Modeling sea ice fracture at very high resolution with VP rheologies
Recent high resolution pan-Arctic sea ice simulations show fracture patterns that are typical of granular materials but with intersection (fracture) angles wider than those observed from high-resolution satellite images (with a modal value of θ = 20º). In this work, we investigate the mechanism of formation and parameter dependencies of ice fracture in simple numerical uni-axial test on a 8 km x 25 km ice floe at an unprecedented resolution of 25 m for two different Visco-Plastic (VP) yield curves: an elliptical (standard in sea ice models) and a coulombic yield curve both with normal flow rule. In the standard VP model, the simulated angle of fracture is θ=33.9º. The dependence of the angle of fracture on the ice shear strength is also contrary to that of typical granular materials with larger angle of fracture for higher shear strength. In this model, the divergence along the fracture lines (or LKFs) is entirely dictated by the ice shear strength with high shear strength resulting in convergence along LKFs and low shear strength resulting in divergence along LKFs. This is again contrary to typical granular materials. Moreover, the angle of fracture depends on the confining pressure in the uni-axial test with more convergence as the confining pressure increases, again contrary to granular material. In the Coulombic model, the angle of fracture is smaller (θ=23.5º), but the solution is unstable because of the discontinuity between the straight limbs of the yield curve and the elliptical capping. Our results show that while the VP model gives angles of fracture that are visually correct, the bias in the magnitude of the angle of fracture and the physical dependencies of the angle of fracture on mechanical strength parameters and stress fields couple the sea ice mechanical strength parameters, the sea-ice drift, sea-ice deformation (strain-rate) field in an inconsistent way. We consider this evidence to move away from the elliptical yield curve and associative (normal) flow rule.
Helmholtz Research Programs > CHANGING EARTH (2021-2027) > PT2:Ocean and Cryosphere in Climate > ST2.2: Variability and Extremes