Crop Systems Simulation Platform for Strategic Investment Decisions.

The HarvestChoice initiative seeks to generate improved knowledge products to inform strategic investment choices for improving the productivity of smallholder crop production systems in sub-Saharan Africa (SSA). One key analytical component of the HarvestChoice research framework (SChEF: the Spatial Characterization and Evaluation Framework) is a grid-based regional-scale crop systems simulation platform, Simplr. Calibrated around the best available representation of crop production systems in SSA, Simplr employs the DSSAT Crop Systems Model and provides a virtual experimentation capacity to explore the potential impacts of a range of technology, management and environmental changes on future crop productivity (main product and residues) and input use profitability.

This paper discusses the development of Simplr and its several use cases. For each 5 arc-minute (about 10 km) grid cell in SSA, a stack of model input datasets has been compiled from analyses of existing global and regional databases as well as consultations with experts at CGIAR centers. Computational burden has been dramatically reduced by utilizing a high-performance computing cluster. Presented use cases include the regional assessment of 1) simulated maize yield responses to nutrient and water managements, 2) profitability of inorganic fertilizer application on maize, 3) risk of adopting new maize varieties, 4) economics of integrated soil fertility management on cereal crops, and 5) climate change impacts. Ongoing challenges, advances, and partnerships with respect to the development and management of the platform, datasets, and tools, as well as the proposed strategy of model performance evaluation will be discussed.