A comparison of vertical land motion observed by GPS and Space Gravity

Abstract

The Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow-On (-FO) missions monitor the Earth's temporal gravity field, allowing to track and study terrestrial water storage changes, groundwater depletion, ice-mass variations, ocean bottom pressure changes, sea level fluctuations, and solid Earth deformation. When possible, succeeding satellite missions become operational prior to decommissioning of the mission they are replacing. Meaning there is an overlap of measurements, which allow one mission to be validated against each other. This was not the case with the GRACE /GRACE-FO missions as there was a ~12-month gap between the missions. My thesis investigates the continuity of two gravity missions by comparing its measurements of geophysical processes with ground observations (GPS).

Height changes at the surface of the Earth are caused by multiple processes, including elastic deformation caused by changes in hydrological mass loads and viscoelastic deformation caused by Glacial Isostatic Adjustment (GIA). The geophysical signals observed by the gravity missions are the sum of both deformation signals. I estimated a GIA signal from a least squares inversion using GPS heights data and gravity data. This process enabled me to separate the elastic and viscoelastic components of the gravity measurements. Subsequently, I reconstructed the height timeseries by calculating the elastic deformation and adding it to the viscoelastic deformation. Finally, I compared the reconstructed timeseries against the observed GPS height changes to validate the accuracy of my GIA estimate.

This validation approach was found to be a quality assessment tool, not only for evaluating my estimated viscoelastic model, but also for assessing existing GIA models. This allows us to assess the accuracy of models and identify the areas where they require improvement. One of the models assessed included the ICE6G_D model, I found a pattern of underestimated GIA signal on the south and west coast of Greenland. Considering that the ICE6G_D model is currently utilized to correct GIA for gravity data, this finding holds significant implications. An underestimation of GIA by the ICE6G_D model causes an underestimation of mass balances in Greenland. Given Greenland's accelerated melt rates, accurately assessing mass balance changes requires precise GIA estimation.

This method relies on sufficient GPS observations in areas with detectable hydrology and GIA signals. This method can be applied to validate models in Fennoscandia, Laurentide and Greenland. Antarctica, however, has not been assessed as there are not sufficient GPS available for the approach to work. As it relies on the presence of GPS, this approach cannot estimate a regional GIA model unless the spacing between GPS is 200 km; however, it can accurately estimate the signal at the location of the GPS.