 |
| Titel |
Reconstructions of the Weichselian ice sheet, a comparative study of a thermo-mechanical approach to GIA driven models. |
| VerfasserIn |
Peter Schmidt, Björn Lund, Jens-Ove Näslund, James Fastook |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250097221
|
| Publikation (Nr.) |
 EGU/EGU2014-12776.pdf |
|
|
|
|
|
| Zusammenfassung |
| Observations of glacial isostatic adjustment (GIA) have been used both to study the
mechanical properties of the Earth and to invert for Northern Hemisphere palaeo-ice-sheets.
This is typically done by solving the sea-level equation using simplified scaling laws to
control ice-sheet thickness. However, past ice-sheets can also be reconstructed based on
thermo-mechanical modelling driven by palaeo-climate data, invoking simple analytical
models to account for the Earth’s response. Commonly, both approaches use dated geological
markers to constrain the ice-sheet margin location. Irrespective of the approach,
the resulting ice-sheet reconstruction depends on the earth response, although the
interdependence between the ice model and the earth model differs and therefore the
two types of reconstructions could provide complementary information on Earth
properties.
We compare a thermo-mechanical reconstruction of the Weichselian ice-sheet using the
UMISM model (Näslund, 2010) to two GIA driven reconstructions, ANU (Lambeck et al.,
2010) and ICE-5G (Peltier & Fairbanks, 2006), commonly used in GIA modelling. We
evaluate the three reconstructions both in terms of ice-sheet configurations and predicted
Fennoscandian surface deformation
ICE-5G comprise the largest reconstructed ice-sheet whereas ANU and UMISM are more
similar in volume and areal extent. Significant differences still exists between ANU and
UMISM, especially during the final deglaciation phase. Prior to the final retreat of the
ice-sheet, ICE-5G is displays a massive and more or less constant ice-sheet configuration,
while both ANU and UMISM fluctuates with at times almost ice-free conditions, such as
during MIS3. This results in ICE-5G being close to isostatic equilibrium at LGM,
whereas ANU and UMISM are not. Hence, the pre-LGM evolution of the Weichselian
ice-sheet needs to be considered in GIA studies. For example, perturbing the ANU or
UMISM reconstructions we find that changes more recent than 36 kyr BP may
change the predicted uplift velocities by more than 0.1 mm/yr, while changes more
recent than 55 kyr BP may change the predicted uplift 10 kyr ago by more than 5
m.
Despite their differences we find that all three reconstructions can equally well fit
observations of the present day uplift in Fennoscandia, as well as the observed sea-level curve
along the Ångerman river, Sweden, albeit with different optimal earth models. However,
only for ANU can a single optimal earth model be determined as a bifurcation in
the optimal viscosity arises from the generally faster present day rebound rates in
ICE-5G and UMISM, resulting in a range of well-fitting earth models for the latter
reconstructions.
Studying models with a reasonable fit to observed present day uplift velocities we find
general trends of over- and under-prediction, indicating that all three ice-sheet reconstructions
need improvement. In general, all three reconstructions tend to over-predict the uplift rates in
southwestern Fennoscandia, whereas over Finland ICE-5G generally over-predicts and ANU
generally under-predicts the uplift rates. UMISM tend to under-predict the velocities over
central to northern Sweden and similar trends can also be seen in ANU and ICE-5G. |
|
|
|
|
|
|