Grid-scale crop dynamic layout optimization model considering stakeholders' cropping preferences and practice behaviours

Rapid population growth has led to an increasing demand for food, accompanied by an increase in the resources invested in producing food, and the extensive food production practices have placed a huge burden on the environment. There is an urgent desire of need to find a sustainable model of food production. Therefore, this study attempted to develop a dynamic regional grid-scale crop layout model (ABCAMODP-Crop) combining agent-based model, multi-objective programming, dynamic programming and cellular automata model to promote more sustainable food production from a supply-side perspective. The model can: (1) take into account the planting preferences and practical behaviours of government and farmer as stakeholders in crop layout; (2) further weigh the conflicts among the various objectives that stakeholders are concerned with from resource, environmental, nutritional and economic dimensions; (3) consider the impact of the previous year's crop layout on the next year's crop placement; and (4) obtain grid-scale accurate crop layout solutions by considering the spatial variability of optimization parameters and the neighborhood effects of crop layout cells. The model was applied to the Shiyang River Basin in China to verify its usefulness. Over the entire planning period, crop layout schemes optimized with the objective of integrating irrigation water productivity, environmental impact, nutrient availability and economic benefits have higher comprehensive benefits than the original scheme, with an increase in average irrigation water productivity per unit area of 14.55%, an increase in total yield of 14.45%, an increase in total carbohydrate availability of 7.39%, an increase in total calcium availability of 18.74%, an increase in total net carbon sequestration of 8.05%, an increase in total net benefit of 20.08%. The model developed allows for a holistic view of crop layout and guides decision-makers in a rational distribution of regional crops, thus promoting more sustainable food production. It can equally be extended to other regions as a generic modelling framework.