Laurence Gray

Glacier velocities and dynamic ice discharge from the Queen Elizabeth Islands, Nunavut, Canada

Recent studies indicate an increase in glacier mass loss from the Canadian Arctic Archipelago as a result of warmer summer air temperatures. However, no complete assessment of dynamic ice discharge from this region exists. We present the first complete surface velocity mapping of all ice masses in the Queen Elizabeth Islands and show that these ice masses discharged ~2.6 ± 0.8 Gt a−1 of ice to the oceans in winter 2012. Approximately 50% of the dynamic discharge was channeled through non surge-type Trinity and Wykeham Glaciers alone. Dynamic discharge of the surge-type Mittie Glacier varied from 0.90 ± 0.09 Gt a−1 during its 2003 surge to 0.02 ± 0.02 Gt a−1 during quiescence in 2012, highlighting the importance of surge-type glaciers for interannual variability in regional mass loss. Queen Elizabeth Islands glaciers currently account for ~7.5% of reported dynamic discharge from Arctic ice masses outside Greenland.

Characterizing interannual variability of glacier dynamics and dynamic discharge (1999–2015) for the ice masses of Ellesmere and Axel Heiberg Islands, Nunavut, Canada

Landsat 7 and RADARSAT-1/RADARSAT-2 satellite images are used to produce the most comprehensive record of glacier motion in the Canadian High Arctic to date and to characterize spatial and temporal variability in ice flow over the past ~15 years. This allows us to assess whether dynamically driven glacier change can be attributed to “surging” or “pulsing,” or whether other mechanisms are involved. RADAR velocity mapping allows annual regional dynamic discharge (iceberg calving) to be calculated for 2000 and the period 2011–2015 (yielding a mean regional discharge of 2.21 ± 0.68 Gt a−1), and velocities derived from feature tracking of optical imagery allow for annual dynamic discharge to be calculated for select glaciers from 1999 to 2010. Since ~2011, several of the major tidewater-terminating glaciers within the region have decelerated and their dynamic discharge has decreased. Trinity and Wykeham Glaciers (Prince of Wales Icefield) represent a notable departure from this pattern as they have generally accelerated over the study period. The resulting increase in dynamic discharge from these glaciers entirely compensates (within error limits) for the decrease in discharge from the other tidewater glaciers across the study region. These two glaciers accounted for ~62% of total regional dynamic discharge in winter 2015 (compared to ~22% in 2000), demonstrating that total ice discharge from the Canadian High Arctic can be sensitive to variations in flow of just a few tidewater glaciers.

Surface Velocities of Glaciers in Western Canada from Speckle-Tracking of ALOS PALSAR and RADARSAT-2 data

ABSTRACT Speckle-tracking of historically acquired ALOS PALSAR and RADARSAT-2 datasets are used to determine the dynamics of major glaciers and ice masses in western Canada over the past decade. For the icefields of the St. Elias Mountains and those that fringe the northern British Columbia/Alaska border, our results are largely consistent with earlier studies that used the same data, but different speckle-tracking techniques, to derive ice motion. However, our results are generally more spatially comprehensive than those previously published, in particular in fast-flowing regions such as Hubbard, Seward, Tweedsmuir and Lowell glaciers. We also produce new velocity maps for the icefields located in the Coast Mountains of southwestern British Columbia and for the Chaba, Clemenceau and Columbia icefields of the Rocky Mountains. Generally, faster flow is present on large ocean- and land-terminating outlet glaciers, particularly those in high accumulation maritime regions. These results, taken together with velocity maps of the Canadian Arctic and Yukon produced in previous studies, mean that baseline maps of glacier velocities determined from speckle tracking of SAR datasets are now available for nearly all the major ice masses of Canada.

Validation of CryoSat-2 SARIn Data over Austfonna Ice Cap Using Airborne Laser Scanner Measurements

The study presented here is focused on the assessment of surface elevations derived from CryoSat-2 SARIn level 1b data over the Austfonna ice cap, Svalbard, in 2016. The processing chain that must be applied to the CryoSat-2 waveforms to derive heights is non-trivial, and consists of multiple steps, all requiring subjective choices of methods such as the choice of retracker, geo-relocation, and outlier rejection. Here, we compare six CryoSat-2 level-2 type data sets of surface elevations derived using different SARIn processing chains. These data sets are validated against surface elevation data collected from an airborne laser scanner, during a dedicated CryoSat validation experiment field campaign carried out in April 2016. The flight pattern of the airborne campaign was designed so that elevations were measured in a grid pattern rather than along single lines, as has previously been the standard procedure. The flight grid pattern was chosen to optimize the comparison with the CryoSat-2 SARIn elevation data, the location of which can deviate from nadir by several kilometers due to topography within the satellite footprint. The processing chains behind the six data sets include different outlier/error rejection approaches, and do not produce the same number of data points in our region of interest. To make a consistent analysis, we provide statistics from the validation of both the full data sets from each processing chain, and on only those data that all the six data sets provide a geo-located elevation estimate for. We find that the CryoSat-2 data sets that agree best with the validation data are those derived from dedicated land ice processing schemes. This study may serve as a benchmark for future CryoSat-2 retracker developments, and the evaluation software and data set are made publicly available.