The sensitivity of GNSS measurements in Fennoscandia to distinct three-dimensional upper-mantle structures

H Steffen,P Wu

doi:10.5194/se-5-557-2014

Abstract

Abstract. The sensitivity of global navigation satellite system (GNSS) measurements in Fennoscandia to nearby viscosity variations in the upper mantle is investigated using a 3-D finite element model of glacial isostatic adjustment (GIA). Based on the lateral viscosity structure inferred from seismic tomography and the location of the ice margin at the last glacial maximum (LGM), the GIA earth model is subdivided into four layers, where each of them contains an amalgamation of about 20 blocks of different shapes in the central area. The sensitivity kernels of the three velocity components at 10 selected GNSS stations are then computed for all the blocks. We find that GNSS stations within the formerly glaciated area are most sensitive to mantle viscosities below and in its near proximity, i.e., within about 250 km in general. However, this can be as large as 1000 km if the stations lie near the center of uplift. The sensitivity of all stations to regions outside the ice margin during the LGM is generally negligible. In addition, it is shown that prominent structures in the second (250–450 km depth) and third layers (450–550 km depth) of the upper mantle may be readily detected by GNSS measurements, while the viscosity in the first mantle layer below the lithosphere (70–250 km depth) along the Norwegian coast, which is related to lateral lithospheric thickness variation there, can also be detected but with limited sensitivity. For future investigations on the lateral viscosity structure, preference should be on GNSS stations within the LGM ice margin. But these stations can be grouped into clusters to improve the inference of viscosity in a specific area. However, the GNSS measurements used in such inversion should be weighted according to their sensitivity. Such weighting should also be applied when they are used in combination with other GIA data (e.g., relative sea-level and gravity data) for the inference of mantle viscosity.

Highlights

It is well known that observations of the glacial isostatic adjustment (GIA) process allow us to determine the earth’s viscosity structure, especially that beneath formerly glaciated areas such as Fennoscandia and North America
An interesting result is that the horizontal components at many global navigation satellite system (GNSS) stations – even those on the other side of the former ice margin – have enough sensitivity in almost all the layers in the upper mantle below the Norwegian Atlantic coast
This paper includes realistic structures of lateral viscosity variation under Fennoscandia to investigate the sensitivity of GNSS measurements in 10 selected stations

Summary

Introduction

It is well known that observations of the glacial isostatic adjustment (GIA) process allow us to determine the earth’s viscosity structure, especially that beneath formerly glaciated areas such as Fennoscandia and North America. Sensitivity kernels of GIA observations to radial changes in viscosity have been calculated in Mitrovica and Peltier (1991, 1993, 1995), Peltier and Jiang (1996a, b), and Peltier (1998) These studies showed that sensitivity is generally higher in the upper mantle than in the lower mantle. This is especially true for data in Fennoscandia, where the resolving power of GIA observations is too low to provide accurate inference of lower-mantle viscosity (Mitrovica and Peltier, 1993; Steffen and Wu, 2011). The sensitivity kernels for selected stations of the BIFROST (Baseline Inferences for Fennoscandian Rebound Observations, Sea Level, and Tectonics) project to radial viscosity variations have been studied by Milne et al (2004) They found sufficiently high sensitivities for the lower mantle.

Modeling

Results

Conclusions