Improving performance of Agro-Ecological Zone (AEZ) modeling by cross-scale model coupling: An application to japonica rice production in Northeast China

The challenges to food security posed by climate change require unprecedented efforts and ability to simulate and predict the interactions between crop growth dynamics, and the environment and crop management at various scales. This calls for model coupling and fusion efforts, which aims to explore the interaction of agro-ecological processes across different scales. In this research, we proposed a coupling framework between two widely used crop models, the process-based and site-specific Decision Support System for Agro-Technology Transfer (DSSAT) model, and the cropping zone centered Agro-Ecological Zone (AEZ) model, with the intention to establish a coupling procedure between them, and to consequently enhance the micro foundation and improve the performance of the AEZ model. The procedure takes three major steps: (1) derive, calibrate and validate the key cultivar parameters using DSSAT, (2) translate these cultivar parameters into AEZ eco-physiological parameters and validate them using AEZ and DSSAT, (3) apply AEZ with these enhanced eco-physiological parameters and compare the new results with the old ones. An illustrative application of this procedure to japonica rice production in Northeast China is carried out for individual year between 1980 and 1999. The application results in a significant improvement in the spatial performance of the AEZ model. Calibration of the crop genetic parameters increases regional average potential yield from 6.5 t/ha, which is substantially lower than the observed yield of 7.3 t/ha in 2000 to 9.3 t/ha. Predicted rice planting areas using the refined AEZ parameterization expands significantly to coincide with the paddy field map of 2000 generated by remote sensing applications. Importantly, the procedure presents a convenient way to update the AEZ model with calibrated genetic parameters, which reflecting observed technological progresses at farm sites.