干旱区生态环境知识资源中心

Evaporation, as a water loss, is important in water resources development and management. In arid zones where water resources are scarce, the estimation of this loss becomes more interesting in the planning and management of irrigation practices. This paper investigates the abilities of Adaptive Neuro-Fuzzy Inference System (ANFIS) technique to improve the accuracy of daily evaporation estimation. Different ANFIS models comprising various combinations of daily climatic variables, that is, air temperature, daily sunshine hours, wind speed, and relative humidity are developed to evaluate degree of effect of each mentioned variables on evaporation for two stations located in central part if I.R. Iran. A comparison is made between the estimates provided by the ANFIS models and the multiple lineal regressions. Various statistic measures are used to evaluate the performance of the models. Based on the comparisons, it was revealed that the ANFIS computing technique could be employed successfully in modeling of evaporation process from the available climatic data. The ANFIS also increased the accuracy of evaporation estimation compare to the multiple linear regression models.