Closing yield gaps under climate change in a global computable equilibrium model of the world economy

Land use and land use change is determined as much by economic and institutional drivers as it depends on the bio-physical conditions. The demand for different uses of biomass is increasing and thus land use change and the expansion of farming areas into natural habitats may threaten ecosystems and their services. Factors influencing this process include climate and demographic change but also an increasing globalisation of agricultural markets paired with an increasing divergence between regional supply and demand of biomass. On the supply side, land productivity considerably increased over the last 6 decades, since in this period the food production was doubled while agricultural land only increased by 10%. However, agricultural yields as well as production stability are threatened by a changing climate. In addition, agricultural production and agricultural markets highly depend on numerous political interventions, e.g. on the support of biofuels. Socio-economic changes, such as increasing incomes and changes in consumption patterns already lead to increases e.g. in demand for animal feed that outgrows food demand for direct consumption. In the medium to long-run these consumption patterns may also alter the processes of global change. Such future pathways of socio-economic and environmental systems can only be assessed with scenarios which describe possible future paths of development. For this numeric models are one important tool. Food supply and accessibility depends not only on the ability to produce a sufficient quantity and quality of food, but also on the food price level. Thus, projections of future food supply and prices need to consider various biophysical and socio-economic parameters at the global and local scale. While the increase in global agricultural yields allowed global food production to keep pace with population growth, in many regions of the world agricultural yields are still low and the exploitable gap is therefore high. Changes in a...