Item

Abstract

Since 2002, the Gravity Recovery and Climate Experience (GRACE) satellite mission measured the time variable gravity field with an unmatched accuracy, leading to global observations of total water storage anomalies (TWSA). However, for climate change attribution and other applications such as closing the sea level budget, multi-decade TWSA time series are required. This has triggered several studies on reconstructing TWSA prior to 2002 via regression approaches or machine learning techniques, with predictors such as rainfall or temperature. However, the reconstructed signal properties of data-driven methods depend heavily on the selected predictors (climate data). Especially the linear trend of TWSA time series, which is mainly attributed to anthropogenic effects, is not or only partially represented by climate data and therefore not captured by the reconstructions. We combine a data-driven approach with satellite laser ranging (SLR) (Löcher and Kusche, 2021) with the aim of reconstructing all components of TWSA for the years 1990 onward. The Löcher and Kusche (2021) approach provides large-scale gravity information (including trends) from SLR ranges in a GRACE empirical orthogonal function (EOF) basis for the pre-GRACE time frame. Detailed information is added by a data-driven reconstruction, based on GRACE EOFs and regression techniques. Although our technique works globally, we will focus in particular on European basins.