Prediction of Irrigation Water Requirements for Green Beans-Based Machine Learning Algorithm Models in Arid Region

• DOI: 10.1007/s11269-023-03443-x
• Publication Date: 2023-03-10T15:02:50Z
• Authors: Ali Mokhtar; Nadhir Al-Ansari; Wessam El-Ssawy; Renata Graf; Pouya Aghelpour; Hongming He; Salma M. Hafez; Mohamed Abuarab
• Keywords: 2. Zero hunger; Evapotranspiration; 13. Climate action; Water resources management; Long short-term memory; Climate change; Geoteknik och teknisk geologi; 15. Life on land; Geotechnical Engineering and Engineering Geology; 6. Clean water; Hybrid models

Abstract Water scarcity is the most obstacle faced by irrigation water requirements, likewise, limited available meteorological data to calculate reference evapotranspiration. Consequently, focal aims of investigation are assess potential machine learning models in forecasting requirements (IWR) snap beans evolving multi-scenarios inputs parameters figure out impact meteorological, crop, and soil on IWR. Six were applied, support vector regressor (SVR), random forest (RF), deep neural networks (DNN), convolutional (CNN), long short-term memory (LSTM), Hybrid CNN-LSTM. Ten variables including maximum minimum temperature, Relative humidity, wind speed, precipitation, root depth, basal crop coefficient, evaporation, a fraction surface wetted and, exposed used as input for with their combination, 8 scenarios designed. Overall models, best scenario was 4 (relative evaporation), however, DNN RF model 7 (root wetted, fraction). While weakest one group climatic factors 6 (maximum relative precipitation). Among hybrid LTSM & CNN accurate SVR had lowest estimation accuracy. The outcomes this research work could set up modeling strategy that would motion improvement efforts identify shortages IWR forecasting, which sequentially may alleviation strategies such policies sustainable use resources management. current approach promising has value other similar regions.