On this article, we calculated international land evapotranspiration for 2003 to 2019 utilizing a mass-balance strategy. To do that, we calculated evapotranspiration because the residual of the water steadiness, utilizing an ensemble of datasets for precipitation, discharge and complete water storage change. We made an error in calculating the worldwide imply precipitation: we used arithmetic averaging to calculate the imply, as a substitute of calculating a spatially weighted imply to account for the altering grid field measurement with latitude. In consequence, the magnitudes of the worldwide imply precipitation time sequence have been underestimated. This impacted the next calculation of world imply evapotranspiration, ensuing within the imply evapotranspiration values being underestimated and altering some outcomes. We’re due to this fact retracting this text. We thank Ning Ma and others for bringing this error to our consideration.