Monthly reference evapotranspiration for Brazil

The set of monthly ETo data referred to as monthly reference evapotranspiration for Brazil. It has a spatial resolution of 30 seconds (~ 1 km²) and a temporal resolution of 1 month. The data set grid is in GeoTIFF format, and corresponds perfectly to WorldClim. It uses the geographic coordinate reference system, with WGS84 projection (EPSG: 4326). The files are named as YEAR MONTH. Reference evapotranspiration (ETo) is a fundamental parameter for hydrological studies and irrigation management. The Penman-Monteith method is the standard for estimating ETo and requires several meteorological elements. Free remote sensing products with evapotranspiration information are rare. The objective of this study was to estimate the monthly ETo from the potential evapotranspiration (PET) made available by the MOD16 product. The monthly ETo estimated by the Penman-Monteith method was considered the standard. For this, data were acquired from the 265 meteorological station of the National Institute of Meteorology (INMET), throughout Brazil, in the period from 2000 to 2014 (15 years). Using machine learning algorithms, MOD16 images and WorldClim information as covariates, the ETo. All machine learning models were effective in improving the performance of the metrics evaluated. Cubist was the model that presented the best metrics for r² (0.91), NSE (0.90) and nRMSE (8.54%) and should be the preferred one for ETo prediction. The use of monthly ETo is recommended, which opens up possibilities for its use in numerous other studies. The article: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0245834