A 10% increase in global land evapotranspiration from 2003 to 2019

We calculate an ensemble of global land evapotranspiration (ET) for 2003 to 2019 over global land using a water-budget approach. We use 4 publicly available precipitation datasets (GPCPv2.3, MERRA-2, ERA-5 and NOAA-NCEP), 5 discharge estimates (JRA-55, and 4 independently calculated ocean -mass balance global discharge estimates) and water storage change derived from the Gravity Recovery and Climate Experiment (GRACE) and GRACE-Follow On (GRACE-FO) missions. We calculate 20 different estimates of global land evapotranspiration using all combinations of the precipitation and discharge datasets, and one estimate of total water storage change (computed using backward difference method and the GRACE/GRACE-FO total water storage change from JPLRL06). The data is presented as a timeseries from 2003 to 2019 at a monthly time step (in units of mm per year). We also provide an estimate of monthly uncertainty based on error in the precipitation data sets (defined as the standard deviation across the precipitation data), error in discharge (defined as standard deviation across the discharge data) and error in the water storage change (this is calculated using the GRACE formal error product) as well as the total error (from summing in quadrature the mean component errors) ('Global-land-ET-error-budget'). The primary dataset is an estimate for all land areas including the ice-sheets ('Global-land-ET'). We also provide two separate estimates of global land ET that: i) do not include the contribution of the ice sheets (Greenland and Antarctica) ('Global-land-ET-without-icesheets'), ii) do not include the contribution of Antarctica ('Global-land-ET-without-Antarctica'). For each ensemble member of ET, the data variable contains the name of the precipitation data set and discharge data set used. We also include the data that has been smoothed and gap filled using bootstrapping methods ('Global-land-ET-smoothing-bootstrap'). All data is monthly and in units of mm/year. 

We also include the global discharge ocean mass balance estimates that were used to estimate global land evapotranspiration. The data set of global discharge is available for 2002 to 2019 using an ocean mass balance approach. The data was created using ocean altimetry (AVISO/DUACS), ocean steric information (EN4), combined with ocean precipitation (GPCPv2.2, CMAP), ocean evaporation (OAFLUX), and also estimates of precipitation - evaporation calculated from ocean atmospheric moisture budget (MERRA-2, ERA-5). The discharge includes runoff from all land masses including the ice sheets. The data was created by H. Chandanpurkar, and is an updated version from Chandanpurkar et al. (2017). Details are available at: Chandanpurkar, H. A., Reager, J. T., Famiglietti, J. S., & Syed, T. H. (2017). Satellite-and reanalysis-based mass balance estimates of global continental discharge (1993–2015). Journal of Climate, 30(21), 8481-8495.