An investigation of Glacial Isostatic Adjustment over the Amundsen Sea sector, West Antarctica
Highlights
► The study focuses on the region of the Amundsen Sea sector (West Antarctica). ► We assessed the GIA effect from a combination of ICESat and GRACE data. ► Inferred GIA effect is significantly larger than current GIA model predictions. ► The GIA results were validated using GPS-derived vertical crustal deformation.
Introduction
The determination of the mass balance of the ice sheets and of their contribution to global sea-level changes is one of the most challenging tasks of Earth system research. To reach more reliable estimates of those changes, one can utilise present-day gravity and altimetry satellite missions which provide much more suitable data than previously. The Gravity Recovery and Climate Experiment (GRACE) (Tapley et al., 2004) observes temporal gravity changes from which surface mass changes over a given region can be deduced (gravity-change method). However, GRACE is sensitive to the integral signal of all mass changes. Therefore, it is not possible to distinguish between mass changes caused by present-day ice-mass variations and other superimposed mass signals like the Glacial Isostatic Adjustment (GIA). In this paper, GIA denotes the description of the visco-elastic response of the Earth to changing ice loads through glacial history. The solid Earth reaction to present ice-mass changes should be described by an immediate elastic response and not be included in the GIA model. For the entire Antarctic Ice Sheet (AIS) the expected mass change effect due to GIA is of similar order of magnitude as the ongoing present-day ice-mass change (e.g. Horwath and Dietrich, 2009). Therefore, a sound knowledge of the GIA signal is of crucial importance for a precise ice-mass change estimation.
The Geoscience Laser Altimeter System (GLAS) on board the Ice, Cloud and Land Elevation Satellite (ICESat) (Zwally et al., 2002) measures temporal changes of the ice surface elevation (altimetry method). This method requires additional information on the density of the ice-firn layers if the observed height or volume changes are to be converted into ice-mass changes. For the altimetry method, in the critical regions the GIA correction is about two to three orders smaller than the measured elevation change. Therefore, it is much less sensitive to uncertainties in the GIA correction than the gravity-change method.
Available GIA models consist of a visco-elastic Earth model and a glacial history. For the AIS, the latter is poorly constrained, since indications on the changing sea level and on former ice margins are still sparse (e.g. Larter et al., 2007). Additionally, there exist substantial differences in the Earth's structure between East and West Antarctica (Morelli and Danesi, 2004) which are not yet reflected in the adopted rheological models. Therefore, GIA predictions still have large uncertainties, especially for West Antarctica. Several authors proposed the combination of GRACE and ICESat observations to infer present-day GIA predictions which do not depend on the glacial history of the AIS. The iterative approach suggested by Velicogna and Wahr (2002) deduces the spatial GIA pattern from the overall GRACE geoid trend by subtracting the geoid signal of the ICESat-derived ice-mass change. In the next iteration step, the ICESat observations are corrected for the previously inferred height changes due to GIA. Finally, the derived vertical crustal deformations induced by GIA are compared to those observed at continuous Global Positioning System (GPS) sites. This allows to relate possible discrepancies to an error in the applied ice-firn density assumption. Riva et al. (2009) presented a single step approach which combines GRACE, ICESat and density models of the ice-firn layers and of the Earth's mantle. This approach benefits from the density contrast between the ice-firn density and mantle density, thus allowing to distinguish between ice and solid Earth mass changes.
The present study focuses on the Amundsen Sea embayment, West Antarctica (cf. Fig. 1). There, the major part of the ice-mass loss of the entire AIS takes place, which causes a significant contribution to global sea-level rise (e.g. Rignot et al., 2008, Horwath and Dietrich, 2009). This ice-mass loss can be observed especially in the drainage basins of Pine Island Glacier, Thwaites Glacier and Smith Glacier (PITS). Altogether, about 25% of the West Antarctic Ice Sheet (WAIS) drains into the Amundsen Sea embayment (Larter et al., 2009). Utilising GRACE and ICESat observations for the concurrent six-year observation period, robust regional estimates of the mass and volume changes for the PITS area are derived. Combining these results with reasonable assumptions for the ice-firn density and mantle density we come to a model-independent average estimate of the GIA-induced vertical crustal deformation (“observed GIA”). Finally, we validate our findings comparing these deformation rates with those derived by GPS observations at three bedrock sites in the area under investigation.
Section snippets
Present-day mass- and volume-changes
For any specific region of the AIS the total observable (e.g. by means of satellite gravimetry) present-day mass change with respect to a reference state as well as its linear temporal change can be decomposed into the following components:
While denotes the GIA-induced mass change over the region of interest, the term in brackets summarises all mass changes related to present-day changes of the ice sheet's mass. In detail, this
GIA models
Usually, GIA model predictions are used to correct satellite observations for superimposed GIA signals. In this study we use GIA predictions by the regional model IJ05 (Ivins and James, 2005) and the global model ICE-5G (VM2) (Peltier, 2004) to compare them with our observed GIA corrections. The model predictions shown in Fig. 2a (kindly provided by E. Ivins) are based on the IJ05 ice-load history for Antarctica. The underlying compressible Earth model comprises an elastic lithosphere of 100 km
Results
The combined linear and seasonal model (Fig. 3a—black curve) which was fitted to the GRACE-derived mass change time series (Fig. 3a—green curve) depicts a dominating linear mass decrease and a seasonal signal with a peak-to-trough difference of 37.0 Gt. As described in Section 2.3, the GRACE-derived mass trend includes the effect of present-day ice-mass changes and mass change due to GIA and amounts to . Our overall error estimate for the
Discussion
The comparison of our derived GIA results with the model predictions listed in Table 1 reveals that our numbers are significantly larger. To validate our results we compare them to the observed uplift rates at the three GPS campaign sites corrected for the elastic uplift due to present-day ice-mass changes. The corrected observed vertical deformation rates (cf. Table 3) indicate a non-uniform, spatially varying GIA pattern for the PITS basin. Nevertheless, the mean uplift rate of 19.5 mm/yr
Acknowledgments
This research was supported by the German Research Foundation (DFG) and partly by the European Science Foundation COST Action ES0701 “Improved constraints on models of Glacial Isostatic Adjustment”. The field work was carried out during the R/V “Polarstern” cruises ANT-XXIII/4 and ANT-XXVI/3. We like to thank the chief scientist Karsten Gohl, the helicopter crew and all participants for their support. All figures were prepared using the GMT graphics package (Wessel and Smith, 1998). We thank
References (59)
- et al.
Plate kinematics and deformation status of the Antarctic Peninsula based on GPS
Global and Planetary Change
(2004) - et al.
Rapid crustal uplift in Patagonia due to enhanced ice loss
Earth and Planetary Science Letters
(2010) - et al.
A global hydrological model for deriving water availability indicators: model tuning and validation
Journal of Hydrology
(2003) - et al.
Preliminary reference Earth model
Physics of the Earth and Planetary Interiors
(1981) - et al.
Volume and mass changes of the Greenland ice sheet inferred from ICESat and GRACE
Journal of Geodynamics
(2012) - et al.
Rapid viscoelastic uplift in southeast Alaska caused by post-Little Ice Age glacial retreat
Earth and Planetary Science Letters
(2005) - et al.
Modeling of firn compaction for estimating ice-sheet mass change from observed ice-sheet elevation change
Annals of Glaciology
(2011) - et al.
Seismological imaging of the Antarctic continental lithosphere: a review
Global and Planetary Change
(2004) - et al.
Glacial Isostatic Adjustment over Antarctica from combined ICESat and GRACE satellite data
Earth and Planetary Science Letters
(2009) - et al.
SELEN: a Fortran 90 program for solving the ‘sea-level equation’
Computers and Geosciences
(2007)
ICESat's laser measurements of polar ice, atmosphere, ocean, and land
Journal of Geodynamics
Rise of the Ellsworth mountains and parts of the East Antarctic coast observed with GPS
Geophysical Research Letters
Isolating the PGR signal in the GRACE data: impact on mass balance estimates in Antarctica and Greenland
Geophysical Journal International
Geodetic measurements of vertical crustal velocity in West Antarctica and the implications for ice mass balance
Geochemistry Geophysics Geosystems
Low degree spherical harmonic influences on Gravity Recovery and Climate Experiment (GRACE) water storage estimates
Geophysical Research Letters
Effect of gravitational consistency and mass conservation on seasonal surface mass loading models
Geophysical Research Letters
Determination of vertical and horizontal deformations of the Earth's crust and of ice shelves in West Antarctica by GPS observations
Present-day vertical crustal deformations in West Greenland from repeated GPS observations
Geophysical Journal International
Precise analysis of ICESat altimetry data and assessment of the hydrostatic equilibrium for subglacial Lake Vostok, East Antarctica
Geophysical Journal International
Deformation of the Earth by surface loads
Reviews of Geophysics and Space Physics
AOD1B product description document
Technical Report. Potsdam
Release notes for GFZ Potsdam GRACE L-2 products — version RL04
Physical Geodesy
Errors of regional mass variations inferred from GRACE monthly solutions
Geophysical Research Letters
Signal and error in mass change inferences from GRACE: the case of Antarctica
Geophysical Journal International
Bedrock response to Llanquihue Holocene and present-day glaciation in southernmost South America
Geophysical Research Letters
Antarctic glacial isostatic adjustment: a new assessment
Antarctic Science
On-land ice loss and glacial isostatic adjustment at the Drake Passage: 2003–2009
Journal of Geophysical Research
Cited by (0)
- 1
Now at: Bundesamt für Kartographie und Geodäsie, Karl-Rothe-Straße 10–14, 04105 Leipzig, Germany.