Projection of net irrigation requirement over India using a bias-corrected spatially downscaled dataset

Abstract

Abstract With the country's economy directly dependent on food grain production, accurate estimation of the net irrigation requirement (NIR) of crops is essential for maximizing production. The daily IMD precipitation (0.25° × 0.25°) and temperature (1° × 1°) gridded datasets (1971–2010) were considered for the present scenario. Two satellite-based crop coefficient layers were prepared for the study: (i) kc_EEFlux, the evaporative fraction layer from earth engine evapotranspiration flux (EEFlux) and (ii) kc_FEWS, the ratio of the famine early warning system network (FEWSNET) monthly actual evapotranspiration (ET) (ETa) to ET0_FEWSNET layers. The NIR was estimated for three different cropping pixels derived from the land-use land cover (LULC) map of India, namely, Kharif-only, Rabi-only, and double/triple pixels using ‘ET0 and NIR toolbar’. The National Aeronautics and Space Administration (NASA)'s Earth Exchange Global Daily Downscaled Projections dataset were employed for projecting the future scenario. Results showed discrepancies in the extreme values of FAO56-Penman–Monteith (PM) and Hargreaves–Samani's estimated reference ET (ET0) but followed a similar trend in all months (R2 = 0.88). Results also suggested that FEWSNET offered more reliable estimates of NIR over EEFlux. The projected ensemble NIR indicated a probable negative relative change during the Kharif season (RCP4.5: from −14.47 to −7.25%; RCP8.5: from −11.45 to −5.59%) while a probable positive relative change during the Rabi season (RCP4.5: 3.30–5.21%; RCP8.5: 2.34–8.83%).