GCL-Mascon2024: a novel satellite gravimetry mascon solution using the short-arc approach

Abstract. This paper reports an innovative mass concentration (mascon) solution obtained with the short-arc approach, named “GCL-Mascon2024”, for estimating spatially enhanced mass variations on the Earth's surface by analyzing K-/Ka Band Ranging satellite-to-satellite tracking data collected by the Gravity Recovery And Climate Experiment (GRACE) mission. Compared to contemporary GRACE mascon solutions, this contribution has three notable and distinct features: First, this solution recovery process incorporates frequency-dependent data weighting techniques to reduce the influence of low-frequency noise in observations. Second, this solution uses variable-shaped mascon geometry with physical constraints such as coastline and basin boundary geometries to more accurately capture temporal gravity signals while minimizing signal leakage. Finally, we employ a solution regularization scheme that integrates climate factors and cryospheric elevation models to alleviate the ill-posed nature of the GRACE mascon inversion problem. Our research has led to the following conclusions: (a) the temporal signals from GCL-Mascon2024 exhibit 6.5 %−20.4 % lower residuals over the continental regions, as compared with the (Release) RL06 versions of other contemporary mascon solutions from GSFC, CSR, and JPL; (b) in Greenland and global hydrologic basins, the correlation coefficients of estimated mass changes between GCL-Mascon2024 and other RL06 mascon solutions exceed 95.0 %, with comparable amplitudes; especially over non-humid river basins, the GCL-Mascon2024 suppresses random noise by 36.7 % compared to contemporary mascon products; and (c) in desert regions, the analysis of residuals calculated after removing the climatological components from the mass variations indicates that the GCL-Mascon2024 solution achieves noise reductions of over 28.1 % as compared to the GSFC and CSR RL06 mascon solutions. The GCL-Mascon2024 gravity field solution (Yan and Ran, 2024) is available at https://doi.org/10.5281/zenodo.14008167.