GRACE Science Team Meeting

Session B.2a: Understanding GIA

Erik Ivins

(Ice Loss Rates and GIA Constrains Using Simultaneous GPS and GRACE data: Patagonia and Graham Land, Antarctic Peninsula)
Erik Ivins

(Global Simultaneous Estimation of Present-Day Surface Mass Trend and GIA Using GRACE and Surface Geodetic Data Combination)

(A Model for Removing the Post-Glacial-Rebound Signal from GRACE Data)
John Wahr

(Glacial Isostatic Adjustment and GRACE - Status and Future)
Holger Steffen

(Understanding Glacial Isostatic Adjustment: A Joint DynaQlim/GGOS Workshop)
Richard Gross


(Contribution of GRACE gravity rates to a model for GIA across Fennoscandia)
Emma Hill

Session: B.2.a - Theme: Understanding GIA
Title: Ice Loss Rates and GIA Constrains Using Simultaneous GPS and GRACE data: Patagonia and Graham Land, Antarctic Peninsula
First Author: Erik Ivins
Presenter: Erik Ivins
Co-Authors: M.M. Watkins; D-N. Yuan; R. Dietrich; G. Casassa; A. Rülke

Abstract: Understanding glacial isostatic adjustment (GIA) in Antarctica has been an elusive problem due to the lack of observations sufficient to constrain upper mantle viscosity. Here, however, we show that simultaneous use of GPS and GRACE data can provide highly useful bounds on both ice mass trends over 6.5 years and the elusive GIA component. For the Patagonian Ice Fields (PIF) and the northern Antarctic Peninsula (NAP) the level 2 products distributed by the various analysis centers are useful, but they are not designed to optimize resolution for these regions. In PIF and NAP we desire resolution that approaches the 50 km scale. Although such resolution is not achievable using mascon techniques, they prove to have far tighter error bounds when Gaussian smoothed with a 400 km filter radius. The superior resolution and accuracy then pave the way for the construction of an iterative procedure using GPS and the global mascons of Watkins and Yuan (2007, EoS). GRACE plus GPS trends can then be employed to retrieve ice loss rate and GIA water height equivalent change. The GPS data for the PIF region is specifically designed for this simultaneous retrieval (Dietrich et al. 2009, EPSL) and GRACE-mascons retrieve -26 ± 6 Gt/yr for PIF and -41.5 ± 10 Gt/yr for all the NAP north of Palmer Land and Alexander Is. (70 deg. S), respectively, during 2003-2009. These loss values are consistent with radar derived out-flux estimates and with ICESat height change measurements.

Back to top

Session: A.1 - Analysis Techniques
Title: Global Simultaneous Estimation of Present-Day Surface Mass Trend and GIA Using GRACE and Surface Geodetic Data Combination
First Author: Xiaoping Wu
Presenter: Xiaoping Wu
Co-Authors: M. Heflin, H. Schotman, L. L. A.Vermeersen, D. Dong, R. Gross, E. Ivins, A. Moore, S. Owen

Abstract: Separating geodetic signatures of present-day surface mass trend and Glacial Isostatic Adjustment (GIA) requires multi-data types of different physical characteristics. We take a kinematic approach to the global simultaneous estimation problem. Three sets of global spherical harmonic coefficients from degree 1 to 60 of the present-day surface mass trend, vertical and horizontal GIA induced surface velocity fields, as well as rotation vectors of 15 major tectonic plates are solved for. To a good approximation, the GIA induced geoid coefficients are related to the vertical coefficients except for the rotational feedback effect. The estimation is carried out using GRACE geoid trend, 3-dimensional geodetic velocities, and the data-assimilated JPL ECCO ocean bottom pressure model. The ICE-5G/IJ05 (VM2) predictions are used as a priori GIA mean model. An a priori covariance matrix is constructed in the spherical harmonic domain for the GIA model by perturbing and propagating the covariance matrices of plausible errors in the dynamic parameters. Unprecedented high-precision results are achieved. For example, geocenter velocities due to present-day surface mass trend and due to GIA are both determined to uncertainties of better than 0.1 mm/yr without using direct geodetic geocenter information. The rotational feedback effect on GIA is addressed observationally. Future improvements, and benefits from new data will also be explored in the global inverse framework.

Back to top

Session: B.2.a - Theme: Understanding GIA
Title: A Model for Removing the Post-Glacial-Rebound Signal from GRACE Data
First Author: John Wahr
Presenter: John Wahr
Co-Authors: Shijie Zhong, Archie Paulson

Abstract: We describe Paulson et al’s (2007) global model of post-glacial rebound. The model’s predictions of the secular change in the geoid are publicly available on the GRACE Tellus website. The model uses the ICE-5G ice deglaciation history and VM2 viscosity profile of Peltier (2004). There is good agreement between the model’s predicted secular change and the secular change fit to the existing GRACE data over Canada and Scandinavia, the two major regions of Pleistocene glaciation that are not still covered with ice. We include a discussion of the effects of polar wander feedback on the C21 and S21 geoid coefficients.

Back to top

Session: B.2.a - Theme: Understanding GIA
Title: Glacial Isostatic Adjustment and GRACE - Status and Future
First Author: Holger Steffen
Presenter: Holger Steffen
Co-Authors: W. van der Wal, R. Riva, P. Wu, B. Vermeersen, H. S. Wang

Abstract: Analysis of data of the GRACE satellite mission has clearly identified the long-term mass changes in regions of Glacial Isostatic Adjustment (GIA) such as North America, Fennoscandia, Greenland and Antarctica. After more than 7 years, the determined trends are quite robust. To isolate the GIA signal from the integral effect of processes in the atmosphere, cryosphere, hydrosphere and geosphere, e.g. hydrological contributions in North America and Fennoscandia and ice mass changes in Greenland and Antarctica have to be removed, which results in high demands on the analysis method, the filtering technique and the reduction as well as combination with other models and data.

In both Fennoscandia and North America, the GRACE solution fits well to results from terrestrial measurements and from modelings. In Fennoscandia, the uplift center is located in the Gulf of Bothnia, and the uplift area comprises the Scandinavian Peninsula and Finland. A still unknown contribution arises from hydrology, which cannot be sufficiently separated. This also applies to GIA over Northern America, where two prominent domes west and south-east of the Hudson Bay are determined. Recent global hydrology models that have been used for reduction of the hydrological effect, however, have to be revised, as they already show non-negligible differences among each other. Over Greenland and Antarctica, both regions of strong interaction between GIA and ice mass change, GRACE is of immense importance to determine the mass balance of the ice sheet. While GIA is on short (years to decades) timescales a linear process, the ice sheet mass balance is characterized by periodic signals and possibly experience significant deviations from the secular trend.

In future, a prolongation of the gravity time series is definitely necessary as it will improve the separation of GIA, ice balance and hydrology effects. A clear GIA signal will then give insight in Earth's 3D structure, tectonic processes and topographic changes.

Back to top

Session: B.2.a - Theme: Understanding GIA
Title: Understanding Glacial Isostatic Adjustment: A Joint DynaQlim/GGOS Workshop
First Author: Richard Gross
Presenter: Richard Gross
Co-Authors: M. Poutanen

Abstract: The phenomenon of glacial isostatic adjustment (GIA) with its unique temporal and spatial signatures provides one of the great opportunities in the geosciences for obtaining information about dynamic Earth processes. GIA contains information about recent climate forcing, being dependent on the geologically recent loading and unloading of ice sheets; it presents a unique opportunity to study the dynamics and rheology of the lithosphere and mantle; is responsible for substantial sea-level variations; and is of fundamental importance to geodesy because site positions, reference frames, Earth rotation and gravity are all influenced by it.

The Global Geodetic Observing System (GGOS) of the International Association of Geodesy and the International Lithosphere ProgramÕs Regional Co-ordination Committee DynaQlim (Upper Mantle Dynamics and Quaternary Climate in Cratonic Areas) jointly organized a workshop on "Understanding Glacial Isostatic Adjustment" that attracted 36 international scientists to Espoo, Finland during June 23-26, 2009. The objectives of this workshop were to: (1) review the current state of the science in modeling GIA, (2) review the use of geodetic measurements to both constrain and to test GIA models, (3) identify obstacles to improving GIA models, and (4) identify improvements to the global geodetic observing system that are required to better understand GIA.

Isolating the GIA signal in geodetic observations is an important prerequisite to advancing our understanding of GIA. Workshop participants identified both improved GIA models and gravity observations from GRACE (the Gravity Recovery and Climate Experiment) as being of key importance to accomplishing this. The spatial patterns of GIA and present-day ice mass change can be used to help separate these from other signals in GRACE observations. Continuing without interruption time variable gravity observations from space by GRACE-like missions was strongly recommended by workshop participants.

Back to top

Session: B.2.a - Theme: Understanding GIA
Title: Contribution of GRACE gravity rates to a model for GIA across Fennoscandia
First Author: Emma Hill
Presenter: Emma Hill
Co-Authors: J. L. Davis; M. E. Tamisiea

Abstract: We have developed a technique to use geodetic data to update an a priori model for Glacial Isostatic Adjustment (GIA). We use this technique to update a model for Fennoscandian GIA using GPS, tide-gauge, and GRACE gravity rates. The technique allows us to investigate the individual contributions from these data sets in a self-consistent manner. In this presentation we focus on the contribution from GRACE to our updated model, and compare models produced with assimilation of the CSR and GFZ GRACE data sets. As part of our assimilation routine we simultaneously estimate a rate of local eustatic sea-level rise. The estimated rate (~1.8 mm/yr) is consistent for all solutions, regardless of which data sets are assimilated. However, this is only the case if a uniform regional gravity rate is also estimated for the assimilated GRACE data. Without GRACE data assimilated this eustatic sea-level rate is estimated to be 1.6 ± 0.3 mm/yr. With GRACE data assimilated, but without estimation of a uniform regional gravity rate, this estimate increases to 3.3 ± 0.1 mm/yr (CSR) or 3.7 ± 0.1 mm/yr (GFZ). Including a uniform gravity rate (estimated at 0.18 ± 0.02 microGal/yr for CSR and 0.25 ± 0.02 microGal/yr for GFZ) eliminates much of the bias between solutions with and without the GRACE data assimilated, resulting in more consistent estimates for regional sea level of 1.8 ± 0.2 mm/yr (CSR) and 1.5 ± 0.2 mm/yr (GFZ). As well as being more consistent with the eustatic sea-level rate estimated without GRACE data, these estimates are more consistent with each other. There are several possible explanations for the requirement of this uniform gravity rate. One possibility is that simplifications in modeling the Laurentia ice sheet over North America (the effects of which are included in our a priori GIA models and covariance) are causing a bias across Fennoscandia that appears uniform due to the small size of the region. Alternatively, this could be due to aliasing errors associated with mismodeling of the ocean tides, or mismodeling of the hydrology signal over Fennoscandia. We also show non-uniform differences between the GIA model produced with assimilation of CSR and GFZ data, which have an approximately north-south gradient across the region.

Back to top




UT Sponsored Travel

Sessions & Abstracts

Meeting Program

Meeting Proceedings

Past Meetings