Book Review Climate Change Geology Glaciers Remote Sensing Sea-level Rise

Glacial crevasses: how deep do they go? What does it mean?

Morlighem, M., E. Rignot, J. Mouginot, H. Seroussi, and E. Larour, (2014). Deeply incised submarine glacial valleys beneath the Greenland ice sheet. Nature Geoscience. V.7, pp. 418-422. Doi: 10.1038/NGEO2167


Greenland ice sheets are big and beautiful and breathtaking. They are also mysterious and there are major gaps in what is known about the ice thickness in Greenland, especially along the coast. The missing information would be a helpful key for understanding the relationship between sea level and glacial retreat.

Commonly, ice thickness is determined by airborne radar sounding. The technique has many caveats including the expense, the need for numerous surveys over a small area, limitations of depth, and limitations on accuracy. The technique has however proven to work well for interpreting inland ice thickness. In contrast, and perhaps expected based on the caveats, the method has not been usefully applicable to the coastal areas. In the research reviewed here the scientists thought that analyzing data sets of ice thickness and flow with respect to each other, and focusing on bed topography, would enable them to interpret glacial-ocean interaction in Greenland. Information gained could then be used to refine sea-ice and glacial retreat models.


In short, researchers input data into an algorithm and used the results to calculated bed topography along Greenland’s east, west, and north coasts. The two inputs used were ice thicknesses from airborne radar soundings and ice flow rates from high resolution ice motion data derived from satellite interferometric synthetic aperture radar (InSAR).  The information was input into a mass conservation algorithm to calculate ice thickness. Finally, bed topography was determined by subtracting the calculated ice thickness from an available digital ice-surface elevation data set.

The shaded boxes are the areas focused on in this study.
Figure 1: The shaded boxes are the areas focused on in this study. This map depicts bed topography elevation determined from the MC algorithm technique.


Researchers identified previously undetected U-shaped crevasses extending below sea level from the terminus to tens to hundreds of kilometers (km) inland that house fast-flow ice. The features are evidence that the glacial motion has been consistent for some time because they are so large and very well shaped.

Observations varied depending on what section of Greenland studied (Figure 1). In West Greenland there were three troughs all below sea level and all cutting ~100 km inland.   Northern Greenland hosts troughs that extended 50 km inland and a couple hundred meters below sea level, and there were also submarine valleys that did not have ice cover. East Greenland had troughs but they did not extend below sea level, a phenomenon attribute to hard bedrock and colder ice sheets.

The study found that modeling the landscape in this fashion predicts between 100-300% more ice below sea level in Greenland than predicted in early investigations.


The real reason to care about the presence of a network of crevasses is because they are drains to the ocean; only 8% of the identified crevasse network contacts seawater, but it controls 88% of the ice discharged from Greenland. Increased contact of glaciers with the ocean implies Greenland’s ice retreat influenced by the oceans is more susceptible than previously understood; it is possible that a faster than expected retreat inward may occur.  For example retreat could be increased from ocean heat flow 300-400 m below the sea surface.   There is also potential for subtropical water from the North Atlantic to make its way into fjords and contribute to an accelerated retreat.

Investigations like this provide important improvements to existing predictions and models.   With each new study of this type region there is literally going to be a deepened knowledge of glacial dynamics, how scientists predict glaciers will be affected by environmental changes, and how their relationship with the ocean with evolve.

Similar approaches could be applied towards exploring unseen topography below ice sheets on our own planet, and maybe even Jupiter’s ice covered sixth moon, Europa!

One thought on “Glacial crevasses: how deep do they go? What does it mean?

Leave a Reply

Your email address will not be published.