Estimating of Evapotranspiration Using Artificial Neural Network

Abstract

This study investigates the application of artificial neural networks (ANNs) on the prediction of daily grass reference crop evapotranspiration (ET0) and compares the performance of ANNs with the conventional method (Penman-Monteith). The use of ANNs was examined number of hidden layers and the activation function were also tested. The best ANN architecture for estimation of daily ET0 was obtained for different data set for Nubaria. Using these data, the networks were trained with daily climatic data (maximum and minimum temperature, dew point and wind speed) as input and the Penman-Monteith (PM) estimated ET0 as output. The analysis was carried out with MATLAB software. Feed forward onelayer networks with sigmoid function were used. Performance evaluation of the models have been carried out by calculating root mean square error (RMSE), The network were selected based on maximized R and R 2 value and minimized RMSE values which were 0.98, 0.957 and 0.44 mm/day, respectively in testing. The optimal ANN (4-12-1) for Nubaria regions showed a satisfactory performance in the ET0 estimation. These ANN models may therefore be adopted for estimating ET0 in the study area with reasonable degree of accuracy.

Performance Evaluation of Artificial Neural Networks for Estimating Refer-ence Evapotranspiration in Shahat, Libya using limited climatic data

“…Where the values of RMSE, R 2 and NSE were, 0.26, 0.98 and 0.97 respectively. These results correspond to the results of (Genaidy, 2020), (Heramb et al, 2023), (Rajput et al, 2023), (Abdel-Fattah et al, 2023), (Tunalı et al, 2023).…”

“…Several types of research have been conducted using ANNs to estimate evapotranspiration as a function of climatic elements (Kumar et al, 2002), (Sudheer et al, 2003), (Odhiambo et al, 2001), (Trajkovic et al, 2003), (Achite et al, 2022), (Genaidy, 2020), (Heramb et al, 2023), (Rajput et al, 2023), (Abdel-Fattah et al, 2023), (Tunalı et al, 2023), (Ekhmaj, 2012) and (Ekhmaj et al, 2013). These researches found satisfactory results, compared with those obtained from the FPM method.…”

Artificial Neural Networks for the Prediction of the Reference Evapotranspiration of the Peloponnese Peninsula, Greece

“…The tested architecture was based on the trial-and-error method and the ANNs were trained using the Levenberg-Marquardt algorithm [40,41]. Hyperbolic tangent (Equation (1); [85]) was the utilized activation function, based on the literature [71,86]. The hyperbolic tangent, along with the sigmoid function, is a non-linear function widely used as an activation function in ANNs [85].…”

Remote-Sensing Inversion Method for Evapotranspiration by Fusing Knowledge and Multisource Data

“…A study by Genaidy et al [19] was one of the first studies that received wide attention on the use of neural networks for evapotranspiration estimation. Yang et al [20] used a support vector machine approach to successfully estimate the evapotranspiration for the contiguous United States at 8 d using surface temperature, enhanced vegetation index, and surface cover in combination with incident shortwave radiation based on remotely sensed data and observations from 22 AmeriFlux sites.…”