One of the hydrological components of the cycle is evaporation, which has actual quantities that are challenging to quantify in the field. As a result, estimations of the evaporation rate's value are made using empirical relationships derived from data on climate components. Several applications of water resources, including hydrological, hydraulic, and an optimal agricultural irrigation system, depend heavily on accurate estimation of evaporation losses. Accurately estimating and forecasting hydrological phenomena is thought to be one of the most critical aspects of managing and developing water resources, as well as creating future water plans that consider various climate change scenarios. The Artificial Neural Network (ANN) and Support Vector Regression (SVR) methods are cutting-edge models that have been employed in several recent research to estimate various hydrological parameters. In the current study, the evaporation rate of Haditha Dam Lake on the Euphrates River in the Al-Anbar Governorate, Iraq, was predicted using ANN and SVR methods. It was designed to receive daily meteorological data, such as temperature, sunshine duration, wind speed, and humidity levels. Evaporation was chosen as the network's output. The present study presented several input scenarios with different input variables to examine the performance of the proposed models. Several statistical indicators have been used to evaluate the prediction results which are root mean square error (RMSE), Nash-Sutcliffe efficiency (NSE), mean absolute error (MAE), and correlation (R2) the prediction accuracy. The outcomes demonstrated that ANN could predict evaporation value with a high degree of accuracy better than the SVR method. The best prediction model achieved high correlation and mean error between actual and predicted data.

One of the hydrological components of the cycle is evaporation, which has actual quantities that are challenging to quantify in the field. As a result, estimations of the evaporation rate's value are made using empirical relationships derived from data on climate components. Several applications of water resources, including hydrological, hydraulic, and an optimal agricultural irrigation system, depend heavily on accurate estimation of evaporation losses. Accurately estimating and forecasting hydrological phenomena is thought to be one of the most critical aspects of managing and developing water resources, as well as creating future water plans that consider various climate change scenarios. The Artificial Neural Network (ANN) and Support Vector Regression (SVR) methods are cutting-edge models that have been employed in several recent research to estimate various hydrological parameters. In the current study, the evaporation rate of Haditha Dam Lake on the Euphrates River in the Al-Anbar Governorate, Iraq, was predicted using ANN and SVR methods. It was designed to receive daily meteorological data, such as temperature, sunshine duration, wind speed, and humidity levels. Evaporation was chosen as the network's output. The present study presented several input scenarios with different input variables to examine the performance of the proposed models. Several statistical indicators have been used to evaluate the prediction results which are root mean square error (RMSE), Nash-Sutcliffe efficiency (NSE), mean absolute error (MAE), and correlation (R2) the prediction accuracy. The outcomes demonstrated that ANN could predict evaporation value with a high degree of accuracy better than the SVR method. The best prediction model achieved high correlation and mean error between actual and predicted data.