Cocoa (Theobroma cacao L.) is one of the world's most important agricultural commodity tree crops with the largest share of global production concentrated in West Africa. Current on-farm yields in this region are low and are expected to decrease in response to climate change through decreasing climate suitability. Previous studies identified numerous climate, soil and agronomic management factors limiting cocoa yields on farmer fields, however, the relative importance of these factors in explaining variation in yields which is relevant for prioritizing interventions is largely unknown. Additionally, effects of temperature, rainfall and atmospheric carbon dioxide concentration [CO2] on cocoa tree physiology and productivity are poorly understood. As a consequence, possible implications of climate change for cocoa productivity and adaptationshave not yet been considered. Climate-induced geographic shifts in the West African cocoa belt may have serious implications for farmers, cocoa supply and forests. This thesis therefore aimed to: (1) assess how current climate, soil and management factors affect current cocoa yields; (2) quantify the cocoa yield gap and the factors that can narrow the gap; and (3) assess the impacts of projected changes in climate and the underlying rise in atmospheric concentration [CO2] on future cocoa production. In this thesis, on-farm data on cocoa yield were combined with simulations with a cocoa crop growth model.It was found that under current climate, agronomic management was the dominant determinant of on-farm cocoa yields in Ghana, more so than environmental (climate and soil) conditions whilst climate effects on yields were stronger than soil effects. Nonetheless, the role of environmental conditions on cocoa yield becomes more important with increasing yields, such that the most productive cocoa farms tend to be the ones whose yields are most climate sensitive. Large cocoa yield gaps were found on farms revealing large opportunities to increase yield beyond current levels. Maximum water-limited yield gaps were much larger than yield gaps attainable in high-input and low-input systems. Climate factors were the important drivers of the absolute maximum water-limited and attainable yield gaps in high-input systems, but not in low-input systems. Relative yield gaps (maximum water-limited, attainable in high-and lowinput systems) were reduced by management practices, particularly cocoa tree density and black pod control. This shows that improved agronomic practices offer opportunities to substantially vi increase production of present-day cocoa plantations.Under future climate by mid-century ( 2060), large increases in potential water-limited yields and gains in area suitable for growing cocoa were expected, particularly when assuming full effects of elevated CO2 and under wetter climate-change scenarios. Impacts were expected to follow a (south) east -west gradient with projected yield increases and in area suitable for cocoa being most positive for Cameroon, followed by Nigeria (excep...