Examining links between glacier velocities and calving behaviours in Svalbard
In a nutshell
Aim: To understand the links between sliding and calving behaviour in marine-terminating glaciers using Kronebreen, Svalbard as an example system.
1. Are calving speed-ups the cause or the product of changes in glacier velocity?
2. Do changes in glacier velocity propagate up-glacier or down-glacier? Why?
3. How important is the presence of water in basal sliding and glacier velocity?
4. Does the injection of melt from the surface have a significant influence on glacier velocity?
Techniques: stereoscopic and monoscopic time-lapse photogrammetry, hot water borehole drilling, UAV surveying
Supervisors: Nick Hulton (University of Edinburgh/UNIS), Doug Benn (University of St. Andrews/UNIS) and Bryn Hubbard (Aberystywth University)
Associated with the Calving Rates And Impact On Sea Level (CRIOS) research group at UNIS
Calving glaciers in Svalbard
Svalbard is an archipelago located in the Arctic Ocean (76°N, 16°E), approximately halfway between mainland Norway and the North Pole. Of its total area, 60% of Svalbard is covered by glaciers, making it one of the largest glaciated areas in the Arctic. These glaciers consist of large continuous ice masses that are divided by mountain ridges and nunataks.
Svalbard glaciers have largely been retreating from their last maximum extent during the Little Ice Age. This retreat has been reasonably steady, so they have currently contributed little to current sea level rise. However, they are very sensitive to climate change because of the influence of the North Atlantic Ocean current system bringing warmer water to the Arctic. Predictions indicate that Arctic glacier will have a larger contribution to sea level rise in the near-future (Blaszcyk, Jania and Hagen, 2009).
The most significant mechanism by which glaciers contribute to sea level rise is calving. Calving is the complex phenomenon involving the discharge of ice from the front of a glacier, forming icebergs. These can be dramatic events where massive chunks of ice fall off, or small instances that resemble crumbling. Calving was once seen as a minor component of glacier dynamics, but is now recognised as a key player in sea level rise predictions. The rate of calving is predicted to increases in the coming century, but exactly how and when this will occur is uncertain (Benn, Warren and Mottram, 2007).
One of the fastest calving glaciers in Svalbard is Kronebreen, which is an ocean-terminating glacier that lies on the west coast close to the research town called Ny-Ålesund (78°53’N, 12°30’E). The glacier is 50 km long, fed from an upper ice catchment called Holtedahlfonna. The width of the glacier at the front (where the ice meets the ocean) is 5 km, forming a dramatic calving front. The glacier surface is very crevassed, partly because it is persistently fast-flowing. Kronebreen has been known to flow up to 4 metres per day, equating to an average annual velocity between 300-800 metres per year (Eiken and Sund, 2012). Recent increase in the calving rate at the front is suspected to be related to an increase in movement upglacier. However, both velocity speed-ups and slow-downs have been observed after calving events (Nuth et al., 2012).
Long-term monitoring of calving is challenging. Calving events can be distinguished in seismic records, but it can be difficult to distinguish the size and type of calving (Köhler et al., 2012). Satellite monitoring is often constrained by the repeat pass time, so although is good for examining long-term changes in calving rate, cannot be used to monitor short-term changes and individual calving events. Time-lapse photography offers an alternative, providing high-resolution data at a high frame rate (1 image every 10 minutes, for instance). This can be used to effectively observe the size and type of calving events, and measure surface velocities at the same resolution.
What’s been done so far?
Seven time-lapse cameras were deployed from May to September in 2014, with the aim to obtain images covering the entire glacier surface. Five out of seven of these cameras successfully worked throughout the season, capturing images every 30 minutes. This equates to 6000 images per camera, 40 000 images in total, and a full hard-drive!
From these, we have been able to use monoscopic techniques (i.e. analysis using images from one camera) to obtain surface velocities through feature tracking and calving rates at the glacier front. We have also been able to monitor the release of meltwater to the bed via surface lakes, which is understood to be an important influence on sliding and vertical displacement (Stevens et al., 2015).
Along with this, we have obtained a 14-month bed pressure record which is useful for examining links between bed water pressure with surface velocity at a high resolution. Two pressure sensors were installed at the bed of Kronebreen in September 2013 via hot water borehole drilling. The two sensors were stacked on top of each another, with one directly on the bed and the other a couple of metres above. These sensors logged the local pressure every two hours which is relayed through a transmitter at the surface of the glacier, so the data can be accessed remotely. The sensor directly on the bed stopped logging in April 2014 whilst the upper sensor lasted till December 2014, producing the 14-month record. Tight links between the sensors suggest little variation in the vertical column, so we are confident the upper sensor is representative of the bed conditions.
The time-lapse cameras were re-deployed in May 2015. The cameras were re-configured as pairs to enable stereoscopic analysis (i.e. analysis using images from two cameras) for this season. Some of the original sites were occupied along with several new sites, and an additional camera was placed on the south mountain ridge (Collethøgda) to make 8 time-lapse camera sites. All cameras are set to take photos every 30 minutes with the exception of the two cameras focused on the calving front. These are set to take photos every 10 minutes in order to capture detailed information on the type of calving that occurs. We are currently in the process of developing our own software for stereoscopic photogrammetry analysis, which will hopefully be ready when the cameras are collected in September 2015. The software is called PyTrx. It is hoped that this software will perform image matching, registration and feature tracking from a user-friendly interface to determine surface velocities and enable automated calving event detection.
Benn, Warren and Mottram (2007) Earth-Sci Rev. 82
– Key paper on current understanding of calving
Blaszcyk, Jania and Hagen (2009) Polish Polar Research 30
– Estimates of calving fluxes at glaciers in Svalbard
Eiken and Sund (2012) Polar Res. 31
– Previous time-lapse work on Kronebreen (and other Svalbard glaciers) using feature tracking to determine surface velocities
Köhler et al. (2012) The Cryosphere 6
– Calving detected at Kronebreen using seismic records
Nuth et al. (2012) J. Glaciol. 58
– Estimates of long-term calving flux at Kronebreen
Stevens et al. (2015) Nature 522
– Monitoring supraglacial lake drainage on Greenland