Future climate change potentially can have a strong impact on the African continent. Of special concern are the effects on food security and the restricted adaptive capacity of Africa's poverty stricken population. Targeted policy interventions are, therefore, of vital importance. While there is a broad consensus on selection of climate and agricultural indicators, a coherent spatial representation of the populations' vulnerability is still subject to debate, basically because important drivers at household and institutional level are captured at the coarser (sub)-national level only. This paper aims to address this shortcoming by capitalizing on available spatially explicit information on households, food security institutions and natural resources to identify and characterize vulnerable groups in climate change prone areas of East and West Africa. First, we identify and localize groups with varying degrees of vulnerability, using food security and health indicators from georeferenced household surveys. Second, we characterize these vulnerable groups using statistical techniques that report on the frequency of occurrence of household characteristics, social bonding, remittances and agro-ecological endowments. Third we localize areas where climate change conditions affect production of major staple crops even after a maximum adaptation of crop rotations. Fourth, we characterize the vulnerable groups in the climate change affected areas and compare their profiles with the overall assessment to elucidate whether generic or climate change targeted policies are required. Since climate change will impact predominantly on agricultural production, our analysis focuses on the rural areas. For West Africa, we find that vulnerable groups in areas likely to be affected by climate change do not fundamentally differ from vulnerable groups in the study area in general. However, in East Africa there are remarkable differences between these groups which leads to the conclusion that in this part of Africa, poverty reducing strategies for climate change affected areas should differ from generic ones.