Balancing livestock needs and soil conservation: Assessment of opportunities in intensifying cereal?legume?livestock systems in West Africa Final Report SYSTEMWIDE LIVESTOCK PROGRAMME slp C G I A R S yst e m w i d e L i v e s t o c k P rogr a m m e ILRI INTERNATIONAL LIVESTOCK RESEARCH INSTITUTE Balancing livestock needs and soil conservation: Assessment of opportunities in intensifying cereal?legume?livestock systems in West Africa FINAL REPORT SYSTEMWIDE LIVESTOCK PROGRAMME ? 2010 ILRI (International Livestock Research Institute). Editing, design and layout?ILRI Editorial and Publishing Services, Addis Ababa, Ethiopia. Correct citation: IITA. 2010. Balancing Livestock Needs and Soil Conservation: Assessment of Opportunities in Intensifying Cereal-Legume-Livestock Systems in West Africa: CGIAR Systemwide Livestock Programme, Project Final Report. Addis Ababa (Ethiopia). 56 pp. iiiBalancing livestock needs and soil conservation in West Africa Table of Contents List of Tables iv List of Figures v A. Project information 1 1 Title of project 1 2 Goals and objectives 1 3 Specific objectives 1 4 Project outputs 1 B. Investigators and collaborating institutions 2 5 Lead principal investigator (PI) and contact details 2 6 Principal investigators and institutional affiliation 2 7 Collaborators and institutional affiliation 2 C. Final report 3 8 Summary 3 9 Site description 4 10 Implemented activities and results per output 6 Output 1 Analysis of the factors affecting farmers? decisions on land management and trade-offs in the uses of crop residues 6 Output 1.1 Socio-economic characteristics and allocation of crop residues by farm households in the subhumid and semi-arid savannahs of West Africa 6 Output 1.2 Trade-offs in agricultural uses of stover and haulms in the dry savannahs of Ghana, Nigeria and Niger 16 Output 2 Identifying areas of intervention and entry points through which appropriate crop?livestock integration technologies can stimulate the intensification of crop?livestock systems 35 Output 2.1 Identifying entry points for improving the productivity of cereal?legume?livestock systems: The NUTMON approach 35 11 Capacity building/training 50 12 Presentations in conferences/meeting 50 13 Problems and measures taken 50 14 Linkages with other research 51 iv Balancing livestock needs and soil conservation in West Africa List of Tables Table 1.1.1 Maximum likelihood estimates of potential adoption of crop residue technology 12 Table 1.2.1 Physical and chemical properties of soils at the study sites 17 Table 1.2.2 Crop residue application rates 18 Table 1.2.3 Chemical characteristics of crop residues incorporated into soil 19 Table 1.2.4 Chemical composition and digestibility of crop residues fed 21 Table 1.2.5 Apparent trade-offs for the crop residues use?scenarios tested at Farm 1 in Cheyohi 25 Table 1.2.6 True trade-offs for the crop residues use?scenarios tested in farm 1 at Cheyohi 25 Table 1.2.7 Apparent trade-offs for the crop residues use?scenarios tested at Farm 2 in Sarauniya 26 Table 1.2.8 True trade-offs for the crop residues use ? scenarios tested at Farm 2 in Sarauniya 26 Table 1.2.9 Apparent trade-offs for the crop residues use?scenarios tested at Farm 1 in Garin Labo 27 Table 1.2.10 True trade-offs for the crop residues use ?scenarios tested at Farm 1 in Garin Labo 28 Table 1.2.11 Correlation coefficient (r) for crop produces at Farm 1 in Cheyohi 28 Table 1.2.12 Correlation coefficient (r) for livestock produces at Farm 1 in Cheyohi 29 Table 2.1.1 Resource profile of households? categories 37 vBalancing livestock needs and soil conservation in West Africa List of Figures Figure 1 Locations of the study 4 Figure 1.1.1 Conceptual framework of alternative uses of crop residues in crop?livestock farming system 8 Figure 1.1.2 Distribution of socio-economic characteristics and allocation of crop residues 10 Figure 1.1.3 Allocation of crop residues in crop?livestock farming system 11 Figure 1.2.1 Grain yield and live weights measured in the farms at Cheyohi 23 Figure 1.2.2 Grain yield and live weights measured at the farms in Sarauniya 24 Figure 1.2.3 Grain yield and live weights measured in the farms at Garin Labo 24 Figure 1.2.4 Revenue?trade-off relationships 29 Figure 1.2.5 Rainfall pattern and crop water requirement 30 Figure 2.1.1 Conceptual framework for nutrient cycling in smallholder cereal?legume?livestock systems 36 Figure 2.1.2a Nitrogen flows in cereal?legume?livestock systems at farm level in Garin Labo 40 Figure 2.1.2b Nitrogen flows in cereal?legume?livestock systems at village level 41 Figure 2.1.3a N balances at farm level in Garin Labo with N fertilizer application 41 Figure 2.1.3b N balances at farm level in Garin Labo without N fertilizer application 42 Figure 2.1.4a N balances at village-level with N fertilizer application 42 Figure 2.1.4b N balances at village-level without N fertilizer application 42 Figure 2.1.5a P flows in cereal-legume-livestock systems at farm-level in Garin Labo 43 Figure 2.1.5b P flows in cereal-legume-livestock systems at village-level 43 Figure 2.1.6a P balances at farm-level in Garin Labo with N fertilizer application 44 Figure 2.1.6b P balances at farm-level in Garin Labo without N fertilizer application 44 Figure 2.1.7a P balances at the village-level with N fertilizer application 44 Figure 2.1.7b P balances at village-level without N fertilizer application.Entry points for improving cereal?legume?livestock productivity 45 Figure 2.1.8 Hot spots for research interventions in cereal?legume?livestock farms in Garin Labo 45 1Balancing livestock needs and soil conservation in West Africa A Project information 1 Title of project Balancing livestock needs and soil conservation: Assessment of opportunities in intensifying cereal?livestock systems in West Africa 2 Goals and objectives The project?s goal was to identify key areas where research can make a difference in balancing trade-offs among livestock, soils and crops, while taking advantage of synergies in evolving crop?livestock systems. 3 Specific objectives The specific objectives were to: Quantify trade-off effects between the use of biomass as a livestock feed and its use in ? improving soil fertility; Identify the key driving forces and areas of intervention and entry points through which ? research can facilitate synergies during the intensification of crop?livestock systems; Create better institutional linkages between the different actors involved in research, ? extension and policy issues related to mixed farming systems. 4 Project outputs Analysis of the factors affecting farmers? decisions on land management and trade-offs in ? the uses of crop residues; The identification of areas of intervention and entry points through which appropriate ? crop?livestock integration technologies can stimulate the intensification of crop?livestock systems; and Enhanced institutional and partnership linkages between policy, extension, research and ? private sector actors and farmers for effectively addressing constraints faced in evolving mixed farming systems in a holistic way. 2 Balancing livestock needs and soil conservation in West Africa B Investigators and collaborating institutions 5 Lead principal investigator (PI) and contact details R Abaidoo (IITA) 6 Principal investigators and institutional affiliation D Chikoye, Birte Junge, E Berkhout and N Nziguheba (IITA) 7 Collaborators and institutional affiliation Okike E Gonzales, H Mario, and E Grings? International Livestock Research Institute (ILRI) B Gerard, Fatondji, Dougbedji ? International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) M Nouri ? Insitut National de Recherches Agronomiques du Niger (INRAN), Niger ENO Iwuafor? Institute for Agricultural Research (IAR), Samaru, Nigeria H Hansen ? The Royal Veterinary and Agricultural University (KVL), Denmark N Karbo ? Animal Research Institute (ARI), Ghana 3Balancing livestock needs and soil conservation in West Africa C Final report 8 Summary The system-wide livestock project ?Balancing livestock needs and soil conservation: Assessment of opportunities in intensifying cereal?livestock systems in West Africa? was carried out in Ghana, Nigeria and Niger. It was a project led by IITA in collaboration with ILRI, ICRISAT, INRAN, IAR, KVL and ARI. The general objective of the project was to identify key areas where research can make a difference in balancing trade-offs among livestock, soil, and crops, while taking advantage of synergies in evolving crop?livestock systems. The project focused on the identification of socio- economic factors influencing decision-making on crop residue uses, quantification of trade-offs in using crop residues as soil amendment or livestock feed, and the identification of entry points for improving the productivity of cereal?legume?livestock systems. The project was implemented in villages along crop?livestock integration gradient from northern Guinea savannah in Ghana through the Sudan savannah in Nigeria to the Sahel savannah in Niger. Field surveys and experiments were conducted from June 2007 to December 2008. The key socio-economic determinants of the uses of crop residues by a household were education, household size, agriculture extension visits. Contrary to popular notion that risk perception of farmers is the major driver for residue allocation, the study observed that risk perception of farmers had no influence on the decision taken by farmers on the use of their crop residues. The trade-off analysis provided a useful insight on the profitability of crop residue allocation options. The use of crop residues as fodder for livestock increased livestock productivity but had little or no effect on crop productivity when used as soil amendments. Farmers in northern Guinea savannah achieved the optimum farm revenue by using 25% of haulm and 75% of stover as fodder, and the remaining as soil amendment. On the contrary, farmers in the Sudan savannah used 75% of haulm and 25% of stover as fodder in order to attain the optimum farm revenue. However, in the Sahel savannah, higher farm revenues were achieved by feeding all residues to livestock and incorporating none into the soil. The potential pathways for improving the prevailing trade-offs identified in the study were as follows: the use of improved dual purpose legumes in appropriate rotation to legume biomass yield, application of surface mulch and tied ridging to improve soil water storage, and processing of stover to enhance its palatability and intake. Entry points for improving the productivity of cereal?legume?livestock systems were identified by auditing nutrient flows and calculating nutrient balances at the farm and village-levels. Regardless of the regimen of N fertilizer used, N balances were negative. Positive P balances were achieved when the recommended application rates of P fertilizers were used. Key entry points identified were: 4 Balancing livestock needs and soil conservation in West Africa Reducing the use of crop residues for fuel and construction purposes and substituting ? with other locally available materials; Quantifying the short and long-term benefits of crop residue retention and packaging the ? technology to boost its adoption; Developing cost-effective options for improving the quality of manure; and ? Developing cost-effective technologies to control leaching.? 9 Site description Characterization of the study area The studies were conducted in the northern Guinea savannah of Ghana, Sudan savannah of Nigeria, and Sahel savannah of Niger. These agro-ecological zones represent a gradient of crop? livestock integration in West Africa, with low integration in the northern Guinea Savannah and high integration in the Sudan and Sahel savannahs. The villages selected for the studies were located at the Tolon-Kumbungu District of Ghana, Kano state of Nigeria, and Maradi region of Niger (Figure 1). Source: Geographical Information Systems unit, IITA, Ibadan, Nigeria. Figure 1. Locations of study. Location: The Tolon-Kumbungu district lies within longitude 0?0.5?W and 1?20?W, latitude 9?7.5?N and 0?10?N, and covers a total land area of 304.5 km 2 . Kano lies on longitude 8?30?E, Agro-ecological zones Arid Sahel Savannah Derived Savannah Desert Humid Forest Semi-arid/Sudan Savannah Mid altitude northern Guinea Savannah Northern Guinea Savannah Southern Guinea Savannah National boundry 300 150 0 300 600 900 km 5Balancing livestock needs and soil conservation in West Africa latitude 11?30?N, and has a land area of 20,760 km 2 . The Maradi region is located on longitude 7?07?E, latitude 13?35?N, and has land area of 35,100 km?. Rainfall: The Tolon-Kumbungu district has a mono-modal rain pattern that starts in May and ends in October. The mean annual rainfall of the district is 913.6 mm yr ?1. The rainfall pattern in Kano is also mono-modal with a mean of 686 mm yr ?1 . The Maradi region has the least mean rainfall of 402 mm yr ?1 . Temperature: At Tolon-Kumbungu the daily average temperature is highest in March through to April/May just before the rains begin. Lowest temperature occurs in December?January which is usually associated with harmattan conditions. The district has a mean annual temperature of 26.0?C. The mean annual temperature in Kano is 26.7?C while Maradi has a mean annual temperature of 28.7?C. Soils: The soils in Tolon-Kumbungu consist of several soil series that includes the Tingoli, Nyankpala, Kumayili, Kpalesangu, Changnayili, and the Volta series. The dominant soil in the district is Ferric Luvisols (FAO and UNESCO 1994). The dominant soils in Kano are Regosols (FAO and UNESCO 1994), while the dominant soils of Maradi department are Eutric Gleysols (FAO and UNESCO 1994). Vegetation: The vegetation of Tolon-Kumbungu is generally woody savannah with the common trees being the neem (Azdirachta indica), dawadawa (Parkia biglobosa), shea (Vitellaria paradoxa), and Mango (Mangifera indica). Some common grasses are Heteropogon contortus, Imperata cylindrica, Cynodon dactylon and Andropogon spp. All the tree types are lopped in the wet season to feed to the tethered sheep and goats. The major plant species of Kano State are Combretum spp, Acacia spp, Terminalia spps and Andropogon gayanus. Plant species such as Acacia spp, Commiphora spp, and Cenchrus spp are common in the Maradi department. 6 Balancing livestock needs and soil conservation in West Africa 10 Implemented activities and results per output Output 1 Analysis of the factors affecting farmers? decisions on land management and trade-offs in the uses of crop residues Output 1.1 Socio-economic characteristics and allocation of crop residues by farm households in the subhumid and semi-arid savannahs of West Africa T Adesiyan, T Abdoulaye, EO Idowu and R Abaidoo Introduction In recent years, greater part of land in the subhumid and semi-arid agro-ecological zones of West Africa have been characterized by significant amount of land degradation and conversion caused mainly by overgrazing and agricultural activities (Oldeman et al. 1990). Soil fertility depletion in smallholder farming is the biophysical root cause of the declining per capita food production and which largely contribute to poverty and food insecurity. As noted by Barbier (2000), many rural farm households in Africa tend to respond to land productivity decline by abandoning their existing degraded farmland and moving to new land for cultivation. However, rapid growth in human population and the problems associated with land tenure systems in West Africa have necessitated the intensification of farmland use. In order to get improved land management practices, farmers in the savannah region of West Africa are moving from shifting cultivation, bush fallowing, and pastoralism towards mixed crop?livestock systems (Tiffen 2004). Kristjanson and Thorton (2001) pointed out that crop?livestock systems are the most important means for producing food across sub-Saharan Africa (SSA). About 438 million people (70% of the total SSA human population), 92 million cattle (80% of SSA cattle), and 194 million sheep and goat (80% of the total SSA sheep and goat) are found in these systems (Thornton et al. 2002). Smith et al. (1997) also noted that there would be increasing integration of crops and livestock over the next few decades due to increasing demand for crop and livestock products. However, investment in research on the challenges and potential rewards of improving integrated crop?livestock systems has been inadequate. Also, the advantages of crop?livestock systems for efficient use of natural resources and for meeting the food demand of developing countries have not been adequately promoted (Amogu 2004). The relevance of crop residue used in crop?livestock farming system in developing nations cannot be over emphasized. Several studies (e.g. Kristjanson et al. 2002) have considered crop residues and crop?livestock farming systems independently but none has quantified the trade-offs involved in the alternative uses of crop residue. Also, no study has been undertaken in the study area to examine farm 7Balancing livestock needs and soil conservation in West Africa households? decisions to use crop residues in crop or livestock production. This study explores the socio-economic factors which influence intensification of crop residues use in crop? livestock farming systems. The trade-offs, risks, and inefficiencies associated with the alternative uses of crop residues in the intensification technology among farm households in the subhumid and semi-arid savannah of West Africa are investigated. Research questions The questions addressed in the study were as follows: What is the current re-allocation of crop residues between crop and livestock production ? in Ghana, Nigeria, and Niger? Do farmers? risk attitude and inefficiencies influence the decision to choose between ? alternative uses of crop residue? What other factors influence the intensification of crop residues in crop?livestock farming ? systems in Ghana, Nigeria, and Niger? Objectives of the study The main objective of this study was to examine the socio-economic factors which affect the re- allocation of crop residues between crop and livestock production for improved rural livelihood. The specific objectives were to: Characterize the current allocation of crop residue use among farm households in Ghana, ? Nigeria, and Niger; and Analyse the socio-economic factors which influence the potential adoption intensity of ? crop residue for crop production. Conceptual framework Crop?livestock farming systems could be conceptualized as a complex interaction of different rural population, crop and livestock enterprises, land resources, environment, market, and government policies. As population increases in rural areas and agricultural land becomes scarce, the proportion of land available for farm households diminishes. This puts pressure on farm lands and results in land degradation. The productive ability of degraded land can be renewed either through natural regeneration (fallow) or application of fertilizer (organic or inorganic or both). In crop?livestock farming systems, crop residues and animal dung are good sources of organic fertilizer; and the livelihood of farm households largely depends on crop yields and livestock farming enterprises such as animal traction and manure. Crop residues are mostly fed to livestock or incorporated into the soil as a soil fertility improvement measure. They can also be sold or used for other domestic purposes. Rural farmers engaged in crop?livestock farming systems have multiple objectives of producing crops and livestock by maximizing economic benefits from current production, minimizing the cost of labour input, and ensuring sustainability of their resource base. 8 Balancing livestock needs and soil conservation in West Africa Farm households undertake three major decisions in the maximization of economic benefit from crop and livestock production. These include decisions on crop production, livestock production, and crop residues allocation. Since crop residues have alternative uses, farm households are faced with trade-offs in the benefits accruable from the use of crop residues. The production risk perception, risk attitude, and technical efficiency of farm households may have significant effect on the decision to adopt crop residues for alternative uses in crop?livestock farming system. Hence, any intensification technology to be adopted by the farm households must be assessed based on these and relevant factors. Any public policy formulation, therefore, should consider not only the marginal contribution of crop residue use to the mean of output but also the marginal reduction in variance of output and inefficiency. Figure 1.1.1. Conceptual framework of alternative uses of crop residues in crop?livestock farming system. Crop production Land degradation Livestock production Organic fertilizer Soil fertility Socio-economic factors Adoption of crop residue technology Fuel and other domestic uses Population pressure Crop residues Farm household decisions Inorganic fertilizer Sales of crop residues Production inefficiency Natural regeneration Animal dung Farm household livelihood/economic benefits Animal traction Crop production risk Farmers risk attitude/risk perception 9Balancing livestock needs and soil conservation in West Africa Empirical specification The Tobit Model (Greene 2008) was specified to analyse the socio-economic factors which influence the intensification of crop residues for crop production: ij ij ij YZbe=+ ** * if 0 0 if 0 ij ij ij ij Y Y Y ? > ? = ? ? ? ? where: ij Y denotes crop residue used for purpose j by farmer i Z is a vector of independent variables such as age, education, farm size, household labour, use of credit, crop extension visits, livestock extension visits, risk perception index and inefficiency index b denotes parameters to be estimated i e is assumed to be NID (0, 2 s ) and independent of i x Data description The study was undertaken in Ghana, Nigeria, and Niger in West Africa. The northern region of Ghana is located in the subhumid savannah zone while Kano State in Nigeria and Maradi District in Niger are found in the semi-arid savannah zones. In terms of crop?livestock farming in West Africa, maize?sorghum?cattle system is mostly practised in the northern region of Ghana, and the pearl millet?cowpea?cattle system is mostly practised in Kano and Maradi (Manyong 2002). Multi-stage sampling technique was used. In Ghana, 12 villages were randomly selected from 3 districts in the northern region. Similarly, 12 villages were selected from 3 Departments in Maradi. Five farm households each were selected from each typology making a total of 180 farm households each for Ghana and Niger. 1 In Kano, six local government areas (LGAs) were selected at random (i.e. two 2 LGAs from each of the three agricultural zones of the state) after which four villages were randomly selected within each LGA making a total of 24 villages. Fifteen farm households were selected randomly from each of the sampled villages making a total sample size of 360 in Nigeria. A set of well-structured questionnaires were then used to collect the relevant data after pre- testing the questionnaire. Questionnaire administration was done through the assistance of interpreters who translated the questionnaires to the farmers in Hausa (Niger and Nigeria) and Dagbani languages (Ghana). 1. The socio-economic typologies we considered were based on resource endowments of households, namely: Type 1: crop farmers (those farmers who produce only crops or combine livestock of less than 1 Tropical Livestock Unit (TLU) with crop production; Type 2: crop?livestock non-equipped or less resource endowed farmers (those farmers who combine between 1 and 2 TLU of livestock with crop production); and Type 3: crop?livestock equipped or resource endowed farmers (those who combine more than 2 TLU of livestock with crop production). 10 Balancing livestock needs and soil conservation in West Africa The coordinates of farmer-specific locations were collected through the use of Geo-Positioning System (GPS) equipment. Bio-physical data like rainfall, temperature, and soil types of farmer- specific locations (GPS) were also generated from the FAO soil type, rainfall, and temperature data in IITA database. Results and discussion This section briefly discusses some of the socio-economic characteristics and allocation of crop residues among farm households who practice crop?livestock farming systems in Ghana, Nigeria, and Niger. These distributions have been indicated in Figure 1.1.2. The allocation of cereal and leguminous crop residues for either crop production or livestock production are shown in Figure 1.1.3. Source: Cross-country household survey (2009). Figure 1.1.2. Distribution of socio-economic characteristics and allocation of crop residues. As indicated in Figure 1.1.2, not much difference was found between age, farming experience, education, household size, crop extension, and livestock extension visits of the sampled farm households in the three subregions. Age?ranges from 42 to 48 years? Farming experience?28 to 34 years? Education?two to four years? Household size?9 to 10 members? Visit by crop extension and livestock of extension?three to five times per production ? season Socio-economic characterstics of respondents Measurement (%) 0 20 40 60 80 100 Age (years) Farming experience (years) Education (years) Household size Plot size (ha) Total farm size (ha) Off-farm income ($) Non-farm expenditure ($) Visit by crop extension Visit by livestock extension Distance time (minutes) Ghana Nigeria Niger 47.8 47.8 42.4 30.4 34.2 28 30.4 34.2 2 10 4 9 0.66 1.9 1.8 158 4.1 4.1 495.9 1203.9 256.6 504.1 2574.3 533.6 3 5 3 3 3 3 39.4 20.5 15.5 Variable Harvested crop resudue among respondents Ghana Nigeria Niger T ype of residue (kg) Maize residue Millet residue Sorghum residue Total cereal residue Co wpea residue Groundnut residue So ybeans residue 0 50,000 100,000 150,000 200,000 1453.8 8.4 174.4 1636.6 0 13.5 47.6 3610.8 2429 0 3132 11412 14544 0 394.1 383.2 0 0 0 159320 16530 Niger Nigeria Ghana Quantity of residue (kg) 11Balancing livestock needs and soil conservation in West Africa Figure 1.1.3. Allocation of crop residues in crop?livestock farming system. Significant differences, however, exist in their off-farm income and non-farm expenditure; Nigeria recording the highest values of USD 1203.9 and USD 2574.3, respectively. In terms of farm size, farmers in Nigeria and Niger had the highest land holdings of 4.1 ha each and Ghanaian farmers had the smallest holdings of 1.58 ha. Also, it is interesting to note the distribution of harvested crop residue among the farm households across the three regions. The farmers mainly harvested maize, millet, sorghum, cereal, cowpea, groundnut, and soybean crop residues. Nigeria holds the record of having the highest sorghum and cereal residues. As depicted in Figure 1.1.3, the location of farm household tends to influence the trade-offs in the crop?livestock farming system. With the exception of Ghanaian farm households who harvested soybean residues, allocation of legume residues for livestock production appears to 1. Allocation of cerial residues Allocation of maize residues by farm households Allocation of millet residues by farm households Allocation of sorghum residues by farm households % residues allocation % residues allocation % residues allocation Country Country Country 0 100 14.4 76.8 62.5 0.0 Ghana Nigeria Niger Ghana Nigeria Niger Ghana Nigeria Niger Livestock Soil Livestock Soil Livestock Soil 20 0 50 100 64.7 54.7 10.4 75.7 22.3 0 50 100 58.2 0 55.7 25 66.4 3.5 2. Allocation of legume residues Allocation of cowpea residues by farm households Allocation of groundnut residues by farm households Allocation of soybean residues by farm households Country Country Country 0 50 100 0 50 100 0 50 100 0 85.6 1.1 0 92.7 3.9 97 2.2 81.3 3.4 85.7 2.9 Ghana Nigeria Niger Ghana Nigeria Niger Ghana Nigeria Niger Livestock Soil Livestock Soil Livestock Soil 17.2 82.6 65.7 5.2 0 0 % residues allocation % residues allocation % residues allocation 12 Balancing livestock needs and soil conservation in West Africa be high in all the three regions than for crop production. Also, with the exception of Ghana, allocation of cereal residues to feed livestock was higher than for crop production. These findings, thus, suggest that farm households in Ghana (65?68%) compared to those in Nigeria and Niger tend to intensify their crop residue use in soil fertility improvement for crop production than for livestock production. Intensification of crop residues for livestock production in Nigeria and Niger is not surprising because of the historical antecedents of households in these semi-arid savannah zones to engage in heavy livestock production. As noted by Amogu (2004), several attempts are now being made by international and multilateral funding agencies to promote programs that enhance crop?livestock integration in Nigeria, for instance. Empirical results The Tobit estimates which explain the adoption intensity of crop residue technology by sampled farm households in Ghana, Nigeria, and Niger are shown in Table 1.1.1. The efficiency variable is negative and significant for farm households located in all the three countries. This indicates that farmers with lower technical efficiency tend to use higher proportion of their harvested crop residues for crop production. Table 1.1.1. Maximum likelihood estimates of potential adoption of crop residue technology Variable Ghana Nigeria Niger coefficient coefficient coefficient Constant 1.022 (9.624)*** ?0.045 (?0.299) 1.060 (4.321)*** Efficiency ?0.195 (?1.904)* ?0.351 (?1.842)* ?1.307 (?4.550)*** Risk 0.306 (0.310) 0.399 (1.638) 0.231 (1.300) Age 0.002 (0.899) 0.002 (0.975) ?0.007 (?1.579) Household size 0.004 (0.573) 0.016 (2.433)** 0.001 (0.066) Experience ?0.001 (?0.675) ?0.002 (?1.038) 0.001 (1.264) Education 0.003 (0.688) 0.015 (3.210)*** ?0.003 (?0.345) Farm size ?0.004 (?0.155) 0.036 (3.414)*** 0.058 (2.937)*** Off farm income 0.727E?07 (0.513) ?0.142E?06 (?1.148) ?0.101E?06 (?0.734) Expenditure ?0.101E?05 (?0.409) ?0.248E?06 (?2.535)** ?0.226E?06 (?0.940) Crop extension 0.010 (1.067) ?0.003 (?1.028) ?0.015 (?2.088)** Livestock extension ?0.017 (?3.104)*** ?0.002 (?0.634) 0.0001 (0.617) Log likelihood fn ?36.603 ?129.704 ?94.116 ? 0.268 (17.841)*** 0.286 (15.586)*** 0.374 (12.992)*** No. of observations 180 360 180 T-ratio in parentheses; *** Significant at 1%; ** significant at 5%; * significant at 10%. Source: Author?s computation (2010). The empirical finding is also consistent with the hypothesis that farmers with low technical efficiency may want to improve their resource-use efficiency leading to changes in land management practices. On average, each additional decrease in farmers? technical efficiency increased the proportion of crop residues adoption by 4.4%. Therefore, efforts to increase crop production by farmers with low technical efficiency should focus more on increasing current users. The risk variable is positive as expected but it is not significant for any of the agro- ecological zones considered in the study. 13Balancing livestock needs and soil conservation in West Africa Household size which was used to proxy household labour has a positive and significant influence on the intensity of crop residue use for crop production in Nigeria but is statistically insignificant for Ghana and Niger. The empirical result indicates that larger farm households in Nigeria adopt the use of crop residues for crop production more than those with smaller household sizes. This finding also reveals the effect of household size on household resource allocation behaviour. Larger farm households endowed with available labour assist in the transportation and incorporation of crop residues into the soil. Each additional unit increase in household size increased the proportion of crop residues for crop production by 7.5%. The education variable is positive as expected and significant for Nigeria but insignificant for Ghana and Niger. This is consistent with the human capital theory that households with higher human capital are in a position to understand and appreciate new innovations better and would intensify their crop residue use for crop production compared to those with lower human capital. The intensity of crop residue use for crop production is positively and significantly influenced by farm sizes of households in Nigeria and Niger but in Ghana, it is statistically insignificant. The crop extension variable is positive for Ghana as expected but not in Nigeria and Niger. Since livestock production is heaviest in Maradi and Kano districts, we expect ceteris-paribus, the crop?livestock intensification trade-off to shift from livestock to crop production if the households receive more crop extension visits. The expenditure variable also exhibit the negative apriori sign but it is significant only for Nigeria. The livestock extension variable is negative and significant only for Ghana indicating that farmers who receive less extension in livestock production tend to intensify their crop residue use for crop production at the expense of livestock production. Although livestock production is very relevant to farm households in northern Ghana but relatively not as high as compared to Maradi of Niger and Kano of Nigeria, it was not surprising that farmers in Ghana who receive less livestock extension services tend to shift their intensification strategy from livestock production to crop production. Conclusion This study has analysed the socio-economic characterization, the trade-offs of crop residue allocation between crop and livestock production, and the adoption of crop residue for crop production among farm households who practice crop?livestock farming systems. In particular, the study examined the factors which influence the households? crop residue use intensity for crop production in the subhumid savannah zone of Ghana and semi-arid savannah zones of Nigeria and Niger. The northern region of Ghana is noted for maize?sorghum?cattle system while the pearl millet?cowpea?cattle type of crop?livestock farming is practised by farm households in Kano and Maradi districts. Farm households in Ghana tend to intensify their crop residue use for crop production than for livestock production compared to those in Nigeria and Niger. The empirical results show that 14 Balancing livestock needs and soil conservation in West Africa factors such as efficiency of farm households, education, household size, livestock, and crop extension visits influence the adoption intensity of crop residue use for crop production. The main findings of the study are that efforts to increase crop production through farm households with low technical efficiency should focus more on increasing current users. In Nigeria, farm households endowed with available labour supply tend to facilitate the transportation and incorporation of crop residues into the soil as a soil fertility improvement measure. The trade-off between crop residue use for crop and livestock production among farm households in Nigeria shifts toward crop production when households have high human capital. Although livestock production is very relevant in northern Ghana, less livestock extension visits to farm households shift adoption of crop residue for livestock production to crop production. Although some studies have shown risk attitudes of farmers to influence their decision to choose between alternative uses of crop residue in crop?livestock farming system, our empirical analysis did not find any evidence of this among the sampled farm households in Nigeria, Ghana, and Niger. References Amogu U. 2004. Emerging socio-economic, institutional and policy factors likely to influence future intensification of crop?livestock systems in Nigeria. In: Williams TO, Tarawali SA, Hiernaux P and Fernandez-Rivera S (eds), Sustainable crop?livestock production for improved livelihoods and natural resource management in West Africa. Proceedings of an international conference held at the International Institute of Tropical Agriculture (IITA) Ibadan, Nigeria, 19?22 November 2001. IITA (International Institute of Tropical Agriculture), Ibadan, Nigeria. Barbier EB. 2000. The economic linkages between rural poverty and land degradation: Some evidence from Africa. Agriculture, Ecosystems and Environment 82:355?370. Greene WH. 2008. Econometric analysis. 6th Edition. Upper Saddle River, Prentice Hall, New Jersey, USA. Kristjanson P, Tarawali S, Okike I, Singh BB, Thornton PK, Manyong VM, Kruskal RL and Hoogenboom G. 2002. Genetically improved dual-purpose cowpea: Assessment of adoption and impact in the dry savannah region of West Africa. ILRI Impact Assessment Series No. 9. ILRI (International Livestock Research Institute), Nairobi, Kenya. Manyong VM. 2002. Economic research at IITA for the improvement of agriculture in the subhumid and humid zones of West Africa. In: Sakurai T, Furuya J and Takagi H (eds), Economic analyses of agricultural technologies and rural institutions in West Africa: Achievements, challenges, and application to rice farming research. Proceedings of JIRCAS International Workshop, 12?13 July 2001, Tsubuka, Japan. JIRCAS (Japan International Research Center for Agricultural Sciences) Working Paper Report 25. JIRCAS, Tsubuka, Japan. pp. 37?58. Oldeman LR, van Engelen VWP and Pulles JHM. 1990. Extent of human-induced soil 15Balancing livestock needs and soil conservation in West Africa degradation. In: Oldeman LR, Hakkeling RTA and Sombroek WG (eds), World map of the status of human-induced soil Erosion: An explanatory note. 2nd Edition, Annex 5. International Soil Reference and Information Centre, Wageningen, the Netherlands. Smith JW, Naazie A, Larbi K, Agyemang K and Tarawali S. 1997. Integrated crop?livestock systems in sub-Saharan Africa: An option or an imperative? Addis Ababa, Ethiopia. 5 pp. Thornton PK, Kruska RL, Henninger N, Kristjanson PM, Reid RS, Atieno F, Odero A and Ndegwa T. 2002. Mapping poverty and livestock in the developing world. ILRI (International Livestock Research Institute), Nairobi, Kenya. 124 pp. Tiffen M. 2004. Population pressure, migration and urbanization: Impacts on crop?livestock systems development in West Africa. In: Williams TO, Tarawali SA, Hiernaux P and Fernandez-Rivera S (eds), Sustainable crop?livestock production for improved livelihoods and natural resource management in West Africa. Proceedings of an international conference held at IITA, Ibadan, Nigeria, 19?22 November 2001. IITA (International Institute of Tropical Agriculture), Ibadan, Nigeria. pp. 3?27. 16 Balancing livestock needs and soil conservation in West Africa Output 1.2 Trade-offs in agricultural uses of stover and haulms in the dry savannahs of Ghana, Nigeria and Niger A Opoku, R Abaidoo, B Gerard, M Nouri, E Iwuafor, N Karbo, E Grings and EY Safo Introduction Crop residues have enormous potential to improve soil fertility (Bationo et al. 1995) but their use have been limited by keen competition for them as fodder and large amounts needed to achieve optimum crop yields. Currently, there are conflicting reports on the partition of crop residues for either crop or livestock production. Although Delve et al. (2001) recommended that high quality plant materials should be used as mulch and low quality plant materials as fodder, Larbi et al. (2002) maintained that 50?75% of crop residues, regardless of their quality, should be used as mulch and the remaining 25?50% as fodder. Farmers, on the other hand, prefer to feed all the high quality residues to their livestock. To enable farmers make informed decisions on the allocation of crop residues, it is imperative to provide them with information on the quantities of crop or livestock products they give away (trade-off) for allocating more crop residues into livestock or crop production. Trade-off refers to the opportunity costs of selecting one production alternative rather than the other. Smallholder farmers face multiple trade-offs when deciding on the allocation of their available financial, labour and nutrient resources to competing production activities within their farms (Tittonell et al. 2007). Crissman et al. (1998) proposed trade-off analysis as a tool for providing quantitative information to support decision-making on agricultural production systems. Indeed, trade-off analysis has been used to streamline resource allocation in peri-urban vegetable production (Francisco and Ali 2006), resource and labour allocation by smallholder farmers (Tittonell et al. 2007), investments in nitrogen fertilization and weed control (Dimes et al. 2001), and potato productivity and environmental quality (Stoorvogel et al. 2004). Until now, no study has been conducted to quantify the benefits that a crop?livestock farmer may gain or forfeit (trade-offs) for using crop residues as either fodder or mulch. The challenge for stakeholders in sustainable crop?livestock production is to quantify these trade-offs and recommend optimum rates of crop residues for soil application and livestock feeding. There is a widespread non-adoption of previous technologies on the use of organic materials for soil fertility in sub-Saharan Africa (Palm et al. 1997; Nandwa and Bekunda 1998; Palm et al. 2001). Conventionally, the evaluation of many agricultural technologies has been based on agronomic efficiency, yet agronomic effectiveness alone does not determine the actual usefulness of a technology to a farmer. Certainly, to motivate farmers to inculcate emerging ?best-fit? technologies into their practices, there is an urgent need to involve them in assessing the sustainability of these technologies in terms of their agronomic superiority, economic viability, environmental friendliness, and social acceptability. 17Balancing livestock needs and soil conservation in West Africa Materials and methods Two researcher-managed on-farm experiments were conducted to quantify crop?livestock benefit trade-offs. These involve measuring the benefits that a farmer loses from crop production for a unit gain in livestock produce by feeding more crop residues to livestock rather than incorporating them into the soil. Five scenarios of allocating legume haulms (H) and cereal stover (S) for soil application (SA) and livestock feeding (LF) were evaluated as follows: Scenario 1: 0% SA (0% H, 0% S) vs. 100% LF (100% H, 100% S) ? Scenario 2: 50% SA (25% H, 75% S) vs. 50% LF (75% H, 25% S)? Scenario 3: 50% SA (50% H, 50% S) vs. 50% LF (50% H, 50% S)? Scenario 4: 50% SA (75% H, 25% S) vs. 50% LF (25% H, 75% S)? Scenario 5: 100% SA (100% H, 100% S) vs. 0% LF (0% H, 0% S)? The first study monitored the impact of incorporating crop residues into the soil on the productivity of the cropping system while the second study assessed the effect of feeding crop residues to livestock on the productivity of the livestock unit of the farm. Study 1: Effect of crop residue incorporation on productivity of cereal? legume cropping system Sites The study was conducted at two farms each at Cheyohi, Sarauniya, and Garin Labo. The soils at Cheyohi, were Ferric Luvisols (FAO and UNESCO 1994) with loamy texture. At Sarauniya, the soils were Regosol (FAO and UNESCO 1994) with sandy loam texture while in Garin Labo, soils were Eutric Gleysols (FAO and UNESCO 1994) with sandy texture. The selected physical and chemical properties of the soils (0?15 cm) at the beginning of the study are given in Table (1.2.1). Table 1.2.1. Physical and chemical properties of soils at the study sites Soil parameters Location Cheyohi Sarauniya Garin Labo Farmer 1 Farmer 2 Farmer 1 Farmer 2 Farmer 1 Farmer 2 Bulk density (g cm ?3 ) pH (H 2 O) 5.95 6.43 6.14 5.90 6.46 6.13 Organic C (%) 0.49 0.57 0.43 0.48 0.24 0.29 Total N (%) 0.04 0.04 0.03 0.03 0.01 0.01 NO 3 (mg kg ?1 ) 4.74 4.69 4.34 5.23 3.28 7.65 NH 4 (mg kg ?1 ) 0.69 0.74 1.66 1.45 0.68 0.62 Avail P (mg kg ?1 ) 4.41 6.88 12.90 10.03 1.83 3.74 Ca (cmol kg ?1 ) 2.25 2.29 2.04 2.22 1.96 2.01 Mg (cmol kg ?1 ) 1.01 1.04 0.79 0.67 0.47 0.54 K (cmol kg ?1 ) 0.35 0.35 0.32 0.34 0.20 0.20 Na (cmol kg ?1 ) 0.21 0.22 0.21 0.21 0.18 0.19 CEC (cmol kg ?1 ) 3.81 3.91 3.35 3.44 2.81 2.94 18 Balancing livestock needs and soil conservation in West Africa Experimental design Five treatments of legume haulms and cereal stover mix were incorporated into the soil: 0% H 0% S (T1); 25% H 75% S (T2); 50% H 50% S (T3); 75% H 25% S (T4); and 100% H 100% S (T5). The design was a randomized complete block design (RCBD) with three replications. Plot sizes were 20 m ? 10 m at Cheyohi, 30 m ? 4 m at Sarauniya, and 20 m ? 6 m at Garin Labo. Adjacent plots within the blocks were separated by 1 m wide access while blocks were separated by 2 m wide access. Crop residue incorporation Crop residues used for the study were obtained from the selected farms at the end of the cropping season in 2007. Crop residues were weighed into the appropriate proportions, spread evenly on the designated plots, and incorporated manually into the soil. The amounts of crop residues incorporated at various locations are shown in Table 1.2.2. Table 1.2.2. Crop residue application rates Location Crop residue Application rate (kg/ha) 0H 0S 25H 75S 50H 50S 75H 25S 100H 100S Sarauniya Farmer 2 Groundnut haulms 0.0 375.0 750.0 1125.0 1500.0 Maize stover 0.0 1755.0 1170.0 585.0 2340.0 Farmer 3 Groundnut haulms 0.0 525.0 1050.0 1575.0 2100.0 Maize stover 0.0 2648.3 1765.5 882.8 3531.0 Cheyohi Farmer 2 Cowpea haulms 0.0 201.6 403.2 604.8 806.4 Maize stover 0.0 1458.0 972.0 486.0 1944.0 Maize husk 0.0 216.0 144.0 72.0 288.0 Farmer 3 Cowpea haulms 0.0 243.6 487.2 730.8 974.4 Maize stover 0.0 1966.2 1310.8 655.4 2621.6 Maize husk 0.0 326.3 217.5 108.8 435.0 Garin Labo Farmer 2 Cowpea haulms 0.0 266.7 533.3 800.0 1066.7 Millet stover 0.0 2400.0 1600.0 800.0 3200.0 Farmer 3 Cowpea haulms 0.0 250.0 500.0 750.0 1000.0 Millet stover 0.0 2250.0 1500.0 750.0 3000.0 Land preparation and crop management Animal-drawn mould board ploughs and tine harrows were used to prepare plots for seeding during the major rainy season of 2008. The cropping systems practised were maize?cowpea intercropping at Cheyohi; maize?groundnut sole cropping at Sauraniya; and millet?cowpea intercropping at Garin Labo. The amounts of N, P 2 O 5 , and K 2 O applied to cereals and legumes at the selected sites represented two-thirds of the national NPK recommendations specific to the selected systems. Diseases and pests of economic importance to the crops were not encountered during study; consequently, no herbicides or pesticides were applied to the crops. 19Balancing livestock needs and soil conservation in West Africa Biochemical analysis of plant materials Dry matter content of plant samples were determined by drying plant materials at 105?C for 16 h (AOAC 1990). Plant materials were ashed in a muffle furnace at 550?C for 8 hours to determine ash content. Plant samples were wet-digested with a mixture of H 2 SO 4 , selenium, and salicylic acid, the P and N concentrations in the digest were determined using the automated analytical (Technicon Auto-Analyzer II) method of Novozamsky et al. (1983). The total C was determined by the modified wet combustion technique described by Nelson and Sommers (1982). Acid detergent fiber (ADF), the fraction of plant materials containing lignin, cellulose and ash, was obtained by boiling plant samples with sulphuric acid? cetyltrimethyl ammonium bromide solution under reflux conditions for 1 h. Lignin was then determined by oxidizing the ADF with buffered potassium permanganate solution (Anderson and Ingram 1993). The total extractable polyphenols (consisting of hydrolysable tannins, condensed tannins and non-tannin polyphenolics) was determined by the Folin-Denin method (Anderson and Ingram 1993). Total K was extracted by 1 M ammonium acetate and determined by flame emission spectroscopy. Plant samples were categorized using the index proposed by Tian et al. (1995) as PRQI = [1/ (0.423 C/N + 0.439 lignin + 0.138 polyphenols)] ? 100, with the coefficients of C:N, lignin and polyphenol representing their relative contributions to the index. The chemical characteristics and plant residue quality index (PRQI) of the crop residues incorporated into the soils are shown in Table 1.2.3. Table 1.2.3. Chemical characteristics of crop residues incorporated into soil Location Crop residue Quality parameter (%) C/N PRQI Total C Total N Total P Total K Lig* Phenol Sarauniya Farmer 1 Groundnut haulms 47.6 2.2 0.14 2.4 12.4 5.7 21.6 6.5 Maize stover 48.6 0.6 0.03 2.3 7.5 5.1 81.0 2.6 Farmer 2 Groundnut haulms 48.0 2.3 0.24 1.7 12.1 6.2 20.9 6.7 Maize stover 48.1 0.6 0.05 1.3 8.3 6.8 80.2 2.6 Cheyohi Farmer 1 Cowpea haulms 45.0 1.7 0.14 1.4 13.0 6.0 26.4 5.6 Maize stover 45.4 0.3 0.03 0.6 10.0 7.3 151.4 1.4 Maize husk 47.7 0.4 0.05 0.7 5.9 5.4 119.3 1.9 Farmer 2 Cowpea haulms 46.9 1.6 0.27 2.6 13.8 5.3 29.3 5.2 Maize stover 47.9 0.6 0.06 1.5 9.2 5.0 79.9 2.6 Maize husk 47.2 0.3 0.05 0.6 5.1 4.3 157.3 1.4 Garin Labo Farmer 1 Cowpea haulms 42.5 1.4 0.07 1.0 14.6 3.3 30.4 5.1 Millet stover 48.9 0.4 0.05 2.2 11.4 3.8 122.3 1.7 Farmer 2 Cowpea haulms 50.4 2.1 0.12 1.1 9.0 2.6 24.0 6.9 Millet stover 50.1 0.3 0.03 2.6 10.9 3.1 167.0 1.3 * Lig = Lignin. 20 Balancing livestock needs and soil conservation in West Africa Soil physical and chemical analysis Air-dried soil samples were passed through a 2 mm sieve and analysed for particle size ? distribution by the Bouyoucos hydrometer method (Bouyoucos 1926). Bulk density was determined by the core method (Blake and Hartge 1986). ? Soil pH was determined in water (1:1 soil?water ratio). ? Soil organic carbon was determined by the wet combustion method (Nelson and Sommer ? 1975). Soil samples for NH? 4 -N and NO 3 -N determination were extracted with 2 M KCl and analysed with the Technicon Auto-Analyzer II. Total N was analysed by the auto-analyzer after digesting with a mixture of H 2 SO 4, selenium, and salicylic acid. Phosphorus was extracted by Bray 1 method and determined with the auto-analyzer.? Exchangeable acidity was determined in 1 M, KCl extracts by titrating with 0.01 N ? NaOH. The Al in the titrate was complexed with NaF and back titrated with 0.01 N HCl to determine Al levels. Exchangeable bases were extracted with 1 M ammonium acetate. The amounts of Na ? and K in the extract were determined by flame photometry, while atomic absorption spectrophotometry was used to determine the concentrations of Ca and Mg in the extract. Statistical analysis Measured variables and estimated parameters were subjected to analysis of variance for RCBD with three replicates using GenStat discovery edition 12 (Payne et al. 2009). Orthogonal contrast was used to separate treatment means. Study 2: Effect of crop residues intake on productivity of livestock Acquisition of experimental animals This study was conducted in the homesteads of the selected farmers during the dry season of 2007 (December 2007 to March 2008). Thirty male Sahelian sheep (initial live weight = 26.0 (? 2.5)) were bought from a livestock market in Maradi for the study at Garin Labo, while 30 male goats (initial live weight = 11.9 (? 1.4)) were bought from a similar market at Bejuwa (Jigawa State, Nigeria) for the study at Sarauniya. The 30 male sheep (initial live weight = 13.1 (? 1.4)) used for the study at Cheyohi were bought from a livestock market at Savelugu. All the test animals were aged between 12 and 18 months. Experimental design, feeding and management At each farm, 15 animals were blocked according to their initial live weights and assigned to the 5 dietary treatments of legume haulms and cereal stover mix [0% H 0% S (T1); 25% H 75% S (T2); 50% H 50% S (T3); 75% H 25% S (T4); and 100% H 100% S (T5)] corresponding to the 21Balancing livestock needs and soil conservation in West Africa proportion of crop residues not used for soil incorporation. The biochemical composition of the crop residues fed is shown in Table 1.2.4. Table 1.2.4. Chemical composition and digestibility of crop residues fed Location Crop residue Chemical constituents 1 (%) OMD Ash DM CP NDF ADF Lig* Cell Hcel NDFL Sarauniya Farmer 1 Groundnut haulms 11.3 89.3 13.6 52.2 41.4 12.4 28.0 10.8 23.8 555.4 Maize stover 5.9 91.2 4.0 75.0 46.8 7.5 37.5 28.2 10.0 519.3 Farmer 2 Groundnut haulms 10.8 88.7 14.1 46.4 39.0 12.1 25.6 7.4 26.1 573.9 Maize stover 6.5 91.2 3.6 74.9 44.9 8.3 35.5 30.0 11.1 505.5 Cheyohi Farmer 1 Cowpea haulms 6.2 90.0 10.4 51.8 37.6 13.0 26.1 14.2 25.1 553.9 Maize stover 5.6 92.9 2.1 75.6 48.7 10.0 24.0 26.9 13.2 479.4 Maize husk 2.6 90.6 2.1 83.5 40.0 5.9 32.8 43.5 7.1 524.2 Farmer 2 Cowpea haulms 7.4 90.2 9.9 55.0 40.5 13.8 29.9 14.5 25.1 539.5 Maize stover 4.4 91.9 3.9 76.7 46.2 9.2 35.5 30.5 12.0 483.9 Maize husk 1.7 90.2 2.8 83.1 37.2 5.1 31.7 45.9 6.1 558.7 Garin Labo Farmer 1 Cowpea haulms 5.1 89.9 8.8 60.2 47.2 14.6 33.0 13.0 24.3 516.4 Millet stover 5.1 91.4 2.3 76.7 50.0 11.4 38.4 26.7 14.9 460.1 Farmer 2 Cowpea haulms 6.8 89.5 13.3 47.2 29.4 9.0 22.8 17.8 19.1 591.6 Millet stover 5.0 91.9 1.8 80.7 50.9 10.9 40.9 29.8 13.5 443.6 * Lig = Lignin. The experimental design was a RCBD with three replications. Animals were housed individually in roofed pens of 1 m ? 2 m floor spacing. The animals underwent standard quarantine procedures for 14 days before the start of the experiment during which they were injected with antibiotic, drenched with anti-helminths, and treated against acaricides. Crop residues were offered daily at a rate of 50 g DM per kg live weight (Tanner et al. 2001). Haulms and stover were supplied in separate feeders. Crop residues were offered to test animals at 8:00 h; control animals were herded on range lands from 8:00 h to 17:00 h. Water and mineral lick were supplied ad libitum. Rations for all experimental animals were supplemented with 100 g of wheat bran daily except those at Cheyohi. The duration of the feeding trial ranged from 34 to 58 days depending on the amount of crop residues produced. Measurements The quantity of crop residues offered was recorded daily during the study period. The refusals (orts) were collected from the feeders and the floor, and weighed before the morning feeding (0800 hr). After every 14 days, animals were weighed in the morning before feed was supplied, and fitted with fecal bags. The fecal matter collected over 24 hours were emptied into plastic bags, air dried, and stored for chemical analysis. 22 Balancing livestock needs and soil conservation in West Africa Mean animal live weight change per day was determined from the bi-weekly live weights after the two-week adaptation period. Fecal organic matter (FOM) excretion was calculated from the organic matter intake (IOM) and the average organic matter digestibility (OMD) of the stover and haulms as: FOM ( )OMDIOM ??= 1 . The OMD was estimated with the transfer function OMD (g kg ?1 ) ( )NDFLNDF 1479700042.06.607 2 +??= developed by Coleman et al. (2003), where NDF (g kg ?1 ) is the neutral detergent fiber and NDFL is the lignin content of NDF expressed as g lignin kg ?1 NDF. Laboratory analysis Crop residues were analysed for DM, ash contents, ADF, lignin, cellulose, phenols, and total N concentration. Crude protein (CP) was determined as N ? 6.25. Neutral detergent fiber (NDF), a measure of hemicellulose and ADF was analysed using the method of van Soest and Robertson (1985). Hemicellulose contents were calculated as the differences between NDF and ADF. Fecal samples were analysed for DM, N, P, and K. Quantification of trade-offs Trade-off related to the quantities of crop produce was sacrificed by a farmer for a unit benefit from livestock by allocating less than optimum amount of the crop residues into crop production. To account for the contributions from products and by-products of the farm, apparent and true values of the trade-offs were estimated. The apparent trade-offs (ATO) referred to the quantities of grains sacrificed for a unit gain in live weight and was calculated as: ATO where P G is price of a unit quantity of grain; P LW is price of a unit live weight of livestock; Gy max is the mean grain yield attained by applying the optimum amount of crop residues; Gy i is the grain yield attained by applying a given amount of crop residues LWG 1-i is weight gained by feeding the remaining amount of crop residues to livestock The true trade-off (TTO) referred to the quantities of grains and crop residues sacrificed for a unit gain in live weight and manure voided and was calculated as: TTO where P R is price of a unit quantity of crop residues; P M is the price of a unit quantity of manure; Ry max is the mean crop residue yield attained by applying the optimum amount of crop residues; () iLWiG LWGPGyGyP ? ??= 100max ()()()() iMiLWiRiG MPLWGPRyRyPGyGyP ?? ?+??+?= 100100maxmax 23Balancing livestock needs and soil conservation in West Africa Ry i is the crop residue yield attained by applying a given amount of crop residues; and M 1-i is the manure voided by feeding the remaining amount of crop residues to livestock. Farm revenue was calculated as function of the revenue accruing from the sales of grains, crop residues, live weight, and manure. Statistical analysis Data on DM intake, weight gain, and nutrient concentration in fecal samples were subjected to analysis of variance for RCBD with three replicates using GenStat discovery edition 12 (Payne et al. 2009). Orthogonal contrast was used to separate treatment means. Results Effect of crop residue use on grain yield and live weight At Cheyohi, the incorporation of maize stover, maize husk and cowpea haulm gave rise to significantly higher grain yields of maize but had no effect on grain yield of cowpea (Figure 1.2.1) in Farm 1. The amount of crop residues generated by Farmer 1 supported livestock feeding for 48 days. During this period, sheep fed on the rangeland lost 8.3 g/day while those fed crop residues significantly (P < 0.05) increased their live weight by 15?40 g/day (Figure 1.2.1). Figure 1.2.1. Grain yield and live weights measured in the farms at Cheyohi. In Farm 2, the application of maize stover and cowpea haulm had no significant (P > 0.05) effect on the grain yields of maize. No cowpea grain yield was recorded on this farm as the farmer harvested the crop earlier than expected (Figure 1.2.1). The amount of crop residues generated by Farmer 2 supported livestock feeding for 58 days. Sheep fed with crop residues increased their live weight significantly (P < 0.05) by 21?41 g/day compared to animals grazed on the rangelands. 24 Balancing livestock needs and soil conservation in West Africa At Sarauniya, the incorporation of maize stover and groundnut haulms had no significant (P > 0.05) effect on the grain yields of both maize and groundnut (Figure 1.2.2). At Farm 1 in Sarauniya, the amount of crop residues obtained supported livestock feeding for 56 days. Goats used in the study increased their live weights regardless of the source of feed. Weights gained by animals fed with 100%, 75% and 50% were significantly (P < 0.05) higher than animals fed on the rangeland (Figure 1.2.2). Figure 1.2.2. Grain yield and live weights measured at the farms in Sarauniya. At Farm 2, the incorporation of maize stover and groundnut haulms had no significant (P > 0.05) effect on the grain yields of both maize and groundnut (Figure 1.2.2). The amount of crop residues obtained supported livestock feeding for 56 days. Animals grazed on the rangeland attained a marginal growth rate of 3.3 g/day while those fed on maize stover and groundnut haulm grew significantly by 15?58 g/day. At Garin Labo, the incorporation of millet stover and cowpea haulms had no significant (P > 0.05) effect on the grain yields of both millet and cowpea (Figure 1.2.3). No cowpea grain yield was recorded on Farm 2 as the farmer harvested the crop earlier than expected. The amount of crop residues obtained from the study farm supported sheep feeding for a period of 30 to 32 days. As indicated in Figure 1.2.3, weights gained by animals herded on the rangelands were comparable to weights gained by animals fed on the crop residues. Figure 1.2.3. Grain yield and live weights measured in the farms at Garin Labo. S1 S2 S3 S4 S5 0 10 20 30 40 50 60 0C 0M 25C 75M 50C 50M 75C 25M 100C 100M Crop residue incorporated into soil (%) 0 200 400 600 800 1000 1200 S 1 S 2 S 3 S 4 S 5 0 5 10 15 20 25 30 35 40 45 50 0C 0M 25C 75M 50C 50M 75C 25M 1 00C 1 00M 0 1 00 200 300 400 500 600 700 800 Farm 1 (SE) Live weight Maize Groundnut (SE) Live weight Maize Groundnut Crop residue incorporated into soil (%) Liveweight (kg TLU -1 ) Farm 2 Grain yield (kg ha -1 ) Liveweight (kg TLU -1 ) Grain yield (kg ha -1 ) S 1 S 2 S 3 S 4 S 5 0 5 10 15 20 25 30 35 40 0C 0M 25C 75M 50C 50M 75C 25M 100C 100M 0 1 00 200 300 400 500 600 S1 S2 S3 S4 S5 0 5 10 15 20 25 30 35 0 C 0 M 2 5C 75M 50 C 50 M 75C 2 5M 10 0 C 10 0 M 0 100 200 300 400 500 600 700 800 Farm 1 Live weight Millet Cowpea Live weight Millet Liveweight (kg TLU ?1 ) Farm 2 Liveweight (kg TLU ?1 ) Grain yield (kg ha ?1 ) Grain yield (kg ha ?1 ) Crop residue incorporated into soil (%) Crop residue incorporated into soil (%) 25Balancing livestock needs and soil conservation in West Africa Quantification of trade-offs in using crop residue as soil amendment or fodder The quantities of maize and cowpea grains sacrificed and live weights gained by allocating more crop residues into either crop or livestock production at Farm 1 are shown in Table 1.2.5. Allocation of crop residues had no significant (P > 0.05) effect on the apparent trade-offs and the TTO calculated for Farm 1 (Tables 1.2.5 and 1.2.6) and Farm 2 (values not shown). On the basis of the apparent trade-offs assessment, the best case scenario was the incorporation of 25% haulm and 75% stover into the soil; and the feeding of 75% haulm 25% stover to livestock (scenario 2). However, the TTO appraisal identified the incorporation of 75% haulm and 25% stover into the soil; and feeding of 25% haulm 75% stover to livestock (scenario 4) as the best case scenario (Table 1.2.6). Table 1.2.5. Apparent trade-offs for the crop residues use?scenarios tested at Farm 1 in Cheyohi Scenario Grain yields sacrificed Live weight gained Apparent Trade-offs (Kg 200 m -2 ) (? 200 m -2 ) (Kg head -1 ) (? head -1 ) (Kg kg -1) (?/?) Maize Cowpea Totals Maize Cowpea Totals 1 7.27 0.92 8.19 2.18 0.50 2.68 2.00 3.62 4.09 0.74 2 4.25 0.25 4.51 1.28 0.14 1.41 1.40 2.53 3.22 0.56 3 3.09 0.59 3.67 0.93 0.32 1.24 1.03 1.87 3.55 0.66 4 2.73 0.00 2.73 0.82 0.00 0.82 0.73 1.33 3.73 0.62 5 0.00 0.85 0.85 0.00 0.46 0.46 ?0.40 ?0.72 ?2.13 ?0.64 Contrast probabilities F pr 0.089 0.858 0.355 0.089 0.857 0.355 <0.001 <0.001 0.439 0.443 1 vrs 2+3+4+5 0.024 0.541 0.077 0.024 0.541 0.077 <0.001 <0.001 0.49 0.489 2+3+4 vrs 5 0.094 0.496 0.443 0.093 0.496 0.443 <0.001 <0.001 0.092 0.093 3 vrs 2+4 0.833 0.604 0.895 0.834 0.602 0.895 0.665 0.652 0.96 0.965 2 vrs 4 0.502 0.803 0.588 0.501 0.799 0.588 <.001 <0.001 0.939 0.982 SE 1.528 0.70 0.75 0.46 0.38 0.75 0.06 0.11 2.82 0.60 Exchange rate in October 2008 was USD 1 = 1.01 GH ?. Table 1.2.6. True trade-offs for the crop residues use?scenarios tested in farm 1 at Cheyohi Scenario Crop products sacrificed Livestock products benefited True trade-offs (?/?) (Kg 200 m ?2 ) (? 200 m ?2 ) (Kg head ?1 ) (? head ?1 ) Stover Haulms Stover Haulms Grain Total Manure Manure LWG Total 1 17.13 1.63 0.81 0.34 2.68 3.83 15.30 0.59 3.62 4.21 0.91 2 11.66 0.67 0.55 0.14 1.41 2.10 6.93 0.26 2.53 2.79 0.75 3 6.26 0.87 0.29 0.18 1.24 1.72 7.46 0.32 1.87 2.19 0.78 4 6.17 0.00 0.29 0.00 0.82 1.11 9.45 0.36 1.33 1.68 0.66 5 0.00 1.22 0.00 0.26 0.46 0.72 2.47 0.10 ?0.72 ?0.62 ?1.16 Contrast probabilities F pr 0.159 0.807 0.163 0.804 0.355 0.381 <0.001 <0.001 <0.001<0.0010.596 1 vs. 2+3+4+50.052 0.412 0.054 0.414 0.077 0.085 <0.001 <0.001 <0.001<0.0010.575 2+3+4 vs. 5 0.149 0.548 0.149 0.539 0.443 0.487 <0.001 <0.001 <0.001<0.0010.148 3 vs. 2+4 0.632 0.669 0.624 0.667 0.895 0.935 0.151 0.595 0.652 0.79 0.981 2 vs. 4 0.399 0.638 0.403 0.633 0.588 0.541 0.001 0.027 <0.001<0.0010.93 SE 4.35 0.975 0.20550.2041 0.747 1.099 0.376 0.027 0.1103 0.12861.096 Exchange rate in October 2008 was USD 1 = 1.01 GH ?. 26 Balancing livestock needs and soil conservation in West Africa In both analyses, the use of all crop residues as soil amendment and none as fodder (scenario 5), which mimicked the standard farmer practices of leaving all crop residues on the field, was the worst case scenario. Depending on the amount of crop residues incorporated or fed, the farmer sacrificed 66 pesewas to 99 pesewas of crop grains and residues for a cedi benefit from live weight and manure. In scenario 5, where the animals grazed on the rangeland, the farmer sacrificed 72 pesewas of crop grains and residues but lost 62 pesewas of livestock produce (Table 1.2.6). In Sarauniya, the allocation of crop residues had no significant (P > 0.05) effect on the apparent trade-offs and TTO calculated for Farms 1 (data not shown) and 2 (Tables 1.2.7 and 1.2.8). Table 1.2.7. Apparent trade-offs for the crop residues use?scenarios tested at Farm 2 in Sarauniya Scenario Grain yields sacrificed Live weight gained Apparent trade- offs (Kg 200 m ?2 ) (N 200 m ?2 ) Maize Groundnut Total Maize Groundnut Total (Kg head ?1 ) (N head ?1 ) (Kg kg ?1 ) (N/N) 1 2.51 2.69 5.20 136.90 172.03 308.94 2.33 519.78 2.23 0.59 2 0.99 0.79 1.78 54.06 50.43 104.50 1.90 423.25 0.94 0.25 3 0.10 1.79 1.89 5.23 114.69 119.92 1.43 319.29 1.32 0.38 4 0.72 0.76 1.48 39.24 48.64 87.88 0.60 133.66 2.47 0.66 5 0.00 0.00 0.00 0.00 0.00 0.00 0.13 29.70 0.00 0.00 Contrast probabilities F pr 0.604 0.499 0.377 0.602 0.501 0.372 0.014 0.014 0.769 0.579 1 vs. 2+3+4+5 0.161 0.167 0.078 0.161 0.168 0.078 0.012 0.012 0.676 0.600 2+3+4 vs. 5 0.674 0.402 0.416 0.673 0.404 0.403 0.022 0.022 0.941 0.314 3 vs. 2+4 0.618 0.468 0.907 0.617 0.469 0.854 0.689 0.689 0.712 0.269 2 vs. 4 0.877 0.985 0.906 0.876 0.986 0.911 0.034 0.034 0.26 0.685 SE 1.193 1.092 1.731 65 70 102.2 0.361 80.5 1.133 0.2186 Exchange rate in October 2008 was USD 1 = 125.1 N. Table 1.2.8. True trade-offs for the crop residues use ? scenarios tested at Farm 2 in Sarauniya Scenario Crop products sacrificed Livestock products benefited True trade- offs (N/N) (Kg 200 m ?2 ) (N 200 m ?2 ) (Kg head ?1 ) (N head ?1 ) Stover Haulms Stover Haulms Grain Total ManureManureLWG Total 1 4.86 3.42 18.99 98.47 308.94 426.39 11.08 47.59 519.78 567.37 0.75 2 1.32 1.48 5.16 42.73 104.50 152.38 5.46 21.46 423.25 444.71 0.34 3 0.12 2.36 0.47 67.95 119.92 188.34 5.68 27.87 319.29 347.17 0.54 4 1.23 1.06 4.82 30.52 87.88 123.22 5.81 24.57 133.66 158.23 0.78 5 0.00 0.00 0.00 0.00 0.00 0.00 2.32 10.00 29.70 39.71 0.00 Contrast probabilities F pr 0.614 0.516 0.614 0.516 0.372 0.376 <0.001 <0.001 0.014 0.01 0.548 1 vs. 2+3+4+5 0.15 0.193 0.151 0.193 0.078 0.085 <0.001 <0.001 0.012 0.008 0.910 2+3+4 vs. 5 0.795 0.336 0.795 0.336 0.403 0.373 <0.001 <0.001 0.022 0.018 0.743 3 vs. 2+4 0.67 0.538 0.67 0.538 0.854 0.78 0.9 0.004 0.689 0.657 0.587 2 vs. 4 0.98 0.834 0.98 0.834 0.911 0.888 0.371 0.056 0.034 0.037 0.131 SE 2.36 1.382 9.21 39.8 102.2 141.2 0.261 0.983 80.5 80.9 0.572 Exchange rate in October 2008 was USD 1 = 125.1 N. 27Balancing livestock needs and soil conservation in West Africa In Farm 2, both the apparent trade-off and TTO analyses found the use of 25% haulm 75% stover as soil amendment and 75% haulm 25% stover as fodder (scenario 2) to be the best case scenario. The incorporation of all crop residues into the soil and feeding of none to livestock (scenario 5) was found to be the worst case scenario. Depending on the amount of crop residues incorporated or fed, the farmer sacrificed 35 to 78 kobos of grains and crop residues for a naira benefit from live weight and manure (Table 1.2.8). Where animals grazed on the rangeland, the farmer sacrificed neither grains nor crop residues and got 40 naira worth of livestock produce. In Garin Labo, the allocation of crop residues had no significant (P > 0.05) effect on the apparent trade-offs and TTO calculated for the two farms. The standard farmer practices of feeding all crop residues to livestock and leaving none on the field for soil application (scenario 1) was found to be the best case scenario by both apparent trade-offs and TTO assessments. The incorporation of 75% haulm and 25% stover; and feeding 25% haulm and 75% stover to livestock (scenario 4) was found to be the worst scenario. Depending on the amount of crop residues incorporated or fed, the farmer sacrificed 8 to 26 CFA cents of grains and crop residues for 1 CFA franc benefited from live weight and manure (Table 1.2.9). Table 1.2.9. Apparent trade-offs for the crop residues use?scenarios tested at Farm 1 in Garin Labo Scenario Grain yields sacrificed Live weight gained Apparent trade-offs (Kg 200 m ?2 ) (CFA 200 m ?2 ) Millet Cowpea Total Millet Cowpea Total (Kg head ?1 ) (CFA head ?1 ) (kg kg ?1 ) (CFA/CFA) 1 0.29 0.26 0.56 59.33 92.17 151.50 2.27 2241.92 0.25 0.07 2 0.77 0.00 0.77 154.67 0.00 154.67 1.73 1714.41 0.45 0.09 3 0.63 0.02 0.65 126.00 7.00 133.00 1.33 1318.78 0.49 0.10 4 0.00 0.85 0.85 0.00 297.50 297.50 1.47 1450.65 0.58 0.21 5 0.29 0.32 0.61 58.67 112.00 170.67 1.20 1186.90 0.51 0.14 Contrast probabilities F pr 0.567 0.917 0.95 0.568 0.917 0.98 0.564 0.564 0.757 0.594 1 vs. 2+3+4+5 0.639 0.969 0.727 0.64 0.969 0.806 0.155 0.155 0.995 0.891 2+3+4 vs. 5 0.422 0.972 0.603 0.424 0.972 0.742 0.586 0.586 0.233 0.167 3 vs. 2+4 0.993 0.66 0.713 0.992 0.66 0.673 0.659 0.659 0.891 0.868 2 vs. 4 0.178 0.43 0.763 0.179 0.43 0.899 0.702 0.702 0.673 0.47 SE 0.594 0.723 0.883 118.8 253.2 267.1 0.475 469.8 1.347 0.1741 Exchange rate in October, 2008 was 1$ = 437.8 CFA franc. 28 Balancing livestock needs and soil conservation in West Africa Table 1.2.10. True trade-offs for the crop residues use ?scenarios tested at Farm 1 in Garin Labo Scenario Crop products sacrificed Livestock products benefited True trade-offs (CFA / CFA) (Kg 200 m ?2 ) (CFA 200 m ?2 ) (Kg head ?1 ) (CFA head ?1 ) Stover Haulms Stover Haulms Grain Total Manure Manure LWG Total 1 1.80 0.37 13.24 31.02 151.50 195.76 11.78 127.40 2241.92 2369.32 0.08 2 2.63 0.00 9.79 0.00 154.67 164.46 5.66 75.12 1714.41 1789.53 0.09 3 2.07 0.27 5.15 22.56 133.00 160.71 5.96 75.01 1318.78 1393.79 0.12 4 0.00 1.23 0.00 104.34 297.50 401.84 6.25 66.15 1450.65 1516.80 0.26 5 0.71 0.47 1.77 39.48 170.67 211.92 3.71 45.67 1186.90 1232.57 0.17 Contrast probabilities F pr 0.668 0.948 0.163 0.948 0.98 0.984 <0.001 <0.001 0.564 0.511 0.799 1 vs. 2+3+4+5 0.578 0.923 0.448 0.923 0.806 0.863 <0.001 <0.001 0.155 0.132 0.978 2+3+4 vs. 5 0.525 0.98 0.294 0.98 0.742 0.785 <0.001 <0.001 0.586 0.557 0.253 3 vs. 2+4 0.96 0.806 0.444 0.806 0.673 0.692 0.988 0.409 0.659 0.666 0.813 2 vs. 4 0.234 0.461 0.038 0.461 0.899 0.797 0.146 0.16 0.702 0.694 0.829 SE 1.851 1.126 5.84 95.2 267.1 354.8 0.256 4.1 469.8 471.9 0.444 Exchange rate in October 2008 was USD 1 = 437.8 CFA franc. Trade-offs and farm revenue relations A strong negative relationship (P < 0.001) was found between TTO and the farm revenue in all the selected farms. At Farm 1 in Cheyohi and Farm 2 in Sarauniya, the trade-offs accounted for 87% of variations in the farm revenues accruing from the scenarios tested (Figure 1.2.4). On the other hand, 80% of fluctuations in the farm revenue of Farm 1 in Garin Labo could be attributed to the trade-offs. Trade-off components and crop residue input relations The proportion of maize stover and husk incorporated into the soil did not affect the grain yield and crop residue yield (Table 1.2.11). The amount of haulm incorporated and the quantities of N, P, and K supplied through crop residue application significantly correlated with maize grain yield, total grain yield, and total crop residue yield. While about 93% of the variations in maize yield could be attributed to the linear effect of the amount of haulm incorporated, only 2% of the variation in cowpea yield was due to the incorporation of haulm. Table 1.2.11. Correlation coefficient (r) for crop produces at Farm 1 in Cheyohi Parameter Amount of crop residue and nutrient applied (kg 200 m ?2 ) Stover Haulm Husk Total N Total P Total K Maize grain (kg 200 m ?2 ) 0.781 0.963 0.781 0.997 0.966 0.975 (0.119) (0.008) (0.119) (<.001) (0.007) (0.005) Cowpea grain (kg 200 m ?2 ) ?0.048 0.156 ?0.048 0.096 0.051 0.060 (0.939) (0.803) (0.939) (0.878) (0.935) (0.924) Total grain yield (kg 200 m ?2 ) 0.752 0.959 0.752 0.983 0.946 0.956 (0.143) (0.010) (0.143) (0.003) (0.015) (0.011) Total residue yield (kg 200 m ?2 ) 0.685 0.977 0.685 0.973 0.915 0.929 (0.202) (0.004) (0.202) (0.005) (0.029) (0.023) 29Balancing livestock needs and soil conservation in West Africa A = Farm 1 in Cheyohi; B = Farm 2 in Sarauniya; C = Farm 1 in Garin Labo. Figure 1.2.4. Revenue?trade-off relationships. Both the amount of haulm offered to livestock and the quantity ingested correlated significantly with live weight but not with fecal output. Feeding of stover, though had no effect on live weight, correlated significantly with fecal output. The quantities of CP and NDF ingested correlated significantly with both live weight and fecal output (Table 1.2.12). However, CP intake exerted a stronger effect (r 2 = 0.97) on live weight than fecal output (r 2 = 0.79). On the contrary, the linear effect of NDF ingested was stronger on fecal output (r 2 = 0.98) than live weight (r 2 = 0.77). Table 1.2.12. Correlation coefficient (r) for livestock produces at Farm 1 in Cheyohi Parameter Offer rate (g day ?1 ) Intake rate (g day ?1 ) Stover Haulm Husk Stover Haulm Husk CP NDF Weight gain (kg h ?1 ) 0.732 0.970 0.732 0.694 0.964 0.737 0.985 0.877 (0.160) (0.006) (0.160) (0.193) (0.008) (0.155) (0.002) (0.051) Manure (kg h ?1 ) 0.956 0.785 0.956 0.921 0.793 0.970 0.896 0.990 (0.011) (0.116) (0.011) (0.026) (0.110) (0.006) (0.040) (0.001) Rainfall pattern and crop water requirement The total rainfall collected during the cropping season in Garin Labo was 376 mm. The total requirement for a 100-day millet crop was 480 mm leading 21% moisture deficit in millet water y = -2.23 (0.28)x + 13.30 (0.26) p < 0.001; R2 = 0.87 0 2 4 6 8 10 12 14 16 ?1 ?0.5 0 0.5 1 1.5 2 True tradeoffs (GHc/GHc) y = -401.2 (45.5) x + 1638.6 (40.5) p < 0.001; R2 = 0.87 0 500 1000 1500 2000 ?0.6 ?0.4 ?0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 True tradeoff (Naira/Naira) Farmer revenue (GHC) Farmer revenue (Naira) y = ?2578.6 (377)x + 4441.5 (166) p < 0.001; R2 = 0.7957 0 1000 2000 3000 4000 5000 6000 7000 ?0.5 0 0.5 1 True tradeoff (CFA/CFA) Farmer revenue (CFA) A B C 30 Balancing livestock needs and soil conservation in West Africa requirement (Figure 1.2.5). A cowpea crop intercropped with the millet a month after sowing accessed only 205 mm of rainfall; yet, a 95-day cowpea crop required 356 mm of rainfall for its growth and development. Consequently, by practicing intercropping, only 57% of the water requirement of cowpea was met (Figure 1.2.5). However, farmers could satisfy the water requirement of cowpea by planting the crop as sole crop together with millet crop at the onset of the growing season. Figure 1.2.5. Rainfall pattern and crop water requirement. Discussion The observation that grain yield of maize increased with increasing amount of crop residues in Cheyohi (northern Guinea savannah) affirms the findings of Larbi et al. (2002) that along the transect from humid forest to the northern Guinea savannah, grain yield of maize increased with mulching rate. The lack of response of millet and cowpea to crop residue application in the Sahel savannah supports the conclusion of Giller et al. (2009) that crop residue management can result in yield benefits in the long-term. However, in the short-term, yield losses or no yield benefits may result. In studies where positive responses to crop yield were observed in the short-term, they were attributed to the improved rainwater use efficiency through improved infiltration and reduced evaporative water losses (Giller et al. 2009) and mobilization of soil P through the release of organic acids from the decomposing residue (Hue 1991). Nutrient immobilization (Larbi et al. 2002), occurrence of residue-borne diseases, and poor germination (Giller et al. 2009) have been cited as factors responsible for the often-observed short-term yield reductions. The results of our study indicated that the application of haulms could be a viable strategy for increasing the grain yield of maize in the northern Guinea savannah. However, approaches 0 20 40 60 80 100 120 0 20 40 60 80 100 120 Days after planting ETc Rainfall 1 2 3 4 0 20 40 60 80 100 120 0 20 40 60 80 100 120 140 Days after planting ETc intercrop Rainfall ETc sole crop 1 2 3 4 1 2 3 4 Initial growth period Crop development stage Mid-season period Late-season period Rainfall evapo- transpiration (mm) Millet Rainfall evapo- transpiration (mm) Cowpea 31Balancing livestock needs and soil conservation in West Africa other than crop residue management may be required to increase the grain yield of cowpea. The strong positive correlation between haulm intake and live weight found in our study while confirming the findings of Faftine et al. (1998) and Ayantunde et al. (2007) suggest that mutton production could be increased dramatically by increasing the proportion of haulm fed to small ruminants. In addition, CP intake influenced live weight better than NDF. Following the report by Savadogo et al. (2000) that upper part of the cereal stover is more digestible and has a higher concentration of CP than the lower parts, improvement in live weights could be achieved by selective removal of ?stover tops? from the field for livestock feeding. On the other hand, less nutritious ?stover bottoms? are retained on the field to replenish the organic matter of the soil. The trade-offs estimated in this study had a strong negative relationship with farm revenue affirming the fact that the smaller the trade-off, the better the crop residue allocation option. The legume haulm by virtue of the low C:N ratio, high concentration of CP, and high digestibility exerted a significant impact on both crop and livestock production units of the farm. The trade- offs indicated that farmers in the northern Guinea savannah where crop?livestock integration is low obtained the highest farm revenue by allocating lower amount (25%) of the haulm for livestock feeding and retaining a higher amount for soil incorporation (75%). Due to the lack of response to crop residue incorporation in the dry savannah agro-ecological zones, the highest farm revenue was obtained when more haulm (75% in Sudan savannah and 100% in the Sahel) was fed to livestock rather than incorporating it into the soil. In addition to the well-known lack of response to crop residue application in the short-term (Giller et al. 2009), the poor workability of the soils in savannahs during the dry season made manual incorporation of crop residues ineffective and allowed free roaming animals to graze the residues applied. The current trade-offs for allocating crop residues between the crop and livestock units of the farm may be improved by adopting proactive measures, which would increase the productivity of the two units. Firstly, by planting improved dual purpose legumes in rotation with other crops rather than as intercrop, the water requirement of the legume could be satisfied to supply farmers with quality crop residues for both soil application and livestock feeding. Secondly, as the quantity and distribution of rainfall is a major biophysical constraint to agriculture in the dry savannahs, improved soil water conservation practices (i.e. surface mulching and tied ridging) are important to improve crop productivity. Lastly, intake of stover in our study was 30?52% as opposed to 80?100% intake of haulm. Considering that stover forms the bulk of the crop residues at the disposal of farmers, strategies such as milling or chopping and treating stover with palatable feed ingredients are warranted to improve stover intake. Short-term benefits are important to attract farmers to crop residue management; yet, a significant effect of the application of crop residues on crop yield may require several seasons of continuous practice. Livestock, on the other hand, respond instantaneously to crop residues 32 Balancing livestock needs and soil conservation in West Africa rations. Besides, while the residual effect of crop residues on the crop yields may last for seasons, no such residual effects are found on the live weights of livestock. The vital importance of research on the trade-offs in the alternative uses of crop residues is to determine the appropriate time-frame that would allow the impact of crop residue application on the cropping system to be evaluated in a holistic manner. Conclusion Trade-off analysis is a useful decision-making tool for providing information on the profitability of crop residue allocation options. The use of crop residues as fodder for livestock increased livestock productivity; however, soil amendment crop residues had little or no effect on crop productivity. Though the trade-offs calculated for the five scenarios were not significantly different on all the study farms, farmers in northern Guinea could improve their farm revenue by using 25% of haulm and 75% of stover as fodder; and 75% of haulm and 25% of stover as soil amendment. In Sudan savannah, farm revenue could increase by using 75% of haulm and 25% of stover as fodder; and 25% of haulm and 75% of stover as soil amendment. Finally, in the Sahel savannah, higher farm revenues were achieved by feeding all residues to livestock and incorporating none into the soil. The potential pathways for improving the prevailing trade-offs identified in the study were: the use of improved dual purpose legumes in appropriate rotation to legume biomass yield; application of surface mulch and tied ridging to improve soil water storage; and processing of stover to enhance palatability and intake. References Anderson JM and Ingram JSI. 1993. Tropical soil biology and fertility: A handbook of methods. CAB International, Oxon, UK. AOAC (Association of Official Analytical Chemists). 1990. Official methods of analysis. 15th edition. AOAC, Washington, DC, USA. pp. 69?88. Ayantunde AA, Delfosse P, Fernandez-Rivera S, Gerard B and Dan-Gomma A. 2007. Supplementation with groundnut haulms for sheep fattening in the West African Sahel. Tropical Animal Health Production 39:207?216. Bationo A, Buerkert A, Sedogo MP, Christianson BC and Mokwunye AU. 1995. A critical review of crop residue use as soil amendment in the West African semi-arid tropics. In: Powell JM, Fernandez-Rivera S, Williams TO and Renard C (eds), Livestock and sustainable nutrient cycling in mixed farming systems of sub-Saharan Africa. Volume II. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. pp. 305?322. 33Balancing livestock needs and soil conservation in West Africa Blake GR and Hartge KH. 1986. Bulk density. In: Klute A (ed), Methods of soil analysis. Part I. Physical and mineralogical methods. Agronomy monograph No. 9 (2nd ed.). pp. 363?375. Bouyoucos GJ. 1926. The hydrometer as a new method for the mechanical analysis of soils. Soil Science 23:334?352. Crissman CC, Espinosa P, Ducrot C, Cole DC and Carpio F. 1998. The case study site: Physical, health and potato farming systems in Carchi Province. In: Crissman CC, Antle JM and Capalbo SM (eds), Economic, environmental, and health trade-offs in agriculture: Pesticides and the sustainability of andean potato production. Kluwer Academic Publishers, Boston, USA. pp. 85?120. Delve RJ, Cadisch G, Tanner JC, Thorpe W, Thorne PJ and Giller KE. 2001. Implications of livestock feeding management on soil fertility in the smallholder farming systems of sub- Saharan Africa. Agric. Ecosystem Environment 84:227?243. Dimes J, Muza L, Malunga G and Snapp S. 2001. Trade-off between investments in nitrogen and weeding: On-farm experimentation and simulation analysis in Malawi and Zimbabwe. 7 th Eastern and Southern Africa regional maize conference. pp. 452?456. Faftine O, Muir JP and Massaete E. 1998. Dry season supplementation of goats tethered on range supplemented with cowpea or groundnut residues with or without fresh leucaena leaves. Animal Feed Science Technology 76:1?8. FAO and UNESCO. 1994. Soil map of the world. (revised). World Resources Report 60. FAO, Rome, Italy. Francisco SR and Ali MM. 2006. Resource allocation trade-offs in Manila?s peri-urban vegetable production systems: An application of multiple objective programming. Agricultural Systems 87:147?168. Franzluebbers K, Juo ASR and Manu A. 1994. Decomposition of cowpea and millet amendments to a sandy Alfisol in Niger. Plant and Soil 167:255?265. Giller KE, Witter E, Corbeels M and Tittonell P. 2009. Conservation agriculture and smallholder farming in Africa: The heretics? view. Field Crops Research 114:23?34. Hue NV. 1991. Effects of organic acids/anions on P sorption and phytoavailability in soils with different mineralogies. Soil Science 152:463?471. Larbi A, Smith JW, Adekunle IO, Agyare WA, Gbaraneh LD, Tanko RJ, Akinlade J, Omokaye AT, Karbo N and Aboh A. 2002. Crop residues for mulch and feed in crop?livestock systems: Impact on maize grain yield and soil properties in the west African humid forest and savanna zones. Expl Agric. 38: 253?264. Nandwa SM and Bekunda MA.1998. Research on nutrient flows and balances in East and Southern Africa: State-of-the-Art. Agriculture, Ecosystems and Environment 71:5?8. Nelson DW and Sommers LE. 1975. A rapid and accurate method for estimating organic carbon in soil. Proceedings of the Indiana Academy of Science 84:456?462. 34 Balancing livestock needs and soil conservation in West Africa Nelson DW and Sommers LE. 1982. Total carbon, organic carbon and organic matter. In: Page AL, Miller RH and Keeney DR (eds), Methods of soil analysis. Part 2. American Society of Agronomy, Madison, USA. pp. 539?579. Novozamsky I, Houba VJG, van Eck R and van Vark W. 1983. A novel digestion technique for multi-element plant analysis. Communications in Soil Science and Plant Analysis 14(3):239? 248. Palm CA, Myers RJK and Nandwa SM. 1997. Organic?inorganic nutrient interactions in soil fertility replenishment. In: Buresh RJ, Sanchez PA and Calhoun F (eds), Replenishing soil fertility in Africa. Special publication no. 51. Soil Science Society of America. Madison, Wisconsin, USA. pp. 193?218. Palm CA, Gachengo CN, Delve RJ, Cadisch G and Giller KE. 2001. Organic inputs for soil fertility management in tropical agro-ecosystems: Application of an organic resource database. Agric. Ecosyst. Environ 83:27?42. Payne RW, Murray DA, Harding SA, Baird DB and Soutar DM. 2009. GenStat for Windows (12th Edition) Introduction. VSN International, Hemel Hempstead, UK. 310 pp. Savadogo M. 2000. Crop residue management in relation to sustainable land use. PhD thesis. Wageningen Agricultural University, the Netherlands. van Soest PJ and Robertson JB. 1985. Analysis of forages and fibrous foods. A laboratory manual for animal science. Cornel University, Ithaca, New York, USA. 202 pp. Stoorvogel JJ, Antle JM and Crissman CC. 2004. Trade-off analysis in the Northern Andes to study the dynamics in agricultural land use. Journal of Environmental Management 72:23? 33. Tian G, Brussaard L and Kang BT. 1995. An index for assessing the quality of plant residues and evaluating their effects on soil and crop in the (sub-) humid tropics. Appl Soil Ecol 2:25?32. Tittonell P, van Wijk MT, Rufino MC, Vrugt JA and Giller KE. 2007. Analyzing trade-offs in resource and labor allocation by smallholder farmers using inverse modeling techniques: A case-study from Kakamega district, western Kenya. Agricultural Systems 95:76?95. 35Balancing livestock needs and soil conservation in West Africa Output 2 Identifying areas of intervention and entry points through which appropriate crop?livestock integration technologies can stimulate the intensification of crop?livestock systems Output 2.1 Identifying entry points for improving the productivity of cereal?legume?livestock systems: The NUTMON approach A Opoku, R Abaidoo, M Nouri, E Iwuafor, N Karbo, E Grings and EY Safo Introduction A revolution in the productivity of smallholder farms is required to redress the current deficits in food production and breaking the present poverty cycle of low input?low production?low income (World Bank 2007). Apart from the prevailing farming constraints in the savannahs of West Africa, which compel smallholder farmers to rely on low-external inputs strategies for crop and livestock productions, the alarming rate of nutrient mining is a major setback to agricultural productivity (Sanchez et al. 1996; Smaling et al. 1996). On a continental scale, Africa consumes 0.8 million tonnnes of N, 0.26 million tonnes of P, and 0.2 million tonnes of K (FAO 1995) and losses as much as 4.4 million tonnes of N, 0.5 million tonnes of P, and 3 million tonnes of K from its cultivated lands annually (Sanchez et al. 1996). A quantitative knowledge on nutrient flows in such a farming system offers a credible insight into the sustainability of the system, facilitates the identification of the main losses of nutrients from the system; hence, serves as a diagnostic tool to identify entry points through which research could stimulate agricultural productivity. Accordingly, many studies in sub-Saharan Africa (SSA) in the last decade have focused on the quantification and estimation of nutrients that enter and leave the farming systems (Smaling et al. 1996; van den Bosch et al. 1998; Kanmegne et al. 2006). Most of these studies, however, provided quick balance sheet, based on a short time-frame exercise, and depended on a number of assumptions relating to system dynamics. Of concerned is the validity of such assumptions, their degree of reliability, and capability to provide insight into these dynamic processes. Scoones and Toulmin (1998) questioned the credibility of nutrient balance analysis to provide reliable directions and support for policy formulation on resource management. On the contrary, Lynam et al. (1998) provided convincing evidence that nutrient balance formed a template for economic budgeting; hence, a useful tool for understanding the determinants of soil management decisions undertaken by a farmer. The nutrient monitoring (NUTMON) framework is an integrated, multidisciplinary methodology that targets different actors in the process of managing natural resources and is useful in 36 Balancing livestock needs and soil conservation in West Africa assessing soil nutrients balances at the farm-scale (Smaling et al. 1996; van den Bosch et al. 1998). In sum, a thorough audit of nutrient flows in these farming systems and judicious manipulation of the flows to redress the nutrient imbalances may be a plausible pathway for identifying efficient farming technologies and increase agricultural productivity. Conceptual framework A cereal?legume?livestock system is conceptualized as a farming system comprised of a cereal? legume production unit, a livestock production unit, and a homestead through which nutrient transfers take place (Figure 2.1.1). Nutrients may be imported into the farm primarily through feed concentrate, mineral fertilizers, and biological N fixation while export occurs through the sales of livestock and crop products (Watson et al. 2005). In the savannahs of West Africa, deposition of harmattan dust is another important nutrient input into the farming system (Harris 1999). Additional nutrients losses may occur through leaching, erosion, and dentrification (de Jager et al. 1998). Figure 2.1.1. Conceptual framework for nutrient cycling in smallholder cereal?legume?livestock systems. Crop residues RANEGLAND HOMESTEAD Fuel wood Harvest food Fuel wood Intake Feed concentrates Grazing SOIL Manure Symbiotic nitrogen fixation LIVESTOCK Crop residues left in field Ask/ compound waste Compost Inorganic fertilizer Erosion Leaching Harmattan dust CEREAL?LEGUME A B C Sales Sales Nutrient uptake A = Feed degradation by ruminants B = Partitioning of nutrients by ruminants C = Mineralization of nutrients 37Balancing livestock needs and soil conservation in West Africa Nutrients in crop?livestock systems are cycled in several stages, and losses at each stage may decrease the amount of useful output. For example, crop residues may be fed to livestock and the manure generated returned to the cropland. Turner and Hiernaux (2002) found rangeland to be an integral component of the daily grazing orbit of livestock in the dry savannahs as animals are typically kept on a free range. As a result, livestock grazing on rangelands may import nutrients onto croplands when the manure deposited in confinement either through kraaling or night parking is used in crop production (Harris 2002). Alternatively, nutrients in crop residues may be taken up by the subsequent crop to produce biomass and grain when left on the field after harvest (Powell et al. 2004). Nonetheless, in the dry savannahs, a substantial amount of crop residues left on the field may be lost as a result of bush fires, strong winds, termites, free roaming animals, or transhumant herds of Fulani cattle. Carsky and Ndikawa (1998) reported that 4 Mg ha ?1 of Mucuna biomass disappeared during the dry season due to wind and termite activities. Materials and methods Characterization of households Nine case study farms were selected in Garin Labo to represent three socio-economic groups of farmers (rich, medium, and poor-resource) with three farmers in each group. In Saurniya and Cheyohi, the study was conducted on three case study farms with one farmer from each socio-economic group. The test farmers were selected on the basis of their resource endowment, interest in learning, and capacity to exchange information with their peers. Categorization of households into socio-economic groups was based on local wealth ranking exercise centred on ownership of draught oxen, donkeys, livestock herds, and cultivated crop land (Table 2.1.1). Differentiation of households into the socio-economic group was undertaken before data collection. The rich farmers group was also called ?equipped crop? livestock farmers? while the medium-resourced farmers were referred to as unequipped crop? livestock farmers. The poor-resourced farmers, on the other hand, were referred to as ?crop only farmer.? Table 2.1.1. Resource profile of households? categories Criteria Rich Medium Poor Draught animal + equipments 2 units 1units 0 Cattle (number) >2 1?2 0 Small ruminants (number) >20 11?20 0?10 Total herd size (TLU) >2 1?2 <1 Total land holding (ha) >5 2?5 1?2 Quantification of nutrient flows Nutrient flows managed by farmers A survey was conducted from March to October 2007 in the 15 selected households to collect information on nutrient flows managed by farmers. Farmers gave information on 38 Balancing livestock needs and soil conservation in West Africa the different production units, land use, major farm products, and their destinations. The inflows investigated were the quantities and types of mineral fertilizers (IN 1) and manure, feedstuffs, and concentrates entering the farm annually (IN 2). The outflows were crop products (OUT 1) and residues (OUT 2) leaving the farm annually for the homestead use, sold, or given as gifts. Farmers generally gave quantities in their own units, such as sacks, bags and buckets, which were converted to standard metric amounts. Samples of the different inputs and products were collected and analysed for their N, P, and K concentrations. Environmental nutrient inflows Nutrient inflows such as atmospheric deposition and biological nitrogen fixation were estimated from transfer functions derived from climate and soil data of the sites. The combined wet and dry atmospheric deposition (IN 3), was calculated using the transfer function developed by Stoorvogel and Smaling (1990), in which IN 3 N , IN 3 P , IN 3 K is the input of N, P and K (kg ha ?1 yr ?1 ) and p is the mean annual precipitation (mm yr ?1 ), as follows: IN 3 N = IN 3 P = IN 3 K = Biological nitrogen fixation (IN 4) in the crop production systems was estimated from the general equation: IN 4 (N) = [(A L ? IN 4a) + (A F ? IN 4b)] ? [A F ] ?1 Where A L is the area of legume field, A F is the farm size, IN 4a is the symbiotically fixed and IN 4b the non-symbiotically fixed nitrogen. It was assumed that 60% of the total N demands of groundnut and cowpea are supplied through symbiotic nitrogen fixation (Stoorvogel and Smaling 1990). IN 4a = Non-symbiotic nitrogen fixation was estimated from the function (Smaling et al. 1993): IN 4b = where: N G and N H are quantities of N accumulated in grain and haulm, respectively; and Y G and Y H being grain yield and haulm yield, respectively. 2/1 14.0 p 2/1 023.0 p 2/1 092.0 p ()(){}(){} [ ]005.0135026.0 ??++??+? pYNYN HHGG (){}[]005.013502 ??+ p 39Balancing livestock needs and soil conservation in West Africa Estimation of environmental nutrient outflows Leaching of soil N and K below the root zone (OUT 3) were calculated. In tropical soils P is tightly bound to soil particles; as a result, P outflow due to leaching was assumed to be negligible. The quantities of N lost annually through leaching (kg ha ?1 yr ?1 ) was estimated from the transfer function developed by de Willigen (2000): OUT 3 N = where: p is annual precipitation (mm yr ?1 ); C is the clay content of the top soil (%); L is rooting depth (m); Nf is N derived from the application of mineral and organic fertilizer (kg ha ?1 ); Oc is organic carbon content of the top soil (%); and Nu is N uptake by the crop (kg ha ?1 yr ?1 ). The amount of K lost annually through leaching (kg ha ?1 yr ?1 ) was calculated using the transfer function developed by Smaling (1993) as follows: OUT 3 K = where Ke is the exchangeable K (cmol kg ?1 ) in the top soil and Kf is the amount of K derived from mineral fertilizer. The loss of gaseous N (kg ha ?1 yr ?1 ) from the soil (OUT 4) was calculated by multiplying the percentage of N lost through denitrification (DN) by the amount of N supplied through fertilizer application and soil mineralization as follows: OUT 4 = where Ns is mineralized N in the rootable zone (kg ha ?1 ), Nf is N applied with mineral and organic fertilizer (kg ha ?1 ). Ns is determined from soil total N and the annual relative mineralization rate (M) estimated at 3% (Nye and Greenland 1960). DN is a function of clay content of the top soil, C (%), and the annual rainfall p (mm yr ?1 ), through the transfer function (Smaling et al. 1993): DN = () )NuOcNfL C p ???+???+ 00362.00000601.00037.037.21 ()()41.000029.0 +??+ pKfKe ()DNNfNs ?+ MNtotNs ??= 20 pC 01.013.04.9 +?+? 40 Balancing livestock needs and soil conservation in West Africa Nutrient balance The nutrient balance was calculated without nutrient input through sedimentation (IN 5) since the cropping systems in study did not employ irrigation. Also, nutrient losses through erosion (OUT 5) were not included as slope angles measured on the test farms were less than 0.5%. The nutrient balance was estimated as: Nutrient balance = IN 1 + IN 2 + IN 3 + IN 4 ? OUT 1 + OU T2 + OUT 3 + OUT 4 Results Nitrogen flows and balances in cereal?legume?livestock systems All the selected farmers in Garin Labo received 50 kg/ha of urea from the SLP team in Niger. As a result, the socio-economic status of the farmers had no significant effect on their N inputs although equipped crop?livestock farmers supplied more N through manure than the other farmer groups (Figure 2.1.2a). Equipped crop?livestock farmers also lost significantly higher amount of N (21 kg/ha) through crop residue than crop only farmers (15 kg/ha). All farmers regardless of their socio-economic status suffered similar losses of N from crop produce and leaching. Figure 2.1.2a. Nitrogen flows in cereal?legume?livestock systems at farm level in Garin Labo. Following the existing fertilizer recommendations for the study locations, farmers in Sarauniya applied more N through mineral fertilizers than those in Cheyohi and Garin Labo (Figure 2.1.2b). Nitrogen inputs through the manure application, atmospheric deposition, and BNF also differed significantly across the study locations (Figure 2.1.2b). Groundnuts supplied significantly higher amount of N through BNF than cowpea in either Cheyohi or Garin Labo. Farmers in Sarauniya lost significantly higher amount of N through harvested crop produces and residues than farmers in Cheyohi and Garin Labo (Figure 2.1.2b). OUT 1 IN4IN3IN2IN1 ?30 ?20 ?10 0 20 30 OUT 2 OUT 3 OUT 4 Crop only Crop?livestock unequipped Crop-livestock equipped (LSD 0.05 ?OUT2) N (kg ha ?1 yr ?1 ) Nutrient flows 10 41Balancing livestock needs and soil conservation in West Africa IN 1 IN 2IN 3 IN 4 OUT 1 OUT 1 OUT 1 OUT 1 N (kg ha ?1 yr ?1 ) (LSD 0.05 ) Nutrient flows Cheyohi Sarauniya Garin labo ?120 ?80 ?40 0 40 80 120 Figure 2.1.2b. Nitrogen flows in cereal?legume?livestock systems at village level. Under the current farmer practice, where all crop residues are removed from the field, N balances were negative (?6.9 to ?18.6 kg/ha) on all farms (Figure 2.1.3a) in Garin Labo. In a scenario where Farmers 4 and 9 incorporated half of their residues, Farmer 4 defray the negative balance by 8 kg/ha while Farmer 9 attained a positive balance (Figure 2.1.3a). In the absence of fertilizer application, highly negative (?20.3 to ?40.2 kg/ha) N balances were obtained on all fields (Figure 2.1.3b). Figure 2.1.3a. N balances at farm level in Garin Labo with N fertilizer application. As shown in Figure 2.1.4a, N balance across the study locations was more negative in Sarauniya (?22.0 kg/ha) than in either Cheyohi (?6.5 kg/ha) or Garin Labo (?10.8 kg/ha). In scenarios where farmers do not apply mineral fertilizer, highly negative balances (?33.83 to ?81.85 kg/ha) were obtained (Figure 2.1.4b). Whether farmers applied mineral fertilizers or not, the N balances estimated for these villages improved dramatically with the incorporation of half of the crop residue produced (Figures 2.1.4a and 2.1.4b). Typo 3Typo 2Typo 1 9* 9 4* 4 8 7 65 3 21 ?100 ?80 ?60 ?40 ?20 0 20 40 60 80 Farmers Total IN Total OUT BAL N (kg ha ?1 yr ?1 ) *Farmers incorporated half of the crop residue into the soil 42 Balancing livestock needs and soil conservation in West Africa Figure 2.1.3b. N balances at farm level in Garin Labo without N fertilizer application. Figure 2.1.4a. N balances at village-level with N fertilizer application. Figure 2.1.4b. N balances at village-level without N fertilizer application. N (kg ha -1 yr -1 ) GL* SN* CH* SN CH GL (LSD 0.05 total input) ?160 ?120 ?80 ?40 0 40 80 120 160 Selected villages Total IN Total OUT BAL CH = Cheyohi SN = Sarauniya GL = Garin Labo *Farmers incorporated half of the crop residue into the soil N (kg ha ?1 yr ?1 ) Selected villages Total IN Total OUT BAL GL CH SN CH* SN* GL* ?160 ?120 ?80 ?40 0 40 80 120 160 (LSD 0.5 total input) (LSD 0.5 total input) CH = Cheyohi SN = Sarauniya GL = Garin Labo *Farmers incorporated half of the crop residue into the soil Total IN Total OUT BAL N (kg ha ?1 yr ?1 ) Typo 3Typo 2Typo 1 9* 9 4* 4 8 7 65 3 21 ?100 ?80 ?60 ?40 ?20 0 20 40 60 80 Farmers *Farmers incorporated half of the crop residue into the soil 43Balancing livestock needs and soil conservation in West Africa Phosphorus flows and balances in cereal legume livestock systems The socio-economic status of the farmers had no significant effect on the amount of P supplied by the farmers into the cereal?legume unit of the farm (Figure 2.1.5a). All farmers, regardless of their socio-economic status suffered similar losses of P from crop produce and crop residue. Figure 2.1.5a. P flows in cereal-legume-livestock systems at farm-level in Garin Labo. In accordance with the existing fertilizer recommendations for the study locations, farmers in Sarauniya applied more P through mineral fertilizers than those in Cheyohi and Garin Labo (Figure 2.1.5b). Phosphorus inputs through manure application in Sarauniya and Garin Labo differed significantly from P input via manure in Cheyohi (Figure 2.1.5b). Figure 2.1.5b. P flows in cereal-legume-livestock systems at village-level. Even under the current farmer practice of total crop residue removal, P balances in Garin Labo were positive (3.3 to 7.4 kg/ha) (Figure 2.1.6a). However, by incorporating half of the residues of Farmers 4 and 9, only a marginal improvement in the P balances of Farmer 4 and Farmer 9 were observed. In the absence of SSP application, negative (?0.5 to ?4.6 kg/ha) P balances were obtained on all fields (Figure 2.1.6b). P (kg ha ?1 yr ?1 ) OUT 1 OUT 2 IN 3IN 2IN 1 ?4 ?2 0 2 4 6 8 10 Nutrient flows Crop only Crop livestock unequiped Crop livestock equiped P (kg ha ?1 yr ?1 ) IN 1 IN 2 IN 3 IN 4 OUT 1 OUT 2 ?10 ?5 0 5 10 15 20 25 30 Nutrient flows Cheyohi Sarauniya Garin Labo 44 Balancing livestock needs and soil conservation in West Africa Figure 2.1.6a. P balances at farm-level in Garin Labo with N fertilizer application. Figure 2.1.6b. P balances at farm-level in Garin Labo without N fertilizer application. As indicated in Figure 2.1.7a, P balance across all locations was more positive in Sarauniya (15.7 kg/ha) than in either Cheyohi (9.7 kg/ha) or Garin Labo (5.4 kg /ha). In scenarios where farmers do not apply mineral fertilizer, negative balances (?2.4 to ?7.9 kg /ha) were obtained (Figure 2.1.7b). Figure 2.1.7a. P balances at the village-level with N fertilizer application. P (kg ha -1 yr -1 ) Total IN Total OUT BAL 1 2 3 5 6 7 8 4 4* 9 9* Typo 1 Typo 2 Typo3 ?10 ?5 0 5 10 15 Farmers P (kg ha ?1 yr ?1 ) Total IN Total OUT BAL GL* SN* CH* GL SN CH (LSD 0.95 input) ?20 ?15 ?10 ?5 0 5 10 15 20 25 30 Selected villages (LSD 0.05-bal) P (kg ha ?1 yr ?1 ) 1 2 3 5 6 7 8 4 4* 9 9* Typo 1 Typo 2 Typo3 ?10 ?5 0 5 10 15 Farmers Total IN Total OUT BAL 45Balancing livestock needs and soil conservation in West Africa Figure 2.1.7b. P balances at village-level without N fertilizer application. Entry points for improving cereal?legume?livestock productivity Farmers in Garin Labo used about 20% of the total crop residues generated to satisfy their fuel wood and raw material needs for the construction of granaries, fences, and roofing mats. It leads to an export of 3?4.2 kg/ha of N annually (Figure 2.1.8). Figure 2.1.8. Hot spots for research interventions in cereal?legume?livestock farms in Garin Labo. P (kg ha ?1 yr ?1 ) (LSD 0.05-bal) GL* SN* CH* GL SN CH ?20 ?15 ?10 ?5 0 5 10 15 20 25 30 Selected villages Total IN Total OUT BAL Harvest produce N P K 19.6 20.6 24.3 3.0 2.9 3.1 7.4 7.0 7.7 Sold crop N P K 3.9 4.1 4.9 0.6 0.6 0.6 1.5 1.4 1.5Construction/fuel material N P K 3.0 3.6 4.2 0.2 0.4 0.5 11.8 13.3 13.6 HOMESTEAD N P K 3.9 3.9 3.9 3.9 3.9 3.9 3.9 3.9 3.9 P M R Nitrogen fixation P M R P M R P M R N P M R 8.1 7.6 11.5 CEREAL-LEGUME P M R N P K 11.9 14.6 16.9 0.9 47.2 1.8 53.3 2.0 54.4 P M R N P K 34.5 38.8 45.4 4.2 5.1 5.7 66.4 73.7 75.7 Mulch P M R 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Uptake Mineral fertilizer P M R N P K 23.0 7.9 0.0 23.0 7.9 0.0 23.0 7.9 0.0 1 2 3 4 LIVESTOCK SOIL N P K % % 1.8 0.3 1.2 ? ? ? N P K 0.8 0.1 0.7 Household waste Leaching P M R N P K 20.8 - 4.0 20.2 - 5.2 20.6 - 7.0 Compost Manure Household consumption Fodder P = Poor resourced farmers M = Medium resourced farmers R = Rich resourced farmers 46 Balancing livestock needs and soil conservation in West Africa Also, farmers generally do not retain crop residues on the field for soil fertility restoration. As a result, there is severe nutrient mining of 35?45 kg/ha of N and 66?76 kg/ha of K annually (Figure 2.1.8). Farmers heaped manure at place near the kraal without any protection against the rainfall or sunshine and caused 60% of N in the manure to be lost during storage. Finally, about of 20.2?20.8 kg/ha corresponding to 57% of the total N input into the cropping system was lost through leaching. Hence, the four hotspots for research intervention to improve the nutrient cycling efficiency of the farming system shown on Figure 2.1.8 include the following: Reduction of the use of crop residues for fuel wood and construction purposes by ? identifying other locally available sources; Quantification of the short and long-term benefits of crop residue retention and packaging ? the technology appropriately to boost its adoption; Development of cost-effective options for improving the quality of manure; and? Development of cost-effective technologies to control leaching.? Discussion The negative N balance observed at the farms and village levels suggest that annual crop production in these villages relies on soil N stocks to sustain crop production. A depletion of these reserves at the prevailing rate of 7 to 19 kg/ha per year will bring crop production to a halt if remedial measures are not used to reserve the trend. The N balances in this study were better than the average N-balance for sub-Saharan Africa (?22 kg/ha per year) as reported by Stoorvogel and Smaling (1990) when farmers applied the recommended doses of mineral N fertilizers. In the absence of mineral N fertilizer use, N balances became worse than average value. It confirms that although smallholder farmers in the savannahs of West Africa are applying N fertlizers, their application rate were short of the recommended rates. Leaching loses of N (19?22 kg N/ha) compared favourably with the 24.5 to 30 kg N/ha losses found by Wortmann and Kaizzi (1998). The observed marginal losses of N through gaseous exchanges could be due to the low pH and well drained nature of these soils which moderated the processes of denitrification and ammonia volatilization. Following the application of the recommended doses of mineral P fertilizers, positive balance was achieved. The P balances estimated without the use of P fertilizer in Garin Labo (?2.40 kg P/ha per year) agreed with the average P balance in Niger (?2.0 kg P/ha per year) reported by Stoorvogel and Smaling (1990) indicating that smallholder farmers in Niger may not be using P fertilizers in the cropping systems. Compared with average balances for Ghana and Nigeria (Stoorvogel and Smaling 1990), our estimates indicated that smallholder farmers use P fertilizer but at a lower rate than recommended. Retention of half of the residues generated on the field supplied higher amount N (8?26.28 kg N/ha/yr) than P (0.5?2.0 kg P/ha per year) into cropping 47Balancing livestock needs and soil conservation in West Africa system. Considering that N is the most limiting plant nutrient in the soils of the savannahs (Vanlauwe et al. 2002), returning of crop residues may improve crop production greatly. Crop residue is a scarce resource in the savannahs of West Africa as the amount of useable residues produced in the zone would support ruminant population for only 3.1 months/year if all is fed to livestock (Fern?ndez-Rivera et al. 2004). Approaches to promote tree and shrub production such as agroforestry with pollarding and alley farming are, therefore, needed to reduce the dependency on crop residues for fuel and construction purpose. Regardless of immense benefits of crop residues retention on crop production, farmers in the dry savannahs of West Africa find it prudent to remove all from the field. Research efforts should be intensified with the farmer as a key stakeholder to ensure efficient utilization of crop residues. In this regard, a working knowledge on the short and long term benefits of crop residue retention may refine farmers? decisions. As poor handling and storage of manure significantly reduced the fertilizer value, cost-effective strategies (i.e. storing manure in pits rather than heaps; under shed and covering with polyethylene film; and on concrete floors and under roofs) should be evaluated and used appropriately. The high leaching losses found in our study demand a cost-effective integrated approach to curb these losses. Such techniques may focus on increased synchrony and synlocation of nutrient uptake by crops and moderate rate of infiltration. Conclusion The current farmer practices on crop residue allocation in the savannahs of Ghana, Nigeria, and Niger, irrespective of regimen of N fertilizers used, lead to the depletion of soil N. The stocks of P in the soil are increased when the recommended application rates of P fertilizers are followed. As a myriad of factors contributed to the widespread negative N balances, a multifaceted approach is required to reverse the trend. Such a strategy should reduce the use of crop residues for non-agricultural purposes and increase their availability for crop and livestock production. It may identify other locally available materials for fuel wood and construction, promote the retention of crop residues on the field after harvest, and improve the storage of manure while reducing leaching losses simultaneously. Lastly, these management options should be affordable to the farmer and compatible with his/her practice. 48 Balancing livestock needs and soil conservation in West Africa References van den Bosch H, Gitari JN, Ogaro VM, Maobe S and Vlaming J. 1998. Monitoring nutrient flows and economic performance in Africa farming systems (NUTMON). III. Monitoring nutrient flows and balances in the districts in Kenya. Agric Ecosyst. Environ. 71:63?80. Carsky RJ and Ndikawa R. 1998. Screening multiple-use cover crops for the Sudan savanna of northern Cameroon. In: Buckles D et al. (eds), Cover crops in West Africa: Contributing to sustainable agriculture. IDRC (International Development Research Center), Ottawa, Canada. pp. 179?187. FAO (Food and Agriculture Organization of the United Nations). 1995. The state of food and agriculture. FAO, Rome, Italy. Fernandez-Rivera S, Okike I, Manyong V, Willams TO, Kruska RL and Tarawali SA. 2004. Classification and description of the major farming systems incorporating ruminants livestock in West Africa: Lessons learned and emerging research opportunities. In: Williams TO, Tarawali SA, Hiernaux P and Fernandez-Rivera S (eds), Sustainable crop?livestock production for improved livelihoods and natural resource management in West Africa. Proceedings of an international conference held at IITA, Ibadan, Nigeria, 19?22 November 2001. IITA (International Institute of Tropical Agriculture), Ibadan, Nigeria. pp. 89?122. Harris F. 1999. Nutrient management strategies of smallholder farmers in a short-fallow farming system in North-East Nigeria. The Geographical Journal 165(3):275?285. Harris F. 2002. Management of manure in farming system in semi-arid West Africa. Experimental Agriculture 38:131?148. de Jager A, Nandwa SM and Okoth PF. 1998. Monitoring nutrient flows and economic performance in African farming systems (NUTMON). 1. Concepts and methodologies. Ag., Ecosys. Env. 71:37?48. Kanmegne J, Smaling EMA, Brussaard L, Gansop-Kouomegne A and Boukon A. 2006. Nutrient flows in smallholder production systems in the humid forest zone of southern Cameroon. Nutrient Cycling in Agroecosystems 76:233?248. Kwakye PK. 1980. The effects of method of dung storage and its nutrient (NPK) content and crop yield in the north east savanna zone of Ghana. Paper 20. Organic recycling in Africa. FAO Soils Bulletin 43. FAO, Rome, Italy. Lekasi JK, Tanner JC, Kimani SK and Harris PJC. 1999. Manure management in Kenya highlands. Practices and potential. 2 nd ed. Henry Doubleday Research Association (HRDA) Publication, Coventry, UK. 35 pp. Lynam JK and Herdt RW. 1989. Sense and sustainability, sustainability as an objective in international agricultural research. Agricultural Economics 3:381?398. Nye PH and Greenland DJ. 1960. The soil under shifting cultivation. Technical Communication 51. Commonwealth Bureau of Soil, Harpenden, UK. 49Balancing livestock needs and soil conservation in West Africa Powell JM, Pearson RA and Hiernaux PH. 2004. Crop?livestock interaction in the West African drylands. Agron J. 96:469?483. Rufino MC, Tittonell P, van Wijk MT, Castellanos-Navarrete A, Delve RJ, de Ridder N and Giller KE. 2007. Manure as a key resource within smallholder farming systems: Analyzing farm-scale nutrient cycling efficiencies with the NUANCES framework. Livestock Science 112:273?287. Sanchez PA, Izac AMN, Valencia I and Pieri C. 1996. Soil fertility replenishment in Africa. In: Breth SA (ed), Achieving greater impact from research investments in Africa. Sasakawa Africa Association, Mexico City, Mexico. Scoones I and Toulmin C. 1998. Soil nutrient balances: What use for policy? Ag., Ecosys. Env. 71:255?267. Smaling EMA. 1993. An agro-ecological framework for integrated nutrient management with special reference to Kenya. PhD thesis, Wageningen Agricultural University, Wageningen, the Netherlands. Smaling EMA, Stoorvogel JJ and Windmeijer PN. 1993. Calculating soil nutrient balances in Africa at different scales. II District scale. Fert. Res. 35:237?250. Smaling EMA, Fresco LO and de Jager AA. 1996. Classifying, monitoring and improving soil nutrient stocks and flows in Africa agriculture. Ambio 25(8):492?496. Stoorvogel JJ and Smaling EMA. 1990. Assessment of soil nutrient depletion in sub-Saharan Africa, 1983?2000. Report 28. The Winand Centre for Integrated Land, Soil and Water Research (SC-DLO) Wageningen, The Netherlands. Thomsen IK. 2000. C and N transformations in 15 N cross-labelled solid ruminant manure during anaerobic and aerobic storage. Bioresource Technology 72:267?274. Vanlauwe B, Diels J, Aihou K, Iwuafor ENO, Lyasse O, Sanginga N and Merckx R. 2002. Direct interactions between N fertilizer and organic matter: evidence from trials with 15 N- labelled fertilizer. In: Vanlauwe B, Diels J, Sanginga N and Merckx R (eds), Integrated plant nutrient management in sub-Saharan Africa: From concept to practice. CAB International, Wallingford, Oxon, UK. pp. 173?184. Watson CA, Oborn I, Eriksen J and Edwards AC. 2005. Perspectives on nutrient management in mixed farming systems. Soil Use und Management 21:132?140. de Willigen P. 2000. An analysis of the calculation of leaching and denitrification losses as practiced in the NUTMON approach. Report 18. Plant Research International, Wageningen, the Netherlands. World Bank. 2007. World development report 2008. Agriculture for development. Washington, DC, USA. CD-ROM. Wortmann CS and Kaizzi CK. 1998. Nutrient balances and expected effects of alternative practices in farming systems of Uganda. Ag., Ecosys. Env. 71:115?129. 50 Balancing livestock needs and soil conservation in West Africa 11 Capacity building/training Ten participants from Niger, Nigeria, Denmark, Ghana and Kenya attended the training workshop on the use of Integrated Modeling Platform for Mixed Animal Crop Systems (IMPACT) and Reference works (Refworks) from 12 to 16 February 2007 at IITA-Kano station. The IMPACT workshop equipped participants with the relevant skills to collect comprehensive data on crop?livestock systems, create a database, and assess the profitability, labour efficiency and partial nutrient balance of a farming enterprise. It also introduced to the participants the concept of a holistic approach to monitor the food security status of a household. The lessons on the Refworks sharpened the skills of participants in literature search, literature coalition and references citation. The projects also trained two PhD students on the biophysical and socio-economic aspects of crop residue management. The two students designed research protocols and executed it to contribute to the key outputs of the project. The biophysical student is a student from the Kwame Nkrumah University of Science and Technology in Ghana and is writing a thesis on ?Sustainable management of crop residues and manure in smallholder cereal?legume?livestock systems in the savannahs of West Africa.? The socio-economic student is preparing a thesis entitled ?Socio-economic factors influencing crop residues intensification decisions in the subhumid and semi-arid savannahs of West Africa? to be submitted to the Obafemi Awolowo University in Nigeria. 12 Presentations in conferences/meeting Presentation on trade-offs in agricultural uses of crop residues were made by Andrews Opoku and Dr Robert Abaidoo at the joint ILRI?IITA workshop at IITA on 3 November 2009. Dr Tahirou Abdoulaye also made a presentation on ?Balancing livestock needs and soil conservation: Assessment of opportunities in intensifying cereal?legume?livestock systems in West Africa? at the Livestock Program meeting in Addis Ababa on 4 December 2009. 13 Problems and measures taken Challenges were posed by the poor workability of the soil (which made manual incorporation of crop residues less effective) and by the low intake of cereal stover by livestock. In addition, most of the farmers only had a few animals, so it was difficult to quantify the trade-offs at individual farm level. Future studies on crop residue allocation may consider implementing field experiments in areas where control grazing is practised. This will prevent crop residues meant for soil improvement from being grazed by free-roaming animals. To improve stover intake, future studies may chop 51Balancing livestock needs and soil conservation in West Africa or mill the material and fortify it with palatable feed ingredients. Animals were bought from local markets in the study areas to make up for numbers required. Future studies may, however, foster the formation of vibrant farmer groups and entreat members to donate animal each for the study. 14 Linkages with other research The project has worked closely with DGIS Dutch Government?APO at IITA, Kano to develop protocols for baseline data collection and the BMZ/GTZ-Postdoctoral Scientist (Soil Conservation Specialists) at IITA, Ibadan to review literature on past soil conservation projects and practices in Ghana, Benin and Nigeria. Balancing livestock needs and soil conservation: Assessment of opportunities in intensifying cereal?legume?livestock systems in West Africa Final Report SYSTEMWIDE LIVESTOCK PROGRAM slp C G I A R Syste m w i d e L i v e s t o c k P rogr am m e C M Y CM MY CY CMY K Cover_SLP_ProjectReport.pdf 6/3/2010 10:42:43 AM