Antarctic scientists have produced the first detailed three-dimensional map of an ice floe using an underwater sonar, terrestrial laser scanner and snow probe, to make measurements below, above and within the ice floe.
Reporting in Eos Transactions in February, Dr Guy Williams, of the Antarctic Climate and Ecosystems Cooperative Research Centre, said maps generated by combining such technologies would enhance comparison with and calibration of airborne or satellite measurements of sea ice thickness. These measurements are used to monitor climate-related changes in sea ice thickness, distribution and snow cover on a large scale.
Previously, scientists made local or ‘point-based’ measurements of sea ice thickness by drilling hundreds of holes in a floe and measuring the ice depth in each.
‘The limited scope of point measurements makes them difficult to compare with methods that can monitor large spatial extent, such as airborne or satellite surveys,’ Dr Williams said.
‘However, our floe-scale survey method can be scaled up to provide a direct comparison for airborne surveys, bridging the gap between point scale measurements and satellite measurements of sea ice thickness. It can also be downscaled for biological studies such as the effect of sea ice thickness on sea ice algae distribution.’
Dr Williams was part a team operating an autonomous underwater vehicle (AUV) to map the underside of sea ice during the second Sea Ice Physics and Ecosystem eXperiment (SIPEX-II) voyage to East Antarctica, between September and November 2012 (Australian Antarctic Magazine 23: 8, 2012).
The SeaBED-class AUV ‘Jaguar’, developed and operated by the Woods Hole Oceanographic Institution, was equipped with a multibeam sonar that recorded the structure of the under-ice surface by sending out a swath of ‘pings’ and measuring the amount of time it took for the sound to bounce back.
‘We operated at a depth of 20 to 30 metres beneath the ice and drove the AUV in a lawnmower pattern across a 400 by 400 metre grid, with overlapping swaths, to map the ice thickness beneath the ocean surface, at a resolution of better than 0.25 metres,’ Dr Williams says.
‘This resolution enabled us to discriminate individual ridge keels and rafted ice blocks.’
At the same time, on the surface of the ice floe, Dr Ted Maksym of the Woods Hole Oceanographic Institution in the US, and Dr Ernesto Trujillo-Gomez of the École Polytechnique Fédérale de Lausanne in Switzerland, took surface measurements of the snow and ice in a GPS-located 100 x 100m survey grid.
Dr Maksym used an automated snow probe to measure the snow depth and the position of his measurements at 1000–2000 points on the ice floe, with a spatial resolution of 1–3m. These provided point-scale measurements of the snow depth and information about the ice ‘topography’ (surface features) underneath.
Dr Trujillo-Gomez used a terrestrial laser scanner to obtain highly detailed images of the snow topography. The laser had a range of about 300m and scanned objects at an accuracy of within 5cm at 100m, with an even greater resolution closer to the laser. Three scans of the survey area were made at different angles to get a 360 degree view, allowing Dr Trujillo-Gomez to ‘see’ around objects in the snow, such as rafted ice or snow drifts that blocked the laser’s sight of the snow surface behind. He then aligned the different scans using markers that he had precisely located in the survey grid.
His combined scans featured some 50 million data points, most only 2mm apart, generating an image that showed every bump or undulation in the snow and any penguin that stood around for long enough.
By combining their data, the three scientific teams were able to deliver the first complete, coincident, whole-of-floe measurements of sea ice, at modest cost and using logistics typical of a standard sea ice voyage. The maps were also generated quite rapidly, providing information relevant to other projects conducted at the same study site.
‘This work could pave the way for a new era of field experiments exploring ocean and sea ice processes on scales of 10 to 100 kilometres, providing an important link between point-scale and satellite measurements,’ Dr Williams said.
Wendy Pyper
Australian Antarctic Division
