If lake ice is assumed to deform and fail as a linear viscoelastic
material under the action of wind stress, then a simple ice
concentration scaling law can be constructed suitable for
one-dimensional lake models embedded within environmental prediction
systems. Most 1-D lake models assume no ice mechanics at all, while
others adapt the viscous-plastic rheology common in ice-ocean models for
the purpose of estimating ice fraction. Elastic buckling is generally
disregarded as a significant failure mechanism in ice under low stress
conditions at geophysical scales. However, by adding viscosity to the
constitutive equation, the conditions for viscoelastic buckling seem
quite plausible over a wide range of lake size and ice thickness. An ice
concentration scaling law based on this process is evaluated here in
multiannual simulations over North America and found to produce superior
ice phenology statistics compared with simulations based on plastic
failure or no ice mechanics.