis. University of Florida, Gainesville, FL, USA. Janzen, D. 1976. Why bamboos wait so long to flower. Annual Review of Ecology and Systematics 7: 347-391. Janzen, D. 2004. Setting up tropical biodiversity for conservation through non-damaging use: Participation by parataxonomists. Journal of Applied Ecology 41: 181-187. Jennings, S. B., N. D. Brown, D. H. Boshier, T. C. Whitmore, and J. do C. A. Lopes. 2001. Ecology provides a pragmatic solution to the maintenance of genetic diversity in sustainably managed tropical rain forests. Forest Ecology and Management 154: 1-10. Johns, A. D. 1991. Responses of Amazonian rain forest birds to habitat modification. Journal of Tropical Ecology 7: 417-437. Johns, J. S., P. Barreto and C. Uhl. 1996. Logging damage during planned and unplanned logging operations in the Eastern Amazon. Forest Ecology and Management 89: 59-77. Johnson, N. and B. Cabarle. 1993. Surviving the cut. World Resources Institute, Washington, D.C. 71 pp. Kainer, K. A., M. Schmink, A. C. Pinheiro Leite, and M. J. daSilva Fadell. 2003. Experiments in forest-based development in western Amazonia. Society and Natural resources 16: 869-886.. Krueger, W. 2004. Effects of future crop tree flagging and skid trail planning on conventional diameter-limit logging in a Bolivian tropical forest. Forest Ecology and Management 188: 381-393. Lacerda, A. and G. Griffiths. 2005. Conservation and sustainability within Brazilian Amazon forests: The role of botanical identification. Presentation given at the Association of Tropical Biology and Conservation, Uberlandia, Brazil, July 24-28. Lawrence, A. and F. S. Roman. 1996. The role of inventory in the communally managed forests of Quintana Roo, Mexico. Pages 83-110 in J. Carter, ed. Recent approaches to participatory forest resource management. Rural Development Forestry Network, Overseas Development Institute, London, UK. Lentini, M., A. Verissimo and L. Sobral. 2003. Forest facts in the Brazilian Amazon. IMAZON, Belem, Brazil.

PAGE 99

87 Lieberman, M. and D. Lieberman. 1994. Patterns of density and dispersion of forest trees. Pages 106-119 in L. A. McDade, K. S. Bawa, H. A. Hespenheide, and G. S. Hartshorn, editors. La Selva: Ecology and natural history of a neotropical rainforest. University of Chicago Press, Chicago, IL, USA. Lorenzi, H. 2000. Arvores Brasileiras. Instituto Plantarum de Estudos da Flora Ltda., Nova Odessa, Sao Paulo, Brazil. Marcon, E. 2005. Ripley, version 3.1. http://e.marcon.free.fr/Ripley/ accessed December 5, 2005. Margoluis, R. and N. Salafsky. 1997. Measures of success: Designing, managing, and monitoring conservation and development projects. Island Press, Washington, DC, USA. May, P. H. 2002. Forest certification in Brazil: Trade and environmental enhancement. Consumer Choice Council, Washington, DC, USA. Menton, M. C. 2003. Effects of logging on non-timber forest product extraction in the Brazilian Amazon: Community perceptions of change. International Forestry Review 5: 97-105. Metzger, F. and J. Schultz. 1984. Understory response to 50 years of management of a northern hardwood forest in upper Michigan. American Midland Naturalist 112: 209-223. Meyer, J. L. and G. S. Helfman. 1993. The ecological basis of sustainability. Ecological Applications 3: 569-571. Molnar, A. 2003. Forest certification and communities: looking forward to the next decade. Forest Trends, Washington, DC, USA Morellato, P. C. and H. F. Leitao-Filho. 1996. Reproductive phenology of climbers in a southeastern Brazilian forest. Biotropica 28: 180-191. Mller, U. 1998. Effects of refinements on the development of a neotropical rainforest after controlled logging in Suriname. MSc thesis. Wageningen Agricultural University, Wageningen, The Netherlands. Nelson, B. W., A. C. A. Oliveira, G. T. Batista, P. Vidalenc, and M. Silveira. 2001. Modeling biomass in the southwest Amazon by polar ordination of Landsat TM. Pages 1683-1690 in Anais X SBSR, Foz do Iguau, 21-26 April 2001, INPE. Nepstad, D., A. Alencar, A. C. Barros, E. Lima, E. Mendoza, C. A. Ramos, and P. Lefebvre. 2004. Governing the Amazon timber industry. Pages 388-414 in D. Zarin, F. J. Putz, M. Schmink and J. Alavalapati, editors. Working forests in the tropics: Conservation through sustainable management? Columbia university Press, New York, NY, USA.

PAGE 100

88 Nepstad, D., D. McGrath, A. Alencar, A. C. Barros, G. Carvalho, M. Santilli, and M. d. C. Vera Diaz. 2002. Frontier governance in Amazonia. Science 295: 629-631. Nyland, R. D. 2002. Silviculture concepts and applications. McGraw-Hill, New York, USA. Oliveira, M. V. N. d. 2000. Sustainable forest management for small farmers in Acre state in the Brazilian Amazon. PhD dissertation. University of Aberdeen, Aberdeen, Scotland. Oliveira, M. V. N. d, E. M. Braz, D. F. R. P. Burslem, and M. D. Swaine. 1998. Small-scale natural forest management: A new model for small farmers in the Brazilian Amazon. Tropical Forest Update 8: 5-7. Panfil, S. N. and R. E. Gullison. 1998. Short term impacts of experimental timber harvest intensity on forest structure and composition in the Chimanes Forest, Bolivia. Forest Ecology and Management 102: 235-243. Parren, M. P. E. 2003. Lianas and logging in West Africa. PhD dissertation. Wageningen University, Wageningen, The Netherlands. Parren, M. and F. Bongers 2001. Does climber cutting reduce felling damage in southern Cameroon? Forest Ecology and Management 141: 175-188. Pearce, D., F. E. Putz and J. K. Vanclay. 2003. Sustainable forestry in the tropics: Panacea or folly? Forest Ecology and Management 172:229-247. Peres, C. A., C. Baider, P. A. Zuidema, L. H. O. Wadt, K. A. Kainer, D. A. P. Gomes-Silva, R. P. Salomo, L. H. Simes, E. R. N. Franciosi, F. C. Valverde, R. Gribel, G. H. Shepherd, M. Kanashiro, P. Coventry, P. W. Yu, A. R. Watkinson, and R. P. Freckleton. 2003. Demographic threats to the sustainability of Brazil nut exploitation. Science 302: 2112-2114. Peres, C. A. and J. Terborgh. 1995. Amazonian nature reserves: An analysis of the defensibility status of existing conservation units and design criteria for the future. Conservation Biology 9: 34-46. Pereira, R. Jr., J. Zweede, G. P. Asner, and M. Keller. Forest canopy damage and recovery in reduced-impact and conventional selective logging in eastern Para, Brazil. Forest Ecology and Management 168: 77-89. Pinard, M. A. and F. E. Putz. 1996. Retaining forest biomass by reducing logging damage. Biotropica 28: 278-295. Pinard, M. A., F. E. Putz, J. Tay, and T. E. Sullivan. 1995. Creating timber harvesting guidelines for a reduced-impact logging project in Malaysia. Journal of Forestry 93:41-45.

PAGE 101

89 Pinedo-Vasquez, M., D. J. Zarin, K. Coffey, C. Padoch, and F. Rabelo. 2001. Post-boom logging in Amazonia. Human Ecology 29: 219-239. Poore, D., P. Burgess, J. Palmer, S. Rietbergen, and T. Synnott. 1989. No timber without trees: Sustainability in the tropical forest. Earthscan, London, UK. Pretty, J. and Hine, R. 1999. Participatory appraisal for community assessment: principles and methods. Center for Environment and Society, University of Essex, UK. Putz, F. E. 1994. Approaches to sustainable forest management. working paper no. 4, Center for International Forestry Research, Jakarta, Indonesia. Putz, F. E. 1996. Research needs and information gaps. Pages 164-178 in V. M. Viana, editor. Certification of forest products: Issues and perspectives. Island Press, Washington, DC, USA. Putz, F. E. 2004. Are you a conservationist or a logging advocate? Pages 15-30 in D. Zarin, F. J. Putz, M. Schmink and J. Alavalapati, editors. Working forests in the tropics: Conservation through sustainable management? Columbia university Press, New York, NY, USA. Putz, F. E., K. H. Redford, J. G. Robinson, R. Fimbel, and G. M. Blate. 2000. Biodiversity conservation in the context of tropical management. Biodiversity Series Paper no. 75, World Bank, Washington, DC, USA. Putz, F. E. and C. Romero. 2001. Biologists and timber certification. Conservation Biology 15: 313-314. Putz, F. E., L. K. Sirot and M. A. Pinard. 2001. Tropical forest management: Silvicultural effects on forest structure, fruit production and locomotion of arboreal animals. Pages 11-34 in R. A. Fimbel, A. Grajal and J. G. Robinson, eds. The cutting edge: Conserving wildlife in logged tropical forests. Columbia University Press, New York, NY, USA. Putz, F. E. and V. Viana. 1996. Biological challenges for certification of tropical timber. Biotropica 29: 323-330. Putz, F. E. and D. M. Windsor. 1987. Liana phenology on Barro Colorado Island, Panama. Biotropica 19: 334-341. Redford, K. H. and A. M. Stearman. 1993. Forest-dwelling native Amazonians and the conservation of biodiversity: Interests in common of in collision? Conservation Biology 7: 248-255.

PAGE 102

90 Ribeiro, J. E. L.S., M. J. G. Hopkins, A. Vicentini, C. A. Sothers, M. A. S. Costa, J. M. de Brito, M. A. D. de. Souza, L. H. P. Martins, L. G. Lohmann, P. A. C. L. Assuncao, E. da. C. Pereira, C. F. da. Silva, M. R. Mesquita, and L. C. Procopio. 1999. Guia de identificacao das plantas vasculares de uma floresta de terra-firme na Amazonia Central. INPA, DFID, Manaus, Brazil. Rice, R., R. E. Gullison and J. Reed. Can sustainable management save tropical forests? Scientific American 276: 34-39. Romero, C. 1999. Reduced-impact logging effects on commercial non-vascular pendant epiphyte biomass in a tropical montane forest in Costa Rica. Forest Ecology and Management 118: 117-125. Runkle, J. R. 1992. Guidelines and sample protocol for sampling forest gaps. General Technical Report, PNW-GTR-283. United States Department of Agriculture, Forest Service, Pacific Northwest Station,USA. Salafsky, N., B. Cordes, M. Leighton, M. Henderson, W. Watt, and R. Cherry. 1998. Chainsaws as a tool for conservation? A comparison of community-based timber production enterprises in Papua New Guinea and Indonesia. Network paper 22b, Rural Development Forestry Network, London, UK. Salick, J., A. Mejia and T. Anderson. 1995. Non-timber forest products integrated with natural forest management, Rio San Juan, Nicaragua. Ecological Applications 5: 878-895. Santos, M. C. dos. 2000. Adaptive Co-management: A case study. PESACRE, Porto Dias, Acre, Brazil. Report written for CIFOR. Scherr, S. J., A. White and D. Kaimowitz. 2002. Making markets work for forest communities. Forest Trends, Washington, D. C. and Center for International Forestry Research (CIFOR), Bogor, Indonesia. Schmink, M. 2004. Communities, forests, markets, and conservation. Pages 119-129 in D. Zarin, F. J. Putz, M. Schmink and J. Alavalapati, editors. Working forests in the tropics: Conservation through sustainable management? Columbia university Press, New York, NY, USA. Schnitzer, S. A. and F. Bongers. 2002. The ecology of lianas and their role in forests. Trends in Ecology and Evolution 17: 223-230. Schnitzer, S. A., J. W. Dalling and W. P. Carson. 2000. The impact of lianas on tree regeneration in tropical forest canopy gaps: Evidence for an alternative pathway of gap-phase regeneration. Journal of Ecology 88: 655-666. Schulze, M. D. 2003. Ecology and behavior of nine timber tree species in Par, Brazil: Links between species life history and forest management and conservation. PhD dissertation. Pennsylvania State University, State College, USA.

PAGE 103

91 Schupp, E. W. 1992. The Janzen-Connell model for tropical tree diversity: Population implications and the importance of spatial scale. The American Naturalist 140: 526-530. Schwartzman, S., A. Moreira and D. Nepstad. 2000. Rethinking tropical forest conservation: perils in parks. Conservation Biology 14: 1351-1357. Schwarzkopf, L. and B. Rylands. 1989. Primate species richness in relationship to habitat structure in Amazonian rainforest fragments. Biological Conservation 48: 1-12. Shanley, P. and G. R. Gaia. 2002. Equitable ecology: Collaborative learning for local benefit in Amazonia. Agricultural Systems 73: 83-97. Sheil, D. S. and A. Lawrence. 2004. Tropical biologists, local people and conservation: New opportunities for collaboration. Trends in Ecology and Evolution 19: 634-638. Sheil, D. and M. Van Heist. 2000. Ecology for tropical forest management. International Forestry Review 2: 261-270. Sierra, R. 2001. The role of domestic timber markets in tropical deforestation and forest degradation in Ecuador: Implications for conservation planning and policy. Ecological Economics 36: 327-340. Silva, J. N. M., J. O. P. de Carvalho, J. de C. A. Lopes, B. F. de Almeida, D. H. M. Costa, L. C. de Oliveira, J. Vanclay, and J. P. Skovsgaard. 1995. Growth and yield of a tropical rain forest in the Brazilian Amazon. Forest Ecology and Management 71: 267-274. Silveira, M. 2001. A floresta aberta com bambu no sudeste da Amaznia: Padres em processos em mltiplas escalas. PhD dissertation. Universidade de Brasilia, Brasilia, Brazil. Simberloff, D. 1999. The role of science in the preservation of forest biodiversity. Forest Ecology and Management 115: 101-111. Sist, P., R. Fimbel, D. Sheil, R. Nasi, and M.-H. Chevallier. 2003a. Towards sustainable management of mixed dipterocarp forests of Southeast Asia: Moving beyond minimum diameter cutting limits. Environmental Conservation 30: 364-374. Sist, P., T. Nolan, J.-G. Bertault and D. Dykstra. 1998. Harvesting intensity versus sustainability in Indonesia. Forest Ecology and Management 108: 251-260. Sist, P., D. Sheil, K. Kartawinata, and H. Priyadi. 2003b. Reduced-impact logging in Borneo: Some results confirming the need for new silvicultural prescriptions. Forest Ecology and Management 179: 415-427. Smiet, A. C. 1992. Forest ecology on Java: Human impact and vegetation of montane forest. Journal of Tropical Ecology 8: 129-152.

PAGE 104

92 Smith, D. M., B. C. Larson, M. J. Kelly, and P. M. S. Ashton. 1997. The practice of silviculture: Applied forest ecology. John Wiley and Sons, New York, New York, USA. Snook, L. K. 1998. Sustaining harvests of mahogany from natural forests in Mexicos Yucatan Peninsula: Past, present, and future. R. B. Primack, D. Bray, H. A. Galletti, and I. Ponciano, editors. Timber, tourists and temples. Island Press, New York, NY, USA. Steidl, R. J. and L. Thomas. 2001. Power analysis and experimental design. Pages 14-36 in S. M. Scheiner and J. Gurevitch, editors. Design and analysis of ecological experiments. Oxford University Press, Oxford, UK. Stone. S. S. 2003. From tapping to cutting trees: Participation and agency in two community-based timber management projects in Brazil. PhD dissertation. University of Florida, Gainesville, FL, USA. Summers, P. M., J. O. Browder and M. A. Pedlowski. 2004. Tropical forest management and silvicultural practices by small farmers in the Brazilian Amazon: Recent farm-level evidence from Rondnia. Forest Ecology and Management 192: 161-177. Swaine, M. D., D. Lieberman and F. E. Putz. 1987. The dynamics of tree populations in tropical forest: A review. Journal of Tropical Ecology 3: 359-366. Terborgh, J. 1999. Requiem for nature. Island Press, Washington, DC, USA. Terborgh, J. 2000. The fate of tropical forests: A matter of stewardship. Conservation Biology 14: 1358-1361. Thiollay, J.-M. 1992. Influence of selective logging on bird species diversity in a Guianan rain forest. Conservation Biology 6: 47-63. Uhl, C., P. Barreto, A. Verissimo, E. Vidal, P. Amaral, A. Barros, C. J. Souza, J. Johns, and J. Gerwing. 1997. Natural resource management in the Brazilian Amazon. Bioscience 47: 160-168. Uhl, C. and R. Buschbacher. 1988. A disturbing synergism between cattle ranching practices and selective tree harvesting in the eastern Amazon. Biotropica 17: 265-268. Uhl. C. and I. C. G. Vieira. 1989. Ecological impacts of selective logging in the Brazilian Amazon: A case study from the Paragominas region of the state of Par. Biotropica 21: 98-106. Verissimo, A. and P. Barreto. 2004. National forests in the Brazilian Amazon. Pages 31-40 in D. Zarin, F. J. Putz, M. Schmink and J. Alavalapati, editors. Working forests in the tropics: Conservation through sustainable management? Columbia university Press, New York, NY, USA.

PAGE 105

93 Verissimo, A., P. Barreto and M. Mattos. 1992. Logging impacts and prospects for sustainable forest management in an old Amazonian frontier. Forest Ecology and Management 55: 169-184. Verissimo, A., M. Cochrane, C. Souza Jr., and R. Salomo. 2002. Priority areas for establishing national forest. Conservation Ecology 6(1): 4. Vester, H. F. M. and M. A. Navarro-Martnez. 2005. Ecological issues in community tropical forest management in Quintana Roo, Mexico. Pages 183-213 in D. B. Bray, L. Merino-Prez and D. Barry, eds. The community forests of Mexico: Managing for sustainable landscapes. University of Texas Press, Austin, TX, USA. Vidal, E., J. Johns, J. J. Gerwing, P. Barreto, and C. Uhl. 1997. Vine management for reduced-impact logging in eastern Amazonia. Forest Ecology and Management 98: 105-114. Vitt, L. J., T. C. S. Avila-Pires, J. P. Caldwell, and V. Oliveira. 1998. The impact of individual tree harvesting on thermal environments of lizards in Amazonian rain forest. Journal of Conservation Biology 12: 654-664. Walters, B. B. 2005. Ecological effects of small-scale cutting of Philippine mangrove forests. Forest Ecology and Management 206: 331-348. Walters, B. B., C. Sabogal, L. K. Snook, E. de Almeida. 2005. Constraints and opportunities for better silvicultural practice in tropical forestry: An interdisciplinary approach. Forest Ecology and Management 209: 3-18. Webb, E. L. 1997. Canopy removal and residual stand damage during controlled selective logging in lowland swamp forest of northeast Costa Rica. Forest Ecology and Management 95:117-129. Whitman, A. A., N. V. L. Brokaw and J. M. Hagan. 1997. Forest damage caused by selection logging of mahogany (Swietenia macrophylla) in northern Belize. Forest Ecology and Management 92:87-96. Whitmore, T. C. 1984. Tropical rain forests of the Far East. Clarendon Press, Oxford, UK.

PAGE 106

BIOGRAPHICAL SKETCH Cara A. Rockwell was born on May 25, 1971, in Belleville, Illinois, and grew up in Tallahassee, Florida. She received a Bachelor of Science degree in biological sciences from Florida State University in 1994. Following graduation, she traveled for a year in Asia, Australia and the South Pacific, which fueled her interest in conservation and sustainable development issues. After three years of working with a land stewardship initiative in Ann Arbor, Michigan, she joined Peace Corps Paraguay in 1998. She remained in Paraguay for three years, working first as an agroforestry volunteer with a farmers organization and later as the coordinator for the Peace Corps Paraguay environment program. Upon returning to the US in 2002, she interned for Conservation Internationals Conservation Coffee program in Washington, DC. In 2003, she began her graduate program at University of Florida in the School of Forest Resources and Conservation and with the Tropical Conservation Development program. She conducted her masters research in 2004 and 2005 in Acre, Brazil, investigating the impacts of a community-based timber operation on future crop trees. She plans to continue her doctoral work at the University of Florida. 94


Permanent Link: http://ufdc.ufl.edu/UFE0012840/00001

Material Information

Title: Community-Based Timber Management in Acre, Brazil and Its Implications for Sustainable Forest Management
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: UFE0012840:00001

Permanent Link: http://ufdc.ufl.edu/UFE0012840/00001

Material Information

Title: Community-Based Timber Management in Acre, Brazil and Its Implications for Sustainable Forest Management
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: UFE0012840:00001


This item has the following downloads:


Full Text












COMMUNITY-BASED TIMBER MANAGEMENT IN ACRE, BRAZIL, AND ITS
IMPLICATIONS FOR SUSTAINABLE FOREST MANAGEMENT
















By

CARA A. ROCKWELL


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


2005

































Copyright 2005

by

Cara A. Rockwell




























This thesis is dedicated to my friends and colleagues in PAE Porto Dias, who welcomed
me into their homes, and to my husband, Chris Baraloto, for coming this far in the
journey together.














ACKNOWLEDGMENTS

I would like to express my sincerest gratitude to my advisor and committee chair,

Karen Kainer, for all of her unfailing support. The past two and a half years have been a

truly enjoyable experience. My committee members, Marianne Schmink and Jack Putz,

have been excellent sources of encouragement and knowledge. I extend my full

appreciation to their involvement in the development of this thesis.

Thanks go to the School of Forest Resources and Conservation, the Tropical

Conservation and Development Program and Sigma Delta Epsilon/Graduate Women in

Science for the funding and support of this research.

The execution and completion of this project could not have been accomplished

without the participation of many key partners in Brazil. First, I must thank the families

of PAE Porto Dias, the Associagdo Seringueira Porto Dias, and the Projeto Manejo

Florestal Comunitario de Multiplo Uso for their patience as well as their kindness during

my stay in their community. I must particularly thank the following families for

welcoming us inside their homes; such hospitality will never be forgotten: the family of

Francisco Correia da Cunha and Raimunda Feitosa do Nascimento, the family of Adilton

Ferreira de Souza and Maria Alves Barboza, the family of Juscelino da Silva Correia and

Sebastiana do Santo Ven6ncio, and the family of Lazaro da Silva Salgueiro and

Marinalva Souza Amora Salgueiro. Also, a special obrigada goes to the following

community members for their participation in the study: Luzia Amora Salgueiro, Luziani

Amora Salgueiro, Juciclei Ven6ncio Correia, and Juciane Ven6ncio Correia, and thanks









go also to Luciana Roncoletta, who was conducting her own research in Porto Dias, but

still found time to assist with my project. Undoubtedly, the staff of the Centro dos

Trabalhadores da Amaz6nia (CTA) was there for me from beginning to end. They are

exemplary models of how collaborative partnerships can work. I especially would like to

thank my friends and colleagues, Nivea Marcondes, Magna Cunha dos Santos, Evandro

Araujo, Patricia Roth, and Pedro Bruzzi. I received valuable information and support

from the staff of the Federal University of Acre herbarium as well as faculty members

Marcos Silveira and Evandro Ferreira. Lucia Wadt from the Empresa Brasileira de

Pesquisa Agropecudria (EMBRAPA) provided support both in terms of research interest

and in friendship. And across the border in Bolivia, Marielos Pefia-Claros of the Instituto

Boliviano de Investigaci6n Forestal (IBIF) gave me great insight and suggestions

pertaining to the implementation of this study.

I never would have gotten through the crazy moments without the support from so

many fellow students, both in Gainesville and in Acre. If I miss any names, it is only

because there have been so many in the last couple of years. My warmest regards and

thanks go out to the girls of the Kainer lab: Diana Alvira, Jamie Cotta, Rosa Cossio

Solano, Amy Duchelle, Shoana Humphries, Christie Klimas, and Joanna Tucker. I thank

fellow RPCV Paraguayers who have seemed to find their way to University of Florida,

for all of their kind moments, advice, and plentiful terere: Mandy Baily, Maria DiGiano,

Tom Henshaw, and Katie Painter. The hospitality of Rich Wallace and Ana Puentes in

Acre will never be forgotten. I thank Samantha Stone for her introduction to Porto Dias

and CTA. She has set a high standard for all to follow. Other fellow SFRC students who

have given great support along the way and have always been willing listeners are: Julie









Clingerman, Robin Collins, Trina Hofreiter, Louise Loudermilk, George McCaskill, and

Miriam Wyman.

Other faculty and staff at the University of Florida have often provided much-

needed counsel and direction. Their efforts are duly appreciated: Cherie Arias, Hannah

Covert, Jon Dain, Eric Jokela, Ramon Littell, Christine Staudhammer, and Daniel Zarin.

I must not neglect to thank those who assisted both in terms of hospitality and

technical assistance during the writing process in French Guiana: Sabrina Coste, Paul

Arrive, Eric and Darouny Marcon, Francois Mornau, and Jean-Yves Goret.

I never could have gotten to this stage without the support and love of my parents

and my family. And there are no words sufficient to thank my husband, Chris Baraloto.

My appreciation will probably never make up for all of the cuts and stings suffered in the

tabocal forest of Acre while he acted as my botanist and trail cutter, but it has

nonetheless been an incredible journey. His love of the forest has been one of the greatest

educations in this process.
















TABLE OF CONTENTS



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

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

LIST OF FIGURE S .................................................. ... ...... .............. ....

ABSTRACT .............. .................. .......... .............. xi

CHAPTER

1 G EN ER A L IN TR O D U C TIO N ......................................................... .....................1

2 ECOLOGICAL LIMITATIONS OF REDUCED IMPACT LOGGING AT
COMMUNITY AND SMALLHOLDER SCALES ......... ......................................8

Introdu action ............... ... .................... ......... ..... ........................ ....... ..8
Sustainable Forest Management (SFM) and Reduced-Impact Logging (RIL)
M o d e ls .............................. .......... ..... ... .................... . ........... .1 1
Community/Smallholder Scale vs. Industrial Operations .......................................13
To What Extent Does the RIL Model Ensure SFM at the
Sm allholder/Com m unity Level? .......................................... ......... .... ............... 19
Pre-harvest Forest Inventories, Botanical Identification and Selection of Seed
Trees................ ...................................... 23
M inim um D iam eter Cuts ............................................................................. 27
Low Harvest Intensity ............. .... ................................... 30
Skid T rail P planning .............................. ........................ .. ........ .... ............32
R otation C ycles ............................................... ........ ...... 33
L ian a C u ttin g ...............................................................34
Conclusions................. ........................................36
Logging Block Size and Implications for Pre-harvest Inventories ................ 37
M o n ito rin g ................................................................3 9
M moving beyond R IL ........................... ......... ............................. ............... 40

3 EVALUATING LOGGING DAMAGE TO FUTURE CROP TREES IN ACRE,
BRAZIL: IMPLICATIONS FOR MANAGEMENT OF A CERTIFIED
C O M M U N IT Y F O R E ST ................................................................ .....................44

Introduction .............. ..... .. .......... ......... ........... .... ..... ........ 44









Site D description ...................... ...... ...... ........................ ........ .. 47
Forest Management and Harvesting Operations...................................................49
M methods ...................................................... ................................5 1
Future Crop Tree (FCT) Selection ............................ ...... ...................... 51
Experimental Design and Pre-logging Inventory ..............................................52
A rea D istu rb ed ......................... .......... ....... ..... ........................... 52
FCT Location in Bamboo (Guadua)-dominated Forest............... ............... 53
Liana Presence ....................... ......... .......................... .... ....... 53
H harvest Intensity .................. ..................... ......... .......... .. ................53
M parking Treatm ent .................. ...................... ......................... 54
Assessment of FCT Damage ............ ........................................ ....................54
S statistical A n aly ses........... ...... ................................................ .. .... ...... 5 5
Results ...................... .................. ........ .........................55
Overview of the Logging O operation ............................... ................................. 55
F C T D a m a g e ................................................................................................. 5 6
A re a D istu rb e d ............................................................................................... 5 7
Discussion .............. ..................................... .........57
FCT Damage in Guadua-dominated Forest .................................... ..... 59
Impact of Lianas on FCT Damage ........................... ......................... 61
Impact of Harvest Intensity on FCT Damage..........................................62
Impact of Marking Treatment on FCT Damage................................ ..............63
C o n c lu sio n .................................................. .................. ................ 6 4

4 C O N C L U SIO N ......... ......................................................................... ........ .. ..... .. 78

R E F E R E N C E S ............................................................................ 8 1

B IO G R A PH IC A L SK E TCH ..................................................................... ..................94
























viii















LIST OF TABLES
Table
page
2-1 Contrasting tree population parameters from inventories at two different spatial
scales comm only used by sm allholders.. ...................................... ...............42

2-2 An example of spatial aggregation among neotropical tree species harvested by
com m unities in A cre, B razil.......................................................... ............... 43

3-1 Commercial timber species inventoried during 2004 logging operations in the
PAE Porto Dias, Acre, Brazil. For those species harvested in 2004-2005, total
volum e (m 3 ha 1) is given ............................................................. .....................67

3-2 Classification of damage sustained by FCTs (future crop trees) during logging
operations (modified from Jackson et al. 2002 and Krueger 2004).........................69

3-3 Summary of harvesting impacts in 2004 for the three logging blocks at PAE
P orto D ias, A cre, B razil. ........................................ ............................................70

3-4 Results of a logistic regression for the probability of damage to future crop trees
(FCTs) at Porto Dias, Acre, Brazil during 2004 logging operations......................71














LIST OF FIGURES


Figure page


3-1 Photograph taken in PAE Porto Dias, Acre, Brazil, demonstrating the re-curved
thorn on a branch of a Guadua culm ........................................... .. ........... .......... 72

3-2 PAE Porto Dias, with the four landholdings inventoried in the study, including
the three (in blue) harvested in 2004 ............................................. ............... 73

3-3 Logging block of Barrinha 1, showing skid trails, harvested trees, and location
of future crop trees (FCTs) in the zones of impact .............................................74

3-4 Logging block of Barrinha 3, showing skid trails, harvested trees, as well as
location of future crop trees (FCTs) in the zones of impact...............................75

3-5 Logging block of Palestina, showing skid trails, harvested trees, as well as
location of future crop trees (FCTs) in the zones of impact...............................76

3-6 Relationship between size of felled trees and the surface area of the
corresponding logging gap created during felling......................... .................77














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

COMMUNITY-BASED TIMBER MANAGEMENT IN ACRE, BRAZIL, AND ITS
IMPLICATIONS FOR SUSTAINABLE FOREST MANAGEMENT

By

Cara A. Rockwell

December 2005

Chair: Karen Kainer
Major Department: Forest Resources and Conservation

In recent years, there has been much debate about whether conservation goals can

be achieved by logging old-growth tropical forests. Even so, the global demand for

tropical hardwood is not likely to subside anytime soon, making good management of

forests outside of parks and other protected areas essential. Some researchers have

suggested that rural communities may be the most ideal candidates to engage in

sustainable timber management, since they generally harvest at low intensities. Yet, it has

also been noted that the impact of natural resource exploitation at this scale is still not

well known, and that sustainable forest management (SFM) guidelines will necessarily

vary, depending on the location in which they are being implemented.

Field studies were conducted in Acre, Brazil to assess the impacts of the Porto Dias

community timber management project. The main objectives of this investigation were

(1) to determine the extent of area disturbed (treefall gap and skid trail construction) in a

small-scale certified logging operation, (2) to determine a difference in damage incidence









to future crop trees (FCTs) > 20cm dbh between forest with or without bamboo (Guadua

spp.), (3) to determine a difference in damage incidence to FCTs > 20cm dbh between

those associated with lianas and those that are not, (4) to determine to what extent

harvesting can be conducted more intensely (m3ha1), without incurring greater damage to

FCTs > 20cm dbh, and (5) to determine to what extent marking of FCTs diminishes

damage during logging. Related to this experimental study, an examination of key

reduced-impact logging (RIL) harvesting techniques and their potential influences on

population dynamics of tropical tree species in small scale timber operations was also

initiated, with the goal of addressing potential future management strategies for the

smallholder system.

The results from the study indicate that although the effects of marking, liana

presence and location in Guadua-dominated forest were not significantly associated with

FCT damage, the high values reported for area damaged merit attention. Previous studies

have demonstrated that low intensity harvests are often associated with increased damage

levels per tree due to skid trails, suggesting that the gap mosaic of the forest will often

change, even when low-intensity RIL guidelines are followed. Given the series of poorly

understood parameters outlined in this thesis, it is recommended that systematic post-

harvest monitoring accompany logging operations, so that the benefits of recommended

practices may be evaluated by managers at the appropriate scale.














CHAPTER 1
GENERAL INTRODUCTION

Deforestation is one of the greatest environmental challenges of the past few

decades (Food and Agricultural Organization (FAO) 2005). Brazil, which claims 63% of

the Amazon basin (Lentini et al. 2003), has long been at the center of the tropical

conservation debate. The Brazilian Amazon constitutes approximately one-third of the

world's extant tropical forests (Uhl et al. 1997). Since the 1960s, however, the country

has been plagued by destructive development policies that have opened up the region to

industrial logging, mining, ranching, and large-scale colonization projects, contributing to

a 14% forest cover loss to date (Nepstad et al. 2002). Compounding the problem of forest

cover loss, Brazil is simultaneously the world's largest producer and consumer of tropical

timber, most of which is generated through land clearing or unplanned selective logging

(Verissimo and Barreto 2004).

Selective logging is the harvesting method in which individual trees (rather than an

entire group) are removed. When conducted on a rotational basis, selective logging is

generally regarded to have a low impact on the forest (Uhl and Buschbacher 1988, but

see Sist et al. 2003 a,b, Asner et al. 2005). Unplanned selective logging, however,

typically excludes the use of a systematic timber inventory or management plan and often

harvests the same area over short, repeated intervals, frequently removing high volumes

of timber. The primary goal in this case is to extract the largest and most valuable

commercial species with little regard to surface area damaged or future harvests (Schulze

2003). In many instances, this has led to acute (up to 50% or more) canopy reduction,









severely disturbed mineral soils and damage to the residual stand (Uhl and Vieira 1989,

Verissimo et al. 1992, Johns et al. 1996). In response, scientists and practitioners alike

have proposed implementing sustainable forest management (SFM) practices that could

limit canopy cover loss and protect ecosystem function and biodiversity while

maintaining the economic value of the forest (Putz 1994, 1996, Pearce et al. 2003).

Definitions of SFM differ, but the term generally implies the sustenance of many forest

products and services over long periods of time (Pearce et al. 2003). Yet, despite the

recent conservation interest in SFM, there are many who would argue that to link

conservation with "sustainable forest management" is erroneous, and that the only way to

curb deforestation is outright protection (Rice et al. 1997, Bowles et al. 1998, Terborgh

1999).

Recently, there has been a massive undertaking in Brazil to reduce the negative

environmental impacts of regional silvicultural practices, much of which is generated by

large scale industrial operations. Forest management research in the region was first

initiated by FAO in the 1950s (Oliveira 2000), but it was not until the early 1990s that the

Brazilian federal government began advocating reduced-impact logging (RIL) techniques

as a means to combat rampant predatory logging in the Amazon basin (Holmes et al.

2002, Schulze 2003). RIL is an essential component of SFM if timber is to be harvested

from the forest (Pearce et al. 2003), typically employing a pre-harvest forest inventory,

road and skid trail planning, pre-harvest vine cutting, and directional felling (Dykstra and

Heinrich 1996). Indeed, in the absence of secure prices for non-timber forest products

(NTFPs) and infrastructural support for ecotourism efforts, selective logging under the

RIL model at low intensities may be one of the few assured income-generating activities









that can preserve the structure of the forest without compromising the entire ecosystem

(Oliveira 2000).

The SFM model has generated considerable interest for community managers and

smallholders in the region, with the last few years witnessing the creation of a number of

innovative community-based conservation experiments founded on SFM principles

(Kainer et al. 2003, Stone 2003). Extractive reserves, or tracts of forest under community

control with secure usufruct rights, figure prominently among these community

development efforts. The remarkable and often violent history that preceded the

emergence of such initiatives as the extractive reserve (RESEX) program immediately

gave these forest communities widespread international support, but their conservation

goals have since become the focus of intense scrutiny (Nepstad et al. 1992, Redford and

Stearman 1993, Terborgh 1999, 2000). Many researchers, although generally not

contesting the social justice value that such endeavors provide to a historically

disenfranchised population, have questioned the wisdom in holding up the extractive

reserves as models for biodiversity conservation (Redford and Stearman 1993, Terborgh

2000). This issue has become especially pertinent now that many community forests have

shifted from local economies based on non-timber forest products (NTFPs) to that of

timber extraction (Kainer et al. 2003, Schmink 2004). The general consensus among

some researchers is that external economic and cultural pressures, in concert with a

tendency to abuse the resource commons (see Hardin 1968), will hinder the ability of

community forest managers to protect biodiversity (Redford and Stearman 1993,

Terborgh 2000).









In Brazil, though, voluntary forest certification has provided a strong incentive for

both community forest managers and commercial logging companies to adopt SFM

practices (May 2002, Kainer et al. 2003). As of April 2004, Brazil ranked fourth in the

world in number of certified forests and eighth in area (slightly under 1.6 million ha,

529,079 ha of which are considered natural forest) (May 2002). The failure of

government policies to curb illegal logging gave birth in the 1990s to the certification

instrument, which endorses forest products originating from sustainable forestry

operations, a process often executed by a third-party organization (Molnar 2003). Forest

certification's role as a community-based conservation mechanism has received

widespread support (Molnar 2003). This has been particularly true of the Forest

Stewardship Council (FSC), an international organization that has had a strong influence

in the development of certification standards in the Brazilian Amazon (Humphries 2005).

Ultimately, the certification process could provide access not only to the global certified

forest product market but also to silvicultural evaluation and assistance, thereby creating

motivation to conserve the resource base (Putz 1996, Putz and Viana 1996, Putz and

Romero 2001).

Despite their positive associations with the conservation movement, a number of

researchers have criticized forest certification bodies and biologists for producing SFM

guidelines and criteria that fail to recognize the problems that such a system might create

for smallholders (Pretty and Hine 1999, Molnar 2003). Even though community timber

management operators may have the knowledge, "human nature and economics" can

offset the desire to sustainably manage the forest (Salafsky et al. 1998). Smallholders and

community forest managers in the rural tropics have limited financial and technical









resources, so implementing additional silvicultural treatments and monitoring may be

unfeasible for them, unless the proposed activities are seen as adding to their

productivity. Furthermore, given intense competition from less sustainable commercial

industrial timber enterprises and illegal logging, the long-term impact that certification

will have on the industry and deforestation in general is questionable, particularly in

Latin America (Sierra 2001). In Brazil, for example, certified wood is currently meeting

only 2% of the annual demand for logs (Lentini et al. 2003).

Community and smallholder forest enterprises are generally thought to be more

environmentally benign than their industrial counterparts (Salafsky et al. 1998), making

them seemingly ideal candidates to implement the SFM and RIL models. There are,

however, a number of inherent ecological limitations associated with small-scale timber

operations that are rarely addressed in the management literature, with few ecological

investigations regarding the impact of timber extraction having been conducted at the

level of the community or smallholder level (Oliveira 2000, Walters 2005). Although RIL

is, in general, ecologically beneficial to the residual stand, there is evidence that the

magnitude of those benefits varies, depending on the forest being exploited (Fredericksen

and Putz 2003, Schulze 2003) and the specific RIL techniques being used. RIL was

initially developed for large scale industrial operations where harvest intensities and

investment potential are high (Oliveira et al. 1998). Nevertheless, community forest

managers have eagerly adopted RIL standards, even though the ecological impacts of

RIL at this scale are not well-known. Given that these impacts are poorly understood,

there is a need to explore both the limitations and the benefits of RIL for these









community and smallholder-based systems rather than embracing what is held to be the

ideal by researchers working with larger operations.

In this thesis I examine both RIL guidelines as generally implemented at the

community and smallholder levels as well as a specific case study in PAE Porto Dias,

Acre, Brazil, testing the harvesting impacts generated by the activities of the "Multiple-

Use Community Forest Management Project" on the residual stand. This approach

potentially addresses some of the challenges of community timber operations in Acre as

well as examining future crop tree (FCT) damage as sustained during RIL operations.

In Chapter Two, I examine some of the ecological constraints associated with

timber operations at the community and smallholder scales. To do so, I evaluate several

assumptions of selected RIL recommendations and their implications for smallholder

management. After a brief introduction to SFM and RIL, particularly as they relate to the

Brazilian Amazon, I discuss eleven key differences between industrial and

community/smallholder operations. I then ask to what extent the RIL model contributes

to SFM at the smallholder/community level, with an examination of RIL harvesting

techniques and their potential influences on the population dynamics of tropical tree

species in smallholder systems, with a focus on the Brazilian Amazon. I conclude the

paper with a discussion of potential future management strategies for smallholders.

In Chapter Three, I present research investigating the impacts that forest type

(bamboo/non-bamboo), liana presence, harvesting intensity, and marking have on FCT

damage during logging operations in PAE Porto Dias. The study focuses on the adult and

sub-adult size classes (> 20 cm dbh) of commercial timber species in the hopes of






7


capturing information about the availability of timber at the time of the next harvest (30

years).

This thesis has been organized such that the second and third chapters are two

individually and fully structured papers for publication. Each one of these chapters has its

own conclusions, while Chapter 4 summarizes the main findings of the entire study and

provides a discussion of the future directions for small-scale timber management

research.














CHAPTER 2
ECOLOGICAL LIMITATIONS OF REDUCED IMPACT LOGGING AT
COMMUNITY AND SMALLHOLDER SCALES

Introduction

Researchers and practitioners alike have proposed that large tracts of tropical

forests will not be conserved without the involvement of local forest residents (Scherr et

al. 2002, Shanley and Gaia 2002). Already, approximately one quarter of the world's

forests are community-owned or managed, a proportion that is expected to double in the

next 15 years (Molnar 2003). Some have advised that these forests will stand a greater

chance in surviving when placed under the jurisdiction of communities, assuming they

will have a desire to conserve their forest resource base (Camino and Alfaros 2000). In

concert with a growing smallholder interest in formal timber production and a need for

sustainable forest management (SFM) practices outside of protected areas (Putz et al.

2000), it is likely that community and smallholder forests will play an increasingly

important role in global conservation.

For various reasons, including an international market demand for sustainably

harvested timber products, many of these community and smallholder operations are

becoming third-party certified (Molnar 2003). This voluntary process emphasizes

rigorous social and environmental standards for production forests, potentially providing

a strong economic incentive for local forest managers to value the resource base (Putz

1996, Putz and Viana 1996). The reduced impact logging (RIL) management models

favored by the certification instrument, however, have failed to develop situation-









appropriate strategies that accommodate the ecological and social complexities of small-

scale commercial timber extraction. Instead, the tendency has been to impose

management practices of large commercial operations onto smallholder systems (Oliveira

2000), causing some forest operations to take on corporate features usually relegated to

industrial timber enterprises (Putz 2004). It has been suggested that further research and

operational training are needed to evaluate how variations in forest type and size of

logging operations affect the performance of RIL systems (Holmes et al. 2002).

Without much empirical information suggesting otherwise, it is often assumed that

because many smallholders harvest at low intensities and utilize animal traction (Oliveira

2000) or small machines, they are more environmentally benign than industrial logging

enterprises (Salafsky et al. 1998). Nevertheless, there are a number of inherent ecological

limitations associated with small-scale timber operations and low intensity harvests,

which have been shown to have substantial impacts on forest structure, composition and

regeneration (Smiet 1992, Gullison and Hardner 1993, Boot and Gullison 1995, Whitman

et al. 1997, Awasthi et al. 2003, Walters 2005). Community and smallholder-owned

forests generally encompass smaller areas compared to their industrial counterparts (<

500 ha) (Lentini et al. 2003). When managing for timber in small, non-contiguous

management units, ecological constraints to regeneration, such as insufficient pollen

transfer and/or inbreeding (Ghazoul et al. 1998), dispersal limitation (Dalling et al. 2002),

and predator satiation (Schupp 1992) can become exacerbated, as population sizes of all

species are correspondingly small (Putz et al. 2000).

RIL's emphasis on environmental sustainability is often assumed to be a natural fit

for rural forest dwellers (Braz and Oliveira 1996). However, it is becoming increasingly









clear that the shift from traditional management systems to that of third-party certified

commercial timber extraction is challenging. Smallholders and communities are typically

required to implement RIL techniques to meet certification standards, compelling them to

invest in both equipment and human capital not normally required for their daily

subsistence needs (Stone 2003).

Although the RIL model has been demonstrated to have many ecological benefits

in tropical forests, there is a need to explore both the limitations and the benefits of RIL

specifically for smallholder systems. In this paper, I evaluate key assumptions of several

RIL recommendations and their implications for smallholder management. I use the

terms smallholderr" and "community" interchangeably throughout, though there is a

distinction (albeit broad) between smallholders living on individual lots in colonization

projects and traditional or indigenous forest-dwelling communities, such as those living

in or near forests under some form of protection. As well, when referring to the terms

"community" and smallholderr" management in this paper, I am specifically addressing

locally-run operations, not to be confused with commercial logging concessions on

smallholder properties (e.g. Nepstad et al. 2004). After a brief introduction to SFM and

RIL, I discuss eleven key differences between industrial and community/smallholder

operations. I then ask to what extent the RIL model ensures SFM at the

smallholder/community level, with an examination of RIL harvesting techniques and

their potential influences on the population dynamics of tropical tree species in small

scale systems, with a focus on the Brazilian Amazon. I conclude the paper with a

discussion of potential future management strategies for the smallholder system.









Sustainable Forest Management (SFM) and Reduced-Impact Logging (RIL) Models

Selective logging when applied on a long rotational basis and at low harvest

intensities is generally regarded to have a low impact on the residual stand (Uhl and

Buschbacher 1988, but see Sist et al. 2003a,b, Asner et al. 2005). Selective logging,

however, can be one of the first steps in forest conversion, providing a subsidy for

pasture formation as in the Brazilian Amazon case (Oliveira 2000). Additionally, very

few commercial timber operations in this region employ sustainable selective logging

techniques, with an estimated 95% of all timber extracted through unplanned

conventional logging (Verissimo et al. 2002).

The extent to which selective logging causes permanent damage can be controlled

in some ways by planned management activities, as has been suggested by the SFM

model. SFM can be defined based on the following assumptions: (1) management can be

exercised in a manner compatible with the maintenance of biodiversity, (2) management

of tropical forests is economically viable, and (3) management can result in sustained

timber yield over the long term (Bawa and Seidler 1998). Translating these concepts into

practice, Bruenig (1996) maintains that when discussing timber extraction, SFM should

encompass the following: "canopy openings should be kept within the limits of natural

gap formation; stand and soil damage must be minimized; felling cycles must be

sufficiently long and tree marking so designed that a selection forestry canopy structure

and a self regulating stand table are maintained without, or with very little, silvicultural

manipulation; (and) production of timber should aim for high quality and versatility".

Bawa and Seidler (1998) conclude, however, that too often, SFM is promoted without

rigorous examination of the interrelatedness between biodiversity maintenance, economic

viability, and long-term yields, causing some researchers to doubt the validity of SFM









and calling for the outright protection of forested regions instead (see Rice et al. 1997,

Bowles et al. 1998, Terborgh 1999). As well, some researchers have questioned whether

timber harvests in at least some forests can be sustained without silvicultural intervention

(Daube et al. 2005). Nevertheless, the SFM philosophy has been embraced by multiple

forest actors. In Brazil, for example, the federal government, state institutions, NGOs and

research organizations have actively promoted SFM, hoping that it will quell unchecked

land conversion and predatory logging activities (Schwartzman et al. 2000, Nepstad et al.

2002, Fearnside 2003, Kainer et al. 2003).

RIL is considered a necessary component to achieve the goals of SFM, though its

implementation alone cannot maintain biodiversity and ecosystem functions nor even

secure sustained timber yield (referring to the objective of maintaining timber production

for future harvests) (Webb 1997, Putz et al. 2000, Sist et al. 2003b). RIL implies a series

of pre- and post-logging guidelines (see Pinard et al. 1995, Dykstra and Heinrich 1996)

which are designed to protect advanced regeneration (i.e., seedlings, saplings and larger

future crop trees), minimize soil damage, limit damage to non-target populations

(i.e.wildlife and non-timber plants species), protect water quality, and preserve ecosystem

processes (i.e. carbon sequestration) (Putz et al. 2000). Operations under these guidelines

typically employ long-term cutting cycles, pre-harvest forest inventories, road and skid

trail planning, pre-harvest vine cutting, and directional felling (Dykstra and Heinrich

1996). RIL has been shown to reduce damage to the residual stand by 30-50% and has

also demonstrated an increased efficiency of logging operations (see Johns et al. 1996,

Pinard and Putz 1996, Bertault and Sist 1997, Barreto et al. 1998, Sist et al. 1998, Holmes

et al. 2002).









Under pressure from both domestic and international concerns to improve the

ecological sustainability of the tropical hardwood industry, Brazil began to adopt RIL

techniques in the early 1990s, following the model that was originally developed in

Southeast Asia and Suriname a decade earlier. Since that time, many Brazilian timber

companies have agreed to abide by RIL regulations at the behest of federal government

regulations, and in some cases, third-party forest certification standards, hoping to tap a

high-end niche market (Schulze 2003).

Community/Smallholder Scale vs. Industrial Operations

The SFM model (and implicitly, RIL) has generated considerable interest for

community forest managers and smallholders (Braz and Oliveria 1996). Researchers

often point to local peoples' "long-term commitment to place" and their intimate

knowledge of ecosystems (Schmink 2004) as evidence of their ability to both manage and

(when necessary) restrict access to forest resources (Agrawal and Gibson 1999). Yet, it

has also been noted that the implementation of a systematic management regime can

often be overwhelming for communities and smallholders, requiring significantly

different conditions for long-term success than large-scale commercial enterprises

(Schmink 2004). Indeed, to mitigate the challenges faced by these operations, some

suggest that community and smallholder forest management systems must be considered

within case-specific ecological, socio-economic, political, and cultural settings (Berkes

and Folke 2001, Schmink 2004). This viewpoint recognizes the unique character of

individual community and smallholder systems and also further emphasizes the need to

distinguish these enterprises from their industrial counterparts.

Several general operational characteristics distinguish community and smallholder

timber enterprises from the large-scale commercial timber industry, building on a list of









five proposed by Salafsky et al. (1998): (1) Ownership: Industrial operations usually buy

logging rights via a concession, or the timber company may own the land. More than

likely, the proprietors live offsite. In contrast, community and smallholder timber

harvesting rights are owned or, at the very least, held by local inhabitants (Salafsky et al.

1998). For example, in the Brazilian Amazon, one third of the land area is inhabited by

smallholder and indigenous communities with the rights to manage their own forests

(Shanley and Gaia 2002). Due to a strong sense of ownership and multigenerational

responsibilities, community forest managers may implement more sustainable

management practices than their industrial counterparts (Walters et al. 2005). Indeed, in

those community cases where secure land tenure is lacking, there appear to be fewer

incentives to manage the resource sustainably (Banana and Bombya-Ssembajjwe 2000,

Gibson and Becker 2000). (2) Timber as integrated livelihood component: Community

and smallholder forest managers rely on the forest for a suite of goods and services,

viewing timber management as one component of an integrated system rather than a sole

income-generating activity (Salafsky et al. 1998). Since timber harvesting is a seasonal

activity (usually during the driest months), local managers pursue other livelihood

activities throughout the remainder of the year (Oliveira 2000). Even so, timber

management and other livelihood activities are not always compatible, especially when

harvesting is conducted by those from outside the community. In one Eastern Amazonian

case, community members from the Tapaj6s-Arapiuns Extractive Reserve believed that

commercial logging negatively impacted NTFP collection (due to incidental damage) and

hunting (through easier game access by logging roads) (Menton 2003). Researchers also

have reported a lack of documented systems explicitly integrating timber and NTFPs in









management plans (Bawa and Seidler 1998, but see Salick et al. 1995, Romero 1999).

Local forest managers who wish to protect a multitude of forest services (NTFPs, game

species, watershed) may be more inclined to adopt RIL guidelines geared to protect

components of the forest beyond future timber harvests. (3) Scale: Community and

smallholder enterprises generally operate on a smaller scale than their industrial

counterparts, the latter tending to rely on capital intensive techniques and heavy

machinery rather than local labor (Salafsky et al. 1998). Many techniques developed

specifically for communities have been initiated in secondary forests, where the use of

animal traction is ideal for the extraction of small diameter logs (Pinedo-Vasquez et al.

2001). Smallholder logging in primary forests has become more common in recent years,

particularly in the Brazilian state of Acre (see Kainer et al. 2003, Stone 2003), but the

need for heavy machinery in these systems and the subsequent competition with larger

enterprises has constrained community participation in this type of extraction (Oliveira

2000). However, smaller scale operations may facilitate more intensive management for a

suite of services and products, something that might be impractical in a larger

management unit. Also, in Brazilian community timber efforts, harvest intensities tend to

be low, usually no more than 10 m3 (or two trees) ha-1 (Oliveira et al. 1998), which (in

theory) should be compatible with RIL guidelines. In comparison, most Brazilian

conventional forest management systems harvest anywhere from 30 to 40 m3 ha- yr-

(Oliveira 2000). (4) Added-value: Industrial logging operations harvest roundwood logs

that are then transported to a centralized offsite-processing mill. Conversely, community

and smallholder operations often seek to add value to the lumber onsite, transforming the

raw material into finished or semi-finished products (Salafsky et al. 1998). This









potentially provides greater income to local residents, such as the case of at least one

certified community operations in Acre, Brazil that is producing artisan objects from

locally harvested wood (pers. obs.). (5) Local reinvestment: In community-based

systems, capital is generally reinvested locally and there is a greater long-term incentive

for sustainability (Salafsky et al. 1998). In comparison, industrial timber enterprises tend

to move profits to other localities or sectors of the economy, thereby reducing any

potential social benefits from forestry activities (Salafsky et al. 1998). For example, in

Oaxaca, Mexico, Bray et al. (2003) note community use of logging profits to encourage

more benign forest enterprises (i.e., water bottling, ecotourism, and resin-tapping) and

build more community assets (i.e., potable-water networks, schools, and clinics). And

even though timber extraction has subsidized cattle pasture expansion in the Brazilian

Amazon (Oliveira 2000), Summers et al. (2004) found that smallholder timber managers

in Rond6nia tended to clear less forest and had smaller cattle herds than their neighbors.

(6) Adaptive management/traditional governance: Communities and smallholders

sometimes develop adaptive management systems that permit local users to identify

internal and external threats to the forest (Margoluis and Salafsky 1998, Salafsky et al.

1998). Some researchers have recognized this as a benefit of traditional governance, or

locally-defined institutions and regulations that guide decision making processes (Banana

and Gombya-Ssembajjwe 2000, Becker and Le6n 2000, Bray et al. 2003). For instance,

Becker and Le6n (2000) found that the Yuracare of the Bolivian Amazon have clear,

locally-defined rules regarding forest resources, upon which they can control both access

and use of the forest at multiple scales. (7) Value of forest to future generations:

Intergenerational values or investment in the forest for future generations also distinguish









community timber enterprises from their industrial counterparts, a factor often

overlooked (see Bowles et al. 1998) when discussing excessive production costs of

community forests (Bray et al. 2003). In several studies conducted in the Brazilian

Amazon, rural respondents identified silvicultural activities as positively impacting the

future of their children and grandchildren (Walters et al. 2005). (8) Traditional

ecological knowledge: Rural peoples who have often lived in the same geographic

region for generations have accumulated considerable knowledge about the natural

resource base through daily observations in their subsistence activities, usually

demonstrating a comprehensive understanding of ecosystem linkages (Folke et al.1998,

Berkes 1999, Calheiros et al. 2000, Berkes and Folke 2002). For example, it has been

noted that despite the highly technical silvicultural expertise available in Finland, the

main factor explaining the excellent forest health in that country is that conservation

practices have been based on TEK passed from one generation of forest managers to

another (Banuri and Apffel-Marglin 1993). Accordingly, TEK can be applied to a

systematic forest resource management plan, providing details on timber quantity and

location, regeneration requirements, and other requisites for successful implementation of

RIL guidelines. (9) Commitment to forest conservation: Mexican communities in

Oaxaca and Quintana Roo consistently logged below the authorized volume (some by as

much as 37%), suggesting a willingness to reduce the volume of extraction when

inventories indicate external threats to the resource base. Some of these same

communities also prohibit or limit hunting in local forests, displaying an increasing

commitment to biodiversity conservation (Bray et al. 2003) or perhaps, a clear

understanding of sustainable game harvest limits. Other forest managers in Acre, Brazil









have developed local indicators for advanced regeneration as determinants for harvest.

By identifying a sufficient number of what they term as "daughters" and

"granddaughters" of the tree(s) to be harvested, they acknowledge the importance of

leaving several individuals of differing size classes of a particular species in the

management unit, not just reproductive adults (P. Roth, pers. comm.). (10) Participatory

monitoring: Local participants, especially those termed "parabiologists" or

"parataxonomists", have been instrumental in monitoring environmental impacts and

planning formal conservation initiatives, and, in some cases, have actually improved the

data quality of the academically-trained researcher (Janzen 2004, Sheil and Lawrence

2004). In one Brazilian case, local participants monitored the populations and phenology

of selected fruit tree species in collaboration with professional biologists, eventually

determining that the environmental impact of timber extraction was too great a risk to the

long-term NTFP value of a particular fruit species (Shanley and Gaia 2000). These

examples suggest that local forest dwellers are very aware of forest dynamics and capable

of executing environmental monitoring (especially those post-logging surveys

encouraged by the RIL paradigm) when provided with some capacity building and

equipment (Sheil and Lawrence 2004). (11) Influence/funding from outside

organizations: Particularly since the early 1990s, a network of community forests in the

Brazilian Amazon have received financial and infrastructural support of a number of

NGOs and governmental institutions (Stone 2003). This support has led to the voluntary

third-party certification of at least 7 community operations in the region, fortifying the

competitiveness of community forest management in a demanding market dominated by

industrial giants. Indeed, two of the principal barriers faced by community and









smallholder managers are the small scale production and consequent marketing

difficulties, limitations that certification may be able to improve, primarily through

access to a niche market (May 2002). Yet, Stone (2003) also warns of the increasing

reliance of certified community timber projects on outside funding and technical

assistance, mostly because of the complex requirements inherent with RIL adoption, such

as investment in expensive equipment and the writing of management plans.

To What Extent Does the RIL Model Ensure SFM
at the Smallholder/Community Level?

Prior to the recent emphasis on the RIL model, traditional smallholder forest

exploitation in the Brazilian Amazon was based on low technology methods that

produced minimal financial returns, but also resulted in low environmental impacts

(Oliveira 2000). Indigenous inhabitants have been managing forests in the region since

pre-Columbian times (see Heckenberger et al. 2003) for a suite of subsistence products

and services (Summers et al. 2004), a pattern that has continued to the present. Rural

Amazonian inhabitants typically integrate wood production into their existing swidden-

fallow systems, resulting in a highly patchy landscape and allowing smallholders to

maintain favorable environments for several different species (Pinedo-Vasquez et al.

2001). It was not until the early 1990s, amidst a deepening international concern for

global deforestation and the creation of the Program for the Conservation of the Brazilian

Rain Forest (PPG7) in 1992, that the formal establishment of SFM projects for

community and smallholder forests emerged (Stone 2003), eventually leading to the first

RIL project for privately-owned smallholder forests in the state of Para in 1993 (Amaral

and Neto 2000). By the mid 1990s, the Brazilian federal government also began to take

an interest in community-based SFM models and in 1998, the Brazilian Institute of the









Environment and Renewable Resources (IBAMA) (the federal agency directly

responsible for the execution of federal forest policy) implemented the PMFSimples

(Simple Sustainable Forest Management Plan) protocol, allowing smallholders to manage

up to 500 ha for timber (Stone 2003). It also enabled community forest initiatives to

submit one management plan for all participating properties, rather than do this

individually. Since 1995, at least 7 smallholder and community forest projects have been

established in the region, under a wide range of property regimes extractivee reserves,

national and state forests, indigenous reserves, and agricultural colonization projects)

(Stone 2003). Community and smallholder interest in timber has been motivated in great

part by a need to increase and diversify livelihoods, fueled in particular by low net returns

for NTFPs (Kainer et al. 2003).

However, as the sophistication and intensity of community and smallholder forest

management systems increases, it has also become evident that silvicultural prescriptions

appropriate to local conditions are needed, especially those that take into account a

shortage of investment and human labor (Oliveira 2000). The Brazilian RIL model was

initially developed for large scale industrial operations where investment potential is high

(Oliveira et al. 1998). For example, since 1995, training in RIL techniques for Brazilian

operations has been a primary focus of the Tropical Forest Foundation (TFF) and its

Brazilian subsidiary, Fundacgo Floresta Tropical (FFT) (Holmes et al. 2002). This effort

has resulted in the development of several model systems throughout the Amazon basin,

but most of these examples are considered large-scale operations. Not surprisingly, these

types of models have been eagerly adopted by community forest managers, even though









the RIL ecological impacts on the overall forest structure at this scale are not well-known

(Oliveira et al. 1998).

In its original conception, one of the basic tenets of RIL mandated that logging

should mimic the so-called gap-phase dynamics of the neotropical rainforest (see Brokaw

1982) by creating small-scale disturbances around the extracted crop tree (Hartshorn

1989). Natural small-scale disturbances (usually involving the death and subsequent fall

of a neighboring dominant tree) permit access to limiting light (and other) resources. By

attempting to mimic this type of disturbance during RIL operations, it is hoped that

residual stand damage will decrease and that growth rates of surviving trees will increase.

Natural disturbances, however, are qualitatively and quantitatively different from those

logging-induced disturbances (Boot and Gullison 1995, Vitt et al. 1998, Dickinson et al.

2000). Boot and Gullison (1995) note that mortality rates of trees in forests with only

natural disturbances are on the order of 1-2% (see Swaine et al. 1987), while

conventional selective timber harvests can damage as much as 55% of the residual stems.

Additionally, they point to other sources of residual stand degradation, such as soil

compaction, erosion, and changes in drainage, that are not part of a natural small-scale

disturbance.

Although RIL techniques are designed to minimize this type of aggravated impact,

some species of commercially-valuable trees require larger disturbances for reproduction

than is mandated by minimal impact logging techniques, which usually emphasize low

harvest intensities (Fredericksen and Putz 2003). Careful planning may maintain timber

volume but the removal of scarce, high value logs could result in replacement by less

important species, ultimately reducing forest biodiversity and dooming the second cutting









cycle to timber not yet recognized by the market (Schulze 2003). In this case, the gap-

phase management paradigm could be in direct opposition to the ecological role of

catastrophic natural and anthropogenic disturbances in forests, conditions to which

particular species have adapted over time (Deneven 1992, Bush and Colinvaux 1994,

Fredericksen and Putz 2003). Fredericksen and Putz (2003) point to examples of

neotropical trees that may not benefit from a low-impact management regime, such as

Swietenia macrophylla, Bertholletia excelsa and Cedrela spp., species which appear to

establish in large disturbed areas.

Another of the underlying assumptions of many RIL guidelines is that advanced

regeneration of valuable commercial timber species is abundant and exists uniformly

throughout the stand (Schulze 2003). However, given the suite of differing growth rates,

wood densities, population dynamics, and modes of seed dispersal, both between and

within commercial species in the Amazon basin (Schulze 2003), it may be that

knowledge of tree autecology is one of the greatest limiting factors in implementing SFM

(Hammond et al. 1996, Guariguata and Pinard 1998). Currently, there is a lack of

sufficient ecological information to help guide forest management decisions in the

Brazilian Amazon, which is one of the reasons forestry methods have not advanced much

beyond RIL (Schulze 2003). Yet, it is prudent to evaluate small-scale, RIL-based logging

operations to date, and discuss how some of the basic tenets of RIL (pre-harvest forest

inventories, selection of harvest and seed trees, minimum diameter cuts, low harvest

intensity, skid trail planning, rotation cycles, and liana cutting) have worked at the

smallholder level and how they can be modified.









Pre-harvest Forest Inventories, Botanical Identification, and Selection of Seed Trees

A full pre-harvest inventory provides managers with detailed information on

distribution, quantity, and size of potential harvest trees and future crop trees (Schulze

2003). Unfortunately, for many small operations (as well as some large ones), 100%

inventories are cost prohibitive, especially when the entire management unit (as is the

case of those community management projects in the extractive settlement projects and

extractive reserves in Amazonia) is as large as 300 ha. In at least one study in Quintana

Roo, Mexico, local rural people identified time and finances as two of the biggest

constraints for the inventory process (Lawrence and Roman 1996).

In response, it has been common to inventory only the logging block for the following

year, as the inventories are generally carried out one year prior to harvesting activities.

Although this may eliminate the need to pay for a costly comprehensive inventory, it

limits the manager's ability to harvest with respect to species population biology (see

Table 2-1). For example, in a 10 ha logging block chosen from a 300 ha landholding,

several individuals of a particularly valuable species are identified during the 10 ha

inventory. For management purposes, these individuals are then the sole representatives

of that species within the 300 ha landholding, since a full inventory of adult trees has not

been carried out. Many commercial tropical tree species are characterized by highly

clumped spatial distributions (Hubbell 1979; Table 2-2) and are self-incompatible (Bawa

et al. 1985), so harvesting even half of the inventoried individuals could potentially lead

to inbreeding and genetic drift (Jennings et al. 2001).

Ghazoul et al. (1998) demonstrated that the reproductive success of ,\l/,i',t

siamensis was limited by compatible pollen transfer at a reduced population density after

logging. This Allee effect (see Allee 1949) can increase the probability of extinction,









particularly in small populations. For those species occurring in widely spaced clumps, a

reduced density in aggregations may result in a decrease in pollen exchange between

clumps, severely limiting genetic variability within aggregations and thus, their ability to

adapt to changing conditions (Ghazoul et al. 1998).

When asked about this type of problem, local forest managers often respond that

they know their forest well enough to have a general understanding of species

distribution patterns (pers. obs.). This might be true in many cases, but at least one study

conducted in Para, Brazil revealed that during Rapid Rural Appraisal interviews, local

residents grossly overestimated the densities and distributions of game-attracting fruit

species, presumably because they were favoring forest groves where the tree species

occurred in high densities (Shanley and Gaia 2002). Even if the landowner or community

cannot afford to pay for a 100% inventory, other forms of surveys (like rapid-assessment

type inventories) should be tested to give the forest manager a more accurate idea of the

availability of species and their distributions across the landholding. For example, in the

ejido system of Quintana Roo, Mexico, Lawrence and Roman (1996) report that local

inventory specialists prefer small circular plots as opposed to a 100% inventory design.

Although these circular plots may lack some of the statistical accuracy of a full-scale

inventory, they satisfy local needs for ease of implementation and available financial

resources.

Botanical identification during the inventory is crucial for making the most basic

management decisions and should be considered one of the first steps in SFM. In

Brazilian community forest settings, much of the identification is carried out by a forest

manager or tree identifier (mateiro) from the community. Usually in these cases, initial









identifications are recorded as common names. Thanks to the substantial botanical

knowledge base in the community, most mateiros are able to identify species to common

name with few difficulties, using a series of morphological features (i.e., bark, slash, and

odor). The greater problem arises when the inventory sheet is sent back to the office of a

local NGO or state agency and entered into a database. At this point, the common names

are converted to scientific epithets based on lists (often created by a state or federal

agency distant from the management region) that have been circulated between agencies.

Too often, these lists do not correspond correctly to all common names, as the master list

has been generated by a different set of mateiros. The assumption that these common

names will then correspond with each other is highly optimistic, as even mateiros in the

same community will often use different names for the same species. In Lacerda and

Griffiths' (2005) study from the Tapaj6s forest in Para, for example, more than 90% of

common names were technically correct, but once the inventory sheet arrived in the

office and was converted to corresponding scientific names, botanical accuracy was

reduced to less than 40% correct. This type of error can have serious repercussions for

both the selection of harvest trees and post-harvest population dynamics.

The pre-harvest inventory and subsequent botanical identification have strong

implications for the selection of seed trees. Some RIL guidelines require that "mother

trees", or seed trees, are maintained in the management unit at the "appropriate" spacing

and density, based on means of dispersal, individual size, and tree diameter at first

reproduction, as suggested by Guariguata and Pinard (1998). The seed tree concept does

not elude forest managers on the ground. One landowner in a community timber

management project in Acre, Brazil expressed his reluctance at cutting down large-









crowned individuals ofDipteryx spp. (even though such trees would bring an excellent

price on the market), believing them to be good seed sources (as well as being concerned

about the subsequent large gap size created when felled). He would instead take a smaller

conspecific, or what is locally known as a "daughter" tree (F. Correa da Cunha, pers.

comm.).

Target mother tree densities are often around 3-4 trees ha-1, or (as in accordance

with federal Brazilian law) 10% of the harvest compartment, but given that

reproductively mature conspecific densities in many neotropical forests are much lower

(Lieberman and Lieberman 1994, Guariguata and Pinard 1998), these target goals may be

inadequate. As well, if individuals cannot be properly identified to species, their selection

as seed trees remains dubious. For example, current community timber operations in Acre

identify Dipteryx spp. to one common name, cumaruferro, even though at least 7 species

of this genus exist in the region (C. Baraloto, unpublished data). Consequently, in a

hypothetical situation where 30 mature cumaruferro in a 10 ha logging block have been

left as seed trees, it is highly probable that this group entails several species that may be

isolated from other conspecific reproductive adults.

In general, the dilemmas of implementing a pre-harvest inventory, identifying

scientific species, and selecting seed trees can occur at both the community and industrial

scales. Brazilian federal law requires all companies to survey their logging concessions

prior to harvest, regardless if they are certified or not. The difference for community and

smallholder operations is that a pre-harvest botanical inventory will require long-term

budgetary commitments not normally required on the part of a community-based

organization or forest-dwelling individuals (Janzen 2004). Local parataxonomists, during









the time of the inventories, are taken away from other responsibilities with families and

subsistence activities, potentially compromising their situations at home. As such, they

may need to be compensated financially by the forest management group, which can

cause resentment among untrained community members who do not benefit directly from

the botanical survey.

Minimum Diameter Cuts

The most common harvesting method in the tropics is selective logging, in which

only the largest, most valuable individuals are removed from the forest (Webb 1997).

One of the justifications used to support the practice of employing a minimum diameter

cut is that smaller individuals of the same species will grow to replace the harvested trees.

However, application of minimum-diameter cuts (often termed "high grading", or taking

the biggest and the best) is riddled with undesirable outcomes, and has thus been

criticized as a "short-term gain for long-term loss" (Nyland 2002). As minimum diameter

cuts are often defined by market demands rather than tree biology, Sist et al. (2003a)

suggest that diameter limits be based instead on the structure, density and diameter at

reproduction of target species. Minimum diameter cuts are not necessarily an inherent

part of RIL, but their implementation is linked to the polycyclic cutting cycle, which is a

common RIL recommendation.

High grading typically dooms the next harvest cycle to small, poorly-formed trees

of lesser commercial value or, at the very least, leaves inferior genetic stock to produce

the next generation (Smith et al. 1997, Fredericksen and Putz 2003). For example, Vester

and Navarro-Martinez (2005) point out that selective logging of the largest trees may

deplete a population of the most favorable genotypes, eventually leading to genetic

erosion. The extent to which this selection will affect populations depends on a number









of factors, including harvest intensity, the contribution of the residual trees to

regeneration, and the degree to which undesirable traits are genotypic rather than

phenotypic (Jennings et al. 2001). As well, strong correlations have been drawn between

tree size and fecundity (Chapman et al. 1992), so the removal of a stand's largest

individuals often eliminates the most prolific seed producers, important for both wildlife

as well as regeneration (Thiollay 1992).

Most tropical rain forests have more individuals below the minimum diameter

cutting limit than above. When the reverse is true, the population is often assumed to be

in decline (Jennings et al. 2001). One forest type representing the latter category is the

arborescent bamboo (Guadua spp.)-dominated forest of southwestern Amazonia, which

has been overlooked by the literature for management considerations (see Chapter 3).

Griscom (2003) hypothesizes that if Guadua achieves dominance in managed forests,

either the majority of tree stems will become competitively excluded, or a substantial

proportion of the remaining stems will become deformed as a result of mass loading

(referring to the added physical mass of the bamboo, which can displace aboveground

tree structure (Griscom 2003)). He notes that it is not necessarily the inability of juvenile

trees to germinate and establish in Guadua stands that distinguishes the reduced above-

ground biomass in these forests, but rather the failure of seedlings and saplings to recruit

into larger size classes. These factors present a few challenges to the minimum diameter

cut, such as lack of advanced regeneration (one of the requisites for RIL) and removal of

a substantial portion of what little canopy exists in the form of large trees (pers. obs.). As

such, the stand dynamics of this particular type of forest might lead to an increased









probability of high grading. Yet many smallholders, because they are limited to their

landholdings, may be forced to harvest in these areas.

Another major drawback of the minimum diameter cut affecting both large and

small-scale operations is that knowledge of tree size at first reproduction (which is

essential for maintaining adequate regeneration levels) is very limited in the neotropics

(Guariguata and Pinard 1998). In the Yucatan forests of Mexico, for example, an original

60 cm dbh diameter limit for Swietenia macrophylla (later reduced to 55 cm) was based

not on ecological constraints, but rather, market demand (Snook 1998). Even with the

new 55 cm limit, it is assumed that it will take 75 years for the next generation of S.

macrophylla to reach that minimum, an assumption that has been met with much

skepticism (Vester and Navarro-Martinez 2005).

Brazilian forest law requires that all harvested timber trees must be > 45 cm

diameter at breast height (dbh) (Schulze 2003). This is considered to be a fairly

conservative lower limit by many forest managers, and some impose their own higher

diameter limits on certain species, based on what they think is the ideal sapwood-

heartwood ratio of the tree (F. Correa da Cunha, pers. comm.). Whereas a few species are

characterized by little sapwood (e.g. Peltogyne spp.) and thus are considered worthwhile

to harvest at diameters as small as 45 cm dbh, many species are not harvested unless they

are much larger. In one community operation in Acre, Brazil, Tabebuia spp. and

Hymenaea spp. must reach 70 cm dbh before they are cut, whereas other taxa, such as

Dipteryx spp., must attain 80 or 90 cm dbh before they can be considered for harvest.

This management strategy is also employed in operations that manage vast forest

concessions, and in either case, such a minimum diameter limit could remove many of









the largest trees on the site. Schulze (2003) suggests that by increasing the minimum

diameter cut to 60 cm dbh, one runs the risk of removing the most vigorous individuals

on site, not necessarily the oldest. In this scenario, it would be best to employ a

proportional cutting across adult size classes rather than concentrate efforts on a

minimum harvestable size, thereby ensuring the maintenance of fast-growing individuals

for the next cutting cycle.

Low Harvest Intensity

Low harvest intensities (few trees harvested per hectare) are common in

smallholder systems, and indeed, low intensity selective harvesting is attractive to those

conservationists promoting SFM (Whitman et al. 1997). In small community

management units in Acre, Brazil, the desire to maintain NTFP resources coupled with

local investment in timber certification initiatives have encouraged communities to try

and create as little damage as possible to the residual forest during logging operations

(pers. obs.). This aspiration is also apparently reflected in Oaxaca and Quintana Roo,

Mexico, where community forest managers generally log below the authorized volume

(Bray et al. 2003).

However, low harvest intensities associated with these smallholder timber

operations may actually aggravate logging disturbance. Several studies have documented

that as intensity of harvest increased, fewer skid trails and logging roads were required,

resulting in a declining per-tree damage rate (Gullison and Hardner 1993, Whitman et al.

1997, Panfil and Gullison 1998). Panfil and Gullison, in their 1998 study of short term

impacts of experimental timber harvest intensities in the Chimanes Forest of Bolivia,

cautiously argue for an intensive (albeit at a fixed volume), localized harvest rather than

an extensive extraction over a larger area. They suggest that the primary benefit of doing









so is the reduction in mortality of smaller size classes due to skid trail construction and

use. The tradeoffs involved with this scenario include the following: first, by logging one

area of a large management unit more intensely, there is an increased chance of reducing

local population levels. Second, even if the area damaged due to skid trail and road

construction is decreased, there will be more concentrated disturbance in the treefall area

in the form of large gaps, which are more susceptible to small-scale disturbances, such as

windthrow and fire (Webb 1997). Third, the threshold values for the release of pioneer

and vine species after such harvests is unknown for many forests, necessitating

monitoring plots for this potentially site-specific information (Panfil and Gullison 1998).

Fourth, the impacts of logging damage should ideally consider biodiversity maintenance

in the event of a more intensive and localized extraction (Panfil and Gullison 1998).

Tradeoffs associated with low harvest intensities will necessarily vary, depending

on site and operational constraints. One of the primary factors that should be taken into

consideration is the location and abundance of commercial species in a given logging

block and to compare these values with other management units (if they exist) in the

project. If considerable skid trail construction is required to reach one individual of a

desired species, the destruction and cost necessary to harvest it may outweigh the benefit

of acquiring the log. In this case, it may be possible to take another individual of the same

species in a management unit better suited to its harvest (i.e. greater abundance of the

targeted species or closer proximity), if there are multiple landowners harvesting

simultaneously in one year as a community effort. It should not be assumed that just

because an operation is extracting at a low intensity that it is having minimal impact on









the forest. It is possible to have few commercial-sized trees injured during logging while

simultaneously incurring substantial ground and smaller stem damage (see Chapter 3).

Skid Trail Planning

One way to minimize extraction damage is to reuse old skid trails, but this method can

prove challenging for smallholders, depending on the management system being

implemented. For example, in the "Multiple-Use Community Forest Management

Project" in the extractive settlement of Porto Dias in Acre, Brazil, harvests are conducted

in 10 ha (200 m x 500 m) logging blocks within individual landholdings (approximately

300 ha), with at least five landholdings participating in a given year. The majority of the

first logging blocks (from 2000) in the participating landholdings were located near

existing roads, which greatly facilitated the removal of stems from the logged areas and

the associated construction of log decks. Logging blocks for the subsequent years were

then positioned next to these initial logged sites so that old skid trails from previous years

were reused. This is beneficial to the forest managers in that they do not need to construct

an extra length of primary skid trail (although it does require them to open up these trails

again every year), and it is favorable to the forest because the amount of superfluous skid

trails is minimized. Using this method, though, means that future tree regeneration will

never become permanently established along the length of these reused skid trails. It also

requires the constant perturbation of the previous logging blocks.

In comparison, the forest management project of Sdo Luis de Remanso (another

extractive settlement project in the state of Acre) utilizes 50 ha logging blocks, slightly

reducing the need to reenter the management area in sequential years (depending on the

number of members that remain in the project and the yearly execution of management

plans). Instead of several different landholdings simultaneously participating in the









harvesting (as is the case in Porto Dias), only one landholding (in general), or 50 ha, is

logged per year. In comparison to the Porto Dias system, this setup may be more

economically viable, as the forest managers usually harvest in just one landholding per

year, so ground efforts are not dispersed. Conversely, in Porto Dias, if five landholdings

are logged within a given year, the logging crew is compelled to maintain roads and

construct/reopen skid trails in several sites (N. Marcondes, pers. comm.), adding extra

cost and effort to the operation.

Rotation Cycles

One prerequisite for RIL is the implementation of growth and yield data for

different management regimes to guide the choice of rotation cycle (Silva et al. 1995). In

the Brazilian Amazon, the rotation cycle generally ranges from 30-35 years, based on an

average diameter growth rate of 0.5 cm year-1 (Silva et al. 1995). Silva et al. (1995),

however, caution that ongoing measurement and analysis of permanent plots over the

next 30 years are needed before forest managers embrace an ironclad management

prescription. They point to the variable diameter increments in any given class of trees,

based in part on environmental conditions, growth habits, and genotypes.

Another drawback of the Brazilian harvest rotation is the application of silvicultural

recommendations generated in one region to another. By 1989, approximately 105 ha of

permanent plots had been established in the Brazilian Amazon, but most of them are

located in the state of Para (Silva et al. 1995). Meyer and Helfman (1993) point to the

need for sustainable resource management plans based on an understanding of specific

local conditions. Many of the Brazilian community timber operations are emerging in the

western state of Acre (see Kainer et al. 2003), yet these management systems tend to base

their diameter and volume increment projections on eastern Amazonian data. One









exception is the permanent plot study that has been established in the Pedro Peixoto

Settlement Project by the Brazilian Agricultural Research Corporation (EMBRAPA),

which has based the smallholder management system on locally-generated growth and

yield data. Drawing from the results generated by the project, Oliveira et al. (1998)

propose that extraction of fewer trees per hectare at shorter intervals, in concert with

silvicultural treatments, will maintain pre-intervention forest structure and biodiversity

levels by creating a gap mosaic of different ages. In other words, a short rotation cycle (in

this case, ten years whereby four hectares are harvested annually on a 40-ha stand) will

substitute for the larger impact caused by more intensive interventions.

Yet, even if site-specific growth and yield models state otherwise, a short rotation

could severely degrade the quality of the resource base. Webb (1997) warns that although

increased light levels due to removal of harvested individuals will generally promote the

rapid growth of residual trees, repeated harvests over short rotation periods will

potentially result in low levels of hardwood regeneration as well as a highly

discontinuous canopy. Oliveira (2000), though, counters that damage levels from short

interval cycles can be mitigated to a certain extent by reusing old logging roads and skid

trails. As well, he makes the observation that the Brazilian SFM system is difficult to

apply to the smallholder setting given that the required rotation cycles (20-30 years)

make the clearing of forest for agriculture and pasture a more enticing option since the

profits from those activities are much more immediate.

Liana Cutting

Proponents of RIL regard the pre-harvest cutting of lianas on crop trees essential if

canopy and future crop tree damage are to be minimized and lianas are abundant (Fox

1968, Appanah and Putz 1984, Vidal et al. 1997, Gerwing and Vidal 2002, Pereira Jr. et









al. 2002). On average, lianas connect each canopy tree to anywhere from three to nine

others in Amazonian terra firme forest (Vidal et al. 1997). During tree felling, this liana

connectivity inadvertently can bring down several other trees, creating a far larger gap

than would occur if lianas associated with harvest trees were pre-cut (Appanah and Putz

1984, Vidal et al. 1997). This suggests that directional felling (a basic tenet of most RIL

guidelines whereby harvest trees are felled in a specified direction to minimize residual

stand damage), is less effective when interconnected vines are not eliminated (Fox 1968).

Furthermore, as these uncut lianas fall with the crop tree, they typically resprout

vigorously, thus competing with new tree recruits and reducing establishment success in

gaps (Schnitzer et al. 2000, Gerwing 2001).

Yet, there is little doubt that lianas are essential components of tropical forests,

providing an important food source for wildlife species as well as supplying a canopy

pathway for some arboreal fauna (Putz et al. 2001). Lianas also contribute to canopy

closure after treefall and stabilize the microclimate of the subcanopy layer (Schnitzer and

Bongers 2002). When asked about the disadvantages of cutting lianas, local forest

managers in Acre, Brazil articulated concern for game species because many liana fruits

are available when tree fruits are not (pers. obs.), an observation that is corroborated by

several scientific studies (Heideman 1989, Morellato and Leitao-Filho 1996). These same

forest managers have also observed that lianas constitute pathways between individual

tree crowns for many animals (especially primates), something that has also been

substantiated in the literature (Schwarzkopf and Rylands 1989, Bobadilla and Ferrari

2000). Consequentially, local logging crews from some Brazilian community forests have

expressed reluctance at cutting lianas prior to the felling cycle (pers. obs.).









Implementing this particular RIL recommendation might present a few quandaries

for the community or smallholder forest manager who is looking to maintain more than

just the next cycle's timber yield, a consideration that may not manifest itself with an

industrial commercial logging operation. Yet, not cutting lianas presents an obvious

tradeoff, as they negatively influence growth, leaf production and sexual reproduction of

host trees (Parren 2003). This is an important consideration, particularly in a landholding

that is limited in size, and, consequently, in future crop tree abundance. Perhaps more

importantly, though, is the shear risk that the logging crew takes when felling a tree that

has not been freed of climbing vines (Amaral et al. 1998), a factor that is valid at both the

smallholder and industrial scales. Parren and Bongers (2001) propose that the benefits of

cutting or not cutting lianas prior to logging is heavily site dependent, so it may be that

local forest managers must experiment with this facet of the RIL paradigm before making

a final decision as to pre-harvest silvicultural prescriptions.

Conclusions

In this review, I have attempted to highlight some of the ecological constraints of

RIL as practiced at the community and smallholder scales, mostly drawing upon

examples from the Brazilian Amazon region. Many of the limitations outlined here can

also be applied to the industrial commercial setting, such as the importance of a pre-

harvest botanical inventory. However, a key distinction with smallholders is that these

forest managers often extract timber from undersized management units, necessarily

reducing the size and often the density of the populations from which they are harvesting.

Ghazoul et al. (1998) point out that SFM (and presumably RIL) requires the maintenance

of viable tree populations, but admit that minimum effective population sizes and









densities are rarely known. Logging, even under the RIL model, can reduce population

viability when conspecifics are isolated from each other (Ghazoul et al. 1998).

Although RIL is, in general, ecologically beneficial to the residual stand, there is

evidence that the magnitude of benefits varies, depending on which forest system is being

exploited (Fredericksen and Putz 2003, Schulze 2003). Some logging guidelines outlined

in this review are unlikely to change in the near future, simply because they are part of

Brazilian federal law, such as the requisite rotation cycle and the minimum diameter cut.

Nor perhaps should they, given the alternative options, such as reentering previously

harvested stands after a short time or taking undersized trees. Given that many of these

small scale timber operations must continue to function within the confines of their

realities, I see the following points as potential recommendations for smallholder timber

management following RIL guidelines.

Logging Block Size and Implications for Pre-harvest Inventories

A diversity of smallholder systems has been described in this review, illuminating the

need to consider details of each one (such as logging block size) when planning timber

harvests. In the state of Acre alone, several logging block sizes are used, from the four-ha

harvest compartment in Pedro Peixoto, to the 10- and 50-ha systems of Porto Dias and

Sdo Luis de Remanso, respectively. As logging operations are currently executed, I

believe that the 50-ha system is best suited to maintain timber species population viability

simply because managers implement a pre-harvest inventory in a larger unit, providing a

more accurate assessment of species distributions than with a 10-ha block (see Table 2-

1). Also at this scale, logging crews will not need to reenter the 50-ha area as frequently

as is required in the Porto Dias system, in which skidders often return annually to

previously harvested logging blocks, reopening primary skid trails. Also, by harvesting









just one 50-ha unit per year, the logging crew is arguably exerting less effort in person

hours and equipment usage, as compared to harvesting several non-contiguous 10-ha

blocks in a given year.

Even so, political, cultural and socio-economic reasons exist for choosing one system

over another. As such, although harvesting a 10-ha block in one year may not be the most

ecologically viable option, it is what certain communities have to deal with, along with

all of the associated environmental disadvantages. The 10-ha logging block system is also

likely to grow in usage, as it is reportedly being favored by new communities wishing to

manage for certified timber in Acre (N. Marcondes, pers. comm.). This reality suggests

the need for full scale inventories of the entire landholding projected for timber

management to minimize potential dangers of inbreeding and inadequate seed dispersal.

The measurement of genetic variation is impractical for forest managers, but the potential

for genetic drift can be reduced by basic knowledge of species populations generated by a

landscape scale inventory (Jennings et al. 2001). Failing implementation of a 100%

survey for those trees above at least the minimum diameter cut, rapid assessment

inventories should be considered.

Botanical identification is a critical component of logging inventories, for which local

parataxonomists are key partners. Before inventories take place, however, common

names should be agreed upon, so that the same nomenclature is applied throughout the

survey process. To ensure that these common names then correspond with their correct

scientific counterparts, care must be taken to have samples identified with known

herbarium examples, if possible. At the very least, care should be taken to not solely rely

on databases generated by other organizations, whose own tree identifiers may also be









using common names that differ from those being used by a local parataxonomist from

the timber project.

Monitoring

Recognizing that smallholder and community forests must establish their own

management strategies while building on RIL guidelines, I suggest that systematic post-

harvesting monitoring become integrated into logging operations, so that the benefits of

recommended practices (such as pre-harvest liana cutting) may be tested at the

appropriate scale. I recognize that smallholders in the rural tropics have limited financial

and technical resources, and that monitoring may be unfeasible for them, unless they

perceive that proposed activities add to their livelihood productivity. Accordingly,

Ghazoul (2001) concedes that given associated high costs, an adaptive management

approach, entailing collaboration between biologists and local forest managers, would

allow for sustainable management to be based on the technical skills, ecological

knowledge base, equipment, and funds available to the managers. This collaborative

approach to forestry requires considerable communication between local managers and

professional forestry technicians and an assessment of existing local ecological

knowledge and traditional management systems (Carter 1996).

To this end, collaboration between smallholders, the agencies that support them and

researchers who may be able to contribute technical expertise to the process should be

fully encouraged. I recognize that working outside of traditional field sites in community

settings obliges researchers and practitioners to approach forest management and

monitoring differently, requiring them to consider social, political, and economic factors

that affect the way that local peoples interact with the natural resource base. Beyond this

consideration, we need to also think about the dynamics of the forest managed for a suite









of products and services at the scale discussed in this review, providing communities and

smallholders with valuable feedback on their management systems and further providing

a link between biodiversity conservation and SFM.

Moving beyond RIL

It is indisputable that RIL methods are a vast improvement over the much more

destructive and haphazard conventional selective logging practices that are,

unfortunately, still the norm (Schulze 2003). Yet, forest management under RIL

guidelines alone could be at odds with the SFM model, which demands forest

management for a suite of goods and services as well as the maintenance of ecological

functions (Fredericksen and Putz 2003). Arguably, it is this broader suite of outputs that

many communities target in their forest management schemes. Rural communities

depend on wildlife, NTFPs, watersheds, and soil properties for their livelihoods,

underscoring the importance of looking beyond sustained timber production as the most

important forest management objective. Even if seed sources and dispersal mechanisms

remain intact, thereby (in theory) ensuring the viability of future timber regeneration

(Baraloto and Forget 2004), the long term logging impacts on these other livelihood and

conservation elements must not be overlooked. It seems particularly appropriate then to

move beyond an emphasis on RIL guidelines for smallholders engaged in timber

management, while still incorporating its basic tenets into practical application.

It is highly probable that the so-called sustainability of intensive timber

management may not be known until at least three cycles of harvest have been completed

(Poore et al. 1989). This observation bolsters Simberloff s (1999) suggestion that any

proposed silvicultural system designed to maintain biodiversity while producing timber

should be treated as a hypothesis, due to limited empirical evidence supporting its






41


viability. Until then, though, we must continue to focus on forest management research

that is not just rooted in tree reproduction biology (Guariguata and Pinard 1998), but also

consider the application of such investigations at the smallholder and community levels,

necessarily incorporating the local ecological knowledge that is often overlooked by

academic researchers yet when it is acknowledged. In this way, we might be able to avoid

the pitfalls of relying on RIL as proxy for SFM, especially in terms of the smallholder

and community settings, which necessarily require considerations beyond the biological

context.












Table 2-1. Contrasting tree population parameters from inventories at two different spatial scales commonly used by smallholders.
Shown are the results of a 50 ha inventory of commercial species stems > 35 cm DBH from PAE Sdo Luis de Remanso,
Acre, Brazil, along with the range of values (5 and 95%) obtained by 500 simulations of randomly-placed 10 ha plots
within the larger area. Species were included only if they were represented by at least 10 trees in the plot, and their
botanical identification had been verified. Bold cases indicate that the distribution of values from the 10 ha simulations
exceeded the actual value at 50 ha, suggesting that management decisions would change depending on the scale at which
inventories are conducted.
Common Harvestable Density'
Species Came Total Density (ind ha ) (He D sity
Name (ind ha )
50 ha 10 ha 50 ha 10 ha
Aspidosperma vargasii Amareldo 0.92 0.5- 1.3 0.36 0.1 -0.5
Myroxylon balsamum Balsamo 0.31 0.2 0.6 0.15 0.1 0.4
Breu-
Tetragastris panamensis vermelho 4.62 2.6 8.6 0.82 0.2 2.2
Cedro-melho
Cedrela odorata er 0.87 0.3- 1.7 0.67 0.3- 1.3
vermelho
Amburana cearensis Cerejeira 0.56 0.1 0.9 0.41 0.1 0.9
Cumaru-
Apuleia leiocarpa Cumaru 0.97 0.7- 1.7 0.79 0.5- 1.3
cetim
Barnabydendron riedelii Guaribeiro 0.82 0.3 1.3 0.67 0.2 1.0
Tabebuia serratifolia Ip6-amarelo 0.23 0.1- 0.4 0.08 0 0.2
Hymenaea parvifolia Jutai 1.21 0.7 2.1 0.64 0.2 1.2
1 refers to minimum size harvested, which varies among species from 45 60 cm DBH.









Table 2-2. An example of spatial aggregation among neotropical tree species harvested
by communities in Acre, Brazil. Shown are the minimum distance, if any, for
each species at which individuals (stems > 35 cm DBH) are significantly
clustered (Ripley's L-statistic exceeding 95% confidence interval, based on
1000 bootstrapped values at 5 m distance intervals using Besag's edge
correction conducted in Ripley software package; Marcon 2005). Data are
from a 50 ha inventory of commercial species from PAE Sdo Luis de
Remanso, Acre (Centro dos Trabalhadores da Amazonia, unpublished data).
Species were included only if they were represented by at least 10 trees in the
plot, and if their botanical identification had been verified.
Species Common Name Spatial Distribution
Clumped? Distance
Aspidosperma vargasii Amareldo No
Myroxylon balsamum Balsamo No
Tetragastris panamensis Breu-vermelho Yes 6 m
Cedrela odorata Cedro-vermelho Yes 8 m
Amburana cearensis Cerejeira No
Apuleia leiocarpa Cumaru-cetim No
Barnabydendron riedelii Guaribeiro Yes 15 m
Tabebuia serratifolia Ip6-amarelo No
Hymenaea parvifolia Jutai No














CHAPTER 3
EVALUATING LOGGING DAMAGE TO FUTURE CROP TREES IN ACRE,
BRAZIL: IMPLICATIONS FOR MANAGEMENT OF A CERTIFIED COMMUNITY
FOREST

Introduction

The guiding principle of sustainability dictates that a commercial forest should be

managed in a way that limits as much undesirable damage as possible to the ecosystem.

This is essential for not only ecological reasons, but economic ones as well (Miller

1998). For example, it is well known that unplanned (conventional) logging results in

unnecessary damage to the residual forest and reduces the chances for future timber

production (Pinard et al. 1995). Johnson and Cabarle (1993) estimate that anywhere from

26 to 75% of a given tropical stand of future crop trees (FCTs) are injured during

conventional logging operations. Damage to residual stands is often correlated with

logging intensity, which varies greatly, depending on spatial and temporal factors, access,

equipment, terrain, climate, supplemental treatments, and market acceptance of lesser-

known species (Metzger and Schultz 1984, Sist et al. 1998, Putz et al. 2000).

The most common harvesting method in the tropics is selective logging, whereby

only scattered trees of sufficient economic value are extracted from the forest. In

unplanned operations, selective logging has been observed to cause extensive canopy

cover removal (Johns et al. 1996, Webb 1997, Whitman et al. 1997, Pereira et al. 2002),

soil compaction (Whitman et al. 1997), long-term changes in tree species composition

(Thiollay 1992), and reduced faunal diversity (Johns 1991, Thiollay 1992). Accordingly,









harvesting techniques and intensities need to be controlled to reduce residual stand

damage (Webb 1997).

Controlled selective logging, or reduced-impact logging (RIL), utilizes methods

that minimize damage to the residual stand, such as pre-harvest inventory, mapping,

directional felling, vine cutting, and planning of skid trails, log decks, and roads (Uhl et

al. 1997), potentially avoiding some of the post-logging damage noted by other authors

(Pinard and Putz 1996, Johns et al. 1996). RIL techniques have been shown to reduce

damage to residual stands in mixed dipterocarp forests in Southeast Asia by as much as

30-50% (Pinard and Putz 1996, Bertault and Sist 1997, Sist et al. 1998). Furthermore,

RIL methods have demonstrated an increased efficiency of logging operations and a

reduction in the amount of timber wasted during a typical harvest (Barreto et al. 1998,

Holmes et al. 2002).

Even with implementation of RIL methods, sustainability of harvesting will

ultimately depend upon advanced regeneration to replace the larger size classes during

the next cutting cycle (Putz et al. 2000). This assumption is challenged by the damage

incurred by the residual stand during logging operations. For example, Jackson et al.

(2002) report that RIL operations in a FSC-certified timber concession in Bolivia killed

or severely damaged an average of 22 trees (> 10 cm dbh) for every one extracted. In the

same forest, Krueger (2004) found that flagging reduced damage to FCTs by 20% in

felling gaps and by 10% along skid trails receiving 2-10 skidder passes, demonstrating

that marking, or flagging, of FCTs can reduce damage to the residual stand by improving

visibility during on-the-ground operations (Dykstra and Heinrich 1996). Currently,

general Brazilian forest management practices do not employ this method.









Mechanisms underlying these residual stand impacts need to be understood in the

context of specific local conditions (Sheil and Van Heist 2000). Commercial species

distribution and human exploitation varies greatly between neotropical regions,

supporting the argument that silvicultural treatments appropriate in one location will not

necessarily be applicable in another (Putz 1996). For example, Webb (1997) points out

that difficulties encountered in Carapa nicaraguensis forest swamp in Costa Rica

required management techniques not employed in other logging operations. Given the

need to develop and evaluate site-specific management strategies that will mitigate the

negative effects of timber harvesting, it is essential to look at several variables across

locales.

One forest type on which little has been written about in the management literature

is the bamboo-dominated forest of southwestern Amazonia, which covers almost 180,000

km2 (Griscom and Ashton 2003). This forest is characterized by a mix of structurally

heterogeneous stands without bamboo and patches of trees scattered within a canopy

dominated by dense stands of Guadua species (with some species of Guadua reaching

heights of 30 m) (Griscom 2003). Similar to other mast seeding bamboo genera (Janzen

1976), Guadua spp. undergo a single synchronized reproduction event followed by

synchronized mortality, at 25-30 year intervals (Nelson et al. 2001). At least two species,

G. sarcocarpa and G. weberbaueri have been linked with reduced forest basal area as

well as tree species richness (Silveira 2001, Griscom 2003). This phenomena is strongly

linked to the species' ecology: branches of G. sarcocarpa and G. weberbaueri culms are

endowed with curved barbs which act as grappling hooks (see Fig. 3-1), enabling the

bamboo to be partially or fully supported by trees for vertical growth (Griscom 2003),









weighing them down, and increasing the likelihood of blow-downs (Griscom and Ashton

2003). In concert with the common tree blow-downs in this region, Griscom and Ashton

(2003) propose that low recruitment of trees into the larger size classes explain the

persistence of Guadua domination, rather than the catastrophic disturbances of fire and

agricultural intervention cited by other authors. If Guadua achieves dominance in

managed forests, commercial trees may become competitively excluded and a substantial

proportion of the remaining stems may become deformed, due to mass loading (Griscom

2003). The low visibility in some areas of the Guadua dominated forest also presents an

impediment to sawyers trained in directional felling (F. Correa da Cunha, pers. comm.).

Combined, these factors present some unique challenges for silvicultural management

and timber extraction in Guadua-dominated forest.

This study evaluated logging damage to FCTs in a certified community forest in

Acre, Brazil. The main objectives of this investigation were (1) to determine the extent of

area disturbed (canopy gaps and skid trails) during a small-scale certified logging

operations, (2) to determine a difference in damage incidence to FCTs between forest

with and without bamboo (Guadua), (3) to determine if likelihood of FCT damage is

increased when lianas are present, (4) to determine to what extent harvesting can be

conducted more intensely (m3ha-) without incurring greater residual stand damage in

general and damage to FCTs > 20 cm dbh in particular, and (5) to determine to what

extent marking of FCTs diminishes damage and/or mortality of FCTs (>20 cm dbh)

during logging.

Site Description

The Porto Dias Extractive Settlement Project (PAE) (S 1000'39, 9", W

66 46'26,4") is situated in the northeastern corer of the Brazilian state of Acre, in the









municipality of Acreldndia. It is bordered by various rubber estates, cattle ranches, and

colonization projects (see Fig. 3-2), and is separated from Bolivia by the Abuna River

(Stone 2003). The site is defined by red-yellow latisols of low fertility and relatively flat

topography, with an annual precipitation of 1890 mm yr1, most of which falls between

November and March. As of 2001, approximately 97% of the reserve was still forested

(CTA, unpublished data), but it is one of the last large contiguous pieces of forest in the

municipality (N. Marcondes, pers. comm.). As such, it is under constant pressure from

the influences of agricultural conversion within the PAE as well as illegal clearing on its

borders. Older residents recall Swietenia macrophylla (mahogany) being logged in the

region and transported via the Abuna River (Stone 2003), but mahogany is currently

extremely rare in the PAE. Three tropical humid terrafirme forest types are generally

recognized in the reserve: open canopy forest; open canopy forest mixed with arborescent

bamboo (Guadua spp.); and closed canopy forest.

Three species of Guadua are found in the study site: G. weberbaueri, G.

sarcocarpa, and one as-yet unnamed species, which occurs in riverine forest along the

Abuna River (M. Silveira, pers. comm.). At the end of 2004, when logging operations

commenced, G. sarcocarpa underwent a monocarpic dieoff that affected much of the

study area. Both G. weberbaueri, and G. sarcocarpa are thought to be synchronized on a

30 year masting cycle across large areas (100-100,000 km2) (Nelson et al. 2001). The two

species exhibit similar autecological characteristics, with an apparent lack of seed

dormancy, germination occurring in both gaps and non-gaps, and partial shade tolerance

among seedlings (Griscom 2003). In at least one study in Madre de Dios, Peru, G.

weberbaueri demonstrated a lower mean stem diameter (cm) in comparison with G.









sarcocarpa (4.04 vs. 6.67, respectively), but exhibited higher stem density (stems/100

m2) (34.2 vs. 23.8) (Griscom 2003).

The reserve encompasses approximately 22,145 ha, and is divided among

coloca5oes, or individual family landholdings constructed around pre-existing rubber

trails (approximately 300 ha each). There are around 100 registered families who live in

the PAE, in addition to an estimated 20 families who are squatting illegally in the reserve

(Santos 2000). Traditional family livelihoods in PAE Porto Dias include forest extraction,

hunting, subsistence agriculture, and small-scale animal husbandry (Stone 2003). Cash

was customarily earned from rubber and Brazil nut harvests, but in recent years rubber

tappers have been turning to logging as an alternative income source. The Center for

Amazonian Workers (CTA), a non-governmental organization based in Acre, has been

instrumental in developing the management framework of the Multiple-Use Community

Forest Management Project. The project was initiated in 1995, but the first timber harvest

was not harvested until 2000. SmartWood certified the logging operation using Forest

Stewardship Council (FSC) criteria in 2002, distinguishing it as the second FSC-certified

community timber management project in Brazil. FSC certification requires that the

project implement a special monitoring program in the bamboo-dominated forest, as there

is concern that logging damage may be greater there than in forest lacking bamboo (N.

Marcondes, pers. comm.). Additionally, Guadua-dominated stands have been identified

as important habitat for many valuable wildlife species, making them focal areas of

interest for conservation efforts (Griscom 2003).

Forest Management and Harvesting Operations

Landholdings (approximately 300 ha each) of reserve residents serve as the timber

management areas, and therefore do not favor the compartmentalization typical in









Brazilian industrial timber operations (Braz and Oliveira 1996). In compliance with

federal regulation, 10% of the landholdings may be cleared for agricultural activities and

5% must be set aside for preservation (no harvesting of trees). Thus, most residents have

at least 255 ha available for timber management, result in a 25-year logging rotation (255

ha/10 ha yr-1= 25.5 years). The original ten members of the project all harvested 10 ha per

year, but within a few years after the first timber harvest, it was decided that only five

landholdings per year would be harvested. In the latter case, project members began

harvesting every two years in their respective landholdings. However, given that some

families have since dropped out of the timber project, other members have still harvested

in sequential years (i.e. more than one or two years in a row), further confounding the

question of the cycle length.

Inventories and mapping of all tree species >35 cm dbh are carried out one year

prior to harvesting in 10 ha blocks within each landholding to be logged; as of 2005,

though, a 100% inventory of the entire property had not been executed in any of the

participating landholdings. After selecting the crop trees, the project must submit its

logging plan to the Brazilian Institute of the Environment and Renewable Resources

(IBAMA), which will then approve or disapprove the plan, based on an acceptable basal

area removed, desired residual stand species distribution and crop tree proximity to water

sources. Only trees > 45 cm dbh may be legally harvested. Logging is generally carried

out in the driest part of the year, from July to November. Twelve major timber species are

presently exploited (see Table 3-1), with a low intensity of cutting (1-3 trees ha-1 or

approximately 10 m3 ha-1). Since the timber project members began logging biannually,

the total annual harvest for the project is around 500 m3, or 100 m3 landholding-1. Sawn









board and artisan products from the community sawmill are primarily sold to the

domestic certified market in Southern Brazil. The project does not market logs.

Members of the timber project (as of August 2005 there were seven participating

families) conduct most of the logging activities. These techniques include inventories,

selection of crop trees, assessment of defects in crop trees, directional felling, and skid

trail planning, RIL skills for which they have been extensively trained over the last ten

years by the Technological Foundation of Acre (FUNTAC) and the Tropical Forest

Foundation (FFT) in Para, and for which they now train other forest managers. The only

activity that they do not directly participate in is skidding logs to the sawmill, a process

contracted to heavy equipment owners and operators from outside the community.

During logging operations, the landowner typically assists the construction of skid trails

by flagging the path that the skidder or tractor operator will use.

Methods

Future Crop Tree (FCT) Selection

Fifty-two tree species were selected for the study, based on their potential

importance as commercial timber sources (see Table 3-1; Ribeiro et al. 1999, Lorenzi

2000). Due to external market demands, however, only about 10-12 species are exploited

on a regular basis in Porto Dias, most commonly Amburana cearensis, Cedrela spp.,

Peltogyne spp., Dipteryx spp., Apuleia leiocarpa, Tabebuia spp., Aspidosperma vargasii,

and Hymenaea intermedia. For purposes of this study, the definition of FCTs was

broadened to account for all species of commercial value, both current and future (see

Table 3-1). When proper field identification was questionable, species verifications were

conducted at the Federal University of Acre herbarium in Rio Branco, Acre, Brazil, but

some trees were only identified to genus (see Table 3-1).









Experimental Design and Pre-logging Inventory

The greatest amount of damage to FCTs during logging occurs in areas closest to

felled crop trees and skid trails (Johns et al. 1996). Therefore, observations of logging

impacts were concentrated in a 50 m radius (hereafter referred to as "zone of impact")

around each designated harvest tree in each of the 10 ha (200 x 500 m) logging blocks in

four landholdings scheduled to be harvested in 2004-2005 (Barrinha 1, Barrinha 3,

Palestina, and Sao Pedro). This number was eventually reduced to three, as the rainy

season began before Sao Pedro could be logged. A full inventory of all commercial FCTs

>20 cm dbh was conducted in the zones of impact. Sample sizes in the logging blocks

(i.e. the number of zones of impact) depended on the abundance of crop trees, with a total

of 18 zones of impact and 161 FCTs in Barrinha I, 20 zones of impact and 216 FCTs in

Barrinha III, and 12 zones of impact and 109 FCTs in Palestina, including some overlap

between zones. In 2004, before logging operations commenced, the following data were

noted for FCTs: species, location (x, y coordinates), dbh (1.3 m or above the buttresses);

total estimated height; qualitative trunk and crown quality ("good", "tolerable", or

"inferior"); crown position ("dominant", "intermediate", "suppressed"); presence in

bamboo-dominated stand ("yes"/"no"), and presence of at least one liana to the canopy

("yes"/"no").

Area Disturbed

All primary and secondary skid trails and treefall gaps were mapped using a

compass and meter tape. The size of treefall gaps was estimated using the "center-point"

system (modified from Runkle 1992). From a point in the approximate center of each

treefall gap, compass angles in eight different directions were measured to the edge of

intact canopy (Johns et al. 1996). This information was then entered into ArcGIS 9.0









(ESRI, Redlands, CA), to calculate area of the resulting polygon. The ground area

transformed by the total length of skid trails was determined using modified methods

from Johns et al. (1996), with skid trail width being measured at the beginning and end of

each straight segment of trail. Trail width measurements were also taken at each bend to

account for potential bulldozer maneuvering. This information was then combined with

total skid trail length (estimated using pre-harvest inventory maps) in ArcGIS 9.0.

FCT Location in Bamboo (Guadua)-dominated Forest

Zones of impact (both in areas of high and low harvest intensity) naturally fell in

locations where bamboo was present and in areas where it was absent. Location in

Guadua-dominated stands was defined as at least one culm of Guadua spp. within 2 m of

the FCT trunk.

Liana Presence

To test the effect of presence of lianas on FCT damage probability, liana

occurrence on a FCT was denoted as at least one vine connecting to the canopy.

Harvest Intensity

To detect the impact that a range of timber harvest intensities could have on FCTs

in Porto Dias, zones of impact were established in locations scheduled to receive a locally

high intensity of cutting (>3 trees ha-1), and also in areas of locally low intensity cutting

(1-2 trees ha-1). This design allowed for establishing a harvest intensity gradient in each

logging block, but did not necessarily allocate the treatments evenly across the logging

blocks, as the abundance and distribution of crop trees varied greatly between

landholdings.









Marking Treatment

To test the effect of marking, about half of all inventoried FCTs in each of the four

logging blocks were marked. All FCTs within the zone of impact of a tree to be harvested

were either marked or left unmarked, stratifying the treatment between regions of low

and high crop tree density (see Figs. 3-3-5). Marking entailed a single band of orange

paint 20 cm wide at approximately 1.7 m above the ground (adapted from Krueger 2004).

Initially, the random assignment of the marking treatment was considered when

designing this experiment. However, it was decided that the highly visible presence of

even one marked FCT in the vicinity of unmarked FCTs could induce the logging crew to

alter their activities in some way as to favor neighboring FCTs. This potential source of

bias suggests that statistical independence of the FCT treatments would have been

difficult to achieve even with this type of experimental design.

Assessment of FCT Damage

Inventory data were used to construct a spatial database in ArcGIS 9.0. For each

FCT, distance to nearest felled tree, nearest gap edge and nearest skid trail were

measured. Harvest intensity (trees ha1) was estimated for each FCT as the volume

exploited in the surrounding 1 ha (circumscribed by a radius of 56.4 m). Local FCT

density was estimated as all FCTs > 20 cm dbh within a 25 m radius of a given FCT. This

measurement was included to control for variation explained by local density when

testing for other factors such as marking. Seven months after logging activities were

completed, in June and July, 2005, all previously inventoried FCTs were revisited and

classified according to levels of damage or mortality (see Table 3-2).









Statistical Analyses

Quantifying FCT damage was considered for this study. A logistic regression

analysis was used to test the binary response variable (damaged/not damaged) by treating

each FCT as an independent statistical unit to which a series of independent variables was

assigned: local harvest intensity (continuous); distance to nearest felled crop tree

(continuous); distance to nearest skid trail, (continuous); distance to nearest gap

(continuous); neighborhood FCT density (continuous); presence of bamboo (binary);

marking (binary); and presence of lianas (binary). A regression analysis was performed to

determine the influence of crop tree dbh on felling gap area. Differences were considered

significant at P > 0.05.

Results

Overview of the Logging Operation

The mean tree (> 35 cm dbh) basal area was determined to have the following

values for the three landholdings: 7.7 m2 ha-1 in Barrinha I, 8.3 m2 ha-1 in Barrinha III,

and 8.8 m2 ha-1 in Palestina. All harvesting occurred between October and November

2004, and a bulldozer (Valmet 128) was used to extract logs. Two community-based

sawyers participated in the operations: one felled trees only in Palestina and the other cut

in both Barrinha I and Barrinha III. No new roads were created to facilitate logging

operations, and pre-existing skid trails in logging blocks from the project's previous

harvesting activities were reopened to access new areas. All three logging blocks were

located close to existing roads, so boles were eventually skidded to log decks adjacent to

these routes.

A total of 22.3 ha of 30 ha were sampled in the three logging blocks. Forty five

trees were extracted from all of the landholdings, with 38 actually cut and 7 that fell









under natural circumstances at least one year prior to harvesting, for a total harvested

volume of 237.7 m3, or approximately 7.9 m3 ha Felled tree (n=38) dbh (cm) ranged

from 60-114, with a mean of 71.7 (s.d. 14.73). Average harvested tree density was

approximately equal between the three logging blocks (see Table 3-3), suggesting that

any differences in damage incurred by FCTs was not due to differences in logging

intensity (Johns et al. 1996). Some of the trees originally selected were not cut for various

reasons (usually they were found to be hollow prior to harvesting), so loggers harvested

several substitute trees. As a result, some of the trees designated as experimental FCTs

the previous year were not affected by logging operations, as they were far from any skid

trails or felling gaps (> 60 m from felling gap, cut stump, or skid trail). These trees (a

total of 18) were not considered for the analysis. At the same time, damaged trees that

were near the newly substituted crop tree sites were not tallied after harvesting, because

no pre-logging data were gathered for them.

FCT Damage

A total of 70 FCTs out of 468 were damaged during logging operations, or

approximately 15% of the experimental population. The most common damage types

were slightly damaged crowns, followed by scraped boles and roots, classified as "minor"

damage (see Table 3-3). Tree felling operations damaged a greater proportion of trees

than skid trail construction and use, but the results were not statistically significant.

Marking did not significantly reduce overall FCT damage in the Porto Dias

operations (see Table 3-4). Furthermore, the incidence of FCT damage levels was not

higher in Guadua-dominated forest, nor was there a significant interaction between a

FCT's presence in a Guadua-dominated stand and the marking treatment (interaction

term in Table 3-4). Fifty-six percent of the inventoried FCTs carried lianas, but damage









probability due to liana presence on the FCT was not highly significant (P = 0.10), but

many damaged FCTs had at least one large liana attached to its crown (67%).

The closer a FCT was to a skid trail or logging gap, the greater the likelihood of it

being damaged (see Table 3-4). In contrast, distance to cut stump and local FCT

neighborhood density had no significant effect. Harvest intensity (see Table 3-4) also did

not influence the probability of FCT damage, implying that in this study, the probability

of FCT damage did not increase with the number of trees harvested in the surrounding

hectare.

Area Disturbed

The percentage of total sampled area affected by harvesting was 28.9%, with the

majority in felling gaps (16.8%). Average area per felling gap amongst all three logging

blocks (mean s.d.) was estimated to be 393 m2 181. Regression analysis revealed a

significant positive effect of crop tree dbh in single tree gaps area (see Fig. 3-6). Skidding

damaged 11.8% of the total surface area.

Discussion

Timber harvesting, as conducted by the Multiple-Use Community Forest

Management Project in PAE Porto Dias, had minimal impacts on residual stands, at least

on commercial individuals > 20 cm dbh. Additionally, potential damage was avoided by

not having to construct new log landings and log roads. Even so, the high values reported

for area damaged, both skid trails and felling gaps, merit analysis (see Table 3-3). The

values for felling gap area per tree cut were in some cases almost twice that of other

reported neotropical studies (see Johns et al. 1996). In many instances, increased gap

sizes can be explained by the size of the trees that were felled. For example, one of the

most valuable taxa in Porto Dias, Dipteryx spp., comprised some of the largest









individuals amongst harvested trees (up to 114 cm dbh), contributing to the strong

positive correlation between logging gap size (m2) and dbh of crop trees (see Figure 6). In

the case of Palestina, more than 60% of the harvested trees were > 70 cm dbh, explaining

in part why this particular logging block had the greatest mean gap size (433 m2) amongst

the three landholdings. When considering exceptionally large trees for harvest, forest

managers might do well to consider the tradeoffs between creating a very large gap

(thereby potentially increasing fire and windthrow risk) and the substantial profit for the

sake of the wood.

It is also worth noting that only those species valued for their merchantable timber

were inventoried for the experiment. For example, locally important NTFP species such

as Bertholletia excelsa and Hevea brasiliensis were not tallied during the original

inventories, but during the post-logging survey, damage to several trees of these species

was observed. For future studies, it would be prudent to address some of the issues of

impact of timber extraction on other livelihood activities, such as NTFP collection,

particularly in such regions as the Brazilian Amazon where many communities are

attempting to integrate a suite of income-generating activities. Commercial logging, with

potentially high financial yields, is unlikely to disappear in community forests.

Nevertheless, interest persists in the harvesting of non-timber products such as Brazil

nuts, rubber, and Copaifera and Carapa oils, species that may be adversely affected by

timber harvests. When considering these factors, can timber harvesting and other

livelihood strategies be made more compatible and if so, what are the tradeoffs (see

Salick et al. 1995, Romero 1999, Menton 2003)?









FCT Damage in Guadua-dominated Forest

Given the low visibility in Guadua-dominated forest, I expected that FCT damage

rates would be higher in these areas, but results proved otherwise. There could be a

number of explanations for this outcome: even though poor visibility in Guadua groves

has been cited as an impediment in the felling of large trees because neighboring FCTs

are not always easily seen, FCT damage might be minimal simply because there are

fewer FCTs to damage, as demonstrated by the low basal area estimates cited by other

authors (Silveira 2001), or, more specifically, by the low representation of smaller stems

in those same estimates. Guadua frequently outcompetes tree seedlings and saplings such

that they fail to recruit into larger size classes (Griscom 2003). Therefore, one could

conclude that there are comparatively more FCTs from the lower diameter classes to

damage in forests without bamboo, or at the very least, there are more trees that operators

must avoid when skidding or felling. It is also uncertain as to how the synchronized

mortality of G. sarcocarpa may have directly or indirectly affected felling operations,

although visibility may have improved in these stands because mature bamboo culms

were drying out and falling to the ground in some areas.

Even though I did not find that selective logging increased the probability of FCT

damage in Guadua-dominated stands, the potential for other types of long-term residual

stand damage warrants future investigations. We observed dramatically increased light

levels in post-logging bamboo-dominated forest, possibly attributable to the loss of the

biggest canopy trees in the block, of which there were few to begin with. There are a few

potential problems associated with this. First, by opening large areas during harvesting,

the chances of repeated small-scale disturbances, such as windthrow and fire, are

increased (Webb 1997). Given that Guadua benefits from intermediate and large canopy









disturbances (Griscom 2003), once a gap is opened, bamboo may proliferate and

competitively exclude the majority of tree stems. One example of this phenomenon is

Whitmore's (1984) observation of the reduction in value of Malayan forests that were

invaded by arborescent bamboo after unplanned logging operations. The high light

intensity characteristic of large gaps might also select for a suite of different species as

compared to small gaps (Denslow 1980), a dilemma if the regeneration of small gap-

preferring species is desired by the forest managers. In general, Guadua-dominated

stands are already characterized by a discontinuous canopy, which may amplify the

effects of openings created by logging activities (i.e. increased light levels). As such, for

future operations, forest managers may want to select crop trees based on the presence of

large neighboring trees, ensuring that some sort of basic canopy structure is kept intact

around the logging gap if smaller openings are desired.

Another concern in this forest type is the potential for high grading, or the type of

harvesting that selects for the largest and most valuable individuals. One of the

underlying assumptions of RIL is that advanced regeneration of valuable commercial

timber species exists uniformly throughout the stand so that future harvests are not

compromised by the present one (Schulze 2003). Yet in Guadua-dominated stands, the

lack of advanced regeneration potentially increases the chances of having small, poorly-

formed trees of lesser commercial value dominate the next harvest. For the operation

described in this paper, FSC certification standards, and indeed, Brazilian federal law,

require that forest managers leave at least 10% of the harvested species' populations as

seed trees. This perhaps ensures harvests for generations to come, but it does not

guarantee a sustainable timber yield for the immediate second cycle. As well, even









though Brazilian federal law imposes a minimum diameter cut of 45 cm dbh, many

species in this operation are not harvested until they are much larger (see Chapter 2),

since taxa such as Dipteryx spp. are characterized by high levels of sapwood and thus

attain commercial value at a greater girth. As a result, there is often an understandable

temptation to take the biggest and the best individuals from a site, although at least one

landowner in Porto Dias admitted hesitation in taking large Dipteryx spp. individuals, as

they are often the most prolific seed sources (F. Correa da Cunha, pers. comm.). The

complexities associated with timber harvesting in Guadua-dominated forest may require

a much different approach to silvicultural techniques than developed in other parts of

Amazonia, and underscore the need for future monitoring of harvesting activities and

residual stand dynamics as well as extreme caution in the removal of the largest trees on

site.

Impact of Lianas on FCT Damage

Lianas connecting the crowns of trees will often cause a domino-like effect during

felling operations (Vidal et al. 1997), although few adjacent trees were completely

brought down by felled trees during this operation. Other neotropical studies (see Fox

1968, Appanah and Putz 1984, Johns et al. 1996, Vidal et al. 1997, Pereira Jr. et al. 2002,

Gerwing and Uhl 2003), have demonstrated increased damage during logging operations

when vines were present. Although I expected vines to increase FCT damage, I did not

observe any statistically significant effects at the ca=0.05 level, though given results in

other studies, it is probable that the magnitude of the outcome is consequential from a

management standpoint (Steidl and Thomas 2001). It was evident from post-logging

observations in the field that parts of crowns were pulled down by bulldozers at they









moved through the forest. Undoubtedly, if lianas associated with FCTs had been cut at

least one year prior to harvesting, some of this damage could have been avoided. Still,

Porto Dias forest managers expressed reluctance to cut lianas prior to the felling cycle

(pers. obs.), citing their importance as a food source for wildlife species. Yet, it has been

shown that lianas negatively influence the growth, leaf production and sexual

reproduction of host trees (Parren 2003), and when not cut prior to harvesting operations,

increase the potential danger for logging crews (Amaral et al. 1998). Parren and Bongers

(2001) propose that the benefits of cutting or not cutting lianas prior to logging is site

dependent, so it may be that local forest managers must experiment with this particular

RIL guideline before making a final decision as to pre-harvest silvicultural prescriptions.

Impact of Harvest Intensity on FCT Damage

Local harvest intensities in the logging operation described here removed 1-4 trees

ha-1, seemingly with no increased levels of residual stand damage in the > 20 cm dbh

class as the maximum ceiling of harvest intensity (4 trees ha-) was reached. This type of

low-intensity selective logging is attractive for conservationists espousing the

implementation of sustainable forest management as a way to limit canopy cover loss and

protect ecosystem function and biodiversity (Whitman et al. 1997). Even so, Gullison and

Hardner (1993) have pointed out that as the number of trees removed increases, fewer

new skid trails are required (although the damaged area per tree due to logging gaps

remains constant). In the case of the operation described here, even though there was no

evidence supporting increased levels of FCT damage as harvest intensity increased, the

average area (m2) damaged per tree extracted due to skid trail construction was quite high

(see Table 3-3) as compared to other neotropical studies, relative to the number of trees

extracted per hectare. For instance, Johns et al. (1996) found that unplanned logging









operations using a bulldozer to skid trees damaged 119 m2 per bole removed in relation to

skid trails (with an average harvest intensity of 5.6 trees ha-1). The high levels of area

damaged per tree extracted in this study suggest that there might be a tradeoff associated

with trying to harvest isolated trees across even a relatively small area of 10 ha, both in

terms of cost and as a function of area and trees damaged.

One way to offset this effect of low intensity harvests would be to increase

localized harvest intensity at a fixed volume. The primary benefits of this method would

be reduction in mortality of smaller size classes as well as fewer skid trails, not

necessarily reduction in mortality or damage to FCTs (Panfil and Gullison 1998).

Another consideration is that large gaps allow for more solar radiation to reach the forest

floor, drying remnant organic matter such as leaves and increasing the amounts of fuel in

post-logging sites (Johns et al. 1996). Furthermore, the release thresholds of pioneer

species, lianas and bamboo in the large gaps resulting from this scenario are unknown

(although conditions might be ideal for the regeneration of such species as S.

macrophylla) (Panfil and Gullison 1998). As such, the implementation of an increased

localized harvest intensity is risky, especially for a landowner limited to a management

unit of 300 ha.

Impact of Marking Treatment on FCT Damage

Despite positive results of marking FCTs in at least one other study (see Krueger

2004), marking did not appear to reduce FCT damage in the Porto Dias timber

operations. That FCT marking had no effect on damage levels along skid trails was

surprising, given that machine operators were from outside the community and therefore

not as familiar with the forest. Even so, the owners of the landholdings were responsible

for flagging the path of the skid trails prior to extraction, so as to avoid any damage









associated with skidding. This precaution could explain why marking had no effect on

FCT damage along skid trails in this particular study, as bulldozer operators were already

able to move without hesitation, especially when creating secondary and tertiary trails. As

well, most logs were bucked to about 5 m prior to skidding, perhaps avoiding some of the

damage that occurs when tree-length logs are skidded. Two local sawyers were also in

charge of felling operations, perhaps adding to the effectiveness of the operation, as they

are both experienced harvesters in Porto Dias, in addition to having lived for many years

in the reserve.

Conclusion

As more traditional forest dwellers begin to incorporate formalized timber

management into their livelihood strategies (which in the recent past have been based on

the exploitation of NTFPs), it will become increasingly important to monitor and

document these systems. Historically, most of the neotropical research and forest

legislation has focused on large-scale operations (Oliveira 2000). Smallholders will

necessarily differ from their industrial counterparts, given their low harvest intensities

and management regimes that favor a suite of forest products and services beyond timber.

Even though many general sustainable logging guidelines exist, one of the greatest

barriers to good silvicultural management is developing site-specific management

objectives and monitoring principles that recognize the heterogeneity inherent in tropical

forest systems (Putz 1996). As was shown in this study, the gap mosaic of the forest will

change dramatically in comparison to its pre-logging state, even when low intensity, RIL

guidelines are followed (Webb 1997). Accordingly, community forest managers must

establish their own management paradigm while building on the RIL model. Given the

series of poorly understood parameters outlined in this study (pre-harvest liana cutting,









post-logging stand dynamics in bamboo-dominated forest, etc.), it is suggested that

systematic post-harvesting monitoring accompany logging operations (already one of the

requirements for FSC certification), so that the benefits of recommended practices may be

evaluated by the managers at the appropriate scale. Pretty and Hine (1999) argue that

monitoring methods must be sensitive to local needs in order to work in the long term,

primarily due to limited financial and technical resources. It stands to reason then that for

monitoring purposes, local peoples' efforts may be better placed in protecting the larger

size classes of FCTs, particularly if the concessions are on a relatively short (30 years in

this case) cutting cycle. This particular study focuses on the adult and sub-adult size

classes (> 20 cm dbh) in the hopes of capturing information about the next harvest

generation. Many previous logging experiments have concentrated on seedling

regeneration, but these seedlings are unlikely to be recruited to harvestable size classes

within at least two subsequent harvest cycles (Clark and Clark 1992, Connell and Green

2000). There may be more of a solid commitment to participatory monitoring if the

individuals in question could easily project a young tree's value across thirty years (which

would be in their children's lifetime), as opposed to a seedling that may take up to 100

years to get to a harvestable size, if it survives at all.

Reducing damage is one of the most important components of a sustainable forest

management program (Johns et al. 1996). Direct damage to FCTs was shown to be slight

for this study, and indeed, it was considerably less than in other neotropical investigations.

Even so, damage could be reduced even further. The Porto Dias timber management

project is harvesting lightly (7.92 m3 ha-1), but the implementation of liana cutting,

cautious selection of both crop and seed trees and avoidance of stands heavily dominated






66


by Guadua could all help approach sustainable forest management, a concept which may

have the best chances of being executed by those who directly benefit from the natural

resource base on a daily basis.












Table 3-1. Commercial timber species inventoried during 2004 logging operations in the
PAE Porto Dias, Acre, Brazil. For those species harvested in 2004-2005, total
volume (m3 ha-1) is given.
Total volume
Scientific Name Common Name Family harvested (m3)
2004-2005


cereieira Fabaceae 56.6


Amburana cearensis.
Andira sp.
Apuleia leiocarpa
Aspidosperma
megalocarpon
Aspidosperma sp.
Aspidosperma spp.
Aspidosperma vargasii
Astronium lecointeii
Astronium sp. 1
Brosimium uleanum
Carapa guianensis
Cariniana sp. 1
Cedrela cf.fissilis
Cedrela odorata.
Clarisia racemosa
Copaifera cf langsdorfii
Copaifera multijuga
Couratari spp.
Diplotropis spp.
Dipteryx sp. 1
Dipteryx spp.
Enterolobium
shomburgkii

Geissospermum sp.
Hymenaea intermedia
Hymenaea oblongifolia
Hymenaea sp.
Hymenolobium cf.
excelsum
Hymenolobium sp.
Hymenolobium sp.
Hymenolobium sp.
Jacaranda copaia
Manilkara spp.


Fabaceae
Caesalpiniaceae
Apocynaceae

Apocynaceae

Apocynaceae
Apocynaceae
Anacardiaceae
Anacardiaceae
Moracae
Meliaceae
Lecythidaceae
Meliaceae
Meliaceae
Moraceae
Caesalpiniaceae
Caesalpiniaceae
Lecythidaceae
Fabaceae
Fabaceae
Fabaceae
Mimosaceae

Apocynanceae

Fabaceae
Fabaceae


angelim branca
cumaru cetim
carapanauba preta

carapanauba
amarela
pereiro
amareldo
aroeira
aroeira preta
manit6
andiroba
tauari churu
cedro branco
cedro rosa
guarifiba amarela
copaiba preta
copaiba branca
tauari
sucupira preta
cumaruzinho
cumaru ferro
fava orelhinha

quariquara
amarelo
jatobai
jutai
jutai da folha
grande
angelim preto

angelim da mata
angelim pedra
favela preta
marupa
macaranduba


Fabaceae

Fabaceae
Fabaceae
Fabaceae
Bignoniaceae
Sapotaceae


23.4


9.7
1.8

9.4


1.7
4.4






6.1
65.0
7.7


19.5


14.1*





6.5


Fabaceae










Continued Table 3-1.
Total volume
Scientific Name Common Name Family harvested (m3)
2004-2005
Martiodendrom elatum pororoca Caesalpiniaceae
Mezilaurus itauba itauba Lauraceae
Minquartia guianensis? quariquara branca Olacaceae
Myroxylon balsamum balsamo Fabaceae 2.5
Parkia pendula angico vermelho Mimosaceae
roxinho da folha
Peltogyne paniculatum r Caesalpiniaceae
grande
Peltogyne sp. roxinho Caesalpiniaceae
Pouteria cf reticulata abiurana preta Sapotaceae
Pouteria sp. 1 abiurana roxo Sapotaceae
Pouteria sp. 2 abiurana abiu Sapotaceae
Qualea cf tesmanni catuaba Vochysiaceae 3.5
Simarouba amara marmpa preto Simaroubaceae
Tabebuia cf impetigenosa ipe roxo Bignoniaceae
Tabebuia spp. ipe amarelo Bignoniaceae 2.91
Tetragastris altissima breu vermelho Burseraceae
Vatairea spp. sucupira amarela Fabaceae
Vochysia sp. 1 cedrinho Vochysiaceae
Vochysia spp. guaruba Vochysiaceae
* volume recorded under one common name, but likely reflects more than one species of
Hymenolobium.






69



Table 3-2. Classification of damage sustained by FCTs (future crop trees) during logging
operations (modified from Jackson et al. 2002 and Krueger 2004).
Damage type Bole Root Crown
Severe Snapped at base, Uprooted Loss > 66% crown
bent or leaning
Moderate Exposed cambium Exposed cambium Loss 33-66% crown
tissue
Minor bark scrape root scrape Loss < 33% of
crown











Table 3-3. Summary of harvesting impacts in 2004 for the three logging blocks at PAE
Porto Dias, Acre, Brazil.
Impact variable Barrinha Barrinha Palestina
I III
Area sampled (ha) 8.0 8.4 5.4
Harvest density
Number of harvested trees ha' 1.5 1.6 1.4
Percent of harvested trees > 70 cm dbh 27 50 64
Volume harvested (m3 ha-) 7.6 8.5 7.7
Disturbance dimensions
Number of logging gaps 14 13 11
Felling gap area (excluding bole) (m2 ha1) 545.6 433.4 476.6
Proportion of area damaged due to felling gap 5 4 5
(excluding bole) (%)
Mean gap size SD, excluding bole (m2) 389.7 361.2 433.3
S119 249 167
Skid trail area (m2 ha') 519.4 318.6 347.0
Proportion of area damaged due to skid trail (%) 5 3 4
Total area damaged (ha) 1.15 0.83 0.89
FCT damage rate
Total number of damaged FCTs 18 37 15
Basal area of damaged FCTs (m2) 0.34 0.63 0.48
Proportion of FCTs damaged (%) 12 17 15
Frequency of damage categories (%)
Minor or moderate crown, bole, or root damage 78 70 80
Severe crown damage 6 11 0
Severe bole damage 22 24 7
Severe root damage 6 3 27










Table 3-4. Results of a logistic regression for the probability of damage to future crop
trees (FCTs) at Porto Dias, Acre, Brazil during 2004 logging operations.
Shown are the estimates for five continuous covariates, including distance to
nearest stump, felling gap, and skid trail, neighborhood FCT density (number
of stems within a 25 m radius) and local felling intensity (number of trees
felled in the surrounding hectare); and for three categorical independent
variables, including liana presence, location in bamboo-dominated forest, and
marking treatment. Statistically significant variables at P < 0.05 are shown in
bold.
Variable Estimate Wald p
Log distance nearest stump -0.07 0.01 0.91
Log distance nearest felling -1.48 22.96
gap
Log distance nearest skid 5
-2.25 28.37 _0.001
trail
Log neighborhood FCT 0
-0.26 0.24 0.62
density
Local felling intensity 0.16 0.88 0.35
Liana presence (L) 0.31 2.79 0.09
Location in bamboo (G) -0.02 0.01 0.90
Marking treatment (M) -0.14 0.57 0.45
L*G 0.14 0.52 0.47
L*M 0.24 1.55 0.21
G*M -0.10 0.29 0.59
L*G*M 0.11 0.32 0.57




































Figure 3-1. Photograph taken in PAE Porto Dias, Acre, Brazil, demonstrating the re-
curved thorn on a branch of a Guadua culm.
































PAE Porto Dias, Acrelandia, Acre, Brasil


LEGEND
S Primary Road
- Secondary Road
- Creek & River
SBarrinha 1
| Barrinha 3
| Palestina
SSao Pedro


SKilometers
0 2.5 5 10


Figure 3-2. PAE Porto Dias, with the four landholdings inventoried in the study, including the three (in blue) harvested in 2004.











































Figure 3-3. Logging block of Barrinha 1, showing skid trails, harvested trees, and location of future crop trees (FCTs) in the zones of
impact.







































Figure 3-4. Logging block of Barrinha 3, showing skid trails, harvested trees, as well as location of future crop trees (FCTs) in the
zones of impact.







































Figure 3-5. Logging block of Palestina, showing skid trails, harvested trees, as well as location of future crop trees (FCTs) in the zones
of impact.













900


800


700


600


500 -


400 -


300


50 60


70 80 90 100


DBH of harvested tree (cm)



Figure 3-6. Relationship between size of felled trees and the surface area of the
corresponding logging gap created during felling.


r = 0.26, P = 0.001
*


00







0 0

S* *
0
*
0
*














CHAPTER 4
CONCLUSION

Redford and Stearman (1993) propose that despite claims that low-impact

extraction is compatible with biodiversity conservation, if a full range of genetic, species

and ecosystem diversity is to be maintained, then any significant activity by humans must

not be allowed. Advocates of SFM have countered this argument by stating that large-

scale preserves have a limited chance of succeeding in the face of restricted financial

resources, human population growth, and economic pressures (Putz et al. 2000, Pearce et

al. 2003). Who is correct? Or is there no one right answer? Instead, should areas managed

for timber resources be but one component of an interconnected model of complementary

nature, production and indigenous reserves, allowing for various degrees of protection

(Peres and Terborgh 1995)? The forest is going to be exploited, whether so-called

preservationists like it or not. The extent to which that exploitation may incur permanent

damage can be controlled in some ways by planned management activities. Even though

a strong link between biodiversity conservation and SFM may never be fully satisfied, the

global demand for tropical hardwood is not likely to subside anytime soon. This is

especially true of Brazil, which is simultaneously the world's largest producer and

consumer of tropical hardwood (May 2002).

Questions about economic and environmental sustainability still beleaguer

community-based timber projects' quest for staying power, though many see the timber

projects and forest managers as key assets in protecting forest ecosystems from

encroaching mechanized agriculture and cattle production, two of the greatest threats to









Amazon biodiversity (Nepstad et al. 2002, Kainer et al. 2003). With future studies, it will

be challenging to detect changes in the resource base given that many plant and animal

species are harvested simultaneously in tropical forests; ecological tradeoffs associated

with these multispecies systems are so complex as to warrant long-term trial-and-error

experiments (Becker and Ostrom 1995). Thus, complex community level changes may be

difficult to asses (Hall and Bawa 1993). Also, a critical factor to consider in this debate is

such ambiguous vocabulary as "sustainable," "biodiversity" and "ecosystem

management" (Ghazoul 2001). The inability of biologists to agree upon this lexicon

makes the practical application of sustainability guidelines by community forest

managers even more challenging. Consequently, there is a greater risk in resource

degradation if specific ecological problems are not readily identifiable.

The conclusions reached in this thesis suggest that forest dwellers may be able to

sustainably use common resources of the forest, but whether or not they do so will

depend on a wider setting, that of a shifting national and international politico-economic

climate (Cardoso 2002). Fortunately, particularly in the state of Acre, the government has

been extremely supportive of forest dweller communities, in addition to providing a

strong facilitative role. With the continued presence of external agencies, these

communities will stand a better chance at conserving forest resources.

One issue that is frequently mentioned in the literature but still lacks substantial

investigation is the idea that economic sustainability is not always consistent with

ecological sustainability (Hall and Bawa 1993). The links between sustainable

development and resource management must be studied further, as the rural poor do

indeed have an invested interest in sustaining the resource base, but at the same time they






80


must also increase and diversify their overall livelihood (Cardoso 2002). Ultimately, the

success of SFM will be contingent upon addressing the drivers of forest degradation and

deforestation. Conservation will not occur through SFM and forest certification alone;

local, regional, national, and international policy makers must also be engaged in this

process. Otherwise, SFM as a tool for biodiversity conservation has little chance of

success.















REFERENCES


Agrawal, A. A. and C. C. Gibson. 1999. Enchantment and disenchantment: The role of
community in natural resource conservation. World Development 27: 629-649.

Alder, D. and J. N. M. Silva. 2000. An empirical cohort model for management of terra
firme forests in the Brazilian Amazon. Forest Ecology and Management 130: 141-
157.

Allee, W. C. 1949. Group survival value for Philodina roseola, a rotifer. Ecology 30:
395-397.

Amaral, P. and M. A. Neto. 2000. Manejo florestal comunitario na Amaz6nia Brasileira:
Situacgo atual, desafios e perspectives. Brasilia: Instituto Internacional de
Educacgo do Brasil (IIEB). Electronic document, http://www.tech-
inform.de/taller regional/bibliografia/bibliografia.htm, accessed November 30,
2005.

Amaral, P. A. Verissimo, P. Barreto, and E. Vidal. 1998. Floresta para sempre: Um
manual para a produgco de madeira na Amaz6nia. World Wildlife
Fund/IMAZON/USAID, Belem, Para, Brazil.

Appanah, S. and F. E. Putz. 1984. Climber abundance in virgin Dipterocarp forest and the
effect of pre-felling climber cutting on logging damage. The Malaysian Forester
47: 335-342.

Asner, G. P., D. E. Knapp, E. N. Broadbent, P. J. C. Oliveira, M. Keller, and J. N. Silva.
2005. Selective logging in the Brazilian Amazon. Science 21: 480-482.

Awasthi, A., S. K. Uniyal, G. S. Rawat, and A. Rajvanshi. 2003. Forest resource
availability and its use by migratory villages of Uttatkashi Garwhal Himalaya
(India). Forest Ecology and Management 174: 13-24.

Banana, A. Y. and W. Gombya-Ssembajjwe. 2000. Successful forest management: The
importance of security of tenure and rule enforcement in Ugandan forests. Pages
87-98 in C. C. Gibson, M. A. McKean and E. Ostrom, eds. People and forests:
Communities, institutions, and governance. MIT Press, Cambridge, MA, USA.

Banuri, T. and F. Apffel-Marglin. 1993. Who will save the forest? Knowledge, power
and environmental destruction. Zed Books, London, UK.









Baraloto, C. and P.-M. Forget. 2004. Effects of post-logging microhabitats on
regeneration from seed: Implications for sustainable forest management. Pages
191-203 in S. Gourlet-Fleury, J.-M. Guehl, and O. Laroussinie, eds. Ecology and
management of a neotropical rainforest: Lessons drawn from Paracou, a long-term
experimental research site in French Guiana. Elsevier, Paris, France.

Barreto, P., P. Amaral, E. Vidal, and C. Uhl. 1998. Costs and benefits of forest
management for timber production in eastern Amazonia. Forest Ecology and
Management 108: 9-26.

Barreto, P., C. Uhl and G. Yared. 1993. O potential de producgo sustentavel de Madeira
em Paragominas, Para na Amaz6nia Oriental: Considerac6es ec6logicas e
econ6micas. 7 Congresso Florestal Brasileiro, Anais da Sociedade Brasileiro de
Engenheiros Florestais, pp. 387-392.

Bawa, K. S., D. R. Perry and J. H. Beach. 1985. Reproductive biology of tropical lowland
rainforest trees. I. Sexual systems and incompatibility mechanisms. American
Journal of Botany 72: 331-345.

Bawa, K. S. and R. Seidler. 1998. Natural forest management and conservation of
biodiversity in tropical forests. Conservation Biology 12: 46-55.

Becker, C. D. and R. Le6n. 2000. Indigenous forest management in the Bolivian
Amazon: Lessons from the Yuracare people. Pages 163-191 in C. C. Gibson, M. A.
McKean and E. Ostrom, editors. People and forests: Communities, institutions, and
governance. MIT Press, Cambridge, MA, USA.

Becker, C. D. and E. Ostrom. 1995. Human ecology and resource sustainability: The
importance of institutional diversity. Annual Review of Ecology and Systematics
26: 113-133.

Berkes, F. 1999. Sacred ecology: Traditional ecological knowledge and resource
management. Taylor and Francis, Philadelphia, Pennsylvania, USA.

Berkes, F. and C. Folke. 2002. Back to the future: Ecosystem dynamics and local
knowledge. Pages 121-146 in L.H. Gunderson and C.S. Holling, editors. Panarchy:
Understanding transformations in human and natural systems. Island Press,
Washington, USA.

Bertault, J.-G. and P. Sist. 1997. An experimental comparison of different harvesting
intensities with reduced-impact and conventional logging in East Kalimantan,
Indonesia. Forest Ecology and Management 94: 209-218.

Bobadilla, U. L. and S. F. Ferrari. 2000. Habitat use by Chiroptes satanas utahicki and
Syntopic platyrrhines in eastern Amazonia. American Journal of Primatology 50:
215-224.









Boot, R. G. A. and R. E. Gullison. 1995. Approaches to developing sustainable extraction
systems for tropical forest products. Ecological Applications 5: 896-903.

Bowles, I., R. Rice, R. Mittermeier, and A. daFonseca. 1998. Logging and tropical
conservation. Science 280: 1899-1900.

Bray, D. B., L. Merino-Perez, P. Negreros-Castillo, G. Segura-Wamholtz, J. M. Torres-
Rojo, and H. F. M. Vester. 2003. Mexico's community-managed forests as a global
model for sustainable landscapes. Conservation Biology 17: 672-673.

Braz, E. M. and M. V. N. d'Oliveira. 1996. Planning to reduce damage. Tropical Forest
Update 6: 13-18.

Brokaw, N. V. I. 1982. The definition of treefall gap and its effects on measures of forest
dynamics. Biotropica 11: 158-160.

Bruenig, F. E. 1996. Conservation and management of tropical rainforests: An integrated
approach to sustainability. CAB International, New York, USA.

Bush, M. B. and P. A. Colinvaux. 1994. A paleoecological perspective of tropical forest
disturbance: Records from Darien, Panama. Ecology 75: 1761-1768.

Calheiros, D. F., A. F. Seidl and C. J. A. Ferreira. Participatory methods in environmental
science: Local and scientific knowledge of a limnological phenomenon in the
Pantanal wetland of Brazil. Journal of Applied Ecology 37: 684-696.

Camino, R. and M. Alfaros. 2000. Certification in Latin America: Experience to date.
Forest, Trees and People Newsletter 43: 25-27.

Cardoso, C. A. S. 2002. Extractive reserves in Brazilian Amazonia: Local resource
management and the global political economy. Ashgate Studies in Environmental
Policy and Practice, Hampshire, UK.

Carter, J. 1996. Recent approaches to participatory forest resource management. Rural
Development Forestry Network, Overseas Development Institute, London, UK.

Chapman, C. A., L. J. Chapman, R. Wangham, K. Hunt, D. Gebo, and L. Gardner. 1992.
Estimators of fruit abundance of tropical trees. Biotropica 24: 527-531.

Clark, D. A. and D. B. Clark. 1992. Life history diversity of canopy and emergent trees in
a neotropical rain forest. Ecological Monographs 62: 315-344.

Connell, J. H. and P. T. Green. 2000. Seedling dynamics over thirty-two years in a
tropical rain forest tree. Ecology 81: 568-584.

Dalling, J. W., H. C. Muller-Landau, S. J. Wright, and S. P. Hubbell. 2002. Role of
dispersal in the recruitment limitation of neotropical pioneer species. Journal of
Ecology 90: 714-727.









Daube, E., T. Fredericksen and M. Pefia. 2005. Sustainability of timber harvesting in
Bolivian tropical forests. Forest Ecology and Management 214: 294-304.

Deneven, W. M. 1992. The pristine myth: The landscape of the Americas in 1492.
Annals of the Association of American Geographers 82: 369-385.

Denslow, J. S. 1980. Gap partitioning among tropical rainforest trees. Biotropica 12: 47-
55.

Dickinson, M. B., D. F. Whigham and S. M. Hermann. 2000. Tree regeneration in felling
and natural treefall disturbances in a semideciduous tropical forest in Mexico.
Forest Ecology and Management 134: 137-151.

Dykstra, D. P., and R. Heinrich. 1996. FAO model code of forest harvesting practice.
Food and Agricultural Organization of the United Nations, Rome, Italy.

Food and Agricultural Organization (FAO). 2005. State of the world's forests. Food and
Agricultural Organization of the United Nations, Rome, Italy.

Fearnside, P. M. 2003. Conservation policy in Brazilian Amazonia: Understanding the
dilemmas. World Development 31: 757-779.

Folke, C., F. Berkes and J. Colding 1998. Ecological practices and social mechanisms for
building resilience and sustainability. Pages 414-436 in F. Berkes and C. Folke,
editors. Linking social and ecological systems. Cambridge University Press,
London, UK.

Fox, J. E. D. 1968. Logging damage and the influence of climber cutting prior to logging
in the lowland Dipterocarp forest of Sabah. The Malayan Forester 31: 326-347.

Fredericksen, T. S. and F. E. Putz. 2003. Silvicultural intensification for tropical forest
conservation. Biodiversity and Conservation 12: 1445-1453.

Gerwing, J. J. 2001. Testing liana cutting and controlled burning as silvicultural
treatments for a logged forest in the eastern Amazon. Journal of Applied Ecology
38: 1264-1278.

Gerwing, J. J. and E. Vidal. 2002. Changes in liana abundance and species diversity eight
years after liana cutting and logging in an eastern Amazonian forest. Conservation
Biology 16: 544-548.

Ghazoul, J. 2001. Barriers to biodiversity conservation in forest conservation.
Conservation Biology 15: 315-317.

Ghazoul, J., K. A. Liston and T. J. B. Boyle. 1998. Disturbance-induced density-
dependent seed set in .\lh,mii siamensis (Dipterocarpaceae), a tropical forest tree.
Journal of Ecology 86: 462-473.









Gibson, C. C. and C. D. Becker. 2000. A lack of institutional demand: Why a strong local
community in western Ecuador fails to protect its forest. Pages 135-161 in C. C.
Gibson, M. A. McKean and E. Ostrom, editors. People and forests: Communities,
institutions, and governance. MIT Press, Cambridge, MA, USA.

Griscom, B. W. 2003. The influence of bamboo (Guadua sarcocarpa and Guadua
weberbaueri) on stand dynamics in lowland terrafirme forests of southeastern
Peru. PhD dissertation. Yale University, New Haven, CT, USA.

Griscom, B. W. and P. M. S. Ashton. 2003. Bamboo control of forest succession: Guadua
sarcocarpa in southeastern Peru. Forest Ecology and Management 175: 445-454.

Guariguata, M. R. and M. A. Pinard. 1998. Ecological knowledge of regeneration from
seed in neotropical forest trees: Implications for natural forest management. Forest
Ecology and Management 112: 87-99.

Gullison, R. E. and Hardner, J. J. 1993. The effects of road design and harvest intensity
on forest damage caused by selective logging: empirical results and a simulation
model for the Bosque Chimanes, Bolivia. Forest Ecology and Management 59: 1-
14.

Hall, P. and K. Bawa. 1993. Methods to assess the impact of extraction of non-timber
tropical forest products on plant populations. Economic Botany 47: 234-247.

Hammond, D. S., S. Gourlet-Fleury, P. van der Hout, H. ter Steege, and V. K. Brown.
1996. A compilation of known Guianan timber trees and the significance of their
dispersal mode, seed size and taxonomic affinity to tropical rainforest management.
Forest Ecology and Management 83: 99-116.

Hardin, G. 1968. The tragedy of the commons. Science 162: 1243-1248.

Hartshorn, G. S. 1989. Application of gap theory to tropical forest management: Natural
regeneration in strip clearcuts in the Peruvian Amazon. Ecology 70: 567-569.

Heckenberger, M. J., A. Kuikuro, U. T. Kuikuro, J. C. Russell, M. Schmidt, C. Fausto,
and B. Franchetto. Amazonia 1492: Pristine forest or cultural parkland? Science
301: 1710-1714.

Heideman, P. D. 1989. Temporal and spatial variation in the phenology of flowering and
fruiting in a tropical rainforest. Journal of Ecology 77: 1059-1079.

Holmes, T. P., G. M. Blate, J. C. Zweede, R. Pereira, P. Barreto, F. Boltz, and R. Bauch.
2002. Financial costs and benefits of reduced-impact logging relative to
conventional logging in the eastern Amazon. Forest Ecology and Management 163:
93-110.

Hubbell, S. P. 1979. Tree dispersion, abundance, and diversity in a tropical dry forest.
Science 203: 1299-1309.









Humphries, S. 2005. Forest certification for community-based forest enterprises in
Brazil's Western Amazon: Local stakeholders' perceptions of negative and positive
aspects and the certification process. MSc Thesis. University of Florida,
Gainesville, FL, USA.

Janzen, D. 1976. Why bamboos wait so long to flower. Annual Review of Ecology and
Systematics 7: 347-391.

Janzen, D. 2004. Setting up tropical biodiversity for conservation through non-damaging
use: Participation by parataxonomists. Journal of Applied Ecology 41: 181-187.

Jennings, S. B., N. D. Brown, D. H. Boshier, T. C. Whitmore, and J. do C. A. Lopes.
2001. Ecology provides a pragmatic solution to the maintenance of genetic
diversity in sustainably managed tropical rain forests. Forest Ecology and
Management 154: 1-10.

Johns, A. D. 1991. Responses of Amazonian rain forest birds to habitat modification.
Journal of Tropical Ecology 7: 417-437.

Johns, J. S., P. Barreto and C. Uhl. 1996. Logging damage during planned and unplanned
logging operations in the Eastern Amazon. Forest Ecology and Management 89:
59-77.

Johnson, N. and B. Cabarle. 1993. Surviving the cut. World Resources Institute,
Washington, D.C. 71 pp.

Kainer, K. A., M. Schmink, A. C. Pinheiro Leite, and M. J. daSilva Fadell. 2003.
Experiments in forest-based development in western Amazonia. Society and
Natural resources 16: 869-886..

Krueger, W. 2004. Effects of future crop tree flagging and skid trail planning on
conventional diameter-limit logging in a Bolivian tropical forest. Forest Ecology
and Management 188: 381-393.

Lacerda, A. and G. Griffiths. 2005. Conservation and sustainability within Brazilian
Amazon forests: The role of botanical identification. Presentation given at the
Association of Tropical Biology and Conservation, Uberlandia, Brazil, July 24-28.

Lawrence, A. and F. S. Roman. 1996. The role of inventory in the communally managed
forests of Quintana Roo, Mexico. Pages 83-110 in J. Carter, ed. Recent approaches
to participatory forest resource management. Rural Development Forestry
Network, Overseas Development Institute, London, UK.

Lentini, M., A. Verissimo and L. Sobral. 2003. Forest facts in the Brazilian Amazon.
IMAZON, Belem, Brazil.









Lieberman, M. and D. Lieberman. 1994. Patterns of density and dispersion of forest trees.
Pages 106-119 in L. A. McDade, K. S. Bawa, H. A. Hespenheide, and G. S.
Hartshorn, editors. La Selva: Ecology and natural history of a neotropical
rainforest. University of Chicago Press, Chicago, IL, USA.

Lorenzi, H. 2000. Arvores Brasileiras. Institute Plantarum de Estudos da Flora Ltda.,
Nova Odessa, Sao Paulo, Brazil.

Marcon, E. 2005. Ripley, version 3.1. http://e.marcon.free.fr/Ripley/, accessed December
5, 2005.

Margoluis, R. and N. Salafsky. 1997. Measures of success: Designing, managing, and
monitoring conservation and development projects. Island Press, Washington, DC,
USA.

May, P. H. 2002. Forest certification in Brazil: Trade and environmental enhancement.
Consumer Choice Council, Washington, DC, USA.

Menton, M. C. 2003. Effects of logging on non-timber forest product extraction in the
Brazilian Amazon: Community perceptions of change. International Forestry
Review 5: 97-105.

Metzger, F. and J. Schultz. 1984. Understory response to 50 years of management of a
northern hardwood forest in upper Michigan. American Midland Naturalist 112:
209-223.

Meyer, J. L. and G. S. Helfman. 1993. The ecological basis of sustainability. Ecological
Applications 3: 569-571.

Molnar, A. 2003. Forest certification and communities: looking forward to the next
decade. Forest Trends, Washington, DC, USA

Morellato, P. C. and H. F. Leitao-Filho. 1996. Reproductive phenology of climbers in a
southeastern Brazilian forest. Biotropica 28: 180-191.

Miller, U. 1998. Effects of refinements on the development of a neotropical rainforest
after controlled logging in Suriname. MSc thesis. Wageningen Agricultural
University, Wageningen, The Netherlands.

Nelson, B. W., A. C. A. Oliveira, G. T. Batista, P. Vidalenc, and M. Silveira. 2001.
Modeling biomass in the southwest Amazon by polar ordination of Landsat TM.
Pages 1683-1690 in Anais X SBSR, Foz do Iguacu, 21-26 April 2001, INPE.

Nepstad, D., A. Alencar, A. C. Barros, E. Lima, E. Mendoza, C. A. Ramos, and P.
Lefebvre. 2004. Governing the Amazon timber industry. Pages 388-414 in D.
Zarin, F. J. Putz, M. Schmink and J. Alavalapati, editors. Working forests in the
tropics: Conservation through sustainable management? Columbia university Press,
New York, NY, USA.









Nepstad, D., D. McGrath, A. Alencar, A. C. Barros, G. Carvalho, M. Santilli, and M. d.
C. Vera Diaz. 2002. Frontier governance in Amazonia. Science 295: 629-631.

Nyland, R. D. 2002. Silviculture concepts and applications. McGraw-Hill, New York,
USA.

Oliveira, M. V. N. d'. 2000. Sustainable forest management for small farmers in Acre
state in the Brazilian Amazon. PhD dissertation. University of Aberdeen,
Aberdeen, Scotland.

Oliveira, M. V. N. d', E. M. Braz, D. F. R. P. Burslem, and M. D. Swaine. 1998. Small-
scale natural forest management: A new model for small farmers in the Brazilian
Amazon. Tropical Forest Update 8: 5-7.

Panfil, S. N. and R. E. Gullison. 1998. Short term impacts of experimental timber harvest
intensity on forest structure and composition in the Chimanes Forest, Bolivia.
Forest Ecology and Management 102: 235-243.

Parren, M. P. E. 2003. Lianas and logging in West Africa. PhD dissertation. Wageningen
University, Wageningen, The Netherlands.

Parren, M. and F. Bongers 2001. Does climber cutting reduce felling damage in southern
Cameroon? Forest Ecology and Management 141: 175-188.

Pearce, D., F. E. Putz and J. K. Vanclay. 2003. Sustainable forestry in the tropics:
Panacea or folly? Forest Ecology and Management 172:229-247.

Peres, C. A., C. Baider, P. A. Zuidema, L. H. O. Wadt, K. A. Kainer, D. A. P. Gomes-
Silva, R. P. Salomdo, L. H. Sim6es, E. R. N. Franciosi, F. C. Valverde, R. Gribel,
G. H. Shepherd, M. Kanashiro, P. Coventry, P. W. Yu, A. R. Watkinson, and R. P.
Freckleton. 2003. Demographic threats to the sustainability of Brazil nut
exploitation. Science 302: 2112-2114.

Peres, C. A. and J. Terborgh. 1995. Amazonian nature reserves: An analysis of the
defensibility status of existing conservation units and design criteria for the future.
Conservation Biology 9: 34-46.

Pereira, R. Jr., J. Zweede, G. P. Asner, and M. Keller. Forest canopy damage and
recovery in reduced-impact and conventional selective logging in eastern Para,
Brazil. Forest Ecology and Management 168: 77-89.

Pinard, M. A. and F. E. Putz. 1996. Retaining forest biomass by reducing logging
damage. Biotropica 28: 278-295.

Pinard, M. A., F. E. Putz, J. Tay, and T. E. Sullivan. 1995. Creating timber harvesting
guidelines for a reduced-impact logging project in Malaysia. Journal of Forestry
93:41-45.