M. Amiri
S. Eslamian
J Abedi-Koupai
M. Khozaei

There are many available methods to estimate evaporation from a water surface, comprising both direct and indirect methods. The probabilistic, statistical, and stochastic approaches require large amounts of data for the modeling purposes and therefore are not practical in local evaporation studies. It is necessary to adopt a better approach for evaporation modeling, which is the fuzzy sets and adaptive neural based Fuzzy regression method as used in this paper. Daily pan evaporation estimates have been achieved by a suitable fuzzy regression models for the meteorologicaldata recorded from five gauging stations which are located in Esfahan Province. The daily climatic data of the five gauging stations, including maximum and minimum temperature, maximum and minimum relative humidity, wind speedand sunny hours are introduced as input data and evaporation (EP) as output data. Results of fuzzy regression models approaches were analyzed and compared with measured daily pan evaporation values. The estimated EP values from a fuzzy regression model using five input parameters, including maximum and minimum temperature, mean relative humidity, wind speed and sunny hours were obtained with RMSE=0.77 mm/day, R2 = 0.89. Thus, in this study thefuzzy regression is work well for the data set used

Evaporation, Fuzzy regression, climatic data, modeling, indirect method