ESA ESM 3.0

ESA Earth System Model 3.0

A high-resolution synthetic Earth system model and realistic de-aliasing error dataset for closed-loop satellite gravimetry simulations.

Overview

The realistic assessment of candidate constellations for future satellite gravity missions (such as GRACE-C, NGGM, and MAGIC) requires comprehensive end-to-end numerical simulations. These simulations model satellite orbits based on realistic global mass redistributions, retrieve time-variable gravity solutions, and evaluate post-processing techniques to mitigate spatially anisotropic errors.

To support these mission studies, the ESA Earth System Model 3.0 (ESA ESM 3.0) provides a major update over its predecessor (ESM 2.0). ESM 3.0 delivers a high-resolution, multi-component representation of global mass transport across Earth's atmosphere, oceans, hydrosphere, cryosphere, sea level, solid Earth, earthquakes, and deep ocean processes.

Figure : Standard deviation in cm EWH (a) and trend in cm/a EWH (b) of the combined AOHISBED coefficients after performing the spherical harmonic synthesis.

The variability of AOHISBED (Figure a) is dominated by variations in terrestrial water storage over land and to a smaller extent oceanic bottom pressure signals over the oceans. Prominently visible are also the effects of the co- and post-seismic signals from the largest earthquakes, particularly the Sumatra-Andaman and Tohoku-Oki events.

The global trend pattern of AOHISBED is dominated by GIA signals at the position of the former Laurentide Ice-Sheet as well as over Fennoscandia (Figure b). In addition, ice mass changes in both the continental ice-sheets and mountain glaciers are clearly visible. Hydrological trends, albeit smaller, are also pronounced in the Amazon and Congo river basins. Additionally, the co-and post-seismic deformation of the Sumatra-Andaman and Tohoku-Oki earthquakes contribute significantly to the local trend signal.

Key specifications

- - Time Coverage: 14 years (2007 to 2020)
  - Temporal Resolution: 6-hour time sampling
  - Spatial Resolution: Spherical harmonic expansion up to degree and order 180
  - Extended Time Series: Can be concatenated with ESM 2.0 (1995 to 2006) via dedicated bias fields to create a 26-year continuous dataset
  - Combined Mass Stream: Full AOHISBED layer representation (Stokes coefficients)

Subsystem model upgrades (ESM 2.0 vs. ESM 3.0):

A priori de-aliasing error models (AOe07 & HIe07)

A crucial requirement for closed-loop simulations is the availability of realistically noisy background models to reproduce high-frequency mass variability that would otherwise alias into monthly gravity retrievals. ESM 3.0 introduces two updated error streams:

- - AOe07 (Atmosphere and Ocean Error): Generated from 6-hourly surface pressure differences between ERA5 and MERRA2 reanalyses, combined with an MPIOM ocean model ensemble accounting for forcing uncertainties and mesoscale turbulence.
  - HIe07 (Hydrology Error): Derived from daily terrestrial water storage differences between LISFLOOD forced with ERA5 versus MERRA2.

Data access and citation

The ESA ESM 3.0 dataset (Stokes coefficients up to degree and order 180 at 6-hour resolution) and its associated error streams are freely available for download on the GFZ Information System and Data Center (ISDC):

Data Repository: https://isdc-data.gfz.de/future_missions/esaesm/ESAESM3/
Digital Object Identifier (DOI): 10.5880/GFZ.UEOS.2026.001

Project details and consortium

- - Funding Agency: ESA
  - Contract Name: SING (Contract No. 4000145265/24/NL/SC)
  - Lead Institution: GFZ (Potsdam, Germany)

Partner Institutions:

German Research Centre for Geosciences (GFZ, Germany); Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research (AWI, Germany); Delft University of Technology (TU Delft, Netherlands); The University of Tokyo (UTokyo, Japan); NASA Goddard Space Flight Center (NASA GSFC, USA); The Ohio State University (OSU, USA); Università degli Studi di Trieste (UniTS, Italy); Finnish Geospatial Research Institute (FGI, Finland).

Report abuse