GLOBAL SENSITIVITY ANALYSIS OF CROP GROWTH SUCROS MODEL APPLIED TO HUSK TOMATO

Abstract

Given that the husk tomato crop (Physalis ixocarpa Brot.), is relevant for Mexican horticulture, since is fourth in vegetables acreage, the SUCROS (a Simple and Universal Crop Growth Simulator) model for simulation of potential growth was used to implement a sensitivity analysis and model calibration. A crop growth experiment was carried out in order to quantify changes in biomass and leaf area. To identify which model’s parameters affect more the predictions of total dry weight, leaf weight and fruits weight a global sensitivity analysis was performed. Uniform probability density functions were defined for each parameter and the extended Fourier amplitude sensitivity test and Sobol methods were used to calculate first order and total effects indices, at the beginning, at the middle and at the end of the crop cultivation period. For the first method 5000 Monte Carlo Simulations were performed and 2700 for the second. The model was calibrated by means of the weighted least squares procedure using their more sensitive parameters. It was found that parameters: relative growth rate of leaf area, base temperature, specific leaf area, potential CO₂ assimilation rate, light use efficiency and radiation extinction coefficient, influence to a greater extent the behavior of the SUCROS model applied to husk tomato crop. The high correlation coefficients 0.99, 0.99, 0.96, 0.89 and 0.95 to total dry weight, stem dry weight, fruits dry weight and leaf area index, respectively, indicate a good fit of the model for overall cultivation period. The error between the estimated and measured yield was 6.2 t ha^-1.