Search

Soil fertility gradients develop on smallholder farms due to preferential allocation of inputs. A multi-location on-farm trial was conducted in Meru South, Central Kenya whose overall aim was to test minimum tillage and crop-residue retention practices in socio-ecological niches across heterogeneous smallholder farms. We identified three soil fertility classes together with the farmers, namely: good, medium, and poor. In each soil fertility class, two tillage (minimum or regular) and two crop residue (removed or retained) practices were tested for four consecutive seasons. Maize (Zea mays L.) grain yields in the good fields were above 2.5 Mg ha-1 across cropping seasons and cumulated yields were not influenced by tillage or crop residue management. The grain yields in the medium fields ranged between 1.3 and 5.4 Mg ha-1 and were greater with crop residue retention. In the poor fields, grain yield was

Agricultural development is complex, highly dynamic and differs among varying contexts. Decision-making for sustainable agricultural development cannot be based on generalized science-based knowledge, but should include context-specific knowledge and values of local stakeholders. Computer models seem a useful tool to integrate scientific knowledge include local-specific data, and explore local-specific solutions. In this paper we study whether and how a multiple goal linear program (MGLP) model could enhance learning for sustainable development. According to the learning theory, multi-actor learning is only productive when it consists of first-order (experiential) learning and second-order (social) learning. We applied an action-research approach and explored the value of an MGLP model SHARES (SHAred RESources) for learning by agricultural extension staff and farmers in an integrated rural development project in Burkina Faso. Fieldwork showed the main value of SHARES in the capacity to generate farm scenarios and trigger second-order learning about tacit frames-of-reference. People rarely engage in secondorder learning, but pursue different objectives and often remain trapped in confusing discussions and action. SHARES was a critical boundary-spanning object that facilitated communication between farmers and agricultural staff, enhanced mutual understanding, and the determination of area- and category-specific farm development goals.

African farming systems are highly heterogeneous: between agroecological and socioeconomic environments, in the wide variability in farmers’ resource endowments and in farm management. This means that single solutions (or ‘silver bullets’) for improving farm productivity do not exist. Yet to date few approaches to understand constraints and explore options for change have tackled the bewildering complexity of African farming systems. In this paper we describe the Nutrient Use in Animal and Cropping systems – Efficiencies and Scales (NUANCES) framework. NUANCES offers a structured approach to unravel and understand the complexity of African farming to identify what we term ‘best-fit’ technologies – technologies targeted to specific types of farmers and to specific niches within their farms. The NUANCES framework is not ‘just another computer model’! We combine the tools of systems analysis and experimentation, detailed field observations and surveys, incorporate expert knowledge (local knowledge and results of research), generate databases, and apply simulation models to analyse performance of farms, and the impacts of introducing new technologies. We have analysed and described complexity of farming systems, their external drivers and some of the mechanisms that result in (in)efficient use of scarce resources. Studying sites across sub-Saharan Africa has provided insights in the trajectories of change in farming systems in response to population growth, economic conditions and climate variability (cycles of drier and wetter years) and climate change. In regions where human population is dense and land scarce, farm typologies have proven useful to target technologies between farmers of different production objectives and resource endowment (notably in terms of land, labour and capacity for investment). In such regions we could categorise types of fields on the basis of their responsiveness to soil improving technologies along soil fertility gradients, relying on local indicators to differentiate those that may be managed through ‘maintenance fertilization’ from fields that are highly-responsive to fertilizers and fields that require rehabilitation before yields can improved. Where human population pressure on the land is less intense, farm and field types are harder to discern, without clear patterns. Nutrient cycling through livestock is in principle not efficient for increasing food production due to increased nutrient losses, but is attractive for farmers due to the multiple functions of livestock. We identified trade-offs between income generation, soil conservation and community agreements through optimising concurrent objectives at farm and village levels. These examples show that future analyses must focus at farm and farming system level and not at the level of individual fields to achieve appropriate targeting of technologies – both between locations and between farms at any given location. The approach for integrated assessment described here can be used ex ante to explore the potential of best-fit technologies and the ways they can be best combined at farm level. The dynamic and integrated nature of the framework allows the impact of changes in external drivers such as climate change or development policy to be analysed. Fundamental questions for integrated analysis relate to the site-specific knowledge and the simplification of processes required to integrate and move from one level to the next. Keywords: Crop–livestock systems; Soil fertility; Smallholders; Farm types; Simulation modelling

During participatory prototyping activities in Vihiga, western Kenya, farmers designed what they considered to be the ideal farm [Waithaka, M.M., Thornton, P.K., Herrero, M., Shepherd, K.D., 2006. Bio-economic evaluation of farmers’ perceptions of viable farms in western Kenya. Agric. Syst. 90, 243–271]: one in which high productivity is achieved through optimising crop-livestock interactions. We selected four case study crop-livestock farms of different resource endowment (Type 1–4 – excluding the poorest farmers, Type 5, who do not own livestock) and quantified all relevant physical flows through and within them. With this information we parameterised a dynamic, farm-scale simulation model to investigate (i) current differences in resource use efficiencies and degree of crop-livestock interactions across farm types; and (ii) the impact of different interventions in farm Types 3 and 4 on producing the desired shifts in productivity towards the ideal farm. Assuming no resource constraints, changes in the current farm systems were introduced stepwise, as both intensification of external input use (fertilisers and fodder) and qualitative changes in the configuration of the farms (i.e. changing land use towards fodder production, improving manure handling and/or changing cattle breeds). In 10-year simulations of the baseline, current scenario using historical weather data the wealthiest farms Type 2 achieved food self-sufficiency (FSS) in 20% of the seasons due to rainfall variability, whereas the poorer Type 4 only achieved FSS in 0 to 30% of the seasons; soil organic C decreased during the simulations at annual rates of −0.54, −0.73, −0.85 and −0.84 t C ha−1 on farms of Type 1–4, respectively; large differences in productivity and recycling efficiency between farm types indicated that there is ample room to improve the physical performance of the poorer farms (e.g. light and water use efficiency was 2–3 times larger on wealthier farms). Simulating different intensification scenarios indicated that household FSS can be achieved in all farm types through input intensification, e.g. using P fertilisers at rates as small as 15 kg farm−1 season−1 (i.e. from 7 to 28 kg ha−1). Increasing the area under Napier grass from c. 20 to 40% and reducing the area of maize, beans and sweet potato in farms of Type 3 and 4 increased their primary productivity by c. 1 t ha−1 season−1, their milk production by 156 and 45 L season−1, respectively, but decreased the production of edible energy (by 2000 and 250 MJ ha−1 season−1) and protein (by 20 and 3 kg ha−1 season−1). By bringing in a more productive cow the primary productivity increased even further in Farm Type 3 (up to 5 t ha−1 season−1), as did milk production (up to c. 1000 L season−1), edible energy (up to c. 10,000 MJ ha−1 season−1) and protein (up to c. 100 kg ha−1 season−1). The impact of livestock management on the recycling of nutrients and on the efficiency of nutrient use at farm scale can be large, provided that enough nutrients are present in or enter the system to be redistributed. An increase in N cycling efficiency through improved manure handling from 25 to 50% would increase the amount of N cycled in the case study farms of Type 1 and 2 by only ca. 10 kg season−1, and only 1–2 kg season−1 in Type 3 and 4. The various alternatives simulated when disregarding resource constraints contributed to narrow the productivity and efficiency gaps between poorer and wealthier farms. However, the feasibility of implementing such interventions on a large number of farms is questionable. Implications for system (re-)design and intensification strategies are discussed.

Efficient water use in rice cultivation is a prerequisite for sustaining food security for the rice consuming population of India. Novel rice production practices, including water-saving techniques, modifications in transplanting, spacing, weeding and nutrient management, have been developed and shown to be effective on farm, but adoption of these techniques by farmers has remained restricted. Potential constraints include technical difficulties with new practices, and labour and gender issues which differ between farms. On the basis of a rapid survey of 100 rice-based farms, four farm types were identified based on their socio-economic and biophysical characteristics. Detailed farm surveys were conducted on three representative farms of each farm type to evaluate land use patterns, use of inputs such as water, labour, nutrient, capital and machinery, income from crop and animal production and off-farm activities. Opportunities exist for one or more new rice cultivation techniques to be adopted in all the four farm types. For all farm types, however, the opportunities for use of water-saving irrigation were the least promising. In general, adoption of water-saving irrigation will not improve farmers’ livelihoods despite its importance in reducing water scarcity problems at regional scale. At farm scale, the potential for adoption of water-saving irrigation depends on the season, location of fields and the irrigation source. Changes in government policies such as rules and regulations, pricing, institution building and infrastructure development, as well as training and education of farmers are needed to improve the adoption of modified methods for rice cultivation.

In South Patagonia, Argentina, sweet cherry is the main fruit-tree crop grown for export, resulting in a highly seasonal labour demand. Managers of deciduous perennial fruit orchards must consider both biological and economic relationships in selecting crop species and orchard design. This makes decisions at the farm-level extremely complex, as especially in such perennial crops, strategic (`what to plant¿, `with which technology¿ and `how much area of each activity¿, i.e. the final design) and tactical (`when, what and how to plant in time¿, the pathway to the planned farm) decisions have a long-term effect. The objective of this study was to explore the consequences of different strategic and tactical decisions at farm scale in fruit production systems of South Patagonia, considering the variation in interests and aims of different stakeholders, and using a sensitivity analysis to evaluate the consequences of possible changes in external conditions. A dynamic farm-scale optimization model called OPTIFROP was developed to generate alternative farm development plans, by allocating, in the course of the time horizon of the run, production activities to different land units, while optimising different objective functions, subject to several constraints. Although time-dependent, dynamic, mathematical programming models for analysing farming systems have been described in literature, the dynamic aspects of long-term decision-making in orchard design and their impact on the sequential (annual) nature of orchards in different growth phases (i.e. medium-term decision-making), need a higher time-staged dynamic approach with a staircase matrix structure. The model includes two objective functions at farm level: (1) maximization of the present value of cumulative financial result, which is the main objective for growers, and (2) maximization of cumulative farm labour, which is an objective often mentioned by policy makers. The inter-months deviation for labour demand (during the period of high labour demand, November¿April) was included as an upper-bound. Input and output coefficients for the land use options considered in OPTIFROP were quantified using the Technical Coefficient Generator FRUPAT. Model results indicated that the present value of cumulative financial result and the cumulative farm labour are conflicting to a very limited extent. Timing and feasibility of implementing certain combinations of production technologies are affected by resource endowments and initial conditions, but these factors do not influence land use selection in the long term. Land use selection is driven by the objectives of the stakeholders. OPTIFROP showed that, through introduction of alternative crops, substantial reductions in labour peaks in the period November¿April could be achieved with a relatively small reduction in farm income. The sensitivity of the model solution to the cherry price suggests that the fruit production sector of South Patagonia should pay more attention to the robustness of their land use plans and take preventive measures to avoid being caught by a possible crisis due to changes in the context.

The looming water crisis and water-intensive nature of rice cultivation are driving the search for alternative management methods to increase water productivity in rice cultivation. Experiments were conducted under on-station and on-farm conditions to compare rice production using modified methods of irrigation, planting, weeding and nutrient management with conventional methods of cultivation. Farm surveys were used to evaluate adoption of modified rice cultivation method. On-station experiments showed that, a combination of water-saving irrigation, young seedling or direct seeding, mechanical weeding and green manure application increased the rice water productivity though the largest yields were obtained for a combination of conventional irrigation, young seedling or direct seeding, mechanical weeding and green manure application. On-farm experiments demonstrated a yield advantage of 1.5 t ha −1 for the modified method over conventional method. We found, however, that yield advantages were not the sole factor driving adoption. Associated changes required in management, including the increased labour demand for modified planting, unwillingness of agricultural labourers to change practices, difficulties with modified nursery preparation and the need to replace cheaper women’s labour for hand weeding with more costly men’s labour for mechanical weeding, all reduced the chance of adopting the modified rice cultivation method. Risks associated with water-saving irrigation, such as uncertainty about the timing and amount of water release for irrigation affect adoption adversely as well. There was no incentive for farmers to adopt water-saving irrigation as water from reservoirs and electricity for pumping well-water are both free of charge. To date farmers continue to experiment with the modified cultivation method on a small part of their farms, but are unlikely to adopt the modified method on a large-scale unless policies governing water management are changed.

International audience; Conservation agriculture (CA) is promoted as a promising technology to stabilize or improve crop yields in Africa and Madagascar. However, small-scale farmers face difficulties to retain soil cover, mainly because of competing uses for the biomass produced, especially to feed cattle. To explore the relation between dairy production and CA we developed an optimisation model at farm level. Our aim was to explore trade-offs between CA practices and the size of dairy cow herds. Our model includes three main components: the farm, the crops and the cattle herd. The optimisation was made on the total net income for three years. Biomass produced by cropping activities can either serve as mulch or to feed cows. We applied a constraint on the minimum soil cover % to keep at the end of each year for CA fields: from 30% to 95%. We simulated two scenarios of milk market: a small milk market with low forage price and an open milk market scenario with higher price of forage. Three prototypes of farms were simulated with different proportion and size of four kinds of field. These three prototypes were: medium-sized farm with hillsides dominating, medium-sized farm with paddy fields dominating and small-sized farm with hillsides. Changing the degree of soil cover to be retained on CA plots did not significantly modify the total net farm income. It was more strongly influenced by the characteristics of the milk market. In case of a limited milk market it was not profitable to have more than seven cows because the expenses were not compensated by animal production. When setting minimum soil cover to 30% then all of the simulated results include biomass coming from CA cropping system even with 12 cows/farm. Conversely when setting this constraint to 95%, above 6/7 cows/farm forage come only from conventional fields. In all of the situations simulated even with 6 cows, with the current and twice the price for forage, it was possible to keep at least 50% of soil cover on 30-60% of the total farm area. CA was not feasible for farms with no irrigated paddy fields or when forage fetched a high price regardless of the constraint for % of soil cover to be kept on CA fields. Overall, CA systems can be beneficial for dairy cow farmers due to the forage produced, although the milk market and thus the value of biomass for forage, has a strong influence on CA practice at field level.

There are numerous examples of technologies with great potential that have not been accepted by smallholder farmers. Quite often, these technologies do not fit well into smallholder systems due to the inherent high level of heterogeneity of these systems. For example, despite their great potential, the adoption of legumes by smallholder farmers in many parts of sub-Saharan Africa has remained poor. A wide range of biophysical (e.g. climate, soil fertility, etc.) and socio-economic variables (e.g. preferences, prices, production objectives etc.) influence the use of legumes in smallholder systems. While some of these variables constrain the adoption of some legumes, others offer opportunities for beneficial use of other legumes in the same system. Therefore, widespread adoption of legumes in smallholder systems can only be achieved if all of the major biophysical and socio-economic constraints are simultaneously identified and addressed. The `socio-ecological niche¿ concept proposed in this paper provides the framework through which this might be achieved. The socio-ecological niche, in any given region of agricultural activity, is created by the convergence of agro-ecological, socio-cultural, economic and ecological factors, to describe a multidimensional environment for which compatible technologies can be predicted. The socio-ecological niche concept can be applied in many different contexts in technology development. However, this paper discusses its use with respect to the development of legume technologies. Two case studies are presented to illustrate the concept and to demonstrate its practical significance. The concept is being used in on-going research on legumes in smallholder farming systems in western Kenya.

In smallholder farms throughout subSaharan Africa, it is a common pattern for some fields to receive substantial inputs of fertilisers and manure, but others to receive nutrient inputs infrequently or never. A soil and crop model, SCAN, based on annual summary of carbon, nitrogen and phosphorus dynamics was calibrated against chronosequences on coarse soils in semi-arid Zimbabwe, giving satisfactory predictions of organic matter loss and crop yield responses to amendments. This model was then used to explore the effects of different nutrient resource allocation strategies across farms on field- and farm-scale production and the development of soil fertility gradients. Regular applications of manure to only part of a farm, common on farms with a limited amount, rapidly lead to large gradients in crop yield. However, the first version of the model indicated that spreading manure evenly at a lower rate would give greater whole-farm yields, so it seems illogical for farmers to be concentrating their nutrient resources onto infields. Even within an existing soil fertility gradient, greater total farm yields were obtained by spreading a given amount of fertiliser across the fields than by concentrating this fertiliser onto more fertile fields. However, this result was caused by the failure of the SCAN model to adequately simulate the poor response to fertilisers typically seen in outfields. A second version of the model modified to increase runoff and reduce nutrient availability on soils depleted in carbon gave a more realistic decline in crop yields on fields receiving inadequate nutrient inputs. The results of the second version of the model showed that the optimal allocation strategy is dependent on the existing soil fertility gradient. Fertiliser should only be applied on fields where grain yield is responsive to higher nutrient inputs, and not on infields which are nutrient saturated or on degraded outfields. To improve the relevance of fertiliser and manure recommendations, it is necessary to consider resource limitations and production at the farm scale, and the effects of applying nutrient resources not only on current crop yield, but also on the development of soil fertility on different fields. Simplified dynamic modelling approaches are highly suitable for exploring such questions of resource allocation at farm scale.

Agricultural intensification of farming systems in sub-Saharan Africa is a prerequisite to alleviate rural poverty and improve livelihoods. Legumes have shown great potential to enhance system productivity. On-farm experiments were conducted in different agro-ecological zones (AEZ) in Western Kenya to assess the agronomic and economic benefits of promising legumes. In each zone, trials were established in fields of high, medium and low fertility to assess the effect of soil fertility heterogeneity on legume productivity and subsequent maize yield. Common bean, soybean, groundnut, lima bean, lablab, velvet bean, crotalaria, and jackbean were grown in the short rains season, followed by maize in the long rains season. Alongside, continuous maize treatments fertilised at different rates were established. AEZs and soil fertility gradients within these zones greatly affected crop productivity, returns to land and labour of rotations, as well as the relative performance of rotations. Poorer soil fertility and AEZs with lower rainfall gave smaller legume and maize yields and consequently, smaller returns to land and labour. The cultivation of legumes increased maize yields in the subsequent long rains season compared with continuous maize receiving fertiliser at a similar rate, while the increase of maize after green manure legumes was stronger than that after grain legumes. Maize yield responded strongly to increasing amounts of N applied as legume residues with diminishing returns to legume-N application rates above 100 kg N ha−1. In the low potential zones, factors other than improved N availability likely also stimulated maize yield. Rotations with grain legumes generally provided better returns than those with green manures. Intercropping bean with maize in the long rains season provided an additional bean yield that did not come at the expense of maize yield and improved returns to land and labour, but more so in the high potential zones. The results demonstrate the strong impact of biophysical diversity on the productivity of the legumes and suggest the need for careful targeting of legume technologies to the different biophysical conditions.

Land use changes from perennial to annual crops and vice versa may affect water balances, in particular through changing evapotranspiration rates with great implications for modelling. Direct comparison of evapotranspiration rates published in the literature is hardly possible, since differences in soils and climate make data incomparable. This paper presents literature data on crop parameters and environmental conditions that determine transpiration for four crops (oil palm, cocoa, maize and rice) and evaporation. Transpiration at leaf scale and soil evaporation as well as evaporation of intercepted rainfall have been computed using these data and are scaled-up to canopy scale. The objective is to quantify evapotranspiration of the crops under identical environmental conditions at two temporal scales: daily and annual evapotranspiration. Variation among crops in transpiration at leaf scale owing to crop characteristics is levelled out in scaling up to canopy scale. Differences in annual evapotranspiration between perennial and annual crops mainly are due to the fact that perennial crops transpire during the dry season, although at low rates, but still considerably higher than evaporation rates of bare and dry soils. Apparently, the degree of soil cover with vegetation in space and in time is of major importance to evaluate differences in annual evapotranspiration caused by changes in land use.

Minimum tillage and vegetative barriers can conserve soil and water resources in the steep-sloping highlands of East Africa but there has been little adoption by smallholder farmers. Soil conservation efficiency and short-term economic benefits provided by tillage and vegetative barriers were assessed over four cropping seasons to understand performance under local farming conditions. Minimum tillage was compared with regular tillage and vegetative barriers (leucaena and Napier) with no barriers. Maize and soybean yields were greater with than without vegetative barriers, except with Napier barriers when minimum tillage was practiced where strong root competition occured. Cumulatively for the four cropping seasons, Napier barriers with regular tillage conserved most soil (72%) followed by Napier with minimum tillage (53%). The least soil (1%) was conserved for minimum tillage without barriers and leucaena barriers were intermediate in decreasing soil erosion. The highest positive marginal rate of returns (MRRs) were realized under leucaena barriers with regular tillage (2.09) followed by Napier with regular tillage (1.32). Minimum tillage without barriers had the lowest positive MRRs (0.08). Future increase in the price of key inputs would have greater depressive effect on the MRRs of Napier barriers with regular tillage than leucaena barriers with regular tillage. Minimum tillage without barriers was inefficient in soil conservation particularly when rainfall was intense and had poor MRRs. Leucaena barriers conserved less soil than Napier barriers but were more economically attractive, demonstrating a clear trade-off between soil erosion that is likely to impact crop yields in the long-term and short-term economic benefits. Napier barriers with regular tillage present a win–win scenario due to efficient soil conservation and attractive economic returns provided future prices of labour and Napier cuttings remain stable.

Improving water use and nitrogen efficiencies is of overall importance to society at large - to conserve scarce water resources and prevent environmental pollution. Efficient cultivation practices for rice which had no yield penalty were not adopted by farmers because of the open access to water free of charge. Well-chosen combinations of policy measures are thus needed to stimulate adoption of new cultivation practices. We developed a multi-objective linear programming (MGLP) model to explore the impact of: (i) modified rice cultivation including water-saving irrigation on farm profit; (ii) water pricing and water quota government policies on adoption of modified rice cultivation by farmers; (iii) a combination of (i) and (ii) to achieve the objectives of both farmers and society at large, and (iv) to study the trade-offs between income, water and nitrogen use. The analysis was carried out on four rice-based farm types for the state of Tamil Nadu, South India. Model results showed that observed farm profit of all four farm types could be increased using current practices simply by optimizing land use for specific crops. Adoption of modified rice cultivation further increased farm profit. Water-saving practices were selected only when water pricing was introduced. Farm profits were reduced even at low water prices but were compensated by farmers through adoption of modified rice cultivation. The combination of policies that stimulate adoption of modified rice cultivation was effective in achieving both increased farm income and water savings. The required water prices differed across farm types and seasons and impacted poor resource-endowed farmers the most. Providing water quotas could protect the poor resource-endowed farmers. The model helped to identify the optimal water price and water quota for each farm type to achieve both the objectives of farmers and society at large. Opportunities for reducing water use and avoiding environmental pollution at acceptable profits are available for all farm types, but need to be tailored to the farmers' resource endowments. (C) 2010 Elsevier Ltd. All rights reserved.

The decrease in crop yields at increasing distances from the homesteads within smallholder farms of Sub-Saharan Africa (SSA) is normally ascribed to the existence of within-farm soil fertility gradients. Field observations also suggest that a large part of such variability is concomitantly caused by poor agronomy. To understand the interaction between soil fertility (S factors) and management decisions (M factors) affecting crop variability, we combined field research conducted in western Kenya (Vihiga, Kakamega and Teso districts; rainfall: 1600, 1800 and 1200 mm, respectively) with explorations using the simple dynamic crop/soil model for dynamic simulation of nutrient balances, previously tested for the region. Field measurements indicated within-farm differences in average maize grain yields of 48% (2.7 vs. 1.4 t ha −1 ) in Vihiga and of 60% (1.5 vs. 0.6 t ha −1 ) in Teso, between fields that were close and far from the homestead, respectively. Extreme values ranged widely, e.g. between 4.9 and 0.3 t ha −1 for all the farms surveyed in Vihiga, where the average farm size was 0.6 ha. Maize grain yields tended to increase with increasing contents of soil C, total N, extractable P and exchangeable bases. However, the negative relationship between S factors and distance from the homestead was not as strong as expected, and yield variability was better explained by multiple regression models considering M factors such as planting date, plant density, resource use and weed infestation (40–60% across sites). Then, we analysed the variation in resource (cash, labour, N) use efficiency within farms of different resource endowments with the aid of the simulation model. N balances at plot scale varied from ca. +20 to −18 kg ha −1 , from −9 to −20 kg ha −1 and from −16 to −18 kg ha −1 for the different fields of the high, medium and low resource endowment case-study farms, respectively. Labour productivities ranged between ca. 10 and 38 kg grain man-day −1 across field and farm types. The results indicate the need of considering within farm heterogeneity when designing soil fertility management interventions. Resource use efficiency was strongly affected by soil quality. As farmers invest more effort and resources in the more productive and less risky fields, the interaction between S and M factors leads to farmer-driven resource use efficiency gradients within smallholder farms.

The spatial and temporal changes of land use, cropping patterns and cattle keeping were assessed for the period 1961–1999 in Kyamtwara division, Bukoba district, Tanzania. The assessment was based on interpreting aerial photographs, surveys and a review of historical statistical data. The area of grassland declined by 40% with a concomitant increase in annual crop fields and forest of 225 and 36%, respectively. The cropping pattern changed from a predominance of banana/coffee/beans to a complex mixed cropping of banana/coffee/beans/maize and root crops in the homegarden, and increased cultivation of maize and root crops in pure stands. Farmers stopped cultivating sorghum and finger millet. The population of indigenous cattle decreased by 50% and an equal percentage of dairy cattle was introduced, but cattle-owning households decreased by 85%. Nutrient balances of homegardens ranged between −27 and 17 kg N ha −1 yr −1 , −1 and 7 kg P ha −1 yr −1 and −5 and 12 kg K ha −1 yr −1 , with the positive balances achieved by resource-rich households. Nutrient balances of crops in pure stand ranged between −15 and −2 kg N ha −1 yr −1 , −2 and −1 kg P ha −1 yr −1 and −14 and −1 kg K ha −1 yr −1 , with more negative balances observed with maize, implying that soil nutrient stocks are decreasing. Increasing population density, coupled with an unequal distribution of resources among households, land tenure, economic policies and poor crop markets were identified as major causal factors of the above changes. Reversing soil fertility decline requires external inputs of nutrients. Within the current poor economic situation of the farming community, different potential soil fertility improvement strategies, including exploitation of N 2 -fixing legumes are discussed.

The use of a hierarchy in growth factors (defining, limiting and reducing growth factors), as developed for plant production has shown its usefulness in the analysis and design of plant production systems. This hierarchy presents a theoretical framework for the analysis of biophysical conditions in plant production. We hypothesize that analysis and design of agricultural land use systems is facilitated by development of a similar set of production ecological concepts for animal production, as livestock is often part of such systems. In this paper we present such a hierarchy. We identify growth defining (temperature, daylength, animal genetic characteristics), limiting (water and feed quantity and quality) and reducing (diseases, pollutants and other conditions leading to sub-optimal wellbeing) factors, determining the production of an individual animal, in parallel to their definition for crop production, and aggregate this production to herd scale. We discuss how management intervenes with these factors. Application of the production ecological concepts in design of land use systems ensures that all production systems are based on the prevailing biophysical characteristics and that intrinsic system properties are taken into account. This approach also provides a valuable framework for yield-gap analysis, explaining current production levels, and identifying constraining factors, for setting the research agenda by identifying knowledge gaps and for educational purposes. We illustrate application of the concepts in the exploration and design of alternative animal and mixed plant-animal production systems with two examples. The first example relates to potential production in intensive dairy farming in a temperate climate and the second to feed-limited cattle production in the tropics. (C) 2003 Elsevier Science Ltd. All rights reserved.

Current methods to estimate biomass as a basis for carrying capacity and secondary production include biomass measurements in the field, use of rainfall /biomass relations and simulation models. Each of the methods have their drawbacks. Collection of field data is laborious, rainfall /biomass relations lack precision and models, which simulate processes, are highly-demanding in input data, expertise and advanced equipment. An approach is presented to bridge the gap between those simulation models and the need for more accuracy in simple biomass estimates at less detailed scales. Transfer functions are developed using the results of a process-simulation model called SAHEL. The transfer functions allow estimation of the depth of wetting of the soil profile and percolation in relation to the water content at field capacity (pF 2.5). In comparison with simple rainfall-biomass relations, application of transfer functions allow more accuracy in estimating biomass production at a less detailed scale. With the transfer functions length of growing periods could be established more accurately, allowing for more precision in agro-ecological zoning methods.

Gross energy and crude protein balances of cattle in traditional systems and ranching in eastern Botswana have been estimated. The calculations show that ranching is more productive on a per animal basis while traditional systems are more productive on a per hectare basis. Ranching appears to be more efficient than traditional systems in terms of animal gross energy produced per unit of gross energy used for maintenance. Traditional systems, on the other hand, appear to be more efficient in terms of animal gross energy produced per unit of plant energy produced, and in terms of food chain efficiency. The same conclusions were drawn when crude protein balances were considered. The risk of pasture degradation might be greater at the higher stocking rates found in traditional systems, and feed shortages might occur more often, but lower stocking rates and ranch-type changes in product allocation in these systems would most probably lead to lower productivity per hectare. Consequently, this would lead to lower food production and less employment on a national scale. A socio-economic analysis at a national level is urgently needed to be able to weigh the advantages and disadvantages of the development of ranching at the cost of traditional systems.

Livelihoods of rice farmers depend on the efficient use of scarcely available agricultural resources. Farmers tend to maximize economic output of farming activities that may not necessarily coincide with the optimal use of resources from an ecological perspective. However, improving resource use efficiencies at the regional level is important for society at large. Efficiencies can be enhanced by well-chosen combinations of resource efficient technologies at the farm level and policy interventions at the regional level, thereby obtaining a balance between the objectives of both farmers and society. Rice-based farms in Tamil Nadu, India, were grouped into four farm types based on their biophysical and socio-economic characteristics. Crop and farm level resource use efficiencies of water, labour, capital and nutrients were quantified on three representative farms per farm type. The four farm types differed in water, labour and nutrient productivity and profitability both at crop and farm level. Water productivity was poor on Farm Types 1, 2 and 3 compared with Farm Type 4 clue to the open access to the commonly available canal water on the first types. Labour productivity was highest on Farm Type 2 due to more family labour use and lowest on Farm Type 3 due to the small operational holding. Farm Types 1 and 2 were most profitable and Farm Types 3 and 4 were least profitable directly related to the resource endowments. Farm Type 3 was least efficient in all the resources considered, emphasizing the negative effect of low resource endowments. Possible policy interventions in order to improve the resource use efficiencies and their effect on the farmer livelihoods are discussed. Government policy interventions may influence the farm resource use efficiencies through the adoption of resource efficient technologies. However, an identical set of policy interventions cannot be applicable to all farm types since current resource use efficiencies and adaptability of these farm types for change in policies differed substantially. (C) 2012 Royal Netherlands Society for Agricultural Sciences. Published by Elsevier B.V. All rights reserved.