Ice Masses of the Eastern Canadian Arctic Archipelago 13

Home , Agassiz Ice Cap, Ayles Ice Shelf, Bylot Island, Devon Ice Cap, Leading Island, Penny Ice Cap

Wesley Van Wychen, Luke Copland, and David Burgess

Abstract since *2000 these ice shelves have been undergoing The islands of the Canadian Arctic, known as the significant reductions in their area and volume, often Canadian Arctic Archipelago (CAA), contain the largest caused by episodic calving events that can produce ice area of glacierized terrain outside of the ice sheets of islands with diameters of 10 km or more. Once detached, Greenland and Antarctica. The ice masses are focused in these ice islands can drift within the waters of the CAA the eastern part of the CAA, and stretch from the ice and Arctic Ocean for years to decades and may pose a shelves of northernmost Ellesmere Island to glaciers in threat to Arctic shipping and offshore oil developments. the southernmost parts of Baffin Island, together with a few mountain glaciers in northern Labrador. The majority Keywords of glacier ice (roughly 70%) is contained within the large Glaciers Á Mass balance Á Glacier motion Á Glacier ice masses located on Baffin Island (Penny and Barnes Ice dynamics Á Canadian Arctic Archipelago Caps), Bylot Island (Bylot Island Ice Cap), Devon Island (Devon Ice Cap), Ellesmere Island (Prince of Wales Icefield, Manson Icefield, Sydkap Ice Cap, Agassiz Ice Cap, Northern Ellesmere Icefield) and Axel Heiberg 13.1 Physical Geography of Ice Masses Island (Steacie and Müller Ice Caps). The remaining ice is in the Canadian Arctic stored mainly in smaller ice caps and valley glaciers that skirt the coastline. The region’s major ice caps typically The islands of the eastern Canadian Arctic Archipelago have maximum elevations around 2000 m asl and (CAA) contain *150,000 km2 of glacierized terrain, rep- descend to sea level where outlet glaciers meet the ocean. resenting 28% of the worlds glacierized area outside of the Scientific studies of glaciers in the Canadian Arctic began ice sheets (Sharp et al. 2014) (Fig. 13.1). Approximately in earnest in the 1950s, with previous knowledge largely 107,000 km2 of the glacier area is located in the northern gleaned from historical materials and expedition reports. CAA, in the Queen Elizabeth Islands (QEI: Devon, Axel Systematic in situ surface mass balance measurements Heiberg and Ellesmere Islands), while the remaining ice is initiated in the late 1950s and continue to present day, located in the southern CAA, on Baffin Island providing one of the longest continuous records of glacier (*37,500 km2) and Bylot Island (*5000 km2). The mass balance within the Arctic. The Canadian Arctic also region’s major ice masses are classified either as “ice caps” contains some of the very few remaining northern (broad dome-like ice masses less than 50,000 km2 in area hemisphere ice shelves. Recent analysis indicates that that cover much of the underlying topography), or “ice- fields” (differentiated from ice caps by the presence of W. Van Wychen (&) bedrock nunataks that protrude through the ice in their Department of Geography and Environment Management, interior regions) (Copland 2013). The remaining *10% University of Waterloo, Waterloo, ON, Canada terrestrial ice cover in this region is composed of stagnant e-mail: [email protected] plateau ice caps in the QEI less than *1200 km2 in size, and Á W. Van Wychen L. Copland standalone Alpine cirque glaciers, many of which fringe the Department of Geography, Environment and Geomatics, fi University of Ottawa, Ottawa, ON, Canada coastlines of Baf n Bay, Nares Strait and the Arctic Ocean. Figures 13.2, 13.3, and 13.4 provide field photos of various D. Burgess Natural Resources Canada, 601 Booth Street, glacier features in the Canadian Arctic. Ottawa, ON K1A 0E8, Canada

© Springer Nature Switzerland AG 2020 297 O. Slaymaker and N. Catto (eds.), Landscapes and Landforms of Eastern Canada, World Geomorphological Landscapes, https://doi.org/10.1007/978-3-030-35137-3_13 298 W. Van Wychen et al.

Fig. 13.1 Overview of the major ice masses of the eastern Canadian Icefield; 5 = Sydkap Ice Cap; 6 = Steacie Ice Cap, 7 = Müller Ice Cap; 8 High Arctic (denoted by green shading). 1 = Northern Ellesmere = Devon Ice Cap; 9 = Bylot Island Ice Cap, 10 = Barnes Ice Cap; 11 = Icefield; 2 = Agassiz Ice Cap; 3 = Prince of Wales Icefield; 4 = Manson Penny Ice Cap; 12 = Coastal Glaciers and Ice Caps of Baffin Island

To understand the distribution of glaciers within the source, with the Arctic Ocean a secondary source (Koerner CAA, it is necessary to understand the regional climatology 1979). As a consequence, the distribution of glaciers within of the Arctic Islands. Mean annual air temperatures within the region follows these climatic conditions, with glacierized the Canadian Arctic decrease as a function of latitude and terrain concentrated in high latitude, mountainous areas elevation, such that areas located at high elevations and more adjacent to moisture sources, such as along the east coast of northerly latitudes are colder than those located at low ele- Ellesmere Island and west coast of Axel Heiberg Island vations and latitudes (Sharp et al. 2014). In general, the (Sharp et al. 2014). The major exception to this distribution climate of the CAA is characterized by relatively short, cool is Barnes Ice Cap, which is located on a plateau in the summers followed by longer, colder winters, which provides interior of Baffin Island and is thought to exist as a remnant the climatic conditions necessary for glacier formation and of the Laurentide Ice Sheet Complex (Briner et al. 2009). survival. In terms of precipitation, there is a southeast to Topographically, the major ice masses of the CAA gen- northwest gradient across the Canadian Arctic Archipelago, erally rise to elevations of 1800–2000 m asl at their sum- with annual average precipitation of 420 mm in Iqaluit and mits, with their inland ice descending gradually to margins at 70 mm in Eureka, with most of the region receiving less 400–500 m asl, and their large outlet glaciers often reaching than 200 mm annually. Baffin Bay is the dominant moisture the ocean. Ice cap sectors flowing towards coastlines 13 Ice Masses of the Eastern Canadian Arctic Archipelago 299

Fig. 13.2 The snout of a glacier on southeastern Penny Ice Cap, July 2016. Photograph L. Copland

Fig. 13.3 A land-terminating outlet glacier located on western Axel Heiberg Island, July 2015. Photograph L. Copland

adjacent to the Baffin Bay moisture source tend to be thicker bed elevation profiles (3–4 km in width) along most tide- than their inland counterparts (Koerner 1979; Leuschen et al. water glaciers in the Canadian Arctic (https://nsidc.org/data/ 2017) and are drained primarily by large fast-flowing outlet icebridge). An example from these datasets, combined with glaciers which terminate at sea-level. Locations of maximum high resolution surface elevation data from the ArcticDEM, ice thickness (600–800 m) exist under the head of major reveals detailed surface (Fig. 13.5a) and glacier bed outlet glaciers, where the onset of channelization funnels ice (Fig. 13.5b) topography of the Agassiz Ice Cap, NE Elles- from high elevation accumulation areas to the ocean and has mere Island. eroded the subglacial topography (Dowdeswell et al. 2004). Longitudinal profiles extracted along Cañon Glacier The most recent ice thickness surveys conducted in 2014 highlight the contrasting geometry between the surface and under NASA’s Operation IceBridge program mapped the bed geometry, which is characteristic of many glaciers 300 W. Van Wychen et al.

Fig. 13.4 Photo of the calving front of Trinity Glacier, Prince of Wales Iceﬁeld, August 2016. Photograph L. Copland

across the CAA. The relatively smooth descent in surface (25–75 m y−1) near the equilibrium line altitude (ELA) with elevation of Cañon Glacier as it extends from >2000 m asl to velocities decreasing to stagnation in both an up-glacier and sea level along its 80 km length (Fig. 13.5c) contrasts with down-glacier direction from this point. This pattern also the ice-bed interface, which is controlled by the underlying generally holds true for the small tidewater terminating gla- bedrock and drops sharply to 300–400 m below sea-level for ciers of the southern CAA, where speeds of 40–100 m y−1 are the lowermost 50 km of the glacier (Fig. 13.5d). The fact common near the ELA with ice motion often decreasing to that many tidewater glaciers of the northern CAA are near stagnation both up-glacier and down-glacier from there grounded well below sea level for tens of kilometers inland (Van Wychen et al. 2015). On the other hand, velocities of from their termini make them susceptible to thinning and 30–90 m y−1 are commonly found near the ELA of large acceleration driven by changes in sea level and the pene- tidewater terminating glaciers of the QEI, with velocities tration of warm ocean water at their beds. This can result in increasing in a down-glacier direction typically to 100– enhanced subaqueous melting, floatation, and the likely 250 m y−1 at their terminus (Van Wychen et al. 2014). Of presence of highly deformable marine sediments (Burgess particular note are the Trinity and Wykeham Glaciers of Prince et al. 2005; Dowdeswell et al. 2004). Ongoing analysis of of Wales Icefield, which have recently attained flow speeds high resolution geophysical datasets from airborne and of >1200 m y−1 and >600 m y−1 respectively. These maxi- satellite measurements are critical for improving current mum velocities attained by both glaciers make them currently estimates of mass loss through iceberg calving, changing the fastest flowing glaciers in the Canadian Arctic (Fig. 13.6a). glacier dynamics, and the potential for catastrophic collapse The velocity structure of Belcher Glacier on Devon Ice of glaciers in response to measured and predicted climate Cap can be used as an example of the general patterns of ice and ocean warming (Van Wychen et al. 2016). motion of large outlet glaciers in the northern CAA (Fig. 13.8a, b). Within the upper portion of the ice mass (40–60 km from the terminus), velocities of 5–15 m y−1 are 13.2 Ice Flow of the Eastern CAA common and reflects the fact that ice motion within these areas is likely limited to “internal deformation”, that is, ice 13.2.1 Spatial Pattern of Ice Flow that is frozen to its bed and flowing under its own weight. in the Eastern CAA As the glacier descends to lower elevations and begins to converge towards channelized flow where it enters confined Regional scale velocity mapping has provided significant valleys (15–40 km from the terminus), the convergent flow insight into the spatial variability in motion for the ice masses tends to thicken and accelerate. As a result of increasing of the eastern CAA (Figs. 13.6 and 13.7)(Burgessetal.2005; overburden pressure and longitudinal stress, liquid water is Van Wychen et al. 2012, 2014). Typically, land-terminating generated at the glacier bed which facilitates a transition glaciers in the Canadian Arctic attain their highest velocities towards glacier movement dominated by basal sliding. 13 Ice Masses of the Eastern Canadian Arctic Archipelago 301

Fig. 13.5 a Ice surface elevation of Agassiz Ice Cap (surface glaciers on Agassiz Ice Cap. Data available via NASA’s Operation elevations provided by the ArcticDEM project available online IceBridge. c Extracted centreline ice surface elevations of Cañon https://www.pgc.umn.edu/data/arcticdem/, dashed red line indicates Glacier. Proﬁle location shown in (a). d Extracted centreline ice bottom the centreline of Cañon Glacier. b Ice bottom elevations for select elevation for Cañon Glacier. Proﬁle delineated in (a) 302 W. Van Wychen et al.

Fig. 13.6 Overview of surface ice motion for the main ice masses of Icefield, Manson Icefield, Müller Ice Cap and Steacie Ice Cap. b Sydkap the northern Canadian Arctic Archipelago derived from speckle Ice Cap. c Manson Icefield. d Devon Ice Cap. Red labels indicate tracking of RADARSAT-2 imagery acquired in winter 2016–17 for: names of major outlet glaciers a Northern Ellesmere Icefield, Agassiz Ice Cap, Prince of Wales 13 Ice Masses of the Eastern Canadian Arctic Archipelago 303

Ice velocities in this region increase to 25–100 m y−1, and as the glacier continues its descent towards the ocean, basal sliding becomes an increasingly important component of overall motion. Ice speeds continue to increase to the very lowermost region of the terminus where velocities commonly exceed 200 m y−1 (Fig. 13.8b). In many cases, maximum velocities of 300–500 m y−1 usually occur over low sloping or flat surfaces in the terminus region. This likely indicates that the glacier is overriding deformable sediments that offer little-to-no resistance to ice flow (Burgess et al. 2005). In contrast, the large outlet glaciers of the southern Canadian Arctic generally display a different velocity structure than their northern counterparts, with the velocity structure of Coronation Glacier on Penny Ice Cap providing a good example (Fig. 13.8c, d). In the upper most portions of Coronation Glacier, ice motion is low and indicative of rates of motion expected by internal deformation alone. However, ice velocities peak (*150 m y−1)25–35 km from the terminus, as ice funnels through relatively narrow channels and glacier motion (as indicated above) is likely dominated by basal sliding. However, unlike large outlet glaciers in the north, velocities decrease to 25–50 m y−1 in the lowermost lengths of the glacier indicative of increased basal friction and a reduced importance of sliding as a control on glacier motion. Van Wychen et al. (2015) attributed the asymmetric velocity pattern between the NCAA and SCAA to the fact that outlet glaciers in the NCAA are typically fed by much larger source (i.e. accumulation) areas which help maintain driving stresses that perpetuate glacier flow towards the terminus. In addition, reduced friction of a tidewater glacier as it flows over a subaqueous grounding line into the ocean, compared to a terrestrial terminus, is likely also a factor contributing to the general differences in flow patterns observed between NCAA and SCAA outlet glaciers. It should be noted, however, that these patterns of ice motion are broad gener- alizations only, as glacier surging and pulsing can cause significant deviations from these patterns.

13.2.2 Temporal Variability in Ice Flow

Inter- and intra-annual variability in glacier motion is widespread across the Canadian Arctic, with research as early as the 1960s linking seasonal changes in surface hydrology with accelerated ice motion (Cress and Wyness 1961;Müller and Iken 1973), and identifying surge-type glacier behaviour (Hattersley-Smith 1969;Müller 1969; Løken 1969; Holdsworth 1973, 1977). Beginning in the Fig. 13.7 Overview of surface ice motion for the ice masses of the southern Canadian Arctic Archipelago derived from speckle tracking of early 2000s, a renewed focus was placed on understanding ALOS PALSAR imagery acquired in winter 2007–08 and 2010–11, ice motion and its variability across the CAA, including for: a Bylot Island Ice Cap; b Barnes Ice Cap; and c Penny Ice further investigation of the connections between surface Cap. Red labels indicate names of major outlet glaciers 304 W. Van Wychen et al.

Fig. 13.8 a Overview of surface ice velocities of Belcher Glacier, and centreline surface ice velocities of Coronation Glacier, Penny Ice Cap, b extracted centreline surface ice velocities of Belcher Glacier. Transect and d extracted centreline surface velocities of Coronation Glacier. shown as dashed red line from 1 to 1′ in (a). c Overview of extracted Transect shown as dashed red line from 2 to 2′ in (c) hydrology and seasonal acceleration (Copland et al. 2003a, has helped reveal how large-scale glacier dynamics have b, c; Dowdeswell et al. 2004; Short and Gray 2005; Burgess evolved over time (Williamson et al. 2008; Van Wychen et al. 2005; Bingham et al. 2003, 2006, 2008), as well as et al. 2012, 2016, 2017; Millian et al. 2017; Schaffer et al. systematic cataloguing of surge-type glaciers (Copland et al. 2017; Strozzi et al. 2017), as well as on individual valley 2003a). More recently, repeated regional velocity mapping glaciers (Thomson and Copland 2017, 2018). 13 Ice Masses of the Eastern Canadian Arctic Archipelago 305

From *2000–2015 most glaciers across the Queen network of these glaciers becoming more efficient, produc- Elizabeth Islands showed no detectable change in their ing lower basal water pressures and a reduction in basal dynamics. For example, for the ice masses of Axel Heiberg sliding. Multi-annual velocity variability, perhaps due to and Ellesmere Islands, Van Wychen et al. (2017) found that surging, has also previously been reported for the glaciers of 101 of the 117 (*86%) sampled glaciers showed no Bylot Island Ice Cap (Dowdeswell et al. 2007) and Barnes significant (>20 m y−1) change in velocity based on the Ice Cap (Løken 1969; Holdsworth 1973, 1977), located in available data. Many of these glaciers were small and the southern CAA. As the potential of repeated annual and land-terminating, while significant velocity variability was sub-annual velocity mapping using satellite imagery determined on larger outlet glaciers that terminated in the becomes increasingly feasible, the knowledge of velocity ocean. Eight of these larger outlet glaciers slowed down in variability in this region is likely to improve. the period 2000–2015, while six underwent acceleration followed by deceleration (Van Wychen et al. 2016). Mittie Glacier, which slowed throughout this period, provides a 13.2.3 Dynamic Discharge: Mass Loss via Ice good illustration of the magnitude of velocity change that is Flow possible within the Canadian Arctic. In the early 2000s, the main trunk of the glacier was flowing at speeds of >1000 m Dynamic discharge, or mass loss via iceberg calving from y−1, but slowed to stagnation (<20 m y−1) along the entirety the termini of tidewater glaciers, can be calculated as a of its trunk by 2015. Van Wychen et al. (2016) attributed the function of the surface velocities and measured (or mod- observed deceleration of these eight outlet glaciers to the end elled) ice thicknesses. When combined with the surface mass of their surge cycles. For the glaciers that underwent balance (discussed below in Sect. 13.3), dynamic discharge speed-up followed by slowdown, Van Wychen et al. (2016) allows scientists to refine estimates of total glacier contri- determined that based on the scale, pattern and configuration butions to global sea level rise and associated freshwater of the glacier bed topography, these glaciers were respond- fluxes to the ocean, while providing the information neces- ing to either surging or pulsing mechanisms. This means that sary for determining the dominant processes responsible for the majority of the observed changes in glacier dynamics regional mass loss. For the ice masses of the southern CAA, over the study period were due to regular processes inherent Van Wychen et al. (2015) determined a dynamic discharge − to the glaciers of the region, rather than attributable to estimate of 0.017–0.108 Gt y 1. By comparison, mean external drivers (for example, increases in surface melt). The dynamic discharge (2011–2015) from the northern CAA was − two major exceptions to these trends were the Trinity and *2.47 ± 0.88 Gt y 1 indicating that the northern ice masses Wykeham glaciers, which both at least doubled in surface produce *95% of icebergs within Canada. The data pre- velocity over the period 2000–2015 (Van Wychen et al. sented in Van Wychen et al. (2016, 2017) indicates that only 2016). Rapid acceleration of these glaciers accompanied by 8 glaciers accounted for *75% of total discharge, with significant terminus retreat and substantial thinning, led Van Trinity and Wykeham Glaciers alone accounting for *50% Wychen et al. (2016) to speculate that either ocean or cli- of dynamic discharge (and up to *62% in recent years). mate warming was responsible for the acceleration and Although many of the tidewater terminating glaciers in the retreat of these adjacent ice masses. In addition, because the Canadian High Arctic had reduced discharge between 2000 main trunks of these glaciers are grounded below sea level and 2015, these reductions have been more than compen- for up to 45 km up-glacier from their termini (Leuschen sated by increases in velocity and discharge from the Trinity et al. 2017), they may be prone to continued acceleration and Wykeham Glaciers. These patterns highlight the into the future and catastrophic collapse. importance of routine glacier velocity monitoring as total Although repeat velocity mapping at the annual scale has dynamic discharge from the CAA is sensitive to the varia- not been systematically undertaken in the southern Canadian tions in dynamics of just a few select glaciers in this region. Arctic, recent research has revealed multi-decadal scale ice Millan et al. (2017) compared rates of glacier discharge to motion variability in this region (Schaffer et al. 2017; Strozzi surface mass balance for the northern CAA and found that et al. 2017; Heid and Kääb 2012). For example, Schaffer during the 1991–2005 period dynamic discharge accounted et al. (2017) used optical and RADAR remote sensing for *52% of mass loss, while surface melt accounted for the imagery to determine that ice velocities on six major outlet remainder. However, during the 2005–2014 period, ice glaciers on Penny Ice Cap increased between the mid-1980s discharge only accounted for *10% of total mass loss while and mid-1990s and then subsequently decreased to surface melt accounted for *90% of total mass loss. These 2011/2014. Schaffer et al. (2017) attributed the deceleration results are similar to those found by previous studies of these six glaciers as a response to increasingly negative (Gardner et al. 2011; Van Wychen et al. 2014) and highlight surface mass balances since the mid-1980s, and postulated the fact that surface melt and runoff is increasingly domi- that the higher melt rates have resulted in the basal drainage nating glacier mass losses on a regional scale, although 306 W. Van Wychen et al. dynamic discharge can still be important for specific basins advection of warm air from the northwest Atlantic to the that are drained by large dynamic tidewater glaciers. Canadian High Arctic and increased anti-cyclonic air circulation over the QEI in summer (Gardner et al. 2011; Sharp et al. 2011; Lenaerts et al. 2013). These changes in atmo- 13.3 Glacier Mass Balance Within the CAA spheric circulation have accounted for the longer and more and Recent Trends intense melt seasons within the Canadian Arctic in recent years (Wolken et al. 2009; Dupont et al. 2012; Zdanowicz Total mass balance of glaciers in the CAA is estimated et al. 2012; Fisher et al. 2012; Mortimer et al. 2016). As a through combined knowledge of the mass lost through result, glaciers and ice caps in the CAA were largely in dynamic discharge (as described above) and surface mass balance until the late 1980s, but became significantly neg- balance, where surface mass balance is the difference ative by the mid-1990s. Since the mid-2000s however, melt between mass gained through snowfall, and mass loss due to runoff has increased three-fold relative to the 1960–2004 melt runoff. Surface mass balance is calculated annually over average (Sharp et al. 2011) accounting for 50% of the total a hydrological year referenced to the end of the summer melt mass loss that has occurred over the 57-year period of season which, on the remote High Arctic glaciers, occurs observation. around the end of August or early September. When mass In the *mid-1990s, airborne campaigns to map the gain, or accumulation, exceeds total ablation (i.e. melt runoff surface elevation (and surface elevation change) of Canada’s plus dynamic discharge), glaciers will grow and advance, High Arctic glaciers and ice caps have allowed for more while glaciers will shrink and recede when total ablation regionally expansive inferences of changes in glacier mass to exceeds accumulation. be made augmenting results from the in situ monitoring In situ measurements of glacier surface mass balance over programme (Abdalati et al. 2004). Since the early 2000s, long-term reference glaciers in the Canadian High Arctic space-borne measurement techniques (such as gravimetric (White Glacier, Meighen Ice Cap, and Devon Ice Cap measurements provided by the GRACE (2002–2017) satel- (NW)), indicate that the monitored glaciers in this region lite system and repeated measurements of glacier surface have thinned by 8–12 m since the early-1960s (Fig. 13.9). elevation provided by ICESAT (2003–2009)) have provided Winter accumulation over the monitored glaciers in the insight into regional changes in mass of glaciers and ice caps Canadian high Arctic has remained relatively consistent across the eastern Canadian Arctic. Mass loss from glaciers since the *1960s, which means that total glacier mass in the CAA increased more than two-fold from *25 Gt y−1 balances in the Canadian Arctic have largely been driven by for the 1995–2000 epoch (Abdalati et al. 2004)to60± increased summer melt (Koerner 2005; Sharp et al. 2011). 8Gty−1 between 2004 and 2009 (Gardner et al. 2004). Since the late 1980s, and especially since 2005, changing Jacob et al. (2012) documented similar mass loss estimates summer atmospheric circulation patterns have increased of 67 ± 6Gty−1 between 2003 and 2010, with mass losses

Fig. 13.9 Cumulative thickness change of reference glaciers in the northern Canadian Arctic Archipelago. Mass balance data for Devon (NW) and Meighen Ice Cap is courtesy of the Geological Survey of Canada, mass balance data for White Glacier (Müller Ice Cap) from Trent University (1960–2012; G. Cogley) and University of Ottawa (2013–2015; L. Copland & L. Thomson). See Fig. 13.1 for site locations 13 Ice Masses of the Eastern Canadian Arctic Archipelago 307 continuing to accelerate to 2015 (Harig and Simons 2016). 2017). Arctic ice shelves are further differentiated from their Notwithstanding contributions from the continental ice Southern Hemisphere counterparts by their size, with the sheets, Canadian glaciers and ice caps have become the largest individual arctic ice shelf, the Ward Hunt Ice Shelf of largest contributors to eustatic sea-level rise (i.e. the com- Northern Ellesmere Island, being three orders of magnitude ponent of sea-level change by changes in quantity of global smaller than the large ice shelves of Antarctica (Dowdeswell ocean water) on Earth. Canada’s glaciers and ice caps cur- and Jeffries 2017). rently contribute 0.16 mm a−1 to global sea-level rise, which Evidence based on radiocarbon data indicates that the ice is 23% and 75% of net contributions from the Greenland and shelves of northern Ellesmere Island developed Antarctic ice sheets, respectively (Gardner et al. 2013; *5500 years ago in response to Holocene cooling (England Shepherd et al. 2012). et al. 2017), which allowed for the advance of glaciers Nearly all model projections indicate that glacier surface beyond their current positions and the formation of a large mass balance will continue to be negative for the foreseeable contiguous ice shelf termed the “Ellesmere Ice Shelf” future across the Canadian Arctic Archipelago. Lenaerts et al. (Vincent et al. 2001; Dowdeswell and Jeffries 2017). In (2013) and Marzeion et al. (2012) predict that under moderate 1906, the “Ellesmere Ice Shelf” had an estimated area of warming scenarios published in the Intergovernmental Panel 8597 km2, but as of 2015 only smaller remnants were left in on Climate Change AR5 report, glaciers and ice caps across individual fiords, totalling 535 km2 (Mueller et al. 2017). this region are expected to lose 18% (or 35 mm sea-level The main remaining pieces comprise the Ward Hunt Ice equivalent) of their total mass by 2100. Similarly, Radić et al. Shelf (total 224.3 km2, now split into separate eastern and (2014) estimate that by 2100 mass loss from Canada’s Arctic western portions), the Milne Ice Shelf (187.4 km2) and the glaciers and ice caps will contribute 41–57 ± 18 mm of SLE Petersen Ice Shelf (19.6 km2). Small pieces <10 km2 in area under moderate (RCP4.5) to high (RCP8.5) warming sce- remain in a few other fiords, but almost all of these have narios. Canadian land ice will thus continue to lose mass and been undergoing rapid reductions in extent over the past be an important contributor to global sea-level rise into the decade (Mueller et al. 2017). twenty-second century, with significant implications for ocean circulation and the health and sustainability of marine ecosystems (Böning et al. 2016). 13.4.2 Physical Structure of Arctic Ice Shelves

The ice shelves of Northern Ellesmere Island range in 13.4 Ice Shelves and Ice Islands thickness from *20–100 m (Mortimer et al. 2012; White et al. 2015), compared to antarctic ice shelves that can 13.4.1 Definition and Origin of Arctic Ice approach 1 km in ice thickness (Griggs and Bamber 2011). Shelves They are characterized by an undulating surface topography punctuated by troughs and ridges with wavelengths of *30– The largest concentration of ice shelves in the Arctic is 300 m (Jeffries 2017) and amplitudes up to 7.5 m (Mortimer found along the northernmost coast of the CAA. Ice shelves et al. 2012) (Fig. 13.10). The origin of the rolling topogra- comprise permanent floating ice masses that can fill fiords phy of these ice shelves is somewhat unclear, but Holds- and extend many kilometres into the ocean, and are defined worth (1987) and Jeffries (2017) offer a number of possible by Dowdeswell and Jeffries (2017) as: “an ice mass of explanations including: thermal stress, sea-ice pack pressure, considerable thickness ( 20 m) that is afloat on the ocean stacking of sea ice pressure ridges, glacier compression, tidal but attached to the coast. It is often of great horizontal extent action, snow dune formation due to prevailing wind patterns (many km) and has, typically but not exclusively, a regularly and surface meltwater that pools into elongated lakes due to undulating surface. An ice shelf can form by the seaward the prevailing wind patterns. Although the exact process (or extension of a glacier or glaciers, or by the formation of combination of processes) responsible for these features is multiyear sea ice, or both and thicken further by the accu- unclear, they nevertheless provide a distinctive surface mulation of snow at the top surface and the accretion of ice topography that aids in distinguishing them from adjacent from water at the bottom surface”. Although they are sea ice, glaciers and icebergs. commonly associated with the Antarctic, where they make Dowdeswell and Jeffries (2017) describe four types of ice up about 75% of the coast, ice shelves also exist in the shelves based on their method of formation, which provides Northern Hemisphere where they are primarily concentrated a useful way to classify these features: on the northern coastline of Ellesmere Island (Jeffries 2017), some Eurasian fjords that fringe the Arctic Ocean (Dow- (a) Classical Ice Shelf: this type is commonly associated deswell 2017) and the floating ice tongues of northernmost with those found in Antarctica, and is comprised of Greenland where fast-flowing glaciers reach the ocean (Reeh floating margins of glaciers or an ice sheet, where 308 W. Van Wychen et al.

Fig. 13.10 View across the central portion of the Milne Ice Shelf, July 17, 2014. Photograph L. Copland

glacial ice reaches tidewater and remains intact on the portion in 1983 (Jeffries 1991). The entire contiguous ocean surface. The Milne Ice Shelf (Fig. 13.10)of “Ellesmere Ice Shelf” is thought to have primarily been northern Ellesmere Island, which is comprised of a of this type when it existed at the start of the 20th number of glacier outlets that have coalesced within the century. confines of the steep-sided Milne Fiord, provides the (c) Composite Ice Shelf: this refers to an ice shelf that is closest approximation of this type of ice shelf in the comprised of both glacier ice and sea ice, sometimes Northern Hemisphere. Ground-penetrating radar mea- also with the inclusion of icebergs. The Serson Ice Shelf surements in 2008/9 indicate that it reaches a maximum on northern Ellesmere Island provided a good example thickness of *94 m (Mortimer et al. 2012). of this type of ice shelf prior to its almost complete (b) Sea-ice Ice Shelf: this describes ice shelves which form break-up in 2008 (Lemmen et al. 1988; Dowdeswell due to the thickening of multi-year sea ice which has and Jeffries 2017; Mueller et al. 2017). This ice shelf remained in place in protected inlets and embayments was made up of two main glacier tongues which for long periods of time (centuries to millennia). These extended many kilometres into the ocean, with inter- ice shelves act as a platform for new ice to accumulate vening areas infilled with ice that had originated as very on their surface from snowfall, and on their bottom old multi-year landfast sea ice (Dowdeswell and Jeffries from basal freeze-on of sea water, which allows them to 2017). thicken substantially over time (Lemmen et al. 1988; (d) Glacier Tongues: this refers to the floating margins of Jeffries et al. 1989; Dowdeswell and Jeffries 2017). By glaciers that are narrow relative to their overall lengths, definition, they are at least 20 m thick, but can be confined by either slow moving ice (typically sea ice) or substantially thicker than this. The Ward Hunt Ice Shelf bedrock on their sides and primarily fed by fast-flowing is the best current example of this type of ice shelf in outlet glaciers (Dowdeswell and Jeffries 2017; Hambrey the Northern Hemisphere, and was estimated to be 1994). These features are common on the northern 44.5 m thick from a borehole drilled in its western coastline of Greenland, and a number are found in the 13 Ice Masses of the Eastern Canadian Arctic Archipelago 309

protected fjords of northern Ellesmere Island. Mueller next), or is evacuated from the Arctic Ocean along the et al. (2017) identified eight ice tongues with a total area eastern coastline of Greenland via the Transpolar Drift of 94 km2 in 1959, compared to 9 ice tongues with a ocean current. combined area of 23 km2 in 2015. The number of (2) Archipelago Drift occurs when an ice island drifts into tongues increased over this period, even though their the channels between the islands of the Canadian Arctic area declined by >75%, due to the reclassification of Archipelago and eventually disintegrates within these some ice shelves to ice tongues as the surrounding ice waterways. For example, Hobson’s Choice Island shelf was lost. The largest remaining feature of this type drifted into Queen’s Channel and broke up there in in the Canadian Arctic is the Milne Glacier Tongue, 1992 (Jeffries 1992) (Fig. 13.11). which comprises the northernmost *16 km of the (3) Nares Strait Drift describes the situation when ice Milne Glacier (Hamilton et al. 2017). islands drift eastward immediately after calving and become drawn into Nares Strait, the channel located In terms of mass balance processes, ice shelves gain mass between eastern Ellesmere Island and northwest through surface accumulation of snow and superimposed Greenland. These ice islands eventually drift southward ice, sea ice thickening, water freezing to their base, and through Baffin Bay to the east coast of Newfoundland glacier input (Mueller et al. 2008; Jeffries 2002; Dowdeswell and Labrador, where they melt and break-up. Ice island and Jeffries 2017). The relative importance of each of these WH-5 is the only ice island from northern Ellesmere accumulation processes varies by ice shelf type. Ice shelves Island known to have made this route, between lose mass via surface melt and runoff, sublimation, basal approximately 1961 and 1964 (Nutt 1966). melt at the ice-ocean interface and the calving of icebergs or ice islands. Depending on the drift path that an ice island undertakes it can potentially have a significant lifespan before it breaks up. This is especially true for ice islands that undertake 13.4.3 Ice Islands Arctic Ocean Drift. For example, Ice Island T-3 circulated three times around the Arctic Ocean over a period of A unique feature of the ice shelves of northern Ellesmere *40 years, before eventually exiting the Arctic Ocean Island, compared to any other ice mass in the CAA, is that through Fram Strait and breaking up off southern Greenland they can produce very large ice islands when they calve. Ice in 1984 (Jeffries 1992; Van Wychen and Copland 2017). islands refer to large tabular icebergs, which typically have a Additionally, as ice islands disintegrate they typically freeboard of at least *2–5 m and can reach >500 km2 in become increasingly difficult to track, and may spawn many area (Jeffries 1992; Van Wychen and Copland 2017). They smaller ice islands from one large original feature. For these are frequently distinguishable by their ribbed surface reasons, ice islands can pose a particular threat to offshore topography, inherited from the ice shelf from which they activities and infrastructure as a single original ice island formed. The largest recorded ice island calved from northern may threaten the same region multiple times as it breaks Ellesmere Island in *1950 and was named T-2, and had apart. dimensions of *31 Â 33 km (Van Wychen and Copland 2017; Jeffries 1992). As these ice islands primarily calve from free-floating ice shelves, their ice thicknesses are 13.4.4 Recent Changes of the Northern similar to their parent source and can reach values of 20– Ellesmere Ice Shelves 100 m. Given the large dimensions of ice islands they can pose Between 1906 and 1982, *90% of the ice shelf area that particular threats to shipping and offshore infrastructure, skirted the northern coastline of Ellesmere Island at the start such as oil production platforms, as they circulate within of the 20th century was lost. Over this time, losses mainly Arctic waters. Van Wychen and Copland (2017) catalogued occurred during large calving events prior to the 1950s, historical drift tracks of ice islands that originated from the which released many ice islands into the Arctic Ocean and northern coastline of Ellesmere Island and found that these left intact ice shelves in protected fiords of northern Elles- ice features undertook three general routes (Fig. 13.11). mere Island (Copland et al. 2017; Koenig et al. 1952). A second period of large calving events started in *2005, (1) Arctic Ocean Drift occurs when an ice island drifts with the remaining ice shelves losing *50% of their 2005 westward after detaching from the coastline and area by 2012 (Copland et al. 2017). Of the three large becomes entrained in the Beaufort Gyre, allowing it to remaining ice shelves (Petersen, Milne and Ward Hunt), the circulate within the Arctic Ocean. Eventually the ice Milne thinned by an average of 8.1 ± 2.8 m between 1981 island either undertakes Archipelago Drift (described and 2008/2009 and lost 13% of its 1981 volume (Mortimer 310 W. Van Wychen et al.

Fig. 13.11 Drift paths taken by large ice islands over the past *80 years across the CAA: (1) Arctic Ocean Drift; (2) Archipelago Drift; (3) Nares Strait Drift. Dates in legend refer to earliest possible date of calving. Modified from Van Wychen and Copland (2017) et al. 2012). Point measurements of ice thickness on the freeze-on. The decrease in FDD was identified as par- Ward Hunt Ice Shelf indicate that it thinned from *40– ticularly important leading up to the loss of the Ayles 50 m in the 1950s to *25–30 m in the early 2000s, with Ice Shelf in 2007 (Copland et al. 2007). Although most of the thinning taking place since the 1990s (Braun changes in summer air temperature have been less 2017). Although repeated ice thickness mapping has not apparent in the region than changes in other seasons, been completed on the Petersen Ice Shelf (Fig. 13.12), mass White et al. (2015) showed that losses of the Petersen balance measurements indicate that it is losing mass at a rate Ice Shelf occurred coincident with record-breaking of 28.45 Mt y−1 at the surface, with little evidence that ice is summer air temperatures in 2005, 2008, 2011 and 2012. being replaced via basal freeze-on, and will be entirely lost (2) Reductions in Glacier Input into Ice Shelves: for Clas- by the 2040s or earlier (White et al. 2015). As such, it is sical and Composite ice shelves, there is evidence of important to understand the processes driving the current long-term reductions, or the complete loss, of glacier loss of these ice shelves. input to the Ayles, Milne and Petersen ice shelves since Copland et al. (2017) identified several factors that have *2000 (Mortimer et al. 2012; White et al. 2015; likely led to the recent loss of the northern Ellesmere Ice Copland et al. 2017). For example, five tributary gla- Shelves, including. ciers provided input to the Milne Ice Shelf in 1959, but this was reduced to only one by 2011. These reductions (1) Rising Regional Air Temperatures, particularly in the in ice input reduce the mass delivered to these systems autumn, winter and spring seasons, which have reduced and prime them for break-up and disintegration. the overall number of freezing degree days (FDD) that (3) Long-Term Negative Mass Balances: in situ surface are important for limiting the impacts of summer melt mass balance stakes surveyed on the Ward Hunt Ice and for ice shelf recovery via processes such as basal Shelf indicate that winter accumulation rates have 13 Ice Masses of the Eastern Canadian Arctic Archipelago 311

Fig. 13.12 View across the Petersen Ice Shelf, July 13, 2014. Photograph L. Copland

remained relatively constant since the 1950s, while from those that remain, and projections of continued climate summer surface melt has been much more variable, warming and sea ice loss in the Arctic, implies that the resulting in generally negative surface mass balances long-term survival of these features is doubtful. over this period (Braun 2017). This trend is similar to all glaciers and ice caps in the Canadian Arctic (see Sect. 13.3), as well as for the other ice shelves within 13.5 Conclusions the region (Mortimer et al. 2012). Incursions of warm Arctic Ocean waters are likely to have promoted Knowledge of the mass balance and characteristics of gla- increased basal melt in the 1990s and again from 2005 ciers and ice masses of the eastern Canadian Arctic Islands to 2007 (Beszczynska-Moller et al. 2012; Copland et al. has been primarily based on in situ observations initiated in 2017) which would have further promoted mass loss the 1960s. Since *2000, the increased availability and res- from these ice shelves. olution of remote sensing imagery and methodologies to (4) Changes in Arctic Sea Ice: episodic losses of ice processes these data have helped to further expand our col- shelves have tended to occur when the northern coast- lective understanding of these ice masses on a regional basis. line of Ellesmere Island is free of sea ice, suggesting All of these observations converge to reveal a rapidly that sea ice plays an important role in buttressing ice changing landscape. In terms of glacier dynamics, regional shelves and aiding their stability and long-term survival velocity mapping has shown that dynamic variability within (Copland et al. 2017). the region is now much more widespread than previously thought. Glacier mass balances indicate that snowfall within Based on the above review, and that of Copland et al. the region has remained relatively steady since the 1960s, but (2017), there is not a single factor that can be attributed as that surface melt and runoff have increased substantially, the mechanism driving recent ice shelf losses. Rather, it is beginning in the 1980s and especially since 2005. Addi- likely the combination of all four of these factors, and per- tionally, nearly all models project that the Canadian arctic haps others, that has led to the loss of ice shelves in the glaciers will continue to lose mass and be a major contributor northern CAA and the reductions in mass of the remaining to sea level rise over the next century. Within this context, the intact ice shelves there. Taken together, the rapid loss of ice ice shelves of Northern Ellesmere Island also continue to lose shelf extent since 2005, indications of continued mass loss mass, punctuated by the episodic detachment of large pieces 312 W. Van Wychen et al. of ice shelf and the creation of ice islands. In light of this, Dowdeswell EK, Dowdeswell JA, Cawkwell F (2007) On the glaciers continued monitoring of the ice features of the eastern CAA of Bylot Island, Nunavut, Arctic Canada, Arctic, Antarctic and Alpine Research, 39(3):402–411. https://doi.org/10.1657/1523- is essential for furthering our understanding of how the 0430(05-123)%5bDOWDESWELL%5d2.0.CO;2 region will continue to respond to climate change. Dowdeswell JA (2017) Eurasian Arctic Ice Shelves and Tidewater Ice Margins. In: Copland L, Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch. 3. Springer Nature, Dordrecht Dowdeswell JA, Jeffries MO (2017) Arctic Ice Shelves: an introduc- References tion. In: Copland L, Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch. 1. Springer Nature, Dordrecht Dupont F, Royer A, Langlois A, Gressent A, Picard G, Fily M, Cliché Abdalati W, Krabill W, Frederick E, Manizade S, Martin C, Sonntag J, P, Chum M (2012) Monitoring the melt season length of the Barnes Swift R, Thomas R, Yungel J, Koerner R (2004) Elevation changes Ice Cap over the 1979–2010 period using active and passive of ice caps in the Canadian Arctic Archipelago. J Geophys Res 109. microwave remote sensing data. Hydrol Processes 26:2643–2652 https://doi.org/10.1029/2003jf000045 England JH, Evans DA, Lakeman TR (2017) Holocene history of Beszczynska-Moller A, Fahrbach E, Schauer U, Hansen E (2012) Arctic Ice Shelves. In: Copland L, Mueller D (eds) Arctic Ice Variability in Atlantic water temperature and transport at the Shelves and Ice Islands, Ch. 7. Springer Nature, Dordrecht entrance to the Arctic Ocean, 1997–2010. ICES J Mar Sci 69:852– Fisher D, Zheng J, Burgess D, Zdanowicz C, Kinnard C, Sharp M, 863 Bourgeois J (2012) Recent melt rates of Canadian Arctic ice caps Bingham R, Nienow P, Sharp M (2003) Intra-annual and intra-seasonal are the highest in many millennia. Global Planet Change 84–85. flow dynamics of a High Arctic polythermal valley glacier. Ann https://doi.org/10.1016/j.gloplacha.2011.06.005 Glaciol 37:181–188 Gardner AS, Moholdt G, Wouters B, Wolken GJ, Burgess DO, Bingham RG, Nienow PW, Sharp MJ, Copland L (2006) Hydrology Sharp M, Cogley JG, Braun C, Labine C (2011) Sharply increased and dynamics of a polythermal (mostly cold) High Arctic glacier. mass loss from glaciers and ice caps in the Canadian Arctic Earth Surf Processes 31:1463–1479 Archipelago. Nature 473:357–360 Bingham RG, Hubbard AL, Nienow PW, Sharp MJ (2008) An Gardner AS, Moholdt G, Cogley JG, Wouters B, Arendt AA, Wahr J, investigation into the mechanisms controlling seasonal speed-up Berthier E, Hock R, Pfeffer WT, Kaser G, Ligtenberg SR, Bolch T, events at a High Arctic glacier. J Geophys Res 113. https://doi.org/ Sharp MJ, Hagen J-O, van den Broeke M, Paul F (2013) A 10.1029/2007/jf000832 reconciled estimate of glacier contributions to sea level rise 2003– Böning CW, Behrens E, Biastoch A, Getzlaff K, Bamber JL (2016) 2009. Science 340:852–857 Emerging impact of Greenland meltwater on deepwater formation Griggs JA, Bamber JL (2011) Antarctic ice-shelf thickness from in the North Atlantic Ocean. Nat Geosci 9:523–527 satellite radar altimetry. J Glaciol 57:485–498 Braun C (2017) The surface mass balance of the Ward Hunt Ice Shelf Hambrey MJ (1994) Glacial environments. UCL Press, London and Ward Hunt Ice Rise, Ellesmere Island, Nunavut, Canada. In: Hamilton AK, Laval BE, Mueller DR, Vincent WF, Copland L (2017) Copland L, Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch. Dynamic response of an Arctic epishelf lake to seasonal and 6. Springer Nature, Dordrecht long-term forcing: implications for ice shelf thickness. The Briner JP, Davis PT, Miler GH (2009) Latest Pleistocene and Holocen Cryosphere 11:2189–2211 glaciation of Baffin Island, Arctic Canada: key patterns and Harig C, Simons FJ (2016) Ice mass loss in Greenland, the Gulf of chronologies. Quat Sci Rev 28:21–22 Alaska, and the Canadian Archipelago: seasonal cycles and decadal Burgess D, Sharp M, Mair D, Dowdeswell JA, Benham TJ (2005) Flow trends. Geophys Res Lett 43:3150–3159 dynamics and iceberg calving rates of Devon Ice Cap, Nunavut, Hattersley-Smith G (1969) Recent observations on the surging Otto Canada. J Glaciol 51:219–230 Glaicer, Ellesmere Island. Can J Earth Sci 6:883–889 Copland L, Sharp M, Dowdeswell JA (2003a) The distribution and Heid T, Kääb A (2012) Repeat optical satellite images reveal flow characteristics of surge-type glaciers in the Canadian High widespread and long term decrease in land-terminating glacier Arctic. Ann Glaciol 36:73–81 speeds. The Cryosphere 6:467–478 Copland L, Sharp M, Nienow P (2003b) Links between short-term Holdsworth G (1973) Evidence of a surge on Barnes Ice Cap, Baffin velocity variations and the subglacial hydrology of a predominantly Island. Can J Earth Sci 10:1565–1574 cold polythermal glacier. J Glaciol 49:337–348 Holdsworth G (1977) Surge activity on the Barnes Ice Cap. Nature Copland L, Sharp M, Nienow P, Bingham R (2003c) The distribution 269:588–590 of basal motion beneath a High Arctic polythermal glacier. J Glaciol Holdsworth G (1987) The surface waveforms on the Ellesmere Island 49:407–414 ice shelves and ice islands. In: Workshop on extreme ice features, Copland L, Mueller D, Weir L (2007) Rapid loss of the Ayles Ice Shelf, technical memorandum, 141 (NRCC 28003). National Research Ellesmere Island, Canada. Geophys Res Lett 34. https://doi.org/10. Council of Canada, pp 385–443 1029/2007gl031809 Jacob T, Wahr J, Pfeffer T, Swenson S (2012) Recent contributions of Copland L (2013) Classification of ice masses. In: Shroder JF glaciers and ice caps to sea level rise. Nature 482:514–518 (ed) Treatise on Geomorphology, Ch. 8.4. Elsevier, Oxford, Jeffries MO, Krouse HR, Sackinger WM, Serson HV (1989) pp 45–52 Stable-isotope (18O/19O) tracing of fresh, brackish and sea ice in Copland L, Mortimer C, White A, McCallum MR, Mueller D (2017) multiyear land-fast sea ice, Ellesmere Island, Canada. J Glaciol Factors contributing to recent Arctic ice shelf losses. In: Copland L, 35:9–16 Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch. 10. Springer Jeffries MO (1991) Massive, ancient sea-ice strata and preserved Nature, Dordrecht physical-structural characteristics in the Ward Hunt Ice Shelf. Ann Cress P, Wyness R (1961) The Devon Island expedition, observations Glaciol 15:125–131 of glacial movements. Arctic 14:247–259 Jeffries MO (1992) Arctic ice shelves and ice islands: origin, growth Dowdeswell JA, Benham TJ, Gorman MR, Burgess D, Sharp M (2004) and disintegration physical characteristics, structural-stratigraphic Form and flow of the Devon Island ice cap, Canadian Arctic. variability, and dynamics. Rev Geophys 30:245–267 J Geophys Res 109. https://doi.org/10.1029/2003jf000095 13 Ice Masses of the Eastern Canadian Arctic Archipelago 313

Jeffries MO (2002) Ellesmere Island Ice Shelves and Ice Islands, in SharpM,BurgessDO,CawkwellF,CoplandL,DavisJA, Satellite Image Atlas of Glaciers of the world. In: Williams RS, Dowdeswell EK, Dowdeswell JA, Gardner A, Mair D, Ferrigno JG (eds) Glaciers of North America. US Geological Wang L, Williamson S, Wolken GJ, Wyatt F (2014) Remote Survey, Washington DC, pp J147–J164 sensing of recent glacier changes in the Canadian Arctic. In: Jeffries MO (2017) The Ellesmere Ice Shelves, Nunavut, Canada. In: Kargel JS, Leonard GJ, Bishop MP, KääbA,RaupBH Copland L, Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch. (eds) Global land ice measurements from space, Ch. 9. 2. Springer Nature, Dordrecht Praxis-Springer, pp 205–228 Koenig LS, Greenaway KR, Dunbar M, Hattersley-Smith G (1952) Shepherd A., Ivins ER, Geruo A, Barletta VR, Bentley MJ, Battadpur S, Arctic Ice Islands. Arctic 5:67–103 Briggs KH, Bromwich DH, Forsberg R, Galin N, Horwath M, Koerner RM (1979) Accumulation, ablation, and oxygen isotope Jacobs S, Joughin I, King MA, Lenaerts JTM, Li J, Ligten- variations on the Queen Elizabeth Islands ice caps, Canada. berg SRM, Luckman A, Luthcke SB, McMillan M, Meiser R, J Glaciol 22:25–41 Milne G, Mouginot J, Muir A, Nicolas JP, Paden J, Payne AJ, Koerner RM (2005) Mass balance of glaciers in the Queen Elizabeth Pritchard H, Rignot E, Rott H, Sorensen LS, Scambos TA, Islands, Nunavut, Canada. Ann Glaciol 42:417–423 Scheuchl B, Schrama EJO, Smith B, Sundal AV, van Angelen JH, Lemmen DS, Evans DJ, England J (1988) Ice Shelves of Northern van de Berg WJ, van den Broeke MR, Vaughen DG, Velicogna I, Ellesmere Island, NWT, Canadian landform examples. Can Geogr Wahr J, Whitehouse PL, Wingham DJ, Yi D, Young D, Zwally JH 32:363–367 (2012) A reconciled estimate of Ice-Sheet mass balance. Sci 388 Lenaerts JT, van Angelen JH, van den Broeke MR, Gardner AS, (6111):1183–1189. https://www.doi.org/10.1126/science.1228102 Wouters B, van Meijgaard E (2013) Irreversible mass loss of Short NH, Gray AL (2005) Glacier dynamics in the Canadian High Canadian Arctic Archipelago glaciers. Geophys Res Lett 40:870–874 Arctic from RADARSAT-1 speckle tracking. Can J Remote Sens Leuschen C, Goineni P, Rodriguez-Morales F, Paden J, Allen C (2017) 31:225–239 IceBrdige MCords L2 Ice Thickness, Version 1. Boulder, Strozzi T, Paul F, Wiesmann A, Schellenberger T, Kääb A (2017) Colorado USA. NASA National Snow and Ice Data Center Circum-Arctic changes in the flow of glaciers and ice caps from Distributed Active Archive Center. Acquired 1 April 2014 satellite SAR data between 1990s and 2017. Remote Sens 9. https:// Løken OH (1969) Evidence of surges on the Barnes Ice Cap, Baffin doi.org/10.3390/rs909047 Island. Can J Earth Sci 6:899–901 Thomson L, Copland L (2017) Multi-decadal reduction in glacier Marzeion B, Jarosch AH, Hofer M (2012) Past and future sea-level velocities and mechanisms driving deceleration at polythermal change from the surface mass balance of glaciers. The Cryosphere White Glacier, Arctic Canada. J Glaciol 63:450–463 6:1295–1322 Thomson L, Copland L (2018) Changing contribution of peak velocity Millan R, Mouginot J, Rignot E (2017) Mass budget of the glaciers and events to annual velocities following a multi-decadal slowdown at ice caps of the Queen Elizabeth Islands, Canada, from 1991 to White Glacier. Ann Glaciol 58:145–154 2015. Environ Res Lett 12. https://doi.org/10.1088/1748-9326/ Van Wychen W, Copland L, Gray L, Burgess DO, Danielson B, aa5b04 Sharp M (2012) Spatial and temporal variation of ice motion and ice Mortimer CA, Copland L, Mueller DR (2012) Volume and area changes flux from Devon Ice Cap, Nunavut, Canada. J Glaciol 58:657–664 of the Milne Ice Shelf, Ellesmere Island, Nunavut, Canada, since Van Wychen W, Burgess DO, Gray L, Copland L, Sharp M, 1950. J Geophys Res 117. https://doi.org/10.1029/2011jf002074 Dowdeswell JA, Benham TJ (2014) Glacier velocities and dynamic Mortimer CA, Sharp M, Wouters B (2016) Glacier surface tempera- ice discharge from the Queen Elizabeth Islands, Nunavut, Canada. tures in the Canadian High Arctic, 2000–2015. J Glaciol 62:963– Geophys Res Lett 41:484–490 975 Van Wychen W, Copland L, Burgess D, Gray L, Schaffer N (2015) Mueller D, Copland L, Hamilton A, Stern D (2008) Examining Arctic Glacier velocities and dynamic discharge from the ice masses of ice shelves prior to the 2008 breakup. EOS, Transactions, American Baffin Island and Bylot Island, Nunavut, Canada. Can J Earth Sci Geophysical Union 89(49) 52:980–989 Mueller D, Copland L, Jeffries MO (2017) Changes in Canadian Arctic Van Wychen W, Davis J, Burgess DO, Copland L, Gray L, Sharp M, ice shelf extent since 1906. In: Copland L, Mueller D (eds) Arctic Mortimer C (2016) Characterizing interannual variability of glacier Ice Shelves and Ice Islands, Ch. 5. Springer Nature, Dordrecht dynamics and dynamic discharge (1999–2015) for the ice masses of Müller F (1969) Was the Good Friday Glacier on Axel Heiberg Islad Ellesmere and Axel Heiberg Islands, Nunavut, Canada. J Geophys surging? Can J Earth Sci 6:891–894 Res-Earth Surf 121:39–63 Müller F, Iken A (1973) Velocity fluctuations and water regime of Arctic Van Wychen W, Davis J, Copland L, Burgess D, Gray L, Sharp M, valley glaciers. In: Symposium on the hydrology of glaciers, Dowdeswell JA, Benham T (2017) Variability in ice motion and Cambridge, de l’Association Internationale d’Hydrolgie Scientifique dynamic discharge from Devon Ice Cap, Nunavut, Canada. J Glaciol Nutt DC (1966) The drift of Ice Island WH-5. Arctic 19:244–262 63:436–449 Radić V, Bliss A, Beedlow AC, Hock R, Miles E, Cogley JG (2014) Van Wychen W, Copland L (2017) Ice island drift mechanisms in the Regional and global projections of twenty-first century glacier mass Canadian High Arctic, Ch. 7. In: Copland L, Mueller D (eds) Arctic changes in response to climate scenarios from global climate Ice Shelves and Ice Islands. Springer-Verlag models. Clim Dyn 42:37–58 Vincent WF, Gibson J, Jeffries MO (2001) Ice shelf collapse, climate Reeh N (2017) Greenland Ice Shelves and Ice Tongues. In: Copland L, change, and habitat loss in the Canadian high Arctic. Polar Record Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch. 4. Springer 37:133–142 Nature, Dordrecht White A, Copland L, Mueller D, Van Wychen W (2015) Assessment of Schaffer N, Copland L, Zdanowicz C (2017) Ice velocity changes of historical changes (1959–2012) and the causes of recent break-ups Penny Ice Cap, Baffin Island, since the 1950s. J Glaciol 63:716–730 of the Petersen ice shelf, Nunavut, Canada. Ann Glaciol 56:65–76 Sharp M, Burgess DO, Cogley JG, Ecclestone M, Labine C, Wolken G Williamson S, Sharp M, Dowdeswell J, Benham T (2008) Iceberg (2011) Extreme melt on Canada’s Arctic ice caps in the 21st century. calving rates from northern Ellesmere Island ice caps. Canadian Geophys Res Lett 38. https://doi.org/10.1029/2011gl047381 Arctic, 1999–2003. J Glaciol 54:391–4000 314 W. Van Wychen et al.

Wolken G, Sharp M, Wang L (2009) Snow and ice facies variability Luke Copland is a Professor in the Department of Geography, Environ- – and ice layer formation on Canadian Arctic ice caps, 1999 2005. ment and Geomatics at the University of Ottawa, where he holds the J Geophys Res 114. https://doi.org/10.1029/2008jf001173 University Research Chair in Glaciology. He is also Director of the Labo- Zdanowicz C, Smetny-Sowa A, Fisher D, Schaffer N, Copland L, Eley, ratory for Cryospheric Research (https://cryospheric.org), and works with Dupont F (2012) Summer melt rates on Penny Ice Cap, Bafﬁn his graduate students on projects across northern Canada to investigate Island: Past and recent trends and implications for regional climate. glacier dynamics and the response of ice masses to climate change. J Geophys Res 117. https://doi.org/10.1029/2011jf002248 David Burgess is a Research Scientist at the Geological Survey of Canada where he specializes in the use of remote sensing technology and in-situ methods to study the mass balance and dynamics of ice caps in the Canadian Wesley Van Wychen is the Associate Director of the Canadian Cryospheric Arctic. He leads the Government of Canada’s National Glaciology Pro- Information Network and Polar Data Catalogue at the University of gramme through which he conducts integrated research and assessments of Waterloo where his work focuses on ensuring the long-term integrity of glacier-climate change across Canada’s Arctic and Alpine regions. Dr. Arctic and Antarctic datasets. His research focuses on the development and Burgess is a lead investigator and coordinator for Canada’s involvement in use of novel remote sensing techniques for understanding Arctic environ- calibration and validation of the CryoSat-2 radar altimeter (European Space mental change at the regional scale. He also has a strong interest in Agency) over Canada’s Arctic ice caps. understanding glacier mass balance and the evolution of glacier dynamics in response to climate change.