ABSTRACT
The Sustainable Development Goals (SDGs) encourage nations to substantially increase food production to achieve zero hunger (SDG 2) while preserving life on land (SDG 15). A key question is how to reconcile these potentially competing goals spatially. We use integer linear programming to develop an 'integrated land use planning framework' that identifies the optimal allocation of 17 crops under different hypothetical conservation targets while meeting agricultural demands by 2030. Intensifying existing cropland to maximum yield before allocating new cropland would reduce land requirement by 43% versus cropland expansion without intensification. Even with yield gap closure, tropical and sub-tropical crops still require expansion, primarily allocated to Venezuela, eastern Brazil, Congo Basin, Myanmar and Indonesia. Enforcement of protected areas, via avoiding conversion in 75% of Key Biodiversity Areas and 65% of intact areas, is vital to attain biodiversity targets but bears large opportunity costs, with agricultural rents dropping from $4.1 to $2.8 trillion. Although nationally constrained forest conservation efforts would earn 9% less agricultural rents compared to globally coordinated conservation solutions, they were also able to reduce intact habitat and forest loss (43% and 35% reduction). Our results demonstrate that careful choice of the allocation of future cropland expansion, could dramatically reduce-but not eliminate-the tradeoffs between the SDGs for food production and land biodiversity conservation.