Inventory of EU Marine Climate Change Research

New search

GREENICE - Greenland arctic shelf ice and climate experiment

Summary information

Funding:FP5 - Research project
Total cost:2426334
Ec contribution:1842255
Start date:2002-12-01
End date:2006-06-01
Duration:42 months
Coordinator:Peter Wadhams (
Organisation:University of Cambridge - United Kingdom
Themes:Sea level rise; ice melting
Regio:Arctic; North Atlantic
Project name:GREENICE - Greenland arctic shelf ice and climate experiment
Project summary:Abstract
The overall aim of the project was to study the structure and dynamics of the sea ice cover in a critical region of the Arctic Ocean, north of Greenland, and to relate these to longer-term records of climate variability retrieved from sediment cores.

The ice cover in the region is among the thickest in the Arctic, as the sea ice is forced against the north coast of Greenland and the Canadian Archipelago by the transpolar drift stream. This thick and heavily deformed ice prevents access by even the most powerful icebreakers and has resulted in an almost complete lack of ice, ocean or geological data from the region. The challenge was to determine in what way ice conditions are changing as part of the overall pattern of retreat and thinning seen elsewhere in the Arctic, and at the same time to determine from seabed coring whether such a heavy ice regime, deep inside the ice limits, was ever free of ice during the past two glacial cycles.

The project was an integrated programme of measurements, remote sensing and modelling. Three winter field measurement campaigns were carried out:

Fieldwork in 2003 was aimed at trialling systems and methods for the main ice camp the following year. Efforts centred on the AWI vessel Polarstern, specifically during a two-week period when she was moored to a drifting ice floe in the Yermak Plateau area. Aerial campaigns were conducted with the AWI helicopter-borne electromagnetic induction system (HEM) and the KMS swath laser profilometer, mounted on a Twin Otter aircraft. Activities on the ship included intensive ground-truthing using in situ thickness measurements, both by drilling (ice augers, hot water drill) and sledge-borne EM. The drift also allowed the development of thickness monitoring buoys based on the measurement of the spectrum of flexural-gravity waves in the ice. Concurrent data were obtained from an ice camp (APLIS) north of Alaska, providing long-range comparisons of waves necessary for testing the buoy concept.

2004 saw the project team install and occupy an ice camp in the Lincoln Sea, north of Greenland, using Twin Otter aircraft. The camp was a novel, low-cost, lightweight effort, which provided an excellent platform for science in this otherwise inaccessible region. The camp was placed 280 km north of Alert (85°N, 65°W), and occupied by 10 scientists for two weeks in May. Activities at the camp included geological investigations of the seabed and sub-seafloor, a co-ordinated aerial thickness measurement campaign, in situ measurements of ice thickness and properties, and the deployment of an array of buoys designed to measure both path-integrated ice thickness and drift, hence deformation over the lifetime of the project and beyond.

The opportunity was taken to repeat the HEM and laser measurements north of Alert with a limited campaign in 2005, to examine temporal as well as spatial ice thickness variability in the region.
Project outputs:Scientific achievements

The geology team designed and built a custom lightweight (air-transportable) winch and piston corer to extract sediment samples from the seafloor. Ice camps provide a superb platform for such investigations, as they drift slowly across the ocean beneath and the low-noise environment allows high quality seismic studies to be conducted with relatively simple equipment. The gravel-rich nature of the seabed prevented many samples from being obtained, but seven cores were obtained in all, of which two proved to be very high quality. The cores contained high abundances of sub-polar plankton species which suggest that the study area was ice-free during the last interglacial. This open water area may have occurred as a polynya or may reflect a generally reduced sea ice cover of the interior Arctic Ocean. This is a striking result, since at present the study area is heavily ice-covered, and forecast models of future shrinking Arctic sea-ice cover suggest that this area is one of the least sensitive to warming in the Arctic. The geological results obtained from the GreenICE project challenge this view. The camp also produced the first seismic survey of the shallowest part of the Lomonosov Ridge, across 62 km of drift track. Results show that the top of southern Lomonosov Ridge is bevelled (550 m water depth) and only thin sediments (< 50 ms) cover acoustic basement. About 1 km of sediments is found at the western entrance to the deep passage between southern Lomonosov Ridge and the Lincoln Sea continental margin.

The helicopter-borne EM system was used extensively during the project to map the spatial variability of ice thickness in Fram Strait (2003) and in the Lincoln Sea. A small additional campaign from Alert in 2005 also allowed the investigation of interannual variability. Priority was given to flying lines co-incident with the Twin Otter-mounted laser, to examine the relation between freeboard and thickness, though poor weather during the 2004 ice camp severely limited the number of helicopter flying days. Modal thickness of multiyear ice in the Lincoln Sea appeared to increase from 3.9 m in 2004 to 4.2 m in 2005, with snow thickness also increasing, from 0.18 m (2004) to 0.28 m (2005). A much higher fraction of thick, deformed ice was seen in 2005, compared with 2004. Mean thickness tends to be underestimated by the HEM, since the bird cannot 'see' the full depth of ice keels, due to footprint and porosity effects. Profiles from the HEM and laser data were used to improve the parameterisation of ridging in the AWI sea ice model. Model outputs were compared with field measurements and the optimised model used to examine how ridging, ice thickness and ice drift patterns varied with different atmospheric forcing, representative of past scenarios.

The utility of long period wave measurements in the ice covered Arctic Ocean was examined, with the aim of using the measurements to diagnose the path-integrated ice thickness between the source of the waves (the open ocean beyond Fram Strait) and the measurement point, in this case to the north of Greenland. Autonomous wave measuring buoys were developed and successfully deployed for over two years. Considerable practical and theoretical problems were encountered with the resonant wave theory, however, leading to the evaluation of alternative formulations to extract ice thickness. Viscoelastic parameterisations seemed most promising, and the evaluation of this method is currently underway as part of a follow-up project. The buoys also allowed the drift and dynamics of the ice to be determined. Drift was dominantly southwards, with the ice exiting the region through Nares Strait, between Greenland and Ellesmere Island. Ice draft profiles from a United Kingdom submarine cruise under the GreenICE camp area are also presented. Although this effort was not funded under GreenICE, it provides a useful comparison with the HEM and laser results. Preliminary analysis was carried out from analogue data, as we are still awaiting the detailed digital data to be released.

The Danish National Space Centre (previously KMS) carried out extensive surveys of the sea ice north of Greenland using a ski-equipped chartered Air Greenland Twin-Otter, fitted with a downward-looking swath laser profilometer. This measured ice freeboard on scales of 500 km and efforts focused on relating this to the ice thickness measured by the HEM and in situ drilling, as well as submarine draft profiles. The relation is critical for future satellite missions, which seek to measure ice thickness using freeboard alone. Comparisons between HEM and laser measurements yielded good agreement, though generally higher thickness values are estimated by the laser scanner system. This may be caused by underestimation of the thickest ice, especially ridges, by the HEM system. Laser scanner results in 2003 agreed with in situ observations within the expected accuracy. Comparisons with the submarine are also encouraging, though the laser data tend to have thinner modal thickness. Direct comparisons were also made with the laser measurements from the NASA ICESat satellite. Through the GreenICE airborne campaigns it has been demonstrated that the airborne laser scanner measurements are an effective way to measure sea ice thickness and freeboard over large scales (100 to 1.000 km). Coincident EM measurements, and in situ measures of the freeboard/draft ratio, will continue to be useful to quantify the inherent errors in the freeboard to thickness conversion.

DTU were largely responsible for the acquisition and analysis of satellite remote sensing data. They provided both near-real-time support for the field observations (giving a large-scale overview for aerial and on-the-ground operations) and later, offline, analysis of satellite datasets in comparison with field data. Images were dominantly radar views, from both ENVISAT ASAR and RADARSAT, though extensive use was also made of the optical MODIS images and passive microwave (SMMR, SSM/I) instruments. ASAR data was compared with co-incident laser overflights to establish relations between backscatter and ice freeboard/thickness. Historical data were also used prior to the ice camp to estimate and plan its track. Recent images were used to match submarine tracks, acquired a month earlier, to subsequent laser overflights for draft-freeboard comparisons. Small scale ice dynamics was also examined using high-resolution ASAR data, comparing results with the drift of both Polarstern in 2003 and the buoys in 2004.

The project yielded major discoveries about the periodically ice-free nature of the central Arctic Ocean. In addition, methods of ice thickness determination were inter-compared in an area of high thickness and roughness, and a novel technique of thickness monitoring was found not to function in the way described by its progenitor. The project produced extensive co-operation between groups and demonstrated the valuable insights that a multi-platform approach can achieve. The techniques and rationale developed by the project are finding continued application under follow-up projects such as the European DAMOCLES and CRYOVEX programmes, which will continue to track the evolution of ice thickness in this critical area in coming years.