Annual Report 1999 Project PE-5 Sustainable Systems for Smallholders -~--~---·· ~-- --~- -- .. ~ - -~ --~----- .. - Integrating improved germplasm and resource management for enhanced crop and livestock production systems Project PE-5 Sustainable Systems for Smallholders: \ .. Integrating improved germplasm and resource management for enhan.ced crop and livestock production systems ANNUAL REPORT 1999 • ADB AEZ AMUCAU CA CAP CATIE CGIAR CIAT CIFOR CIMMYT CODESU CORPOICA CSIRO DEPAM ECAG FECONU FM FPR FSP GHG GIS GPS HS IARCS ICRAF IDRC IFDC IIAP ILRI INEI INIA IRRI NITA LAC NARS NGO NRM OMIAP PRGA PTD RIEPT SEAFRAD SLP UCR UNU YAC As ian Development Bank Agro-Ecological Zone ACRONYMS Asociacion de Mujeres Campes inas de Ucayali Central America Communíty Action Plan Centro Agronomico Tropical de lnvestígacion y Enseñanza Consultive Group Intemational Agricultural Research Centro Internacional de Agricultura Tropical Center for International Forestry Research Centro Internacional de Mejoramiento de Maíz y Trigo Consorcio para el Desarrollo Sostenible de Ucayali Corporacion Colombiana de Investigacioin Agropecuaria Commonwealth Scientífic & Industrial Research Organizatíon Proyecto: Desarrollo Participativo Amazonico Escuela Centroamericana de ganadería Federacion de Comunidades Indígenas de Ucayali Forest Margins Farmer Participatory Research Forage for Smallholders Project Green House Gases Geographic Information System Global Positíoning System Hill Sides lnternational Agricultural Research Centers International Council for Research in Agroforestry lnternational Development Research Center Intemational Fertilizer Development Center Peruvian Amazon Research Institute International Livestock Research Institute National Statistics and Census Institute Instituto Nacional de lnvestigacion Agraria (Peru) International Rice Research lnstitute Instituto Veterinario de lnvestigacion Tropical y de Altura Latín America & Caribean National Agricultural Research Systems Non Government Organization Natural Resources Management Organizacion de Mujeres Indígenas de la Amazonia Peruana Systemwide Participatory Research nad Gender Analysis Program Participatory Technology Development Red Internacional de Evaluacion de Pastos Tropicales Southeast Asían Forages and Feed Resources Research and Development Systemwide Livestock Program Universidad de Costa Rica Universidad Nacional de Ucayali V: Home garden; A: fish pond; C: Livestock (Vietnam) • Table of Contents Project Overview Project description 1 Project Log-frame 2 Work breakdown structure 7 Strategy 8 Highlights and progress towards achieving projects objectives lO Output l. Alternative land use options for agricultural systems assessed 17 Activity 1.1. Characterize and diagnose prob/ems and opportunities at the system leve/ 17 1.1.1. Spatial analysis of socioeconomics, the environment, and poverty at Pucallpa, Peru 17 1.1 .2. Hong Ha commune, A'Luoi district, Thua Thien Hue province, Vietnam 20 Activity 1.2. Assess biological and socioeconomic impact of a/ternative land use options 25 1.2.1 Analysis of farmer economic benefits and trade offs, Pucallpa. Peru 25 1.2.2 . Forage tree adoption and use in Asia 27 Activity 1.3. Determine indicators of human nutrition and healtlt, Aguaytia watershed 40 1.3 .1 Health, biodiversity and natural resource use on the Amazon frontier 40 Activity 1.4 Evaluate social and prívate trade-off of alternative /and use options 48 1.4.1. Fallow management strategies and trade-off analysis of fallow periods 48 Activity 1.5. syntltesize ofresultsfrom different outputs and activities 49 1.5 .1. Effect of pasture intensification in the forest margins, Peru 49 1.5 .2. A range of forage options for different farming systems in Southeast Asia 51 1.5.3 . Synthesis of information on legume green manure/cover crops in Eastem Africa 55 Output 2. Component technologies for sustainable production development through 55 farmer participatory research Activity 2.1. Farmer experimenta/ion to adapt tecltnologies 55 2.1.1. Multi-institutional and participatory research in Pucallpa (DEP AM) 55 2.1.2. Farmer rice and banana trials in the Aguaytia watershed, Pucallpa 60 2. 1.3. A farm-level economic model for ex-ante analysis of new technologies 65 2 .1.4. lncreasing food security in Hong Ha, Vietnam with the participation of farmers 66 2 .1 .5. Other interventions- Hong Ha commune, Vietnam 68 Activity 2.2./mproving feed quality and resource management in dual-purpose catt/e 69 production systems 2.2.1. Optimization of forage utilization 69 2.2.1.1. Trials to determine relations between milk production and forage resources 69 2 .2.1 .2. Feeding and cutting trials to refine methods of supplementation 73 2.2.2. Development of new feeding systems through on-farm research 76 2 .2.2.1. Milk production on grass legume pastures Costa Rica 76 2.2.2.2. Milk production on caws supplemented with Cratylia 78 2.2.2.3 . On farm evaluation in Honduras and N icaragua 81 2.2.3 . Evaluation of new feeding altematives to a llow early weaning 83 2 .2.3. 1. Use of Stylosanthes guianensis with pre-weaned calves 83 2.2.3 .2 Rice y ield (Oriza saliva L.) asan indicator ofsoil fertility 84 2.2.4. On-farm trials established for evaluating new forages in the llanos of Colombia 86 Activity 2.3. Evaluatiug legumes f or f eed supply, nutrient cycling all{/ improved f allows 2.3.1. ldentification of species for cover crops and fa llow improvement, Cauca, Colombia 2.3.2. Eval uation of herbaceous legumes as cover crops, Colombia 2.3 .3. Rotat ional effects of ti ve legumes on maize-bean intercrop production, U ganda 2.3.4. Effects of legumes on infection of subsequent crops by rootk.not, Kenya 2.3.5. Crop response to use of of alternative plant materials for soil fertil ity management Activity 2.4. Developing integratedforage-based componentsfor uplandfarming systems 2.4.1 . Selection of improved forages in South Asia Activity 2.5 Developing improved soil management practices in cassava-based systems 2.5.1. New cassava options for Asia 2.5.1. 1. The nutrient requirements of cassava 2.5.1.2. Soil improvement with green manure, intercropping and alley cropping 2.5.1.3. Effective ways to reduce erosion in cassava-based croping systems 2.5. 1.4. Varietal testing 2.5.2. Conduct demonstrations and FPR trials on farmers' fields 2.5.2.1. FPR demonstration plots 2.5.2.2 . FPR trials on farmers' fields 2.5.2.3. Adoption and dissemination of improved practices Activity 2.6. Establish and maintain databases ojinformation and results Output 3. Models/frameworks developed to target research, integrate results, assess impact and extrapolate results Activity 3.1. adapt and evaluate integrated simulation mode/s for smal/holders systems 3.1.1. Validation ofthe DSSAT-CENTURY link 3.1.2. Other modification to DSSAT 3 .1.3. DSSA T application to Honduras maize systems Acivity 3.2. Develop economic models to assess technologylland use options 3.2.1. Economic model for ex-ante evaluation of research interventions Activity 3.3. Develop aframeworkfor monitoring and assessing impact ofresearch in the forest margins Activity 3.4. Develop participatory monitoring and evaluation methods for technology development 3 .4.1 . Participatory monitoring and evaluation of new technologies Activity 3.5. Integrate information on variety adaptation and appropriate teclmologies witlt GIS databases to target germplasm use 3.5.1. Converting the forage database toa graphical p1atform 3.5.2. Use of GIS mpdels for better targeting offorage germplasm for end users Activity 3.6. Develop decision guides on soilfertility management 3.6.1. Conceptual frameworks to guide research and extension efforts for integrated nutrient management Output 4. Increased effectiveness of CIA T and partners to conduct appropriate research for developing productive and sustainable land use practices Activity 4.1. Coordination andfundidng 89 89 91 92 93 94 97 97 101 101 101 104 106 107 109 109 111 114 115 116 116 117 117 11 8 123 123 125 128 128 135 135 137 138 13 8 139 139 4.1. 1. Forest margins ecoregional site- Pucallpa 139 4. 1.2. PE-5 sustainable systems for sma1 lholders 140 4.1.3. Funding 140 4.1.3. 1. Existing special projects 140 4. 1.3 .2. Project near approval 141 4.1.3.3 . In discussion with donors 141 4.1.3.4. Proposals developed 141 Activity 4.2. Facilita/e multi-institutional and participatory researclt 142 4.2.1 . Multi-institutional research in the Aguaytia watershed,Pucallpa, Peru 142 4.2.2. Impact assessment workshop in Pucallpa - the process 144 4.2.3. Facilitating inter-center collaboration in the forest margins, Pucallpa 146 4.2.4. Facilitating community-based natural resources management, Vietnam 153 Activity 4.3. facilitate regional partnerships/networks 154 4.3 . l. Coordination meetings of forage for smallholders project 154 4.3.2 . The Asian Cassava Research Network 154 4.3.3. Tropileche - Workshops and meetings to analyzed and plan research activities 155 Activity 4.4. Compare effectiveness of different institutional models for effecting change 157 in natural resource management 4.4 .1. Participatory systems research: towards the future 157 Activity 4.5. Facilita/e a policy enabling environment that ensures adoption of 162 appropriate policies and tecltnologies 4.5. 1. Institutionalizing a collaborative research approach in Pucallpa, Peru 162 Activity 4. 6. Develop training approacltes and materials on targeting, development and 164 diffusion of new tecltnologies and provide training for partners 4.6. 1. Approaches to targeting and developing new forage technologies 164 4.6.2. Training provided on use of economic model in Peru 166 4.6.3. Farmer-experimentation mini-kits as a tool for promotion of 1egume technology 167 Activity 4. 7. Communicate results tltrouglt networks, workshops and journals 168 4. 7 . l. Communicate activities of the forages for smallho lders project 168 4. 7 .2. Communication of results within the Asian cassava research network 168 4.7.3. Tropileche - dissemination ofresearch results 169 4. 7 .4. Publications 17 1 Overview Project PE-5: Sustainable Systems for Smallholders: Integrating Improved Germplasm and Resources Management for Enhanced Crop and Livestock Production Systems Objective: To collaborate with national organizations in developing generic agricultura! technologies and inforrnation that lead to more productive and sustainable production systems. Outputs: lntegrated land use options that contribute to sustaining the natural resource base. Appropriate technologies for farrning systems particularly where beans, cassava, livestock and rice are components of the system. Decision support tools and other methods for use by farrners, researchers, planners and policy makers from the local to national levels. Increased capacity of national organizations and farrner groups to conduct demand driven research. Gains: lntegration of commodity and natural resource research. New approaches to the development of environmentally sound technologies. lndicators for measuring economic and environmental impact of improved technology at the farm and watershed levels. Methodology to extend results beyond benchmark sites. Milestones: 1998 Workshop to analyze the effectiveness of FPR in the development of new technology options for smallholder systems. Integrated cassava management systems for Asia. 1999 New forage technologies for smallholder systems in Asia and Latín America. Methodology for assessment of socioeconomic and environmental impact at farrn leve l. 2000 Dissemination of new crop and livestock technologies for smallholder systems in Latín America and Southeast Asia commenced. Multi-institutional and participatory models for systems research in place in Latin America and Southeast Asia. 200 1 New crop and livestock technologies for smallholder systems in Latín America, Southeast Asia, for example, increased bean production in low P soils; improved fallow systems for the forest margins, with demonstrated impact ofthe technologies on increased welfare of poor rural families and sustainable land use. An analysis and synthesis of various approaches used for participatory diagnosis of needs and opportunities and in participatory technology development. Users: The research will benefit low-income farrners in Latin America, Asia, and Africa by increasing available food and cash flow to rural households while providing a basis for more sustainable production systems. Adoption of environmentally sound farrning practices will benefit society as a whole. Collaborators: ICRAF, ILRI, IRRI; linkages with national R&D organizations and specialized research organizations. CGIAR system linkages: Protecting the Environment (50%); Crop Systems (20%); Livestock Production Systems (15%); Training (10%); Networks (5%). CIA T project linkages: Conservation of genetic resources; gerrnplasm enhancement in beans, cassava, and tropical forages; natural resource management in areas of land use dynamics, soil processes, and watershed management; strengthening NARS through developing partnerships, participatory research, and impact assessment. Project Log-frame CIAT Area: Natural Resources Project: PE-5 - Sustainable Systems for Smallholders: lntegrating Improved Germplasm and Resource Management for Enhanced Crop and Livestock Production Mana2er: Peter Kerridge Narrative Summary Goal Knowledge, tools, technologies, skills and organizational principies that contribute to the improved management of natural resources are accessible to NARS and beneficiaries Purpose To collaborate with national organizations in developing generic agricultural technologies and information that lead to more productive and sustainable production systems. Measurable lndicators/Milestones - Use of CIA T NRM research outputs in 3 reference si tes in 5 years, related to changes in land management and associated with increases in per capita income and food availability, improved soil-water-nutrient use efficiency, increased biodiversity in production systems, and stakeholders participating in land use planning. -Use ofthe CIAT NRM research outputs beyond the 3 reference sites in the 3 target agroecosystems (savannas, hillsides, forest margins) by stakeholders in 5 years. - CIA T NRM research outputs applied by at least 3 other institutions outside LAC by the end ofthe sth_)'_ear. - % increase in income of smallholders - no. ofnew component technologies - % decrease in soilloss, and increase in soil fertility & water retention - changes in functional aspects of biodiversity - decreases in deforestation and burning - widespread adoption of sustainable practices Mea os of Verification Projects, plans and reports of national sector agencies, donors, NGO's and community-based organization in the 3 reference sites in LAC mandate agroecosystems which refer to use of CIAT's NRM research outputs. Assumptions CIAT's pm1ners are willing to use these research outputs to improve NRM. - impact evaluation studies 1- donor and client support for sustainable land use research __ . ..., .... _ INarrative summarv Eco. Personnel lndicators (lntermediate impact) 1 Milestones Means of Verification ~ssumption s ~utputs l. IA!ternative land use options for Analysis oftrade-offs and policy options ofaltematives for land use and lagricultural systems assessed technology development are available for local and national planners ""'M pw 02 Policy options, for use in local and national planning for the forest Workshop and technical co llnboration and margins, developed from a synthesis of socio-economic and biophysical report to Govemment of intcgration of rcscarch effects of altemative land use in the Aguaytia watershed Peru activitics in Pucallpa !Asia iPK,SF O 1 Plan for community involvement in management of resources of technical report watershed site, Vietnam, accepted by government officials Continucd donor support IH, S IFH 00 Options and incentives necessary to develop and utilize feed resources workshop and technical Continucd collaboration IFM in a sustainable manner in dual-purpose cattle systems in Latín America report with N/\RS partncrs !Asia ws 99 Options for an integrated approach to improving feed resources for final report to donor Maintcnance of close livestock in SE Asia rc lations wi th policy m a J.. crs !Asia ~ 04 New options adopted for R&D activities in cassava production operational plans of 11'-3 continues to support 1 Thailand, Vietnam and China govemment agencies R&D in/\sia 1 IActivities (CIAT collaborators) IMilestones Annual report PE-5 1.1 Characterize and diagnose problems ir:-M SF, DW,PK, 199 Review problems and opportunities in the Aguaytia watershed and opportunities at the system level iDH, GH, RT through community participation ""M p H,DW 199 Analysis of poverty and land use in the Aguaytia watershed (with PE-2, PE-4, SN1, SN3) ~ ~F, CEL ~9 Diagnose opportunities in Puerto Lopez area of Llanos !Asia Univ. Hue ~9 Catchment site in Hong Ha characterized and problems and opportunities ~F, PK diagnosed for research by Hue tearn 1.2 Assess biological and socio-economic FM DW ~9 Analysis of economic benefits to farmers associated with altemative land impact of alternative land use use options in Aguaytia using profit-maximization options (PE-2, PE-4, BP-1) DW po Riverine flood risk mapped in Aguaytia watershed DW po Effect of in come variation u pon farmer-land use decision making in !Asia luniv Hue Aguaytia watershed 199 Economic analyses of crop alt~rnatives in Hong Ha, Vietnam 1.3 Determine indictors of human nutrition FM n-M, SF, DW 199 Describe production and consumption pattems of selected communities in Aguaytia watershed PO Assess diet, nutrient intake and physical condition of selected communities 1.4 Evaluate social and priva te trade-off of IFM DW ~9 Analysis of economic value of fa llow periods in Aguaytia watershed altemative land use options (BP-1) 1.5 Synthesize results from different FM DW, SF, FH 99 Role of intensification in the Forest Margins analyzed ' 1 outputs and activities (PE-4, BP-1) Asia WWS, PH ~9 Range of options for integration of forages into smallholder 1 farming systems in Asia identified Narrative summarv Eco. Personnel lndicators (lntermediate impact) 1 Milestones Means of Verification Assumotions Output 2. Farmers using new technologies that are more productive and conserve Component technologies for natural resources sustainable production developed through farmer participatory research FM SF, DH 02 New technology options developed for Aguaytia watershed - Annual report H FH, PJA 00 30% increase in milk production in dry season, CA - Report Tropileche H MP, PJA 03 lncreased use of forage legumes by small farmers in CA - Annual report S CEL 03 New forage options for Llanos - Annual report Asia wws 99 40 farmers at each of 18 sites in SE Asia using improved forages in SE - Workshop proceedings Asia Asia RH 04 Increased net benefits to farmers at 12 sites with less erosion in cassava- - lmpact evaluation based systems in Asia A frica cw 02 Examples of sustainable bean production systems in East Africa - Technical reports Activities Milestones - Annual report 2.1 Farmer experimentation to adapt FM SF, DH, PK 02 A wide range of land use options being evaluated by farmers in Aguaytia technologies watershed, Pucallpa PK,RH 00 A range of land use options being evaluated by farmers at catchment si tes in Vietnam PK,DH 99 Multi-institutional FPR team operating in Pucallpa DW 99 Application ofeconomic model for DEPAM grants 2.2 Improving feed quality and resource H, S FH,PJA, 00 30% increase in milk production during the dry season in Costa Rica. management in dual-purpose cattle FM CEL 99 Preliminary results from on-farm sites in CA production systems (IP-5) PJA 99 Impact of on-farm research being measured in CA CEL 99 On-farm trials established for evaluating new forages, Llanos 2.3 Evaluating legumes for feed supply, H MP, PJA 02 Increased use of forages legumes by small farmers in CA. nutrient cycling and improved fallows MP, PJA 99 Pilot evaluation sites established. (IP-5, PE-2) A frica cw 00 New cover crop legumes identified for soil fertility improvement 2.4 Developing integrated forage-based Asia PK 00 Plan for extending impact of FSP components for upland farming systems ws 99 An average of 40 farmers at each of 18 si tes in SE Asia using improved (SN-1) forages (FSP) 2.5 Developing improved soil Asia SF, WS 99 Diagnosis of conditions for adoption of forage tree species management practices in cassava-based RH 99 Plan for extending impact ofphase 1 on improved resource management systems in cassava-based systems in Thailand, Vietnam and China 2.6 Establish and maintain databases of FM GH,DW 99 Database established in Pucallpa information and results Narrative summary Eco. Personnel lndicators (lntermediate impact) 1 Milestones Means of Verification Assumptions Output 3. Farmers and technicians using models/approaches that assist decision Models/frameworks developed to target making at the farm and watershed levels research, integrate results, assess impact and extrapolate results H AG 00 lntegrated simulation model of germplasm and management options for - Working document - Collaboration with PE-3 Central America hillsides and runding available FM DW 99 Economic model available for ex-ante evaluation ofresearch proposals in - Working document - Dula avai lablc Pucallpa, Peru FM DW 99 lndicator framework used by all partners in the forest margins, Pucallpa - Working document - Collaboration with Asia WS,PK 00 Participatory evaluation and monitoring framework to assess FPR partncrs technologies in SE Asia - Final report - Nccd pcrceivcd by H MP, LHF O 1 GIS-based DSS of forage adaptation in CA used by extension staff partncrs - CDRom - Partncrs trained Asia RH 00 Decision guides available on soil ferti lity management for use by A frica cw extension workers and farmers - Extension material Activities Milestones - Annual report 1 3. 1. Adapt and evaluate integrated H AG 00 DSSAT Model adapted and applied to evaluate different land use - simulation models for smallholders alternatives in hillsides systems {IP-5, PE-3, PE-4) 99 Calibrations of DSSAT model for various crops in smallholder systems in CA 99 Data collection for smallholder systems in CA 00 Various legume options evaluated for different environmental and management conditions by simulation modeling 99 Legume green manure option included in DSSAT 3.2 Develop economic models to assess FH 99 Proposal developed for calibration of Cornell model technologylland use options FM DW 99 Economic model for ex-ante evaluation of research interventions in Aguaytia watershed FM DW 99 Model to assess potential of riverine areas, Aguaytia watershed 3.3 Develop a framework for monitoring FM DW,SF,GH 99 Impact analysis framework developed for the forest margins and assessing impact of research in DH,AI the forest margins 3.4 Develop participatory monitoring and Asia WS,SF, PK 00 Participatory evaluation and monitoring framework to assess FPR evaluation methods for technology technologies development (SN-1 , PRGA) 99 Results ofPE&M available for Philippines site 3.5 Integrate information on variety H MP, LHF O 1 GIS-based DSS for forage adaptation adaptation and appropriate 00 Secondary sites established for verification technologies with GIS databases to 99 Programming completed target germplasm use (IP-5, PE-4) 3.6 Develop decision guides on soil Asia RH 00 Utility of informal ion on soil fertility management packaged for use by fertility management (PE-2) Africa cw extension staff and farmers evaluated -· -- Narrative summarv Eco. Personnel Indicators (lntermediate impact) 1 Milestones Means of Verification Assumptions Output 4. Organizations with trained staff using new strategies for R&D lncreased effectiveness of CIA T and partners to conduct appropriate All PK 99 Coordinated approach to systems oriented research and funding in place - Annual report, CCER - Conscnsus within team research for developing productive and 00 A multi-institutional and participatory approach to R&D accepted by sustainable land use practices FM PK,DH, SF NARS in Pucallpa and operational at Hue, Vietnam - Site visit - Approach achicves succcsscs All Tea m 00 Functional partnerships- Tropileche, DEP AM, SEAFRAD, Cassava - Annual report - Nccd and collaborotion R&D network, Univ. Hue Asia PK,RH 02 Strategy for scaling up technologies developed using FPR in SE Asia and - Working document - Ncw projects funded Africa WS,CW A frica 99 Manuals on training approaches for applying PR in SE Asia - Manual available - Spccia l projcct fund ing All Team 00 Technicians at pilot and satellite sites trained in participatory technology - Training courses maintaincd development Activities Milestones - Annual report 4.1 Funding and coordina! ion of PE-5 All PK 99 Funding obtained for on-going and new activities 99 Project outputs achieved 4.2 Facilitate multi-institutional research FM DH, SF,PK 99 Regular meetings and collaboration within DEPAM team at the watershed leve! for R&D 99 Inter Center workshop on research in the FM, Pucallpa (SN-1, SN-3) Asia SF, PK 99 Workshop on PR and team building in Hue 4.3 Facilitate regional partnerships/ All Team 99 Coordinated research plans networks (IP-3, IP-5) wws 99 Coordination meetings of FSP FH 99 Coordination meeting ofTropileche 4.4 Compare effectiveness of different FM DH 00 Review successes and failures of different approaches to institutional institutional models for effecting change changes in NRM Asia SF 99 Paper comparing participatory and traditional approaches to technology development- FSP workshop A frica cw 4.5 Facilitate a policy enabling FM SF, OH 99 Review process of developing a collaborative research approach in ' environment that ensures adoption Asia DH Pucallpa 1 of appropriate policies and O 1 Review outcomes of enabling policy changes in selected watershed si tes technologies (PE-3,SN-l, SN-3, BP- DH 02 Review achievements in scaling up technologies developed using FPR 1) 4.6 Develop training approaches and wws 99 Manuals on FPR approaches materials on targeting, development DW 99 Training provided on use of economic model in Peru and diffusion of new technologies Al! FH 99 Training in use of ex-ante model and Jand use systems, and provide cw 99 Diffusion of green manure technologies training for partners (SN-3) DW,SF,PK 99 Workshop on impact analysis, Pucallpa 4.7 Communicate results through networks, workshops and joumals All Team 99 Yearly output of papers, newsletters and intemet ---- ------ - O vervicw Project PE-S. Sustainable systems for smallholders: integrating improved germplasm and resource management for enhanced crop and livestock production Purpose To collaborate with national organizations in developing generic agricultura! technologies and inforrnation that lead to more productive and sustainable production systems. ; o Land use options Componen t Technologies Decision support Capacity building u t Alternative Iand use systems for Models/frameworks developed to Increased effectiveness of CIA T a nd p Component technologies for u agricultural systems assessed sustainable production developed target research, integrate results, partners to conduct appropriatc t through farmer participatory assess impact and extrapolate results research for developing productivc S research and sustainable land use practices 1.1 Characterize and diagnose problems 2.1 Farmer experimentation to adapt 3.1 Adapt and evaluate integrated 4.1 Coordination and fundi ng of PE-5 and opportunities at the system leve! technologies in Pucallpa, Peru simulation models for smallholder 4.2 Facilitate multi-institutional rcsearch at through community participation (with 2.2 Improving feed quality and resources systems (IP-5, PE-3, PE-4) the watershed level for R&D (SN-1 . A PE-2, PE-4, SNI, SN3) management in dual-purpose cattle 3.2 Develop economic models to assess SN-3) e t 1.2 Assess biological and socio-economic systems (IP-5) technology/land use options 4.3 Facilitate regional i impact of altemative land use options 2.3 Evaluating legumes for feed supply, 3.3 Develop a framework for monitoring partnerships/networks (11'-3. 11'-5) V (PE2, PE4, BP-1) nutrient cycling and improved fallows and assessing impact of research in the 4.4 Compare effectiveness of di fTcrcnt i 1.3 Determine indicators of human (IP-5, PE-2) forest margins institutional models for cffccting t nutrition and health in the Aguaytia i watershed 2.4 Developing integrated forage-based 3.4 Develop participatory monitoring and changes in NRM e components for upland farming evaluation methods for technology 4.5 Facilitate a policy cnabling S 1.4 Evaluate social and prívate trade-offs systems (SN-1) development (SN-1, PRGA) environment that ensures adoption of of altemative land use options (BP-1) 2.5 Developing improved soil 3.5 lntegrate information on variety appropriate policies and tcchnologics 1.5 Synthesize of results from different management practices in cassava adaptation and appropriate (PE-3, SN- 1, SN-3, BP-1) outputs and activities (PE-4, BP-1) based systems technologies with GIS databases to 4.6 Develop training approaches and 2.6 Establish and maintain databases of target germplasm use (IP-5, PE-4) materials on targeting. dcvelopmelll information and results 3.6 Develop decision guides on soil and diffusion of ncw tcchnologics and fertility management (PE-2) land use systems, and providc lraining for partners SN-3) 4.7 Communicate results through networks, workshops and journals - - -- • collaboration with other projects 7 1 1 Strategy ClA T's mission is: Project PE-5 Sustainable systems for smallholders: integrating improved germplasm and resource management for enhanced crop and livestock production To contribute to the eradication of hunger and poverty in tropical deve/oping countries Overview by generating technology and information with partners through scientific research and dissemination that leads to lasting increases in agricultura/ productivity while preserving the natural resource base. PE-5 's contributes to this mission by collaborating with a range ofpartners to develop sustainable and productive technologies that integrate improved gennplasm with conservative land management practices. Improved germplasm is taken from germplasm improvement projects at ClA T and other sources and incorporated with improved management practices through a fanner participatory research approach to develop relevant component technologies at the farm leve!. Alternative component technologies from CIAT and other organizations are evaluated at the watershed leve! for their biological, social and economic effects on productivity and resource management. An analysis of trade-offs between productivity and environmental benefit produces a range of integrated options for local and national policy makers. Process Research is conducted in a systems context. In developing component technologies at the farm/community leve!, there is diagnosis and evaluation of opportunities for new technologies within the overall farming system, participatory development of the technology with farmers and local partners, monitoring and impact assessment, feedback to applied researchers, and synthesis and dissemination of the results. Developing a range of integrated options at the watershed leve!, involves facilitating the fonnation and operation ofmulti-disciplinary research teams from organizations working in the area, characterization ofthe area, diagnosis of problems and opportunities, facilitating adaptive research to work on priority issues, monitoring and impact assessment. Concurrently there is an analysis and synthesis of results for community, researchers and policy makers. Research focuses on both improving productivity and environmental protection with a balance between intervention and to evaluate and adapt technologies and modeling to target research, integrate results, assess impact and extrapolate results. We have a team of agronomists, livestock specialist, soil scientist, agricultura) anthropologist, resource economist, systems modeler and specialist in participatory research. Individual team members work on developing component technologies and integrated options at the watershed leve l. The research is dependent on effective partnerships within CIA T and with national and international organizations. 8 Overview Strategic outputs • lntegrated land use options that contribute to sustaining the natural resource base • Appropriate component technologies, particularly in areas for which CIA T has a mandate • Decision support tools and other methods for use by farmers, researchers, planners and policy makers from the local level to nationallevels • Increased capacity of national institutions to conduct component and systems research using participatory approaches Where we work We work in upland areas where there is a high incidence ofpoor rural families, low productivity and problems of resource degradation. At present, PE-5 in in volved in the development of component technologies for cassava, forages and rice in the forest margins, hillsides and savannas of Latín America, Asia and Africa and in watershed leve) research in the forest margins, Pucallpa, Peru and a steep upland watershed in Vietnam. 9 Overview Sustainable Systems for Smallholders Highlights and Progress towards achieving project objectives 1.1 Introduction The objectives, expected outputs, associated activities, required inputs and the strategy being followed have been summarized in the previous pages. However the environment we work in is dynamic and conditioned to a large extent by opportunities for funding. The major portian ofthe funding is obtained through Special Projects that have to be designed to meet specific donor as well as stakeholder objectives and are at different stages of implementation. We need to respond to changing stakeholder demands, new technical information and indigenous knowledge of our clients. Nevertheless the logframe with four broad Outputs allow us to maintain our focus on a common set of goals, exchange ideas and outputs, share resources and integrate somewhat di verse activities towards achieving common goals and contributing to the NRM outputs of CIA T. In this section, we summarize achievements and progress in relation to the subproject areas which, to a large extent, coincide with the Special Projects that fund the activities ofPE-5. This presentation provides a more coherent account of the activities needed from different project Outputs to achieve a desired outcome for developing sustainable production at the component or watershed level, e.g. technology inputs, modeling and training. These achievements are cross-referenced to activity reports for the four broad Outputs of PE-5 and demonstrate coherence within project PE-5 as a whole. Progress towards achieving goals is summarized in relation to (i) development of productive and sustainable component technologies, (ii) improved NRM by communit ies at the watershed level, (iii) cross project outcomes of research, and (iv) project contributions to CIA T NRM outputs. 1.2 Productive and sustainable technologies 1.2.1 Forages for tropical dual-purpose systems in LAC (Tropileche project) Objective: To evaluate legume-based feeding systems to improve the efficiency of milk production and improve NRM in smallholder dual-purpose systems in LAC. Previous research: • Characterized the hillsides and forest margin agro-ecosystems in which the project worked . • Identified potentially useful legumes, in particular, a new shrub legume, Cratylia argentea, adapted to infertile and soils and dry conditions (project IP-5). Practica! methodologies for assessing affect offeeds on milk production • Modified a linear programming model for ex-ante evaluation and • Training for NARS scientists Research in 1999 (see Outputs 2.2; 4.3.3 ; 4.7.3 for full details): • On-farm research in Costa Rica demonstrated the economic advantage of C. argentea to replace purchased supplements when fed fresh or as silage • Farmers integrated this legume into their production system, improving the management ofthe C. argentea-sugar can e feeding system for the dry season • Results are being extended to Nicaragua and Honduras in Central America working with project PE-3. • Website was developed for disseminating information on dual-purpose cattle systems Jmpact: Tropileche is being accepted as a platform for dissemination of new forage germplasm from project IP-5 and component technologies developed in PE-5, a forum for exchange of ideas, a vehicle for undertaking 10 Overview FPR research, strengthening the capacity ofN ARS, providing feedback to CIA T on research priorities, anda mechanism for deve loping bi-lateral projects with NARS. In the coming year, similar research activities wil l be continued and extended to other countries with more emphasis on monitoring and impact assessment. Funding wil l be sought for continuing acti vities beyond 2000. 1.2.2 Forages in upland farming systems in Asia (Forages for Smallholders project) Objective: To integrate forage technology options to improve feed resources and resource management in smallholder farming systems in Southeast Asia, in particular, in Indonesia, Laos, Philippines and Vietnam. Previous research: • Forage species identified for different environments and farming systems • Forage technologies being developed with farmers at 18 on-farm sites • Feedback from on-farm applied research led to strategic research in selection ofBrachiaria species for high seed production and Stylosanthes for disease resistance • FPR teams formed in 4 countries and approach modified for local conditions • Training provided for >200 researchers and extensionists • Externa[ review noted the FPR approach was the key in bringing forages to fanners Research in 1999 (see Outputs 1.2.2; 1.5.2; 2.4. 1; 3.4; 4.3 .1; 4.6.1; 4.7.1 for full details): • Synthesis showed 11 forage technology options were identified for 7 smallholder farming systems • 18 grasses and legumes have been chosen for the region in terms of adaptation to climatic conditions and soils and suitability for various forage technology options • A framework is being developed for participatory monitoring and evaluation in technology development • A survey was carried out in 3 countries to identify indigenous criteria for adoption of shrub legumes and reasons for fanner acceptance or rejection of new forage options developed by the FSP • Rapid impact study is being carried out across 18 sites • Summarized training approaches for developing new forage technologies in three publications • Organized intemational workshop "Working with Farmers". • Continued to support regional SEAFRAD (Southeast Asia Forage and Feed Resources Network) Impact: This Special Project concludes in December 1999. Most activities for the year have focused on the analysis, summary and synthesis ofresults obtained during the 5 years ofthe proj ect. In addition, follow-on projects ha ve been developed to assist and monitor the adaptation and adoption of new forage options and to investigate participatory approaches to scaling up and disseminating the forage technologies. 1.2.3 Developing improved soil management practices in cassava-based systems in Asia Objective: To maintain or improve soil productivity and optimize ferti lizer use in cassava-based cropping systems Previous research: • 25 cassava varieties containing genetic material from CIA T ha ve been released in Asia and are estimated to be grown over 600,000 ha. • This variety improvement was accompanied by the investigation of management practices to reduce soil loss under cassava • Soil fertility maintenance, inter-cropping and erosion barriers all contributed to reduced erosion. • Research teams ofNARS were set up in China, Indonesia, Thailand and Vietnam and trained to conduct on-farm demonstrations and FPR 1 1 Overview • A nucleus group of scientists, extension staff and farmers began to adopt FPR Research in 1999 (see Outputs 2.5; 4.3.2; 4.7.2 for full details): A second phase ofthe project was funded by the NIPPON Foundation in 1999 to: (i) continue FPR technology development on improved management practices, (ii) to scale up and disseminate improved technologies, (iii) to undertake research problems that arise from field experience and (iv) to train larger numbers of national scientists in FPR • New sites have been selected and national teams FPR are being established in China (2), Thailand (5) and Vietnam (5). • Results from 1999 confirm earlier findings: • K deficiency can be expected when soil ex. K is <0.15-0.17 me/1 OOg • Inter-cropping with cassava gives high economic retums and reduces soil erosion • Choice of options for eros ion control depends on location, soils and possible end-uses Impact: There is enthusiasm by national partners to continue collaboration in improvement of soil management cassava-based systems. In part, this is seen as a means of ensuring access to advanced cassava lines and other technologies available from CIA T. Most partners are actively supporting the FPR approach by introducing elements to their research and extension services. The challenge to the new project will be to identify participatory approaches for scaling up new technologies. 1.3 Natural resource management at the watershed level 1.3.1 Systems research in the Forest Margins, Pucallpa, Peru Objective: To contribute to elimination of poverty through development of sustainable production systems that minimize deforestation, loss of biodiversity and gaseous emission to the atmosphere Previous research: • Area characterized in relation to social groups and Iand use and the effect of different land use systems on changes in biodiversity • Commenced facilitation of a multi-institutional and participatory approach to adaptive research • FPR trials on rice improvement commenced Research in 1999 (see Outputs 1.1.1 ; 1.2.1; 1.4.1 ; 1.5.1; 2.1.1; 2. 1.2; 2.1.3; 3.2.1 ; 3.3.1; 3.4.1; 4.2.1; 4.2.2; 4.2.3; 4.5.1; 4.6.2 for full details): • Continued characterization through spatial analysis of poverty in relation to geographical position in the landscape. • Commenced a study on the causal linkages between human health and resource use An analysis of the effect of labor requirement on introduction of new technologies showed that if non- competitive with traditional crops there could be a negative effect on forest cover Analysis of studies on intensification of pasture systems suggested that forest scarcity is a prerequisite for intensification of agricultura! technology • The first FPR trials on rice have demonstrated to national scientists that farrners are interested and can con tribute to evaluation and adaptation of potentially useful technologies • A multi-institutional and participatory team (DEPAM) is taking an active role in testing new ways of working across institutions and with farmers • 11 FPR projects were initiated by DEP AM An agro-enterprise unit from SN-1 has been integrated with the activities ofDEPAM. A farm-leve l economic model was developed for ex-ante analysis of potential new technologies 12 Overview A framework was developed with all partners for monitoring and assessing impact of research This was followed up with the formation of inter-disciplinary and multi-institutional working groups to refine impact indicators and collect data for impact assessment Separate consultations were held with IARC and national partners to share research agendas and develop programs that were collaborative, complementary and directed toa common set of development objectives Clase collaboration was maintained with the CGIAR Systemwide Programs, Alternatives to Slash-and- Burn and the Participatory Research and Gender Analysis Programs. lmpact: CIA T now has an integrated research agenda in Pucallpa that is complementary to national and international partners. Working relationships have been improved to the extent thatjoint research proposals are being developed. CIA T's initiative in introducing participatory approaches through a multi-institutional adaptive research team is leading to closer linkages of researchers with farmers in setting and monitoring research. However, this process is in an early stage of development and needs to aim for greater involvement of government planners and policy makers if incentives are to be developed that favor the environmental protection that is a majar focus oflARC research in the Forest Margins. We view this as a long-term research initiative of CIA T. 1.3.2 Community-based natural resource management, Hong Ha, Vietnam Objective: To improve food security and involve the community in decision making processes that affect maintenance of natural resources. Previous research: • Identified the watershed ofthe Hong Ha commune, A'Luoi district, Thua Hue Province, as a study area for community management of natural resources • Project was developed and funding approved. Research in 1999 (see Outputs 1.1.2; 2.1.4; 2.1.5; 4.2.4 for full details): • The watershed occupied was characterized with respect to physical resources, agriculture, economics, culture and gender issues, and the institutional system • Farmers have begun evaluating new varieties of rice, cassava, fruit and vegetables, green manure crops and improved small livestock practices • Rice yields were increased from 50-1 00% with new varieties • National partners were provided support in problem diagnosis, defining and prioritizing research opportunities, in PFR, provision of germplasm and technical areas. • The national team established linkages with district and provincial officers. lmpact: The main outcomes to date are a good understanding ofthe physical and human resources in the watershed, establishment of empathy with the community and linkages with government officials. In addition farmers have already been involved in evaluating solutions to overcoming the acute food shortage and improving living conditions, such as a fresh water supply. As food security is improved more attention can be focused on NRM issues of forest and waste land management. lt remains a major challenge to identifY acceptable approaches that accept a local community becoming involved in decision making processes that are normally taken at provincial levels. 13 Overview 1.4 Cross project outcomes of research Many outcomes of research in the above areas can be adapted for use across PE-5 as a whole and indeed in other projects in ClA T and by our national and international partners. However, sorne areas of research are more generic than others and are summarized here. 1.4.1 Adaptation and evaluate simulation models for smallholder farming systems Objective: Adapt and evaluate DSSAT for smallholder systems to better understand their strengths and weaknesses and to identify opportunities for their use. Previous research: • The soil organic matter module in CENTUR Y was linked to DSSA T. • A Brachiaria grass module option was developed for DSSA T Research in 1999 (see Output 3.1 for full details): • Validated the CENTURY soil organic module SOM in DSSAT • Successfully simulated maize production at the Hillsides reference site and identified that moisture was more limiting than N. lmpact: Simulation models offer the opportunity to evaluate farm-level management strategies, estímate yields for economic analysis and assess the use of improved germplasm in association with improved NRM. The research conducted here has been directed at supporting the systems research undertaken by project PE-3 in the hillsides of Central America though it has wider application. However, research in PE-5 is likely to be discontinued due to inability to secure Special Project funding. 1.4.2 Legumes identified for feed supply, nutrient cycling and improved fallows (with IP-2 and IP-5) Objective: To identify legumes and other plants that can contribute to maintaining soil fertility and a protein source for livestock. Previous research: There has been a long, though somewhat intermittent study ofthe effect of legume covers on soil improvement. Research continues because of declining soil fertility and the inability of resource-poor farmers to purchase nutrient inputs. Research in 1999 (see Outputs 1.5.3; 2.3.1; 2.3.2; 2.3.4; 2.3.5; 2.4.1; 2.5.1 for fu ll details): • Legumes being evaluated for multiple uses in Central America and Asia • Lower nematode populations were maintained with Crotolaria, Lablab and Mucuna than other legumes • Subsequent crop yields increased by green manure crops in hillsides of LAC and East Africa • Synthesis of information on legume green manure and cover crops for East Africa Impact: Although various cover crop legumes have been shown to have beneficia! effects on subsequent crops, there has not been widespread adoption. Future studies need to focus on the reasons for this and in developing a strategy to encourage more widespread evaluation by farmers. Research could also focus more on developing integrated fertilizer-green manure strategies. 14 Overview 1.4.3 Facilitation of policy enabling environments Objective: To develop approaches to enable in stitutional change that is receptive to improved natural resource management. Previous research: There has been a strong focus on FPR in PE-5 and CIA Tasa whole. However, the concepts of participation at the farrn leve! need to become accepted at higher levels in organizations for research to become more demand driven efficient in use of resources. In particular, this is important for development and implementation of improved NRM practices. Research in 1999 (see Outputs 4.2.1; 4.2.2; 4.4 .1; 4.5.1 ; 4.6.1 for full details): • Review of systems research in CIA T identified a trend for new partnerships anda progression in the development of participatory approaches • New approaches were developed for targeting and developing new forage and cassava technologies • Facilitated the forrnation of integrated multi-institutional teams that use participatory approaches in problem definition and solution. Impact: CIA T is becoming recognized as a leader in enabling institutional changes as evidenced in our systems research in the Forest Margins and Vietnam and in the regional FPR projects developing improved forage and cassava technologies in Asia. More attention needs to be focused on planners and policy makers in addition to community involvement. In addition, social science resources and/or experience need to be increased. 1.5 Contributions to the NRM outputs of CIA T The following are the NRM outputs of CIA T • Improved Land Use Management • Decision Support Tools • Indicators for Sustainability • NRM Technology components and Inforrnation Organizational models and participatory methods • Improved capacity in NARS for NRM Examples of Project outputs that contri bu te to these NRM outputs include: Improved Land Use Management A range offorage options for farming systems in Southeast Asia (1.5 .2) Synthesis of inforrnation on legume green manure and cover crops in Africa ( 1.5.3) • An analysis of intensification and deforestation in the Forest Margins ( 1.5.1) Decision Support Tools • Simulation model adapted for smallholder farrning systems (3 .1.1) • Ex-ante model for evaluation of research interventions (3 .2.1) • GIS model for targeting offorage gerrnplasm for end users (3 .5.2) Jndicators for Sustainability Framework for monitoring and impact assessment in the Forest Margins (3 .3 .1) 15 NRM Technology components and Informal ion • Cratyl ia argentea-sugar cane supplement for increasing milk production in the dry season (2.2.2) Integrated soil management strategy for cassava-based systems in Asia (2.5.1) Organizational models and participatory methods • Model for multi-institutional and participatory research-DEPAM (2.1.1; 4.2.1; 4.5.1) • Approaches to targeting and developing technologies (4.6.1) Improved capacity in NARS for NRM • Research networks (2.4.7) • Collaborative planning ( 4.2.2) • Facilitating multi-disciplinary research ( 4.2.4) 16 Overview Altemative Land Use Options Output l. Alternative Iand use options for agricultura) systems assessed Activity 1.1 Clzaracterize and diagnose problems and opportunities at tite system leve/ througlz community participation 1.1.1 Spatial analysis of socioeconomics, the environment and poverty at Pucallpa, the Forest Margins reference site Higblights • Linking census data to a GIS database allowed us to demonstrate that differences between sorne 400 villages could be related to geographical position in the landscape. Objective: To understand how communities in the forest margins function over a broad area and to use this understanding to focus research and development activities. Rationale: The project will help CIA T and its partners target research and development activities to the poorest of the small farmers in the benchmark si te. A secondary benefit of the project is to give researchers a too! for research on agricultura! systems. Methods: This project focuses on the geographic dimensions of agriculture, environment and population at the village level in the area surrounding Pucallpa, Peru (Figure 1). In 1999 a database was established, training carried out and sorne preliminary analysis undertaken. Global positioning systems (GPS) receivers were used to locate over 400 villages in the area surrounding Pucallpa. lnformation from the 1993 Population census and the 1994 agricultura} census was then linked to digital maps ofthe study area. The census data includes information on population dynamics, govemment services, agricultura! systems and poverty. The training workshop was held Lima for professional staff from the Peruvian Amazon Research Institute (IIAP), the National Statistics and Census lnstitute (INEI), and Ministries of Agriculture, Education and Health. Workshop participants leamed sorne ofthe basics ofthe user-friendly GIS package ArcView in focusing on spatial pattems of poverty and environment. Outputs: Our first results are an analysis ofwhich variables might be useful to understand poverty dynamics in the benchmark site. We calculated basic needs that are unmet (NBI is the Spanish acronym) from the population census. The indicator is made up of 5 socially accepted standards of basic needs. They are school attendance, dependency ratio, house materials, house plumbing and number of persons per room in the house. (Figure 2) suggests that the average number of basic needs unmet in each village discriminate differences between villages in the study area. For the 40,369 rural people in the census survey, the average number ofNBI is 2. We are using this indicator as a dependent variable in statistical analyses where we are trying to find the factors related to poverty at the village leve!. Our next steps will be to make more sophisticated statistical analyses and to interpret the information. This work will be published by INEI as part oftheir institutional research series. 17 - .o.Rf,t. OF INIE RES I c:J DE P~.RIME NT S •i f PEI. ro -e 1 e ro ::2 120 100 80 60 40 20 o e nl ...., Labor Requirements Flat-upland Farms 120 100 ~ 80 ro "'9 60 e ro ::2 40 20 Alternative Land Use Options Labor Requirements Riverine Farms o~~~~~--~~~~----~~ e nl ...., "'5 ...., a. Q) en Figure l. Labor requirements by month of average flat-upland (20 ha) and riverine (18 ha) farms The price variability of primary agricultura! crops (rice, cassava, plantain, maize) is related to the time of harvest. Since few farmers have crop storage capability anda strong desire to quickly sell their harvest prices decline markedly at harvest time. For rice, the harvests are in January (uplands) and September (riverine). 0.6 ~0.5 ~ 0.4 o (J) 0.3 0.2 e nl ...., Monthly Crop Prices (Average 1996,1997) "'5 ...., a. Q) U) > o z Rainfall and River Flooding Ucayali (Average 1984-1995) 21 o .---------------------------,- 150 ......... -¡¡; ro E E .§. 140 +.;.; /-----"r \------------t---+ 145 ~ Cl ~ ·m -~ 70 +--------\-:7't'<:----::-:/---;·'------+ 140 J: ~ ~ > o +-+--+--+-+--+--+--+-+-+-+--+ 135 ii e ro ""') a. Q) (/) > o z Data: Ministry of Agriculture-Ucayali, Universidad Nacional del Ucayali Figure 2. Monthly crop prices for rice, maize, plantain and cassava Figure 3. Rainfall and river flooding in Ucayali Impact: The trade-off between economic and environmental goals are best examined by starting analysis at the farming system leve!. Both agricultura! research and government policy initiative can be modeled in an ex-ante manner to foresee likely implications upon farmer income, household food security and forest cover. Price variability of primary agricultura! crops can be an advantage if crop storage were an alternative for farmers. Future research should examine farmers interest in storage and the benefits versus the costs (potential pest damage, capital investment required). These benefits and costs should also be compared against improving yields oftraditional agronomic crops. Contributors: Douglas White, Ricardo Labarta, SN-1 (CIAT), INIA, MinAg-Ucayali, DEPAM, ICRAF 26 1.2.2 Forage tree adoption and use in Asia Highlights In traditional systems in Bali, farmers grew trees for fodder if: agriculture was intensive; • cattle were penned and fed by cut-and-carry; • agroforestry was a part oftraditional systems; • the system did not depend on shade-intolerant annual crops; trees were superior in providing dry season fodder. Alternative Land Use Options Fanners testing legume fodder in projects were positive in tenns of animal health and weight gain. Purpose: Examine fanners ' traditional and project-introduced adoption and management offorages, especially legume shrubs and trees. Rationale: Slow adoption of legume shrubs by fanners at sites ofthe Forages for Smallholders (FSP) project indicated a need to understand traditional systems in which such forages were already employed; and to understand farmers adoption and non-adoption where the project was facilitating fanner testing and adaptation to local systems. Methods: Ethnographic and participatory evaluation procedures were used to understand mixed agricultura) systems, fanners' animal feeding systems and their perceptions regarding the forages utilized. A small team of researchers from the Faculty of Animal Husbandry at Udanaya University (Den pasar, Bali, Indonesia), the Environmental Bamboo Foundation, and CIA T visited Besakih and Petang in the uplands north ofDenpasar and sites on Nusa Penida an island south ofBali . These sites featured traditional mixed agroforestry and livestock systems ranging in intensity from fully penned animals in Besakih to cut-and-carry combined with tethering in Peteng and Nusah Penida. Fanners were asked about their wet and dry season fodder use and to evaluate the forages used according to their own criteria. Eighteen to twenty-five fanners were individually interviewed at each site. Field observations were recorded. FSP sites were visited in central Vietnam (Xuan Loe near Hue), northern Sumatra (Marenu), and northern Mindanao (Malitbog). At each si te, a small group of researchers collaborating with the FSP project visited both forage adoptors and non-adoptors to discuss forage use and evaluations. The proportion of fodders used by each group interviewed were aggregated for the individual forage uses across the sample. Most fanners had similar numbers of livestock at each si te. Where a few infonnants had larger herds, e are was taken to determine if their forage use proportions were similar to those of their neighbors and to correct the aggregate use as necessary. In tenns of participatory forage evaluations, fanners evaluated forages using matrices showing each respondent's species x each respondent's evaluation criteria with the fanners assigning relative values using beans or maize as counters. Individuals differed in tenns of both forages used and evaluation criteria. Data was aggregated in two ways: (i) by presenting relative values for those planting a particular forage and employing a given evaluation criteria; and (ii) by presenting the percentage oftotal "votes" received by a given species x evaluation criteria combination. The first method over-valued the less frequently encountered species x criteria combinations, i.e. it ignored the negative "votes" of infonnants not using a particular forage and evaluation criteria combination. The second method undervalued the species x evaluation combinations held by the minority. Data aggregated from farmers' individual evaluations were, thus, presented to show both sets of values. 27 Alternative Land Use Options T he matrix method also suffered in that values ass igned toa ce ll could not be less than zero, eliminating re lative degrees of negative evaluations. Farmers were then simply asked to name both positive or "good" and negative or "bad" characteristi cs assoc iated with each forage source. Outputs: Traditional intensive cropllivestock/agroforestry in Bali. Nineteen farmers were interviewed in Besakih. The volcanic slopes u sed by the Besakih farmers extend from so me 1000 to 1500 meters above sea leve!, providing a cool climate and relatively rich soils suited to agroforestry and root crops. Farm size was a mean 0.95 ha, skewed by three extended families . Mean farm size was 0.6 ha for the 16 families (range 0.2-1 .0 ha); while the remaining three families had 2.0-4.0 ha. Coffee and sweet potato were the most important crops followed by cassava, citrus, banana, cloves, coconut, and sorne maize. Albizia sp was grown for timber for the local wood carving industry. Modal number of cattle was two (range 2-6 for the 16 families, 7-20 for the three extended families). Cattle were penned and not grazed or tethered. All feed was provided by cut-and-carry. Farmers relied upon on-farm feed resources ranging from natural grasses or weeds to planted grasses (Pennisetum purpureum ), trees ( Calliandra calothyrsus, Gliricidia sepium, Albizia saman, and jackfriut), and crop residues (sweet potato vines, Jeaves, and tubers) (Table 1). Table l. Farmers' fodder sources (%), wet and dry seasons, Bali Besakih Peteng Nusa Penida "' = = = = "' = = loo o "' o o loo o o u "' o loo "' "' u "' "' s ~ "' u ~ ~ E ~ ~u ~ E u u u loo ~ u u ~ loo ~ ~ ~ loo ~ - ~ ~ ~ - u t' - t' u t' ~ ~ u ~ ~ o Q ~ ~ Q o Q o Tree fodder 37 43 25 37 41 53 G. sepium 781 11 10 100 12 14 100 32 28 C. calothy rsus 100 20 19 28 2 3 A. saman 33 2 7 22 1 1 E. orienta/is 50 4 6 Ficus sp 89 5 23 Sesbania sp 28 4 2 Jackfruit 61 4 7 83 6 13 P1anted grasses P purpureum 94 19 15 100 21 13 Local grasses 89 11 8 94 24 11 94 25 8 Crop residues 13 13 24 30 28 33 Sweet potato (tuber, Ieaf) 89 13 13 Cassava 61 6 8 63 6 6 Banana stalk 100 14 15 83 12 16 Coconut fronds 56 4 7 56 1 10 Bean leaf 50 9 1 TOTAL 80 79 94 91 94 94 1 % of farmers using a particular forage; 2% of fodder coming from a forage type or species 28 Alternative Land Use Options Farms were intensively cultivated, wi th sma ll parcels separated by "live fences" comprised ofa wide mix of trees anda few grasses. Fence row spec ies included the trees C. sepium, C. calothyrsus, A. saman, Erylhrina orientalis, j ackfruit, avocado, salak (a local fruit), and grasses P. purpureum and King grass (a P. purpureum x P. typho ides hybrid). Farmers adm itted that with al! an imals penned, the apparent live "fences" were not established as fences per se. More li kely is that these were "linear fields" establ ished for fodder (and sorne fruit and timber) and having the advantages of ease of harvest and, more importantly, the deflection of much of the abo ve- and below-ground competition of the trees into adjacent pathways, roadsides, and terrace walls. Individual farmers in Besakih evaluated the forages they each used, using criteria each saw as important. When data is aggregated to show entries reflecting the mean score for farrners planting a particular forage and using a given criteria (Table 2), C. calythyrsus, G. sepium, P. purpureum, and sweet patato were judged as somewhat equal and superior toA saman, jackfruit, and local grasses. The most important criteria were yield, palatability, and weight gain. Calliandra scored high in terms of weight gain, yield, an imal health, and fast growth. Although scoring high across most criteria, Gliricidia was especially valued for yield. P purpureum was valued for its high y ield; and sweet pa tato (leaves and tubers) especially high for palatability, weight gain, and fast growth. Table 2. Species evaluation*, Besakih, Bali Evaluation Criteria .e: - ~ .e .:::: e ~ >. "" ~ e ~ "" ~ e E 'i :E <-" = o :.e "' .e: o ~-; e - -; "" .~ ... "" ~ e ·-~ .e: <-" ·-· .. e: "' - ~ E en .e - ~ -; .X ] ~ ·a:¡ !: - e- c -; >. "' ~ "" -~ ~ e ... "' "" c::;¿ o "" .. > ~ < Q ~ ..... ¡..... cz: Use criteria (%) 78 78 72 50 22 22 17 5 C. calothyrsus 100 5 4 6 5 3 3 5 2 33 1 G. sepium 78 5 4 4 4 3 4 4 4 32 1 A. saman 33 3 3 2 2 2 3 3 3 21 5 Jackfruit 61 3 3 3 2 2 2 2 3 20 5 P. purpureum 94 6 5 5 2 3 3 4 2 30 1 Local grasses 89 4 3 4 2 l 3 1 1 19 5 Sweet potato 89 4 7 7 4 2 1 6 2 33 1 TOTAL 30 29 31 21 16 19 25 17 Relative Importance 5 7 5 4 7 * Eighteen farmers each planted different species and used different evaluation criteria. Entries are mean scores for those planting a given species and using a given criteria. Relative scores for species are percent planting x total. Relative importance of criteria are percentages using criteria x total. Only slightly di fferent results emerged when the percentage of "votes" gained by each species x evaluation criteria was considered (Table 3). The criteria ofyie ld, palatability, and weight gain remained the most important; and the forages Calliandra, Pennisetum, and sweet patato were still the highest rated. 29 Altemative Land Use Options Table 3. Species eva/uation *, Besakih, Bali Evaluation Criteria - 1).1) ·= e :E e Q,j e "' C'l o e ·~ Q,j C) "' .e - :e C'l o 1).1)-; = - ";.,e Q,j .~ J.. e .... C'l C'l ..e ~ :e C) ·- J.. ¡s: "' - 1).1) s.::: - Q,j -; ..:.:: ~ C'l ·¡¡ ·- C'l e C'l - =- e -; e Q,j - "' C'l C'l -~ > ~ "' C'l S:~ o C'l e c.. <:::C Q ~ ¡;;,;, E-< ,::::.:: Use criteria (%) 78 78 72 50 22 22 17 5 C. calothyrsus 100 5 5 5 3 <1 21 l G. sepium 78 4 3 3 2 <1 <1 15 4 A. saman 33 1 1 <1 <1 <1 <1 <1 <1 3 7 Jackfruit 61 2 2 2 <1 <1 <1 <1 <1 7 6 P purpureum 94 6 5 5 1 1 1 1 <1 20 1 Local grasses 89 4 3 3 2 <1 1 <1 <1 14 4 Sweet potato 89 4 6 6 3 <1 <1 1 <1 21 1 TOTAL 26 25 24 11 3 5 5 2 100 Relative Importance 4 7 5 5 8 * Entries are percentage of all "votes" for each species x evaluation combination. Species rank and relative importan ce of criteria reflect respective sums of rows or columns. Gliricidia and local grasses followed in popularity. The "less important" evaluation criteria for Besakih farmers were animal health, ease of establishment, fast growth, dry season productivity, and availability of planting material. Moving downslope, 18 farmers were interviewed in Petang. Farm size was a mean 0.6 ha. Cassava was the most important crop, followed by citrus, coffee, banana, cacao, claves, peanut, coconut, ginger, papaya, and maize. As in Besakih, farmers in Peteng relied on their linear fields for tree fodders (G. sepium throughout the year and E. orientalis and jackfriut in the dry season) and P. purpureum, as well as local grasses and crop residues--banana stalk, cassava leaf, and coconut fronds (Table 1). Eighteen farmers were interviewed in Sakti on the small island ofNusa Penida offthe southem coast of Bali. Mean farm size was 2.0 ha (range 0.3-7.0 ha, mode 1.5 ha). The island receives less rainfall, has a drier dry season, and has poorer (limestone) soils than the Balinese uplands. Main crops were cassava, maize, coconut, banana, and beans. Mean number of cattle was three head per family (range 1-5, mode, 2 head). Cattle were fed by tethering (largely on each farmers' own lands, often under coconut) and by cut-and-carry. Farmers relied on tree fodders, including G. sepium throughout the year and Ficus sp in the dry season. Local grasses were abundant in the wet season (accounting for 25% of cut-and-carry fodder); less so in the dry season. Banana stalk was an important feed so urce throughout the year (Table 1 ). Although farmers in Sakti agreed that Sesbania sp was superior to all other forages in terrns ofweight gain, palatability, and animal health, only one fourth ofthe farmers maintained the tree, which accounted for only 4% offeed in the wet season and 2% in the dry. Sesbania was not more widely adopted because of its short life span. On the other hand, although Ficus sp was viewed as providing poor quality fodder, it 30 Alternative Land Use Options served asan "insurance" feed source in the dry season adopted by 50% ofthe farmers and providing 23% of d1y season fodder. Forage project cooperators, non-cooperators, adopters, ami non-adopters in Sumatra, Vietnam, and nortlzern Mindanao. The FSP site in Marenu, Sumatra, is a recently settled transmigration site. Farmers, both FSP project cooperators (n = 1 O) and non-cooperators (n = 8) reported having a mean of one ha; although sorne may have had more land and reported the "official" land holding for settlers. Cooperators hada mean 34 head of sheep; while non-cooperators had 19. Main income sources for cooperators were sheep, upland crops, and off-farm labor. Cooperators additionally claimed lowland rice and oil palm as main in come sources. Cooperators complained of wild pigs, drought/lack of water, lack of capital, lack of job opportunities, and sheep theft as problems. Non-cooperators saw pigs, lack of capital, and drought/Jack of water as problems. lt appeared that non-cooperators had fewer sheep than cooperators, but were more successful in terms of off-farm employment and in the establishment of lowland rice paddies and oil palm plantations. Sheep were fed by combined grazing on commons and cut-and-carry for momings and evenings when animals were penned. Both cooperators and non-cooperators planted grasses and trees. Rates of adoption for severa! grasses were higher for cooperators, with non-cooperators relying more on King grass than cooperators. Half ofthe cooperators compared to none ofthe non-cooperators had sown Stylosanthes guianensis; and more cooperators had adopted and were using G. sepium and L. leucocephala compared to non-cooperators. Non-cooperators relied more upon local grasses than cooperators in the dry season (Table 4). Farmers' evaluations offodder species were also recorded in terms ofpositive and negative qualities of each (see Table 7). Table 4. Forage use, FSP cooperators & non-cooperators, wet and dry seasons, Mareo u, Sumatra Cooperators Non-cooperators 0/o 0/o farmers ws DS farmers ws DS Trees 252 19 13 12 G. sepium 901 10 7 38 4 3 A. saman 50 6 5 50 4 4 L. leucocephala 90 9 7 50 5 5 Legumes S. guianensis 50 8 4 o o o Local grasses 80 12 17 63 19 32 Planted grasses 45+ 50 61 45 P. atratum 90 15 15 100 20 15 P. guenoarum 100 16 12 75 11 8 B. humidicola 60 4 12 25 7 6 B. decumbens 60 7 6 50 9 6 S. sphacelata 20 3 3 13 1 1 King grass 10 <1 2 63 13 9 Total 91 90 93 80 % of fanners using a particular forage; % of fodder coming from a forage type or species 31 Altemative Land Use Options Ten FSP cooperators and 8 non-cooperators were interv iewed in Xuan Loe near Hue in central Vietnam. Besides producing lowland rice and sugar cane, a lmost all farmers were tree planters. Most had fa irly large numbers of fruit trees; a large proportion managed reforestation areas under government contract; anda high proportion had family land similarly sown to plantation forests. Comparing cooperators and non-cooperators, cooperators had more Iand (mean 2.4 ha vs 1.6 ha), but--for families having each enterprise--similar areas of lowland rice (0.2 ha), sugar cane (0.2 ha), areas under family forestry ( 1.3-1.4 ha) and numbers of cattle ( 4.0-4.4). Greater proportions of cooperators, however, had sugar cane, family forestry, and cattle (67% vs 50%). A1though fewer non-cooperators had water buffalo, those having such animals hada higher number per family. More non-cooperators cared for government forest plots, but had smaller areas (7 .5 ha) compared to those cooperators (9.7 ha) having such contracts (Tab1e 5). Table 5. Production assets, Xuan Loe, Vietnam Participants (n=JO) Non-participants (n=S) % sample Mean Range % Sample Mean Range Fann size (ha) 100 2.4 0.3- 8.0 100 1.6 0.3 - 3.4 Paddy area (ha) 83 0.2 0.1-0.3 88 0 .2 0 .1-0.2 Sugar cane (ha) 89 0 .2 0.1-0.3 57 0.2 0.1 - 1.0 Family forestry (ha) 83 1.4 0.1 - 5.0 29 1.3 0.5-2.0 Contract forestry (ha) 27 9.7 6.0-20.0 50 7.5 6.0-10.0 Fruit trees (units) 100 104 23-290 86 140 29-280 Cattle (animals) 67 4.4 1-10 50 4 .0 1-10 Buffalo (animals) 44 1.7 1-3 25 4.0 3-4 Farmers identified problems as lack of water for crops, lack of capital, low soil fertility, and lack of transport--followed by a lack of labor for grazing livestock and a lack of grazing land. Cooperators were just becoming familiar with sorne of the grasses and fewer of the trees through testing (on small plots) as a part ofFSP activities. The main evaluation criteria used by farmers were palatability, "quality", yield, weight gain, and animal health. lfthe evaluations offarmers using particular species and evaluation criteria are compared (without reference to the actual proportions of farmers actually using a given forage and/or evaluation criteria), native grasses were given highest marks dueto high scores in terms of quality, palatability, and yield. P maximum was also rated highly across criteria, and especially in terms ofpalatability. The trees G. sepium and L. leucocephala, although planted by 77% ofthe informants, scored low across criteria. Factoring in proportions offarmers p1anting a given forage and using particular evaluation criteria, native grasses, P. maximum, and S. guianensis (which al! were testing or using) were given the highest ratings (Table 6). Ironically, farmers' tree planting practices appeared to work against the adoption of fodder trees. Most farmers planted a wide range of fruit trees in their home gardens and cared for forest plantations on both their own and on government lands. Introduced fodder trees had to compete with fruit trees in the borne gardens and with commercial timber elsewhere. Because farmers perceived the potential for receiving high (and apparently low-risk) returns to forestry, the enterprise competed with 1ivestock husbandry. 32 Alternative Land Use Options Table 6. Fodder assessment (n=13), Xuan Loe e riteria = "' o ·; L. 'O 'O <1.1 C,!) E :.e <1.1 <1.1 o ..... ";,.e ü ..... .... L. ~ .e e~.~_ ~.:.e ~ 'O el) s.::: "; ..:.c L. ~ .;! ~ ~ 'Gj ·- ~ - = L. - L. = ~ ~ = ~ = <1.1 o ~ o o o ~ o ~l. o > <::Z:: E-o ~ U E-o u~ % used criteria lOO 85 77 70 38 S guianensis lOO 2 3 3 3 3 14 4 14 P maximum 70 6 4 4 4 2 20 2 14 1 B ruziziensis 31 3 7 3 2 2 17 3 5 6 Native grasses 46 7 8 7 4 3 29 1 13 1 G sepium 77 2 2 2 2 1 9 5 7 5 L leucocephala 77 2 2 2 1 3 10 5 8 4 TOTAL 22 26 2 1 16 14 Re1ative importance 2 1 2 4 4 Corrected total 22 22 16 11 5 Relative importance 1 1 3 4 5 • Entries are relative mean scores for those planting a given species and using a given criteria. Ranking of species and re lative importance of criteria were calculated from sums of rows and colurnns, respectively. Corrected totals and ranking reflect proportion ofthose using the species and criteria. Maturing forest plantations also resulted in less available natural fodder for either grazing or cut-and- cany. Sorne farmers had reduced their animal numbers; and the community as a whole may reduce cattle and buffalo numbers further to the point where draft needs are met. Future forage adoption will depend on the relative economic importance of lowland rice, sugar cane, forestry, and livestock. The importance of livestock will depend on needs for draft, the importance of farmyard manure, and the long-term investment advantages of cattle compared to forestry. A guess would be that cattle numbers would either stay the same or decrease. Livestock enterprises may, however, intensify in response to demand from Hue for higher quality feed produced on small on-farm areas. A short period was spent in Maltibog in northem Mindanao in the Philippines. Small farmers have one or two head of cattle fed by tethering and cut-and-cany. Main crops are bananas, maize, and coconut. Although FSP cooperators were testing a range of new forages, many appeared to be interested in the possibility of receiving cattle via govemment dispersa! programs (which traditionally required adoption of new forages as a pre-requisite). The high availability ofbanana stalk and open grazing lands meant that fodder resources were available, a factor working against new forage adoption. One community had a large area of mature L. leucocephala trees, which was not being used as a major fodder so urce. On the other hand, dry-season fodder shortages and increasing demand for meat in the city of Cagayan de Oro may eventually lead to an increase in the genuine adoption of new forages for cattle-fattening enterprises. An evaluation offorage species across sites. Farmers across sites were asked to name positive and negative characteristics associated with their different forage options (Table 7). The results were 33 Altemative Land Use Options Table 7. Farmers' evaluations of fodder species Species Calliandra spp G. sepium Albizia sp L. /eucocepha/a Sesbania sp Ficus sp Jackfruit Erythrina sp Stylosanthes sp P. purpureum P. atratum P. guenoarum P. humidicola Good: Bad: Good: Bad: Good: Bad: Good: Bad: Good: Bad: Good: Bad: Good: Bad: Good: Bad: Good: Bad: Good: Evaluations Weight gain, yield, palatable, animal health Root competition, must mix w/other fodders, reduces animal fertility High yield, weight gain, animal health, milk production, palatable, prevents diarrhea, easy to grow, easy to harvest, cutting tolerant, long life Not palatable iffed to much, must mix, lowers cattle fertility, leaf fall in OS, pests, slow regrowth, pests Commercial wood, weight gain, palatable, fast growing , drought tolerant Oifficult to harvest, diarrhea, not palatable, slow regrowth, excess leads to hair loss Palatable, high milk production, easy to harvest, drought tolerant, cutting tolerant, quick regrowth Pests, must mix, excess causes ewes to bleed Animal health, weight gain Short life Produces in OS, long life Low nutritive value, low yield, one harvest per year, not for animal health, shade competition A vailable in dry season/drought resistant, timber, prevent diarrhea Low nutrient value, constipation Weight gain, palatable, animal health Low yield, diarrhea Nutrition, animal health, good for weanlings, palatable, drought tolerant Old growth not palatable (OGNP), itchy, not drought tolerant, slow regrowth, not palatable Weight gain, fast growing, perennial, yield, palatable, easy to harvest, available Bad: OGNP, must mix, no contribution to animal health, needs fertilizer Good: Bad: Good: Bad: Good: Bad: Quick regrowth, cutting tolerant, drought tolerant, high Jeafyield, easy to harvest, palatable, all parts consumed Sharp edged, OGNP Fast regrowth, cutting tolerant, yield, easy to harvest, palatable, weight gain, produces in OS Not drought tolerant, OGNP, rots if cut too low Orought tolerant, quick regrowth, cutting tolerant, palatable, easy to maintain Less Jeafproduction, OGNP, crop competition, cannot plant other crops on same Jand after 34 Altemative Land Use Options Table 7 Farmers' evaluations of fodder species (continued) Species King grass Setaria sp Local grasses Sweet potato Cassava leaf Banana stalk Good: Bad: Good: Bad: Good: Evaluations Easy to establish, quick regrowth, palatable, easy to harvest, yield OGNP, crop competition, not drought tolerant, difficult to maintain, itchy Quick regrowth, palatable, drought tolerant Short life Natural mixtures for animal health, fast growing, weight gain, palatable, available Bad: Low productivity in dry season Good: Bad: Good: Bad: Good: Bad: Fattening & animal "fmishing" Diarrhea Palatable, weight gain Not palatable 3 days after harvest, bloat Animal health, provides water in DS, palatable, easy to harvest, increases milk production in DS Diarrhea if fed in excess aggregated beca use of the substantial consensus across si tes in the three countries ( albeit, farmers at each si te had a different suite of forages and, therefore díd not evaluate all species ). The legume trees, Calliandra sp, G. sepium, and L. leucocephala were viewed positively in terms ofyield, palatability, animal weight gain, and animal health . Negative characteristics included the need to mix leguminous tree fodder with other fodders, pests, and leaffall in the dry season (G. sepium). Sesbania sp fodder was considered of especially high value in N usa Penida, but was not more widely planted because of its short life span. Although viewed as producing fodder of low nutrient value, Ficus sp and jackfruit were valued for their needed dry season productivity. Vietnamese farmers appeared to prefer to plant fruit rather than fodder trees in their home gardens. Albizia sp and jackfruit were valued for their timber as well as fodder. Although farmers agreed that Stylosanthes spp was good in terms of animal health, nutrition, and weanlings, s low re-growth and itchiness (for farmers harvesting the fodder) were described as problems. lnformants disagreed asto the palatability and drought tolerance of Stylosanthes. Most of the p lanted grasses were found to be desirab le in terms of fast growth, high yield, palatability, weight gain, and ease of harvest. Common complaints about the grasses included that old growth was not palatable and competed with crops. Farmers sought grasses, which were cutting tolerant, drought tolerant, adapted to low soil fertility, and were fast to re-grow. Farmers across sites generally favored natural grass mixtures as being fast growing, good for animal health and weight gain, palatable, and, of course, available. In sorne areas, lowered production in the dry season was mentioned as a problem. 35 Alternative Land Use Options Sweet patato tubers and lea ves were u sed for cattle fattening and "finishing" in sorne of the upland areas of Bali. Cassava leafwas commonly used as fodder in many areas, and was also viewed positively in terms of animal weight gain. Banana stalk was a significant fodder source at several of the sites. Among several positive characteristics was that it a lso provided water in the dry season. Impact: Recent studies range from pessimistic to hesitantly optimistic conceming the adoption oftrees on farrn, especially for fodder. Case studies in Nepal and India have shown that, in spite of increased tree planting for fuelwood and shifts from open grazing to stall feeding, farrners have relied heavily on crop residues and forage grasses to meet animal feed needs. Researchers concluded for these cases "In contrast to the previous analysis of fue!, trees on farrn do not appear to be a viable strategy for livestock feed (Wamer et a/1999). Another review offorage husbandry in the tropics concluded that, "A wide and diverse range oftrees and shrubs are used as fodder, but few are planted. When they are planted, it is seldom primarily to provide forage. Rather, forage is a by-product of fruit trees, live fences, and erosion- control strips, and makes the planting ofthese trees more attractive to farrners" (Bayer and Waters-Bayer 1998:139). Farrners in the highlands ofNicaragua (in an area somewhat similar to the sites visited in Java), on the other hand, used Lacuna spp and G. sepium as fodder sources, and maintained naturally occurring Guazuma ulmifolia and Acacia pennatula trees beca use of their dry-season forage productivity (Nicola Maria Keilbach, personal communication, cited in Bayer and Waters-Bayer 1998). A collection of studies from South Asia and Eastern Africa indicated, in general, that the (albeit few) observed shifts to more intensive on-farrn tree planting was occurring in regions undergoing agricultura! intensification, and that this intensification has taken place in the more arable and productive areas with relatively higher rainfall (Arnold and Dewees, 1997). For the areas visited in this study, severa! factors would appear to affect decisions regarding forage and tree forage adoption by smallholder farrners with mixed crop and livestock systems. Tree adoption was encountered where a combination of relatively high populations o ver a fixed land area had led to agricultura! intensification. In Bali, such intensification featured high to exclusive reliance upon cut-and- carry feeding for penned animals; anda high reliance upon on-farrn planted forages. In these cases, off- farrn commons or open access areas supplying grazing land or fodder for cut-and-carry were not available. lndeed, in Besakih every plant--trees, crops, weeds--was privately owned. Fodder tree adoption also appeared more likely where fanners were already agroforesters growing a range of trees for a variety of purposes. Agroforestry itself also appeared more likely where systems were not largely reliant upon shade intolerant annual crops such as upland rice or maize. Finally, fodder trees were likely to be adopted where a marked dry season significantly decreased the relative availability of fodder from non-tree compared to tree sources. The presence of adequate fodder sources in the forrn of open grasslands, grasslands under coconut, crop residues (e.g., banana stalk in Indonesia and the Philippines), and the growing offield crops for animal feed (sorne ofthe sweet patato in Besakih) would tend to decrease adoption offodder trees. Livestock serve as a "bank account" for many small farrn systems. Family forestry (in Vietnam) served the same purpose and was viewed as a better long-terrn investment, thus "competing" with livestock as an enterprise. Farrners in project areas may also genuinely adopt new forages as they shift from herding and grazing to increased stall feeding (e.g., goats in Marenu) or spuriously in the hope ofreceiving animals through cattle distribution programs (e.g., Malitbog). These factors are synthesized in farrners' decision tree (Figure 1). 36 ls livestock an economically viable on- fann enterprise? .. Js natural forage plentiful in WS but scarce in the DS? Ifyes, can other on-fann fodder sources lill the gap? a) Would trees fill DS needs better than grasses? ~ b) Would trees be too compelitive with crops? ~ e) Are there trees w/long life span and fast regrowth after cutting ? ~ IF a) YES, b) NO & e) YES, PLANTTREES .. .... .. .. ls natural forage plentiful throughout the year? lfno, is the best choice to decrease livestock? + If no, would fann area allow for planting forages? + IF YES, TEST FORAGES .... IF YES TO 80TH, CHANGE TO CA TTLE FATTENTNG t lf yes, would profit be high enough, risk low enough? i Would the new forages allow a change to animal ..... fattening? Alternative Land Use Options IF YES, DO NOT ADOPTNEW FORAGES lfyes, decrease livestock., do not plant forages 1 Would other options be more prolitable than livestock? a) Would grasses compete with crops? ~ b) Are there grasses tolerant to cutting, fast regrowth? 1 e) Are there high yielding grasses? -¡ IF a) NO, b) YES & e) YES PLANT GRASSES Figure l. Farmer forage & tree adoption decision model 37 Altemative Land Use Options lmplications for tlze forages for smal/lzolders project. The FSP is correct in offering farrners at selected sites menus offorage grasses, legumes, and trees; and in facilitating farmer-participatory research in the testing ofthe introduced materials. The fieldwork reported on here provides severa! other suggestions: l. There needs to be continuing emphasis on demand for introduced forages in relation to existing forage resources and the changing relative profitability of livestock over other on- or off-farm enterprises. There may be little opportunity for intensification where livestock simply take advantage of available native forages or where other enterprises such as forestry would "compete" with livestock. Though it is acknowledged that in many cases the FSP has found that farmers like to ha ve small areas of improved forages to use at those times when labor is limiting for collection of existing forages or for herding livestock, even though other fodder sources may be available. 2. Areas undergoing intensification--e.g. where land is becoming less available and penned animals are replacing grazing--would be likely for the adoption of new forages . "Linear fields" such as those encountered in Bali may be appropriate for mixes of trees and grasses where open fields are not available. 3. Farrners in areas with more available land and natural forage resources may still be interested in new grasses and possibly trees ifthere are clear advantages in terms of dry season productivity. Farrners were willing to plant or use fodder trees producing inferior feed as long as dry season production was assured when needed. The El Nino related drought appears to have generated interest in Kalimantan when the new forage species provided the only green to be seen (Werner Stur, personal communication). 4. Farmers expressed a range of perceptions regarding the suitability of legume forages. In general, although good for weight gain and animal health, farmers also thought that legumes needed to be mixed with other foods, that animals refused to eat more than small amounts, and that fertility- related problems could arise. If not already doing so, the project may need to work with farmers willing to experiment with feeding regimes to determine the soundness of such perceptions. Farmers at one FSP site are apparently now more interested in G. sepium after recently finding that their goats would, contrary to previous belief, consume loppings from the tree (Werner Stur, personal communication). 5. Further research is needed on the gender and age distribution of labor for cut-and-carry systems. Although mal e informants generally claimed to con tribute equal shares of labor as women, observations give the impression that women contribute more for cut-and-carry and that children provide more for grazing and tethering. Women may have less involvement in fodder or tree planting decisions; and the opportunity costs of children's labor may be low. Both factors could reduce new forage adoption. 6. Crop residues were a major animal feed source in the areas visited. The FSP may want to integrate crop residues within any on-farrn research. 7. Where natural forages are plentiful, the FSP may want to work with farmers to address the resource use/access issues associated with such forages in order for farmers to beneficially improve management ofthe resource. Communities may be able to work together on enriched natural pastures, for example. 8. Finally, and to repeat severa! points above, farrners did not appear to be worried about the establishment costs in terms of time to productivity and care of seedlings associated with trees. 38 Alternative Land Use Options Competition with crops, longevity, recuperation and regrowth after lopping, tree pests, and fodder suitabili ty were the main concerns. References Arnold, JE and PA Dewees, eds . 1997. Farms, Trees & Farmers: Responses to Agricultura! lntensification. London: Earthscan. Bayer, W andA Waters-Bayer. 1998. Forage Husbandry. London: MacMillan Education LTD. Warner, M, PG Bezkorowajnyj, and JR Witcombe. 1999. Matching Iivelihood systems to tree selection in high potential farming systems: lessons from participatory research in Nepal and India. ODI Agricultura! Research and Extension Network paper 89. 39 Alternative Land Use Options Activity 1.3 Determine the indicators of human nutrition and healtlz in the Aguaytía waterslted 1.3.1. Health, Biodiversity and Natural Resource Use on the Amazon Frontier: An Ecosystem Approach Purpose: To developing and test a participatory ecosystem approach to human health assessment that provides insight into the causallinkages between natural resource use, biodiversity, health and nutrition. The following hypotheses are being evaluated: • A more productive ecosystem results in higher nutritional status • Greater biodiversity on and surrounding the farrn results in higher nutritional status • Nutritional status will vary between native and colonist communities dueto differences in culture and local knowledge of biodiversity and ecological resources • The shift from more subsistence-oriented to commercial production systems does not necessarily result in an increase in nutritional status • The shift to commercial production affects the role of women in the household, their time allocation and involvement in health and nutrition decisions • Nutritional status will vary by age and sex within the household • Nutritional status will vary throughout the year dueto seasonal differences in nutrient intake Rationale: Ucayali and the surrounding upper rainforests are of great biological significance, containing 23% and 44% ofknown tropical plant and bird diversity in the Neotropics (IUCN 1996). Despite the natural diversity and fecundity of the setting, rural communities face a range of nutritional and health problems. Chronic malnutrition affects 47% of rural children below age five (INEI 1997), anemia and vitamin A deficiencies are widespread and incidence of malaria, dengue fever and persistent diarrhea continues to rise (Direccion Regional de Salud 1997). Health research has demonstrated the importance of nutrition to child development and human health (Pelletier et al. 1993; Bailey et al. 1998). In the study area, agricultura( production, fishing and wild food extraction, provide the main sources of food consumed. Nutritional status and human health is therefore dependent on the family 's ability to ensure sufficient nutrient intake through a diversity of production activities. Exploitation of resources in the surrounding ecosystems result in di verse seasonal combinations of farrning, fishing, logging, and hunting and gathering activities. Within such populations, the relationship between household production, income level and nutrition is complex and poorly understood (Leonard et al. 1993). More specifically, the nutritional status of Amazonian populations in Peru has largely been ignored, and the impact on nutrition, and particularly gender relations, of recent shifts to commercial production, has yet to be explored Methods: Current research activities include: • Characterization and assessment of the livelihood systems of riverine and upland communities with respect to: i) ii) iii) diversity of food and income sources; agricultura! diversity and biodiversity of wild foods gathered from surrounding forests; contribution of different land uses (fishing, farrning, hunting and gathering, logging) to dietary quality and food security. 40 Altemative Land Use Options • Assessment of household food security and nutritional status of individual m en, women and children, with emphasis on seasonal changes in nutrient intake and deficiencies of micronutrients (iron, zinc, vitamin A, B and C) • Characterization of child morbidity patterns and disease periodicity • An ethnographic study of local beliefs, knowledge and decision-making processes that affect the selection of foods and the treatment of illness. Once the above field activities are complete, the project will: • determine the significant correlates of nutritional status and identify the causal linkages between natural resource use, health and nutrition through bivariate and multivariate analyses, • develop with each community their own definitions and indicators of individual and community health using participatory methods, • develop and implement Community Action Plans (CAPs), drawing on the problems and opportunities identified through the research and the community's own definitions ofhealth. Building institutional collaboration. From the outset ofthe project, local representatives of government and non-governmental organizations have been involved in refining the research objectives, designing the specific methodology and carrying out the research. Initially, focus groups and key informant interviews were held with all relevant organizations to confirm the key health issues in the regían and provide a richer description of the institutional setting and current efforts in the health arena. These meetings included; the Ministries of Health, Fisheries, and Agriculture, AMUCAU (Asociación de Mujeres Campesinas de Ucayali), AIDESEP and FECONU (Federación de Comunidades Nativas de Ucayali), OMIAP (Organización de Mujeres Indígenas de la Amazonía Peruana) and IIAP. At this time inter- institutional agreements with CIA T were formalized. Tite research team. In addition to the core research team of six full-time staff, a larger team of local representatives was established. This group meets regularly, providing feedback on research activities and advice on future work. More importantly, the direct involvement ofthese representatives in the actual implementation ofthe research (household surveys) has contributed greatly to the exchange of information, training of local researchers, complementarity of research efforts and a greater interest and sense of local ownership over the research and its findings. Through the formal agreement with the Ministry of Health, they ha ve provided an environmental health expert for testing the quality of water sources and taking stool samples for parasitic infections. In addition, a recent graduate from the nursing program is conducting her one-year rural residency with the project. The Director of FECONU is leading the study on the role of the fisheries in nutrition and facilitating community participation in the native sites. And lastly, two health promoters from the governrnent nutrition program (Kusiykulu) and AMUCAU are conducting the anthropometric assessments at all sites. Site selection was a critica! step in the research. Specific criteria were identified in order to capture the heterogeneity of the region and all local partners were involved in the process. Eight si tes were selected based on differences in; • ecosystem type (riverine versus upland forests) • ethnicity (native versus colonist) • access to markets and involvement in market economy • time of settlement ( early versus old frontier) • dominant land use strategy (slash and burn agriculture, fishing, cattle ranching and oil palm plantations). 41 The research sites include: Riverine communities: Upper Ucayali River Aguaytía River Upland communities: Palmeros*(Neshuya-Curimaná) Ganaderos (Campo Verde-Km 80) Hierbas Buenas (Km 38) Altemative Land Use Options Cunchuri (mestizo) Puerto Belén (native) Caco Macaya* (native) Santa Rosa (native) Naranjal* (mestizo) Oil palm growers (mestizo) Cattle ranchers (mestizo) Slash-and-bum agriculture (mestizo) Cao Macaya, Naranjal and Neshuya-Curimaná are sites for the ethnographic study. Two research methods are being used: i) household and community surveys of all eight sites (3-4 days at each site), and ii) ethnographic study of three sites (1 0-12 days at each si te). Researcb visits for the surveys are being carried out for: i) household and community surveys in June-July 1999 (dry season), October-November 1999 (start ofthe rainy season), February-March 2000 (height ofwet season). This captures the seasonal changes that result from the 1Om ríse in ríver levels in the rainy season. The affect ofthe hydrological cycle on food availability, water quality and dísease incidence is critica) to understanding the health and ecological dynamics ofthe rainforest and its inhabitants. ii) ethnographic study, research visits are in; August-September 1999 (dry season), Jan- February 2000 (wet season). Tables 1 and 2 indicate the frequency of collection and períod of recall of each variable measured. Withín tbe household, the primary female and maJe caregivers are tbe main infonnants. Children are divided into four categories (0-23 months, 2-5 years, 6-10 years, 11-18 years) and one child is selected randomly from each category. Within each research site, between 32 and 50 families are interviewed, depending on the size ofthe community. ln the ethnographic study, participatory research metbods are used, with specific emphasis on gender dynamics, intra-household food allocation, local beliefs and knowledge of health and nutrition, ethnographic food classification and decision-making processes regarding the diagnosis and treatment of childhood illness. All community members are participants in the ethnographic study. The Ecosystem Approach. The goal of the Ecosystem Approaches to Human Health Program lnitiative at IDRC is to improve human health by supporting trans-disciplinary research on the structure and function of stressed ecosystems on which people depend for their livelihoods and by applying this knowledge to the development of appropriate and effective interventions and policies. ln contrast to the more traditional medica) model of health, the ecosystem approach focuses on the relationship between health and ecology. This shift in focus to the surrounding ecosystem and its impact on health, forces us to consider for example, how floods and poor water quality affect parasitic infections and incidence of diarrhea, how biodiversity increases dietary quality and reduces susceptibility to micronutrient malnutrition, how soil infertility and land degradation reduce household food security and how the lack of hygiene and sanitation practices facilitates the spread of infectious disease. 42 Alternative Land Use Options The ecosystem approach is guided by three principies; it is hierarchical, participatory and adaptive. Ecosystems exist in nested hierarchies; each comprised of smaller systems while at the same time being part of a larger whole. The different layers (individual, household, community, region) evolve within a variety of ecological and socio-economic contexts and constraints constraints (ABen et al. 1982; 1993; Conway 1987). Research and intervention must therefore target all levels, recognizing that the decision- making constraints imposed by govemance structures at the regional and national leve) impact on the array of options available at the household leve!. An ecosystem approach is inherently participatory. lts ultimate goal is to create a locally-driven sustainable process to assess, analyze and alleviate malnutrition. lt aims to enhance local capacity to cope with key problems affecting human health, increase self-reliance and reduce the need for externa! funds and expertise. A participatory approach allows local stakeholders to defme their own goals and develop a sense of ownership over the research findings and desired interventions. Lastly, an ecosystem approach is adaptive, recognizing that socio-ecological systems are dynamic and often unpredictable. Interventions emphasize the development of social and human capital, with the understanding that the capacity of people and institutions to adapt to sudden and unexpected stresses, is what ultimately leads to a more sustainable and resilient system. Projected Outputs: (i) Report of nutritional and health status of men, women and children in se\ected communities and the significant detenninants of nutritional status (ii) Analysis of the impact of seasonality on resource use activities, food availability, income levels and disease periodicity, and identification of critica! periods when cycles of disease and food insecurity most severely threaten the health of women and children (iii) Community Action Plan (CAPs) for each site, addressing key health and nutrition problems and incorporating local definitions and indicators of human health (iv) Dietary guidelines for each community, based on the availability of local sources of nutrient-rich foods (v) Educational programs and irnproved nutrition programs targeting high-risk groups and critica! periods when the cycles of nutrient intake and disease most adversely affect health. (vi) Small-scale food production projects in each comrnunity, aimed at increasing the diversity of foods consumed, eg. family gardens, small animals, fruit orchards and fish ponds. (vii) Short course developed for local university students in agronomy and health addressing the issues of malnutrition and health in Ucayali, based on the insights gained at the eight research sites. Projected Impacts: This project has two key potential irnpacts. First, the resulting analyses of nutritional status will allow more targeted and effective nutrition rehabilitation and health programs at the regional leve!. At the site leve!, community action plans, dietary guidelines and small-scale food production projects will increase dietary quality and diversity and reduce nutrient deficiencies. Second, the project may play an important role in the assessment of altemative land use options for the Pucallpa benchmark site. Despite a myriad of intemational studies, the link between agricultura! development and health and nutritional status is not yet clear (von Braun et al. 1994). Increases in production and income that result from improved technologies and agricultura! intensification may, but do not necessarily, lead to changes in health and nutritional status. As one ofthe main goals ofPES is to evaluate the biological and socio-economic impact of different land use options, this project provides the methods and data to broaden this evatuation and include the consequences for human health and nutrition. Contributors: Funding Sources: This is a joint CIA T-University of Guelph Project, funded by IDRC's Canadian- CGIAR Collaborative Research Grants Program. In-kind contributions from CIA T, University of Guelph, 43 Altemative Land Use Options Instituto de Investigaciones de la Amazonía Peruana (IIAP), and the Mínístry of Health, Ucayali, include scientist's time, transportation, field equipment and laboratory services. Collaborators: The official collaborating organízations are CIA T; Ministry of Health, Ucayali; Minístry of Fisheries, Ucayali; Instituto de lnvestigacion de la Amazonia Peruana; University of Guelph; and the University of Waterloo. lndividuals involved are: Tamsyn Murray (CIAT Pucallpa!Univ. ofWestern Sydney Hawkesbury); David Waltner-Toews (Univ. ofGuelph); James Kay (Univ. ofWaterloo); Douglas White, José Sanchez Choy, Carolina Rengifo, Yolanda Malqui, Francisco Econaise, Maria Elena De Aguila (CIA T, Pucallpa); Tania Tuesta, Víctor Gonzales (Ministry of Health, Ucayali); Juan Carlos Saavedra (Ministry of Fisheries, Ucayali); Rafael Uriqui, Adolfo Pino (FECONU); Sam Fujisaka (CIAT, Palmira), Peter Berti (PATH Canada). References: Allen, T.F. & Starr, T.B. 1982. Hierarchy: Perspectivesfor Ecological Complexity. University of Chicago Press, Chicago, 31 O pp. Allen, T.F., King, A.W., Milne, B., Johnson, A., Turner, S. 1993. The problem ofscaling. Evolutionary Trends in Plants 7(1): 3-8. Bailey, K., M. de Onis, M. Blossner. 1998. Protein-energy malutrition. In: Murray, C. & A. Lopez (eds) Malnutrition and the Burden of Disease: the global epidemiology ojprotein-energy malnutrition, anaemias and vitamin dejiciencies. Volume 8, The Global Burden of Disease and lnjury Series. Conway, G. R. 1987. The properties of agroecosystems. Agricultura/ Systems 24: 95-117. Direccion Regional de Salud, Direccion de Estadísticas e lnformatica. 1997 INEI. 1997. Población. Mujer y Salud. Resultados de la Encuesta Demograjica y de Salud Familiar IUCN. 1996. The Conservation Atlas ojTropical Forests: The Americas. C.S. Harcourt & J.A. Sayer (eds) Simon and Schuster, New York, 354pp. Leonard, W., K. Dewalt, J Uquillas & B. Dewalt. 1993. Ecological correlates of dietary consumption and nutritional status in coastal Ecuador. Ecology of Food and Nutrition Vol 31, pp 67-85. Pelletier, D., E. Frongillo, & J. Habicht. 1993 . Epidemiologic potentiating effect ofmalnutrition on chi1d morta1ity. American Journal of Public Health 83: 1130-1133. Von Braun, J. & E. Kennedy. (Eds) 1994. Agricultura/ Commercialization, Economic Development and Nutrition. Johns Hopkins Press for IFPRI, Baltimore. 411 pp. 44 Alternative Land Use Options T bl 1 D t C 11 ~ S a e . a a o ectwn or urveys Level lndicators Variables Survey 1 Survey 2 Survey 3 Methods Freq.of collection/ Ju~Ju 99 Oc~No99 MaApOO Period of recall Landscape Ecosystem Soil fertility X Soil tests Once at each site productivity Community Access to food Food prices X X X Observe 3X Food availability X X X Observe 3X Access to health Health services X X X Interview with Once on initial visit/ services available health worker Include changes during study Household Socío~conomic Wealth index X Recall Once status Non-food X Recall Once. Recall period expenditures varíes with item Nutritional status Energy and nutrient X X X 24 hour recall 3X for prior 24 hours Food Security intake Dietary Quality Nutrient intake X X X Food frequency 3X for prior 7 days recall Dietary diversity F ood sources X X X Recall 3X for prior 24 hours lncome leve!; Income by source X X X Recall 3X, round 1 for prior diversity growing season, round 2 and 3 for period b/w visits. Production levels; Production and X X X Recall Sarne as above diversity extraction outputs Food security Crop storage X X Recall 2X, report what is in storage at time of visit Water access; Water Quality X X X Water sarnples Three times contamination Environmental health Hygiene Practices X Recall Once Child Growth and Anthropometric X X X Actual Threetimes development measurements measurement Nutritional status Energy and nutrient X X X Recall 3X for prior 24 hours intake Health status Breastfeeding history X Recall by mother Once on initial visit Diarrhea incidence & X X X Recall by mother 3X, round 1 for prior 6 pattems months, round 2 and 3 for period between visits Incidence and X X X Recall by mother Same as above pattems of respiratory infections Morbidity pattems X Recall by mother Once for prior six months and whole life Mortality X Recall by mother Once, deaths during survey period are recorded lron Status X X Measurement of Twice (PATH anemia haemoglobin strip) Parasitic infectíon X X Stool sarnples Twice 45 Altemative Land Use Options Level lndicators Variables Survey 1 Survey 2 Survey 3 Methods Freq.of coUectionJ Ju-Ju 99 Oc-No99 MaApOO Period of recaH Women Nutritional status Energy and nutrient X X X Recall 3X for prior 24 hours intake Female fertility Reproductive history X Recall Once on initial visit Health status Morbidity pattems X Recall Once, deaths during survey period are recorded Labour demand Time allocation X X X Recall 3X for prior 24 hours M en Nutritional status Energy and nutrient X X X Recall Same as above intake Labor demand Time allocation X X X Re cal! Same as above Table 2: Data Collection for Ethnographic Study Level Variables Summer Winter Methods Pa rticipants Freq.of AuSe99 JaFeOO "' number in brackets collectionJ indicates the # o{ l!roups period of recall Landscape Spatial mapping of X X GPS/GfS Fanners/fishermen/ Twice ecological diversity; hunters and gatherers lakes, swamps, forests, rivers, palm forests etc. Spatial mapping of X X GPS/GIS Farmerslfishermen/ Twice resource use activities; hunters and gatherers fanning, fishing, hunting, gathering and logging Community Sources of information X Observe/key informant Community leaders, Once interviews teachers, students Community organization X Observe/key informant Community Ieaders, Once interviews teachers Community facilities and X X Cornmunity mapping 2 groups of mal e comm. Twice resources members approx. 20 people Hygiene and sanitation X X Observation, 2 groups of female comm. Twice Community mapping & members approx. 20 Health walk people Disease periodicity X X Seasonal calendar 2 groups of female comm. Twice, summer (focus group) members approx. 20 and winter recall people periods Historical information X Timeline (focus Groups of 6-9 people Once, for entire groups) • Elderly ( 1*) history of • Fishermen (2) community • Hunters (2) • Farmers (2) • Women (3) 46 Altemative Land Use Options Level Variables Summer Winter Metbods Participants Freq.of AuSe99 JaFeOO * number in brackets collection/ indicates the # of groups period of recall Individual Diarrhea management X X Key infonnant interv./ Interviews with: Twice in dry and case histories and • 8-10 people wet seasons decision models knowledgeable of health issues • 1 O women with children with a recent diarrhea episode Nutritional ethnography X PiJe sort/food Small groups of 2-3 Once (local classification) attributes/attribute people include: rating • women with children < 5 yrs. (5) • women > 45 yrs (5) • men (2) • adolescents (2) • teachers (2) Health ethnography X Focus groups Focus groups include: Once • women with children < 5 yrs (2) • women > 45 yrs (2) • men (1) • adolescents ( l) Fertility (pregnancy, X Interviews lnterviews with 1 O womeo Once prenatal care, birth, diet, of different ages contraception etc.) Risk Management X X F ocus groups Groups of 6-9 people Twice (recall is to (identification of risks, • Elderly (1) the extent to frequency, predictability • Fishennen (2) which they can and coping strategies) • Hunters (2) remember) • Farmers (2) • Women (3) 47 Alternative Land Use Options Activity 1.4 Evaluate social and private trade-offs of alternative land use options 1.4.1 Fallow management strategies and trade-off analysis of the economic value of fallow periods in the Aguaytia watershed Purpose: To analyze the economic value and trade-off of fallow management strategies. Rationale: In slash-and-burn agricultura! systems, lands are fallowed to rejuvenate soil fertility and to reduce weed proliferation. Yet as land becomes scarce, farrners often shorten fallow periods (Thiele, 1993). However, shortened fallow typically come ata double cost, namely, lower crop yields and hígher labor requirements for weeding. Alternatives to the traditional fallow-cropping cycles of slash-and-bum agriculture systems include improved fallow and secondary forest management. Whereas improved fallow research attempt to make the rotation period shorter with the introduction of trees and leguminous cover crops, secondary forest management concentrates upon the non-agricultura! products provided by the maturing fallow. Both national and international research institutions commonly have initiatives to intensify the use of agricultura! and secondary forest land. lt is seen as a way to increase farrn earnings and reduce pressure upon the forest resources, intensification is seen to be a win-win situation. Y et the economic benefits and costs of the improved fallow and secondary forest management options are not well known. For example, an experiment by (Montagnini and Mendelsohn, 1997) conducted on an agricultural research station touted the potential benefits of improved fallows. An assumed real interest rate of 5% makes the results seem unreal as smallholder farrners typically face high interest rates in order to borrow capital funds. Again the whole issue offarrner adoption ofthe improved fallow technology option was overlooked. Outputs: The research will focus upon the following: the effect of fallow periods and other significant determinants u pon agricultural yields an economic val u e of resting the land i.e. the latent (non-use) val u e of the secondary forest. the sensitivity of agriculture-secondary forest tradeoffs to changes in interest rates and prices of agricultura! and forest products. Impact : The three IARC' s, ICRAF, CIFOR and CIA T, working in the Ecoregional Center in Pucallpa, have identified fallow asan integral component oftheir research agenda. Ajoint study has been planned between the three centers. As a result, planning and implementation has taken longer than expected but with an anticipated outcome of a more widely accepted output. Theoretical and empírica! analyses have commenced and will be completed in 2000. Contrihutors D.White and R. Labarta (CIAT); W. Gusman (ICRAF); J. Smith (CIFOR). 48 Altemative Land Use Options Activity 1.5 Synthesize of results from different outputs and activities 1.5.1 Effect of pasture intensification in the forest margins, Peru, Latin America Highlights: • Empirical evidence supports the hypothesis that forest scarcity is a prerequisite for intensification of agricultural technology. Purpose: To examine the ability of intensive pasture technology to positively or negatively affect deforestation rate and to identify conditions that lead to more adoption. Rationale: As part of a CIFOR conference entitled Agricultura/ Technology lntensification and Deforestation, CIA T presented results of its research experience with improved pastures. Two opposing schools ofthought regarding the effect of improved pasture technologies are found in the literature: either they increase deforestation, or decrease deforestation. Little empirical evidence is used to support either argument. In this paper, in addition to providing empirical results, we add a third perspective: pasture technologies can also have no effect upon forest cover. Methods: The paper employed empirical data from the research and extension consortium Tropileche. Since longitudinal data linking pasture and forest cover are scarce, a synthesis approach was used linking empirical results from three research sites (Costa Rica, Perú and Colombia). These sites provide a stage on which to compare the introduction and effects of an intensive technology option, improved feeding systems for small-scale farmer mil k and beef production, u pon the forest. Outputs: Given that the value offorested land in tropical Latín America (and elsewhere) is low, continued deforestation and extensive cattle production appear to be rational prívate choices. The forest margin regions of the Amazon provide an example of such land use. In the case of older forest margins, which are more developed regions such as in Central America and increasingly in South America, farmers intensify production in order to prevent degradation of pastures and avoid the higher cost option of expanding onto neighboring lands. Hence, the initial question posed by the conference, whether intensification causes increased or decreased deforestation has resulted in an unfortunate altemative hypothesis: forest scarcity is a pre-requisite for technology intensifica/ion. Causallinks between the intensification of pasture technology and deforestation are difficult to determine. The relationship contains the combined issues of land degradation, technology feasibility, adoption, and farmer incentives to preserve forests. Out of a set of bio-physical and socio-economic factors presented for each ofthe study sites, we distinguish one measure, Jand price, which captures the effect oftwo opposing forces: the level of development (i.e. market access) and the amount of forest cover. In areas where land prices are low, markets are immature and forest is relatively abundant. Whereas where Jand prices are high, markets are more developed and forest cover is scarce. Land price also serves as an ex- ante indicator for the feasibility of pasture technology adoption. A land use history framework pennits a linking of empírica) evidence from the case study si tes to test the hypothesis (Figure 1). On one side ofthe continuum lie nascent regions with low land values where intensive pasture technologies are not adopted. In these areas, both continued deforestation and extensive cattle production appear to be rational prívate choices. Hence technology adoption, the first step required to affecting forest cover, is not an attractive option. The Peruvian Amazon is an example of such land use. On the other side ofthe continuum, lie regions where improved pasture technologies are an attractive option and appear to be adopted yet the effect on forest cover is small since little forest remains. 49 Altemative Land Use Options Nevertheless, there may be opportunity for reforesting land areas. An intennediate position along the continuum is where adoption is feasible and may reduce pressure on forest cover. As land use becomes more mature, with less forest and better developed markets, land prices rise. In Colombia and to a greater extent in Costa Rica, fanners intensify production in arder to avoid degradation of pastures and the higher cost option of expanding onto neighboring lands. Figure l. A Laod Use Contiouum with respect to Markets, Laod Prices, Forest Cover aod Technology Adoptioo Nascent / Mature t-- Intennediate Market t-- ' ' Undeveloped Developed Market ,... Market Colombia Costa Rica Perú Increasing Land Price Expensive Land Inexpensive Land Adoption and Scarce F orest Abundant forest deforestation depend Adoption of and deforestation upon land price intensive pasture Rare adoption of relative to cost of technology intensive pasture intensification, technology productivity of altematives and market demand Pohcy mterventtons wtll be needed m arder to achteve long-tenn success m controlhng deforestatwn. In tropical Latín America where land degradation can spur further deforestation, technological advances that make land productive again are a critica) component of a policy response. Yet minimizing deforestation and land degradation are but two environmental objectives. A concurrent issue is the human welfare of those living at the forest margin. Technical advances such as improved pastures will achieve these multi- faceted objectives of human welfare and environmental quality as long as they are properly targeted and coupled with policies that will restrict deforestation or make it an unattractive financia) altemative. Impact: While perhaps unpleasing to sorne, the future of forest margin regions will continue to have cattle beca use of producer and consumer demands. In many frontier regions, fanners search for profitable land use altematives. Examples of desperation are the Pucallpa farmers who establish pastures without having cattle. The mere hope of having cattle is enough to compel the land use decision. Moreover, consumer demand increases for animal products are expected to be especially large. For developing countries, future expansion of the livestock industry is predicted to be so great that an IFPRI/F AO!ILRI, ( 1998) study calls it 'The Next Food Revolution.' Whether the necessary increases in agricultura! and animal production will be extensive or more intensive/commercial remains open to debate (May and Segura Bonilla, 1997). The implications of failing to implicate a relation between deforestation and intensification necessitates a shift from examining the positive or negative impacts of intensification on deforestation to one of how to guide research efforts so that deforestation and extensive land use become a less attractive option for fanners. Together, both technical research to increase land use productivity, and policy research to test ways to increase incentives for forest preservation, become a pressing global need. Contributors: D. White, F.Holmann, S.Fujisaka, and C.Lascano (CIA T); K.Reategui (DEPAM) References: IFPRIIF AO!ILRI ( 1998) Livestock to 2020: The Next Food Revolution. May, P.H. and Segura Bonilla, O. ( 1997) The environmental effects of agricultura! trade liberalization in Latín America: an interpretation. Ecological Economics 22, 5-18. 50 Alternative Land Use Options 1.5.2 A range of forage options for different farming systems in Southeast Asia Highlights A broad range of forage technology options identified for smallholder farms in Southeast Asia Purpose: To integrate forage technology options to improve feed resources and resource management in smallholder farming systems in Southeast Asia, in particular, in Indonesia, Laos, Philippines and Vietnam. Rationale: Livestock provide much needed stability in fragile upland agricultura! systems in Asia. They provide an opportunity to accumulate capital (which can be readily converted into cash in times of need), food, draught power and they contribute substantially to nutrient cycling through manure. Ruminants utilize freely-available local feed resources, such as residues from crop fields and communal grasslands, forests and roadsides. These feed resources are 'free' to the farmer as they cannot be utilized in other ways. With intensification of agrículturallánds, the communal feed resources are becoming scarcer, and, degraded by over-use. The options are to reduce the number of animals or intensify forage production. Once farmers perceive a need for p\anting forages, these can be grown in ways which not only provide feed but al so enhance natural resource management through suppression of weeds, improvement of soil fertility and control of erosion. Methods: During the last five years the Forages for Smallho\der Project (FSP), funded by AusAID, has been using farmer participatory approaches (see Outputs 2.4 and 4.6) and conventional 'forage evaluation' techniques (Output 2.4) to develop sustainable forage technology options for resource-poor upland farms in Southeast Asia. The project has been working at 18 si tes in Southeast Asia (Figure 1 ), which cover the main range of environments and upland farming systems ofthe region. Figure l. Project locations of the Forages for Smallholders Project (includes on-farm, regional and nursery evaluation sites) 51 Alternative Land Use Options Outputs: A summary ofthe forage technology options that are being evaluated and adopted by farmers in different agricultura) systems is shown in Table l. lntensively managed 'cut & carry' grass plots near houses or animal sheds are being evaluated and adopted in almost all areas regardless of the type of agricultura! system. Often fanners identified a lack of labor as a limitation and they see intensively managed plots as an attractive way of reducing the demand on labor of keeping livestock. In many cases farrners intend to use these intensively managed plots only at specific times. Examples are days when they ha ve to go to the market, sorne family members are sic k, or during periods of peak labor demand by other agricultura) activities. In other cases stalling animals facilitates manure collection for use on high return agricultura! crops such as coffee. Table l. Forage technology options for upland farms in Southeast Asia Forage technology options e e o "' ·¡¡; o .e .., -~ "' e; .., o e;: !:: ... ;:3 .., E ... e ... ... .., e O.Cil «S 00 o c ~ o. o. o ¡¡¡ (';- ~ ~ o. ~ "O o o u o .., ... ... u-o .., oC! u ... :> u u "O-"O 00 .., E .., E oC! .., o ~ "' 1 ... c o "' ~ e 00 ... .., :::1!:: ::s.C ::S f 5 ·;; "O c.. :> c.. :> o e OO ... 00 ... :J .., E o e o ... o .., ... .., u o :t u u u o(,) -l C. en 01) o e ~ c. c. o ~ e:- ..2 ~ o. ~ -a o o u e .... ... u u (J U "O o ~ "O 01) u > "O- e u E ~ u e ... ... e o lll :;¡ e f l u lll u 1 s ~ :=..e ::l :; ·;;: c. > c. > 01)- 01) é ... :J .§ o o o ... o u ... u u o u ... u o (J ..J0.05 <0.05 >0.05 <0.05 Stylosanthes guianensis (% ofdiet) o 5.1 10.6 3.0 8.8 482 15 5.5 12.5 3.2 8.9 196 30 5. 1 15.5 3. 1 8.8 321 45 5. 1 16.4 3. 1 8.9 517 Significance level >0 .05 <0.05 >0.05 >0.05 <0.05 These results show that the only variables that were affected by increasing legume content in the diet were urea in the milk and liveweight gain. The increase of urea in mil k had a linear and consistent pattem for both legumes. The higher rate of increase with P. phaseo/oides than S. guianensis reflects the higher N 7 1 Component Technologies concentration in the fonner. The response in liveweight gain was erratic with both legumes. Chemical analysis of the refused forage indicated that there was a slight reduction in total dry matter intake as the legume content of the diet was increased. The reason why an in crease in the protein content of the diet, which is reflected in the increasing urea in the milk, did not increase milk production is not clear. Crossbred animals were used and should have been able to respond to increased protein in the diet. Data from the chemical analysis of forage consumed may assist in the interpretation ofthe results. Im pact. U nder the condítions of energy/protein balance of the animals in this tria!, legume intake increased the leve! of milk urea nitrogen but not milk yield. Thus, urea leve! in milk has potential as a metabolic indicator to protein intake and where the only variable source is legume, then the amount of legume in the diet. Research has not demonstrated the value of grass-legume pastures for díary cattle in Pucallpa Contributors: M. Ara, M. de la Torre and C. Reyes (IVIT A), and l. Unchupaico (Universidad Nacional del Centro), Peru. 72 Component Technologies 2.2.1.2 Feeding, and cutting trials to refine methods of supplementation during the dry season using different forage resources. Highlights • Nitrogen intake was higher with more frequent feeding at high levels of legume supplementation • Frequent cutting of Crat/yia is more important than cutting height for producing a high quality feed supplement (i) In vivo studies on the complementarity between forage basal diet and legume supplement Rationale: Previous results from confined feeding trials suggested that in supplementing available forage sources to overcome nutrient deficiencies in ruminants, it is important to synchronize feeding of the higher quality forage supplement with the basal forage so that energy and protein are available concurrently. On the other hand, synergism among forages may vary not only with the types of forages fed but al so with how they are fed (leve! and frequency). Methods: Eight African type weathers (24 kg L W) fed a low quality grass basal diet were randomly allocated to 4 treatments of supplementation with sugar can e ( 60%) mixed with Craty/ia argente a ( 40% ). Treatments arranged in 4x4 Latin Square design were: Tl- Low level ofsupplementation (0.5% ofBW) fed once a day (AM), T2- High level of supplementation ( 1.0%% of BW) fed once a day (AM), T3- Low \evel of supplementation (0.5% of BW) fed twice a day (AM+PM) and T4- High leve) ofsupplementation (1% ofBW) fed twice a day (AM+PM). Measurements included quality ofthe basal diet and supplements offered intake, digestibility and N balance. Results: The low quality grass basal diet was low in CP (4.8%) and high in cell wall content (NDF 79% and ADF 44%). On the other hand, chopped sugar cane fed asan energy supplement was low in CP (3.1 %) but had low cell wall content (39% NDF and 24%ADF). The legume (lea ves of Craty/ia) fed had high CP (21%) and high levels ofcell wall (67%NDF and 37% ADF). Thus the supplement fed was high in energy and medium in protein (10% ofDM). Intake ofthe basal diet did not differ among treatments, but there were differences in supplement intake due to treatments. As expected, intake of sugar cane and Craty/ia tended to be higher when fed at the highest leve l. However, it is interesting to note that when supplements were offered at the high level, intake of sugar cane and Craty/ia increased with twice-a-day feeding relative to once-a-day feeding (Table 3). These differences in intake of supplements were not reflected in significant changes in DM or cell wall digestibility. Nevertheless, there was a tendency for higher digestibility with a twice a day feeding at both levels of supplementation. As expected, N intake was highest with increased leve) of supplementation (Table 4); however at the high leve) of supplementation N intake was higher with twice a day feeding. Since fecal and urinary N did not change with treatment, N retention was greater when sheep were given the forage-based supplements at the high leve! and twice a day (Table 5). lmpact: With the forage-based supplement fed it was evident that leve! and frequency of feeding hadan effect on nitrogen utilization by growing sheep. However, the results indicate that feeding twice a day would only be justified when high levels of forage-based supplements are offered. When the high leve! 73 Component Technologies ( 1% of BW) of sugarcane/Cratylia was fed twice a day there was a 33% increase in N retention relative to feeding the same amount of supplement once a day. This was not the case when the low level (0.5% of BW) was offered. Table 3. Effect of leve! and frequency of forage-based supplementation on intake and digestion by sheep fed a low quality grass. ltem Intake (gDM/kgBW/d) Basal diet Sugar cane Cratylia Digestibility (%) DM NDF ADF AM** 0.5%BW 25.7 2.5 b 2.0 e 53.0 54.1 51.6 Frequency and level of supplementation• AM AM+PM*** AM + PM 1% BW 0.5% BW 1% BW 25.2 25.2 25.7 3.3 b 2.8 b 4.0 a 3.1 b 2.0 e 3.6 a 53.0 55.4 56.6 54.3 57.7 57. 1 50.7 54.3 53.1 *60% sugar cane + 40% Cratylia argentea (leaves) ••supplement fed once a day at 0.5 or 1% ofBW •••supplement fed twice a day at 0.5 or 1% ofBW a,b,c Means different (P<0.05) SE 0.8 0.4 0.1 1.7 1.8 1.9 Table 4. Effect of frequency and level of forage-based supplementation on nitrogen utilization by ships fed a low quality grass. ltem Frequency and level of supplementation • AM** AM 0.5%BW 1%BW N Intake (g/d) 5.6 a 6.2 b Fecal N (g/d) 3.5 3.5 Fecal N,% N intake 62.5 a 57.1 b Urine N (g/d) 1.4 1.5 Urine N,% N intake 25.4 25.8 Retained N (g/d) 0.8 e 1.2 d *60% sugar cane + 40% Cratylia argentea (leaves) ••supplement fed once a day at 0.5 or 1% ofBW •••supplement fed twice a day at 0.5 or 1% ofBW a,b,c,d,e Means different (P<0.05) AM + PM*** 0.5%BW 5.6 a 3.3 59.9 a 1.5 28.6 0.8 e Contributors: W. Quiñonez, P. Avila, and C. Lascano, CIAT, Colombia. 74 AM + PM 1% BW 6.7 e 3.5 52.0 e 1.6 24.7 1.6 a SE 0.01 0.1 2.0 0.2 3.7 0.2 Component Technologies (ii) Productivity of Cratylia argentea cut at düferent frequency and height in the Central Pacific region of Costa Rica. Rationale. The legume Cratylia argentea has proven to be an altemative to fanners because it provides forage biomass with a high protein content during the dry season. It is necessary to have more details of a rnanagement strategy to maximize the quality and quantity of biomass of Cratylia that guarantees the sustainability ofthe bank in the long-term. Earlier cutting studies had shown that yield increased with high density of planting (1 0,000 to 20,000 plantslha) and that it could be cut as early as 4 months after planting without a long term effect on yield. In feeding trials, it has been observed the crude protein concentration of leaves is higher at a cutting frequency of 90 days (20%) cf.l80 days ( 16% ). As the reason for supplementing with Cratylia is to provide a high protein feed, it was thought useful to investigate the effect of cutting more frequently and also leam if cutting height affected productivity in the longer term. We present the first years results of a cutting trial investigating these factors. The experiment was conducted on-fann under researcher control. Metbods. 21 O plants from a pro te in bank of Craty/ia argente a established in 1996 were divided into 3 groups of70 plants each managed at three height levels: 30, 60, and 90 cm. Each group was then divided into two sub-groups of 3 5 plants each for two harvesting frequencies to give re-growth of 60 and 90 days. Plant spacing was 0.5 m between plants and l m between rows (20,000 plantslha). Chemical analysis for crude protein, non-detergent fiber, acid detergent fiber, and lignin was performed on the edible portions (leaves and thin stems of approximately less than 6 mm thick). Results. Results indicate that there was no significant difference in forage quality between cutting heights (30, 60 and 90 cm above ground leve!) for plants cut at either 60 or 90 days though the nutritional quality (higher CP and lower NDF) for material from plants cut at 60 than 90 days. Yield increased with a higher cutting height and at the lower cutting frequency (Table 5). Table 5. Evaluation of Cratylia argentea at two ages of re-growth and three cutting heights Re- Cutting Height Re- growth height of re- growtb DM CP NDF ADF Ligo in erowth Sboots o¡o o/o o/o (days) (cm) (cm) (#) o/o o¡o 30 0.61 8.58 28 .8 18.7 52.7 44.9 14.6 60 60 0.51 9.64 24.5 18.0 55.4 43 .9 15.4 90 0.70 13.44 22.5 19.1 57.9 43 .9 15.3 Mean* 0.60 10.55 25.3 18.6 55.3 44.1 15.1 30 1.10 9.88 32.3 16.2 57.0 44.1 15.8 90 60 1.15 12.69 34.4 15.2 59.5 46.1 16.5 90 1.24 14.53 33.9 15.9 61.8 44.2 16.0 Mean** 1.16 12.37 35.5 15.8 59.4 44.8 16.1 * Average of 5 cuts; * * Average of 3 cuts Impact. The slight reduction in biomass (8%) at the shorter cutting interval of 60 than 90 days is well compensated by a higher nutritive value ( 19% V s 16% CP). These results need to be made known to farmers growing Cratylia as a protein bank. lt is planned to continue this tria! over 2-3 years to determine if more frequent cutting is sustainable in the long terrn . Contributors: M. Lobo and V. Acuña, MAG, Costa Rica 75 Component Technologíes 2.2.2 Development of new feeding systems through on-farm research Highlights • Forage feeding systems for strategic supplementation during the dry season reduce the need for purchasing animal feed concentrates and increases income in dual-purpose fanns • The use of legumes with an energy so urce during the dry season maintains milk production 2.2.2.1. Milk production with dual-purpose cows grazing Brachiaria brizantha associated withArachis' pintoi and Centrosema brasilianum during the rainy season in the Central Pacific regio o of Costa Rica. Rationale: The overall constraint limiting animal productivity in the Central Pacific region of Costa Rica is the low quality of naturalized grasses. There is a need to develop new feeding altematives to overcome this limitation. In our on-farm work, sorne fanners have chosen to develop fodder banks (e.g., of sugarcane and shrub legumes). For others this did not seem feasible because ofhigh labor demand. Another altemative is to develop more productive grass-legume pastures to improve forage quality. This reports on-fann research with a fanner who did not wish to plant fodder banks and had relatively leve! land on which to establish improved pastures associated with legumes. Previously this fanner had had experience only with pure grass pastures. Metbods: Areas were selected for pure grass pasture and grass-legume associations. At the beginning of the 1996 rain y season, pure grass pastures and associations of 4 ha of B. brizantha with A. pintoi and C. brasilianum were established. There were two paddocks per treatment. The B. brizantha and C. brasilianum were established by seed at a rate of 4 kglha for each species and the A. pintoi with vegetative material at a rate of 1 t/ha. Both legumes were planted in furrows set at a distance 70 cm apart after the grass was established. Pastures were grazed for 5 days grazing by the dairy herd every month. Measurements were made only during the rainy seasons over the two years, 1997 and 1998 when milk yield was measured on days 1, 3, and 5. Results: Tables 6 and 7 show forage biomass availability, botanical composition, and milk yield of cows grazing pure grass and the grass-legume associations. Biomass production was higher during both years in the pasture containing the greatest proportion of legume, which suggests that nitrogen fixation promoted greater grass growth. Likewise, milk yield was 8% higher than pure grass where there was 30% legume and only 3 to 5% higher where there was 20% legume. Table 6. Forage availability and botanical composition of B. brizantha associated with different proportions of legumes during the 1997 and 1998 rainy seasons. Pasture with Pasture with 30% legumes 20% legumes 1997 1998 1997 1998 Biomass (kg DM!ha) 4110 4480 5110 5410 Grass (%) 62 67 57 58 Legumes (%) • 22 18 31 27 Weeds (%) •• 16 15 12 14 *Ca/opogonium muconoides, Zomia spp Aeschinomene spp, C. brasi/ianum and Arachis pintoi. • * Mimosa modest, Amaranthus spp, Borreria spp and others. Table 7. Average milk yield (kg/cow/day) of duaJ-purpose cows grazing pastores with different 76 Component Technologies proportions of legumes during the rainy season in 1997 and 1998. Type of pasture Milk production In crease (kg/cow/d) (%) 1997 1998 1997 1998 Only grass 8.6 7.6 Grass with 20% legumes 9.0 7.8 4.7 2.6 Grass with 30% legumes 9.3 8.2 8.1 7.9 Impact. The increase in milk production on the grass-legume associations occurred despite the fact that cows were receiving a daily supplementation of 5 kg of chicken manure and 1.5 kg of soybean hulls. Thus, the use of legumes in association with grasses increase milk yields, especially when the proportion is around 30% of total edible biomass. This could be due to a combination of higher biomass availability during the rainy season and higher N content in the diet. Contributors: M. Lobo and V. Acuña, MAG, Costa Rica 77 Component Technologies 2.2.2.2. Milk production of dual-purpose cows supplemented with Craty/ia argentea, chicken manure, and sugarcane during the dry season in the sub-humid tropics. Rationale: In ecosystems with prolonged droughts such as the Central Pacific region of Costa Rica, the production and quality of forage biomass from grasses is reduced drastically. Producers sol ve this limitation by supplementing with agricultural by-products such as chicken manure. This by-product was inexpensive several years ago. However, its demand has been growing and therefore, its price in real terms has increased. As a result, producers are interested in the evaluation of alternatives to replace the use of chicken manure. We report three on-farm trials, two evaluating the replacement of chicken manure with Craty/ia argentea and the other tria! evaluating Cratylia fed fresh oras silage. Methods: Experiment l. This tri al was conducted during the dry season ( April 1999) in a farm of a small producer located in Miramar in the Central Pacific region ofCosta Rica. Miramar is atan altitude of250 m.a.s.l. with an annual mean temperature of28°C and mean precipitation of2400 mm. Treatments evaluated were: TI= Control group, grazing only (naturalized andjaragua grasses), T2= Sugarcane + C. argentea + rice polishing + grazing, T3= Sugarcane + chicken manure + rice polishing + grazing. Nutritional characteristics of feeds were: Sugar cane: 2% CP, 3.0 Mea! ME. Chicken manure: 19.5% CP, 1.8 Mea! ME Rice polishings: 12% CP, 3.0 Mea! ME Fresh Cratlyia: 20% CP, 1.8 Mcal ME (from 90 day regrowth cut at 30 cm) Cratylia silage: 16.4%CP, 1.9 Mcal ME (from 180 day regrowth cut at 30 cm) A latín square design was used with the three treatments with two cows in each of them and with selected cows in the second month of lactation. Animals in treatment 2 received 12 kg of sugarcane, 8 kg of C. argentea, 0.6 kg of semolina, and 0.5 kg of molasses per cow in addition to grazing, while animals in treatment 3 received 3 kg of chicken manure instead of C. argentea. The experiment !asted for 30 days, animals were rotated through each treatment for 1 O days (7 days of adjustment and 3 for data collection). Experiment 2: This trial was conducted in a small farm located in Barranca in the Central Pacific region ofCosta Rica. Barranca is atan altitude of280 m.a.s.l. with an annual mean temperature of28°C and mean precipitation of2500 mm. Treatments evaluated were: T 1 = Sugarcane + silage of C. argente a + rice polishing T2= Sugarcane + fresh C. argentea + rice polishing; and T3= Sugarcane + chicken manure + rice polishing. A latín square design was used with three treatments with two cows in each one as in the above experiment. Animals were handled in individual groups and received for treatment 1 an averaged of 12 kg ofsugarcane, 6 kg ofCratylia si1age and 0.6 kg ofrice polishing as fed; treatment 2 received 12 kg of sugarcane, 6 kg of C. argentea, and 0.6 kg ofrice po1ishing; and cows in Treatment 3 received 12 kg of sugarcane, 3 kg of chicken manure and 0.6 kg of rice polishing. The experiment !asted for 30 days, animals rotated on each treatment for 1 O days each (7 days for adjustment and 3 days of data collection). Experiment 3: This experiment was conducted on-farm as part of student thesis program. This tria! was conducted in the middle of the dry season (February to May, 1998) in a small farm located in Barranca in the Central Pacific region ofCosta Rica. Barranca is atan altitude of280 m.a.s.l. with an annual mean temperature of28°C and mean precipitation of2500 mm. Treatments evaluated were: 78 TI= Chicken litter and molasses, T2= Chicken litter, sugarcane, molasses and wheat bran, T3= Chicken litter, molasses, C. argentea, and wheat bran Component Technologies Details ofthe dietare presented in Table 10. The diets were balanced to be isonitrogenous and isocalorific with different quantities of supplements used. In 0 3, C. argentea contributed more than 75% of the CP requirements of animals but a small amount of chicken litter was included in the diet since observed intake of C. argentea was not sufficient to balance N requirements. Animals had access to H rufa pastures. The crude protein (CP) content and in vitro dry matter digestibility (IVDMD) for Cratylia was 19.2% and 58.4%, respectively, and for H. rufa was 3.9% and 33.9%, respectively. A 3x3 Latín square change over design with three replications was used with a total of9 crossbred cows between 60 to 80 days into the lactation. Each ofthe three experimental periods consisted of 10 days of adaptation to treatments and 5 days experimental data collection. Results: Experirnent 1: Table 8 shows milk yield and composition, feeding costs, income from milk, and the benefit: cost ratio for the different treatments evaluated. As can be observed, milk yields between treatments 2 and 3 were not significantly different (p=0,076), which suggests that C. argentea can be used to substitute for chicken manure without a decrease in milk production. In addition, these results indicate that when milk is produced with Cratylia argentea, the feeding costs are lower and the benefit: cost ratio greater, thus making this altemative more attractive economically to producers. Table 8. Average milk production and composition, supplement feeding cost, milk income, and benefit: cost ratio of diets supplemented cows with Cratylia argentea and chicken manure. Mil k Total Supplement yield solids Fat Feeding Cost In come B:C Treatments k2/cow/d (%) (%) ($fk2 milk) ($fk2 milk) Ratio Grazing only 5.45b 11.2 3.1 - 0.24 Grazing + Cratylia 5.85 a b 11.5 3.2 0.11 0.25 2.22 Grazing + Chicken manure 6.29 a 11.2 2.9 0.22 0.24 1.08 Experiment 2: Table 9 show that the feeding altematives based on Cratylia argentea, both as silage and as fresh leaf, were more economical than the altemative to supplement with chicken manure. In addition to Craty/ia rations being more economical, the milk yields of cows consuming Craty/ia either as fresh or as silage were similar to those with chicken manure. However, milk yields from cows consuming Cratylia silage were lower than those from cows consuming fresh Craty/ia. Nevertheless, Craty/ia silage might be used as a substitute for chicken manure because of lower feed costs. Table 9. Average milk yield of dual-purpose cows supplemented with Cratylia argentea either fresh '1 d 'th h' k oras s1 a~e an Wl e 1c en manure. Total Supplement Mil k Milk yield solids Fat Feeding cost loco me B:C Treatments (kg/cow/d) (%) (%) ($fk2) ($/ke) Ratio Cratylia as silage 5,09 b 12,33 3,65 0.17 0.27 1.58 Fresh Cratylia 5.47 a 12,22 3,45 0.11 0.27 2.37 Chicken manure 5.26 a b 11 ,7 1 3,00 0.22 0.26 1.14 79 Component Technologies Experiment 3: Table 10 shows the effect on milk yield by treatment. Milk yield averaged 6.0 kg milklcow/day and there were no significant differences between treatments. Milk fat was lower (2.7%) for the treatment which had a higher amount of sugarcane in the diet, but differences were not significant. The results show that Cratylia can be used as a substitute for chicken manure, at least to 82% of the diet. Table 10. Diets offered to animals and milk yields for all treatments. Diets Diet #1 Chlcken manure Molasses Diet #2 Chicken manure Sugar cane Wheat bran Molasses Diet #3 Chicken manure Mo1asses Wheat bran Cratylia argentea Consumption (kg!animaVday) 6.0 2.5 5.0 5.0 0.7 0.12 1.0 4.12 0.7 6.0 Milk yield (kg!cow/day) 5.9 6.0 6.1 Impact. Most farmers with dual-purpose cattle in recent years have been producing milk during the dry season using chicken manure. Results from this work clearly show that Cratylia can largely replace chicken litter as a protein supplement, without a significant reduction in mílk yíeld. This is a very important finding given that the demand for chicken litter to supplement dairy cows has increased its price in many tropical regions and fanners can no longer afford to huy it. With the introduction of Cratylia argentea to cattle production systems in the Pacific region, the utilization of chicken manure has declined in farms participating in the Tropileche Consortium to such degree that this year practically none was purchased. Therefore, the legume technology being promoted for dual-purpose cattle farms has the possibility of allowing small producers to have access to a farm- grown protein supplement and to increase cash tlow and profits. In addition, spontaneous adoption of Craty/ia argentea is occurring in other locations in Costa Rica. During the last 12 months 84 kg of experimental seed has been sold to 28 farmers located in three different sites in Guanacaste, Nicoya, and around Esparza. The most economical option for a producer in the dry season is to supplement cows with fresh Cratylia. The next best option is to supplement with Cratylia as silage. The least economical option is to supplement cows with chicken manure. As a result, the use of legume silage such as Cratylia argentea is recommended over the use of chicken manure. In addition, the use of legume silage allows producers to utilize smaller areas because more edible biomass is produced for dry season supplementation, especially in situations where the opportunity cost for labor during the rainy season is low. Contributors: M. Lobo, V. Acuña, and J. Araya (MAG), M. lbrahim, M. Franco, D. Pezo, andA. Camero (CA TIE), Costa Rica 80 Component Technologies 2.2.2.3. On- farm evaluation of new forage alternatives in dual-purpose farms in Honduras and Nicaragua. Rationale. The adoption of improved pasture technology is more certain if farmers participate in the eva\uation of selected and advanced pasture gerrnplasm coming from research. Forage components based on adapted grasses and legumes have been identified by CIA T which give increases in animal productivity as well as increasing soil organic matter and activity ofthe so il macrofauna. Tropileche has an on-farm component that allows farmers to participate in the development of new pasture technologies and speed up the adaptive process and subsequently dissemination and adoption. Methods. Component technologies developed in Tropileche have been shared with national institutions that have included their evaluation in their research and development. CIA T has collaborated by providing basic seed for experiments, in advice on pasture establishment and on-farm methodologies. The collaborating institutions have paid for seed and establishment costs and for follow-up activities. Results. During 1998 and 1999, 63 ha ofnew pasture components were successfully established in Honduras and Nicaragua (Table 11). B. brizantha cv. La Libertad and the promising new line, B. brizantha CIA T 2611 O, either in monoculture or associated with Arachis pintoí ( cvs. Porvenir and Maní Forrajero), are the two grasses being demonstrated. Plots have been also established of the shrub legume Cratylia argentea (CIA T 18668) and Leucaena /eucocephala (CIA T 17263). Table 11. Areas planted with new forage options during the period July 1998-August 1999 by collaborators of the Tropileche Project in Nicaragua and Honduras. Country aod Collaborator Nicaragua (Proyecto de Desarrollo Lechero) Honduras (DICTA) Si te Esqui pulas Rivas El Rama Yoro- Yorito Coma yagua Species planted B. brizantha cv. La Libertad + A. pintoi cv. Porvenir cv. La Libertad Cv. La Libertad C. argentea L. leucocephala Cv. La Libertad Cv. La Libertad + A. pintoi B. brizantha 2611 O +A. p intoi Cv. Llanero +A. pintoi C. argentea Cv. La Libertad +A. p intoi Cv. Llanero +A. pintoi 81 Area P laoted 2.8 6.6 21.8 0.7 0.7 7.0 (39.6) 7.5 3.5 2.0 1.0 7.5 2.0 23.5) Farmers Involved # 1 2 4 l 1 (10) 3 2 2 1 (10) Component Technologies Twenty dual-purpose farmers are participating in this activity. The increased forage production from the new forages in comparison with the traditional grasses, such as Jaragua and Paspalum spp.is self-evident. Objective monitoring ofthe impact on animal productivity, in particular, on the increase in milk has not been done yet dueto two reasons: (a) lack ofresources ofthe national scientists, especially transportation and per diem, and (b) the Hurricane Mitch, which damaged a significant proportion ofthe established pastures as well as most rural roads. lmpact. Progress has been made during the last year in relation to the establishment ofnew forage components in collaboration with national institutions and farmers in Honduras and Nicaragua. The Tropileche project has been used as a platfonn to test and promote new forage altematives at the farm leve l. The collaboration and acceptance by farmers has been good, but the documentation of the impact on animals and soils by the collaborating institutions has not been implemented yet. On-farm work is costly and demands not only time and effort, but also resources like transportatíon facilities and operating capital for gas and per diem, which are frequently scarce at the national institution level, especially in poor countries such as Nicaragua and Honduras, both considered with Haití the poorest countries in Latín America. Given this reality, it is suggested that we need to work with these national program in obtaining sufficient resources in the form of operatíng capital to complement on-farm data collection and monítoríng. Contributors: Tito Fariñas (IDR, Nicaragua), Conrado Burgos (DICTA, Honduras), P. J. Argel and F. Holmann (CIA T). 82 Component Technologies 2.2.3 Evaluation of new feeding alternatives to allow early weaning. Highlights • Supplementing pre-weaned calves with legumes during the dry season results in more milk for sale and higher liveweight gain that with current management practice of feeding only grass • The legumes grown for feeding calves contribute to soil N and might be incorporated in a rotational crop-legume fallow system 2.2.3.1 Use of Stylosanthes guianensis with pre-weaned calves in dual-purpose cattle systems in the forest margins of Colombia Rationale: The two main outputs in dual-purpose cattle fanns are (a) milk and (b) weaned calves for fattening. Under traditional management, farmers usually favor selling as much milk as possible to increase their cash flow, but as a result calves suffer from under nutrition and mortality rates are high. Therefore, development of feeding systems that allow farmers to obtain more milk for sale and the same time result in good performance of pre-weaned calves is of high priority in dual-purpose cattle systems. The idea of using Stylosanthes guíanensis for grazing pre-weaned calves has been tested in Pucallpa, Peru. Results indicate that with this altemative fanners can sell almost one more liter of milk/cow/day and still maintain adequate growth oftheir calves, which has important economical benefits. To further test the use of Stylosanthes for pre-weaned calves, we initiated a collaborative study this year with COPRPOICA partners in Macagua!, Caqueta, Colombia. Methods: A small paddock (2 ha) of Stylosanthes guianensis was established in the CORPOICA research station in Macagua!, Caquetá to allow 1 to 3 months calves to freely graze after milking. Calves with access to Stylosanthes also received residual milk (milk remaining in the udder after hand milking) after each milking. Calves in the control treatment received milk from one quarter ofthe udder at milking and had access toa grass pasture after milking. In all cases calves remained with their dam for 3 to 4 hours after milking, before going to the grass or Stylosanthes pastures. Results: The amount of milk for sale resulting from the use of Stylosanthes by pre-weaned calves was 21% higher than recorded with cows that had calves managed in the traditional systems (Table 12). In addition, liveweight gain of calves with access to Stylosanthes was 30% higher than in the control group during a 90-day period. Impact: The results obtained in forest margins of Colombia on the use of Stylosanthes guianensis for pre- weaned calves are in agreement with those obtained in small dairy farrns of Pucallpa, Perú. This technology could be very attractive to smaU dairy producers given that the cost of establishment of this legume is less than the establishment of legume-based pastures for the milking herd and it results in increased cash flow due to the extra milk for sale, without affecting calf performance. In addition, the Stylosanthes technology could form part of a crop-pasture rotation system reducing the need to fallow land for secondary forest regeneration and subsequent slash and bum for annual crop production. This is because well managed Stylosanthes pastures can persist for 3 or 4 years and during this time produce a beneficia! effect on the soil through N fixation and nutrient cycling. 83 Component Technologies Table 12. Mil k for sale and growth of pre-weaned calves with and without access toa Stylosanthes guianensis pastura in Caquetá, Colombia. Item Milk for sale (Vcow/day) Liveweight gain of pre-weaned calves (g/ A/d) 1Six cows with calves 2Six cows with calves Control 3.3 297 Stylosantbes pasturé 4.0 389 Contributors: J. Velasquez, G. Ruiz, and C. Lascano, CORPOICA and CIA T, Colombia. 2.2.3.2. Rice yield (O riza sativa L.) as an indicator of soil fertility with the incorporation of Stylosanthes guianensis stubbles in Pucallpa, Peru. Rationale. It has been demonstrated that pure fodder banks of Stylosanthes guianensis can be used for calf feeding (Tropileche Annual Report 1998). However, Stylosanthes only persists for 2-3 years under grazing and then needs to be replanted. As it is a legume that can also add N to the soil for a following crop, it was thought that its adoption by farmers might be enhanced if it could be demonstrated that it had a dual role, feed for calves and green manure in a rotation system. This study reports the evaluation of Stylosanthes as a legume for a potential crop-legume fallow system. Methods: The experiment was conducted on a farmer's field with low fertility soils, acid pH, and highly saturated with aluminum (ultisoils). Farm is located near Pucallpa which is situated 270 m.a.s.l. and with annual mean precipitation of2000 mm and mean temperature of26°C. The experiment started in October of 1998 and concluded in F ebruary of 1999. Treatrnents were: T 1 = rice without N fertilization, T 2 = rice after S guianensis without N fertilization, T 3 =rice after S guianensís with 50 kg N/ha, T 4 = rice after S guianens ís with 1 00 kg N !ha, Ts = rice after S guianensis with 150 kg N/ha, and T6 = rice after S guianensis with 200 kg Nlha. The variables evaluated were: rice yield, number of tillers/m2, number of spiklets/m2, number of grains per spiklet, and weight per 1000 grains. The rice variety planted was "Chancabanco" with a 90-day vegetative period. All treatrnents received a basal application of 50 kglha of potassium chloride and 50 kglha as rock phosphate. The design was a completely randomized block with 6 treatrnents and 5 replications. Results: Table 13 shows rice yield, number of tillers, and number of spiklets. Rice yield after the incorporation of Stylo stubble was significantly higher than the control. There were no significant differences toN fertilization above 100 kglha N. these results demonstrate that the incorporation of Stylo stubble can make a significant contribution ofN to the soil for a following crop and there may be economical benefits from a rotation system incorporating Stylo. Impact. Stylosanthes guianensis is a legume with the potential to benefit small farmers when used in a rotational system with agricultura! crops because it can be used both as animal feed and increases crop yields through increasing available soi l N. 84 Component Technologies Table 13. EfTect of incorporation of S. guianensis and N fertilization on rice yield, and number of tillers and spiklets in Pucallpa, Peru. Treatments Rice yield (kg/ha) Tillers (#/ml) Spiklets (#/mz) Control 380 e 102 a 136 b Sty/o 520 b 111 a 154 ab Stylo + 50 kglha N 600 b 119 a 154 ab Stylo + 100 kglha N 830 a 125 a 174 ab Stylo + 150 kglha N 870 a 130 a 187 a Sty/o + 200 kglha N 890 a 136 a 185 a • Values in a single column followed by equalletters do not differ significantly (P < 0.05) Contributors: Jorge Vela, IIAP, Peru 85 Component Technologies 2.2.4 On-farm trials established for evaluating new forages in the llanos of Colombia Highlights • Demonstrated the feasibility of establishing Arachis in the degraded pastures on livestock farms of the piedmont of the Colombian llanos, but not on farms in the well-drained savannas with soils of lower fertility Purpose: The main objective of the work is to determine the utility of new ecotypes of legume and grasses developed in the Forage Project (IP-5) to reclaim degraded pastures as compared with commercial cultivars in livestock farms ofthe llanos ofColombia. Rationale: One major limitation for beef and milk production in Neotropical savannas is degradation of introduced grasses, as a result ofnitrogen deficiency and overgrazing. Thus CIAT's Forage Project (IP-5) has been developing improved grasses and legumes that can contribute to reclaim large areas of degraded pastures in tropical regions where livestock is a major land use system. Methods: In collaboration with IP-5 and CORPOICA we initiated in 1998 on-fann eva1uation of new grasses and legumes in representative fanns of the llanos of Colombia. A total of four farms (two in the well-drained savannas and two in the piedmont) were initially selected to eva1uate new ecotypes ofthe pasture legume Arachis pintoi. Selected farms were representative ofthe two sub-ecosystems and have large areas of degraded pastures. In addition, fanners participating in the Project indicated their willingness to cover sorne ofthe cost ofthe work done in their fanns. In each farm at 1east 8 ha of degraded Brachiaria spp pastures were used to establish the following treatments: a) Four ecotypes of Arachis pintoi (CIAT 17434-control and 18744, 18748 and 22160) and, b) Two legume planting densities (3 and 6 kg!ha) Land was prepared using a chisel plow anda disk harrow following overgrazing ofthe Brachiaria pastures. The seed Arachís seed was planted using a conventional grain row planter. The fertilizer used (kg!ha: 250 rock phosphate, 250 dolomite Ca, 150 Potassium chloride, and 25 su1fur) was broadcasted in al! the area. Soil physical and chemical characteristics were measured in al! farms before planting the legume. Measurements were a1so made on the above ground biomass ( cover, botanical composition, forage on offer and presence of pest and diseases ). In order to allow an estímate on changes in the soil o ver time, additional measurements were done in the soil (physical and chemica1) after p1anting Arachis. Post- establishment measurements include: rate of germination of Arachis seed (30 days after planting), cover of the legume (at 45 day intervals), botanical composition and legume and grass yield (in the rainy and dry seasons). Outputs: The establishment of Arachis sown in 1998 in association with Brachiaria spp was only successful in 1 of the 4 farms selected. In a l! farms the accession CIA T 22160 had very low germination and there were no measurable difference in the establishment rate of the other accessions (CIA T 18744 and 18748) relative to the commercial cultivar (CIA T 17434). 86 Component Technologies Pastures reclaimed in 1998- Piedmont: The fann where their was good legume establishment in association with B. humidicola (low to establish) is located in the piedmont with relatively good soils (2.5% OM) anda native-grass vegetation covering the degraded pasture chosen for rehabilitation. After sowing the legume-grass and applying the fertilizer a large proportion ofweeds genninated together with sorne B. decumbens (the original grass introduced in the pasture) but weeds were successfully controlled with a mechanical cutter. Initial grazing began 7 months after establishment with an average stocking rate of2.5 Alha. Legume proportion ofthe pastures was not significant\y affected by sowing rate and after 1 year, the botanical composition ofthe pasture was 34% grass, 40%Arachis and 26% weeds. Animal LWG has been very high (Ikglhead/day) most likely dueto compensatory growth. The second fann located in the piedmont where establishment of Arachis was poor had relatively good soils (2.8% OM), but the area chosen to plant the legume was dominated by B. decumbens in poor condition. Initial establishment ofthe legume following land preparation and fertilization was good, but at the end of the rainy season (December 1998) the proportion was low regardless of ecotype used or sowing rate. This reduction of legume in the pastures was most likely the result of competition from the vigorous grass that genninated after land preparation. It was not possible to minimize this competition given that the farmer did not ha ve animals to graze the pasture when needed. lt was decided to re-establish Arachis (CIA T 18744 and 18748) in the rainy season of 1999. The amount of Arachis is now ranging from 7 to 12% and intennittent grazing is being applied to favor the legume. Pastures rec/aimed in 1998- We/1- drained savannas: Establishment of Arachis was not successful in the two farms located in the well-drained savannas in spite of having good seed germination and having applied fertilizer. The lower soil fertility found in well-drained savannas as compared with the piedmont together with competition from sown grasses and weeds would explain the poor establishment of Arachis in degraded pastures. To obtain good Arachis-based pastures more fertilizer would have to be applied, but it is unlikely that fanners would be willing to pay the extra cost. In the on-farm work carried out so far in the llanos of Colombia to rehabilitate degraded pastures we have learned several things, which are summarized as follows: l . Farmers recognize pasture degradation as a major constraint for increasing milk or beefproduction, but have serious limitations of machinery for land preparation and sowing of new grass and legume species. In addition, farmers ha ve limitations to invest on fertilizers and on seed of legumes, usually more expensive than that of grasses. 2. Security aspects have detennined to a great extent that owners seldom visit their farms and as a consequence we have not been able to involve the owners in the on-farm work as much as we would have liked. 3. Successful establishment of Arachis is dependent not only on soil fertility, but also on the original vegetation in the degraded pasture, on the companion grass used and on grazing management. 4. Failures in the establishment of Arachis pintoi in degraded pastures in the low fertility soils ofthe well- drained savannas have detennined that we look for other legume options. 5. Demand for new grasses with drought tolerance and for shrub legumes to supplement milking cows in the dry season. New plantings: High priority was given in 1999 to the selection of new fanns to plant new grasses and legumes in degraded areas using as a criteria the willingness of resident farmers to actively participate in the evaluation. In Table 1 we summarize the new plantings in farms ofthe piedmont and well-drained savannas. 87 Table l. Plantings during 1999 in farms of the llanos of Colombia Sub-Region Piedmont Fann 1 Fann 2 Fann 3 Fanns 4 and 5 Well-drained savannas Fann 1 and 2 Fann 3 Species planted B. bryzantha CIA T 2611 O B. bryzantha ClA T 263 18 B. bryzantha CIA T 263 18 Desmodiun ova/ifolium (mixture of 5 accessions) + P .phaseo/oides (Kudzu) Cratylia argentea D. ovalifo/ium (mixture of 5 accessions) B. bryzantha ClA T 26318 Component Technologies Area (Ha) 2 3 5 20 V. to 1 ha 7.5 to 48 .5 5 As can be seen in Table l new plantings include ecotypes of Brachiaria bryzantha selected in the Red Colombiana de Brachiaria for high forage yield and drought tolerance. In addition, ecotypes of D. ovalifolium selected in a multilocational tria! in Colombia were sown as the legume component to recuperate degraded pastures in both ecoregions. lt is felt that D. ovalifolium is a better option to recuperate pastures in farrns in well drained savannas as compared with Arachis, given its good adaptation to low fertility soils and low seed rates (400 to 500 glha) required for establishment. The shrub legume Cratylia was established in two farrns as a fodder bank to supplement milking cows in the dry season. lmpact: It is to premature to talk about impact ofthe on-farm work in the llanos ofColombia aimed at testing with farrner's new grass and legume options to rehabilitate degraded pastures. However, it should be kept in mind that the current economic-social conditions in the region do not contribute to the desire of farrners to make large investments on pasture renovation on their properties. Thus the work being executed willlay the groundwork for future interventions. Cootributors: Camilo Plazas (CIA T- PE-5), C.E. Lascan o (CIA T -IP-5/PE-5), J. Miles (CIA T- IP-5), Raul Perez (CORPOICA) and A. Rincon (CORPOICA) 88 Component Technologies Activity 2.3 Evaluating legumes for feed supply, nutrient cycling and improved fallows 2.3.1 Identification of species for cover crops and fallow improvement, Cauca, Colombia Highlights • Com yield was increased l-2 fold following legume cover crops in a degraded hillside soil • Purpose: Identify legumes which in crease yields of subsequent crops with the participation of the community Rationale: Earlier trails have shown that cover crop legumes sown into a fallow can increase the yield of a foUowing maize crop depending on the quality and quantity of biomass of the legume used and the rainfall if there was adequate moisture for the com crop. Methods: Location: Centro Alternativo de Investigacion Agropecuaria "El Nilo" north Cauca, 3°4 'N and 76° 28' W, average temperature 23.5°C and altitud !200m masl. The area is in the foothills ofthe Central Andean range, with soft slopes and erodible terraces. It is used for crops and dual-purpose cattle. Precipitation is bimodal, March to May and October and December, with an average of l600mm/an. The soil is silty clay in texture with pH 5.2, OM 5.7%, very low P (Bray 11) 1.6 ppm, Ex.bases (meq/lOOg) are Ca= 7.2, Mg = 3.2 y K= 0.15. The site used for the experiment was a badly degraded pasture. Treatments were. : a) Legume covers • Mucuna pruriens CIA T 9349 • Pueraria phaseoloides CIA T 7182 • Canavalia brasiliensis CIA T 17009 • Centrosema pubescens CIA T 15160 • Crotalaria juncea CIA T 21709 • Cajanus cajan CIA T 1870 1 • Natural-fallow cover (witness) b) Ferti1izer applied to legumes: 10 and 30 kg!ha P e) Natural fallow which was also fertilized Two 1evels ofP were applied, dueto previous experience ofpoor legume establishment oflegume at similar low soil P levels. The research was carried out with the participation of the indigenous Cabildos from northem Cauca (ACIN) who had been resettled and had little farming experience on the particular site. There main concem was to increase crop yield so asto improve food security. The legumes were planted in October 1998 in plows, followed by maize in May 1999, two weeks after the incorporation ofthe legumes, The maize crop was fertilized with basal (kg!ha element): 50P, 50K, 20Mg, 25 Boronzinco (a mixture ofB, Cu, Zn) and two 1eve1s ofN, O and 100 kg/ha N. Results: Legume Cover: The 1egumes established quickly reaching 70% cover at three months with the exception of Pueraria whose establishment was affected by weed competition and Crotalaria which was attacked by ants. However, after six months at the time of incorporation they were well established. There were no significant differences between the levels of P during the establishment or in biomass produced. There were differences in y ield between legume accessions, the best being C. cajan CIA T 18701 with 4.5 tlha (Table 2). The legumes P. phaseo/oides and C. pubescens require more time to accumulate adequate biomass before incorporation. 89 Component Technologies Labi/e soi/ N at p/anting: The KCI extractable N03-N was increased by the legume covers. There was no significant difference to 30cm depth at the time of incorporation, but considerable differences 30 days after incorporation at of0-10 cm and 20-30 cm (Table 1). Total N was 3.5 times more than that ofnatural fallow. Table l. Totallabile nitrogen (N03+NB.) at time of ~lanting and 30 dais after ~lanting Legume cover Depth At planting 30 days after planting (cm) (ppm) C. brasi/iensis 17009 0-1 0 15.0 20-30 6.7 C. cajan 18701 0-10 13.2 20-30 5.0 C. }unce a 21709 0-10 8.6 20-30 5.4 C. pubescens 15160 0-10 10.7 20-30 5.8 M pruriens 9349 0-10 8.8 20-30 7.8 P. phaseo/oides 7182 0-10 10.4 20-30 5.7 Witness 0-10 6.6 20-30 3.3 LSD P< 0.05 0-10 ns 20-30 ns Table 2. Dry matter yields of legume covers at time of incorporation, and maize (averaged over P and N levels) Treatments Yie1d Covers Maize 21.6 19.9 12.7 14.4 10.9 12.8 12.8 13.7 9.9 12.4 12.5 13.7 6.0 5.8 7.0 8.0 at 6 months Grain Total tlha C. cajan 18701 4.5 5.1 16.7 M. pruriens 9349 2.8 4.4 14.2 C. pubescens 15160 1.4 4.0 12.5 C. brasi/iensis 17009 2.2 3.7 12.6 P. phaseoloides 7182 1.2 3.6 11.5 C. juncea 21709 3.3 2.7 9.6 Control -fallow 3.4 1.6 7.6 LSD P < 0.05 2.4 2.5 Ns Maize yield: There were significant differences in the yield of maize grain with different legume covers. Differences were not significant for total dry matter yield of maize plant MS or between levels of P applied to legume covers and N applied to maize. The y ield of grain was greater following Cajanus cojan, Mucuna pruriens and Centrosema pubescens with yields above 4 tlha. (This is surprising since C. 90 Component Technologies pubescens hada relatively low biomass yield (Table 2). The yields were higher than expected, possibly due to adequate soil moisture. Grain yield of maize increased 2.4 times following a legume cover crop (Table 3). This increase in maize yield can be attributed to the effect ofthe legume and not only the applied N as there was no increase from an application of fertilizer N. In all cases the yield of grain was twice or greater than twice that of the control following incorporation of legume. This confmns earlier results obtained in Cauca. Table 3. Performance of grain and total coro Iield {tlha} after legume im~roved fallow1 Treatment Level ofP Level ofN Grain Total DM Yield kglha tlha Legume1 10 o 3.8 12.5 Control fallow 10 o 1.3 8.3 Legume 10 100 3.9 12.9 Control fa\\ow lO lOO \.6 6.6 Legume 30 o 3.8 12.3 Control fallow 30 o 1.5 6.0 Legume 30 100 4.1 13.4 Control fallow 30 100 2.2 9.5 Average six legumes: Ce , Mp, Pp, Cp, Cj y Cb Impact: The results show that legumes that are well adapted and established rapidly can be used as cover crops to increase yields of a following crop. The community planted and managed the covers and maize and they have indicated interest in evaluating these results on a wider scale There is considerable self regeneration of so me of the legumes and it is planned to monitor the effect of this volunteer green cover. lt remains to be seen whether the legumes will self regenerate in drier years. Contributors: Luis H. Franco, Fredy Tabares and Peter Kerridge (CIA T) 2.3.2. Evaluation of herbaceous legumes as cover crops in rubber and oil palm plantations, Meta, Colombia Rationale: There is a need in the plantation industry ofthe Llanos ofColombia to find sustainable ways to reduce weed infestation, to maintain and improve soil fertility, to control erosion and increase the microfauna biomass. There is currently a trend to promote plantation systems in Colombia. In the rubber plantations the target group for this promotion are small to medium size farmers who want to diversify their fanning operations. In the oil palm plantations plots of up to 5 ha are leased out to landless fanners to manage the oil palms for the oil palm industry. Productive legume covers could reduce labor inputs and con tribute to soil N thus improving the efficiency of production and having an indirect effect on the welfare of resource poor fanners. To evaluate this hypothesis, in 1999, a range of legume accessions ofthe species Arachis pintoi, Desmodium heterocarpon and Pueraria phaseoloides have been sown under shade and no-shade conditions to evaluate them as covers. Contributors: M. Peters (IP-5), C. Plazas (PE-5). 91 Component Technologies 2.3.3. Rotational effects of five legumes on maize-bean intercrop production, Uganda Bighlights • The contribution of Canavalia ensiformis on maize-bean intercrop productivity was greater than for other legumes, due to more N flxation, more recovery of deep soil nitrates, and more accumulation of soil P in organic fonn . Purpose: To evaluate the effectiveness of severallegumes in the improvement of productivity, and to understand the reasons. Rationale: Alternative legumes have a potential role in the maintenance of soil productivity. Severa! have been identified as promising including Canava/ia ensiformis, Mucuna pruriens, Crotalaria ochroleuca, and Lab/ab purpureus. The causes of the rotational effects ha ve not been well understood. As the rhizosphere of sorne vigorously growing legumes is more active than for common food crops, we thought it likely that one or more of these legumes would acquire significant amounts of P from applied Busumbu roe k P. Methods: The above legumes were compared to soybean in a tria! of split plot arrangement over a period of five seasons, 1997a to l999a. Main and sub-plot treatments were legume species and P treatments, respectively. P was applied at 200 kg P20 5 ha·• as rock P, TSP ora mixture. The comparisons were for N- fixation ( 1~ atom excess with grain sorghum as the reference crop), proflle soil water (monitored with neutron pro be), soil nitrates and crop yield in a maize-bean intercrop, green manure rotation in a sub- humid, bi-modal rainfall system ofUganda. Outputs: The legumes and the food crops did not recover sígnificant amounts of P from the rock P. Maize-bean intercrop was most productive following canavalia and produced least after mucuna and soybean (Table 1). These effects generally persisted to the second season following legumes. Table l. Whole plant aod graio yield for intercropped maize and bean as affected by green manure of five legume species in the subsequent and second following season. Species Canavalia Mue una Crotalaria Lablab Soybean LSD 0.05 for legumes Grain yield (kg ha· ) Subsequent season ( 19978 Second following season Maize 3678 2625 2970 3224 2483 493 and 19988) ( l999A) Bean Maize 8ean 658 3747 430 508 3465 347 628 3523 313 490 3192 144 613 2516 443 94 465 97 Canavalia produced the most biomass and accumulated the most soil P in the plant biomass. Mucuna and canavalia fixed the most atmospheric N (67 and 39% Ndfa and 155 and 133 kg N ha·•, respectively); crotalaria fixed little N. Lablab and soybean produced the least biomass. Soil nitrate following a season of maize-bean production was highest where canavalia had been grown previously. Canavalia was the most effective Jegume in uptake ofsoil N, leaving 73 compared to 120 to 148 kg ha·• ofnitrate in 120 cm ofsoil depth, with the difference largely due to less nitrate at depth following canavalia. Profile soil water status was highest under soybean and lowest under crotalaria (cycle I) and canavalia (cycle II). Surface soil moisture content was similar for all species, but differences were evident at depth. Approximately ISO and 300 mm of rain were required to replenish pro file soil water during the intercrop following cycle I of crotalaria and cycle II of canavalia, respectively. 92 Component Technologies Impact: Canavalia and mucuna are promoted for use by farmers through training, distribution of leaflets, demonstrations and farmer-experimentation mini-kits. Canavalia is preferred by sorne for its ease of management. Mucuna is preferred by others who use it to feed dairy cows. Slow adoption is associated with weak extension efforts, but farmers ' knowledge ofthe species is increasing. Contributors: Charles Wortmann, Kayuki Kaizzi (NARO) and Beverly Mclntyre (Rockefeller Foundation). Linked to [1>2 which provided the funding. 2.3.4. Effects of legumes on infection of subsequent crops by rootknot nematode, Kenya Higblights • lnfection by rootknot nematode (Me/oidogyne incognita) was generally less with Crotalaria grahamiana, Mucuna pruriens and Lablab purpureus compared to other legumes, but severe following Sesbania sesban and Tephrosia vogellii. Purpose: To evaluate the rotational effects of several legumes on nematode infection in subsequent crops. RatiooaJe: Altemative legumes have a potential role in the maintenance of soil productivíty, fodder production, weed suppression and pest management. Tephrosia has become popular with farmers in parts of Uganda for control of mole rats. However, tephrosia, sesbania and sorne other legumes are known to be good hosts of meloidogyne nematodes which can be a serious pest ofbean, tobacco, tomato and sorne other crops. Other species, on the other hand, are noted for suppression of nematodes. Better understanding of the legume-nematode relationship is needed for better integration of altemative legumes in farming systems. Metbods: Three trials were conducted where soil was sampled from trial and demonstration sites, which allowed a comparison of legumes for effects on nematodes. Tomato seedlings, which had been established in sterile soil, were transplanted into pots containing these soil samples. At approximately seven weeks after transplanting, nematode infection was as ses sed by counting the number of rootknots per 50 cm of root length and per gram of fresh root weight. Outputs: Results are presented in Tables 1-3. Crotalaria grahamiana appears to very good at suppressing meloidogyne. Results for mucuna were inconsistent, having low levels of nematode infection in two trials but high infection in another where the soil for mucuna was taken from a weedy fallow. lnfection rates following sesbania and tephrosia were very high. Soybean and Crotalaria ochroleuca also were associated with high infection rates. Subsequent observations on farmers ' fields confrrrned that bean had more rootknots when in close proximity to tephros ia and sesbania. Table l. Tbe effects of five anouaJ Iegumes o o iofection of tomato roots by rootknot oematodes Legume species Rootknots Rootknots (50 cm· of roots) (g- of fresh roots) Canava/ia ensiformis 7.5 10.2 Mucuna pruriens 0.7 4.7 Crotalaria ochroleuca 17.2 26.7 Do/ichos lablab 0.5 6.5 Soybean 10.5 12.7 LSD 0.05 ns ns 93 Component Technologies Table 2. Tbe effects of sesbania and tepbrosia on infection of tomato roots by rootknot nematodes Legume species Tephrosia voge/lii Sesbania sesban Mucuna pruriens Weedy fallow LSD0.05 Rootknots (50 cm·1 ofroots) 20.7 26.4 1.5 1.9 9.3 Table 3. Tbe effects oftbree perenniallegumes and mucuna on infection oftomato roots by rootknot nematodes Legume species Tephrosia voge/lii, Kawanda Tephrosia voge/lii, Senge Sesbania sesban, Senge Mucuna pruriens, Senge (weedy) Crotalaria grahamiana, Kawanda Rootknots (50 cm·1 ofroots) 107 193 104 77 o Rootknots (g'1 of fresh roots) 80 93 73 54 o LSD 0.05 44.i 28.41 1 Error tenn is based on sampling error as soil was collected from single plots of legumes; plot effects are confounded with species effects. Impact: Crotalaria grahamiana and, to a lessor extent, mucuna and lablab should be useful in reducing meloidogyne numbers in badly infested fields, or preceding highly susceptible crops. Sesbania and tephrosia must be used with caution; non-susceptible crops should follow fallow improved with these species. Collaborators: Charles S. Wortmann and Imelda Kashiija (NARO). Linked to IP2 which provided the funding. 2.3.5. Crop response to application of altemative plant materials for soil fertility management Higbligbts: Plant parts of severa! naturally occurring species, which are often abundant in U ganda, resulted in increased soil productivity, but the benefit varied with the quality ofthe material. Purpose: To evaluate severa! naturally occurring plant species as resources for the management of soil fertility. Rationale: Resource poor farrners often have diverse but scarce resources which might be used in the management of soil fertility. Sorne naturally occurring herbaceous and shrub species produce abundant biomass of moderate nutrient Jevels. When these plants are near a cultivated field, the biomass can be 94 Component Techno1ogies transferred to supply the crop with nutrients. Earlier work indicated that farmers would be willing to use su eh materials if pro ven effective in improving productivity. Metbods: The effects on crop productivity of application of immature plant materials of Tithonia diversifolia, Lantano camara, Cassia hirsuta and Aspilia kostchyi was evaluated on-station over 6 seasons. The organic materials, applied at 4 t ha·1 dry weight, were compared to inorganic fertilízers. Surface application was compared to incorporation of the materials. Farmer-managed trials were conducted on ten farms over four seasons with two reps per farm. The test crop was maize. Outputs: Tithonia and aspsilia had the highest and lowest N content, respectively (Table 1). Lantana had the highest lignin (16%) and polyphenol (3.4%) content. The organic materials, except for Aspilia, supplied more N and K than the fertilizer but more P was supplied in the fertilizer treatment. Table l. Nutrients (kg) supplied in 4 t ot dry matter by immature plant material of four species N p K Ca Mg C. hirsuta 119.2 7.2 182.4 51.2 16.0 L. camara 107.6 6.4 107.2 34.8 22.4 A. kotschyi 53.2 4.4 160.8 67 .2 10.8 T. diversifo/ia 140.0 11.2 190.0 52.0 20.0 Fertilizer, full 100.0 21.3 104.3 0.0 0.0 Maize yields were highest with the full fertilizer rate but bean yields with cassia and tithonia were similar to the full rate offertilizer (Table 2). Response to the organic materials appeared to be regulated by plant N and polyphenol content, with less response to aspilia and lantana. The effects ofthe organic materials Íncreased with time. Table 2. Maize and bean grain yield (kg!ha) as affected by application of parts ofvarious plant species and inorganic fertilizers Cassia Aspi1ia Lantana Tithonia Species Ferti1izer (recommended rate) Fertilizer (l/2 recommended) Control Mean separarion by DMRT at p < 5% Maize yield mean of 4 seasons (kg ha·) 3350bc 2560d 30l0c 34\0b 3890" 3490b 2030° Method of application did no/ affect crop yield Bean yield mean of 2 seasons (kg ha· ) 980. 710c 740bc 860'b 88o• 790'bc 500d Crop response to application of lantana was less than with fertilizer (Table 3). The effectiveness of lantana increased with continued application indicating residual effects ofthe organic material. Applying half-rates of fertilizer and lantana in combination gave inconsistent results with less yield tban the full fertilizer rate in two seasons and similar yield in the other seasons. Yield response to lantana application may have been constrained by low P availability as the median P level (Oisen method) is 3.5 ppm in this community for fields having annual crops. 95 Component Technologies Table 3. Maize graio yield (kg/ha) as affected by applicatioo of Lantana residues aod ioorgaoic fertilizers (results from 10 oo-farm triaJs in Igaoga District) Treatment Lantana Lantana + Fertilizer (50% rates) Fertilizer Control LSD s% 1997a 2430 2820 2970 1950 710 Maize yield 1997b 1998a 1998b (kglha·) 1190 3820 1870 1990 4630 2320 2410 4700 3130 1010 2760 1190 560 860 780 Impact: The impact is not predictable. Response of fanners has not been encouraging. Expectations that fanners would begin their own experimentation with organic materials were not realized. Higher quality materials such as tithonia, where abundant, might be more attractive to fanners. Collaborators: Kayuki Kaizzi, Charles Wortrnann. Linked to IP2 and ECABREN which funded this work. Technical input ofCheryl Palm ofTSBF is acknowledged. 96 Component Technologies Activity 2.4 Deve/oping integrated forage-based components for upland farming systems 2.4.1 Selection of improved forages in Southeast Asia Highlights • A range of forages areavailable for different climatic and soil conditions and farming systems in Southeast Asia Purpose: To identify forages suitable for different upland farming systems in Southeast Asia Rationale: Livestock are an integral component of most farming systems in Southeast Asia. They are a form of diversification for accumulating capital and providing food and in their contribution to other components by providing draft power and improving nutrient cycling. With intensification, traditional feed resources are becoming scarcer or more difficult to access. Improved forages offer a means of complementing other food resources. However, they need to be well adapted to climate and soil conditions, easily manageable by farmers and able to be integrated with other forms of land use. Farmers need to learn that improved forages require management, as do other crops. Hence it was considered important that farmers be involved in the whole process of selection, evaluation and adapting forages into their particular farming system. Once farmers perceive that a particular species can make a contribution to their Iivestock feed requirement, they experiment with different ways to improve management ofthe sown species and seek other uses. Methods: The Forages for Smallholder Project (FSP) has used both farmer participatory approaches and conventional ' forage evaluation' techniques to develop sustainable forage technology options for resource- poor upland farms in Southeast Asia at 18 sites in Southeast Asia. These broadly cover the main range of envirorunents and upland farming systems of the region. lnitially large numbers of poten ti al species from CIA T and CSIRO were screened in regional nurseries, either on research stations or on-farm but under the close supervision of researchers to identify species with climatic and soil adaptation. Farmer groups were in volved in the planting and management of demonstration nurseries of species that had shown potential in their own villages. Individual farmers then selected species for evaluation on their own farms. Researchers initially provided seed and advice on establishment and worked alongside farmers in monitoring the performance of species. Feedback between farmers results and researchers enhanced learning and progress in selection of species and development of an appropriate technology. An important aspect of the selection process was to ensure that seed or planting material could be readily produced at the local leve\, as there is not a formal seed industry for forages. Outputs: Forage species suited for different uses, climates and soils are presented in Tables 1 and 2. Detailed information on the best varieties of each species are available in a booklet entitled 'Developing forage technologies with smallholder farmers- how to select the best varieties to offer farmers in Southeast Asia' . This information has been collated from farmer and researcher evaluations. These forages are now being evaluated in a genotype by environment experiment to accurately map the environmental adaptation ofthese forages. Final results are expected in 2000. Brachiaria decumbens cv.Basilisk has been identified as a promising grass for areas with a pronounced dry season in Southeast Asia. Unfortunately, this variety produces no or very little seed at low latitudes and an evaluation of a large range of Brachiaria germplasm is being conducted to identify varieties with acceptable seed yields and good dry season performance in northeast Thailand. Severa! promising accessions have been identified: CIAT accession numbers 16835, 6387, 1873, 1737, 16327, 16315, 16871, 16212, 26159, 26318, 16829, 16827 and 6780. Detailed studies ofthe seed production characteristics and 97 Component Technologies dry season performance of these accession are being conducted to identify a small range of accessions for farmer evaluation in a wide range of envimments. Table 1: Forage species for different climates and soils CLIMA TE SOIL FERTILTIY AND ACIDITY ... ~ o o "O o o -5 ,_..._ E G) "ü -e -~ e u,_..._ - "' -5 .g o -"' ~::: e "' e - .E~-~ o -~ o "' "' -¡; u o "' ~o "O "' "' c."' Q > ... >.E o -o ·- 11) :; < "" "' '-' "' - G) Grasses Andropogon gayanus • •• • • • Brachiaria brizantha • •• •• • •• • Brachiaria decumbens • •• •• • •• • Brachiaria humidicola •• • • • • • • Brachiaria ruziziensis •• • •• • Panicum maximum •• • • •• • Paspalum atratum •• • • •• • Pennisetum purpureum and hybrids •• • •• • Setaria sphacelata •• • •• •• • Legumes Arachis pintoi •• • •• • • Calliandra calothyrsus • •• • •• Centrosema macrocarpum •• • • • • • Centrosema pubescens •• • •• • Desmanthus virgatus •• •• • • • Desmodium cinerea • • • • G/iricidia sepium •• •• • •• Leucaena leucocephala •• •• • •• • Stylosanthes guianensis •• •• • • •• •• e e = highly suitable e = possible no stars = not suitable Stylosanthes guianensis CIA T 184 is one of the most promising, broadly adapted legume in the region. Although this variety is highly susceptible to the fungal disease anthracnose in its native range, it has shown remarkable resistance to this disease in Southeast Asia where there are different strains of anthracnose to those in South America. However, there is a danger that this resistance may not be durable and experiments are being conducted in Hainan, P.R. China and Los Banos, Philippines which aim to identify accessions with a higher leve! of resistan ce than CIA T 184. Of 40 Stylosanthes spp. and hybrids with known resistance from CIAT, EMBRAPA and CSIRO, severa! are showing promise for use in Southeast Asia. These include the hybrids GC1579 and GC1463. By the end of 1999, the project hopes to identify a small range of resistant accessions for wider testing in the region. 98 Component Technologies Table 2: Suitability of forages for different uses Forage Tecbnology Options "' Lo ~ e ~ o o e - loo u "' ·¡¡; .g "' 111 o .~ -; 111 Lo -Lo Cll Lo = - 111 - "' e Lo Lo e '-' "' = 111 ~ 111 -; - :t: Cll "' ~ e o 111 Q. .52 .S = "' 6 = loo - o "' "' 111 ~"' ... e o u ~ "' "' > C.CII u "' ~ ~ o = u Cll Lo Q. = :t: "' o o u "' "' ~ Cll ~ o u loo Lo u e u "' Lo > u u "0- 111 ~ 111 ~ u o ... Lo e o 6"0 6 N e ~ Lo u u = Lo = ~ = - Cll ·;;: "' c. > > o- ~= ~ = Lo u 6 o o a.. e 111 u e, ;,:} :z: - u u e, 3 ~-== ...;¡ Grasses Andropogon gayanus •• • • Brachiaria brizantha •• • • Brachiaria decumbens • •• • Brachiaria humidicola •• •• Brachiaria ruziziensis • •• • Panicum maximum •• • • Paspa/um atratum •• •• •• Pennisetum p urpureum and hybrids •• • Setaria sphace/ata •• • • • Legumes Arachis p intoi • •• •• Ca/liandra ca/othyrsus •• • • • Centrosema macrocarpum • •• •• • • Centrosema pubescens • •• •• • • Desmanthus virgatus •• • • • Desmodium cinerea •• •• G/iricidia sepium •• •• • • Leucaena /eucocephala •• • • • • • •• Stylosanthes guianensis •• • • •• •• • • • •• e e = highly suitable e = possible no stars = not suitable Impact: The impact ofthe selection and evaluation process is demonstrated by farmers identifying a new forage option (Table 1, Output 1.5 .2). Selection and evaluation by farmers and identification of new forage options is a continuous and on-go ing process. Successful selection and evaluation resu lts in a forage option for a particular need. Or a new forage option is the outcome of a successful selection and evaluation process. Thus we have only included those species that have been shown to have a role in a farming system in the table of climatic and soil adaptation (Table 2). 99 Component Technologies We consider that the species identified (Table 1) will cover most needs offanners for improved forages in the different upland fanning systems in Southeast Asia. However, as more farmers adopt these forage options, and they are used over an extended period, certain problems and also new opportunities will arise. Such problems could be seed supply, nutrient depletion under 'cut and carry' systems and disease. Research has been initiated on selection of Brachiaria species which will produce high seed yields under the climatic conditions of Southeast Asia. lt is proposed to initiate research with ILRI into nutrient dynamics in livestock systems. Research has also been initiated through an ACIAR project to identify altemative Stylosanthes guianensis species with resistance to anthracnose. New needs may be a demand for green cover legumes for use in cropping systems and altemative shrub legumes for dry season production. Contributors: Wemer Stür, Francisco Gabunada and Louie Orencia (CIA T, FSP Philippines); Peter Home and Phonepaseuth Phengsavanh (CSIRO, FSP Laos); Viengsavanh Phimphachanhvongsod (NAFRI, Laos); Maimunah Tuhulele (DGLS, Indonesia); Ibrahim (Livestock Service, East Kalimantan, Indonesia); Tatang lbrahim (BPTP, North Sumatra, Indonesia); Ed Magboo (PCARRD, Philippines); Willie Nacalaban (LGU, Malitbog, Philippines); Perla Asís, LGU, Cagayan de Oro, Philippines); Le Hoa Binh (NIAH, Vietnam), Le Van An (UAF, Hue, Vietnam), Bui Xuan An (UAF, Ho Chi Minh City, Vietnam); Liu Guodao (CATAS, Hainan, China); Chaisang Phaikaew (DLD, Bangkok, Thailand); Ganda Nakamanee (DLD, Pakchong, Thailand); Bryan Hacker (CSIRO Tropical Agriculture, Brisbane, Australia); Peter Kerridge (CIAT, Cali, Colombia). lOO Component Teclmologies Activity 2.5 Deve/oping improved soil management practices in cassava-based systems 2.5.1. New Cassava Options for Asia 2.5.1.1. The Nutrient Requirements of Cassava Highlights • K is highly important in maintaining high cassava yields and high root starch content. • Responses to K occur when the exchangeable K content of the soil drops below 0.15-0.1 7 me/ 1 OOg • Annual applications of 80-100 kg K20/ha are needed to offset the remo val of K in the root harvest Purpose: To maintain or improve soil productivity and to optimize the efficiency of fertilizer use in cassava-based cropping systems. Rationale: Continued production of cassava on the same land is feasible in Asia beca use of the near absence of diseases and pests. However, if cassava roots (and sometimes stems and leaves) are harvested and removed from the fie\d, this may \ead to nutrient dep\etion and soil degradation. Thus, it is important to determine the nutrient requirements ofthe crop, to establish diagnostic criteria for interpreting soil and plant tissue analyses, and to develop efficient fertilizer practices for maintaining high yields and high starch contents of the roots. Methods: Soil fertility maintenance through the application ofNPK fertilizers. Numerous long-term NPK trials with a uniform design ha ve been conducted with cassava in Asia sin ce 1987. In 1998/99 only four trials were continued in three countries. The trials use an incomplete factorial design with four levels each ofN, P and K in various combinations. Outputs: Figure 1 is an example ofthe response to annual applications ofN, P and K in two cassava varieties grown for the ninth consecutive year in a rather fertile red Oxisol (Eustrotox) at Hung Loe Agric. Research Center in south Vietnam. There were no significant responses to fertilizer application during the first four years. In the ninth year, however, there were significant responses to N and K and to the combined application ofNPK, which more than doubled yields from 15 to 33 tlha. N application tended to decrease, while K application increased the starch content of roots. An annual application of 80 kg N, 40 P20s and 80 K20/ha maintained high yields and high starch contents, even after nine years of continuous cropping. Figure 2 shows the trend over the years in absolute and relative yields (average oftwo varieties) as well as the soil K and P contents. Absolute yields tended to increase over time, partially due to a change from low-yielding local varieties during the first three years to high-yielding new clones. Trends in the relative yield indicate that the response to K became increasingly more pronounced over time; the response to N also increased, but more slow\y, whi\e that to P did not change much over time, as cassava removes relatively líttle P in the root harvest. These trends are also reflected in the soil K and P contents. The response to K became significant in the 5th and highly significant in the 7th year of cropping when the exchangeable K content of the soil had dropped below 0.17 and 0.15 me/ 1 OOg, respectively. This corresponds with previous results obtained in other locations. The available P content decreased only slightly when no P was applied, but increased significantly with annual application of 40 kg P20 5/ha. In both cases the soi\ P content remained well above the criticallevel of 5 ppm. 101 Component Technologies 40 -t~~-4 ,. .. ---+---· /--....... ,t---t---4 - ..... ~ .e: 1 ...... t' ,. 1 1 ~ 1 1 1 - 30 , ... r __ ---- ----------- 1 -- 1 -- "C t/ // - ,y ! ~ ·;:. ,f .... = J) 1 = t -~ P=40 N=ro ¡;¡.. N=ro ~ fl) 10 fl) ~ K=ro K=ro P=40 u o o 40ro 100 o J) 40 o 40 ro 100 0-0-0 ID4()..8) l@ID.l(í) - =S\1936-26 =KI.\fi ~ ............ , .... t---t---· ,. .. --.. ---· ...--.. ---· 1 - 28 1 /---- _ .. ~ ~:r/--- .t ' / - ............... / --- ~ ~--·'! / - / / .... f ........... t/ = ~ -= 24 u 1 .e: P=40 N=ro N=ro u ... ~ - K=ro P=40 00 K=ro J) o 40 ro 100 o J) 40 o 40 ro Figure l . Effeet of annual applieations ofvarious levels ofN,P and K fertilizer on the fresh root yield and stareh eontent of two eassava eultivars grown at Hung Loe Agríe Researeh Center in Dong Nai, Vietnam in 1998/99 (9th year). 102 Component Technologies ...... JS -+-NOPOKO • - -NOPlKl ~ JO .... NlPOKl '0:1 - ·-NlPlKO -¡¡ lS ">. "'*""NlPlKl i lO .. 1S • ~ = 10 ; u S o o 2 l 4 S 6 7 8 9 lOO ...... ~ a o '0:1 -¡¡ ">. 60 .. -~ 40 • -¡¡ IX 10 o o 3 • 5 ' 9 0.25 g e:> 0.2 ~ 0.15 ! Critica! K-level ~ 0. 1 ·¡; o. os m o o 2 J 4 5 6 7 a 9 JO ...... e 20 Q, ,e c:l. ·¡; 10 m Critica! P-level -------------------------------------------· o o l J 4 S 6 7 a 9 Crop cycle Figure 2. Effect of annual applications of N, P an K on cassava root yield, relative yield (yield without the nutrient over the highest yield with the nutrient) and the exchangeable K and available P (Bray 2) content of the soil during nine years of continuous cropping at Hung Loe Agricultural Research Center, Dong Nai, Vietnam. 103 Component Technologies Impact: The long-tenn fertility trials have clearly indicated the importance ofK in maintaining high yields of cassava, and, at least in Asia, the low requirements for P. In soils that are very low in P, such as in Brazil and parts of Colombia, P applications are probably necessary for only a few years. Based on these and other data, the Department of Agriculture in Thailand has changed its fertilizer recommendation for cassava, from a 15-15-15 compound fertilizer to a cheaper one containing 15, 7 and 18% N, P20s and K20 , respectively. Contributors: Mr. Nguyen Huu Hy of lAS in Vietnam, Dr. Nguyen The Dang of Univ. of Agric. and Forestry ofThai Nguyen, Vietnam; Mr. Zang Weite and Mr. Huang Jie of CATAS in Hainan, China; and Mr. Wargiono ofCRIFC in Bogar, Indonesia; R. Howeler, Cl.AT. 2.5.1.2. Soil Improvement and Erosion Control by the Use of Green Manure, lntercropping and Alley Cropping Highlights • In areas of acid soils and irregular rainfall, intercropping cassava with peanut generally gives highest net incorne and helps to control erosion • In high--Ca soils, alley cropping with Leucaena leucocephala or G/iricidia sepium are effective in improving productivity and in reducing erosion. • On slopes, contour hedgerows of vetiver grass, elephant grass or Taiwan grass are effective in reducing erosion, but the latter two rnay compete strongly with nearby cassava plants. Purpose: To evaluate green manure, intercrops and hedgerow systems for their efficiency in increasing yields or income, and for their effectiveness in controlling erosion. Rationale: In areas where chemical ferti lizers are not available or are out of reach for poor cassava farmers, soil fertility can be maintained or improved with the use ofbiological systems, such as legurninous green manure, intercrops or alley crops; these may also help to reduce erosion by quickly establishing a soil cover between cassava or by providing mulch. In addition, various grass species, when grown as contour hedgerows can be very effective in reducing soil erosion when cassava is grown on slopes, but rnay reduce the yield of nearby cassava, mainly through underground competition for water and nutrients. Methods: 1) Several experirnents were established on gentle slopes in Vietnam and Indonesia to evaluate the effectiveness of various green manure, intercrops and alley crops in increasing cassava yields and total net income, as well as in reducing erosion. 2) On 5% slope in Jatikerto, Malang, Indonesia, five grasses and one leguminous species were evaluated for their effect on cassava productivity as well as on soil erosion. Outputs: The results of an intercropping trial conducted on 16% slope at Thai Nguyen University in Vietnam are shown in Table l. Most intercrops slightly reduced cassava yields but increased gross and net income. Peanut and mungbean were most effective in increasing income, while peanut was also most effective in decreasing erosion. Follow-up experiments (data not shown) indicate that two rows ofpeanut between cassava rows (spaced at 1.0 m) produced highest net income, but 3 or 4 rows of peanut between wider-spaced cassava was slightly more effective in reducing erosion. Under the conditions ofNorth Vietnam, with low temperatures in the spring, peanut should be planted 1-2 weeks befare cassava. 104 Component Technologies Table l. Effect of intercropping cassava witb various grain legumes on crop yield, gross and net income and soilloss dueto erosion on 10% slope. Tbai Nguyen, Vietnam. Intercropping Yield (tlha) Gross Costs Net Dry soil treatments income1> fert.+seed 1> in come loss Cassava intercrop -E (mil. dlhaj ~ (tlha) l . Cassava monoculture 18.67 7.47 6.22 1.25 31.2 2. C+peanut 16.50 1.08 12.00 8.77 3.23 24.0 3. e +soybean 18.42 0.15 8.27 7.98 0.29 28.5 4. C+mungbean 20.83 0.27 10.49 7.84 2.65 28.6 5. C+black bean 17.92 0.35 9.62 7.94 1.68 28.6 6. C+cuoc bean 17.67 0.17 7.92 7.87 0.05 28.1 ()Prices ( dong): cassava: 400/kg fresh roots mung bean 8000/kg dry grain peanut 5000/kg dry pods black bean 7000/kg dry grain soybean 6000/kg dry grain cuoc bean 5000/kg dry grain The long-term effect of intercropping, green manuring and alley cropping on soil fertility and cassava yields after seven years of continuous cropping in Hung Loe Center in Vietnam are shown in Table 2. All the soil improvement practices increased soi1 organic matter (OM), P and K, but the leguminous hedgerow species were most effective in this respect. Alley cropping with Leucaena leucocephala and Gliricidia sepium was also most effective in increasing cassava yields or starch contents, resulting in an increase in starch yie1d of 11 .4 and 8.1 %, respectively. Table 2. Effect of leguminous intercrops, green manure and bedgerow species on cassava yield and starch content (cv KM60) and soil fertility parameters following 7 years of continuous cropping Dong Nai, Vietnam, 1998/99. Soi1 characteristics1¡ Cassava Root root starch Starch OM p K yie1d content yie1d Treatments (%) (ppm) (me/IOOg) (t/ha) (%) (t/ha) Cassava monucu1ture 2.5 9.5 0.24 25.00 27.8 6.95 Cassava intercropped with peanut 2.9 10.2 0.27 24.68 28.3 6.98 Cassava intercropped with cowpea 3.0 11.3 0.32 24.00 27.6 6.62 Cassava intercropped with Canava/ia ensiformis ZJ 2.7 9.7 0.27 26.45 26.9 7.11 Cassava al1ey cropped with Leucaena leucocephala 3.0 19.2 0.38 25.89 29.9 7.74 Cassava alley cropped with Gliricidia sepium 2.9 11.3 0.37 27.00 27.8 7.51 '> Soi1 analysis before 6th cropping cycle 2 > Canavalia intercropped with cassava, pulled out and mulched after two months The effects of hedgerows of five grass and one Jeguminous hedgerow species on cassava and intercropped maize yields, on gross income and on runoff and eros ion, when cassava was grown on 5% slope in Jatikerto, Malang, Indonesia are shown in Table 3. Hedgerows ofvetiver grass produced the highest cassava and maize yields, as well as the highest gross income, but elephant grass, Taiwan grass and king grass were more productive and of higher quality as an animal feed. These more productive species, however, also competed more strongly with neighboring cassava plants. All hedgerows were quite 105 Component Technologies effective in reducing runoff and eros ion, but the grass species were generally more effective than the leguminous hedgerows, while Taiwan grass was slightly more effective than the other grass species. Table 3. The effect of con tour hedgerows on yields of cassava and intercropped maize, hedgerow yield, gross income, runoff and soilloss on a 5% slope. Jatikerto, Malang, Indonesia 1997/98. Gross Soil Runoff Treatments Yield (tlha) income•> loss (%of Cassava maize Hedgerows {'OOOR~a) {tlba2 rainfalli> l. No hedgerows 18.853) 3.123) 4.296 68.92 30.5 2. vetiver grass hedgerows 31.15 5.52 13.95 7.227 19.33 20.4 3. e1ephant grass hedgerows 28.39 4.89 34.54 6.538 18.80 18.1 4. king grass hedgerows 26.96 4.71 29.85 6.232 18.96 18.4 5. Taiwan grass hedgerows 28.88 5.17 33.40 6.719 18.08 17.9 6. Benggala grass hedgerows 24.94 4.42 17.66 5.787 20.90 20.0 7. Gliricidia (from stakes) hedgerows 25.19 4.55 8.57 5.875 29.90 23.5 8. Gliricidia ~from seed~ hed~erows 26.69 4.70 6.69 6.182 28.66 22.9 t) Prices (rupiah): cassava 170/kg fresh roots, maize 350/kg dry cobs 2> Rainfall was 1295 mm from Dec'97 to Aug'98 3> Low yields because plots had previously received no fertilizers from 1988 to 1996 2.5.1.3. Effective Ways to Reduce Erosion in Cassava-based Cropping Systems Highlights • lntercropping and con tour hedgerows of vetiver grass or Tephrosia candida are very effective in increasing farmer's income and reducing erosion. Purpose: To develop component technologies that are effective in reducing erosion, maintain soil fertility and increase farm income. Rationale: When cassava is grown on slopes, soil erosion may be serious. Erosion can be significantly reduced by various agronomic practices, but nearly every practice has certain limitations. Research continues to look for options that are more effective in reducing erosion, less costly in establishment or maintenance, and that provide sorne direct income to the farmer. Outputs: The results of an intercropping x hedgerow tria! conducted on 5% slope in Tamanbogo, Lampung, Indonesia are shown in Table 4 . Intercropping with maize and either rice or peanut, followed by soybean or peanut, significantly reduced cassava yields but increased net income and reduced erosion, as compared to growing cassava in monoculture. Among intercropping systems the rice-soybean system was slightly better than the peanut-peanut system in increasing net income, while there were no significant differences in terms of erosion control. Among hedgerows ofvetiver grass, elephant grass and sorghum there were no or only minor differences in terms of net income and erosion control. The results of an eros ion control tria! conducted on 10% slope in Thai Nguyen Univ. in Vietnam are shown in Table 5. Planting cassava in monoculture without ridges or hedgerows resulted in the highest soilloss by erosion and the lowest yield and gross income. Intercropping with peanut markedly increased cassava yíelds, increased gross income, supplied crop residues for soil improvement and decreased erosion by 45%. But eros ion could be further decreased by combining intercropping with the planting of contour hedgerows of either Tephrosia candida or vetiver grass; this reduced erosion to about 20% ofthat obtained with monocropping. 106 eomponent Technologies Table 4. Effect of cropping systems and live barriers on the yields of cassava, maize, rice, peanut and soybean, the gross and net income, as well as the total soilloss due to erosion during a 10-month cropping cycle on 5% slope in Tamanbogo, Lampung, Indonesia, in 1997/98. Yield (tlha) Gross Fertilizer Net Dry soil cassava maize rice peanut soybean in come•> costs2> income3> loss ('000 Rp/ha) (tlha) eassava monoculture 26.9 a 4,307 e 268 4,039 e 23.8 a e mono+vetiver grass hlrows 27.7 a 4,425 e 268 4,157 e 18.9 b e mono+elephant grass h!rows 24.2b 3,868 f 268 3,600 d 17.6 be e mono+sorghum h/rows 27.4 a 4,390 e 268 4,122 e 16.9 e e +M+R-S + vetiver h/rows 13.0 e 0.14 2.51 0.79 6,279 be 664 5,615 b 13.5 d e + M+R-S + elephant gr.hlrows 16.8 d 0.15 2.02 0.77 6,547 ab 664 5,883 ab 12.0 d e + M+R-S + sorghum h/rows 14.7 e 0.35 3 .15 0.77 6,727 a 664 6,063 a 12.8 d e + M+P-P + vetiver hlrows 14.9 e 0.40 1.794) 5,023 d 612 4,416 e 12.5 d e +M+P-P + sorghurn hlrows 19.8 e 0.38 2.024) 6,119 e 612 5,507 b 12.7 d F-test •• •• •• • • l)j)fices (rupiah):cassava 160/kg fresh roots; peanut 1,400/kg dry pods; rice 550/kg paddy maize 350/kg dry grain; soybean 3,500/kg dry grain 2>Prices: urea Rp 350/kg; SP-36 450/kg; Ke l 500/kg 3~et income = Gross income-Fertilizer costs; •>surn oftwo harvests Table 5. Effect of cropping systems and management practices on the yield of cassava, intercropped peanut and hedgerow species, soilloss on 10% slope. Thai Nguyen, Vietnam. 1998. Yield (tlha) Gross Dry soil Qeanut income•> loss erop/soil management treatments cassava dry pods residue (mil. dlha) (tlha) l . e monoculture, no ridges, no hedgerows 17.96 8.98 24.11 2. e +P, no ridges, no hedgerows 21.72 0 .39 5.58 12.81 13.26 3. e +P, no ridges, Tephrosia hedgerows 22.38 0 .27 3.81 12.54 5.60 4. e +P, no ridges, vetiver hedgerows 2 1.50 0.30 3.16 12.25 5.40 5. e +P, contour rid es, Te hrosia+vetiver hed erows 21.04 0 .38 3.11 12.42 5.18 rices (dong): cassava 500/kg fresh roots; peanut 5000/kg dry pods Contributors: Mr. Wargiono ofCRIFC, Bogor and Dr. Wadi Hani Utomo ofBrawijaya University, Malang, Indonesia; Mr. Nguyen Huu Hy ofiAS, Ho Chi Minh city and Dr. Nguyen The Dang ofThai Nguyen Univ. in Thai Nguyen, Vietnam; Dr. Somjate Jantawat ofKasetsart Univ. and Mr. Anuchit Tongglum of DOA, Thailand; Mr. Huang Jie of CATAS, Hainan, China. 2.5.1.4. Varietal Testing Highlight • A sweet variety, KM 98-1, selected from hybrid seed in South Vietnam, had high yield and high starch content when evaluated in Central Vietnam Purpose: To develop high yielding and edible cassava varieties for Central Vietnam, where cassava is still an important food security crop. 107 Component Technologies Rationale: In the mountainous regions of Central Vietnam farmers are extremely poor and suffer from periodic food shortages due to the lack of suitable lowlands for rice production; cassava is their major food staple. Higher yielding and edible varieties, developed at Hung Loe Center in the south and at Thai Nguyen Univ. in the north, could improve food availability, while high-yielding bitter varieties may be useful for on-farm pig feeding or for starch extraction. Methods: 15 cassava varieties were collected in north and south Vietnam and planted for preliminary observation and multiplication ofplanting material in single rows or small plots at Hue Univ. in Central Vietnam. Outputs: Results ofthis observation trial are shown in Table 6. Although most varieties were bitter, developed for starch production and pig feeding, there were a few sweet and high yielding varieties, notably KM 98-l , developed from hybrid seed in south Vietnam. This variety is now being multiplied for future large-scale testing in comparison with the local varieties, and, if successful, for distribution to farmers. Table 6. Results of an observation trial conducted of 15 clones from north and South Vietnam. Hue 1998 Fresh root Starch yield content Clone (t/ha) (%) South Vietnam HL23 23.9 27.9 KM94 34.6 28.0 KM95 23.7 27.7 KM 98-1 40.1 29.4 KM 98-2 24.1 27.1 SM 1447-7 35.2 28.6 CMR2075-18 31.7 28.4 CMR 3608-1 27.8 28.2 OMR 35-08-2 11.8 25.7 OMR 35-15-8 19.4 26.6 OMR 36-06-1 O 20.5 27.9 OMR 36-75-13 16.9 24.5 North Vietnam Xanh Vinh Phu 19.8 27.0 SM 1717-12 29.6 28.0 SM 1829-24 25.8 24.9 t) A = acceptab1e; NA = not acceptable Observations bitter rather bitter sweet, high fiber bitter bitter bitter sweet less branching Contributors: Nguyen Thai Cae (Univeristy of Hue), R. Howeler (CIA T). 108 Farmer acceptance1> NA A NA A NA A A A NA NA NA NA A A NA Component Technologies 2.5.2. Conduct Demonstrations and FPR Trials on Farmers' Fields 2.5.2.1 FPR Demonstration Plots Purpose: To show farmers a wide range of options, from which they can select a few practices that are considered most effective in controlling soil erosion and most suitable for the local conditions. Rationale: Few practices recommended to control soil erosion have been adopted by farmers, because they were either not suitable for the local conditions or they produced no direct benefits to the fanners. By setting up demonstration plots and organizing field days for farmers to visit these plots and discuss the pros and cons of each treatment, farmers are encouraged to evaluate the various practices and select those they consider most useful for further testing on their own fields througb FPR (see below). Metbods: In 1989/99 demonstration plots wi~ 1 O to 16 treatme!_!ts were established in two si tes each in Thailand, Vietnam and China, and one in Indonesia. Plots were laid out side-by-side on a uniform slope, usually without replication. Erosion losses in each treatment were determined by collecting and weighing of soil sediments trapped in plastic covered channels located along the lower side of each plot. Fanners from selected FPR pilot sites were invited to visit the demonstration plots at field days organized during or at the end ofthe crop cycle. Farmers were asked to evaluate each treatment and to select those they considered most useful for their own conditions. Outputs: The results ofa typical demonstration conducted on 8-10% slope at Thai Nguyen Univ., Vietnam are shown in Table 7. Cassava yields were extremely low and eros ion was the most serious when cassava was grown in monoculture without fertilizers or hedgerows; this resulted in a negative net income, and none ofthe farmers considered this a useful practice. Fanners selected as most useful the treatments that included intercropping with peanut, app\ication of fertilizers, and with or without hedgerows of Tephrosia candida; or monocropping with fertilizers but planting at a closer spacing. The latter is a simple and cheap technology, which in this case was very effective in increasing cassava yields and net income, while decreasing erosion. Although vetiver grass hedgerows were the most effective in reducing erosion (also observed in demonstration plots in China and Indonesia) few farmers selected this as a useful practice, mainly because the grass itself is of little use to the farmers (unlike Tephrosia candida, which is considered a useful green manure, while stems are used as fue!), it is difficult to obtain planting material and the grass occupies sorne space (about 10%) ofthe production field in areas where land is very scarce. Interestingly, few farmers selected hedgerows of Tephrosia combined with pineapple, which is a recommended and rather popular practice in Central Vietnam. 109 eomponent Technologies Table 7. Effect ofsoil-crop management treatments on cassava yield, gross and net income and soilloss on 10% slope. FPR demonstration plots at Agro-forestry College ofTbai Nguyen Univ., Tbai Nguyen, Vietnam, 1998. Yield Gross income Production e· SoiVcrop management treatments1> cassava peanut cassava peanut Total l. e monoculture, no fertilizers, 1x0.8m, no hedgerows 2. e, contour ridges, with fert!>, 1x0.8m, no hedgerows 3. e+P, with fert., 1x0.8m, no hedgerows 4. e, with fert., 1x0.8m, vetiver+Tephrosia hedgerows 5. e+P, with fert., 1x0.8m, Tephrosia hedgerows 6. e+P, with fert., 1x0.8m, pineapple+Tephr. hedgerows 7. e+P, with fert., lx0.8m, natural grass hedgerows 8. e+P, with fert., 1x0.8m, vetiver hedgerows 9. e+P, with fert., 1x0.8m, vetiver+Tephr. hedgerows 10. e, with fert. , 0.6x0.8m, no hedgerows 5.75 16.25 15.42 16.17 17.75 16.50 14.83 15.83 16.83 24.58 (tlha) 0.73 0.71 0.65 0.60 0.63 0.77 l)e = cassava (cv. Vinh Phu), P = peanut 21>rices: cassava d 500/kg fresh roots pineapple peanut 5000/kg dry pods Tephrosia seed labor 15000/day peanut seed vetiver grass 20/plant urea (46%N) cassava stake 50/stake · 1 SSP ( 16%P20s) Kel (50%K20) 3 >percent of farmers (out of 35) considering the treatment as "good" 4 >Fertilizer = 60 kg N+ 40 P20s + 120 K20iha 110 2.87 8.13 7.71 3.65 8.09 8.88 3.55 8.25 3.25 7.42 3.00 7.92 3.15 8.42 3.85 12.29 d 100/plant 5000/kg 2.87 8.13 11.36 8.09 12.43 11.50 10.42 11.07 12.27 12.29 7000/kg dry pods 3000/kg 1000/kg 2000/kg labor fert. se• (mil. dlha) 2.62 o 3.02 1.12 3.03 1.12 2.74 - 1.12 4.99 1.12 4.99 1.12 3.03 1.12 4.99 1.12 4.99 1.12 2.79 1.12 0.63 0.63 0.91 0.99 0.97 1.5$ 0.91 1.31 1.37 1.04 Component Technologies 2.5.2.2 FPR Trials on Farmers' Fields Highlights • During 1989/99 132 farmers in eight pilot sites in Thailand, Vietnam, China and Indonesia conducted FPR trials on erosion control, varieties, fertilization and intercropping practices. • For control of soil erosion farmers prefer: • in Thailand planting of con tour hedgerows of vetiver grass; • in Hainan, China, hedgerows of vetiver or sugarcane; • in Vietnam, intercropping with peanut combined with hedgerows of Tephrosia candida; • in Indonesia, hedgerows of elephant grass, Leucaena leucocephala or Gliricidia sepium. Purpose: To develop and then facilitate the adoption of appropriate integrated crop/soil management practices for more sustainable cassava-based cropping system, by involving farm~rs directly in the testing of those practices they consider most use fui under their own conditions. Ratiooale: Continuous production of cassava on poor soils, especially on slopes, without appropriate soiVcrop management results in nutrient depletion and soil erosion Recommended fertilization and eros ion control practices are seldom adopted by cassava farmers beca use they don ' t ha ve the resources, are not aware of degradation or it's importance, or the recommended practices are unsuitable for the local conditions. To enhance the use of more sustainable production practices, and safeguard the productivity of the soil resource for future generations, it is considered important to involve farmers directly in the development of effective soil conserving practices by involving them in the evaluation of selected practices. Once farmers have selected and adapted suitable practices for their own conditions, they are more likely to adopt those practices and convince others of their long-term benefits. Methods: Researchers and extensionists from collaborating national institutions helped farmers in selected pilot sites to set up simple FPR trials on erosion control, varieties, fertilization and intercropping practices. Farmers in the pilot sites volunteered todo these trials and decided as a group about the type of trials to be conducted and the treatments to be tested. Researchers and extensionists supplied sorne of the planting materials and trained farmers in conducting simple trials. Table 8. Types and number ofFPR trials conducted with cassava in Asia in 1998/1999. Thailand Vietnam China Indonesia Type of trial Sahatsakhan Phanom Sarakham Pho Yen Thanh Ba Luong Son Baisha Malang Blitar Kalasin Chachoengsao Thai Nguyen Phu Tho Hoa Binh Hainan East Java East Java Erosion control 7 i > 5 6 3 6 lO Varieties 8 32) 15 13 2 8 F ertilization lO 62) 4 4 3 5 lntercropping 7 Total 25 16 31 23 8 14 15 o 1) planted in early 1999; i) planted in late 1998 Eros ion control experiments, usually with 4-5 treatments without replication, were laid out along the contour on land with uniform slope. Erosion losses in each treatment were determined by collecting and weighing of soil sediments trapped in plastic-covered channels below each plot. At field days during or at the end of the crop cycle, farmers visited each other's trials and were encouraged to discuss and evaluate the usefulness of each treatment. At time of harvest, yield and soil losses were determined and the results of the trials discussed and evaluated with participants and other members of the community; after this, farmers selected the best practices for further testing or for adoption during the following year. Table 8 shows the type and number of FPR trials conducted in each pilot si te during 1989/99. 111 Component Technologies Outputs: Tables 9, 1 O and 11 show results of FPR erosion control trials conducted by farmers in pilot sites in Vietnam, Indonesia and Thailand. In Vietnam, highest net income was obtained with the fanner's traditional practice of cassava monoculture and application of 1 O t/ha of pig manure (Table 9). However, this practice also resulted in the highest level of eros ion. Intercropping cassava with peanuts, applying fertilizers in additíon to manure, and planting hedgerows of either pineapple or vetiver grass slightly decreased net income but decreased erosion by about 90%; these were preferred by farrners. Table 9. Effect of crop management treatments on tbe yield of cassava and intercropped peanut, gross and net income and soilloss in trial conducted by six farmers. Kieu Tung village, Tbanb Ba district, Pbu Tbo province, Vietnam 1998. Dry soil Yield Gross income Prodn. Net Farmers' loss cost lncome ranking Treatments1> cassava peanuf> cassava Peanut total (tlha) (t/ha) (mil. d/ha) l . e monoculture, (TP) 18.4 25.8 10.30 10.30 4.00 6.28 2. e+P 13.3 18.2 0.47 7.26 2.58 9.84 5.90 3.94 3. e+P+fert. 14.0 20.3 0.51 8.13 2.80 10.93 7.42 3.51 4. e+P+fert.+ Tephrosia hlrow 5.3 21.6 0.51 8.64 2.80 11.44 7.67 3.77 5. e+P+fert.+ pineapple hlrow 0.8 23.3 0.73 9.33 4.01 13.34 7.67 5.67 6. C+P+fert.+ vetiver h/row 3.1 26.5 0.38 10.61 2.09 12.70 7.67 5.03 7. e mono+fert.+Tephrosia h/row 6.3 25.1 10.02 10.02 5.77 4.25 1) All plots received lO tlha ofpig manure; TP = fanners traditional practice = cassava mono + lO t/ha ofpig manure fert. = 60 kg N+40 P20 5+ 120 K20/ha; C= cassava, P = intercropped peanut l) Dry pods l) Prices (dong):cassava d 400/kg fresh roots urea (45%N) 3,000/kg -+ 60 N = 0.400 mil. dong peanut 5,500/kg dry pods SSP ( l6%P20 5) l ,000/kg -+ 40 P20 5 = 0.250 mil. dong labor 7,500/day K2SO. (50%K20)2,800/kg -+120 K20 = 0.672 mil. dong pig manure 200/kg (incl.trans.& applic.) Total fertilizers = 1,322 mil. dong 6 5 5 3 2 l 4 In Indonesia ten farrners compared four practices in FPR erosion control trials (Table 10). The net income Table 10. Average results of ten FPR soil erosion control trials conducted by farmers in Sumbersuko village, Dampit, Malang, East Java, Indonesia, 1997/98. Yield Gross Prod. Net Dry Treatments'> income Costs lncome3> soilloss cassava maize (tlha) ('000 Rp/ha) (tlha) l. C+M; farmer practice, in-line mounds followed by up/down ridging 18.9 1.20 6,063 1,200 4,863 18.10 2. C+M; recommended practices, contour ridging,vetiver bedgerows 22.2 1.35 7,060 1,900 5, 160 6.40 3. C+M; recom. practices; contour ridging, lemon grass hedgerows 21.1 1.50 6 ,897 1,900 4,997 9.45 4. C+M+P-Cp2>; recom. practices, contour ridging of cassava row 7.6 0 .85 6,962 2,370 4,592 14.65 l) C = cassava, M = maize, P = peanut, Cp - cowpea 2> Yields of peanut: 620 kglha; cowpea: 360 kglha J) Prices: cassava Rp 270/kg fresh roots; maize 800/kg dry grain on cob peanut 4,500/kg dry grain in pod; cowpea 4,000/kg dry grain 112 Component Technologies was highest with vetiver grass contour hedgerows; this also reduced erosion by about 65% compared with the farmer's traditional practice. Nevertheless, farmers showed little interest in planting vetiver grass hedgerows, because the grass does not provide other benefits to them. lnstead, they preferred planting lemon grass or elephant grass, which are almost equally effective in controlling erosion, while these can be used in cooking andas animal feed, respectively. In Thailand seven farmers tested five common treatments in addition to one treatment of their own individual choice (Table 11). Among the five common treatments, intercropping with sweet corn or planting at closer spacing produced the highest gross income and intermediate levels of erosion. Lowest levels of erosion were measured in two trials where farmers had selected vetiver grass hedgerows as the additional treatment. While vetiver grass produces few other direct benefits, many farmers in Thailand still prefer this grass to other options because once planted it requires relatively little maintenance, is very drought resistant and not very competitive. Table 11. Average results ofseven1> FPR erosion control trials conducted by farmers in Sahatsakhan district, Kalasin, Tbailand, 1998/99. Yield Dry Treatments cassava intercrop Gross income ('0008/ha) soil loss (t/ha) (t/ha) cassava intercrop total (t/ha) l. Farmer' s practice (up/down ridges) 29.74 30.93 30.93 12.48 l. Contour ridges 29.42 30.61 30.61 8.44 3. Lemon grass hedgerows 32.81 34.12 34.12 9.60 t . Closer plant spacing (0.8 x 0.8 m) 35.66 37.09 37.09 10.91 5. Intercropped with sweet com 25.76 7944 ears 26.79 12.17 38.96 13.36 S. Vetiver grass hedgerows 31.35 32.61 32.61 7.35 7. Intercropped with pumpkin 31.19 3400 purnpkins 32.44 18.36 32.44 14.13 8. lntercropped with peanut 15.70 1.20 16.37 18.00 34.37 29.64 l) Only four trials for treatment 7, two for treatment 6 , and one for treatment 8 Tables 12 and 13 show results ofFPR variety trials conducted in Vietnam and Thailand, respectively. In Pho Yen district ofThai Nguyen province 15 farmers tested five promising new clones in comparison with the local variety, Vinh Phu (Table 12). All new varieties had higher yields than Vinh Phu, and farmers were particularly impressed with SM17-17-12, which produced a 50% higher yield than the local variety. This variety has consistently outperformed Vinh Phu and is now rapidly being multiplied for further distribution. Table 12. Average results of 15 FPR variety trials conducted by farmers in Tien Phong and Dac Son villages ofPho Yen district, Tbai Nguyen province, Vietnam in 1998. Cassava Gross Fertilizer Net Farmers' yie1d income•> costs income preference Variety (t/ha) -E (mil. donglha) > (%) l. Xanh Vinh Phu 16.89 8.45 2.90 5.55 7 2. KM60 20.40 10.20 2.90 7.30 65 3. KM 95-3 18.45 9.22 2.90 6.32 o 4. CM 4955-7 24.62 12.3 1 2.90 9.41 82 5. KM94 21.91 10.96 2.90 8.06 50 6. SM 17- 17-1 2 25.44 12.72 2.90 9.82 lOO I)Prices: cassava d 500/kg fresh roots 113 Component Techno1ogies In Sahatsakhan district of Kalasin province in Thailand seven farmers compared four new clones with the traditional variety Rayong 1 (Table 13). All new clones produced higher yields and had higher starch contents than Rayong l. The best new clone, CMR.-81, produced a gross income o ver 30% higher than the local variety; this line is likely to be released this year under the name of Rayong 72. Tab1e 13. Average results of seven FPR variety trials conducted in Sahatsakhan district of Kalasin, Thailand in 1998/99. Cassava root Starch Starch Gross yield content yield income1> Varieties (tlha) (%) (tlha) ('0008/ha) Rayong 1 21.76 26.4 5.74 21.37 Rayong 5 24.61 29.4 7.23 26.33 Rayong 90 23.34 29.6 6.91 24.97 Kasetsart 50 25.72 29.5 7.59 27.52 CMR-81 27.04 28.2 7.62 28 .12 i) Cassava price: Baht 1.1 0/kg fresh roots at 30% starch shows average results of three FPR fertilizer trials conducted in Dong Rang village in North Vietnam (Table 14). Cassava yields and net income increased with application of chemical fertilizers, especially with a well-balanced fertilizer of 40 kg N, 40 P20 5 and 80 K20/ha. Potassium was the most limiting of the three macro-nutrients. Table 14. Average results of three FPR fertilizer trials conducted by farmers in Dong Rang village of Luong Son district, Hoa Bioh province, Vietnam in 1998. Fertilizer treatments •> Cassava yield (tlha) Gross income2> Fertilizer costs2> (mil. donglha) Net income l. Farmer's practice 10.95 4.38 O 4.38 2. Cassava + NP 12.40 4.96 0.518 4.44 3. Cassava + PK 13.40 5.36 0.640 4.72 4.Cassava+NK 15.35 6.14 0.678 5.46 5. Cassava + NPK 16.50 6.60 0.9 18 5.68 1lfarmer's practice = no fertilizer. N = 40 kg Nlha; P = 40 kg P20~a; K = 80 kg K20/ha 2>Prices: cassava d 400/kg fresh roots; urea ( 46%N) 3200/kg Fused Mg-phos (15% P20 5) 900/kg; KCI (50% K20) 2500/kg 2.5.2.3. Adoption and Dissemination of Improved Practices > After conducting FPR trials on their own fields for a few years farrners will often try out sorne selected practices on parts oftheir production fields, make certain adaptations if necessary, and then implement these improved practices on their whole fields as rapidly as possible. In the pilot site in Baisha county in China, farrners have been replacing the local variety SC205 with a range of new higher-yielding clones selected in their FPR variety trials. Among these, the most popular are SC8013, SC8634, SC8639 and OMR35-70-7. Farrners are also applying sorne chemical fertilizers in combination with chicken manure. Contour hedgerows of sugarcane to control eros ion were found to be less effective than vetiver grass. Farmers having land on steep slopes are now increasing 114 Cornponent Technologíes planting material of vetiver grass to be able to expand the areas planted with vetiver hedgerows, and prevent further eros ion of their soil. In the two pilot sites in Malang and Blitar districs of E. Java, Indonesia, most farmers still prefer their local varieties, Menyok (sweet) or Faroka (bitter) and Ijo, respectively, over any ofthe newly introduced clones. In Malang, however, sorne farmers are increasingly planting a bitter and early bulking variety called "Caspro", which produces very high yields on good soils; this variety was introduced many years ago to a neighboring village in an IDRC-sponsored project. In Blitar, more and more farmers are now planting the bitter variety Faroka, introduced from Malang, as it is high yielding and can be used for sale to starch faetones. While farmers in both districts recognize the effeetiveness of vetiver grass for erosion control, they still prefer planting hedgerows of elephant grass, Gliricidia or Leucaena, which are more useful to feed eattle or goats. The limited amount of planting material of vetiver grass presently available is being used to stabilize gulleys or protect natural drainage ways. In the two pilot si tes in Thailand farmers have already replaeed the local variety Rayong 1, as well as previously introdueed Rayong 3 and 60, with the higher yielding varieties Rayong 5 and 90, as well as Kasetsart 50. The soon-to-be released Rayong 72 (CMR-81) will further replace mainly Rayong 60, as it has a higher stareh eontent. Many farmers ·are now applying 150-300 kg!ha of compound fertiíizers, mostly 15-15-15, but increasingly also 15-7-18, as recommended by the Dept. of Agríe. and Agríe. Extension. In W ang Nam Yen district of Sra Kaew province, farmers who had previously planted FPR eros ion control trials are now planting con tour hedgerows of vetiver grass; in 1998 about 50 ha had already been planted. In Soeng Saang distriet ofNakom Ratchasima province, a village adjacent to the former FPR pilot si te observed the effectiveness of vetiver grass in FPR eros ion control trials; subsequently, they requested help from the Dept. of Agríe. Extension to plant 320 ha of eassava fields on sloping land with eontour hedgerows of vetiver; in 1999 about 50 ha were planted. In the three pilot sites in Vietnam, farmers are rapidly incrementing planting material of KM 60, KM 94, KM 95-3, SM 17-17-12 and SM 17-17-40 to replace their local variety Vinh Phu. They also now intererop cassava, mostly with peanut but also with black bean, and apply about 80 kg N, 40 P20s and 80 K20/ha in addition to 5-10 tons ofpig manure. In one ofthe sites, in Dong Rang village, farmers are planting contour hedgerows of Tephrosia candidato control erosion and improve soil fertility. Vetiver grass is planted only on a small scale (partially due to a lack of planting material), and mainly to stabilize gulleys orto protect waterways or paths. Contributors: Mr. Zhang Weite, Mr. Lin Xiong, Mr. Li Kaimian and Mr. Huang Jie of CATAS, China; Dr. Hadi Wani Utomo and Mr. Aldon Sinaga ofBrawijaya Univ.; and Dr. Suyamto, Mr. Abdulla Taufic and Mr. Sudjarwoto of RILET, Malang, Indonesia; Dr. Somjat Jantawat of Kasetsart Univ., Mr. Anuchit Tongglum, Mr. Danai Suparhan and Mr. Somphong Katong ofDOA, Mr. Kaival Klakhaeng, Mr. Somnuek Hemvijit and Mrs. Wilawan Vongkasem ofDOAE, Bangkok, Thailand; and Dr. Nguyen The Dang, Dr. Tran Ngoc Ngoan, Mr. Ly Sy Loi and Mrs. Dinh Ngoc Lan ofThai Nguyen Univ. and Dr. Thai Phien ofNISF in Hanoi, Vietnam; R. Howeler (CIA T). Activity 2. 6 Establish and maintain databas es of information and results Progress in establishing data bases of results of PE~ S The information from trials in Cauca with small producers on legume establishment, P fertilizer management with P and residual effects and use of legumes to improve fallow and erop performance, were analyzed and stored in the forages database. User friendly software its being developed to make it available to other members of CIA T and later on to outside users. The information on Carimagua trials has been organized and analyzed to be included in the database. l\5 Models/Frameworks Output 3. Models/frameworks developed to target research, integrate results, assess impact and extrapolate results Activity 3.1 Adapt and evaluate integrated simulation models for smallholders systems Highlights • Link between DSSAT and the soil-organic-matterlcrop-residue module ofCENTURY validated. It will beco me a new standard option in the next DSSA T version. • New Brachiaria decumbens model option made available to become a new standard crop option in the next DSSA T version. • DSSA T applied to maize systems in Honduras in order to determine the best management options. • Limitations identified for adding a new Arachis pintoi option to DSSAT. Purpose: Adapt DSSA T for smallholder systems and analyze such systems in Honduras, in order to better understand their strengths and weaknesses, and to identify opportunities for improvement. Rationale: Agronomic simulation models are an important too! for [i] understanding system dynamics, [ii] quantifying relationships that are not easily measured [iii] quantifying effects of different management strategies, [iv] predicting problems and results of management strategies and [ v] extrapolating results o ver larger geographical areas. They therefore help to plan research and integrate research results, and enhance collaboration. For smallholder systems in Central America, such a model may help in analyzing different options of land use and farm management. This will result in a better understanding ofhow crop production varíes with environmental conditions (weather, soil), and ofwhich management strategies (e.g. fertilization, planting date) are best under different conditions and most suitable to reduce weather risks. This will allow designing management strategies at the plot or whole-farm leve! that pay maximum careto the biophysical sustainability ofthe agricultura! system and its economíc viability. Modeling is therefore an important too! for research planning and interpretation of results. lt also helps to ensure that CIA T is not reactive but proactive when dealing with long-term sustainability issues. Methods: The Decision Support System for Agrotechnology Transfer (DSSA T) is a widely used agronomic model, which allows evaluating germplasm-by-environment interactions without the need to do expensive and time-consuming multi-site, multi-treatment experiments. Experimental data from one site can thus be extrapolated to other areas, having a different soil type or climate, or where farmers use other crop varieties. In low-input agricultura! systems, plant nutrients mainly come from soil-organic- matter (SOM) decomposition. DSSA T has a module for the simulation of SOM dynamics, but is limited in what is simulated, in particular with respect to the above- and below ground cycling of organic residues. A more complete SOM module in the CENTURY model was linked to DSSAT. Using DSSA T with the modified SOM module, low-input maize-bean systems at different altitudes in the watershed in Honduras, where the Hillsides Project has its main focus, have been analyzed. The simulations considered the optimum management system (e.g. planting date and limited N application), yield levels over a 20-year cycle and the longer-term effect on the system sustainability as reflected in the soil 's organic matter content. 116 Models/Frameworks Outputs: 3.1.1. Validation ofthe DSSAT-CENTURY link The soil-organic-matter (SOM) module ofthe CENTURY model was successfully incorporated into DSSAT. Comparing the original (PAPRAN-model-based) SOM module with the new CENTURY-based module (Figure 1) there was a satisfactory congruence between observed and simulated data with a set of experiments on legume residue decomposition from Brazil. By incorporating the CENTURY-based SOM module into DSSA T, the model should be more suitable for simulating low-input systems where most nutrients come from N2 fixation by legumes or plant residue decomposition. The new module has been accepted by the DSSA T coordinating group to become a standard element of the next model version. 30 20 10 -.. (A) Fallow 0-IS cm ~~ O r-··--.--.--·-.---.---.--.---.-~ ~ ] .$ z l i 30 o so 100 ISO 200 250 300 350 (B) . PAPRAN-base SOM module --- CENTURY-based SOM module Fallow IS-30 cm • Measured O SO 100 ISO 200 250 300 3SO 30 20 10 o 30 20 10 .. o 50 100 ISO : .. ·: .. . . . .. • 200 Fallow + residues 0-15 cm 2SO 300 3SO Fallow + residues IS-30 cm ••• o so 100 !50 200 2SO 300 3SO Time (days) Figure 1: Mineral N content ofthe 0-15 (A) and 15~30-cm (B) soillayers under fallow, without or with residues, as measured ( •) and as simulated with the DSSA T, using the P APRAN-based ( dotted line) and CENTURY-based (fullline) SOM modules. The residues applied were 5907 kg ha-1 of Mucuna spp. tops with a N concentration of2.37%. 3.1.2 Other modifications to DSSAT Modifications were made to the existing peanut model in DSSA T for the perennial forage peanut Arachis pintoi, a species of importance in humid tropical pasture systems and as a cover crop in plantation and fruit crops. Calibrating the new model option with five data sets from different sites in Colombia that were part ofthe lntemational Network for the Evaluation ofTropical Pastures (RIEPT), revealed severa! 117 Models/Frameworks differences between Arachis and the common peanut that have to be dealt with in the model' s source code. The peanut-model's author (Ken Boote- Univ. Florida), who assisted in this modification, has indicated an interest in following up on the problems encountered through student research. The Brachiaria decumbens grass growth option we developed for DSSA T last year has been made available to the DSSAT authors to be included as a standard option in a future version ofthe model. Experiments were carried out in Colombia and Honduras for estimating so-called 'genetic coefficients' of various legume species, data that are needed to simulate these crops' growth and development. These data have been made available to Univ. Georgia and CIMMYT, who are working on a DSSAT crop module for these species. These modules will be applied for analyzing smallholder systems in Central America. 3.1.3 DSSAT application to Honduran maize systems Preparing the input data sets on soils, clim~te and cultivars has been a major effort, because data were either not accessible, incomplete or did not exist. For a long-tenn research effort in Central America, it is important that CIA T give database management sufficient attention. Sin ce there are only two years of weather data available for the watershed where we are working, the new MarkSim weather generator program (Peter Jones, unpublished data, project PE-4) was used to produce daily estimates of maximum and mínimum temperature, precipitation and solar radiation. Daily rainfall data for 20 years was generated for three weather stations at different altitudes within the watershed (San Antonio - 420 m; El Guaco - 790 m; Las Lagunitas -1123 m). Soil data were taken from ear\ier analyses by CIAT-Honduras. Simulated maize planting was done when the soil moisture levels were suitable for planting, i.e. when the topsoil should have had adequate water to ensure gennination and establishment. The question was how to defme 'an adequate soil moisture leve)'? This will, amongst other things, depend on the risk a farmer is willing to take. Three levels were set: 40%, 50% and 60% ofmaximum available-water holding capacity. The planting window allowed was from May 20 to July 10 (i.e. between Julian days 140- 191 ; a 'Julian day' is the day number from 1 till365 - or 366 in leap years- counting from January 1). For San Antonio, simulated planting was mostly done in late May (daysl40-145; Figure 2), a very narrow range. Whether San Antonio farmers indeed plant during this narrow planting window is uncertain, because there often is a dry spell between day 150 and 170, which may kili the just-emerged seedlings. Low-risk farmers may postpone planting till after that date. At the other two sites, planting occurred somewhat later. At El Guaco the range of planting dates was much wider, because the rainfall is more uncertain until around da y 170. Time to maturity for maize is presented in Figure 3, which clearly shows that there was no relationship between planting date and time needed to mature. ln San Antonio, the crop always matured within 120 days and most ofthe time within 100 days, but in El Guaco and Las Lagunitas this may take up to 150 days, even ifplanting was done on the sarne day (e.g. day 140-142). Late maturing gave a much higher yield in these two sites. Planting after day 150 did not give a good yield in El Guaco and Las Lagunítas. At San Antonio, yields were very similar every year (Figure 4), with a mean and median around 1000 kglha and a maximum of 1500 kglha. In one out of four years the yield was only 500 kglha. This narrow yield range may partly reflect that the planting window was very narrow (Figure 2). The two other sites hada wider planting window and consequently a wider yield range. For El Guaco, average yield was 800- 1300 kg/ha, depending on the available-water content at planting; for Las Lagunitas it was 1300 kglha. Peak yields were around 6500 kglha (El Guaco) and 7000 kglha (Las Lagunitas), which occurred only once in 20 years time; this is indeed quite high for unfertilized conditions. The data for El Guaco show an interesting point: farmers who are early planters (at 40% maximum available-water content) and don' t wait until the soil is wetter, loose on average sorne 500 kglha compared with more patient farmers who postpone planting until the soil has 50% or 60% of it maximum available-water content. 118 175 170 ..... 165 Q) .e E 160 ::J e ~ 155 -e e 150 .~ :; -, 145 140 135 o 40 • r ,. :•' '!'' • ~.,.. • • • • • • • • 95% 90% 75% mean median ~~ 10% and 25% over1ap 50 60 40 50 60 40 50 60 Proportion of maximum available-water content (%) Figure 2: Optimum maize planting date, if ' optimum' ís defined as (a) 40%, (b) 50% or (e) 60% of the maximum available-water content of the soil. 10000 9000 8000 7000 • • • ";" 6000 ro • • • ..c. Ol ~ 5000 "C Q) 4000 >= 3000 2000 1000 o 40 50 60 40 50 60 40 50 60 Proportion of maximum available-water content at planting (%) Figure 4: Maize yield that results from the optimum planting conditions of Figure 2. 119 7000 6000 ... 3000 i400) 13000 > 2000 7000 6000 ... 3000 ! 400) ... .... 3000 > 2000 7000 6000 ... 3000 ~400) ¡ 3000 > 2000 1000 ·' , -- ........ - ---- -~- ,. 1 -~--L __ _ PlaadiiJ elote (JIIIIaa doy) • \ "••diiJ date {lh"-- -clay) • • •• 1- ... n -l • 1. " • • .. r" n Models/Frameworks Figure 3: Planting date, maturity time and yield of maize, as it results from the optimwn planting conditions of 50% maximum available-water content. 120 10000 9000 8000 7000 111 6000 .s:; l 5000 :2 ,g¡ >- 4000 3000 - • • • • De~un N H20 N+H20 Models/Frameworks • • • • ¡ • • • • • • De~ult N H20 N+H20 Oe~ult N H20 N+H20 No-stress treatment Figure 5 : Potential maize yield under conditions of no N stress (N), no water stress (H20), and neither N nor water stress (N + H20), as it results from the optimum planting conditions of 50% maximum available-water content. The yields from under unfertilized and non-irrigated conditions are indicated as reference points (Default). Running DSSA T with an option where nitrogen is automatically added every time when there is a need for N (zero N-stress conditions), yields are scarcely increased compared to unfertilized maize (Figure 5; NB: these are runs with 50% maximum available-water content at planting). Only the peak values increase with added N in San Antonio and Las Lagunitas. Frequent irrigation, in contrast, greatly increased yields at aJJ sites; yield variability, which initially was high at El Guaco and Las Lagunitas, was reduced considerably with irrigation. Eliminating both water stress and N stress gave a further yield increase, in particular in Las Lagunitas. Impact: Agronornic field studies cannot cover the wide range of natural conditions that one encounters in an area. Further, how can experimental results frorn one site be interpreted for use at sites with different soil, climate or management conditions? Agronomic simulation models such as DSSA T can play an important role here to ensure that recommendations that follow from research are not of a "blanket" type for a wide area, but are well targeted to the specific local conditions. 121 Models/Frameworks Model calculations are no replacement for experimental research; they give predictions of what is likely to happen given certain environmental and farm-management conditions. Simulation models can supply answers to "what-if' questions (e.g. what is the likely yield going to be ifthe fanner applies fertilizer?). U sed in interaction with researchers who know the systems well and who are able to provide input data for crop management and assess the validity of model predictions compared to actual on-farm results, a simulation model ís a very valuable research tool. lt can be used for evaluating farm-level management strategies and estimating yields for economic analyses. In particular, a simulation model can be used to assess the use of improved germplasm in association with improved natural resource management, an important objective of CIA T' s NRM research. Contributors: • CIA T project PE-5: Atjan Gijsman, Luís Horacio Franco • CIA T project IP-5 (l. M. Rao ): Identifying parameters needed for adding the forage legume Arachis pintoi to DSSA T. • CIA T project IP-5 (Carlos Lascan o): Identífying and analyzing a livestock decision-support model for its suitability to CIA T . · • CIA T project IP-5 (Michael Peters): Carrying out an experiment in Honduras for measuring parameters needed to add new legumes to DSSA T. • CIA T project PE-3 (Rubén Darlo Estrada): Collaboration on adding the Arachis pintoi option to DSSAT, because Estrada is setting up an Arachis seed production project with small fanners. • CIA T project PE-3 (Bruno Barbier & Marco Tulio Trejos): Obtaining DSSAT input data for estimating crop yields under various environmental and management conditions in Honduras. In tum these data provide input to PE-3 for evaluating farm-level management strategies and economic analyses. • CIA T project PE-4 (various): Applying the MarkSim weather simulator and interfacing DSSAT with GIS-georeferenced data on soil and weather conditions. • University ofCaldas (Colombia): Two students who worked on the Arachis pintoi model for DSSAT. • University ofGeorgia and University ofFlorida (USA): Modifying the source code ofDSSAT to incorporare a new SOM module needed for applying the model to low-input smallholder systems. • International Fertilizer Development Center (IFDC; USA): Estimating maize genetic coefficients for DSSA T and estimating climate data. • ILRI (Kenya) and University ofEdinburgh (UK): Collaboration on adding a livestock grazing option to DSSAT. • University of Edinburgh (UK): Invited meto collaborate on writing a book chapter for the Tropical Pasture Management and Utilization Workshop in Bolivia. • CrMMYT (Mexico ): Estimating maize genetic coefficients for DSSAT and collaborating for adding a new crop option for the legume Mucuna pruriens to DSSA T. 122 Models!Frameworks Activity 3.2 Develop economic models to assess technology/land use options 3.2.1 Economic model for ex-ante evaluation of research interventions in Aguaytia watershed Highlights • A user-friendly computer model, usíng common electronic spreadsheets, is functioning and being fine-tuned by local researchers and extensionists. Purpose: To provide an opportunity for local researchers and extensionists to examine interactions among crops within complex slash-and-bum fanning systems via the use of a computer model. Rationale: National and international institutions in Ucayali have developed many technological altematives for small farmers, but the adoption ofthese altematives has been mixed at best. Even though researchers and extensionists have expertise in one or more crops, they commonly have an incomplete understanding ofthe farming system functions with respect to scarce inputs as land, labor and capital. Outputs: Modeling ofthe region's farming systems involved required two stages ofresearch. First, crop- and farm-level data were collected to create enterprise budgets. These data were then used for the second modeling stage of fanning systems. A third stage of model refinement and the interpretation of results are being conducted with local researchers. Crop and farm system infonnation carne from two sources. The frrst source, national and intemational researchers, provided farm management data. The second source was direct information provided by farmers about their resources and economic retums of their farming system. The initial research focused upon three distinct agroecological areas. Surveyed were 21 flat-uplands farmers, 13 riverine farmers and 17 from the upland hills. (Table 1) Table l. Main crops and agricultural and production systems in the Ucayali Region, Peru Flat Uplands Riverine Upland Hills Rice Rice Rice Maize Maize Maize Plantain Plantain Plantain Bean Bean Cotton Cassava Soybean Cassava Pasture Watermelon Pasture Peanuts CamuCamu Cacao Oil palm Fishing Reforestation Poultry Poultry Pineapple Peach palm Citrus The agro-economic data included specific information regarding crop benefits and costs. A key component of the enterprise budgets was information regarding monthly labor requirements. All infonnation was organized and placed in electronic spreadsheets. The agro-economic data serve as parameters for the farming system model. With the objective of maximizing farrn profit subject to restrictions of land, labor and capital, the model determines the optímal 123 Models/Frameworks mix of cropping activities. To reflect the local situation of slash-and-burn farming, the model includes other restrictions such as a minimum land area for subsistence crop production. The model employs the use of two labor sources in order to calculate the maximum labor available in an average farro. The first, family labor is based upon the number of persons older than 14 years who can conduct farro activities. The second source is hired labor. For example, in the Ucayali there are severa! months when farmers need to hire labor (January, February, April, July, August and September). During these months, farmers typically hire two laborers. These model parameters are based upon research experience and personal communication with farmers. Impact: The use of an agro-economic model, which employs profit-maximizing linear programming techniques, permits a more complete understanding ofhow production factor scarcity (land, labor and capital) can affect farming systems. Furthermore, the model measures the impact oftecbnological changes such as new crops and varieties upon farm earnings. Cootributors: Douglas White, Ricardo L~barta, SN-1, INIA: MinAg-Ucayali, DEP AM, ICRAF 124 Models/Frameworks Activity 3.3 Develop a f ramework f or m onitoring and assessing impact of research in the forest margins Highlight: • A three day workshop was convened with national and intemational partners to develop an indicator framework that identified: i) opportunities for institutional collaboration and ii) restrictions inhibiting research and extension efforts from having on-the-ground development impact. Purpose: To improve the effectiveness ofresearch and extension efforts by institutions working in the Forest Margins, Pucallpa Perú. Rationale: Although institutions working in the Pucallpa region have specific mandates, they all strive toward a common goal of sustainable development. Yet at times the target of sustainable development is elusive. Not only it is problematic to defme; it is still more difficult to attain. Institutions often maintain a focus on specific research and extension outputs whose línkages to or impact upon development goals are not clear. A system of impact analysis enables the plarming and prioritization of research and extension activities in a logical and easily understood manner within a framework of sustainable development goals. (Pachico et al. 1998). Such a framework also permits continuous monítoring of research in progress. To measure research impact in and across different benchmark si tes, CIA T is creating an extensíve system of indicators. Although thís is useful for the international center, the outcome is likely to be too complex to be of interest to local national scientists and policymakers. Hence PE-S researchers, together with national and international partners at the Forest Margins reference site, developed a methodology to create a framework of indicators that wíll allow analysis of the impact of collaborative research. This workshop was a follow-up to the 1998 PPO (Participatory Plarming by Objective) entitled "A Common Agenda of Research and Development for the Central Amazon". In the first workshop, institutional representatíves prioritized four specífic themes crucial to the region: biodiversity, marketing of Amazonian products, professional training, and participatory development oftechnologies. To aid in . goal definition and to organize outputs, the logical framework methodology was employed to assist goal defmition and organize outputs. However, even with the logical framework, ít remained difficult to relate institutional efforts with the final goal of sustainable development. Outputs: This second workshop on impact assessment was based upon two pillars, one representing the supply of instítutional outputs and the other the demand of development goals. (Figure 1). The first workshop provided the inputs for the second pillar, i.e. the six development goals. Construction ofthe pillar to represent the instítutional outputs was the task of the impact assessment workshop. Thís required synthesis of numerous, detailed activities of the various institutions to produce the six institutional outputs. Figure 1. The Workshop Pillars and Bridge ~:::::-~ The Bridge->.>~ Technology Development Management and Conservation of Natural Resources lnstitution Strengthening Business Development Human and Environmental Health Policy Analysis and Formulation 1 Key lnst itutio nal Outputs 125 .---------------------------, lncrease Rural Family lncome Reduce lnfant Mortality lncrease Family Nutrition Maintain Biodiversity lncrease Forest Cover lncrease the Sequestration of Carbon 1 Goals of Su stainable Development 1 Models!Frameworks Sin ce not all institutional outputs reach the final goal of having an impact u pon development, there exists a gap between the two pillars. To illustrate via an example, the production of agriculture technologies is a tangible output yet if there are low rates of adoption, little development impact has been made. Thus a minimum requirement for research to have a known impact is that its results can be linked clearly to the development goals. The adding of activities, which increases the probability of an output making development impacts, is what (Smutylo, 1998) terms as improving the "reach" of an output. Hence, as for the case of agricultural technologies, research should identify ways to increase the likelihood of adoption in order to bridge the gap between merely producing research outputs and making sígnificant development impact. T o increase the reach for each of the six key institutional outputs, the workshop participants identified extra activities called intermediate steps. An example helps illustrate the detail required to develop the extra activities that form the bridge (Figure 2). For the case ofthe key institutional output, Policy Ana/ysis and Formulation, it becomes evident that a research output such as a model, which identifies the relationship between policy and land use change, is insufficient to have impact upon the goal of sustainable development. Other activities, which in crease the reach, improve the chances of makíng impact. For example, the policy making process and actors must frrst be understood and identified. Subsequent steps of creating communication channels between the researchers and poticy makers permit the research results to be shared and discussed. Then policy makers can make informed choices to formulate and implement policies that can aid in reaching the environment and well-being goals. Figure 2. Policy Analysis and Formulation Output Research itkntifies. relationships between policy and land use change - a. So en= CD"C .. 1: .!al "CCD CD.r! e: o o .. .E Goals . . .. " Policy maiiñg processes understood .. .:.;.- ..., .J. .:. ~ \S • . ;./~ .•. •. · Ide~tification of influential policy makers ~ .. ':, ·, .':'!<~:·.~ -..~l .. ~._.-/;.,; \; :~:: < .!' --: · ~€bann~ls ofcommunication established ,. ... .. .. ... ,; Information regarding problems and opportunities disseminated Policy makers understand information to make informed choices Formufation and implementation of appropriate policies Follow-up research ofpolicy impact(a feedback loop) w ~ Improved conditions of e11Vironment and human well-being Reach and control are inversely related. As extra activities along the bridge reach further from the original output, control by the researcher/institution wanes. For instance, dependence upon others such as the policy maker is nearly absolute at the policy formulation and implementation stage. The complexity and tangibility ofthese bridges differed according to each key institutional output. For example, while these intermediate steps may be sequential, sorne may need to be taken concurrently. With the key output Business Development, bridge construction was a relatively straightforward endeavor, while Jnstitution Strengthening proved to be more complex and less tangible. Discussion of such minute subtleties had the possibility of derailing progress of the workshop. Thus, the task of refining the sets of intermediate steps and indicators was left for the working groups established for each ofthe six key outputs. \26 Models/Frameworks lmpact: Tbere were three physical outputs ofthe workshop that served as a frrst stage ofinformation sharing and generation for the institutions working in the region. A sumrnary of all the research that is being carried out in the Pucallpa region. Diagrams (bridges) demonstrating the links between key outputs and development goals. A comrnon indicator framework to measure impact upon sustainable development. Ironically throughout most of the workshop, the direct development of indicators was eschewed. Rather, the indicators were a product ofthe local institutions' need to measure success with respect to research outputs, the intermediate steps and development goals. As part of this workshop, the participants also discovered opportunities for inter-institutional collaboration and revealed limitations or challenges that impede activities from having an impact upon development goals. Contributors: Douglas White, Sam Fujisaka, Ricardo Labarta, Dean Holland, Peter Kerridge, ICRAF, CIFOR, CODESU and national res~hers and extension staff. References: Pachico, D., Ashby, J.A., Farrow, A., Fujisaka, S., Johnson, N. and Winograd, M. (1998) Case Study and Empirical Evidence for Assessing Natural Resource Management Research: The Experience of CIAT. Smutylo, Terry. Rethinking lmpact Measurement: IDRC's Experience. 1998. Ottowa Canada, Intemational Development Research Centre. Evaluation Unit. 127 Models/Frameworks Activity 3.4 Develop participatory monitoring and evaluation methods for technology development 3.4.1 Participatory mooitoring and evaluation of new technologies developed with smallholders. Highlíghts • Impact assessment workshops carried out with stakeholders at the national, regional, and farmer leve!. • Víllage leve! spatial mapping and socio-economic indicator census conducted. • Benchmark and forage technology adoptíon surveys conducted. Purpose: l . To develop and apply a framework and methods to assess concurrent and ex-post impacts of forage innovations. 2. To study the process in which farmers experiment with, adapt, and finally adopt forage systems innovations. 3. To compare impacts ofparticipatory versus conventional research approaches. Ratiooale: There have been many projects aimed at reducing rural poverty by increasing productivity and maintaining the natural resource base. While participatory approaches to technology development ensure that new technologíes meet farmers' needs, there has been a lack of assessment of adoption and the impacts or outcomes on farm productivity and the natural resource base. While other studies of impact have focused on key productivity íncreases at the regionallevel there are few ' user-friendly' methods that assess environmental and economic impact at the farm level during early stages of adoption. Any framework for monitoring progress or assessing impacts of new technologies must be related both to the problems/needs expressed by farmers as well as expected outcomes at different scales (farm, community, region). With this capability farmers and researchers can modify technology development to better target and meet both local and regional needs. Methods: The project uses a mix of participatory and conventional approaches in developing a methodology for concurrent and ex-post impact assessment at each of the project village sites (Table 1). Table 1 List of Sites aod Activities "' ¡: ·¡¡; V 01) o E e e ·o. 01) . V (;j :o "' V e: o o. ·,:: e "' o c. u Spatial Mapping X Commu.nity GIS village level F ocus groups at mapping and socio- maps. Socio-economic project villages. economic census by indicators and wealth focus group rankings. Population data for survey sample stratification Problem Diagnosis X Flowcharting, Indicators ofFarming Focus groups at weighting of systems constraints project villages. importance and their relative importance. Impact Assessment X Flowcharting, Indicators of Forage developers, weighting of intermediate and final municipal level importan ce impacts of adoption agricultura! officers and their relative and nationallevel FSP importance. coordinators focus groups. Case Study Interviews X X Semi-structured Farro. group, and Purposively sampled interviews, partial village level farming farmers in project budgets systems information. villages. Participatory Methods X X Participatory National Institution Municipallevel Training approaches, building and training. agricultural officers, interview community techniques development workers, survey enumerators. Benchmark Survey X X Socio-economic Database for Multistage indicators, matrix concurrent and ex-post proportional stratified weighting, adoption analysis of impacts sample of smallholders decisions, farming and technology spread. in project villages. system outputs and Baseline information inputs for monitoring and evaluation of farming system Forage Adoption X X Forage evaluation, Database for All forage adopters in Survey forage concurrent and ex-post project villages. development, analysis of forage and expansion plans species adoption and development. 129 Models/Frameworks Workshops will be held with municipal, regional and nationallevel agricultura! officers to develop and strengthen institutional expertise in impact assessment and project monitoring and evaluatíon. Outputs: l. GIS maps of project villages have been constructed along with spatial databases of socio-economic indicators and wealth rankings. 2. Identification of constraints on the farming system facing smallholder fanners has been obtained from consultations with farmers in the project site (Figure 1). 3. Indicators of impact of forage technologies have been obtained from consultations with the various stakeholders (Figure 2). The identified impacts were collated from each ofthe stakeholder focus groups and each of the farmer groups was asked to weight the impacts according to importance. There was a wide distribution of responses from each si te but severa! impacts stood out as being most important to aH the farmers (Figure3). The importance of these impacts to the smallholders in the project area ha ve been statistica11y analyzed (Figure 4). The impacts that were considered ·very important to the farmers included: • The ability to provide feed to counter seasonal and overa11 shortfa11s, • The opportunity to ensure the safety of animals against theft and accidental death by tethering closer to borne. • The ability of forages to fatten animals, and • The soil erosion control potential of forages . 4. A database has been developed storing the results of the Benchmark Survey and the Adoption Tree Survey. This database will enable statistical analysis of impacts and adoption to be carried out for the project sites. The Adoption Tree Survey is also being conducted at other FSP sites in SouthEast Asia and the results stored in the database as well. The database will be linked with the GIS maps to provide a spatial dimension to the data analysis and presentation. Impact: Apart from the training of local extension staff in participatory methods of project monitoring and evaluation the project impacts at this stage of the project cycle are minimal. While all milestones set out in the project documentation ha ve been achieved, results are expected after the second stage of the project. The likely impacts in elude the development of a methodology for identification of project irnpacts in a participatory framework as well as a methodology for benchmarking, concurrent, and ex-post project monitoring and evaluation. Cootributors: Tim Purcell and Rob Cramb (NRSM, University ofQueensland, Australia); Wemer Stür, Francisco Gabunada and Louie Orencia (CIA T, FSP Philippines); Sam Fujisaka, Peter Kerridge (CIA T, Cali, Columbia); Willie Nacalaban, Gaspar Velasco, Pinky Ojales and Judith Saguinhon (LGU, Malitbog, Philippines ). 130 Mxll4 1 • 11101 (111ta d!latot / eli aq:s) ~yltat:atoon baklid (I.Jlrd Íl st~ ;rd staT)' ) Wlilllf y uta {t-b llrd) Sdl f ertilty (Scil fertlity is 1~ 1--tl lblro(I.Jlc:k d fertlise-) l/llfDJ silo sle (t-b sito sitef a cxmnvity) Models!Frameworks ~ '1 R'dllem[laglosis, Stio PaléVl, Kal~kilg. 17 Jire 19999.X&n Tas arg pesyo sa ¡:a¡.albl (Hgh ¡:ri:e of tasic lll!lldl) Sl*laaht~¡tcp (Mrral d-es) ~ haycp bisafa (Mrrala - b:al treeca-lo.v ~- ni sal& pice) EdJati::n {Tltbn fees are t'ig'l ) Oilmlka- ¡:ae~sa rek (Ca'tt serd ct'il- -o ...... ~ V o V ....... (/) e: .2 Q) "' .~ ~ e Q. :::> ~ Q) e: Q) E -"" ~ u o E e "' o E g' "' e;; .!? O' -o o ·:;: e;; Q) e: i Q. o e: ~ o "' ·e; :;:; o Q) en -o o~ "' E o Q) "' :;:; "' >.0 ¡.: ] ~ :.e o ~ a. .=u "' t; Q) o ...... "' e S ~ a. e: g o u e: 8 • Very lmportant O lmportant • Less lmportant 133 Models/Frameworks Figure 4 Analysis of Importance of lmpacts of Forages in Smallholder Farming Systems lmpact Mean Importance Least Significant ( 1 =lmportant, 1 O=Least lmportant, Difference (p<0.05) >11=Not lm~rtant}_ Provides Feed for Livestock 5.79 X Animal Safety (Theft, Accident) 6.15 X X Fattens Livestock 6.68 X Soil Erosion Control 7.64 X Provides Manure and Fertilizer 8.12 X X Increased Livestock Price 8.33 X X X Cut&Carry prevents straying 8.67 X X X Sell Feed 8.69 X X X Livestock Numbers 8.85 X X X X Soil F ertility 8.97 X X X X Time Savings 8.97 X X X X Help with Finances 9.09 X X X Crop Yield 9.11 X X X Healthy Animals 9.12 X X X X Education and Tuition Fees 9.29 X X X Labor reduction for women and children 9.37 X X X Livestock Dispersa! Program 9.62 X X X Animal/Human Health 9.63 X X X Landscaping 9.99 X X Land Tenure 10.42 X 134 Models/Frameworks Activity 3.5 Integrate information on variety adaptation and appropriate technologies with GIS databases to target germplasm use 3.5.1 Converting the forage database to a graphical platfonn Highlight: • Trial version of database on a graphical platform accessible to selected scientists vía the Intranet Rationale: The Tropical Forage Program in CIA T has generated an information system for the evaluation of germplasm, right from collection or exchange to the release of cultivars by national institutions. A great part ofthis information has been entered into an ORACLE database, which at present is available for CIA T scientists. For the actualization, inquiries and searches of forage data, an information system based on the fourth generation language ORACLE FORMS 3 was developed; this system is available vía the Calima Server. However, in view ofthe technological advances, the requirements ofusers in CIAT and the importance of sharing all the research results with other partners institutions through the internet or vía magnetic means, it is important to convert this information system to a graphical, user-friendly and attractive platform. Methods: The programs in ORACLE FORMS 3 are converted to ORACLE DEVELOPER 2000, with emphasis on user-friendliness and on an attractive platform. In the frrst phase, the old programs with information avaílable in character mode are transferred to graphic mode. In the second phase new modules are added. Once completed, the database will be tested via CIA Ts Intranet and through Internet and CD- Rom versíons developed. Results and discussion: An initial tria! version was developed. In figures 1 and 2 screenshots ofthe current versions are shown. The actual versíon was tested for year 2000 compatibílity. Next steps include the further refinement and testing of the database for lntra- and Internet use and the development of a CD-ROM version. Target group for this CD Rom are mainly Nars and NGO's in Latín America. In the future an English version ofthe database is planned. At present, the database contains a total of 1537 accessions, evaluated in 315 sites. In many ofthese sites, available information is incomplete and therefore a great effort is being made to update the database through capturing information from existing publications and through obtaining data directly from scientists in the forage and Production systems projects. Impact: A preliminary version of the graphically and user-friendly database is available for selected scientists through the Intranet. During 2000 this database will be refined for utilization via the Internet on for a CD-Rom. This database will also be utilized in developing the Decision Support Tool for the targeting of forage germplasm (see section below). Contributors: IP-5 (Martha Herrera, Arturo Franco, Carlos Lascano, Michael Peters), PE-5 (Luis Horacio Franco), SB-1 135 Models/Frameworks Figure l. Main screen ofthe forage database in an attractive graphic platfonn developed in ORACLE DEVELOPER 2000 Figure 2. Second screen showing the type of infonnation available for a specific accession 136 Models/Frameworks 3.5.2 Use of GIS models for better targeting of forage germplasm for end users Rationale: The overall approach which intends to integrate agro-ecological, economic and social information, is based on the following main assumptions: • a wealth of information on the agro-ecological adaptation of forage germplasm is available in CIA T' s- held forage databases but access and hence utilization ofthis information is poor; • there is a need to tak.e account of socio-economic factors that influence adoption as well as adaptation to environmental conditions. Based on these assumptions, a new approach to targeting forage germplasm needs to enhance the utility of existing information, allow for integration of environmental adaptation and socio-economic factors influencing adoption, and for multiple end-uses. lt is anticipated, that this approach will allow a more accurate and client-oriented prediction of entry points for forage gennplasm into farming systems. Metbods: A work.ing group was formed to carry out the work and agreed to follow a step-wise procedure for the development ofthe system: • Incorporate the existing RIEPT (Red Intemational de Evaluación de Pastos Tropicales) database- for regional trials A +B into the GIS system to describe the agro-ecological adaptation of forage germplasm in Latin America • Incorporate supplementary information of agro-ecological adaptation existing in CIA T -held forage databases e.g. the RABAOC database. • Incorporate experiences of CIA T and NARS partners on adaptation • Incorporate socio-economic information from characterization studies on a regional level ( e.g. Central America). Results: Parameters for the description ecological adaptation and agronomic performance of accessions across environments were identified. Currently the group is in the process of revising the classification of agro-ecosystems to be utilized for the database as the classification of agro-ecological zones developed by Thomas T. Cochrane may not be suitable for use in the GIS too! to be developed. Impact: lt is believed that the system to be developed will greatly enhance the availability of integrated information on the agro-ecological and socio-economic adaptation of forage germplasm for multiple uses. The integration of infonnation from different sources will allow the improved targeting of forages to farmer's conditions and demands. As a result it is likely that: • efficiency and client-orientation of future research will be enhanced, and • the dissemination of existing and future research results will be improved. Progress has been made in data preparation ofthe database underlying the GIS model and sorne trials have been run. For better targeting, of climatic and soil conditions the classification of agro-ecosystems is being revised. Contributors: IP-5, PE-4, PE-3 , PE-5 (M. Peters, Glen Hyman, Luis Horacio Franco, Arturo Franco, Belisario Hincapie, Gerardo Ramirez, Alexander Gladkov. P. Jones) 137 Models!Frameworks Activity 3.6 Develop decision guides on soi/fertility management 3.6.1. Conceptual frameworks to guide research and extension efforts for integrated nutrient management. Higblights • NARS researchers from eight countries developed conceptual frarneworks, using a decision guide format, and prioritized activities for research and extension for 14 agro-ecological zones. Purpose: To improve the efficiency of research and extension for integrated nutrient management by applying information available from various sources in consideration ofthe temporally and spatially diverse situations of farmers. Rationale: Small-scale farming systems are often agronomically and biologically diverse. The farmers have diverse but scarce resources to use in soil fertility management. They need to get very high rates of return on invested money, as the opportunity cost is high. Information for an Agro-Ecological Zone (AEZ) is often scarce but information from other places may be applicable. Application of such information to typical situations in the AEZ, coupled with preliminary economic and agronomic analyses, enables the researcher to make reasonable estimates on how best to use scarce resources. Information needs and adequacies are identified, and priority activities for research and extension are identified. Metbods: Members ofthe technical working group ofECAABREN for research on integrated nutrient management met to develop conceptual frameworks for improved soil fertility management for selected AEZ's. Information from diverse sources was integrated and interpreted in consideration of characteristics of farming systems in the AEZ. Preliminary agronomic and economic analyses were done. Conceptual frameworks were developed following a decision guide format. Priority research and extension activities were proposal and operational plans were developed. Impact: Conceptual frarneworks were developed for farming systems of: (i) The Bukoban High Rainfall Zone ofNW Tanzania (ii) Sandy loam soils ofthe Westem Transect ofMount Meru, in N. Tanzania (iii) The Mbozi Palteau in the Southem Highlands ofTanzania (iv) The Ded:za Hills of Malawi (v) Four AEZ' s of South Kiyu, D.R. Congo (vi) Areka, SW Ethiopia (vii) The Andranomanelatra AEZ, Antsirabe, Ethiopia (viii) Gikongoro AEZ, Rwanda (ix) Crete Congo AEZ, Rwanda (x) Lake Victoria Crescent, Uganda (xi) Northem Moist Farmlands, Uganda (xii) Vihiga District, W. Kenya V arious field activities are being implemented in most of these AEZs. Collaborators: C. Wortmann, Kayuki Kaizzi and the technical working group ofECABREN on integrated nutrient management. Linked to IP2. 138 .) t.. ,. ~ Increasing Research Effectiveness Output 4. Increased effectiveness of CIA T and partners to conduct appropriate research for developing productive and sustainable land use practices Activity 4.1 Coordination and Funding This has in volved coordination of CIA T activities at the agro-ecoregional si te for the forest margins in Pucallpa, Peru and activities of project PE-S 4.1.1 Forest margins eco regional site- Pucallpa Highlights • A second Participatory Planning Meeting was organized and held in Pucallpa with all institutional stakeholders to jointly link research thr~:mgh a set of intermediate and development indicators • CIA T research in forest margins gained a new momentum with the location of a resource economist anda specialist in participatory approaches in Pucallpa • A meeting between CIA T, CIFOR and ICRAF researchers to share research results and plan future research was organized and held in Pucallpa Outputs: This year we have continued to develop closer working relationships with both our international and national partners based in Pucallpa. CIA T research efforts have been increased in areas of economics and participatory research (with staff located in site), in land use studies, in agroenterprise development and in the area of community health in relation to land use. A framework was developed for choosing a set of intermediate and development indicators that unify the research and development prograrns of all institutions located in Pucallpa. This led to the formation of working groups to refine both research objectives and indicators so that they were logically linked with a set of common development goals. Research was initiated on the trade-off between productivity and environmental goals ll multi-institutional projects were initiated by the Participatory Research Team (DEPAM). DEP AM is a non-sectarian platform for facilitating demand driven research and making the most efficient use of resources of the locally-based institutions. A training series on participatory research was initiated. Research commenced on an analysis of poverty in relation to location and land use. Market studies have been completed of exotic crops and are commencing on the more traditional crops of the area. A survey has commenced ofthe health of 6 contrasting communities in the forest margins A improved communications system has been set up for the Ecoregional Center using an interna) network and a dedicated telephone line for internet connections Impact: CIA T is leading the research in areas of economics, participatory research, land use, and agro- enterprise development and has facilitated many coordination activities Coordinator: Peter Kerridge (CIAT-Palmira) and Doug White (CIAT-Pucallpa) 139 lncreasing Researcb Effectiveness 4.1.2 PE-5- Sustainable Systems for Smallholders Higbligbts • Systems research at the watershed leve! has been consolidated in the forest rnargins site, Pucallpa, and with at an indigenous cornrnunity site in Vietnam • Commodity oriented research is producing new forage and cassava rnanagement technologies with wide application Outputs: The project workplan was revised, individual workplans developed and assessed rnid-year, and an annual report prepared. The Project Manager visited with scientists at sites in Asia (Laos, Philippines, Thailand and Vietnam) and Latín America ( Peru). Coordination was focussed on consolidating CIA T research at the forest margins reference si te near Pucallpa, Peru, in providing support to a national team in Vietnam facilitating cornrnunity rnanagement of natural resources, in liaison with donors, in developing project proposals for continuing research with a commodity focus in forages and cassava, in seeking integration with other CIA T projects in geographical or commodity areas where the PE-5 is active, and in providing support to younger staff, in particular, those who are outposted. Support was provided to those directly responsible for coordination of the DEPAM project, Tropileche, Forages for Smallholders, Integrated Cassava Cropping Systems projects and Ecosystem health. Monitoring and evaluation activities have been commenced in all areas where the project is active. Impact: PE-5 continues to have effective teams developing component technologies within a systems context for forages and cassava using utilizing new germplasrn from the Cassava and Forage Germplasm Improvement projects. Systems research has been consolidated at the watershed leve! in the forest margins and in Asia. Feedback from staff suggests that they are receiving adequate support in their activities. Coordinator: P. C. Kerridge 4.1.3 Funding 4.1.3.1 Existing sp«ial projects Foragesfor Smallholders (Southeast Asia). CIA T and CSIRO. Funded by AusAID. Jan 1995-Dec 1999. Integrated Cassava-based Cropping Systems in Asia: Farming Practices to Enhance Sustainability. CIAT. Funded by the NIPPON Foundation. Jan 1999-Dec 2004 lmproved Legume-based Feeding Systems for S mal/holder Dual-purpose cattle production in Tropical Latin America. CIA T and ILRl under the SLP. Funded by BID and the SLP. Jan 1998 to Dec 2000. Addressing Poverty and Protecting the Environment in the Peruvian Amazon: The Interaction of Technological Innovation, Policy and Marlcet Opportunities. CIA T and CODESU. Funded by IDRC May 1998 to Dec 2000. Extension expected to Dec 1999. Health, Biodiversity and Natural Resource Use in the Western Amazon Lowlands: an Integrated Agroecosystem approach. CIAT and Univ. Guelph. Funded by IDRC. Dec 1998 to May 2001. 140 Increasing Research Effectiveness Community-based Upland Natural Resource Management in Hong Ha commune, Thua Thien Hue Province, Vietnam. University ofHue and CIAT. Oct 1998 to Jun 2001. Evaluation of Leucaena in Central America and Southeast Asia Contract with Oxford Forestry Institute. Apr 1997 toMar 1999. Participatory Monitoring and Evaluation of New Teclmologies Developed with Smal/hollders CIAT and University ofQueensland, Funded by ACIAR. Jan 1999 to Jun 2000 An assessment of the impact of smallholder cassava production and processing practices on the environment and biodiversity. ClA T. Funded by IF AD 4.1.3.2 Projects near approval Developing Sustainable Forage Technologiesfor Resource-poor Upland Farmers in Asia. CIAT. To be funded by ADB. Jan 2000 to Dec 2003. 4.1.3.3 In discussion witb donors Improving livestock production systems through the integra/ion of forages on smallho/der farms in northern Lao PDR. CIAT and CSIR.O. Donor is AusAID. 4.1.3.4 Proposals deve1oped F orest margins research Collaboration in developing the Altematives to Slash-and-Bum Phase ill propasa! Promoting Partnershipsfor Improved Livelihoods and Environmental Management in the Amazon Forest Margins of Perú. Moriah Fund Concept Note. Not approved. Socio-Economic Analysis of Farming System Technology Options in the Aguaytia Region, Perú. Presented to Winrock Intemational, Lima, Peru by CIA T -ICRAF, Pucallpa. Tropileche Evaluation of options for the development of the dairy industry in tropical Latín America. Submitted to Fontago. Not approved. Integrating biodiversity conserva/ion and smallholder /ivestock production in subhumid tropical landscapes. ILRI and CIA T. Submitted to CRUSA and the SLP. Southeast Asia Integratingforages into uplandfarming systems of Vietnam. Prepared on request of AusAID. AusAID decided not to go ahead with development of the proposal. 141 lncreasing Research Effectiveness Activity 4.2 Faci/itate multi-institutional and participatory research at the watershed leve/ 4.2.1 Multi-institutional research in the Aguaytia watershed through the DEPAM project, Pucallpa, Peru Highlights • A multi-institutional participatory research team is taking an active role in testing new ways of workingjointly across institutions, and with farrners. Purpose: To evaluate processes for developing a multi-institutional and participatory research model involving international and national research centers in Pucallpa, Peru. Rationale: Pucallpa is a difficult place to catalyze development through agricultura! research. lt is dynamic and transient: more than half of the current farrners airived in Pucallpa in the last 1 O years (Fujisaka et al. 1997), yet up toa third of farrners wish to lea ve in search of areas with better soils (Y anggen, 1997). In frontier areas, a hectare of land can cost less than a bag of fertilizer, possibly making intensification a fantasy while cash resources are so low and labor is relatively cheap (White et al. 1999). Effective solutions for dispersed populations in these conditions are likely to be interdisciplinary and community-implemented, yet research has continued to take a disciplinary and supply-driven approach. In response, CIA T has catalyzed a multi-institutiona1 and demand-driven approach to research in Pucallpa in two complementary ways (i) through its nonnal role as a lead or associate agency following a specific research agenda in collaboration with local institutions and (ii) through facilitating the development of a Participatory Research Team (DEP AM) whose 1eadership is in local hands (see Output 2.1.1 ). This presentation focuses on the processes and research activities involved with multi-institutional research through the Participatory Research Tearn. Output 4.5 "Institutionalizing a Collaborative Research Approach in Pucallpa through the project DEP AM" focuses on the parallel management structures. Methods: DEP AM provides a vehicle for influencing the following Ievels of activity relating to agriculture: 1 2 3 4 Farmer Capacity and Researcher Improving Research Organizational Space Control Professional Design and Development Implementation • Supporting the • Too1s for • Support in Creating space to deve1opment of Participation participatory project institutionalize fanner • Processes for design and participatory and multi- organizations Eva1uation imp1ementation discip1inary research • Awareness-raising • Helping people to • Putting of technologies, do participation, participation in the resources and and to know if it picture institutions has worked available • Bringing a gender • Linking fanners focus around needs and • Bringing a resources community focus 142 lncreasing Research Effectiveness The approaches at alllevels are similar, i.e. to open spaces for action, critica! reflection, and deliberate learning strategies based on experience. DEP AM curren ti y focuses its systematic activities on levels 2, 3, and 4, but alllevels are closely linked in practice. The effects of these interventions alone and in combination form the basis of action research into strategies for sharing multi-institutional and participatory approaches in Ecoregional sites. Outputs: At the project/institutionallevel, DEPAM provided seed funds for proposals for small multi- institutional participatory research projects in July 1998, anda second round ofprojects in April 1999. These projects are detailed under Output 2.1.1. Through a competitive system, project proposals were evaluated by both an advisory committee of local researchers, and a committee of farmers who represent producers throughout Ucayali. For professional development, DEPAM facilitated a course in tools and approaches for participatory research in October 1998 with 24 researchers, and is currently responding to researcher demand with a second course with 18 researchers which started August 1999. In addition, the participatory research team selected two members to attend a workshop "Gender and Stakeholder Analysis for Participatory Research", and held a workshop among themselves to feedback and analyze the outcomes ofthe course. CIA T is providing continua! field leve! support and mentoring to DEP AM projects through staff in Pucallpa. Members ofthe DEPAM participatory research team have met on two occasions to evaluate jointly their progress within the projects. They identified the following difficulties with implementing participatory approaches: • Overcoming the pressure to obtain immediate positive results • Lack of institutional freedom • Maintain institutional interest in participatory research (financial interest) • Maintaining the scientific rigor of participatory research • Achieving and maintaining interest and enthusiasm of farmers • Understanding farmers' problems • Pitching participatory research at an appropriate leve! of complexity Impact: Members ofthe participatory research team are planning and carrying out field activities, such as field days that include researchers from severa! institutions. In addition, the participatory research team is starting to make joint decisions regarding the use of DEP AM resources, such as training opportunities. Team members are taking increasing responsibility for their own leaming. They have identified the following learning goals, and are designing a process to achieve these through a "leaming project", which is facilitated by DEP AM: • Simple methods to include both men and women in projects • Techniques for empowerment • Adapting participatory methods to reality • Techniques of social evaluation, including gender aspects and families Projects are experimenting with their processes of research, and changing them as a result. For example, the levels of participation in the small grant projects was evaluated on a scale of 0-1 O and gave an average of 3.8 (range 0-8). Researchers are jointly designing and implementing a conscious leaming strategy to diagnose and improve levels of participation, which currently includes their own upskilling, and consultation with farmers and with research managers. Contributors: Dean Holland, Sam Fujisaka, Peter Kerridge, Douglas White, Glenn Hyman, Scott Cechi, Carlos Bruzone, Carlos Ostertag, Mark Lundy, Ann Braun, Maria Fernandez from: CIA T projects PE-5, PE-4, IP-1, SN-1 , SN-3, inter-center project Participatory Research and Gender Analysis Program. 143 4.2.2 Impact assessment workshop in PucaUpa - the process (See also Output 3.3) Higbligbts Increasing Research Effectiveness • The fonnation of inter-disciplinary multi-institutional working groups in Pucallpa to refme impact indicators and collect required data. Purpose: To facilitate and enable national and intemational research partners to develop a feasible framework for impact analysis of research and development outputs in the Forest Margins reference site, Pucallpa. Rationale: While scientific papers and books containing lists of development and sustainability indicators abound, their actual use is rare (Amazon Cooperation Treaty, 1995; Dumanski et al. 1998). Not only are these works too generalized to represent a particular geographic area, their fonnulation often has omitted to include those who compile and use the data. Hence there ex.ists a detachment between available infonnation frameworks and the needs of local institutions. Outputs: The PE-5 research team developed a process to include personnel from all research and development institutions located in Pucallpa in the development of an impact analysis framework. This " participatory institutional" approach helped address their informational needs by ordering and priortizing a plethora of available data. The identification of incentives so that national institutions were willing to embark upon the process analyzing impact themselves was an implicit but central task ofthe workshop. For example, impact measurement is a useful monitoring and accountability system for local national and intemational directors, be they from govemment or donor institutions. The provision of an information system, which displays accountability and impact, can assist local institutions in acquiring additional funding. While the structure used to organize the workshop and indicators appears like a logical framework matrix, it is distinct. The framework contains five boxes and the order in which to work them is crucial. lnstead of beginning with a supply of institutional activities and then defining the goals of development, in a self- justifying marmer, it was decided to begin with the demand or goals side. Thus the workshop commenced by identifying the components of sustainable development (Box A). Results of the 1998 PPO workshop (see Output 3.2) provided the input for this box. Measures of development to the goal or where the region wants to arrive. Discussion next moved to the supply si de, the research and extension activities of the institutions (Box B). Participants filled out cards representing what they do and placed them on the walls. The following task required generalizing these activities to form key institutional outputs. (Box C). This was accomplished in a group manner by moving the cards into thematic groups signifying the reasons for perfonning the activities. Although a key output from one institution may be an activity to another institution, the exercise distilled the main themes thus providing a common language. The development indicators (Box A) and key outputs (Box C) correspond to the pillars ofFigure 1 in Output 3.3. The next exercise plarmed is to elaborate measures of success for institutional activities. However instead of using details of the specific activities, it was decided that the indicators be selected for the more general level of key institutional outputs. (Box 0). Subsequent refinement would take place at a later stage in the working groups. The final stage of filling the workshop 144 Figure 3 Framework used at t he Workshop Key Output lntermediate Outputs: milestones: Steps: e D E Activities: 1) 2) 3) B Working Order of the Boxes: A: Development lndicators- measures of where to arrive B: Activities- what the institutions do C: Key Outputs- reasons for performing the activities D: Output Milestones- results ofthe key institutional outputs lncreasing Research Effectiveness Development lndicators: A E: Intermediate Steps- the bridge between the activities and the development goals structure required identification of the intermediate steps between key institutional outputs and the goals of sustainable development (Box E). Participants formed six groups to construct a bridge for each the key outputs. lt soon became apparent that more activities were needed to increase the probability of having impact u pon the sustainable development goals. Impact: The most important outcome of the workshop was the processes of facilitating collaborative research. An example was the formation of inter-disciplinary and multi-institutional working groups to refine the impact indicators. Subsequent meetings of these groups ha ve brought together scientists and administrators to discuss how change can be made. Through the process of developing the indicators and impact analysis product, ownership rests within the institutions in Pucallpa. The work of distributing tasks amongst the participants to compile existing data and collect field data remains. Contributors: Douglas White, Sam Fujisaka, Peter Kerridge, Dean Holland, Ricardo Labarta, Mario Lanao-ATINCHIK, ICRAF, CIFOR. CODESU References: Tratado de Cooperación Amazonica Proposal of Criteria and lndicators for Sustainability of the Amazon Forest: Tarapoto. 29, Lima, Peru: TCA: Pro Tempore Secretariat. (1995) Dumanski, J., Pettapiece, W. and McGregor, R. ( 1998) Relevance of Scale Dependent Approaches for lntegrating Biophysical and Socio-Economic Information and Development of Agroecological lndicators. Nutrient Cycling in Agroecosystems 50: 13-22. 145 lncreasing Research Effectiveness 4.2.3 Facilitating Inter-Center collaboration in the forest margins, Pucallpa Highlights • An Inter-Center meeting between CIA T, CIFOR and ICRAF carne up with a resolution and agenda to develop collaborative research projects for joint submission to donors. Rationale: Pucallpa is the reference site for research in CIA T in the forest margins ofthe Amazon, i.e. it is a CIA T ecoregional reference site. lt also is a reference site for the Alternatives to Slash and Burn Program led by ICRAF. CIFOR also has research program on this site. IPGRI has expressed an interest in undertaking research on genetic resources of indigenous crops such as cassava provided it can obtain funding. The Centers individually have collaborative links with the national organizations in Pucallpa. Hence it is logical that the Centers themselves should have strong collaborative links with each other. Outputs: There are two levels of outputs. The most common is informal collaborative research between scientists with common or complementary skills. The second leve! of collaboration have been meetings to acknowledge and make use of complementary skills in order to make efficient use of resources. Joint meetings or more formal activities: • November 1997- Inter-Center meeting ofDG's from CIA T, CIFOR and ICRAF who agreed on a Common Vision for research at the Ecoregional Reference site at Pucallpa. • February 1998- The Centers moved into a common office building on the grounds of INIA, the national research organization under the Ministry of Agriculture and now share other common facilities such an electronic network. • July 1998- Research ofCIAT, CIFOR and ICRAF was put into a common logframe that used the Common Visionas a framework. This allowed everyone to see what others were doing but did not result in any increase in collaboration between scientists. CIA T requested and organized in ... • June 1999- A meeting of research scientists from CIA T, CIFOR and ICRAF to share and discuss research outputs and research agenda. The outcomes of this meeting were: i) a better understanding of our goals that lead to a vision for the direction of future research ii) Better understanding of each other's research with areas for future collaboration identified. iii) An agreement and agenda to develop three research proposals for joint submission and funding. An emerging vision - "Our role is to diminish the undesirable trajectories and increase introduction of diversified land use systems, facilitate sustainable soil and crop management, capacity building and inter-institutional change, produce global goods, undertake "trade-off' analysis and develop political support for desirable policies, and contribute to improved health and nutrition of local people." was developed from the following table which was put together following a short brain storming session. 146 lncreasing Research Effectiveness (i) Visioo of the directioo for future research Desirable Undesirable Diversified land use systems lnappropriate agriculture Land systems based on improved zonification, - Expansion of cattle ranch.i.ng at expense of SF, Mix of perennials and long term falloffs, RF due to poücy signals, Sustainable rotational systems(where demand for Annual food crop production? intensification) Diversified production systems, Unsustainable forest use Agroforestry systems/Forest based production No management ofprimary forest systems in farmland Destructive extractionism, Mosaic pattern of land use systems High quality germplasm available, Deforestation Harvesting the rain forest, ueva ley sobre desarrollo amazonia, Profitable secondary forest production, Tree monoculture Production areas of forest under management by i) forest enterprises and ii) rural communities Global goods delivered Biodiversity Loss Biodiversity protected Monocultivo, lncrease in e sequestration Loss of genetic diversity lmproved soil and crop management Soil degradation Better tools for soil management, Easy alternatives for small farmers Política! support lnappropriate policies Better informed targeted policies New Iaw sobre desarrollo amazonia, Forestry law ofland tenure Vs management lmproved market mechanism Lack of markets Poor access to markets Trade-off analysis available Lack of Trade-off analysis Agricultura! incomes Vs forest cover and biodiversity, Short rotation improved falloffs Vs secondary forest cover Community control Lack of community control Farmer initiated research, Disempowered, apathetic and poor people Community ownership of action Capacity building and institutional change Mistargeting researcb Acopió de mas investigacion, NARS-increased Research defined independent of national capacity in NRM, Centers and others organizations, lnappropriate agriculture, collaborating in R&D Research not used by local producers, poor collaboration among Centers and others lmproved health and nutrition Improved nutrition, Reduced child mortality and child morbidity 147 (ii) Recent, current and proposed research and gaps (Under themes ofthe Common Vision) Increasing Research Effectiveness This summary is presented as a record of the discussions and as a basis on which more definite research plans might be developed. IT IS NOT COMPLETE. l. Policy and institutional environment Recent Pasture technology and deforestation study (CIA T) Current Farm leve) and maximization model w/ ex-ante analysis oftechnical innovations Economics and policy analysis of fallow management strategies Poverty -market access or resource access (CIAT) Nutrition programs with DRS (CIA n lnstitutional structures for multi-disciplinary researcb (CIA n Potential for participation in C markets (CIFOR) Future Riverine farm risk analysis (CIA n Landscape mapping re land use trajectory scenario (CIFOR) Landscape level research (CIFOR, CIA T, ICRAF) Political economy secondary forests and policy (CIFOR) Evaluate farmer organizations (operational mgt) (CIA T) Diagnosing and strengthening of support system for local economic activity (CIA T) Gaps Understanding trade-off and targeting to the appropriate content Forest sector policies 2. Managing biodiversity Recent Plant community biodiversity and land use (CIAT, ICRAF, INIA) ICRAF- Priority setting of agroforestry trees, economic evaluation of agroforestry trees, local knowledge about agroforestry trees, farmers use and management of germplasm, molecular genetic diversity in agroforestry trees, Calycophyllum Current Crop geography and agricultura) development (ICRAF) Ecological footprint (nutrition and biodiversity) (CIA T) Importance of fish in diet of riverine communities (CIA T) ICRAF- Molecular genetic diversity in agroforestry trees- Guazuma crinita Effect of domestication on genetic diversity- lnga edulis Genetic variation among provenance, Calycophyllum, spriceanum, Guazuma crinita Genetic variation among families - Bactris gasipaes ICRAF- Effectiveness of phenotypic selection- Inga edulis, Calycophyllum spriceanum Germplasm production systems -secondary orchards Economic valuation of gerrnplasm production systems Future Ethnobotany and ecology of social and economic important biodiversity (ICRAF) ICRAF - Genetic variation among provinces/families 4 species Genetic diversity within species Reproductive ecology of species G. crinita, C. spruceanum Silvicultura! management oftrees in secondary forest- G. crinita, C. spruceanum Secondary forest- G. crinita, C. spriceanum Indigenous knowledge about trees 148 Gaps Methods to manage and maintain biodiversity lmplications ofbiodiversity change Intellectual property rights Local utility of biodiversity lncreasing Research Effectiveness Ethnobotany and equity of social and economic important biodiversity 4. New options for diversified land use systems Recent CIA T- Proposals for methodologies - market identification, Integrated agroenterprise projects, design of support system Selection of exotic portfolio-agronomic, commercial and economic characterization Inventory of- development projects, native processing, key non-exotic products Establishment of Rural Agroenterprise Development Committee CIFOR Dynamics of secondary forests in new and old colonization areas Inventarios floristicas en punnas mayores de 4 anos en unidades agrarias en tres sectores (semuya, nueva requena, neshuya) Patrones de uso de tierra y su impacto sobre los bosques secundarios Current CIA T On-station rice testing FPR rice testing Participatory evaluation of exotic market options Market study of non-exotic products ICRAF FPR improvement offallows, organic inorganic fertilizer, contour rows, multistrata Applied research (farm site) improved fallows CIFOR Rehabilitation methods for degraded secondary vegetation and pasture lands INWCIFOR Farmer management of secondary forests CIFOR/CATIE/UNALM Composition Floristica V s intensidad de uso de suelos Bosques secundarios en sistemas productivos de pequenos productores - investigaction participativa D Current/ V Cotan Caractization economica fmanciera de unidades agrarias (beneficios/costos) Future CIA T K A P surveys of yuca and beans Evaluacion genotipos de arroz para sistema barrizales Silva-pastoral Participatory design of integrated agro enterprise projects for selected portfolio Analysis of current marketing and distribution system CIFOR -Improving the value of residual forests- silvicultural options to regenerate RF- socioeconomic and policy research Gaps lmproving the value of residual forests - silvicultura! options to regenerate RF - socioeconomic and policies 5. Strengthening individuals, communities and institutions in NRM Recent CIA T Training for farmer participatory research and extension Current CIA T-Farrner participatory rice testing Strategies for integrating participatory and multi-institutional research in PucaJipa Human welfare, poverty and NRM environment and welfare University course on health and nutrition for UNU Dietary guidelines ICRAF- Training in NRM soils, genetic resources of trees 149 Future Fertility capability classification (ICRAF) Analysis ofNRM research policy (CIAT. ICRAF, CIFOR) Gaps Creative space for institutional change Collaborative research con socios y entre IARC's Graduate education for L.A> scientists Research at the community level Training on diversified forest management system Community development- awareness and structures Fortalicimineto de las organizations Analysis ofNRM research policy 6. Assuring and measuring impact Recent None Current CIA T -Impact indicators framework Community indicators of health Assessment ofhealth and nutritional status (food security) Key variables affecting malnutrition CIFOR- C sequestration on land use systems Future CIA T Methodologies for measuring impact of policy research Factors that influence adoption of methodologies ICRAF Valor de beneficios intangibles (e.g. Biodiversidad) Gas emissions FCC classification Increasing Research Effectiveness Vincular estudios de políticas y analysis de impacto (CIA T, CIFOR, ICRAF) Monitoring system for impact assessment of improved land use production systems CIFOR Gaps Factors that influence adoption of methodologies Valor de beneficios intangibles (biodiversidad etc) Appropriate methodologies for measuring of policy research (iii) Issues for discussion that arose from individual presentations: Research areas Vision for future land use Gaps General discussion themes Collaborative Studies In Marketing Study of indigenous knowledge on value of forests Where to focus research -Improving crops is not assisting in replacing the current system Strategy? From intensification of agriculture to emphasis on political economy, non-agricultura) sectors?) Research riverine as well as upland Ecological research on weeds 150 Information sbaring Joint researcb report List of projects Coordinate and list encuestos Data sharing and data base Sharing annual reports Is Pucallpa being integrated witb otber Forest Margin researcb? (iv) Summary of project proposals, collaboration and responsibilities lncreasing Research Effectiveness Tille: Identify elements for improving the va/ue of residual forests, secondary forests, and jallows Team: John Weber, Joyotee Smith, Doug White, Violeta Cotan (Facilitator), Carlos Ostertag, Cesar Sabagal (Leader), Julio Alegre, Ricardo Labarta, Tamsyn Murray, Sam Fujisaka Themes to consider including: i) Management of secondary forests in areas of socio-economics and genetics ii) Evaluation and management of fallow and the trade-off Possible components: Trade-offy policy. Technology and Genetics Tille: The sustainable use and conservation of biodiversity Team: John Weber (Leader), Glenn Hyman, Tamsyn Murray, Sam Fujisaka, Carlos Ostertag, Cesar Sabogal, Violeta Cotan, Joyotee Smith, Carmen (Facilitator) Themes to consider including: i) Use and conservation ofbiodiversity ii) Local conservation and knowledge The project would needs to define biodiversity and identify indicators for monitoring biodiversity in different land use systems Title: Identify factors that affect decision making by producers and policy makers and enable the process of cbange to demand-led researcb and development. Dean Holland (leader); Keneth Reategui (facilitator) Team: Dean Holland (Leader), Kenetb Reategui (Facilitator), Carlos Ostertag, Jobo A viles, Ricardo Labarta, Julio Alegre, Miguel, Violeta CoJan Tbemes to consider including: i) Strengthen inter-institutional capacity ii) Factors that affect adoption iii) Determine the best manner of collaboration Suggested that the project revise adoption in the past to obtain an idea of what methods function Contributors: ICRAF- Dale Bandy, Julio Alegre, John Weber, Polly Ericksen 151 lncreasing Research Effectiveness CIA T- Peter Kerridge, Sam Fujisaka, Dean Holland, Ricardo Labarta, Glen Hyman,Tamsyn Murray, Carlos Ostertag, Doug White, CIFOR- César Sabagol, Joyotee Smith, Violeta Colán Colán, DEP AM - Keneth Reátegui !52 lncreasing Research Effectiveness 4.2.4 Facilitating Community-based Natural Resource Management, Vietnam Objective: To provide technical and mentoring support toa University ofHue team carrying out research to improve food security and ensure long-term stability of natural resources of an ethnic community in the mountainous region of Thua Tien Hue Province, Vietnam. Rationale: CIA T was invited by IDRC and the University of Hue to collaborate in a project on community management of natural resources through providing improved gennplasm, technical input, and training and advice in problem diagnosis, setting research priorities, interpreting data and reviewing research. Output: Specific inputs by CIA T have been: • Reviewing and editing the research proposal • Providing improved germplasm of cassava, forages and beans • Technical input into research design of agronomic experiments • Problem diagnosis and data interpretation • Research planning • Coaching in the use oftools for diagnosis and analysis of data Impact: The University of Hue team has established good rapport with the community and completed characterization ofthe physical and agricultural (forestry, livestock, fish, crops and home gardens) resources, cultural characteristics, social and gender issues, economics, and the institutional setting. lnterventions through participatory research of rice varieties, vegetables and pulses and smalllivestock show promise of increasing productivity. lt is hard to assess the impact of the role of CIA T in the project at this stage. CIA T sees this as a collaborative research venture rather than a consultancy and is gaining experience as well as contributing from our own resources and experience. We see community-based natural resource management as a strength that CIA T can develop and/or support through the skills it has in upland agro no m y in Asia, use of participatory approaches in research, and experience in agro-enterprise developments. Contributors: P. Kerridge, S. Fujisaka, R. Howeler, P. Horne, C. Wheatley (CIA T). The Hue team is listed with Output 1.1.2. 153 lncreasing Research Effectiveness Activity 4.3 Facilitate regional partnerships/ networks 4.3.1 Coordination meetings of Forage for Smallholders Project Highlights • Plans were discussed to synthesize findings at an intemational workshop and to publish them in different languages of the region • A strategy was discussed for continuation of FSP activities with reduced donor support Purpose: Review achievements and plan activities of the project. Rationale: Regional meetings bringing together majar partners in research are a vital part of effective coordination of multilateral research. Outputs: The fourth annual meeting ofthe FSP was held in Nha Trang, Vietnam from 26-28 January 1999 and attended by 33 participants from 9 countries (Australia, China, Colombia, Indonesia, Laos, Malaysia, Philippines, Thailand and Vietnam). The meeting reviewed experiences with on-farm forage development and discussed options for continuing the work at each site after December 1999 when the project concludes. Topics discussed during the two days of presentations and discussions included forage technology development, new forages for particular uses, assessing impact offorage technologies and future directions for the FSP. The participants visited the nearby field site at M' Drak, where more than 90 farmers are planting forages to provide supplementary feed for their cattle at times of year wben feed is scarce or labor is in short supply. lmpact: Tbe regional meetings have proven to be an effective way of exchanging informatíon between scientists in the region more so than through the Newsletter and electronic means. Contributors: FSP staff and national partners from 7 countries. Coordinators: Wemer Stur (CIA T) and Peter Home (CSIRO) 4.3.2 The Asían Cassava Research Network Purpose: To improve the efficiency of cassava research conducted in Asia for the benefit of cassava farmers in the regían. Rationale: Cassava programs in national research institutions in Asia are often small and have very limited budgets for research. By exchanging cassava germplasm and information, and by coordinating the research to be conducted, the efficiency ofthese programs can be markedly improved. Moreover, by stimulating more collaboration between research and extension institutions, the research becomes more focussed on practical problems while the quality of extension improves. Methods: The Cassava Advisory Committee, consisting of one representative from each majar cassava producing country in Asia, with the CIA T Regional Coordinator acting as secretary, decides about the country that will host the next Regional Cassava Workshop and the tapies to be discussed. These workshops have been held every three years since 1984. The 6th Workshop will be held in Ho Chi Minh city, Vietnam, in Feb 2000. The papers presented at these workshops are edited and then printed as Workshop Proceedings. The 5th Proceedings was printed in 1998. !54 lncreasing Research Effectiveness Outputs: During the past ten years the Thai-CIA T cassava breeding program has supplied the breeding programs of other countries with sexual seed. advanced lines or clones in the forrn of tissue culture or occasionally as stakes; this, in addition to the sexual seeds supplied by CIA T/Colombia. After passing through many stages of selection, these seeds have yielded many promising clones, wbich were eventually released as new varieties by national programs. In 1997, new higher-yielding varieties were planted in 660,000 ha in Thailand, corresponding to 63% ofthe cassava growing area. For the whole of Asia, it is estimated that the area under new varieties will surpass 1 million ha in 1999. In the 2nd phase ofthe Nippon Foundation Project on Sustainable Cropping Systems, which started in early 1999, the emphasis will be oo developing and then disseminating more effective production practices that are acceptable to farmers, through farmer participatory research and dissemination. The project will concentrate oo Thailand and Vietnam. In both countries workshops have been organized to plan future activities and to coordinate the work of six collaborating research and extension institutes and universities in Vietnam and six in Thailand; in the latter country ffiSRAM is also a partner. The project is not only strengthening the collaboration among various research and extension organizations within each country, but also'a vertical collaboratioñ among researchers/extensionists at various levels of administration, from national down to the locallevel. Coordinator: R. Howeler (CIA T) 4.3.3. Tropileche - Workshops and meetings to analyze and plan research activities Highlights • The workshop held in Peru in July 1999 established the research agenda for the forest margins for the next two years and set the principies for expanding the collaboration with other partners in Ecuador and Bolivia. Annual Workshops 1999 Rationale: Annual meetings to discuss workplans, current research and future challenges as well as constraints are important in order to in crease the efficiency of research and interchange ideas. This year separate workshops have been organized for South America (July 1999 in Peru) and Central America and the Caribbean (Costa Rica in October 1999). Results: Tropileche held a workshop to plan and discuss present and future activities in South America during June 27 to July 2, 1999, in Moyobamba, Peru. The objectives ofthe workshop were to: (a) present the research achievements obtained by the Consortium and pose future challenges; (b) present research results achieved in Peru and to discuss new activities for 1999 and 2000; (e) participate in a field visit to understand and identify opportunities in current animal production systems in the Moyobamba region of the Peruvian Amazon; ( d) review strategic and participatory research based on needs and constraints; and (e) ana\yze and discuss new forrns of collaboration with other institutions and in other countries of South America, especially Ecuador, Bolivia and Brazil. lnvited participants in the workshop included 22 researchers from Peru, Colombia, Ecuador, Bolivia, and Brazil. The workshop proceedings will be distributed during the month of October. Mayor outcomes of the workshop were: ( 1) Tropi1eche will continue to reduce research activities in Pucatlpa in 1999 and 2000 and in crease activities in Moyobamba. Potential for adoption of new forages activities in Pucallpa is 1ow due to an oversupply offorage from a reduced herd inventory. In addition, the dual-purpose herd has a low 155 Increasing Research Effectiveness genetic potential for milk yield which reduces economic benefits from investing in new forage technologies and the market potential for milk in Pucallpa is severally limited as there is no milk plant. The situation in Moyobamba is the reverse. A cooperative milk processing plant has been opened, farmers are improving the genetic potential of their herds for milk production and there is a demand by farmers for improved feeding systems. (2) The Consejo Transitorio Agrícola Regional (CT AR), through the Fundacion para el Desarrollo Agrícola del Alto Mayo (FUNDAAM) will be our partner in Moyobamba and the institution responsible to carry out all of the research activities agreed duiing the workshop. The CT AR also agreed to invest as matching funds 35% ofthe resources that Tropileche delivers to Moyobamba. (3) There are good possibilities to expand research activities in Ecuador with our partner INIAP by accessing Wold Bank funds bilaterally and in Bolivia with CIAT-Santa Cruz through a collaborative project with DFID. Tropileche agreed with both INIAP and CIAT-Santa Cruz to follow up and develop project proposals for submission to both WB and DFID to access bilateral funds. Collaborators: F. Holmann, C. Lascano. P. Kerridge andA. Ramirez, CIAT, Colombia. !56 lncreasing Research Effectiveness Activity 4.4 Compare effectiveness of different institutional models for effecting change in natural resource management 4.4.1 Participatory systems research: towards the future Highlights • Compilation and publication ofthe book "Systems and Farmer Participatory Research: Developments in Research on natural Res o urce Management" Purpose: The book records presentations at an interna! workshop held at CIA T in November 1997 to review the different facets of systems and participatory research conducted in CIA T. Rationale: Different systems and participatory and systems research had been conducted without an analysis of whafhad been accomplished, what directions were being taken, and what directions appear to be the most appropriate for the future. Methods: Researchers met in a workshop in November 1997 to exchange experiences and ideas. Each participant or team of partícipants were then invited to submit papers for review and inclusion in the book. Considerable follow-up work with authors and researchers led to the publication of the book. Outputs: A concluding chapter synthesized experiences and future directions which is reported here. The general theme through the book is how agricultura! research ís striving to make problem-solving more effective and efficient through integrating germplasm and natural resource management (NRM) research, developing and applying farmer participatory research (FPR) approaches, and establishing new forms of partnerships for research. The chapters pro vide a clear view of what has been achieved and reasons for successes and failures. Expected future impacts are implicit in much ofthe work, as are strategies needed to achieve them. lntegration of germplasm andNRM research. The projects described indicate substantial progress in the integration of germplasm and NRM research. Researchers are introducing new forage materials, especially legumes, and small-scale farmers in Asia and Africa are testing and using them as green manure, cover crops, and in improved fallow rotations. These uses of forages in low-input systems seek to irnprove nutrient cycling, maintain soil structure, combat weeds, and reduce soil erosion. In the Latín American savannas and forest margins, researchers are working with farmers and ranchers to develop animal feeding systems that are both more productive and more sustainable through the introduction and farmer testing ofnew forage mixtures. Work on integrated crop-livestock systems based largely on new multipurpose legumes in the savannas has sought to solve problems assocíated with low soil fertility, pasture degradation, soil compaction and eros ion, build-up of pests and weeds, and water contamination. FPR in cassava-based systems in Latín Arnerica and Asia has successfully combined the testing of new cassava varieties with soil erosíon control and integrated crop and pest management. Successes have been based on: • Research commitment to integrating germplasm and NRM research and to form needed multídiscíplinary teams. • Sufficient resources to support long-term germplasm and strategic research. • Availability of seed or planting materials of new varieties or cultivars identified as potentially appropriate for farmers' systems. • The natural willingness of farmers to test new varieties. !57 Increasing Research Effectiveness Making new varieties available is often an incentive to participate in the testing ofNRM innovations. Also implicit in the research presented is that successful systems research is focussed and conducted in carefully selected appropriate benchmark or reference sites in targeted agroecosystems. The agroecosystems that CIA T targets are the forest margins, Latin American hillsides, and Latin American savannas. Although research is conducted at levels ranging from farm to agroecosystem, much of the fieldwork is done at the watershed level. Failures are less clear. CIA T's commitment to integrated germplasm-NRM research (i.e., to systems research) has contributed directly to the shaping of the systems projects described in this volume. The papers show how the NARS partners were conducting systems research in these same projects, but do not indicate if changes of approach from separate research on germplasm/commodity and NRM to integrated research was taking place in NARS as a whole. The irnplicit, desirable irnpact for the future would be the NARS' adoption of systems approaches; and the similarly implicit strategy would be to continue and strengthen partnership research and demonstrate the successes of systems research approaches. Participatory Research. The chapters in the book reveal a range of participatory approaches and show the development such approaches over time. The FPR approaches include: • Selected farmers individually conduct trials and provide feedback to researchers. Treatments are largely researcher-designed. In the case of the Colombian forest margins, this initial researcher-led FPR led to other farmer-initiated experimentation. • Individual farmers test researcher-developed innovations ( e.g., for soil eros ion control). At the same time, researchers in Cauca, Colombia, tried to make the adoption of soil eros ion control measures more economically attractive by assisting in the commercialization of new products resulting from adoption. Cassava researchers in Asia supplied innovations more immedíately attractive to farmers (new varieties and management options) for testing as a way to generate interest in soil conservation. • A committee for local agricultura} research (CIAL in Spanish) selected by and representing members of a given community, conducts FPR, based on participatory diagnosis. Field days are held to allow wider participation in evaluating results and planning for further research. Work with cassava farmers in Brazil and with small-scale hillside farmers in Colombia demonstrates the effectiveness of this approach. • Researchers provide new materials and information to farmers, who then individually test and adapt innovations to their particular needs and situations. Forage research in Asia and farmers' independent experiments with green manure and cover crops in Uganda have Jed to systems-level innovations unforeseen by researchers. • FPR is extended to innovations in the local institutional structure to further empower local communities and to allow communities to find new, needed ways to interact with each other and with outside agencies. Organizing discussions between upper and lower watershed users in Cauca, Colombia, allowed a change from conflicting resource-use interests to negotiation and working toward mutual benefits. Successful initiation and organizing of small-farmer seed production in Uganda has led to recognizing the need for further institutional change in the seed sector. The chapters in the book document severa! types of success in FPR. First, as the above indicates, researchers working with farmers steadily made innovations in the FPR approach itself. They improved methods and, overall, expanded the idea ofparticipation from feedback from farmer-managed trials to the incorporation of farmers in problem diagnosis, selection of research problems, and community-based evaluation of results, leading to local choice in the direction of further experimentation. The 158 Increasing Research Effectiveness encouragement and documentation of fanners' independent experiments based around introduced "unfmished" technology components reflects a growing respect for fanner traditional knowledge, fanners' natural tendency to experiment to solve problems, and the idea that only fanners can adapt innovations to particular local conditions. Finally, participation has started to encompass institutional change and local empowerment. The second success and impact of FPR has been increased adoption of the approach by partner NARS. The work in Brazil, conducted by a wide range ofNARS, began with a great deal of skepticism and ended with NARS changing policy in favor of using FPR in most, if not all, on-fann research. The research networks and consortia functioning in Asia are expressly working to institutionalize FPR and systems approaches throughout the region. Much of the effort to date has been both in the conduct of field projects and in training (oftrainers, extension workers, researchers, and, importantly, NARS decision makers). The development of committees for local agricultura! research in Colombia has led to both projects and NARS in severa! countries of Latin America adopting the approach. Third, the different fonns of FPR are clearly effective and efficient in solving agricultura! problems in an integrated and locally appropriate fashion; and such effectiveness is starting to have impact through adoption of technologies and the benefits brought about by such adoption. The adaptation and adoption of new crops, varieties, legume-based improved NRM at the local leve!, pest and crop management practices, and livestock feeding systems are described in the various chapters. Fourth, FPR appears to lead to greater fanner confidence in their own abilities, first in the conduct of problem-solving research, and later in new organizational and management skills. Recognition and encouragement of fanners' experiments, organized or not, has led to greater fanner-researcher interaction and increases in research efficiency. The various papers have also identified constraints to effective FPR. Obviously, in many cases, researchers need to pro vide sound, appropriate technical alternatives. Examples are cassava varieties that are actually superior to local varieties in Brazil, Arachis cultivars suited to poorer soils in the Colombian forest margins, or contour hedgerow species such as broom grass and lemon grass, which can bring real economic benefits. Once alternatives are identified as viable and valuable, planting materials often need to be made available. Lack of cassava planting materials in Asia and Iack of Iegume seed in many projects has slowed the rate of FPR work at a time when fanner enthusiasm was relatively high. For projects having a predetennined set oftechnology innovations (e.g., soil erosion control measures, Iegume cover crops, and green manure ), si tes and participating fanners must be carefully selected. Those more likely to participate are farmers suffering from, recognizing, and trying to soive the problem to be addressed (i.e., soil erosion or a need for improved nutrient cycling). Real and effective participation does not occur if fanners are not interested in the given problem. As a variation on the need to properly select si tes and participants, the work in the Colombian forest margins encountered the interesting case of lack of effective participation because farm managers were not owners; the owners being largely absentee. NARS and local development or extension workers must have sufficient interest, training, commitment, and resources to work with fanners . They need to make field visits, provide (at times) fanner training, initially help organize local agricultura! research groups or support individual experiments, and help farmers with evaluation and further planning. Lack of financia! and institutional support can grind FPR projects to a halt. Projects "tied" to particular commodities and/or management innovations may be unable or unprepared to work with farmers on their more pressing problems and needs. Such a limitation may result in farmers' 159 lncreasing Research Effectiveness lack of interest in participating in the project. A solution has been to increase the web and reach of participating institutions. The work in the hillsides of Cauca, Colombia, organized all interested externa! agencies to coordinare efforts and resources to better address the range of local needs and problems. Agricultura! research appears to be at an intennediate stage in tenns of impact. FPR methods are being developed and tested. FPR is being integrated into the work of partner NARS. Fanners in project communities are empowered both to experiment and to deal more effectively with outside agencies. And new technical and organizational innovations are being adopted at project sites. The most important impact, however, is in the future: wider use of FPR beyond project areas, wider adaptation and adoption of technical altematives developed, and local benefits derived by participating fanners and society as a whole. The strategy for reaching such irnpact is to continue projects that lead to strengthening the integration of FPR approaches in NARS; and then, and with more difficulty, to find ways of achieving wider impact with less initial support. That is, the important question remains as to how to achieve wider impact with less initial project support and "hand holding". The move toward devolution ofpower and responsibilities to local communities in many areas of the world, and as a philosophy of increasing numbers of development and donar agencies, may be the needed step in the right direction. New jorms oj team and partnership research. Each of the projects described in the book reflects interdisciplinary teamwork and partnerships with NARS, NGOs, and others. Teams ranged from collaborations of agronomists and geographers/GIS specialists to teams comprised of entomologists, plant breeders, pathologists, social scientists, soil scientists, livestock and forage specialists, and agronomists. Projects have ranged from intemational center collaboration with a majar NARS, to partnerships among intemational centers, various NARS, universities (both local and intemational), NGOs, and local organizations. CIA T researchers in Asia, Africa, and Latín America have also helped to propase, seek funding for, and develop research networks and consortia intended to work on systems, employ FPR methods, and promote both intemational center-NARS partnerships and effective partnerships among participating countries. Partnership research reflects the fact that CIA T and its partners ha ve compara ti ve advantages in different complementary areas, and that partnerships are integral to promoting the sustainability of new ways of conducting research by the NARS. As in the use of systems approaches, the projects described in the book were designed as team and partnership based activities. As such, the papers did not address the lack of such partnerships in other projects. Certainly, however, funding for intemational agricultura! research now supports new and more inclusive partnership relationships. Support has dried up for the lone intemational center scientist conducting independent research. Support is lessening for bilateral center-NARS agreements, but is increasing for proposals for partnerships among intemational centers, the various NARS (e.g., ministries and departments of agriculture, forestry, and natural resources rather than only agriculture), NGOs, and local community groups. The story unfolded in the different chapters indicates that progress is being made and that, in tenns of strategy, although progress may have initially been rather "donar driven", current steps forward are being taken because of the demonstrated benefits. Finally, agricultura! research appears to have reached mid-stream, but is still heading in the right direction. Gennplasm and NRM research will continue to be conducted separately where appropriate ( e.g., in generating new breeding lines or in monitoring the effects of current land uses on greenhouse gas emissions or biodiversity). In helping to solve problems where small-scale fanners want increases in production and in systems' sustainability, and where society also has desired outcomes, the integration of gennplasm and NRM research is natural, appropriate, even obvious. A crucial future goal of systems research (reaching the far side ofthe stream) will be to develop effective ways of dealing with trade-off between local or farrner goals for increased productivity and societal goals of maintaining the global environment intact. 160 lncreasing Research Effectiveness Although not addressed in the book. CIA T' s work on systems also includes analyses of policies that ha ve i) contributed to current resource use parteros at different scales, affected farmer adoption of different types of innovations, and 2) sought a balance between increased productivity and environmental protection. A goal of such research is to be able to inform policymakers as to the effects of policy on NRM decisions at various levels. The projects ha ve shown a trend of improvements in FPR approaches. The challenge will be to make the participation of farmers and local communities more self-sufficient and self-sustaining. While local empowerment in its various forms appears to be the model for the future, considerable "political will" is needed to facilitate wide-reaching genuine change. In forming effective teams and partnerships, however, agricultural research seems to be nearing the far side ofthe stream. New partnerships and new forms of funding are emerging; at frrst only considered acceptable but now desirable and effective by participating re.searchers and institutions. A future volume on this topic will hopefully demonstrate that with few missteps the goals identified here will have been reached and what has already been achieved will have been consolidated and furthered. Contributors: S. Fujisaka and authors. 161 lncreasing Research Effectiveness Activity 4.5 Facilita/e a policy enabling environment that ensures adoption of appropriate policies and technologies 4.5.1 Institutionalizing a Collaborative Research Approach in Pucallpa through the DEP AM project Highlights • DEPAM is developing a multi-level approach to institutionalizing collaborative research among institutions in Pucallpa. Purpose: To investigate structures and processes that allow institutionalization of multi-institutional and participatory research approaches among national and international research centers in Pucallpa. Rationale: Diagnoses with researchers in volved in DEP AM in Pucallpa ha ve found that institutional policies, structures, and freedom often presenta barrier to researchers' abilities to respond flexibly to the needs of farmers and of other institutions. Participatory and multi-institutional research will remain a fringe activity until it is understood and integrated as a valued approach into organizational structures and log-frames. This section focuses on the support and changes to research management. Output 4.2.1 "Multi- institutional research in the Aguaytia watershed through the DEP AM project, Pucallpa, Peru" focuses on the parallel field-level research activities associated with multi-institutional research and Output 2.1.1 describes that research. Methods: The organization ofDEPAM provides a space to test new institutional arrangements for supporting participatory and multi-institutional research, and to investigate the effect of these changes on participating institutions. DEP AM introduces severa! innovations to research management in Pucallpa: • A multi-institutional research team designed to coordinate activities. • A farmer committee to evaluate project proposals and monitor and evaluate ongoing projects • A multi-institutional advisory committee to oversee scientific rigor of activities. • Development of a Rural Agroenterprise unit which can help create the market and business links necessary to take advantage of agricultura! research • Learning resources and a supportive learning environment Outputs: • The DEP AM advisory committee and the farmer committee reviewed 18 project proposals that were submitted following a second call for DEPAM small projects in April1999. • Unity and skills in the farmer committee were strengthened by the committee selecting one of its members to attend a course "participatory plant breeding" in Quito, Ecuador. • There is increased awareness of participatory approaches among researchers and research managers, and managers actively support the involvement of their researchers in the DEPAM participatory research team. Impact: Researchers and research managers are beginning to discuss the effects of the processes, and organization ofresearch on its record offield-level success. Three ofthe 11 projects in DEPAM have substantially changed their focus or organization in response to diagnoses with farmers or with 162 lncreasing Research Effectiveness participating institutions. The field-level effects of such changes will be measured in the following year. This is an embryonic learning system. There is increased support for participatory processes by heads of institutions. Three institution heads have asked for their own course on participatory research (curren ti y in planning) and sorne ha ve wanted to dedícate 7 days to join the course in participatory research for researchers. The courses for researchers and research managers will culminate with a joint day to openly discuss how to integrate multi-institutional and participatory approaches into the institutional framework of organizations in Pucallpa. The system of competitive research grants for sub-projects has been successful in providing DEP AM with an entry-point with local institutions, and many of the sub-projects show great improvement in their incorporation of participatory research. A key indicator is that the projects themselves are able to report their difficulties with participatory methods, and are analyzing their needs and actively requesting assistance both from each other and from DEPAM. However, the system ofresearch grants that was ~sed asan incentive-to participate has also made sorne aspects of inter-institutional collaboration difficult. National institutions have had different degrees of success in integrating the "project-focused" structure of DEP AM with their existing "institution-focused" structure. They have their own research agendas, often set up within a national rather than local context and only a limited amount of resources to carry them out. The DEP AM project members ha ve themselves identified remaining challenges in the limited freedom of many Ucayali research institutions from national mandates, and with a historical non-transparent, personal focus to project management. Again it has been questioned whether research grants should be open to any agenda or only to specifc demand-led agendas. The Farmers Committee has been effective in setting strategic directions for DEPAM, and in advising in the selection of research projects. However, they represent a large area, and have few links with the farmers who are involved in the DEP AM research projects, agreeing that they cannot adequately x.:present these farmers . DEP AM is looking for mechanisms to strengthen the committee, possibly by inclusion of farmer representatives from all DEP AM projects, and supporting their role to evaluate the content and processes ofthe projects. Contributors: Dean Holland, Sam Fujisaka, Peter Kerridge, Douglas White, Glenn Hyman, Scott Cechi, Carlos Bruzone, Carlos Ostertag, Mark Lundy, Ann Braun, Maria Femandez in CIA T projects PE-S, PE- 4, IP-1 , SN-1 , SN-3, and the Inter-Center Participatory Research and Gender Analysis Program. 163 Increasing Research Effectiveness Activity 4. 6 Develop training approaches and materials on targeting, development and diffusion of new technologies and provide training for partners 4.6.1 Approaches to targeting and developing new forage technologies Highlights • Effective participatory approaches have been implemented and adapted in four Southeast Asian countries. 170 national partners have been trained in these approaches since July 1998. Purpose: To improve the effectiveness ofadaptation and adoption offorage technologies by smallholder farmers in Southeast Asia Rationale: Until recently, the approaches used for developing agricultura! technologies on smallholder farms were too prescriptive. The role of research organizations was seen to be to provide technology packages toan extension service, which then delivered them to 'model' farmers with an expectation that the technology would spread to other farmers. Most of this process was controlled or driven by researchers and policy makers with little or no input from farmers . The technologies delivered in this manner were often not appropriate for many of the smallholder farmers and adoption was poor. Improving the adaptation and adoption of agricultura! technologies on smallholder farms required a change in approach to involve farmers more actively in the process oftechnology development. Outputs: The Participatory Technology Development (PTD) approach used by the Forages for Smallholders Project (FSP) in Southeast Asia has been synthesized from similar approaches developed by CIAT. The approach uses many ofthe principies ofParticipatory Rural Appraisal, but extends the involvement of farmers well beyond the initial stage of appraisal through to forage technology development and evaluation on farms (Figure 1). Adaptation & .... _ __ Adoption \ \ \ ' ........... _ --- --- --------------------~ Active Farmer Participation Test forage technologies ----)la.,._ Participatory technology development ----------)lo- Feedback Figure l. The key stages ofParticipatory Technology Development used by the FSP. 164 Increasing Research Effectiveness PTD begins with diagnosis of livestock-related problems by the fanning community, in partnership with field workers from the FSP. Commonly, a problem tree is developed and discussed to help the community define, group and prioritize their main problems. Potential solutions to these problems are then discussed and plans made about how to test and evaluate promising forage technologies. As fanners experiment with these, their experiences are monitored and assessed so that changes can be made to the technologies being evaluated as necessary. The key principie of this process is active, decision-making involvement of fanners at all stages of forage technology development, with technical input and facilitation by government staff. Other important factors that have contributed to successful implementation in the field are: (i) Careful selection of sites and farmers A small quantity of seed distributed to fanners with a gen u in e need for forages will be far more successful than large quantities distributed where there is no need. The FSP has developed criteria for site and fanner selection to ~dentify sites and fanners with the greatest potential to benefit from forages. (ii) Active local partners To work actively with fanners requires considerable commitment from local development workers. A promising si te with enthusiastic fanners can fail if the local development worker is not supportive. (iii) Regular visits to farmers Visiting fanners regularly and providing informal training during the early stages ofPTD is crucial. Working with a small number of enthusiastic fanners whom you can support is usually more successful than working with many fanners with little contact. (iv) Close collaboration with 'farmer advocates ' Sorne fanners are especially active in experimenting with and championing locally-successful forage techno logies. Identifying these 'farmer advocates' and encouraging them in the process of forage development can result in rapid adoption and spread of forage technologies. (v) Linking with national partners and other development agencies Working in partnership with national government and non-government organizations has been essential to achieve sustainable technology development. (vi). On-going development of participatory skills Training courses sensitize development workers to the ' tools ' ofPTD but the essential skills of communicating with farmers can only be gained with time and field experience. Partners in the FSP from all countries have regularly met to share skills, experiences and ideas. PTD is a learning process for all involved. Locally-successful technologies are a result of this process and cannot be easily transferred to other areas without the new fanners also going through a process of evaluating and adapting them. The challenge now is to develop methods of expanding locally-successful forage technologies to wider areas but within a participatory framework. Impact: The new participatory approach has been implemented at 18 sites in four Southeast Asian countries (Indonesia. Laos, Philippines and Vietnam). Since July 1998, 10 training courses have helped 1 70 national partners upgrade their skills in participatory technology development with forages using the training manual "Developing forage technologies with fanners" developed by the FSP. Regional interest in the experiences of the FSP with PTD has resulted in a grant from A CIAR to produce three new books for publication in five languages: 165 lncreasing Researcb Effectiveness • Developing forages with smallholder farmers - how to select the best varieties to offer fanners • Developing forages with smallholder farmers - how to grow, manage and use forages • Developing agricultura! solutions with smallholder fanners - participatory approaches for getting it right the first time These books are in preparation and will be published over the next 12 months. Contributors: Wemer Stür, Francisco Gabunada and Louie Orencia (CIAT, FSP Philippines); Peter Home and Phonepaseuth Phengsavanh (CSIRO, FSP Laos); Viengsavanh Phimphachanhvongsod (NAFRI, Laos); Maimunah Tuhulele (DGLS, Indonesia); Ibrahim (Livestock Service, East K.alimantan, Indonesia); Tatang Ibrahim (BPTP, North Sumatra, Indonesia); Ed Magboo (PCARRD, Philippines); Willie Nacalaban (LGU, Malitbog, Philippines); Perla Asís, LGU, Cagayan de Oro, Philippines); Le Hoa Binh (NIAH, Vietnam), Le Van An (UAF, .Hue, Vietnam), Bui Xuan An (UAF, Ho Chi Minh City, Vietnam), Bryan Hacker (CSIRO Tropical Agriculture, Brisbane, Australia), Peter Kerridge (CIA T, Cali, Colombia). 4.6.2 Training provided on use of economic model in Peru Highlights • The second multi-institution workshop provided feedback on model operation and its parameters. Purpose: To facilitate more interaction amongst Pucallpa-region economists, agronomists and extensionists vía a fanning system model. Rationale: Slash-and-bum farming practices are heterogeneous and complex land use systems. Fanners produce a mix of animals, agriculture crops, tree and forest products. Furthermore crops are annual, semi- perennial or perennial in monocultures or multiple crop associations. Tbese many facets of farming systems provides a challenging environment in which to research the implications of agronomic and policy research. Outputs: After the development of the initial agro-economic model two further steps are necessary, refining of the model with poten tia! users, and training of its use among different local institutions. During a second workshop the model received comments about its potential use ofthe too!, the type of technologies that national institutions want to include and ideas on how to make it more user-friendly. Another workshop is being organized to test the addition of different agriculture technology altematives developed by institutions and to refine parameters used the model parameter according to farmers and researcher experiences. lmpact: The next workshop will presenta deeper explanation ofthe model and its underlying assumptions. It will also include instruction on how organize data before it can be incorporation into the model. A sensitivity analyses will include opportunities for participants to build real-world scenarios and to adjust key parameters based upon their own experiences. The model has have received keen interest from within the Ministry of Agriculture and will be employed by extensionists who are in close contact with farmers. In this manner, the potential impact of agronomic advances can be discemed before they are incorporated in farmer fields. Contributors Douglas White, Ricardo Labarta, SN-1 , INIA, MinAg-Ucayali, DEP AM, ICRAF 166 Increasing Research Effectiveness 4.6.3 Farmer-experimentation mini-kits as a tool for promotion of legume technology Highlights • Mini-kits were distributed to farmers to enable leaming experiences about alternative green manure and cover crops. Purpose: To devise a means of efficiently adapting cover crop (CC) technology to the diverse circumstances of large numbers of farmers, and enabling farmer-to-farmer dissemination at low cost. Rationale: Cover crop technology is difficult to promote because: benefits only appear over time; adjustments of cropping systems are generally needed; technical options are environrnent specific and identification of appropriate niches takes time; farmers ha ve little understanding of the potential of cover crops in soil management; and the extension services generally lack the capacity to promote such technology. Methods: For the l999A season, 2000 kits were prepared in collaboration with the IDEA project for distribution to farmers in six districts of U ganda. Another 1 000 kits were prepared for the second season of 1999 and distributed through various NGOs for use in eight districts. The kits contained seed of canavalia, mucuna and sometimes tephrosia and/or climbing beans. They also contained leaflets on canavalia, mucuna, tephrosia for mole rat control, a listing of best bet options for mid-altitude areas of Uganda, and a decision guide to the use of four legume species. Outcome: The effectiveness of the approach has not yet been evaluated. lmpact: More capable farmers, with an inclination to innovation, will have successfulleaming experiences through these kits and will be able to integrate appropriate legumes into their systems. Eventually, we expect more farmers to leam from their neighbors. Collaborators: Charles Wortmann and Mark Wood, IDEA (Investment for the Development ofExport Agriculture) 167 Increasing Research Effectiveness Activity 4. 7 Communicate resu/ts through networks, workshops and journals 4.7.1 Communication activities of the Forage for Smallholders Project Outputs: News/etter. SEAFRAD, the Southeast Asian Forages and Feed Resources Research and Oevelopment Network is a network of researchers and development workers who are working with forages. The secretariat of SEAFRAD and editorship of the SEAFRAD newsletter rotate between member countries on an annual basis. In 1998, it was based in the Oirectorate General of Livestock Services, Indonesia. In 1999, it moved to the National Institute of Animal Husbandry, Vietnam. Each year, two issues ofthe SEAFRAD newsletter are produced and distributed. SEAFRAD has been sponsored by the Forages for Smallholders Project from 1995-1999. This sponsorship will be continued by ADB from 2000-2003. Books: In 1998/99, the Forages for Smallholders Project published an Indonesian and Vietnamese versions ofthe 'Field experiments with forages and crops: practical tips for getting it right the first time' manual by Cheng, Y. and Home, P. Workshops: The progress offorage technology development in the Philippines and Indonesia was reviewed at workshops held in January and April 1999. This workshop brought together researchers and development workers involved in participatory forage technology development, reviewing progress, sharing experiences and planning activities at FSP sites. In Lao POR, a Workshop "Oeveloping forage technologies with farmers: putting plans into action in northem Lao POR" was held in Luang Phabang on 15 and 16 July 1999, which communicated the results ofthe FSP toa wide audience. The Workshop was attended by 64 participants from the National Agriculture and Forestry Research Institute (NAFRI), Provincial Agriculture and Livestock Offices, Oistrict Agriculture Offices, Govemment-funded rural development programs, Lao Womens Union, Department ofLivestock and Fisheries, and foreign-funded projects. Contributors: Wemer Stür (CIA T, FSP Philippines) and Peter Home (CSIRO, FSP Laos). 4.7.2 Communication of results within the Asian Cassava Research Network Outputs: • Ouring 1999 the papers presented at the Intemational Symposium on Cassava, Starch and Starch Oerivatives, held in Nanning, Guangxi, China, in Nov 1996, were edited and published as Workshop Proceedings. These will be widely distributed to libraries, researchers and the private sector involved in starch processing. • A chapter entitled "Cassava Nutrition and Fertilization" was written as part of a book on cassava to be published by CABI. • A desk study on ''the Effect of Small-scale Cassava Production and Processing on the Environment" commissioned by IF AD in Rome, was coordinated by CIA T and written in collaboration with CIP, liTA, NRl and CIRAD. This will be published and presented at the Global Cassava Forum, to be held in late 1999 or early 2000. Cootributor: R. Howeler (CIA T) 168 Increasing Research Effectiveness 4.7. 3 Tropileche- Dissemination of research results Highlights • The database developed on the web with research results on duaJ-purpose cattle offers researchers a new tool to access infonnation • The production of a videotape about a small producer who adopted new feeding strategies is a practica! way to show producers and extension agents how small farmers can succeed in intensifying production Outputs: Tropileche Newsletter The Consortium has published six newsletters and the seventh ís in print. Publication dates are March and October. The objective of this newsletter is to infonn about the activities of the Consortium, on-goíng research trials, research results being produced at the different benchmark sites, and any other news our partners consider useful to inforrn. These newsletters can be obtained free of charge through the Tropileche HomePage on the Internet (http:/www.ciat.cgiar.org/tropileche/start.htm). Contributors: F. Holmann, CIA T -ILRI, Colombia, and national institutions in Peru, Costa Rica, Nicaragua, and Honduras Data base on research results from dual-purpose cattle In 1996 Tropileche developed a database ofresearch results on dual-purpose cattle systems in tropical Latín America from the year 1960. Themes include nutrition and feeding, forages (grasses and legumes), genetic improvement and reproduction, animal health, economics, and extension, transfer, and adoption of technology. This is maintained. At present there are more than 2,200 references and approximately 100 items are added each month. All references include basic descriptors and 70% of them al so include an abstract. This database was developed in micro CDIISIS and follows the normative ofthe information system AGRIS-CARIS from F AO. This database is available through the Tropileche HomePage on the Internet (http:/www.ciat.cgiar.org/tropileche/start.htm).The average number of users consulting the database during the last year was 4.3 per day. Contributors: A. Medina and F. Holmann, CIA T-ILRI, Colombia Tropileche on Internet The Tropileche Consortium has developed its own HomePage on the Web, which contains the newsletters that have been produced as well as the data base containing research results generated in tropical Latín America. This HomePage can be accessed through the CIA T HomePage (hnph .. ww.ciat.cgiar.or!!ltropileche/start.htm). In addition, this HomePage has a list of researchers with affinities in research on dual purpose cattle in LAC, with contact addresses. Researchers can access the database from anywhere in the world, request information, and communicate and interact with other colleagues. Contributors: F. Holmann andA. Medina, CIA T-ILRI, Colombia 169 lncreasing Research Effectiveness Videotape Production Tropileche in collaboration with the Department ofCommunications ofthe Ministry of Agriculture of Costa Rica developed and produced a 11-min videotape. This is a case study of evolution ofthe feeding system on a dual-purpose fann located in the dry tropics of the Central Pacific region of Costa Rica. This will be made available to extension workers and loaned to fanners and others on request. Antonio Lopez is a small producer who has adopted many of the technologies that the Tropileche consortium has developed in association with the Ministry of Agriculture. Antonio is currently producing more milk on less area, has doubled bis family income, and has released areas from the livestock enterprise to serve as protected areas for timber production and protection of water sources. This videotape will be used to show other producers in Costa Rica and Latin America how one small fanner succeeded in intensifying production with new forage-based technologies that were developed by him with the technical assistance ofMAG ~d Tropileche anda donation of seed for evaluation. Contributors: F. Holmann, C.E. Lascano, P.J. Argel, and R. Goyenaga, CIA T, and MAG, Colombia, Costa Rica. Dean Holland, Sam Fujisaka, Peter Kerridge, Douglas White, Glenn Hyman, Scott Cechi, Carlos Bruzone, Carlos Ostertag, Mark Lundy, Ann Braun, Maria Femandez 4. 7.4. Publications See list in following section. 170 Publications Publications 1999 Journal Papers Argel, P.J. y Lascano, C. E. 1998. Cratylia argentea (Desvaux) O. Kuntze: Una nueva leguminosa arbustiva para suelos ácidos en zonas húmedas tropicales. Pasturas Tropicales 20(1): 37-43. Belay Simane, C. Wortmann and G. Hoogenboom, 1998. Haricot bean agroecology in Ethiopia: assessment using agroclimatic and crop growth simulation models. African Crop Science Joumal 6( 1 ):9-18. Cardenas, E.A., Maass, B.L., Peters, M,., Franco, L.H. 1999. Evaluacion de germoplasma nuevo de Arachis pintoi en Colombia 2. Bosque muy humedo premontano (zona cafetera), Caldas. Pasturas Tropicales volumen 21(2) (in press) Chee, Y.K., Chong, D.T., Ahmad Faiz, Tajuddin lsmail, Stür, W.W. and Shelton, H.M. (1998). Forage resources in rubber plantations in Central Malaysia. Malaysian Joumal of Animal Science 3, 28-36. Duarte Osear, Pulido José, Silva Jorge, and Holmann Federico. 1998. Analysis of the current situation and technological altematives of agricultura) production systems in the Ceasar' s Valley Watershed of Colombia. Revista Corpoica (in press). Fischler, M. and C.S. Wortmann, 1999. Green manure research in eastem Uganda- a participatory approach. Agroforestry Systems (in press) Fischler, M., C.S. Wortmann, and B. Feil, 1998. Crotalaria (C. ochroleuca G. Don) as a green manure in maize-bean cropping systems in Uganda. Field Crops Research 6112:97-107. Fujisaka, S. 1999. "Towards a new institutional model offarmer participation in research on natural resource management and germplasm improvement. In: Fujisaka, S., ed. Fujisaka, Sam, German Escobar, and Erik Veneklaas. 1998. P1ant community diversity relative to human land uses in an Amazon forest colony. Biodiversity and Conservation Vol. 7:41-57 Fujisaka, S, C Castilla, G Escobar, V Rodrigues, E Veneklaas, R Thomas, and M Fisher. 1998. Impacts of forest conversion: estimates of carbon emissions and plant species loss in a Brazilian Amazon colony. Agriculture,Environment, and Ecosystems. 69:17-26 Fujisaka, Sam, and D. White. 1998 Pasture or permanent crops after slash and bum cultivation? Land use choice in three Amazon forest colonist. Agroforestry Systems. 42:45-59 Fujisaka, Sam, German Escobar and Erik Veneklaas. 1998. Plant community diversity relative to human land use in an Amazon forest colony. Biodiversity and Conservation 7:41-57. Fujisaka, S, C. Castilla, G. Escobar, V. Rodrigues, E. Veneklaas, R. Thomas and M. Fisher. 1998. The effects of forest conversion to annual crops and pastores: estima tes of carbon emissions and plant species loss in a Brazilian Amazon colony. Agriculture, Ecosystem and Environment 69:17-26. Gijsman A.J. and Sanz J.I. 1998 Soil Organic Matter Protection in a Volcanic-Ash Soil under Fallowing or Cultivation with Applied Chicken Manure. European Joumal ofSoil Science 49:427-436. 171 Publications Giraldo L.M., Lizcano L.J., Gijsman A.J., Rivera B., Franco L.H. 1998 Adaptación del modelo DSSAT para simular la producción del Brachiaria decumbens. (Adapting the DSSA T model for simulating the production of Brachiaria decumbens ). Pasturas Tropicales 20:2-12. Hacker, J.B., Phimphachanhvongsod, V., Novaba, S., Kordnavong, P., Veldkamp, J. and Simon, B.K. (1998). A guide to the grasses ofXieng Khouang Province, Lao POR and sorne notes on ecology of grazing lands in the province. Genetic Resources Communication No. 28. CSIRO Tropical Agriculture, St. Lucia, Australia, 89 p. Hess H. D., F1orez H., Lascano C.E., Baquero L.A .. , Becerra and Ramos J. 1999. Fuentes de variación en composición de la leche y niveles de urea en sangre y leche de vacas en sistemas doble proposito en el tropico bajo de Colombia. Pastura Tropicales 21 (1): 33-42 Holmann, F. 1999. Ex-ante analysis ofnew forage alternatives in Peru, Costa Rica, and Nicaragua. Joumal ofLivestock Research for Rural Development (in press). Ho1mann, F. 1999. Evaluación ex-ante de nuevas alternativas forrajeras en Perú, Costa Rica, y Nicaragua. Pasturas Tropicales (en imprenta). Keller-Grein, G., Schultze-Kraft, R., Franco, L.H., Ramnirez, G., 1999. Multilocational agronomic evaluation of selected Centrosema pubescens on acid soils. Tropical Grasslands (in press) Otsyula, R.M., S.I. Ajanga, R.A. Buruchara and C.S. Wortmann, 1998. Development of an integrated bean root rot control strategy for westem Kenya. African Crop Science Joumal 6( 1 ):61-68. Pezo Danilo, Holmann Federico, and Arze José. 1998. Bioeconomic evaluation of a dairy production system based on the intensive use of ferti1ized grasses in the humid tropics of Costa Rica .. Agronomía Costarricense (in press). Rivas, L. and F. Holmann. 1999. Early Adoption of Arachis pintoi in the humid tropics: The case ofthe dual-purpose systems in Caquetá, Colombia. Tropical Grasslands (submitted). Rivas, L. and F. Holmann. 1999. Adopción temprana de Arachis pintoi en el trópico húmedo: El caso de los sistemas de doble propósito en Caquetá, Colombia. Pasturas Tropicales 21 ( 1 ): 2-17. Wortmann, C.S., M. Silver-Rwakaikara and J. Lynch. 1998. Efficiency ofnitrogen acquisition and utilization in common bean (Phaseolus vulgaris L.) in Uganda. African Crop Science Joumal 6(3):273-282. Wortmann, C.S. 1998. An adaptation breeding strategy for water deficit in bean developed with the application of the DSSA T3 Drybean Model. African Crop Science Joumal 6(3 ):215-225. Yi Kexian, Lascano C. E., Kerridge P.C. and Avila P. 1998. The effect ofthree tropical shrub legumes on intake rate and acceptability by small ruminants. Pasturas Tropicales 20 (3): 31-35 Book/Book chapter Argel, P.J. and Paton, C.J. 1999. Overcoming Legume Hardseededness. In: Loch, D.S. and Ferguson, J.E. (eds). Forage Seed Production, Volume 2: Tropical and Subtropical Species. CAB INTERNATIONAL. p. 247-265. 172 Publications Bhuktan, J, G Denning, and S Fujisaka. 1999. Rice cropping practices in Nepal: indigenous adaptation to adverse and difficult environments. In: Prain, G, S Fujisaka, and DM Warren, eds. Cheng, Y. and Home, P.M. (1998). Field experiments with forages and crops: practical tips for getting it right the first time. (FSP, Laos) 48p. (lndonesian Version). Cheng, Y. and Home, P.M. (1999). Field experiments with forages and crops: practica} tips for getting it right the first time. (FSP, Laos) 48p. (Vietnamese Version). Ferguson, J.E. (1998). Experiences offorage seed systems from Latin Arnerica. In: Home, P.M., Phaikaew, C. and Stür, W.W. (eds.). Forage Seed Supply Systems: Proceedings ofan Intemational Workshop held in Khon Kaen, Thailand, 31 October to 1 November 1996. CIAT Working Document Number 175, Los Banos, Philippines, 23-34. Fujjsaka, S, ed w/ collªboration of A Jones. 1999. Systems and Farmer Participatory Research: Developments in Research in Natural Resource Management. Cali: CIA T. Fujisaka, S. 1999. "Side stepped by the Oreen Revolution: farmers' traditional rice cultivars in the uplands and rainfed lowlands". In: Prain, G, S Fujisaka, and DM Warren, eds. Fujisaka, S. 1999. "Towards a new institutional model offarmer participation in research on natural resource management and germplasm improvement. In: Fujisaka, S., ed. Fujisaka, S, O Madrid, L Hurtado, HUsma, A Riese, Y Flores, F !drogo, J Barbaran, L Arevalo, and R Labarta. 1999. "Land use systems and dynamics in Pucallpa, Peru". /n: Fujisaka, ed. Fujisaka, S. 1999. "Participatory systems research: towards the future". /n: Fujisaka, S, ed. Holmann, F. and C. Lascano. 1998. A new strategy to improve the dual purpose cattle production systems in the tropics: The Tropileche Consortium. In G. Gonzalez-Stagnaro, N. Madrid-Bury, and E. Soto Belloso, eds., Improvement of crossbred cattle in dual purpose systems. University of Zulia. Maracaibo, Venezuela. Home, P.M., Phaikaew, C. and Stür, W.W. (eds.) (1998). Forage Seed Supply Systems: Proceedings ofan lntemational Workshop held in Khon Kaen, Thailand, 31 October to 1 November 1996. CIA T Working Document Number 175, Los Banos, Philippines, 80 pp. Howeler, R.H. (Ed.). 1998. Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held Nov 3-8, 1996 in Dangzhou, Hainan, China. 539 pp. Lascano, C. E., W. Stur and P. Home. 1998. Small farm production systems for rumiant animals in tropical regions. In: Proceedings ofthe 8111 World Conference on Animal Production. Symposium Series 1, Seoul, Korea, p. 377-39 1. Prain, G, S Fujisaka, and MD Warren. 1999. Biological and Cultural Diversity: The Role of lndigenous Agricultura/ Experimentation in Development. London: Intermediate Technology Publications. Stür, W.W. and Home, P.M. (1999). Towards farmer participation - Developing forage technologies with smallholder farmers in East Kalimantan, Indonesia. In Fujisaka, S. (ed). Systems Research and Participatory Research: Developments in Natural Resource Management. CIA T, Cali, Colombia. 173 Publications Stür, W.W. and Horne, P.M. (eds.). (1998). Developing Forage Techno1ogies with Farmers- A Training Manual. Forages for Smallholders Project, Los Baños, Philippines. Stür, W.W., Owen, J .A., Kerridge, P.C., Horne, P.M. and Hacker, J.B. (eds). 1998. Proceedings ofthe Second Regional Meeting ofthe Forages for Smallholders Project, held at the Chinese Academy of Tropical Agricultural Science, Danzhou, Hainan, P.R. China, 19-24 January 1997. CIAT Working Document No. 173, Los Baños, Philippines, 101 pp. White, D., F. Holmann, S. Fujisaka, K. Reategui, and Carlos Lascano. 1999. Does intensification of pasture technologies affect forest cover in tropical Latin Arnerica?: Inverting the question. /n D. Kaimowitz andA. Angilsen, eds., Agricultura} Technology Intensification and Deforestation. Commonwealth Agricultural Bureau (CAB). Forthcomming. Wortmann, C.S. and C.A. Eledu, 1999. Uganda' s Agroecological Zones: A Guide for Planners and Policy Makers. CIA T Publication. Wortmann, C.S., R.A. Kirkby,C.A. Eledu and D.J. Allen, 1998. An Atlas on Common Bean (Phaseolus vulgaris L.) production in Africa. Cali, Colombia: Centro Internacional de Agricultura Tropical. Wortmann, C.S., C.K. Kaizzi and M. Fischler, 1999. Farmers' experimentation on green manure/cover crops: A component ofparticipatory research for improvement ofUgandan farming systems. In: Systems and farmer participatory research: developments in natural resource management. (Ed. S. Fujisaka). CIAT, Cali, Colombia, pp 118-127. Wortmann, C.S. and R.K. Kirkby, 1998. Bean production in Africa: systems, constraints and research opportunities. In: Voysest, O. (ed.). An ecoregional framework for bean germplasm development. Documento de Trabajo, Centro Internacional de Agricultural Tropical (CIA T), Cali, Colombia, pp 29-54. Workshop and conference papers Abdullah, A., Wong, C.C. and Mohd Yusof, C.E. (1998). Experiences with on-farm forage evaluation on the east coast ofPeninsular Malaysia. In: W.W. Stür, J.A. Owen, P.C. Kerridge, P.M. Horne and J.B. Hacker (eds). Proceedings ofthe Second Regional Meeting ofthe Forages for Smallholders Project, held at the Chinese Academy ofTropical Agricultura} Science, Danzhou, Hainan, P.R. China, 19-24 January 1997. CIA T Working Document No. 173, Los Baños, Philippines, 81-86. Arca M, White O, O.Rios, Alegre J. The Forest Margins ofthe Peruvian Amazon, presentation at the Slash and Burn Agriculture: A Global Synthesis Symposium ofthe American Society of Agronomy MeetingsNovember 1- 5, 1999 Salt Lake City, Utah Argel, P.J. and Pérez, G. 1998. Adaptation ofnew species ofLeucaena in Costa Rica, Central America- Preliminary results. In: Shelton, H. M., Gutteridge, R. C., Mullen, B. F. and Bray, R. A. (eds). Leucaena- Adaptation, Quality and Farming Systems. ACIAR Proceedings No. 86, Canberra, Australia. p. 146-149. Argel, P.J., Lascano, C. E. and Ramírez, L. 1998. Leucaena in Latin American Farming Systems: Challenges for Development. In: Shelton, H. M., Gutteridge, R. C., Mullen, B. F. and Bray, R. A. (eds). Leucaena- Adaptation, Quality and Farming Systems. ACIAR Proceedings No. 86, Camberra, Australia. p. 319-323 . 174 Publications Argel, P.J. 1999. Tecnologías Forrajeras para el Desarrollo de una Ganadería más Productiva en el Trópico bajo de Centroamérica. Contribución del CIA T. In: Pomareda, C. ( ed). Intensificación de la Ganadería en Centroamérica: Beneficios Económicos y Ambientales (Memorias). FAO/CATIE, mayo, 1999. (In press ). Asis, P. and Orencia, L. (1998). Developing forage technologies with farmers in Cagayan de Oro, Mindanao, Philippines. Proceedings ofthe Regional Meeting ofthe FAO Forage Regional Working Group on Grazing and Feed Resources ofSoutheast Asia, 5-9 October 1998, Bicol, Philippines. Binh, L.H. (1998). The FSP in Vietnam- Progress and Plans. In: W.W. Stür, J.A. Owen, P.C. Kerridge, P.M. Horne and J.B. Hacker (eds). Proceedings ofthe Second Regional Meeting ofthe Forages for Smallholders Project, held at the Chinese Academy of Tropical Agricultural Science, Danzhou, Hainan, P.R. China, 19-24 January 1997. CIA T Working Document No. 173, Los Baños, Philippines, 9-14. Dmji, S. (1998). Mechanisms offodder seed supply systems in Bbutan. Australia's experíence with tropical forage seed supply systems. In: Horne, P.M., Phaikaew, C. and Stür, W.W. (eds.). Forage Seed Supply Systems: Proceedings of an International Workshop held in Khon Kaen, Thailand, 31 October to 1 November 1996. CIA T Working Document Number 175, Los Banos, Philippines, 73- 78. Fujisaka, S. 1999. "A retrospection ofsoil conservation in Clavería, Philippines" . In: McDonald, M and K Brown, eds. Issues and Options in the Design ofSoil and Water Conservation Projects. Bangor: University ofWales; Proceedings ofworkshop held in Llandudno, Conwy, UK, 1-3 Feb 1999. Fujisaka, S and Survey teams, 1999. "Forage tree adoption and use in SE Asia" Paper presented at workshop, "Working with Farmers: the Key to Adoption ofForage Technologies" Cagayan de Oro, Phi lippines, 12-15 October 1999. Fujisaka, S. 1999. "Research: help or hindrance to good farmers in high risk systems?" Paper, ASA Symposium "Agroecology and SoiVCrop Management among lndigenous Cultures" ASA/CSSA/SSSA Annual Meetings, Salt Lake City, Utah 31 Oct-4 Nov 1999. Fujisaka, Sam. 1998. Challenges facing soil science: a view from the outside. Opening Session ofthe 16Th World Congres of Soil Science ("The Aims of Soil Science, Challenges to be Met by Soil Science, and the Services Soil Science Can Render"). 20 August 1998, Montepellier, France. Gabunada, F., Balbarino, E. and Obusa, A. (1998). Farmer participatory research in forage in Matalom, Leyte. In: W.W. Stür, J.A. Owen, P.C. Kerridge, P.M. Horne and J.B. Hacker (eds). Proceedings of the Second Regional Meeting ofthe Forages for Smallholders Project, held at the Chinese Academy ofTropical Agricultural Science, Danzhou, Hainan, P.R. China, 19-24 January 1997. CIAT Working Document No. 173, Los Baños, Philippines, 59-79. Gabunada, F and Stür, W.W. (1998). Agronomic evaluation ofthe Leucaena foundation collection: 2. Hum id tropics in the Philippines. In: Proceedings of an intemational workshop ' Leucaena- Adaptation, Quality and Farming Systems' held in Hanoi, Vietnam, 9-1 4 February 1998. ACIAR Proceedings No. 86, Canberra, Australia, 113-115. 175 Publications Gijsman A.J., Hoogenboom G. and Parton W.J. 1999 Linking DSSAT and CENTURY for improved simulation of smallholder agricultura! systems. Proceedings of the Intemational Symposium on Modelling Cropping Systems, Lleida (Spain), 21-23 June 1999, p. 189-190. Gijsman A. and Kerridge P. 1998 Integrating experiments with agronomic models and geographic information systems to better target research and the extension of research results. Proceedings of the Workshop on Ecoregional Research at ILRI (P.K. Thornton and A.N. Odero, Eds.), p.83-85. International Livestock Research Institute, Addis Abeba, October 5-8, 1998. Gijsman A.J. and Bowen W.T. 1999 Simulating crop production in low-input agricultura! systems with DSSA T linked to the CENTURY soil-organic-matter module. In preparation for the Third International Symposium on Systems Approaches for Agricultura! Development (SAAD-ill), November 8-1 O, 1999. Ouodao, L., Changjun, B. and Huang Huide (1998). Forage seed supply systems in Hainan, PR China. In: Horne, P.M., Phaikaew, C. and Stür, W.W. (eds.). Forage Seed Supply Systems: Proceedings of an International Workshop he1d in Khon Kaen, Thailand, 31 October to 1 November 1996. CIA T Working Document Number 175, Los Banos, Philippines, 17-22. Guodao, L., Changjun, B., Changshun, J and Jiashao, W. (1998). The FSP activities in China. In: W.W. Stür, J.A. Owen, P.C. Kerridge, P.M. Horne and J.B. Hacker (eds). Proceedings ofthe Second Regional Meeting ofthe Forages for Smallholders Project, held at the Chinese Academy ofTropical Agricultura! Science, Danzhou, Hainan, P.R. China, 19-24 January 1997. CIAT Working Document No. 173, Los Baños, Philippines, 3-8. Holmann, F., C. Lascano, andA. Ramirez. Workshop on Research Progress achieved by the Tropileche Consortiurn. Moyobamba, Peru. June 28-30. 1999. Holmann, F. Study of options for the development and intensification of milk production systems in tropical Latin America. Presented at the Workshop "Development ofmilk production in tropical Latin America", November 18-21, 1998, Maracay, Venezuela. Hopkinson, J.M. (1998). Australia's experience with tropical forage seed supply systems. In: Horne, P.M., Phaikaew, C. and Stür, W.W. (eds.). Forage Seed Supp1y Systems: Proceedings ofan International Workshop held in Khon Kaen, Thai1and, 31 October to 1 November 1996. CIA T Working Document Number 175, Los Banos, Philippines, 35-46. Horne, P.M. (1998) Securing the livelihoods offarmers in upland areas ofLao POR: the role oflivestock and opportunities for forage development. Proceedings ofthe ACIAR workshop "Upland farming systems in the Lao POR: Prob1ems and opportunities for livestock", Vientiane, 19-23 May, 1997. (ACIAR, Canberra; in press) Horne, P.M. (1998) Let them eat grass: experiences in using participatory approaches to develop forage technologies. Proceedings ofthe ACIAR meeting "Red soi1 forages- from research into practice", Jianyang, Fujian Province, China, 6 to 9 October 1997 (ACIAR, Canberra; in press) Howeler, R.H. 1998. Cassava Agronomy Research in Asia. - An Overview 1993-1996. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Dangzhou, Hainan, China. Nov 3-8 1996 pp 335-375 Howeler, R. H. 1999. Developing sustainable cassava production systems with farmer's 176 Publications involvement in Asia. In: Proc. CIA T's Intemal Workshop on "CIA T's Experience with Systems Research and its Future Direction", held at CIAT, Cali, Colombia, Dec 1-2, 1997. (in press) Howeler, R.H. 1999. Cassava production practices- Can they maintain soil productivity? In: Proc. Intem. Sympj. on Cassava, Starch and Starch Derivatives, held in Nanning, Guangxi, China. Nov 11-15, 1996. (in press) Howeler R.H. and G. Henry. 1998. Farmer Participatory research for cassava technology transfer in Asia.- Constraints and opportunities. In: R.H. Howeler (Ed.). Cassava Breeding, Agronomy and Farmer Participatory Research in Asia. Proc. 5th Regional Workshop, held in Danzhou, Hainan, China. Nov 3-8, 1996. pp. 497-514. Howeler, R.H., Horne, P.M. and Stor, W.W. CIAT projects on natural resources management in Asia with special reference to the Red River Basin of Vietnam. Towards an Ecoregional Apporache for Natural Resources Management in the Red River Basin of Vietnam. Selected papers of a planníng workshop held in Hanoi, Vietnam, 6-9 October 1997. pp 95-118. Howeler, R.H, Nguyen The Dang and W. Vongkasem. 1998. Farrner participatory selection ofvetiver grass as the most effective way to control erosion in cassava-based cropping systems in Vietnam and Thailand. In: Proc. First lntem. Conference on Vetiver: A Miracle Grass, held Feb 4-8, 1996 in Chiang Raí, Thailand. Office Royal Devel. Projects Board, Bangkok, Thailand. pp. 259-272. Howeler, R.H., A. Tongglum S. Jantawat and W.H. Utomo. 1999. The use offorages for soil fertility maintenance and erosion control in cassava in Asia. Proc. Third Annual Meeting ofForages for Smallholder Project (FSP), held in Samarinda, East Kalimantan, Indonesia. March 22-27, 1998. pp. lbrahim, T. (1998). Participatory research on forages with smallholder farrners in North Sumatra, Indonesia. In: W.W. Stür, J.A. Owen, P.C. Kerridge, P.M. Home and J.B. Hacker (eds). Proceedings ofthe Second Regional Meeting ofthe Forages for Smallholders Project, held at the Chinese Academy ofTropical Agricultural Science, Danzhou, Hainan, P.R. China, 19-24 January 1997. CIA T Working Document No. 173, Los Baños, Philippines, 27-33 . lbrahim, M.A., F. Holmann, M. Hemandez, andA. Camero. 1999. The contribution of Erythrina as a pro te in bank. with banana waste for the improvement of animal production systems in the humid tropics. Agroforestry Systems (in press). Krishnan, K. ( 1998). The Kerala experience with forage seed production and supply system. In: Home, P.M., Phaikaew, C. and Stür, W.W. (eds.). Forage Seed Supply Systems: Proceedings ofan lntemational Workshop held in Khon Kaen, Thailand, 31 Octoberto 1 November 1996. ClAT Working Document Number 175, Los Banos, Philippines, 47-56. Lascano Carlos E. 1999. Se1ective grazing on grass-legume mixtures in tropical pastures. In: Proceedings ofthe Intemational Symposium on Grassland Ecophysiology and Grazing Ecology. Curitiba, Parana, Brazil- 24-26 August, 1999. p 151-164 Lascano Carlos E. 1999. Desarrollo de especies forrajeras para sistemas de producción animal en America tropical. In: Proceedings Simposio Internacional de forrajeras subtropicales. Tucumán, Argentina- September l-3 . p 65-70 Lascano, C.E., Stür, W.W. and Home, P.M. (1998). Small farrn production systems for ruminant animals in tropical regions. Prodeedings ofthe 8th World Conference of Animal Production. Symposium Series 1, Seoul, Korea, 377-391. 177 Publications Magboo, E., Gabunada, F. and Faylon, P.S. (1998). Collaborative forage R & O program in the Philippines - the Forages for Smallholders Project. In: W.W. Stür, J.A. Owen, P.C. Kerridge, P.M. Home and J.B. Hacker (eds). Proceedings ofthe Second Regional Meeting ofthe Forages for Smallholders Project, held at the Chinese Academy ofTropical Agricultural Science, Oanzhou, Hainan, P.R. China, 19-24 January 1997. CIAT Working OocumentNo. 173, Los Baños, Philippines, 3 5-45. Mullen, B.F., Gabunada, F., Shelton, H.M., Stür, W.W. and Napompeth, B. (1998). Psyllid resistance in Leucaena. Proceedings of an intemational workshop ' Leucaena - Adaptation, Quality and Farming Systems' held in Hanoi, Vietnam, 914 February 1998. ACIAR Proceedings No.86, Canberra, Australia, 51-60. Nacalaban, W., Gabunada, F. and Magboo, E. (1998). Farmer participatory approach in developing systems to integrate forages and fodder trees in smallholder crop-based farming systems of Malitbog, Bukidnon. Proceedings ofthe Regional Meeting ofthe FAO Forage Regional Working Group on Grazing and Feed Resources of Southeast Asia, 5-9 October 1998, Bicol, Philippines. Peters, M., P. Home, A. Schmidt, F. Holmann, P. Kerridge, S.A. Tarawali, R. Schultze -Kraft, C.E. Lascano, P. Argel, W. Stür, S. Fujisaka, K. Müller-Samann and C. Wortmann The role offorages in reducing poverty and degradation of natural resources in tropical production systems. Presented in the" Poverty Workshop", September 10-14, 1999. IICA, San Jose, Costa Rica. Phaikaew, C. and Hare, M. (1998). Thailand's experience with forage seed supply systems. In: Home, P.M., Phaikaew, C. and Stür, W.W. (eds.). Forage Seed Supply Systems: Proceedings of an Intemational Workshop held in Khon Kaen, Thailand, 31 October to 1 November 1996. CIAT Working Oocument Number 175, Los Banos, Philippines, 7-16. Phaikaew, C., Klum-em, K. and Nakamanee, G. (1998). Forages for Smallholders Project in Thailand. In: W.W. Stür, J.A. Owen, P.C. Kerridge, P.M. Home and J.B. Hacker (eds). Proceedings ofthe Second Regional Meeting ofthe Forages for Smallholders Project, held at the Chinese Academy of Tropical Agricultural Science, Oanzhou, Hainan, P.R. China, 19-24 January 1997. CIA T Working Oocument No. 173, Los Baños, Philippines, 53-57. Phaikaew, C. and Stür, W. (1998). Forage Seed Production and Supply Systems in Southeast Asia. Proceedings ofthe Regional Meeting ofthe FAO Forage Regional Working Group on Grazing and Feed Resources of Southeast Asia, 5-9 October 1998, Bicol, Philippines. Phimphachanhvongsod, V. (1998). The FSP in Lao POR - progress and plans. In: W.W. Stür, J.A. Owen, P.C. Kerridge, P.M. Home and J.B. Hacker (eds). Proceedings ofthe Second Regional Meeting of the Forages for Smallholders Project, held at the Chinese Academy ofTropical Agricultural Science, Oanzhou, Hainan, P.R. China, 19-24 January 1997. CIAT Working OocumentNo. 173, Los Baños, Philippines, 47-52. Phimphachanhvongsod, V., Phengsavanh, P. and Home, P. (1999). Oeveloping forage technologies with farmers: putting plans into action in northem Lao POR. Proceedings of a meeting held in Luang Phabang, 15-16 July 1999. Forages for Smallholders Project, Vientiane, Lao POR. Tuhulele, M. (1998). Progress report on the FSP in Indonesia. In: W.W. Stür, J.A. Owen, P.C. Kerridge, P.M. Home and J.B. Hacker (eds). Proceedings ofthe Second Regional Meeting ofthe Forages for 178 Publications Smallholders Project, held at the Chinese Academy ofTropical Agricultura! Science, Danzhou, Hainan, P.R. China, 19-24 January 1997. CIAT Working DocumentNo. 173, Los Baños, Philippines, 15-25. Turton, C. and Baumann, P. (1998). Beyond the formal sector: Fodder seed networks in India. In: Horne, P.M., Phaikaew, C. and Stür, W.W. (eds.). Forage Seed Supply Systems: Proceedings of an International Workshop held in Khon Kaen, Thailand, 31 October to 1 November 1996. CIAT Working Document Number 175, Los Banos, Philippines, 57-72. White, D., F. Holmann, S. Fujisaka, K. Reategui, and Carlos Lascano. Does intensification of pasture technologies affect forest cover in tropical Latín Arnerica?: Inverting the question. Presented at the International Workshop "Agricultura! Technology Intensification and Deforestation". March 11- 13, 1999. CATIE, Turrialba, Costa Rica. Wortrnann, C. S., 1999. Nutrient budgets: understanding the problems, causes and trends of soil resource degradation. Presented at the Soil Fertility Management Workshop ofthe Soil and Water Conservation Society ofUganda, 5-6 May, 1999. Wortrnann, C.S. and C.K. Kaizzi. 1998. Fanner involvement in research for improved soil management. Presented at PRIAM synthesis workshop in Nazareth, Ethiopia. Xiaoliang, L. ( 1998). The use of improved grasses and legumes in Guandong. In: W. W. Stür, J .A. Owen, P.C. Kerridge, P.M. Horne and J.B. Hacker (eds). Proceedings ofthe Second Regional Meeting of the Forages for Smallholders Project, held at the Chinese Academy ofTropical Agricultura! Science, Danzhou, Hainan, P.R. China, 19-24 January 1997. CIAT Working DocumentNo. 173, Los Baños, Philippines, 91 . Zhiping, Q, Sumei, T and Caixang, H. ( 1998). Phosphorus application in creases the yield of Stylosanthes guianensis in Hainan. In: W .W. Stür, J .A. Owen, P.C. Kerridge, P.M . Horne and J.B . Hacker (eds). Proceedings ofthe Second Regional Meeting ofthe Forages for Smallholders Project, held at the Chinese Academy ofTropical Agricultura! Science, Danzhou, Hainan, P.R. China, 19-24 January 1997. CIAT Working Document No. 173, Los Baños, Philippines, 87-90. Otber publications Argel, P.J. 1999. Maní Forrajero: Una Leguminosa de Uso Múltiple para el Sector Agropecuario de Costa Rica. Montecillos (Costa Rica). Año XV, No. 102. p. 12-13. Rachier, G .O., C.S.Wortmann and J.S . Tenywa, 1998. Phosphorus tolerance in common bean as affected by root architecture. Annual Report of the Bean Improvement Cooperative 41 :206-207. Tropileche News Letters. No 4 (March 98); NoS (October 98); No6 (March 99). Wortmann, C.S., M. Fischler, F. Alifugani and C.K. Kaizzi, 1998. Accomplishments ofparticipatory research for systems improvement in Iganga District, 1993-1997. CIA T Occasional Paper No. 27, Kampala, Uganda. Wortmann, C. S. and C.K. Kaizzi, 1998. Nutrient balances and expected effects of alternative practices in farming systems of Uganda. IN: Nutrient Balances as Indicators ofProductivity and Sustainability in sub-Saharan African Agriculture. Agricultural Ecosystems and Environment (Ed. Smaling, E.M.A.} 71(1 /2/3): 115-130. 179 Publications Poster Papers Argel, P.J. 1999. Maní Forrajero: Una Leguminosa de Uso Múltiple para el Sector Agropecuario de Costa Rica. Montecillos (Costa Rica). Año XV, No. 102. p. 12-13. Argel, P.J. Lobo di Palma, M. Romero F. Gonzalez J. Lascano C.E. Kerridge P.C. Holmann F. 1999. The leguminous shrub, Cratylia argentea: a dry season feeding alternative for the humid tropics Le Van An, Le Quang Bao, Le Due Ngoan, Nguyen Thi Cach, Nguyen Xuan Hong, Nguyem Minh Hieu, Hoang Huu Hoa, Peter Kerridge, Sam Fujisaka, Reinhardt Howeler, Peter Home, John Graham 1999. Community-based Natural Resource Management in Hong Ha, Hue, Vietnam. Brisbane (Australia) Video tapes The Farm of Antonio Lopez (11 minutes). CIAT, MAG, Costa Rica 1999. 180 List of Donors AusAID Funding: Main contact: Forages for Smallholders Project (FSP) Mr. Tony Coughland Manager, South East Asia Regional Section AusAID GPO Box 887, Canberra ACT 2601 Australia Donors Colombia- The Government of Colombia Funding: Production systems components Main contact: Dr. Alvaro Uribe IDRC Funding: Main contact: IDRC Funding: Main Contact: IDRC Funding: Main Contact: Director Unidad Estudios Agrarios Departamento de Planeación Nacional Santafé de Bogotá, Colombia DEPAM Project-Participatory Technology Development, Pucallpa. Dr. Simon Carter Casilla de correo 63 79 Montevideo, 111 000 Uruguay Health, diversity and Natural Resources use in the Westem Amazonian Lowlands An Integrated agro-ecosystem approach Jo-Ann St. Louis 250 Albert St. Ottawa Ontario Canada Supplementary funds to assist University ofHue Community-based Upland Natural Resources Management in Hong Ha, Hue, Vietnam John Graham Tanglin P.O. Box 101 Singapore Interamerican Development Bank (BID) Funding: Legume-based forage systems for dual-purpose cattle farm Main contact: Dr. Rubén Echeverría Economist, Agriculture Div. PRA 1300 New York Avenue NW Washington OC 20577 USA 181 Donors Oxford Forestry Institute (OFI) Funding: Supplementary funds evaluation of Leucaena in Asia and LAC from DFID Main contact:: Mr. Alan Pottinger Oxford Forestry Institute Dept. of Plant Sciences, University of Oxford South Parks Road, Oxford OX1 3RB United l<.ingdom Systemswide Livestock Program Funding : Legume-based forage systems for dual-purpose cattle Main contact: Dr. J. Smith Intemational Livestock Research Center, ILRI P.O. Box 5689 Addis Ababa, Ethiopia The Nippon Foundation, Tokyo, Japan Funding: Integrated Cassava-based Croping Systems in Asia: Farming Practices to enhance Sustainability Main contact: Ms Ayako Sono Chairperson The Nippon Foundation Senpaku Shinko Bldg 1-15-16 Toranomon, Minato-ku, Tokyo 105 Japan U niversity of Queensland Funding: Supporting Leucaena evaluations as part of ACIAR PN 9433 (''New Leucaenas for Southeast Asian, Pacific and Australian Agriculture Main contact:: Max Shelton Department of Agriculture The University ofQueensland Brisbane, Qld 4072 Australia 182 StaffList Senior Staff PETER C. KERRIDGE, Ph.D., Agronomist, Project Leader SAMUEL FUnSAK.A, Ph.D., Agricultura/ Anthropologist FEDERICO HOLMANN, Ph.D., Animal Scientist, agricultura/ economist PETER HORNE, Ph.D., Forage Agronomist (Stationed in Vientiane, Lao POR - CSIRO) REINHARD HOWELER, Ph.D., Agronomist (Stationed in Bangkok, Thai/and) CARLOS E. LASCANO, Ph.D., Ruminant Nutritionist WERNER STUR, Ph.D., Forage Agronomist (Stationed in Los Baños, Philippines) DOUGLAS WHITE, Ph.D. Resource Economist DEAN HOLAND, Ph.D. Agronomist Participatory Research Consultant PEDRO ARGEL, Ph.D., Agronomist (Stationed in San José Costa Rica) Research Fellow ARJAN GIJSMAN, Ph.D., Soi/ Scientist/Model/er Jr Research Fellow TAMSYN MURRAY Research Associates RICARDO LABARTA, Econimist LE VAN AN BUIXUAN AN Research Assistants PATRICIA AVILA, Animal Scientist GERMAN ESCOBAR, Biologist FRANCISCO GABUNADA Jr., Agronomist LUIS H. FRANCO, Agronomist EMMA LUISA ORENCIA, Agronomist PHONEPASEUTH PHENGSAV ANH, Agronomist GERARDO RAMIREZ, Statistician GUSTAVO RUIZ, Animal Scientist 183 Staff Administrative Assistant MAruAD~EMABO~LA CARMENANDI FERNANDO HINCAPIE ANDERSON MEDINA Technicians FREO Y T ABARES HERMANUSMA JOHN AVILES JOSE ALEXANDER HOYOS Secretaries YUVIZA BARONA CHINT ANA CHANHDENG DEABONILA Staff 184