Determination of the Earth's structure in Fennoscandia from GRACE and implications for the optimal post-processing of GRACE data

• Published in: Geophysical Journal International
• Main Authors: Steffen, Holger, Wu, Patrick, Wang, Hansheng
• Format: Text
• Language: English
• Published: Oxford University Press 2010
• Subjects: Gravity; Geodesy and Tides; Fennoscandia; Fennoscandian
• Online Access: http://gji.oxfordjournals.org/cgi/content/short/182/3/1295; https://doi.org/10.1111/j.1365-246X.2010.04718.x

Analysis of data from the Gravity Recovery and Climate Experiment (GRACE) satellite mission allows us to identify regions of long-term mass changes such as the areas of Glacial Isostatic Adjustment (GIA) in North America and Fennoscandia. As there are now more than 7 yr of data available, the determined trends are robust enough for the inference of viscosity structure of the Earth's mantle. In this study, we focus on the Fennoscandian rebound area as there are abundant high-quality terrestrial data to use as ground-truth. In the first step, GRACE data are taken to fix the optimal radial (1-D) viscosity profile and the lithospheric thickness combination, which are needed as background parameters in 3-D earth modelling. The results are in basic agreement with results based upon relative sea level and GPS data, showing a lithospheric thickness in Fennoscandia between 90 and 160 km and an upper mantle viscosity of about [2–4]× 1020 Pa s. The lower mantle viscosity is poorly resolved, however. In the second step, GRACE data are used to constrain the 3-D viscosity using spherical finite element modelling. In this case, the results also agree with past investigations, but GRACE data alone cannot discriminate between lateral heterogeneities in the mantle that are thermal in origin from those due to changes in chemical composition. More notably, we treat in detail GRACE-related questions such as implementation of an adequate Level-2 filter technique and identification of the best reduction method for hydrological mass change signals. It turns out that the Gaussian filter technique is the best for this type of investigation. Even the best global hydrology models used in GRACE investigations still fail to improve the mismatches—thus one should be careful not to blindly use them for ‘improving’ GIA models in North America or other centres of rebound. In conclusion, our study clearly shows that GRACE data greatly complement the study of GIA. As there are new GRACE releases in progress, and also in light of a new generation ...