A Remote Sensing Approach to Map Crop Stress and Optimize Irrigation through Crop-Water Production Functions

Abstract

Irrigated agriculture consumes over 70 % of global freshwater, intensifying competition among agricultural, urban, industrial and environmental needs. This study presents a geospatial framework that integrates remote sensing–derived evapotranspiration (ET) with crop–water production (CWP) functions to map wheat stress and optimize irrigation across Allahabad district, India. Using surface energy-balance modelling (METRIC/SEBAL) and FAO 56 reference-ET, we estimated actual (ETa) and potential (ETp) evapotranspiration for the December 2015–April 2016 season. Remote sensing–based biomass and harvest-index models yielded actual and potential grain output, enabling stage-wise yield response factor (Ky), calculation - the key coefficient in many CWP formulations via the Doorenbos–Kassam approach (FAO 33). At the district scale, wheat exhibited Ky values of 0.6, 1.5 and 1.1 for vegetative, flowering and grain-filling stages, respectively (overall Ky = 1.1), indicating high sensitivity to water deficits. Tehsil-level analyses revealed pronounced spatial variability in Ky, driven by differences in irrigation timing, volume and management practices. The Ky functions obtained using predicted yield can pinpoint “hot spots” where targeted irrigation at critical growth stages can maximize yield per drop. Embedding these spatially explicit CWP functions into an operational decision-support tool allows growers and agencies to schedule water more precisely, reallocate scarce supplies across fields, and bolster regional water security. This remote sensing–CWP fusion offers a scalable strategy for precision irrigation that enhances crop water productivity under constrained resources.