C21C-01 INVITED
Arctic Sea Ice Thickness - Past, Present And Future
In November 2005 the International Workshop on Arctic Sea Ice Thickness: Past, Present and Future was held at Rungstedgaard Conference Center, near Copenhagen, Denmark. The proceedings of the Workshop were subsequently published as a book by the European Commission. In this review we summarise the conclusions of the Workshop on the techniques which show the greatest promise for thickness monitoring on different spatial and temporal scales, and for different purposes. Sonic methods, EM techniques, buoys and satellite methods will be considered. Some copies of the book will be available at the lecture, and others can be ordered from the European Commission. The paper goes on to consider early results from some of the latest measurements on Arctic sea ice thickness done in 2007. These comprise a trans-Arctic voyage by a UK submarine, HMS "Tireless", equipped with a Kongsberg 3002 multibeam sonar which generates a 3-D digital terrain map of the ice underside; and experiments at the APLIS ice station in the Beaufort Sea carried out by the Gavia AUV equipped with a GeoSwath interferometric sonar. In both cases 3-D mapping of sea ice constitutes a new step forward in sea ice data collection, but in the case of the submarine the purpose is to map change in ice thickness (comparing results with a 2004 "Tireless" cruise and with US and UK data prior to 2000), while for the small AUV the purpose is intensive local mapping of a few ridges to improve our knowledge of their structure, as part of a multisensor programme
C21C-02 INVITED
In Situ Thickness Observations of Sea Ice and Snow in the Fram Strait
The Fram Strait, at the end of the transpolar drift is the main gateway for multiyear sea ice exiting the Arctic Basin. Knowledge about sea ice in the Fram Strait gives information on the general state of high-arctic sea ice. In particular, sea ice mass balance can be derived from knowing ice concentration, extent and thickness. Since 2003, we regularly carry out in situ sea ice surveys in the Fram Strait. Sea ice thickness data were collected yearly in September and during two spring cruises in 2005 and 2007. Ice and snow thickness is measured along profiles by ground-based electromagnetics and drillings. In May 2005, additional profiles were obtained using airborne helicopter electromagnetics. Sea ice observed in September in the Fram Strait is commonly multiyear ice with modal total thicknesses (ice plus snow) between 2.25 and 3.25 m. Snow thickness is generally small, since only a small amount of new snow accumulated after the summer melt (modal snow thickness between 0.02 and 0.11 m). Whereas the majority of the ice drifts relatively fast in a SSW direction, grounded icebergs function as islands on the Greenland Shelf, resulting in multiyear fast ice forming in their vicinity. Spring surveys reveal substantial amounts of snow covering both first- and multiyear ice. With airborne thickness profiling, a gradient of increasing modal total thickness from east (1.65 m) to west (3.25 m) could be identified. The different sea ice regimes in the Fram Strait represent ice of different age and origin. It is planned to monitor possible trends in ice thickness by continuing the surveys, combined with continuous ice draft data obtained from moored upward looking sonars.
C21C-03 INVITED
ICESat: Sea ice freeboard, snow depth, and thickness
Total freeboard (snow and ice) and thickness of the Arctic Ocean sea ice cover are derived from ICESat data for two 35-day periods: one during the fall (Oct-Nov) of 2005 and the other during the winter (Feb-Mar) of 2006. Our freeboard retrieval approach is based on reflectivity and the expected statistics of freeboard variability from combined analysis of RADARSAT/ICESat data. Results suggest that our retrieval procedures could provide consistent freeboard estimates along 25-km segments with uncertainties of better than several centimeters. With a climatology of snow density, ECMWF snowfall is used to construct a time-varying field of snow depth for the conversion of freeboard to sea ice thickness. The derived ice thickness estimates are compared with ice draft observations from moored upward looking sonar data and the snow depth/thickness data from mass balance buoys in the Beaufort Sea. Preliminary results show that the estimated ICESat thickness estimates are within 0.5 m of the ice drafts reported by moorings. In this talk, we highlight some of the issues associated with the process of freeboard retrieval, thickness estimation, and quality assessment due to the disparity of spatial resolution between the ICESat footprint and those from in-situ measurements.
C21C-04 INVITED
Continued thinning of sea ice at the North Pole and in the Transpolar Drift?
The extent of Arctic sea ice is strongly decreasing, with a new record summer minimum observed in 2007. However, the interpretation of this rapid decline in the context of climate change is hampered by the lack of ice thickness data, as it is unclear how much of the retreat is due to variations in thermodynamic or dynamic boundary conditions. A 350 km long ice thickness profile was obtained in April 2007 between the North Pole and 87N, 58W by means of helicopter-borne electromagnetic profiling. The measurements revealed the presence of predominantly second-year ice with a mean total (snow plus ice) thickness of 3.31 +/- 1.51 m and a modal total thickness of 2.35 m. There was no strong ice thickness trend towards the coast of Ellesmere Island. Ground-based measurements showed a bi-modal snow thickness distribution, with 0.05 m of snow on first-year ice and 0.2 to 0.4 m on second-year ice. If compared with earlier sporadic measurements in the same region between 1991 and 2001, the results indicate a continued ice thinning of 0.3 to 0.7 m between 2001 and 2007. However, the comparison is complicated by the unknown amplitude of the seasonal thickness cycle, as previous measurements were only performed during August. Therefore, the paper discusses the use of additional data like e.g. from ice mass balance buoys which can give insights into the thermodynamic components of sea ice mass balance, and outlines the urgent need for more systematic snow and ice thickness observations. This need has unfortunately only barely been met so far during the ongoing International Polar Year. The paper will also present plans for future, more systematic airborne campaigns, which could cover the whole region of the Arctic Ocean. These include an Arctic airship crossing of the Total Pole Airship project, and an Arctic circumnavigation with a new, German BT67 airplane. http://www.jeanlouisetienne.fr/poleairship
C21C-05
A decade of sea ice thickness mapping by airborne lidar between Greenland and the North Pole
Airborne laser altimetry provides a direct measurement of sea ice freeboard, when combined with a precise geoid model and a lowest-level filtering algorithm to take into account residual errors in GPS-positioning, ocean dynamic topography, tides etc. Using swath laser scanning, the method additionally gives detailed information on the geometry of leads, ridges and the distribution of thin ice and open water. The conversion of sea ice freeboard heights to thickness is based on the assumption of equilibrium, with major errors sources relating to snow depth and density of sea ice. In the paper we describe results of measurements with airborne laser north of Greenland, Ellesmere Island and in the Fram Strait region, carried out on a yearly basis since 1998, in the first years using a single beam laser, and since 2001 using swath laser scanning giving a resolution of approximately 1 m in the ice features. The campaigns have mostly been done in the spring period, typically in connection with airborne gravity surveys or CryoSat calibration and validation activities. Observed secular changes in the sea ice freeboard heights are masked by limited spatial and temporal extent of campaigns, as well as interannual variability in the sea ice regime of the region. To address the error sources in the lidar thickness determination, a number of in-situ and helicopter EM comparisons have been carried out, e.g latest in April 2007 around the Tara drifting station beyond the North Pole, as part of the Damocles project. In cooperation with ESA and APL, coincident Ku-band radar and laser systems have also been flown, giving a unique opportunity for airborne measurement of snow depth as well.
C21C-06
Spatial and temporal variability of Arctic sea-ice draft, 1975-2005
Sea-ice draft measurements made by upward looking sonars are used to estimate the mean drafts of nominal 50 km horizontal segments acquired on 34 U.S. Navy submarine cruises from 1975 to 2000. The cruises are distributed equally in spring and autumn, and the 2203 mean drafts in the data release area provide coverage of approximately half of the Arctic Ocean. The spatial and temporal variability is modeled by multiple regression as a sum of contributions by a time-average spatial field, a mean annual cycle that is independent of position, a long-term time variation, and a residual. Taken together, the spatial, annual and long-term components explain 79% of the total variance in the data. The overall mean is 2.97 m. The amplitude of the annual cycle is 1.06 m. The time-independent spatial map has a minimum draft of 2.2 m near Alaska and a maximum of 4 m at the edge of the data release area 200 miles north of Ellesmere Island. The long-term variation exhibits thinning ice with a peak draft of 3.42 m in 1980 and a minimum draft of 2.29 m in 2000, a decrease of 1.13 m. The maximum rate of thinning is -0.08 m/yr in 1990. The long-term variation nearly levels off at -0.007 m/yr by the end of the record in 2000. Similar multiple regression analyses are performed on a smaller region centered on the North Pole (NP), where the coverage provided by submarine-mounted sonars is denser than average. The annual cycle and long-term variation for this NP region are similar to corresponding parameters estimated for the entire data release area. However there is little coherent spatial variability in NP and the total variance explained by the model is reduced to 65%. Some of the unexplained variance is related to substantial variability in the summer ice draft. Ice draft measurements from sonars moored at the North Pole Environmental Observatory are used to extend coverage in the NP region to spring 2005. The extended data support the "leveling off" of the long term variation, seen at the end of the submarine-only analysis that ends in 2000. The extended data set does not indicate a continuation of thinning in the NP region from 2000 to 2005.
C21C-07
Arctic sea ice melt in summer 2007: Surface and bottom ice ablation
Satellite observations indicate a record minimum in Arctic sea ice extent in September 2007, with a particularly large retreat in the East Siberian, Chukchi, and Beaufort Seas. Insights on the nature of this retreat, and of the summer melt season, can be gleaned from 7 autonomous buoys that were monitoring the thermodynamic mass balance of the ice during the melt season. The buoys were located in the Beaufort Sea and near the North Pole. The total surface ablation over the summer was 0.5 m to 0.7 m at more southern buoys and 0.1 m to 0.3 m for the buoys further north. These values are comparable to observations made in other years. Bottom albation in the Beaufort Sea was extreme, ranging from 0.6 m to 1.2 m. At one site (75 N, 141 W), ice that was 3.2 m thick in May, had completely melted by the end of August, with a 8:5 ratio of bottom to surface melting. There was an anomalously large amount of bottom ablation for this region, requiring an average ocean heat flux of approximately 70 W m-2 from mid-June through August. These observations suggest that the ocean heat flux played a significant role in the dramatic reduction of the 2007 summer sea ice extent in the Beaufort Sea region.
C21C-08
Sea ice thickness measurements - data management considerations
IPY is in full swing with numerous ice stations, voyages, and autonomous systems. Thickness is a key variable being measured intensively during this period to assess the mass balance of sea ice and its rate of change under increasingly warming conditions. Within this context, quantification of sea ice thickness variability at a multiple of scales is being investigated and recorded in both hemispheres by scores of scientists from around the globe. While this measured variability will lead to new understanding of sea ice and its relationship to the global climate, variability in data structures used to record and archive these results will only lead to inconsistencies, incompatibilities, and a lack of spatiotemporal coherence. The sea ice community (small and adaptable as it is) needs to develop a consensus of critical thickness parameters and associated meta data criterion that can profit from two main sources of data gathering: operations (which usually produce high data volume with minimal analysis) and science projects (offering low data volume, but considerable analysis). Consensus between the operational and science community on key data and meta data requirements will make the difference in both the quality and volume of a long-term sea ice thickness archive. This presentation explores ways to address these issues by considering a fundamental set of protocol/format requirements when recording thickness measurements. In particular, records of the sampling rates and spatial scales, instrument type and uncertainties, etc. Essentially, we need to develop and follow a standard format for archiving sea ice thickness measurements. This does not mean "ASCII" or "CSV", rather a clearly defined parameter set which covers major scales, procedures, and algorithms used to define thickness and associated uncertainties. Efforts on this topic can be used to spearhead a white paper to circulate throughout the community to build a common set of protocols for archiving sea ice thickness from the drilled point measurement to the satellite footprint. We present this topic to stimulate discussion within both operational and research communities on how best to proceed.