Model runs over Europe were conducted within the ESM project (www.esm-project.net/) for the Frontier Simulations supporting the water and matter fluxes from the European landmass to receiving water bodies (Baltic Sea, Atlantic Ocean and the Mediterranean Sea). Daily discharge from the mesoscale Hydrologic Model (mHM; Samaniego et al., 2010; Kumar et al., 2013; Code version: git.ufz.de/mhm/mhm git version: 35b5cb1) operated at the spatial resolution of 1/16deg for the simulation period from 1.1.1960-31.12.2022 across the European domain (Longitude -11 to 41 Latitude 35 to 72). Model runs were conducted within the ESM project (www.esm-project.net/) for the Frontier Simulations supporting the water and matter fluxes from the European landmass to receiving water bodies (Baltic Sea, Atlantic Ocean and Mediterranian Sea). Special consideration was given to the coastal cells by filtering out those (bordering) grid cells that do not have 100% landmass (i.e., cells with a significant proportion of water bodies/sea/ocean coverage). Meteorological forcing data are based on the E-OBS v21e (daily precipitation, temperature, Hofstra et al. 2009), potential evapotranspiration is based on the Hargreaves-Samani method. Soil characteristics are obtained from the global SoilGrids database (Hengtl et al. 2014; the land cover is derived from the Globcover_V2 (http://due.esrin.esa.int/page_globcover.php); geomorphological features are based on the GMTED2010 (Danielson et al., 2011). Model parameterization was constrained using the observed discharge time series from the GRDC stations (https://portal.grdc.bafg.de/), satisfying the following three conditions: gauge LAT>48degN, area> 5000km2, area <170000km2. Multi-basin calibration and validation were employed to check the consistency of model simulations following Rakovec et al., 2016 and Samaniego et al. 2019, as follows. Calibration objective function using KGE, DDS algorithm with 500 iterations, to account for uncertainty in the calibration process and the basin selections, 50 random initial conditions were randomly drawn sub-set of basins (N=6basins). The best parameter set in the cross-validations across 1201 basins was selected for the final run (ID: 542). A static 2D file of flow direction over Europe at the routing resolution 1/16deg. Internal upscaling to 1/16deg from the higher resolution (1/512deg) done within mHM (Code version: mesoscale Hydrologic Model (git.ufz.de/mhm/mhm git version: 35b5cb1). Special consideration was given to the coastal cells by filtering out those (bordering) grid cells that do not have 100% landmass (i.e., cells with a significant proportion of water bodies/sea/ocean coverage). Flow direction network (lat,lon) and routed runoff (time,lat,lon) at 1/16deg are provided as separate datasets. Meteorological forcing data of the mHM model from 1.1.2020 to 31.12.2022 are based on the E-OBS v26e.

Model runs over Europe were conducted within the ESM project (www.esm-project.net/) for the Frontier Simulations supporting the water and matter fluxes from the European landmass to receiving water bodies (Baltic Sea, Atlantic Ocean and the Mediterranean Sea). Daily discharge from the mesoscale Hydrologic Model (mHM; Samaniego et al., 2010; Kumar et al., 2013; Code version: git.ufz.de/mhm/mhm git version: 35b5cb1) operated at the spatial resolution of 1/16deg for the simulation period from 1.1.1960-31.12.2019 across the European domain (Longitude -11 to 41 Latitude 35 to 72). Model runs were conducted within the ESM project (www.esm-project.net/) for the Frontier Simulations supporting the water and matter fluxes from the European landmass to receiving water bodies (Baltic Sea, Atlantic Ocean and Mediterranian Sea). Special consideration was given to the coastal cells by filtering out those (bordering) grid cells that do not have 100% landmass (i.e., cells with a significant proportion of water bodies/sea/ocean coverage). Meteorological forcing data are based on the E-OBS v21e (daily precipitation, temperature, Hofstra et al. 2009), potential evapotranspiration is based on the Hargreaves-Samani method. Soil characteristics are obtained from the global SoilGrids database (Hengtl et al. 2014; the land cover is derived from the Globcover_V2 (http://due.esrin.esa.int/page_globcover.php); geomorphological features are based on the GMTED2010 (Danielson et al., 2011). Model parameterization was constrained using the observed discharge time series from the GRDC stations (https://portal.grdc.bafg.de/), satisfying the following three conditions: gauge LAT>48degN, area> 5000km2, area <170000km2. Multi-basin calibration and validation were employed to check the consistency of model simulations following Rakovec et al., 2016 and Samaniego et al. 2019, as follows. Calibration objective function using KGE, DDS algorithm with 500 iterations, to account for uncertainty in the calibration process and the basin selections, 50 random initial conditions were randomly drawn sub-set of basins (N=6basins). The best parameter set in the cross-validations across 1201 basins was selected for the final run (ID: 542). A static 2D file of flow direction over Europe at the routing resolution 1/16deg. Internal upscaling to 1/16deg from the higher resolution (1/512deg) done within mHM (Code version: mesoscale Hydrologic Model (git.ufz.de/mhm/mhm git version: 35b5cb1). Special consideration was given to the coastal cells by filtering out those (bordering) grid cells that do not have 100% landmass (i.e., cells with a significant proportion of water bodies/sea/ocean coverage). Flow direction network (lat,lon) and routed runoff (time,lat,lon) at 1/16deg are provided as separate datasets.

This experiment contains an instance of the analog weather generator unseen-awg (Version 1.0) and over 10,000 years of artificial time series generated with unseen-awg for European weather under present-day climate conditions. Long simulations of artificial weather data help studying the weather-related risks across many sectors. The simulations are composed of 500 21-year-long daily time series and stored as look-up tables. They can be expanded into multivariate spatiotemporal data using the provided reforecast dataset of impact-relevant meteorological variables. The provided unseen-awg generator instance uses default parameter settings. It can be loaded using the corresponding Python class. The instance includes a large dataset of pre-computed similarities. Compared to using unseen-awg without pre-computation, this enables substantially faster simulations.

The file “GCOS_EHI_1960-2020_Continental_Heat_Content_data.nc” presents an updated estimate of the global continental heat storage for the period 1960-2020. For the first time, the continental heat storage is assessed as composed by: ground heat storage due to changes in subsurface temperatures, inland water heat storage due to the warming of inland water bodies, and permafrost heat storage due to thawing of ground ice in the Arctic. Furthermore, we argue that all three components of the continental heat storage should be monitored independently of their relative magnitude, as heat gain in the three components alters several important climate phenomena affecting society and ecosystems. This file contains the total continental heat storage relative to 1960. The ground heat storage has been estimated by inverting 1079 subsurface temperature profiles form the Xibalbá database (https://figshare.com/articles/dataset/Xibalb_Underground_Temperature_Database/13516487) and a bootstrap technique to aggregate the Singular Value Decomposition (SVD) inversions of each profile (Cuesta-Valero et al., 2022a). The data are used in Cuesta-Valero et al. (2022b) and von Schuckmann et al. (2022).

The file “GCOS_EHI_1960-2020_Continental_Heat_Content_data.nc” presents an updated estimate of the global continental heat storage for the period 1960-2020. For the first time, the continental heat storage is assessed as composed by: ground heat storage due to changes in subsurface temperatures, inland water heat storage due to the warming of inland water bodies, and permafrost heat storage due to thawing of ground ice in the Arctic. Furthermore, we argue that all three components of the continental heat storage should be monitored independently of their relative magnitude, as heat gain in the three components alters several important climate phenomena affecting society and ecosystems. This file contains the total continental heat storage relative to 1960. The ground heat storage has been estimated by inverting 1079 subsurface temperature profiles form the Xibalbá database (https://figshare.com/articles/dataset/Xibalb_Underground_Temperature_Database/13516487) and a bootstrap technique to aggregate the Singular Value Decomposition (SVD) inversions of each profile (Cuesta-Valero et al., 2022a). The data are used in Cuesta-Valero et al. (2022b) and von Schuckmann et al. (2022). This version includes an update of continental heat content uncertainty, where the standard deviation has been corrected from the precedent version to consider properly the value from permafrost heat storage uncertainty.

This version of the fully coupled catchment simulation features the atmospheric model COSMO run at 1.1km (0.01°rotlat/lon grid), the land surface model CLM and the groundwater model Parflow, both run at 400m (regular lat/lon grid). Coupled with OASIS3-MCT.