Providing food and water security for a population expected to exceed 9 billion by 2050 while conserving natural resources requires achieving high yields on every hectare of currently used arable land. This is especially relevant for sub-Saharan Africa (SSA), where food production is not keeping pace with population growth. While recognizing there are other aspects to food security than production alone (e.g. distribution, demand, waste, governance, population), efficiently increasing production on existing farmland forms an essential component of the sustainable intensification paradigm. In SSA 80% of the food is currently produced by smallholder farmers, and rural population is projected to increase while average farm size will decrease in most SSA countries. Therefore, smallholder farms must be part of the solution to local and global food security. However, smallholder production systems across SSA are extremely diverse in terms of agro-ecology (climate, soil) and socio-economic conditions. Characterizing this diversity is essential for targeting research and policy interventions in the context of food security. The Global Yield Gap Atlas project (GYGA, www.yieldgaps.org) has developed a new global agro-climatic zonation scheme, whereby zones are homogeneous enough in terms of climate relevant for crop growth, not too detailed to prevent local data collection on climate, soils and cropping system, and covering most important current cropping areas so yield gap assessments can be upscaled from local to regional and national level. In this presentation we propose to make use of improved digital soil information available from the Africa Soil Information Service (AfSIS) in combination with the GYGA agro-climatic zonation scheme and upscaling protocol for yield gap assessment at local to national scale. These yield gap assessments can be used for national food security assessments as well as to target agronomic research and/or policy interventions.