Mathematical modelling of crop water productivity for processing tomato

Abstract

Crop water productivity models (CWPMs) are of great importance in evaluating different irrigation programs. The mean goal of the study was to evaluate the performance of the Jensen, Minhas, Blank, Stewart and Rao CWPMs in predicting fruit yield of processing tomato. Field experiments were conducted for two consecutive growing seasons. The soil water stress sensitivity indices of the CWPMs were determined using experimental data from the second crop growing season. Yields simulated by the CWPMs were compared with the experimental data for the first season. The sensitivity indices for the crop growth stages were taken into account as appropriate weights of the soil water sensitivity of the vegetative, flowering, yield formation and ripening stages of the processing tomato crop. The results give evidence that processing tomato is much more sensitive to soil water stress during flowering and yield formation stages whereas the adverse impact of water stress on yield is very limited at vegetative stage. The highest modelling efficiency (0.96) between field-measured and simulated yield by the model, the lowest arithmetic mean of errors (0.04), mean absolute deviation (0.07), mean square error (0.02), absolute percentage error (12.76), root mean square error (0.15) and coefficient of residual mass (0.05) were achieved by Minhas model and followed by Rao model based on same parameters of statistical analyses. Both the Minhas and the Rao models with their soil water stress sensitivity indices generated for the different growth stages obtained in this study are recommended for the processing tomato in the sub-humid environments.