Options for intensification of mixed crop-livestock systems in rural communities of south-western Niger: Prospects and constraints

In semi-arid south-western Niger, climatic variability, low productivity, poverty and institutional constraints limit the scope of agricultural production to meet the rising needs of the growing human population. The integration of crop and livestock production activities has long been seen as the most promising means for intensifying agriculture and improving productivity and food security. Livestock play key roles in mediating soil fertility through excreting manure and grazing crop residues and distant pastures. They are ideal complements to the often advocated 'external fertiliser inputs' for maintaining soil fertility. Major local trends associated with human and animal population growth include continuous cropping, which has resulted in the breakdown of the fallow system; declining soil fertility and yields; migration; animal transhumance and reduced availability of high-quality pastures. This study explored the impacts of human and animal population growth under farming intensification on the performance and livelihoods of differently endowed farms and on the agro-ecosystem. Comparative analysis of different cropping intensity scenarios, at three sites of the study area was conducted. Explorative bio-economic optimisation models used to quantify and compare selected performance indicators of various farm types under different conditions revealed important socio-economic and ecological trade-offs. A recursive approach was used to project the analysis to 2020. The study showed that soil fertility may not always or irreversibly deteriorate with intensification. Local coping strategies could still be effective - though not indefinitely. Owning livestock could allow some farmers to achieve food security, while maintaining soil fertility by capturing and mobilising soil nutrients. Intensification will bring improvements - in labour status and incomes - but its benefits will be unevenly