Crop Water Footprint Estimation and Modeling Using an Artificial Neural Network Approach in the Nile Delta, Egypt
Ahmed Elbeltagi,Jinsong Deng,Ke Wang,Yang Hong
DOI: https://doi.org/10.1016/j.agwat.2020.106080
IF: 6.7
2020-01-01
Agricultural Water Management
Abstract:Modeling crop water use as a useful decision tool for agricultural policy decision-makers to improve water-use efficiency has been increasing. This study aimed to estimate, forecast, and model the green and blue water footprints (WFg and WFb, respectively) of maize by using an artificial neural network (ANN). Three Egyptian Nile Delta governorates were selected as major maize-producing sites: Ad Daqahliyah, Al Gharbiyah, and Ash Sharqiyah. The monthly data of minimum temperature (T-min), maximum temperature (T-max), precipitation (P), solar radiation (SR), soil moisture (SM), wind speed (WS), and vapor pressure deficit (VPD) data were obtained from open access data over the period from 2006 to 2016. The analyzed data were divided into two parts from 2006 to 2012 and from 2013 to 2016 for model training and testing, respectively. To predict WFb in the three governorates, the results show that the models with SR, humidity (H) and VPD; T-mean crop coefficient (K-c), and H; and SM, WS, VPD, and K-c were the best ANNs with different hidden layers (5, 3), (2, 6) and (7, 3), respectively. Furthermore, the findings showed that the optimal ANN for forecasting WFg included T-mean, WS, and P; P, WS, VPD, and SR; WS, T-max, and VPD with hidden neuron layers (7, 3), (7, 5) and (8, 5), respectively, for the three locations. The calculated WF values achieved a high statistical significant versus those simulated in the three sites with the lowest distributional variations, and the accuracy and coefficients of determination were close to 1. Moreover, for model testing, the findings indicated that the deviations between the actual and predicted WFs ranged from - 2.6 to 6.63 % and from - 2.4 to 3.16 % for the blue and green WFs, respectively. Thus, the developed models generated relatively better results and can help promote the decision-making process for both water managers and development planners.