Item

Abstract

Satellite gravimetry observations have been widely used to study mass change effects due to large earthquakes in the recent two decades. Since the launch of Gravity Recovery and Climate Experiment (GRACE) mission in 2002, monthly gravity field solutions from GRACE and GRACE Follow-On (GRACE-FO) have been helping quantify deformation effects and better understand geodynamics of gigantic earthquakes, such as the 2004 Sumatra-Andaman Mw9.1, 2010 Chile Mw8.8 and 2011 Tohoku-Oki Mw9.0 earthquakes. Usually the spherical harmonic (SH) coefficients in GRACE/GRACE-FO monthly gravity field solutions are contaminated by stripe errors, which hinders reliable detection and interpretation of seismic signals. Thus dedicated filters are necessary to suppress the noise in the retrieval of seismic signals from GRACE observations. Due to the similarity of the co-seismic signals and stripe errors in spatial patterns, the commonly used filters tend to attenuate or distort signals dramatically while suppressing noise. In this study, we propose a new filter by examining the spectral feature of co-seismic signals in the SH domain, which adopts a V-shape weighting matrix to filter the stripe noise, rather than the commonly used isotropic Gaussian weighting matrix. Case studies of the above gigantic earthquakes indicate that the proposed V-shape filter is able to improve the signal to noise ratio (SNR) and better retrieve co-seismic signals. We also carry out tests on the earthquakes with relatively smaller magnitudes (Mw 8.0~8.2) during the GRACE-FO observing period, and find that the new filter surpasses commonly used filters in detecting relatively weaker co-seismic gravity change effects.