A Remote sensing-based geospatial framework for assessing crop water dynamics

Abstract

Food security depends on reliable agricultural production, and effective water management is essential for achieving good crop yields. Irrigated agriculture covers approximately 20% of cultivated land and produces about 40% of global food, while agriculture accounts for approximately 69% of global freshwater withdrawals (FAO AQUASTAT). Crop water requirements are commonly estimated using the FAO-56 approach, which uses crop coefficients (Kc) for different growth stages. However, applying this approach across large agricultural areas requires detailed information on crop type and development. This study combines a maize crop map with satellite-based Normalized Difference Vegetation Index (NDVI) data to estimate maize water requirements. The crop map identified maize fields; monthly NDVI described their development. A linear relationship between NDVI and Kc was used to estimate monthly Kc values. Crop evapotranspiration (ETc) was then calculated using Kc and reference evapotranspiration (ETo) and compared with satellite-derived actual evapotranspiration (ETa). The approach was applied to maizegrowing areas in Guanajuato, Mexico, from 2021 to 2024 at 30-m spatial resolution. Seasonal Kc patterns identified growing periods. Areas where ETa was lower than ETc were identified as potential crop water deficit areas. Seasonal ETc of 658-792 mm against ETa of 514-567 mm gave mean deficits of 110-226 mm over 91-97% of the maize area. These differences may indicate water stress but do not directly confirm stress because of uncertainties in the input data and water-use estimates. The approach provides spatially detailed information on maize water requirements and potential deficits, supporting improved crop water monitoring and irrigation management.