Wheat is the main winter crop in Rio Grande do Sul State, Brazil, and the calculation of soil water balance is an important element in simulation models for crop growth, development and yield. The objective of this study was to improve the water balance estimation in soil under wheat by two modified models. The soil water content was measured by the gravimetric method during the growing season of two wheat cultivars sown on three dates in Santa Maria, RS, Brazil. Soil water for crop was represented as the fraction of plant-available soil water (FASW). The performance of the original and the modified version of two water balance models, namely the Campbell & Diaz model and the Amir & Sinclair model, was evaluated by the root of the mean square error (RMSE). The modified Campbell & Diaz model is more realistic and has a greater potential for performing well in regions other than where it was developed, while the modified Amir & Sinclair model simulated the available soil water for a wheat crop better in the study region. A maximum root depth of 0.3 m is most appropriate to simulate the fraction of plant-available water in the São Pedro soil unit.
Triticum aestivum L.; phenology; environment; water stress
