Sea ice deforms under convergent and shear motion, causing rafting and ridging. This results in thicker ice than could be formed by thermodynamic growth only. About two-thirds of the Arctic sea ice volume consists of deformed sea ice. Three different approaches to simulating the formation of pressure ridges in a dynamic-thermodynamic continuum model are considered. They are compared with and evaluated by airborne laser profiles of the sea ice surface roughness. The different ridging schemes are (1) an additional prognostic equation for the deformation energy from which ridge parameters are derived, (2) a redistribution model with two ice categories, level and ridged ice, combined with a Monte Carlo simulation of ridging and (3) additional prognostic equations for ridge density and height resulting in the formation of ridged ice volume. The model results show that the ridge density is typically related to the state of ice motion whereas the mean sail height is related to the thickness of the ice involved in the ridging. In general, all of the three models reproduce realistic spatial distributions of ridges. Finally, the second ridging scheme is regarded to be best appropriate for climate modelling while the third scheme has advantages in sea ice forecasting.