Hydro-Physiological Controls of Crop Water Stress Under Salinity and Deficit Irrigation: An ANN-Based Framework for Sustainable Irrigation Management in a Changing Climate

Climate change is intensifying soil moisture variability, atmospheric evaporative demand, and salinity intrusion in agricultural landscapes, creating new challenges for sustainable food production. Understanding how soil hydrology and plant physiological stress interact under these conditions is essential for designing resilient irrigation strategies. This study presents a hydro-physiological assessment of wheat and maize grown under controlled combinations of soil salinity and deficit irrigation, and introduces an Artificial Neural Network (ANN) based Crop Water Stress Index (CWSI) model for real-time decision support in semi-arid farming systems of northern India.
Field experiments (2023–2025) were conducted to measure canopy temperature, air temperature, relative humidity, vapor pressure deficit (VPD), and soil moisture under varying salinity (EC levels) and irrigation regimes. These data were used to develop whole-season and stage-specific ANN models capable of capturing non-linear interactions between soil hydrology, crop physiology, and atmospheric demand. The ANN-based CWSI successfully distinguished mild-to-severe stress transitions and detected early-stage water stress acceleration during periods of high VPD, indicating a propensity toward flash drought development under combined salinity–moisture constraints.
Results show that salinity amplifies crop water stress by reducing effective root-zone moisture availability, leading to higher canopy–air temperature gradients and elevated CWSI values even under moderate irrigation. Stage-specific ANN models achieved strong performance (R² = 0.87–0.94), particularly during flowering and grain filling, where hydrological stress most affects yield. The framework demonstrates how data-driven CWSI modeling can translate complex soil–plant–atmosphere interactions into actionable irrigation insights for farmers.
This work highlights a scalable approach to precision irrigation scheduling, enabling reduced water use without compromising crop health in regions vulnerable to hydrological extremes and sociohydrological pressures. By linking soil hydrology, irrigation management, and physiologically informed stress indicators, the study contributes to sustainable food production strategies in a global climate change context.