Testing APSIM in a complex saline coastal cropping environment

Due to the seasonal increase in soil salt accumulation after cessation of monsoon rains, the simulation of cropping system performance becomes highly challenging in coastal saline cropping areas. Rapidly changing groundwater (GW) dynamics during this period (GW depth and salinity) drive changes in capillary soil moisture rise, soil evaporation and consequent deposition of salts in the crop root zone. Difficulty in simulating this complex cropping environment makes model-based examination of optimal cropping patterns and agronomic management difficult, as one season can be very different to the next. The performance of crops is also difficult to predict under future climate scenarios in these regions, as the impact of both changes in climate and groundwater salinity dynamics on soil status in the crop root-zone changes in complex fashion. No previous simulation study has sought to combine such dynamic and complex elements in simulating crop performance. We calibrated and validated the APSIM model for simulating a broad range of experimental treatments in a rice-pulse cropping system over two seasons, using the example of coastal saline West Bengal, India. This represents a novel evaluation of the APSIM model in simulating the complex mechanisms of seasonal soil water and salinity behavior (as driven solely by daily climate and a dynamic shallow saline water-table), together with the associated crop responses. The model performed well in simulating the observed soil chloride content (CC) and soil water content (SWC) with a high coefficient of determination (R2) for both calibration and validation datasets (CC, R2 = 0.91** and 0.89**; SWC, R2 = 0.90** and 0.93** respectively) and also acceptable RMSE values. These were well within the bounds of observed experimental error, indicating that the model was simulating system behaviour acceptably. APSIM then successfully simulated the observed crop performance in response to these soil dynamics across 24 unique environmental situations. This illustrates that crop performance in such complex environments can be robustly simulated, and that models like APSIM are a useful tool to translate outputs from other models at different scales (for example climate change from general circulation model's (GCM's), and future changes to groundwater depth and salinity dynamics from regional hydrology models) into changes in cropping system performance. This positions APSIM strongly as a robust research tool for climate change stud