Gap-filling GRACE and GRACE-FO data with a climate adjustment scheme using Singular Spectrum analysis

Abstract

• A new approach with climate adjustment scheme using SSA (climSSA) for filling GRACE/-FO data gap is proposed. • The climSSA approach can significantly improve the representation of sub-seasonal and inter-annual signals in GRACE/-FO TWSA. • The performance of climSSA-filled TWSA is similar to or better than recent products. The satellite missions of Gravity Recovery and Climate Experiment (GRACE) and its follow-on (GRACE-FO) have provided two decades’ observations for global hydrological cycle and climate change communities. However, there exists 33-month data gaps, especially the 11-month gap during 2017–2018 within the missions, introducing significant biases and uncertainties in monitoring long-term trends and identifying extreme events from GRACE-based terrestrial water storage anomaly (TWSA). Although previous efforts to fill these gaps, a global gap-filled product that fulfills increasing desirable accuracy is still lacking. Here, we propose a novel approach to fill in the global TWSA data gap with climate adjustment scheme using Singular Spectrum Analysis (climSSA) together with the climate drought index (SPEI-6) as input. The climSSA has significantly increased the representation of sub-seasonal and inter-annual signals in TWSA relative to traditional SSA. Cross-validation analysis for a 24-month testing period suggests that the Nash-Sutcliffe Efficiency values between climSSA and GRACE-FO are significantly positive (≥0.3) in > 90 % global land and larger than 0.8 in > 50 % global land. Its performance remains steady after removing the contribution from seasonal cycles, and is similar to or better than three recent products and geodetic satellite laser ranging (SLR) satellites-derived data. Our approach suggests the merit of using climate drought index in reconstructing a long-term GRACE-like TWSA over the GRACE-gap and pre-GRACE periods.