A comparison of VegSyst and mod-VegSyst models in predicting dry matter, nitrogen uptake and transpiration of greenhouse-grown tomatoes

Abstract

The dynamic model in discrete time VegSyst for vegetable crops predicts thermal time (TT), daily dry-matter production (DMP), crop nitrogen uptake (Nup) and crop transpiration (ETc). Because this model makes photosynthetically active radiation (PAR) and ETc calculations that are very complex, a simplified model has recently been proposed. In mod-VegSyst, photo-thermal time (PTT) is proposed as a new state variable. Whereas VegSyst calculates ETc according to the FAO method, as the product of reference evapotranspiration and crop coefficients, in mod-VegSyst, ETc is calculated by the Baille transpiration model for greenhouse crops. Also, mod-VegSyst uses a simpler exponential light interception function based on the leaf area index (LAI), which is a function of PTT, to calculate DMP and Nup. The aim of the current work was to compare the quality of predictions of the two VegSyst models using data for global solar radiation, air temperature and relative humidity from an experiment carried out in Chapingo, Mexico, during autumn-winter 2015, in which tomatoes (Solanum lycopersicum L) were grown in an experimental greenhouse. Firstly, the parameters of both models were calibrated using nonlinear least squares estimation. The performance of the two models was then evaluated using squared root mean squared error (RMSE) and mean absolute error (MAE) statistics. The results showed that DMP and ETc rates are better predicted by mod-VegSyst than by VegSyst, whilst Nup was predicted similarly by the two models.