Transforming GRACE mascon TWS-L3 data into verticaldisplacements using a regression approach:a case study of the Türkiye region

Abstract

The Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow-Onsatellite missions have been monitoring hydrological events on Earth's surface for nearlytwo decades. Monthly gravity solutions from these satellites are available as Level-2 (L2)spherical harmonic coefficients or as ready-to-use Level-3 (L3) data, typically representingTotal Water Storage (TWS) variations. L3 data, such as Goddard Space Flight Center (GSFC)mascon data, include essential corrections like post-glacial rebound and signal-leakage,and precisely represent TWS variations for specific regions, such as river basins, withoutadditional corrections. However, certain geopotential changes, such as groundwater-inducedvertical displacements, gravity anomalies and geoid height changes cannot be directly obtainedfrom these data. To evaluate these geopotential changes, L3 data needs to be transformedinto harmonic coefficient solutions. While this method is more computationally demandingcompared to adapting L2 data with necessary corrections, the question remains: How canL3 TWS data be directly transformed into other potential changes? In this study, we proposea regression approach for the T & uuml;rkiye region, using approximately hundred GSFC-masconblocks to convert TWS into groundwater-induced vertical displacements. Transformationparameters are estimated by considering outcomes from L2 data, specifically selecting DDK2-filtered data. The ratio between vertical displacement and TWS for each mascon is modeledby a quadratic function based on TWS magnitudes. Investigating residuals reveals a time-dependent pattern, which requires a second regression to model this aspect. This two-stepregression approach successfully transforms TWS into vertical displacements, with a root-mean-square error of about half a millimeter, providing satisfactory results for the region.