Work at the Norwegian Polar Institute
January 29, 2006Now that Christmas is over it is time to get back to work again. I am starting work at the Norwegian Polar Institute (NP) in Tromsø. Moving to a new place is a little hectic and stressful but things how calmed down now and I have started to work on my new project: calculating englacial water content of Holtedalfonna on Svalbard. As this work is ongoing and I have just started, this entry will be a brief overview of using radar-echo sounding data to image glaciers.
Glaciers, like trees, have annual accumulation layers. The layers are formed by snowfall that falls each year and which eventually is compressed to ice due to pressure from the new snow above it. Each layer of accumulation, one year of snowfall, thins and spreads out from the weight of the new snowfall. Additionally, due to gravity, glaciers flow. Although we generally think of ice as a solid, it is relatively close to its melting point on Earth and therefore behaves quite similarly to liquid in many cases, only on a longer time scale. We see that the ice flows like a liquid when we measure the stakes on glaciers that were allowed to sit over a winter. They have moved, often at different rates, showing that different areas of a glacier move with different speeds. When examining glaciers or ice sheets, such as those on Greenland and Antarctica, we can see annual accumulation layers and identify properties in them. As the ice is compressed and flows the layers deep down become thinner than those on the surface and interpretation becomes more difficult, instead of dealing with a foot of ice as representative of one year, we may be dealing with centimeters if ice or less as representing one layer.
Ice cores are usually used to examine the annual layers. We drill into the ice and pull out a core that contains layer upon layer of ice. Bubbles of air, and if there is more pressure ice-air clathrates, trapped in the ice give us a picture of the atmospheric composition in that accumulation layer and therefore at a definite time in the past. Furthermore, radioactive isotopes and volcanic dust provide us with dates on layers that can confirm our interpretation. A core is, however, a picture of the ice in one place. We can only be sure of our conclusions where the core is drilled and thus have a picture of the history of the ice sheet at that one spot. This is where ice-penetrating radar comes into the picture.
When operating ice penetrating radar we place a radar transmitting antenna and receiving antenna on the ice surface. The light waves are then transmitted into the ice. Each layer has slightly different dielectric properties and therefore acts as a lens with a slightly different characteristic. The reflections from these ice lenses, representing annual accumulation layers, are clearly visible on a radargram, a visualization of radar data. If we take radar data where a core is drilled we can quickly extend the results from a core several hundred kilometers across an ice sheet and give a large-scale picture of past climate. Furthermore, radar data allow us to examine dynamics inside a glacier. We can see cavities, conduits, water, and the ice-bedrock interface using radar since the optical properties of these objects are different. Furthermore, we can also see the dynamics of the ice. The bending of layers conforming to bedrock topography is visible. We can also see stress and strain in the ice layers. Radar data thus gives us a much more complete picture of the internal workings of an ice sheet and is unique in that it can image large areas of the ice sheet easily without a large amount of bulky equipment.
My specific project is looking at radar data from a glacier called Holtedalfonna on Svalbard. I will calculate and spatially map the englacial water content. This work is just starting and pictured below is some radar data with several digitized layers.
Click on the image for a larger version.


Traveling by ship on Isfjorden.