Developing Water Management Strategies using Water Consumption Uniformity, Relative Evapotranspiration and Crop Water Productivity

Abstract

Accurate and reliable information on irrigation practices and crop water use is essential for evaluating and improving existing water management strategies. This study assesses irrigation equity, adequacy, and crop water productivity for two water-intensive crops—Jute and Rice—cultivated across West Bengal from 2012 to 2022. Actual crop evapotranspiration (ETa) was derived using the Google Earth Engine Evapotranspiration Flux product. Irrigation performance indicators were computed using Relative Evapotranspiration (RET) for equity and the Coefficient of Variation (CV) for adequacy. Crop type classification was carried out using the Random Forest Algorithm, a supervised machine learning technique. Crop water productivity (CWP) was estimated as the ratio of crop yield to crop water use, with crop yield variability inferred from Normalized Difference Vegetation Index (NDVI) patterns. For jute and rice, the relative errors in the evapotranspiration estimates ranged from 7–27% and 0–3%, respectively. Spatial fluctuation in ETa varied between 10% and 35%, reflecting substantial heterogeneity in water use across the study region. Higher values of RET indicate optimal irrigation, while lower values signify water stress. The analysis revealed that approximately 32.5% of the jute-growing area and 13% of the rice-growing area experienced excess water application, whereas 41.73% of jute and 37.30% of the rice areas were waterdeficient. These findings indicate significant spatial non-uniformity in irrigation practices, which may contribute to localized water scarcity and reduced water-use efficiency if not addressed through improved water management interventions.