Knowing the material properties of glacier ice is essential for accurately modelling the flow of the Greenland and Antarctic ice sheets. One way this can be done is to consider the overall alignment of ice crystals. Referred to as ice ‘fabric’, this alignment is generated by cumulative stresses placed upon the ice over millennia. Along with ice temperature, ice fabric exerts a strong influence on the viscosity of glacier ice. Unlike ice temperature, ice fabric has a direction-dependent effect on ice viscosity. This effect is not typically incorporated into ice flow models. Consequently, there is a lack of understanding regarding its overall impact on ice sheet evolution and predicted sea level rise. This is particularly true of ice streams: fast-flowing corridors that transport ice to the oceans. Using data collected by the British Antarctic Survey from the Rutford Ice Stream, West Antarctica, the Marie Skłodowska-Curie Actions supported BRISRES project developed new methods to uncover spatial variation in the ice fabric and viscosity. “By comparing these data with satellite observations of the ice surface velocity, we assessed how the microstructural fabric influences large-scale ice flow behaviour,” says Tom Jordan, previously from the University of Bristol, the project host. The Marie Skłodowska-Curie fellow showed that the fabric leads to ice becoming softer due to shearing and compression within the ice stream, thereby enhancing the ice flow.
Introducing new methods
Information about ice fabric traditionally comes from ice core samples taken from the slow-flowing interior of ice sheets. By studying crystal alignment, researchers can reveal information about both past- and present-day ice flow. Another approach is to use geophysical techniques such as radio-echo sounding (radar), which exploits polarisation, the vibrational direction of radio waves. This technique can measure vertical and lateral variation in ice fabric, essential to understanding the overall impact on ice dynamics. “Radar data is attractive as it can be quickly collected on the ground or from airplanes, enabling continental-scale mapping,” explains Jordan. “However, previous methods sometimes resulted in ambiguous data and didn’t incorporate measurements into ice flow models.” BRISRES developed a new method to extract ice fabric information from polarised radar data based on interferometry, where the wave phase is used to take precise measurements of the crystal alignment. A workflow was then developed for inputting these fabric measurements into ice flow models, specifying direction-dependence of the viscosity. “A field team on an ice sheet can now quickly estimate how the viscosity of ice is impacted by the fabric beneath them,” notes Jordan. The methods were developed in slow-flowing regions of Greenland, where they could be tested against ice core data, then applied to fast-flowing regions of Antarctica. “A major surprise was the ice fabric complexity observed within the Rutford Ice Stream. This included rapid fabric rotations within the ice column, with the fabric in deeper ice being very different from the near-surface. This suggests that a simplified vertical model won’t capture the relevant material properties of the ice,” says Jordan.
Towards improved ice flow models
By constraining ice viscosity, the BRISRES techniques enable more accurate ice flow models. Furthermore, better understanding dynamically critical regions of the ice sheets, such as ice streams, will help researchers explore ice sheet stability, ultimately benefitting sea level rise mitigation strategies. The BRISRES methodologies have already been adopted by glaciology teams, such as those investigating Thwaites Glacier, one of the most unstable glaciers in West Antarctica.
BRISRES, ice flow, ice fabric, core, ice sheet, Antarctic, Greenland, sea level rise, ice crystal, ice viscosity