The physical, radiative and microwave scattering characteristics of melt ponds on Arctic landfast sea ice
Melt ponds are an important characteristic of Arctic sea ice because of their control on the surface radiation balance. Little is known about the physical nature of these features and to date there is no operational method for detection of their formation or estimation of their aerial fraction. Coincident in situ observations, aerial surveys and synthetic aperture radar data from a field site in Arctic Canada are compared in an evaluation of the physical, radiative and electrical properties of melt ponds on first-year and multiyear sea ice. Results show that the interrelationships between the thermal diffusivity and conductivity of the snow cover control the mechanisms of snow ablation. Aerial fractions of snow patches, and light and dark coloured melt ponds, show considerable variation both as a function of proximity to land and due to ice type. First-year sea ice is shown to have a water background with discrete snow patches distributed throughout. Multiyear sea ice consists of discrete 'particles' within a snow background. Morphological measurements indicate that snow patches range in size with average areas of from 5 to 20m2 . Pond sizes over multiyear sea ice are also highly variable with averages ranging from 15 to 20m2. The integrated shortwave albedo was measured in the field and averaged to: snow patches (0.64 0.07); light melt ponds (0.29 0.04); and dark melt ponds (0.14 0.03). Snow patch size statistics explained a statistically significant proportion of the surface shortwave albedo. We found that microwave scattering could be used to obtain a measure of the onset ofmelt and had utility in detecting subtle details ofthe thermodynamic transition from winter through early melt into pond formation. We formalized a statistical relationship between microwave scattering and surface climatological albedo (sigma-alpha relationship). We found the relationship valid only for landfast firstyear sea ice under windy conditions. We conclude with a discussion of the role of surface wind stress and diurnal cycling in specification of the sigma-alpha relationship.