Are Weather Generators Robust Tools to Study Daily Reference Evapotranspiration and Irrigation Requirement?

Abstract

Reference evapotranspiration (ETo) is one of the driving forces in crop simulation models and is very important to be estimated accurately. Moreover, weather generator (WG) models are widely used in combination with these crop models. As the quality of model output is related to the quality of weather data used as input, the evaluation of the sensitivity of model outputs to the quality of generated weather data is essential. In this study, eight different weather generator models were assessed and their outputs were used to estimate daily reference evapotranspiration and irrigation requirement. Two daily weather generator algorithms were combined with a monthly weather generator and/or an adjustment algorithm for low-frequency variances. Precipitation occurrence series was generated by an independent semi-empirical distribution. The daily weather generators outperformed the monthly models in reproducing daily statistics, while the monthly models performed better in simulating the monthly and yearly variations. After analyzing the model performances in simulating climatic variables, more assessments were carried out on ETo and irrigation requirement. The results depicted the strength of all the models in simulating daily ETo and irrigation requirement. Although all the studied models have comparable performances in simulating these two daily variables on daily and monthly scales, the monthly WGs outperform the daily models on yearly time scales and have better performances in simulating standard deviation values of yearly mean ETo and irrigation requirement. It can be concluded that WG models are robust tools for estimating these two daily variables if they can at least reproduce daily statistics (i.e. mean and standard deviation) well. But it must be taken in considerations that each WG model (including the one studied here) has different weaknesses and strengths and the best choice must be done according to the requirements.