Inter-comparison of ice sheet mass balance products from GRACE: ESA CCI Round Robin results

• Main Authors: Groth , A., Horwath, M., Horvath, A., Meister, Rakia, Sørensen, Louise Sandberg, Forsberg, René, Su, X, Shang, K., Shum, C.K., Ditmar, P., Ran, J., Klees, R., Wouters, B., Schrama, E., Shepherd, A.
• Format: Conference Object
• Language: English
• Published: 2016
• Subjects: /dk/atira/pure/sustainabledevelopmentgoals/climate_action; SDG 13 - Climate Action; Antarctic; Greenland; The Antarctic; Antarc*; Ice Sheet
• Online Access: https://orbit.dtu.dk/en/publications/e4c9dc96-70af-4d62-98dd-736b4bda4c55; http://lps16.esa.int/page_session185.php#1689p

Both the Antarctic Ice Sheet (AIS) and the Greenland Ice Sheet (GIS) have been identified as key parameters, so called Essential Climate Variables (ECV), in the climate system. Within the framework of the Climate Change Initiative (CCI) of the European Space Agency (ESA), reliable long-term satellite-based data products are generated for selected ECVs. Since ice sheet mass balance is an ECV parameter of highest interest, both the AIS_cci and the GIS_cci project will provide mass balance products based on satellite gravimetry data: (a) time series of monthly mass changes for individual drainage basins, and (b) gridded mass changes covering the entire ice sheet.Gravimetry Mass Balance (GMB) products are derived from data acquired by the GRACE (Gravity Recovery and Climate Experiment) mission. Although GRACE data have the advantage of being directly sensitive to mass changes, their limited spatial resolution and complex error structures place particular demands on the applied processing strategy. To choose the most suitable algorithm which minimizes the impact of GRACE errors and signal leakage errors on GMB products, an open Round Robin experiment was set up. Participants were asked to provide GMB products according to the specifications of the official products by the ESA CCI using their preferred processing strategy and GRACE release. In addition, the same algorithms should be applied to a series of synthetic datasets, which realistically mimic mass variations in various subsystems of the Earth (e.g. cryoshpere, ocean, continental hydrology). By comparing the derived synthetic mass changes with the a priori known 'synthetic truth', leakage errors can be quantified.Here we inter-compare the Round Robin results from six individual contributions, relying on different processing strategies, including regional integration approaches, mass inversion strategies, and a forward modeling approach. Time series of basin averaged and gridded products are compared with respect to their specific noise level. The minimization ...