An attempt to quantify the temporal variability in the volume composition of Arctic sea ice is presented. Categories of sea ice in the Transpolar Drift in Fram Strait are derived from monthly ice thickness distributions obtained by moored sonars . The inflection points on each side of the old ice modal peak are used to separate modal ice from ice which is thinner and thicker than ice in the modal range. The volume composition is then quantified through the relative amount of ice belonging to each of the three categories thin, modal, and thick ice in the monthly ice thickness distributions. The trend of thin ice was estimated to be negative at 28.8% per decade (relative to the long-term mean), which was compensated for by increasing trends in modal and thick ice of 7.9% and 4.7% per decade, respectively. A 7-8 year cycle is apparent in the thin and thick ice records, which may explain a loss of deformed ice since 2007. We also quantify how the categories contribute to the mean ice thickness over time. Thick (predominantly deformed) ice dominates the mean ice thickness, constituting on average 66% of the total mean. Following the loss of deformed ice since 2007, the contribution of thick ice to the mean decreased from 75% to 52% at the end of the record. Thin deformed ice did not contribute to this reduction; it was pressure ridges thicker than 5 m that were lost and hence caused the decrease in mean ice thickness.

Ice ridge keel geometry was studied by analyzing one year of upward looking sonar data collected in the Transpolar drift stream at 79°N, 6.5°W in 2008/2009. Ridges were identified using the Rayleigh criterion with a threshold value of 2.5 m and a minimum draft of 5 m. The keel shape was studied after the identification of ridges from temporal data. On average ridge keels were symmetric both with respect to the centroid of the keel and the keel crest location. By quantifying the ratio between observed keel area and the keel area of an assumed triangular keel shape (often assumed for first year ridges) we observed that in 79% of the cases the ridge cross sectional area would be underestimated by a triangular keel shape. Because keel loads on ships and structures increase with keel draft and keel area it is important that an assumed keel shape maintains the observed keel area. Thus we suggest that a better generalization of the shape of first year ridges is a trapezoidal keel shape rather than triangular. Based on the observations the mean trapezoidal keel, representing both first year ridges and old ridges, has a keel bottom-width which on average is 17% of the keel width. For the deepest keels (N 15 m) the mean keel bottom width was 12% of the keel width. The mean keel draft was 7.3 m and the deepest ridge was 25 m. The temporal data was converted to spatial data based on an ice drift speed estimate which assumed free drift. From the spatial data we found that the mean keel width was 28 m and the mean keel cross sectional area was 164 m 2 .

Upward looking sonar (ULS) instruments have been used for several decades to provide continuous measurements of ice draft. The time resolution of the ice draft observations is typically 1-2 seconds. When fused with ice drift speed observations, a high horizontal spatial resolution can be realized. Such a high resolution allows for the identification of individual ice keel features and an analysis of their spatial characteristics. Many methods are available for transforming the ice draft series from an equispaced time domain to an equidistant spatial domain. This paper analyzed the sensitivity of ice keel statistics to three transformation methods applied to ULS sea ice measurements in the Beaufort Sea and North Chukchi Sea. Although differences were found between the methods, these were related to episodes when the sampling frequency is not high enough to profile an ice draft feature travelling with a high drift speed. Knowledge of maximum drift speeds in the region of a measurement location along with the enhanced power and storage capacities of modern ice profilers enable sampling configurations which avoid this scenario.