Upscaling modelled crop yields to regional scale: A case study using DSSAT for spring wheat on the Canadian Prairies.
Huffman, E.C., Qian, B., DeJong, R., Liu, J., Wang, H., McConkey, B.G., Brierley, J.A., and Yang, J.Y. (2015). "Upscaling modelled crop yields to regional scale: A case study using DSSAT for spring wheat on the Canadian Prairies.", Canadian Journal of Soil Science, 95(1), pp. 49-61. doi : 10.4141/cjss-2014-076 Access to full text
Dynamic crop models are often operated at the plot or field scale. Upscaling is necessary when the process-based crop models are used for regional applications, such as forecasting regional crop yields and assessing climate change impacts on regional crop productivity. Dynamic crop models often require detailed input data for climate, soil and crop management; thus, their reliability may decrease at the regional scale as the uncertainty of simulation results might increase due to uncertainties in the input data. In this study, we modelled spring wheat yields at the level of numerous individual soils using the CERES–Wheat model in the Decision Support System for Agrotechnology Transfer (DSSAT) and then aggregated the simulated yields from individual soils to regions where crop yields were reported. A comparison between the aggregated and the reported yields was performed to examine the potential of using dynamic crop models with individual soils in a region for the simulation of regional crop yields. The regionally aggregated simulated yields demonstrated reasonable agreement with the reported data, with a correlation coefficient of 0.71 and a root-mean-square error of 266 kg ha-1(i.e., 15% of the average yield) over 40 regions on the Canadian prairies. Our conclusion is that aggregating simulated crop yields on individual soils with a crop model can be reliable for the estimation of regional crop yields. This demonstrated its potential as a useful approach for using crop models to assess climate change impacts on regional crop productivity.
- Date modified: