Modification of CSM-CROPGRO-Cotton model for simulating cotton growth and yield under various deficit irrigation strategies

Abstract

Cotton (Gossypium hirsutum L.) is one of the main crops in Southern Xinjiang of China, and it is the major component of the regional economy. However, cotton cultivation in this region is facing severe challenges, e.g., the projected increase in the frequency of droughts and the decrease in precipitation in the future. Development and evaluation of deficit irrigation strategies in this region can potentially conserve irrigation water while maintaining cotton yields. In this study, the actual soil evaporation was introduced into the calculation process of stress factors in the Decision Support System for Agrotechnology Transfer (DSSAT) Cropping System Model (CSM-CROPGRO-Cotton) to simulate and evaluate the growth and yield of drip-irrigated cotton under mulching in response to various deficit irrigation strategies. The modified CSM-CROPGRO-Cotton model was calibrated using field data under full irrigation (100% reference evapotranspiration (ET0)) and validated by deficit irrigation data (0.4 ET0, 0.6 ET0 and 0.8 ET0 at the bud, flowering and boll stages) during the two cotton growing seasons of 2018 and 2019. The results showed that the modified model CSM-CROPGRO-Cotton simulated the growth index of cotton well under various deficit irrigation conditions. The index of agreement (d) of the simulated and measured biomass was greater than 0.845, and the coefficient of determination (R2) was greater than 0.967 (P < 0.01). Although the simulation of leaf area index was not as good as biomass, it also yielded satisfactory estimates, with d > 0.934 and R2 > 0.817 (P < 0.01). The simulated photoperiods in various deficit irrigation treatments were consistent with the observations. The averaged absolute relative error (ARE) in seed cotton yield over the deficit irrigation treatments was 2.1%, indicating high simulation accuracy. The results showed that the modified CSM-CROPGRO-Cotton model was appropriate for simulating cotton growth and yield under various water deficit strategies, which is a reasonable decision support system in managing irrigation for sustainable agriculture. The simulations also indicated that adopting deficit irrigation practices under normal weather conditions could conserve water without adversely affecting seed cotton yield, but there was a risk of increased yield loss under dry conditions.