Modeling and routing of surface evaporation from the Amir Kabir reservoir using the Mann-Kendall and neural network technology

Evaporation as a natural parameter due to the release of water from the upper part of mankind has always been of interest to scholars and researchers. In this study, we try to apply the artificial neural network model to estimate evaporation from the Amir Kabir dam and to evaluate the model accuracy. In this context, 18 years data from 1997 to 2014 were used and after consecutive try and error, the best structure for computing the amount of evaporation from the surface of the dam was selected. This structure has five neurons in the first, fourth and second layers that showed the best result in 1000 replications. Also, statistical coefficients obtained from the analysis using artificial neural network was considered in choosing the best structure with the amount of 0.9365 which was the highest amount among other tests and the amount of test and training data error were 0.0321 and 0.0311, respectively. In addition, general trend of effective data on evaporation was determined, using Mann-Kendall test on 15 years daily data. In Mann-Kendall method, temperature changes, wind speed and precipitation graphs had no significand trend and showed -1.69< U