Enhancing water scarcity resilience in Egypt through machine learning-driven phenological crop mapping and water use efficiency analysis

Abstract

Agriculture forms the backbone of Egypt’s economy, with the Nile Valley and Delta serving as key production zones for crops like wheat, rice, and clover. However, the sector faces mounting pressure from water scarcity, as it depends almost entirely on the Nile for irrigation, making it necessary to map major crops for assessing Water Use Efficiency (WUE) and informing agricultural planning. In this study, we used machine learning (ML) techniques—specifically Support Vector Machine (SVM) to time-series phenological data and optical indices (Enhanced Vegetation Index (EVI), Bare Soil Index (BSI), Land Surface Water Index (LSWI), Normalized Difference Vegetation Index (NDVI), and Plant Senescence Reflectance Index (PSRI)) to map major crop types—specifically rice (a summer crop),wheat and clover (winter crops) —across entire Nile Basin in Egypt. Training and testing showed satisfactory performance, with testing accuracy ranging from 0.73 to 0.82 and training accuracy from 0.70 to 0.90. In addition, this study evaluates responsiveness of crop WUE to Vapor Pressure Deficit (VPD) and other meteorological and biophysical factors—including solar radiation, precipitation, maximum temperature, gross primary productivity, and evapotranspiration. Our findings confirm VPD as dominant factor affecting WUE, with a 3.5 kPa threshold beyond which WUE no longer responds, signaling a physiological limit for water management. The projected VPD trend, based on ensemble analysis of Coupled Model Intercomparison Project Phase 6 models under SSP245 and SSP585 scenarios, indicates an increase in number of months with high VPD in future, reinforcing the need for adaptive irrigation strategies in the region.