Estimation of Daily Evaporation Rate using Artificial Neural Network in Shiraz and Zarghan Cities

Evaporation is one of the most important components of the hydrological cycle that plays a very important role in the management of water resources and the environment. Knowing the amount of water lost due to the evaporation process in an area, especially in arid and semi-arid areas that face shortages of water resources, is one of the most important management principles in regional planning. The aim of this study was to evaluate the accuracy of artificial neural network method in estimating daily evaporation in Shiraz meteorological station and its generalizability in Zarghan meteorological station located in Fars province. For this purpose, 1775 data on a daily scale from meteorological factors including temperature, relative humidity, wind speed, sunshine were collected and then the amount of daily evaporation was estimated using 4 models of artificial neural network. For modeling in this study, multilayer perceptron neural network and sigmoid function were used. The results obtained from four models of artificial neural network were evaluated based on the criteria of coefficient of determination (R2), Nash-Sutcliffe coefficient (NSC) and Root Mean Square Error (RMSe). The results showed that in Shiraz meteorological station, model 4 with a structure of 5-6-1 neurons has less RMSe and higher R2 and NSC in both training and testing stages than other models, so as a superior model to predict the rate of evaporation Was selected daily at Shiraz meteorological station. The results of the generalizability of Model 4 with 5-6-1 structure in Zarghan meteorological station also show the high accuracy of this model in predicting daily evaporation in this station, so it can be used as a suitable model to predict daily evaporation values in This station was used during periods when evaporation was not measured.