Performance evaluation of the BUDGET model in simulating cotton and wheat yield and soil moisture in Fergana valley

Cotton (Gossypium hirsutum L.) and wheat (Triticum aestivum L.) are major crops grown in Uzbekistan and water shortage is considered as the main limiting factor for crop growth as well as sustainable economic development. The objective of this study was to adapt and test the ability of the soil water balance model BUDGET (ver. 6.2) to simulate cotton as well as wheat yield and soil water content under current agronomic practices in the Fergana Valley. Crop yield and soil moisture content data, collected and measured from sites in 2010 and 2011, were compared with model simulations. Results showed that the BUDGET can be used to predict cotton yield and soil water content with acceptable accuracy using the minimum approach. However, predicted wheat yield was high compared to the observed and reported yield. Overall, relationship between the observed and predicted cotton and wheat yield for both sites combined produced R2 of 0.91 and 0.15, RMSE of 0.24 and 1.64 t ha−1, relative Nash-Sutcliffe efficiency (Erel) of 0.71 and -5.68 and index of agreement (d) of 0.48 and -0.54, respectively. Similarly, comparison of the observed and simulated soil moisture contents at the top 0-30 cm soil layer and soil water contents in 90 cm profile yielded R2 of 0.88 and 0.71-0.88, RMSE of 2.74 %vol. and 21.4-28.7 mm, Erel of 0.87 and 0.53-0.81, respectively and d around 1.0. Consequently, the BUDGET can be a valuable tool for simulating both cotton yield and soil water content, particularly considering the fact that the model requires relatively minimal input data. Predicted soil water balance can be used to improve current practice of irrigation water management, whereas simulated soil moisture content can be used to estimate capillary rise from groundwater in the UPFLOW model. However, performance of the model has to be evaluated under a wider range of agro-climatic and soil conditions in the future.