Global Highly Generalized Land Evapotranspiration Dataset (HG-Land v1.0)

The global Highly Generalized Land Evapotranspiration (ET) dataset (HG-Land v1.0) employs the state-of-the-art Deep Forest machine-learning algorithm, incorporating meteorological and vegetation variables as input data. The HG-Land dataset provides global monthly estimations, measured in W/m2, with a spatial resolution of 0.5°. It covers the period from January 1982 to December 2018. The data is stored in a network Common Data Form (netCDF) format, conveniently within a single file. This file comprises five variables: time, longitude, latitude, ET, and ET standard deviation.