Models used for designing cropping systems and for responding to cropping problems caused by climate variations must generate accurate predictions. Here, we describe the SIMPIÑA model, which simulates the development and growth of the "Queen Victoria" pineapple cultivar and which accounts for stress resulting from nitrogen and water deficiencies. We present the calibration and the validation of SIMPIÑA with 15 independent data sets derived from experiments carried out on Réunion Island and covering wide ranges of climatic conditions and management practices. Comparison of simulations with data sets shows that the predictive accuracy of SIMPIÑA is very good, with relative RMSE values ranging from 0.06 to 0.19for plant fresh biomass; such precision is sufficient for informing management decisions. Interestingly,there was no bias between observed and simulated values. A process-removal approach allowed us to determine how stress processes resulting from water or nitrogen deficiency influence the predictive capacity of the model across a broad range of climatic conditions. There was no clear trend for the effect of climate on model error in comparisons of the model with stress processes removed. When stress processes were partially removed from the model, fruit biomass error was particularly high when the effect of stress was removed from the radiation conversion efficiency and from biomass remobilization.Given its ability to correctly predict crop dynamics under contrasting conditions, SIMPIÑA appears to include the essential processes at the correct level of complexity.