ICESat over Arctic sea ice: Estimation of snow depth and ice thickness

Starting with retrieved freeboards from four ICESat campaigns (ON05, October/November 2005; FM06, February/March 2006; ON06, October/November 2006; and MA07, March/April 2007) we estimate their ice thicknesses using constructed fields of daily snow depth and compare them with ice drafts from moored upward-looking sonars. The methodologies, considerations, and assumptions used in the conversion of freeboard to ice thickness are discussed. The thickness distributions of the Arctic multiyear and seasonal ice covers are contrasted. Broadly, the resulting fields seem seasonally and interannually consistent in terms of thickness, growth and ice production. We find mean thicknesses of 2.15/2.46m in ON05/FM06 and an overall thinner ice cover of 1.96/2.37min ON06/MA07. This represents a growth of similar to 0.3 m and similar to 0.4 m during the similar to 4-month intervals of the ON05-FM06 and ON06-MA07 campaigns, respectively. After accounting for data gaps, we compute overall Arctic Ocean ice volumes of 11,318, 14,075, 10,626, and 13,891 km 3 for the ON05, FM06, ON06, and MA07 campaigns, respectively. The higher total volume in ON05 (versus ON06) can be attributed to the higher multiyear ice coverage that fall: 37% versus 31%. However, the higher estimated ice production (less export) during the second year (3265 versus 2757 km3) is likely due to the higher growth rate over the larger expanse of seasonal sea ice during the fall and winter. Inside a 25-km radius of two mooring locations in the Beaufort Sea, the estimated mean ICESat ice drafts from ON05 and FM06 are within 0.5 m of those measured at the moorings.