Spatial variability quantification of maize water consumption based on Google EEflux tool

Abstract

The evapotranspiration (ET) and crop coefficient (Kc) spatial variabilities are disregarded in traditional methods of evapotranspiration estimation based on lysimeters. With the development of remote sensing techniques, the estimative of ET on agricultural areas, in a specialized way, has become possible through the use of algorithms based on the surface energy balance such as the METRIC and its automated version, featured on the Google Earth Engine Evapotranspiration Flux (EEFlux) platform. This study was carried out at a center pivot irrigated area located in the city of Primavera do Leste, MT, Brazil. One growing season (2016) of the specie Zea mays (maize) was analyzed. Using processed images from the Landsat 8 satellite, within the EEFlux platform, the spatial variability of the actual evapotranspiration (ETa) and the Kc curve of this crop was determined. The water use efficiency (WUE) was also determined. A comparative analysis was performed using different statistical indices: root mean square error (RMSE), the mean bios error (MBE) and the index of agreement (d). The ETa for maize ranged from 1.3–4.1 mm d−1 and the Kc obtained ranged from 0.3 to 1.2. The average WUE of maize was 1.13 kg m-3. The method of estimation of ETa and Kc spatialized using the Google EEFlux platform made possible the understanding the spatial variability of these two variables and, therefore, this application has high potential to estimate the ETa and Kc on different stages of maize crop growth cycle.