Gravity–topography regression across Antarctica: implications for isostatic regimes and subglacial crustal structure

Renata Regina Constantino , Kirsty Tinto, Leonardo Uieda , Carla Braitenberg

We investigate the relationship between Bouguer disturbance and equivalent topography across Antarctica using satellite gravity data, under the assumption that a linear relationship is expected at long wavelengths for Airy-type isostatic compensation. Equivalent topography ensures consistency across continental, marine, and ice-covered domains by expressing bathymetry and ice loads as crustal-equivalent height. The analysis is based on the GOCO06S satellite-only gravity model and BedMachine v3 datasets. Bouguer disturbance is computed through forward modelling of all relevant mass contributions using prism integration. Spatial variations in the gravity–topography relationship are quantified using moving circular windows (264 km diameter, 75% overlap), consistent with the effective resolution of the gravity model and the wavelengths at which local compensation is expected. For each window, slope, intercept, correlation coefficient (r), coefficient of determination (R²), and RMSE are estimated. Regions such as the Antarctic Peninsula and the southern Transantarctic Mountains exhibit high R² and strong negative correlation, indicating a coherent long-wavelength coupling between gravity and topography. While consistent with Airy-type compensation, this behaviour may also reflect contributions from mantle density variations associated with thermal anomalies. In contrast, large sectors of East Antarctica show positive correlation and low to moderate R² values. Combined with seismic evidence for thick and cold lithosphere, this pattern indicates that gravity variations are influenced by lithospheric rigidity and regional compensation rather than local Airy-type crustal thickening. The joint interpretation of regression parameters further delineates candidate subglacial sedimentary basins, in agreement with independent probabilistic models, and identifies new regions of interest for future geophysical investigation.