Transcript Comparison of Post-Glacial Rebound Model Predictions

Towards a standard model for present-day
signals due to postglacial rebound
H.-P. Plag, C. Kreemer
Nevada Bureau of Mines and Geology and Seismological Laboratory
University of Nevada, Reno, Nevada, USA
David Lavallée
University of Newcastle upon Tyne, U.K.
Towards a standard model for present-day
signals due to postglacial rebound
Background
Comparison of PGR model predictions
The secular surface velocity fields
Separating rigid body motion from PGR
Is there consistency between observation and model predictions?
Background
Motivation:
There is a variety of PGR models with large differences
The uncertainties in the prediction of the present-day signal are poorly
known
Many applications in geosciences need to correct for PGR
IERS Conventions are not explicit in how to handle PGR
The SBL Project:
Goal is to set up, if possible, a standard model for the present-day
PGR signal with solid error bars
In 2005, Call for Submission of predictions of the present-day PGR
signal in sea level, 3-D surface displacements, gravity field, and Earth
Rotation
Establishment of a web page with the submissions and the results of
model inter-comparison (partly finished)
Model intercomparison is under way
Comparison of model to observations just started
Comparison of Post-Glacial Rebound Model Predictions
Comparison of Post-Glacial Rebound Model Predictions
Comparison of Post-Glacial Rebound Model Predictions
3-D displacements
VM2
REF
ALT
JXM
Comparison of Post-Glacial Rebound Model Predictions
3-D displacements
VM4
REF
ALT
JXM
Comparison of Post-Glacial Rebound Model Predictions
Standard deviation with
respect to global mean
Comparison of Post-Glacial Rebound Model Predictions
Cross correlations
Comparison of Post-Glacial Rebound Model Predictions
Normalized Scalar Product of 3-D displacements for VM4 and the other models
VM2
REF
ALT
JXM
Comparison of Post-Glacial Rebound Model Predictions
Normalized Scalar Product of 3-D displacements for VM4 and the other models
VM2
REF
ALT
JXM
Plan for Comparison and Validation
Intercomparison of all quantities
(3-D, LSL, geoid, Earth rotation, free air anomaly
Comparison to observations:
- 3-D to GPS, ...
- LSL to tide gauges
- Geoid to GRACE
- Earth rotation to IERS
Initial Step for Validation
Separation of PGR and rigid plate motion:
Plag et al. (2002): include PGR in the determination of rigid plate
motion
Kierulf and Plag (2003): significant improvement for Eurasia
Kreemer et al. (2006): ...
Initial Step for Validation
Total of 376 points
Combination of weekly
global and regional
solutions
1999 - 2005
Initial Step for Validation
For comparison, observed velocities need to be in same frame as predictions
For each PGR model, we calculating a scale and translation rate from a least
square fit of the 220 vertical velocities for sites on 15 tectonic plates.
All models suggest a translation of the GPS velocities of ~1.2-2.1 mm/yr
towards western Europe, and a scale change of a factor between 1 and 2.
Initial Step for Validation
What is Next?
5 X 5 degrees
Total of 222 grids elements
78 elements with multiple values
Conclusions
Regional intermodel differences larger than the uncertainties in
the observed velocity field, particularly for North America and
Eurasia.
Space-geodetic observations provide valuable constraints for
these models.
ICE-5G history inconsistent with the observed velocity field in
North America.
Accounting for the PRG signal in the determination of the rigid
body rotation improves the estimates for N.A. and Eurasia
For plates in the far-field of the former ice loads, the
improvement is either small or negligible.
There, PGR signal may be below the error of the observed
velocity field or erroneous for several reasons (including the
effect of lateral heterogeneities in the solid Earth).