Sea Ice Rheology From In-situ And Remote Sensing Observations: Fracture and Friction
Abstract
Sea ice mechanics and dynamics control, along with thermodynamics, sea ice thickness distribution and mass balance, which themselves play an important role in regional and global climate via, e.g., the ice-albedo feedback. Therefore, a correct understanding of sea ice rheology and deformation is an important task. Current sea ice models are based on a continuum mechanics framework and describe the ice cover in terms of elastic-plastic (EP) or viscous-plastic (VP) mechanics and, accordingly, invoked the concept of a rate- and scale-independent failure envelope. Within that scheme, stress states inside the envelope lead to either elastic (EP) or viscous/creep (VP) behavior, whilst stress states that reach the envelope lead to terminal failure through plastic flow. The plasticity so invoked is not dislocation-based, von-Mises plasticity in which volume is conserved and the "strain vector" is normal to the envelope, but a more soil-like plasticity in which volume is not conserved and the hydrostatic component of the stress tensor plays a major role in setting the terminal failure stress. In the VP model, viscous behaviour is assumed to represent the average effect of various small-scale "plastic" events. An implication of the VP model is that for loading paths that end inside the envelope, the strain rate is proportional to the stress. The analysis presented here is based on (i) in-situ stress measurements (provided by CRREL) recorded during the SHEBA experiment, and (ii) RGPS estimates of strain-rate over a zone covering SHEBA. We show that no correlation exists between the average strain-rate over the zone and the stress averaged over all the field sensors installed inside this zone. On the other hand, the in-situ stress states, when plotted in the principal stresses space, lie inside an envelope with Coulombic branches, whereas highly biaxial compressive stress states are absent. However, this failure envelope is as a scale-dependent concept. The in-situ envelope appears as a contracted version of the laboratory envelope, at least for the Coulombic branches, contracted by a degree that reflects larger stress concentrators and reduced cohesive strength. Similarly, the in-situ envelope expands as one records stresses over a longer period, as we increases the probability to record the activation of a small concentrator in the vicinity of the sensor. Therefore, instead of flowing in a viscous manner, even a moderately loaded ice cover appears to deform in a brittle manner. We propose a new framework in which fracture and frictional sliding govern inelastic deformation even under small stresses and over all spatial and time scales, and characterized by scaling properties that preclude homogenization or averaging procedures to be used to properly model sea ice deformation. This difficulty to link small to large scales reveals the need for new in-situ as well as remote sensing measurements of mechanical parameters such as stress, strain-rate or seismic waves resulting from fracturing events.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2006
- Bibcode:
- 2006AGUFM.C33B1273W
- Keywords:
-
- 0750 Sea ice (4540);
- 0754 Leads (4540);
- 0794 Instruments and techniques;
- 3902 Creep and deformation;
- 5104 Fracture and flow