Abstract
•CSM-CROPGRO-Cotton model was calibrated over a wide range of planting dates.•GLUE and sensitivity analysis were found helpful for genetic coefficients estimation.•Multi-year field trials data of cotton phenology, growth, and yield were used.•Model has potential to simulate phenology, growth and seed cotton yield.•Multi-year analysis (1980–2013) revealed decrease in yield after 20-April planting.
Simulation models are widely used for making decision support especially under sub-optimal climatic conditions for yield improvement. The impact of uncertain weather conditions on cotton production can be explored with the aid of such models. Cotton, being the queen of fibers, enjoys itself a predominant position amongst all other cash crops and has the potential to narrow the gap between production and consumption of fiber and edible oil. The goal of the study was to evaluate the performance of Cropping System Model CROPGRO-Cotton for examining temporal variation in cultivars and to determine the potential impact of planting dates. The model was calibrated using a diverse range of field observation for phenology, growth and seed cotton yield (SCY) and its components of 20-April planting date, genetic coefficients were estimated using Generalized Likelihood Uncertainty Estimation (GLUE) and sensitivity analysis sub modules in CSM-DSSAT. Calibrated and evaluated results were reasonably good for phenology, growth, boll weight, SCY and yield attributes with good statistics. Simulation and observed data revealed the decreasing trend with delayed planting for all cultivars. Phenological phases were predicted well with high d-index (>0.94 and >0.85), and the dynamics of time series for LAI and biomass were predicted also well (d>0.96 and >0.98), respectively during model evaluation for all planting dates and cultivars. Seed cotton yield was simulated with lower RMSE (137–382) at final picking, while dynamics of time series for SCY were predicted reasonably well with high values of d-index (0.95–0.99) for all cultivars during model evaluation. Planting date analyses of the years (1980–2013) agreed with observed SCY trend, showed decrease of 27% for delayed planting from 20-April to 21-June while first too early planting (10-March) had also faced 8% reduction for all cultivars. Cultivars [MNH-886, NIAB-9811 (NIAB-Kiran) and NIAB-112] outperformed with higher SCY predominantly due to longer appropriate growing season by utilizing optimal weather conditions, attained optimum growth at all key phenological phases. Cultivar NIAB-112 performed well being a short duration and attained high SCY for late planting (1-June) as well during the growing years. Cotton cultivars planting during 1-April–10-May can be suggested for the farmer’s field to avoid weather stress and to improve utilization of resources for sustainable cotton production in the region. Result demonstrated the model potential for decision support for cotton management practices in the region, including identifying optimum planting dates for cotton production.