Reducing leakage error in GRACE-observed long-term ice mass change: a case study in West Antarctica

Spatial leakage is a major limitation for quantitative interpretation of satellite gravity measurements from the gravity recovery and climate experiment (GRACE). Using synthetic data to simulate ice mass changes in the Amundsen Sea Embayment and Antarctic Peninsula, we analyze quantitatively the effects of a limited range of spherical harmonics (SH) coefficients and additional filtering, which in combination can significantly attenuate signal amplitudes. We present details of a forward modeling algorithm and show that it is capable of removing these biases from GRACE estimates. Examples show how to implement the method by constraining locations of presumed mass changes, or leaving these locations unspecified within a continental region. Our analysis indicates that leakage effects from far-field mass signals (e.g., terrestrial water storage change and glacial melting over other continents) on Antarctic mass rate estimates appear to be negligible. However, leakage from long-term ocean bottom pressure change in the surrounding Antarctic Circumpolar Current regions may bias Antarctic mass rate estimates by up to 20 Gigatonne per year (Gt/year). Experiments based on proxy GRACE measurement noise indicate that the effects of GRACE spatial noise on estimated Antarctic mass rates via constrained and unconstrained forward modelings are $$\sim $$∼5 and 15 Gt/year, respectively.