Validation of the AMBAV soil-water balance model to simulate soil water content in a grass field

The validation of soil water balance models and the evaluation of the quality of the model predictions at field-scale require time-series of in situ measured model outputs. Therefore, the objectives of this study were comparison of the simulated soil water content (SWC) by AMBAV (Agrarmeteorologisches Modell zur Berechnung der Aktuellen Verdunstung) model as a cheap and reliable method with the corresponding results by soil samples (observed data) and validation of AMBAV soil water balance model in a grass field in Braunschweig, Germany. AMBAV model use synoptic data sets and Penman-Monteith formula. This experimental dataset includes a period of 7 months. Three statistical tests, which are R2 (coefficient of determination), MARE (mean absolute relative error) and PE (prediction efficiency) index were performed. Differences between observed and simulated active root zone SWC were not significant (pr > t = 0.097). Low MARE values and high values of PE and R2 indicate that the soil water balance simulation model can be used safely to simulate the soil-water content in the active root zone of the grass. The AMBAV model was able to find the optimum date for grass-specific irrigation as an agro-meteorological.