Sea-ice roughness and drag coefficients in a dynamic–thermodynamic sea-ice model for the Arctic

A quantitative relationship between observed sea-ice roughness and simulated large-scale deformation work is established in order to provide new means for model validation and a better representation of the sea-ice component in climate modelling. Sea-ice roughness is introduced as an additional prognostic variable in a dynamic–thermodynamic sea-ice model with a viscous-plastic rheology. It is defined as the accumulated work of internal forces acting upon an ice volume, given in energy per area. A fraction of this total deformation work is transferred to the potential energy stored in pressure ridges. Using ridge geometries and distribution functions from observations, observable quantities like mean pressure ridge height, ridge frequency as well as volumetric and areal fractions of deformed ice are derived from the simulated ice roughness. Comparisons of these simulated quantities with measurements (submarine-borne sonars, laser altimeters on helicopters) show good agreement. Satellite-borne observations of sea-ice roughness now under development will provide an even larger data set which will be used for model verification. Additionally roughness-dependent drag coefficients are introduced to account for the effect on the momentum exchange between ocean and atmosphere due to the form drag of roughness elements. The simulations indicate that the inclusion of sea-ice roughness provides for a more realistic representation of the boundary layer processes in climate models.