Comparison of Evapotranspiration Calculated and Estimated By Different High-Resolution Gridded Analyses For Northeast Brazil

Objective: to present the results of a climatological analysis of evapotranspiration (ET) in the Brazilian Northeast (NEB) between 1961 and 2020, highlighting its seasonality, the identification of homogeneous groups by cluster analysis, and positive trends over the decades. In addition, the text compares the accuracy of two datasets (MODGroETa and ERA5-Land), emphasizing the superiority of ERA5-Land for ET estimates, especially after bias correction. The central purpose is to highlight the relevance of continuously monitoring and evaluating ET data for different regions and time scales. Method: For this study, meteorological data from the National Institute of Meteorology (INMET) and high-resolution gridded analyses (0.1° x 0.1°) adjusted by elevation were used. These data underwent rigorous quality control and altitude correction. It used atmospheric reanalysis data with high spatial resolution (9 km) and hourly temporal frequency, based on observations from satellites, surface stations and snow data. The remote sensing model derived from MOD16 was used to estimate evapotranspiration (ET) based on the Penman-Monteith equation. It is available with a spatial resolution of 0.1° and a decadal time scale. Multivariate statistical technique applied to identify homogeneous areas in Northeast Brazil with similar ET characteristics, based on the monthly distribution of data. Results and Discussion: The climatology of ET (1961-2020) in the NEB showed strong seasonality. Between January and April, the highest values (>4 mm/day) are concentrated in the semi-arid region, decreasing in the west of Piauí and Maranhão (<3.5 mm/day). From May to July, ET decreases in the east, while from September to December, the interior registers maximums (>5.5 mm/day). Cluster analysis identified six homogeneous regions, independent of rainfall patterns, with higher ET from September to November and lower values between April and June, showing significant seasonal variations in the NEB. Research implications: The research contributes to water management, agricultural zoning and environmental conservation in the Brazilian Northeast (NEB). It identifies trends in evapotranspiration (ET) associated with climate change, provides subsidies for improving climate models and public policies, and fosters future research into adaptation and mitigation of impacts on ecosystems and communities. Originality/Value: The study is original in identifying regional patterns of evapotranspiration in the NEB, combining cluster analysis and multiple data sources. Its value lies in its practical application for water management, agricultural zoning, environmental conservation and public policies, strengthening climate adaptation strategies and mitigating impacts on vulnerable ecosystems and communities.