<%BANNER%>

Exploration of Food Security Via a Meso Scale Modeling Approach

Permanent Link: http://ufdc.ufl.edu/UFE0021174/00001

Material Information

Title: Exploration of Food Security Via a Meso Scale Modeling Approach Jamaica as an Illustrative Case
Physical Description: 1 online resource (143 p.)
Language: english
Creator: Justice, Lauren
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: dss, food, interdisciplinary, jamaica, model, qnd, systems
Interdisciplinary Ecology -- Dissertations, Academic -- UF
Genre: Interdisciplinary Ecology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The purpose of my study is to examine the interactions within a food system and between that system and its socioecological environment in order to reveal management or policy strategies that strengthen food security conditions. Using Jamaica as a case study my interdisciplinary project developed and tested a computer model, QnD:Jamaica, that fuses technology with lessons from complex systems theory in order to promote adaptive management of food systems. An iterative collaborative design process based on soft systems approaches was employed during model design. The problem situation as understood by key stakeholders was investigated through informal strategic dialogues. These qualitative descriptions were conceptualized by a formal framework to identify a limited number of quantifiable parameters and drivers of interest for the scenario model. The model incorporates ecosystem and sociopolitical issues into a user-friendly framework to analyze the spatially explicit dynamics of food, money and people in terms of their effects on food security conditions. Illustrative scenario simulations were performed and the implications for systemic features were explored. The results of my study provide insight into conceptualizing food systems in general and the Jamaican food system specifically. My thesis serves as an explicit example of integrated soft research methods and analytical tool development and will guide future modeling efforts.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lauren Justice.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Kiker, Gregory.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021174:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021174/00001

Material Information

Title: Exploration of Food Security Via a Meso Scale Modeling Approach Jamaica as an Illustrative Case
Physical Description: 1 online resource (143 p.)
Language: english
Creator: Justice, Lauren
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: dss, food, interdisciplinary, jamaica, model, qnd, systems
Interdisciplinary Ecology -- Dissertations, Academic -- UF
Genre: Interdisciplinary Ecology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The purpose of my study is to examine the interactions within a food system and between that system and its socioecological environment in order to reveal management or policy strategies that strengthen food security conditions. Using Jamaica as a case study my interdisciplinary project developed and tested a computer model, QnD:Jamaica, that fuses technology with lessons from complex systems theory in order to promote adaptive management of food systems. An iterative collaborative design process based on soft systems approaches was employed during model design. The problem situation as understood by key stakeholders was investigated through informal strategic dialogues. These qualitative descriptions were conceptualized by a formal framework to identify a limited number of quantifiable parameters and drivers of interest for the scenario model. The model incorporates ecosystem and sociopolitical issues into a user-friendly framework to analyze the spatially explicit dynamics of food, money and people in terms of their effects on food security conditions. Illustrative scenario simulations were performed and the implications for systemic features were explored. The results of my study provide insight into conceptualizing food systems in general and the Jamaican food system specifically. My thesis serves as an explicit example of integrated soft research methods and analytical tool development and will guide future modeling efforts.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lauren Justice.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Kiker, Gregory.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021174:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20110404_AAAADM INGEST_TIME 2011-04-04T21:33:12Z PACKAGE UFE0021174_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 10996 DFID F20110404_AACLCK ORIGIN DEPOSITOR PATH justice_l_Page_143.pro GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
737403e46110c4a2c25688ce9982ece6
SHA-1
51a80c4a325a032a846e4bdf4c0931856eb39a97
1841 F20110404_AACKXE justice_l_Page_117.txt
0d12038c15d2c977dc02afe54ceeb825
69b1c9025e2c796d72b1c19b8514093e4a818cea
43829 F20110404_AACLBW justice_l_Page_128.pro
f6cc3cd2a6522a6a0694a40d4ddfa32b
79abf4dd9ebffe8e86052c59330cf10244c85016
725 F20110404_AACKWQ justice_l_Page_100.txt
74216d683bbaceea632ccc123c467768
d5c2c50e4c3b615fe47fbc7924af11e02550b28f
91034 F20110404_AACLDA justice_l_Page_010.jpg
d94f9538ad480d28b6c99cbb8ce96e4e
eec19e7908569ae4cfc54ef4ba7225bf2299e2c9
25814 F20110404_AACLCL justice_l_Page_001.jpg
8dcb4d9615c66c081897be4a55ffee0d
e77e933edc2ad27e1658be2677c47eb7cfb993b2
1302 F20110404_AACKXF justice_l_Page_118.txt
967abcb2d84f1592b96f833077e7fc7e
09e3ebb2ac5684b5d50fa4db02287263fdbb4ab0
18202 F20110404_AACLBX justice_l_Page_129.pro
fcbf55c7e8cfcd411c1b1a5380f54c23
d89b8e0b02c22dcbd957baad7353f98fb19e5110
1582 F20110404_AACKWR justice_l_Page_101.txt
c49be2f67ef2245c8a9f3fc24f510d36
94940f782bf2e8a5dc1c15a13e8984600d202022
26777 F20110404_AACLDB justice_l_Page_010.QC.jpg
04f9c739d248fe1dfe6c5caba0c59693
67729fa8cff1f4f3ef54ad617680e0029d90fbad
1259 F20110404_AACLCM justice_l_Page_002.QC.jpg
fe80c07d0fb65914cbdd32e5d4c03b34
0b3f1a70b4d5d253a75f227669f8689ddfc1e4f2
378 F20110404_AACKXG justice_l_Page_120.txt
41bc8977bf97291af2e3c080c45c4fea
01ee6a39b541bc9e81df38a8a57468f9078cf44c
51868 F20110404_AACLBY justice_l_Page_130.pro
816f05257b300845b7f2ad335f54e9bd
c9014c3a8f923f7093e3b0a4558c54e6ca0794e0
579 F20110404_AACKWS justice_l_Page_102.txt
a39342b43fe7fc2fcc5955b666faf1c6
a40f4b3a7cfcfbde502f9f792434e026de4b9c20
94719 F20110404_AACLCN justice_l_Page_003.jpg
a6172f16210b0608be9931442f9e34b0
ab078ca6f242ce08e30e916a89f1e7599d6b4cf4
2246 F20110404_AACKXH justice_l_Page_123.txt
11db90469a800f0ccb15df1dde782f8a
08c082242110c1cc91b42aca0cd780a5b89ac9c4
56008 F20110404_AACLBZ justice_l_Page_131.pro
c967e983610213b37ceb2b1773e054ec
1d4318d216a7b2d4bb350ccd2b2ee95e984fa45c
1351 F20110404_AACKWT justice_l_Page_103.txt
3adf2c287cc4c51b02a2969aa6a37edb
7e3006873424cce9728dd530fc784352247f9a39
12750 F20110404_AACLDC justice_l_Page_011.jpg
e31d3fe1f27ca661fd48f5264ddea1fe
b7044ad72151c287d76b3738db25732aacd3749b
30988 F20110404_AACLCO justice_l_Page_003.QC.jpg
3e315332513e7321f8d292bf7b6cdd5e
3a8ad4eed869e6d045ab000ed9980c4a72be616d
1376 F20110404_AACKXI justice_l_Page_124.txt
e0103dd16153ea06518be3a44309b336
0d192281c7681e3153caaed4ca9bed85b4a1cc53
779 F20110404_AACKWU justice_l_Page_104.txt
26db4335e506cb102197a5dece5834ae
13264c4a3ba9fc5f5c8ff35257aead6d832c6893
3840 F20110404_AACLDD justice_l_Page_011.QC.jpg
3a3278360427d0a03941e65af818c705
27ab7822ec9751da5523486ee817f9f3ac1626ba
101369 F20110404_AACLCP justice_l_Page_004.jpg
43819b0cc0e01523dfcb11e88dd0434e
7bd6f3ead807324fec06d044e4a7c929ff1411fd
1693 F20110404_AACKXJ justice_l_Page_125.txt
1b0a9bde3a43e6c71335d7b65abc7313
ace03f547a8861e6615b3007375352f1c9495b0f
1227 F20110404_AACKWV justice_l_Page_105.txt
705d21a1a23a5bda126054b327e01e12
5ebadf8ba98a0943ac3d3e1f76ecf16be5126887
28397 F20110404_AACLDE justice_l_Page_012.QC.jpg
44b6c4b1e8a3e6286afea83c62cfd6bf
7c2f5e6e44e7ebd09389dc4394bf0ae32b00257e
24252 F20110404_AACLCQ justice_l_Page_004.QC.jpg
a5304df707ff6a3a70daf94e37a9307a
2cb7bcb6c11de63757279b1e156dd8d149a93254
F20110404_AACKXK justice_l_Page_126.txt
438fe181dc6ede86fd1ef9781d89a825
8f08f1d69a2a9932c8c30f93b3acd55557b3a3f6
1209 F20110404_AACKWW justice_l_Page_106.txt
9d5168d2ed1c8d56c2f003361994e9b4
ff2d77fa8b245d7bbab37393b82ae01b9152955f
115482 F20110404_AACLDF justice_l_Page_014.jpg
5d599a2fe2915e8bb7819511e9867d78
15528b05edb8cf998b7336d065e31f071c961c4a
119341 F20110404_AACLCR justice_l_Page_005.jpg
7cd8ca9bfd170a651428b1b27c3b00c4
b165314fc41ccfbe6c178031a7fabb7d65433203
1922 F20110404_AACKXL justice_l_Page_128.txt
3b3b06ed568f05176f4866f71b73a8ae
bf088140bde965548e2f94b6ae1379c32f053dd7
1645 F20110404_AACKWX justice_l_Page_107.txt
4e1eff578934b014c5c99174217cb24d
4ea27e02f3d1199fce746a4100a58bbfdde37d7a
34708 F20110404_AACLDG justice_l_Page_014.QC.jpg
1f66b271df90f68002537511d9bad333
ad4a1efd0512a2ff74444f21dfaadef071d86899
121865 F20110404_AACKYA justice_l_Page_005.pro
79d2b1d25ab72ca1aa016389a537bd46
bd7a28bfbfc6d9350b3b4f31fda8a767c454823c
25121 F20110404_AACLCS justice_l_Page_005.QC.jpg
f0f527d06b03cc6ab5b6b15d48fcb59a
c2be9c9867f7b0135e42adac353d6bb8f4f21340
723 F20110404_AACKXM justice_l_Page_129.txt
b15b675bb0e1323492e519b0b1b7417a
e65f0ba19893111d5ad25460dd964813feae9b52
2191 F20110404_AACKWY justice_l_Page_108.txt
e5204241d7718c9cfbd55672bf725481
6f7518d00368cd2f3fd415fb0af4d1f0b646d686
19938 F20110404_AACKYB justice_l_Page_006.pro
9a0ddda00d33f0aaa2b3c8c506c0b14a
a22cffd838cf7b5ac9fead332ae528b7e5e33b1d
22617 F20110404_AACLCT justice_l_Page_006.jpg
4430d38863eef3c520599dba380df717
f3ddf7be07e4c053180235459dd4f8b78a4c18ea
2150 F20110404_AACKXN justice_l_Page_130.txt
9b25155719a3d13735502b6bb7caed6b
073adf41a540b4425f8201e6a196b92e12dc02e0
1534 F20110404_AACKWZ justice_l_Page_109.txt
b3477e15204aad542187202f92923d8f
725a036b474114b5971555c967d6e1e2c7af20e8
110430 F20110404_AACLDH justice_l_Page_015.jpg
42093f8d793937a587fcd90099db233e
0829f3fe02e9943fec5ae1f03891fcb1cc75792f
48164 F20110404_AACKYC justice_l_Page_007.pro
cb37a07b6fe5d0db7bcd7f9510a05c77
b6073864f1eb4a2699c2b65117bba971fbec7cf5
5592 F20110404_AACLCU justice_l_Page_006.QC.jpg
3b729b670fc573231404ae4969c0a75f
d21c3d62e3399a77338fd9a637156d03b6f12cb7
2230 F20110404_AACKXO justice_l_Page_131.txt
15d44332afb0405fbb6113f732e7eecf
9c8398f8f75d072c6e7b2adfcdee9ea99ac9676d
36400 F20110404_AACLDI justice_l_Page_015.QC.jpg
5650a9736e9680bffd95881c9e5a4477
d232c1ef47a5770c161f74022e55d2df12c3d7b2
72796 F20110404_AACKYD justice_l_Page_008.pro
17d61fa93aa9d7733eeca4fa14fa64e2
8a59c52e54fcafde188999359e409d67c35dbb59
57107 F20110404_AACLCV justice_l_Page_007.jpg
da98a9c9442e895cde608815a212eccc
9f530e1f94687159f76d43ebb8d60b489b1fb1a2
620 F20110404_AACKXP justice_l_Page_132.txt
d053d1787e48228b59aa4f337e4161e8
e0ec6eaf755773570e295f90c42d839506d1fefa
36783 F20110404_AACLDJ justice_l_Page_016.QC.jpg
ec94f73f5a00fe8e9f5a8a644f278a48
e35b80357a42a5da8987ddc6f779055b9b90a453
78216 F20110404_AACKYE justice_l_Page_009.pro
5cd527958ed11456b6fbd8a437d6c2c7
7882da2082c0f7827b7c3888bd77ae041fa8618c
16273 F20110404_AACLCW justice_l_Page_007.QC.jpg
8d002a58957f40bd52a7bd0250242f9f
c062893ac194183e06c359c402e675c34dc761a3
2507 F20110404_AACKXQ justice_l_Page_133.txt
b62762214d65fde96f2fe59346b94ba3
037774d19cf4cfc0507cacea2458547da1afcfe7
63864 F20110404_AACLDK justice_l_Page_017.jpg
4309bb2b91d456cb4ca09f14971fc3ad
45d679e0a6ca097f3c4f97e908796da2efacaf51
76993 F20110404_AACKYF justice_l_Page_010.pro
dabe2066a4b861b8e9f7ec86ef640c5c
4d0c8a85e7342df42c211ef9955eca354cc4a15a
89474 F20110404_AACLCX justice_l_Page_008.jpg
2fcb44dba0028703ed7ea511914ddf14
e05056b28464f4c2e13479c5970b3f452804abf4
2412 F20110404_AACKXR justice_l_Page_135.txt
229b618e2b98e48661a6de1e2284d064
5b28f9af0bfa721bfc6ff4e14e82dcc14cb362c8
35883 F20110404_AACLEA justice_l_Page_029.QC.jpg
e0941fcace893dcda67dd86c611d1bab
6835b4828f4bf6d2b94fb93839f4e5ca648ef2b3
21092 F20110404_AACLDL justice_l_Page_017.QC.jpg
a6f6e44c82d42a311dde8b7a0ffc87cf
778bf7793005945209240a3ba7d0bde7ba0b841b
43414 F20110404_AACKYG justice_l_Page_012.pro
812bc44e2a6dd497f94d8b19f85a7469
dd1c6c4afd1c4952b21dfb6a96b59e923448592e
26196 F20110404_AACLCY justice_l_Page_008.QC.jpg
e157b8a5bb301c44b474b2db3be0ab07
acb389aca8e6043e59c401fc206c3608dbc6cae2
2216 F20110404_AACKXS justice_l_Page_136.txt
2f5e5124b031d30ea167bd77c40a3715
2c220b939d5cbd6181ee1c2a3d07851bb1058aa2
94342 F20110404_AACLEB justice_l_Page_030.jpg
fc2d6f5dc54693f392846b15b7dab63a
65f55bff28a383ebc732e0bf5126956a15e15a97
33621 F20110404_AACLDM justice_l_Page_018.QC.jpg
2b114c351fffca7e7ac3a88ecfd3312a
1ed8ea4ac1e347b317345abdcca09551a5376170
5034 F20110404_AACKYH justice_l_Page_013.pro
3d0b5f3d891cefce8c404827bba7a4eb
2b7e4ccc515aac6e36befa734c3873ed8e76f948
26125 F20110404_AACLCZ justice_l_Page_009.QC.jpg
440fae1f18376d6129b45f49d72d39ba
aa5e0f76e148c14aa6bb8f58cdf356b7c73ac399
2558 F20110404_AACKXT justice_l_Page_137.txt
00457b0e55e4e34b6850964496387a76
df9581c2b94a8596bcacc9c4471f858e7d5c5efd
31981 F20110404_AACLEC justice_l_Page_030.QC.jpg
e69fb7f0eb031a2e31ff6e6ddf35fc79
f20ddb60e84ee779c604e129a7853eb7a1cf8965
100350 F20110404_AACLDN justice_l_Page_019.jpg
73cf684e2859b0bc29fe2c73de36f1ff
897261caebf189986a77de54701230bbc63b9b42
56861 F20110404_AACKYI justice_l_Page_014.pro
349fbb45ce8d41a2e05d74ad2aa40d16
98c11e6aa54eea818e02b0c9aff95b783f6311cf
2715 F20110404_AACKXU justice_l_Page_139.txt
b202f07f41841a9c0ec65e5aeda58ff0
e531a17e7ba7df138683840e8c0dcdb38dbbfe38
92633 F20110404_AACLED justice_l_Page_031.jpg
f8b2e703327453b13a20dd30c2a3297a
fb6f27e9cbaa31f5b8b5ad0b66db78439dfca104
107067 F20110404_AACLDO justice_l_Page_020.jpg
f8db242d93c876f87bad61a2b381f332
98d9c9075dca06b7231a6dc09a42c3d832d25bb2
54475 F20110404_AACKYJ justice_l_Page_015.pro
2752bb1c911dd5a9b8921795d30cc17d
cc0a50c249ab59b37518ddbc93ff9875ae6205a6
2281 F20110404_AACKXV justice_l_Page_140.txt
d42e3516ea3d3f6aec22ffde14d87bd8
fd6d3cc7e23ebd13fb297a94d90e94cf21a50e8b
30027 F20110404_AACLEE justice_l_Page_031.QC.jpg
7015c6c7469c78fc24a4f054a221f494
b73a9f44e3eddb224cf527208e1977189e8939b7
34802 F20110404_AACLDP justice_l_Page_020.QC.jpg
c3388435f7b731b465bc40fe948826d1
50fd30b92136cd3eba2f34e337afe36c290c3aad
54240 F20110404_AACKYK justice_l_Page_016.pro
41c39f2e691d72832103c1e660cc77cf
bc46b6ff9029e423cfd2d807f1a013a48c7b262a
476 F20110404_AACKXW justice_l_Page_143.txt
853788438e81396da1c75dad8c397064
8ec7168550921b3f8e9f2840a83cce1895c10da7
23309 F20110404_AACLEF justice_l_Page_032.jpg
ade0d06319b99bfc9e9dc31d387610b4
87d445a7a8f7a3cebfe51472b9adb4721a4be0cc
109080 F20110404_AACLDQ justice_l_Page_021.jpg
8fed7d89203fed67a842468902a41ea1
24ef33909f9d3d18af61a36a34a8935f4cc86889
31408 F20110404_AACKYL justice_l_Page_017.pro
48cc2dd608670575c7df7fe0cc899210
5142c3dd700959930f9d0131f2cc4262bb4ee7ff
8573 F20110404_AACKXX justice_l_Page_001.pro
93e35e34b79eb90000215a43727cdb71
dbae09fe1e5852ac326659f01041314c71a66114
7598 F20110404_AACLEG justice_l_Page_032.QC.jpg
d4f91951ed2d9d992585355409289ba9
15d5ea7c16dbf0cc25c546109ece73ea96b000f4
35746 F20110404_AACLDR justice_l_Page_021.QC.jpg
b657ec15ccf08ad75c2dae6e022f1aaa
5897af8604b3b9b9cf7bdc59039b290cff2844bb
50124 F20110404_AACKYM justice_l_Page_018.pro
c0c7174ba1f635a949f7c78acee6cd1c
0f6c309d951790c3cdb12d7695d2994a6f4fab32
1043 F20110404_AACKXY justice_l_Page_002.pro
23bb1789384de05b955bd98b91a38425
6e55d7cdbcdcaf0ad2e6f53fa451f8e4650c58c8
110391 F20110404_AACLEH justice_l_Page_033.jpg
db63e02a9c5f9992de769bb05e4114a0
f957fad601bee3bb56ae9ab9bef299c12a7f3fa7
54521 F20110404_AACKZA justice_l_Page_036.pro
e200af35e01a7d8c0ffe50c7bfdd2987
df1eb6e48258e0a3a983e0dfb068b1ca375af44d
109989 F20110404_AACLDS justice_l_Page_023.jpg
eb15420f281ce4fd5bdf0b7299fd9b11
7e1f99eef0262ea4f1f29403a2e3f6affa0db457
49332 F20110404_AACKYN justice_l_Page_019.pro
eaadd5c024576af9b30be3c60056a00e
e2f827bc6b4f3aa1abfc3e4d481045e10cd027be
96089 F20110404_AACKXZ justice_l_Page_004.pro
b8627d94a5216c595574a058d3f3b9c1
f621050884e946c93542bdd88c8bed7cf869f2de
55154 F20110404_AACKZB justice_l_Page_037.pro
0b2866d63ddda002df5a3de74e7c041e
02c80ceae3cfd9bec979b2746c2f96f8f8e5570d
116211 F20110404_AACLDT justice_l_Page_024.jpg
ac49367b1aa4d77010bf7f2bbbd4a45f
783ccaf9dda807d67c412ad270b48981046943d5
53650 F20110404_AACKYO justice_l_Page_021.pro
73e43b99c8bb6a5bbfe287f18d3bd69d
11bb66606b4e487b661185eb48e4b5479d820bce
101185 F20110404_AACLEI justice_l_Page_034.jpg
7de044b68d5b73bf30ace21b54e8db44
b715f73d8ace56d696d0278cdd46d938853c1a46
47298 F20110404_AACKZC justice_l_Page_039.pro
c8607262f3ef3b153a9ef4e7be53b3d7
778172d069c14ef4041805a08bc27fed7b4352a6
38434 F20110404_AACLDU justice_l_Page_024.QC.jpg
90270f1505de2112765a6a4cb3d679fb
5fe707053d5259e356fa12c2215d5a08d4346025
51476 F20110404_AACKYP justice_l_Page_022.pro
1f5a2c805790784cb2d58ef10694fc34
3acff3c48f4c3953c237c86d94ce3da778e67ff6
32491 F20110404_AACLEJ justice_l_Page_035.QC.jpg
5998c442244e2eaf5231eb577829a476
76f0d29a83b1dcff741689a8ca2234ad4b9ff7aa
51700 F20110404_AACKZD justice_l_Page_040.pro
f12f3a00c3d5ba2aa09fca51ef907015
77f85a87d08b0e97cc9ca2a7acf9546f83999f29
98448 F20110404_AACLDV justice_l_Page_025.jpg
03241ac6fa8f470991ba8a3ac27386bb
94b168449e64266b4eb9633504c8608d64e462be
58231 F20110404_AACKYQ justice_l_Page_024.pro
e7bad290d028d3f66c9297edc5874cc1
966f713bcf51816f19661a34144da4e13401ecc2
109661 F20110404_AACLEK justice_l_Page_036.jpg
4cce57106443bd6abd6e2d3e9103a1e2
fde633ce7f9c20d8090fd4c043c5806ee7857f13
53659 F20110404_AACKZE justice_l_Page_041.pro
9b118104f8c982f1d8155d1afcf7a8d4
a602e30d0d875d78a72a9771d41906de55443020
34718 F20110404_AACLDW justice_l_Page_026.QC.jpg
30fd8bdb69e09f47d0284f8a0025a114
1f010ebb1ac85153ea17ac244eda8907f1900d45
47604 F20110404_AACKYR justice_l_Page_025.pro
7c1a1cad915bbdaf845effe23f831193
9d95dd178d442d91c63ebcb5e42430b95eed40af
106777 F20110404_AACLFA justice_l_Page_046.jpg
58e0447577bb9a9f4c6acdf2218100a7
07dfd9c890d16fe0d31fc0b7817aeb0df6327e04
36674 F20110404_AACLEL justice_l_Page_036.QC.jpg
b86cba385508593fae41a406e96af1d5
53bb76cfedbe307fced4d7df81059126fd4e3500
52103 F20110404_AACKZF justice_l_Page_042.pro
26632aac2e964288059723c0a4e8ee26
e049c66e724f8caa804dc296bc4e8dd3899997d0
36535 F20110404_AACLDX justice_l_Page_027.QC.jpg
9386a845444c466b3446f31510fc8b5f
3e76a5173ace0b9783ff80576b89731477c15f7d
58703 F20110404_AACKYS justice_l_Page_026.pro
96bf69b381bac9ce25d7e1ab262d8fae
336b07e94b10edd863a866277717987a1cf628cf
35893 F20110404_AACLFB justice_l_Page_046.QC.jpg
24c1102155d4a25816f07b14f1a13e27
1f7ef6eb93abb742b0e7e0d678c6983c546d9159
34810 F20110404_AACLEM justice_l_Page_037.QC.jpg
e533c482f42f5f7b27db752ba88b93ed
986ae44fcab0e89ce81d9f3f293d77a59e6aeec1
53924 F20110404_AACKZG justice_l_Page_043.pro
fef315f3f6db9a4496e2d98a8299c47a
8d949b9445c6a531882f4ec8d8a68a24f489c91b
36639 F20110404_AACLDY justice_l_Page_028.QC.jpg
8b2105966a507c78c193709c644859a2
89154af087957c3fdc199506700cf3da7a462615
54678 F20110404_AACKYT justice_l_Page_027.pro
158ccee22ba8ef05dc9c87c9a4e27bc4
c792f5b62f4678f4bf20a27144c54f18b92e3285
105125 F20110404_AACLFC justice_l_Page_047.jpg
7b08898b1e921483771216ae73166eba
0c591c304faa7e6da278f1aaf6df8468639cd05e
103205 F20110404_AACLEN justice_l_Page_038.jpg
a4a5bc20cb3a7fb402af2b8702b3e154
31c2f8afc600b92eb0a34a9cbf175ceaf56c7590
50881 F20110404_AACKZH justice_l_Page_044.pro
aeee1caba1b728dee9696abcbb52e736
82e4947d17ca1c1c962123aec904a67eca3c7cfc
109800 F20110404_AACLDZ justice_l_Page_029.jpg
1dacd4d08376c0365192b45efde51b17
3a67f573ee002e864f0bf099d75416ed889b11f0
45771 F20110404_AACKYU justice_l_Page_030.pro
dead4b61b44861640d6774b5d7d069a2
0b965b996615dafbdd7905864c2f13b06f20da2d
34850 F20110404_AACLFD justice_l_Page_047.QC.jpg
54783e9bfa4a1dcf88fc12eea040f7ea
9ecae3dd4323e5b705fb5139a43637620d111ec1
31550 F20110404_AACLEO justice_l_Page_038.QC.jpg
54500ec95e6cfb8bd419ab8d9a41cb86
c086b385316cad19008fb3e9daa8ae7208827ed1
52802 F20110404_AACKZI justice_l_Page_045.pro
7991e5d7fbb76bdc3ff4b71e8e2b9105
02ae64cb75c5dcf477dd00f6ed1a0787208f250e
37173 F20110404_AACKYV justice_l_Page_031.pro
8c3207b8f54fea6bd5e9a0d3c3370dfc
83c464bb911cc2490e30f7eb6c22640c6de16be3
32418 F20110404_AACLFE justice_l_Page_048.QC.jpg
cd4a3542b7f251627a45669505eabe04
92a2b34c06c87654c1c4df39dd6c16bf8a1b2a7f
97381 F20110404_AACLEP justice_l_Page_039.jpg
c3bda1ce9061d1484615a05539d0b21a
2856e1e099db03713641a46080f87b629521f14a
53502 F20110404_AACKZJ justice_l_Page_046.pro
5f6d9af59a088da18147449826a86d3c
e30623221c4528bb56fb0861264ae1ec679743e9
9640 F20110404_AACKYW justice_l_Page_032.pro
ad354af29be71c7e577087c887276e10
7a7ad5d61ba27c10357b6e99d54aea53c4345bfe
112372 F20110404_AACLFF justice_l_Page_049.jpg
ea551d506d3649394dc85047ba94f0bd
5c535ee5fc7873e0dfd48251c0cee0dd5473af1e
32218 F20110404_AACLEQ justice_l_Page_039.QC.jpg
fabfd5a2c194a7bb172dd9d99365b5b1
7145e994f442ea2690763dd887c528afe9c3bd25
52078 F20110404_AACKZK justice_l_Page_047.pro
43c8f1ff188552bd1bafe3ea80f8be03
d5ed2b96e4be634ea5446932f913d91afa39584b
54072 F20110404_AACKYX justice_l_Page_033.pro
c17f73b46c9821e2c086f84edea6e83b
6037042fe78b12fc195c506e7c418699f13489ef
34716 F20110404_AACLFG justice_l_Page_049.QC.jpg
c44114fd4be883ccac2e8b47ccbe6a05
58148cd49ca5594abbbdbbee8fc8e0b4552ea428
107136 F20110404_AACLER justice_l_Page_040.jpg
f07fb460b2b38b677f9f631eade2d5db
c50fac8b11d828541e0a7abe9df310ea00e2cda2
31070 F20110404_AACKZL justice_l_Page_048.pro
e7ddcd4b2bd5dbb58f33b28f07fb371b
da6fe6be19e4be34267b19cd96e33bce7c3c879c
50041 F20110404_AACKYY justice_l_Page_034.pro
7d19727659405a75642fb70fa13c68e6
fbb9468a42442283ef31a90174e543b52cbd4dae
107332 F20110404_AACLFH justice_l_Page_050.jpg
ac2ca479dcf95976d5eb0103e0a295b4
01ff4f9141846de8280d97112c55fa92ba688b75
108577 F20110404_AACLES justice_l_Page_041.jpg
66db8ed8208fae45d7058c45fadc9893
046cb7251959501b1c7b1dc776b66effe3f2597a
60565 F20110404_AACKZM justice_l_Page_049.pro
80b48a635e33a0138758ec2dbfa0ace9
524ef5843aa4e919d9ebb1f34a98d504af682ac4
50607 F20110404_AACKYZ justice_l_Page_035.pro
f00c78c334bce7615e67aa13c7fcbc2f
cbc420dfc52ec9e3c40235ee3f251145c945b499
35028 F20110404_AACLFI justice_l_Page_050.QC.jpg
eec203a9348c700e4f5f4862f12443c5
12b5313cc475e11544743e2aff1913a99f7583f7
36526 F20110404_AACLET justice_l_Page_041.QC.jpg
a75389123d472adac2785cc102a55a25
0f1035647db2ad494f610b8c7119c1b5a3df89b7
53283 F20110404_AACKZN justice_l_Page_050.pro
a351ce8e07d1730ececc19f0f04c5453
5535975767aca2f1c4d764f110bec8847f54a3e2
106570 F20110404_AACLEU justice_l_Page_042.jpg
94d23ecc03646f1fbe9d4115b27e82ed
6a70cc2329ba70a662aff5b2cda6b6ae65bc2050
37067 F20110404_AACKZO justice_l_Page_051.pro
8ef915103aeb44271ba34e22200204ef
78bc8ca62018956d4a364d55d09c9869d0a01a04
86054 F20110404_AACLFJ justice_l_Page_051.jpg
4b410fb4cb40f9b01581efd2d5d2b4ed
ee2cd34fc27eacade7de25879a5f22e7fed2d208
109044 F20110404_AACLEV justice_l_Page_043.jpg
a3470c836399c62dae102f5f128ed7ae
40e38733b79bfceef99cdeaf51a8a206f6b89572
40480 F20110404_AACKZP justice_l_Page_052.pro
8edc61dbd2fb38fd99d1116850e1a754
45eef33ac01a1ebc071f56a5d5ebec625fe7ed82
28989 F20110404_AACLFK justice_l_Page_051.QC.jpg
a2ed2baf20e6199c8b7237905e2de131
5e67b8677524c96fbb8cc199ae354e0e4f18e51b
103963 F20110404_AACLEW justice_l_Page_044.jpg
f3e88dae5dedc98e0e5835e42034590f
dd7694bea7c0da1af00c687b48372ab8d0062f7c
8607 F20110404_AACKZQ justice_l_Page_053.pro
ad50cf2beb1466e0ff7c6e15faa3e9f3
72edd15bb76acd76f03d9f4287a0552a55d3f4e7
75851 F20110404_AACLFL justice_l_Page_052.jpg
9a793b495b4524cfc928a91e853bde02
0757bf884d8575cce6b58cdd4d6540da2d279216
33774 F20110404_AACLEX justice_l_Page_044.QC.jpg
5e40d217304bf354a8bd7c41a68e6a23
2ced155b0b78b18298f443f7d0787876867db5ea
49952 F20110404_AACKZR justice_l_Page_054.pro
1822eba1e859217ca860faa65295cfd9
6e1eda1d933f7260481a2fe092c067096c733dc1
103163 F20110404_AACLGA justice_l_Page_062.jpg
c2064e51f659ec123ccd55c7480fde51
5c2e8e13641b404d96b1b70015742753b22743f4
24626 F20110404_AACLFM justice_l_Page_052.QC.jpg
28afb4677140c582a5ac8d284ed44573
796cf42737a7bae7cdd06c0517eb0d3025829a7c
108179 F20110404_AACLEY justice_l_Page_045.jpg
75086c0fe5fb7f9b9d6f7335eb64aca0
c38c68a5233c99974d9f720b2b58c8c57d1ccced
49212 F20110404_AACKZS justice_l_Page_055.pro
31bdf72005c35525cb6df19fdbd7498c
f8fadb930e18b59086a42221c0ccd7774d5f92b6
34290 F20110404_AACLGB justice_l_Page_062.QC.jpg
4c8bbf7c798e9f2205fa975acc5e4a21
b05b858a6e5c001f5b3466816a8a126c7a20978b
100928 F20110404_AACLFN justice_l_Page_053.jpg
705aa311da20d3956f7e8e996a82c8aa
75952ad83861d4c11ddc8ca8612bfd058f0c855b
35670 F20110404_AACLEZ justice_l_Page_045.QC.jpg
5c9f5e8bfe0be5939d4f367e24fc7d84
700afa78d85eb07118799988cfece243b16ede6b
33726 F20110404_AACKZT justice_l_Page_056.pro
eeb7297439dcdbf68a6b0ded5f9cd730
49162c340d2de9dbaab04e42587657d66327f36e
103552 F20110404_AACLGC justice_l_Page_063.jpg
d1d977f6bb29cb0664d759399e4611da
2d1b15d216999bd831febd5b08baeae8946ac873
29837 F20110404_AACLFO justice_l_Page_053.QC.jpg
847c35dd103c27c70033260f99c45e20
f8c273afd06803b336f41ae9aafb78a4d2945e63
40155 F20110404_AACKZU justice_l_Page_057.pro
a200b7126f7de5133264065e355d6606
9eb25d5c73fb7319574a38505d89d058bca425f7
33270 F20110404_AACLGD justice_l_Page_063.QC.jpg
789a24eee47688ac20093e411d75a769
9f2e19ef38019815dd14a73f2ad7f1d0746a325e
101359 F20110404_AACLFP justice_l_Page_054.jpg
504f7aacc7bcc7f4c36f2e7135a1b630
7dc215e0e0d9e3c6cff53f0e1fb13c1ebec1aff9
53668 F20110404_AACKZV justice_l_Page_059.pro
662e973e11b0d0222573adad923ab09d
0e2a75e33004bcf404df152888ad8d217458be05
113366 F20110404_AACLGE justice_l_Page_064.jpg
c1153becc053d0dfeb791c4d3876dcc5
e6c38bd8866866dd76f2bbbc6cf6ebdf8480a48a
33946 F20110404_AACLFQ justice_l_Page_055.QC.jpg
d63e9f853e4a77f2a8af076291032ac9
eef655bfc0a5ab108ed80c79694be6fc071dba53
53772 F20110404_AACKZW justice_l_Page_061.pro
9fae057e4f41baafe81018b15ac64175
2f0b73508017a1e157ae54bf2342bae1a62a810b
36857 F20110404_AACLGF justice_l_Page_064.QC.jpg
81f40643c6176875719792ce8d9ecc5f
e21efc7a01013d0e4a088df8ba09c013e71b0c16
70002 F20110404_AACLFR justice_l_Page_056.jpg
40281feebcd4c10bc011011c17bbec60
6175ec9ebe73cd3d88dc3e2dcdda426864f22a18
52842 F20110404_AACKZX justice_l_Page_062.pro
136ba157f7af12624f291cdb6a6274c0
e20b1a9576ca675bb0a1119263408b4c40831ba6
111412 F20110404_AACLGG justice_l_Page_065.jpg
f5070289e24e4232d8dcf44bc35ed9d7
0c046ed62bc1a0abb96c7202675da5049e5c404b
97199 F20110404_AACLFS justice_l_Page_057.jpg
c8c186a19cac3c6008b714e5f76e9c06
3c7558660a7a2b9d4aeb30fedbc0a694cffd073c
55429 F20110404_AACKZY justice_l_Page_064.pro
2bc0941547de6440fcc2e8359327cd2e
2cd78d21bdc6730e62f124e9bcdebc67c5e55933
106849 F20110404_AACLGH justice_l_Page_066.jpg
065ddf69b20749078a2524e289d6e529
c0dbed6c9dc815092f245a45482514dd1259133e
87980 F20110404_AACLFT justice_l_Page_058.jpg
e820132826ae98ec642e615a0ed9e8a1
339c3cbab1947982f0cfa3d1b8d7078a7a0e7f32
54170 F20110404_AACKZZ justice_l_Page_065.pro
ae2448cdcff359698c2c6f163f4c260e
8b46b9b0d0a4bf8c3d0038787ca45b7ea0e91b8f
34805 F20110404_AACLGI justice_l_Page_066.QC.jpg
4f0703244bbe6c0af6ca240af40543dd
95ec0660bdc8a38be585160ca73b05139e3f0e2d
109390 F20110404_AACLFU justice_l_Page_059.jpg
9e4652977cf8fe3324c13e906ae51d29
414a5d42fc4f05a1f733c32162cb6f0949b678b9
34198 F20110404_AACLGJ justice_l_Page_067.QC.jpg
7859c6645a728c61ec19c497788b3991
9a5cd4c4752995c9e53ed7ed27d38fcd535ab3d1
36372 F20110404_AACLFV justice_l_Page_059.QC.jpg
7130d441e0efa2a305205ef64a4d3321
e9a6bfefbc99ff9c9112d984543b74fec8b33c32
125100 F20110404_AACLFW justice_l_Page_060.jpg
00a78f6f0603ec5ebb87a9639b6b4d54
416333746aba262a7efc00459f19d6b3ec63043f
110943 F20110404_AACLGK justice_l_Page_068.jpg
33f98c140513fa61de1a6ed255b5085b
4a09d2a9e257ffb4f98b4e930454fe4a62bde6b6
39219 F20110404_AACLFX justice_l_Page_060.QC.jpg
3239509626760f18a5f129fbc9b637a9
4bb5df892110398e503f37780a1aa4f942d085fd
23586 F20110404_AACLHA justice_l_Page_076.QC.jpg
0f52d62c6409bd7fa7af35432b6af906
87e1199a6cf251f2daf0157062afc9bd174ce80a
36226 F20110404_AACLGL justice_l_Page_068.QC.jpg
40368b0ed925d9eed23e84c8ee8ce4f5
30ea5e784d3c9fe287071a32368fbf905a57ef20
106526 F20110404_AACLFY justice_l_Page_061.jpg
5e901550e37554e2626089a452301733
b2395bc0a409e6650f3ec6bd1871ef8c4645b991
109635 F20110404_AACLHB justice_l_Page_077.jpg
eff349acb38c5c5cb8fbe82688ef0047
f3d47237b59b0f8031518c819643d13c2b4354de
120125 F20110404_AACLGM justice_l_Page_069.jpg
0896a66bded112d8aefb6482106ce55c
f92708c09acccbaba78ae262dac65570383158bc
35193 F20110404_AACLFZ justice_l_Page_061.QC.jpg
44d58866a7ed0f1903e2fcca3496b9cd
4d61e7f554c722e723e225abe651ffd31c84003a
34787 F20110404_AACLHC justice_l_Page_077.QC.jpg
86b3d4ef5fbc665eecbc4786faad3d33
50399a374e8072d68fa1eaf6afcf62b03e6168ac
36481 F20110404_AACLGN justice_l_Page_069.QC.jpg
01947bf64e5a54bf898be24ede754069
8ce93c8cb8103bdbfe6838e981823d0ba1eae4b3
91965 F20110404_AACLHD justice_l_Page_078.jpg
e2fae470d2015484f7f229490f8e40ff
ffb9d9e5ce658f8efa6f7f799b14c370da809708
50344 F20110404_AACLGO justice_l_Page_070.jpg
d25cb24d25412780c493df3b13b4116e
f011cf044e7364cc418ac8cc9bcc410684645fb7
29302 F20110404_AACLHE justice_l_Page_078.QC.jpg
412590ca81054be926f1d26b2538ade1
946d809d070ae5137eb998a76951ae57fc39b830
16741 F20110404_AACLGP justice_l_Page_070.QC.jpg
8c2954d9a1cdf41ab8c840e228a09011
79c5f2667c26e7bd4a189fee46a60d90fb758853
90439 F20110404_AACLHF justice_l_Page_079.jpg
e5ccc1d6b2f584f3d109cd2827b527f3
140bc3e2516658a9a15cc48108399c56feb4cc4a
106612 F20110404_AACLGQ justice_l_Page_071.jpg
0568d1b9322397f36a5aef42c5645ecb
f86347dde19d5be5de7049953b515012919b115a
29893 F20110404_AACLHG justice_l_Page_079.QC.jpg
b4451c14d828ed49b7627cde15a723e0
a94a7cc5e4c3363e2dd6ed9af73b8da6d65e8e40
35412 F20110404_AACLGR justice_l_Page_071.QC.jpg
c69c0692f965cf55a5b80a381cf6d80f
922039ee24fb76f3c891a535365201ed686081f5
77912 F20110404_AACLHH justice_l_Page_080.jpg
e910cd598396439f524f17168548df7b
47c2fd8431c299143d2ecac16854df724169ed1c
90086 F20110404_AACLGS justice_l_Page_072.jpg
a41f5046f403c058af2663f717157292
7484c1b5e291505fee79a52fe56f953853bd9f55
25460 F20110404_AACLHI justice_l_Page_080.QC.jpg
12d351c84be465606d057496336e4669
f3f9de10de167b8e36f938dd261f3a0f1e13e0f0
29842 F20110404_AACLGT justice_l_Page_072.QC.jpg
ab7b020ea3b1fb19cb9d09517bf35489
a36f882f8c1347253dd3f7f1ebce6a24fe954b5b
31382 F20110404_AACLHJ justice_l_Page_081.QC.jpg
f4d16b11e0daf931b022c69e7450ec7d
61b402d6b8c39bec6c9e569cade72768984ce3bf
31936 F20110404_AACLGU justice_l_Page_073.QC.jpg
d49b723d0e9b64b7ea7cdc8b50cc6f17
f0ace95042342358d3b959076fedc1fc7ae768dd
105599 F20110404_AACLHK justice_l_Page_082.jpg
f5c6bb39fbbcd36c7686edd3b1dd0fe6
9bdb022a0531eb55d92ac54f53171608d5e2b8c2
110749 F20110404_AACLGV justice_l_Page_074.jpg
92dbcdef9152bb1ad28a86840cc8a2d7
c8e545ee5c094bff063da3958ef23073a803ff8f
36352 F20110404_AACLGW justice_l_Page_074.QC.jpg
c1eea4323a5db3ce65064357703ff801
085df29936e8fc56fdbc819283e3b2b2917eff6f
34314 F20110404_AACLHL justice_l_Page_082.QC.jpg
d924e3d8877d03b570a03b70d1eda9f6
c9362360e71b6c58bdbcf0a364edc950c346b762
74876 F20110404_AACLGX justice_l_Page_075.jpg
9a36e3493e4a22aa0e0c9dd01b8ba8a7
8417332ef69506d4847609f69e36d91fbe087584
49084 F20110404_AACLIA justice_l_Page_090.jpg
c2ceac7a643a5f0ed8af5db0a4b2dbe5
66b1a05b5d6096bc0ed50491e455c5a2ba921c5c
47144 F20110404_AACLHM justice_l_Page_083.jpg
2e1e8d827483c41764c21c751c315549
1fa52bd9239fb3317d6f0c61b28b308df6766247
23977 F20110404_AACLGY justice_l_Page_075.QC.jpg
69678cbfb825705b0c1f71db1509fba0
6f2bf477dd3b4ef96d54d35519c25ee7af3054ce
16253 F20110404_AACLIB justice_l_Page_090.QC.jpg
106023403c28f542342750abd87e1045
9c153f411f968bf9c7e09c4b57d4ee413e4dead6
16555 F20110404_AACLHN justice_l_Page_083.QC.jpg
6081c8b255d84aeccdf804559046fff4
79c8ebb9e1323a58c8f87e6fd5b894cdebf82a7b
70111 F20110404_AACLGZ justice_l_Page_076.jpg
c5669fbdd09bc3708a050fb4aa3ecf2f
0cd658a919f4728bab3503a5150ef6448a70d458
63812 F20110404_AACLIC justice_l_Page_091.jpg
cb72e144bde3aa4bef56efa40e0c5d42
e537467a1218e08feabfa3223a7f99a6ebd70fdd
52372 F20110404_AACLHO justice_l_Page_084.jpg
4e36ac8520744e3d5c658d944e1500c2
6ea64b0fe7801344f57f5077bc939edb4270a18d
21113 F20110404_AACLID justice_l_Page_091.QC.jpg
89895bff2b2d15ff26aa97ed1365fd5a
2d08a570588d164db8f57e64da1eedef36035e84
17866 F20110404_AACLHP justice_l_Page_084.QC.jpg
3db9c8d013225f09a7d7b44e8cbc2181
c025e4b38bd2aeafcec83637d12878ab5dead2c9
49046 F20110404_AACLIE justice_l_Page_093.jpg
0587ec67d0da6cdcda397f5b79b90866
0a1c09ce116865a54d3164579fed36af22b3802e
48607 F20110404_AACLHQ justice_l_Page_085.jpg
071b6c1f420efba4a1bf51561a7c40eb
f96617dbaa2f5b7f6ee9b7c7b88404f5a2d18564
16420 F20110404_AACLIF justice_l_Page_093.QC.jpg
3cf0b6c94027909b5a359348a5f82acd
ec482417359f575dc91a81d7f280fcb22e193ad9
15645 F20110404_AACLHR justice_l_Page_085.QC.jpg
8b3a6e5dded726540e8a11d8652089fc
553020da9de0d37cfc2997f621526f78c28f5296
47701 F20110404_AACLIG justice_l_Page_094.jpg
37dfb6439ca747768855a5173e5e1c0a
d51fec3acc031b78adf467fe7ee0ab72c0cde3f4
55838 F20110404_AACLHS justice_l_Page_086.jpg
db7fb49c208fb94a0dd6722947b5696a
5c65f61df07190f7d06cb21a03a894121b2b2ac3
15696 F20110404_AACLIH justice_l_Page_094.QC.jpg
5a42327cb0195f30ef34f0535b6ecca3
ff3aef62b167196480533d16fe49b6c08917dd3d
18280 F20110404_AACLHT justice_l_Page_086.QC.jpg
d4ca69cfe172a9cf1a63f41f689cd8b6
11967bef19eb957ab76c1459fb95280a7a560e4e
29218 F20110404_AACLII justice_l_Page_095.QC.jpg
9adab8d03f71c8b9f710d341ce2cc82f
6e50d5737ecacc18ce6c547990686462817fc036
52132 F20110404_AACLHU justice_l_Page_087.jpg
e618401cee3d60a6311298effa95d63a
048fa80f829300214af39d6fcce2687c1427a3ee
106448 F20110404_AACLIJ justice_l_Page_096.jpg
02a14206777bcae8f2c400074c4210d8
f82574eac6a5ea97af6ed581d557fb4f0e03c86c
17938 F20110404_AACLHV justice_l_Page_087.QC.jpg
e2aabaf13a2716a5b582caf8de34100e
df60f54f4c649e7d5cd93749c9b092bc8a9809fe
35190 F20110404_AACLIK justice_l_Page_096.QC.jpg
af6155ecac671ff6671820315e41e427
006ca40249e3863c0d783d85ffd985c2cfc2b60c
73779 F20110404_AACLHW justice_l_Page_088.jpg
7aba707dc812323a0b7b47b5a124a542
ce61d4108240e83222c27b890428bb6355a957d7
53811 F20110404_AACLIL justice_l_Page_097.jpg
4f127a372cf9a8977b43935e1783ecc1
5d49ab955388e83f5666c31fd06f28098d3fd718
23865 F20110404_AACLHX justice_l_Page_088.QC.jpg
19239145a245e951fa386169cbc37590
4e29fae24794acd7504ca3fd5158f1168b2bb605
58809 F20110404_AACLJA justice_l_Page_105.jpg
29f70cc71d5db0641854d917561eb390
abc13d3e5688e1f2d4f217717959da8ba72b8596
70398 F20110404_AACLHY justice_l_Page_089.jpg
af53f16e0b8afbd187a3e60516264e5f
4ff9433212efa67bf19821460d1105437d42b09d
19940 F20110404_AACLJB justice_l_Page_105.QC.jpg
52e9bea56d349989ca19f2a9447674e5
1c11d2a7f8b890850078ff076725fd7951fa0667
62404 F20110404_AACLIM justice_l_Page_098.jpg
a0e22a38cbc9ee5fccd35800a7824be9
928cdb83de883c4a93b67b6205431893f927c104
21710 F20110404_AACLHZ justice_l_Page_089.QC.jpg
7bca450c999a77ca25a37a2e2ab150eb
0d2e5cfad513d5a1a04be989094c35b7ab42eaf0
68066 F20110404_AACLJC justice_l_Page_106.jpg
ed35734c1b35681a5d54eda7d0d23c37
6d3bf034a23802af7c9e378aabbbc7599b74e84c
21124 F20110404_AACLIN justice_l_Page_098.QC.jpg
6ed049fed1e251c4c571923f2128b836
fe46d470ee8c5fc21e1afcdf07199132c702c29f
84977 F20110404_AACLJD justice_l_Page_107.jpg
25d25037f96eed424e9c1c94404cadcb
4eb8eab23228120e02d4d39d21314b5720dea3cc
49151 F20110404_AACLIO justice_l_Page_099.jpg
6e421dd0a53162cea9b8e05d36d382c9
397bec90331c2aba32ff8ba24351f2dde681c19b
28274 F20110404_AACLJE justice_l_Page_107.QC.jpg
3a2fb739632beee14756fcbd000ce9c0
8d5884ea12adb8d7b79d1bb54db93f1df4c88b8d
16706 F20110404_AACLIP justice_l_Page_099.QC.jpg
c62a8571286595a5142e2ca8fd81ef11
f9a4baf85a1f945c65b34ddd93ccf1116d8dec92
101910 F20110404_AACLJF justice_l_Page_108.jpg
5c9e8da8ceeaa3b9aa0acaeec7fd1ea0
477a67968d82ccb65e94cd504491911e7dd27b3d
44589 F20110404_AACLIQ justice_l_Page_100.jpg
87e0ce751ddbc06763d5449bb685c76a
9a5f07855abfab9a8adb3dc9d5b031d4737bae70
33332 F20110404_AACLJG justice_l_Page_108.QC.jpg
1f83c57facc60bf586350e6753f08e2e
3a55c45f9195a7060970ca306ee06a42974d9c89
15371 F20110404_AACLIR justice_l_Page_100.QC.jpg
e121f36b93b0a95e283dd20d30352732
bb9930ce004bfde10d7326a60dd1cb38fe6ac9c4
82634 F20110404_AACLJH justice_l_Page_109.jpg
147b1a9f43e8ea556dbd292c2efb5ec3
9729ce552e7866242d7ebf4de166eb957a48741c
74567 F20110404_AACLIS justice_l_Page_101.jpg
f45895c186bf137d5a584d968332d569
d776c4a44655496061775f596d68c8ee51d2b73d
26162 F20110404_AACLJI justice_l_Page_109.QC.jpg
12780c1697768c3123fb629ee88eb9fb
418fb63dca2ffcbe044acfbd1762b7cc95b4ca99
24332 F20110404_AACLIT justice_l_Page_101.QC.jpg
9838185c4d69d40faefd3a0c6fdf61a1
a7d6d4004ae4565bb6d44a61ad277dcb708a6b3e
99973 F20110404_AACLJJ justice_l_Page_110.jpg
523b5c1958618deec159a08f1bc0fd7d
e3bd2eb3ac480449be500377df38ae8000029b77
44929 F20110404_AACLIU justice_l_Page_102.jpg
7b92772a2f23e0e50bc2ca204c880927
9236e0984f356d5f75a223a6dcb04c4fb1d6fa91
31377 F20110404_AACLJK justice_l_Page_110.QC.jpg
ee6e3736dcbd059d8470e951dd39d716
aaf1b292c50adf94a4db54659a8adfb0e06f3148
14700 F20110404_AACLIV justice_l_Page_102.QC.jpg
4abf8506d9cb4b7c81e6effdc66d962a
3a155ebafce8d08cf5f0983d76358f8cbf078593
90906 F20110404_AACLJL justice_l_Page_111.jpg
043cf3c7d2ec80360a56639232931e59
fd2cf5f3617bd634544c78f630eb54b482a02001
77184 F20110404_AACLIW justice_l_Page_103.jpg
2dfba246a832a714bcee1e77687fda88
155ea48ee2172a6d3de185fb64c21e6271cf3f57
9087 F20110404_AACLKA justice_l_Page_120.QC.jpg
04d6b3f2b85b13600102e15b98dd48a2
1967fa6c2f98ce0e3fc8bc5434cb014004a40edb
28746 F20110404_AACLJM justice_l_Page_111.QC.jpg
88256333f003ae02f6ee0fa4188f2e92
760f40fa2dfcc294663fb410f6aaeb2c39857f4c
24754 F20110404_AACLIX justice_l_Page_103.QC.jpg
8ac2c581cabaa2dee0dbd7747a574955
bfe5f5f24a13c56773013ff95b8f727553647acc
99892 F20110404_AACLKB justice_l_Page_121.jpg
306b6514e16f43ca752f3d64d994aeb8
d5b7514d542abd06ff3aa075a7677a4bcf1e7ea3
52589 F20110404_AACLIY justice_l_Page_104.jpg
2dadf320ad88798ff3c55f63b19ef8c9
56b8a5d94a4ad7fbcfe2635fe3546fc8d37ad8ec
31999 F20110404_AACLKC justice_l_Page_121.QC.jpg
02918daeaf98ee393a0b09e90d445a33
b236c5df874eaf40baa9695a590ac2c8f9bdc649
100734 F20110404_AACLJN justice_l_Page_113.jpg
c19452276a4bc08f370befd744a3addb
8e48262803267ea569429f1376930d42524a3a4e
19409 F20110404_AACLIZ justice_l_Page_104.QC.jpg
03adc142dde7b76b1ddffc808516605f
a1d036679330ab10d124c9d80bbbb19069e1f0a2
105364 F20110404_AACLKD justice_l_Page_122.jpg
1cff09c8a6cb508324fafe80f4bdebec
62f9f621cd86bd7d403e848a5bae424f032d4a02
100241 F20110404_AACLJO justice_l_Page_114.jpg
325c3086bafe2d6389136ed9ef99aee7
80f530363c8d6cb9f369b50081377a528a045652
113051 F20110404_AACLKE justice_l_Page_123.jpg
b58719b4a5c63f844adbe98508bef398
6eb117e7f5c17c5de93ad30d2cb003b40bd23273
31342 F20110404_AACLJP justice_l_Page_114.QC.jpg
c1c988487ae14ca14d2e67dfe4d00434
04568551d5323cf7c0623d8b04f7eae0283f8b5f
36671 F20110404_AACLKF justice_l_Page_123.QC.jpg
3da9deaa595fc72fd0e5cc529447728e
0f530c5004f3f0e639b40a0f9c50c76fd8289946
91113 F20110404_AACLJQ justice_l_Page_115.jpg
91b8c286d52e96717f0f92cd87232897
e77d549273e6ce1f436f74a54425ba64943d940a
71581 F20110404_AACLKG justice_l_Page_124.jpg
6337674d80b17bce8807ef7368e76e56
30c6edc96c9e26297a8b5d386aba1cb5ef2a7468
29597 F20110404_AACLJR justice_l_Page_115.QC.jpg
de3aecbff116d34fd902c656ba2a0e73
fc00a364deb2fe536fa30f48f0a899f74ecaf97b
22355 F20110404_AACLKH justice_l_Page_124.QC.jpg
fbd5fcef73b7ec0f50d5fb747d85b3b5
ad8c731e3032164f60d68b48865db6b44abea627
91810 F20110404_AACLJS justice_l_Page_116.jpg
801752a8ed4ce18a37cdbcbb4719cae5
bf499a95697ca5d58ba626d7b7c651352ccf60f8
77516 F20110404_AACLKI justice_l_Page_125.jpg
ad02f18f35ed8325cbe802b5a338be23
9ef08016bc25baf4d1e53fe96f4f9de098c21e1d
29475 F20110404_AACLJT justice_l_Page_116.QC.jpg
b1244eaeaafc719a0d779a1a826a30dc
eee332fd16fbec4db03ced2e3c9cd90ebdc67f9e
24821 F20110404_AACLKJ justice_l_Page_125.QC.jpg
f43368df18976b2507cad9d65b974f01
5627510e83570ef5b5504ed12c0e3090840334e4
89484 F20110404_AACLJU justice_l_Page_117.jpg
f9fd8fea4977c6f57b9398e7c934cf80
bdba3b367416fcc3f92390f8b44a74ac3bdc426a
82203 F20110404_AACLKK justice_l_Page_126.jpg
35bace94e517fd5d28b8c5126d7c2747
ed6b4a5852ab3f14cca34b86b5355c1b730d94e4
29822 F20110404_AACLJV justice_l_Page_117.QC.jpg
c69ea7f7407c76aa59f746128749ef17
b34c84bfc74aac05c23689ed3e4eb21330020ed7
26828 F20110404_AACLKL justice_l_Page_126.QC.jpg
fb174e7ffbd45ae56cb5631d0634339c
f3e5b1c21cc3895364129f65647e53cd2ac73749
22926 F20110404_AACLJW justice_l_Page_118.QC.jpg
1fe7fc0074648de997a0e6dd9f15eac5
0b7bad5052b99596c29e53e480407cf0ac3d3d92
52017 F20110404_AACLKM justice_l_Page_127.jpg
a656bf98e81e08ce360e9488f1439ad0
9f95b34f654e2dc03eb94ff06340fbf77132b4bf
60264 F20110404_AACLJX justice_l_Page_119.jpg
ac5ca02157ab0b8d4e8bbc552b376bec
fab1d6d2b550da07470675776fac597ddc4539e1
35502 F20110404_AACLLA justice_l_Page_134.QC.jpg
20062cca1823c89d721d67b23ab16891
28ddf1acbcaac5c359e3d67ae699756bda2f8ca1
18209 F20110404_AACLKN justice_l_Page_127.QC.jpg
4a3ba8dd751c2f67ad09a36236dfc346
a7b9e657084a1e14686fe019ddbdb6b40b2a7738
19904 F20110404_AACLJY justice_l_Page_119.QC.jpg
e2b50e6e5ab479218aac5e60d809ee56
6cc7731ebfc03562f7aa54a3a4878683d6b8b526
120709 F20110404_AACLLB justice_l_Page_135.jpg
d91bcf2d6acbb839ea5582cfc3f54f06
d3cf319ea1ad6eb36c5a605345ca8252a1e87350
27416 F20110404_AACLJZ justice_l_Page_120.jpg
8a8453d6490d9a8d40a7bfd644a6e6a1
a6979c064bfbc4ca056018925f8b31024e5c0cf3
35561 F20110404_AACLLC justice_l_Page_135.QC.jpg
daa5e2fc111d9290243be3fc6c179152
d54b834584fa519818503e6368fe4bd9aa377d18
94383 F20110404_AACLKO justice_l_Page_128.jpg
0bfc693cdc5253de6a5646f9fa6bc5ab
cb25fae39eefa442cb8030ea6d0d5180036a54fc
114741 F20110404_AACLLD justice_l_Page_136.jpg
2198d1b5c172b0003763110542b4e354
e9374bde626d1d3312eb90d08816f60a0b51a497
29566 F20110404_AACLKP justice_l_Page_128.QC.jpg
c9aff95f37924ec5e426ebcb86e7b257
0902522a408b21031d6fc4af3d7d9881911953a7
34643 F20110404_AACLLE justice_l_Page_136.QC.jpg
45f64a79ee25cd7832e547c50333afb8
501978ea36750b298a432beec137cf375b772d68
39029 F20110404_AACLKQ justice_l_Page_129.jpg
ea14973d2b9d91de094b9693a999e6c6
5fb42eadf0bf2790e99bb4c8b6d567e919384ef0
135890 F20110404_AACLLF justice_l_Page_137.jpg
e36e8f695b1ffb3a0b9c3436433d4ba5
d1bc72a80e686078667ef83717bbbdf605609da7
13582 F20110404_AACLKR justice_l_Page_129.QC.jpg
953493b9e12fd804e466c6d3c55c1063
610527955866c85d4daed6ee3f33f765bcf5b4ed
37944 F20110404_AACLLG justice_l_Page_137.QC.jpg
da53a7320f02685594e78861d4f0b3ed
94e1171d78451e5d34bdaa30df97ee960db29ecf
106399 F20110404_AACLKS justice_l_Page_130.jpg
5463d52d7eff205498b2ca95779e12eb
d35c29dc34e9b50c094b8df9d4478f1e7a3c1319
34934 F20110404_AACLLH justice_l_Page_138.QC.jpg
3f2e549406e096780c97e81499f8a524
5c5f7c0c70750d560679e9f0768a9598b0ab2b9e
34766 F20110404_AACLKT justice_l_Page_130.QC.jpg
bd627a0e3585acf328549ce68bed8395
88739ac074cfef3a7a1a43d807180236533ae447
134831 F20110404_AACLLI justice_l_Page_139.jpg
c4f95fa4a3b6f0b04dc56cb8a4e7456b
592bb2f9244655a9c3b15c9954f19194d3ac592b
112225 F20110404_AACLKU justice_l_Page_131.jpg
3bea1a3eb41f7e46df4be7fb2870c21b
93ea3a759adcdcdcb5c41d6c36d78092c113cf5f
38700 F20110404_AACLLJ justice_l_Page_139.QC.jpg
f5ea4eeb5d27a638dc85f58b8592a582
98ac60c62f183646a03cccebd47bfbba396a5803
36449 F20110404_AACLKV justice_l_Page_131.QC.jpg
1b29b1463f2ca51e1cc00c6ece85441c
6255b18a71d7860f52c0a2d3005141d1608a54ff
116192 F20110404_AACLLK justice_l_Page_140.jpg
4c84c938ca46bdd881084544af55a300
c879a13a2632de0ce449e6ac16c52f22ede12894
34333 F20110404_AACLKW justice_l_Page_132.jpg
a1ef5fd36bac7f5d74ddc69760cd156d
11672e1cdab14011d000f85b365e62d46b7d3a4c
34706 F20110404_AACLLL justice_l_Page_140.QC.jpg
a3eb5ea83c4b63996607001654ad6dfd
436321f5304c4aeb0a7e98741ec9c51d0bc8bed6
116650 F20110404_AACLMA justice_l_Page_013.jp2
1563628e7f133cdded9897c5798a09e9
f86a30537961cdfaefe17cad0cb268cabd97e424
121718 F20110404_AACLLM justice_l_Page_141.jpg
5141fb6ab7be06dcbde134a9dceee4fd
d9b5621f216d2d02799d0b0454311b2b98c8ed9b
11874 F20110404_AACLKX justice_l_Page_132.QC.jpg
0865125393dff8a584378ec78cde8eb1
8a8afb96f16b2e3dc529c49d49abd383886e2995
1051945 F20110404_AACLMB justice_l_Page_014.jp2
9bdc120f235194b9ad39d3e638560ff4
ec0e76fa6ed731a2670fad94993c3dc4d1da92ca
35422 F20110404_AACLLN justice_l_Page_141.QC.jpg
b3e34427c39b132c5b2cbd4e7883ca13
019b5c7ae3ad43f49f887ef2329397f56b1cca7e
37381 F20110404_AACLKY justice_l_Page_133.QC.jpg
d169965c6ef7b2e3dbd664e9d6ced19a
99960c39b8a7ff874b90c64cd0242b7dfe1ef95a
5029 F20110404_AACKJA justice_l_Page_013.QC.jpg
083b0096a14d68c17f7a1797dba55481
0e326b63db74570c6af0aed7dd019e61893f4516
1051983 F20110404_AACLMC justice_l_Page_015.jp2
86ff3904129c30ce7aca16b84e9ea4e2
d23977fda5d96e8404cc92c6ea789322c14edd55
95571 F20110404_AACLLO justice_l_Page_142.jpg
d88cc2a5b4ec973f176120b1d52abe0a
39c5885e033119a15a98ec954aee46daf35b1f35
116825 F20110404_AACLKZ justice_l_Page_134.jpg
7e04ef991c9c6ba97807ea608dc1feb7
684f15c355ff927da01e7a8f4c9d82bba79d042e
22973 F20110404_AACKJB justice_l_Page_056.QC.jpg
90a61565d5134bf96377839109fcc218
de5590f3c6fe6cc7ff15c5970322c428f6722a2e
698689 F20110404_AACLMD justice_l_Page_017.jp2
3ef32b209afc2782b9a0f5b3aa54c6a4
9d7851632dea16ff37a4ef2b2395c313b3e292b7
1520 F20110404_AACKJC justice_l_Page_031.txt
b92e88de28d4d029b675ded82cfda171
9f698d33afbbba27bf4e3d587949421b968d9668
1051923 F20110404_AACLME justice_l_Page_018.jp2
16b6b2f3a3f85c05751ce27601d59883
a8c9f893d8addc8dbe0b398e675c15071d641ca6
27384 F20110404_AACLLP justice_l_Page_142.QC.jpg
988a3a298935f767cfb5fb30d8dcc563
e76c48702962706ddb0f7d81aafedaf3650a0cdf
54601 F20110404_AACKJD justice_l_Page_023.pro
d8ae13a4b22d47fd35f8eb73d266c8d5
7eba5d713ff2458c64fde377182edbc0e02a78f4
F20110404_AACLMF justice_l_Page_019.jp2
a5c2bf7dfef2c628c2c2948c4c000116
044305d0e7e3cecfbb399d5ec6f2a310af3fa650
27300 F20110404_AACLLQ justice_l_Page_143.jpg
13478c848ba1d80291fe8a6b74740b02
4d7fac60580a3f0200734d8074260d559367fe5c
35791 F20110404_AACKJE justice_l_Page_112.QC.jpg
dbc33fa97ab262ce8eee9a35c9d291b5
168a13366379221ba2eea27d0661e4186b93f984
1051941 F20110404_AACLMG justice_l_Page_020.jp2
17815a8569ed21373dabc71c923d6c42
51064031dee5b680175ea23dd4ba250606ff6431
853 F20110404_AACKIP justice_l_Page_119.txt
e83971ec284e82d6326c259e259987d9
d6457d8dff49d765976e6c7cfbdad13012603a30
255625 F20110404_AACLLR justice_l_Page_001.jp2
01da9b2d572a2102dccea6a38c8bf43a
e755f6eacdd4fbaf54f4a2f4f7661d44d59624c9
8423998 F20110404_AACKJF justice_l_Page_098.tif
b69bf3da079c42c068c6a48ca521b9e3
f164e7696c5462427109ecf58b56d4ef01dc9651
1051986 F20110404_AACLMH justice_l_Page_021.jp2
7668a14b1a4beff5d4db5f4210f7c32f
8848239439ba238c9b60cee13c9c34fcdb56688c
36248 F20110404_AACKIQ justice_l_Page_033.QC.jpg
3cd9cbf8d9e067a01fbee750c59a7773
2d0d7987391084d3d0004674e9635e827b004a52
24534 F20110404_AACLLS justice_l_Page_002.jp2
2e5747a7f57c25b6cd7dd17cb4718a55
db58854c579c6b5d8af63af20aa4edb363d963bf
8177 F20110404_AACKJG justice_l_Page_019thm.jpg
8d63f3f7873378321e93b30cb389d2a9
ee91c5c73abd58bfb2333b765b40e8a22844997c
1051982 F20110404_AACLMI justice_l_Page_022.jp2
c2b63b7c83daa10b1632a1074f3bd44a
b0d34db944bf6fe056439737bdb6084ad0443a22
F20110404_AACKIR justice_l_Page_001.tif
2306131dd87b70db60627b1f0b25b220
9413297cbde54b9161fa6ce0c485ffd9c2d8ca79
1046100 F20110404_AACLLT justice_l_Page_004.jp2
99702d2f9dc4cc8dcf6665e9af1cbff4
c4876f014cc194a1e5629d74816916c6154ff2b9
F20110404_AACKJH justice_l_Page_096.tif
4dcc744f51dc470d47a65a070a044c57
71bdedc817261899eeec85396dd1ba88dba6cf9f
1051976 F20110404_AACLMJ justice_l_Page_023.jp2
6e058db31e624bd42030679415102b0b
046827aac4e1563a59125e22b1cf2b54a122a4ff
36780 F20110404_AACKIS justice_l_Page_043.QC.jpg
786d3d23a1a21ac56cb67637cb70e508
6c5e22e418c3a1d326e018828676ae549ee1516e
1051977 F20110404_AACLLU justice_l_Page_005.jp2
9031516c71209dc68cc0ea4932064607
70df5013c0a96480176f14081820f34afcd6f135
109476 F20110404_AACKJI justice_l_Page_027.jpg
5593c5b713e1e886952d64b70db25ec9
6b3bea18ae4cd93f85d585abec9365147ad2b25e
1051964 F20110404_AACLMK justice_l_Page_024.jp2
33cd560e63262c9a3cee94a71425fa5f
d32a5654ab45b5a2fc4a3e3667aa4ef785680b15
968682 F20110404_AACKIT justice_l_Page_078.jp2
bfb93adffc87ca313210c51d6b8bf104
1c035a7e3138747150776aef5c38345cb553467d
191756 F20110404_AACLLV justice_l_Page_006.jp2
aeac6c7449f425e4d525944ba6355f69
0eee31cee422107d4b948ef3e738db5f3476677e
1051961 F20110404_AACKJJ justice_l_Page_082.jp2
8a4a39f39004c9e58e673778c2dc7f47
ab1873e361cce482a585261345a16d61a7830223
1051933 F20110404_AACLML justice_l_Page_025.jp2
c4917148f957b0aa52b9f0daf8f62b4b
1f97d470b362fd1c3a281c3a1dd619088d4b1499
52451 F20110404_AACKIU justice_l_Page_020.pro
ae41c2f4b8f41002a4bde4fb4ffbb0e6
46a6a8eba49fb6ee814bc7cc8d784a291e770d47
563745 F20110404_AACLLW justice_l_Page_007.jp2
9b08341868a19de301ebedb2a86facfc
208f93abe9be327b6b5864536d9371f38a715940
1051985 F20110404_AACLNA justice_l_Page_043.jp2
07c3a090db0092a95765019db619f8a5
f611f9ddf315ef28c5511b067cbbec6bf34c363a
F20110404_AACKJK justice_l_Page_139.tif
7f4db9d53c927c9cf136e41fb28c9910
ff84e845566daa12aad543bc0bb9e86445bb330f
1051974 F20110404_AACLMM justice_l_Page_027.jp2
7fd573f58c1d34f573698b22bbc02fa4
a9c1b455e6c2af4926b37f05fcb2d54bc1fea785
2128 F20110404_AACKIV justice_l_Page_122.txt
9fe9bba5a91a67e92de815d1d1f33751
449a2e305175f7c702ccb98ec47a422e7f7171ab
978278 F20110404_AACLLX justice_l_Page_009.jp2
f1425b422eda567713d68fd6c419cf7f
fba61886111f8e6812a972eba0457e56b3ca557e
1051915 F20110404_AACLNB justice_l_Page_044.jp2
d6abf991e47ee06e40fc7d4f26372bd8
91ff5cf4072a5c0b85236086647ab29fbf0396f2
31242 F20110404_AACKJL justice_l_Page_034.QC.jpg
e7fc546609cc1c9385b40bd392649150
cae78b8c5b16ea93859525bb1890e5670c1e27b7
1051970 F20110404_AACLMN justice_l_Page_028.jp2
4b5fb244637c069adc37761a895c63cd
ac55c9da5a341726e1b00d8988226ef3477a12e0
34380 F20110404_AACKIW justice_l_Page_042.QC.jpg
b1a993b55cc0c57c7854afd315cf7306
32597e6c77e137c7320dbac98c7194979abd9d59
109388 F20110404_AACLLY justice_l_Page_011.jp2
932097fe3ebbf8523cbaa3c647688042
5e3701c3a3cf6712ca32e16821647950c46ed6e8
1051967 F20110404_AACLNC justice_l_Page_045.jp2
d6786e5b9f11212645706355d13ef6cb
59955290e75ae3a32dac25bd730d411b32903ad1
8885 F20110404_AACKJM justice_l_Page_015thm.jpg
79bdcbdf22e23ed37e7a86b311bf11af
c5889deb61051cc4eebc69a815d07dd5a892ad6e
1051975 F20110404_AACLMO justice_l_Page_029.jp2
e98ab118aabe0d967010a89f4f9fa551
38e8fc415a83cf0342d39f5e75a99b70abb46f31
103778 F20110404_AACKIX justice_l_Page_035.jpg
ef046a3a0b46614e01f3d7976641c802
741938591d3c4809ffb26e7a870dffa8819008f3
1004485 F20110404_AACLLZ justice_l_Page_012.jp2
3b8fefd10d9788f24eb095085d8cbc3d
0bedea9f8e4e002c5343e2cc6e365477a713f98f
31454 F20110404_AACKKA justice_l_Page_057.QC.jpg
75d9fd2856987062859cfd7a7f093d11
813cf7601787d293f8efebd8f55aa86e4388a9e8
F20110404_AACLND justice_l_Page_046.jp2
798609f0ba3f789bc9c3055305a52e19
ca7c723fca1c31fcb2d33f46b47856065a9464ae
100405 F20110404_AACKJN justice_l_Page_073.jpg
a13862ae555574dcd2c36044519c44b5
0cd4a93651fb2c04f9a24f10fb14a056e0190fc4
1044589 F20110404_AACLMP justice_l_Page_030.jp2
552660d5f6b21be84fd7a7ccf5148443
15e06bd11ece90231697d025d742cfcdf3d281fb
930765 F20110404_AACKIY justice_l_Page_008.jp2
561f908238d537c58871afabf6f2d571
012587b13cc1fc300004f5bf7ca24f18e2ac2ee9
8225 F20110404_AACKKB justice_l_Page_057thm.jpg
51ca9c8d1a5e1686712125bf34909154
2605f863b578deebd8a4b56d0cb85e39e985af1d
1051943 F20110404_AACLNE justice_l_Page_047.jp2
b4ab7819a8d3109b5e6b5bd6c8a3d497
3f3f1283fea4f31e101dbc850c8a45c391b2a82a
22810 F20110404_AACKIZ justice_l_Page_106.QC.jpg
e6494713f8d620e3c10b9da0343b214b
38e64b7954ff564da6c43850352db5b978044a60
23173 F20110404_AACKKC justice_l_Page_097.pro
d6ad6ced281bee765ba92cdd50761fb7
561e7bb0983e636ec72821a53f88081b2f165ffb
F20110404_AACLNF justice_l_Page_048.jp2
dddcf2d1173b616ea4a22cb7f15e1ce3
cf10da053141d01aca099338cbed70a1c2e8c956
607030 F20110404_AACKJO justice_l_Page_098.jp2
e2ef6180db1bfe5bc0f4629edd7dba99
9f19d91821ede3668389a7bf78600280353375ac
1016914 F20110404_AACLMQ justice_l_Page_031.jp2
00ffb0fc26adb56080f49d02376b1e8a
0692c311e420aaf1e0deb6bacde58582ebf6b4be
2248 F20110404_AACKKD justice_l_Page_116.txt
92e53e4116ee41440ff570d853883b40
c94d5d487443f73b71570fff91dfd0263daab84e
1051984 F20110404_AACLNG justice_l_Page_049.jp2
34d83a667aeb3c1b9c74b2ca8ab9c87f
10d257cc08a6aa9b20eaddf1b9481f421223b894
2439 F20110404_AACKJP justice_l_Page_110.txt
cba8c1eebff3a3c3989e295fab99656d
f1fbcd330f8df92a96b2ce37d5473e3597f46c8e
1051978 F20110404_AACLMR justice_l_Page_033.jp2
f3434686c0603be20a45b43e5623cebd
e4d4231684e483d95cc42b642e533fa97323f467
1016919 F20110404_AACKKE justice_l_Page_003.jp2
7fe33bf6dbb4ae1df08295aaff6b408c
2f21ae31593d2d8cc693e249e49b0735a649b93e
F20110404_AACLNH justice_l_Page_050.jp2
0a82d8fcf8506b9ee22172d536d9f161
98d763077c0eb9629f9b6f16fa402d231e9b02d2
15449 F20110404_AACKJQ justice_l_Page_092.pro
47d8835adee2cb76dc71c2512dc79de6
500fb4d8f491443ed2e0353b8b61b162f3df747b
1051979 F20110404_AACLMS justice_l_Page_034.jp2
396dcb49282fc578ad7cdba096e6dabf
b7e0495778ec1c92c5f3fca87b578eabd5953b1f
F20110404_AACKKF justice_l_Page_113.tif
de43b95ccb02f20a1b0ddea479abd798
992fefb6334aee837cb49dc5c11346718e46d6f0
915492 F20110404_AACLNI justice_l_Page_051.jp2
84ed4e11a5c356031ce4b24f75b84797
0d34cd2b40e947dba366c73a7d82cdb9dd8f0d5c
6063 F20110404_AACKJR justice_l_Page_089thm.jpg
3336b96fe4391c4ae08f878ad68b1f3e
ec7639f1e7e96ad2a35e08fdb8431144529eeb00
F20110404_AACLMT justice_l_Page_035.jp2
00f880ca0e6dd175be757f7eb22463fc
d4254ebea848d4a7780c726ac303f1186644128e
F20110404_AACKKG justice_l_Page_075.tif
0b44a2bdc806a883a1bcab4568d2e383
016351dbe37d177f1a0db791525ec4564cf42800
822002 F20110404_AACLNJ justice_l_Page_052.jp2
6b0149f12916e1eddded88a0902dbcb2
86aca2968526de2c3f743941d5808eedeff9cdc0
1051942 F20110404_AACKJS justice_l_Page_113.jp2
74af593fe0218af6bfced9fbebf88b3f
80c093a7fba0fb779faf01ec3485bf9ac4a481bd
F20110404_AACLMU justice_l_Page_036.jp2
07b6228f7e776e288ef183cf30f32966
e6dbdc561fab87e234511b49a2d1b629f24e99c4
29606 F20110404_AACKKH justice_l_Page_058.QC.jpg
470044b1ee5e61805147378dd79ce9c9
f975e43d73dba92d0789c3f3b8a9268c49ca7183
1051733 F20110404_AACLNK justice_l_Page_053.jp2
f85de43a42443108e83e5c563047e6c1
e61afb6a4d8d9f01a5d193badfe157c5c39f68c2
1051973 F20110404_AACLMV justice_l_Page_038.jp2
5cc8a60a15de26a7fcf9734067dc09dc
475a5bb49d29d754d509c18c0280122133679319
8642 F20110404_AACKJT justice_l_Page_062thm.jpg
efb008f87f4333a3a19af70c0add9dd0
494ef8981e7dc8b27279a9854fd6fd5d9b1b25b0
85521 F20110404_AACKKI justice_l_Page_095.jpg
24f7c757df8f808f9c446f29a8c32629
5952fe1d91143e5fee08f0e0a1067095fe343e9d
1051934 F20110404_AACLNL justice_l_Page_054.jp2
15813151353ac18b98e497b5fc307abe
ad27daa3664ff9025e82d5f0af40752e7f295c91
1051980 F20110404_AACLMW justice_l_Page_039.jp2
cfe90f51ab9048a71569083ed747b02d
7a38c6366ad8fbaf89f9ab25d140db0e7ae2d799
105065 F20110404_AACKJU justice_l_Page_067.jpg
06829d9ecb13578e224783f28af34aa6
1c0f10763e0924edff6474bdd721b8bc2efabcee
F20110404_AACKKJ justice_l_Page_088.tif
abfab3dc2d2ee1ebe53f4d018c3adcce
a7ae2c6cc901e446ff375ffe8ce4fd6f7d7c1023
529760 F20110404_AACLOA justice_l_Page_070.jp2
3aaa35b0e0a98b004598d61a820843bf
491b5a321b5911cb5e6b461ff81bfa267ca8cc4f
F20110404_AACLNM justice_l_Page_055.jp2
e729c09909480bf5218a2934fbd4580a
af759e80c03538d9e4ffe69980d3cb703695314d
F20110404_AACLMX justice_l_Page_040.jp2
d31ac9c6f77664fc0a05d213fc2c9a56
590815fd54aa143de68460d35553c0783d4fb755
113474 F20110404_AACKJV justice_l_Page_026.jpg
b8d21924b8c58c14bf136f886b312ac3
3f9fe83a661d0b1b4786644f6043f7ef76d8f164
54459 F20110404_AACKKK justice_l_Page_029.pro
38aa070007c57e958186f5928bf9c7d4
73e91c6b05a9bd1a12fb3b060719fc6e124027a3
1051877 F20110404_AACLOB justice_l_Page_071.jp2
965fa65c5dc3bcfb61b396b0c734e7d1
687014283649f24d1f285acfc441e3065d00794f
763420 F20110404_AACLNN justice_l_Page_056.jp2
065dd0cccd74922eff100152d2532949
4d23413793745e4c1c754641c7b205c156c6c854
1051981 F20110404_AACLMY justice_l_Page_041.jp2
05ba890b5a81cd9f4c0cffdbc55d60e1
58c83d008800028c6720883a56e2d188b653e86a
2097 F20110404_AACKJW justice_l_Page_050.txt
73ed053055393fb6067b15c4bf346787
4ce60138bc76636bcae596b1fdf65ac962560873
121248 F20110404_AACKKL justice_l_Page_138.jpg
d8cede6133babcd50eee052856fbbfa5
855c679ecbe171304d61710230178fd6db29ec15
997019 F20110404_AACLOC justice_l_Page_072.jp2
22562efcd8fcea34302a984a2a3ec5bb
e15b96e7a848fab3e4367cfa02f1da2d76820786
993102 F20110404_AACLNO justice_l_Page_057.jp2
30028df88b4d92ce6553745963153e22
0c65a4d9f2da16f3b489afc4f19d7c34ba462995
1051871 F20110404_AACLMZ justice_l_Page_042.jp2
f0e4a31ce3767ee97ace3ae06311dfc0
e5767ed20108bc4af45aa46f518de55aca200308
71309 F20110404_AACKJX justice_l_Page_118.jpg
538a3c8640903e06fa1b7c7393a33a65
7acd43382de340711e884021bca37110ff8f6276
929 F20110404_AACKLA justice_l_Page_006.txt
dd0243ef9af46673a3414a16aa894ac7
2877c81cf103740dfd907247d39641353e74a451
F20110404_AACKKM justice_l_Page_054.tif
cc02205089b990e1f24c67dfbf2fd4b3
00aadac04639659448479e4cdab03238e83c5cf0
F20110404_AACLOD justice_l_Page_073.jp2
1ad82faa69f4064415cc0df00548f150
020b8a495404ca07916a59c8b2be1512c9b70fc8
873516 F20110404_AACLNP justice_l_Page_058.jp2
a10f50f7f9da74daf50c5c66838d782f
5c6f1557ce447d5279c758feead8ce19086d7b5b
F20110404_AACKJY justice_l_Page_026.jp2
b33dc7f58ce3fd8a88d512adac47a888
31f72bcac9a4169792dbbc16605f61c6273daa48
1797 F20110404_AACKLB justice_l_Page_115.txt
9bfd597b2eb28b813061f6b661139431
32266a55513f85c23a3a9e0dbb92ae698aafbc90
8738 F20110404_AACKKN justice_l_Page_096thm.jpg
baf99ebc2c116763cd8a513652ed4b98
417fe2957f4a1fb9848ad3c5cd72cc49b5030e7d
F20110404_AACLOE justice_l_Page_074.jp2
c4b343ff60c87670dc737d67d8ecd071
938e760537c9a793a39667f2e289b94e6bd2fcbe
F20110404_AACLNQ justice_l_Page_059.jp2
8d65f925dce8bea14c43f84cbe3f7cbe
6685e80bfcd2b6cc21c5c4b08f8c4d8122ffe6f4
1051968 F20110404_AACKJZ justice_l_Page_016.jp2
32d1251e8d4e5d9af5cdbdbd62d3ad8c
7a7d615b614ddfb5db1dc5301120f1a78a8bd9ae
38566 F20110404_AACKLC justice_l_Page_109.pro
dfc8227b338df369d9bdf1e03cc60116
f9002ed9aa0a645c3bc19810ae92852386e35de6
2350 F20110404_AACKKO justice_l_Page_141.txt
6ec98fdd99856d1e01f70f9c2a6d4271
72d3944f72ba29e57bcabb61c577d1ec714e6f6b
790014 F20110404_AACLOF justice_l_Page_075.jp2
75f3fafd0f5c35daca7a46e76f0be164
078665c3e9a2fe048a596ede04728eb695d66045
F20110404_AACKLD justice_l_Page_034.tif
6093305b4b5ae87c673d34db679d7f33
36d1c8d0ae93aef1a6035bad31d0037791d1f0b0
709431 F20110404_AACLOG justice_l_Page_076.jp2
dd037bbe3dfc5bcfc5d69cfbbbb6bcf2
7293245cf688c2c58daa29d956596a6b2445a833
F20110404_AACLNR justice_l_Page_060.jp2
ddb9e43aba5b27e6462965c766d11124
4100177de2e970dac3dae6d584a80d8f1b1c4193
F20110404_AACKLE justice_l_Page_138.tif
b1a2cadc2633aa72262c61ac8f5cdda1
954665be3b71b66baa4a2120d7011f094f3fe1da
1347 F20110404_AACKKP justice_l_Page_098.txt
b3d555b69f3cb4fe96fa4acb1093f459
e3f75d537cc8d05a6d4b3726e29e799c4e9b5bf4
1051916 F20110404_AACLOH justice_l_Page_077.jp2
7711fd275bcc3dc68287a06d4b731e4a
9883e9f46576521b5b41d82bda113e51a3af6371
1051954 F20110404_AACLNS justice_l_Page_061.jp2
47cd9f25342f218614eb1cca27bdd0c7
5a420ad51afe7ea41f08a85853f8b2d251b296dd
714112 F20110404_AACKLF justice_l_Page_089.jp2
94e9180e5b1bb70ddc7eea70d5572638
13eacdf413981e98b6f875826523bf40274db7d2
104223 F20110404_AACKKQ justice_l_Page_022.jpg
1211e5951412180015075d36a0d4891f
56dc1c71db3a6f6e19ae314f58c03b777cbd6661
952551 F20110404_AACLOI justice_l_Page_079.jp2
ac8dc8dbf4167f9c7954d3bd11d8702c
7362a7dc6efb29451f11f74def42db72b63358dd
F20110404_AACLNT justice_l_Page_062.jp2
6f35200e5629a53c19644baa4f4ff238
6ab52f868b6aeb23b0c54bd0ccb89ed3ef26ce7e
110583 F20110404_AACKLG justice_l_Page_055.jpg
43bb1211ae54c28a1d458c4cb8ce4edc
84fcf7d45e0cc39f6bf1f369f41a9668e64ef4b4
23537 F20110404_AACKKR justice_l_Page_070.pro
f271a6400c05396f5fccb4557d8d6b05
280f522f5c32059a597acd4b5433a8334ccd6ed8
804813 F20110404_AACLOJ justice_l_Page_080.jp2
61be82a3133d6d5d12a3b7dc43a69f56
5325aa008227f54c1784922cff623815cc331046
1051963 F20110404_AACLNU justice_l_Page_063.jp2
8b18b4147f4e7d297e67fc5d776c7243
a2162728575a6796a9a1a88860b2890bd5442791
F20110404_AACKLH justice_l_Page_059.tif
661e3602d38b6c00ac969bd157596728
8574b6581ef4bd0c73d4b1a9fa9b727712d1ed94
F20110404_AACKKS justice_l_Page_072.tif
5a2f99f425f51676eb05709cc5390629
1b645dd1cc02237c3b07a021337c02f20a233f18
986389 F20110404_AACLOK justice_l_Page_081.jp2
5ceb867fd2cbdb66e1d2425ae35f828b
55fc0987f671026d8e2d39ecdc88251723778723
1051955 F20110404_AACLNV justice_l_Page_064.jp2
96e201e4e28d83ae647a90a193c38efd
5ba14d0c0998ea8dbd3452943f3d1bdb9063b079
32565 F20110404_AACKLI justice_l_Page_113.QC.jpg
29e12c33a706a2d6156eb73b09330bfc
e66f9955c1539ec8751752196ab7e8dc89630a14
49381 F20110404_AACKKT justice_l_Page_092.jpg
92acf51411bad891d9fc2db1717b5fb1
07f44a31b79d6b308a8d30478eb002ac1092d152
462644 F20110404_AACLOL justice_l_Page_083.jp2
ffd5faf57ed05c5656faee3ba4092a41
2125e506ea84576b6b9564c168ccf6f9182bb253
1051962 F20110404_AACLNW justice_l_Page_065.jp2
917a3964c168d83460fa29c9af31e305
b9bed52dcfa782e298dc1e8b50117714f59b2d7a
17816 F20110404_AACKLJ justice_l_Page_097.QC.jpg
bcb582a578dce8f93a0cdd52f9a34f45
73da22fd668585adbf37ce008da8b10b1119d0c0
F20110404_AACKKU justice_l_Page_002.tif
af23d60c7e0d595b0ed7ef8f06dfdcb2
e973229fe40b6b8ade1734c74c16b4cd0a7fa74e
413225 F20110404_AACLPA justice_l_Page_102.jp2
ad351df62cd3bd10cd84c56cd4b90a78
1d523f036833ed3e34c97f1912503f0be1dc4ec4
516679 F20110404_AACLOM justice_l_Page_084.jp2
16a179914d2253cabbf9de050c9b4cf0
cf39d52ea4d1b33f6254cd4a21ad101125a731c0
F20110404_AACLNX justice_l_Page_067.jp2
68a71d97503f0b0845fa7ebb27c45f6a
fb0590fc6474fa04a28b60e3f9d6b9a1c538fad8
55851 F20110404_AACKLK justice_l_Page_028.pro
f503335ca5e3794ff6982f2c14320364
ad0a0d1ab3158ca451b3b09d2cf50d0ef32b55e7
133931 F20110404_AACKKV justice_l_Page_133.jpg
9e805b4f78fe576c8758b4041b25a195
fc6265ecdec4ed76050ceddc8cfafd26acd996c4
751665 F20110404_AACLPB justice_l_Page_103.jp2
a3973488b9c4a5090d21bdd0de13415a
f7066d6f291604efd2046d6bec3ff31b7f600cdd
437256 F20110404_AACLON justice_l_Page_085.jp2
d3de9208e20ba0ddd51ca0398a59f86a
4c0b1cceed25f6a4caead01cfe7d0c109866742b
F20110404_AACLNY justice_l_Page_068.jp2
95d871b5f3ed992369757c6a7d80c2fd
29f427ebdf4bec0326a8c41d0b1cf6e215572218
2790 F20110404_AACKLL justice_l_Page_077.txt
039eea99d047e201ab3c0788541a5c12
00d81d26639f5834653ba9d19411e6c9ddc1b05e
107258 F20110404_AACKKW justice_l_Page_037.jpg
c9103aaa1f9f0f01875639ad8a9715df
e4805e77934d73805d8e09302054616b61d1cd2e
506333 F20110404_AACLPC justice_l_Page_104.jp2
65cb6da6e3f77752e7ec21fb0862a339
b3df0c4af1352f8f7c408a7aca69399e74e3c628
553905 F20110404_AACLOO justice_l_Page_086.jp2
1f6e15258d9bb9d254e53aa589e4ff38
b83b7b1631660663e9c4c29fe3d9312dacf9884a
F20110404_AACLNZ justice_l_Page_069.jp2
a41d0f5117509670c8bdc3d66d6a3f87
045ac31909d86deb492b7306d6c7e4a0dd68bf68
F20110404_AACKMA justice_l_Page_082.tif
df4175b8903c632e74c7cbb6ed27aec4
834962e94624b165e543db901e4edc0bc64c2e38
736 F20110404_AACKLM justice_l_Page_127.txt
422cb5c0b630b0eceb0fbe6b9760c789
6d8227b2be189df3c3dac5c5d1829b49704b68cf
2168 F20110404_AACKKX justice_l_Page_023.txt
1123ea2f8dcd0a1ee18433642427177e
0fbff020c83fd063bdde3429bc5c285d3f8800e2
892551 F20110404_AACLPD justice_l_Page_107.jp2
dfb85b32667f421f15d74fd6f7ac2547
ee03527b3b218d67a0aef4e33227abd95bafb9f2
765983 F20110404_AACLOP justice_l_Page_088.jp2
3b4b33bb9d0bba17ad5f26accdc52bed
7c74ebef4dc7d2203882965f747d545ae0ad9b47
8887 F20110404_AACKMB justice_l_Page_023thm.jpg
a5b4828bbfa4244a3d0fc1781530a2e5
a92ca746aaa2065fd2b60e8412728b327019a30d
50797 F20110404_AACKLN justice_l_Page_066.pro
0a877ce0a5a507253ca7d399890905e9
be9736d2583fa87b8f3c50396ea467bdc38a43a2
4320 F20110404_AACKKY justice_l_Page_002.jpg
20707afab7f6b7f96cfb1602e9d1c499
a2b7a89361895210f921d73b921ce21e88da6b0d
F20110404_AACLPE justice_l_Page_108.jp2
60fa1460c1f4bb9c37ab11ca382c265a
ad4b087df0d0fbd8e46093ae8f1f5d7e93374632
441590 F20110404_AACLOQ justice_l_Page_090.jp2
608b264a99ce45fb2841e02973b08ba4
d8b0635dd51bff6b137c7380fdd9d89866e3553a
1051911 F20110404_AACKMC justice_l_Page_066.jp2
c127996e22ed98230e2790176fd651bc
9439cb6e3692abaeb4a7fbd68013563dbca7f4f1
F20110404_AACKLO justice_l_Page_105.tif
df6053724b7db31ab39659d003e71425
694d902203237cc195fd9cb2337ac94d04dc8bea
43863 F20110404_AACKKZ justice_l_Page_072.pro
f4b333673c3d93feef8d7064cf277110
61e3f4230b01597c0d4dad1970b6f251bb28a855
890823 F20110404_AACLPF justice_l_Page_109.jp2
a2b1f1cb0bc2397b8a281a7ccb6ba3bb
01ff517fb759f8f50bac2564c2c3999ab1c8f211
618486 F20110404_AACLOR justice_l_Page_091.jp2
8c758ece7c85153b2c602b7a7133748d
ee8a62056469f5d4ebeb0ebfd60de52fba6e86e8
F20110404_AACKMD justice_l_Page_004.tif
dad221de08b47d703643727a794f0a33
59e1136fe63cd8f08b30da4fc601b7f347a9d8b0
F20110404_AACKLP justice_l_Page_102.tif
41304a8f7056e8a55eb8c3430f03c3f2
197db4c0f8c0f1ec4305b0a934d2f6e98f1890a1
1049732 F20110404_AACLPG justice_l_Page_110.jp2
f848634060bf873037fdd8ccd25ebab5
d002b4522b265448b2b100d41c7d9c02f39c8afc
12399 F20110404_AACKME justice_l_Page_127.pro
223a415434821a924bd4da89be7c1668
9c82689410abde92e58dba4e73603586adad5796
957193 F20110404_AACLPH justice_l_Page_111.jp2
742b7dc4c707e054c76abb78a612f252
d67c1309d1fe7fe2ef5421908e1d5b8636be9f97
448562 F20110404_AACLOS justice_l_Page_092.jp2
baf4fc95f1d47850220c08197f543694
47ef4f7dddc6edf515c035ee48460c6ea535641f
739820 F20110404_AACKMF justice_l_Page_101.jp2
f87dfd07c9d675a5ac26fca116025a0e
93b66526afb26a7c8add96660896dc34c6f21aac
1976 F20110404_AACKLQ justice_l_Page_054.txt
9accbd6ec57ee20763010ff98f4a5130
16923204e851c66e33cc9b7c78ac4a43e386cba6
F20110404_AACLPI justice_l_Page_112.jp2
17ec68ed435bed1f55a9bbfa84db5a7a
498eea7bec596e3e4e37dfe2fe7f20327f622e52
471268 F20110404_AACLOT justice_l_Page_093.jp2
38b408517ac19b1db1065ac84e99ada1
4140371c684ec693002755573bc06c3b1bc53f01
8222 F20110404_AACKMG justice_l_Page_114thm.jpg
aaffcdec3bad6cef139585e964debd76
d808b921bccff13f88ca4c4ae7dffef66b54263a
8337 F20110404_AACKLR justice_l_Page_034thm.jpg
08507e8f8a0685ee4b8c5229561ddac4
151a5bfca3c4a2ce27ca681c0e04d83aa0ec919e
F20110404_AACLPJ justice_l_Page_114.jp2
872abae61f4acf32ee4ca3a2fe5d2d42
3933b0fa0445d6663f0ce591a9628985e71751ba
450064 F20110404_AACLOU justice_l_Page_094.jp2
8f0f186c4757b8d58b0f1baa6268d44c
257b34d9e70c67c5568e1b4fdd6cd00eb4664f11
113019 F20110404_AACKMH justice_l_Page_028.jpg
259040466ab16e00689ba7a043d10d60
d54cccd4d2494d7908b52383dd510e5f2f56811a
2115 F20110404_AACKLS justice_l_Page_121.txt
0886aec5023a858723e75f4e830ed1f0
733c1c0e7b4ed0451dd8b3d778dd0ff11449d426
1013891 F20110404_AACLPK justice_l_Page_115.jp2
3363e3da990a898d58ca1fd903555e10
067e4dd81ab301a3d30f5b88a982fa6aafc69a2e
901714 F20110404_AACLOV justice_l_Page_095.jp2
3b2a8b4f62a009ba0ae29f734ef5dbde
08303e46a9887c24cd56f0c8201d1d16e4c04b98
1051910 F20110404_AACKMI justice_l_Page_133.jp2
3beb1695f0a0dc2df978f078ccb1cadd
164de7c6e43710f6b4aa97017aed6d2dca78a194
104437 F20110404_AACKLT justice_l_Page_018.jpg
1b6cd5a99d7983a5209af87ad879927a
e19944e83469ec5fb3774baa9139b940f6cc1c19
974348 F20110404_AACLPL justice_l_Page_116.jp2
05490bfb8ae7cf3a977bfbed0112a7a6
9109ce6177ec38acf99752ee930acd8687058187
1051948 F20110404_AACLOW justice_l_Page_096.jp2
74d7cb8d07f39a25bdadc1b96f126b48
804ff2105035081c36a6813793f618ce83ff0e3b
2011 F20110404_AACKMJ justice_l_Page_035.txt
8a6c2a89607ea71604880d4f39624a60
647f27aef7c8ec13e6b369f6a55ca6a5674da65f
51147 F20110404_AACKLU justice_l_Page_063.pro
21e7b1e50635c8812994e705bef748a7
e153fc1b388f3696e08780fd28af1066b037eb6d
353564 F20110404_AACLQA justice_l_Page_132.jp2
fddc14714052c2552aa7c327be1931a2
b91263e33d886ac1f06ec71ecdb5e54ea6acbfe2
1009798 F20110404_AACLPM justice_l_Page_117.jp2
6a301aa7b61e36d284808da092928a03
db558c8a5d0ec79b8275c5b819a92e268462dae4
518510 F20110404_AACLOX justice_l_Page_097.jp2
6d214dd2fa4c89769ec3001c659d0b18
49a5481c5da128c8752ada2a9acb676675249c78
455687 F20110404_AACKLV justice_l_Page_087.jp2
c871ebeeb0d9754361fd88ee472365d1
750a0a1395af75e4133c4d6e4dfa996f3e86482a
554228 F20110404_AACKMK justice_l_Page_105.jp2
003ffd8bcc41f0972f69b18b47b28169
8cec0613a41cede8124727cbaa17ad0030de38fb
1051949 F20110404_AACLQB justice_l_Page_134.jp2
7b15e794658184a27426c25386866bd5
1f8578fb7460bab39fecfccda2ae1698c9b838ca
765181 F20110404_AACLPN justice_l_Page_118.jp2
63f40ae163266fb92b5885e29e22682d
3225d545b8fcad982131b7070503d21852fba73d
455738 F20110404_AACLOY justice_l_Page_099.jp2
78d4701b6166289b51de9eddc3bb763c
6bf063d4dca2415c350f98096970fd9282de3b90
1957 F20110404_AACKLW justice_l_Page_019.txt
28dbe62b975bf40b12ed89f293859bf6
8547323a1762e59e07c71469421b5b91e0f9394b
7591 F20110404_AACKML justice_l_Page_111thm.jpg
84a378d0323d79f830cec1e444e396df
d0225b1f87ed4a19c3d9e12ae53eb9812e5608de
F20110404_AACLQC justice_l_Page_135.jp2
bd884828f320a71567c12c5731ff1c02
a7ff94177478101c5e4d0ea97a3cab99d4cb050d
554493 F20110404_AACLPO justice_l_Page_119.jp2
577087c0292329aa10edb432c38f0c25
a69bc8983a3e4adcfcd13de9b36df82389380bd6
387421 F20110404_AACLOZ justice_l_Page_100.jp2
18bd6e29b3572dd6fff1afe4b653a68c
997f60fab9010ce83ed2c86e565e77f5ed5dba7b
35453 F20110404_AACKLX justice_l_Page_065.QC.jpg
ce8fd26a9d04fd55d5f0970b39ccd044
f147237a617e424035934375815f456ca518647a
4495 F20110404_AACKNA justice_l_Page_093thm.jpg
eb503ab43efd207af10b702bc10848eb
221144c41a611fc08162ea78293113a4cf2678af
53621 F20110404_AACKMM justice_l_Page_110.pro
725a3112582e3c6f489fb06f2674b9f0
cb688d278b1656d3da2390aaa6dd29b7ed747075
F20110404_AACLQD justice_l_Page_136.jp2
ca07cec5e2e80b1279a232a30be8b284
6eac65b8ad11686ac6b1e50084dda395e279465d
235666 F20110404_AACLPP justice_l_Page_120.jp2
7f5f26d179055ce3857c25cacdda331e
6d9d1a99567ba349abd83eb82ba8d2f191d4e2f4
9287 F20110404_AACKLY justice_l_Page_060thm.jpg
35fa4f153c6c2a56ff68c65d8e969087
9814cb20dba2fec0ee87f69df4d0a7b283e28fd7
8655 F20110404_AACKNB justice_l_Page_011.pro
e2f56603a7f322480f799b6f0cbf4f47
212fb92485e589fe443d9f1e134eec04eaad6571
225616 F20110404_AACKMN justice_l_Page_032.jp2
a030afe72aec2cf4fa6bdfff20200e47
ee0e033f7c57475dc28b96654776089af2cb4872
F20110404_AACLQE justice_l_Page_137.jp2
d4d1831542ba0226d8b01f247db46cfc
f93e36f9558e63df8192f9e8e7a66bedfe59338c
F20110404_AACLPQ justice_l_Page_121.jp2
cc63eaa7f10a1cf6d268695794ad0626
870374e322584473aa7fb79adc55438ede047441
16579 F20110404_AACKLZ justice_l_Page_092.QC.jpg
aee07436ab8edcb02da02eef68f49b28
81918f40597044412d49f633a73add213d1f271e
2144 F20110404_AACKNC justice_l_Page_015.txt
11b6a38c5c815a04f169335e999525bf
3123ed7c9323c82f76aa6b5d68c9aa2dcecd048f
31761 F20110404_AACKMO justice_l_Page_025.QC.jpg
4fe8861a7581fdcac8910d3217bcd2d5
1252a72fd447c532fff0e8602c8a6ac216e2772e
1051953 F20110404_AACLQF justice_l_Page_138.jp2
5b8ab1816af3d945caf86d145e20087d
12a442f45f65fe08439b9d2c20ae2386417a6df9
F20110404_AACLPR justice_l_Page_122.jp2
f43ba3f5eaa2e258f76f97c9d6351fb3
e63299691c51d9dd8e9afdc68a3666f22a7cd605
F20110404_AACKND justice_l_Page_008.tif
97778297d9723a0a18b97661eb561ae6
ad8e076cf4d77220837e924f567eed8f4a7cf2b5
F20110404_AACKMP justice_l_Page_084.tif
3eb3e251fbf2bfaed441c513d3955f0c
6ecfd0bdd2599d8942f622b0f2476b0f26f134a5
F20110404_AACLQG justice_l_Page_139.jp2
60d352510d2ffd30caf4f7869920b466
86fd18f6ecc350439f2ef69dda401b3ec618a3d9
1051936 F20110404_AACLPS justice_l_Page_123.jp2
07bfaee954d7424c9d50442bc321ff99
c56ac452ad17b1e45f6fd00fdb00e92dc2e32516
1846 F20110404_AACKNE justice_l_Page_142.txt
c89f637b392e8a17176cb7c6d11b93a4
ede04cc8607239468a8d503ba1461808d8b7752d
36490 F20110404_AACKMQ justice_l_Page_023.QC.jpg
d667c6f44994212bfa291807f8d25594
6eb9bd6bfc7fccf872148cfc225f3fdb64c47d45
F20110404_AACLQH justice_l_Page_140.jp2
8e4fe94585c25489c251aa6de7b443b1
28cd506c1ca6acf9c58b3afda36173156749d6ed
7847 F20110404_AACKNF justice_l_Page_078thm.jpg
5e73d55958d38d7b772ecfb3887d9379
f26e7b05f2c36c5eb16203f5ef4c8e2d74cefb7e
1051926 F20110404_AACLQI justice_l_Page_141.jp2
d89a85383b7ab817214e6d9c9291e2a6
2a37dd49d2c1800db6858b5065d6a3c3f36aa90f
756574 F20110404_AACLPT justice_l_Page_124.jp2
260564cc2ed579b08bd7074f3bbea157
773c67d8da47a97656a5c8061bf494656679b8db
32628 F20110404_AACKNG justice_l_Page_058.pro
e8576ee49b279db86fa7208ad653042e
1be6a4e7b84b7497d4626fdd3b0b21474b0592c0
5955 F20110404_AACKMR justice_l_Page_124thm.jpg
6fbb5b3b8290b3bcd26b375552c9d2e1
a96364543a011f974800d8856e5a6a36e1cb3fd5
1026146 F20110404_AACLQJ justice_l_Page_142.jp2
d1e89bf9962e0bf9878a11e6eb16fec4
710ce7d3765021169c198450af387c8cef6b9b19
784839 F20110404_AACLPU justice_l_Page_125.jp2
536b3e8318013aee261412768888c7be
86a4c36831ec894b2b9b70eb16944b295f7a77ac
F20110404_AACKNH justice_l_Page_006.tif
ad5d1c38143662f854dd19766f342135
adc1e2f8769cecfccae76697940329e87ee649d4
90715 F20110404_AACKMS justice_l_Page_012.jpg
8cb96bd8faac15be85cc98bffc8463ca
ab9e86b62af25dfc7460eed30b110d568ec704de
259348 F20110404_AACLQK justice_l_Page_143.jp2
a2d7f77a8596b67f1ccbe933a8e75086
1c192bf0d13375fa6937ff7f9ad7308c51bd0c3f
866499 F20110404_AACLPV justice_l_Page_126.jp2
854fdfdf6b36e0727469461c901096ae
80774b52900e630368a6b9b51a793b0a069df576
961365 F20110404_AACKNI justice_l_Page_010.jp2
6508693e9b2b180b606b7f8626d7c518
adc6c71df2a29d7beb7f2ce1ffa1ed9251fac6ac
95596 F20110404_AACKMT justice_l_Page_081.jpg
bf19ea9f077fc4b323cf43888e928033
922b709d890d8241dbe12a1bc162d88679644668
2349 F20110404_AACLQL justice_l_Page_001thm.jpg
060d330c985922d02ee5b9cbc3b12ebb
532f06a0f770dbe2063850b88daa4f39f547817e
491312 F20110404_AACLPW justice_l_Page_127.jp2
daeb0695827d8f28514105c28244fab2
84febc371dba84c386116427a77d2766db4885f7
92220 F20110404_AACKNJ justice_l_Page_009.jpg
accb2d5c40f3c373ec24fdd5c12144bf
9315f283d772fa341d52cb8d7950fb29b7b75d13
5112 F20110404_AACKMU justice_l_Page_097thm.jpg
65f9f0802796891a5ed5b7fbd70a795d
e3ccfba226a2cc47de452553ce3a59e20e609a38
5678 F20110404_AACLRA justice_l_Page_017thm.jpg
3ff4c28effab4e141e07d0a07c267048
2ac183e7f7a417ed82dd4005241f62eb087fdaae
527 F20110404_AACLQM justice_l_Page_002thm.jpg
3ff76ba8f5f762a073cf27dea2535fbc
845136f14d7a48a4a77bd49486fba59ce05dc17c
1017946 F20110404_AACLPX justice_l_Page_128.jp2
1fda90a5023f5b1e65f5e4b71057d769
9e0fc2086d3d17b015eccadd21a13097e50e5cc9
2340 F20110404_AACKNK justice_l_Page_138.txt
a35bc153704d06aa387de4ed0bca1f73
b44eb892fb6af8a9d6405ae7cf5d8cee2226ff47
67116 F20110404_AACKMV justice_l_Page_060.pro
b839fdc0beed4ab7f4cbf18dae7d7fa1
3ce9ffa9c93455de867935f88f61021442bfe203
9016 F20110404_AACLRB justice_l_Page_020thm.jpg
09bd689a5d79e5a6e5b2458b6e9fb360
cbd3c3cec86a70a0264b0933f2469d67d3358465
7357 F20110404_AACLQN justice_l_Page_003thm.jpg
28544720c8a4101da9b4b87e354c887d
5e186cd1f14250ac429524b12f0df30d465226ca
407713 F20110404_AACLPY justice_l_Page_129.jp2
4358abb3dc6b04fd0e906c7359a3d612
a6c328932f15cd4546cc9ff03f5514c921197456
2285 F20110404_AACKNL justice_l_Page_024.txt
677c21bcb94dd4bb8667d1e17e0edc14
c35c50ff7c2e0a8db857500c8f7325c36bc267e6
1947 F20110404_AACKMW justice_l_Page_073.txt
c355ee0a518cca1e676f247520607ea2
31d62666b7a54162543174f2fe198609b643db9b
8919 F20110404_AACLRC justice_l_Page_021thm.jpg
df449c19170c42bfcd660363bc485393
39adc52d0c90cc4abd80da2669c22cbf43c2564f
5898 F20110404_AACLQO justice_l_Page_004thm.jpg
d1f896584a2f03c5946c0856eb5c8f6b
74848aca3d1b97a7ae2c533ae7ca4ba0817e002f
F20110404_AACLPZ justice_l_Page_131.jp2
e8cc74f25c2e93f8f824f1a1eb4831cf
b217cd8bdae832e9152217f5516092d714ddbd48
98974 F20110404_AACKOA justice_l_Page_048.jpg
c9014707a58ca8eac3c121c131974b48
8e9dfbec6bdc948c084153889b507d5f28f786e7
13599 F20110404_AACKNM justice_l_Page_013.jpg
a460a27e858905fd6b70fd01b9b13f5a
5cc6629c4373f5058a712d341058101970cfa74a
34040 F20110404_AACKMX justice_l_Page_022.QC.jpg
afcfcf00f397a0574234b30a40ae7764
bf21100b9c72dd95e220f09a7fea1b9fe1dccac9
8734 F20110404_AACLRD justice_l_Page_022thm.jpg
ef008f94eb3f2c25f3d86cea827f2fce
4bed352f4b57d8adf3939b81fb3252cd6a8a1b08
5833 F20110404_AACLQP justice_l_Page_005thm.jpg
83377db949cfdb209634261f145ce10c
c111f6e1a0e8bd8251e406cd4efd8d2321860aa4
2145 F20110404_AACKOB justice_l_Page_027.txt
4bd16ba6743976c56216430214c84bb1
d032fcbf6240da3f96c2e88462a861e31746d93b
43762 F20110404_AACKNN justice_l_Page_081.pro
37387c94f937580029ad7ef569305d74
782c45b03829a6a9cb950685aa80ab3c55f2fb78
50661 F20110404_AACKMY justice_l_Page_038.pro
02c9773334510f97dfbbabdc950aa222
c428118a25b7b9492072c2dbec786d996b128101
7803 F20110404_AACLRE justice_l_Page_025thm.jpg
52f8dd24a9c67498c5b433ed3c58abde
20505ba70ccc22522af5b5ea181d3a83d121e8b8
1631 F20110404_AACLQQ justice_l_Page_006thm.jpg
65ff3259486916b948d45f4c7fe70058
fbc64f8b0e9d72306d9152f87a30632e49212779
2337 F20110404_AACKOC justice_l_Page_134.txt
ad1adef1606258bd2e48872eb9249683
d04f0d80cc3392195213e0c7786fdf26ffbfeee0
33334 F20110404_AACKNO justice_l_Page_019.QC.jpg
11b7e268ef00f3558d5fde98fba36751
b46d709f6c2ff0761c643d643ef494d8a7fcd498
8889 F20110404_AACKMZ justice_l_Page_143.QC.jpg
75035d0742e32a7b7db53104794e45f2
0f6d1297eaf34b08e36775654cd12d0546d74039
8576 F20110404_AACLRF justice_l_Page_026thm.jpg
e1b155bf81f21e20d18e984984a29616
4d0b536b937e51c762b8ed3d133b0c1448cef34b
3981 F20110404_AACLQR justice_l_Page_007thm.jpg
dff26c5a0ac95b86bba58b9f6e773183
28ba980728a06d3662f10bbf4a395880915df31c
F20110404_AACKOD justice_l_Page_130.jp2
b61cd37a1daa18817a5abc0cb363f639
9c57d8945fcd8263eff171764877969908e43daf
32922 F20110404_AACKNP justice_l_Page_054.QC.jpg
9395c69825862c5f361f7f4fbf46da0f
5718d2eb00a864cc7808e91472fb1c8303ca9da4
8418 F20110404_AACLRG justice_l_Page_027thm.jpg
ba269cf42264e99123bdac6ae5bc9df4
d3560a73dbfad9c409d14c780c941e87a2d101fd
6127 F20110404_AACLQS justice_l_Page_008thm.jpg
b1a8dc08b1ce3197e141e33d9be6a62e
a390d140f848c338482598aa28a6e89cf7cfa904
52914 F20110404_AACKOE justice_l_Page_096.pro
8c952e8c773bc6e576d798f1ecfd4532
3686260eb6c44e8d4f17b5990f6e897a2c77ce36
4858 F20110404_AACKNQ justice_l_Page_084thm.jpg
c35c4d8d794f084b07d0f46b97286a9d
2119b7b0900140442f3f8b4a7128109d7f81ac31
8971 F20110404_AACLRH justice_l_Page_028thm.jpg
1e2ccbb5ff8128d3067e64b1f3153a95
01310ee3327e6061dc8813d2fb4353922319e647
6259 F20110404_AACLQT justice_l_Page_009thm.jpg
7424f1e41f782aa6bf54bfd8b9e42fbc
724fd15eeeab59376229b8182897073facbf898f
109155 F20110404_AACKOF justice_l_Page_112.jpg
49efa2fe01b8fc99ff7355b51e78d27b
1763a10ce6ef0a29db549fae9681ec030f57d046
111464 F20110404_AACKNR justice_l_Page_016.jpg
cd7e317d0105fb5ab2692700b55abd69
de6d7cd52e239bdc6feccd0a0657760db155e47f
8798 F20110404_AACLRI justice_l_Page_029thm.jpg
23a5f4328ae839044f668202eebe17a5
a780db52dc3aa0333eb329be9800262f61dacf41
8300 F20110404_AACKOG justice_l_Page_018thm.jpg
4a16ad2ffcd67a0f2cf1dcfb9ff630b9
6dde6d4f1c4730dce0ad0ff26753259d346ba859
7675 F20110404_AACLRJ justice_l_Page_030thm.jpg
75083f961537999cc46c6f01b893696e
de3ed538bcec535482740ebedc70535e5dc088ef
6384 F20110404_AACLQU justice_l_Page_010thm.jpg
36fe2b276b2c4adeff2c3d9c42e7a414
3cc70454b2e8c2cad7205e8ef45880ff400c7819
21951 F20110404_AACKOH justice_l_Page_086.pro
6c59b061edc9c4c6626da2968992b759
ee9353aacb351c7377d6e3e2fcdd6281538bd39f
F20110404_AACKNS justice_l_Page_124.tif
06bce29017f79c575e1fc3921ad55eed
51e96a047860e144ad7a96adab43e09082cc63e9
7722 F20110404_AACLRK justice_l_Page_031thm.jpg
83507cae2b724651542a03901fb64812
f0b15e04dc38ff8c4c3da2e0c5d618c1041f2773
1051 F20110404_AACLQV justice_l_Page_011thm.jpg
20dd6ee2a3636c3273bc98f687b9f889
7a31702bfe137d9fcf7bce129dfb9f034882c1fd
9125 F20110404_AACKOI justice_l_Page_024thm.jpg
da72b4ec6f3d8d7b421ff68b2d22a38c
1bed5836a7baac090ad6a02d8ccd127d64e31da4
35000 F20110404_AACKNT justice_l_Page_040.QC.jpg
2276cf6cced0e417223ce07f3b61349b
3d83d962292574b1484a889bdead70458b48fc03
2137 F20110404_AACLRL justice_l_Page_032thm.jpg
e56c1f9a190583965a9a89948b6afa46
1a0b83671853f986e8d31bceba4cde2230e7e9ac
7172 F20110404_AACLQW justice_l_Page_012thm.jpg
d887d00f5a51e7246d5a0b9637bb17a9
c1ebf9e10953ce2a997c29f67c690f9dfc4cbc6b
51123 F20110404_AACKOJ justice_l_Page_111.pro
d40fe74040789d7107d835445e818049
7a9054c8942a97f2b9ae2462d9355b7c3528db94
4701 F20110404_AACKNU justice_l_Page_092thm.jpg
85a45acc8c8464b48350d28a3b896c67
f112041752b059e8940300dae853fc27d65eb90c
8611 F20110404_AACLSA justice_l_Page_048thm.jpg
c64ce3c6d7678907407ff56652f8a807
dff614c0faacf5f11b31caf8332c1391a71b581a
8645 F20110404_AACLRM justice_l_Page_033thm.jpg
11c7f12d79eeb019cdc796d396cc1106
dcfe9ae4460813b8c22e7620d22a3531949fad71
1265 F20110404_AACLQX justice_l_Page_013thm.jpg
9f04732cda9649b300fed425f52a562b
cf8cfa10b91b9a577c99203c9e4b3131cbdb98bc
F20110404_AACKOK justice_l_Page_037.jp2
3f3ff4fc65eba37d6adfa331972b2459
27284d56dae76947c33d8577c60cfc92ff63bcba
F20110404_AACKNV justice_l_Page_119.tif
de72a3675115d5eb9d352056b6b4eab6
02d1eff9539ca9b497ce5e0cbb90da7adf26c23f
8536 F20110404_AACLSB justice_l_Page_049thm.jpg
dd527e7bceb729b1dfdd16f8f136d29d
78fe0ed91623cc6d317d68beaa19cd49d2913ba8
8191 F20110404_AACLRN justice_l_Page_035thm.jpg
2042014bf8147352459972023764826c
33f8878d371fc301b53360270c58f604381599b0
8609 F20110404_AACLQY justice_l_Page_014thm.jpg
e87e39dbcfbfdad221b01591deced424
cef2338e26bc230eb5abf2b4ce89fa2d69add913
660740 F20110404_AACKOL justice_l_Page_106.jp2
defacc0e995ccd38fcffc82eb6412399
74870dacf7cddd6e741e96b9fb7aef5cf486e669
2142 F20110404_AACKNW justice_l_Page_029.txt
0a0855f547cff77d210687c10528e8ad
02e0780375c0fd5447d8b377b8d48db14c9034a6
8441 F20110404_AACLSC justice_l_Page_050thm.jpg
a9c1b889edf7bc177d4ff80fcfe9f705
01fd5c2b2594096f2ef60b5d4047750df551af13
8669 F20110404_AACLRO justice_l_Page_036thm.jpg
21ce27e5a9d6e8fe663e8e5bd378f255
ded5a5e16069df0d9301a645c0c5c2ac92fb685f
8604 F20110404_AACLQZ justice_l_Page_016thm.jpg
a9256dcd9b7b5fad3c0ba5771285a113
d29ff57bbabf79ea107d3a10d24ff8423356d921
F20110404_AACKOM justice_l_Page_089.tif
20d74f1a8cdac51d63609e7ba20a502b
41c4ad46a05a3b70c43f4a2c1a37bd00aad27cab
27847 F20110404_AACKNX justice_l_Page_103.pro
80a7c0b6957df9b5d5acaa294458c2fe
afabb75d083d469c99e02d8704e7d1b306eab577
F20110404_AACKPA justice_l_Page_010.tif
4312d00c9a7e24c730754be2de89b0ee
d78f1ea6f8b21ae53deab780d2bea97e5e8f503f
7561 F20110404_AACLSD justice_l_Page_051thm.jpg
6c0e044312d028265c2e096ccba62a93
0024bcfedb2e918c756b67b5baaa83ca2f1cc99d
F20110404_AACLRP justice_l_Page_037thm.jpg
2b57e65610970f6e95eca7b4e31168c9
2e0b716f28b9e658575fdd20cb0c10d7a455959a
1775 F20110404_AACKON justice_l_Page_072.txt
5694f1c738b5de5f6e8e3a78296219e0
cfcd26b698b98858f872a6cb6b1313f160a51b87
F20110404_AACKNY justice_l_Page_126.tif
5dcc8b42358a5e9db23a14ba3806239f
fb4748f44ec3fb141286bedb795956e601c479c9
F20110404_AACKPB justice_l_Page_011.tif
88e41f7a9b7babc30b8abaf4a7983844
6d73f4b29452648d20501cfae1832a79ee85a962
6141 F20110404_AACLSE justice_l_Page_052thm.jpg
9fb61a7bd908209ada1fe3355b1bbe5a
425a77fe9a841a76c6981440aa9dcfd67563ff3a
8262 F20110404_AACLRQ justice_l_Page_038thm.jpg
32c5985c814beaad41a357caea6408cc
3a6d8763e2fee8c0066e9405056ad258cf8b5a15
1909 F20110404_AACKOO justice_l_Page_012.txt
16cb52f9beaa4c25b56856f2875617f5
992455c77e94dbbd5f80268832e785b83cc9ec18
48659 F20110404_AACKNZ justice_l_Page_079.pro
b4ad3826df48edcd43faf4457a019adf
2abfb9bfddc82506b990e6e0f89dd70e3354dd9a
F20110404_AACKPC justice_l_Page_012.tif
9a082da3e50bdb1a0bb90d990a455005
260a3067962917dc00d40c226d4ffd691cf5b21d
7633 F20110404_AACLSF justice_l_Page_053thm.jpg
1d927ba3497059756999b914bc4edbe6
b90e2659ab6cb49fe807da6094e1ce00b691b0b6
7703 F20110404_AACLRR justice_l_Page_039thm.jpg
ec1b7f5246e621aed34f0e7eb7d90f74
4e839d49e8ec072e062af3c9f6b90c9f66102ac2
56639 F20110404_AACKOP justice_l_Page_140.pro
81b1f1455a9aee91f96f9a7de0538f6e
de97b918c9b2a4969700a0f42bf4be26cf836c66
F20110404_AACKPD justice_l_Page_013.tif
a96a6224de78fcc4ef0d4d0371296099
859df4b896481b89ffdf276bd9da93ab632a4a65
8345 F20110404_AACLSG justice_l_Page_054thm.jpg
f8bb0d2e95287241a546a0b8363751ba
a6d084a59be02f4c44f58f8dd510883d7abf17d3
8636 F20110404_AACLRS justice_l_Page_040thm.jpg
50f9a7dfbfb346cff1cb72fe84432bf5
96e9d556e8b8518655c2b4300554faa9c4c8fe40
2836 F20110404_AACKOQ justice_l_Page_132thm.jpg
30f4e28eb6a9dae31dd59dbf91103766
8d6753c678cde655b012bf8e2170a68d3f8ea6e5
F20110404_AACKPE justice_l_Page_014.tif
0afa50cb28179b703514217ffa4e13e9
20affefa3edcecebda94dbf31d16f22b9f4cb3b2
8733 F20110404_AACLSH justice_l_Page_055thm.jpg
e92df6b631852f427b30b4353bc1f327
0023c8f7998db159299e7c7c5a48e41853b0bded
8850 F20110404_AACLRT justice_l_Page_041thm.jpg
b0c193314a03ed61c16ea13920d325c1
4b8652d33610a1b7ed65d53d1cfe3975b6d6c391
44769 F20110404_AACKOR justice_l_Page_003.pro
d8b82be31167859a7370b3ca998f5a5e
15ce9fd80f9430e136ec9ec408a1dff8dfc7fa6b
F20110404_AACKPF justice_l_Page_015.tif
362a4723efe62d8bcebdd3371508ab34
cf6b4a0a0ad44ef20b3f5e417e48ad70fb08ac35
5888 F20110404_AACLSI justice_l_Page_056thm.jpg
21c605ff0b7b57389e5479676b60a165
e899d2d8024dd6a35a35cff37c30edeb1650b97b
8450 F20110404_AACLRU justice_l_Page_042thm.jpg
640800b67e91d2491e2641254aad6e3a
bb76dd014356ea561c9f5d1b00fec888a04d0303
34140 F20110404_AACKOS justice_l_Page_122.QC.jpg
898c5af638152a65efd34d57eeb04acb
917876fc06b6bd0991a02f096a7d1b6a86f39269
F20110404_AACKPG justice_l_Page_016.tif
2fd37ac2b7513b7c62c2953084a27def
08af1511815db31af2046986393ca48757da65f7
7957 F20110404_AACLSJ justice_l_Page_058thm.jpg
8741ca5b468a7fdec30272350341357a
6e4b3844aac087833e4cac9e874e5a65ddd4b414
F20110404_AACKPH justice_l_Page_017.tif
43c5c03a6719205da1fccd8a285079f3
41d81a540ee8d6f70dbaf47537618f50972fbb4a
F20110404_AACLSK justice_l_Page_059thm.jpg
9ce0f50100eafac96d81957b62e8b7f8
5f5e4524becfbf283bd21a871392d9b7b76f514e
8698 F20110404_AACLRV justice_l_Page_043thm.jpg
434244fed44c6b4f5ce3438731983cd1
a7d1abaf10ced146dd6a0f4801a5ab428efa499c
231712 F20110404_AACKOT UFE0021174_00001.xml FULL
5b7826bc90adee1251b8ad46cee928e2
723ceab7b1958c14868c5368ef5436775fd6bb9e
BROKEN_LINK
justice_l.pdf
F20110404_AACKPI justice_l_Page_018.tif
9cb12474f2533b69a6ae63f87b27ad38
956520af4dc6d39447c1fa2a553284d2e10b668a
8748 F20110404_AACLSL justice_l_Page_061thm.jpg
dd342cbf3d3c533ac263957d05d4cf83
4bec9bdd6f2e3b4dd437e6651930d50eaf2299b7
8277 F20110404_AACLRW justice_l_Page_044thm.jpg
20e66153e8208a0f0fd6ed359bc1d133
6b10b13eb37243160a6d5a35ed1f5a12cd956a93
F20110404_AACKPJ justice_l_Page_019.tif
00007b40ebc9d6451a92e32168db1e71
5a63b8e760b6f50c243c6bb942c7e60b98e0aec4
9170 F20110404_AACLTA justice_l_Page_077thm.jpg
bf209e9c2963a5737e2b734010dbad9c
c19a50047fd71f92eb8ba83aec769213538b09fe
8620 F20110404_AACLSM justice_l_Page_063thm.jpg
2bbcef597679836ef6c6b3291f6d8e69
d8ada74d2098f0a6ba279eb32feb86e1b4318f37
8792 F20110404_AACLRX justice_l_Page_045thm.jpg
e1d0d1297fb37870f3f5ebb59973f2e5
23a8bc24dec7a25aafc2775f8aa48a6f76feecdc
F20110404_AACKPK justice_l_Page_020.tif
d276e77becff73ad4046ba42ca877367
f99dfbbd1efdd3db2e77f692c552b52a4242daa5
7581 F20110404_AACLTB justice_l_Page_079thm.jpg
12ca3ba7651c3f86405113cbfd7362f5
1bafaa72a0fe96c99763df0987f771198a02e5c7
9020 F20110404_AACLSN justice_l_Page_064thm.jpg
c21fcfd4c10c7a7d20bf86606d5461db
ad2a2a038ac6deb992a610d026f59008f37e58e7
8529 F20110404_AACLRY justice_l_Page_046thm.jpg
4740fc218924cb0f75ff23f9451534b9
ac624cc681517610fd40fbd114000532283f71b3
F20110404_AACKOW justice_l_Page_003.tif
49e064db2d56471b34ea125228d72f63
359ac5e91bee5723c9e0bb9c52726fa493d7b4fc
F20110404_AACKPL justice_l_Page_021.tif
2da200ae93d1dc7d9e81fae8680ea0aa
dfa76c06d5c83e619e4f5ecc7e3c96c3d0b56493
6690 F20110404_AACLTC justice_l_Page_080thm.jpg
062ea78f43a5b645c4e8c3276c19e485
f03510efad88ab2ef4a7ee46febf57d8cbac56b8
8752 F20110404_AACLSO justice_l_Page_065thm.jpg
c2bbe8d5a3f32a8428f25784b87b1d27
a5933767955bc4d267fff0aeddeebf76edf08f24
8494 F20110404_AACLRZ justice_l_Page_047thm.jpg
3516f456eb8226759666e08ff6058e17
355ad7707d12e1ff44c03142c30f6b58faba5bc9
F20110404_AACKOX justice_l_Page_005.tif
303acf5e1cbfeaf78de72c94e8271504
47b095a0e9217d1c6d686ff6274a6cdf47fd542d
F20110404_AACKQA justice_l_Page_037.tif
6b08f731e8d78994a509bd53c95b8d40
dd02713c40da281e02d9632252b70f126ecd4643
F20110404_AACKPM justice_l_Page_022.tif
81dca42a7c49d1be052842bc79b48564
9f30cbcad417a9606cd0998d1803e78bda3ff95d
7689 F20110404_AACLTD justice_l_Page_081thm.jpg
acbe5d497e86e6f19d193070b9b7db17
de3ccde5125d027315d5b90386b5ef2181c4a9e3
8637 F20110404_AACLSP justice_l_Page_066thm.jpg
bdad35faa5d25797314f4946896775f7
fe019c76b0719d49d268ccbc8168385c80195d7b
F20110404_AACKOY justice_l_Page_007.tif
233845d37875d0960736837e5809a282
7d36f7519f93ed2085d9af4adf82b69f1f84e0c2
F20110404_AACKQB justice_l_Page_038.tif
e0ff8ff06b320d8aaef8e41019645f65
0f37aff4df1b2ac2b62ca5e94019a99881a428cd
F20110404_AACKPN justice_l_Page_023.tif
409ca180873a0f802d10ba46818c64f1
c6cdae7e28bd52971874b434315757aa9ec9678e
8623 F20110404_AACLTE justice_l_Page_082thm.jpg
b28ecb0a3f5fb460c4d9e5560cbc310a
4e0ac0f2ff835a1f06e5249e9e55c2b56ff45baf
8378 F20110404_AACLSQ justice_l_Page_067thm.jpg
241c8cd61c090e58749c35e976eb24b6
920ce59a7519fb1d7aa56d0684eebbe9d91f0951
F20110404_AACKOZ justice_l_Page_009.tif
b2d5b0befc87e637b33d3bd5b98d5451
71c42edda90605bbdf7dad45bc57267492021cca
F20110404_AACKQC justice_l_Page_039.tif
37d0dd17750e0abae1da7d1698b79778
055470aa953795b6461e7f2614c258be49844a3f
F20110404_AACKPO justice_l_Page_024.tif
43f25883931cd577bd1e30598d8f3cac
972b5c731796f0213ec9e1d68e837008ee3b94d6
4799 F20110404_AACLTF justice_l_Page_083thm.jpg
41db190b27d205fd1fc5648102afbaf3
6be6ef7a0be456e85bf447b714b5e514cb205a50
8803 F20110404_AACLSR justice_l_Page_068thm.jpg
6aecb84da910de28b623da069de62d4b
3a5f50fadfcbf02544001cbe465e8423f2b0bd41
F20110404_AACKQD justice_l_Page_040.tif
9c37757f868f608df2311a15e3e7e953
531e3a13296d5284e63de4077c8ada42c7f2649a
F20110404_AACKPP justice_l_Page_025.tif
cdc5434b7e05a5632b5f4a60513978dd
addc0a0caa2f6a227438075a49346c50fd09f2d2
4643 F20110404_AACLTG justice_l_Page_085thm.jpg
c61e4a4d69994ba5a64dd5b33a0accb5
7a2e0a948de40cb21bf445d891939b1bece99ec3
8874 F20110404_AACLSS justice_l_Page_069thm.jpg
e9d45514523e5cc5284d37adb9935002
11d1e3418163be9c2245fe8f758e55a133b47a74
F20110404_AACKQE justice_l_Page_041.tif
5a3c0c90e2fb52c747f2cbcd17ec6994
bc0d178de7cf2e5b2554e88ce74ae5eb8ab2eb51
F20110404_AACKPQ justice_l_Page_026.tif
1d139e2b331197adb0e62ea6add33e7a
fa4ba503251e482d86ecdc6c9b71272dface9e83
4911 F20110404_AACLTH justice_l_Page_086thm.jpg
42cecdc500a6d785def31b6f95fb07f2
84b45239a181b70854e0697fa1af49e5245c5141
4355 F20110404_AACLST justice_l_Page_070thm.jpg
badaecbce5adfe8ab5cae5cee8b63d73
8b2a1e800ac9454c9748c4b91a2fffa10d3cbd11
F20110404_AACKQF justice_l_Page_042.tif
06d72e86c785ad60e654612e061ec4d5
6db10df0c2a710b49da0c1c643241fe50d3559aa
F20110404_AACKPR justice_l_Page_027.tif
9a59276269bdce352c464a3acf65f711
49c8851c6ffc1fd0d1df1fcf51d6bdfe33270112
F20110404_AACLTI justice_l_Page_087thm.jpg
df1a2238b944c156eb7f403dab9bad06
8b0fd44ced0ee8fbe80ba929a161d3ba615b98f8
8428 F20110404_AACLSU justice_l_Page_071thm.jpg
965c5d6db50002b9ac0c5088ba0415b9
96e03382b804bd3f64d997a1d5d86e207e8cc150
F20110404_AACKQG justice_l_Page_043.tif
7860803f7777e02c6dab2e32295db802
1787b251c226676890d779207e057c675dfd08d3
F20110404_AACKPS justice_l_Page_028.tif
7173aeefd00553adf4a267c567bb307a
e8001886398f1ce67d5dd4375ed8a3c86314036c
6555 F20110404_AACLTJ justice_l_Page_088thm.jpg
6ddb9a6270220805df936a05d708d13e
ec34c8cd2bca08fdbe504610b32b9dd12e9f8d32
7012 F20110404_AACLSV justice_l_Page_072thm.jpg
b97d1dd25df2df442afe63781b28abb7
da12533aa7b3e31568278dcb7cf1b91a73463e5c
F20110404_AACKQH justice_l_Page_044.tif
167e542035f83a81225ef55afa23e0d4
54b47e90b33b956aada69154632c83285979d819
F20110404_AACKPT justice_l_Page_029.tif
1379965e9719ba3de67866d1aac8d483
0868fb4658639799fa10e8a5a1dcfb6fc3ebc42a
4680 F20110404_AACLTK justice_l_Page_090thm.jpg
c3cf4f386fdff84916b6911f64b6ab7e
3ecd23f33aaa8653e27c964b8a1a9f03274cdda8
F20110404_AACKQI justice_l_Page_045.tif
ac6261c28b27e7363b219adc48de8b36
f7cf4b31c476098e12c74ca30ded9a08cf4568de
5577 F20110404_AACLTL justice_l_Page_091thm.jpg
e04bc8da0d5653493f86d39c5b29dc4e
230cd5890aaa9183a483804a374f146576811053
7736 F20110404_AACLSW justice_l_Page_073thm.jpg
813ffe65f3420bffff02f8a09abaa756
b134a6cd7fd688168c8b7d04c9cb970a06685dfa
F20110404_AACKQJ justice_l_Page_046.tif
44a1432ba309a68499fc76adbbb83b41
e83074474667dccfc5a11ce0ecff481e800bb0e1
F20110404_AACKPU justice_l_Page_030.tif
a284635802cfb5edd331d5d031cf6d3b
a4e6f81aa820226bde6b5805f96b7f77e9afe680
7903 F20110404_AACLUA justice_l_Page_110thm.jpg
9d7056472ae199b726f4d92e5efeed69
cf15492984c792c2e23671644796475b967a4835
4400 F20110404_AACLTM justice_l_Page_094thm.jpg
ce0f29e7f9d2cdc0323a49da73bd16f8
8f0e528df6deb67a2954d892a7d5f533c560acbb
8759 F20110404_AACLSX justice_l_Page_074thm.jpg
baa01f1a6fe1b85d1d6ea209dc6f1aa2
fdb614c7a233640eec8f7c44f266e36d9ceb6e33
F20110404_AACKQK justice_l_Page_047.tif
9a493cb57706860f5455cbb7705ca443
714a4530fedc67a4fbe4a5368137fc278b1705fb
F20110404_AACKPV justice_l_Page_031.tif
93396b5494bdcc8e142476ad0eab0978
a65adfc121298663945cc289af9d2fa1b9347653
8940 F20110404_AACLUB justice_l_Page_112thm.jpg
2817eb26f5a5bc71a11b5e24f01433f4
66eb762e0e40f37312e65bd4534e4505d827ba57
7363 F20110404_AACLTN justice_l_Page_095thm.jpg
97d7f5ebdd34ddbe6446dfc914eeb9c5
d9596ca77ebc20c9e7e697055cb13a119fbac7ab
6045 F20110404_AACLSY justice_l_Page_075thm.jpg
25ef7e3ca73923af9f03d716a7513f2b
ae9079f5ac3e8b197f9ace2184c373f5fce01af0
F20110404_AACKQL justice_l_Page_048.tif
798ad82981cca1416e760280fb5dc25d
dbeddadb3ef8be390d14ffac959672a3a39b831e
F20110404_AACKPW justice_l_Page_032.tif
6247f3f60f6d8a0aef12885734ef47ba
fe67563989f094c26534f0fde700bb8741972476
8178 F20110404_AACLUC justice_l_Page_113thm.jpg
d611f00617befa11999e5d8cb5120b83
dce2f26ebf60f38252c8539a31c19028203a1378
5854 F20110404_AACLTO justice_l_Page_098thm.jpg
0344a18edef92fe1710b12d8bec7d63e
b6f75a83a6e86032a8ef6ade16b483da3d82f490
6464 F20110404_AACLSZ justice_l_Page_076thm.jpg
a1ec7c24a8b539aefff9690c3c6d3be3
521cc65c758e579bba0e069e701904017e95682c
F20110404_AACKRA justice_l_Page_065.tif
5ba438b965a0b37e8ce04fb63d9da3a3
6dd91209f8eaaaf3239d0e9c2017b0403fe2f883
F20110404_AACKQM justice_l_Page_049.tif
06eb713b19b3e21e024fef1619fe6de1
0e4fa7e7832b164b7e036da8aeb2f3f30628da98
F20110404_AACKPX justice_l_Page_033.tif
3008cf98c3164800708c5cf0a4229cf5
59fe1cc8d48a71dafee81d2c322c76170f0f691b
7610 F20110404_AACLUD justice_l_Page_115thm.jpg
4d0d70a3149a638a6df2a90c0f883026
3ce3eef76f7240621011a753039387e6c20a16e8
4871 F20110404_AACLTP justice_l_Page_099thm.jpg
2a93104c107d299bd03846dce741c90a
f60826b11b31577e201ccf8a196c99eebdb1a9ed
F20110404_AACKRB justice_l_Page_066.tif
c76d170752711938d86a19570680c15d
7f96d01bbcba34ba64d3475158af2850dea53262
F20110404_AACKQN justice_l_Page_050.tif
208d28c94d6e39767f212077daf6c1fb
11d44d294d6891eb54758373b52c2b12a3bcc81c
F20110404_AACKPY justice_l_Page_035.tif
ff503b05b5273540a99618d07b79182a
c367b1c46fcb19dc0d418909f9ffef2d9e667d09
7335 F20110404_AACLUE justice_l_Page_116thm.jpg
f6bdce9da591556a1acca8f50c6ae003
6299d68278883f6fef31a32f8c98898d4cfcaf33
4760 F20110404_AACLTQ justice_l_Page_100thm.jpg
a1f161b131aad961cda842fa94c146bc
02e5226f66b4add3aee4f5a1168d1019b03c8e06
F20110404_AACKRC justice_l_Page_067.tif
8de188a25bcb14f3e87be6749b667f2c
b702107db185e29a202b86ec6727879c282399fc
F20110404_AACKQO justice_l_Page_051.tif
020531f7742c69c47fa3f89aff99e0e2
e165b3853704f189ab9f1badd345bdb268248725
F20110404_AACKPZ justice_l_Page_036.tif
65cb44aa93ddb7423785652d82a5eae6
2d30a8ec65cdee4b66fc07de86484aa7ec46e663
F20110404_AACLUF justice_l_Page_117thm.jpg
66d8b784bb7a417a6917f2a648ff457a
dfab39c24a31e792724d49206b2dfee0f51ee8cd
6480 F20110404_AACLTR justice_l_Page_101thm.jpg
208de25224faed8c845d1aa54730bcb9
e240bab07bc09a0bbbd476348ac0bed9fc40fb31
F20110404_AACKRD justice_l_Page_068.tif
1f560db74d5ca805b4db09fffd10f494
d412e9a10cf044d2cfd92feeb8329a39c209c6f3
F20110404_AACKQP justice_l_Page_052.tif
9c5b971d140888b19e941a3570cd2757
8a4c19be7981c3f6b260c97980c3fb11e4fd0ee1
5868 F20110404_AACLUG justice_l_Page_118thm.jpg
45b78bbb6a1b4060a8cb3962425a7706
19fd4d5b53ab7eb143350265eca99639c2b22529
4451 F20110404_AACLTS justice_l_Page_102thm.jpg
f2a45de1a5f1267c5b1e1e65b2d7e57d
6c723432ddf28a522e26c3649a9a0baf30076e6b
F20110404_AACKRE justice_l_Page_069.tif
76ca84be22624898d7464553fbe85b8f
d3af308f52c00722b7a08c0352ed29f7ac4feb33
F20110404_AACKQQ justice_l_Page_053.tif
8b3d8b8eb3db58eb6b2db92e2e8bdb3c
134534a70e517bb4ed22e17feb29202c95c59f17
5178 F20110404_AACLUH justice_l_Page_119thm.jpg
5303e0184d3c7502630d6983fa338a49
f0baec0708a7ae6f372c3518d47d6396d1686d5a
6605 F20110404_AACLTT justice_l_Page_103thm.jpg
85492badf0a05088b987869b278cb39c
3701f887aa82ad956870ab912a6e04479a498ecd
F20110404_AACKRF justice_l_Page_070.tif
9a5b90c391d07ea568f6eeb4eeac41d0
5737470d7aa62bfe69cace7eda6a789760dc00ae
F20110404_AACKQR justice_l_Page_055.tif
80a6ef5fee398fc5361c588e67029d3e
9cea6223a6e057bb3725e20dbfdc6f2e00e1d8e0
2396 F20110404_AACLUI justice_l_Page_120thm.jpg
d42870e20e01dee0398934083ce926ed
26142da584636a5201d9d86ce1054071ae64be85
5677 F20110404_AACLTU justice_l_Page_104thm.jpg
9c5af4136f3407959c6d1b5dd62c71a9
018d9d0199a3a509e6a58aabcd5655cd389c603f
F20110404_AACKRG justice_l_Page_071.tif
995086618907349c2e53470289cbea5b
0575b65877e06b9ec5d64c8405328929f40a19b0
F20110404_AACKQS justice_l_Page_056.tif
fb19259fd89ffa8b68b0ea19f57594a4
958d3eae388c6bf7597fe009ee58dbc47bb6d8d9
8316 F20110404_AACLUJ justice_l_Page_121thm.jpg
051520ec0a99a18ad46456c12ddb784f
16201fdd6105608bc188204e203a64813ce3b9c1
5602 F20110404_AACLTV justice_l_Page_105thm.jpg
2bb3f4970e17d2de1d1b47872e4ea5c5
a4ca0f4edd2db504d916e4f4db8d3b5ae8582d72
F20110404_AACKQT justice_l_Page_057.tif
8ef2918180c9f67e01302ecb741597fa
1550e5831b8174ea2eff107d12c271b104e33c13
F20110404_AACKRH justice_l_Page_073.tif
62e8d173d72634bfb50372d89a43b9f2
bf4362dbb0078c1b7b824cfe48343fd19bf7d631
8601 F20110404_AACLUK justice_l_Page_122thm.jpg
e7c83293e61d3dd787f66692be732e7d
d0155cc980ff381000a8fc5b15c306b8931c5cf4
6349 F20110404_AACLTW justice_l_Page_106thm.jpg
e781cb8dd2000faaef9c2030f42958c2
f607676adb4121f58fe8fadc25aca2b2c85c486c
F20110404_AACKQU justice_l_Page_058.tif
810e4a431d93fc4249e4238a3210d43b
b5d53678ec6e2d7cabb86886f950f66ab08db0ea
F20110404_AACKRI justice_l_Page_074.tif
130c579f067aabfda19c4643a86a8b33
c948f30a69c68b500f2c6e5fc11538e966bb2a9a
8897 F20110404_AACLUL justice_l_Page_123thm.jpg
ad052889d4f58aabb05ee8d0019df3f9
2b4b480616f2f92e1134f143f8831a0961ace464
F20110404_AACKRJ justice_l_Page_076.tif
0ea07b478266afb4c4c22cd39fee10ac
94db89d75a1e335c644a50033150a62746f6ea32
8758 F20110404_AACLVA justice_l_Page_140thm.jpg
12b711fd712977044575e8be4a661bad
e5852b9688e5eb0aba0f1f523f81995f91bb3e83
6474 F20110404_AACLUM justice_l_Page_125thm.jpg
fb939c1a20feea1e6f8466467711fdbc
d1e6b8d5a2a7bd7a7349e9acfb7b7d02994e90ec
7159 F20110404_AACLTX justice_l_Page_107thm.jpg
3403dfcb275575d41a370bd9d7d05c88
ddf0c2b5313c1cbd611e927748febb22cff0e21e
F20110404_AACKQV justice_l_Page_060.tif
62d831932957479fd9395fc13631842b
f1f6ad1a19805417dca82e710431e64deaf77a6b
F20110404_AACKRK justice_l_Page_077.tif
b54abdbd53f55ee476792ab7b9379711
11b8b51203cb823a7294e5174173fb2808680e37
8779 F20110404_AACLVB justice_l_Page_141thm.jpg
9f85d9a40168c1772d1690fa1b13f51f
7faa2fb361b8b2967343934875e7341b465719f6
6940 F20110404_AACLUN justice_l_Page_126thm.jpg
796b58d97e558f083129a974555c4f5b
5a6e0bd3ddfbc5e98afc2f108039de5c06759ac0
8578 F20110404_AACLTY justice_l_Page_108thm.jpg
d6f2146611a2c24eedb923fa002f0bbf
bfb5eab901a1fd846323fbac9dbb02b8790e5ba5
F20110404_AACKQW justice_l_Page_061.tif
03c1804b8a045149103c326f3161e0cd
66ca53329183e0582697a1019ff10074d80e7518
F20110404_AACKRL justice_l_Page_078.tif
77ca49d8ba4d626f92c224730fa4c9ac
6b186d56439127c86d96c64f70bd894c413398cc
6699 F20110404_AACLVC justice_l_Page_142thm.jpg
2eec711bbb7fcc2b5ef5cea29a56b7f6
1b7bc8ff928a19d36e4d9fb6a9e713fbaae05e23
5924 F20110404_AACLUO justice_l_Page_127thm.jpg
a65a81cf6350bd2f3551a206e62405db
85fb6032f0e061eab888ddf707c25af4c48d03ed
6415 F20110404_AACLTZ justice_l_Page_109thm.jpg
bb70aa1d1e9b534f232ebca754fd5437
9248e178287de97c9dc8a534b8f19e9feda0f917
F20110404_AACKQX justice_l_Page_062.tif
9982f2ce4dd31aea21b6d52da3ac9468
9ff694a1001fe906952cfb98a8eb40f1b069c69a
F20110404_AACKSA justice_l_Page_099.tif
50525bc5951fa7dce190657286106f0f
8edc0e72568a3214b9300319f325e30b4c78497b
F20110404_AACKRM justice_l_Page_079.tif
1cd022e960304f0d1f13ee4444bc06fa
3ea6556b14db86fe321f548944b6294db1eea3c8
2152 F20110404_AACLVD justice_l_Page_143thm.jpg
19260af214c0573d8b7671753de0c174
a60cdbd4c171d03c0b4ad293ef5e0c8f0397de02
F20110404_AACLUP justice_l_Page_128thm.jpg
054f67355921bf49f162f3c0bed5ac72
78338dc483571245627bed2ff4c26a87c48b034d
F20110404_AACKQY justice_l_Page_063.tif
b827de6939fa11afd418580a2f854191
fa079e0b4a76545f03d3fe59dc66fe39bb77a7c6
F20110404_AACKSB justice_l_Page_100.tif
9cdfdd5a5f936c18a95c59f066ecff8b
bdc3179ff8e2e85c2b3fab0c1f7684aba0b742e7
F20110404_AACKRN justice_l_Page_080.tif
ad209594a6da40f9464df6d14949179d
c6a2d606b1bba46bd4fe78f0d1152f72402d243a
169257 F20110404_AACLVE UFE0021174_00001.mets
3fc9f3ecc9d9d9bbff4b6adb3776abaa
08f67210932e982e3facd4444703dbbe7598f2fa
justice_l.pdf
3254 F20110404_AACLUQ justice_l_Page_129thm.jpg
d256e9a488012ae8afd7b16fe8896c3e
40dfcee677858c5ea3b930e3195d9ad42a2f793d
F20110404_AACKQZ justice_l_Page_064.tif
52e8f9eec4e5d1bbe89950e42b1f93d6
102da0578ea980ad9b2ac33c17cf49861fc7f4e9
F20110404_AACKSC justice_l_Page_101.tif
738c2dcf17d97cdd08e0660ce6214951
4932fc8d245279e3f264819b5f258e7503924879
F20110404_AACKRO justice_l_Page_081.tif
d29192656b238e78e415373815bb28f6
6572ab1eb1472aec29ccc1eeafbf3e44c68ccf3f
8323 F20110404_AACLUR justice_l_Page_130thm.jpg
95f94007bba2c93e95a6aa76960ca8c9
28640b22fd7227cf130c21499836218684e68811
F20110404_AACKSD justice_l_Page_103.tif
f3eb76ab8e76ecccf1273b073c5a80fc
3adcebe9ef5dfd4491b11569061de71baa0333b6
F20110404_AACKRP justice_l_Page_083.tif
6f072416da936cd79cf679d10188c22e
6e5a5a30a97e8bea1ebcba4f5a30ffafa59bb810
8817 F20110404_AACLUS justice_l_Page_131thm.jpg
aac5f40dc32eb55ddf5cc7681bb06d85
16b2a7e3437dac72c5db60459a495804b5c52e47
F20110404_AACKSE justice_l_Page_104.tif
d8ebd977444ebe7ec1c59846bdcb4ded
e636bae45b59edd2e617229f97d30681411ffb74
F20110404_AACKRQ justice_l_Page_085.tif
9749cbd8080eb9f7f032b0f5553fd286
33d73232c19653c46963b226807e4bc282bdab93
9066 F20110404_AACLUT justice_l_Page_133thm.jpg
6779bc3772b977a51a367878dd7494e0
ec303214d483bcc2effb61b8e8e5b0649be6cc6c
F20110404_AACKSF justice_l_Page_106.tif
a1ec12e3080a3075b396b0d41405ed98
084fcc1f4e2e40728b5eee618b508ab56617f983
F20110404_AACKRR justice_l_Page_086.tif
3d9f4b23e6d7f1618dc140e6eea70030
a980e8210c864368b21c01fe7210ce7fa4869b83
8767 F20110404_AACLUU justice_l_Page_134thm.jpg
4c28e4614d1fd3b37ecd21998c9acfde
bfdb0567a94fa198f2a822310aab196eeb3eba04
F20110404_AACKSG justice_l_Page_107.tif
f063d473868693a38b773c3e27a6be1c
1803680ff80f3c7c2c45f5d83ad88601c12c2649
F20110404_AACKRS justice_l_Page_087.tif
5e22c5ee7d96faacf428ded129f41645
0a16427b89fdcbfb328e9507e4ffa936fdd9cfdd
F20110404_AACLUV justice_l_Page_135thm.jpg
fdd2339a8cf3618774b5ed9c997a3c73
447efcb4dccb7792b87189775ceae92ea857f9aa
F20110404_AACKSH justice_l_Page_108.tif
6c24fc92a8302a67128f903a15906573
2b5ff07510ac01e381f5165d7f6d6b11d5e3bb67
F20110404_AACKRT justice_l_Page_090.tif
a6ef92e216f3919bcb60bae30b0d4c80
63ac151bd45f967fac6d107b47caa2117ab98e51
8661 F20110404_AACLUW justice_l_Page_136thm.jpg
535ca4e090f50d03d49622ee53e0ff74
845728cbf385054ce04c639a78ddbecc5cd89dcd
F20110404_AACKSI justice_l_Page_109.tif
f3f6b91a497a5a2a866ce166e26df65f
28ff679ff949c90e0623ef73e0b12e73ae8c87c7
F20110404_AACKRU justice_l_Page_091.tif
f8b740c7dac380fc7f1bba991d8678f7
50d47a28784dae42059a77c827321343c9c07b98
9334 F20110404_AACLUX justice_l_Page_137thm.jpg
258bde3562d211e0d842241165a6ffb5
ef576cc5cfdf599cf2b7a4f9eba3ceda4737ec65
F20110404_AACKSJ justice_l_Page_110.tif
b341d6e6881cd0ebe15a7250cc0d42ab
2c000a543294240c64c049bd500e11059f6a5d80
F20110404_AACKRV justice_l_Page_092.tif
4f7bd089b4ba8dd843c7a13184505062
a69fdf8406167b1846662eb7bca3dcfc11723e34
F20110404_AACKSK justice_l_Page_111.tif
2eabbc5c09407a4c1b9475399fda410b
7ce60080cf620887b4d65f1f7e5e28b67cd8d155
9053 F20110404_AACLUY justice_l_Page_138thm.jpg
924985587d7a9f3072a6b5c6b5a4e4d0
5b52db851b9d7754f9366e0df62cea41d2d750ff
F20110404_AACKSL justice_l_Page_112.tif
81a8e81ad0ff07339894a6837cd4cc0c
15302fe1e0c06797d0da04d6cef7345c488b4a87
F20110404_AACKRW justice_l_Page_093.tif
6dc152b9dd9e6bdc78bbd783858f8d1a
08e198e57ed85b2098c77860c00f1e20b861d486
9422 F20110404_AACLUZ justice_l_Page_139thm.jpg
69317bb60ee8033b371ff122233d4ca8
c00a6ed9df71ec3abffff79f463b3a65fa06b0b5
F20110404_AACKTA justice_l_Page_131.tif
f127371884da2d24fe9cc5519c3085f5
7376c0ccdcde0281d4490537d94dd0ef50bbf601
F20110404_AACKSM justice_l_Page_114.tif
13ae4a2da0115e3ff2c26a59bc7ecd34
20aa54e79a85e36d9821316367b514373e698bdc
F20110404_AACKRX justice_l_Page_094.tif
5eb2eda11b44711867907200f96cde95
7c805e95cff926d97ee414afb77512216f9ce460
F20110404_AACKTB justice_l_Page_132.tif
40ad778d047723d3bf47534fde7a0684
4285b8a68dc41bb58e1dfc86a9ebf288fe6ff770
F20110404_AACKSN justice_l_Page_115.tif
a9133f84b093565276a981766e57ed48
dfda503a4277158e6c96b796bb27e060689d9e14
F20110404_AACKRY justice_l_Page_095.tif
15ec52a90a7322761e32af7459573c8e
32c73a831a9eef0241032cb51766961df3dd287d
F20110404_AACKTC justice_l_Page_133.tif
d8a83ef41d017f11f51f4c09961a7d86
f2070ca9788b5dc8c840d3fa2502c1c2023b94f5
F20110404_AACKSO justice_l_Page_116.tif
28d6686b27d0c4ab37bb6aad8f1e3a3c
c1a69ec082c5817a29e106dc545200f4265e8224
F20110404_AACKRZ justice_l_Page_097.tif
624b5dd561bbe0a9bb48bb58aa1f14df
913abfdd8d6de62e890909c9c13e4b1c84ba178d
F20110404_AACKTD justice_l_Page_134.tif
ad341d8c5df5602fa421d6c6fd285897
ffc4ec3134838abb13d5f3c3ac03e7133e444e57
F20110404_AACKSP justice_l_Page_117.tif
015d8be855d2cc7e71e763640cb7214a
2a19381869ec1c6ba4cb863bbd298797b32ac839
F20110404_AACKTE justice_l_Page_135.tif
1b5db67d30807dcbff22b5cc54aa14db
570303c4ea3960f53f30ef64478e3f0d3ca0a83e
F20110404_AACKSQ justice_l_Page_118.tif
04accdb59d9be92fa50728c419114099
d3f542acf0e03d6d06710c083d680449749cde1c
F20110404_AACKTF justice_l_Page_136.tif
a0284c815dcd4886e76a3bf291c9533e
6232af53ccb14924beff25c5fe01b413643f9153
F20110404_AACKSR justice_l_Page_120.tif
756d17716def9e573cd3db872d113b20
92403bd109592829581f8782a068efa40eaae89c
F20110404_AACKTG justice_l_Page_137.tif
6b90477809438b8e7cd4640aa9130f65
12ec45682e3ca0d913f9c5eb16786b5542e8966f
F20110404_AACKSS justice_l_Page_121.tif
34bc875fc00be35d2c06a68e8922b1ae
dc68cd300da6872449fa409ca48003eb1bf8dfca
F20110404_AACKTH justice_l_Page_140.tif
6ef3955d73f8f57650bbaca322d0ec49
2d10d1547a23979e0eb635704014af2ac64e7c6e
F20110404_AACKST justice_l_Page_122.tif
179c62aae5ec0195fe6de2fcfab841ad
c181c6664ffc7431ba4314878e1853c2a6bbad47
F20110404_AACKTI justice_l_Page_141.tif
bf3982afa051606897f2759de69285f1
8e958f64485cb862617cdc5bdc7af91391a81695
F20110404_AACKSU justice_l_Page_123.tif
d7ad140552652e51f23705847c3316a5
1773bd0a4999bdeb8826f9dac2ffc249564adc82
F20110404_AACKTJ justice_l_Page_142.tif
683c3f5366033cef878987ff9dbc88a2
044419289edc20ed79a90ca7f78bb444f3b69590
F20110404_AACKSV justice_l_Page_125.tif
8590edff423926f3fed98892d025dd5d
7309285e98d1352c3b5a1d3fb23e37b37181e59f
F20110404_AACKTK justice_l_Page_143.tif
7b0b84ad7c15544728053ba5874f17d9
53a3e13d79b375bb05049e01f5090465f0eb6ecc
F20110404_AACKSW justice_l_Page_127.tif
5b24765aa58290de07e0641ebeef9c30
2ef9d534c034906ddb58c5cdbd757b59dfec6203
507 F20110404_AACKTL justice_l_Page_001.txt
e37460dcd95bc0a318bd44ce44790ce0
5f1676b0afbc0dcda6151e371b1a19b00ba42901
96 F20110404_AACKTM justice_l_Page_002.txt
bb7558858c12e650ffc6a98edee0e1c2
d660ff687c4bdca177daf81754ebec14813d7a0a
F20110404_AACKSX justice_l_Page_128.tif
c42a04118512b2c7185ff08ae543fbca
da298e7442549dbfc53c845969232e58d4bb48ff
2066 F20110404_AACKUA justice_l_Page_020.txt
ad412c8500eb2fbda3d5896092762c83
31bd27d888c3fbff664e5847b7791bdf86efc842
1814 F20110404_AACKTN justice_l_Page_003.txt
a3d4da556f8c8439e8e79e167c9ebaa6
1797fd5bfab2a4629c8fd351a9763e99f9bc33c1
F20110404_AACKSY justice_l_Page_129.tif
dad623c069f1c4f3551114a9e546227e
e5f495080bd8e70a3a84a2adeb78c5292282beb0
2110 F20110404_AACKUB justice_l_Page_021.txt
4243fa6c7523604d3b48c222e98536c8
026f209337ad64835a00fca5cac64127415ad481
3973 F20110404_AACKTO justice_l_Page_004.txt
1af134a5d1d43c1e28f88750a9b4c293
149de64df9be65d08a037dfbda32fde989f37b58
F20110404_AACKSZ justice_l_Page_130.tif
8b4ead97750e7feb0a111c36eaab09de
99d98e68f19b923a588d52cbd3ae1d7e555d5eba
2037 F20110404_AACKUC justice_l_Page_022.txt
50fc17122917b9f67aaeea96c4b8e8a6
908f89454b98b4bea114fc4de559df793015c88f
4993 F20110404_AACKTP justice_l_Page_005.txt
61e3fb2b4a208b29d097800a5a561bec
33d312755d54dd248749723b6623d8f2f0c2ed00
1886 F20110404_AACKUD justice_l_Page_025.txt
ef0c963aeb9d9beac07d269ba753090a
1e3a1c1e7b00b4ed915d3385502c1bdd35d77fce
2035 F20110404_AACKTQ justice_l_Page_007.txt
dd458848fac1e94f7eb8d0c094936ff2
abd506f6bfed7f2751085f33ae2d23bc789d2334
2315 F20110404_AACKUE justice_l_Page_026.txt
58136341bad377bdea9495417bd37e40
35c3e7ec9c8baf988d5b2822889db13bff8aaf95
2855 F20110404_AACKTR justice_l_Page_008.txt
e14102d35764a08a789f926d23a84f61
d02e31c4a0e75c7c66a1153c87ff7ebee04e3c2e
50554 F20110404_AACLAA justice_l_Page_067.pro
371413793fa648c84e498b85b13495e5
109bfe1e6c2f8b501d032f0e84760cbcda38c1a4
2199 F20110404_AACKUF justice_l_Page_028.txt
e67f4724e977200ba9ed4d05268b6be8
c55bed79513b2fec49703ef6955ddd4ece079b62
3028 F20110404_AACKTS justice_l_Page_009.txt
24c7dc8f476ade8a49e8b05441596a3e
c494ef72ec426182d55ad3d1594941671f05685c
54864 F20110404_AACLAB justice_l_Page_068.pro
4884f9bf767568f373c969ef34ac7280
cf5a66aea2d1a531fe320c9f60da2a00e1bdc060
1818 F20110404_AACKUG justice_l_Page_030.txt
dc3abaa63d7e13326210baa5af9590bd
df98e9504ab08511177fb3c1b59f1bc570cb23dc
2993 F20110404_AACKTT justice_l_Page_010.txt
8210152389cb71c3bdc30400781e67b7
d0efe8c4aa25061d5c6ba65004cc48f8a69fd1da
67987 F20110404_AACLAC justice_l_Page_069.pro
b97f4f59335068c279391e429d11ec84
c1269a31c6e0bee2dc49acca45fd97380ac720e8
384 F20110404_AACKUH justice_l_Page_032.txt
3b6e967aca94a1dfb7b9d16fad526a11
067498bcf8f79948a53faf6f4b6042f3c8143859
350 F20110404_AACKTU justice_l_Page_011.txt
e36b449ce4dadd1ea03709882c631ff1
71cea00f761ccdd4d086db79a9eaf27e6ee8ed18
50698 F20110404_AACLAD justice_l_Page_071.pro
04361e2c87f338abb191e23875107dce
ec12afc3aba679bb36d0c16e1444bf9694d9a132
F20110404_AACKUI justice_l_Page_033.txt
d0b5aab7391de3f1b5deffcf3ee0b6ca
f3738f49a51411fd7a6a7c6477bb8870e351202f
202 F20110404_AACKTV justice_l_Page_013.txt
e4d2ce901ee13640c89cc2fae30999c9
ccef68e12c1da9d63d57511e7bcb1c09105febff
2075 F20110404_AACKUJ justice_l_Page_034.txt
6bc98a793a3d5ab24b2414d1cde27b64
abb3299f8cea8423504f2a1fa2610bb59d521693
2362 F20110404_AACKTW justice_l_Page_014.txt
74188391d0f6419ca330d4f6449148bd
abddbfe33b64b17f3567e6313dec49d80aefeef4
48099 F20110404_AACLAE justice_l_Page_073.pro
34bb6b534cf2dd5b05549d425b33a07c
1fcfd00450b6444052bab9f3b5e460c70038dafc
F20110404_AACKUK justice_l_Page_036.txt
758ee83adb76e6f6e6206f421c483e62
3f30745eb0825469455d3fec1f496cf128a6ba63
2166 F20110404_AACKTX justice_l_Page_016.txt
68896bf4f88238b858b364cccfe0ffb0
5aebf97d360148b8e5f17b8bf74e2537e0541464
54490 F20110404_AACLAF justice_l_Page_074.pro
6c09f21cdfdeea9f010be6ef1e1fbe2f
04f6e975fa3214dee62598adbc778e396c15dd1f
F20110404_AACKUL justice_l_Page_037.txt
8a2d2465224587d13668945e5714efc6
d77d5b516a1f21e68a0989aef380ed7426d7aa07
39267 F20110404_AACLAG justice_l_Page_075.pro
cd38c9a38a5a7cdced044e31f4a3b818
5dbbb37d1a8c9277cf6875aff37b41cd4544c978
339 F20110404_AACKVA justice_l_Page_053.txt
ad68e40566fb1c41c9943fa798f9e770
cea82a8d587570120aa0299f952c9cb2070bf945
2366 F20110404_AACKUM justice_l_Page_038.txt
68e81b20547ae9098704e9786fd3a513
19b29750742d981dba719fc520203360f3e74f56
1325 F20110404_AACKTY justice_l_Page_017.txt
84da49373fc9c33e945cd0060a5fccd7
95f252c03fd2df2b7e570abee4ded9670fbd9cd9
29750 F20110404_AACLAH justice_l_Page_076.pro
0f2cbd472387e91e23fb06e50b6ca421
37e3bec22adfb7ccf2cdf49be8a19cd75d2726ed
2170 F20110404_AACKVB justice_l_Page_055.txt
8aad573580992c9167cb2b01d8d8bbdc
bc16d231e711482d4bbd57602e21f2eba2e237d6
1879 F20110404_AACKUN justice_l_Page_039.txt
af1fc4f42f94c3850e28ea54db39ac8d
2fdb592c95e56c65eb07be7e2398a686f15eba5a
2064 F20110404_AACKTZ justice_l_Page_018.txt
c515a7b08c38685110da4a42c81d7d36
f5424519c7261e19d10049acab0537bf05631404
58950 F20110404_AACLAI justice_l_Page_077.pro
523ca4fd8c8360386d54948d54f23ad5
2224a84744f143f40a1787957bbdbe2a77a9a36d
1343 F20110404_AACKVC justice_l_Page_056.txt
9d4ddbc373e5c045e3aaad8f6b6a31b0
0de5cb110aa5fb95cd60a063665cb43372feeeb8
2129 F20110404_AACKUO justice_l_Page_040.txt
56f5b74f383801a58b1dacb96e1b7ffe
70c576dc10f8df0fcc2d6baf5d9775f2bd786d11
43493 F20110404_AACLAJ justice_l_Page_078.pro
4474e41fa1337caad4831aa8c8e99458
1536a371a75e4dac6795893e20dadba0e674253b
1759 F20110404_AACKVD justice_l_Page_057.txt
9291d75a58dc28f08c8b59a93774dd17
6a67e459301384517736c9c33ad49f6d34a7037f
F20110404_AACKUP justice_l_Page_041.txt
bc1115520d25b965ef4191d141922ab5
ed5cd250308d5536c9f809ab04dc5d665fe1c6b0
48039 F20110404_AACLAK justice_l_Page_080.pro
e7916bf560dceeec21a02b17a76d2fc8
28a8f8cf9d7f2c427c50034d10b1c2a860d9f3bb
1429 F20110404_AACKVE justice_l_Page_058.txt
328ae2bdd296e96b1aee72b2088afa17
01a98899e090998ae442b4e704d0ca4271bb4521
2054 F20110404_AACKUQ justice_l_Page_042.txt
486ec9087f8b2266004bdeef7f43f12d
6aa7895841e31274b7301421eccae4f63ca9ebc7
52199 F20110404_AACLAL justice_l_Page_082.pro
4a952a36634092365ce2375393699d3c
108b3a2994de98d6fa89ed4eb38040cb9fa90732
2119 F20110404_AACKVF justice_l_Page_059.txt
eb23905e175111b27fb7d5d174df689b
a049732be807b07add354c65646897c342e092d0
2125 F20110404_AACKUR justice_l_Page_043.txt
f3301a640fac51d7d27c422626a1766e
2f84204b9d77f2d1a358159966f9be330f93c4b2
33971 F20110404_AACLBA justice_l_Page_101.pro
b7a59a74b0490ac5ec095544213f2532
c1c46729f9319f7786b331823ad7d3224f1d01bf
18117 F20110404_AACLAM justice_l_Page_083.pro
7a4edb5794a8d78bb7f16be324d67c46
5b1f1a209635038afe51ef5080ffbfeff4c77748
2603 F20110404_AACKVG justice_l_Page_060.txt
ea5e063d711a89e734b3fb5705cc04f2
16f9379d294357e1f2b99eaee5735536c3f51e8b
2002 F20110404_AACKUS justice_l_Page_044.txt
62632d983c398ea2d64ab73e03d50081
d873e4649b500e1c4d8e1dc8fa1ed929c11fcfb3
11888 F20110404_AACLBB justice_l_Page_102.pro
f9a30cdee246faa85eed5ec139cea59a
001c5db04fac60b6738fdaa2b9806b580d353d9c
20821 F20110404_AACLAN justice_l_Page_084.pro
3cd1bda3b472ca6f5f07aef271b17070
d004fcf0395f58fcd45e4aa1a1612568b699d79f
F20110404_AACKVH justice_l_Page_061.txt
23a35f50a43a7586804f6d5c4a464b2c
92ce80047bf7aed404ce59ba3d8b8a7959194455
2081 F20110404_AACKUT justice_l_Page_045.txt
1d3994450017424bb8d95924e7ebc5bf
818114bccb3754d7d491f4e5e93ab904d4e67941
16384 F20110404_AACLBC justice_l_Page_104.pro
8b2cb17d97852dd994ca1a123eb5511c
5e89cbcd047607a39bd83cfe53a1dd05accd4225
14454 F20110404_AACLAO justice_l_Page_085.pro
f8d81865b1579b8a25b02f235790ecbc
03fd89213dd8ade848f2fb78f0c7752aa4185d0b
F20110404_AACKVI justice_l_Page_062.txt
bc31f7f4cc3b467b4139b521f72e1044
247aa2ede6caf825d974f3eb0203570376c165b1
2104 F20110404_AACKUU justice_l_Page_046.txt
38716a52640c4e2d5539a09a297410bd
0b22ec3e2f02fcb97c3edd27bd151083f438bfc5
23668 F20110404_AACLBD justice_l_Page_105.pro
0ee7b6d1225f89b46d64d8a8fa96bafd
4b9b09b983c59c050c83471daadbde7a0a34d3ef
15493 F20110404_AACLAP justice_l_Page_087.pro
144a77513fb922a12d620f08c22cb20e
f5c1c43b6c0def7b52513a97be27c935c2dcc344
2025 F20110404_AACKVJ justice_l_Page_063.txt
f236033ce00129a99345d23ccc42d5dc
80df8e24077b3895563edf551633ab99cc7124cc
2071 F20110404_AACKUV justice_l_Page_047.txt
2952f5b3e895914c3955d560b758de4a
775ef3291eb641413eb8fbd1f0e76298e38305c1
25741 F20110404_AACLBE justice_l_Page_106.pro
3d808f94eb42252cada0895dee304546
c3bcbdfc391d1e2dd2a4f9f5f4999a27b9f634b6
34092 F20110404_AACLAQ justice_l_Page_088.pro
391993c44d4dd4228d7f7a9c83b13e2a
931232c44ca4332619ae4416dda0e36c1cf5ed00
2181 F20110404_AACKVK justice_l_Page_064.txt
77f4978b02126398f3653d6a80bf6445
e4daad89064f9394975d8254bc83ed43a81d8c57
1256 F20110404_AACKUW justice_l_Page_048.txt
218938a89cce3383e84dd1ab9fc51cac
097e9abbcc81c44252ecfee5af4cb4e1f60d49e8
29351 F20110404_AACLAR justice_l_Page_089.pro
9ad25d995480fb202bae52b2232d64a0
e0d0305adefbe1233db359116537797cb766621e
2131 F20110404_AACKVL justice_l_Page_065.txt
fe66334b5019f71a36316713a31105b6
61454271eb166d19331f0c993f06dfba5723fd4a
2528 F20110404_AACKUX justice_l_Page_049.txt
c38e997c1ad6a8b8e1791fcef644d9bd
c33e55b16daad169f3bff9bd352cea9f0f63631c
39277 F20110404_AACLBF justice_l_Page_107.pro
2b30c3d9f7b49f408259579b8661ddaf
30001734f0b43af411ebc73cef60427c82c40d9b
898 F20110404_AACKWA justice_l_Page_083.txt
6c4586c6927fb022dc43a197904b1694
e838e29575da8fd7766351d2838a3ecfc3c7f4f5
15880 F20110404_AACLAS justice_l_Page_090.pro
54047a55dbce017161345c937ad33879
81e4d9849d08367b1bc22a52bc04c35ca2daeb65
2003 F20110404_AACKVM justice_l_Page_066.txt
8d37df73948993806875eab7a40939c9
afb4fc34037a8e49fe5c2a1a53dc6033b1852413
1709 F20110404_AACKUY justice_l_Page_051.txt
79d2f89b98ad995da09b30d37b131b98
11b7c6c3af1f1173c62a341dc5d3f0e8814c2d99
51424 F20110404_AACLBG justice_l_Page_108.pro
eb2bdb8f81c6d92f880e0529425f80db
6109b75dc9e1640176f1b9cf340c58c2c726b209
986 F20110404_AACKWB justice_l_Page_084.txt
58c262de3a798fdaf2a665b6e9f58a5a
d8db9e794042dbdab796847e1303439b19984e40
25347 F20110404_AACLAT justice_l_Page_091.pro
95c58b39089fbd696377e8c49c4b4d20
e001204f623a8556983a95203897aa6ed67552b7
1999 F20110404_AACKVN justice_l_Page_067.txt
49132490bc3fda7540a794983d655260
e4dea9a9c8e95ef054b656b29091037076d6e4c5
53925 F20110404_AACLBH justice_l_Page_112.pro
655b188c59e1b16de70ba0ec35d4d28f
6e598ce94c68ab6a046cea4f1eeb194ab64a2fb1
715 F20110404_AACKWC justice_l_Page_085.txt
c629b288d1b397d4379fbd9fd30d11a3
9f3f0bd270d17892cca37d775bb3eb31eff09f80
19987 F20110404_AACLAU justice_l_Page_093.pro
8843fe9c0bbe1cd6fe937e0769683f6c
f21ac803b4cdcf38f481662fd3dcd2b5ccdf00ee
2158 F20110404_AACKVO justice_l_Page_068.txt
5d8e65a8401fcc4cc304d7813ae23b3a
415f5b54a08384d1a50d6eab9732affee87c2deb
1901 F20110404_AACKUZ justice_l_Page_052.txt
1e1faaff47efce5ad6e777d3e1ec83e3
31895ef0b7c73e352822381849f31821c1531b5f
50273 F20110404_AACLBI justice_l_Page_113.pro
d7fe9172321569e6c6eca68c24572508
a8c632058bfde9c03dfa4d74fee617abf886fce5
1063 F20110404_AACKWD justice_l_Page_086.txt
c7eb83cb83c2d4001f9e5a4dfb0ca662
26abb67ec7d032b165441c6b3bfa7f9c999b0d2e
15704 F20110404_AACLAV justice_l_Page_094.pro
f033bc82fadb25fe68bf0c8756d5976b
e7c601328526643511247e211003673c8b86811d
2618 F20110404_AACKVP justice_l_Page_069.txt
822d3951dc7f265d2410f5c66574b16d
242f48d0d7963afbf4763efa5fbf8fa28c873ef2
56554 F20110404_AACLBJ justice_l_Page_114.pro
82639997b12fec39515053796a2f1be0
301ec50af5d674287721d6f57f6fc19483c0a0b1
39014 F20110404_AACLAW justice_l_Page_095.pro
e6cff07511ac195de21e38690d738f72
7f874a75ed9d11ab8a979cf57bd4a1fa50f86d98
939 F20110404_AACKVQ justice_l_Page_070.txt
48410164896a3c9507050872b6ada2e9
d5f66cd743be53d3d4ddbf4dc22ea1229627c897
43011 F20110404_AACLBK justice_l_Page_115.pro
f77a3738d161ef335fdb95e2c74b98cc
853f32243c1b9ab641bcc2d08e7eedda7441d186
777 F20110404_AACKWE justice_l_Page_087.txt
91077506d86712067dfcc3edc879666d
df8648dbb8eb9e691b3b70daac1f9ea6f3cdb12b
25151 F20110404_AACLAX justice_l_Page_098.pro
6da7c26421265657e83e997cf3d8161a
22b5e0f89d92e8c7d3ae4781de4ece9cc4f59728
2095 F20110404_AACKVR justice_l_Page_071.txt
b9e56062fd0882dcd858fc3529551b45
4527f55d73c6077d5eabd7c611dbb81e83f31d67
15405 F20110404_AACLCA justice_l_Page_132.pro
9b32c77e9dcc3217b512c961314d574d
0b8305ec1af1c3fe4c478072960ef85a323fa1a2
43719 F20110404_AACLBL justice_l_Page_116.pro
fd6b803a2be8110d038a770ebaaa45f8
9ddad717d82fd8281fca68519c9e8da11a1a1335
1484 F20110404_AACKWF justice_l_Page_088.txt
833632870386ce2e3260bf77247f970c
6cf10680bd36e7ca4380766368173fe2610d7e8e
16813 F20110404_AACLAY justice_l_Page_099.pro
98c52f4b40a010b5f415420f4189f6c4
e79c28b9f235abc2c2e753b94f4769152b80d361
F20110404_AACKVS justice_l_Page_074.txt
40a9b77e3be55508be4275d80e10e361
19c08bee92fd502b714f9de74c092e12760e9e58
62271 F20110404_AACLCB justice_l_Page_133.pro
ae4154a3e8142c371e92e86abb4b907f
ebffd581c480429287932bf683952d191a0799e0
44892 F20110404_AACLBM justice_l_Page_117.pro
028d261524da562ba9c2a4d8946a27e3
49004c434bd18db7e8d0fb45151f36711d247c35
1330 F20110404_AACKWG justice_l_Page_089.txt
2456a815bcc81f0c927709089ffbf64a
9799402646d22075952d4a93a4c4b88b518ebbbd
11704 F20110404_AACLAZ justice_l_Page_100.pro
d9afd9e095eb9f6826fa4048058010e4
266bbf1cc84b2c9250e54114d78ec3a2cdf4ef78
1785 F20110404_AACKVT justice_l_Page_075.txt
122f7869e9fd66342d1febf7a975d60a
d6d87aa0e7ed30ce748c4f099ea455edb7353269
58066 F20110404_AACLCC justice_l_Page_134.pro
24da901f97742ecb9622abb7ff76167a
3a285322303a9d04228f2e560d6669e1213ee518
32170 F20110404_AACLBN justice_l_Page_118.pro
876e36f8d8af40e8fcf4198fa10ecd42
82db0b19852b83ef8a900455bdfd8a915464dfd6
942 F20110404_AACKWH justice_l_Page_090.txt
cd1598af9a1f386c68406ef5a1c4e31c
8ae2a8349657a4d3a2474b7f92ae3043607c5204
1413 F20110404_AACKVU justice_l_Page_076.txt
daed2515813b464a6830bb30417c3d33
e44ae395ba2fbbe5ade7f5b7fff7c521a0f13be1
59506 F20110404_AACLCD justice_l_Page_135.pro
80ad8f621144749b425fc43b5e432453
6b8b7cd74ebe94cc9597101932a7da7c479b38f8
14095 F20110404_AACLBO justice_l_Page_119.pro
6a4262aded3ce7d3c2f177f2a544e0d7
f4bc60047f2a0d3142a58399a9f4114b510ff526
1189 F20110404_AACKWI justice_l_Page_091.txt
dabf5c57ba21ffb76b4e97e6fb5ba550
cb72479b973ef3a8ec3948f9a918d183b3245977
1878 F20110404_AACKVV justice_l_Page_078.txt
2830fde3909b2c3a6534bfab5d3d4267
814e42e790e521017b291b1965f16cdce06fb51c
55039 F20110404_AACLCE justice_l_Page_136.pro
40ed3654e47fa8bfa48cfc9fba217077
c950882364304c2f57b48c2c22533c9b492556ab
7226 F20110404_AACLBP justice_l_Page_120.pro
de45b45129f1cc6317b58b725faad515
e43b90775e43ae509446867527589a313e2fa194
823 F20110404_AACKWJ justice_l_Page_092.txt
15950d76c9dc8e9be87eeb0ceea12bc3
a9ba0db733e0deb3ce1d996e672eee88b8527070
2208 F20110404_AACKVW justice_l_Page_079.txt
05f401c6fb760a3146dca416c596025e
9c1192cb9f1c798116ccf9fcd6f826e9eae82c55
63502 F20110404_AACLCF justice_l_Page_137.pro
e2da17194c7a765b12ed6ab3aeb3a8f0
e356f5e366a9ba84a6205778927b19e72eefb03b
51373 F20110404_AACLBQ justice_l_Page_121.pro
ec03efabf38f23e6ab260d024a48f9a8
32c884113af054dc22e51f5bb9f79edbe4a6ee62
959 F20110404_AACKWK justice_l_Page_093.txt
3d404342a7e3c2038eca2d7cafc4c2c6
f81f240219728cfafc444ae9cdb86f01ef199478
2416 F20110404_AACKVX justice_l_Page_080.txt
e2f6682df370ffe0a55763246998a11e
492c5f4b9a8b3bd949351f6fdfb1fbfb8243cfa6
53195 F20110404_AACLBR justice_l_Page_122.pro
76ede2156981a8d7c8192366ee967546
cd2d12dde0093e236d45ab121550bc17072e6bb4
799 F20110404_AACKWL justice_l_Page_094.txt
4a4ad350218a846b7afbc05ab6fd8da0
2f21c06eafb25901f3d203939d1792be55374c8d
1773 F20110404_AACKVY justice_l_Page_081.txt
196038c8a93a9271c4e1641eab5567d1
9c9c6571e904569575364b70d0129ac517527b93
57848 F20110404_AACLCG justice_l_Page_138.pro
b7a0b71f194a7cb7970b7042c393606b
05e39668c57d2322a39d3be5997fc752a1e70311
2443 F20110404_AACKXA justice_l_Page_111.txt
5224974cd42122e418a566a3be6af2ec
4d64b1d4a66ac7806a32cb5432541ba2c9eed112
56538 F20110404_AACLBS justice_l_Page_123.pro
1609d9aa094656e58be70fbca6f7c037
405feafa2623dc1fafad4d4c46c6db70e5d0d3b8
F20110404_AACKWM justice_l_Page_095.txt
3d5028c9ad943a58d094719a789ba959
4056af47b41f58a0090ab62e58e3e74231119065
F20110404_AACKVZ justice_l_Page_082.txt
98f3528caee1f86776151136ad067d91
3d04b60d29b1f95793fdeb60c881d109eb0a45fd
67541 F20110404_AACLCH justice_l_Page_139.pro
bb5115a40afb0a8f431e7381823e3bae
ffede3f896e8f68bfe516d2a80c89e6d8d2f4b8c
2153 F20110404_AACKXB justice_l_Page_112.txt
22babd0ea968956749ad3e24efc439f1
ea13fe9400b1c2069f3c748cbabccc393b572410
28159 F20110404_AACLBT justice_l_Page_124.pro
aa1bdd6963979ceaca3b40829978084f
f1e87f70cb53d6d97e5efedf4a98be39f002d0f3
2082 F20110404_AACKWN justice_l_Page_096.txt
4fb886957f1a686f0250e48e58a44d6c
a2557615c5a336eacff2f8bb71d751645e227ecd
57835 F20110404_AACLCI justice_l_Page_141.pro
d42253a0868b17e6c8e4655299e8a268
f35f2af137c6dc8b58b16dafa51ce8f4ae4b1a79
2005 F20110404_AACKXC justice_l_Page_113.txt
6cc38b8e52efcbf17c8c3757d4ea49ee
1332fae1ac707a7b1f444f7aa39770e8267b32eb
34321 F20110404_AACLBU justice_l_Page_125.pro
962d47bf31b23f459fab00048b459286
7b3df491e18618ad667be9778d8177bbbe665a60
1245 F20110404_AACKWO justice_l_Page_097.txt
8005f9575ff21f7da58d5e8a00eb0b1c
24530927dcb637587a1f0ca9aad291b4175cfb98
45160 F20110404_AACLCJ justice_l_Page_142.pro
04596c0a4d43db53a47068036167373f
d7b8ed866c2e3a2ae917ecf274db3194c0b0b678
2587 F20110404_AACKXD justice_l_Page_114.txt
bca1e5f9d2e612be52210b21bc1a6c33
d63d1bc079eb35c932442f5d685e2f69442a80ce
31135 F20110404_AACLBV justice_l_Page_126.pro
7faf2a176f72df31e8fac372aa0b734c
cef31d7bf0c28c57a430333e0c11b4515ea02b03
874 F20110404_AACKWP justice_l_Page_099.txt
ac5391a1adcc781850e81e78e0a0bcd2
89f5f7a998d361dbce7165582a1dda5f38a34655











EXPLORATION OF FOOD SECURITY VIA A MESO SCALE MODELING APPROACH:
JAMAICA AS AN ILLUSTRATIVE CASE




















By

LAUREN COLLEEN JUSTICE


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

2007

































2007 Lauren Colleen Justice









ACKNOWLEDGMENTS

I extend infinite gratitude to my advisor, Dr. Greg Kiker, for supporting me both

intellectually and financially throughout this project. His positive attitude always provided

reassurance in this project and myself. My committee members, Dr. Peter Hildebrand and Dr.

Clyde Kiker, have also provided much appreciated support. I also thank Mrs. Mindy Kiker for

her generous guidance. In Jamaica I thank the University of the West Indies Mona, in particular

Dr. David Dodman, Dr. David Barker, Dr. Balfour Spence, Dr. Michael Witter and especially

Ballayram. Additionally I express appreciation to all of my field collaborators. Thank you to

GECAFS for their inspiration to pursue this project and the additional resources they provided.

The Center for Latin American Studies deserves recognition for supplying the funding to

complete the necessary field work. I express gratitude to The School of Natural Resources,

especially Dr. Steve Humphrey for guiding a challenging and innovative degree program.

On the home front I thank Jessica Wielgosz for reminding me that every student deserves a

break now and again. I am grateful to my Mother, Betheda Gramling, and Vernon Gramling for

their financial support but even more so for their understanding and hugs. I thank my Dad, Jay

Justice, for reminding me that science is exhaustively slow and no one alone will solve the

world's problems. I appreciate my brother, Nicholas Justice, for his inspirational conversations.

I also thank Bonnie Byrd, and John and Vera Gramling for their care and attention. Finally I

express my utmost respect and appreciation to Ward Bradshaw for his motivational attitude and

limitless emotional support.









TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ...............................................................................................................3

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

L IST O F FIG U RE S ............................................................................... 8

ABSTRAC T ............................... ..................... 12

1 IN T R O D U C T IO N ............................................................................................................... 14

Dynamic Collaborative Models for Food System Exploration ........................................ 14
Jam aica as a C a se S tu dy ................................................................................................... 16
O bje ctiv e s ................... ...................1...................7..........

2 PR OBLEM BA CK GR OU N D ......................................................................18

Hard and Soft Methods for Understanding Complex Systems ............................................ 18
W ick e d P ro b le m s....................................................................................................... 19
Complex System s Theory............................................................ ...... .. 19
Two Social Facets of Complex Systems Management ........................................... 21
Complex systems inherent social features .................................. .............21
M managing the m managers ..................................................................................... 22
Food System s as Com plex System s ................................................................ .. ............... 23
Agricultural and Food Policy Decision Support ........................................ ....24
Modeling Complex Systems for Decision Support ............................... ...............26
D decision Support Paradigm Shift ....................................................... ............. .... 26
Modeling ..................... .......................... 27
Hard and Soft Methods for Model Development ..........................................................29
Soft system s m methods .......................................................................... ............... 29
Iterative cycles of tool develop ent ................................................... ............... 30
Cross Disciplinary Convergence and Systems Modeling .........................................31

3 AN APPROACH TO FOOD SYSTEM MODELING...................................33

F o o d S y stem M o d els ................................ ........ ........................................................ 3 3
A pproach...... ................................ .......................................................34
Soft System s M methods .............. ...................................................................35
QnD M modeling System ............... ....... .......................................... .. .............. 36
Informing Decision Support Tool Development ...........................................................38

4 JAMAICAN FOOD SYSTEM: AN ILLUSTRATIVE CASE............................................. 40

H historical R review ............ ..... .......... .... .. ............... 40
The Exchange Rate, Inflation and Food Imports................................. ...............41


4









Economic Development and Poverty Alleviation ................................ .....................43
C lim atic Shocks ................................. .................................. .......... 44
H u rrican e G ilb ert........... .................................................................... ........ .. ....... .. 4 4
Hurricane Ivan ................................. ................................ ....... 45
H holistic Exploration................. ........................................... .... .....46
Rich Picture Reveals Conceptual Framework ................................................... ........... ......47
Empirical Review of Conceptual Categories for Model Parameterization ..........................52
Population Typology ............................ ........ .. ............... ........ 54
Land use and Agricultural Productivity ........................................ ....... ............... 57
D om estic Food D distribution .................................................. .............................. 59
Food Marketing ............................ .......................... 62
Im ported Food D distribution and M marketing ........................................ .....................64
The Jam aican C onsum er.............................................. .................... ............... 65

5 JAMAICAN FOOD SYSTEM MODEL DESIGN AND TESTING.................... ........ 71

A p p ro a c h ..................................................................................... .7 1
Population Typology Representation .................................. .....................................72
Land U se Representation ......... .. ........................................... .... .................. 74
P productivity R presentation .......................................................................... ....................79
Productivity Calibration ........................................ ... .... ........ ......... 82
V vegetables ........... .......... ......... ...... ... ..................................... 82
R oot crop s ...........................................................................84
L e g u m e s .............................................................................8 6
Livestock ..................................................... ................... ........ 88
Testing Productivity Sim ulations ............................................................................. 89
V eg etab les produ ctiv ity ......................................... .............................................89
R oot crops productivity .................... ........ ..................................... ............... 91
Legum es productivity .......... .... .............. ........ ...... .. ..................... 93
L ivestock production ......... .............................................................. ..... .... .. 94
Food Im portation ................... ......... ......................... ...... .. ..... ............. 95
Food Distribution and Marketing Representation ...................................... ............... 95
Representing N national Farmgate Prices ........................................ ....... ............... 96
C ro p p ricin g ............................................................................................ .... 9 6
C rop pricing calib ration ........................................ .............................................9 8
Livestock pricing ................. .. ................................ ................... 101
Price Testing ......................... .................. .......... ........... 102
Representing Producer-Higgler Food Transfer Volumes ........................................106
R presenting Consum er Prices...................... .... ................................. ............... 108
M ark etin g co sts ................................................................... 10 9
Transport costs .......................... ........... .............. ... .......... 110
Jam aican Consum er R epresentation...................... .............. ..................... ............... 112
Consum option Testing ............... ............... .................. ...... ........ .. ........ .... 118

6 AN ILLUSTRATIVE SCENARIO SIMULATION .................................... ............... 121

Biofuel Dem and and Rising Corn Prices............................................................ ............ 121









D discussion of Scenario Sim ulations ............................................. ............................ 123

7 CON CLU SION S......... .......... .................................. ...... .... ........... 130

M odeling Jam aican Food System s .............................................. ............................. 130
B roader L lessons L earned ........................................................................... ....................13 1

L IST O F R E F E R E N C E S .................................................................................... ...................133

B IO G R A PH IC A L SK E T C H ......................................................................... .. ...................... 143













































6









LIST OF TABLES

Table page

3-1 Project phases. ...................................................................34

3-2 Q nD processes. ...................................................................37

4-1 TWOACE characteristics of farmer remuneration system..................................................49

4-2 Jamaican food system interactions that affect availability conditions. .................................52

5-1 R ural population typology. .......................................................................... ......................73

5-2 Percent of Jamaica's land uses by parish.............................................................................77

5-3 M maximum m monthly yield coefficients (tons/ha) ........................................ .....................79

5-4 P rodu activity coefficients.............................................................................. .....................80

5-5 Crop sell amount factors (kg/capita/month).......................................................................108

5-6 M marketing cost factors. ........................................... .......... ........ ........ .. 110

5-7 Transport infrastructure index. .................................................................................111

5-8 B ase preferential consum ption basket. ..................................................................... ....... 14

5-9 Descriptions of one serving for each food type...............................................................114

5-10 Maximum food group expenditures by geographic region. ...............................................115

5-11 Food type expenditure factor ...................... .... ......... .......................... ............... 116




















7









LIST OF FIGURES


Figure p e

2-1 Iterative collaborative cycles in a tool's development process ............................................31

4-1 Rich picture of the key themes of Jamaican food security ..................................................48

4-2 Food system activities and actors. ...... ........................... ........................................ 51

4-3 Depiction of QnD:Jamaica's hierarchical object oriented structure................... ............53

4-4 Depiction of Local Components descriptive data............................................... ...........53

4-5 Primary income source of small farms by parish. .............. ..........................................55

4-6 Landownership of rural economic units by parish. .................................... ............... 57

4-7 Small farm land use in term s of crop type................................................................. 58

5-1 Land use structure in QnD :Jam aica ...................................................................... 75

5-2 FA O cultivated area by crop group. .............................................. ............................. 76

5-3 M O A cultivated area by crop group. ........................................ .......................................76

5-4 Estimated land areas that act as drivers in QnD:Jamaica..................................................... 78

5-5 Vegetables productivity coefficient relationship .......................................... .............81

5-6 Vegetables productivity calibration by date. ............................................... ............... 83

5-7 Vegetables productivity calibration difference from known ............................................83

5-8 V vegetables volum e calibration by date........................................................ ............... 84

5-9 Vegetables volume calibration difference from known. .............................. ................84

5-10 Root crops productivity calibration by date. ............................................. ............... 85

5-11 Root crops productivity calibration difference from known. .............................................85

5-12 Root crops volum e calibration by date. ........................................ ........................... 85

5-13 Root crops volume calibration difference from known ...................................................86

5-14 Legumes productivity calibration by date. ........................................ ....................... 86

5-15 Legumes productivity calibration difference from known ................................................87









5-16 Legum es volume e calibration by date.............................. ........................ ............... 87

5-17 Legumes volume calibration difference from known................................ ...............87

5-18 Livestock production relationship. .............................................. ............................. 88

5-19 Livestock production calibration by date. ........................................ ........................ 89

5-20 Livestock production calibration difference from known. ................................................89

5-21 V vegetables productivity testing by date........................................ ............................ 90

5-22 Vegetables productivity testing difference from known. .............................................. 90

5-23 V vegetables volum e testing by date. ........................................................... .....................90

5-24 Vegetables volume testing difference from known................................... .................91

5-25 Root crops productivity testing by date ........................................ .......................... 91

5-26 Root crops productivity testing difference from known..................................................92

5-27 R oot crops volum e testing by date. ........................................................... .....................92

5-28 Root crops volume testing difference from known. .................................. .................92

5-29 Legum es productivity testing by date........................................................ ............... 93

5-30 Legumes productivity testing difference from known. .............................. ................93

5-31 Legum es volume e testing by date. ............................................... .............................. 94

5-32 Livestock production testing by date ................................................................. ............94

5-33 Livestock production testing difference from known................................ ...............95

5-34 National vegetables farmgate price relationship. ..................................... ............... 97

5-35 National root crops farmgate price relationship. ...................................... ............... 97

5-36 National legumes' farmgate price relationship.................................. ....................... 98

5-37 V vegetable price calibration by date. ............................................ ............................. 98

5-38 Vegetable price calibration difference from known. ...................................................99

5-39 R oot crops price calibration by date. ........................................ ......................................99

5-40 Root crops price calibration difference from known ........................................................ 100









5-41 Legum es price calibration by date ........................................................ ............... 100

5-42 Legumes price calibration difference from known.................................. ...............100

5-43 N national livestock price relationship. ........................................................................... 101

5-44 Livestock price calibration by date ........................................................ ............... 101

5-45 Livestock price calibration difference from known. .................................. .................102

5-46 Vegetable price testing by date ........................................................................ 102

5-47 Vegetable price testing difference from known. ..................................... ............... 103

5-48 Change in known vegetables price index vs. change in simulated nominal vegetables
p rice ......... ....... ... .......... ................. ................................................ 10 4

5-49 R oot crops price testing by date. ............................................................. .....................104

5-50 Root crops price testing difference from known. .................................... .................104

5-51 Root crops simulated volume and real prices .............................................................105

5-52 Legum es price testing by date. ................................................ ............................... 105

5-53 Livestock price testing by date. ............................................... ............................... 106

5-54 Livestock price testing difference from known. ....................................... ............... 106

5-55 P percent of production sold ............................................................................... ......... 107

5-56 Maximum and minimum rural carbohydrate price ratios across parishes..........................119

5-57 Sim ulated vegetable purchases ............................................................................ ..... 119

5-58 Im ported staples purchased. .................................................................... ..................... 120

6-1 Historical scenario of domestic livestock production and demand ..................................... 124

6-2 Scenario 1 of domestic livestock production and demand. ...............................................124

6-3 Scenario 2 of domestic livestock production and demand. ...............................................125

6-4 Scenario 3 of domestic livestock production and demand. ...............................................125

6-5 Regional price shocks to domestic livestock in historical scenario.............................. 126

6-6 Internal structural design noise ............................................................................. ....... 127









6-7 Domestic and imported livestock purchases of Saint James farmers in historical
scene ario ........................................................ .................................12 8

6-8 T total im ported anim al m eat................................................................................ ......... 128










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

EXPLORATION OF FOOD SECURITY VIA A MESO SCALE MODELING APPROACH:
JAMAICA AS AN ILLUSTRATIVE CASE

By

Lauren Colleen Justice

December 2007

Chair: Gregory A. Kiker
Major: Interdisciplinary Ecology

The purpose of my study is to examine the interactions within a food system and between

that system and its socioecological environment in order to reveal management or policy

strategies that strengthen food security conditions. Using Jamaica as a case study my

interdisciplinary project developed and tested a computer model, QnD:Jamaica, that fuses

technology with lessons from complex systems theory in order to promote adaptive management

of food systems. An iterative collaborative design process based on soft systems approaches was

employed during model design. The problem situation as understood by key stakeholders was

investigated through informal strategic dialogues. These qualitative descriptions were

conceptualized by a formal framework to identify a limited number of quantifiable parameters

and drivers of interest for the scenario model. The model incorporates ecosystem and

sociopolitical issues into a user-friendly framework to analyze the spatially explicit dynamics of

food, money and people in terms of their effects on food security conditions. Illustrative

scenario simulations were performed and the implications for systemic features were explored.

The results of my study provide insight into conceptualizing food systems in general and the









Jamaican food system specifically. My thesis serves as an explicit example of integrated soft

research methods and analytical tool development and will guide future modeling efforts.









CHAPTER 1
INTRODUCTION

This modern mind sees only half of the horse-the half which may become a dynamo, or an
automobile, or any other horse powered machine. If this mind had much respect for the full-
dimensioned, grass eating horse, it would never have invented the engine which represents only
half of him. Allen Tate, in Berry, 2002

"...de IMF a de tiefingest ting-dem lend you money fi buy tin mackerel, but not fi help small
farmers plant. Every country where dem go it mash up. Smallfarmer in St. Mary, Jamaica,
2001(Weiss, 2004a)
In our current global food system, most countries are dependent on trade with distant agro-

ecosystems for their food supply. While international trade is beneficial for many societies and

100% reliance on local food is neither practical nor desirable, it is important to recognize and

manage the trade offs that occur between local and external food supply dependence (Sundkvist

et al., 2005; Lyson and Green, 1999; Costanza et al., 1995). Food systems by their nature

embody ecological, social and economic issues and the global changes taking place both

climatically, culturally and economically are having profound effects on how food is produced

transported, understood, valued, prepared, and consumed. Food security offers itself as an

interdisciplinary set of criteria for exploring the complex issues involved in food systems. Food

security encompasses all of the mechanisms that allow individuals to maintain command over

their food consumption (Sen, 1981; Corral et al., 2000). This definition allows food security to

be viewed in its broadest sense as a rich picture of societal well-being and natural resource

viability.

Dynamic Collaborative Models for Food System Exploration

The direct and indirect synergistic relationships within the process of food procurement

from field to fork are highly unknown and unpredictable and make exploration of food system

vulnerabilities a complex task. New tools for assessing, planning and communicating about food

security should embrace this uncertainty and aim to facilitate creative collaborative exploration









of the system and its contextual environment (D. H. Walker, 2002). Methods that investigate

patterns between interactions and emergent properties in problematic situations promise insight

into improved management strategies (Kasperson and Kasperson, 2001; Walker et al., 2004).

Adaptive management strategies were first developed from the recognition that systemic

socioeconomic and ecological knowledge is and always will be imperfect and incomplete

(Berkes et al., 2003; Dyball et al., 2005; Holling and Gunderson, 2002). Unforeseen drivers,

interactions and outcomes are virtually guaranteed. Rigid policies prevent adjustment to these

changing situations and can create conditions where unsustainable and highly deleterious

behaviors continue. Adaptive collaborative techniques offer a methodological structure with

flexibility and space to explore, evaluate and select the most beneficial policies as unpredicted

issues arise. This structure is iterative in nature and includes cycles of policy experimentation,

reflection, evaluation and modification (Keen and Mahanty, 2005).

Many have suggested scenario modeling as a tool that could aid exploration of possibilities

and encourage creative thinking and collaboration amongst stakeholders (Walker et al. 2002;

Carpenter et al., 1999; Janssen et al., 2006). Scenario-based models can provide analyses of

trade-offs between socioeconomic and environmental goals. Scenario modeling does not aim to

predict the future but rather to explore the potential ramifications of a range of policy options

(Baker et al., 1999).

The proposed research project is one of the first national food security analyses to employ

a simulation model nested in a collaborative design process in order to reveal patterns in a food

system's socioecological interactions (Pothukuchi and Kaufman, 2000). This approach is based

on the adaptive management argument that the emergent properties which characterize complex

systems, food systems included, are so dynamic that they currently can not be managed statically









or for single indices. Rather, management requires planning tools that envision multiple

relationships at once and a range of possible states of these relationships.

Using Jamaica as a case study, this interdisciplinary project developed and tested an

exploratory computer model that fuses technology with lessons from complex systems theory in

order to promote adaptive management of food systems. This model, QnD:Jamaica, uses the

Questions and Decisions (QnD) modeling system that allows simulation of biological and abiotic

components of a system (Kiker and Linkov, 2006). QnD is also appropriate to this project

because of it facilitates collaborative discussions through its easily manipulated structure and

game-like format. This thesis takes the approach that when a broad definition of food security

which includes a rich picture of societal well being is applied to food system management, the

most appropriate policy options will consider diverse socioecological dynamics together at the

meso scale.

Jamaica as a Case Study

Like many Caribbean nations, Jamaica has a small open economy remnant of the

plantation system that is vulnerable to major changes in the world economy (Beckford, 1972).

Throughout the 1980s and 1990s the dominate prescriptions of international financial institutions

revolved around the theory that free markets will increase food security by ensuring cheaper and

more consistent food supplies (Weiss, 2004a). However, Jamaica is now one of the most

indebted nations, both in terms of financial and material imbalances. Historically driven by cash

crop exportation, two main symptoms indicate negative shifts in the Jamaican food system:

increasing loss of market for exports as well as large-scale importation of cheap food leading to

Jamaican farmers' inability to compete in their own domestic market (FAO, 2003).

As Jamaica's Minister of Finance has conceded, the country is now operating in 'crisis

mode' (Davies, 2000 in Weiss, 2004a). Assumptions about Jamaica's comparative advantages









in certain sectors come into question with the present and potential international trade shifts and

global climate changes. The Jamaican food system needs to be reevaluated with a cross-

disciplinary multi-scalar lens that captures the interplay between bottom up and top down

features present in a small open economy.

Objectives

The purpose of this study is to explore the interactions within Jamaica's food system and

between that system and its socioecological environment in order to reveal management

strategies that strengthen food security conditions. Of concern are the implications for

agricultural production, income opportunities and nutritional consumption patterns.

Accordingly this research has six objectives:

* Conduct a literature and data review of Jamaican food security

* Use soft systems analysis to survey interests involved in problematic situation

* Translate the interview results, literature concepts and available data into a functioning
simulation model

* Test the model against selected historic conditions

* Explore plausible scenario simulations to enrich understanding of food system links

* Provide insights into future model development and use









CHAPTER 2
PROBLEM BACKGROUND

This section reviews complex systems theory and the parallel development of softer

approaches for exploration into complex systems that seek to generate relevant, timely and

adopted knowledge and its associated technologies and policies. A discussion of complex

systems theory as applied specifically to the management of food systems follows. Finally, the

evolution of decision support technology, specifically dynamic modeling, as it seeks to apply the

complex systems paradigm is reviewed.

Hard and Soft Methods for Understanding Complex Systems

The contemporary research community is not only internally fragmented but also

increasingly distant from the institutions at large who apply research findings (Pidd, 2004;

Douthwaite et al., 2000; D. H. Walker, 2002). This distance inhibits the efficiency and

practicality of the application of research findings (Ibid.). Misunderstandings, distrust, non-

adoption or conflict generally characterize the ineffective knowledge transfer between research

and implementation communities around the world (Ibid.).

The interdisciplinary movement has intensified the discussion of how a particular research

strategy can either assist or inhibit the eventual uptake and practical application of newly

acquired information by the community at large (Douthwaite et al., 2000; D. H. Walker, 2002;

Kiker et al., 2003). This discussion offers new research paradigms that aim to decrease the

inefficiencies that plague the integration of the research, development, and implementation

communities involved in socioecological systems management (Roux et al., 2006; Gunderson et

al., 1995). The combination of soft and hard methods in research approaches is being explored

as one way to bridge divides and inefficiencies by providing both decision and process support

throughout a project (Eden and Ackermannn, 2004).









Wicked Problems

Scholars of ecology as well as engineering and management science agree that large scale

management of socioecological systems has been largely unsuccessful at maintaining resilient

communities (Holling, 1996; Costanza et al., 1995). This failure can be attributed to

shortcomings within both (1) the classical reductionist scientific process and (2) society's

institutional capacity to translate scientific knowledge into real change (Simon, 1962; Forrester,

1961, Ackoff, 1974; Meadows and Robinson; 1985). Both of these problematic issues are

intrinsically linked by the nature of the managed situations as "wicked problems" or "messes"

(Rittel and Weber, 1973; Ackoff, 1974). According to Conklin and Weill a wicked problem

exhibits the following (Pidd, 2004):

* The problem is ever evolving and cannot be confined to a single definition.

* Because there is no definitive Problem, there is no definitive Solution

* There are many stakeholders involved, making the problem "solving" process
fundamentally social.

* The physical and social constraints on the solution change over time

Complex Systems Theory

Attempts to deal with these problematic situations in an integrated manner have been

driving a core paradigm shift across management communities, from natural resource to

operations management for the past fifty years (Folke, 2006). The new paradigm, referred to

here as complex systems, developed from general systems theory. It is a reaction against the

classical reductionist approach to science and management that assumes that complex

phenomena can be broken down into smaller units that can be managed independently

(Churchman, 1979; von Bertalanffy, 1968; Ackoff, 1974; Simon, 1962; Forrester, 1968). Rather,

general systems theory views the whole as greater than the sum of its parts. This shift is from a









"bounded rationality", "positivist", "determinist", or "hard" approach to a "soft" or "procedural

rationality" framework for knowledge formulation and application (Simon, 1962; Checkland,

1979; Douthwaite et al., 2001; Lyons, 2004).

Since the conception of complex systems theory, researchers have struggled to develop a

systematic strategy that can address both the biophysical and the human aspects of complex

systems. Obviously there are fundamental differences between each type of system; biophysical

systems are inherently spatial and transfers occur in material and energy terms, social systems

are often non-spatial and transfer knowledge and information. Some of the early attempts at

formal integrated analysis guidelines focused on both systems' hierarchical structure (Simon,

1962; van Bertallanffy, 1968).

Forrester and H.T. Odum are two of the major contributors to complex systems theory

(Pavoa-Zuckerman, 2000). Both scientists developed symbolic languages to represent system

components and dynamics in order to facilitate modeling of the components in complex systems

(Ibid.). Forrester, an engineer and computer scientist, combined ideas from cybernetics, control

engineering, and organizational theory to develop a guiding philosophy for simulating complex,

non-linear, multi-loop feedback systems initially applied to the management of industrial firms

(1961). Odum, an ecologist drawing from Forrester, used similar theories to explore humans

within natural systems (1983).

The complex systems framework assumes that all materials and information are transferred

through positive and negative synergistic couplings that produce emergent effects such as delays

and thresholds (Forrester, 1961). Additionally, complex systems are understood as open systems

that interact freely with their environment, resulting in disequilibrium and an unbounded set of

interactions (Holling, 1986). Despite the fact that complex systems theory does not have an









agreed upon method for inquiry, it still provides the most helpful perspective to address

management of messy issues (van der leeuw and Aschan-Leygonie, 2005).

Two Social Facets of Complex Systems Management

The complex systems framework has strong implications for the types of tools and

techniques used to explore problematic situations. For example, ecology scholars are working to

develop a common currency between natural and social interactions in order create a language

that allows for more holistic assessments (Folke, 2006). Similarly, social scientists have been

investigating the development of new soft methods of inquiry as well as the possible integration

of these methods with classical hard methods in order to uncover and assess complex problems

(Pidd, 2004).

The complex systems framework acknowledges that not only are problematic

socioecological situations integrated by nature, but also the political and institutional capacity of

each domain is highly linked to the others even if not intentionally (Schipper and Pelling, 2006).

These two types of integration reveal two social facets of complex systems management arising

from: (1) The internal inherent features of socioecological systems (2) the social institutional

structure that exists to facilitate exploration and management of socioecological systems.

Complex systems inherent social features

Prior to the mid 1990s most ecosystem management studies were framed exclusively in

"natural" or "human" terms (the former usually occurring in biology or ecology, the latter in

economics) (van der leeuw and Aschan-Leygonie, 2005). However, following increasing

evidence that neither the social nor the natural system is paramount, the focus shifted to the

interactions between the equal but heterogeneous components of what became termed as

socioecological systems (Gunderson et al., 1995; Holling and Gunderson, 2002; Berkes et al.,

2003; Walker et al., 2004). This progression from restrictive to holistic understandings has been









described as an evolution of the role of humans in socioecological relations moving from re-

active prior to the 1980s, through pro-active during the 1990s, and residing currently as an inter-

active understanding (van der leeuw and Aschan-Leygonie, 2005).

Holling's work in ecological resiliency has fueled much of the methodological theory for

integrating social and natural systems' studies (Holling, 1973; Folke, 2006). The importance of

temporal and spatial scales of analysis and the mechanisms that transcend these levels is at the

core of resilience concepts. Holling posits that only two criteria must be met for resilience

concepts to be applied to studies beyond ecology: (1) the systems need to be described in

dynamic terms, and (2) the systems must have more than one potential state of equilibrium

(1985).

While these developments supply concepts for research into complex systems, they do not

sufficiently address the second social facet of complex systems management. This additional

side to the management problem rests in the process support needed to guide the institutional

machine performing and applying the research.

Managing the managers

Management of issues in complex socioecological systems (i.e. food security) involves not

only a conceptualized system's variables and their interactions but also the issues that arise from

stakeholders' dynamic perceptions of the situation. Communities must strive to navigate the

logistical social aspects of managing a problematic situation by being "conscious of and

competent at dealing with, the differences between individuals (or collective interests) that

comprise the group with respect to a host of issues surrounding the situation" (Bawden, 1995 in

Ison et al., 1997, 266).

Research in the social management disciplines has used complex systems theory to inform

new methods of inquiry that transcend the reductionist model of research in order to offer









process as well as decision support (Eden and Ackermannn, 2004; Pidd, 2004). Rather than

attempting to discover, quantify, predict and control the situations that arise from complex

systems by way of hard methods, soft frameworks offer flexible options for problem exploration

and management. As Ison et al. concisely summarized, "A given problem does not occur in

isolation or exist objectively-'out there'. It is socially constructed on the basis of different

perspectives and is part of an interrelated network of problems" (1997, 260). In accordance with

the notion of wicked problems, preference for a particular strategy to address such a problem

thus reflects a certain situational understanding. Obviously this understanding will differ and

change between individuals, communities, and disciplines making it difficult to agree on a single

strategy of action. However, using the existence of multiple perspectives as the ideology itself,

as the complex systems paradigm assumes, provides for a system of analysis that "prepares for

mutual understanding and negotiation of the problems) in question" (Ison et al., 1997, 260).

Collaborative modeling is one type of decision tool that can apply the complex systems

framework to support research within a holistic socioecological problem structure while

simultaneously alleviating the inefficiencies that arise from the social nature of problem solving.

Food Systems as Complex Systems

Food systems can be understood as one type of complex system where systemic

vulnerability is embodied in temporary conditions of food insecurity. These potential conditions

emerge from a spectrum of possible interactions between internal and external socioeconomic

components of the food system. Like all complex systems, food systems behave synergistically.

Actors at all levels of the food system must deal with both changing ecological and

socioeconomic conditions. Additionally these actors create change leading to feed back within

the system both to themselves and to actors at levels above and below them.









Food security issues occupy a unique meso scale location in the planning field because

food and agriculture possess distinctive qualities regarding quality of life, cultural decisions and

environmental and land use effects. Recognizing that poverty and hunger occur at the household

level as family decisions often determine how limited resources are utilized, much recent effort

has been put into micro level household scale studies to determine how communities deal with

limited resources and how shifts in access efficiencies or practices would change resource use

and quality of life. Yet macro scale indices remain the most common measure of

community/national health in the planning arena (Pothukuchi and Kaufman, 2000).

In order to evaluate food security the linkages or pressure points between the field and the

fork must be holistically assessed (Ericksen, 2006a and 2006b). Increasing food security

requires asking the following questions: Who are the actors being relied on at all levels from

production to consumption of the food system? How does reliance on those actors affect a

population's interests? How could each actor transform or change? And what are the potential

adaptations/coping strategies of a population in the face of the actors' potential transformations?

Agricultural and Food Policy Decision Support

Traditional agricultural research and food policy has failed to produce a global food system

with adequate levels of food security, leading the food and agricultural research community, like

other socioecological sectors, to move away from methods and institutions based on reductionist

frameworks (Fraser et al., 2005; Sundkvist et al., 2005; O'Brien and Leichenko, 2000; Costanza

et al., 1995; Meadows, 1985). With the advent of sustainable development principles, a gradual

shift is taking place whereby issues of agriculture are no longer analyzed separately but are

integrated into the assessment of the entire food system in terms of the roles it plays in rural

welfare, environmental effects and consumer welfare (Eakin, 2005; Andreatta, 1998; Ford,

1992). In this context, the biological and socioeconomic transmission mechanisms that govern









the availability and accessibility of food can be analyzed within a coherent framework (Fraser et

al., 2005).

Twenty five years ago, Amartya Sen introduced new ideas to address poverty and food

security. He moved the concept of food accessibility to the forefront of the food security

discussion (1980). Twenty years ago, Donella Meadows, a systems analyst, described the

chronically inadequate food systems that failed to support much of the world's population

despite the existence of thousands of detailed formal analyses (1985). She made the crucial

insight that this malfunctioning was due to posing questions that were "not being applied to the

total system" and that "in their underlying assumptions and limitations cannot lead to the

answers one really wants to find" (1985, 161). As an example of a proper perspective, Meadows

cites a Dutch international food system simulation model called MOIRA that integrated

environmental and socioeconomic change in 1979! However movement since then towards the

posing of the types of holistic questions that could lead to better analyses and management of

food systems has been slow.

Despite these early promises of success, agricultural research and food policy communities

largely continue to manage by way of limited single goals such as aggregate calorie per capital

availability and overall crop productivity. Just as Holling showed that the use of static target

variables in ecosystem management leads to pathologically vulnerable ecosystems, scholars of

food systems have begun to show how the failure to use integrated social and environmental

assessments have increased food insecurity (1978, O'Brien and Leichenko, 2000; Schipper and

Pelling, 2006). Dynamic modeling is a potential tool for more integrated assessments of food

systems.









Modeling Complex Systems for Decision Support

The systems engineering community that develops technologies to bridge the divide

between research and policy institutions has been traversing a paradigm shift parallel to

ecological and social research scholars (Pidd, 2004; Lyons, 2004; Eden and Ackermannn, 2004).

As the complexity of problematic situations under investigation is increasingly recognized,

technology's role in guiding formulation of strategies for situational improvement is changing.

For example, a feature of complex systems is their unpredictability, tools that claim to make

predictions become irrelevant (Lyons, 2004). In the same vein, the increasing recognition of the

importance to account for competing versions of reality or worldviews in order to create

knowledge that is relevant and adopted, necessitates technology that aids in formulation of

cohesive community problem definition or at least increases understanding between the various

stakeholders of different perspectives (Eden and Ackermannn, 2004; Douthwaite et al., 2001).

Decision Support Paradigm Shift

Widespread disenchantment characterizes the general sentiment towards decision support

systems, (DSS)1 (D.H. Walker, 2002). Research reveals non-delivery, irrelevance, inflexibility,

inaccessibility, lack of confidence, institutional and political boundaries, and negative impacts on

decision making processes as reasons for decision support systems' limited adoption and

perceived minimal impact on decision making processes (D. H. Walker, 2002; Cox, 1996; Davis,

1989; Meadows and Robinson, 1985).

These findings are not surprising as the classical view of manager is one of impartial

controller (Stacey, 2001; Lyons, 2004). The new managerial paradigm emphasizes flexibility


1 Although a decision support system is traditionally understood as a probabilistic computer tool that aims to reveal
optimized solutions to decision makers and predict the impact of a given policy on a single variable, DSS is
understood in this thesis in its broadest sense as an overarching situational strategy that aids in decision making
through multiple flexible tools and methods as defined in D. H. Walker (2002).









rather than control as well as the need to develop approaches and tools which are relevant in the

wider decision-making context (D. H. Walker, 2002). The DSSs must not only aim to deepen

our understanding of a particular system's components and dynamics but should also serve as

devices that bring cohesion to a group of diversified stakeholder groups. A key feature of a

particular DSS should be to broaden understandings of other interests involved. Its functions

should include exploration of how others view the problem situation, followed by facilitating

initial agreement upon problem situation. However, it should contain provisions whereby this

initial vision can be modified as viewpoints or interests change or increased technical knowledge

surrounding the problem is introduced. Often DSS tools are considered successful even when

they are not widely employed because they facilitate learning between developers and managers

(D. H. Walker, 2002). In fact, DSS that fail to facilitate experiential learning have limited utility

(Ison et al., 1997).

Modeling

Certain modeling strategies can transcend the traditional difficulties associated with DSS

relatively easily. Models' inherent qualities as representations of perceived reality allows them

to serve as "transitional objects" that help people see their own and others' mental models

(Morecroft, 2004). Not only can a model provide insights into the system under question but

the very creation and design process can serve as tools to learn about the system as well as the

managerial stakeholders involved (Walker et al., 2002).

It is important to note that there is no "right" model of a system, just more and less useful

ones (Janssen et al., 2006). Because models are merely simplifications of the real system they

cannot test hypotheses they can only describe the reality the modeler wishes to describe (Levins,

1966 in Pavoa-Zuckerman, 2000). While they are not able to prove or disprove any theory, a

model's greatest utility lies in its ability to serve as a learning tool.









Models are useful when integrating multiple disciplinary concepts "because models serve

as conceptual tools in the formulation of theories, helping one to visualize the patterns and

processes of the ecosystems being represented" (Pavoa-Zuckerman, 2000, 36). Models allow us

to transcend our restrictive understanding of what is possible and help us "expand our linear

expectations to include all the possible futures we may encounter" (Farrell, 1998 in Lyons, 2004,

39). Model exercises can reveal trends in the expected importance of certain system features for

implicating resiliency to certain events (Janssen et al., 2006). These models can also reveal

systematic processes for collecting future data about socioecological systems (Ibid.).

Models should attempt to "identify thresholds, their nature, and what determines their

positions along the driving variables" (Walker et al., 2002). They should be simple, multi-scalar

and temporal, and scenario based (Ibid.). They should 1) Act as tools of negotiation, 2)

Create/unearth choice, and 3) Define a context for trade offs (Schrage, 2000). Developers of

models should be aware of how stakeholder interests and assumptions are integrated with system

features before they can adequately develop a tool for systematic learning about a problematic

situation and potential improvement options (Bunch, 2003; Zhu and Dale, 2000).

This larger scene in which model development is nested is eloquently described by

Meadows and Robinson, "Modelers are human beings with human strengths and failures. To the

extent that their work can be structured to bring out their strengths and correct for their failures,

they can contribute immensely to human understanding of how the world works, to social

decision-making, and to the shaping of an uncertain future" (1985, 15). One such modeling

structure is the explicit employment of soft methods consisting of iterative cycles of inquiry and

reflection that revolve around how the stakeholders seeking to learn about and improve a

problematic situation understand the issues involved (van der Heij den, 1996).









Hard and Soft Methods for Model Development

Multiple frameworks exist to guide systems' engineers towards creating tools of greater

relevance and impact (Parker et al., 2002; D. H. Walker, 2002; van Ittersum et al., 1998;

Meadows and Robinson, 1985). These all stress the importance of iterative flexible

development, transparency, integration of scales, issues and stakeholders, qualitative as well as

quantitative concepts and communication facilitation (Ibid.). Most importantly these

frameworks emphasize the need for continual monitoring and investigation into the process of

technology development in order to move towards best practice guidelines.

Soft systems methods

Checkland and Holwell make the astute distinction between hard and soft approaches and

hard and soft skills (2004). While a soft approach assumes an infinite number of possibilities, a

hard approach assumes only one. In this sense a soft perspective is the only appropriate

approach for wicked problems. Soft approaches, however, can accommodate hard skills within

them. When assessing soft vs. hard skills, such as qualitative vs. quantitative research

techniques, there is the possibility of fluid and equal integration between soft and hard (Ibid.;

Pidd, 2004; Eden and Ackermann, 2004). Eden and Ackermann propose the use of combined

soft and hard methods in order to foster both decision and process support (2004). This type of

approach provides a space where group and individual interests can be explored simultaneously

just as a forest can be seen as a whole without losing site of the trees.

In traditional system management theory the objectives of a given system were assumed

obvious (Checkland, 1979). If the management objectives were not met, it was believed to be a

result of system inefficiency, and planners set out to engineer solutions. However, as this

reductionist thinking failed to solve the increasing problems occurring between humans and their

environment (Capra, 1996), managers began to wonder if they should not only be asking, "how









should the system work?", but also "what should the system provide? And why?" (Checkland &

Scholes, 1990). The answers to these questions are never "right" and always contain degrees of

inherent uncertainty (van der Heijden, 1996). Soft systems methods such as scenario planning

(van der Heijden, 1996) and the soft systems methodology, SSM (Checkland, 1979), directly

address the existence of multiple worldviews and provide ways to engage with them (Checkland,

1999). The systems modeling method nested in a softer methods approach offers potential to

overcome traditional DSS limitations. (Bunch, 2003; Zhu and Dale, 2000)

Iterative cycles of tool development

Douthwaite et al. characterize traditional technology development as positivist-realism

associated with a uni-directional information flow from the engineer or knowledge possessor to

the user (2001). This approach is built on the hard science paradigm that assumes "an

independent reality" (Ibid., 2) that can be experimented upon to produce quantifiable results.

Douthwaite has developed an iterative tool development process structured in cycles of learning

rather than uni-direction information flow (Ibid.). This approach is known as the Follow the

Technology approach as opposed to the Transfer of Technology approach employed in

traditional tool development (Ibid.).

At any step during tool development new information could come to light, or forward

movements become stalled, and revisiting a previous step becomes necessary. In order to

explicitly account for these inevitable situations, this approach utilizes ongoing cyclical

exploration and analysis. These cycles of innovation are best described in Figure 2-1 based on

Douthwaite's tool fitness diagram.


























I Time I
Figure 2-1 Iterative collaborative cycles in a tool's development process (after Douthwaite et al.,
2001).

Cross Disciplinary Convergence and Systems Modeling

It is interesting to note the convergence of the ideas contained in the frameworks from

technology engineering with those from other disciplines that also seek to address complex

system management. For instance in the field of vulnerability analysis researchers are moving

away from black box analytical models of process and outcome that do not allow for the

exploration of the dynamic and integrative nature of risk (Adger, 2006). The idea of continual

evaluation of processes stressed above forms the backbone to the entire theory of ecological

adaptive management (Holling, 1978; Kiker et al., 2003). Researchers of resilience theory,

mostly ecologists by training, have illuminated four steps for complex system assessment that

mirror the frameworks suggested by the engineering community. These include 1) development

of a conceptual model; 2) Identification of possible future scenarios: 3) Exploration of

interactions; 4) Stakeholder evaluation. (Walker et al., 2002).

The similarities between frameworks across disciplines have positive implications for the

continued and improved cooperation necessary in complex systems management. Likewise


Additional
Strategic Dialogue









agreement within the DSS community as well as across many other disciplines about the

appropriateness of modeling nested in collaborative methods for revealing concepts in complex

systems promises enhanced technological decision support experiences (Walker et al. 2002;

Carpenter et al., 1999; Janssen et al.,. 2006; van der Leeuw and Aschan-Leygonie, 2005).









CHAPTER 3
AN APPROACH TO FOOD SYSTEM MODELING

Food System Models

A number of model-based analyses have been performed to assess socioecological

interactions involved in food production and consumption (Young and Kantor, 1999; Duchin,

2005). Duchin, developer of a global food system analytical model, The World Trade Model,

describes how public food security is intimately tied to nutritional, environmental and economic

policy, yet all have failed it (2005). Nutritional and environmental educational campaigns often

have less than desired impact while political elements face competing forces of public and

private interests. She stresses that in this tension, the solution lies: the environmental and

nutritional communities must work together to bridge the public, governmental and corporate

divides. Duchin suggests simulation modeling as a way to organize the interests of these groups.

She described the difficult resolution decisions that modelers everywhere often face during

the design and development processes. For instance, in the case of designing a diet scenario she

wondered if it should contain 30, 300 or 3000 food categories. She applied what natural

ecologists have said about modeling complex environmental systems to the creation of stylized

models of human production and consumption activities, "Greater abstraction, or sacrifice of

detail, is necessary to describe the workings of much larger complex systems... Creating

templates to describe alternative diets or production strategies requires levels abstraction not

generally required for representations based only on direct measurement" (2005, 110). An

average quantitative description works because it "is a stylized representation that succeeds when

it captures the characteristics most important for the analysis" (2005, 110). In her words,

successful sets of variables require "attending to the distinctive properties of different variables"

(2005, 110). This identification of the qualitative "distinctive properties" of variables allows one









to "tunnel" to the appropriate inclusion or exclusion of systemic features and avoid unnecessary

duplication.2

Soft systems methods along with the review of empirical evidence guided the

identification of the appropriate systemic features to represent the multiple interests of the

Jamaican food system in the simulation model, QnD:Jamaica.

Approach

The systematic study of how interactions between external influence and structural

mechanisms of a socioecological system affect emergent properties is a new approach for policy

planning methods. More specifically, assessment of food systems as socioecological units is a

new approach to food security management (Pothukuchi, and Kaufman 2000). The purpose of

this study is to explore the interactions within Jamaica's food system and between that system

and its socioecological environment. Of concern are the implications for food production,

income opportunities and nutritional consumption across rural and urban population groups. In

order to address these issues, a complex systems framework was employed. This framework was

applied through a dynamic modeling exercise that employed soft system methods for model

development.

This research effort has two distinct objective phases: 1) Research Phase and 2) Design and

Evaluation Phase shown in Table 3-1.

Table 3-1 Project phases.
I. Research phase II. Design and evaluation phase
Uncover key food security concerns as perceived by Design and develop a computer- based
stakeholders scenario simulation model
Formally translate these concerns to a parameters and Test model's parameters through
drivers format scenario simulations

2 However, as ecologists have recently noted, a feature that only appears to be a duplicate
might be the one key that possesses unique qualities that provide systemic resilience.










Soft Systems Methods

Soft systems approaches use iterative and collaborative techniques to envision the driving

forces which could move a system one way or another (van der Heijden 1996). These approaches

such as scenario planning (van der Heijden 1996) and the Soft Systems Methodology (SSM)

(Checkland 1979, 1999), provide tools to tease out a system's potential pressure points and

explore how possible interventions around these points might improve the problem situation

(Checkland 1999). By allowing modelers to actively engage with stakeholders' assumptions and

the inherent uncertainties in complex systems, soft systems approaches are conceptually

compatible with complex socioecological systems frameworks, and provide a useful research

methodology for interacting with stakeholders and guiding researchers as they carefully seek to

structure the problem situation. These methods assist researchers to elicit perceptions of a real

world situation and then organize the results into mental models of key themes that provide

starting points to explore the context of change within the system.

A soft systems approach first requires expression of concerns about the system in a non-

systems manner. This is accomplished through informal "strategic dialogues" (individually or in

groups) with stakeholders comprised of a broad range of interests in the system (van der Heijden

1996; van der Heijden et al. 2002). To begin these collaborations Quinn and Voyer suggest

asking, "If there was one core question that underlies all the strategic challenges you face what

would it be?" (1996). van der Heijden (1996) suggests the following questions:

* What concerns you most about the future?
* What would you ask an 'Oracle'?
* What would be a good/bad scenario?
* What would lead to a good/bad scenario?
* What critical decisions have to be made soon?
* What was learned from past surprises?









The strategic dialogues that begin a soft systems inquiry into a problem situation

differentiate the soft approach from traditional hard approaches which seek to narrow and define

problem statements early on in a research process. A soft systems approach purposefully avoids

narrowing the problem definition too soon, instead seeking to gather the many views, values,

perceptions, and expectations that characterize a complex issue. Not only do these methods

distill various interests, but they also reveal the multiple scales at which an issue can be framed.

Churchman suggests that researchers "sweep in" a wide variety of stakeholders, both friends and

enemies, in order to reveal the many assumptions about the nature of the problem and what

would constitute improvement (1979).

Interview results along with appropriate literature review findings are mapped in an

ordered database that provides an overview of the key themes in stakeholder frameworks

clustered according to their qualities in the situation. In the first stage of mapping, emergent

issues are unrestrictedly clustered according to all of their potential interactions. Next, clusters

are refined according to common and divergent world views and assumptions. The issues should

also be clustered in terms of their properties of internal or external influence. The resulting

thematic database can also be represented as a rich picture, a graphical conceptualization of the

situation. Transformations of key issues that would improve the problem situation are identified

from the rich picture. A TWOACE analysis (described in detail in chapter four) is applied in

order to specify the world views that give this transformation meaning, the actors involved, and

the environmental constraints.

QnD Modeling System

The Questions andDecisions, QnD, modeling system is a problem exploration object

oriented tool that increases understanding of potential behaviors and management options of a

particular socioecological system (Kiker et al., 2006; Kiker and Linkov, 2006). Through a user-









friendly graphical interface stakeholders can "play" their system by manipulating institutional

and ecological components of interest. Results differ from the combinations of various

environmental and economic drivers and the player's response to them via management options

implemented in different spatial areas and over different time periods. As various scenarios are

played, the interactions between institutional and ecological parameters are exposed and future

possibilities of the system can be envisioned.

The QnD system not only helps stakeholders deepen their understanding of a particular

system's components and dynamics but also acts as a device to bring cohesion to a stakeholder

community. Throughout its iterative design process, a QnD system engages stakeholders to

accomplish multiple tasks: facilitating initial agreement on key forces and themes, broadening

understandings of others' interests, developing scenarios and analyzing various policy options.

QnD was created as a technical tool that is complemented by and works in tandem with an

iterative exploration process. The system's design process was intended to be compatible with

several social science methodologies, such as the Soft Systems Methodolgy (SSM), scenario

planning, or cognitive mapping. In order to apply the QnD modeling system to Jamaica's food

security issues, quantifiable components, processes, and drivers were revealed through soft

systems methods.

Eight of the simulation processes available in the QnD modeling system were used in this

research project. Table 3-2 describes the operations performed by these processes.

Table 3-2 QnD processes.
Process name Operation performed
PAddValue Input one is added to input two
PSubtractValue Input two is subtracted from input one
PMultiplyValue Input one is multiplied by input two
PDivideValue Input one is divided by input two
PSetValue Output is set to input
PRelationship A two column table of values relates any









number of pairs of points as independent
and dependent variables. All points between
defined pairs fall on a linear relationship.
PIfGreaterThan If input two is greater than input one than
perform following subprocess. If not greater
jump to next main process.
PTemporalRunningAverageValue The values for the input from the time step
prior to any number of consecutive previous
periods are averaged.

The QnD modeling system was applied to the Jamaican food system by translating the

results of a soft systems analysis into a conceptual object format (described in detail in chapter

four). This structure forms the backbone of QnD:Jamaica, a functioning food system simulation

model. The next section describes the specific methods employed during field work to build a

rich picture that could inform model design.

Informing Decision Support Tool Development

An initial list of potential collaborators identified general stakeholder groups according to

their structural location within the system. A review of Jamaican food system literature

suggested that tourism, small scale agriculture, international trade, government and

environmental interests are major subsystems influencing food security. A general list of

participants included small farming communities, tourist industry actors (traditional and eco-

based), informal and formal food distribution actors, agricultural, nutritional, and environmental

NGOs, and Ministries involved in food system management. To this general list were added

specific individuals that had proactively explored and/or participated directly in management

issues related to Jamaican food security as evidenced in academic papers and governmental

reports as well the leadership lists for organizations involved in food security. Where feasible,

potential participants were contacted via email to ascertain their availability in Jamaica during









July 2006. As it was not possible to identify by name or contact many stakeholders from outside

of the country, they were solicited in person once field work began.

The first interviewees included three professors from the University of West Indies (UWI-

Mona) and a researcher at the Caribbean Food and Nutrition Institute. During the first interviews

specific individuals from the above general stakeholder categories were suggested as additional

participants in a "snowball sampling" manner. These included a divisional director within the

Ministry of Agriculture, a rural development field officer, an ecotourism outfit owner, members

of a parish-level farming cooperative, a livestock farmer, and a manager of a local supermarket.

During the interviews, the problem situation as understood by the stakeholders was

investigated through the scenario planning approach (van der Heij den, 1996) and the Soft

Systems Methodology (Checkland, 1979, 1999). First, concerns about the system were

expressed in a general manner. The collaborations began with one general question similar to

the one suggested by Quinn and Voyer (1996): "What do you believe are the biggest challenges

to Jamaican food security?" The ensuing discussion loosely followed the question structure as

proposed by van der Heijden (1996). The conversation was often supplemented by questions

directly related to the interviewee's field of expertise. The interviews were recorded and

reviewed later to determine what and why key issues were of concern and attributed to which

stakeholders. In accordance with the scenario methods described above (van der Heijden, 1996)

a thematic database was created that clustered the main issues affecting food security by their

understood internal or external influence on the situation.









CHAPTER 4
JAMAICAN FOOD SYSTEM: AN ILLUSTRATIVE CASE

The pathways that couple production and consumption in small open economies were

revealed in order to be represented in the food system model QnD:Jamaica. The actors and

interactions involved in the Jamaican food system were reviewed within the qualitative

framework provided from the results of soft methods research performed in the field, and

organized into conceptual groups that loosely followed the framework suggested by the research

group GECAFS. The qualities examined were socioeconomic populations, land uses, food

pricing, distribution and marketing structures, and consumption issues as well as how each

subject is embedded in the historical context. This systematic analysis structure provides an

approach that can guide future meso scale explorations of other food systems and be used in

distilling features the meso scale of food systems in general.

Within this framework, empirical data were reviewed for the period of 1986-2004 to

develop quantitative descriptions for the objects that structure QnD:Jamaica. This period was

chosen because data existed for occurrence of multiple political, economic, and climatic events.

These include internal structural adjustments made in accordance with International Monetary

Fund (IMF) agreements such as exchange rate deregulation and market liberalization, hurricanes

Gilbert and Ivan, additional known drought and flood events, and the September 11th attacks on

the US, which affected international socioeconomic conditions.

Historical Review

The 1970s in Jamaica were characterized by a strong national state that utilized extensive

price controls, subsidies and nationalization of primary industries to solve poverty issues. Prime

Minister Michael Manley's People's National Party (PNP) pursued a policy in the early 1970s of

import substitution in efforts to strengthen the Jamaican economy. However, by the late 1970s









the country was heavily indebted to international financial institutions and began a series of

extreme economic reforms in the 1980s and 1990s such as devaluation of the Jamaican dollar,

elimination of state marketing boards, trade liberalization, and privatization of state and public

owned industries.

The Exchange Rate, Inflation and Food Imports

Between 1981 and 1985, government support for agriculture was heavily withdrawn. This

was due to overall decreases in government spending as well as a decrease in the percentage of

spending on agriculture from 19% in 1975 to less than 4 % by 1990 (Weiss, 2004a). In addition

to reduced internal supports, increased input costs, and the overall rise in the cost of living,

Jamaican agriculture also suffered from the reform policies of AGRO-21.

Designed with help from USAID, the policies of AGRO-21 aimed to free up labor from

agriculture and make it available to sectors of the economy in which Jamaica was believed to

have a comparative advantage: mass tourism and manufacturing. However, as labor was shed

without increases in mechanization or other productivity supports, agricultural production

volumes began to fall (Witter, 1989 in Weiss, 2004a). Additionally, the new industries proved

unable to support the extra labor. During the late 1980s and early 1990s there was a mass

outward migration of one in twelve Jamaicans (Anderson and Witter, 1994). This experience

illustrates Weiss' conclusion that in contrast with free market theory, "the decline of one sector

often produces serious dislocations without concurrent gains elsewhere" (2004, 474). Jamaica

clearly exemplifies this with the trade deficit quadrupling between 1980 and 1990 and then

almost tripling again between 1990 and 2004.

Between 1989 and 1992, monetary policy makers in Jamaica aimed to stabilize prices

through macro economic tools (Robinson and Robinson, 1997; Planning Institute of Jamaica

1990-1994). Bound by IMF agreements made during the late 1980s, the Jamaican government









followed a policy of currency devaluation that, in fact, actually increased the prices of consumer

goods. Between 1990 and 1991 the Jamaican dollar depreciated against the American dollar by

60.9 % compared to 25.7% between 1989 and 1990 (Planning Institute of Jamaica, 1991).

According to the financial transmission model put forth by Allen et al., the exchange rate

exhibits the most significant effects on consumer prices in Jamaica and is linked to the

importance imported goods in the economy (2002).

Currency devaluations were expected to increase domestic foods' competitiveness. The

rising costs of imported production supplies, however, constrained the Jamaican farmer, and

food prices rose (Planning Institute of Jamaica, 1992). As the purchasing power of the Jamaican

dollar plummeted between 1989 and 1992, consumer prices steadily rose increasing 268% point

to point. Of the eight regularly tracked household consumption groups, the food and drink

category is weighted highest at 55.5% within the Jamaican CPI. Thus, food price increases

generally have a large influence on the overall inflation rate.

An emergency Food Assistance Program was introduced in the early 1990s to prevent food

riots over the high cost of food. This only served to strengthen the pattern of importation by

subsidizing imported food staples rather than local ones. "The bias in consumption habits

toward foreign foods" was fueled by making it "rational from the point of view of households

maximizing nutrients per dollar to buy imported instead of local foods" (Witter, 1989 in Weiss,

2004a, 468).

The 1990s also saw reductions in trade barriers furthering the upward spiral in cheap

imported food. Between 1991 and 2001 Jamaica's total food and beverage imports increased by

two and a half times. The largest increases were in cereals, feeds, and meat products which

account for half of the value of Jamaica's agro-imports (Weiss, 2004a). While domestic poultry









production grew, it was fueled by US subsidized imported feedstuffs. Also, imported livestock

amounts increased more than twice as fast during the 1990s. Per capital meat consumption

increased by 55% between 1980 and 2000 moving to the center of the diet where it was

previously relatively rare (Weiss, 2003).

Economic Development and Poverty Alleviation

Records indicate a decrease in poverty, from 30.5% in 1989 to 16.9% in 1999, while

unemployment remained fairly constant around 15-17% (Osei, 2002; Le Franc and Downes,

2001; Salmon, 2002; King, 1998). This conundrum addresses the real drivers behind poverty

reduction in the 1990s and supports the model that Jamaican consumption and production

activities are highly affected by the external environment. While Salmon puts some faith in the

success of the government's National Poverty Eradication Program at limiting poverty, Osei,

King, and Le Franc and Downes are more skeptical about the government's role in increased

consumption levels (Ibid.). Real price decreases and real wage increases, neither caused by

governmental policy, played the largest role in increased consumption levels (Ibid.). The

informal sector, in terms of domestic and international trade (both legal and illegal) and

remittances from abroad, were the key factors to reducing the numbers of extreme poor.

Le Franc and Downes use the strong correlation between reduction in the poverty ratio as

well as the inflation rate as evidence that falling prices have been the main driver of increased

consumption (2001). They also make the important point that price changes not only affect

income's purchasing power but also have substitution effects in which consumption patterns, and

thus nutritional patterns, are changed.

These conclusions agree with earlier findings that absolute poverty was reduced by a

recovery in real wages through informal activities, not economic reforms (King, 1998). The

informal sector grew from 14% to 29% of the labor force between 1972 and 1997, and accounted









for 30% of the labor force in 1999 (36% of total employment) (Osei, 2002). Decreases in

unemployment have been shown to be the result of a decline in the number registered in the

work force rather than an actual increase in employment (ECLAC, 2004). As Weiss describes,

"Jamaica's economy now rests on a precarious balance of external migration/remittances,

tourism, drugs, and borrowing that is layered upon a notoriously high level of inequality" (2004,

483).

Climatic Shocks

The Jamaican population's food security was also affected by climatic events. These

natural disaster events affected food accessibility by increasing food costs as well as other

consumption expenses in the form of household repairs and through the loss of jobs. If

consumption expenditures return to pre-disaster levels before domestic production recovers,

there can be a sharp increase in the relative importance of imported goods in the total

consumption budget (Brown, 1994). For these reasons, the components and processes of the

Jamaican food system must be evaluated within the dynamic context of climatic shocks.

Hurricane Gilbert

Hurricane Gilbert passed directly over the center of the island on September 12, 1988.

Damage from this category five hurricane was estimated at $US 4 billion, with the agricultural

sector accounting for 40% of this amount (PAHO, 1988). Gilbert affected 810,000 people and

damage amounted to 65% of GDP (ECLAC, 2004).

Hurricanes have major economic effects on the balance of trade because of the increased

expenditures on capital goods, most of which are imported, as well as the destruction of export

goods that generate foreign exchange earnings (Brown, 1994; ECLAC, 2004). In 1988, an

inflation rate of 7% was expected (Brown, 1994). Prior to Gilbert, the inflation rate was at 3.1%,

on target to stay within the projected limit (Ibid.). However, the shortfalls in balance of trade









caused by increased influx of food and rebuilding materials compounded with the decrease in

exported agricultural goods from crop damage caused increased inflation in the post-hurricane

months (Ibid.). In November, a rate of 14% was projected, however, by the end of the fiscal year

the rates had recovered and final inflation for the year was 9.1% (Ibid.).

Hurricane Ivan

The rate of inflation in 2002 and 2003 was 7.3% and 14.1 % respectively, the highest in six

years. This increase is believed to be caused by further depreciation in the exchange rate

compounded by increased oil prices (ECLAC, 2004). At the beginning of 2004 a decrease in

inflation was projected for the year from 14% to 9%. In the first half of the year domestic

agriculture experienced a 6.5% gross decline due to dry conditions in the central and western

parishes where the majority of domestic food is produced. This led to food shortages and the

food and drink category drove 80% of the inflation in the second quarter of 2004 as opposed to

28% in the first quarter (Planning Institute of Jamaica, 2004).

Category 4 Hurricane Ivan passed from east to west, just south of Jamaica in September

2004. Most affected were the parishes of St. Catherine, Clarendon, Manchester, St. James,

Hanover and St. Mary. A total of 102,000 people or 14% reported damage to households

(ECLAC, 2004). Of those evaluated, 9% were completely destroyed, 75% severely damaged

and 14% had only minor damage. Total damage was estimated at $J 11,163 million with $J

3,666 million of the total reconstruction costs coming from imported goods. An estimated $J

137, 235 (or $US 2,287 in 2004 dollars) reconstruction expenditure per house was estimated.

Two groups most significantly affected by Ivan were farmers and fishers, particularly those

on the southern coasts (ECLAC, 2004). An estimated 11,100 hectares of domestic agricultural

land was affected with 117,700 farmers sustaining losses. There was $J 2,632 million damage to









the domestic sector with legumes sustaining $J 43.2 million, vegetables J$ 396.4 million, and

root crops J$ 570.6 million in losses (2004 J$).

Banana production for the year was entirely destroyed. This sector's devastation put 8000

persons out of work for a period of six to nine months (Ibid.). Full recovery of tourism

infrastructure occurred three months after the event (Ibid.). Efforts were made to employ

workers in rehabilitation efforts, but temporary loss of employment was inevitable due to the

decrease in tourists (Ibid.).

Road infrastructure was another of the significant areas to sustain damage. However,

direct damage costs were much less than the indirect losses sustained because of the missing

support that the infrastructure provides to other sectors (Ibid.). Floods and landslides cut off

entire sections of roads, damaged retaining walls and bridge approaches, and deposited silt on

roadways. Slippage of entire sections of roads occurred from the saturation of land and raging

rivers scoured road edges. However, a single lane of traffic was ensured on all major roadways

by the end of the month. Although an assessment of the indirect unit transport costs increases

was not available, the increases were expected to be "very high in monetary terms-that will have

a negative bearing on the population's well-being." (Ibid., 36).

Holistic Exploration

After a quarter century of pursuing a failing model of an export led economy, Jamaica has

now become what Witter terms a 'consuming appendage' of the US economy (2000 in Wess,

2004). Propositions that consider a more balanced and diversified model of economic

development are presently being put forth (MOA, 2005). However, changes to the current

internal food system must proceed with caution. Policy designed to improve or adjust the current

Jamaican food system is hemmed in by the large numbers of the Jamaican population who

depend directly on informal food production and distribution for their livelihoods. The creation









in 1963 of the Agricultural Marketing Corporation (AMC) to consolidate distribution,

wholesaling, and retailing operations appeared to be a solution to the inefficiencies of the

Jamaican food system. Instead, the corporation resulted in massive livelihood losses to sectors

of the traditional system. Similarly, campaigns to "clean up Kingston" and remove informal

traders were also unsuccessful (Espeut, 2003).

The current food production and distribution system persists because it functions as a last

resort informal income source pursued by the large amounts of surplus labor in the country

(Lewis, 1954). In other words it is a symptom of larger societal structure and cannot be

superficially reformed. The internal mechanisms that transmit economic, social and

environmental effects and feedbacks within the Jamaican food system must be explored in

tandem so that the outcomes of complex interacting components can be weighed in a holistic

context.

The next section describes the rich picture that resulted from the soft systems analysis

which aims to provide a more holistic conceptualization of the issues involved in Jamaican food

security. This is followed by a detailed empirical review of the five explicit conceptual

categories revealed through the rich picture and chosen to be represented in QnD:Jamaica.

These categories include population typology, land use, food distribution, food pricing, and

nutritional consumption.

Rich Picture Reveals Conceptual Framework

Main areas of concern affecting food security revealed by stakeholders through the soft

systems approach could be clustered by internal and external socioeconomic and ecological

drivers. External issues included changes in tourism visits, changes in export markets (closing:

bananas, opening: niche), prices of imported food, decreasing nutritional and cultural value of

imported food, and hurricane effects on infrastructure, tourism and productivity and rainfall










effects on productivity. Each of these external concerns also relates to key internal issue

clusters. For instance, land use change decisions are internal drivers but are closely tied to

external drivers such as closing banana markets. Changed utilization of this newly available land

raises new concerns in terms of maintenance of ecological integrity and economic exploitation.

If niche markets open, what institutional and infrastructural organization is necessary to take

advantage? How can tourism be exploited without sacrificing ecological assets? What about the

current production system reduces food security? What rural economic activities besides

agriculture are options? These issues in addition to others are graphically represented in the rich

picture (Figure 4-1).

I I


Figure 4-1 Rich picture of the key themes of Jamaican food security.

One of the above key issues that lends itself to a potential transformation for decreasing

food system vulnerability is the farmer/market integration concern cluster. This cluster includes


Regional Integration?


m Reefs, Fish stocks
S& Fishing culture










concern over small farmers' returns from crop sales. The TWOACE conceptualization of the

farmer remuneration system is shown in Table 4-1.

Table 4-1 TWOACE characteristics of farmer remuneration system.
Farmer remuneration system
T transformation: What input transformed into Farmers receive little share of market price
output? ->Farmers greater share of market price
W orldview: What world view makes Domestic agriculture strengthens food security.
transformation meaningful?
0 owners: Who could abolish system? International institutions, Jamaican government,
higglers/middlemen
A ctors: Who will perform activities to make Non-governmental rural development advocacy
transformation happen? groups, farmers cooperatives
C customers: Who will benefit/suffer from system? Farmers, citizens of Jamaica
E environmental constraints: What constrains the Minimal tariffs on imported food, farmers have little
system performing the transformation? clout, opportunities for farmers to sell directly are
limited

Other themes for transformation where the TWOACE analysis method was applied

include: tourism returns to community, hurricane destruction on agriculture, value added

processing in agriculture, regulation of distribution, coordination of production decisions,

improved foreign marketing system.

Often during a TWOACE analysis of a potential transformation, underlying influences

surface to an extent where examination must stop so the issue could be deconstructed further.

For instance while "improving returns to farmers" was the initial issue distilled from the

farmer/market integration cluster, when the contextual environment that constrained this

transformation was made explicit, new transformations that would need to occur prior to or in

tandem with the examined transformation were revealed. In this case, the lack of remuneration

is constrained externally by WTO trade agreements that limit the amount of protect ional tariffs

Jamaica can levy on imported goods (often produced by multinational organizations under

economies of scale or agricultural subsidization). Increases to remunerations are constrained

internally by the low incentive for distributors to pay farmers more. Related to this is lack of









clout farmers wield in order to demand more. However farmers are often reluctant to organize.

These new concerns now become new key areas for potential transformation to be examined.

The qualitative descriptions were conceptualized by a formal framework to identify a

limited number of quantifiable parameters and drivers of interest for the scenario model. The

formal framework chosen to organize the scenarios and parameters of interest was developed by

the international research group Global Climate Change and Food Systems (GECAFS). This

group studies new policy formulation and management methods that could increase food and

livelihood security around the world in the face of changing environmental conditions. By

studying multiple national and regional food systems, they hope to uncover general trends in

how certain structural features of food systems benefit or harm aspects of food security. Inspired

by the underlying assumptions of nested systems' emergent properties, the GECAFS framework

aims to create food system descriptions that account for the interactions between components at

multiple spatial and temporal scales (Ericksen 2006a, 2006b).

This framework assumes that by analyzing the variation that occurs in a specific system's

internal interactions and outcomes as it is exposed to different drivers and driver strengths key

features that provide systemic strength or weakness under given scenarios can be revealed. This

framework divides the areas of interest into activities, drivers, and outcomes and explores

interactions between them. Activities refer the tangible structures and actors involved in

producing, processing, distributing and consuming of food. Outcomes refer not only to food

security but also to effects on social welfare and environmental security and the interactions

between these qualities. Interactions between socioecological contextual drivers and activities

produce varying conditions of food security in terms of availability, access and utilization.









Patterns in these interactions are revealed by evaluating each food security component:

utilization, access, and availability.

Having identified key actors in the Jamaican food system through the soft systems

approach it was simple to apply the GECAFS conceptual model and make explicit the activities

that affect food security outcomes. The large degree to which these activities occur external to

Jamaica is evident (Figure 4-2).



JAMAIC A External

small r---6---le |
l dcnSmall [T Large- scale,
Producing Farmers Farmers



Manufacturers

Informal Formal IInpli le.I
1liddlemen Distributors [' %pIn lei


Consuming Rural UrbanTourist




Figure 4-2 Food system activities and actors.

The GECAFS framework builds upon some of the soft system results to bring new insights

to the situation. When assessing the interactions of areas of key concern on availability

conditions in terms of production, distribution and exchange, key features of the Jamaican

system that directly influenced model development were revealed (Table 4-2). It became evident

that in Jamaica availability is influenced differently in terms of food type, domestic/imported and

population type, rural/urban. This indicates that the food system activities-driver interactions

should be explored separately for their effects on the two food and population types in the

decision support tool.










Table 4-2 Jamaican food system interactions that affect availability conditions.
Food type Location of Rural population Urban population
interactions
Production Water Water
Land Land
Climate shocks Climate shocks
Production decisions Production decisions

Distribution Few Road Infrastructure
O
SHiggler infrastructure
Exchange Few Relative import/domestic prices
Production External changes External changes
Distribution Road infrastructure Road infrastructure
0 Market infrastructure Market infrastructure

Exchange Domestic production Domestic production (to a lesser
Extent than in rural)


Empirical Review of Conceptual Categories for Model Parameterization

The elements of interaction that were revealed during Phase one were organized following

the object oriented approach in which anything that can be conceptualized can be viewed as an

object. The concepts in object oriented approaches are nested in hierarchies that capture the

multi-scalar nature of reality. QnD:Jamaica was structured by arranging the drivers, activities,

and outcomes as descriptive data objects in a hierarchy that reflects five conceptual categories of

concern: population typology, land use, food distribution, food pricing, and nutritional

consumption (Figures 4-3 and 4-4). The sections that follow discuss these five categories'

detailed empirical backgrounds which were used to parameterize QnD:Jamaica.





Object
iObject Object/Component Descriptive Data
Hierarchy


Jamaica


Hurricanes
Exchange rate


Figure 4-3 Depiction of QnD:Jamaica's hierarchical object oriented structure.
Hi.e ObjectComponent Descriptive Data
thr. h..&th.


Figure 4-4 Depiction of Local Components descriptive data.


;I-









Population Typology

The small farmer communities in Jamaica (and the whole of the Caribbean) possess a

unique set of qualities that distinguish them from other small scale agriculturalists around the

world. The Caribbean "peasantry" should be understood as a "reconstituted" peasantry (Mintz,

1974) that emerged from an industrial form of production: the plantation. This agricultural

structure which was "capitalist and commercial in its economic relations and urban in its value

orientations" (Le Franc, 1981, 1) has had a number of implications for the present structure of

the Caribbean agriculturalist and rural society as a whole.

Subsistence farming as conventionally defined does not exist in Jamaica. While the small

farm often provides direct sustenance to its cultivator, there is always a market orientation even

for the smallest farm producing the smallest bit of crops (Spence, 1996). This current production

situation is built upon the remnants of the provisional gardens and trade network that originated

during the colonial period. While the small scale farming sector supplies a majority of domestic

food, the large scale export sector continues to benefit from higher quality land and infrastructure

access, fueling an antagonistic tension between the dualistic agricultural sectors (Beckford, 1972;

Edwards, 1961).

80% of Jamaican farmers (800,000) work less than 25% of the land (400,000 hectares) on

farms less than 5 hectares each. These farmers provide the vast majority of domestically grown

food (STATINJA, 1996; Goldsmith, 1981). Additionally, Spence has shown the impressive

extent to which small farmers engage not only in domestic market production but also in export

orientation (1996, 1999). Not only do small farmers diversify between domestic and export

production, but they also expand their income to include a variety of non-farming sources.










Data from the most recent Census of Agriculture corroborates the qualitative expressions

of the existence of "baskets" of livelihood strategies employed by small farmers (1996). Figure

4-5 shows the diversification of small farms' primary income sources across parishes.3


100%
90%
90% O No work
80%
70%I I - Other wage
S60% 0 Agriculturalwage
2 50% P Farm sales
n 40%
30%
20%
10%
0%




Parish

Figure 4-5 Primary income source of small farms by parish. (STATINJA, 1996)

However, this diversification of income strategies appears to be a means to finance the

main intended activity of farming (Ishemo et al. 2006). Le Franc explains that this variety of

diversification strategies serves as the key starting point for developing a typology for

examination of the socioeconomic system:

"Off farm employment provides much-needed work capital and funds for domestic
consumption, particularly during slack periods...because of the need to earn income from a
variety of sources, it makes little sense to try to understand the social and economic
divisions simply in terms of land ownership. Instead it is only a guide and it is necessary
to describe the socio-economic strata and the hierarchy they form in terms of particular
combinations of economic pursuits" (1981)

The key reason behind these diversifications is the inability of agricultural sales alone to

generate enough income to support a family within the increasingly cash dominated environment


3 There are also significant differences between income strategies employed by female versus male headed
households, i.e. service wage earning females vs. agricultural wage earning males. Only combined sex averages are
shown in Figure 4-1.









(LeFranc, 1981). Historically, the free exchange of labor between households was a common

practice in Jamaica and bolstered agricultural profit margins. However, the extent that exchange

can be utilized has been changing over time and differs across parishes exemplifying the

dynamic pressures of the cash economy (Ishemo et al., 2006). Presently, the use of wage labor

by small farmers even when sufficient family labor exists on farm is common (LeFranc, 1981;

Ishemo et al., 2006). While 76% of surveyed migrant farmers moving from other parishes into

the Rio Grande Valley (an area in north eastern Jamaica dominated by both small and large scale

export banana production) claimed the use exchange labor prior to their move, 71% explained

that they currently relied instead on wage labor (Ishemo et al., 2006). Many farms, however, do

not possess the means to transition from exchange labor to wage labor resulting in labor

shortages while unemployment persists (Ibid.).

There are differences not only in the types of income diversification available across the

island but also in the levels of land ownership within rural populations. Figure 4-6 displays the

distribution of rural economic units by their landownership status at the time of the last census

(STATINJA, 1996).










25000

20 Landed Economic Units
Prs 20000 -
C2000 Non-landed economic units

E15000 -
C
o 10000 -
LJ
4--
"5000 -
-0





Parish S S

Figure 4-6 Landownership of rural economic units by parish. (STATINJA, 1996)

Land use and Agricultural Productivity

There are a number of grey areas with regard to interpreting survey data on how small

farmers utilize their land (Spence, 1996). In particular, the area of land in ruinate or pasture must

be interpreted flexibly as farmers often use the term pasture for any areas of grass whether used

for livestock or not, and land in ruinate can either be truly in fallow or might be brought into use

at any point when needed (Ibid.). An equally slippery slope is the very line drawn between

export and domestic crop areas as intercropped food forests are common.

Despite these grey areas, attempts to quantify the export/domestic split on small farms

exist (Spence, 1996, 1999; Davis-Morrison and Barker, 1997; STATINJA, 1996). Spence

documents an export oriented bias of land allocation among small scale farmers in the parish of

Clarendon which he describes as "being representative of the island as a whole" (1996, 133).

This corroborates previous descriptions of this bias despite the small scale sector's role in

supplying the majority of domestic food (Goldsmith, 1981). Spence attributes the bias, to not

only a more favorable export infrastructure but also to the social factors related to the "structural









dynamics of small farming" (1996, 133). The first factor includes access to better credit and

ensured markets but even more reliable and timely transport and payment aspects. The second

influential factor relates to the crop types' labor intensity, environmental needs and level of long

term investment/commitment. He cites farmers' age as the single most important factor

contributing to land use decisions because of its influence on land accumulation, physical

capabilities and level of commitment (1996).

The 1996 Agricultural Census data show that the export bias on small farms does not hold

for all parishes (Figure 4-7). St. Elizabeth, St. Ann, Manchester and Trelawny all show a greater

proportion of land uses in domestic crop production.

12000

10000 Not cultivated
E Domestic crops
8000 El Export crops
S6000

4 0 0 0 -

2000

0




Parish

Figure 4-7 Small farm land use in terms of crop type. (STATINJA, 1996)

The extent to which farmers respond to market changes and alter the types or amounts of

crops in production varies throughout the literature. A microeconomic modeling study in

Mexico showed small scale farmers modifying production in "perverse ways" (Dyer et al. 2006,









279) in response to market price shocks because of interactions with previously unaccounted for

social factors.

In Jamaica, opinions about how and why farmers make decisions are diverse. Spence

concludes his studies discussed above by explaining that Jamaican farmer's generally respond

little to price changes. However, Davis-Morrison and Barker document the dynamic patterns of

production shifts on small farms in Portland parish since the 1970s (1997). They describe how

the collapse of the export banana market in combination with increased support for domestic

crops in the late 1970s and early 1980s led to the dominance of yam cultivation on small farms.

However, the heavy destruction of this crop (as well as the bamboo needed to stake the yam

vines) during Hurricane Gilbert in combination with an increasing export market for fresh

produce caused farmers to turn to dasheen farming in the late 1980s. Alas, an extreme market

glut in 1993 caused this market to fall out and farmers switched back to bananas.

Ballayram and Davis suggest that Jamaica's export supply responds to price incentives

albeit slowly and at varying rates by crop (2003). They point out that price variability has been

high during their study period and that while nominal value has increased significantly, real crop

prices have remained stable or declined, likely do to movements in the exchange rate and the

general price level. An important insight they provide into the Jamaican producer's logic is their

low ability to mitigate risk. Vulnerability encourages non-movement in times of high variability.

The newly competitive, and risky, market environment likely served to dampen producers' price

responsiveness as compared to guaranteed markets and price the security that existed during the

protectionist 1960s and 1970s.

Domestic Food Distribution

Approximately 80% of domestic produce is distributed and marketed by the informal

higglering system (LeFranc, 1981). The use of the term 'higglering' for petty trading activity









has a uniquely Jamaican origin (Mintz, 1974). Higglers purchase excess supply from small

farmers and transport and sell the food staples in various market establishments. There are three

types of domestic produce higglers: 1) Higglers who sell only their own produce; 2) Higglers

who sell their own as well as others' produce; and 3) Higglers who only buy and sell (FAO,

2003). Most higglers live in rural areas and are located at the lower end of the income spectrum

earning an average of less than $J 50 per week in 1977 (Witter, 1988). Today's higglering

situation is very similar to the colonial marketing network that developed when Jamaican slaves

were required to produce a portion of their food in provisional gardens and allowed to sell any

surplus crops4. By 1735 the Sunday markets were formally recognized as places where slaves

could sell "fresh fish, milk, poultry, and other small stock of all kind" (Mintz and Hall, 1960).

Jamaican producers are generally at the mercy of the higglers in terms of how much they

sell, and at what price. Ruijs et al. explain that developing countries' domestic food supply is

often price inelastic (at least in the short run) because farmers or traders cannot respond quickly

enough due to low storage abilities among other factors (2004). These features challenge

traditional economic theory that expects reductions in transport and marketing costs to decrease

consumer prices and increase farmer prices leading to increased trade. Similarly, Wong implies

that in Jamaica, and other developing countries, the "'laws of supply and demand' might not

hold" (1996, 518) due to the inherent nature of the higgler as a petty trader. Like a subsistence

farmer, the higgler is trying to maximize highly limited resources.

Estimates range between 25-50% for regular post harvest losses and up to 100% for semi-

regular losses (FAO, 2003). Hundreds of kilograms of food rot in the fields during seasonal

4 The ruling English maintained their tastes for European food and imported the majority of their foodstuffs at high
costs. The plantation owners had a large aversion towards putting potential cash crop land into foodstuff production
for on island consumption. Thus provisional slave gardens were a way to reduce operating costs. This cultural
phenomenon was the original source of the 'consuming appendage' situation that exists today. Since, the islands
inception there has been little effort to promote import substitution, excluding M. Manley's early terms.









gluts when prices in nearby urban markets are unrepresentatively high (JSDNP). It is unclear if

these situations result from logistical collection obstacles, real lack of market and/or intentionally

maintained low supply levels by higglers to earn a greater profit at the expense of farmer's

earnings.

The potential of the first explanatory factor is clear upon observation of the egg carton-like

cockpit country of the Jamaican interior. Jamaica's parishes are linked by a higher quality, Class

A, outer ring road that loops the island and by lesser quality, Class B, roads that link north to

south through the hilly interior. Class C roads with limited paving connect isolated communities

throughout the rugged countryside to the greater road network. Although the Jamaican road

network is extensive, it is in very poor condition. Road quality deteriorates as one progresses

further from the main arteries often turning to dirt paths that become impassable in heavy rains.

Higglers often charge more to transport crops from areas with poor roads and some exporters

will not travel to certain regions due to bad road conditions.

The high dispersal of the small farming sector across this rugged landscape with each farm

producing just a small amount for sale has been blamed often for the obvious challenge it poses

for would be food distributors (Mintz, 1974; Norton and Symanski, 1975). Katzin relays a

typical historical scenario with a higgler traversing across paths accessible only by foot or

donkey for a daily collection of nineteen bunches of carrots, twelve scallions, five of beets and

five of turnips (1959 in Norton and Symanski, 1975).

Over the past decade the price differentials between domestic goods and subsidized

industrially produced foreign foods have driven certain sectors of Jamaican agriculture out of

business particularly onions, Irish potatoes, and the dairy industry (FAOSTATS). At present,

certain vegetables such as tomatoes, cabbage, and carrots are facing similar price competition.









However, the majority of Jamaica's vegetables are traditional crops such as calalloo and choo

choo, which do not face this direct competition. Although, even in these sectors, there is

evidence that the prices of traditional crops are being driven down as consumers substitute for

cheaper non-traditional vegetables and as preferences shift. Many farmers complain that the low

prices of imported foods relative to domestic crops make it less desirable for higglers to collect

their produce. Farmers go further to explain that the strength of the underground import market

lures farmers out of agriculture and into illegal food importing. Youth express concern explicitly

about the troubled marketing system when they explain why they will not pursue farming

(JSDNP).

Especially since 2001, when the cell phone market was deregulated,5 higglers know what

the consumer market prices for both domestic and imported foods are and will set their buy

prices for farmers accordingly. A MOA official related that higglers will often limit the amount

of produce they take to market in order to drive prices up. The fixed handling costs charged by

higglers are on a per container basis even though there is no uniform container size and include

transport and stall fees but are assumed to be low (Wong, 1996). Additionally, farmers are

reluctant to sign contracts with distributors because the farmers expect them to buy outside of the

contract in order to force the farmer's price down (FAO, 2003).

Food Marketing

The Jamaican marketing infrastructure consists of modern retail outlets as well as

government maintained parochial markets and informal street vendors (Norton and Symanski,

1975). Although modern grocery stores are transforming the retail sector in larger urban areas




5 The cell phone has revolutionized both farmers and higglers information access and networking ability (Hourst,
Martin 2006). Approximately 2 million Jamaicans have cell phones, that is three per household.









such as Kingston, Mandeville, May Pen, and Montego Bay, traditional markets are still the major

source of local foodstuff trade outside of these areas.

Norton and Symanski categorize the traditional parochial markets by the number of

higglers present (1975). Small markets usually have less than twenty higglers, while upwards of

sixty will be present in a large market. They also show a strong correlation between the size of a

community and the extent of its market. They showed that only four of Jamaica's twenty-nine

large markets occur in communities with less than two thousand inhabitants. In the same way,

only three of the thirty-five small markets are held in places with more than three thousand

residents (Ibid.). The markets are often weekly but occur more often in some places depending

on supply and demand (Ibid.).

Each parish has one or more centralized markets in its major urban areas in addition to

smaller markets throughout. The central markets are more likely to serve as retail and wholesale

centers where country higglers bring in food from rural areas and sell to town higglers who will

sell in peripheral suburban markets. This is particularly true of the biggest single market,

Coronation Market in Kingston, which serves as a hub for distribution to other parishes as well

as at least eight other daily urban and suburban markets in Kingston (Norton and Symanski,

1975). Witter estimates that 15,000 higglers serve the Kingston area alone (1988).

Norton and Symanski illustrate Coronation's significance with a survey showing that

only fifteen of sixty-four markets nationwide have no higglers with supplies from Kingston's

central market (1975). They also list factors that serve to guide development of a nationwide

market typology: distance to Coronation market, size and nature of local market, regional

specialization and seasonal supply and demand factors. Other parishes also have central









wholesale markets that serve the wider countryside, though not on the scale of Coronation. These

include Savanna-la-Mar in St. Elizabeth, and Montego Bay in Saint James (Ibid).

Most Jamaican markets occur around the same time usually towards the end of the week

making it difficult for higglers to work the national circuit and efficiently disperse supply to

spots of higher demand. This often leads to a wide range of prices throughout the country (FAO,

2003). The price differences even between Coronation Market and Kingston's secondary

markets such as Papine and Constant Spring are high and there is a thriving town higgler system

of buying from one market (even modern supermarkets!) and selling in secondary suburban

markets. Coronation Market's size often ensures lower prices, certainly in Kingston and often

compared to other parish markets as well.

Imported Food Distribution and Marketing

Many retailers have increased the share of imported to domestically produced foods

because of ensured supplies, reduced tariffs and a stable exchange rate (USDA, 2003). Some

additional conditions that give US products an advantage in Jamaica include massive product

advertisement exposure through increased cable and internet access, and increased demand for

ready to eat products which are not manufactured domestically. Both traditional Jamaican food

outlets and supermarkets use a middle man distributor to acquire the majority of their imported

foods. However, Jamaican grocers are increasingly aligning with international chain stores and

product lines to consolidate the distribution and marketing sector both vertically and

horizontally, although direct importing by large retailers still contributes only 5% to their total

imports (Ibid.). A challenge to realizing the benefits of direct importation is limited storage

space therefore retailers are still largely supplied by local distributors.

Recently, grocers have been unable to meet operational costs through sales earnings

because prices are so low in part from the highly competitive market (Gordan 2006). Not









surprisingly, the Jamaican consumer is highly price responsive and the extreme price

competition between retail and wholesale outlets is causing the line between their roles to blur

(USDA, 2003). The competition from supermarket chains like HiLo and SuperPlus are leading

independent retailers to form alliances and also expand into rural towns to tap new markets

(Gordan, 2006).

Wholesalers and informal commercial importers also supply the traditional outlets such

as the parochial markets and small grocers with imported foodstuffs. These operate mostly in

rural and lower class urban areas but face increased competition from the encroaching larger

retailers. This sector is expected to be displaced by chain and independent large scale retailers in

the medium to long term (USDA, 2003).

The Jamaican Consumer

Calorie intake is not a complete measure of nutrition. The amount of energy supplied by

each macronutrient: carbohydrates, protein, fat etc., has important implications for a population's

food security. Dietary choices are not a cold function of cost per calorie. Rather cultural and

innate human preferences for variety as well as sugars and fats play a large role in a consumer' s

food buying strategy. Popkin coined the term 'nutrition transition' to describe the dietary shifts

that usually occur along with typical demographic and epidemiological changes in a country's

development process (2003). The two main features of the nutrition transition are the reduced

role of carbohydrates and the increased role of animal fats in providing caloric energy supply.

This increased role of animal fats has important implications for international grain markets and

ecological resources world wide. These transitions occur due to a number of factors including

urbanization, technological changes, economic growth and cultural shifts.

National caloric availability data does provide insight into present macroeconomic self-

sufficiency issues in Jamaica indicating that the amount of imports in the country alone are









enough to satisfy annual per capital caloric needs (Ballayram and Henry, 2004). However, it is

important to 'unpack' the imported supply as well as domestic foods in terms of what they

provide nutritionally and how they are distributed geographically, demographically and

temporally throughout Jamaica.

Current data records in Jamaica account for national availability of calories, or Dietary

Energy Supply (DES), by food group and imported vs. domestic sourcing, as well as income

level food expenditure amounts by food group (FAO, 2003; FAOSTAT; Ballayram and Henry,

2004; Survey of Living Conditions, 2002, 2003). Much Jamaican developmental literature

suggests that between 1989 and 1998 the Jamaican diet was comprised mainly of domestic

products (MOA, 2001). These documents suggest that only in 1999 did consumption of

domestic foods fall below that of imported foods with 269.3kg to 307.3 kg/yr/capita respectively

(Ibid.). This data is actually describing only potential intake of domestic foods (availability) and,

as described in the 'Transport and Marketing' section, a lot of domestic produce during this

period wasted in the fields (JSDNP).

The Survey of Living Conditions (JSLC) household expenditure data provide insight into

financial aspects of consumption but because detailed quality and volumes are not surveyed,

actual intake changes can be examined. For example the 2003 JSLC reveals that the lower

income quintile spent 16.9% of its household food budget on cereals as opposed to 10.3% spent

by the upper quintile. Average food expenditures as a percent of total consumption in 2003 are

documented at KMA6=35%, other towns =40%, and rural=45%. Between 1990 and 1992 the

average food expenditures were KMA=50%, other towns =55% and rural=60%. Yet, the

qualitative differences such a calorie/J$ or fiber content are hidden and real dietary distribution is


6 Kingston metropolitan area









masked. The decrease in relative amount spent on food suggests increased food accessibility, but

evidence suggests that the types of food purchased have been changing which might have

implications for the utilization component of food security.

In 1998 the Caribbean Food and Nutrition Institute (CFNI) revealed that 31.8% of the

Jamaican population possesses pre-obesity conditions with 19.7% already obese (1999). In a

1991 survey in Spanish town, Wilks found 30.7% of men and 64.7% of the women to be

overweight (1998). Between 1964 and 2000 the share of fat in the total DES increased from

19.9% to 25.9% (FAO, 2003). Carbohydrate percentages dropped from 69% to 62.8% over the

same period. This evidence agrees with the expected characteristic of a nutrition transition in

developing countries (Popkin, 2003). Likewise, the share of calorie availability from meat

products and vegetable oils has greatly increased since 1964 from 175 and 190 to 250 and 400

kcal/capita/day respectively. Prior to 1984 sweeteners did not compose a significant proportion

of imported food, in the 1996-1998 period imported sweeteners supplied approximately 18% of

DES.

The top five dishes in Jamaica are plain rice, fried chicken, boiled yellow yam, rice and

peas and boiled dumplings. In 1989 the contributions to calorie intake by food category were:

45% carbohydrates, 22% animal products, 14% fats and oils, 12% sugar, and 7% legumes. In a

Spanish Town study nutrients contribution to DES for males was 57.7% carbohydrates, 13.05%

protein, and 30.6% fats. For females it was 59.6% carbohydrates, 13.1& protein, and 30.6% fat

(FAO, 2003).

Poor households in many developing countries exercise similar strategies to cope with

resource shocks and maintain consumption (Handa and King, 2003). These include informal

credit markets and extended sharing networks (international remittances as well as rural to urban









links and neighbor to neighbor sharing). In times of strife, traditions of communities reaching

out to the least fortunate are varied. Some practice 'len han' or lending a hand to those in need

(ECLAC, 2004). Handa and King studied the different coping mechanisms between urban and

rural households through one of the most common food security indicators: child wasting.

Clarke and Howard, indicate increasing investment in Kingston's slum buildings as a factor that

has lessened differences in resilience between in rural and urban communities (2006).

Household consumption surveys are the major source of permanent income data for

Jamaica and many countries around the world. In developing countries these surveys often

suggest that food expenditures (purchased and home grown) account for over half of the

household's budget. The survey data also indicate a positive relationship between increased

income and calorie intake, but at a declining rate as income increases (Alderman, 1986). The

calorie-income elasticity for lower income groups was traditionally believed to be high, ranging

from .4-.8 indicating a high amount of responsiveness to lower prices or increased income

(Ibid.). These findings validate intuitive notions that when one has little of something, and if

more of that item becomes accessible, either through decreased price or increased income, one

will choose to acquire that item. This led policy makers to adopt poverty reduction strategies of

increasing income or regulating prices. However, those approaches failed to view the situation

with a wide enough lens. That view is based on the traditional and cursory definitions of food

security, that calories are the primary nutrient limiting improved nutrition in developing

countries rather than other deficiencies or surpluses that occur from intake of non-staple foods

(Levin et al., 1993; Behrman, 1995).

In a study of the reliability of survey data for predicting food expenditures in Kenya and

the Philippines, it was shown that the causal relationship between increased income and









increased calorie intake is much weaker with elasticity ranging from .2 to even less than .01

(Bouis and Haddad, 1992). They attributed the weak relationship to the nature of the surveying

methods which measured availability much more than actual intake. However, Bouis went

further to establish a behavioral explanation for the low responsiveness of calorie intake amongst

the poor (1996). One would expect a poor person to spend every bit of additional purchasing

power (due to price falls or income increases) on staples. However, this assumes away the

humanity of that person. Humans are not completely rational and will likely spend a "surprise"

windfall on pleasure rather than security. He attributes this to one's inherent humanity. As

Timmer has stated, an appropriate food policy model must approach demand behavior of lower

income people in terms of their resignation to a state of chronic need (2000). In other words, the

utility derived from a certain food over another, or the combined utility of all food consumed is

derived from a combination of energy, variety and taste. It appears irrational to purchase a more

expensive starch because it tastes better than the cheaper even though there might be a sacrifice

in energy intake but this behavior is common in households that have a brief gain in financial

resources (Ibid.).

This concept agrees with evidence from Indonesia that a marginal increase in a staple's

price will produce a shift to another staple or just an over all decline in consumption. If the

relative prices between nutritionally different but similarly taste valued products fall, a consumer

will spend any extra income on the item with greater taste value thus sacrificing nutrition. Once



7 Bouis also points to a possible phenomenon that could serve to dampen the apparent elasticity of higher income
groups while simultaneously over exaggerate the elasticity observed by the lower classes. This phenomenon is the
nature of cultural relations during times of strife. During these periods, rich households might actually be
purchasing more and transferring these amounts to the poor who are purchasing less. In this case actual calorie
intake would be maintained, but it would go unmeasured by traditional methods. This type of informal transfers and
sharing are likely common in Jamaica. This also agrees with a common phenomenon, that counter the capitalistic
model whereby in times of scarcity, price is actually cheapest as the good is ultimately shared freely rather than
profited upon.









a base level of calories is supplied, tastes play a more significant role. In general, in low income

households, the marginal utility that can be derived from energy trumps taste, or variety,

whereby at middle incomes, variety has the greatest value, and for the highest income brackets

taste is the highest priority when making decisions between foods.

This chapter provided a historical context for the Jamaican food system and described the

rich picture that was developed from soft systems methods in the field. This was followed by a

discussion of how the rich picture was used to reveal five conceptual categories that framed the

object oriented structure of QnD:Jamaica. These categories were then reviewed in terms of their

available empirical background in order to parameterize the model. The next chapter describes

how these categories are explicitly represented in QnD:Jamaica.









CHAPTER 5
JAMAICAN FOOD SYSTEM MODEL DESIGN AND TESTING

Approach

As the simulated food system involves transmission mechanisms that channel shocks

through the market, a number of economic models that aim to give insight into consumer

demand, producer supply response, and price transfers in developing countries were reviewed

(Robinson and Robinson, 1997; Alderman, 1986; Timmer, 1980, 2000; Bouis et al., 1992;

Skoufias, 2003; Handa and King, 2003; Rouijs et al., 2004). A general conclusion of critics of

the financial strategies pursued by Jamaica and other developing countries during the past three

decades has been the persistence of "black box" models to shape developmental policy

(Bernanke, 1995). These macroeconomic models often fail to explain why people participate in

foreign trade under adverse conditions. A. Walker, in response, presents microeconomic models

that claim to look "at the way in which the market is organized, the resulting interaction of

participants and the impact that they have on price movements" (2002, 4) in an effort to more

thoroughly explain production and consumption behavior in developing countries.

However, as one of the main goals of this type of modeling exercise is to bridge cross-

disciplinary and sector communication divides, the level of disciplinary jargon in most of the

reviewed models fails to meet QnD:Jamaica's needs. Additionally, neither a wholly macro- nor

micro- approach will fit this project's lens of discovery. Finally, there is limited agreement even

within the economic development community on the efficacy of these models (Berhrman, 1995;

Bouis, 1994; World Bank, 2003).

Empirical, theoretical and qualitative data informed the design of QnD:Jamaica's

simulation structures. In designing the processes that simulate the dynamic transfers of food

volumes and values, simple mathematical functions were employed to capture the unique









production and consumption mechanisms in Jamaica but also highlight generalizable features of

these mechanisms across developing countries. In this manner the QnD modeling system could

be applied to other national, regional or international socioecological situations/food systems.

Therefore the submodels' dynamics are limited to simplified and intuitive understandings of the

reviewed theoretical models in the interest of convenient collaborations with stakeholders from

varied expertise levels.

QnD:Jamaica is concerned with three flows: food, money and people. The data that

describe the representations of these flows is organized in an object oriented hierarchical

structure. Simulation occurs on a monthly timestep. The subprocesses of QnD:Jamaica are

presented below along with the associated structural assumptions that relate to the empirical

review presented in the previous chapter. Additionally, the simulation results of three

comparative index categories, productivity, pricing and consumption, are shown against

historical data to build confidence in QnD:Jamaica's simulation representations.

Population Typology Representation

The population characteristics as described in chapter four made it difficult to distill a rural

typology by only land ownership or income source. Therefore in QnD:Jamaica, each parish's

rural population is distinguished by both landownership and major income source. Capturing

this level of heterogeneity of the rural demographic landscape is vital for food security and other

developmental studies (Dyer et al., 2006). The rural population typology in QnD:Jamaica

consists often CLocalComponents shown in Table 5-1.











Table 5-1 Rural population typology.
Landowners NonLandowners
CFarmer CRuralAgWageEamer
CAgWageEamerWithLand CRuralTouristWorker
CRuralTouristWorkerWithLand CRuralOtherWorker
CRuralOtherWorkerWithLand CRuralUnemployed
CRuralUnemployedWithLand CRuralFisher

As the major income source distribution does not vary significantly across small farms of

0-5 hectares, it was assumed that all small farm land owning units in a parish own the same

amount of land (STATINJA, 1996).

The urban population typology is simpler than the rural because food security is often

closely linked to agricultural activities and land ownership. The urban population consists of

two types of wage earners: CUrbanTouristWorkers, CUrbanOtherWorkers. Also included in the

urban typology are urban fishers, and urban unemployed. Both rural and urban unemployed

have monthly earnings (from the informal sector) these are assumed to be less than the minimum

wage.

Urban tourist workers wage is calculated from a PRelationship process with number of

tourist per month acting as the independent variable in a negatively sloping linear relationship.

This independent variable value is represented as a timeseries driver in the simulation. However,

a hurricane occurrence will reduce this value. Urban other workers' wages are calculated by a

linear relationship driven by GDP values. The average weekly earnings for a general laborer

were J$895 and $J1600 for a tourist worker in 1992 (Planning Institute of Jamaica, 1992).

The understanding of interactions between wages and prices in Jamaica is somewhat

muddled (Craigwell, 1991). It is unclear whether a cost-push relationship exists whereby rising

wages drive price increases in an inflationary spiral (Bourne, 1977) or if prices are so









exogenously determined and labor unions so powerless that workers seek to "catch up" with

price inflation (Craigwell, 1991; Downes and McClean, 1982). Craigwell's study of Jamaica's

wages and export prices over the period of 1957-1984 revealed the possibility of a positive

feedback between wages and prices supporting the inflationary spiral theory (1991). However,

similar tests on Barbados and Trinindad and Tobago suggest only uni-directional flow from price

to wage (Ibid.). Additionally that study period was largely comprised of a peculiar time in

Jamaican history when governmental regulation of wages was high.

In Portland parish, rates as high as 79% of small farmers receive remittances from abroad

with 74% of recipients receiving cash on a monthly basis (Ishemo et al., 2006). Some 67% of

recipients received less than $J5000 (-$US83) per month and 90% less than $J10000 ($US 166)

per month (Ibid.). The ratio or remittances to total income increased throughout the 1990s

(World Bank, 2003). All population groups in QnD:Jamaica receive equal remittance amounts.

The total remittance value entering Jamaica is a driver inputted as a linearly increasing timeseries

file.

Land Use Representation

Land use in QnD:Jamaica is divided first by rural and urban uses (Figure 5-1). Rural land

is further distinguished as non-agricultural lands, estate lands: farms >100ha producing mainly

permanent export crops, and small farm lands: farms <5ha. Middle size farms between 5-100ha

are not included in this model version. Estate lands are further categorized as cultivated and

non-cultivated. Major estate crops include bananas, sugar cane and coffee, however, the model

groups these types into the one category described as export crops. While small farms are

described by the amount under cultivation in four main crop types: root crops, legumes,

vegetables and permanent crops, as well as by the amount that is uncultivated. These crop type

categories were chosen because they have environmental, economic, nutritional and cultural











significance as evidenced by their use in the Agricultural Census, the Jamaican Survey of Living


Conditions and the Economic and Social Survey of Jamaica.


Rural Land Urban Land


Estate Small Other
Lands Farms Land
>100ha <5ha Parks,

Cultivated Root crops

SUncultivate Legumes

Vegetables

Permanent
Crons
Uncultivate

Figure 5-1 Land use structure in QnD:Jamaica.

Despite the various accounts of Jamaican producers' price responsiveness described in the


previous chapter, for the short time periods simulated by this model (5-20 years), it is assumed


that land uses are not substitutable. That is, land does not switch between cultivation of export


crops, food crops, livestock and ruinate. This is in accord with producer response models used


for similar market types in other regions (Ruijs et al., 2004). Instead the amount of land in each


land use is driven by a timeseries file of trended known changes in area cultivated.8


The FAO and MOA estimate slightly different but similarly trended land areas under


cultivation in the crop types of concern for 1995-2004 (Figures 5-2 and 5-3).














8 A supply response function driven by price elasticity could be included in future scenario simulations whereby the
total amount of land in production as well as the spread of crop types (domestic v. traditional export, domestic v.
non-traditional export) varies. This could also include a degradation factor if land in ruinate is cultivated.












25000 Pulses,Total
Roots and Tubers,Total
20000 Vegetables Primary

15000
Ha
10000

5000

0
1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
Year

Figure 5-2 FAO cultivated area by crop group. (FAOSTATS)


16000
14000
12000
10000
Ha 8000
6000
E Legumes
4000 Vegetables
2000 --Yams
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Year

Figure 5-3 MOA cultivated area by crop group. (MOA, 2001)

Each of the three crop types exhibit distinctive trends. Legume area decreased steadily

throughout 1986-2001 with a slight increase in recent years. Root crop land exhibited a fairly

constant level until 1996, except for sharp decreases in 1988 and 1989 due to Hurricane Gilbert.

After 1996, root crop land declined sharply again for two years. It held constant in 1999 and

declined again in 2000. Since then it has been fairly steady. Vegetable area was fairly constant

until 1991 excluding a decrease in 1988. Like root crop land, vegetable area increased steadily

until 1996. Following a sharp decrease in 1997, area has been fairly constant except for a slight

decrease in 2000.









The 1996 Agricultural Census facilitated the distillation of national agricultural land areas

into proper divisions by crop type on small farms at the parish level. It was assumed that as

national areas changed the relative crop type distributions within parishes held constant over the

years (i.e. the small farms of St. Ann and Trelawny still produce the majority of the nation's

yams). These percentages are shown in Table 5-2.

Table 5-2 Percent of Jamaica's land uses by parish.
Percent of Percent of Percent of Percent of Percent of Percent of
all all all legume all all root all
cultivated permanent land in vegetable crop land uncultivated
land in crop land Jamaica land in in Jamaica land in
Jamaica in Jamaica Jamaica Jamaica
St. Andrew 4% 5% 5% 4% 2% 2%
St. Thomas 7% 8% 9% 11% 2% 3%
Portland 5% 7% 1% 2% 3% 4%
St. Mary 9% 13% 2% 7% 4% 5%
St. Ann 9% 4% 10% 19% 16% 13%
Trelawny 5% 3% 2% 2% 16% 4%
St. James 4% 5% 1% 2% 3% 5%
Hanover 5% 5% 3% 3% 7% 7%
Westmoreland 7% 7% 6% 5% 6% 11%
St. Elizabeth 9% 4% 34% 14% 5% 16%
Manchester 8% 4% 12% 11% 13% 13%
Clarendon 16% 20% 10% 11% 14% 10%
St. Catherine 12% 15% 6% 8% 9% 7%

It was originally thought that these distribution percentages could be multiplied by the

national total land areas reported by the FAO and MOA to simulate proper parish level

distributions while also accounting for the annual changes. However, the areas reported by the

FAO and MOA needed to be modified before this distribution could be calculated. The land area

data reported by the FAO and the MOA, lowered production in 1988, 1997, 2000, and 2004

appears to be a function of the amount of area harvested-not planted. In fact, each of these years

correlates to a major environmental event that decreased production (Hurricane Gilbert, Tropical

Storm Marco, drought and Hurricane Ivan respectively). Thus, while productivity appears










constant in these years, it, in fact, decreased when the amount of actually harvested land was

different than the amount originally planted. Thus, it is unknown how much of the decreased

production is due to producers' land use choices prior to planting versus uncontrollable

environmental factors post planting. This difference is critical for this study because of the

interest in hurricane and flooding damage estimation.

In order to overcome what is believed to be an underestimation of land put into

cultivation at the beginning of the season, trend lines were estimated for each crop type's

national area.9 Using this trend line as a driver represents producers' decisions but controls for

the highly unknown feedbacks that drive those decisions. The land area values under contention

were removed, and the trends for apparent periods were estimated separately (Figure 5-4).




25000 ---VEG
ROOTS
20000 LEGS

HA 15000
10000

5000

0
1985 1988 1993 1997 2000 2003
YEAR

Figure 5-4 Estimated land areas that act as drivers in QnD:Jamaica.

Land use changes on the large plantations occupying the majority of Jamaica's agricultural

land would have large economic implications in terms of wage earnings (or small farm earnings

if land were distributed as such) (Weiss 1999, 2004b). Changes in these land uses would also

have vast effects on the ecology of the island. Like small farms, cultivated land area on estates


9 In the future, rather than a trend line, land use changes will be tied to market demand and price so that the
dynamics could be internalized to the model.









simulated in QnD:Jamaica change through a driver file. This driver file was estimated using

known decreases in sugar and banana land areas on large plantations.

Productivity Representation

The crop yield subprocesses in QnD:Jamaica are based on three PRelationship processes

with the simulated month and rainfall amount serving as the driving independent variables. The

following equation describes domestic crops' monthly yield (tons/ha) calculation and includes

maximum yield, productivity, and flood damage coefficients in described in detail below.

Yield=Max monthly yield coefficient X Productivity coefficient X Flood damage factor

Each month is related to a maximum productivity due to seasonal variability. Coefficients

for each month's maximum yield (tons/ha) were calculated from known typical annual yields for

each crop type. Generally maximum productivity occurs in the early spring months (Table 5-3).

Table 5-3 Maximum monthly yield coefficients (tons/ha)
Month Vegetables Root crops Legumes
J 0.9 1.5 0.15
F 1.0 1.45 0.15
M 1.0 1.45 0.15
A 0.9 1.4 0.15
M 0.9 1.4 0.15
J 0.9 1.35 0.15
J 0.9 1.2 0.15
A 0.9 1.2 0.15
S 0.9 1.25 0.15
O 0.9 1.35 0.15
N 0.9 1.45 0.15
D 0.9 1.5 0.15
Annual Yield 11 16.5 1.8

The Jamaican Meteorological Service possesses 30 year average parish level rainfall

records as well as monthly parish level records for 1992-2002. This range of records served to

define the time scale of QnD:Jamaica's base scenario. The crop productivity and flood effect

generator was calibrated using known rainfall amounts from 1992-1997 including historical









drought and flood events, crop yields and welfare effects. Initially the crop productivity

coefficient was calculated with a relative rainfall index based on a month's real rainfall amount

(mm) divided by that month's thirty year average amount (mm). However, because Jamaica's

weather has distinct dry and wet seasons, this method failed to capture decreased productivity

caused by abnormal rainfall in previous months of the growing season.

The use of a relative rainfall index based on a four month running rainfall average (as

opposed to one) divided by the current month's thirty year rainfall average to determine

productivity coefficients proved more appropriate. These productivity coefficients are shown in

Table 5-4.

Table 5-4 Productivity coefficients
Vegetables Roots Legumes
Relative Productivity Relative Productivity Relative Productivity
rainfall coefficient rainfall coefficient rainfall coefficient
index index index
<0.5 0.75 <0.5 0.25 <0.5 0.75
0.6 0.9 1.25 1.0 0.6 0.9
0.699999 0.95 1.5 1.5 0.699999 0.95
0.7 1.0 1.75 1.0 0.7 1.0
0.85 1.0 1.75 0.25 0.85 1.0
1.0 1.1 1.0 1.1
1.39999 1.2 1.39999 1.2
1.4 1.5 1.4 1.5
1.74999 1.5 1.74999 1.5
1.75 0.8 1.75 0.8
1.89999 0.8 1.89999 0.8
1.9 1.0 1.9 1.0
10.0 1.0 10.0 1.0


Figure 5-5 graphically depicts vegetables' productivity coefficient PRelationship process

shown in Table 5-4 above.











1.6
1.4
. 1.2
S1
S0.8
S0.6
0.4
0- 0.2
0


0 0.5 1 1.5 2 2.5 3 3.5
Four Month Relative Rainfall Index

Figure 5-5 Vegetables productivity coefficient relationship

The four month running relative rainfall subprocess captures productivity decreases due to

drought. However, in order to simulate sudden shocks such as a massive flood or hurricane, a

separate subprocess also tracks one month relative rainfall. Floods in Jamaica are common

because of high runoff coefficients across the island, excluding the porous Cockpit country

(Nkemdirim, 1979). High runoff levels are principally correlated with: Annual rainfall amounts

and seasonality and to a lesser extent slope, drainage density and elevation (Ibid). In mapping

flooding hazards, Jamaica's Office of Disaster Preparedness and Emergency Management first

mapped landslide risk with three geological and slope characteristics and then associated high

flood risk with these areas that were less than 40 m high (CDERA, 2003). Presently,

QnD:Jamaica simulates floods based only on rainfall amounts.

If the one month relative rainfall is greater than 3.25, indicating that the month's rainfall

was 325% above normal, crop yields are reduced by a flood damage factor of 20%. As

mentioned above, topography plays a strong role in flood occurrences. However, in

QnD:Jamaica the flood damage trigger (a relative rainfall greater than 3.25) is ubiquitous across

all parishes. In reality this trigger would be lower in places prone to flood (low lying areas, poor

drainage etc.) and higher in landscapes that can withstand heavier rainfall. This uncaptured









variation is likely one reason for the following calibration issue. 1993 and 1996 experienced

similarly high monthly rainfall events, yet greater productivity damage was reported in1993. In

order to maintain high levels of productivity during 1996 (the year of highest recorded

productivities) the flood trigger was set above both 1993 and 1996 monthly relative rainfall.

Productivity Calibration

The food production sub models were calibrated using data from the six year period

between 1992 and 1997. The calibration period was chosen because it contained data for

drought, floods and normal rains. Monthly parish level rainfall data was only available for the

years between 1992 and 2002.

Vegetables

Vegetable productivity between 1992 and 1996 experienced a gradual increase from 9.49 to

12.50 tons/ha. In 1997 there was a sharp decline due to a major drought. The productivity in

1995 was only slightly greater than 1994 but 1996 experienced a large increase due to ideal

rainfall conditions. There were two reported tropical storms during this period, Gordan (1994)

and Marco (1996). Additionally, floods from heavy rainfall were reported during 1993.

Only annual historical productivity values were available. Thus, only inter annual

productivity could be calibrated. The model's monthly outputs were summed and divided by the

respective year's cultivated land area as simulated.

1992 and 1993 produce unrealistically high productivity levels. This is possibly due to

increased maximum productivities throughout the time period from other advances made in

production techniques such as fertilizers and pest controls. In order to account for the changes in

maximum attainable productivity, additional parameters would need to be included in the model.

Unrealistically high production levels in 1992 could also be due to lingering drought effects from

1991. Additionally, to reiterate 1993 experienced heavy floods which were not captured in the









flood simulator so that the damage from1996 floods could also go unregistered. Productivity in

1994 and 1997 was slightly over estimated, while, productivity in 1996 was slightly

underestimated. Simulated production volumes reflect similar levels of accuracy (Figures 5-6, 5-

7, 5-8 and 5-9).


13

12
0
S11

10
-o 9 Known Productivity
n o Simulated Productivity
8i i i i
1992 1993 1994 1995 1996 1997
Year
Figure 5-6 Vegetables productivity calibration by date.

13

12 *
0l0
U- 11



0 1 9
9 10 11 12 13
Known Productivity (tons/ha)
Figure 5-7 Vegetables productivity calibration difference from known.











2.0E+08
1.9E+08
-" 1.8E+08
1.7E+08
ID 1.6E+08
z 1.5E+08
> 1.4E+08
1.3E+08 / Known Volume
1.2E+08 -- Simulated Volume
1992 1993 1994 1995 1996 1997
Year

Figure 5-8 Vegetables volume calibration by date.

2.OE+08
c 1.9E+08
w 1.8E+08
E 1.7E+08
0 1.6E+08
-0 1.5E+08 -
o 1.4E+08
E 1.3E+08
CO 1.2E+08
1.2E+08 1.4E+08 1.6E+08 1.8E+08 2.0E+08

Real Volume (kgs)

Figure 5-9 Vegetables volume calibration difference from known.

Root crops

Root crops productivity could not be calibrated to as high a level of accuracy as

vegetables'. The productivities in 1996 and 1997 were underestimated (Figures 5-10 and 5-11).

Despite imprecision in productivity simulation, the root crop volumes reflect a reasonable degree

of accuracy (Figures 5-12 and 5-13).


















Known Productivity
-- Simulated Productivity


1992 1993 1994 1995 1996 1997
Year


Figure 5-10 Root crops productivity calibration by date.


19

18
-0
S 17

S16

S 15
E
o5
14


Figure 5-11


3.9E+08
3.7E+08
3.5E+08
3.3E+08
3.1E+08
2.9E+08
2.7E+08
2.5E+08


14 16 18
Known Productivity (tons/ha)

Root crops productivity calibration difference from known.


Known Volume
Simulated Volume

1992 1993 1994 1995 1996 1997
Year


Figure 5-12 Root crops volume calibration by date.










) 3.9E+08
3.7E+08
E 3.5E+08
S3.3E+08
3.1E+08
S2.9E+08
D 2.7E+08
&5 2.5E+08
2.5E+08 3.0E+08 3.5E+08 4.0E+08
Known Volume (Kgs)
Figure 5-13 Root crops volume calibration difference from known.

Legumes

Legumes productivity values experienced the highest degree of error. Values in 1992-

1994 are greatly over estimated, while values in 1996, and to a greater extent 1997, are

underestimated. However, the inter-annual changes do show reasonably similar patterns as

known changes. Despite productivity in 1997 being grossly underestimated, simulated volume

reflects greater precision. (Figures 5-14, 5-15, 5-16, and 5-17).



2

1.9

-I 1.8

0 1.7
1.6 Known Productivity
--Simulated Productivity
1.5 .
1992 1993 1994 1995 1996
Year
Figure 5-14 Legumes productivity calibration by date.












o 2
n 1.9

0 0
tS >1.7
S 1.6
5c 1.5
1.5 1.7 1.9 2.1
Known Productivity (kgs/ha)
Figure 5-15 Legumes productivity calibration difference from known.


1.4E+07
1.2E+07
1.0E+07
8.0E+06
6.0E+06
4.0E+06
2.0E+06
0.0E+00


1992 1993 1994 1995 1996 1997
Year
Figure 5-16 Legumes volume calibration by date.


1.4E+07

U 1.3E+07 -
0)
1.2E+07

| 1.1E+07

E 9.5E+06
8.5E+06
7.5E+06 *
7.5E+06 9.5E+06 1.2E+07 1.4E+07
Known Kgs
Figure 5-17 Legumes volume calibration difference from known.


-- Known Volume
- Simulated Volume


-









Livestock

Although small farmers produce a fair amount of the national livestock, QnD:Jamaica

simulates livestock production as a large scale industrialized process. This is a fair

representation of the poultry industry in particular. As the vast majority of animal feed is

imported, the border price of feedstuffs directly influences domestic livestock production

quantities and can be used as a model driver. The strongest relationship between feedstuff price

and production volumes was found between real Jamaican dollar values and total volumes. In

order to calculate values in real Jamaican dollars, the external feedstuff price is multiplied by the

exchange rate and divided by the Consumer Price Index, CPI, both of which are also included are

timeseries driver files.

The real J$ value serves as the independent variable in a PRelationship process to generate

livestock production volumes (Figure 5-18). Livestock production was calibrated within 10% of

real known volumes (Figures 5-19 and 5-20).


-- 1.00E+07
0) Simulated Relationship
S8.E+06 Known Values
0 8.00E+06

6.00E+06

o E+06
o 4.00E+06


2.E+06
"J 2.00E+06


0 1000 2000 3000
Feedstuff Price (Real J$/ton)
Figure 5-18 Livestock production relationship.


___ __










70000
-, 60000 _
c 50000
S40000
E 30000
Known Production
o 20000 -- Simulated Production
10000
0
1992 1993 1994 1995 1996 1997
Year
Figure 5-19 Livestock production calibration by date.


80000
C
0
S 70000

S, 60000
a_ c:
50000

.E 40000
30000
30000 40000 50000 60000 70000
Known Production (tons)
Figure 5-20 Livestock production calibration difference from known.

Testing Productivity Simulations

Vegetables productivity

As in the simulation of vegetable production in 1996, the model was not sensitive enough

to produce high levels of productivity in 2001. Productivity was also underestimated in 2000.

However, this was likely due to the model being overly sensitive to drought occurrences during

that year. In the years 1998 and 1999, poor and good years of productivity respectively, were

simulated near historical values. Volume levels reflected similar trends as productivity

indicating no strong errors in land area estimation. The inter-annual shifts are also similar to

known trends (Figures 5-21, 5-22, 5-23, and 5-24).










13
_1 12

cc 11 5 7
> 10
Known Productivity
D3 Simulated Productivity
2 8
1998 1999 2000 2001 2002
Year
Figure 5-21 Vegetables productivity testing by date.


13
12.5
12 *
2 11.5
0- 2 11
g 10.5
10
10
E 9.5
c 9

9 10 11 12 13
Known Productivity (tons/ha)
Figure 5-22 Vegetables productivity testing difference from known.

2.E+08 Known Volume
-Simulated Volume
2.E+08
0)
2.E+08

o
E 1.E+08

> 1.E+08
1.E+08
1998 1999 2000 2001 2002
Year
Figure 5-23 Vegetables volume testing by date.










) 1.7E+08
-Z
(D 1.6E+08
E
2 1.5E+08
S1.4E+08
1.3E+08

n 1.2E+08
1.2E+08 1.4E+08 1.6E+08
Real Volume (kgs)
Figure 5-24 Vegetables volume testing difference from known.

Root crops productivity

Root crops productivity was tested with the greatest degree of accuracy out of the crop

types. The year 2000 was the only year where productivity was greatly over estimated. Root

crop production levels lend themselves more readily to course productivity simulation. This

could be due to the nature of root crop production or to the fact that there is less variation within

the root crop category as opposed to vegetables which contain a wider variety of species and

cultivation techniques (Figures 5-25, 5-26, 5-27, and 5-28).

17

16
g 15

> 14
:| 14 ---- V--
z- 13 -- Known Productivity
o Simulated Productivity
S12
1998 1999 2000 2001 2002
Year
Figure 5-25 Root crops productivity testing by date.











19
18
o -M 17
1a 16
I15
14
35 13
13 15 17
Known Productivity (tons/ha)

Figure 5-26 Root crops productivity testing difference from known.


2.9E+08
2.7E+08
0) 2.5E+08
2.3E+08
E 2.1E+08
5 1.9E+08 -- Known Volume
1.7E+08 Simulated Volume
1.5E+08
1998 1999 2000 2001 2002
Year

Figure 5-27 Root crops volume testing by date.

3.2E+08
0) 3.0E+08
D 2.8E+08
E
2.6E+08
0
> 2.4E+08
| 2.2E+08
E 2.0E+08
S 2E+08 2.2E+08 2.4E+08 2.6E+08 2.8E+08 3E+08

Known Volume (Kgs)

Figure 5-28 Root crops volume testing difference from known.










Legumes productivity

The legumes productivity simulator generated reasonably accurate inter-annual dynamics.

Productivity in 2001 was underestimated and in 2002, overestimated. The increase in error

between 2002 productivity calculation and volume calculation indicates error in the amount of

estimated land in legume cultivation (Figures 5-29, 5-30, and 5-31).


3
2.5
S2
.-c
1.5
0
I- 1
0.5
0


1998 1999 2000 2001 2(
Year

Figure 5-29 Legumes productivity testing by date.


002


N 2.5

2
|io 2 -- 7--



0
I 1.5

E
y) 1 ,
1 1.5 2 2.5 3
Known Productivity (tons/ha)
Figure 5-30 Legumes productivity testing difference from known.


-- Known Productivity
SSimulated Productivity


I










1.0E+07

-C8.0E+06 o

S6.OE+06

z 4.0E+06
o Known Volume
2.0E+06 --Simulated Volume

0.OE+00 1
1998 1999 2000 2001 2002
Year
Figure 5-31 Legumes volume testing by date.

Livestock production

The livestock production process greatly underestimated production volume for 2000-

2002. This is due to historically high production despite high feed prices (Figures 5-32 and 5-

33).

90000 -- Known Production
E 85000 --Simulated Production
o _80000
0 75000
D 70000
0
65000
60000
1998 1999 2000 2001 2002
Year
Figure 5-32 Livestock production testing by date.










c 85000
80000 *

". 75000

: 70000

65000 *
E
60000
60000 70000 80000
Known Production (tons)
Figure 5-33 Livestock production testing difference from known.


Food Importation

Ideally the amount of food imported would be a result of feed back loops. The driving

influences could be the amount of imported product purchased in previous time periods, either

the immediately prior period or a running average of demand. Or, the imported food amounts

could be driven by the supply present in the current time period. Each method would generate

different degrees of responsiveness.

In this version of the model, however, imported food is represented as an unlimited supply.

The amount actually purchased is limited by income restraints and imported food's prices

relative to substitutable domestic foods (described in greater detail in 'Consumption' section

below). In fact, the notion of inexhaustible imported food is fairly representative of the historical

simulated period. It will be most important to develop feedback loops for additional scenario

simulations beyond this historical period.

Food Distribution and Marketing Representation

The Jamaican food distribution and marketing landscape is understood as a single

national system due to the island's small size and the presence of an extensive road network that

makes it possible to visit every parish in a single day. However despite these features, Jamaica









experiences moderate spatial and temporal price variation as a consequence of parish level

supply, demand and infrastructural differences. Temporal shifts result from domestic seasonality

and other climatic effects as well as external trade conditions. Spatial price variation is a

function of transport and marketing infrastructure differences such as a parish's distance to ports,

road conditions and rural population density. The expanding modern distribution and retail

sector intertwined with the traditional higgler system present unique challenges to managing

Jamaica's food distribution network. The following section discusses how the qualitative

systemic structures described in chapter four are represented as specific sub models to simulate

prices in QnD:Jamaica.

Representing National Farmgate Prices

Simple linear relationships were estimated using past known production volumes and their

associated real farmgate prices within the study period to simulate the mechanisms acting on

national real prices. The reviewed evidence in chapter four suggests that parish level prices

should be increased above this base national level by factors representing higglers' transportation

and marketing costs. The design logic of QnD:Jamaica's pricing processes is discussed below.

Crop pricing

It was assumed that the volumes reported by the MOA and FAO were the total production

volumes rather than the actual volumes at market because the volume, productivity, and the

amount of land in production agree.

Vegetable prices were simulated with a different process than root crops and legumes for

the sake of exploration. While the real prices of vegetables were generated from a PRelationship

process with production volume as the independent variable, the real prices of root crops and

legumes were generated through a multiple step process in order to capture price elasticity. In

those price generation processes, the percent change in the current month's production volume










from the running twelve year average production volume served as the independent variable in a

PRelationship process. These relationships are shown for the three main domestic crops

modeled in QnD:Jamaica (Figures 5-34, 5-35, and 5-36). Nominal prices were calculated by

multiplying the real price values by the CPI driver amount.


- 4.5
. Known Values
._ 4 Simulation Relationship
() --


3
5
5- ---" ------



2-
5
9000000 11000000 13000000 15000000 170
Harvest Volume (Kgs)

34 National vegetables farmgate price relati


4.
4
3.


y = -4.4017x + 2.8745 2.
R2 = 0.8326 2

1.


00000


onship.


-0.3 -0.2 -0.1 0 0.1
Change in Harvest Volume (%)

Figure 5-35 National root crops farmgate price relationship.


3



E
Lu 1.




Figure 5-






co
a


0
a_


,.a










b.
-_ 6

5.5
---


---4.5
Sy=-1.5948x+ 5.1433 4
2 R2= 0.2652 3.5
3
-0.4 -0.2 0 0.2 0.4
Change in Harvest Volume (%)
Figure 5-36 National legumes' farmgate price relationship.

Each price relationship function is bound by minimum and maximum values established

just outside of the known historical values. The ranges are limited in real J$/kg to 2.5 4.2 for

root crops, 1.75-3.3 for vegetables, and 4.4-6.3 for legumes.

Crop pricing calibration

Simulated vegetable prices in 1993 and 1995 were consistently lower than known values.

As vegetable prices are calculated from a direct relationship with harvest volume, price error is a

result of error in the production simulator. Figures 5-37 and 5-38.


3.4
S3.2
6c 3
2.8
S2.6
( 2.4
Lc) -- Known
0 2.2 -*-Simulated
2
1992 1993 1994 1995 1996 1997
Year
Figure 5-37 Vegetable price calibration by date.










3.4
3.2
3
2.8
S2.6
2.4
2.2
2
1.5 2 2.5 3 3.5
Known (J$/Kg)


Figure 5-38 Vegetable price calibration difference from known.

Root crops' price simulation experienced a greater degree of error than vegetables price.

Using the current subprocess design, high price levels could not be generated without over

estimating prices in 1996 and 1997 (Figures 5-39 and 5-40). Simulated inter annual legumes

prices were relatively stable (Figures 5-41 and 5-42).


4.5
Known
4
Simulated
2 3.5

3

c 2.5
a_
2
1992 1993 1994 1995 1996 1997
Year
Figure 5-39 Root crops price calibration by date.






















21
2 3 4 5
Known (Real J$/kg)

Figure 5-40 Root crops price calibration difference from known.


0)

, 6
-5
()
04
(D
)c 3
n_


1992 1993 1994 1995 1996 1997
Year

Figure 5-41 Legumes price calibration by date.


40

S30

E 20
0
S- Known
10
0 --Simulated

S0
1992 1993 1994 1995 1996 1997
Time

Figure 5-42 Legumes price calibration difference from known.






100










Livestock pricing

It is easy to distinguish external feedstuff prices as a clear driver for domestic livestock

prices. After production volumes are simulated using external feedstuff prices as the

independent driver, domestic poultry prices are calculated using a PRelationship process similar

to the one employed for vegetables' prices that relates gross volumes to real prices (Figure 5-43).

Livestock prices are also bound by limits just beyond the historical range: 4.6-7.3 real J$.


8
3 Known Values
S 7 *
--Simulated
.o -0 g6 Relationship


0
M 4
z
3
3.0E+06 4.0E+06 5.0E+06 6.0E+06 7.0E+06 8.0E+06
Production Volume (Kgs)

Figure 5-43 National livestock price relationship.

Livestock prices were calibrated within adequate levels for all years except 1992 and 1993.

In those first years, prices were much higher than any for the rest of the period, but production

levels did not reflect the expected decrease (Figures 5-44 and 5-45).


*o 7.5
a 7 -- Known
8o 6.5 ---Simulated
Cn tf 6
(D 4.5
>-
-F u 5 .5 i i
-Fu 0 5 ------------
0 4.5
z
S4 --------------

1992 1993 1994 1995 1996 1997
Year

Figure 5-44 Livestock price calibration by date.



















4 5 6 7 8
Known Price (Real J$)
Figure 5-45 Livestock price calibration difference from known.

Price Testing

Average annual vegetable prices were consistently over estimated in the testing period

simulations. However, inter-annual simulated dynamics are nearly similar to known changes

(Figures 5-46 and 5-47).


Figure 5-46


1998 1999 2000 2
Year
Vegetable price testing by date.


001 2002


- Known
-i- Simulated










3.6
5 3.4
3.2
3

_o 2.6 Z
E 2.4
05 2.2
2
1.5 2 2.5 3 3.5
Known (J$/Kg)
Figure 5-47 Vegetable price testing difference from known.

Figure 5-48 shows the monthly change in simulated nominal prices against the change in

the known vegetable consumer price index. Simulated price changes were within reasonable

levels; however, they did not display significant similarity to known prices changes. There is

some evidence that the monthly price subprocess is moving closer to simulating reasonable

prices. For example, QnD:Jamaica captures the large price jump in early 1993 even though it is

simulated earlier than it occurred in reality. Also the simulation of the known large decline

towards the end of 1993 followed by a degree of stability throughout early 1994 builds

confidence in the model. Finally in late 1994, the simulated and known inter-monthly dynamics

are similar.


20
15
c 10
5
0 0
C: -5
S-10
(D -15
-2 5 --Known Change Simulated Change
-20 --
-25

y Y 0o' Y Y '
Time









Figure 5-48 Change in known vegetables price index vs. change in simulated nominal
vegetables price.

The testing simulation of root crops' prices reflects the model's inability to generate

dynamic inter-annual changes for this crop which was revealed during calibration (Figures 5-49

and 5-50). Figure 5-51, however, shows that the model is generating dynamic intra-annual

prices in accordance with monthly volume changes.


1998 1999 2000
Year
Figure 5-49 Root crops price testing by date.


2001 2002


2 2.5 3 3.5
Known
Figure 5-50 Root crops price testing difference from known.











3.50E+07 4.5

3.00E+07 2 -

2.50E+07 3. N


0
I 2.00E+07 -- Y V 1 X25 -

> 1.50E+07 -- 2
1.5
I 1.00E+075
-c- Volume 1 1
5.00E+06 -- Price 0.5

O.OOE+00 0
Jan-92 Jan-94 Jan-96 Jan-98 Jan-00 Jan-02
Time

Figure 5-51 Root crops simulated volume and real prices.

Inter-annual legumes prices lack strong inter-annual dynamics as expected from the

calibration period simulation (Figure 5-52). Livestock price simulation is generally similar to

historic amounts (Figures 5-53 and 5-54).


5.6
,, 5.4
68 5.2
5
Q 4.8
( 4.6 --- Known
n 4.4 --Simulated
4.2
1998 1999 2000 2001 2002
Year
Figure 5-52 Legumes price testing by date.











5.5



4.5

wD 4
.-0
a- 3.5 Known
--Simulated
3
1998 1999 2000 2001 2002
Year

Figure 5-53 Livestock price testing by date.

6 6
0)




c i
4 -4.5


4 4.5 5 5.5
Known Price (Real J$)

Figure 5-54 Livestock price testing difference from known.

Representing Producer-Higgler Food Transfer Volumes

All land owning population groups harvest their entire available crop of each food type

(Root crops, Vegetables, Legumes and Permanent Crops). The total amount, DHarvestedCrop is

added to their DCropStore by the process PMonthlyHarvest. Next the farmer will transfer a

certain amount of this store to the DNationalCrogQuantity. The process simulating the transfer

of food into the market is guided by a producer population's production level relative to both

other producer population's production levels across Jamaica and to the producer population's

household consumption demands for the crop under question.










The features that establish the producer populations' food selling relationship process are

based on the per capital monthly production amounts within each population group of each

parish. The key points of interest are: 1) the average per capital production amount in the parish

population with the least production amount, 2) the average national production amount, 3) the

midpoint between the average smallest and national average, 4) the average production amount

from the population group producing the most per capital, and 5) the largest production amount

overall on a per capital basis (Figure 5-55).


90
S75

3 55 a r Absolute Largest
'00 50 _Avg Largest

S35 /


10 -

Avg Smallest
-Production Amount (kgs/capita/month)

Figure 5-55 Percent of production sold.

It is assumed that producer populations' with the smallest production levels would only

sell 10% of their crop because they are producing much less than the consumption demands for

that crop. That is to say, they are producing so little it is assumed to be largely for subsistence

rather than market. The average national production amount was equated with a 50% sell rate.

Along the same lines, those producers growing the largest amount would on average sell 75% of

their crop, but never more than 90%.

QnD:Jamaica's crop production simulation outputs were used to determine the values for

each of the five points of interest above. The amounts for each crop are shown in Table 5-5.









Table 5-5 Crop sell amount factors (kg/capita/month).
Average smallest Midpoint Average Average Absolute
(sell 10%) (sell 35%) national largest largest
(sell 50%) (sell 75%) (sell 90%)
Root crops 9.0 19.0 28.0 90.0 132.0
Vegetables 6.0 10.5 15.0 27.0 40.0
Legumes 0.18 0.54 0.9 2.19 3.6

Representing Consumer Prices

As discussed in chapter four, the Jamaican food distribution infrastructure often isolates

rural communities. This is true not only in terms of collection of their produce, but also in terms

of making imported food available to those populations. Thus, just as prices increase for

domestic foods as distance from farm increases, so do imported food prices increase with

distance from ports and urban centers. Transportation and marketing costs of a given

population's food are captured in QnD:Jamaica through index factors that increase food prices

above their base national price. Every parish has associated marketing and transport index

factors for its rural and urban areas.

The base national price for domestic foods is the nominal farmgate price. The base

national price for imported foods is set by a driver file containing external nominal prices in

US$/kg. These prices are then multiplied by the exchange rate in that time period to establish the

national base price in nominal J$. Tariffs on imported foods have been largely dismantled as

required by WTO negotiations. However, higher customs duties are still employed for sixteen

fresh vegetables, and certain processed fruits, meats and sugar products reflecting their domestic

sensitivity. Under CARICOM agreements, agricultural imports from member countries face

zero tariffs. It is assumed in QnD:Jamaica that imported vegetables and poultry are the only

foods that experience a border duty. These tariffs are set at 200% and 86% respectively in the

base scenario.









Marketing costs

As places with lower demand due to population or income have smaller and less

diversified markets, it is safe to assume certain features of parish level marketing costs based on

the surrounding demographic features (Norton and Symanski, 1975). The costs associated with

rural marketing are based on rural population density while urban marketing costs are based on

the existence and size of a central parish market. Likewise, domestic and imported foods are

assumed to have different marketing costs because of the differing characteristics of their

distribution systems. The DMarketingCostFactor for imported foods in rural areas is based on a

linear relationship established by assuming a 50% price markup in areas with the highest rural

population density (Saint Andrew/Kingston) and a 200% price markup in areas with the lowest

rural population density (Trelawny). Urban imported foods' DMarketingCostFactors were

estimated by associating the largest market, Coronation Market (easily accessed by both Saint

Catherine and Saint Andrew/Kingston parishes) with a 50% price markup and other parishes

containing large markets with 60% price markups. While imported foods' prices in parishes

without large urban markets have 75% marketing mark ups. Domestic food in rural areas does

not incur a marketing cost and in urban areas it experiences only 1/8 of that area's imported

foods marketing cost. The DMarketingCostFactors are shown in Table 5-6.











Table 5-6 Marketing cost factors.
Parish Rural Major market
population
density
(capita/km2)


Rural
Domestic Imported
foods foods


Urban
Domestic Imported
foods foods


St. Andrew 246.5 Coronation 1.0 1.5 1.06 1.5
St. Thomas 91.7 1.0 1.9 1.09 1.75
Portland 77.6 Port Antonio 1.0 1.9 1.075 1.6
St. Mary 150.2 Ocho Rios 1.0 1.8 1.075 1.6
St. Ann 103.7 Browns Town 1.0 1.9 1.075 1.6
Trelawny 68.1 1.0 2.0 1.09 1.75
St. James 146.1 Montego Bay 1.0 1.8 1.06 1.5
Hanover 138.5 1.0 1.8 1.09 1.75
Westmoreland 132.3 Negril 1.0 1.8 1.075 1.6
St. Elizabeth 104.4 Black River 1.0 1.9 1.08 1.65
Manchester 155.5 Mandeville 1.0 1.8 1.075 1.6
Clarendon 215.5 May Pen 1.0 1.6 1.075 1.6
St. Catherine 115.0 Coronation 1.0 1.9 1.06 1.5

Transport costs

The model developed by Ruijs et al. to examine food transport costs across different

regions of Burkina Faso guided QnD:Jamaica's food transport pricing structure (2004). That

model estimated graduated per ton per kilometer increases over main highway transport costs of

75% for less busy paved roads, 150% for unpaved roads, and 200% for dirt roads. During the

rainy season a further 20% and 60% increase was levied on unpaved and dirt road transport costs

respectively. In QnD:Jamaica, estimated food transport distances were established for each

parish. Domestically produced foods were assumed to be transported 15 km for both rural and

urban markets in all parishes. The distance that imported foods are transported varies by parish

according to the average number of kilometers from its closest importing port (Kingston or

Montego Bay). Each distance was then multiplied by the fuel cost per kilometer in that time

period which is a function of world oil prices. The resulting fuel cost along with the marketing









cost associated with each food type in each geographic region is then added to the national

farmgate price.

In QnD:Jamaica the quality of road was quantified for each parish by calculating the

proportion of Class C road length to total road length. 10 This factor,

DTransportInfrastructurelndex, was set to 1.0 for all urban areas as well as the rural Saint

Andrew/Kingston area which has the lowest percentage of Class C roads at 52%. A linear

relationship was established by associating the parish with the lowest proportion with an index of

1.0 and the highest with 1.1. This index represents a range of zero to ten percent increase in food

prices already containing marketing and fuel costs due only to the quality of road in a given

parish (Table 5-7). The base index is increased by five to one hundred percent when road quality

is temporarily reduced by hurricanes or floods depending on the strength of the natural hazard.

Table 5-7 Transport infrastructure index.
DTransportInfrastructurelndex
Parish Proportion of Average distance to Rural Urban
Class C to total import port
roads
St. Andrew 0.52 0 1 1.0
St. Thomas 0.82 50 1.1 1.0
Portland 0.61 98 1.03 1.0
St. Mary 0.58 35 1.02 1.0
St. Ann 0.53 60 (Either Port) 1 1.0
Trelawny 0.61 40 (Montego Bay) 1.03 1.0
St. James 0.83 0 (Montego Bay) 1.1 1.0
Hanover 0.60 30 (Montego Bay) 1.03 1.0
Westmoreland 0.64 45 1.04 1.0
St. Elizabeth 0.79 80 1.09 1.0
Manchester 0.72 98 1.07 1.0
Clarendon 0.61 74 1.03 1.0
St. Catherine 0.75 15 1.07 1.0





10 The Ministry of Transport characterizes 80% of Class C roads as in poor condition. Rural areas are mostly
comprised of Class C roads.









The DRoadFloodDamage index is determined with a relationship function that uses the

DMonthlyRainfallRatio as the independent variable. The relative rainfall in a given parish is the

ratio of that month's total rainfall to the thirty year average rainfall in that month. A ratio of less

than 1.3 (30% above average rainfall) does not incur a damage factor. A ratio of 1.3 sets the

road damage index at 1.05. The index increases linearly until the relative rainfall is 3.0 (300%

above average rainfall, indicating a major flood) at which point the damage index is 1.1. If the

relative rainfall is 4.0 the associated damage index is 1.5. The flood damage index is multiplied

by the infrastructure index. In this way floods in areas where the roads are already of poor

quality (as represented by a high transport infrastructure index) compound the already high

transportation costs.

Jamaican Consumer Representation

Both economic and nutritional limitations guide the food purchase process logic in

QnD:Jamaica. Nutritional demands are set by typical dietary intakes as represented in five major

food groups. Budget limitations are guided by historical household expenditure data.

Additionally, substitutability due to price or cultural preferences structures the food purchase

process.

It is a fair assumption that perceived value, rather than brand recognition or domestic

loyalty, is the most important decision in Jamaican consumers food buying decisions especially

for the lower income groups (USDA, 2003; Timmer, 2000). Thus dietary outcomes in Jamaica

are foremost a function of food prices relative to income. However, other preferences are

important as well as the increasing role these preferences have in higher income echelons.

In the case of carbohydrates, the traditional domestic starch, the yam, is actually in

competition with cereals, rice, and potatoes for the staple food "slot" in the Jamaican diet. Rice

has long been a staple of Jamaican food culture although it has always been largely imported.









Imported cereals and potatoes, on the other hand, have only recently become staples in the

Jamaican diet and their volume per capital has been on the rise.

These foods are usually imported as refined wheat and corn flours, and processed bread

and pasta products for the cereal group and as frozen potatoes used largely for French fries. All

of these imported products generally have much lower nutritional values than domestic yams.

Thus, the competition that yams face from cereals and potatoes has major implications for the

Jamaican diet in terms of nutritional characteristics and cultural value not to mention the

Jamaican small farmer.

Because of these nutritional effects incurred from the price competition between different

products in similar roles, all dietary choices were simplified into five food group categories:

carbohydrates, vegetables, legumes, meat and fish. The average Jamaican diet is composed of a

spread of the five categories, each of which can be satisfied by two different but substitutable

food types distinguished by location of origin (domestic or imported) and nutritional qualities.

By setting minimum required/desired amounts for each food category within the total food

basket, it is ensured that the price line is not the only factor that influences consumer decisions.

At this stage in the simulation each population group has earned an income represented as

their DPrePurchaseWealthStore and they are ready use this store to purchase food. The food

types that can be purchased are:

* Carbohydrates: CRootCrops, CImportedStaples
* Vegetables: CVegetables, CImportedVegetables
* Legumes: CLegumes, CImportedLegumes
* Animal Meat: CLivestock, CImportedAnimalMeat
* Fish: CFish

The base preferential consumption basket in terms of number of servings per food group is

shown in Table 5-8. The qualities of each serving are described in Table 5-9.









Table 5-8 Base preferential consumption basket.
Base preferential Staple carbohydrates Vegetables Meat Legumes
basket by food 7 servings 4 servings 2 servings 1
groups serving
Calories/domestic 700 120 538 60
daily serving
Calories/imported 1624 120 538 60
daily serving

Table 5-9 Descriptions of one serving for each food type.
Staple carbohydrates Vegetables Animal meat Legumes
Size 100g 100g 100g 40g
Food type Domestic Imported Domestic Imported Domestic Import
(yam) (dumplings/ (cabbage) (cabbage) (fried (fried
cereal flour) chicken) chicken)
100gserved
=50g flour

Calories/ 100 232 34 34 269 269 60
100g
Fatg 0.0 0.7 0 0 15 15
Fiber g 3.7 1.7 2 2 0 0
Sugar g 1.0 10.2 3 3 0 0
Protein g 2.0 5.3 1 1 29 29

Unrealistic amounts of money could still be spent trying to fulfill the minimum of a certain

category requirement in some situations. For instance the price of vegetables could increase

above a threshold (and supply was still at a level where a significant amount could be purchased)

where consumers in reality would not buy any. To account for these situations a maximum

expenditure for each category was set. This is based off maximum historical expenditure

percentiles by commodity group. These generally came from 1992 data, when both food

expenditure as a percent of total budget as well as the annual percent increase in the food and

drink CPI (77.30%) was at its highest for the study period (JSLCs 1990-2003). 11


11 Interesting exceptions include greater proportion spent on starches in 1990 compared to 1992. It appears that this
was largely compensated by a decrease in cereal purchases in the lowest quintiles and by a cut back in meals away
from home by the middle class. Fruit and vegetable expenditures in 2003 comprised greater shares of the budget
than in 1992 for all regions.









The maximum food expenditures as a percentage of total consumption by region were

rural=65%, urban=60% and Kingston/Saint Andrew (KSA)=55%. Table 5-10 shows the

maximum expenditures by food category used in QnD:Jamaica.

Table 5-10 Maximum food group expenditures by geographic region.
KSA Towns Rural
Meat & dairy 31% 30% 39%
Carbohydrates 16% 18% 24%
Fruits & 14% 10% 9%
vegetables
Meals away 24% 26% 20%
Other 15% 16% 8%

The maximum all food expenditure is declared as the DFoodBudgetFactor for each population

group based on their regional location (rural, urban or KSA). Likewise, the maximum

percentage of food budget allotted to each food group is declared for each population group as

DMaxFoodGroupBudgetFactor.

These factors are used to constrain the desired purchase within reasonable expenditure

limits. The DFoodBudgetFactor is multiplied by the population group's

DPrePurchaseWealthStore to arrive at the DMaxFoodBudget. This budget is then appropriated

to DMaxFoodGroupBudgets. These are further divided into MaxFoodTypeBudgets by

multiplying the DMaxFoodGroupBudget by the related food type's expenditure factor.

The different transportation and marketing factors present in rural and urban areas result in

different regional/habitat level food prices in a given parish. Therefore, the regional price

differentials, DRuralFoodGroupPriceRatio and DUrbanFoodGroupPriceRatio, between local

and imported FoodTypes are calculated separately:

1) DHabitatConsumerlmportedFoodTvpePrice / DHabitatConsumerDomesticFoodTvpePrice =

DHabitatFoodGroupPriceRatio









The resulting DHabitatFoodGroupPriceRatio is used in a simple linear PRelationship

function to determine the percentage of a FoodGroup's demand that will be filled by its

associated local FoodType (Table 5-11).

Table 5-11 Food type expenditure factor.
X Y
DHabitatFood DDomesticFoodTypeExp Notes
GroupPriceRat enditureFactor
io
0 .25 ImportedFoodTypePrice is anything <50%
of the DomesticFoodTypePrice in same
FoodGroup
.5 .25 ImportedFoodTypePrice is half of the
DomesticFoodTypePrice in same
FoodGroup
1 .5 ImportedFoodTypePrice is equal to the
DomesticFoodTypePrice in same
FoodGroup
2 .75 ImportedFoodTypePrice is twice of the
DomesticFoodTypePrice in same
FoodGroup
>2 .75 ImportedFoodTypePrice is more than twice
of the DomesticFoodTypePrice in same
FoodGroup

The above relationship function assumes that the consumption amount of a particular

FoodType is directly proportional to its price relative to substitutable FoodTypes within a given

FoodGroup. These are the values currently in the ComponentDetails for all food groups except

vegetables. The current relationship for vegetables describes a less elastic price responsiveness.

No matter how much cheaper imported vegetables are than domestic ones, imported vegetables

will never make up more than 55% of the vegetables' budget as opposed to a maximum of 75%

for the other imported food types. The relationship could be modified to show either a preference

for domestic or imported foods within a food group when the price ratio equals one or other

"tipping points" of interest.










Having accounted for consumer price responsiveness with the above relationship,

consumer dietary demands as captured in the five food groups are then calculated. Basic food

group needs are declared in the "GLOBAL" section of ComponentDetails file as

DMonthlyFoodGroupNeeds (kg/capita/mo). The total needs of each food group for a

population group are calculated:

3) DTotalMonthlyFoodGroupNeeds = DMonthlyFoodGroupNeeds X DPopulation X DAverageFamilySize

The unsold homegrown food available to landowning population groups is then accounted

for:

4a) DFoodTvpeStore = DPrePurchaseFoodTypeStore

4b) DTotalMonthlyFoodGroupNeeds DPrePurchaseFoodTpe Store = DFoodGroupPurchaseDemand

Next, the total demand for each FoodGroup is distributed into its associated local and

imported FoodType demands using the above expenditure factor based on price differences

between substitutable products:

5a) DDesired(Domestic)FoodTvpePurchase = DFoodGroupPurchaseDemand X D(Habitat)FoodTvpe

ExpenditureFactor

5b) DDesired(Imported)FoodTvpePurchase = DFoodGroupPurchaseDemand -

DDesired(Domestic)FoodTvpePurchase

The cost of each desired amount of food is then calculated by multiplying it by its

associated price.

6) DDesiredFoodTvpePurchaseDemand X DHabitatConsumerFoodTpePrice = DDesiredFoodTvpeCost

In order to constrain the amount of money spent on a food group, this cost is then

compared to the maximum allowed expenditure (described above) for a particular food group12

If the DDesiredFoodTvpeCost is more than the allowed expenditure than the amount to be spent on


12 This accounts for intra FoodGroup substitutability while the ExpenditureFactor accounts for inter FoodGroup
substitutability.









the food type is instead set to the DFoodTypeBudget. This expenditure, whether

DDesiredFoodTvpeCost or DFoodTypeBudget, is tracked as DMoneyPaidForFoodType.

In case the DDesiredFoodTypeCost (and the associated DDesiredFoodTypePurchase) was

unreasonably large and constrained by the maximum budget, the real amount to be purchased

must be recalculated:

7) DFoodTvpePurchase = DMoneyPaidForFoodType DHabitatConsumerFoodTpePrice

This amount is then transferred from the DNationalFoodType Store into that population's

DFoodTypeStore and the DMoneyPaidForFoodType is transferred from their DWealthStore into

the CUrbanMiddleFolk's DWealthStore

Consumption Testing

Figure 5-56 shows the large geographical spread in the imported vs. domestic

carbohydrates price ratios generated by the marginal marketing and transportation cost factors.

Also shown is the decrease in this ratio over time, which agrees with historical evidence,

indicating that imported food prices have decreased relative to domestic food. Figure 5-57

shows the simulated annual totals of imported vegetables purchased vs. the known imported

levels. This food type shows more realistic simulation amounts than imported staples levels

shown in Figure 5-58.












S0.90

0.80
E
0
O
z 0.70

E 0.60
0
2 0.50
0.)
o 0.40
E
S0.30

S0.20
.o Min Price Ratio
S0.10

0.00 -----
NC NM Mo I- 'I- LO LO (D (D r- r- 00 0 G Go o 0 C4 C


Year

Figure 5-56 Maximum and minimum rural carbohydrate price ratios across parishes.


18000000
S_ 16000000 -Known
S14000000 -Simulated
S- 12000000
,U 10000000
S^ 8000000
f 6000000
0 z 4000000





Year
Figure 5-57 Simulated vegetable purchases.

Year

Figure 5-57 Simulated vegetable purchases.










-0
a,
c-
0)
_0




0
0-


"t
0
-E


Figure 5-58 Imported staples purchased.


500000000
450000000
400000000
350000000
300000000 -
250000000
200000000
150000000 -*-- Known
100000000
5-- Simulated
50000000
0
1992 1993 1994 1995 1996 1997
Year









CHAPTER SIX
AN ILLUSTRATIVE SCENARIO SIMULATION

The following is a discussion of a simulation exercise completed with QnD:Jamaica. It

serves as an illustration to guide and inspire future scenario simulations. In this exercise only

one external parameter is manipulated while all other components are held constant. This

approach is one traditional method in scenario modeling and can be understood as an overlay of

one new quality against a known historical background. In this way, this simulation exercise is

essentially asking, "What if one external change of concern had occurred during this known

historical period?" To expand this exercise, one internal parameter was also manipulated for a

total of four scenarios including the unmodified base historical one. Changing these two

parameters only allows one to perform a coarse sensitivity analysis. As QnD:Jamaica is in its

relatively early stages of development it is important to isolate influences to really examine the

model's responsiveness. Future simulations using QnD:Jamaica will move beyond this approach

and describe a variety changing parameters in complex scenario descriptions.

Biofuel Demand and Rising Corn Prices

There has been much recent discussion surrounding the ramifications of rising corn and

soybean prices and the role of growing ethanol biofuel demand in these increases13. The recent

rise in corn tortilla prices in Mexico and the resulting food security effects has been well

documented (Roig-Franzia, 2007). In the US, there has been a 70% increase in poultry feed

prices over the past six months resulting in a 14% increase in broiler prices over the same period

(Sauser, 2007).




13 It is important to note that rising corn prices are likely an emergent property of other feedbacks in the global
system besides ethanol demand. Potential activities of influence include droughts on multiple continents and rising
international wheat costs also.









In Jamaica, these price increases are also a major concern for the poultry industry, as much

of the feed is imported corn (Edwards, 2007). The Jamaican consumer's chicken prices have

steadily increased over the past year to compensate for the increased production costs (Thame,

2007). The higher costs of this important source of protein in the Jamaican diet will likely have

important nutritional effects on the population. In addition to these most direct food security

concerns, downturns in the large poultry industry would have ripple effects across the economy.

It is important to note that these increases are occurring against a backdrop of historically low

corn and feed stuff prices. While ethanol production is a new element in corn prices and its

effects on value is a large concern, the diminished prices throughout the 1990s and early 2000s

could arguably be considered unrealistically low.

Jamaican corn-based feed prices increased by 25% in 2006 and are expected to increase

another 35% by the end of 2007 (Thame, 2007). These values were used to estimate an external

feedstuff price driver file that was used in place of the original timeseries file containing known

historical prices. This file describes a linear feed price increase of 25% per year from the base

price in 1992. To expand upon this simulation, the internal parameter describing consumers'

ideal poultry demand total was increased from 3kgs/capita/month to 4kgs/capita/month in two

scenarios. This represents changing preference, or utilization, patterns.

* Historical Scenario: 1990s feedstuff prices and historical 3kgs/capita/month livestock
desired utilization

* Scenario 1: 25% annual increase in feedstuff prices and historical 3kgs/capita/month
livestock desired utilization

* Scenario 2: 25% annual increase in feedstuff prices and increase in desired livestock
utilization to 4kgs/capita/month

* Scenario 3: Historical feedstuff prices and increase in desired livestock utilization to
4kgs/capita/month









Discussion of Scenario Simulations

National domestic livestock production volumes, purchase volumes and consumer desired

purchase volumes are the three output parameters examined in this discussion of the four

scenario runs. These subjects were chosen because they embody the key qualities of food

security: availability, accessibility and utilization. Total production amounts are described by the

"Available" curve in the figures below. In these scenarios, the maximum desired amounts never

exceeded the maximum budgeted amounts, thus, the "Accessible" curve describes the demand

for domestic livestock amount and is unlimited by budget constraints. The "Purchase" curve

shows what was actually bought by consumers. Also shown in the figures is the known real

annual domestic livestock production volume as a reference (Figures 6-1, 6-2, 6-3 and 6-4).

In the historical scenario, the available amount is similar to the historically known

production level, especially in the first five years of simulation. This scenario is showing an

available level that does not meet the national needs and purchasing ability in the first four years.

After 1996, however, accessibility becomes the limiting factor with purchased amounts declining

below available levels as well as earlier purchase levels. Scenario 1 clearly shows a lower level

of available chicken than the historical scenario. This is expected do to the rising costs of

production incurred by the increased feed costs. Accessibility is similar to the historical scenario

during the early years of simulation, but is lower in later years because the rising costs of

production are being passed on to the consumer. These increased domestic chicken prices are

causing the relative price between domestic and imported chicken to be greater which is the

model's mechanism to limit consumer's domestic purchasing power. This decreased

accessibility is especially apparent in years 5, 7, 8, 9, 10, and 11 of Scenario 1. However,

lowered domestic availability is still a greater limitation than accessibility in all months except a

few in year 7 and most months in year 9.













S8000000
7000000
1 6000000
0> 5000000
S4000000
o 3000000Available
a. Available
.2 2000000 Accessible
SPurchased
W 1000000 Known Availability
E
o 0
0


Year

Figure 6-1 Historical scenario of domestic livestock production and demand.


8000000
0 7000000
0
' 6000000
> -5000000
S 4000000
- --'3000000
a 2000000
E
o 1000000
0


- ---



-Available
-Accessible
Purchased
Known Production


1 2 3 4 5 6 7 8 9 10 11
Year

Figure 6-2 Scenario 1 of domestic livestock production and demand.

In Scenario 2, accessibility is always greater than the low amounts of available chicken. If

poultry production decreased to levels simulated in this scenario, there would be a strong amount

of unmet demand which would have strong implications for protein intake and shifts in diet

composition. Scenario 3 shows purchasing levels most similar to known historical production

amounts indicating that the Jamaican consumer's poultry demand is likely closer to

4kgs/capita/mo than 3kgs/capita/mo.












8000000
0 7000000 -1
6000000
> 5000000 -
g) 4000000 -
3000000 -" J
0 2000000 -- Availabile
E Accessibile
0 1000000 Purchased
0
1 2 3 4 5 6 7 8 9 10 11
Year

Figure 6-3 Scenario 2 of domestic livestock production and demand.



9000000
0 8000000
0- 7000000 M
0 6000000
( f 5000000 -
.0 4000000 -
"U 3000000 Available
S 2000000 Accessible
O 1000000 Purchased
0
1 2 3 4 5 6 7 8 9 1011
Year

Figure 6-4 Scenario 3 of domestic livestock production and demand.

Figure 6-5 shows national base domestic livestock prices as well as the regional prices for

rural and urban habitats in two parishes. It is not surprising that the urban area in Saint James

generally shows the lowest prices as it has the lowest marketing coefficient being closest to a

large market, Montego Bay. The sharp price increases throughout the scenario in all regions are

due to transport infrastructure shocks caused by floods. It is interesting to note that there is never

an event where one of these parish's habitat areas, either rural or urban, experiences a spike

without the other habitat area.












National
Saint James Rural
o 100 Saint James Urban
Z Saint Elizabeth Rural
80 -- Saint Elizabeth Urban ___
80 -
I-




640

S20
E
o
0
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Year

Figure 6-5 Regional price shocks to domestic livestock in historical scenario.

Figure 6-6 shows a technical error in QnD:Jamaica. The purchase amount process is using

the budget data value from the previous month rather than the current month's data value as

designed. This is resulting in what appears to be overly dynamic purchase output. Figure 6-7

also shows the Saint James farmer population's purchase amounts of imported and domestic

chicken alongside their total poultry demand/needs. The output data shown in this graph has

been modified to remove the noise errors that QnD:Jamaica produced. However, the most

important results can be seen in both figures. The model is showing purchasing observed in the

historical data records: decreasing domestic poultry purchases and increasing imported poultry

purchases. This similarity to known trends is evidence that builds confidence in the mechanisms

employed in QnD:Jamaica.













80000


70000


60000


50000


40000


30000 ..


20000


10000 -


0
1992 1993 1994 1995 1996 1997 1998


Figure 6-6 Internal structural design noise.




60000 1


50000




40000




S30000




20000




10000


1992


Needs
Domestic Purchased
Imports Purchased
Total

1999 2000 2001 2002


- Desired Domestic Livestock
- Desired Imported Livestock
Total


1993 1994 1995 1996 1997 1998 1999 2000
Year


2001 2002


n 1l









Figure 6-7 Domestic and imported livestock purchases of Saint James farmers in historical
scenario.

Figure 6-8 describes the volume of imported animal meat purchased in the various

scenarios. As expected, the purchased import amount is lowest in the historical scenario when

domestic production was relatively high, prices relatively low and demand only

3kgs/capita/month. Similarly when the preference parameter is set at the lower value, imported

livestock is still relatively low, but higher than in the historical scenario because domestic

livestock prices were relatively higher compared to imported animal meat prices. Imported

purchase quantities are generally highest in the second scenario when demand is high yet

domestic production shows decreasing output due to higher feedstuff prices. In the first five

years of the scenarios, imported purchased do not differ greatly from the amounts produced in

the historical scenario. This is likely a result of the combination of domestic livestock prices'

relative competitiveness to imported prices in the early years as well as relatively small value's

in real internal feedstuff prices despite the great external increases.


12000000
1 -- Historical Scenario
10000000 -- Scenario 1
8 Scenario 2
Ee 8000000 -- Scenario 3

C 6000000 --
2 | 4000000 ..j'i W
E 2000000
0 -
1 2 3 4 5 6 7 8 9 10 11
Time
Figure 6-8 Total imported animal meat.

The changing demand for imported animal meat as domestic livestock prices and supply

change as illustrated in the above scenarios indicates the need to consider the potential qualities









of the imported animal meat in the event of a global feedstuff increase. As this situation is

occurring in the global food system not just in the Jamaican context, at least one other immediate

external parameter is important to this scenario exploration. Because Jamaica imports a fair

share of its total poultry supply, around 25% annually throughout the 1990s, external poultry

prices are important in the Jamaican food system. The rising cost of corn is also being passed

onto US chicken prices (Sauser, 2007). The next step in this scenario exercise should be to

examine the combined effects of rising imported poultry prices would have in combination with

rising feedstuff prices.









CHAPTER 7
CONCLUSIONS

Modeling Jamaican Food Systems

This thesis represents the second developmental cycle in what hopes to be a chain of

constructive model design and learning. The first iteration was completed during the summer of

2005 during multiple workshops led by GECAFS. During the phase of model development

documented in this thesis, the structure of the base version was fleshed out into greater detail.

Discussions in the field revealed a model structure that aims to depict the Jamaican food system

in a manner most appropriate to those interests most involved in the situation. In depth empirical

research guided parameterization of the model as many of the previously desired relationships

and data were not well documented enough to support simulation.

There are many potential future applications of QnD:Jamaica. The most obvious is greater

exploration of consumption activities in terms of the availability and accessibility of food. The

assumptions and effects of the current expenditure factor and price ratio processes need to be

explored further to determine their necessity and/or implications. Because changing food

preferences is a key concern of Jamaica food security one of the user's management options

could be to modify the proportions of desired foods within a consumer's food basket, i.e.

increase daily vegetables demand from 400g to 500g. This represents desired effects of

nutritional education policies or cultural changes. Additionally an ideal preference distribution

based on the CFNI's recommended calorie source distribution by food category could be used to

guide this parameter. Each month's consumption basket could measured against these

proportions to show how simulated nutrition levels compare to desired levels.

Another obvious next direction for the model is the development of parameters based more

directly on natural resources. These should include such indices as forest health (i.e. canopy









cover, biodiversity), fisheries health (reef diversity, fish stocks), erosion indices, water quality

(availability, stormwater and agricultural runoff), etc. Another necessary addition to the model is

development of more internal feedback loops (rather than so many driver files) such as farmers'

land use decisions, internal migration flows, and tying wages to economic sector indicators.

These should also be tied into the ecological dynamics and indices.

Broader Lessons Learned

The most useful application of the knowledge gained during this research will be in

discussions of more formal guidelines for conducting interdisciplinary meso scale research. One

of the key struggles faced during this project was interpreting and manipulating available macro

and micro level data in a manner that revealed a limited number of cross-scalar and cross-

disciplinary linkages and systemic features in an effort to remain fairly transparent. The impact

of this project has both practical and theoretical applications and advances multiple disciplines

separately and the interdisciplinary cause itself. As development planners and the decision

support community seek to apply the adaptive management paradigm, practical systematic

methods that implement the proposed principles must be developed and tested. Traditionally

these two fields have experienced tension due to communication and goal assessment

difficulties, among others. This project serves as an example of one innovative method that aims

to limit these difficulties. The knowledge gained during this research provides these disciplines

with a case study demonstrating an explicitly collaborative, interdisciplinary and iterative

framework for development and implementation of an adaptive management planning tool.

A future comparison of this project's site specific evidence with case studies from other

food systems, will give insight into general structural patterns and "rules" in food systems.

These patterns will help food system planners better maneuver through the many unpredictable

variables as they formulate food security policy.









When a holistic definition is applied to food security, it becomes nearly impossible to

ignore the previously hidden impacts that the current international food system has on all

socioecological aspects of our global society. Traditionally underrepresented groups will be

empowered when a rich definition of food security that necessitates a complex systems approach

to food management is fostered. The food security evaluation framework that this project

promotes has the broadest impact of raising awareness of the extensive and unseen ramifications

of our global food system.









LIST OF REFERENCES


Ackoff, R.L. 1974. Redesigning the Future: A Systems Approach to Societal Planning. John
Wiley and Sons, New York, USA.

Adger, W.N. 2006. Vulnerability. Global Environmental Change 16:268-281.

Alderman, H. 1986. The Effect of Food Price and Income Changes on the Acquisition of Food
by Low-Income Households. International Food Policy Research Institute.

Allen, C., H. Hall, and W. Robinson. 2002. Estimating a Small Scale Macroeconomic Model of
the Jamaican Economy: Some Preliminary Results. Research and Economic Programming
Division Bank of Jamaica.

Andreatta, S. 1998. Transformation of the agro-food sector: lessons form the Caribbean.
Human Organization 57(4) 414-429.

Anderson, P., and M. Witter. 1994. Crisis, adjustment and social change: a case study of
Jamaica, Pages 1-55 in E. Le Franc, editor. Consequences of Structural Adjustment: A
Review of the Jamaican Experience. Canoe, Kingston.

Ballayram and C. Davis. 2003. Economic reforms and agricultural supply response in Jamaica.
Farm and Business: The Journal of the Caribbean Agro-Economic Society 6(1):1-20.

Ballayram and F. Henry. 2004. The Importance of Integrating Food and Nutrition Security
Plans in the Economic Strtegy of the Region. Paper presented at the CARDI/CTA/CFNI
Regional Seminar, Belize City.

Baker, J., H. Jenkins, and H. V. Bonavita. 1999. Basics steps in scenario planning creating
alternative futures, Curtin University of Technology, Perth, Western Australia, 1-51.

Beckford, G. 1972. Persistent Poverty: Underdevelopment in Plantation Economies of the Third
World. New York, USA: Oxford University Press.

Behrman, J. 1995. Micronutrients and household behavior: what we know and what we don't
know. Working Paper Series on Agricultural Strategies for Micronutrients Number 2.
International Food Policy Research Institute, Washington D.C.

Berkes, F., J. Colding, and C. Folke, editors. 2003. Navigating social-ecological systems:
building resilience for complexity and change. Cambridge University Press, Cambridge,
UK.

Bernanke, B., and M. Gertler. 1995. Inside the Black Box: The Credit Channel of Monetary
Policy Transmission. Journal of Economic Perspectives 9 (4).

Berry, W. 2002. The whole horse: the preservation of the agrarian mind. Pages 7-19 In A.
Kimbrell, editor. Fatal harvest: the tragedy of industrial agriculture.









Bouis, H., L. Haddad, and E. Kennedy. 1992. Does it matter how we survey demand for food?
Food Policy 17(5):349-360.

Bouis, H., L. 1994. The effect of income on demand for food in poor countries: are our
databases giving us reliable estimates? Journal of Development Economics 44(1): 199-226.

Bourne, C. 1977. Inflation in the Caribbean. Institute of Social and Economic Research,
Jamaica.

Brown, H. A. 1994. Economics of Disasters with Special Reference to the Jamaican
Experience. University of the West Indies Centre for Environment and Development,
Jamaica.

Bunch, M.J. 2003. Soft Systems Methodology and ecosystem approach: A system study of the
Cooum River and environs in Chennai, India. Environmental Management 31(2): 182-197.

Capra, F. 1996. The Web of Life: A New Scientific Understanding of Living Systems. Anchor
Books, New York, USA.

Caribbean Disaster Emergency Response Agency (CDERA). 2003. Status of Hazard Maps
Vulnerability Assessments and Digital Maps: Jamaica Country Report. CDERA, Jamaica.

Carpenter, S., W. Brock and P. Hanson. 1999. Ecological and social dynamics in simple models
of ecosystem management. Conservation Ecology 3(2):4. [online] URL:
http://www.consecol.org/vol3/iss2/art4/.

Checkland, P.B. 1999. Soft Systems Methodology in Action. John Wiley & Sons, Toronto.

Checkland, P.B. 1979. Techniques in 'soft' systems practice part 2: building conceptual
models. Journal of Applied Systems Analysis 6:33-40.

Checkland, P.B., and S. Holwell. 2004. "Classic" OR and "soft" OR- and asymmetric
complementarity. Pages 45-60 in M. Podd, editor. Systems modelling: theory and practice.
John Wiley and Sons, UK.

Checkland, P.B. and J. Scholes. 1990. Soft Systems Methodology in Action. John Wiley &
Sons, Toronto.

Churchman, C. W. 1979. The Systems Approach. New York, USA: Laurel.

Clarke, C., and D. Howard. 2006. Contradictory socio-economic consequences of structural
adjustment in Kingston, Jamaica. The Geographical Journal 172(2):106-129.

Corral, L., P. Winters, and G. Gordillo. 2000. Food insecurity and vulnerability in Latin
America and the Caribbean. Working Paper Series in Agricultural and Resource
Economics. University of New England.









Costanza, R., J. Audley, R. Borden, P. Ekins, C. Folke, S.O. Funtowicz, and J. Harris. 1995.
Sustainable trade- a new paradigm for world welfare. Environment 37(5): 16-20.

Cox, P., 1996. Some issues in the design of agricultural decision support systems. Agricultural
Systems 52(2-3):355-381.

Craigwell, R. 1991. Wages and prices: what causes what? Social and Economic Studies
40(1):93-103.

Davis, F. D. 1989. Perceived usefulness, perceived ease of use and user acceptance of
information technology. MIS Quarterly 13:319-340.

Davis-Morrison, V., and D. Barker. 1997. Resource management, environmental knowledge, and
decision making in small farming systems in the Rio Grande valley, Jamaica. Caribbean
Geography 8(2):96-106.

Douthwaite, B., N. C. de Haan, V. Manyong, and D. Keatinge. 2001. Blending "hard" and
"soft" science: the "follow-the-technology" approach to catalyzing and evaluating
technology change. Conservation Ecology 5(2):13 [online] URL:
http://www.consecol.org/vol5/iss2/artl3/

Duchin, F. 2005. Sustainable consumption of food. Journal of Industrial Ecology 9(1-2):99-
114.

Dyball, R., S. Beavis, and S. Kaufman. 2005. Complex adaptive systems: Constructing mental
models. Pages 41-62 in M. Keen, V.A. Brown and R. Dyball, editors. Social Learning in
Environmental Management: Towards a Sustainable Future. Earthscan: London, UK.

Dyer, G. A., S. Boucher, and J. E. Taylor. 2006. Subsistence response to market shocks.
American Journal of Agricultural Economics 88(2): 279-291.

Eakin, H. 2005. Institutional change, climate risk, and rural development vulnerability: cases
from central Mexico. World Development 33(11):1923-1938.

Economic Comission for Latin Ameruca and the Caribbean (ECLAC). 2004. Assessment of the
Socioeconomic and Environmental Impact of Hurricane Ivan on Jamaica.

Eden C., and F. Ackermann. 2004. Use of "soft" OR models by clients-what do they want from
them? Pages 146-163 in M. Pidd, editor. Systems modelling: theory and practice. John
Wiley and Sons, England.

Edwards, A. February 23, 2007. Cost of imported grain impacts Jamaica Broilers' profits. The
Jamaica Observer.

Edwards, D. 1961. An Economic Study of Small Farming in Jamaica. Institute of Social and
Economic Research: Mona, Jamaica.









Ericksen, P.J. 2006a. Conceptualizing Food Systems for Global Environmental Change
Research. Global Environmental Change and Food Systems: Wallingford, UK.

Ericksen, P.J. 2006b. Assessing the Vulnerability of Food Systems to Global Environmental
Change: A Conceptual and Methodological Review. Global Environmental Change and
Food Systems: Wallingford, UK.

Espeut, P. December 10, 2003. The drama of Christmas vending. The Jamaica Gleaner.

FAO. 2003. Nutrition Country Profile: Jamaica. Food and Agriculture Organization of the
United Nations.

FAO, FAOSTATS, 2007, http://www.faostats.org, accessed on 3/07.

Folke, C. 2006. Resilience: The emergence of a perspective for social-ecological systems
analysis. Global Environmental Change 16:253-267.

Ford, J. R. 1992. Guyana's food performance in a Caribbean context: lessons for food security
policy. Food Policy October:326-336.

Forrester, J. W. 1961. Industrial dynamics. MIT Press, Cambridge, M.A.

Fraser, E.D., W. Mabee, and F. Figge. 2005. A framework for assessing the vulnerability of
food systems to future shocks. Futures 27:465-479.

Goldsmith, A. 1981. The politics of rural stagnation: development policy, local organizations and
agrarian change in Jamaica. UMI, Ann Arbor, Michigan.

Gordan, S. November 24, 2006. Competing supermarket chains still in expansion mode. The
Jamaica Gleaner.

Gunderson, L H., C. S. Holling, and S.S. Light. 1995. Barriers and bridges to the renewal of
ecosystems and institutions. Columbia University Press, New York, USA.

Handa, S., and D. King. 2003. Adjustment with a human face? Evidence from Jamaica. World
Development 31(7):1125-1145.

Holling, C.S. 1973. Resilience and stability of ecological systems. Annual review of Ecology
and Systematics 4:1-23.

Holling, C.S., editor. 1978. Adaptive environmental assessment and management. Wiley,
London, UK.

Holling, C.S. 1986. The resilence of terrestrial ecosystems:local surprise and global change.
Pages 292-317 in W. C. Clark and R.E. Munn, editors. Sustainable development of the
Biospere. Cambridge University Press, UK.









Holling, C.S. 1996. Engineering resilience versus ecological resilience. Pages 31-44 in P.
Schulze, editor. Engineering within ecological constraints. National Academy Press,
Washington, D.C., USA.

Holling, C. S., and L. H. Gunderson. 2002. Resilience and adaptive cycles. Pages 25-62 in C.S.
Holling and L. H. Gunderson, editors. Panarchy: Understanding transformations in human
and natural systems. Island Press, Washington D.C.

Hourts, H., and D. Miller. 2006. The cell phone: and anthropology of communication. Berg
Press, New York, USA.

Ishemo, A., H. Semple, and E. Thomas-Hope. 2006. Population mobility and the survival of
small farming in the Rio Grande Valley, Jamaica. The Geographical Journal 172(4):318-
330.

Ison, R.L., P.T. Maiteny, and S. Carr. 1997. Systems methodologies for sustainable natural
resources research and development. Agricultural Systems 55(2):257-272.

Jamaican Meteorological Society. Parish Rainfall, 2007, http://www.metservice.gov.jm/,
accessed on 2/07.

Jamaican Ministry of Agriculture (MOA). 2001. Update on the State of Food Security in
Jamaica. Ministry of Agriculture, Kingston, Jamaica.

Jamaican Ministry of Agriculture (MOA). 2005. Agricultural Development Strategy. Ministry
of Agriculture, Kingston, Jamaica.

Jamaica Sustainable Development Network (JSDNP). Jamaica Sustainable Development
Networking Programme. 2005, http://www.jsdnp.org.jm/JSDNPdotORGdotJM/home.htm,
accessed on 10/06.

Janssen, M.A., O. Bodin, J.M. Anderies, T. Elmquist, H. Ernstson, B. McAllister, P. Olsson, and
P. Ryan. 2006. Toward a network perspective of the study of resilience in socio-
ecological systems. Ecology and Society 11(1).

Kasperson, J.X. and R.E. Kasperson. 2001. Global Environmental Risk. Earthscan and UNU
Press, London, UK.

Keen, M. and S. Mahanty. 2005. Collaborative learning: bridging scales and interests. Pages
104-122 in Keen, M., V.A. Brown and R. Dyball, editors. Social Learning in
Environmental Management: Towards a Sustainable Future. Earthscan, London.

Kiker, C.F, J.W. Milon, and A.W. Hodges. 2003. Adaptive learning from science-based policy:
the Everglades restoration. Ecological Economics 37:403-416.

Kiker, G.A., T.S. Bridges, I. Linkov, A. Varghese, and T. Seager. 2006. Application of multi-
criteria decision analysis in environmental decision-making. Integrated Environmental
Assessment and Management 1(2):1-14.









Kiker, G.A. and I. Linkov. 2006. The QnD model/game system: integrating questions and
decisions for multiple stressors. in Arapis, G., N. Goncharova, and P. Baveye, editors.
Ecotoxicology, Ecological Risk Assessment, and Multiple Stressors. Springer,
Netherlands.

King, D. 1998. Adjustment, stabilization, and poverty in Jamaica: 1989-1995. University of the
West Indies, Mona, Jamaica.

Le Franc, E. 1981. Social structure, land use, and food availability in the Caribbean. Food
Nutrition Bulletin 3(4).

Le Franc, E., and A. Downes. 2001. Measuring human development in countries with invisible
economies: challenges posed by the informal and remittance sectors in Jamaica. Social
and Economc Studies 50(1).

Lewis, W. A. 1954. Economic development with unlimited supplies of labor. Manchester
School of Economic and Social Studies 22:139-191.

Lyons, M. 2004 Insights from complexity: organizational change and systems modelling.
Pages 21-44 in M. Pidd, editor. Systems Modelling: Theory and Practice. John Wiley and
Sons, Chichester, UK.

Lyson, C., and J. Green. 1999. The agricultural marketscape: A framework for sustaining
agriculture and communities in the Northeast. Journal of Sustainable Agriculture
15(2):133-150.

Meadows, D.H. 1985. System analysis and the world food system. European Journal of
Operational Research 20:158-167.

Meadows, D.H. and J.M. Robinson. 1985. The Electronic Oracle: Computer Models and Social
Decisions. John Wiley and Sons, Chichester, UK.

Mintz, S. 1974. Caribbean transformations. Aldine Publishing, Chicago, USA.

Mintz, S., and D. Hall. 1960. The origin of the Jamaican internal marketing system. Yale
University Press, New Haven, USA.

Morecroft, J. 2004. Pages 101-126 in M. Pidd, editor. Systems modelling: theory and practice.
John Wiley and Sons, UK.

Morgan, M.G., B. Fischhoff, A. Bostrom, and C.J. Atman. 2002. Risk communication: a mental
models approach. Cambridge University Press.

Nkemdirim, L. C. 1979. Spatial and seasonal distribution of rainfall and runoff in Jamaica.
Geographical Review 69(3):288-301.

Norton A., and R. Symanski. 1975. The internal marketing system of Jamaica. Geographical
Review 65(4):461-475.









O'Brien, K.L., and R.M. Leichenko. 2000. Double exposure: assessing the impacts of climate
change within the context of economic globalization. Global Environmental Change
10:221-232.

Odum, H. 1983. Systems Ecology: an Introduction. John Wiley, New York, USA.

Osei, P. D. 2002. A critical assessment of Jamaica's national poverty eradication programme.
Journal ofInternational Development 14:773-788.

Ostrom, E. 1999. Coping with tragedies of the commons. Annual Review ofPolitical Science
2:493-535.

Pan American Health Organization (PAHO). Emergency Preparedness and Disaster Relief
Coordination Program, Washington D.C., US Disaster Reports, 5. 1988.
http://www.crid.or.cr/digitalizacion/pdf/eng/doc6803/doc6803.htm, accessed on 10/06.

Parker P., R. Letcher, A. Jakeman, M.B. Beck, G. Harris R.M. Argent, M. Hare, C. Pahl-Wostl,
A. Voinov, M. Janssen, P. Sullivan, M. Scoccimarro, A. Friend, M. Sonnenshein, D.
Barker, L. Matejicek, D. Odulaja, P. Deadman, K. Lim, G. Larocque, P. Tarikhi, C.
Fletcher, A. Put, T. Maxwell, A. Charles, H. Breeze, N. Nakatani, S. Mudgal, W. Naito, O.
Osidele, I. Eriksson, U. Kautsky, E. Kautsky, B. Naeslund, L. Kumblad, R. Park, S.
Maltagliati, P. Girardin, A. Rizzoli, D. Mauriello, R. Hoch, D. Pelletier, J. Reilly, R.
Olafsdottir, and S. Bin. 2002. Progress in integrated assessment and modelling.
Environmental Modelling and Software 17:209-217.

Pavao-Zuckerman, M.A. 2000. The conceptual utility of models in human ecology. Journal of
Ecological Anthropology 4:37-56.

Pidd, M. 2004. Complementarity in systems modeling Pages 1-19 in M. Pidd, editor. Systems
Modelling: Theory and Practice. John Wiley and Sons, England.

Pirot, Jean-Yves, P. Meynell and D. Elder. 2000. Ecosystem Management: Lessons from
around the world: a guide for development and conservation practitioners. IUCN, France.

Planning Institute of Jamaica. 1990-2004. Economic and Social Survey of Jamaica, Kingston.

Planning Institute of Jamaica and Statistical Institute of Jamaica. 1990-2003. Jamaica Survey of
Living Conditions, Kingston.

Popkin, B.M. 2003. The nutrition transition in the developing world. Development Policy
Review 21(5-6):581-597.

Pothukuchi, K and J.L. Kaufman. 2000. The food system: A stranger to the planning field.
Journal of the American Planning Association 6(2): 113-124.

Quinn, J. B. & J. Voyer. 1996. Logical incrementalism: Managing strategy formation. Pages
95-101 in H. Mintzberg and J.B. Quinn, editors. The Strategy Process: Concepts,
Contexts, Cases. Prentice Hall, Englewood Cliffs.











Rittel, H. and M. Webber. 1973. Dilemmas in a general theory of planning. Policy Sciences
4:155-169.

Robinson, W. and J.W. Robinson. 1997. The transmission mechanism of monetary policy in the
Jamaican economy. Annual Conference of the Caribbean Centre for Monetary Studies.

Roig-Franzia, M. January 27, 2007. A culinary and cultural staple in crisis. Washington Post.

Roux, D.J., K. H. Rogers, H. C. Biggs, P. J. Ashton and A. Sergeant. 2006. Bridging the
science-management divide: moving from unidirectional knowledge transfer to knowledge
interfacing and sharing. Ecology and Society 11(1): 4 [online] URL:
http://www.ecologyandsociety.org/vol 1 1/issl/art4/

Ruijs, A., C. Schweigman, and C. Lutz. 2004. The impact of transport- and transaction-cost
reductions on foods markets in developing countries: evidence from tempered expectations
for Burkina Faso. Agricultural Economics 31:219-228.

Salmon, J. U. 2002. Poverty eradication: the Jamaican experience. Social and Economic Studies
51(4):63-94.

Sauser, B. February 13, 2007. Ethanol demand threatens food prices. Technology Review.

Schipper, L., and M. Pelling. 2006. Disaster risk, climate change and international
development: scope for and challenges to, integration. Disasters 30(1):19-38.

Schrage, M. 2000. Serious Play: How the World's Best Companies Simulate to

Innovate. Harvard Business School Press, Boston, USA.

Sen, A. 1980. Poverty and Famines: An Essay on Entitlement and Deprivation. University
Press, Oxford.

Simon, H.A. 1962. The architecture of complexity. Proceedings of the American Philosophical
Society 106(6):467-482.

Skoufias, E. 2003. Is the calorie-income elasticity sensitive to price changes? Evidence from
Indonesia. World Development 31(7): 1291-1307.

Spence, B. 1996. The influence of small farmers' land-use decisions on the status of domestic
food security in Jamaica. Caribbean Geography 7(2):132-142.

Spence, B. 1999. Spatio-evolutionary model of Jamaican small farming. The Geographical
Journal 165(3):296-305.

Stacey, R.D. 2001. Complex responsive processes in organizations: learning and knowledge
creation. Routledge, London.









Statistical Institute of Jamaica (STATINJA). 1996. Census of Agriculture. Kingston.

Sundkvist, A., R. Milestad, and A. Jansson. 2005. On the importance of tightening feedback
loops for sustainable development of food systems. Food Policy 30:224-239.

Thame, C. July 18, 2007. Poultry, bread prices to increase by 15-20%, says FAS. The Jamaican
Observer.

Timmer, C.P. 1980. Food prices and food policy analysis in LDCs. Food Policy August: 188-
199.

Timmer, C.P. 2000. The macro dimensions of food security: economic growth, equitable
distribution, and food price stability. Food Policy 25:283-295.

USDA Foreign Agricultural Service. 2003. Jamaica and dep Retail food sector report. Global
Agriculture Information Network.

van der Heijden, K. 1996. Scenarios. The Art of Strategic Conversation. John Wiley and Sons,
Chichester, England.

van der Heijden, K., R. Bradfield, G. Burt, G. Cairns, G. and G. Wright. 2002. The sixth sense
accelerating organizational learning with scenarios. John Wiley & Sons Ltd.,Chichester,
England.

van der Leeuw, S. E., and C. Aschan-Leygonie. 2005. A long term perspective on resilience in
socio-natural systems. Pages 227-264 in H. Liljenstrom and U. Svedin, editors. Micro,
meso macro: addressing complex systems couplings. World Scientific, Singapore.

van Bertalanffy, L. 1968. General system theory: foundation, development, applications. George
Braziller, New York, USA.

van Ittersum, M. K., R. Rabbinge, and H.C. van Latesteijn. 1998. Exploratory land use studies
and their role in strategic policy making. Agricultural Systems 58(3):309-330.

Walker, A. 2002. The Microstructure of the Jamaican Foreign Exchange Market: Volumes,
Volatility and Spreads. Working Paper Monetary Analysis & Programming Dept.
Research and Economic Programming Division Bank of Jamaica.

Walker, D.H. 2002. Decision support, learning and rural resource management. Agricultural
Systems 73:113-127.

Walker, B., S. Carpenter, J. Anderies, N. Abel, G. Cumming, M. Janssen, L. Lebel, J. Norberg,
G.D. Peterson, and R. Pritchard. 2002. Resilience management in socio-ecological
systems: a working hypothesis for a participatory approach. Conservation Ecology
6(1):14.









Walker, B.H., C.S. Holling, S.C. Carpenter, and A.P. Kinzig. 2004. Resilience, adaptability and
transformability. Ecology and Society 9(2):5 [online] URL:
http://www.ecologyandsociety.org/vol9/iss2/art5/

Weiss, T. 2004a. Restructuring and redundancy: the impacts and illogic of neoliberal
agricultural reforms in Jamaica. Journal of Agrarian Change 4(4):461-491.

Weiss, T. 2004b. (Re-)making the case for land reform in Jamaica. Social and Economic
Studies 53(1):35-72.

Weiss, T. 2006. A precarious balance: neoliberalism, crisis management, and the social
implosion in Jamaica. Capital and Class 85:115-147.

Witter, M. 1988. The role of higglers/sidewalk vendors/informal commercial traders in the
development of the Jamaican economy. Unpublished manuscript, Department of
Economics. University of West Indies.

Wong, D. C. 1996. A theory of petty trading: the Jamaican higgler. The Economic Journal
106(435):507-518.

World Bank. 2003. Jamaica: The Road to Sustained Growth. Country Economic Memorandum.
Report number 26088-JM.

Young, C.E., and L.S. Kantor. 1999. Moving toward the Food Guide Pyramid: Implications for
U.S. agriculture. Pages 403-423 In E. Frazao, editor. America's eating habits: Changes
and consequences USDA Economic Research Service, Washington D.C., USA.

Young, O.R., F. Berkhout, G.C. Gallopin, M.A. Janssen, E. Ostrom, E. and S. van der Leeuw.
2006. The globalization of socio-ecological systems: an agenda for scientific research.
Global Environmental Change 16:304-316.

Zhu, X. and A.P. Dale. 2000. Identifying opportunities for decision support systems in support
of regional resource use planning: an approach through soft systems methodology.
Environmental Management 26(4): 371-384.









BIOGRAPHICAL SKETCH

Lauren C. Justice was born in Atlanta, Georgia on November 19, 1980 to Joseph B. Justice

and Betheda F. Gramling. She graduated from Druid Hills High School in 1999. In 2003 she

graduated from Emory University in Atlanta, Georgia with a Bachelor of Arts in Environmental

Studies. In 2005 she enrolled at the University of Florida to pursue a Master of Science in

Interdisciplinary Ecology.





PAGE 1

1 EXPLORATION OF FOOD SECURITY VIA A MESO SCALE MODELING APPROACH: JAMAICA AS AN ILLUSTRATIVE CASE By LAUREN COLLEEN JUSTICE 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 2007

PAGE 2

2 2007 Lauren Colleen Justice

PAGE 3

3 ACKNOWLEDGMENTS I extend infinite gratitude to my adviso r, Dr. Greg Kiker, for supporting me both intellectually and financially th roughout this project. His positive attitude always provided reassurance in this project and myself. My committee members, Dr. Peter Hildebrand and Dr. Clyde Kiker, have also provided much apprecia ted support. I also th ank Mrs. Mindy Kiker for her generous guidance. In Jamaica I thank the University of the West Indies Mona, in particular Dr. David Dodman, Dr. David Barker, Dr. Balfour Spence, Dr. Michael Witter and especially Ballayram. Additionally I express appreciation to all of my field collaborators. Thank you to GECAFS for their inspiration to pursue this project and the additional resources they provided. The Center for Latin American Studies dese rves recognition for s upplying the funding to complete the necessary field work. I express gratitude to The School of Natural Resources, especially Dr. Steve Humphrey for guiding a ch allenging and innovative degree program. On the home front I thank Jessica Wielgosz for reminding me that every student deserves a break now and again. I am grateful to my Mother, Betheda Gramling, and Vernon Gramling for their financial support but even more so for their understanding and hugs. I thank my Dad, Jay Justice, for reminding me that science is exhaustively slow and no one alone will solve the worlds problems. I appreciate my brother, Nichol as Justice, for his insp irational conversations. I also thank Bonnie Byrd, and John and Vera Gram ling for their care and attention. Finally I express my utmost respect and appreciation to Wa rd Bradshaw for his motivational attitude and limitless emotional support.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................3 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 ABSTRACT...................................................................................................................................12 1 INTRODUCTION................................................................................................................... ..14 Dynamic Collaborative Models for Food System Exploration .............................................. 14 Jamaica as a Case Study.........................................................................................................16 Objectives...............................................................................................................................17 2 PROBLEM BACKGROUND...................................................................................................18 Hard and Soft Methods for Understanding Complex Systems...............................................18 Wicked Problems............................................................................................................. 19 Complex Systems Theory................................................................................................19 Two Social Facets of Complex Systems Management................................................... 21 Complex systems inherent social features............................................................... 21 Managing the managers........................................................................................... 22 Food Systems as Complex Systems....................................................................................... 23 Agricultural and Food Policy Decision Support.............................................................24 Modeling Complex Systems for Decision Support................................................................ 26 Decision Support Paradigm Shift.................................................................................... 26 Modeling....................................................................................................................... ...27 Hard and Soft Methods for Model Development............................................................ 29 Soft systems methods............................................................................................... 29 Iterative cycles of tool development........................................................................ 30 Cross Disciplinary Convergence and Systems Modeling............................................... 31 3 AN APPROACH TO FOOD SYSTEM MODELING.............................................................. 33 Food System Models..............................................................................................................33 Approach....................................................................................................................... ..........34 Soft Systems Methods.....................................................................................................35 QnD Modeling System....................................................................................................36 Informing Decision Support Tool Development............................................................. 38 4 JAMAICAN FOOD SYSTEM: AN ILLUSTRATIVE CASE.................................................. 40 Historical Review.............................................................................................................. .....40 The Exchange Rate, Inflation and Food Imports............................................................. 41

PAGE 5

5 Economic Development and Poverty Alleviation........................................................... 43 Climatic Shocks............................................................................................................... 44 Hurricane Gilbert......................................................................................................44 Hurricane Ivan..........................................................................................................45 Holistic Exploration.........................................................................................................46 Rich Picture Reveals Conceptual Framework........................................................................ 47 Empirical Review of Conceptual Categories for Model Param eterization............................ 52 Population Typology....................................................................................................... 54 Land use and Agricultural Productivity.......................................................................... 57 Domestic Food Distribution............................................................................................ 59 Food Marketing...............................................................................................................62 Imported Food Distribution and Marketing.................................................................... 64 The Jamaican Consumer.................................................................................................. 65 5 JAMAICAN FOOD SYSTEM MODEL DESIGN AND TESTING........................................ 71 Approach....................................................................................................................... ..........71 Population Typology Representation..................................................................................... 72 Land Use Representation........................................................................................................ 74 Productivity Representation.................................................................................................... 79 Productivity Calibration.................................................................................................. 82 Vegetables................................................................................................................82 Root crops................................................................................................................84 Legumes...................................................................................................................86 Livestock.................................................................................................................. 88 Testing Productivity Simulations.................................................................................... 89 Vegetables productivity............................................................................................ 89 Root crops productivity............................................................................................ 91 Legumes productivity............................................................................................... 93 Livestock production................................................................................................94 Food Importation............................................................................................................. 95 Food Distribution and Marketing Representation..................................................................95 Representing National Farmgate Prices.......................................................................... 96 Crop pricing..............................................................................................................96 Crop pricing calibration...........................................................................................98 Livestock pricing....................................................................................................101 Price Testing..................................................................................................................102 Representing Producer-Higgler Food Transfer Volum es..............................................106 Representing Consumer Prices...................................................................................... 108 Marketing costs......................................................................................................109 Transport costs....................................................................................................... 110 Jamaican Consumer Representation..................................................................................... 112 Consumption Testing..................................................................................................... 118 6 AN ILLUSTRATIVE SCENARIO SIMULATION ............................................................... 121 Biofuel Demand and Rising Corn Prices..............................................................................121

PAGE 6

6 Discussion of Scenario Simulations..................................................................................... 123 7 CONCLUSIONS......................................................................................................................130 Modeling Jamaican Food Systems....................................................................................... 130 Broader Lessons Learned.....................................................................................................131 LIST OF REFERENCES.............................................................................................................133 BIOGRAPHICAL SKETCH.......................................................................................................143

PAGE 7

7 LIST OF TABLES Table page 3-1 Project phases............................................................................................................ .............34 3-2 QnD processes........................................................................................................................37 4-1 TWOACE characteristics of farm er remuneration system..................................................... 49 4-2 Jamaican food system interactions that affect availability conditions. .................................. 52 5-1 Rural population typology......................................................................................................73 5-2 Percent of Jamaicas land uses by parish................................................................................ 77 5-3 Maximum monthly yiel d coefficients (tons/ha) ..................................................................... 79 5-4 Productivity coefficients................................................................................................. ........80 5-5 Crop sell amount fact ors (kg/capita/m onth)......................................................................... 108 5-6 Marketing cost factors..........................................................................................................110 5-7 Transport infrastructure index............................................................................................ ..111 5-8 Base preferential consumption basket.................................................................................. 114 5-9 Descriptions of one se rving for each food type. ...................................................................114 5-10 Maximum food group expend itures by geographic region. ............................................... 115 5-11 Food type expenditure factor..............................................................................................116

PAGE 8

8 LIST OF FIGURES Figure page 2-1 Iterative collaborative cycles in a tools developm ent process .............................................. 31 4-1 Rich picture of the key them es of Jamaican food security..................................................... 48 4-2 Food system activities and actors......................................................................................... ..51 4-3 Depiction of QnD:Jamaicas hierar chical object oriented structure. ....................................... 53 4-4 Depiction of Local Co m ponents descriptive data................................................................... 53 4-5 Primary income source of small farms by parish. ................................................................. 55 4-6 Landownership of rural economic units by parish. ............................................................... 57 4-7 Small farm land use in term s of crop type.............................................................................. 58 5-1 Land use structure in QnD:Jamaica.........................................................................................75 5-2 FAO cultivated area by crop group........................................................................................ 76 5-3 MOA cultivated area by crop group....................................................................................... 76 5-4 Estimated land areas that act as drivers in Q nD:Jamaica.......................................................78 5-5 Vegetables productivity coefficient relationship .................................................................... 81 5-6 Vegetables productiv ity calibration by date. .......................................................................... 83 5-7 Vegetables productivity cal ibration difference from known..................................................83 5-8 Vegetables volume calibration by date................................................................................... 84 5-9 Vegetables volume calibration difference from known......................................................... 84 5-10 Root crops productivity calibration by date......................................................................... 85 5-11 Root crops productivity cal ibration difference from known................................................85 5-12 Root crops volume calibration by date................................................................................. 85 5-13 Root crops volume calib ration difference from known........................................................ 86 5-14 Legumes productivity calibration by date............................................................................ 86 5-15 Legumes productivity calibra tion difference from known...................................................87

PAGE 9

9 5-16 Legumes volume calibration by date.................................................................................... 87 5-17 Legumes volume calibra tion difference from known. .......................................................... 87 5-18 Livestock production relationship........................................................................................88 5-19 Livestock production calibration by date.............................................................................89 5-20 Livestock production calibra tion difference from known.................................................... 89 5-21 Vegetables productivity testing by date................................................................................90 5-22 Vegetables productivity te sting difference from known......................................................90 5-23 Vegetables volume testing by date....................................................................................... 90 5-24 Vegetables volume testing difference from known..............................................................91 5-25 Root crops producti vity testing by date. ............................................................................... 91 5-26 Root crops productivity testing difference from known....................................................... 92 5-27 Root crops volume testing by date....................................................................................... 92 5-28 Root crops volume test ing difference from known.............................................................. 92 5-29 Legumes productivity testing by date................................................................................... 93 5-30 Legumes productivity tes ting difference from known......................................................... 93 5-31 Legumes volume testing by date.......................................................................................... 94 5-32 Livestock production testing by date....................................................................................94 5-33 Livestock production testing difference from known...........................................................95 5-34 National vegetables farm gate price relationship.................................................................. 97 5-35 National root crops farmgate price relationship................................................................... 97 5-36 National legumes farmgate price relationship..................................................................... 98 5-37 Vegetable price calibration by date...................................................................................... 98 5-38 Vegetable price calibration difference from known.............................................................99 5-39 Root crops price calibration by date..................................................................................... 99 5-40 Root crops price calibration difference from known..........................................................100

PAGE 10

10 5-41 Legumes price calibration by date...................................................................................... 100 5-42 Legumes price calibration difference from known............................................................. 100 5-43 National livestoc k price relationship. ................................................................................. 101 5-44 Livestock price calibration by date.....................................................................................101 5-45 Livestock price calibration difference from known........................................................... 102 5-46 Vegetable price testing by date.......................................................................................... .102 5-47 Vegetable price testi ng difference from known.................................................................103 5-48 Change in known vegetables price index vs. change in sim ulated nominal vegetables price..................................................................................................................................104 5-49 Root crops price testing by date.........................................................................................104 5-50 Root crops price tes ting difference from known................................................................ 104 5-51 Root crops simulated volum e and real prices..................................................................... 105 5-52 Legumes price testing by date............................................................................................ 105 5-53 Livestock price testing by date...........................................................................................106 5-54 Livestock price testi ng difference from known..................................................................106 5-55 Percent of production sold.............................................................................................. ...107 5-56 Maximum and minimum rural carbohydr ate price ratios across parishes. ......................... 119 5-57 Simulated vegetable purchases........................................................................................... 119 5-58 Imported staples purchased................................................................................................120 6-1 Historical scenario of domes tic livestock production and dem and...................................... 124 6-2 Scenario 1 of domestic liv estock production and dem and...................................................124 6-3 Scenario 2 of domestic liv estock production and dem and...................................................125 6-4 Scenario 3 of domestic liv estock production and dem and...................................................125 6-5 Regional price shocks to domestic livestock in historical scenario...................................... 126 6-6 Internal structural design noise.............................................................................................127

PAGE 11

11 6-7 Domestic and imported livestock purchas es o f Saint James farmers in historical scenario............................................................................................................................128 6-8 Total imported animal meat................................................................................................ ..128

PAGE 12

12 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EXPLORATION OF FOOD SECURITY VIA A MESO SCALE MODELING APPROACH: JAMAICA AS AN ILLUSTRATIVE CASE By Lauren Colleen Justice December 2007 Chair: Gregory A. Kiker Major: Interdisciplinary Ecology The purpose of my study is to examine the in teractions within a food system and between that system and its socioecological environmen t in order to reveal management or policy strategies that strengthen food security conditions. Using Jamaica as a case study my interdisciplinary project developed and tested a computer model, QnD:Jamaica, that fuses technology with lessons from complex systems theo ry in order to promote adaptive management of food systems. An iterative collaborative design process based on soft systems approaches was employed during model design. The problem situ ation as understood by key stakeholders was investigated through informal strategic dialogu es. These qualitativ e descriptions were conceptualized by a formal framework to identify a limited number of quantifiable parameters and drivers of interest for the scenario m odel. The model incorporates ecosystem and sociopolitical issues into a use r-friendly framework to analyze th e spatially exp licit dynamics of food, money and people in terms of their effects on food secur ity conditions. Illustrative scenario simulations were performed and the impli cations for systemic features were explored. The results of my study provide insight into conceptualizing f ood systems in general and the

PAGE 13

13 Jamaican food system specifically. My thesis serv es as an explicit exampl e of integrated soft research methods and analytical tool development and will guide future modeling efforts.

PAGE 14

14 CHAPTER 1 INTRODUCTION This modern mind sees only half of the ho rse-the half which may become a dynamo, or an automobile, or any other horse powered machine. If this mind had much respect for the fulldimensioned, grass eating horse, it would never have invented the engine which represents only half of him. Allen Tate, in Berry, 2002 de IMF a de tiefingest ting-dem lend you mon ey fi buy tin mackerel, but not fi help small farmers plant. Every country where dem go it mash up. Small farmer in St. Mary, Jamaica, 2001(Weiss, 2004a) In our current global food system, most countri es are dependent on trad e with distant agroecosystems for their food supply. While internati onal trade is beneficial for many societies and 100% reliance on local food is neither practical nor desi rable, it is important to recognize and manage the trade offs that occur between lo cal and external food s upply dependence (Sundkvist et al., 2005; Lyson and Green, 1999; Costanza et al., 1995). Food syst ems by their nature embody ecological, social and economic issues and the global change s taking place both climatically, culturally and economically are having profound effects on how food is produced transported, understood, valued, pr epared, and consumed. Food s ecurity offers itself as an interdisciplinary set of criteria for exploring the complex issues involved in food systems. Food security encompasses all of the mechanisms that allow individuals to maintain command over their food consumption (Sen, 1981; Corral et al., 2000). This definition allows food security to be viewed in its broadest sense as a rich pi cture of societal well-b eing and natural resource viability. Dynamic Collaborative Models for Food System Exploration The direct and indirect synerg istic relationships within th e process of food procurement from field to fork are highly unknown and unpred ictable and make exploration of food system vulnerabilities a complex task. New tools fo r assessing, planning and communicating about food security should embrace this uncertainty and aim to facilitate cr eative collaborative exploration

PAGE 15

15 of the system and its contextual environment (D. H. Walker, 2002). Me thods that investigate patterns between interactions and emergent properties in problematic situations promise insight into improved management strategies (Kasperson and Kasperson, 2001; Walker et al., 2004). Adaptive management strategies were first developed from the recognition that systemic socioeconomic and ecological knowledge is and always will be imperfect and incomplete (Berkes et al., 2003; Dyball et al., 2005; Holli ng and Gunderson, 2002). Unforeseen drivers, interactions and outcomes are vi rtually guaranteed. Rigid policie s prevent adjustment to these changing situations and can cr eate conditions where unsustain able and highly deleterious behaviors continue. Adaptive collaborative te chniques offer a methodol ogical structure with flexibility and space to explore, evaluate and sel ect the most beneficial policies as unpredicted issues arise. This structure is iterative in nature and includes cycles of policy experimentation, reflection, evaluation and modifi cation (Keen and Mahanty, 2005). Many have suggested scenario modeling as a tool that could aid exploration of possibilities and encourage creative thinking and collaborati on amongst stakeholders (Walker et al. 2002; Carpenter et al., 1999; Janssen et al., 2006). Scen ario-based models can provide analyses of trade-offs between socioeconomic and environmen tal goals. Scenario modeling does not aim to predict the future but rather to explore the poten tial ramifications of a range of policy options (Baker et al., 1999). The proposed research project is one of the fi rst national food security analyses to employ a simulation model nested in a collaborative desi gn process in order to re veal patterns in a food systems socioecological interactions (Pothukuchi and Kaufman, 2000). This approach is based on the adaptive management argument that the emergent properties which characterize complex systems, food systems included, are so dynamic that they currently can not be managed statically

PAGE 16

16 or for single indices. Rather, management requires planning tools that envision multiple relationships at once and a range of po ssible states of these relationships. Using Jamaica as a case study, this interdis ciplinary project developed and tested an exploratory computer model that fuses technology with lessons fr om complex systems theory in order to promote adaptive management of food sy stems. This model, QnD:Jamaica, uses the Questions and Decisions (QnD) modeling system th at allows simulation of biological and abiotic components of a system (Kiker and Linkov, 2006). QnD is also appropriate to this project because of it facilitates collaborative discussi ons through its easily manipulated structure and game-like format. This thesis takes the appr oach that when a broad definition of food security which includes a rich picture of societal well being is applied to food system management, the most appropriate policy options will consider diverse socioecol ogical dynamics together at the meso scale. Jamaica as a Case Study Like many Caribbean nations, Jamaica ha s a small open economy remnant of the plantation system that is vulnerable to major changes in the world ec onomy (Beckford, 1972). Throughout the 1980s and 1990s the dominate prescrip tions of international financial institutions revolved around the theory that fr ee markets will increase food security by ensuring cheaper and more consistent food supplies (Weiss, 2004a). However, Jamaica is now one of the most indebted nations, both in terms of financial and mate rial imbalances. Historically driven by cash crop exportation, two main symptoms indicate negative shifts in the Jamaican food system: increasing loss of market for exports as well as large-scale importation of cheap food leading to Jamaican farmers inability to compete in their own domestic market (FAO, 2003). As Jamaicas Minister of Finance has conceded, the country is now operating in crisis mode (Davies, 2000 in Weiss, 2004a). Assump tions about Jamaicas co mparative advantages

PAGE 17

17 in certain sectors come into question with the pr esent and potential international trade shifts and global climate changes. The Jamaican food sy stem needs to be reevaluated with a crossdisciplinary multi-scalar lens that captures the interplay between bottom up and top down features present in a small open economy. Objectives The purpose of this study is to explore the interactions within Jamaicas food system and between that system and its socioecological environment in order to reveal management strategies that strengthen food security cond itions. Of concern are the implications for agricultural production, income opportunities and nutritiona l consumption patterns. Accordingly this research has six objectives: Conduct a literature and data revi ew of Jamaican food security Use soft systems analysis to survey in terests involved in problematic situation Translate the interview results, literature c oncepts and available da ta into a functioning simulation model Test the model against selected historic conditions Explore plausible scenario simulations to enrich understanding of food system links Provide insights into future model development and use

PAGE 18

18 CHAPTER 2 PROBLEM BACKGROUND This section reviews complex systems theory and the parallel development of softer approaches for exploration into complex system s that seek to generate relevant, timely and adopted knowledge and its asso ciated technologies and policie s. A discussion of complex systems theory as applied specifically to the ma nagement of food systems follows. Finally, the evolution of decision support technology, specifica lly dynamic modeling, as it seeks to apply the complex systems paradigm is reviewed. Hard and Soft Methods for Understanding Complex Systems The contemporary research community is no t only internally fragmented but also increasingly distant from the institutions at large who appl y research findings (Pidd, 2004; Douthwaite et al., 2000; D. H. Walker, 2002). This distance inhibits the efficiency and practicality of the application of research findings (Ibid.). Misunderstandings, distrust, nonadoption or conflict generally ch aracterize the ineffective knowle dge transfer between research and implementation communities around the world (Ibid.). The interdisciplinary movement has intensified the discussion of how a particular research strategy can either assist or inhibit the eventu al uptake and practical application of newly acquired information by the community at large (Douthwaite et al., 2000; D. H. Walker, 2002; Kiker et al., 2003). This discus sion offers new research paradigms that aim to decrease the inefficiencies that plague the integration of the research, development, and implementation communities involved in socioecological system s management (Roux et al., 2006; Gunderson et al., 1995). The combination of soft and hard methods in research approaches is being explored as one way to bridge divides and inef ficiencies by providing both decision and process support throughout a project (Eden a nd Ackermannn, 2004).

PAGE 19

19 Wicked Problems Scholars of ecology as well as engineering and management sc ience agree that large scale management of socioecological systems has been largely unsuccessful at maintaining resilient communities (Holling, 1996; Costanza et al., 1995) This failure can be attributed to shortcomings within both (1) th e classical reductionist scient ific process and (2) societys institutional capacity to transl ate scientific knowledge into re al change (Simon, 1962; Forrester, 1961, Ackoff, 1974; Meadows and Robinson; 1985). Both of these problematic issues are intrinsically linked by the nature of the managed situations as wicked problems or messes (Rittel and Weber, 1973; Ackoff, 1974). Acco rding to Conklin and Weill a wicked problem exhibits the following (Pidd, 2004): The problem is ever evolving and cannot be confined to a single definition. Because there is no definitive Problem, there is no definitive Solution There are many stakeholders involved, making the problem solving process fundamentally social. The physical and social constraints on the solution change over time Complex Systems Theory Attempts to deal with these problematic situations in an integrated manner have been driving a core paradigm shift across manage ment communities, from natural resource to operations management for the past fifty year s (Folke, 2006). The new paradigm, referred to here as complex systems, developed from genera l systems theory. It is a reaction against the classical reductionist approach to science and management that assumes that complex phenomena can be broken down into smaller uni ts that can be ma naged independently (Churchman, 1979; von Bertalanffy, 1968; Ackoff, 1974; Simon, 1962; Forrester, 1968). Rather, general systems theory views the whol e as greater than the sum of its pa rts. This shift is from a

PAGE 20

20 bounded rationality, positivist, determinist, or hard approach to a soft or procedural rationality framework for knowledge formula tion and application (Simon, 1962; Checkland, 1979; Douthwaite et al., 2001; Lyons, 2004). Since the conception of complex systems theor y, researchers have struggled to develop a systematic strategy that can address both the biophysical and the human aspects of complex systems. Obviously there are fundamental diffe rences between each type of system; biophysical systems are inherently spatial and transfers occu r in material and energy terms, social systems are often non-spatia l and transfer knowledge and informati on. Some of the early attempts at formal integrated analysis gui delines focused on both systems hierarchical structure (Simon, 1962; van Bertallanffy, 1968). Forrester and H.T. Odum are two of the majo r contributors to complex systems theory (Pavoa-Zuckerman, 2000). Both scientists deve loped symbolic languages to represent system components and dynamics in order to facilitate modeling of the components in complex systems (Ibid.). Forrester, an engineer and computer sc ientist, combined ideas from cybernetics, control engineering, and organizational theory to de velop a guiding philosophy for simulating complex, non-linear, multi-loop feedback systems initially app lied to the management of industrial firms (1961). Odum, an ecologist drawing from Forrest er, used similar theories to explore humans within natural systems (1983). The complex systems framework assumes that a ll materials and information are transferred through positive and negative synergis tic couplings that produce emer gent effects such as delays and thresholds (Forrester, 1961). Additionally, complex systems are understood as open systems that interact freely with their environment, resulting in disequilibrium and an unbounded set of interactions (Holling, 1986). De spite the fact that complex systems theory does not have an

PAGE 21

21 agreed upon method for inquiry, it still provides the most helpful perspective to address management of messy issues (van der leeuw and Aschan-Leygonie, 2005). Two Social Facets of Complex Systems Management The complex systems framework has strong im plications for the types of tools and techniques used to explore problematic situati ons. For example, ecology scholars are working to develop a common currency between natural and social interactions in order create a language that allows for more holistic assessments (Folke 2006). Similarly, social scientists have been investigating the de velopment of new soft met hods of inquiry as well as the possible integration of these methods with classica l hard methods in order to uncover and assess complex problems (Pidd, 2004). The complex systems framework acknowledges that not only are problematic socioecological situations integrated by nature, bu t also the political and institutional capacity of each domain is highly linked to the others even if not intentionally (Schipper and Pelling, 2006). These two types of integration re veal two social facets of comp lex systems management arising from: (1) The internal inherent features of socioecological systems (2) the social institutional structure that exists to facilitate exploration and management of socioecological systems. Complex systems inherent social features Prior to the mid 1990s most ecosystem management studies were framed exclusively in natural or human terms (the former usually occurring in biology or ecology, the latter in economics) (van der leeuw and Aschan-Leygoni e, 2005). However, following increasing evidence that neither the social nor the natural system is paramount, the focus shifted to the interactions between the equal but heterogeneous components of what became termed as socioecological systems (Gunderson et al., 1995 ; Holling and Gunderson, 2002; Berkes et al., 2003; Walker et al., 2004). This progression from restrictive to ho listic understand ings has been

PAGE 22

22 described as an evolution of th e role of humans in socioecolo gical relations moving from reactive prior to the 1980s, through pr o-active during the 1990s, and resi ding currently as an interactive understanding (van der leeuw and Aschan-Leygonie, 2005). Hollings work in ecological re siliency has fueled much of the methodological theory for integrating social and natural systems studies (Holling, 1973; Folke, 2006). The importance of temporal and spatial scales of analysis and the m echanisms that transcend these levels is at the core of resilience concepts. Holling posits that only two criteria must be met for resilience concepts to be applied to st udies beyond ecology: (1) the systems need to be described in dynamic terms, and (2) the systems must have mo re than one potential state of equilibrium (1985). While these developments supply concepts for research into complex systems, they do not sufficiently address the second social facet of complex systems management. This additional side to the management problem rests in the process support needed to guide the institutional machine performing and a pplying the research. Managing the managers Management of issues in complex socioecological systems (i.e. food security) involves not only a conceptualized systems variables and their in teractions but also the issues that arise from stakeholders dynamic perceptions of the situatio n. Communities must st rive to navigate the logistical social aspects of managing a problematic situat ion by being conscious of and competent at dealing with, the differences betw een individuals (or collective interests) that comprise the group with respect to a host of issues surrounding the situation (Bawden, 1995 in Ison et al., 1997, 266). Research in the social management disciplines has used comp lex systems theory to inform new methods of inquiry that tran scend the reductionist model of research in order to offer

PAGE 23

23 process as well as decision support (Eden a nd Ackermannn, 2004; Pidd, 2004). Rather than attempting to discover, quantify, predict and cont rol the situations that arise from complex systems by way of hard methods, soft frameworks offer flexible options for problem exploration and management. As Ison et al. concisely su mmarized, A given problem does not occur in isolation or exist objectively-out there. It is socially constructed on the basis of different perspectives and is part of an interrelated network of problem s (1997, 260). In accordance with the notion of wicked problems, preference for a particular strategy to address such a problem thus reflects a certain situational understand ing. Obviously this understanding will differ and change between individuals, communities, and di sciplines making it difficult to agree on a single strategy of action. However, using the existence of multiple perspectives as the ideology itself, as the complex systems paradigm assumes, provides for a system of analysis that prepares for mutual understanding and negotia tion of the problem(s) in question (Ison et al., 1997, 260). Collaborative modeling is one type of decision tool that can apply the complex systems framework to support research within a holis tic socioecological pr oblem structure while simultaneously alleviating the inefficiencies that arise from the social nature of problem solving. Food Systems as Complex Systems Food systems can be understood as one type of complex system where systemic vulnerability is embodied in temporary conditions of food insecurity. These potential conditions emerge from a spectrum of possible interactions between internal and external socioeconomic components of the food system. Like all complex systems, food systems behave synergistically. Actors at all levels of the food system must deal with both changing ecological and socioeconomic conditions. Additionally these actors create change leading to feed back within the system both to themselv es and to actors at levels above and below them.

PAGE 24

24 Food security issues occupy a unique meso s cale location in the planning field because food and agriculture possess distinctive qualities re garding quality of life, cultural decisions and environmental and land use effects. Recognizi ng that poverty and hunger occur at the household level as family decisions often determine how lim ited resources are utilized much recent effort has been put into micro level household scale st udies to determine how communities deal with limited resources and how shifts in access effici encies or practices would change resource use and quality of life. Yet macro scale i ndices remain the most common measure of community/national health in the pla nning arena (Pothukuchi and Kaufman, 2000). In order to evaluate food security the linkage s or pressure points be tween the field and the fork must be holistically assessed (Erick sen, 2006a and 2006b). Incr easing food security requires asking the following questions: Who are th e actors being relied on at all levels from production to consumption of the food system? How does reliance on th ose actors affect a populations interests? How could each actor tran sform or change? And what are the potential adaptations/coping strategies of a population in th e face of the actors potential transformations? Agricultural and Food Policy Decision Support Traditional agricultural resear ch and food policy has failed to produce a global food system with adequate levels of food s ecurity, leading the food and agricu ltural research community, like other socioecological sectors, to move away fr om methods and institutio ns based on reductionist frameworks (Fraser et al., 2005; Sundkvist et al., 2005; OBr ien and Leichenko, 2000; Costanza et al., 1995; Meadows, 1985). With the advent of sustainable developmen t principles, a gradual shift is taking place whereby issues of agricult ure are no longer analyzed separately but are integrated into the assessment of the entire food system in term s of the roles it plays in rural welfare, environmental effects and consumer welfare (Eakin, 2005; Andreatta, 1998; Ford, 1992). In this context, the biological and so cioeconomic transmission mechanisms that govern

PAGE 25

25 the availability and acce ssibility of food can be analyzed within a coherent framework (Fraser et al., 2005). Twenty five years ago, Amartya Sen introduced new ideas to address poverty and food security. He moved the concept of food accessib ility to the forefront of the food security discussion (1980). Twenty years ago, Donella Meadows, a systems an alyst, described the chronically inadequate food systems that fail ed to support much of the worlds population despite the existence of thousands of detailed formal analyses (1985). She made the crucial insight that this malfunctioning was due to posing questions that were not being applied to the total system and that in their underlying assu mptions and limitations cannot lead to the answers one really wants to find (1985, 161). As an example of a proper perspective, Meadows cites a Dutch international food system simu lation model called MOIRA that integrated environmental and socioeconomic change in 1979! However movement since then towards the posing of the types of holistic questions that could lead to better analys es and management of food systems has been slow. Despite these early promises of success, agricultural resear ch and food policy communities largely continue to manage by way of limited si ngle goals such as aggregate calorie per capita availability and overall crop producti vity. Just as Holling showed that the use of static target variables in ecosystem management leads to pa thologically vulnerable ec osystems, scholars of food systems have begun to show how the failure to use integrated social and environmental assessments have increased food insecurity (1978, OBrien and Leichenko, 2000; Schipper and Pelling, 2006). Dynamic modeling is a potential tool for more in tegrated assessments of food systems.

PAGE 26

26 Modeling Complex Systems for Decision Support The systems engineering community that deve lops technologies to bridge the divide between research and po licy institutions has been traversi ng a paradigm shift parallel to ecological and social research scholars (Pi dd, 2004; Lyons, 2004; Eden and Ackermannn, 2004). As the complexity of problematic situations under investigation is increasingly recognized, technologys role in guiding formulation of strate gies for situational improvement is changing. For example, a feature of complex systems is th eir unpredictability, tools that claim to make predictions become irrelevant (Lyons, 2004). In the same vein, the increasing recognition of the importance to account for competing versions of reality or worldviews in order to create knowledge that is relevant and adopted, necessita tes technology that aids in formulation of cohesive community problem defi nition or at least increases unde rstanding between the various stakeholders of different perspectives (Ede n and Ackermannn, 2004; D outhwaite et al., 2001). Decision Support Paradigm Shift Widespread disenchantment ch aracterizes the general sentiment towards decision support systems, (DSS)1 (D.H. Walker, 2002). Research reveals non-delivery, irrelevance, inflexibility, inaccessibility, lack of confiden ce, institutional and political boundaries, and negative impacts on decision making processes as reasons for d ecision support systems limited adoption and perceived minimal impact on decision making pr ocesses (D. H. Walker, 2002; Cox, 1996; Davis, 1989; Meadows and Robinson, 1985). These findings are not surprising as the classical view of manager is one of impartial controller (Stacey, 2001; Lyons, 2004). The new ma nagerial paradigm emphasizes flexibility 1 Although a decision support system is traditionally understo od as a probabilistic computer tool that aims to reveal optimized solutions to decision makers and predict the impact of a given policy on a single variable, DSS is understood in this thesis in its broadest sense as an overarching situational strategy that aids in decision making through multiple flexible tools and methods as defined in D. H. Walker (2002).

PAGE 27

27 rather than control as well as the need to develo p approaches and tools wh ich are relevant in the wider decision-making context (D. H. Walker, 20 02). The DSSs must not only aim to deepen our understanding of a particular systems comp onents and dynamics but should also serve as devices that bring cohesion to a group of diversified stakehol der groups. A key feature of a particular DSS should be to broaden understandings of other interests involved. Its functions should include exploration of how others view the problem situation, fo llowed by facilitating initial agreement upon problem situation. However, it should contain provisions whereby this initial vision can be modified as viewpoints or interests change or increased technical knowledge surrounding the problem is introduced. Often DSS tools are considered successful even when they are not widely employed because they facil itate learning between developers and managers (D. H. Walker, 2002). In fact, DSS that fail to f acilitate experien tial learning have limited utility (Ison et al., 1997). Modeling Certain modeling strategies can transcend the traditional difficulties associated with DSS relatively easily. Models inherent qualities as representations of perceived reality allows them to serve as transitional object s that help people see their own and others mental models (Morecroft, 2004). Not only can a model provide insights in to the system under question but the very creation and design proces s can serve as tools to learn a bout the system as well as the managerial stakeholders invol ved (Walker et al., 2002). It is important to note that there is no right model of a system, just more and less useful ones (Janssen et al., 2006). Because models are merely simplifications of the real system they cannot test hypotheses they can onl y describe the reality the modele r wishes to describe (Levins, 1966 in Pavoa-Zuckerman, 2000). While they are not able to prove or disprove any theory, a models greatest utility li es in its ability to serve as a learning tool.

PAGE 28

28 Models are useful when integrating multiple disciplinary concepts because models serve as conceptual tools in the formulation of theo ries, helping one to visualize the patterns and processes of the ecosystems being represented (Pavoa-Zuckerman, 2000, 36). Models allow us to transcend our restrictive understanding of wh at is possible and help us expand our linear expectations to include all th e possible futures we may encounter (Farrell, 1998 in Lyons, 2004, 39). Model exercises can reveal trends in the ex pected importance of certain system features for implicating resiliency to certain events (Janssen et al., 2006). These models can also reveal systematic processes for collecting future da ta about socioecological systems (Ibid.). Models should attempt to identify thresholds, their nature, and what determines their positions along the driving variables (Walker et al., 2002). They should be simple, multi-scalar and temporal, and scenario based (Ibid.). They should 1) Act as tools of negotiation, 2) Create/unearth choice, and 3) Define a context for trade offs (Schrage, 2000). Developers of models should be aware of how st akeholder interests a nd assumptions are integrated with system features before they can adequa tely develop a tool for systema tic learning about a problematic situation and potential improvement options (Bunch, 2003; Zhu and Dale, 2000). This larger scene in which model developm ent is nested is eloquently described by Meadows and Robinson, Modelers are human beings with human strengths and failures. To the extent that their work can be structured to bri ng out their strengths and correct for their failures, they can contribute immensely to human understanding of how the world works, to social decision-making, and to the shaping of an un certain future (1985, 15). One such modeling structure is the explicit employment of soft methods consisting of iterative cycles of inquiry and reflection that revolve around how the stakeholders seeking to learn about and improve a problematic situation understand the issu es involved (van der Heijden, 1996).

PAGE 29

29 Hard and Soft Methods for Model Development Multiple frameworks exist to guide systems engineers towards creating tools of greater relevance and impact (Parker et al., 2002; D. H. Walker, 2002; van I ttersum et al., 1998; Meadows and Robinson, 1985). These all stress the importance of iterative flexible development, transparency, integration of scales, issues and stakeholders, qualitative as well as quantitative concepts and comm unication facilitation (Ibid.). Most importantly these frameworks emphasize the need for continual mon itoring and investigation into the process of technology development in order to move towards best practice guidelines. Soft systems methods Checkland and Holwell make the astute distinction between hard and soft approaches and hard and soft skills (2004). While a soft approa ch assumes an infinite number of possibilities, a hard approach assumes only one. In this sens e a soft perspective is the only appropriate approach for wicked problems. Soft approaches, however, can accommodat e hard skills within them. When assessing soft vs. hard skills, such as qualitative vs quantitative research techniques, there is the possibili ty of fluid and equal integrati on between soft and hard (Ibid.; Pidd, 2004; Eden and Ackermann, 2004). Eden a nd Ackermann propose the use of combined soft and hard methods in order to foster both de cision and process support (2004). This type of approach provides a space where group and indivi dual interests can be ex plored simultaneously just as a forest can be seen as a whol e without losing site of the trees. In traditional system manageme nt theory the objectives of a given system were assumed obvious (Checkland, 1979). If the management object ives were not met, it was believed to be a result of system inefficiency, and planners set out to engineer solutions. However, as this reductionist thinking failed to solve the increas ing problems occurring between humans and their environment (Capra, 1996), managers began to wonder if they should not only be asking, how

PAGE 30

30 should the system work?, but also what should the system provide? And why? (Checkland & Scholes, 1990). The answers to these questions ar e never right and alwa ys contain degrees of inherent uncertainty (van der Heijden, 1996). Soft systems met hods such as scenario planning (van der Heijden, 1996) and the soft syst ems methodology, SSM (Checkland, 1979), directly address the existence of multiple worldviews and provide ways to engage with them (Checkland, 1999). The systems modeling method nested in a softer methods approach offers potential to overcome traditional DSS limitations. (Bunch, 2003; Zhu and Dale, 2000) Iterative cycles of tool development Douthwaite et al. characterize traditiona l technology development as positivist-realism associated with a uni-directional information fl ow from the engineer or knowledge possessor to the user (2001). This approach is built on the hard science paradigm that assumes an independent reality (Ibid., 2) th at can be experimented upon to produce quantifiable results. Douthwaite has developed an iter ative tool development process st ructured in cycles of learning rather than uni-direction information flow (Ibi d.). This approach is known as the Follow the Technology approach as opposed to the Tran sfer of Technology approach employed in traditional tool development (Ibid.). At any step during tool deve lopment new information could come to light, or forward movements become stalled, and revisiting a prev ious step becomes necessary. In order to explicitly account for these in evitable situations, this approa ch utilizes ongoing cyclical exploration and analysis. These cycles of innovation are best de scribed in Figure 2-1 based on Douthwaites tool fitness diagram.

PAGE 31

31 Figure 2-1 Iterative collaborative cy cles in a tools development pr ocess (after Douthwaite et al., 2001). Cross Disciplinary Convergence and Systems Modeling It is interesting to note the convergence of the ideas contained in the frameworks from technology engineering with those from other disciplines that al so seek to address complex system management. For instance in the field of vulnerability analysis researchers are moving away from black box analytical models of pr ocess and outcome that do not allow for the exploration of the dynamic and in tegrative nature of risk (Adg er, 2006). The idea of continual evaluation of processes stressed above forms th e backbone to the entire theory of ecological adaptive management (Holling, 1978; Kiker et al., 2003). Researchers of resilience theory, mostly ecologists by training, have illuminated f our steps for complex system assessment that mirror the frameworks suggested by the engineeri ng community. These include 1) development of a conceptual model; 2) Identification of possible future scenarios: 3) Exploration of interactions; 4) Stakeholder eval uation. (Walker et al., 2002). The similarities between frameworks across di sciplines have positive implications for the continued and improved cooperation necessary in complex systems management. Likewise Suitability of Tool/Model Time Additional Strategic Dialogue Cycles Discussion Initial cycles: Creating a rich picture to envision model parameters Suitability of Model Model developers best guessversion Later cycles: Refining model detailsTimeInput from developers knowledge is minimal Input from collaborator knowledge Additional Strategic Dialogue Cycles Field Discussions Conceptualize situation Refine Design & simulations discussion Initial cycle: Creating a rich picture to envision model parametersFinal version Suitability of Tool/Model Time Additional Strategic Dialogue Cycles Discussion Initial cycles: Creating a rich picture to envision model parameters Suitability of Model Model developers best guessversion Later cycles: Refining model detailsTimeInput from developers knowledge is minimal Input from collaborator knowledge Additional Strategic Dialogue Cycles Field Discussions Conceptualize situation Refine Design & simulations discussion Initial cycle: Creating a rich picture to envision model parametersFinal version Suitability of Tool/Model Time Additional Strategic Dialogue Cycles Discussion Initial cycles: Creating a rich picture to envision model parameters Suitability of Model Model developers best guessversion Later cycles: Refining model detailsTimeInput from developers knowledge is minimal Input from collaborator knowledge Input from collaborator knowledge Additional Strategic Dialogue Cycles Field Discussions Conceptualize situation Refine Design & simulations discussion Initial cycle: Creating a rich picture to envision model parametersFinal version

PAGE 32

32 agreement within the DSS community as well as across many other disciplines about the appropriateness of modeling nested in collaborative methods for revealing concepts in complex systems promises enhanced technological deci sion support experiences (Walker et al. 2002; Carpenter et al., 1999; Janssen et al.,. 2006; van der Leeuw and Aschan-Leygonie, 2005).

PAGE 33

33 CHAPTER 3 AN APPROACH TO FOOD SYSTEM MODELING Food System Models A number of model-based analyses have been performed to assess socioecological interactions involved in f ood production and consumption (Y oung and Kantor, 1999; Duchin, 2005). Duchin, developer of a global food system analytical model, The World Trade Model, describes how public food security is intimately tied to nutritional, environmental and economic policy, yet all have failed it (2005 ). Nutritional and environmen tal educational campaigns often have less than desired impact while political elements face competing forces of public and private interests. She stresses that in this tension, the solu tion lies: the environmental and nutritional communities must work together to bridge the pub lic, governmental and corporate divides. Duchin suggests simula tion modeling as a way to organize the interests of these groups. She described the difficult resolution decisions that modelers everywhere often face during the design and development processes. For instan ce, in the case of designi ng a diet scenario she wondered if it should contain 30, 300 or 3000 food categories. She applied what natural ecologists have said about modeling complex envi ronmental systems to the creation of stylized models of human production and consumption activ ities, Greater abstraction, or sacrifice of detail, is necessary to descri be the workings of much la rger complex systemsCreating templates to describe alternative diets or produc tion strategies requires levels abstraction not generally required for representations base d only on direct measurement (2005, 110). An average quantitative description works because it is a stylized representation that succeeds when it captures the characteristics most important for the analysis (2005, 110). In her words, successful sets of variables requir e attending to the distinctive properties of different variables (2005, 110). This identification of the qualitative distinctive properties of variables allows one

PAGE 34

34 to tunnel to the appropriate inclusion or exclusion of systemic features and avoid unnecessary duplication.2 Soft systems methods along with the re view of empirical evidence guided the identification of the appropriate systemic features to represent the multiple interests of the Jamaican food system in the simulation model, QnD:Jamaica. Approach The systematic study of how interactions be tween external influence and structural mechanisms of a socioecological system affect emergent properties is a new approach for policy planning methods. More specifically, assessment of food systems as socioecological units is a new approach to food security management (Pothukuchi, and Kaufman 2000). The purpose of this study is to explore the inte ractions within Jamaicas food system and between that system and its socioecological environment. Of concern are the implications for food production, income opportunities and nutriti onal consumption across rural and urban population groups. In order to address these issues, a complex system s framework was employed. This framework was applied through a dynamic modeling exercise that employed soft system methods for model development. This research effort has two distinct objective phases: 1) Research Phase and 2) Design and Evaluation Phase shown in Table 3-1. Table 3-1 Project phases. I. Research phase II. De sign and evaluation phase Uncover key food security concerns as perceived by stakeholders Design and develop a computerbased scenario simulation model Formally translate these concerns to a parameters and drivers format Test models parameters through scenario simulations 2 However, as ecologists have recently noted, a feature that only appears to be a duplicate might be the one key that possesses unique quali ties that provide systemic resilience.

PAGE 35

35 Soft Systems Methods Soft systems approaches use iterative and co llaborative techniques to envision the driving forces which could move a system one way or another (van der Heijden 1996). These approaches such as scenario planning (van der Heijd en 1996) and the Soft Systems Methodology (SSM) (Checkland 1979, 1999), provide tools to tease out a systems pot ential pressure points and explore how possible interventions around these points might improve the problem situation (Checkland 1999). By allowing modele rs to actively engage with stakeholders assumptions and the inherent uncertainties in complex systems, soft systems approaches are conceptually compatible with complex socioecological system s frameworks, and provide a useful research methodology for interacting with st akeholders and guiding researcher s as they carefully seek to structure the problem situation. Th ese methods assist researchers to elicit perceptions of a real world situation and then organize the results in to mental models of key themes that provide starting points to explore the context of change within the system. A soft systems approach first requires expression of concerns about the system in a nonsystems manner. This is accomplished through inform al strategic dialogues (individually or in groups) with stakeholders comprised of a broad range of interests in the system (van der Heijden 1996; van der Heijden et al. 2002). To begin these collaborations Quinn and Voyer suggest asking, If there was one core question that unde rlies all the strategic challenges you face what would it be? (1996). van der Heijden (1996) suggests the following questions: What concerns you most about the future? What would you ask an Oracle? What would be a good/bad scenario? What would lead to a good/bad scenario? What critical decisions have to be made soon? What was learned from past surprises?

PAGE 36

36 The strategic dialogues that begin a soft systems inquiry into a problem situation differentiate the soft approach from traditional hard approaches which seek to narrow and define problem statements early on in a research process. A soft systems approach purposefully avoids narrowing the problem definition too soon, instead seeking to gather the many views, values, perceptions, and expectations that characteri ze a complex issue. Not only do these methods distill various interests, but they also reveal the multiple scales at which an issue can be framed. Churchman suggests that researchers sweep in a wide variety of stakeholders, both friends and enemies, in order to reveal the many assumpti ons about the nature of the problem and what would constitute improvement (1979). Interview results along with a ppropriate literature review findings are mapped in an ordered database that provides an overview of the key themes in stakeholder frameworks clustered according to their quali ties in the situation. In the first stage of mapping, emergent issues are unrestrictedly clustere d according to all of their potentia l interactions. Next, clusters are refined according to common and divergent world views and a ssumptions. The issues should also be clustered in terms of their properties of internal or external influence. The resulting thematic database can also be represented as a rich picture a graphical conceptualization of the situation. Transformations of key issues that would improve the problem situation are identified from the rich picture. A TWOACE analysis (described in de tail in chapter four) is applied in order to specify the world views that give this transformation m eaning, the actors involved, and the environmental constraints. QnD Modeling System The Questions and Decisions QnD, modeling system is a problem exploration object oriented tool that increases understanding of potential behaviors and management options of a particular socioecological system (Kiker et al ., 2006; Kiker and Linkov, 2006). Through a user-

PAGE 37

37 friendly graphical interf ace stakeholders can play their system by manipulating institutional and ecological components of interest. Result s differ from the combinations of various environmental and economic drivers and the playe rs response to them via management options implemented in different spatial areas and over di fferent time periods. As various scenarios are played, the interactions between institutional and ecological parameters are exposed and future possibilities of the system can be envisioned. The QnD system not only helps stakeholders deepen their understand ing of a particular systems components and dynamics but also acts as a device to bring cohe sion to a stakeholder community. Throughout its iterative design proc ess, a QnD system engages stakeholders to accomplish multiple tasks: facil itating initial agreement on key forces and themes, broadening understandings of others intere sts, developing scenarios and an alyzing various policy options. QnD was created as a technical tool that is complemented by and works in tandem with an iterative exploration process. The systems design process was intended to be compatible with several social science methodologies, such as the Soft Systems Methodolgy (SSM), scenario planning, or cognitive mapping. In order to apply the QnD modeling system to Jamaicas food security issues, quantif iable components, processes, and drivers were revealed through soft systems methods. Eight of the simulation processes available in the QnD modeling system were used in this research project. Table 3-2 describes th e operations performed by these processes. Table 3-2 QnD processes. Process name Operation performed PAddValue Input one is added to input two PSubtractValue Input two is subtracted from input one PMultiplyValue Input one is multiplied by input two PDivideValue Input one is divided by input two PSetValue Output is set to input PRelationship A two column table of values relates any

PAGE 38

38 number of pairs of points as independent and dependent variables. All points between defined pairs fall on a linear relationship. PIfGreaterThan If input two is greater than input one than perform following subprocess. If not greater jump to next main process. PTemporalRunningAverageValue The values for the input from the time step prior to any number of consecutive previous periods are averaged. The QnD modeling system was applied to the Jamaican food system by translating the results of a soft systems analysis into a conceptu al object format (described in detail in chapter four). This structure forms the backbone of Qn D:Jamaica, a functioning food system simulation model. The next section describes the specifi c methods employed during field work to build a rich picture that could inform model design. Informing Decision Support Tool Development An initial list of potential collaborators iden tified general stakeholde r groups according to their structural locati on within the system. A review of Jamaican food system literature suggested that tourism, small scale agricu lture, international tr ade, government and environmental interests are major subsystems influencing food securit y. A general list of participants included small farming communities, tourist industry actors (traditional and ecobased), informal and formal food distribution acto rs, agricultural, nutritional, and environmental NGOs, and Ministries involved in food system management. To this general list were added specific individuals that had proa ctively explored and/or particip ated directly in management issues related to Jamaican food security as evidenced in academic papers and governmental reports as well the leadership lists for organizat ions involved in food security. Where feasible, potential participants were cont acted via email to ascertain their availability in Jamaica during

PAGE 39

39 July 2006. As it was not possible to identify by name or contact many stakeholders from outside of the country, they were solicited in person once field work began. The first interviewees included three professors from the University of West Indies (UWIMona) and a researcher at the Caribbean Food and Nutrition Institute. During the first interviews specific individuals from the above general stakeholder categories were suggested as additional participants in a snowball sampling manner. These included a divisional director within the Ministry of Agriculture, a rural development fi eld officer, an ecotourism outfit owner, members of a parish-level farming cooperative, a livestock farmer, and a manager of a local supermarket. During the interviews, the problem situa tion as understood by the stakeholders was investigated through the scenar io planning approach (van de r Heijden, 1996) and the Soft Systems Methodology (Checkland, 1979, 1999). First, concerns about the system were expressed in a general manner. The collaborations began with one general question similar to the one suggested by Quinn and Voyer (1996): Wh at do you believe are the biggest challenges to Jamaican food security? The ensuing discu ssion loosely followed the question structure as proposed by van der Heijden (1996). The conve rsation was often supplemented by questions directly related to the interviewees field of expertise. The interviews were recorded and reviewed later to determine what and why key is sues were of concern a nd attributed to which stakeholders. In accordance with the scenario methods described above (van der Heijden, 1996) a thematic database was created that clustered the main issues affecti ng food security by their understood internal or external influence on the situation.

PAGE 40

40 CHAPTER 4 JAMAICAN FOOD SYSTEM: AN ILLUSTRATIVE CASE The pathways that couple production and c onsumption in small open economies were revealed in order to be represented in the food system model QnD:Jamaica. The actors and interactions involved in the Jamaican food sy stem were reviewed within the qualitative framework provided from the results of soft methods research performed in the field, and organized into conceptual groups that loosely followed the framework suggested by the research group GECAFS. The qualities examined were socioeconomic populations, land uses, food pricing, distribution and marke ting structures, and consumpti on issues as well as how each subject is embedded in the historical context. This systematic analysis structure provides an approach that can guide future meso scale explorations of other food systems and be used in distilling features the meso scal e of food systems in general. Within this framework, empirical data were reviewed for the period of 1986-2004 to develop quantitative descriptions for the objects that structure QnD:Jamaica. This period was chosen because data existed for occurrence of mu ltiple political, economic, and climatic events. These include internal structural adjustments made in accordance with International Monetary Fund (IMF) agreements such as exchange rate de regulation and market liberalization, hurricanes Gilbert and Ivan, additional known drought and flood events, and the September 11th attacks on the US, which affected international socioeconomic conditions. Historical Review The 1970s in Jamaica were characterized by a st rong national state that utilized extensive price controls, subsidies and natio nalization of primary industries to solve poverty issues. Prime Minister Michael Manleys Peopl es National Party (PNP) pursued a policy in the early 1970s of import substitution in efforts to strengthen th e Jamaican economy. However, by the late 1970s

PAGE 41

41 the country was heavily indebted to internationa l financial institutions and began a series of extreme economic reforms in the 1980s and 1990s such as devaluation of the Jamaican dollar, elimination of state marketing boa rds, trade liber alization, and privatizat ion of state and public owned industries. The Exchange Rate, Inflation and Food Imports Between 1981 and 1985, government support for agri culture was heavily withdrawn. This was due to overall decreases in government spendi ng as well as a decrease in the percentage of spending on agriculture from 19% in 1975 to less than 4 % by 1990 (Weiss, 2004a). In addition to reduced internal supports, increased input cost s, and the overall rise in the cost of living, Jamaican agriculture also suffered from the reform policies of AGRO-21. Designed with help from USAID, the policies of AGRO-21 aimed to free up labor from agriculture and make it availabl e to sectors of the economy in which Jamaica was believed to have a comparative advantage: mass tourism and manufacturing. However, as labor was shed without increases in mechanization or other productivity supports, agricultural production volumes began to fall (Witter, 1989 in Weiss, 2004a). Additionally, the new industries proved unable to support the extra labo r. During the late 1980s a nd early 1990s there was a mass outward migration of one in twelve Jamaican s (Anderson and Witter, 1994). This experience illustrates Weiss conclusion that in contrast with free market theory, the decline of one sector often produces serious dislocations without concurrent gains el sewhere (2004, 474). Jamaica clearly exemplifies this with the trade deficit quadrupli ng between 1980 and 1990 and then almost tripling again between 1990 and 2004. Between 1989 and 1992, monetary policy makers in Jamaica aimed to stabilize prices through macro economic tools (Robinson and R obinson, 1997; Planning Institute of Jamaica 1990-1994). Bound by IMF agreements made duri ng the late 1980s, the Jamaican government

PAGE 42

42 followed a policy of currency devalu ation that, in fact, actually increased the prices of consumer goods. Between 1990 and 1991 the Jamaican dollar depreciated against the American dollar by 60.9 % compared to 25.7% between 1989 and 1990 (Planning Institute of Jamaica, 1991). According to the financial transmission model put forth by Allen et al ., the exchange rate exhibits the most significant effects on consum er prices in Jamaica and is linked to the importance imported goods in the economy (2002). Currency devaluations were expected to in crease domestic foods competitiveness. The rising costs of imported production supplies, how ever, constrained the Jamaican farmer, and food prices rose (Planning Instit ute of Jamaica, 1992). As the pur chasing power of the Jamaican dollar plummeted between 1989 and 1992, consumer pr ices steadily rose increasing 268% point to point. Of the eight regul arly tracked household consumption groups, the food and drink category is weighted highest at 55.5% within the Jamaican CPI. Thus, food price increases generally have a large influence on the overall inflation rate. An emergency Food Assistance Program was in troduced in the early 1990s to prevent food riots over the high cost of food. This only served to strengthen the pattern of importation by subsidizing imported food staples rather than lo cal ones. The bias in consumption habits toward foreign foods was fueled by making it ra tional from the point of view of households maximizing nutrients per dollar to buy imported instead of local foods (Witter, 1989 in Weiss, 2004a, 468). The 1990s also saw reductions in trade barriers furthering the upward spiral in cheap imported food. Between 1991 and 2001 Jamaicas tota l food and beverage imports increased by two and a half times. The larg est increases were in cereals, feeds, and meat products which account for half of the value of Jamaicas agro-imports (Weiss, 2004a). While domestic poultry

PAGE 43

43 production grew, it was fueled by US subsidized imported feedstuffs. Also, imported livestock amounts increased more than twice as fast during the 1990s. Per capita meat consumption increased by 55% between 1980 and 2000 moving to the center of the diet where it was previously relatively rare (Weiss, 2003). Economic Development and Poverty Alleviation Records indicate a decrease in povert y, from 30.5% in 1989 to 16.9% in 1999, while unemployment remained fairly constant ar ound 15-17% (Osei, 2002; Le Franc and Downes, 2001; Salmon, 2002; King, 1998). This conundrum addresses the real drivers behind poverty reduction in the 1990s and supports the model that Jamaican consumption and production activities are highly affected by the external envi ronment. While Salmon puts some faith in the success of the governments National Poverty Er adication Program at limiting poverty, Osei, King, and Le Franc and Downes are more skepti cal about the governments role in increased consumption levels (Ibid.). R eal price decreases and real wage increases, neither caused by governmental policy, played the largest role in increased consumption levels (Ibid.). The informal sector, in terms of domestic and in ternational trade (both legal and illegal) and remittances from abroad, were the key factor s to reducing the numbers of extreme poor. Le Franc and Downes use the strong correlation between reduction in the poverty ratio as well as the inflation rate as evidence that falling prices have been the main driver of increased consumption (2001). They also make the importa nt point that price changes not only affect incomes purchasing power but also have substitu tion effects in which consumption patterns, and thus nutritional patterns, are changed. These conclusions agree with earlier findings that absolute poverty was reduced by a recovery in real wages through informal activ ities, not economic reforms (King, 1998). The informal sector grew from 14% to 29% of the labor force between 1972 and 1997, and accounted

PAGE 44

44 for 30% of the labor force in 1999 (36% of tota l employment) (Osei, 200 2). Decreases in unemployment have been shown to be the result of a decline in the number registered in the work force rather than an actual increase in employment (ECLAC, 2004). As Weiss describes, Jamaicas economy now rests on a precarious ba lance of external migration/remittances, tourism, drugs, and borrowing that is layered upon a notoriously high level of inequality (2004, 483). Climatic Shocks The Jamaican populations food security was also affected by climatic events. These natural disaster events affected food accessibility by increasing food costs as well as other consumption expenses in the form of househol d repairs and through the loss of jobs. If consumption expenditures return to pre-disast er levels before domestic production recovers, there can be a sharp increase in the relative importance of imported goods in the total consumption budget (Brown, 1994). For these re asons, the components and processes of the Jamaican food system must be evaluated with in the dynamic context of climatic shocks. Hurricane Gilbert Hurricane Gilbert passed dire ctly over the center of th e island on September 12, 1988. Damage from this category five hurricane was estimated at $US 4 billion, with the agricultural sector accounting for 40% of this amount (P AHO, 1988). Gilbert affected 810,000 people and damage amounted to 65% of GDP (ECLAC, 2004). Hurricanes have major economic effects on the balance of trade because of the increased expenditures on capital goods, most of which are imported, as well as the destruction of export goods that generate foreign exchange earnings (Brown, 1994; ECLAC, 2004). In 1988, an inflation rate of 7% was expected (Brown, 1994). Prior to Gilbert, the inflation rate was at 3.1%, on target to stay within the projected limit (Ibid.) However, the shortfalls in balance of trade

PAGE 45

45 caused by increased influx of f ood and rebuilding materials compounded with the decrease in exported agricultural goods from crop damage cau sed increased inflation in the post-hurricane months (Ibid.). In November, a rate of 14% was projected, however, by the end of the fiscal year the rates had recovered and final inflation for the year was 9.1% (Ibid.). Hurricane Ivan The rate of inflation in 2002 and 2003 was 7.3% and 14.1 % respectively, the highest in six years. This increase is believed to be cause d by further depreciation in the exchange rate compounded by increased oil prices (ECLAC, 2 004). At the beginning of 2004 a decrease in inflation was projected for the year from 14% to 9%. In the first half of the year domestic agriculture experienced a 6.5% gr oss decline due to dry conditions in the central and western parishes where the majority of domestic food is produced. This led to food shortages and the food and drink category drove 80% of the infla tion in the second quarter of 2004 as opposed to 28% in the first quarter (Planning Institute of Jamaica, 2004). Category 4 Hurricane Ivan passed from east to west, just south of Jamaica in September 2004. Most affected were the parishes of St Catherine, Clarendon, Manchester, St. James, Hanover and St. Mary. A total of 102,000 peopl e or 14% reported damage to households (ECLAC, 2004). Of those eval uated, 9% were completely de stroyed, 75% severely damaged and 14% had only minor damage. Total damage was estimated at $J 11,163 million with $J 3,666 million of the total reconstruction costs co ming from imported goods. An estimated $J 137, 235 (or $US 2,287 in 2004 dollars) reconstruction expenditure per house was estimated. Two groups most significantly aff ected by Ivan were farmers and fishers, particularly those on the southern coasts (ECLAC, 2004). An estimated 11,100 hectares of domestic agricultural land was affected with 117,700 farmers sustaining losses. There was $J 2,632 million damage to

PAGE 46

46 the domestic sector with legumes sustaining $J 43.2 million, vegetables J$ 396.4 million, and root crops J$ 570.6 million in losses (2004 J$). Banana production for the year was entirely de stroyed. This sectors devastation put 8000 persons out of work for a period of six to ni ne months (Ibid.). Full recovery of tourism infrastructure occurred three months after th e event (Ibid.). Effort s were made to employ workers in rehabilitation efforts, but temporar y loss of employment was inevitable due to the decrease in tourists (Ibid.). Road infrastructure was another of the signi ficant areas to sustain damage. However, direct damage costs were much less than the indirect losses sustaine d because of the missing support that the infrastructure provides to other sectors (Ibid.) Floods and landslides cut off entire sections of roads, damaged retaining wa lls and bridge approaches, and deposited silt on roadways. Slippage of entire sections of roads occurred from the saturation of land and raging rivers scoured road edges. However, a single la ne of traffic was ensured on all major roadways by the end of the month. Although an assessment of the indirect unit transport costs increases was not available, the increases were expected to be very high in monetary terms-that will have a negative bearing on the populatio ns well-being. (Ibid., 36). Holistic Exploration After a quarter century of pursuing a failing model of an export led economy, Jamaica has now become what Witter terms a consuming ap pendage of the US economy (2000 in Wess, 2004). Propositions that consider a more ba lanced and diversified model of economic development are presently being put forth (MOA, 2005). However, changes to the current internal food system must proceed with caution. Policy designed to improve or adjust the current Jamaican food system is hemmed in by the la rge numbers of the Jamaican population who depend directly on informal food production and di stribution for their li velihoods. The creation

PAGE 47

47 in 1963 of the Agricultural Marketing Cor poration (AMC) to cons olidate distribution, wholesaling, and retailing operati ons appeared to be a solution to the inefficiencies of the Jamaican food system. Instead, the corporation resulted in massive livelihood losses to sectors of the traditional system. Sim ilarly, campaigns to clean up Ki ngston and remove informal traders were also unsuccessful (Espeut, 2003). The current food production and distribution system persists because it functions as a last resort informal income source pursued by the large amounts of surplus labor in the country (Lewis, 1954). In other words it is a symptom of larger societal structure and cannot be superficially reformed. The internal mechanisms that transmit economic, social and environmental effects and feedbacks within the Jamaican food system must be explored in tandem so that the outcomes of complex intera cting components can be weighed in a holistic context. The next section describes the rich picture that resulted from the soft systems analysis which aims to provide a more holistic conceptualization of the issues involv ed in Jamaican food security. This is followed by a detailed empi rical review of the fi ve explicit conceptual categories revealed through the rich picture and chosen to be represented in QnD:Jamaica. These categories include populat ion typology, land use, food distribution, food pricing, and nutritional consumption. Rich Picture Reveals Conceptual Framework Main areas of concern affec ting food security revealed by stakeholders through the soft systems approach could be clustered by intern al and external socio economic and ecological drivers. External issues included changes in to urism visits, changes in export markets (closing: bananas, opening: niche), prices of imported food, decreasing nutritiona l and cultural value of imported food, and hurricane effects on infrastr ucture, tourism and productivity and rainfall

PAGE 48

48 effects on productivity. Each of these external concerns also relates to key internal issue clusters. For instance, land use change decisions are internal drivers but are closely tied to external drivers such as closing banana markets. Changed utilization of this newly available land raises new concerns in terms of maintenance of ecological inte grity and economic exploitation. If niche markets open, what institutional and infr astructural organization is necessary to take advantage? How can tourism be exploited without sacrificing ecol ogical assets? What about the current production system reduces food securi ty? What rural economic activities besides agriculture are options? These issu es in addition to others are graphically represented in the rich picture (Figure 4-1). Figure 4-1 Rich picture of the key themes of Jamaican food security. One of the above key issues that lends itself to a potential transfor mation for decreasing food system vulnerability is the farmer/market inte gration concern cluster. This cluster includes Multinational Agribusiness Tourism Climate Market Cheap food $$$ Wildlife attracts tourism Rural-Urban migration Tourism degrades wildlife Road network Local Food Urban stresses: Informal settlements, sprawl, water use, landfills Reefs, Fish stocks & Fishing culture $$ returns to community ? ?Farmer integration? Farmer/market integration? Nutrition? Safety? Regional Integration? Export Market? Traditional? Niche? Remittances from abroad Multinational Agribusiness Tourism Climate Market Cheap food $$$ Wildlife attracts tourism Rural-Urban migration Tourism degrades wildlife Road network Local Food Urban stresses: Informal settlements, sprawl, water use, landfills Reefs, Fish stocks & Fishing culture $$ returns to community ? ?Farmer integration? Farmer/market integration? Nutrition? Safety? Regional Integration? Export Market? Traditional? Niche? Multinational Agribusiness Tourism Climate Market Cheap food $$$ Wildlife attracts tourism Rural-Urban migration Tourism degrades wildlife Road network Local Food Urban stresses: Informal settlements, sprawl, water use, landfills Reefs, Fish stocks & Fishing culture $$ returns to community ? ?Farmer integration? Farmer/market integration? Nutrition? Safety? Regional Integration? Export Market? Traditional? Niche? Remittances from abroad Multinational Agribusiness Tourism Climate Market Cheap food $$$ Wildlife attracts tourism Rural-Urban migration Tourism degrades wildlife Road network Local Food Urban stresses: Informal settlements, sprawl, water use, landfills Reefs, Fish stocks & Fishing culture $$ returns to community ? ?Farmer integration? Farmer/market integration? Nutrition? Safety? Regional Integration? Export Market? Traditional? Niche? Remittances from abroad Multinational Agribusiness Tourism Climate Market Cheap food $$$ Wildlife attracts tourism Rural-Urban migration Tourism degrades wildlife Road network Local Food Urban stresses: Informal settlements, sprawl, water use, landfills Reefs, Fish stocks & Fishing culture $$ returns to community ? ?Farmer integration? Farmer/market integration? Nutrition? Safety? Regional Integration? Export Market? Traditional? Niche? Multinational Agribusiness Tourism Climate Market Cheap food $$$ Wildlife attracts tourism Rural-Urban migration Tourism degrades wildlife Road network Local Food Urban stresses: Informal settlements, sprawl, water use, landfills Reefs, Fish stocks & Fishing culture $$ returns to community ? ?Farmer integration? Farmer/market integration? Nutrition? Safety? Regional Integration? Export Market? Traditional? Niche? Remittances from abroad

PAGE 49

49 concern over small farmers returns from crop sales. The TWOACE conceptualization of the farmer remuneration system is shown in Table 4-1. Table 4-1 TWOACE characteristic s of farmer remuneration system. Farmer remuneration system T ransformation: What input transformed into output? Farmers receive little share of market price Farmers greater share of market price W orldview: What world view makes transformation meaningful? Domestic agriculture strengthens food security. O wners: Who could abolish system? International institutions, Jamaican government, higglers/middlemen A ctors: Who will perform activities to make transformation happen? Non-governmental rural development advocacy groups, farmers cooperatives C ustomers: Who will benefit/suffer from system? Farmers, citizens of Jamaica E nvironmental constraints: What constrains the system performing the transformation? Minimal tariffs on imported food, farmers have little clout, opportunities for farmers to sell directly are limited Other themes for transformation where the TWOACE analysis method was applied include: tourism returns to community, hurrica ne destruction on agri culture, value added processing in agriculture, regulation of distribution, coordina tion of production decisions, improved foreign marketing system. Often during a TWOACE analysis of a pot ential transformation, underlying influences surface to an extent where examination must stop so the issue could be deconstructed further. For instance while improving returns to farmers was the initial issue distilled from the farmer/market integration cluster, when the co ntextual environment that constrained this transformation was made explicit, new transformati ons that would need to occur prior to or in tandem with the examined transformation were rev ealed. In this case, the lack of remuneration is constrained externally by WTO trade agreements that limit the amount of protect ional tariffs Jamaica can levy on imported goods (often produced by multinational organizations under economies of scale or agricultural subsidization). Increases to remunerations are constrained internally by the low incentive for distributors to pay farmers more. Related to this is lack of

PAGE 50

50 clout farmers wield in order to demand more. Ho wever farmers are often re luctant to organize. These new concerns now become new key areas for potential transformation to be examined. The qualitative descriptions were conceptua lized by a formal framework to identify a limited number of quantifiable parameters and driver s of interest for the scenario model. The formal framework chosen to organize the scenar ios and parameters of interest was developed by the international research gr oup Global Climate Change and Food Systems (GECAFS). This group studies new policy formulation and manage ment methods that could increase food and livelihood security around the world in the f ace of changing environmental conditions. By studying multiple national and regional food system s, they hope to uncover general trends in how certain structural features of food systems benefit or harm aspects of food security. Inspired by the underlying assumptions of nested systems emergent properties, the GECAFS framework aims to create food system descriptions that acco unt for the interactions between components at multiple spatial and temporal sc ales (Ericksen 2006a, 2006b). This framework assumes that by analyzing the va riation that occurs in a specific systems internal interactions and outcome s as it is exposed to different drivers and driver strengths key features that provide systemic strength or wea kness under given scenarios can be revealed. This framework divides the areas of interest into activities, drivers, and outcomes and explores interactions between them. Ac tivities refer the tangible struct ures and actors involved in producing, processing, distributing and consuming of food. Outc omes refer not only to food security but also to effects on social welfare and environmental security and the interactions between these qualities. Interact ions between socioeco logical contextual drivers and activities produce varying conditions of f ood security in terms of ava ilability, access and utilization.

PAGE 51

51 Patterns in these interactions are revealed by evaluating e ach food security component: utilization, access, a nd availability. Having identified key actors in the Jamaican food system through the soft systems approach it was simple to appl y the GECAFS conceptual model and make explicit the activities that affect food security outcomes. The large de gree to which these activities occur external to Jamaica is evident (Figure 4-2). Figure 4-2 Food system activities and actors. The GECAFS framework builds upon some of the soft system results to bring new insights to the situation. When assessi ng the interactions of areas of key concern on availability conditions in terms of production, distribution an d exchange, key features of the Jamaican system that directly influenced model development were revealed (T able 4-2). It became evident that in Jamaica availability is influenced differe ntly in terms of food type, domestic/imported and population type, rural/urban. This indicates that the food system activities-driver interactions should be explored separately for their eff ects on the two food and population types in the decision support tool. Small Farmers Farmers Informal Middlemen Rural Urban Importer/ Exporter Jamaica Processing Consuming Distributing Producing Manufacturers Manufacturers Largescale Farmers Tourist External Formal Distributors

PAGE 52

52 Table 4-2 Jamaican food system interactions that affect availability conditions. Food type Location of interactions Rural population Urban population Production Water Land Climate shocks Production decisions Water Land Climate shocks Production decisions Distribution Few Road Infrastructure Higgler infrastructure Domestic Exchange Few Relative import/domestic prices Production External changes External changes Distribution Road infrastructure Market infrastructure Road infrastructure Market infrastructure Imported Exchange Domestic production Domestic production (to a lesser extent than in rural) Empirical Review of Conceptual Ca tegories for Model Parameterization The elements of interaction that were reve aled during Phase one were organized following the object oriented approach in which anything that can be conceptualized can be viewed as an object. The concepts in object oriented approach es are nested in hierarchies that capture the multi-scalar nature of reality. QnD:Jamaica was structured by arranging the drivers, activities, and outcomes as descriptive data objects in a hierarchy that reflect s five conceptual categories of concern: population typology, land use, f ood distribution, food pr icing, and nutritional consumption (Figures 4-3 and 4-4). The sections that follow discuss these five categories detailed empirical backgrounds which were used to parameterize QnD:Jamaica.

PAGE 53

53 Figure 4-3 Depiction of QnD:Jamaicas hier archical object oriented structure. Figure 4-4 Depiction of Local Components descriptive data. Infrastructure Landuse Population

PAGE 54

54 Population Typology The small farmer communities in Jamaica (and the whole of the Caribbean) possess a unique set of qualities that distinguish them from other small scale agriculturalists around the world. The Caribbean peasantry should be unde rstood as a reconstituted peasantry (Mintz, 1974) that emerged from an industrial form of production: the plantation. This agricultural structure which was capitalist and commercial in its economic relations and urban in its value orientations (Le Franc, 1981, 1) has had a number of implications for th e present structure of the Caribbean agriculturalist and rural society as a whole. Subsistence farming as conventionally defined does not exist in Jamaica. While the small farm often provides direct sustenance to its cultiv ator, there is always a market orientation even for the smallest farm producing the smallest bit of crops (Spence, 1996). This current production situation is built upon the remnants of the provisio nal gardens and trade network that originated during the colonial period. While the small scale farming sector supplies a majority of domestic food, the large scale export sector continues to benefit from higher quality land and infrastructure access, fueling an antagonistic tension between the dualistic agricultural sectors (Beckford, 1972; Edwards, 1961). 80% of Jamaican farmers (800,000) work less than 25% of the land (400,000 hectares) on farms less than 5 hectares each. These farmers pr ovide the vast majority of domestically grown food (STATINJA, 1996; Goldsmith, 1981). Additionally, Spence has shown the impressive extent to which small farmers engage not only in domestic market product ion but also in export orientation (1996, 1999). Not only do small farm ers diversify between domestic and export production, but they also expand their income to include a variety of non-farming sources.

PAGE 55

55 Data from the most recent Census of Agricu lture corroborates the qua litative expressions of the existence of baskets of livelihood strategies employed by small farmers (1996). Figure 4-5 shows the diversification of small farms primary income sources across parishes.3 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%JAM S t. Andrew S t Thomas P o r t l an d S t Mary St. Ann Trel a wny S t James Han ov e r Wes t moreland St. Elizabeth Manchest e r Clarend o n St C a theri n eParishPercent No work Other wage Agricultural wage Farm sales Figure 4-5 Primary income source of sm all farms by parish. (STATINJA, 1996) However, this diversification of income strategies appears to be a means to finance the main intended activity of farming (Ishemo et al. 2006). Le Franc explains that this variety of diversification strategi es serves as the key starting point for devel oping a typology for examination of the socioeconomic system: Off farm employment provides much-needed work capital and funds for domestic consumption, particularly during slack periodsbecause of the need to earn income from a variety of sources, it makes little sense to try to understand the social and economic divisions simply in terms of land ownership. Instead it is only a guide and it is necessary to describe the socio-economic strata and the hierarchy they form in terms of particular combinations of economic pursuits (1981) The key reason behind these diversifications is the inability of agricultural sales alone to generate enough income to suppor t a family within the increasi ngly cash dominated environment 3 There are also significant differen ces between income strategies employed by female versus male headed households, i.e. service wage earning females vs. agricultural wage earning males. Only combined sex averages are shown in Figure 4-1.

PAGE 56

56 (LeFranc, 1981). Historically, the free excha nge of labor between households was a common practice in Jamaica and bolstered agricultural profit margins. However, the extent that exchange can be utilized has been changing over time and differs across pari shes exemplifying the dynamic pressures of the cash economy (Ishemo et al., 2006). Presently, the use of wage labor by small farmers even when sufficient family labor exists on farm is common (LeFranc, 1981; Ishemo et al., 2006). While 76% of surveyed mi grant farmers moving from other parishes into the Rio Grande Valley (an area in north eastern Jamaica dominated by both small and large scale export banana production) claimed the use excha nge labor prior to their move, 71% explained that they currently relied instead on wage labo r (Ishemo et al., 2006). Many farms, however, do not possess the means to transition from excha nge labor to wage labor resulting in labor shortages while unemployment persists (Ibid.). There are differences not only in the types of income diversification available across the island but also in the levels of land ownership wi thin rural populations. Figure 4-6 displays the distribution of rural economic units by their lando wnership status at the time of the last census (STATINJA, 1996).

PAGE 57

57 0 5000 10000 15000 20000 25000St. Andr e w St. Th o m a s P ort l and S t M a r y S t An n T r e l aw ny S t James H a nover Westm o r el an d S t E l i za beth Manc hester Clarendon St. CatherineParishNumber of Economic Unit s Landed Economic Units Non-landed economic units Figure 4-6 Landownership of rural econom ic units by parish. (STATINJA, 1996) Land use and Agricultural Productivity There are a number of grey areas with regard to interpreting survey data on how small farmers utilize their land (Spence, 1996). In particular, the area of land in ruinate or pasture must be interpreted flexibly as farmers often use the term pasture for any areas of grass whether used for livestock or not, and land in ru inate can either be truly in fa llow or might be brought into use at any point when needed (Ibid.). An equally slippery slope is the very line drawn between export and domestic crop areas as intercropped food forests are common. Despite these grey areas, attempts to quan tify the export/domestic split on small farms exist (Spence, 1996, 1999; Davis-Morrison and Barker, 1997; STATINJA, 1996). Spence documents an export oriented bias of land alloca tion among small scale farmers in the parish of Clarendon which he describes as being representa tive of the island as a whole (1996, 133). This corroborates previous descriptions of this bias despite the small scale sectors role in supplying the majority of domes tic food (Goldsmith, 1981). Spence attributes the bias, to not only a more favorable export infrastr ucture but also to the social factors related to the structural

PAGE 58

58 dynamics of small farming (1996, 133). The fi rst factor includes access to better credit and ensured markets but even more reliable and time ly transport and payment aspects. The second influential factor relates to th e crop types labor intensity, environmental needs and level of long term investment/commitment. He cites farm ers age as the single most important factor contributing to land use decisions because of its influence on land accumulation, physical capabilities and level of commitment (1996). The 1996 Agricultural Census data show that the export bias on small farms does not hold for all parishes (Figure 4-7). St. Elizabeth, St. Ann, Manchester and Trelawny all show a greater proportion of land uses in domestic crop production. 0 2000 4000 6000 8000 10000 12000S t A n drew St. Th om a s P ort l and S t M ary St. Ann Trelawny S t James Hanover Westmoreland S t Eliz abe t h Manc hes te r Clarendo n St. C atherin eParishHa Not cultivated Domestic crops Export crops Figure 4-7 Small farm land use in terms of crop type. (STATINJA, 1996) The extent to which farmers respond to market changes and alter the types or amounts of crops in production varies th roughout the literature. A mi croeconomic modeling study in Mexico showed small scale farmers modifying production in perverse ways (Dyer et al. 2006,

PAGE 59

59 279) in response to market price shocks because of interactions with previously unaccounted for social factors. In Jamaica, opinions about how and why farm ers make decisions are diverse. Spence concludes his studies discussed above by explaining that Jamai can farmers generally respond little to price changes. However, Davis-Morri son and Barker document the dynamic patterns of production shifts on small farms in Portland parish since the 1970s (1997). They describe how the collapse of the export banana market in combination with increased support for domestic crops in the late 1970s and early 1980s led to the dominance of yam cultivation on small farms. However, the heavy destruction of this crop (a s well as the bamboo needed to stake the yam vines) during Hurricane Gilbert in combination with an increasing export market for fresh produce caused farmers to turn to dasheen farming in the late 1980s. Alas, an extreme market glut in 1993 caused this market to fall out and farmers switched back to bananas. Ballayram and Davis suggest that Jamaica s export supply responds to price incentives albeit slowly and at varying rates by crop (2003). They point out that price variability has been high during their study period and th at while nominal value has increased significantly, real crop prices have remained stable or declined, likely do to movements in the exchange rate and the general price level. An important insight they provide into the Jamaican producers logic is their low ability to mitigate risk. Vuln erability encourages non-movement in times of high variability. The newly competitive, and risky, market environm ent likely served to dampen producers price responsiveness as compared to guaranteed markets and price the s ecurity that ex isted during the protectionist 1960s and 1970s. Domestic Food Distribution Approximately 80% of domestic produce is distributed and marketed by the informal higglering system (LeFranc, 1981). The use of th e term higglering fo r petty trading activity

PAGE 60

60 has a uniquely Jamaican origin (Mintz, 1974). Higglers purchase excess supply from small farmers and transport and sell the food staples in various market es tablishments. There are three types of domestic produce higglers: 1) Higglers who sell only their ow n produce; 2) Higglers who sell their own as well as others produce; and 3) Higglers who only buy and sell (FAO, 2003). Most higglers live in rural areas and are lo cated at the lower end of the income spectrum earning an average of less than $J 50 per week in 1977 (Witter, 1988). Todays higglering situation is very similar to the colonial marke ting network that developed when Jamaican slaves were required to produce a portion of their food in provisional gardens and allowed to sell any surplus crops4. By 1735 the Sunday markets were forma lly recognized as places where slaves could sell fresh fish, milk, poultry, and other sma ll stock of all kind (Mintz and Hall, 1960). Jamaican producers are generally at the mercy of the higglers in te rms of how much they sell, and at what price. Ruijs et al. explain that developing countries do mestic food supply is often price inelastic (at least in the short run) because farmers or traders cannot respond quickly enough due to low storage abilities among other factors (2004). Thes e features challenge traditional economic theory that expects reductions in transport and marketing costs to decrease consumer prices and increase farmer prices leadin g to increased trade. Similarly, Wong implies that in Jamaica, and other developing countri es, the laws of supply and demand might not hold (1996, 518) due to the inherent nature of the higgler as a pe tty trader. Like a subsistence farmer, the higgler is trying to maximize highly limited resources. Estimates range between 25-50% for regular post harvest losses and up to 100% for semiregular losses (FAO, 2003). Hundreds of kilogr ams of food rot in the fields during seasonal 4 The ruling English maintained their tastes for European fo od and imported the majority of their foodstuffs at high costs. The plantation owners had a large aversion towa rds putting potential cash crop land into foodstuff production for on island consumption. Thus provisional slave gardens were a way to reduce operating costs. This cultural phenomenon was the original source of the consuming appendage situation that exists today. Since, the islands inception there has been little effort to promote import substitution, excluding M. Manleys early terms.

PAGE 61

61 gluts when prices in nearby urban markets are unrepresentatively hi gh (JSDNP). It is unclear if these situations result from logist ical collection obstacles, real lack of market and/ or intentionally maintained low supply levels by higglers to earn a greater profit at th e expense of farmers earnings. The potential of the first expl anatory factor is clear upon obser vation of the egg carton-like cockpit country of the Jamaican interior. Jamai cas parishes are linked by a higher quality, Class A, outer ring road that loops the island and by lesser quality, Class B, ro ads that link north to south through the hilly interior. Class C road s with limited paving connect isolated communities throughout the rugged countryside to the greate r road network. Although the Jamaican road network is extensive, it is in ve ry poor condition. Road quality deteriorates as one progresses further from the main arteries often turning to di rt paths that become impassable in heavy rains. Higglers often charge more to transport crops from areas with poor roads and some exporters will not travel to certain regions due to bad road conditions. The high dispersal of the small farming sector across this rugged landscape with each farm producing just a small amount for sale has been blamed often for the obvious challenge it poses for would be food distributors (Mintz, 1974; Norton and Symanski, 1975). Katzin relays a typical historical scenario with a higgler tr aversing across paths accessible only by foot or donkey for a daily collection of nineteen bunches of carrots, twelve scallions, five of beets and five of turnips (1959 in No rton and Symanski, 1975). Over the past decade the price differe ntials between domestic goods and subsidized industrially produced foreign foods have driven certain sectors of Jamaican agriculture out of business particularly onions, Irish potatoes, and the dairy indust ry (FAOSTATS). At present, certain vegetables such as tomatoes, cabbage, and carrots are facing similar price competition.

PAGE 62

62 However, the majority of Jamaicas vegetables are traditional crops su ch as calalloo and choo choo, which do not face this direct competition. Although, even in these sectors, there is evidence that the prices of traditional crops ar e being driven down as consumers substitute for cheaper non-traditional vegetables and as preferences shift. Many farmers complain that the low prices of imported foods relative to domestic crops make it less de sirable for higglers to collect their produce. Farmers go further to explain th at the strength of the underground import market lures farmers out of agriculture and into illega l food importing. Youth expr ess concern explicitly about the troubled marketing system when th ey explain why they will not pursue farming (JSDNP). Especially since 2001, when the cell phone market was deregulated,5 higglers know what the consumer market prices for both domestic and imported foods are and will set their buy prices for farmers accordingly. A MOA official re lated that higglers will often limit the amount of produce they take to market in order to dr ive prices up. The fixed handling costs charged by higglers are on a per container ba sis even though there is no unifo rm container size and include transport and stall fees but are assumed to be low (Wong, 1996). Additionally, farmers are reluctant to sign contracts with distributors because the farmers expect them to buy outside of the contract in order to force the farm ers price down (FAO, 2003). Food Marketing The Jamaican marketing infrastructure consists of modern retail outlets as well as government maintained parochial markets and in formal street vendors (Norton and Symanski, 1975). Although modern grocery stores are transforming the retail s ector in larger urban areas 5 The cell phone has revolutionized both farmers and higglers information access and networking ability (Hourst, Martin 2006). Approximately 2 million Jamaicans have cell phones, that is three per household.

PAGE 63

63 such as Kingston, Mandeville, May Pen, and Monte go Bay, traditional market s are still the major source of local foodstuff trade outside of these areas. Norton and Symanski categorize the traditional parochial markets by the number of higglers present (1975). Small markets usually have less than twenty higg lers, while upwards of sixty will be present in a large market. They also show a strong correlation between the size of a community and the extent of its market. They showed that only four of Jamaicas twenty-nine large markets occur in communitie s with less than two thousand i nhabitants. In the same way, only three of the thirty-five small markets are held in places with more than three thousand residents (Ibid.). The markets are often weekly but occur more often in some places depending on supply and demand (Ibid.). Each parish has one or more centralized ma rkets in its major urban areas in addition to smaller markets throughout. The ce ntral markets are more likely to serve as retail and wholesale centers where country higglers bri ng in food from rural areas and sell to town higglers who will sell in peripheral suburban markets. This is particularly true of the biggest single market, Coronation Market in Kingston, wh ich serves as a hub for distribu tion to other parishes as well as at least eight other daily urban and suburban markets in Kingston (Norton and Symanski, 1975). Witter estimates that 15,000 higglers se rve the Kingston area alone (1988). Norton and Symanski illustrate Coronations significance with a survey showing that only fifteen of sixty-four mark ets nationwide have no higglers with supplies from Kingstons central market (1975). They also list factors that se rve to guide development of a nationwide market typology: distance to Coronation market size and nature of local market, regional specialization and seasonal supply and demand factors. Ot her parishes also have central

PAGE 64

64 wholesale markets that serve the wider country side, though not on the scale of Coronation. These include Savanna-la-Mar in St. Elizabeth, and Montego Bay in Saint James (Ibid). Most Jamaican markets occur around the same time usually towards the end of the week making it difficult for higglers to work the nationa l circuit and efficiently disperse supply to spots of higher demand. This often leads to a wi de range of prices thro ughout the country (FAO, 2003). The price differences even between Coronation Market and Kingstons secondary markets such as Papine and Constant Spring are high and there is a thriving town higgler system of buying from one market (even modern supe rmarkets!) and selling in secondary suburban markets. Coronation Markets size often ensure s lower prices, certainl y in Kingston and often compared to other parish markets as well. Imported Food Distribution and Marketing Many retailers have increased the share of imported to domestically produced foods because of ensured supplies, reduced tariffs a nd a stable exchange rate (USDA, 2003). Some additional conditions that give US products an advantage in Jamaica include massive product advertisement exposure through increased cable and internet access, a nd increased demand for ready to eat products which are not manufactured domestically. Both traditional Jamaican food outlets and supermarkets use a middle man distribut or to acquire the majority of their imported foods. However, Jamaican grocers are increasingl y aligning with international chain stores and product lines to consolidate the distribution and marketing s ector both vertically and horizontally, although direct importin g by large retailers still contri butes only 5% to their total imports (Ibid.). A challenge to realizing the benefits of direct importation is limited storage space therefore retailers ar e still largely supplied by local distributors. Recently, grocers have been unable to m eet operational costs through sales earnings because prices are so low in part from th e highly competitive market (Gordan 2006). Not

PAGE 65

65 surprisingly, the Jamaican consumer is hi ghly price responsive a nd the extreme price competition between retail and wholesale outlets is causing the line between their roles to blur (USDA, 2003). The competition from supermarke t chains like HiLo and SuperPlus are leading independent retailers to form alliances and also expand into rural towns to tap new markets (Gordan, 2006). Wholesalers and informal commercial importe rs also supply the trad itional outlets such as the parochial markets and small grocers with imported foodstuffs. These operate mostly in rural and lower class urban areas but face increa sed competition from the encroaching larger retailers. This sector is expected to be displaced by chain and i ndependent large scale retailers in the medium to long term (USDA, 2003). The Jamaican Consumer Calorie intake is not a complete measure of nutrition. The amount of energy supplied by each macronutrient: carbohydrates, protein, fat etc., has important implications for a populations food security. Dietary choices are not a cold func tion of cost per calorie. Rather cultural and innate human preferences for variet y as well as sugars and fats play a large role in a consumers food buying strategy. Popkin coined th e term nutrition transition to describe the dietary shifts that usually occur along with typical demographi c and epidemiological changes in a countrys development process (2003). The two main features of the nutr ition transition are the reduced role of carbohydrates and the increased role of animal fats in providi ng caloric energy supply. This increased role of animal fats has important implications for internat ional grain markets and ecological resources world wide. These transiti ons occur due to a numbe r of factors including urbanization, technological changes, economic growth and cultural shifts. National caloric availability data does provide insight into present macroeconomic selfsufficiency issues in Jamaica indicating that the amount of imports in the country alone are

PAGE 66

66 enough to satisfy annual per capita caloric needs (Ballayram and Henry, 2004). However, it is important to unpack the imported supply as well as domestic foods in terms of what they provide nutritionally and how they are distributed geographically, demographically and temporally throughout Jamaica. Current data records in Jamaica account for na tional availability of calories, or Dietary Energy Supply (DES), by food group and imported vs. domestic sourcing, as well as income level food expenditure amounts by food group (F AO, 2003; FAOSTAT; Ballayram and Henry, 2004; Survey of Living Conditions, 2002, 2003). Mu ch Jamaican developmental literature suggests that between 1989 and 1998 the Jamaican diet was comprised mainly of domestic products (MOA, 2001). These documents sugge st that only in 1999 did consumption of domestic foods fall below that of imported foods with 269.3kg to 307.3 kg/yr/capita respectively (Ibid.). This data is actually describing only potential intake of domestic foods (availability) and, as described in the Transport and Marketing section, a lot of domes tic produce during this period wasted in the fields (JSDNP). The Survey of Living Conditions (JSLC) house hold expenditure data provide insight into financial aspects of consumption but because detailed quality and volum es are not surveyed, actual intake changes can be examined. For example the 2003 JSLC re veals that the lower income quintile spent 16.9% of its household food budget on cereals as op posed to 10.3% spent by the upper quintile. Average food expenditures as a percent of total consumption in 2003 are documented at KMA6=35%, other towns =40%, and rura l=45%. Between 1990 and 1992 the average food expenditures were KMA=50%, other towns =55% and rural=60%. Yet, the qualitative differences such a calor ie/J$ or fiber content are hidden and real dietary distribution is 6 Kingston metropolitan area

PAGE 67

67 masked. The decrease in relative amount spen t on food suggests increased food accessibility, but evidence suggests that the type s of food purchased have been changing which might have implications for the utilization component of food security. In 1998 the Caribbean Food and Nutrition Inst itute (CFNI) revealed that 31.8% of the Jamaican population possesses pre-obesity conditions with 19.7% already obese (1999). In a 1991 survey in Spanish town, Wilks found 30.7 % of men and 64.7% of the women to be overweight (1998). Between 1964 and 2000 the share of fat in the total DES increased from 19.9% to 25.9% (FAO, 2003). Carbohydrate perc entages dropped from 69% to 62.8% over the same period. This evidence agrees with the exp ected characteristic of a nutrition transition in developing countries (Popkin, 2003). Likewise, the share of calor ie availability from meat products and vegetable oils has greatly in creased since 1964 from 175 and 190 to 250 and 400 kcal/capita/day respectively. Prior to 1984 sw eeteners did not compose a significant proportion of imported food, in the 1996-1998 period imported sweeteners supplied approximately 18% of DES. The top five dishes in Jamaica are plain ri ce, fried chicken, boiled yellow yam, rice and peas and boiled dumplings. In 1989 the contributions to calorie intake by food category were: 45% carbohydrates, 22% animal products, 14% fats a nd oils, 12% sugar, and 7% legumes. In a Spanish Town study nutrients contribution to DES for males was 57.7% carbohydrates, 13.05% protein, and 30.6% fats. For females it was 59.6% carbohydrates, 13.1& protein, and 30.6% fat (FAO, 2003). Poor households in many developing countries exercise similar strategies to cope with resource shocks and maintain consumption (H anda and King, 2003). These include informal credit markets and extended sharing networks (international remittances as well as rural to urban

PAGE 68

68 links and neighbor to neighbor sharing). In ti mes of strife, traditions of communities reaching out to the least fortunate are varied. Some practice len han or lending a hand to those in need (ECLAC, 2004). Handa and King studied the diffe rent coping mechanisms between urban and rural households through one of the most comm on food security indicato rs: child wasting. Clarke and Howard, indicate increas ing investment in Kingstons slum buildings as a factor that has lessened differences in resilience be tween in rural and urban communities (2006). Household consumption surveys are the major source of permanent income data for Jamaica and many countries around the world. In developing countries these surveys often suggest that food expenditures ( purchased and home grown) a ccount for over half of the households budget. The survey data also indi cate a positive relationship between increased income and calorie intake, but at a declining rate as income increases (Alderman, 1986). The calorie-income elasticity for lower income group s was traditionally believed to be high, ranging from .4-.8 indicating a high amount of responsiven ess to lower prices or increased income (Ibid.). These findings validate intuitive notions that when one has little of something, and if more of that item becomes accessible, either th rough decreased price or increased income, one will choose to acquire that item. This led polic y makers to adopt poverty reduction strategies of increasing income or regulating prices. However, those approaches failed to view the situation with a wide enough lens. That view is based on the traditional and cursory definitions of food security, that calories are the primary nutri ent limiting improved nutrition in developing countries rather than other deficiencies or surpluses that occur from intake of non-staple foods (Levin et al., 1993; Behrman, 1995). In a study of the reliability of survey data for predicting food expe nditures in Kenya and the Philippines, it was shown that the causa l relationship between increased income and

PAGE 69

69 increased calorie intake is much weaker with elasticity ranging from .2 to even less than .01 (Bouis and Haddad, 1992). They attributed the weak relationship to the nature of the surveying methods which measured availability much more than actual intake.7 However, Bouis went further to establish a behavioral explanation for the low responsiveness of calorie intake amongst the poor (1996). One would expect a poor person to spend every bit of additional purchasing power (due to price falls or income increases) on staples. However, this assumes away the humanity of that person. Humans are not comp letely rational and will likely spend a surprise windfall on pleasure rather than security. He attributes this to ones inherent humanity. As Timmer has stated, an appropriate food policy m odel must approach demand behavior of lower income people in terms of their resignation to a st ate of chronic need (2000) In other words, the utility derived from a certain f ood over another, or the combined utility of all food consumed is derived from a combination of energy, variety and taste. It appears irrational to purchase a more expensive starch because it tastes better than the cheaper even though there might be a sacrifice in energy intake but this behavior is common in househol ds that have a brief gain in financial resources (Ibid.). This concept agrees with evidence from Indonesi a that a marginal increase in a staples price will produce a shift to another staple or ju st an over all decline in consumption. If the relative prices between nut ritionally different but si milarly taste valued products fall, a consumer will spend any extra income on the item with greater taste value t hus sacrificing nutrition. Once 7 Bouis also points to a possible phenomenon that could se rve to dampen the apparent elasticity of higher income groups while simultaneously over exaggerate the elasticity observed by the lower classes. This phenomenon is the nature of cultural relations during times of strife. Du ring these periods, rich households might actually be purchasing more and transferring these amounts to the poor who are purchasing less. In this case actual calorie intake would be maintained, but it would go unmeasured by tr aditional methods. This type of informal transfers and sharing are likely common in Jamaica. This also agrees with a common phenomenon, that counter the capitalistic model whereby in times of scarcity, price is actually cheap est as the good is ultimately shared freely rather than profited upon.

PAGE 70

70 a base level of calories is supplie d, tastes play a more significant ro le. In general, in low income households, the marginal utility that can be de rived from energy trumps taste, or variety, whereby at middle incomes, variety has the greate st value, and for the highest income brackets taste is the highest priority wh en making decisions between foods. This chapter provided a historical context fo r the Jamaican food system and described the rich picture that was developed from soft systems methods in th e field. This was followed by a discussion of how the rich picture was used to rev eal five conceptual categories that framed the object oriented structure of QnD: Jamaica. These categories were th en reviewed in terms of their available empirical background in order to parame terize the model. The next chapter describes how these categories are explicitly represented in QnD:Jamaica.

PAGE 71

71 CHAPTER 5 JAMAICAN FOOD SYSTEM MODEL DESIGN AND TESTING Approach As the simulated food system involves transmission mechanisms that channel shocks through the market, a number of economic models that aim to give insight into consumer demand, producer supply response, and price tran sfers in developing coun tries were reviewed (Robinson and Robinson, 1997; Alderman, 1986; Timmer, 1980, 2000; Bouis et al., 1992; Skoufias, 2003; Handa and King, 2003; Rouijs et al ., 2004). A general conc lusion of critics of the financial strategies pursued by Jamaica and other developing countries during the past three decades has been the persistence of black box models to shape developmental policy (Bernanke, 1995). These macroeconomic models of ten fail to explain why people participate in foreign trade under adverse conditio ns. A. Walker, in response, presents microeconomic models that claim to look at the way in which the mark et is organized, the resulting interaction of participants and the impact that they have on price movements (2002, 4) in an effort to more thoroughly explain production a nd consumption behavior in developing countries. However, as one of the main goals of this type of modeling exercise is to bridge crossdisciplinary and sector communication divides, the level of disciplinary jargon in most of the reviewed models fails to meet QnD:Jamaicas ne eds. Additionally, neither a wholly macronor microapproach will fit this projects lens of discovery. Finally, there is limited agreement even within the economic development community on the efficacy of these models (Berhrman, 1995; Bouis, 1994; World Bank, 2003). Empirical, theoretical and qualitative data informed the design of QnD:Jamaicas simulation structures. In desi gning the processes that simulate the dynamic transfers of food volumes and values, simple mathematical f unctions were employed to capture the unique

PAGE 72

72 production and consumption mechanisms in Jamaica but also highlight gene ralizable features of these mechanisms across developing countries. In this manner the QnD modeling system could be applied to other national, regional or international socioecological situations/food systems. Therefore the submodels dynamics are limited to simplified and intuitive understandings of the reviewed theoretical models in the interest of convenient collaborations with stakeholders from varied expertise levels. QnD:Jamaica is concerned with three flows: food, money and people. The data that describe the representations of these flows is organized in an object oriented hierarchical structure. Simulation occurs on a monthly ti mestep. The subprocesses of QnD:Jamaica are presented below along with the asso ciated structural assumptions that relate to the empirical review presented in the previous chapter. Additionally, the simula tion results of three comparative index categories, productivity, pricing and consumption, are shown against historical data to build confidence in QnD:Jamaicas simulation representations. Population Typology Representation The population characteristics as described in chapter four made it difficult to distill a rural typology by only land ownership or in come source. Therefore in QnD:Jamaica, each parishs rural population is distinguished by both landowne rship and major income source. Capturing this level of heterogeneity of the rural demogra phic landscape is vital for food security and other developmental studies (Dyer et al., 2006). Th e rural population typology in QnD:Jamaica consists of ten CLocalComponents shown in Table 5-1.

PAGE 73

73 Table 5-1 Rural population typology. Landowners NonLandowners CFarmer CRuralAgWageEarner CAgWageEarnerWithLand CRuralTouristWorker CRuralTouristWorkerWithLand CRuralOtherWorker CRuralOtherWorkerWithLand CRuralUnemployed CRuralUnemployedWithLand CRuralFisher As the major income source distribution does no t vary significantly across small farms of 0-5 hectares, it was assumed that all small farm land owning units in a parish own the same amount of land (STATINJA, 1996). The urban population typology is simpler than the rural because food security is often closely linked to agricultural activities and land ownership. The urban population consists of two types of wage earners: CUrbanTouristWorkers, CUrbanOtherWorkers. Also included in the urban typology are urban fisher s, and urban unemployed. Both rural and urban unemployed have monthly earnings (from the informal sector) these are assumed to be less than the minimum wage. Urban tourist workers wage is calculated fr om a PRelationship process with number of tourist per month acting as the independent variable in a negatively sloping lin ear relationship. This independent variable value is represented as a timeseries driver in the simulation. However, a hurricane occurrence will reduce this value. Ur ban other workers wages are calculated by a linear relationship driven by GDP values. The average weekly earnings for a general laborer were J$895 and $J1600 for a tourist worker in 19 92 (Planning Institute of Jamaica, 1992). The understanding of interactions between wa ges and prices in Jamaica is somewhat muddled (Craigwell, 1991). It is unclear whether a cost-push relationship exists whereby rising wages drive price increases in an inflationary spiral (Bourne 1977) or if prices are so

PAGE 74

74 exogenously determined and labor unions so power less that workers seek to catch up with price inflation (Craigwell, 1991; Downes and Mc Clean, 1982). Craigwe lls study of Jamaicas wages and export prices over the period of 1957-1984 revealed the possibility of a positive feedback between wages and pric es supporting the inflationary sp iral theory (1991). However, similar tests on Barbados and Trinindad and Toba go suggest only uni-directional flow from price to wage (Ibid.). Additionally that study period was largely comprised of a peculiar time in Jamaican history when governmental regulation of wages was high. In Portland parish, rates as high as 79% of sm all farmers receive remittances from abroad with 74% of recipients receiving cash on a monthl y basis (Ishemo et al., 2006). Some 67% of recipients received less than $J5000 (~$US83) per month and 90% less than $J10000 ($US 166) per month (Ibid.). The ratio or remittances to total income increased throughout the 1990s (World Bank, 2003). All population groups in Qn D:Jamaica receive equal remittance amounts. The total remittance value entering Jamaica is a driver inputted as a linearly increasing timeseries file. Land Use Representation Land use in QnD:Jamaica is divided first by ru ral and urban uses (Figure 5-1). Rural land is further distinguished as non-agricultural lands estate lands: farms >100ha producing mainly permanent export crops, and small farm lands: farms <5ha. Middle size farms between 5-100ha are not included in this model version. Estate lands are further categorized as cultivated and non-cultivated. Major estate crops include bana nas, sugar cane and coffee, however, the model groups these types into the one category descri bed as export crops. While small farms are described by the amount under cultivation in four main crop types: root crops, legumes, vegetables and permanent crops, as well as by the amount that is uncultivated. These crop type categories were chosen because they have e nvironmental, economic, nutritional and cultural

PAGE 75

75 significance as evidenced by their use in the Agri cultural Census, the Jamaican Survey of Living Conditions and the Economic and Social Survey of Jamaica. Figure 5-1 Land use structure in QnD:Jamaica. Despite the various accounts of Jamaican producers price responsiveness described in the previous chapter, for the short time periods simulated by this model (5-20 years), it is assumed that land uses are not substitutable. That is, la nd does not switch between cultivation of export crops, food crops, livestock and ruinate. This is in accord with producer response models used for similar market types in other regions (Ruijs et al., 2004). Instead the amount of land in each land use is driven by a timeseries file of trended known changes in area cultivated.8 The FAO and MOA estimate slightly differe nt but similarly trended land areas under cultivation in the crop types of concer n for 1995-2004 (Figures 5-2 and 5-3). 8 A supply response function driven by price elasticity could be included in future scenario simulations whereby the total amount of land in production as well as the spread of crop types (domestic v. traditional export, domestic v. non-traditional export) varies. This could also include a degradation factor if land in ruinate is cultivated. Rural Land Urban Land Estate Lands: >100ha Small Farms: <5ha Other Land: Parks Cultivated Root crops Legumes Vegetables Uncultivate d Uncultivate d Permanent Crops

PAGE 76

76 Figure 5-2 FAO cultivated area by crop group. (FAOSTATS) Figure 5-3 MOA cultivated area by crop group. (MOA, 2001) Each of the three crop types exhibit distinctive trends. Legume area decreased steadily throughout 1986-2001 with a slight increase in recen t years. Root crop land exhibited a fairly constant level until 1996, except for sharp decrea ses in 1988 and 1989 due to Hurricane Gilbert. After 1996, root crop land declined sharply again for two years. It held constant in 1999 and declined again in 2000. Since then it has been fa irly steady. Vegetable ar ea was fairly constant until 1991 excluding a decrease in 1988. Like root crop land, vegetable area increased steadily until 1996. Following a sharp decrease in 1997, area ha s been fairly constant except for a slight decrease in 2000. 0 2000 4000 6000 8000 10000 12000 14000 16000 1995199619971998199920002001200220032004 Y ea r Ha Legumes Vegetables Yams 0 5000 10000 15000 20000 25000 19851987 1989199119931995 1997199920012003Year Ha Pulses,Total Roots and Tubers,Total Vegetables Primar y

PAGE 77

77 The 1996 Agricultural Census facilitated the dist illation of national ag ricultural land areas into proper divisions by crop type on small farms at the parish level. It was assumed that as national areas changed the relative crop type distributions within parishes held constant over the years (i.e. the small farms of St. Ann and Trel awny still produce the ma jority of the nations yams). These percentages are shown in Table 5-2. Table 5-2 Percent of Jamaicas land uses by parish. Percent of all cultivated land in Jamaica Percent of all permanent crop land in Jamaica Percent of all legume land in Jamaica Percent of all vegetable land in Jamaica Percent of all root crop land in Jamaica Percent of all uncultivated land in Jamaica St. Andrew 4% 5% 5% 4% 2% 2% St. Thomas 7% 8% 9% 11% 2% 3% Portland 5% 7% 1% 2% 3% 4% St. Mary 9% 13% 2% 7% 4% 5% St. Ann 9% 4% 10% 19% 16% 13% Trelawny 5% 3% 2% 2% 16% 4% St. James 4% 5% 1% 2% 3% 5% Hanover 5% 5% 3% 3% 7% 7% Westmoreland 7% 7% 6% 5% 6% 11% St. Elizabeth 9% 4% 34% 14% 5% 16% Manchester 8% 4% 12% 11% 13% 13% Clarendon 16% 20% 10% 11% 14% 10% St. Catherine 12% 15% 6% 8% 9% 7% It was originally thought that these dist ribution percentages could be multiplied by the national total land areas reported by the FAO an d MOA to simulate proper parish level distributions while also accounti ng for the annual changes. Ho wever, the areas reported by the FAO and MOA needed to be modified before this distribution could be ca lculated. The land area data reported by the FAO and the MOA, lowered production in 1988, 1997, 2000, and 2004 appears to be a function of the am ount of area harvested-not planted. In fact, each of these years correlates to a major environmental event that decreased production (Hur ricane Gilbert, Tropical Storm Marco, drought and Hurricane Ivan respectively). Thus, while productivity appears

PAGE 78

78 constant in these years, it, in fact, decreased when the amount of actually harvested land was different than the amount originally planted. Thus, it is unknown how mu ch of the decreased production is due to producers land use choi ces prior to planting versus uncontrollable environmental factors post planting. This differe nce is critical for this study because of the interest in hurricane and flooding damage estimation. In order to overcome what is believed to be an underestimation of land put into cultivation at the beginning of the season, tre nd lines were estimated for each crop types national area.9 Using this trend line as a driver represents producers decisions but controls for the highly unknown feedbacks that drive those decisions. The land area values under contention were removed, and the trends for apparent period s were estimated separately (Figure 5-4). Figure 5-4 Estimated land areas that act as drivers in QnD:Jamaica. Land use changes on the large plantations occupy ing the majority of Jamaicas agricultural land would have large economic im plications in terms of wage earnings (or small farm earnings if land were distributed as such) (Weiss 1999, 2004b) Changes in these land uses would also have vast effects on the ecology of the island. Like small farms, cu ltivated land ar ea on estates 9 In the future, rather than a trend line, land use changes will be tied to market demand and price so that the dynamics could be internalized to the model. 0 5000 10000 15000 20000 25000 1985 1988 1993 199720002003 Y EAR HA VEG ROOTS LEGS

PAGE 79

79 simulated in QnD:Jamaica change through a driver file. This dr iver file was estimated using known decreases in sugar and banana land areas on large plantations. Productivity Representation The crop yield subprocesses in QnD:Jamaica are based on three PRelationship processes with the simulated month and rainfall amount serv ing as the driving independent variables. The following equation describes domestic crops mont hly yield (tons/ha) calculation and includes maximum yield, productivity, and flood damage coefficients in described in detail below. Yield=Max monthly yield coe fficient X Productivity coeffici ent X Flood damage factor Each month is related to a maximum productivity due to seasonal variability. Coefficients for each months maximum yield (tons/ha) were calculated from known typical annual yields for each crop type. Generally maximum productivity occu rs in the early spring months (Table 5-3). Table 5-3 Maximum monthly yield coefficients (tons/ha) Month Vegetables Root crops Legumes J 0.9 1.5 0.15 F 1.0 1.45 0.15 M 1.0 1.45 0.15 A 0.9 1.4 0.15 M 0.9 1.4 0.15 J 0.9 1.35 0.15 J 0.9 1.2 0.15 A 0.9 1.2 0.15 S 0.9 1.25 0.15 O 0.9 1.35 0.15 N 0.9 1.45 0.15 D 0.9 1.5 0.15 Annual Yield 11 16.5 1.8 The Jamaican Meteorological Service posse sses 30 year average pa rish level rainfall records as well as monthly parish level records for 1992-2002. This range of records served to define the time scale of QnD:Jamaicas base sc enario. The crop productivity and flood effect generator was calibrated using known rainfall amounts from 1992-1997 including historical

PAGE 80

80 drought and flood events, crop yiel ds and welfare effects. Initially the crop productivity coefficient was calculated with a relative rainfa ll index based on a month s real rainfall amount (mm) divided by that months thirty year aver age amount (mm). However, because Jamaicas weather has distinct dry and we t seasons, this method failed to capture decreased productivity caused by abnormal rainfall in previous months of the growing season. The use of a relative rainfall index based on a four month running rainfall average (as opposed to one) divided by the current months thirty year rainfall average to determine productivity coefficients proved more appropriate. These productivity coefficients are shown in Table 5-4. Table 5-4 Productivity coefficients Vegetables Roots Legumes Relative rainfall index Productivity coefficient Relative rainfall index Productivity coefficient Relative rainfall index Productivity coefficient <0.5 0.75 <0.5 0.25 <0.5 0.75 0.6 0.9 1.25 1.0 0.6 0.9 0.699999 0.95 1.5 1.5 0.699999 0.95 0.7 1.0 1.75 1.0 0.7 1.0 0.85 1.0 1.75 0.25 0.85 1.0 1.0 1.1 1.0 1.1 1.39999 1.2 1.39999 1.2 1.4 1.5 1.4 1.5 1.74999 1.5 1.74999 1.5 1.75 0.8 1.75 0.8 1.89999 0.8 1.89999 0.8 1.9 1.0 1.9 1.0 10.0 1.0 10.0 1.0 Figure 5-5 graphically depicts vegetables pr oductivity coefficient PRelationship process shown in Table 5-4 above.

PAGE 81

81 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 00.511.522.533.5 Four Month Relative Rainfall IndexProductivity Coefficient Figure 5-5 Vegetabl es productivity coe fficient relationship The four month running relative rainfall subprocess captures pr oductivity decreases due to drought. However, in order to simulate sudden shocks such as a massive flood or hurricane, a separate subprocess also tracks one month relative rainfall. Floods in Jamaica are common because of high runoff coefficients across the island, excluding the porous Cockpit country (Nkemdirim, 1979). High runoff levels are princi pally correlated with: Annual rainfall amounts and seasonality and to a lesser extent slope, drainage density and elevation (Ibid). In mapping flooding hazards, Jamaicas Office of Disaster Preparedness and Emergency Management first mapped landslide risk with three geological and slope characteristics and then associated high flood risk with these areas that were less than 40 m high (CDERA, 2003). Presently, QnD:Jamaica simulates floods based only on rainfall amounts. If the one month relative rainfall is greater than 3.25, indicating th at the months rainfall was 325% above normal, crop yields are reduced by a flood damage factor of 20%. As mentioned above, topography plays a strong ro le in flood occurrences. However, in QnD:Jamaica the flood damage trigger (a relative rainfall greater than 3.25) is ubiquitous across all parishes. In reality this tr igger would be lower in places pr one to flood (low lying areas, poor drainage etc.) and higher in la ndscapes that can withstand heavier rainfall. This uncaptured

PAGE 82

82 variation is likely one reason for the follo wing calibration issue. 1993 and 1996 experienced similarly high monthly rainfall events, yet grea ter productivity damage was reported in1993. In order to maintain high levels of productivity during 1996 (the year of highest recorded productivities) the flood trigge r was set above both 1993 and 1996 monthly relative rainfall. Productivity Calibration The food production sub models were calibrate d using data from the six year period between 1992 and 1997. The calibration period wa s chosen because it contained data for drought, floods and normal rains. Monthly parish level rainfall data was only available for the years between 1992 and 2002. Vegetables Vegetable productivity between 1992 and 1996 e xperienced a gradual increase from 9.49 to 12.50 tons/ha. In 1997 there was a sharp decline due to a major drought. The productivity in 1995 was only slightly greater than 1994 but 1996 experienced a large in crease due to ideal rainfall conditions. There were two reported tropical storms during this period, Gordan (1994) and Marco (1996). Additionally, floods from h eavy rainfall were reported during 1993. Only annual historical productivity values were available. Thus, only inter annual productivity could be calibrate d. The models monthly outputs were summed and divided by the respective years cultivated land area as simulated. 1992 and 1993 produce unrealistically high productiv ity levels. This is possibly due to increased maximum productivities throughout the time period from other advances made in production techniques such as fertilizers and pest cont rols. In order to account for the changes in maximum attainable productivity, additional paramete rs would need to be included in the model. Unrealistically high production levels in 1992 could also be due to lingering drought effects from 1991. Additionally, to reiterate 1993 experienced heavy floods which were not captured in the

PAGE 83

83 flood simulator so that the damage from1996 floods could also go unregistered. Productivity in 1994 and 1997 was slightly over estimated, while, productivity in 1996 was slightly underestimated. Simulated production volumes reflect similar levels of a ccuracy (Figures 5-6, 57, 5-8 and 5-9). 8 9 10 11 12 13 199219931994199519961997 YearProductivity (tons/ha ) Known Productivity Simulated Productivity Figure 5-6 Vegetables produc tivity calibration by date. 9 10 11 12 13 910111213 Known Productivity (tons/ha)Simulated Productivit y (tons/ha) Figure 5-7 Vegetables productivity calibration difference from known.

PAGE 84

84 1.2E+08 1.3E+08 1.4E+08 1.5E+08 1.6E+08 1.7E+08 1.8E+08 1.9E+08 2.0E+08 199219931994199519961997YearVolume (kgs) Known Volume Simulated Volume Figure 5-8 Vegetables volume calibration by date. 1.2E+08 1.3E+08 1.4E+08 1.5E+08 1.6E+08 1.7E+08 1.8E+08 1.9E+08 2.0E+08 1.2E+081.4E+081.6E+081.8E+082.0E+08Real Volume (kgs)Simulated Volume (kgs) Figure 5-9 Vegetables volume ca libration difference from known. Root crops Root crops productivity could not be calib rated to as high a level of accuracy as vegetables. The productivities in 1996 and 1997 were underestimat ed (Figures 5-10 and 5-11). Despite imprecision in productivity simulation, the root crop volume s reflect a reasonable degree of accuracy (Figures 5-12 and 5-13).

PAGE 85

85 12 13 14 15 16 17 18 199219931994199519961997 YearProductivity (tons/ha) Known Productivity Simulated Productivity Figure 5-10 Root crops produc tivity calibration by date. 14 15 16 17 18 19 141618 Known Productivity (tons/ha)Simulated Productivit y (tons/ha) Figure 5-11 Root crops productivity calibration difference from known. 2.5E+08 2.7E+08 2.9E+08 3.1E+08 3.3E+08 3.5E+08 3.7E+08 3.9E+08 199219931994199519961997YearVolume (kgs) Known Volume Simulated Volume Figure 5-12 Root crops volume calibration by date.

PAGE 86

86 2.5E+08 2.7E+08 2.9E+08 3.1E+08 3.3E+08 3.5E+08 3.7E+08 3.9E+08 2.5E+083.0E+083.5E+084.0E+08Known Volume (Kgs)Simulated Volume (kgs ) Figure 5-13 Root crops volume ca libration difference from known. Legumes Legumes productivity values experienced the hi ghest degree of error. Values in 19921994 are greatly over estimated, while values in 1996, and to a gr eater extent 1997, are underestimated. However, the inter-annual chan ges do show reasonably similar patterns as known changes. Despite productivity in 1997 be ing grossly underestimated, simulated volume reflects greater precisi on. (Figures 5-14, 5-15, 5-16, and 5-17). 1.5 1.6 1.7 1.8 1.9 2 19921993199419951996 YearTons/Ha Known Productivity Simulated Productivity Figure 5-14 Legumes productiv ity calibration by date.

PAGE 87

87 1.5 1.6 1.7 1.8 1.9 2 2.1 1.51.71.92.1 Known Productivity (kgs/ha)Simulated Productivity (Tons/ha) Figure 5-15 Legumes productivity calibration difference from known. 0.0E+00 2.0E+06 4.0E+06 6.0E+06 8.0E+06 1.0E+07 1.2E+07 1.4E+07 199219931994199519961997 YearVolume (Kgs/Ha) Known Volume Simulated Volume Figure 5-16 Legumes volume calibration by date. 7.5E+06 8.5E+06 9.5E+06 1.1E+07 1.2E+07 1.3E+07 1.4E+07 7.5E+069.5E+061.2E+071.4E+07Known KgsSimulated Kgs Figure 5-17 Legumes volume ca libration difference from known.

PAGE 88

88 Livestock Although small farmers produce a fair amount of the national livestock, QnD:Jamaica simulates livestock production as a large scal e industrialized process. This is a fair representation of the poultry industr y in particular. As the vast majority of animal feed is imported, the border price of feedstuffs direct ly influences domestic livestock production quantities and can be used as a model driver. The strongest relationship between feedstuff price and production volumes was found between real Jama ican dollar values and total volumes. In order to calculate values in real Jamaican dollars the external feedstuff price is multiplied by the exchange rate and divided by the Consumer Price Index, CPI, both of which are also included are timeseries driver files. The real J$ value serves as th e independent variable in a PRelationship process to generate livestock production volumes (Figure 5-18). Livestock production was calibrated within 10% of real known volumes (Figures 5-19 and 5-20). 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 0100020003000 Feedstuff Price (Real J$/ton)Livestock Production (Kgs) Simulated Relationship Known Values Figure 5-18 Livestock production relationship.

PAGE 89

89 0 10000 20000 30000 40000 50000 60000 70000 199219931994199519961997 YearVolume (tons) Known Production Simulated Production Figure 5-19 Livestock pr oduction calibration by date. 30000 40000 50000 60000 70000 80000 3000040000500006000070000 Known Production (tons)Simulated Production (tons) Figure 5-20 Livestock production calibration difference from known. Testing Productivity Simulations Vegetables productivity As in the simulation of vegetable producti on in 1996, the model wa s not sensitive enough to produce high levels of productivity in 2001. Productivity was also underestimated in 2000. However, this was likely due to the model bein g overly sensitive to drought occurrences during that year. In the years 1998 and 1999, poor and good years of productivity respectively, were simulated near historical valu es. Volume levels reflected similar trends as productivity indicating no strong errors in land area estimation. The inter-annual shifts are also similar to known trends (Figures 5-21, 5-22, 5-23, and 5-24).

PAGE 90

90 8 9 10 11 12 13 19981999200020012002 YearProductivity (tons/ha ) Known Productivity Simulated Productivity Figure 5-21 Vegetables pr oductivity testing by date. 9 9.5 10 10.5 11 11.5 12 12.5 13 910111213 Known Productivity (tons/ha)Simulated Productivity (tons/ha) Figure 5-22 Vegetables productivity testing difference from known. 1.E+08 1.E+08 1.E+08 2.E+08 2.E+08 2.E+08 19981999200020012002 YearVolume (kgs) Known Volume Simulated Volume Figure 5-23 Vegetables volume testing by date.

PAGE 91

91 1.2E+08 1.3E+08 1.4E+08 1.5E+08 1.6E+08 1.7E+08 1.2E+081.4E+081.6E+08 Real Volume (kgs)Predicted Volume (kgs) Figure 5-24 Vegetables volume testing difference from known. Root crops productivity Root crops productivity was tested with the greatest degree of accuracy out of the crop types. The year 2000 was the only year where productivity was greatly over estimated. Root crop production levels lend themselves more readily to course productivity simulation. This could be due to the nature of root crop production or to the fact that there is less variation within the root crop category as opposed to vegetables which contain a wider variety of species and cultivation techniques (Figures 5-25, 5-26, 5-27, and 5-28). 12 13 14 15 16 17 19981999200020012002 YearProductivity (tons/ha ) Known Productivity Simulated Productivity Figure 5-25 Root crops pr oductivity testing by date.

PAGE 92

92 13 14 15 16 17 18 19 131517 Known Productivity (tons/ha)Simulated Productivit y (tons/ha) Figure 5-26 Root crops productivity testing difference from known. 1.5E+08 1.7E+08 1.9E+08 2.1E+08 2.3E+08 2.5E+08 2.7E+08 2.9E+0819981999200020012002 YearVolume (kgs) Known Volume Simulated Volume Figure 5-27 Root crops volume testing by date. 2.0E+08 2.2E+08 2.4E+08 2.6E+08 2.8E+08 3.0E+08 3.2E+082E+082.2E+082.4E+082.6E+082.8E+083E+08Known Volume (Kgs)Simulated Volume (kgs ) Figure 5-28 Root crops volume testing difference from known.

PAGE 93

93 Legumes productivity The legumes productivity simulator generate d reasonably accurate inter-annual dynamics. Productivity in 2001 was underestim ated and in 2002, overestimate d. The increase in error between 2002 productivity calcula tion and volume calculation indi cates error in the amount of estimated land in legume cultivat ion (Figures 5-29, 5-30, and 5-31). 0 0.5 1 1.5 2 2.5 3 19981999200020012002 YearTons/ha Known Productivity Simulated Productivity Figure 5-29 Legumes productivity testing by date. 1 1.5 2 2.5 11.522.53 Known Productivity (tons/ha)Simulated Productivity (tons/ha) Figure 5-30 Legumes productivity testing difference from known.

PAGE 94

94 0.0E+00 2.0E+06 4.0E+06 6.0E+06 8.0E+06 1.0E+07 19981999200020012002 YearVolume (tons) Known Volume Simulated Volume Figure 5-31 Legumes volume testing by date. Livestock production The livestock production process greatly underestimated produc tion volume for 20002002. This is due to historically high production despite high feed prices (Figures 5-32 and 533). 60000 65000 70000 75000 80000 85000 90000 19981999200020012002 YearProduction Volum e (tons) Known Production Simulated Production Figure 5-32 Livestock pr oduction testing by date.

PAGE 95

95 60000 65000 70000 75000 80000 85000 600007000080000 Known Production (tons)Simulated Production (tons) Figure 5-33 Livestock production testing difference from known. Food Importation Ideally the amount of food imported would be a result of feed back loops. The driving influences could be the amount of imported pro duct purchased in previo us time periods, either the immediately prior period or a running av erage of demand. Or, the imported food amounts could be driven by the supply present in the cu rrent time period. Each method would generate different degrees of responsiveness. In this version of the model, however, importe d food is represented as an unlimited supply. The amount actually purchased is limited by income restraints and imported foods prices relative to substitutable domestic foods (described in greater detail in Consumption section below). In fact, the notion of inexhaustible importe d food is fairly representa tive of the historical simulated period. It will be most important to develop feedback loops for additional scenario simulations beyond this historical period. Food Distribution and Marketing Representation The Jamaican food distribution and mark eting landscape is understood as a single national system due to the islands small size and the presence of an extens ive road network that makes it possible to visit every parish in a sing le day. However despite these features, Jamaica

PAGE 96

96 experiences moderate spatial a nd temporal price variation as a consequence of parish level supply, demand and infrastructural differences. Temporal shifts result from domestic seasonality and other climatic effects as we ll as external trade conditions. Spatial price variation is a function of transport and marketing infrastructure differences such as a parishs distance to ports, road conditions and rural popula tion density. The expanding mode rn distribution and retail sector intertwined with the tr aditional higgler system presen t unique challenges to managing Jamaicas food distribution network. The follo wing section discusses how the qualitative systemic structures described in chapter four ar e represented as specific sub models to simulate prices in QnD:Jamaica. Representing National Farmgate Prices Simple linear relationships were estimated using past known production volumes and their associated real farmgate prices within the study period to simulate the mechanisms acting on national real prices. The reviewed evidence in chapter four suggests that parish level prices should be increased above this base national level by factors representing hi gglers transportation and marketing costs. The design logic of QnD:Ja maicas pricing processes is discussed below. Crop pricing It was assumed that the volumes reported by the MOA and FAO were the total production volumes rather than the actual volumes at ma rket because the volume, productivity, and the amount of land in production agree. Vegetable prices were simulated with a differe nt process than root crops and legumes for the sake of exploration. While the real prices of vegetables were generated from a PRelationship process with production volume as the independent variable, the real pric es of root crops and legumes were generated through a multiple step pro cess in order to capture price elasticity. In those price generation processes, the percent change in the current months production volume

PAGE 97

97 from the running twelve year average production volume served as the independent variable in a PRelationship process. These relationships are shown for the three main domestic crops modeled in QnD:Jamaica (Figures 5-34, 5-35, an d 5-36). Nominal prices were calculated by multiplying the real price valu es by the CPI driver amount. 1.5 2 2.5 3 3.5 4 4.5900000011000000130000001500000017000000Harvest Volume (Kgs)Farmgate Price (Real J$/kg) Known Values Simulation Relationship Figure 5-34 National vegetables farmgate price relationship. y = -4.4017x + 2.8745 R2 = 0.83261.5 2 2.5 3 3.5 4 4.5 -0.3-0.2-0.100.1 Change in Harvest Volume (%)Producer Price (Real J$/kg) Figure 5-35 National root crops farmgate price relationship.

PAGE 98

98 y = -1.5948x + 5.1433 R2 = 0.26523 3.5 4 4.5 5 5.5 6 6.5 -0.4-0.200.20.4 Change in Harvest Volume (%)Producer Price (Real J$/kg) Figure 5-36 National legumes farmgate price relationship. Each price relationship function is bound by minimum and maximum values established just outside of the known historic al values. The ranges are limited in real J$/kg to 2.5 4.2 for root crops, 1.75-3.3 for vegetables, and 4.4-6.3 for legumes. Crop pricing calibration Simulated vegetable prices in 1993 and 1995 were consistently lower than known values. As vegetable prices are calculate d from a direct relations hip with harvest volume, price error is a result of error in the production simulator. Figures 5-37 and 5-38. 2 2.2 2.4 2.6 2.8 3 3.2 3.4 199219931994199519961997 YearPrice (Real J$/Kg) Known Simulated Figure 5-37 Vegetable pr ice calibration by date.

PAGE 99

99 2 2.2 2.4 2.6 2.8 3 3.2 3.4 1.522.533.5 Known (J$/Kg)Simulated (J$/Kg) Figure 5-38 Vegetable price cal ibration difference from known. Root crops price simulation expe rienced a greater degree of erro r than vegetables price. Using the current subprocess de sign, high price levels could no t be generate d without over estimating prices in 1996 and 1997 (Figures 5-39 and 5-40). Simulated inter annual legumes prices were relatively stab le (Figures 5-41 and 5-42). 2 2.5 3 3.5 4 4.5 199219931994199519961997 YearPrice (Real J$/Kg) Known Simulated Figure 5-39 Root crops price calibration by date.

PAGE 100

100 2 2.5 3 3.5 4 4.5 2345 Known (Real J$/kg)Simulated (Real J$/kg ) Figure 5-40 Root crops price cal ibration difference from known. 2 3 4 5 6 7 199219931994199519961997 YearPrice (Real J$/Kg) Known Simulated Figure 5-41 Legumes price calibration by date. 0 10 20 30 40 199219931994199519961997 TimePrice (Nominal J$/Kg) Known Simulated Figure 5-42 Legumes price calib ration difference from known.

PAGE 101

101 Livestock pricing It is easy to distinguish extern al feedstuff prices as a clear driver for domestic livestock prices. After production volumes are simulate d using external feedstuff prices as the independent driver, domestic poultr y prices are calculated using a PRelationship process similar to the one employed for vegetables prices that relates gross volumes to real prices (Figure 5-43). Livestock prices are also bound by limits just beyond the historical range: 4.6-7.3 real J$. 3 4 5 6 7 83.0E+064.0E+065.0E+066.0E+067.0E+068.0E+06Production Volume (Kgs)National Price (Rea l J$/Kg) Known Values Simulated Relationship Figure 5-43 National livesto ck price relationship. Livestock prices were calibrated within adequa te levels for all years except 1992 and 1993. In those first years, prices were much higher th an any for the rest of the period, but production levels did not reflect the expected decrease (Figures 5-44 and 5-45). 4 4.5 5 5.5 6 6.5 7 7.5 199219931994199519961997 YearNational Livestock Pric e (Real J$/kg) Known Simulated Figure 5-44 Livestock pr ice calibration by date.

PAGE 102

102 4 4.5 5 5.5 6 6.5 7 7.5 45678 Known Price (Real J$)Simulated Livestock Price (Real J$/kg) Figure 5-45 Livestock price cal ibration difference from known. Price Testing Average annual vegetable prices were consis tently over estimated in the testing period simulations. However, inter-annual simulated dynamics are nearly similar to known changes (Figures 5-46 and 5-47). 2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 19981999200020012002 YearPrice (Real J$/Kg) Known Simulated Figure 5-46 Vegetable price testing by date.

PAGE 103

103 2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 1.522.533.5 Known (J$/Kg)Simulated (J$/Kg) Figure 5-47 Vegetable price testing difference from known. Figure 5-48 shows the monthly change in simula ted nominal prices against the change in the known vegetable consumer price index. Simu lated price changes were within reasonable levels; however, they did not disp lay significant similarity to know n prices changes. There is some evidence that the monthly price subprocess is moving closer to simulating reasonable prices. For example, QnD:Jamaica captures the large price jump in ear ly 1993 even though it is simulated earlier than it occurr ed in reality. Also the simu lation of the know n large decline towards the end of 1993 followed by a degr ee of stability throughout early 1994 builds confidence in the model. Finally in late 1994, the simulated and known inter-monthly dynamics are similar. -25 -20 -15 -10 -5 0 5 10 15 20J-9 2 A-9 2 J-9 2 O-92 J-93 A93 J-93 O-9 3 J -94 A-94 J-94 O-94TimePercent Change Known Change Simulated Change

PAGE 104

104 Figure 5-48 Change in known vegetables pr ice index vs. change in simulated nominal vegetables price. The testing simulation of root crops prices reflects the models inability to generate dynamic inter-annual changes for this crop which was revealed during calibration (Figures 5-49 and 5-50). Figure 5-51, however, shows that the model is generating dynamic intra-annual prices in accordance with monthly volume changes. 2 2.5 3 3.5 4 4.5 19981999200020012002 YearPrice (Real J$/Kg) Known Simulated Figure 5-49 Root crops price testing by date. 2 2.5 3 3.5 4 4.5 22.533.544.5 KnownSimulated Figure 5-50 Root crops price testing difference from known.

PAGE 105

105 0.00E+00 5.00E+06 1.00E+07 1.50E+07 2.00E+07 2.50E+07 3.00E+07 3.50E+07 Jan-92Jan-94Jan-96Jan-98Jan-00Jan-02 TimeHarvest Volume (Kgs ) 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5Price (RealJ$/Kg) Volume Price Figure 5-51 Root crops simula ted volume and real prices. Inter-annual legumes prices lack strong in ter-annual dynamics as expected from the calibration period simulation (Figur e 5-52). Livestock price simula tion is generally similar to historic amounts (Figures 5-53 and 5-54). 4.2 4.4 4.6 4.8 5 5.2 5.4 5.6 19981999200020012002 YearPrice (Real J$/Kg) Known Simulated Figure 5-52 Legumes price testing by date.

PAGE 106

106 3 3.5 4 4.5 5 5.5 19981999200020012002 YearPrice (Real J$) Known Simulated Figure 5-53 Livestock price testing by date. 4 4.5 5 5.5 6 44 555 5Known Price (Real J$)Simulated Price (Real J$) Figure 5-54 Livestock price testing difference from known. Representing Producer-Higgler Food Transfer Volumes All land owning population groups harvest their entire available crop of each food type (Root crops, Vegetables, Legumes and Perman ent Crops). The total amount, DHarvested Crop is added to their D Crop Store by the process PMonthlyHarvest. Next the farmer will transfer a certain amount of this store to the DNational Crop Quantity. The process simulating the transfer of food into the market is guided by a producer propulations production level relative to both other producer populations production levels across Jamaica and to the producer populations household consumption demands for the crop under question.

PAGE 107

107 The features that establish the producer populations food sell ing relationship process are based on the per capita monthly production amounts within each population group of each parish. The key points of interest are: 1) the average per capita production amount in the parish population with the leas t production amount, 2) the average national production amount, 3) the midpoint between the average smallest and natio nal average, 4) the average production amount from the population group producing the most pe r capita, and 5) the la rgest production amount overall on a per capita basis (Figure 5-55). Absolute Largest Avg. Largest Avg. Smallest Midpoint Avg. NationalProduction Amount (kgs/capita/month)Percent of Production Sold 35 90 50 75 0 10 Figure 5-55 Percent of production sold. It is assumed that producer populations with the smallest producti on levels would only sell 10% of their crop because th ey are producing much less than the consumption demands for that crop. That is to say, they are producing so little it is assumed to be largely for subsistence rather than market. The average national produc tion amount was equated with a 50% sell rate. Along the same lines, those producers growing th e largest amount would on average sell 75% of their crop, but never more than 90%. QnD:Jamaicas crop production simulation outputs were used to determine the values for each of the five points of interest above. The amounts for each crop are shown in Table 5-5.

PAGE 108

108 Table 5-5 Crop sell amount factors (kg/capita/month). Representing Consumer Prices As discussed in chapter four, the Jamaican f ood distribution infrastr ucture often isolates rural communities. This is true not only in terms of collection of their produce, but also in terms of making imported food availabl e to those populations. Thus, just as prices increase for domestic foods as distance from farm increas es, so do imported food prices increase with distance from ports and urban centers. Tran sportation and marketing costs of a given populations food are captured in QnD:Jamaica through index factors that increase food prices above their base national price. Every parish has associated marke ting and transport index factors for its rural and urban areas. The base national price for domestic foods is the nominal farmgate price. The base national price for imported foods is set by a driver file containing extern al nominal prices in US$/kg. These prices are then multi plied by the exchange rate in th at time period to establish the national base price in nominal J$. Tariffs on imported foods have been largely dismantled as required by WTO negotiations. Howe ver, higher customs duties ar e still employed for sixteen fresh vegetables, and certain processed fruits, meats and sugar products reflecting their domestic sensitivity. Under CARICOM agreements, agri cultural imports from member countries face zero tariffs. It is assumed in QnD:Jamaica th at imported vegetables and poultry are the only foods that experience a border dut y. These tariffs are set at 200% and 86% respectively in the base scenario. Average smallest (sell 10%) Midpoint (sell 35%) Average national (sell 50%) Average largest (sell 75%) Absolute largest (sell 90%) Root crops 9.0 19.0 28.0 90.0 132.0 Vegetables 6.0 10.5 15.0 27.0 40.0 Legumes 0.18 0.54 0.9 2.19 3.6

PAGE 109

109 Marketing costs As places with lower demand due to populat ion or income have smaller and less diversified markets, it is safe to assume certain features of parish level marketing costs based on the surrounding demographic features (Norton and Symanski, 1975). The costs associated with rural marketing are based on rural population de nsity while urban marketing costs are based on the existence and size of a central parish mark et. Likewise, domestic and imported foods are assumed to have different marketing costs beca use of the differing characteristics of their distribution systems. The DMarketingCostFactor for imported foods in rural areas is based on a linear relationship established by assuming a 50% price markup in areas with the highest rural population density (Saint Andrew/K ingston) and a 200% price mar kup in areas with the lowest rural population density (Trela wny). Urban imported foods DMarketingCostFactors were estimated by associating the largest market, Co ronation Market (easily accessed by both Saint Catherine and Saint Andrew/Kingston parishes) with a 50% price markup and other parishes containing large markets with 60% price markups While imported foods prices in parishes without large urban markets have 75% marketing mark ups. Domestic food in rural areas does not incur a marketing cost and in urban areas it experiences only 1/8 of that areas imported foods marketing cost. The DMarketi ngCostFactors are shown in Table 5-6.

PAGE 110

110 Table 5-6 Marketing cost factors. Rural Urban Parish Rural population density (capita/km) Major market Domestic foods Imported foods Domestic foods Imported foods St. Andrew 246.5 Coronation 1.0 1.5 1.06 1.5 St. Thomas 91.7 1.0 1.9 1.09 1.75 Portland 77.6 Port Antonio 1.0 1.9 1.075 1.6 St. Mary 150.2 Ocho Rios 1.0 1.8 1.075 1.6 St. Ann 103.7 Browns Town 1.0 1.9 1.075 1.6 Trelawny 68.1 1.0 2.0 1.09 1.75 St. James 146.1 Montego Bay 1.0 1.8 1.06 1.5 Hanover 138.5 1.0 1.8 1.09 1.75 Westmoreland 132.3 Negril 1.0 1.8 1.075 1.6 St. Elizabeth 104.4 Black River 1.0 1.9 1.08 1.65 Manchester 155.5 Mandeville 1.0 1.8 1.075 1.6 Clarendon 215.5 May Pen 1.0 1.6 1.075 1.6 St. Catherine 115.0 Coronation 1.0 1.9 1.06 1.5 Transport costs The model developed by Ruijs et al. to examine food transport costs across different regions of Burkina Faso guided QnD:Jamaicas food transport pricing st ructure (2004). That model estimated graduated per ton per kilometer in creases over main highway transport costs of 75% for less busy paved roads, 150% for unpaved ro ads, and 200% for dirt roads. During the rainy season a further 20% and 60% increase was levied on unpaved and dirt road transport costs respectively. In QnD:Jamaica, estimated food transport distances were established for each parish. Domestically produced foods were assumed to be transported 15 km for both rural and urban markets in all parishes. The distance that imported foods ar e transported varies by parish according to the average number of kilometers from its closest importing port (Kingston or Montego Bay). Each distance was then multiplied by the fuel cost per kilometer in that time period which is a function of world oil prices. Th e resulting fuel cost al ong with the marketing

PAGE 111

111 cost associated with each food type in each geographic region is then added to the national farmgate price. In QnD:Jamaica the quality of road was quantified for each parish by calculating the proportion of Class C road le ngth to total road length.10 This factor, DTransportInfrastructureIndex, was set to 1.0 for all urban areas as well as the rural Saint Andrew/Kingston area which has the lowest pe rcentage of Class C roads at 52%. A linear relationship was established by asso ciating the parish with the lowe st proportion with an index of 1.0 and the highest with 1.1. This index represents a ra nge of zero to ten pe rcent increase in food prices already containing marke ting and fuel costs due only to the quality of road in a given parish (Table 5-7). The base i ndex is increased by five to one hundred percent when road quality is temporarily reduced by hurricanes or floods de pending on the strength of the natural hazard. Table 5-7 Transport infrastructure index. DTransportInfrastructureIndex Parish Proportion of Class C to total roads Average distance to import port Rural Urban St. Andrew 0.52 0 1 1.0 St. Thomas 0.82 50 1.1 1.0 Portland 0.61 98 1.03 1.0 St. Mary 0.58 35 1.02 1.0 St. Ann 0.53 60 (Either Port) 1 1.0 Trelawny 0.61 40 (Montego Bay) 1.03 1.0 St. James 0.83 0 (Montego Bay) 1.1 1.0 Hanover 0.60 30 (Montego Bay) 1.03 1.0 Westmoreland 0.64 45 1.04 1.0 St. Elizabeth 0.79 80 1.09 1.0 Manchester 0.72 98 1.07 1.0 Clarendon 0.61 74 1.03 1.0 St. Catherine 0.75 15 1.07 1.0 10 The Ministry of Transport characteri zes 80% of Class C roads as in poor condition. Rural areas are mostly comprised of Class C roads.

PAGE 112

112 The DRoadFloodDamage index is determined w ith a relationship function that uses the DMonthlyRainfallRatio as the inde pendent variable. The relative ra infall in a given parish is the ratio of that months total rainfall to the thirty ye ar average rainfall in that month. A ratio of less than 1.3 (30% above average rainfall) does not in cur a damage factor. A ratio of 1.3 sets the road damage index at 1.05. The index increases linearly until the relative rainfall is 3.0 (300% above average rainfall, indicating a major flood) at which point the damage index is 1.1. If the relative rainfall is 4.0 the associated damage i ndex is 1.5. The flood damage index is multiplied by the infrastructure index. In this way floods in areas where the ro ads are already of poor quality (as represented by a high transport in frastructure index) compound the already high transportation costs. Jamaican Consumer Representation Both economic and nutritional limitations guide the food purchase process logic in QnD:Jamaica. Nutritional demands are set by typical dietary intakes as represented in five major food groups. Budget limitations are guided by historical household expenditure data. Additionally, substitutability due to price or cu ltural preferences structures the food purchase process. It is a fair assumption that perceived value, rather than brand recognition or domestic loyalty, is the most important decision in Ja maican consumers food buying decisions especially for the lower income groups (USDA, 2003; Timm er, 2000). Thus dietary outcomes in Jamaica are foremost a function of food prices relative to income. However, other preferences are important as well as the increasing role these preferences have in higher income echelons. In the case of carbohydrates, the traditional domestic starch, the yam, is actually in competition with cereals, rice, and potatoes for the staple food slot in the Jamaican diet. Rice has long been a staple of Jamaican food culture although it has always been largely imported.

PAGE 113

113 Imported cereals and potatoes, on the other hand, have only recently become staples in the Jamaican diet and their volume per capita has been on the rise. These foods are usually imported as refined wheat and corn flours, and processed bread and pasta products for the cereal group and as frozen potatoes used largely for French fries. All of these imported products generally have much lower nutritional values than domestic yams. Thus, the competition that yams face from cereal s and potatoes has major implications for the Jamaican diet in terms of nutritional characte ristics and cultural value not to mention the Jamaican small farmer. Because of these nutritional effects incurred from the price competition between different products in similar roles, all dietary choices were simplified into five food group categories: carbohydrates, vegetables, legumes, meat and fish. The average Jamaican diet is composed of a spread of the five categories, each of which can be satisfied by two different but substitutable food types distinguished by locati on of origin (domestic or impor ted) and nutritional qualities. By setting minimum required/desired amounts for each food category within the total food basket, it is ensured that the price li ne is not the only factor that in fluences consumer decisions. At this stage in the simulation each populat ion group has earned an income represented as their DPrePurchaseWealthStore and they are re ady use this store to purchase food. The food types that can be purchased are: Carbohydrates: CRootCrops, CImportedStaples Vegetables: CVegetables, CImportedVegetables Legumes: CLegumes, CImportedLegumes Animal Meat: CLivestock, CImportedAnimalMeat Fish: CFish The base preferential consumption basket in te rms of number of servings per food group is shown in Table 5-8. The qualities of each serving are described in Table 5-9.

PAGE 114

114 Table 5-8 Base preferen tial consumption basket. Base preferential basket by food groups Staple carbohydrates 7 servings Vegetables 4 servings Meat 2 servings Legumes 1 serving Calories/domestic daily serving 700 120 538 60 Calories/imported daily serving 1624 120 538 60 Table 5-9 Descriptions of one serving for each food type. Staple carbohydrates Vegetables Animal meat Legumes Size 100g 100g 100g 40g Food type Domestic (yam) Imported (dumplings/ cereal flour) 100gserved =50g flour Domestic (cabbage) Imported (cabbage) Domestic (fried chicken) Import (fried chicken) Calories/ 100g 100 232 34 34 269 269 60 Fat g 0.0 0.7 0 0 15 15 Fiber g 3.7 1.7 2 2 0 0 Sugar g 1.0 10.2 3 3 0 0 Protein g 2.0 5.3 1 1 29 29 Unrealistic amounts of money coul d still be spent trying to fulfill the minimum of a certain category requirement in some situations. For in stance the price of vege tables could increase above a threshold (and supply was still at a level where a significant amount could be purchased) where consumers in reality would not buy any. To account for these situations a maximum expenditure for each category was set. This is based off maximum hi storical expenditure percentiles by commodity group. These ge nerally came from 1992 data, when both food expenditure as a percent of total budget as well as the annual pe rcent increase in the food and drink CPI (77.30%) was at its highest for the study period (JSLCs 1990-2003).11 11 Interesting exceptions include greater proportion spent on starches in 1990 compared to 1992. It appears that this was largely compensated by a decrease in cereal purchases in the lowest quintiles and by a cut back in meals away from home by the middle class. Fruit and vegetable expenditures in 2003 comprised greater shares of the budget than in 1992 for all regions.

PAGE 115

115 The maximum food expenditures as a percenta ge of total consumption by region were rural=65%, urban=60% and Ki ngston/Saint Andrew (KSA)=55%. Table 5-10 shows the maximum expenditures by food cate gory used in QnD:Jamaica. Table 5-10 Maximum food group e xpenditures by geographic region. KSA Towns Rural Meat & dairy 31% 30% 39% Carbohydrates 16% 18% 24% Fruits & vegetables 14% 10% 9% Meals away 24% 26% 20% Other 15% 16% 8% The maximum all food expenditure is declared as the DFoodBudgetFactor for each population group based on their regional lo cation (rural, urban or KSA). Likewise, the maximum percentage of food budget allotted to each f ood group is declared for each population group as DMax FoodGroup BudgetFactor. These factors are used to constrain the desi red purchase within reasonable expenditure limits. The DFoodBudgetFactor is multiplied by the population groups DPrePurchaseWealthStore to arrive at the DM axFoodBudget. This budget is then appropriated to DMax FoodGroup Budgets. These are further divided into Max FoodType Budgets by multiplying the DMax FoodGroup Budget by the related food types expenditure factor. The different transportation and marketing factor s present in rural and urban areas result in different regional/habitat level f ood prices in a given parish. Therefore, the regional price differentials, DRural FoodGroup PriceRatio and DUrban FoodGroup PriceRatio, between local and imported FoodTypes are calculated separately: 1) DHabitat Consumer ImportedFoodType Price / D Habitat Consumer DomesticFoodType Price = D HabitatFoodGroup PriceRatio

PAGE 116

116 The resulting D HabitatFoodGroup PriceRatio is used in a simple linear PRelationship function to determine the percentage of a FoodGroups demand that will be filled by its associated local FoodType (Table 5-11). Table 5-11 Food type expenditure factor. X Y D HabitatFood GroupPriceRat io D DomesticFoodTypeExp enditureFactor Notes 0 .25 ImportedFoodTypePrice is anything <50% of the DomesticFoodT ypePrice in same FoodGroup .5 .25 ImportedFoodTypePrice is half of the DomesticFoodTypePrice in same FoodGroup 1 .5 ImportedFoodTypePrice is equal to the DomesticFoodTypePrice in same FoodGroup 2 .75 ImportedFoodTypePrice is twice of the DomesticFoodTypePrice in same FoodGroup >2 .75 ImportedFoodTypePrice is more than twice of the DomesticFoodT ypePrice in same FoodGroup The above relationship function assumes that the consumption amount of a particular FoodType is directly proportional to its price relative to substitut able FoodTypes within a given FoodGroup. These are the values currently in the ComponentDetails for all food groups except vegetables. The current relations hip for vegetables describes a le ss elastic price responsiveness. No matter how much cheaper impor ted vegetables are than domestic ones, imported vegetables will never make up more than 55% of the vege tables budget as opposed to a maximum of 75% for the other imported food types. The relationship could be modified to s how either a preference for domestic or imported foods within a food gr oup when the price ratio equals one or other tipping points of interest.

PAGE 117

117 Having accounted for consumer price responsiveness with the above relationship, consumer dietary demands as captured in the fi ve food groups are then calculated. Basic food group needs are declared in the GLOBAL section of Componen tDetails file as DMonthlyFoodGroup Needs (kg/capita/mo). The tota l needs of each food group for a population group are calculated: 3) DTotalMonthly FoodGroup Needs = DMonthly FoodGroup Needs X DPopulation X DAverageFamilySize The unsold homegrown food available to la ndowning population groups is then accounted for: 4a) D FoodType Store = DPrePurchase FoodType Store 4b) DTotalMonthly FoodGroup Needs DPrePurchase FoodType Store = D FoodGroup PurchaseDemand Next, the total demand for each FoodGroup is di stributed into its as sociated local and imported FoodType demands using the above expe nditure factor based on price differences between substitutable products: 5a) DDesired(Domestic) FoodType Purchase = DFoodGroup PurchaseDemand X D(Habitat) FoodType ExpenditureFactor 5b) DDesired(Imported) FoodType Purchase = D FoodGroup PurchaseDemand DDesired(Domestic) FoodType Purchase The cost of each desired amount of food is then calculated by multiplying it by its associated price. 6) DDesiredFoodType PurchaseDemand X D Habitat Consumer FoodType Price = DDesired FoodType Cost In order to constrain the amount of money spent on a food group, this cost is then compared to the maximum allowed expenditure (described above) for a particular food group12. If the DDesired FoodType Cost is more than the allowed expenditure than the amount to be spent on 12 This accounts for intra FoodGroup substitutability while the ExpenditureF actor accounts for inter FoodGroup substitutability.

PAGE 118

118 the food type is instead set to the D FoodType Budget. This expenditure, whether DDesired FoodType Cost or D FoodType Budget, is tracked as DMoneyPaidFor FoodType In case the DDesired FoodType Cost (and the associated DDesired FoodType Purchase) was unreasonably large and constraine d by the maximum budget, the real amount to be purchased must be recalculated: 7) D FoodType Purchase = DMoneyPaidFor FoodType / D Habitat Consumer FoodType Price This amount is then transferred from the DNational FoodType Store into that populations D FoodType Store and the DMoneyPaidFor FoodType is transferred from their DWealthStore into the CUrbanMiddleFolks DWealthStore Consumption Testing Figure 5-56 shows the large geographical spread in the imported vs. domestic carbohydrates price ratios generated by the marginal marketing and transportation cost factors. Also shown is the decrease in this ratio over time, which agrees with historical evidence, indicating that imported food pri ces have decreased relative to domestic food. Figure 5-57 shows the simulated annual totals of importe d vegetables purchased vs. the known imported levels. This food type shows more realistic simulation amounts than im ported staples levels shown in Figure 5-58.

PAGE 119

119 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90Jan-92 Jul-92 Jan-93 Jul-93 Jan-94 Jul-94 Jan-95 Jul-95 Jan-96 Jul-96 Jan-97 Jul-97 Jan-98 Jul-98 Jan-99 Jul-99 Jan-00 Jul-00 Jan-01 Jul-01 Jan-02 Jul-02YearRural Price Ratio (Imported to Domestic NominalJ$ ) Max Price Ratio Min Price Ratio Figure 5-56 Maximum and minimum rural carbohydrate price ratios across parishes. 0 2000000 4000000 6000000 8000000 10000000 12000000 14000000 16000000 180000001992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002YearImported Vegetables Purchased (Kgs) Known Simulated Figure 5-57 Simulated vegetable purchases.

PAGE 120

120 0 50000000 100000000 150000000 200000000 250000000 300000000 350000000 400000000 450000000 500000000 199219931994199519961997 YearImported Staples Purchased (Kgs) Known Simulated Figure 5-58 Imported staples purchased.

PAGE 121

121 CHAPTER SIX AN ILLUSTRATIVE SCENARIO SIMULATION The following is a discussion of a simulation ex ercise completed with QnD:Jamaica. It serves as an illustration to guide and inspire futu re scenario simulations. In this exercise only one external parameter is manipulated while al l other components are held constant. This approach is one traditional method in scenario modeling and can be understood as an overlay of one new quality against a known historical backgr ound. In this way, this simulation exercise is essentially asking, What if one external change of concern had occurr ed during this known historical period? To expand this exercise, one internal para meter was also manipulated for a total of four scenarios including the unmodified base histor ical one. Changing these two parameters only allows one to perform a coarse se nsitivity analysis. As QnD:Jamaica is in its relatively early stages of developm ent it is important to isolate influences to really examine the models responsiveness. Future simulations using QnD:Jamaica will move beyond this approach and describe a variety changing parameters in complex scenario descriptions. Biofuel Demand and Rising Corn Prices There has been much recent discussion surr ounding the ramifications of rising corn and soybean prices and the role of growing et hanol biofuel demand in these increases13. The recent rise in corn tortilla prices in Mexico and the resulting food security effects has been well documented (Roig-Franzia, 2007). In the US, th ere has been a 70% incr ease in poultry feed prices over the past six months resulting in a 14% incr ease in broiler prices over the same period (Sauser, 2007). 13 It is important to note that rising corn prices are likely an emergent property of other feedbacks in the global system besides ethanol demand. Potential activities of influence include droughts on multiple continents and rising international wheat costs also.

PAGE 122

122 In Jamaica, these price increases are also a major concern for the poultry industry, as much of the feed is imported corn (Edwards, 2007). The Jamaican consumers chicken prices have steadily increased over the past year to compen sate for the increased production costs (Thame, 2007). The higher costs of this im portant source of protein in the Jamaican diet will likely have important nutritional effects on th e population. In addition to th ese most direct food security concerns, downturns in the larg e poultry industry would have ri pple effects across the economy. It is important to note that these increases are occurring against a backdr op of historically low corn and feed stuff prices. While ethanol produc tion is a new element in corn prices and its effects on value is a large con cern, the diminished prices th roughout the 1990s and early 2000s could arguably be considered unrealistically low. Jamaican corn-based feed prices increased by 25% in 2006 and are expected to increase another 35% by the end of 2007 (Thame, 2007). These values were used to estimate an external feedstuff price driver file that was used in pl ace of the original times eries file containing known historical prices. This file describes a linear f eed price increase of 25% per year from the base price in 1992. To expand upon this simulation, th e internal parameter describing consumers ideal poultry demand total was increased from 3kgs/capita/month to 4kgs /capita/month in two scenarios. This represents changing pr eference, or utiliz ation, patterns. Historical Scenario : 1990s feedstuff prices and hist orical 3kgs/capita/month livestock desired utilization Scenario 1 : 25% annual increase in feedstuff pr ices and historical 3kgs/capita/month livestock desired utilization Scenario 2 : 25% annual increase in feedstuff pric es and increase in desired livestock utilization to 4kgs /capita/month Scenario 3 : Historical feedstuff prices and increa se in desired livestock utilization to 4kgs /capita/month

PAGE 123

123 Discussion of Scenario Simulations National domestic livestock production volumes purchase volumes and consumer desired purchase volumes are the three output parameters examined in this discussion of the four scenario runs. These subjects were chosen because they embody the key qualities of food security: availabili ty, accessibility and utiliz ation. Total production amounts are described by the Available curve in the figures below. In thes e scenarios, the maximum desired amounts never exceeded the maximum budgeted amounts, thus, th e Accessible curve describes the demand for domestic livestock amount and is unlimite d by budget constraints. The Purchase curve shows what was actually bought by consumers. Al so shown in the figures is the known real annual domestic livestock production volume as a reference (Figures 6-1, 6-2, 6-3 and 6-4). In the historical scenario, the available amount is similar to the historically known production level, especially in th e first five years of simulation. This scenario is showing an available level that does not meet the national needs and purchasing ability in the first four years. After 1996, however, accessibility becomes the limiting factor with purchased amounts declining below available levels as well as earlier purchase levels. Scenario 1 clearly shows a lower level of available chicken than the hi storical scenario. This is ex pected do to the rising costs of production incurred by the increased feed costs. Acce ssibility is similar to the historical scenario during the early years of simulation, but is lowe r in later years because the rising costs of production are being passed on to the consumer. These increased domestic chicken prices are causing the relative price between domestic and imported chicken to be greater which is the models mechanism to limit consumers domestic purchasing power. This decreased accessibility is especially apparent in years 5, 7, 8, 9, 10, and 11 of Scenario 1. However, lowered domestic availability is still a greater li mitation than accessibility in all months except a few in year 7 and most months in year 9.

PAGE 124

124 0 1000000 2000000 3000000 4000000 5000000 6000000 7000000 80000001992 1993 1994 1 995 1996 1997 1998 1999 2000 2001 2002YearDomestic Poultry Volume (Kgs ) Available Accessible Purchased Known Availabilit y Figure 6-1 Historical sc enario of domestic livestock production and demand. 0 1000000 2000000 3000000 4000000 5000000 6000000 7000000 8000000 1234567891011YearDomestic Livestock (Kgs) Available Accessible Purchased Known Production Figure 6-2 Scenario 1 of domes tic livestock production and demand. In Scenario 2, accessibility is always greater than the low am ounts of available chicken. If poultry production decreased to leve ls simulated in this scenario, there would be a strong amount of unmet demand which would have strong implica tions for protein intake and shifts in diet composition. Scenario 3 shows purchasing levels most similar to known historical production amounts indicating that the Jamaican consum ers poultry demand is likely closer to 4kgs/capita/mo than 3kgs/capita/mo.

PAGE 125

125 0 1000000 2000000 3000000 4000000 5000000 6000000 7000000 8000000 1234567891011YearDomestic Livestock (Kgs) Availabile Accessibile Purchased Figure 6-3 Scenario 2 of domes tic livestock production and demand. 0 1000000 2000000 3000000 4000000 5000000 6000000 7000000 8000000 9000000 1234567891011YearDomestic Livestock (Kgs) Available Accessible Purchased Figure 6-4 Scenario 3 of domes tic livestock production and demand. Figure 6-5 shows national base dom estic livestock prices as well as the regional prices for rural and urban habitats in two pa rishes. It is not surprising th at the urban area in Saint James generally shows the lowest prices as it has the lowest marketing coefficient being closest to a large market, Montego Bay. The sharp price increas es throughout the scenar io in all regions are due to transport infrastructure shocks caused by floods It is interesting to note that there is never an event where one of these parishs habitat ar eas, either rural or urban, experiences a spike without the other habitat area.

PAGE 126

126 0 20 40 60 80 100 120 19921993199419951996199719981999200020012002YearDomestic Livestock Price (Nominal J$/Kg) National Saint James Rural Saint James Urban Saint Elizabeth Rural Saint Elizabeth Urban Figure 6-5 Regional price shocks to dome stic livestock in historical scenario. Figure 6-6 shows a technical error in QnD:Jamaica. The purchase amount process is using the budget data value from the previous month ra ther than the current months data value as designed. This is resulting in what appears to be overly dynamic purchase output. Figure 6-7 also shows the Saint James farmer populations purchase amounts of imported and domestic chicken alongside their total pou ltry demand/needs. The output data shown in this graph has been modified to remove the noise errors that QnD:Jamaica produced. However, the most important results can be seen in both figures. The model is showing purchasing observed in the historical data records: decr easing domestic poultry purchases and increasing imported poultry purchases. This similarity to known trends is ev idence that builds confidence in the mechanisms employed in QnD:Jamaica.

PAGE 127

127 0 10000 20000 30000 40000 50000 60000 70000 80000 19921993199419951996199719981999200020012002Kgs Needs Domestic Purchased Imports Purchased Total Figure 6-6 Internal structural design noise. 0 10000 20000 30000 40000 50000 60000 19921993199419951996199719981999200020012002YearKgs Desired Domestic Livestock Desired Imported Livestock Total

PAGE 128

128 Figure 6-7 Domestic and imported livestock purchases of Saint James farmers in historical scenario. Figure 6-8 describes the volume of imported animal meat purchased in the various scenarios. As expected, the pur chased import amount is lowest in the historical scenario when domestic production was re latively high, prices rela tively low and demand only 3kgs/capita/month. Similarly when the preference parameter is se t at the lower value, imported livestock is still relatively low, but higher than in the historical scenario because domestic livestock prices were relatively higher compar ed to imported animal meat prices. Imported purchase quantities are generally highest in the second scenario when demand is high yet domestic production shows decreasing output due to higher feedstuff prices. In the first five years of the scenarios, imported purchased do not differ greatly from the amounts produced in the historical scenario. This is likely a result of the combina tion of domestic livestock prices relative competitiveness to imported prices in the early years as well as relatively small values in real internal feedstuff prices despite the great external increases. 0 2000000 4000000 6000000 8000000 10000000 12000000 1234567891011TimeImported Animal Meat Purchased (tons) Historical Scenario Scenario 1 Scenario 2 Scenario 3 Figure 6-8 Total imported animal meat. The changing demand for imported animal meat as domestic livestock prices and supply change as illustrated in the above scenarios indicates the need to consider the potential qualities

PAGE 129

129 of the imported animal meat in the event of a gl obal feedstuff increase. As this situation is occurring in the global food system not just in the Jamaican contex t, at least one other immediate external parameter is important to this scenar io exploration. Because Jamaica imports a fair share of its total poultry suppl y, around 25% annually throughout the 1990s, external poultry prices are important in the Jamaic an food system. The rising cost of corn is also being passed onto US chicken prices (Sauser, 2007). The next step in this scenario exercise should be to examine the combined effects of rising imported poultry prices would have in combination with rising feedstuff prices.

PAGE 130

130 CHAPTER 7 CONCLUSIONS Modeling Jamaican Food Systems This thesis represents the second developmen tal cycle in what hopes to be a chain of constructive model design and learning. The firs t iteration was complete d during the summer of 2005 during multiple workshops led by GECAFS. During the phase of model development documented in this thesis, the structure of the base version was fleshed out into greater detail. Discussions in the field revealed a model structure that aims to depict the Jamaican food system in a manner most appropriate to those interests mo st involved in the situation. In depth empirical research guided parameterization of the model as many of the previously desired relationships and data were not well documented enough to support simulation. There are many potential future applications of QnD:Jamaica. The most obvious is greater exploration of consumption activi ties in terms of the availability and accessibility of food. The assumptions and effects of the current expenditure factor and price ratio processes need to be explored further to determine their necessity and/or implications. Because changing food preferences is a key concern of Jamaica food security one of the users management options could be to modify the proporti ons of desired foods within a consumers food basket, i.e. increase daily vegetables demand from 400g to 500g. This represents desired effects of nutritional education policies or cultural changes. Additionally an ideal preference distribution based on the CFNIs recommended calorie source distribution by food category could be used to guide this parameter. Each months consum ption basket could measured against these proportions to show how simulated nutriti on levels compare to desired levels. Another obvious next direction for the model is the developmen t of parameters based more directly on natural resources. These should incl ude such indices as fo rest health (i.e. canopy

PAGE 131

131 cover, biodiversity), fisheries h ealth (reef diversity, fish stocks), erosion indices, water quality (availability, stormwater and agri cultural runoff), etc. Another necessary addition to the model is development of more internal feedback loops (rathe r than so many driver files) such as farmers land use decisions, internal migration flows, a nd tying wages to economic sector indicators. These should also be tied into th e ecological dynamics and indices. Broader Lessons Learned The most useful application of the knowledge gained during this research will be in discussions of more formal guidelines for conducti ng interdisciplinary meso scale research. One of the key struggles faced duri ng this project was in terpreting and manipula ting available macro and micro level data in a manner that revealed a limited number of cross-scalar and crossdisciplinary linkages and systemic f eatures in an effort to remain fairly transparent. The impact of this project has both practical and theoretical applications and advances multiple disciplines separately and the interdisciplinary cause itself. As development planners and the decision support community seek to apply the adaptive management paradigm, practical systematic methods that implement the proposed principles must be developed and tested. Traditionally these two fields have experienced tension due to communication and goal assessment difficulties, among others. This project serves as an example of one innovative method that aims to limit these difficulties. The know ledge gained during this resear ch provides these disciplines with a case study demonstrating an explicitly collaborative, in terdisciplinary and iterative framework for development and implementation of an adaptive management planning tool. A future comparison of this projects site specific evidence with case studies from other food systems, will give insight into general stru ctural patterns and rul es in food systems. These patterns will help food system planners better maneuver through the many unpredictable variables as they formulat e food security policy.

PAGE 132

132 When a holistic definition is applied to food security, it becomes nearly impossible to ignore the previously hidden impacts that th e current international food system has on all socioecological aspects of our global society. Traditionally underrepresented groups will be empowered when a rich definition of food security that necessitates a co mplex systems approach to food management is fostered. The food secu rity evaluation framework that this project promotes has the broadest impact of raising aw areness of the extensive and unseen ramifications of our global food system.

PAGE 133

133 LIST OF REFERENCES Ackoff, R.L. 1974. Redesigning th e Future: A Systems Approach to Societal Planning. John Wiley and Sons, New York, USA. Adger, W.N. 2006. Vulnerability. Global Environmental Change 16:268-281. Alderman, H. 1986. The Effect of Food Price and Income Changes on the Acquisition of Food by Low-Income Households. International Food Policy Research Institute. Allen, C., H. Hall, and W. Robinson. 2002. Estima ting a Small Scale Macroeconomic Model of the Jamaican Economy: Some Preliminary Re sults. Research and Economic Programming Division Bank of Jamaica. Andreatta, S. 1998. Transformation of the ag ro-food sector: lessons form the Caribbean. Human Organization 57 (4) 414-429. Anderson, P., and M. Witter. 1994. Crisis, adju stment and social ch ange: a case study of Jamaica, Pages 1-55 in E. Le Franc, editor. Consequences of Structural Adjustment: A Review of the Jamaican Experience. Canoe, Kingston. Ballayram and C. Davis. 2003. Economic reforms and agricultural supply response in Jamaica. Farm and Business: The Journal of the Caribbean Agro-Economic Society 6(1):1-20. Ballayram and F. Henry. 2004. The Importance of Integrating Food a nd Nutrition Security Plans in the Economic Strtegy of the Region. Paper presented at the CARDI/CTA/CFNI Regional Seminar, Belize City. Baker, J., H. Jenkins, and H. V. Bonavita. 1999. Basics steps in scenario planning creating alternative futures, Curtin University of Technology, Perth, Western Australia, 1-51. Beckford, G. 1972. Persistent Poverty: Underd evelopment in Plantation Economies of the Third World. New York, USA: Oxford University Press. Behrman, J. 1995. Micronutrients and household behavior: what we know and what we dont know. Working Paper Series on Agricultural Strategies for Micronu trients Number 2. International Food Policy Research Institute, Washington D.C. Berkes, F., J. Colding, and C. Folke, editors. 2003. Navigating social-ecological systems: building resilience for complex ity and change. Cambridge Un iversity Press, Cambridge, UK. Bernanke, B., and M. Gertler. 1995. Inside th e Black Box: The Credit Channel of Monetary Policy Transmission. Journal of Economic Perspectives 9 (4). Berry, W. 2002. The whole horse: the preserva tion of the agrarian mind. Pages 7-19 In A. Kimbrell, editor. Fatal harvest: the tragedy of industrial agriculture.

PAGE 134

134 Bouis, H., L. Haddad, and E. Kennedy. 1992. Does it matter how we survey demand for food? Food Policy 17(5):349-360. Bouis, H., L. 1994. The effect of income on demand for food in poor countries: are our databases giving us reliable estimates ? Journal of Development Economics 44(1):199-226. Bourne, C. 1977. Inflation in the Caribbean. Institute of Social and Economic Research, Jamaica. Brown, H. A. 1994. Economics of Disasters with Special Reference to the Jamaican Experience. University of the West Indies Centre for Environment and Development, Jamaica. Bunch, M.J. 2003. Soft Systems Methodology and ecosystem approach: A system study of the Cooum River and environs in Chennai, India. Environmental Management 31(2): 182-197. Capra, F. 1996. The Web of Life : A New Scientific Understandi ng of Living Systems. Anchor Books, New York, USA. Caribbean Disaster Emergency Response Agency (CDERA). 2003. Status of Hazard Maps Vulnerability Assessments and Digital Maps: Jamaica Country Report. CDERA, Jamaica. Carpenter, S., W. Brock and P. Hanson. 1999. Ec ological and social dynamics in simple models of ecosystem management Conservation Ecology 3(2):4. [online] URL: http://www.consecol.org/vol3/iss2/art4/ Checkland, P.B. 1999. Soft System s Methodolog y in Action. John Wiley & Sons, Toronto. Checkland, P.B. 1979. Techniques in soft sy stems practice part 2: building conceptual models. Journal of Applied Systems Analysis 6:33. Checkland, P.B., and S. Holwell. 2004. Cla ssic OR and soft ORand asymmetric complementarity. Pages 45-60 in M. Podd, editor. Systems modelling: theory and practice. John Wiley and Sons, UK. Checkland, P.B. and J. Scholes. 1990. Soft Systems Methodology in Action. John Wiley & Sons, Toronto. Churchman, C. W. 1979. The Systems Approach. New York, USA: Laurel. Clarke, C., and D. Howard. 2006. Contradictor y socio-economic consequences of structural adjustment in Kingston, Jamaica. The Geographical Journal 172(2):106-129. Corral, L., P. Winters, and G. Gordillo. 2000. Food insecurity and vulnerability in Latin America and the Caribbean. Working Pape r Series in Agricultural and Resource Economics. University of New England.

PAGE 135

135 Costanza, R., J. Audley, R. Borden, P. Ekins, C. Folke, S.O. Funtowicz, and J. Harris. 1995. Sustainable tradea new paradigm for world welfare. Environment 37(5):16-20. Cox, P., 1996. Some issues in the design of ag ricultural decision support systems. Agricultural Systems 52(2-3):355-381. Craigwell, R. 1991. Wages and prices: what causes what? Social and Economic Studies 40(1):93-103. Davis, F. D. 1989. Perceived usefulness, pe rceived ease of use a nd user acceptance of information technology. MIS Quarterly 13:319-340. Davis-Morrison, V., and D. Barker. 1997. Resour ce management, environmental knowledge, and decision making in small farming systems in the Rio Grande valley, Jamaica. Caribbean Geography 8(2):96-106. Douthwaite, B., N. C. de Haan, V. Manyong, a nd D. Keatinge. 2001. Blending hard and soft science: the follo w-the-technology approach to catalyzing and evaluating technology change. Conservation Ecology 5(2):13 [online] URL: http://www.consecol.org/vol5/iss2/art13/ Duchin, F. 2005. Sustainable consumpti on of food. Journal of Industrial Ecology 9(1-2):99 114. Dyball, R., S. Beavis, and S. Kaufman. 2005. Complex adaptive systems: Constructing mental models. Pages 41-62 in M. Keen, V.A. Brown a nd R. Dyball, editors. Social Learning in Environmental Management: Towards a Sust ainable Future. Ea rthscan: London, UK. Dyer, G. A., S. Boucher, and J. E. Taylor. 2006. Subsistence response to market shocks. American Journal of Agricultural Economics 88(2): 279-291. Eakin, H. 2005. Institutional change, climate ri sk, and rural developmen t vulnerability: cases from central Mexico. World Development 33(11):1923-1938. Economic Comission for Latin Ameruca and the Ca ribbean (ECLAC). 2004. Assessment of the Socioeconomic and Environmental Impact of Hurricane Ivan on Jamaica. Eden C., and F. Ackermann. 2004. Use of soft OR models by clients-what do they want from them? Pages 146-163 in M. Pidd, editor. Syst ems modelling: theory and practice. John Wiley and Sons, England. Edwards, A. February 23, 2007. Cost of importe d grain impacts Jamaica Broilers profits. The Jamaica Observer. Edwards, D. 1961. An Economic Study of Small Farming in Jamaica. Institute of Social and Economic Research: Mona, Jamaica.

PAGE 136

136 Ericksen, P.J. 2006a. Conceptualizing Food Systems for Global Environmental Change Research. Global Environmental Cha nge and Food Systems: Wallingford, UK. Ericksen, P.J. 2006b. Assessing the Vulnerability of Food Systems to Global Environmental Change: A Conceptual and Methodological Re view. Global Environmental Change and Food Systems: Wallingford, UK. Espeut, P. December 10, 2003. The drama of Christmas vending. The Jamaica Gleaner. FAO. 2003. Nutrition Country Profile: Jama ica. Food and Agriculture Organization of the United Nations. FAO, FAOSTATS, 2007, http://www.faostats.org accessed o n 3/07. Folke, C. 2006. Resilience: The emergence of a perspective for soci al-ecological systems analysis. Global Environmental Change 16:253-267. Ford, J. R. 1992. Guyanas food performance in a Caribbean context: lessons for food security policy. Food Policy October:326-336. Forrester, J. W. 1961. Industrial dyna mics. MIT Press, Cambridge, M.A. Fraser, E.D., W. Mabee, and F. Figge. 2005. A framework for assessing the vulnerability of food systems to future shocks. Futures 27:465-479. Goldsmith, A. 1981. The politics of rural stagnation: development policy, local organizations and agrarian change in Jamaica. UMI, Ann Arbor, Michigan. Gordan, S. November 24, 2006. Competing supermar ket chains still in expansion mode. The Jamaica Gleaner. Gunderson, L H., C. S. Holling, and S.S. Light. 1995. Barriers and bridges to the renewal of ecosystems and institutions. Columbia University Press, New York, USA. Handa, S., and D. King. 2003. Adjustment with a human face? Evidence from Jamaica. World Development 31(7):1125-1145. Holling, C.S. 1973. Resilience and stability of ec ological systems. Annual review of Ecology and Systematics 4 :1-23. Holling, C.S., editor. 1978. Adaptive environm ental assessment and management. Wiley, London, UK. Holling, C.S. 1986. The resilence of terrestrial ecosystems:local surprise and global change. Pages 292-317 in W. C. Clark and R.E. Munn, editors. Sustainable development of the Biospere. Cambridge University Press, UK.

PAGE 137

137 Holling, C.S. 1996. Engineering resilience versus ecological resilience. Pages 31-44 in P. Schulze, editor. Engineering within ecologi cal constraints. National Academy Press, Washington, D.C., USA. Holling, C. S., and L. H. Gunderson. 2002. Resilien ce and adaptive cycles. Pages 25-62 in C.S. Holling and L. H. Gunderson, editors. Panarc hy: Understanding transformations in human and natural systems. Island Press, Washington D.C. Hourts, H., and D. Miller. 2006. The cell phone : and anthropology of communication. Berg Press, New York, USA. Ishemo, A., H. Semple, and E. Thomas-Hope. 2006. Population mobility and the survival of small farming in the Rio Grande Valle y, Jamaica. The Geographical Journal 172(4):318330. Ison, R.L., P.T. Maiteny, and S. Carr. 1997. Sy stems methodologies fo r sustainable natural resources research and developm ent. Agricultural Systems 55(2):257-272. Jamaican Meteorological Soci ety. Parish Rainfall, 2007, http://www.metservice.gov.jm/ accessed on 2/07. Jam aican Ministry of Agriculture (MOA). 2001. Update on the State of Food Security in Jamaica. Ministry of Agriculture, Kingston, Jamaica. Jamaican Ministry of Agriculture (MOA). 2005. Agricultural Developmen t Strategy. Ministry of Agriculture, Kingston, Jamaica. Jamaica Sustainable Development Network (JSDNP). Jamaica Sustainable Development Networking Programme. 2005, http://www.jsdnp.org.jm/JSDNPdotORGdotJM/home.htm accessed on 10/06. Janssen, M. A., O. Bodin, J.M. Anderies, T. Elmquist, H. Ernstson, B. McAllister, P. Olsson, and P. Ryan. 2006. Toward a network perspec tive of the study of resilience in socioecological systems. Ecology and Society 11 (1). Kasperson, J.X. and R.E. Kasperson. 2001. Gl obal Environmental Risk. Earthscan and UNU Press, London, UK. Keen, M. and S. Mahanty. 2005. Collaborative l earning: bridging scales and interests. Pages 104-122 in Keen, M., V.A. Brown and R. D yball, editors. Social Learning in Environmental Management: Towards a Su stainable Future. Earthscan, London. Kiker, C.F, J.W. Milon, and A.W. Hodges. 2003. Adaptive learning from science-based policy: the Everglades restoration. Ecological Economics 37:403-416. Kiker, G.A., T.S. Bridges, I. Linkov, A. Varghe se, and T. Seager. 2006. Application of multicriteria decision analysis in environmenta l decision-making. Integrated Environmental Assessment and Management 1(2):1-14.

PAGE 138

138 Kiker, G.A. and I. Linkov. 2006. The QnD m odel/game system: integrating questions and decisions for multiple stressors. in Arapis, G., N. Goncharova, and P. Baveye, editors. Ecotoxicology, Ecological Risk Assessment, and Multiple Stressors. Springer, Netherlands. King, D. 1998. Adjustment, stabilization, and povert y in Jamaica: 1989-1995. University of the West Indies, Mona, Jamaica. Le Franc, E. 1981. Social structure, land use, and food availability in the Caribbean. Food Nutrition Bulletin 3(4). Le Franc, E., and A. Downes. 2001. Measuring human development in countries with invisible economies: challenges posed by the informal a nd remittance sectors in Jamaica. Social and Economc Studies 50 (1). Lewis, W. A. 1954. Economic development w ith unlimited supplies of labor. Manchester School of Economic and Social Studies 22:139-191. Lyons, M. 2004 Insights from complexity: organizational change and systems modelling. Pages 21-44 in M. Pidd, editor. Systems Mode lling: Theory and Practice. John Wiley and Sons, Chichester, UK. Lyson, C., and J. Green. 1999. The agricultura l marketscape: A framework for sustaining agriculture and communities in the Northeas t. Journal of Sustainable Agriculture 15(2):133-150. Meadows, D.H. 1985. System analysis and th e world food system. European Journal of Operational Research 20:158-167. Meadows, D.H. and J.M. Robinson. 1985. The Electronic Oracle: Computer Models and Social Decisions. John Wiley and Sons, Chichester, UK. Mintz, S. 1974. Caribbean transformati ons. Aldine Publishing, Chicago, USA. Mintz, S., and D. Hall. 1960. The origin of th e Jamaican internal marketing system. Yale University Press, New Haven, USA. Morecroft, J. 2004. Pages 101-126 in M. Pidd, edito r. Systems modelling: theory and practice. John Wiley and Sons, UK. Morgan, M.G., B. Fischhoff, A. Bostro m, and C.J. Atman. 2002. Risk communication: a mental models approach. Cambridge University Press. Nkemdirim, L. C. 1979. Spatial and seasonal di stribution of rainfall and runoff in Jamaica. Geographical Review 69 (3):288-301. Norton A., and R. Symanski. 1975. The inte rnal marketing system of Jamaica. Geographical Review 65(4):461-475.

PAGE 139

139 OBrien, K.L., and R.M. Leichenko. 2000. Double exposure: assessing the impacts of climate change within the context of economic globalization. Global Environmental Change 10:221-232. Odum, H. 1983. Systems Ecology: an Introduction John Wiley, New York, USA. Osei, P. D. 2002. A critical assessment of Ja maicas national poverty eradication programme. Journal of International Development 14:773-788. Ostrom, E. 1999. Coping with tragedies of the commons. Annual Review of Political Science 2:493-535. Pan American Health Organization (PAHO). Emergency Preparedness and Disaster Relief Coordination Program, Washington D.C., US Disaster Reports, 5. 1988. http://www.crid.or.cr/digita lizacion/pdf/eng/doc6803/doc6803.htm accessed on 10/0 6. Parker P., R. Letcher, A. Jakeman, M.B. Beck, G. Harris R.M. Argent, M. Hare, C. Pahl-Wostl, A. Voinov, M. Janssen, P. Sullivan, M. Sc occimarro, A. Friend, M. Sonnenshein, D. Barker, L. Matejicek, D. Odul aja, P. Deadman, K. Lim, G. Larocque, P. Tarikhi, C. Fletcher, A. Put, T. Maxwell, A. Charles, H. Breeze, N. Nakatani, S. Mudgal, W. Naito, O. Osidele I. Eriksson, U. Kautsky, E. Kaut sky, B. Naeslund, L. Kumblad, R. Park, S. Maltagliati, P. Girardin, A. Rizzoli, D. Ma uriello, R. Hoch, D. Pelletier, J. Reilly, R. Olafsdottir, and S. Bin. 2002. Progress in integrated assessme nt and modelling. Environmental Modelling and Software 17:209-217. Pavao-Zuckerman, M.A. 2000. The conceptual util ity of models in human ecology. Journal of Ecological Anthropology 4:37-56. Pidd, M. 2004. Complementarity in systems mode ling Pages 1-19 in M. Pidd, editor. Systems Modelling: Theory and Practice. John Wiley and Sons, England. Pirot, Jean-Yves, P. Meynell and D. Elder. 2000. Ecosystem Management: Lessons from around the world: a guide for development and conservation practitioners. IUCN, France. Planning Institute of Jamaica. 1990-2004. Economic and Social Survey of Jamaica, Kingston. Planning Institute of Jamaica and Statistical Inst itute of Jamaica. 1990-2003. Jamaica Survey of Living Conditions, Kingston. Popkin, B.M. 2003. The nutrition transition in the developing world. Development Policy Review 21(5-6):581-597. Pothukuchi, K and J.L. Kaufman. 2000. The fo od system: A stranger to the planning field. Journal of the American Planning Association 6(2):113-124. Quinn, J. B. & J. Voyer. 1996. Logical incr ementalism: Managing strategy formation. Pages 95-101 in H. Mintzberg and J.B. Quinn, ed itors. The Strategy Process: Concepts, Contexts, Cases. Prentice Hall, Englewood Cliffs.

PAGE 140

140 Rittel, H. and M. Webber. 1973. Dilemmas in a general theory of planning. Policy Sciences 4:155-169. Robinson, W. and J.W. Robinson. 1997. The transm ission mechanism of monetary policy in the Jamaican economy. Annual Conference of the Caribbean Centre for Monetary Studies. Roig-Franzia, M. January 27, 2007. A culinary and cultural staple in crisis. Washington Post. Roux, D.J., K. H. Rogers, H. C. Biggs, P. J. Ashton and A. Sergeant. 2006. Bridging the science-management divide: moving from unidi rectional knowledge tr ansfer to knowledge interfacing and sharing. Ecology and Society 11 (1): 4 [online] URL: http://www.ecologyandsocie ty.org/vol11/iss1/art4/ Ruijs, A., C. Schweigm an, and C. Lutz. 2004. The impact of transportand transaction-cost reductions on foods markets in developing count ries: evidence from tempered expectations for Burkina Faso. Agricultural Economics 31:219-228. Salmon, J. U. 2002. Poverty eradication: the Jamai can experience. Social and Economic Studies 51(4):63-94. Sauser, B. February 13, 2007. Ethanol dema nd threatens food prices. Technology Review. Schipper, L., and M. Pelling. 2006. Disaster risk, climat e change and international development: scope for and challeng es to, integration. Disasters 30 (1):19-38. Schrage, M. 2000. Serious Play: How th e Worlds Best Companies Simulate to Innovate. Harvard Business School Press, Boston, USA. Sen, A. 1980. Poverty and Famines: An Essay on Entitlement and Deprivation. University Press, Oxford. Simon, H.A. 1962. The architectu re of complexity. Proceedings of the American Philosophical Society 106 (6):467-482. Skoufias, E. 2003. Is the calorie-income elasticit y sensitive to price ch anges? Evidence from Indonesia. World Development 31(7):1291-1307. Spence, B. 1996. The influence of small farmers land-use decisions on the status of domestic food security in Jamaica. Caribbean Geography 7(2):132-142. Spence, B. 1999. Spatio-evolutionary model of Jamaican small farming. The Geographical Journal 165 (3):296-305. Stacey, R.D. 2001. Complex responsive processes in organizations: learning and knowledge creation. Routledge, London.

PAGE 141

141 Statistical Institute of Jamaica (STATINJA) 1996. Census of Agriculture. Kingston. Sundkvist, A., R. Milestad, and A. Jansson. 200 5. On the importance of tightening feedback loops for sustainable development of food systems. Food Policy 30:224-239. Thame, C. July 18, 2007. Poultry, bread prices to increase by 15-20%, says FAS. The Jamaican Observer. Timmer, C.P. 1980. Food prices and food policy analysis in LDCs. Food Policy August:188199. Timmer, C.P. 2000. The macro dimensions of food security: economic growth, equitable distribution, and food price stability. Food Policy 25:283-295. USDA Foreign Agricultural Service. 2003. Jama ica and dep Retail food s ector report. Global Agriculture Information Network. van der Heijden, K. 1996. Scenarios. The Art of Strategic Conversation. John Wiley and Sons, Chichester, England. van der Heijden, K., R. Bradfield, G. Burt, G. Ca irns, G. and G. Wright. 2002. The sixth sense accelerating organizational learni ng with scenarios. John Wi ley & Sons Ltd.,Chichester, England. van der Leeuw, S. E., and C. Aschan-Leygonie. 2005. A long term perspective on resilience in socio-natural systems. Pages 227-264 in H. Liljenstrom and U. Svedin, editors. Micro, meso macro: addressing complex systems c ouplings. World Scientific, Singapore. van Bertalanffy, L. 1968. General system theory : foundation, development, applications. George Braziller, New York, USA. van Ittersum, M. K., R. Rabbinge, and H.C. van Latesteijn. 1998. Explor atory land use studies and their role in strategic policy making. Agricultural Systems 58(3):309-330. Walker, A. 2002. The Microstructure of the Ja maican Foreign Exchange Market: Volumes, Volatility and Spreads. Working Paper Monetary Analysis & Programming Dept. Research and Economic Programming Division Bank of Jamaica. Walker, D.H. 2002 Decision support, learning and rural resource management. Agricultural Systems 73:113-127. Walker, B., S. Carpenter, J. Anderies, N. Abel G. Cumming, M. Janssen, L. Lebel, J. Norberg, G.D. Peterson, and R. Pritchard. 2002. Re silience management in socio-ecological systems: a working hypothesis for a partic ipatory approach. Conservation Ecology 6(1):14.

PAGE 142

142 Walker, B.H., C.S. Holling, S.C. Carpenter, and A. P. Kinzig. 2004. Resilience, adaptability and transformability. Ecology and Society 9(2):5 [online] URL: http://www.ecologyandsocie ty.org/vol9/iss2/art5/ Weiss, T. 2004a. Restructuring and redundanc y: the im pacts and illogic of neoliberal agricultural reforms in Jamaica. Journal of Agrarian Change 4(4):461-491. Weiss, T. 2004b. (Re-)making the case for land reform in Jamaica. Social and Economic Studies 53(1):35-72. Weiss, T. 2006. A precarious balance: neolib eralism, crisis management, and the social implosion in Jamaica. Capital and Class 85:115-147. Witter, M. 1988. The role of higglers/sidewalk vendors/informal commercial traders in the development of the Jamaican economy. Unpublished manuscript, Department of Economics. University of West Indies. Wong, D. C. 1996. A theory of petty trading: the Jamaican higgler. The Economic Journal 106(435):507-518. World Bank. 2003. Jamaica: The Road to Sust ained Growth. Country Economic Memorandum. Report number 26088-JM. Young, C.E., and L.S. Kantor. 1999. Moving toward the Food Guide Pyramid: Implications for U.S. agriculture. Pages 403 In E. Frazao, editor. America's eating habits: Changes and consequences USDA Economic Research Service, Washington D.C., USA. Young, O.R., F. Berkhout, G.C. Gallopin, M.A. Jan ssen, E. Ostrom, E. and S. van der Leeuw. 2006. The globalization of socio-ecological systems: an agenda for scientific research. Global Environmental Change 16:304-316. Zhu, X. and A.P. Dale. 2000. Identifying opportun ities for decision support systems in support of regional resource use pl anning: an approach thr ough soft systems methodology. Environmental Management 26(4):371-384.

PAGE 143

143 BIOGRAPHICAL SKETCH Lauren C. Justice was born in Atlanta, Georgia on November 19, 1980 to Joseph B. Justice and Betheda F. Gramling. She graduated from Druid Hills High School in 1999. In 2003 she graduated from Emory University in Atlanta, Geor gia with a Bachelor of Arts in Environmental Studies. In 2005 she enrolled at the University of Florida to pursue a M aster of Science in Interdisciplinary Ecology.