The Bioeconomic Modelling System MPMAS_XN: Simulating Short and Long-term Feedback Between Crop growth, Crop Management and Farm Development under Climate Change

Climate effects on crop growth trigger management adaptations at three different levels: At an operational level, farmers adapt timing and quantity of sowing, fertilisation, plant protection, and harvesting while observing weather conditions and plant growth during the season. At a tactical level, they consider observed changes in yields, timing, and required input quantities in seasonal production planning. This affects their selection and spatial allocation of crops, rotations and production techniques. At a strategic level, farmers adapt farm structure and machinery, rent land, or abandon farming. This not only alters the way how plants are grown, but also where and if at all. Altered plant growth practices affect soil fertility through altering soil carbon and nitrogen stocks, which may feed back into management decisions. We present the integration of the agent-based farm simulator MPMAS with the agroecological modelling framework Expert-N (XN). The latter incorporates the coupled process models GECROS, LEACHN, the six-compartment carbon turnover module of DAISY and the water flow module of HYDRUS to simulate plant growth, soil carbon, nitrogen and water balances on a sub-daily resolution. Intra-seasonal adaptation of crop management is based on triggers dependent on simulated plant/soil state and is directly executed within the XN routines. MPMAS simulates yearly farmer strategic and tactical decisions that, among others, determine choice and spatio-temporal distribution of crops and crop management, as well as the triggers used within the operational decision level. MPMAS and XN are coupled in a single executable, which facilitates High Performance Computing (HPC) and allows for parallelised simulations of farmer adaptation decisions and plant growth of all farms and plots for entire regions with the potential to include farm-to-farm interactions (e.g. learning). We run simulations for Southwest Germany and demonstrate the potential of the model system to capture feedbacks between crop growth and farm management.