The climate of West Africa is subject to some of the most variable rainfall patterns observed anywhere in the world. In the past, the region has suffered several decades of severe droughts, whilst more recently major flood events have struck a number of the region's rapidly expanding cities. The consequences of these climatic extremes for the population have been particularly pronounced due to widespread and severe poverty. Global climate change, coming on top of such a variable and unpredictable regional climate, poses a major threat to the populations and economies of West Africa. Although the pathway from climate change to human suffering in West Africa is very short, there are some key bottlenecks to using climate projections to mitigate against risks to the population. Critical gaps exist in knowledge of how West African climate will change over the course of the 21st century, and the uncertainties make it almost impossible for agencies to deliver well-informed plans for the coming decades in critical areas such as food security, urban development and water. Even with the best climate information, it remains a significant challenge to integrate the scientific knowledge into planning and management structures. This collaborative project between scientists and policy makers in West Africa and Europe will, on the one hand, increase understanding of the regional climate and how it will change, and on the other, apply that knowledge to practical development questions. One of the key challenges for climate science is to understand how the changing composition of the atmosphere (notably CO2) will impact on the frequency and intensity of extreme events such as floods and droughts. In West Africa, these events are tied to the behaviour of convective rain storms; when storms are particularly intense or occur in rapid succession, devastating floods may result, whilst a week or two without storms during the wet season can trigger crop failure. Climate scientists rely on computer simulations of the global atmosphere, oceans and continents, yet these models have a very crude description of convective storms. For the first time, a new generation of regional climate models is emerging which realistically depict storms, and critically, how storms respond to factors such as land and ocean conditions, and increases in CO2. AMMA-2050 will use these new computer simulations alongside conventional climate models and historical observations, to understand why the statistics of key climate extremes are changing, and what this tells us about climate and its extremes in future decades. The outputs from the models will be used to examine impacts on key sectors in West African society, notably water and agriculture. Adaptation options will be explored, for example through the use of alternative crops, taking account of the inherent uncertainties in climate information, and the ways in which it is interpreted by decision-makers. We will focus on two questions. Firstly, in Senegal we will identify sustainable agricultural adaptation strategies and the policy frameworks to support those options. Secondly, we will examine how climate changes are likely to affect flooding in the rapidly growing city of Ouagadougou in Burkina Faso. The research and capacity building work of AMMA-2050 will help develop a new generation of African researchers and decision-makers, well-placed to respond to the requirements of West African nations. Within AMMA-2050, end-users have an important role, and their needs are embedded in project design and delivery, such that outputs will be responsive to their needs, and delivered in a format that is easily used. Enhanced resilience is an important aim of the project: it starts with improving our understanding of the climate signal over West Africa and leads through to decisions being made in specific pilot studies that showcase the importance of using improved and impact-sensitive science outputs.