Last modified on Apr 07, 2026 07:33

Contributors: E. Boergens, C. Dahle, J. Haas, H. Dobslaw, F. Flechtner, A. Güntner

GFZ Helmholtz Centre for Geosciences

Issued by: GFZ / E. Boergens

Date: 23/09/2025

Ref: C3S2_313c_EODC_WP3-DDP-TWSA-v1_202507_PUGS

Official reference number service contract: ECMWF/COPERNICUS/2024/C3S2_313c_EODC

History of modifications

List of datasets covered by this document

Acronyms

General definitions

Gravity Recovery and Climate Experiment (GRACE): Twin satellite mission observing the time variable gravity field of the Earth from which water mass redistribution can be inferred. The mission was active between 2002 and 2017 and was a joint US-German project. 

GRACE Follow-On (GRACE-FO): Identically constructed successor satellite mission of GRACE, launched in 2018. The mission is again a joint US-German project.

Terrestrial Water Storage Anomaly (TWSA): TWSA represents the deviation or anomaly of terrestrial water storage (TWS) in a certain epoch (e.g., a month) from the long-term (2003-01 - 2022-12) mean TWS in all hydrological water storage compartments: groundwater storage, soil moisture, surface water storage, snow storage, and glaciers. Globally, TWSA can only be observed with GRACE and GRACE-FO. 

Uncertainty: Uncertainty measures jointly the precision and accuracy of the TWSA data, expressed as a standard deviation. The uncertainties of the TWSA grids are influenced, among others, by the uncertainties of the sensors on the satellites, uncertainties of background models applied in the processing chain, the orbit configuration of the missions (near-polar orbit), or environmental effects such as solar activity.

Leakage: Leakage refers to the inability to precisely localise signals in the GRACE-derived data sets, due to factors such as the band-limited resolution of GRACE and filtering. Together, leakage leads to apparent signal loss (leakage out) or gain (leakage in) inside a given integration region.

Glacial Isostatic Adjustment (GIA): Glacial isostatic adjustment describes the deformational behaviour of the solid Earth in response to glacial loading processes, particularly resulting in surface displacements due to stress, gravity, and sea level changes.

Spherical Harmonics (SH): Mathematical functions defined on the surface of a sphere, commonly used to express potential fields such as the Earth's gravity field by solving Laplace's equation.

Spherical Harmonic Coefficients: Commonly used representation of a global gravity field model based on the Spherical Harmonics basis functions. They are jointly estimated by least-squares adjustment, together with other parameters (instrument and orbit), using satellite observations provided by GRACE and GRACE-FO.

Level-2 data: Time-variable, usually monthly, gravity fields given in spherical harmonic (SH) coefficients.

Level-2B data: Post-processed Level-2 data, still given in SH coefficients. The post-processing includes several corrections that are applied to obtain as accurate as possible TWSA data.

Level-3 data: Time-variable, usually monthly, gravity fields given in the spatial domain as gridded data, based on Level-2B data. TWSA data is a Level-3 data set. 

GCOS Product Target Requirements: For each ECV, measurable requirements of the product performance are defined. These requirements encompass quantitative criteria along several dimensions, including spatial resolution, temporal resolution, measurement uncertainty, stability over time (bias drift), and timeliness (latency of data availability). For each target requirement, a threshold and a goal value are defined. Threshold is the minimum acceptable requirement. If a product fails this, it is presumably not useful for climate monitoring. The goal is the ideal level. Once this is met, further improvements in that criterion are not strictly necessary for most applications. 

Executive summary

This Product User Guide and Specifications (PUGS) is a self-contained document that gathers all necessary information for using and applying the Terrestrial Water Storage Anomaly (TWSA) dataset, as delivered within the Copernicus Climate Change Service (C3S).

The Copernicus Climate Change Service (C3S) Terrestrial Water Storage (C3S TWS) provides the operational service to generate data sets of the Essential Climate Variables (ECV) TWSA and groundwater storage change (GWSC) for a wide variety of users in the climate change community. This document covers the TWSA data product. The TWSA dataset distributed via C3S aims to provide users with ready-to-use TWSA data without the need for further pre-processing, such as corrections or filtering.

This document provides a brief summary of the data processing involved in generating the TWSA dataset in Section 1.1. A more extensive description of the data processing algorithm can be found in the Algorithm Theoretical Basis Document (ATBD,  Boergens et al., 2025a). This section also includes details on the spatial and temporal characteristics of the product. Section 1.2 then details the Global Climate Observing System (GCOS) user requirements and their fulfilment with the dataset. This is followed by an example visualisation of key variables in Section 1.3. Section 1.4 collects data usage information. This includes, importantly, the known issues and limitations of the TWSA dataset, discussed in Section 1.4.4, which are especially relevant for users unfamiliar with TWSA data. This document concludes with Section 2, which provides information on data access.

Product Description

Terrestrial Water Storage Anomaly (TWSA) description 

Terrestrial Water Storage Anomaly (TWSA) refers to changes in the total storage across all terrestrial water compartments. These include storage from groundwater, soil moisture, surface water bodies, to snow and ice. TWSA is therefore an essential state variable in the global hydrological cycle and has been designated as an Essential Climate Variable (ECV) by the World Meteorological Organization (WMO). Figure 1 provides a schematic illustration of TWSA’s role in the global hydrological cycle. TWSA is a key indicator of how the hydrological cycle responds to natural and anthropogenic variability. TWSA highlights changes in water availability, extreme events such as droughts and floods, and it helps to link climate variability, precipitation, and evapotranspiration to land water storage. By capturing these imbalances, TWSA plays a crucial role in understanding water cycle dynamics, water security, and climate–hydrology interactions on regional to global scales. 

Currently, only the satellite missions GRACE (Gravity Recovery and Climate Experiment, 2002–2017) and GRACE-FO (GRACE Follow-On, since 2018) are capable of measuring TWSA. GRACE and GRACE-FO detect temporal and spatial variations in the Earth’s gravity field, which, over land, are primarily caused by the redistribution of water masses. TWSA is thus directly inferred from these variations in the gravity field.  

Figure 1: The role of TWSA in the global hydrological cycle. It represents deviations from the average amount of water stored on land in components such as soil moisture, groundwater, snow, ice, and surface water. Thus, it indicates changes in the overall water availability.

TWSA data have been used in numerous hydrological and glaciological applications in recent years. The global availability of TWSA data enables assessments of changes in the magnitude and frequency of extreme events such as droughts and floods. Moreover, TWSA enhances our understanding of regional hydrological processes. Also, the glacier ice mass loss in regions such as Alaska, Patagonia, and the Himalayas can be quantified using TWSA. These are just a few examples of its application range.

Data processing algorithm

The main inputs to the TWSA production system are the Combination Service for Time-variable Gravity Fields (COST-G) spherical harmonic (SH) coefficients (Level-2 data) from GRACE and GRACE-FO (Meyer et al., 2023, 2025). Figure 2 presents an overview of the processing steps required for producing the TWSA data. Comprehensive details of the algorithm employed in the TWSA production system are available in the ATBD (Boergens et al., 2025a). The current version of the algorithm is TWSA v1.0.

Figure 2: Overview of the processing steps from GRACE and GRACE-FO Level-2 data to the TWSA (Level-3) data product. The processing steps leading to COST-G Level-2 data are not part of the C3S TWSA production system. The processing from Level-2 to Level-2B is conducted in the spherical harmonic domain (indicated in green), whereas the subsequent processing steps from Level-2B to the final TWSA product are performed in the spatial domain (indicated in blue). Additional independent input data required for the processing of TWSA are referred to as auxiliary data.

Product characteristics

Processing level

The processing level of this product is Level-3 according to the definition by NASA EarthDATA¹ .

Spatial information

The Level-3 data are provided on a regular 0.5° x 0.5° latitude-longitude grid. This results in a dimension of 720 grid cells in longitude and 360 in latitude for the data fields. Of these 259200 grid points in total, 96054 are land points according to the applied land-ocean mask, which is derived from the European Space Agency - Climate Change Initiative (ESA-CCI) global map of open water bodies v4.0 (CCI WB v4.0, Lamarche et al., 2017)².

Longitude and latitude values are expressed with respect to the WGS84 ellipsoid. The given coordinates describe the central point of each grid cell.

Temporal information 

The time variable represents the mean day of the data collection period used for each solution. Typically, the data product has a monthly temporal resolution; therefore, the time stamp corresponds to the 15th or 16th of the respective month.

However, due to mission operations or instrument issues, some monthly solutions are unavailable, of lower data quality, or based on data collection periods that deviate from one month. Additionally, owing to the gap between the two missions, GRACE and GRACE-FO, no data are available between July 2017 and May 2018. Figure 3 shows all available monthly solutions, while Table 1 provides details of the missing months (excluding the data gap), months with reduced data quality, and solutions derived from non-standard data collection periods. The given data collection periods refer to the dates of the first and last days of observations used in any of the solutions provided by the different analysis centres (AC), i.e., some solutions may use a shorter data collection period.

Figure 3: Collection of all available TWSA monthly fields for Climate Data Record (CDR) v1.0. 

Table 1: Details on the deviations from the monthly solutions, including missing data, monthly solutions with poorer data quality, and solutions where the data collection deviated from the monthly time frame.

Timeliness and update frequency

The entire data processing chain, from data collection to the generation of Level-3 TWSA data, requires approximately three months to complete. Consequently, at the time of publication of a CDR or an ICDR, the most recent available time step dates back at least three months.

An ICDR will be published every six months.

Overview of Product Target Requirements

The product target requirements have been defined by the GCOS (WMO, 2025). Table 2 summarises the requirements and characteristics of the C3S TWSA product.

For each requirement, a threshold and a goal value are defined as listed in Table 2. The goal is the ideal level. Once this is met, further improvements in that criterion are not strictly necessary for most applications.  Threshold is the minimum acceptable requirement. 

Table 2: GCOS goal and threshold requirements for the ECV TWSA. Colours: green - within goal, yellow - within threshold, red - does not meet threshold, grey - not applicable

Example visualisation of key variables

Figure 4 shows, as an example, (a) TWSA and (b) TWSA uncertainty for March 2008. In (c), the flag_filter field is shown for the entire time series. More details on the variables are described in the ATBD (Boergens et al., 2025a) and in the case of flag_filter, in subsection 1.4.2.

Figure 4: Fields of the final data product. (a) shows TWSA from the final data product for the example month of March 2008 stored in the twsa variable in the netcdf file; (b) shows the uncertainty of TWSA for the same month, stored in the twsa_uncertainty variable of the netcdf file; (c) displays the time series of the used filter strength for the interannual signal, which is stored in the flag_filter variable of the netcdf file.

Data usage information

Data format and file naming 

The data file is in "NetCDF-4" format and is CF-Convention v1.8 compliant.

File naming convention:

The data set file name for both CDRs and ICDRs follows the following naming convention, with the tokens of the filename provided in Table 3:

GGG_HHH_III_JJJ_KKK_LLLMMM_NNNOOO_PPP.nc

Table 3: Tokens of the filename

Example:

C3S_TWSA_GLOBAL_MONTHLY_200204_202503_v1.0.nc

Contains the global grid of monthly TWSA data, version 1.0. The temporal coverage spans from April 2002 to March 2025.

Quality flags and data masks

Due to the specific error characteristics of the GRACE and GRACE-FO data, a dedicated time-variable DDK (VDK) filter must be applied during processing (see ATBD, Boergens et al., 2025a). Although these filters adjust their strength according to the covariances of each solution, in some months, they may not sufficiently reduce noise. In such cases, a stronger filter is applied—for example, VDK2 instead of VDK3. Figure 5 shows an example of this problem. In February 2015, the solution was very noisy and even after filtering the data with the VDK3 filter, strong striping artefacts remained. Thus, the stronger VDK2 had to be applied.

Figure 5: Comparison between the different filter strengths. February 2015 is one of the months, where the usually applied VDK3 filter is too weak to remove the striping errors. Instead, the stronger VDK2 has to be used, which is documented in the variable flag_filter.

The filter applied for each solution is indicated in the data file by the variable flag_filter. While months processed with the stronger filter remain usable, they have reduced spatial resolution.

File contents

The data is provided as a single netCDF file for the entire time series.

File dimensions

The file dimensions are detailed in Table 4 below.

Table 4: File dimensions

File variables

The variables in the netCDF file are summarised in Table 5, and more details are provided below.

Table 5: Variables in the TWSA product.

time

Data type: double

Dimensions: (time)

Description: The mean day of the data collection used to estimate the TWSA data field

lon

Data type: float

Dimensions: (lon)

Description: Central longitude of each grid cell

lat

Data type: float

Dimensions: (lat)

Description: Central longitude of each grid cell

twsa

Data type: double

Dimensions: (time, lat, lon)

Description: Gravity-based terrestrial water storage anomaly given as equivalent water height

std_twsa

Data type: double

Dimensions: (time, lat, lon)

Description: Uncertainty of the terrestrial water storage anomaly given as equivalent water height

flag_filter

Data type: double

Dimensions: (time)

Description: Flag indicating the filter strength (VDKx) used for the residual interannual signal in each month. For example: 2=VDK2 used, 3=VDK3 used

File metadata

The netCDF file contains descriptive metadata as global attributes (Table 6 for all attributes and their values). 

Table 6: Global attributes included in the product files: Values set in italic are valid for the CDR v1.0 data version but will change with future CDR and ICDR publications.

Examples of known climate applications and best practices

As a new C3S dataset, the C3S TWSA product has not yet been used in climate applications. However, earlier TWSA products have been widely applied in various climate-related studies. A TWSA dataset processed in a manner similar to the C3S product has been included in the annual WMO State of Global Water Resources Report since 2021 (https://wmo.int/publication-series/state-of-global-water-resources, last accessed 15/09/2025). Likewise, TWSA has also been featured in the Global Water Report (https://www.globalwater.online/, last accessed 15/09/2025), which is published annually and with shorter latency than the WMO report.

Additionally, numerous climate applications of TWSA are presented at https://www.globalwaterstorage.info/en/ (last accessed September 15, 2025), an information portal dedicated to the GRACE and GRACE-FO satellite missions and their applications. These applications are showcased through blog articles written by scientists from the community, as well as a regularly featured “picture of the month".

Known Issues and Limitations

The principal limitation of TWSA data is its spatial resolution of approximately 250–300 km. This limited spatial resolution is primarily due to the attenuation of the gravitational field with increasing distance from the Earth’s surface. As a result, the Earth’s gravity field, as measured by GRACE and GRACE-FO at an orbital altitude of approximately 490 km, is already significantly smoothed compared to measurements taken at the surface. In addition, the data has to undergo filtering during processing (see Figure 2, step 2), which effectively acts as a spatial low-pass filter. This filtering cuts off features smaller than approximately 250km, depending on latitude and spatial orientation. Several studies have shown that, as a rule of thumb, vertical resolution of TWSA exhibits an uncertainty of approximately 15 mm equivalent water height over regions of about 200,000 km² (e.g., Landerer and Swenson, 2012; Vishwakarma et al., 2018).

A challenge closely associated with limited spatial resolution is the phenomenon known as leakage. In essence, leakage refers to the imprecise localisation of signals within the TWSA dataset. Leakage is inherent to the observation methodology of satellite gravimetry and the subsequent filtering mentioned above. Consequently, part of the hydrological signal in a given region is also detected in adjacent regions. Simultaneously, signals from neighbouring regions can leak into the region of interest. Together, these effects result in an apparent loss or gain of signal within a given area. A well-known example of leakage is the apparent mass loss towards the ocean adjacent to major ice mass losses, as observed in regions such as Greenland, Antarctica, or Alaska. It is well established in glaciology that glaciers are losing substantial amounts of ice mass; however, part of this signal leaks into the ocean and becomes obscured during TWSA data processing.

Another key limitation of TWSA data is its coarse temporal resolution, which is approximately one month. Typically, a full month of data is needed to estimate full global gravity field with adequate spatial resolution and accuracy. As a result, hydrological signals with shorter timescales cannot be reliably identified in TWSA data. Moreover, due to mission operations or instrument anomalies, certain monthly solutions may be entirely unavailable, of reduced quality, or require more than 30 days of data collection for estimation. In addition, a data gap exists between the two missions, with no data available from July 2017 to May 2018. Further details regarding the temporal availability of the data are provided in Section 1.1.2.3.

Data access information

TWSA data

The TWSA data will be made available through the Copernicus Climate Data Store (CDS) web-based service (https://cds.climate.copernicus.eu/). To access the CDS and all its toolbox software, a free-of-charge registration is required. Data can be directly downloaded from the portal and used under the License to Use Copernicus Products (also included on the download page). All requests for user support shall be directed via the ECMWF Support Portal.

Input data

The COST-G Level-2 input data, provided as spherical harmonic coefficients, are published by Meyer et al. (2025) and available at https://icgem.gfz-potsdam.de/series/10.5880/COST-G.ICGEM_02_L2, URL last at 14/07/2025.

Auxiliary data

Low-Degree Spherical Harmonic Coefficients

The low-degree spherical harmonic coefficients used in the processing are published by Dahle et al. (2025) and can be downloaded from https://isdc-data.gfz.de/grace/GravIS/COST-G/Level-2B/aux_data/ (URL last accessed 26/06/2025).

Model of Glacial Isostatic Adjustment

The correction model can be downloaded from the Supporting Information of Peltier et al. (2018), https://agupubs.onlinelibrary.wiley.com/action/downloadSupplement?doi=10.1002%2F2016JB013844&file=jgrb52450-sup-0003-Data_S3.txt, URL last accessed 25/06/2025.

Land-Ocean mask

The applied land-ocean mask has been derived from the European Space Agency (ESA) Climate Change Initiative (CCI) global map of open water bodies v4.0 (CCI WB v4.0), which is accessible from https://maps.elie.ucl.ac.be/CCI/viewer/download.php (after registration, URL last accessed 26/6/2025). The land-ocean masked applied in the processing of the TWSA data is available upon request to the user-support portal (https://confluence.ecmwf.int/site/support).

Data citation requirements

Same as landing page once this is agreed with ECMWF/CDS

User Support

A dedicated support portal (https://confluence.ecmwf.int/site/support) has been set up by the Copernicus User Support (CUS) team, which provides support to users of the Copernicus Atmosphere Monitoring Service (CAMS) and C3S services at ECMWF. All enquiries about the TWSA dataset can be submitted through the service desk where appropriate agents will deal with it. Once submitted, the user may add comments or further information to the issue, including responding to questions / requests for additional information from the support team.

References

• Boergens, E., Dahle, C., Haas, J., Dobslaw, H., Flechtner, Güntner, A., 2025a: Terrestrial Water Storage Anomaly v1.0: Algorithm Theoretical Basis Document (ATBD). Document ref. C3S2_313c_EODC_WP3-DDP-TWSA-v1_202507_ATBD_i3
• Dahle, Christoph; Murböck, Michael (2025): Post-processed GRACE/GRACE-FO Geopotential GSM Coefficients COST-G RL02 (Level-2B Product). V. 0001. GFZ Data Services. https://doi.org/10.5880/COST-G.GRAVIS_02_L2B 
• Landerer, F.W., Swenson, S.C., 2012. Accuracy of scaled GRACE terrestrial water storage estimates: ACCURACY OF GRACE-TWS. Water Resour. Res. 48. https://doi.org/10.1029/2011WR011453 [URL last accessed 16/07/2025]

• Lamarche, C., Santoro, M., Bontemps, S., D’Andrimont, R., Radoux, J., Giustarini, L., Brockmann, C., Wevers, J., Defourny, P., Arino, O., 2017. Compilation and Validation of SAR and Optical Data Products for a Complete and Global Map of Inland/Ocean Water Tailored to the Climate Modeling Community. Remote Sensing 9, 36. https://doi.org/10.3390/rs9010036 [URL last accessed 17/07/2025]

• Meyer, U., Lasser, M., Dahle, C., Förste, C., Behzadpour, S., Koch, I., Jäggi, A., 2023. Combined monthly GRACE-FO gravity fields for a global gravity-based groundwater product. Geophysical Journal International ggad437. https://doi.org/10.1093/gji/ggad437 [URL last accessed 16/07/2025]
• Meyer, Ulrich; Jaeggi, Adrian; Dahle, Christoph; Flechtner, Frank; Kvas, Andreas; Behzadpour, Saniya; Oehlinger, Felix; Mayer-Guerr, Torsten; Lemoine, Jean-Michel; Bourgogne, Stephane; Lasser, Martin; Koch, Igor; Flury, Jakob; Chen, Qiujie; Wang, Changqing; Yan, Zhengwen; Zhou, Hao; Feng, Wei (2025): International Combination Service for Time-variable Gravity Fields (COST-G) Monthly GRACE/GRACE-FO RL02 Series. V. 2.1. GFZ Data Services. https://icgem.gfz-potsdam.de/series/10.5880/COST-G.ICGEM_02_L2 [URL last accessed 16/07/2025]

• Peltier, R.W., Argus, D.F., Drummond, R., 2018. Comment on “An Assessment of the ICE-6G_C (VM5a) Glacial Isostatic Adjustment Model” by Purcell et al.: The ICE-6G_C (VM5a) GIA model. J. Geophys. Res. Solid Earth 123, 2019–2028. https://doi.org/10.1002/2016JB013844 [URL last accessed 15/09/2025]

• Vishwakarma, B.D., Devaraju, B., Sneeuw, N., 2018. What Is the Spatial Resolution of GRACE Satellite Products for Hydrology? Remote Sensing 10, 852. https://doi.org/10.3390/rs10060852 [URL last accessed 16/07/2025]
• WMO, 2025. The 2022 GCOS ECVs Requirements, 2022 edition - Updated in 2025, https://library.wmo.int/idurl/4/58111 [URL last accessed 16/07/2025]

Related articles

• Page:
  TWSA v01: Product User Guide and Specification (PUGS) (Copernicus Knowledge Base)

  • twsa
• Page:
  TWSA v01: Product Quality Assessment Report (PQAR) (Copernicus Knowledge Base)

  • twsa