13Ice Masses of the Eastern Canadian ArcticArchipelago

Wesley Van Wychen, Luke Copland, and David Burgess

AbstractThe islands of the Canadian Arctic, known as theCanadian Arctic Archipelago (CAA), contain the largestarea of glacierized terrain outside of the ice sheets ofGreenland and Antarctica. The ice masses are focused inthe eastern part of the CAA, and stretch from the iceshelves of northernmost Ellesmere Island to glaciers inthe southernmost parts of Baffin Island, together with afew mountain glaciers in northern Labrador. The majorityof glacier ice (roughly 70%) is contained within the largeice masses located on Baffin Island (Penny and Barnes IceCaps), Bylot Island (Bylot Island Ice Cap), Devon Island(Devon Ice Cap), Ellesmere Island (Prince of WalesIcefield, Manson Icefield, Sydkap Ice Cap, Agassiz IceCap, Northern Ellesmere Icefield) and Axel HeibergIsland (Steacie and Müller Ice Caps). The remaining ice isstored mainly in smaller ice caps and valley glaciers thatskirt the coastline. The region’s major ice caps typicallyhave maximum elevations around 2000 m asl anddescend to sea level where outlet glaciers meet the ocean.Scientific studies of glaciers in the Canadian Arctic beganin earnest in the 1950s, with previous knowledge largelygleaned from historical materials and expedition reports.Systematic in situ surface mass balance measurementsinitiated in the late 1950s and continue to present day,providing one of the longest continuous records of glaciermass balance within the Arctic. The Canadian Arctic alsocontains some of the very few remaining northernhemisphere ice shelves. Recent analysis indicates that

since *2000 these ice shelves have been undergoingsignificant reductions in their area and volume, oftencaused by episodic calving events that can produce iceislands with diameters of 10 km or more. Once detached,these ice islands can drift within the waters of the CAAand Arctic Ocean for years to decades and may pose athreat to Arctic shipping and offshore oil developments.

KeywordsGlaciers � Mass balance � Glacier motion � Glacierdynamics � Canadian Arctic Archipelago

13.1 Physical Geography of Ice Massesin the Canadian Arctic

The islands of the eastern Canadian Arctic Archipelago(CAA) contain *150,000 km2 of glacierized terrain, rep-resenting 28% of the worlds glacierized area outside of theice sheets (Sharp et al. 2014) (Fig. 13.1). Approximately107,000 km2 of the glacier area is located in the northernCAA, in the Queen Elizabeth Islands (QEI: Devon, AxelHeiberg and Ellesmere Islands), while the remaining ice islocated in the southern CAA, on Baffin Island(*37,500 km2) and Bylot Island (*5000 km2). Theregion’s major ice masses are classified either as “ice caps”(broad dome-like ice masses less than 50,000 km2 in areathat cover much of the underlying topography), or “ice-fields” (differentiated from ice caps by the presence ofbedrock nunataks that protrude through the ice in theirinterior regions) (Copland 2013). The remaining *10%terrestrial ice cover in this region is composed of stagnantplateau ice caps in the QEI less than*1200 km2 in size, andstandalone Alpine cirque glaciers, many of which fringe thecoastlines of Baffin Bay, Nares Strait and the Arctic Ocean.Figures 13.2, 13.3, and 13.4 provide field photos of variousglacier features in the Canadian Arctic.

W. Van Wychen (&)Department of Geography and Environment Management,University of Waterloo, Waterloo, ON, Canadae-mail: wvanwychen@uwaterloo.ca

W. Van Wychen � L. CoplandDepartment of Geography, Environment and Geomatics,University of Ottawa, Ottawa, ON, Canada

D. BurgessNatural Resources Canada, 601 Booth Street,Ottawa, ON K1A 0E8, Canada

© Springer Nature Switzerland AG 2020O. 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

297

To understand the distribution of glaciers within theCAA, it is necessary to understand the regional climatologyof the Arctic Islands. Mean annual air temperatures withinthe Canadian Arctic decrease as a function of latitude andelevation, such that areas located at high elevations and morenortherly latitudes are colder than those located at low ele-vations and latitudes (Sharp et al. 2014). In general, theclimate of the CAA is characterized by relatively short, coolsummers followed by longer, colder winters, which providesthe climatic conditions necessary for glacier formation andsurvival. In terms of precipitation, there is a southeast tonorthwest gradient across the Canadian Arctic Archipelago,with annual average precipitation of 420 mm in Iqaluit and70 mm in Eureka, with most of the region receiving lessthan 200 mm annually. Baffin Bay is the dominant moisture

source, with the Arctic Ocean a secondary source (Koerner1979). As a consequence, the distribution of glaciers withinthe region follows these climatic conditions, with glacierizedterrain concentrated in high latitude, mountainous areasadjacent to moisture sources, such as along the east coast ofEllesmere Island and west coast of Axel Heiberg Island(Sharp et al. 2014). The major exception to this distributionis Barnes Ice Cap, which is located on a plateau in theinterior of Baffin Island and is thought to exist as a remnantof the Laurentide Ice Sheet Complex (Briner et al. 2009).

Topographically, the major ice masses of the CAA gen-erally rise to elevations of 1800–2000 m asl at their sum-mits, with their inland ice descending gradually to margins at400–500 m asl, and their large outlet glaciers often reachingthe ocean. Ice cap sectors flowing towards coastlines

Fig. 13.1 Overview of the major ice masses of the eastern CanadianHigh Arctic (denoted by green shading). 1 = Northern EllesmereIcefield; 2 = Agassiz Ice Cap; 3 = Prince of Wales Icefield; 4 = Manson

Icefield; 5 = Sydkap Ice Cap; 6 = Steacie Ice Cap, 7 = Müller Ice Cap; 8= Devon Ice Cap; 9 = Bylot Island Ice Cap, 10 = Barnes Ice Cap; 11 =Penny Ice Cap; 12 = Coastal Glaciers and Ice Caps of Baffin Island

298 W. Van Wychen et al.

adjacent to the Baffin Bay moisture source tend to be thickerthan their inland counterparts (Koerner 1979; Leuschen et al.2017) and are drained primarily by large fast-flowing outletglaciers which terminate at sea-level. Locations of maximumice thickness (600–800 m) exist under the head of majoroutlet glaciers, where the onset of channelization funnels icefrom high elevation accumulation areas to the ocean and haseroded the subglacial topography (Dowdeswell et al. 2004).The most recent ice thickness surveys conducted in 2014under NASA’s Operation IceBridge program mapped the

bed elevation profiles (3–4 km in width) along most tide-water glaciers in the Canadian Arctic (https://nsidc.org/data/icebridge). An example from these datasets, combined withhigh resolution surface elevation data from the ArcticDEM,reveals detailed surface (Fig. 13.5a) and glacier bed(Fig. 13.5b) topography of the Agassiz Ice Cap, NE Elles-mere Island.

Longitudinal profiles extracted along Cañon Glacierhighlight the contrasting geometry between the surface andbed geometry, which is characteristic of many glaciers

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

Fig. 13.3 A land-terminatingoutlet glacier located on westernAxel Heiberg Island, July 2015.Photograph L. Copland

13 Ice Masses of the Eastern Canadian Arctic Archipelago 299

across the CAA. The relatively smooth descent in surfaceelevation of Cañon Glacier as it extends from >2000 m asl tosea level along its 80 km length (Fig. 13.5c) contrasts withthe ice-bed interface, which is controlled by the underlyingbedrock and drops sharply to 300–400 m below sea-level forthe lowermost 50 km of the glacier (Fig. 13.5d). The factthat many tidewater glaciers of the northern CAA aregrounded well below sea level for tens of kilometers inlandfrom their termini make them susceptible to thinning andacceleration driven by changes in sea level and the pene-tration of warm ocean water at their beds. This can result inenhanced subaqueous melting, floatation, and the likelypresence of highly deformable marine sediments (Burgesset al. 2005; Dowdeswell et al. 2004). Ongoing analysis ofhigh resolution geophysical datasets from airborne andsatellite measurements are critical for improving currentestimates of mass loss through iceberg calving, changingglacier dynamics, and the potential for catastrophic collapseof glaciers in response to measured and predicted climateand ocean warming (Van Wychen et al. 2016).

13.2 Ice Flow of the Eastern CAA

13.2.1 Spatial Pattern of Ice Flowin the Eastern CAA

Regional scale velocity mapping has provided significantinsight into the spatial variability in motion for the ice massesof the eastern CAA (Figs. 13.6 and 13.7) (Burgess et al. 2005;Van Wychen et al. 2012, 2014). Typically, land-terminatingglaciers in the Canadian Arctic attain their highest velocities

(25–75 m y−1) near the equilibrium line altitude (ELA) withvelocities decreasing to stagnation in both an up-glacier anddown-glacier direction from this point. This pattern alsogenerally holds true for the small tidewater terminating gla-ciers of the southern CAA, where speeds of 40–100 m y−1 arecommon near the ELA with ice motion often decreasing tonear stagnation both up-glacier and down-glacier from there(Van Wychen et al. 2015). On the other hand, velocities of30–90 m y−1 are commonly found near the ELA of largetidewater terminating glaciers of the QEI, with velocitiesincreasing in a down-glacier direction typically to 100–250 m y−1 at their terminus (Van Wychen et al. 2014). Ofparticular note are the Trinity andWykeham Glaciers of Princeof Wales Icefield, which have recently attained flow speedsof >1200 m y−1 and >600 m y−1 respectively. These maxi-mum velocities attained by both glaciers make them currentlythe fastest flowing glaciers in the Canadian Arctic (Fig. 13.6a).

The velocity structure of Belcher Glacier on Devon IceCap can be used as an example of the general patterns of icemotion 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 arecommon and reflects the fact that ice motion within theseareas is likely limited to “internal deformation”, that is, icethat is frozen to its bed and flowing under its own weight.As the glacier descends to lower elevations and begins toconverge towards channelized flow where it enters confinedvalleys (15–40 km from the terminus), the convergent flowtends to thicken and accelerate. As a result of increasingoverburden pressure and longitudinal stress, liquid water isgenerated at the glacier bed which facilitates a transitiontowards glacier movement dominated by basal sliding.

Fig. 13.4 Photo of the calvingfront of Trinity Glacier, Prince ofWales Icefield, August 2016.Photograph L. Copland

300 W. Van Wychen et al.

Fig. 13.5 a Ice surface elevation of Agassiz Ice Cap (surfaceelevations provided by the ArcticDEM project available onlinehttps://www.pgc.umn.edu/data/arcticdem/, dashed red line indicatesthe centreline of Cañon Glacier. b Ice bottom elevations for select

glaciers on Agassiz Ice Cap. Data available via NASA’s OperationIceBridge. c Extracted centreline ice surface elevations of CañonGlacier. Profile location shown in (a). d Extracted centreline ice bottomelevation for Cañon Glacier. Profile delineated in (a)

13 Ice Masses of the Eastern Canadian Arctic Archipelago 301

Fig. 13.6 Overview of surface ice motion for the main ice masses ofthe northern Canadian Arctic Archipelago derived from speckletracking of RADARSAT-2 imagery acquired in winter 2016–17 for:a Northern Ellesmere Icefield, Agassiz Ice Cap, Prince of Wales

Icefield, Manson Icefield, Müller Ice Cap and Steacie Ice Cap. b SydkapIce Cap. c Manson Icefield. d Devon Ice Cap. Red labels indicatenames of major outlet glaciers

302 W. Van Wychen et al.

Ice velocities in this region increase to 25–100 m y−1, andas the glacier continues its descent towards the ocean, basalsliding becomes an increasingly important component ofoverall motion. Ice speeds continue to increase to the verylowermost region of the terminus where velocities com-monly exceed 200 m y−1 (Fig. 13.8b). In many cases,maximum velocities of 300–500 m y−1 usually occur overlow sloping or flat surfaces in the terminus region. Thislikely indicates that the glacier is overriding deformablesediments that offer little-to-no resistance to ice flow(Burgess et al. 2005).

In contrast, the large outlet glaciers of the southernCanadian Arctic generally display a different velocitystructure than their northern counterparts, with the velocitystructure of Coronation Glacier on Penny Ice Cap providinga good example (Fig. 13.8c, d). In the upper most portionsof Coronation Glacier, ice motion is low and indicative ofrates of motion expected by internal deformation alone.However, ice velocities peak (*150 m y−1) 25–35 km fromthe terminus, as ice funnels through relatively narrowchannels and glacier motion (as indicated above) is likelydominated by basal sliding. However, unlike large outletglaciers in the north, velocities decrease to 25–50 m y−1 inthe lowermost lengths of the glacier indicative of increasedbasal friction and a reduced importance of sliding as acontrol on glacier motion. Van Wychen et al. (2015)attributed the asymmetric velocity pattern between theNCAA and SCAA to the fact that outlet glaciers in theNCAA are typically fed by much larger source (i.e. accu-mulation) areas which help maintain driving stresses thatperpetuate glacier flow towards the terminus. In addition,reduced friction of a tidewater glacier as it flows over asubaqueous grounding line into the ocean, compared to aterrestrial terminus, is likely also a factor contributing to thegeneral differences in flow patterns observed betweenNCAA 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 causesignificant deviations from these patterns.

13.2.2 Temporal Variability in Ice Flow

Inter- and intra-annual variability in glacier motion iswidespread across the Canadian Arctic, with research asearly as the 1960s linking seasonal changes in surfacehydrology with accelerated ice motion (Cress and Wyness1961; Müller and Iken 1973), and identifying surge-typeglacier behaviour (Hattersley-Smith 1969; Müller 1969;Løken 1969; Holdsworth 1973, 1977). Beginning in theearly 2000s, a renewed focus was placed on understandingice motion and its variability across the CAA, includingfurther investigation of the connections between surface

Fig. 13.7 Overview of surface ice motion for the ice masses of thesouthern Canadian Arctic Archipelago derived from speckle tracking ofALOS PALSAR imagery acquired in winter 2007–08 and 2010–11,for: a Bylot Island Ice Cap; b Barnes Ice Cap; and c Penny IceCap. Red labels indicate names of major outlet glaciers

13 Ice Masses of the Eastern Canadian Arctic Archipelago 303

hydrology and seasonal acceleration (Copland et al. 2003a,b, c; Dowdeswell et al. 2004; Short and Gray 2005; Burgesset al. 2005; Bingham et al. 2003, 2006, 2008), as well assystematic cataloguing of surge-type glaciers (Copland et al.2003a). More recently, repeated regional velocity mapping

has helped reveal how large-scale glacier dynamics haveevolved over time (Williamson et al. 2008; Van Wychenet al. 2012, 2016, 2017; Millian et al. 2017; Schaffer et al.2017; Strozzi et al. 2017), as well as on individual valleyglaciers (Thomson and Copland 2017, 2018).

Fig. 13.8 a Overview of surface ice velocities of Belcher Glacier, andb extracted centreline surface ice velocities of Belcher Glacier. Transectshown as dashed red line from 1 to 1′ in (a). c Overview of extracted

centreline surface ice velocities of Coronation Glacier, Penny Ice Cap,and d extracted centreline surface velocities of Coronation Glacier.Transect shown as dashed red line from 2 to 2′ in (c)

304 W. Van Wychen et al.

From *2000–2015 most glaciers across the QueenElizabeth Islands showed no detectable change in theirdynamics. For example, for the ice masses of Axel Heibergand Ellesmere Islands, Van Wychen et al. (2017) found that101 of the 117 (*86%) sampled glaciers showed nosignificant (>20 m y−1) change in velocity based on theavailable data. Many of these glaciers were small andland-terminating, while significant velocity variability wasdetermined on larger outlet glaciers that terminated in theocean. Eight of these larger outlet glaciers slowed down inthe period 2000–2015, while six underwent accelerationfollowed by deceleration (Van Wychen et al. 2016). MittieGlacier, which slowed throughout this period, provides agood illustration of the magnitude of velocity change that ispossible within the Canadian Arctic. In the early 2000s, themain trunk of the glacier was flowing at speeds of >1000 my−1, but slowed to stagnation (<20 m y−1) along the entiretyof its trunk by 2015. Van Wychen et al. (2016) attributed theobserved deceleration of these eight outlet glaciers to the endof their surge cycles. For the glaciers that underwentspeed-up followed by slowdown, Van Wychen et al. (2016)determined that based on the scale, pattern and configurationof the glacier bed topography, these glaciers were respond-ing to either surging or pulsing mechanisms. This means thatthe majority of the observed changes in glacier dynamicsover the study period were due to regular processes inherentto the glaciers of the region, rather than attributable toexternal drivers (for example, increases in surface melt). Thetwo major exceptions to these trends were the Trinity andWykeham glaciers, which both at least doubled in surfacevelocity over the period 2000–2015 (Van Wychen et al.2016). Rapid acceleration of these glaciers accompanied bysignificant terminus retreat and substantial thinning, led VanWychen et al. (2016) to speculate that either ocean or cli-mate warming was responsible for the acceleration andretreat of these adjacent ice masses. In addition, because themain trunks of these glaciers are grounded below sea levelfor up to 45 km up-glacier from their termini (Leuschenet al. 2017), they may be prone to continued accelerationinto the future and catastrophic collapse.

Although repeat velocity mapping at the annual scale hasnot been systematically undertaken in the southern CanadianArctic, recent research has revealed multi-decadal scale icemotion variability in this region (Schaffer et al. 2017; Strozziet al. 2017; Heid and Kääb 2012). For example, Schafferet al. (2017) used optical and RADAR remote sensingimagery to determine that ice velocities on six major outletglaciers on Penny Ice Cap increased between the mid-1980sand mid-1990s and then subsequently decreased to2011/2014. Schaffer et al. (2017) attributed the decelerationof these six glaciers as a response to increasingly negativesurface mass balances since the mid-1980s, and postulatedthat the higher melt rates have resulted in the basal drainage

network of these glaciers becoming more efficient, produc-ing lower basal water pressures and a reduction in basalsliding. Multi-annual velocity variability, perhaps due tosurging, has also previously been reported for the glaciers ofBylot Island Ice Cap (Dowdeswell et al. 2007) and BarnesIce Cap (Løken 1969; Holdsworth 1973, 1977), located inthe southern CAA. As the potential of repeated annual andsub-annual velocity mapping using satellite imagerybecomes increasingly feasible, the knowledge of velocityvariability in this region is likely to improve.

13.2.3 Dynamic Discharge: Mass Loss via IceFlow

Dynamic discharge, or mass loss via iceberg calving fromthe termini of tidewater glaciers, can be calculated as afunction of the surface velocities and measured (or mod-elled) ice thicknesses. When combined with the surface massbalance (discussed below in Sect. 13.3), dynamic dischargeallows scientists to refine estimates of total glacier contri-butions to global sea level rise and associated freshwaterfluxes to the ocean, while providing the information neces-sary for determining the dominant processes responsible forregional mass loss. For the ice masses of the southern CAA,Van Wychen et al. (2015) determined a dynamic dischargeestimate of 0.017–0.108 Gt y−1. By comparison, meandynamic discharge (2011–2015) from the northern CAA was*2.47 ± 0.88 Gt y−1 indicating that the northern ice massesproduce *95% of icebergs within Canada. The data pre-sented in Van Wychen et al. (2016, 2017) indicates that only8 glaciers accounted for *75% of total discharge, withTrinity and Wykeham Glaciers alone accounting for *50%of dynamic discharge (and up to *62% in recent years).Although many of the tidewater terminating glaciers in theCanadian High Arctic had reduced discharge between 2000and 2015, these reductions have been more than compen-sated by increases in velocity and discharge from the Trinityand Wykeham Glaciers. These patterns highlight theimportance of routine glacier velocity monitoring as totaldynamic discharge from the CAA is sensitive to the varia-tions in dynamics of just a few select glaciers in this region.

Millan et al. (2017) compared rates of glacier discharge tosurface mass balance for the northern CAA and found thatduring the 1991–2005 period dynamic discharge accountedfor*52% of mass loss, while surface melt accounted for theremainder. However, during the 2005–2014 period, icedischarge only accounted for *10% of total mass loss whilesurface melt accounted for *90% of total mass loss. Theseresults are similar to those found by previous studies(Gardner et al. 2011; Van Wychen et al. 2014) and highlightthe fact that surface melt and runoff is increasingly domi-nating glacier mass losses on a regional scale, although

13 Ice Masses of the Eastern Canadian Arctic Archipelago 305

dynamic discharge can still be important for specific basinsthat are drained by large dynamic tidewater glaciers.

13.3 Glacier Mass Balance Within the CAAand Recent Trends

Total mass balance of glaciers in the CAA is estimatedthrough combined knowledge of the mass lost throughdynamic discharge (as described above) and surface massbalance, where surface mass balance is the differencebetween mass gained through snowfall, and mass loss due tomelt runoff. Surface mass balance is calculated annually overa hydrological year referenced to the end of the summer meltseason which, on the remote High Arctic glaciers, occursaround the end of August or early September. When massgain, or accumulation, exceeds total ablation (i.e. melt runoffplus dynamic discharge), glaciers will grow and advance,while glaciers will shrink and recede when total ablationexceeds accumulation.

In situ measurements of glacier surface mass balance overlong-term reference glaciers in the Canadian High Arctic(White Glacier, Meighen Ice Cap, and Devon Ice Cap(NW)), indicate that the monitored glaciers in this regionhave thinned by 8–12 m since the early-1960s (Fig. 13.9).Winter accumulation over the monitored glaciers in theCanadian high Arctic has remained relatively consistentsince the *1960s, which means that total glacier massbalances in the Canadian Arctic have largely been driven byincreased summer melt (Koerner 2005; Sharp et al. 2011).Since the late 1980s, and especially since 2005, changingsummer atmospheric circulation patterns have increased

advection of warm air from the northwest Atlantic to theCanadian High Arctic and increased anti-cyclonic air cir-culation over the QEI in summer (Gardner et al. 2011; Sharpet al. 2011; Lenaerts et al. 2013). These changes in atmo-spheric circulation have accounted for the longer and moreintense melt seasons within the Canadian Arctic in recentyears (Wolken et al. 2009; Dupont et al. 2012; Zdanowiczet al. 2012; Fisher et al. 2012; Mortimer et al. 2016). As aresult, glaciers and ice caps in the CAA were largely inbalance until the late 1980s, but became significantly neg-ative by the mid-1990s. Since the mid-2000s however, meltrunoff has increased three-fold relative to the 1960–2004average (Sharp et al. 2011) accounting for 50% of the totalmass loss that has occurred over the 57-year period ofobservation.

In the *mid-1990s, airborne campaigns to map thesurface elevation (and surface elevation change) of Canada’sHigh Arctic glaciers and ice caps have allowed for moreregionally expansive inferences of changes in glacier mass tobe made augmenting results from the in situ monitoringprogramme (Abdalati et al. 2004). Since the early 2000s,space-borne measurement techniques (such as gravimetricmeasurements provided by the GRACE (2002–2017) satel-lite system and repeated measurements of glacier surfaceelevation provided by ICESAT (2003–2009)) have providedinsight into regional changes in mass of glaciers and ice capsacross the eastern Canadian Arctic. Mass loss from glaciersin the CAA increased more than two-fold from *25 Gt y−1

for the 1995–2000 epoch (Abdalati et al. 2004) to 60 ±

8 Gt y−1 between 2004 and 2009 (Gardner et al. 2004).Jacob et al. (2012) documented similar mass loss estimatesof 67 ± 6 Gt y−1 between 2003 and 2010, with mass losses

Fig. 13.9 Cumulative thicknesschange of reference glaciers in thenorthern Canadian ArcticArchipelago. Mass balance datafor Devon (NW) and Meighen IceCap is courtesy of the GeologicalSurvey of Canada, mass balancedata for White Glacier (Müller IceCap) from Trent University(1960–2012; G. Cogley) andUniversity of Ottawa(2013–2015; L. Copland &L. Thomson). See Fig. 13.1 forsite locations

306 W. Van Wychen et al.

continuing to accelerate to 2015 (Harig and Simons 2016).Notwithstanding contributions from the continental icesheets, Canadian glaciers and ice caps have become thelargest contributors to eustatic sea-level rise (i.e. the com-ponent of sea-level change by changes in quantity of globalocean water) on Earth. Canada’s glaciers and ice caps cur-rently contribute 0.16 mm a−1 to global sea-level rise, whichis 23% and 75% of net contributions from the Greenland andAntarctic ice sheets, respectively (Gardner et al. 2013;Shepherd et al. 2012).

Nearly all model projections indicate that glacier surfacemass balance will continue to be negative for the foreseeablefuture across the Canadian Arctic Archipelago. Lenaerts et al.(2013) andMarzeion et al. (2012) predict that under moderatewarming scenarios published in the Intergovernmental Panelon Climate Change AR5 report, glaciers and ice caps acrossthis region are expected to lose 18% (or 35 mm sea-levelequivalent) of their total mass by 2100. Similarly, Radić et al.(2014) estimate that by 2100 mass loss from Canada’s Arcticglaciers and ice caps will contribute 41–57 ± 18 mm of SLEunder moderate (RCP4.5) to high (RCP8.5) warming sce-narios. Canadian land ice will thus continue to lose mass andbe an important contributor to global sea-level rise into thetwenty-second century, with significant implications forocean circulation and the health and sustainability of marineecosystems (Böning et al. 2016).

13.4 Ice Shelves and Ice Islands

13.4.1 Definition and Origin of Arctic IceShelves

The largest concentration of ice shelves in the Arctic isfound along the northernmost coast of the CAA. Ice shelvescomprise permanent floating ice masses that can fill fiordsand extend many kilometres into the ocean, and are definedby Dowdeswell and Jeffries (2017) as: “an ice mass ofconsiderable thickness (� 20 m) that is afloat on the oceanbut attached to the coast. It is often of great horizontal extent(many km) and has, typically but not exclusively, a regularlyundulating surface. An ice shelf can form by the seawardextension of a glacier or glaciers, or by the formation ofmultiyear sea ice, or both and thicken further by the accu-mulation of snow at the top surface and the accretion of icefrom water at the bottom surface”. Although they arecommonly associated with the Antarctic, where they makeup about 75% of the coast, ice shelves also exist in theNorthern Hemisphere where they are primarily concentratedon the northern coastline of Ellesmere Island (Jeffries 2017),some Eurasian fjords that fringe the Arctic Ocean (Dow-deswell 2017) and the floating ice tongues of northernmostGreenland where fast-flowing glaciers reach the ocean (Reeh

2017). Arctic ice shelves are further differentiated from theirSouthern Hemisphere counterparts by their size, with thelargest individual arctic ice shelf, the Ward Hunt Ice Shelf ofNorthern Ellesmere Island, being three orders of magnitudesmaller than the large ice shelves of Antarctica (Dowdeswelland Jeffries 2017).

Evidence based on radiocarbon data indicates that the iceshelves of northern Ellesmere Island developed*5500 years ago in response to Holocene cooling (Englandet al. 2017), which allowed for the advance of glaciersbeyond their current positions and the formation of a largecontiguous ice shelf termed the “Ellesmere Ice Shelf”(Vincent et al. 2001; Dowdeswell and Jeffries 2017). In1906, the “Ellesmere Ice Shelf” had an estimated area of8597 km2, but as of 2015 only smaller remnants were left inindividual fiords, totalling 535 km2 (Mueller et al. 2017).The main remaining pieces comprise the Ward Hunt IceShelf (total 224.3 km2, now split into separate eastern andwestern portions), the Milne Ice Shelf (187.4 km2) and thePetersen Ice Shelf (19.6 km2). Small pieces <10 km2 in arearemain in a few other fiords, but almost all of these havebeen undergoing rapid reductions in extent over the pastdecade (Mueller et al. 2017).

13.4.2 Physical Structure of Arctic Ice Shelves

The ice shelves of Northern Ellesmere Island range inthickness from *20–100 m (Mortimer et al. 2012; Whiteet al. 2015), compared to antarctic ice shelves that canapproach 1 km in ice thickness (Griggs and Bamber 2011).They are characterized by an undulating surface topographypunctuated by troughs and ridges with wavelengths of*30–300 m (Jeffries 2017) and amplitudes up to 7.5 m (Mortimeret al. 2012) (Fig. 13.10). The origin of the rolling topogra-phy of these ice shelves is somewhat unclear, but Holds-worth (1987) and Jeffries (2017) offer a number of possibleexplanations including: thermal stress, sea-ice pack pressure,stacking of sea ice pressure ridges, glacier compression, tidalaction, snow dune formation due to prevailing wind patternsand surface meltwater that pools into elongated lakes due tothe prevailing wind patterns. Although the exact process (orcombination of processes) responsible for these features isunclear, they nevertheless provide a distinctive surfacetopography that aids in distinguishing them from adjacentsea ice, glaciers and icebergs.

Dowdeswell and Jeffries (2017) describe four types of iceshelves based on their method of formation, which providesa useful way to classify these features:

(a) Classical Ice Shelf: this type is commonly associatedwith those found in Antarctica, and is comprised offloating margins of glaciers or an ice sheet, where

13 Ice Masses of the Eastern Canadian Arctic Archipelago 307

glacial ice reaches tidewater and remains intact on theocean surface. The Milne Ice Shelf (Fig. 13.10) ofnorthern Ellesmere Island, which is comprised of anumber of glacier outlets that have coalesced within theconfines of the steep-sided Milne Fiord, provides theclosest approximation of this type of ice shelf in theNorthern Hemisphere. Ground-penetrating radar mea-surements in 2008/9 indicate that it reaches a maximumthickness of *94 m (Mortimer et al. 2012).

(b) Sea-ice Ice Shelf: this describes ice shelves which formdue to the thickening of multi-year sea ice which hasremained in place in protected inlets and embaymentsfor long periods of time (centuries to millennia). Theseice shelves act as a platform for new ice to accumulateon their surface from snowfall, and on their bottomfrom basal freeze-on of sea water, which allows them tothicken substantially over time (Lemmen et al. 1988;Jeffries et al. 1989; Dowdeswell and Jeffries 2017). Bydefinition, they are at least 20 m thick, but can besubstantially thicker than this. The Ward Hunt Ice Shelfis the best current example of this type of ice shelf inthe Northern Hemisphere, and was estimated to be44.5 m thick from a borehole drilled in its western

portion in 1983 (Jeffries 1991). The entire contiguous“Ellesmere Ice Shelf” is thought to have primarily beenof this type when it existed at the start of the 20thcentury.

(c) Composite Ice Shelf: this refers to an ice shelf that iscomprised of both glacier ice and sea ice, sometimesalso with the inclusion of icebergs. The Serson Ice Shelfon northern Ellesmere Island provided a good exampleof this type of ice shelf prior to its almost completebreak-up in 2008 (Lemmen et al. 1988; Dowdeswelland Jeffries 2017; Mueller et al. 2017). This ice shelfwas made up of two main glacier tongues whichextended many kilometres into the ocean, with inter-vening areas infilled with ice that had originated as veryold multi-year landfast sea ice (Dowdeswell and Jeffries2017).

(d) Glacier Tongues: this refers to the floating margins ofglaciers that are narrow relative to their overall lengths,confined by either slow moving ice (typically sea ice) orbedrock on their sides and primarily fed by fast-flowingoutlet glaciers (Dowdeswell and Jeffries 2017; Hambrey1994). These features are common on the northerncoastline of Greenland, and a number are found in the

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

308 W. Van Wychen et al.

protected fjords of northern Ellesmere Island. Muelleret al. (2017) identified eight ice tongues with a total areaof 94 km2 in 1959, compared to 9 ice tongues with acombined area of 23 km2 in 2015. The number oftongues increased over this period, even though theirarea declined by >75%, due to the reclassification ofsome ice shelves to ice tongues as the surrounding iceshelf was lost. The largest remaining feature of this typein the Canadian Arctic is the Milne Glacier Tongue,which comprises the northernmost *16 km of theMilne Glacier (Hamilton et al. 2017).

In terms of mass balance processes, ice shelves gain massthrough surface accumulation of snow and superimposedice, sea ice thickening, water freezing to their base, andglacier input (Mueller et al. 2008; Jeffries 2002; Dowdeswelland Jeffries 2017). The relative importance of each of theseaccumulation processes varies by ice shelf type. Ice shelveslose mass via surface melt and runoff, sublimation, basalmelt at the ice-ocean interface and the calving of icebergs orice islands.

13.4.3 Ice Islands

A unique feature of the ice shelves of northern EllesmereIsland, compared to any other ice mass in the CAA, is thatthey can produce very large ice islands when they calve. Iceislands refer to large tabular icebergs, which typically have afreeboard of at least *2–5 m and can reach >500 km2 inarea (Jeffries 1992; Van Wychen and Copland 2017). Theyare frequently distinguishable by their ribbed surfacetopography, inherited from the ice shelf from which theyformed. The largest recorded ice island calved from northernEllesmere Island in *1950 and was named T-2, and haddimensions of *31 � 33 km (Van Wychen and Copland2017; Jeffries 1992). As these ice islands primarily calvefrom free-floating ice shelves, their ice thicknesses aresimilar to their parent source and can reach values of 20–100 m.

Given the large dimensions of ice islands they can poseparticular threats to shipping and offshore infrastructure,such as oil production platforms, as they circulate withinArctic waters. Van Wychen and Copland (2017) cataloguedhistorical drift tracks of ice islands that originated from thenorthern coastline of Ellesmere Island and found that theseice features undertook three general routes (Fig. 13.11).

(1) Arctic Ocean Drift occurs when an ice island driftswestward after detaching from the coastline andbecomes entrained in the Beaufort Gyre, allowing it tocirculate within the Arctic Ocean. Eventually the iceisland either undertakes Archipelago Drift (described

next), or is evacuated from the Arctic Ocean along theeastern coastline of Greenland via the Transpolar Driftocean current.

(2) Archipelago Drift occurs when an ice island drifts intothe channels between the islands of the Canadian ArcticArchipelago and eventually disintegrates within thesewaterways. For example, Hobson’s Choice Islanddrifted into Queen’s Channel and broke up there in1992 (Jeffries 1992) (Fig. 13.11).

(3) Nares Strait Drift describes the situation when iceislands drift eastward immediately after calving andbecome drawn into Nares Strait, the channel locatedbetween eastern Ellesmere Island and northwestGreenland. These ice islands eventually drift southwardthrough Baffin Bay to the east coast of Newfoundlandand Labrador, where they melt and break-up. Ice islandWH-5 is the only ice island from northern EllesmereIsland known to have made this route, betweenapproximately 1961 and 1964 (Nutt 1966).

Depending on the drift path that an ice island undertakesit can potentially have a significant lifespan before it breaksup. This is especially true for ice islands that undertakeArctic Ocean Drift. For example, Ice Island T-3 circulatedthree times around the Arctic Ocean over a period of*40 years, before eventually exiting the Arctic Oceanthrough Fram Strait and breaking up off southern Greenlandin 1984 (Jeffries 1992; Van Wychen and Copland 2017).Additionally, as ice islands disintegrate they typicallybecome increasingly difficult to track, and may spawn manysmaller ice islands from one large original feature. For thesereasons, ice islands can pose a particular threat to offshoreactivities and infrastructure as a single original ice islandmay threaten the same region multiple times as it breaksapart.

13.4.4 Recent Changes of the NorthernEllesmere Ice Shelves

Between 1906 and 1982, *90% of the ice shelf area thatskirted the northern coastline of Ellesmere Island at the startof the 20th century was lost. Over this time, losses mainlyoccurred during large calving events prior to the 1950s,which released many ice islands into the Arctic Ocean andleft intact ice shelves in protected fiords of northern Elles-mere Island (Copland et al. 2017; Koenig et al. 1952).A second period of large calving events started in *2005,with the remaining ice shelves losing *50% of their 2005area by 2012 (Copland et al. 2017). Of the three largeremaining ice shelves (Petersen, Milne and Ward Hunt), theMilne thinned by an average of 8.1 ± 2.8 m between 1981and 2008/2009 and lost 13% of its 1981 volume (Mortimer

13 Ice Masses of the Eastern Canadian Arctic Archipelago 309

et al. 2012). Point measurements of ice thickness on theWard Hunt Ice Shelf indicate that it thinned from *40–50 m in the 1950s to *25–30 m in the early 2000s, withmost of the thinning taking place since the 1990s (Braun2017). Although repeated ice thickness mapping has notbeen completed on the Petersen Ice Shelf (Fig. 13.12), massbalance measurements indicate that it is losing mass at a rateof 28.45 Mt y−1 at the surface, with little evidence that ice isbeing replaced via basal freeze-on, and will be entirely lostby the 2040s or earlier (White et al. 2015). As such, it isimportant to understand the processes driving the currentloss of these ice shelves.

Copland et al. (2017) identified several factors that havelikely led to the recent loss of the northern Ellesmere IceShelves, including.

(1) Rising Regional Air Temperatures, particularly in theautumn, winter and spring seasons, which have reducedthe overall number of freezing degree days (FDD) thatare important for limiting the impacts of summer meltand for ice shelf recovery via processes such as basal

freeze-on. The decrease in FDD was identified as par-ticularly important leading up to the loss of the AylesIce Shelf in 2007 (Copland et al. 2007). Althoughchanges in summer air temperature have been lessapparent in the region than changes in other seasons,White et al. (2015) showed that losses of the PetersenIce Shelf occurred coincident with record-breakingsummer air temperatures in 2005, 2008, 2011 and 2012.

(2) Reductions in Glacier Input into Ice Shelves: for Clas-sical and Composite ice shelves, there is evidence oflong-term reductions, or the complete loss, of glacierinput to the Ayles, Milne and Petersen ice shelves since*2000 (Mortimer et al. 2012; White et al. 2015;Copland et al. 2017). For example, five tributary gla-ciers provided input to the Milne Ice Shelf in 1959, butthis was reduced to only one by 2011. These reductionsin ice input reduce the mass delivered to these systemsand prime them for break-up and disintegration.

(3) Long-Term Negative Mass Balances: in situ surfacemass balance stakes surveyed on the Ward Hunt IceShelf indicate that winter accumulation rates have

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)

310 W. Van Wychen et al.

remained relatively constant since the 1950s, whilesummer surface melt has been much more variable,resulting in generally negative surface mass balancesover this period (Braun 2017). This trend is similar toall glaciers and ice caps in the Canadian Arctic (seeSect. 13.3), as well as for the other ice shelves withinthe region (Mortimer et al. 2012). Incursions of warmArctic Ocean waters are likely to have promotedincreased basal melt in the 1990s and again from 2005to 2007 (Beszczynska-Moller et al. 2012; Copland et al.2017) which would have further promoted mass lossfrom these ice shelves.

(4) Changes in Arctic Sea Ice: episodic losses of iceshelves have tended to occur when the northern coast-line of Ellesmere Island is free of sea ice, suggestingthat sea ice plays an important role in buttressing iceshelves and aiding their stability and long-term survival(Copland et al. 2017).

Based on the above review, and that of Copland et al.(2017), there is not a single factor that can be attributed asthe mechanism driving recent ice shelf losses. Rather, it islikely the combination of all four of these factors, and per-haps others, that has led to the loss of ice shelves in thenorthern CAA and the reductions in mass of the remainingintact ice shelves there. Taken together, the rapid loss of iceshelf extent since 2005, indications of continued mass loss

from those that remain, and projections of continued climatewarming and sea ice loss in the Arctic, implies that thelong-term survival of these features is doubtful.

13.5 Conclusions

Knowledge of the mass balance and characteristics of gla-ciers and ice masses of the eastern Canadian Arctic Islandshas been primarily based on in situ observations initiated inthe 1960s. Since *2000, the increased availability and res-olution of remote sensing imagery and methodologies toprocesses these data have helped to further expand our col-lective understanding of these ice masses on a regional basis.All of these observations converge to reveal a rapidlychanging landscape. In terms of glacier dynamics, regionalvelocity mapping has shown that dynamic variability withinthe region is now much more widespread than previouslythought. Glacier mass balances indicate that snowfall withinthe region has remained relatively steady since the 1960s, butthat surface melt and runoff have increased substantially,beginning in the 1980s and especially since 2005. Addi-tionally, nearly all models project that the Canadian arcticglaciers will continue to lose mass and be a major contributorto sea level rise over the next century. Within this context, theice shelves of Northern Ellesmere Island also continue to losemass, punctuated by the episodic detachment of large pieces

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

13 Ice Masses of the Eastern Canadian Arctic Archipelago 311

of ice shelf and the creation of ice islands. In light of this,continued monitoring of the ice features of the eastern CAAis essential for furthering our understanding of how theregion will continue to respond to climate change.

References

Abdalati W, Krabill W, Frederick E, Manizade S, Martin C, Sonntag J,Swift R, Thomas R, Yungel J, Koerner R (2004) Elevation changesof ice caps in the Canadian Arctic Archipelago. J Geophys Res 109.https://doi.org/10.1029/2003jf000045

Beszczynska-Moller A, Fahrbach E, Schauer U, Hansen E (2012)Variability in Atlantic water temperature and transport at theentrance to the Arctic Ocean, 1997–2010. ICES J Mar Sci 69:852–863

Bingham R, Nienow P, Sharp M (2003) Intra-annual and intra-seasonalflow dynamics of a High Arctic polythermal valley glacier. AnnGlaciol 37:181–188

Bingham RG, Nienow PW, Sharp MJ, Copland L (2006) Hydrologyand dynamics of a polythermal (mostly cold) High Arctic glacier.Earth Surf Processes 31:1463–1479

Bingham RG, Hubbard AL, Nienow PW, Sharp MJ (2008) Aninvestigation into the mechanisms controlling seasonal speed-upevents at a High Arctic glacier. J Geophys Res 113. https://doi.org/10.1029/2007/jf000832

Böning CW, Behrens E, Biastoch A, Getzlaff K, Bamber JL (2016)Emerging impact of Greenland meltwater on deepwater formationin the North Atlantic Ocean. Nat Geosci 9:523–527

Braun C (2017) The surface mass balance of the Ward Hunt Ice Shelfand Ward Hunt Ice Rise, Ellesmere Island, Nunavut, Canada. In:Copland L, Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch.6. Springer Nature, Dordrecht

Briner JP, Davis PT, Miler GH (2009) Latest Pleistocene and Holocenglaciation of Baffin Island, Arctic Canada: key patterns andchronologies. Quat Sci Rev 28:21–22

Burgess D, Sharp M, Mair D, Dowdeswell JA, Benham TJ (2005) Flowdynamics and iceberg calving rates of Devon Ice Cap, Nunavut,Canada. J Glaciol 51:219–230

Copland L, Sharp M, Dowdeswell JA (2003a) The distribution andflow characteristics of surge-type glaciers in the Canadian HighArctic. Ann Glaciol 36:73–81

Copland L, Sharp M, Nienow P (2003b) Links between short-termvelocity variations and the subglacial hydrology of a predominantlycold polythermal glacier. J Glaciol 49:337–348

Copland L, Sharp M, Nienow P, Bingham R (2003c) The distributionof basal motion beneath a High Arctic polythermal glacier. J Glaciol49:407–414

Copland L, Mueller D, Weir L (2007) Rapid loss of the Ayles Ice Shelf,Ellesmere Island, Canada. Geophys Res Lett 34. https://doi.org/10.1029/2007gl031809

Copland L (2013) Classification of ice masses. In: Shroder JF(ed) Treatise on Geomorphology, Ch. 8.4. Elsevier, Oxford,pp 45–52

Copland L, Mortimer C, White A, McCallum MR, Mueller D (2017)Factors contributing to recent Arctic ice shelf losses. In: Copland L,Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch. 10. SpringerNature, Dordrecht

Cress P, Wyness R (1961) The Devon Island expedition, observationsof glacial movements. Arctic 14:247–259

Dowdeswell JA, Benham TJ, Gorman MR, Burgess D, Sharp M (2004)Form and flow of the Devon Island ice cap, Canadian Arctic.J Geophys Res 109. https://doi.org/10.1029/2003jf000095

Dowdeswell EK, Dowdeswell JA, Cawkwell F (2007) On the glaciersof Bylot Island, Nunavut, Arctic Canada, Arctic, Antarctic andAlpine Research, 39(3):402–411. https://doi.org/10.1657/1523-0430(05-123)%5bDOWDESWELL%5d2.0.CO;2

Dowdeswell JA (2017) Eurasian Arctic Ice Shelves and Tidewater IceMargins. In: Copland L, Mueller D (eds) Arctic Ice Shelves and IceIslands, Ch. 3. Springer Nature, Dordrecht

Dowdeswell JA, Jeffries MO (2017) Arctic Ice Shelves: an introduc-tion. In: Copland L, Mueller D (eds) Arctic Ice Shelves and IceIslands, Ch. 1. Springer Nature, Dordrecht

Dupont F, Royer A, Langlois A, Gressent A, Picard G, Fily M, ClichéP, Chum M (2012) Monitoring the melt season length of the BarnesIce Cap over the 1979–2010 period using active and passivemicrowave remote sensing data. Hydrol Processes 26:2643–2652

England JH, Evans DA, Lakeman TR (2017) Holocene history ofArctic Ice Shelves. In: Copland L, Mueller D (eds) Arctic IceShelves and Ice Islands, Ch. 7. Springer Nature, Dordrecht

Fisher D, Zheng J, Burgess D, Zdanowicz C, Kinnard C, Sharp M,Bourgeois J (2012) Recent melt rates of Canadian Arctic ice capsare the highest in many millennia. Global Planet Change 84–85.https://doi.org/10.1016/j.gloplacha.2011.06.005

Gardner AS, Moholdt G, Wouters B, Wolken GJ, Burgess DO,Sharp M, Cogley JG, Braun C, Labine C (2011) Sharply increasedmass loss from glaciers and ice caps in the Canadian ArcticArchipelago. Nature 473:357–360

Gardner AS, Moholdt G, Cogley JG, Wouters B, Arendt AA, Wahr J,Berthier E, Hock R, Pfeffer WT, Kaser G, Ligtenberg SR, Bolch T,Sharp MJ, Hagen J-O, van den Broeke M, Paul F (2013) Areconciled estimate of glacier contributions to sea level rise 2003–2009. Science 340:852–857

Griggs JA, Bamber JL (2011) Antarctic ice-shelf thickness fromsatellite radar altimetry. J Glaciol 57:485–498

Hambrey MJ (1994) Glacial environments. UCL Press, LondonHamilton AK, Laval BE, Mueller DR, Vincent WF, Copland L (2017)

Dynamic response of an Arctic epishelf lake to seasonal andlong-term forcing: implications for ice shelf thickness. TheCryosphere 11:2189–2211

Harig C, Simons FJ (2016) Ice mass loss in Greenland, the Gulf ofAlaska, and the Canadian Archipelago: seasonal cycles and decadaltrends. Geophys Res Lett 43:3150–3159

Hattersley-Smith G (1969) Recent observations on the surging OttoGlaicer, Ellesmere Island. Can J Earth Sci 6:883–889

Heid T, Kääb A (2012) Repeat optical satellite images revealwidespread and long term decrease in land-terminating glacierspeeds. The Cryosphere 6:467–478

Holdsworth G (1973) Evidence of a surge on Barnes Ice Cap, BaffinIsland. Can J Earth Sci 10:1565–1574

Holdsworth G (1977) Surge activity on the Barnes Ice Cap. Nature269:588–590

Holdsworth G (1987) The surface waveforms on the Ellesmere Islandice shelves and ice islands. In: Workshop on extreme ice features,technical memorandum, 141 (NRCC 28003). National ResearchCouncil of Canada, pp 385–443

Jacob T, Wahr J, Pfeffer T, Swenson S (2012) Recent contributions ofglaciers and ice caps to sea level rise. Nature 482:514–518

Jeffries MO, Krouse HR, Sackinger WM, Serson HV (1989)Stable-isotope (18O/19O) tracing of fresh, brackish and sea ice inmultiyear land-fast sea ice, Ellesmere Island, Canada. J Glaciol35:9–16

Jeffries MO (1991) Massive, ancient sea-ice strata and preservedphysical-structural characteristics in the Ward Hunt Ice Shelf. AnnGlaciol 15:125–131

Jeffries MO (1992) Arctic ice shelves and ice islands: origin, growthand disintegration physical characteristics, structural-stratigraphicvariability, and dynamics. Rev Geophys 30:245–267

312 W. Van Wychen et al.

Jeffries MO (2002) Ellesmere Island Ice Shelves and Ice Islands, inSatellite Image Atlas of Glaciers of the world. In: Williams RS,Ferrigno JG (eds) Glaciers of North America. US GeologicalSurvey, Washington DC, pp J147–J164

Jeffries MO (2017) The Ellesmere Ice Shelves, Nunavut, Canada. In:Copland L, Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch.2. Springer Nature, Dordrecht

Koenig LS, Greenaway KR, Dunbar M, Hattersley-Smith G (1952)Arctic Ice Islands. Arctic 5:67–103

Koerner RM (1979) Accumulation, ablation, and oxygen isotopevariations on the Queen Elizabeth Islands ice caps, Canada.J Glaciol 22:25–41

Koerner RM (2005) Mass balance of glaciers in the Queen ElizabethIslands, Nunavut, Canada. Ann Glaciol 42:417–423

Lemmen DS, Evans DJ, England J (1988) Ice Shelves of NorthernEllesmere Island, NWT, Canadian landform examples. Can Geogr32:363–367

Lenaerts JT, van Angelen JH, van den Broeke MR, Gardner AS,Wouters B, van Meijgaard E (2013) Irreversible mass loss ofCanadian Arctic Archipelago glaciers. Geophys Res Lett 40:870–874

Leuschen C, Goineni P, Rodriguez-Morales F, Paden J, Allen C (2017)IceBrdige MCords L2 Ice Thickness, Version 1. Boulder,Colorado USA. NASA National Snow and Ice Data CenterDistributed Active Archive Center. Acquired 1 April 2014

Løken OH (1969) Evidence of surges on the Barnes Ice Cap, BaffinIsland. Can J Earth Sci 6:899–901

Marzeion B, Jarosch AH, Hofer M (2012) Past and future sea-levelchange from the surface mass balance of glaciers. The Cryosphere6:1295–1322

Millan R, Mouginot J, Rignot E (2017) Mass budget of the glaciers andice caps of the Queen Elizabeth Islands, Canada, from 1991 to2015. Environ Res Lett 12. https://doi.org/10.1088/1748-9326/aa5b04

Mortimer CA, Copland L, Mueller DR (2012) Volume and area changesof the Milne Ice Shelf, Ellesmere Island, Nunavut, Canada, since1950. J Geophys Res 117. https://doi.org/10.1029/2011jf002074

Mortimer CA, Sharp M, Wouters B (2016) Glacier surface tempera-tures in the Canadian High Arctic, 2000–2015. J Glaciol 62:963–975

Mueller D, Copland L, Hamilton A, Stern D (2008) Examining Arcticice shelves prior to the 2008 breakup. EOS, Transactions, AmericanGeophysical Union 89(49)

Mueller D, Copland L, Jeffries MO (2017) Changes in Canadian Arcticice shelf extent since 1906. In: Copland L, Mueller D (eds) ArcticIce Shelves and Ice Islands, Ch. 5. Springer Nature, Dordrecht

Müller F (1969) Was the Good Friday Glacier on Axel Heiberg Isladsurging? Can J Earth Sci 6:891–894

Müller F, Iken A (1973) Velocity fluctuations and water regime of Arcticvalley glaciers. In: Symposium on the hydrology of glaciers,Cambridge, de l’Association Internationale d’Hydrolgie Scientifique

Nutt DC (1966) The drift of Ice Island WH-5. Arctic 19:244–262Radić V, Bliss A, Beedlow AC, Hock R, Miles E, Cogley JG (2014)

Regional and global projections of twenty-first century glacier masschanges in response to climate scenarios from global climatemodels. Clim Dyn 42:37–58

Reeh N (2017) Greenland Ice Shelves and Ice Tongues. In: Copland L,Mueller D (eds) Arctic Ice Shelves and Ice Islands, Ch. 4. SpringerNature, Dordrecht

Schaffer N, Copland L, Zdanowicz C (2017) Ice velocity changes ofPenny Ice Cap, Baffin Island, since the 1950s. J Glaciol 63:716–730

Sharp M, Burgess DO, Cogley JG, Ecclestone M, Labine C, Wolken G(2011) Extreme melt on Canada’s Arctic ice caps in the 21st century.Geophys Res Lett 38. https://doi.org/10.1029/2011gl047381

Sharp M, Burgess DO, Cawkwell F, Copland L, Davis JA,Dowdeswell EK, Dowdeswell JA, Gardner A, Mair D,Wang L, Williamson S, Wolken GJ, Wyatt F (2014) Remotesensing of recent glacier changes in the Canadian Arctic. In:Kargel JS, Leonard GJ, Bishop MP, Kääb A, Raup BH(eds) Global land ice measurements from space, Ch. 9.Praxis-Springer, pp 205–228

Shepherd A., Ivins ER, Geruo A, Barletta VR, Bentley MJ, Battadpur S,Briggs KH, Bromwich DH, Forsberg R, Galin N, Horwath M,Jacobs S, Joughin I, King MA, Lenaerts JTM, Li J, Ligten-berg SRM, Luckman A, Luthcke SB, McMillan M, Meiser R,Milne G, Mouginot J, Muir A, Nicolas JP, Paden J, Payne AJ,Pritchard H, Rignot E, Rott H, Sorensen LS, Scambos TA,Scheuchl B, Schrama EJO, Smith B, Sundal AV, van Angelen JH,van de Berg WJ, van den Broeke MR, Vaughen DG, Velicogna I,Wahr J, Whitehouse PL, Wingham DJ, Yi D, Young D, Zwally JH(2012) A reconciled estimate of Ice-Sheet mass balance. Sci 388(6111):1183–1189. https://www.doi.org/10.1126/science.1228102

Short NH, Gray AL (2005) Glacier dynamics in the Canadian HighArctic from RADARSAT-1 speckle tracking. Can J Remote Sens31:225–239

Strozzi T, Paul F, Wiesmann A, Schellenberger T, Kääb A (2017)Circum-Arctic changes in the flow of glaciers and ice caps fromsatellite SAR data between 1990s and 2017. Remote Sens 9. https://doi.org/10.3390/rs909047

Thomson L, Copland L (2017) Multi-decadal reduction in glaciervelocities and mechanisms driving deceleration at polythermalWhite Glacier, Arctic Canada. J Glaciol 63:450–463

Thomson L, Copland L (2018) Changing contribution of peak velocityevents to annual velocities following a multi-decadal slowdown atWhite Glacier. Ann Glaciol 58:145–154

Van Wychen W, Copland L, Gray L, Burgess DO, Danielson B,Sharp M (2012) Spatial and temporal variation of ice motion and iceflux from Devon Ice Cap, Nunavut, Canada. J Glaciol 58:657–664

Van Wychen W, Burgess DO, Gray L, Copland L, Sharp M,Dowdeswell JA, Benham TJ (2014) Glacier velocities and dynamicice discharge from the Queen Elizabeth Islands, Nunavut, Canada.Geophys Res Lett 41:484–490

Van Wychen W, Copland L, Burgess D, Gray L, Schaffer N (2015)Glacier velocities and dynamic discharge from the ice masses ofBaffin Island and Bylot Island, Nunavut, Canada. Can J Earth Sci52:980–989

Van Wychen W, Davis J, Burgess DO, Copland L, Gray L, Sharp M,Mortimer C (2016) Characterizing interannual variability of glacierdynamics and dynamic discharge (1999–2015) for the ice masses ofEllesmere and Axel Heiberg Islands, Nunavut, Canada. J GeophysRes-Earth Surf 121:39–63

Van Wychen W, Davis J, Copland L, Burgess D, Gray L, Sharp M,Dowdeswell JA, Benham T (2017) Variability in ice motion anddynamic discharge from Devon Ice Cap, Nunavut, Canada. J Glaciol63:436–449

Van Wychen W, Copland L (2017) Ice island drift mechanisms in theCanadian High Arctic, Ch. 7. In: Copland L, Mueller D (eds) ArcticIce Shelves and Ice Islands. Springer-Verlag

Vincent WF, Gibson J, Jeffries MO (2001) Ice shelf collapse, climatechange, and habitat loss in the Canadian high Arctic. Polar Record37:133–142

White A, Copland L, Mueller D, Van Wychen W (2015) Assessment ofhistorical changes (1959–2012) and the causes of recent break-upsof the Petersen ice shelf, Nunavut, Canada. Ann Glaciol 56:65–76

Williamson S, Sharp M, Dowdeswell J, Benham T (2008) Icebergcalving rates from northern Ellesmere Island ice caps. CanadianArctic, 1999–2003. J Glaciol 54:391–4000

13 Ice Masses of the Eastern Canadian Arctic Archipelago 313

Wolken G, Sharp M, Wang L (2009) Snow and ice facies variabilityand ice layer formation on Canadian Arctic ice caps, 1999–2005.J Geophys Res 114. https://doi.org/10.1029/2008jf001173

Zdanowicz C, Smetny-Sowa A, Fisher D, Schaffer N, Copland L, Eley,Dupont F (2012) Summer melt rates on Penny Ice Cap, BaffinIsland: Past and recent trends and implications for regional climate.J Geophys Res 117. https://doi.org/10.1029/2011jf002248

Wesley Van Wychen is the Associate Director of the Canadian CryosphericInformation Network and Polar Data Catalogue at the University ofWaterloo where his work focuses on ensuring the long-term integrity ofArctic and Antarctic datasets. His research focuses on the development anduse of novel remote sensing techniques for understanding Arctic environ-mental change at the regional scale. He also has a strong interest inunderstanding glacier mass balance and the evolution of glacier dynamics inresponse to climate change.

Luke Copland is a Professor in the Department of Geography, Environ-ment and Geomatics at the University of Ottawa, where he holds theUniversity Research Chair in Glaciology. He is also Director of the Labo-ratory for Cryospheric Research (https://cryospheric.org), and works withhis graduate students on projects across northern Canada to investigateglacier dynamics and the response of ice masses to climate change.

David Burgess is a Research Scientist at the Geological Survey of Canadawhere he specializes in the use of remote sensing technology and in-situmethods to study the mass balance and dynamics of ice caps in the CanadianArctic. He leads the Government of Canada’s National Glaciology Pro-gramme through which he conducts integrated research and assessments ofglacier-climate change across Canada’s Arctic and Alpine regions. Dr.Burgess is a lead investigator and coordinator for Canada’s involvement incalibration and validation of the CryoSat-2 radar altimeter (European SpaceAgency) over Canada’s Arctic ice caps.

314 W. Van Wychen et al.