|
Titel |
Identifying viscosities implicit in current firn-densification models: a step toward a physical-process-based constitutive relation for firn |
VerfasserIn |
Brita I. Horlings, Edwin D. Waddington, C. Max Stevens |
Konferenz |
EGU General Assembly 2017
|
Medientyp |
Artikel
|
Sprache |
en
|
Digitales Dokument |
PDF |
Erschienen |
In: GRA - Volume 19 (2017) |
Datensatznummer |
250142467
|
Publikation (Nr.) |
EGU/EGU2017-6094.pdf |
|
|
|
Zusammenfassung |
Firn densification rate determines the age of air in bubbles in ice cores, and is a primary
uncertainty in conversion of volume changes to mass changes for altimetry surveys of the
polar ice sheets. However, most current firn models are empirically tuned to local temperature
and accumulation rate because not all physical processes that affect firn densification at the
grain scale are adequately understood or measured. With generally fewer tunable parameters
than active physical processes, impacts of such missing processes have instead typically been
lumped together. As the next step toward developing a microphysical-process-based
constitutive relation for firn, effective viscosity is used as the constitutive parameter in
eight firn-densification models. Effective viscosity is implicitly identified in each
model, and is dependent on factors such as stress, strain rate, temperature, and
grain-growth processes. The models are forced with temperature and accumulation rate
to find the effective viscosities. Preliminary results indicate that there are major
qualitative differences in some of the models’ viscosity definitions, which indicate
discrepancies in implicit descriptions of physical processes. Such differences include
discontinuous viscosity at the zone 1-2 boundary (at density of 550 kg m−3), and
discrepant viscosity trends that potentially indicate lumping of multiple processes.
Each firn model describes densification uniquely, and these results confirm that
processes are either oversimplified in different ways or are missing in the definitions.
Future work will involve developing a viscosity function based explicitly on multiple
processes which will help to prioritize future field measurements and lab studies. |
|
|
|
|
|