Exposure to coastal hazards in a rapidly expanding northern urban centre, Iqaluit, Nunavut. (2024)

ABSTRACT. The City of Iqaluit, Nunavut, is an expanding urbancentre with important infrastructure located in the coastal zone. Thisstudy investigates the exposure of this infrastructure to coastalhazards (rising mean sea level, extreme water levels, wave run-up, andsea ice). Using a coastal digital elevation model, we evaluate theinundation and flooding that may result from projected sea level rise.Some public and private infrastructure is already subject to floodingduring extreme high water events. Using a near upper-limit scenario of0.7 m for relative sea level rise from 2010 to 2100, we estimate thatcritical infrastructure will have a remaining freeboard of 0.3-0.8 mabove high spring tide, and some subsistence infrastructure will beinundated. The large tidal range, limited over-water fetch, and wideintertidal flats reduce the risk of wave impacts. When present, theshorefast ice foot provides protection for coastal infrastructure. Theice-free season has expanded by 1.0-1.5 days per year since 1979,increasing the opportunity for storm-wave generation and thus exposureto wave run-up. Overtopping of critical infrastructure and displacementby flooding of subsistence infrastructure are potential issues requiringbetter projections of relative sea level change and extreme high waterlevels. These results can inform decisions on adaptation, providingmeasurable limits for safe development.

Key words: Arctic coast; adaptation planning; infrastructure; sealevel rise; flooding; sea ice; climate change; coastal management

RESUME. La ville d'lqaluit, au Nunavut, est un centre urbainen plein essor dote d'infrastructures importantes sur la zonecotiere. Cette etude se penche sur l'exposition de cetteinfrastructure aux risques cotiers (niveau de la mer montant, niveauxd'eau extremes, vagues et glace de mer). A l'aide d'unmodele numerique de Pelevation cotiere, nous evaluons les inondations etles submersions susceptibles de decouler de la montee projetee du niveaude la mer. Certaines infrastructures publiques et privees sont deja lacible d'inondations en presence de tres hautes eaux. En nousappuyant sur un scenario dont la limite superieure est de pres de 0,7 mpour la hausse relative du niveau de la mer de 2010 a 2100, nousestimons que les infrastructures critiques auront un franc bord de 0,3 a0,8 m au-dessus de la maree haute de vives-eaux, et une partie desinfrastructures de subsistance sera inondee. La grande amplitude de lamaree, le fetch limite sur l'eau et les larges batturesintertidales reduisent le risque de Timpact des vagues. Lorsqu'elleest presente, la glace de rive offre une protection aux infrastructurescoheres. Depuis 1979, la saison sans glace s'est prolongee de 1,0 a1,5 jour par annee, ce qui augmente la possibility de la formation devagues de tempete et, par consequent, Texposition aux jets de rive. Lasubmersion des infrastructures critiques et le deplacement desinfrastructures de subsistance par les inondations constituent desenjeux potentiels qui doivent faire l'objet de meilleuresprojections du changement relatif du niveau de la mer et des niveauxd'eau extremes. Ces resultats pourront eclairer les decisions enmatiere d'adaptation, ce qui permettra d'obtenir des limitesmesurables en vue d'amenagements securitaires.

Mots cles : cote de l'Arctique; planification del'adaptation; infrastructure; montee du niveau de. la mer;inondation; glace de mer; changement climatique; gestion cotiere

Traduit pour la revue Arctic par Nicole Giguere.

INTRODUCTION

Recent rapid changes within the Arctic climate system, such asrising temperatures and increased storm waves during open water, haveexacted heavy tolls on the infrastructure of some Arctic coastalcommunities (Arehart, 2012). The effects of polar climate amplificationmean that parts of the Arctic are warming at higher rates than otherregions of the globe (Serreze and Barry, 2011). NRTEE (2009) projectedsignificant monetary costs for the replacement and maintenance of agingphysical infrastructure at risk from climate change in northern Canada.Billions of dollars will be invested in new infrastructure in the comingdecades, highlighting the need for appropriate adaptation strategies.Larsen et al. (2008) calculated an additional $5.6-$7.6 billion would berequired, in excess of regular maintenance investment, to repair Alaskaninfrastructure if projected climate change persists to 2030. Notaccounting for climate change, Iqaluit's infrastructure deficit isestimated at $40 million, and there is growing demand for newinfrastructure to meet the needs of a rapidly expanding urban centre(Forbes et al., 2012). The environment, isolation, and transportationlogistics of the Arctic raise costs, making infrastructure expensive tobuild and maintain (Forbes, 2011:34). Clearly, an understanding of howprojected environmental change will affect infrastructure is needed toimprove design, minimize risk, and develop sustainable northerncommunities.

Since most Canadian Inuit communities are on the coast, so is alarge proportion of Arctic infrastructure. Atmospheric warming hasalready led to coastal changes, such as increased thermal abrasion andcoastal erosion (Are et al., 2008; Forbes, 2011). The threat to coastalinfrastructure in the Arctic from changing coastal dynamics is only oneamong many: others include thaw subsidence, wind, increased intenseprecipitation, or impeded drainage (e.g., Forbes et al., 2014; Smith,2014; Smith and Forbes, 2014). In some places, potential impacts havealready emerged as hazards, leading to relocation or retreat (Catto andParewick, 2008). These challenges are exacerbated by sparse data overshort time series, which inhibit our ability to predict future hazardconditions (NRTEE, 2009; Strzelecki, 2011). There is pressure to adaptto change and protect key infrastructure. Decisions are made on thebasis of available knowledge, including scientific research (Ford etal., 2010; Forbes, 2011). Appropriate responses depend on the nature ofthe hazard and the infrastructure at risk.

Iqaluit, the capital city of Nunavut, is contending with naturalhazards from exposure on many fronts. Thaw subsidence in permafrost hasdamaged city infrastructure (Nielsen, 2007), food networks of thecommunity are strained by environmental change and an expandingpopulation (Lardeau et al., 2011), and occasional coastal flooding hasoccurred in the past (Fig. 1). These issues are exacerbated by climatechange (Forbes et al., 2012). Adaptation planning is ongoing; it ismandated at both territorial and municipal levels (City of Iqaluit,2010) and incorporated into the city's Sustainability Plan (City ofIqaluit, 2014). Recent rapid population growth complicates this effort,as the existing infrastructure deficit creates an added burden forinvestment in solutions. In this context, previous work has identifiedhazards at the coast, including sea level change, extreme water levels,and changing sea ice patterns, as a topic requiring furtherinvestigation to better define the associated exposure and risk(Shirley, 2005; Nielsen, 2007; City of Iqaluit, 2010; Hatcher et al.,2011).

Marine flooding in Iqaluit was reported in 2003 (Fig. 1).Photographs indicate that this flooding happened in calm conditions withno storm influence, leading to questions about its cause. Water-levelrecords indicate that higher flooding occurred during an extreme eventin 1964, but there was little or no damage at that time because thecurrent urban development along the shore did not yet exist (Fig. 2).The public therefore has little awareness of hazard events that couldendanger the extensive residential, commercial, public, and subsistenceinfrastructure put in place over the last three decades. However,archival water-level data provide some insight into the probability offlood recurrence. We may also ask whether the 1964 and 2003 floodingevents resulted from unusually high tide and, if so, what theimplications would be of coincident storm conditions, or occurrenceduring freeze-up or breakup of the coastal sea ice.

In this paper, we examine the natural hazards associated with thecoastal setting of Iqaluit under present and future conditions. Hazardsconsidered include storm waves, sea ice ride-up and pile-up, and marineflooding associated with storm surges and extreme high tides. We use thelatest projections of local sea level rise, incorporating results fromthe Intergovernmental Panel on Climate Change (IPCC) Fifth AssessmentReport (Church et al., 2013; IPCC, 2013), as well as measured crustalmotion at Iqaluit and the gravitational effects on sea level ofproximity to the Greenland Ice Sheet and local glaciers and ice caps onBaffin Island (James et al., 2014), to consider infrastructure elevationand exposure to flooding with respect to mean and extreme water levelsnow and in the future. Local projections of sea ice concentration, stormwinds, and waves in Frobisher Bay are beyond the scope of the study, andchanges in storm-wave climate are considered only in the context ofrecent regional trends in the length of the open water season.

As part of an international project on responding to environmentalchange in coastal communities (Lane et al., 2013), this study wasinitiated to address the knowledge gap on coastal hazards. The secondauthor had long-term experience in the community, and consultations withlocal, territorial, and federal agencies preceded the study. Thesecontacts included City of Iqaluit planning staff, the Amarok Fluntersand Trappers Association, the Nunavut Research Institute, the Governmentof Nunavut Department of Environment, the Canada-Nunavut GeoscienceOffice, and individual residents. Since the study was designed to informthe sustainability planning process and the next revision of the generalplan, we collaborated closely with the Director of Engineering andSustainability and the Sustainability Coordinator (Forbes et al., 2012).

The study objectives were to (1) identify natural hazards thatpresent a risk to coastal infrastructure in Iqaluit, (2) quantify thewaterfront exposure in the context of observed trends and sea levelprojections, and (3) identify coastal infrastructure at risk in Iqaluit.We derived the data needed to address these goals from a number ofsources, including archival climate and water-level data, anecdotalinformation, conversations with city staff and other residents, mooredinstrument data, and field surveys (Hatcher et al., 2014).

STUDY AREA

Environmental Setting

Iqaluit sits at the head of Koojesse Inlet (63.7[degrees] N,68.5[degrees]W) in the northwest corner of Frobisher Bay on BaffinIsland (Fig. 3). The study area encompasses the full shoreline of theinlet between Inuit Head in the southwest and the old settlement of Apex(now the eastern suburb of Iqaluit) in the east (Fig. 2). The workfocused on hazards to coastal infrastructure along three stretches ofcoastline: the Iqaluit waterfront, the old cemetery, and Apex beach(Fig. 2). The landscape is a product of glacial erosion, which formed anumber of rock ridges trending from northwest to southeast with thintill or shallow marine deposits in the intervening depressions (Hodgson,2005; Allard et al., 2012). The rock is granitic and resistant toerosion. There are no trees, and many of the rock ridges areunvegetated. Permafrost (defined as ground at a temperature <0[degrees]C for two years or more) is ubiquitous above the high-tideline, and excess ground ice is present near the surface in many places,producing distinctive small-scale landforms and leading to thawsubsidence where the near-surface thermal regime is disturbed byconstruction or other human activities (Short et al., 2012).

The inlet is macrotidal, with a semi-diurnal tide and spring tidalrange of 12.4 m (CHS, 2001). The shore is bare rock in many places, withhigh-tide beaches of mixed sand and gravel at Apex and along much of thedowntown waterfront. Very extensive boulder-strewn tidal flats form awide intertidal zone in most of the study area (Fig. 4A). Similar tidalflats with innumerable boulders on the surface are found along much ofthe northern coast of Frobisher Bay. At the seaward limit of the flats,the seabed falls off into the deeper waters of the harbour.

Wave action is limited in Koojesse Inlet by a number of factors.The inlet opens to the southeast, and Long Island sits at the entrance,providing some shelter from incident waves (Fig. 3). The islands thatseparate inner and outer Frobisher Bay lie roughly 54 km (straight linedistance) from the mouth of the inlet (Fig. 3) and block all oceanswell. Therefore, the wave field is locally forced and limited by themaximum over-water fetch (less if ice is present). Storms capable ofproducing waves that can affect the coast are restricted to a narrowsouth-east fetch exposure. They occur predominantly in fall, whenextra-tropical cyclones passing through the Labrador Sea and towardBaffin Bay tend to move westward over southern Baffin Island, bringingwarmer air masses north and producing precipitation (Maxwell, 1981;Hatcher, 2014).

Coastal retreat is minimal because much of the shore consists ofresistant bedrock, and at least until very recently, the site has beenemergent (falling relative sea level) as a result of glacial-isostaticuplift exceeding sea level rise. The rate of downcutting on the tidalflats is unclear. Sediment movement is dominated by sea ice dynamics(McCann et ah, 1981; McCann and Dale, 1986; Leech, 1998; Dale et ah,2002). Ice prevails for an average of nine months of the year. Duringthe ice season, thick intertidal ice is repeatedly lifted and droppedonto the flats by the cycle of tides. Sediment is entrained throughbasal adfreezing, but the ice does not move offshore. This patternoccurs because of preferential thawing of sediment-laden ice in thespring and retention of melting intertidal ice over the tidal flats bythe solid landfast ice in deeper water (McCann and Dale, 1986).

We are aware of no previous trend analysis of seasonal sea iceduration in this region. Previous work on sea ice in the area hasfocused exclusively on the dynamics of breakup, which is a criticalannual event for deposition of rafted sediment (McCann and Dale, 1986;Leech, 1998); however, we know less about the process of freezeup.Community members describe it as a period of change when the sea ice"sets up" on the coast. During this time, access is extremelydifficult because the ice is thin and constantly shifting in theintertidal zone over the tidal flats.

Part of the process of freeze-up involves forming the ice foot, anice accumulation near the high-tide line, where it is frozen to thesubstrate for the winter. Subsequent inundation during high spring tidesbuilds thickness further and contributes to the development of a flatice terrace (Forbes and Hansom, 2011). The edge of the ice foot where itmeets with mobile intertidal ice acts as a hinge point and is a locus ofdiscontinuous sea ice ride-up and pile-up (Fig. 5A).

Though the coast was formed during a long period of fallingrelative sea level, the current trend at Iqaluit is unclear. The sitehas probably been very slowly emergent in recent decades. The crust inthis area is still undergoing postglacial rebound, with an uplift rateof 3.97 [+ or -] 0.65 mm/yr (about 40 cm per century), as indicated by4.3 years of continuous GPS measurement (James et al., 2014). Thisuplift is at least partially offset by local sea level rise, but therate of rise is moderated by the gravitational effects of proximity tothe Greenland Ice Sheet and ice masses on Baffin Island. On the otherhand, the effects of ice mass loss in Antarctica will be slightlyenhanced in this region (Mitrovica et al., 2001). Relative sea level isknown to have risen in recent years in outer Frobisher Bay (farthereast), as indicated by flooded habitations of the ancestral Inuit Thuleculture (M.E. Thomas, pers. comm. 2009) and geological evidence (Milleret al., 1980).

Urban Development

The present City of Iqaluit began in the mid-20th century as ahybrid settlement around the United States Strategic Air Command base atthe head of the inlet (Fig. 6A). Inuit would seasonally occupy the beachin order to take advantage of both employment at the base and goodfishing in the inlet (Eno, 2003). Iqaluit is an Inuktitut word thattranslates to 'place of many fish.' The airbase acted as anucleus of development, but infrastructure expanded to the shoreline inorder to support the landing of supplies arriving by ship (Figs. 2, 6B).As development on the eastern side of Iqaluit grew to the coast, thehamlet of Apex developed 4 km to the east, connected by a road to thecore of Iqaluit (Fig. 3). The entire built-up area now falls within thecity boundary.

The City of Iqaluit is home to about 7000 people, and thepopulation has been growing for many years, particularly since becomingthe capital of Nunavut in 1999. The rate of growth from 2006 to 2011 was8.3% (Statistics Canada, 2014). The large commercial and institutionalbuildings along the waterfront have all been built since 1970. Amidstthis government and private sector infrastructure in the backshore,traditional activities and a subsistence economy continue, resulting ina proliferation of small wooden sheds and repurposed shipping containersdirectly adjacent to the high-tide line.

The developed waterfront areas in Iqaluit and Apex were divided forplanning purposes into two zones according to specifications in theIqaluit General Plan (City of Iqaluit, 2010). They are defined byhorizontal setback distances of 30.5 m and ~75 m from the high-tideline. The first is designated "Open Space" and the second is arough delineation of the coastal planning zone (designated the SijjaangaDistrict in the General Plan, sijjaanga being Inuktitut for beach areaor waterfront). This zone, while intended to restrict commercial andinstitutional development, includes major commercial and transportationinfrastructure. In this study, to quantify the flood hazard, thelandward limit of the Sijjaanga District, which is not formally mapped,was set at 75 m inland from the high-tide line. This limit encloses theinfrastructure mentioned in the General Plan as belonging within theSijjaanga District.

The physical character of the coast can be subdivided into twotypes: beaches and bedrock outcrop. Waterfront development has occurredalmost exclusively on the low-slope beaches and emerged relict beachesbetween rock headlands (Fig. 4B). The main waterfront of the city is alarge bay-head beach. Confined to the north side of the inlet, thiswaterfront is now fully backed by urban infrastructure. To the east atApex, the beach is short (less than 500 m) and located adjacent to theoutlet of the Apex River. At this site, there is one residence amid acluster of heritage Hudson's Bay Company buildings, which date from1949.

Key infrastructure facilities, including municipal utilitybuildings and residential structures, are located in the backshore ofIqaluit's main beach. The subsistence support structures (the mostabundant coastal infrastructure in both number and extent) sit on top ofthe beach crest along the entire length of the waterfront. Most of thesestructures are close to the spring high-tide line, between the city andthe sea. Subsistence infrastructure is reported to have been flooded in2003 (Shirley, 2005).

Figure 7 shows the major components of infrastructure within thecoastal zone of the three primary study areas: the Iqaluit urbanwaterfront, cemetery beach, and Apex beach. If we exclude the fueltransfer facility, causeway, and dump across the harbour, the urbanwaterfront of Iqaluit begins at the head of the inlet, where a riverflows onto the tidal flats just east of the sewage lagoon. The latter isretained by two dams. East of this lagoon is the sealift barge-landingfacility and the Canadian Coast Guard property. Farther along are a boatyard, housing, and subsistence infrastructure on the west side ofPumping Station 2. From that point, the unpaved coastal access lane runseast along the backshore, with subsistence infrastructure on the seawardside, to the Elders Centre and the North Mart shopping complex. Thecoastal access lane then continues eastward between the formercourthouse and subsistence infrastructure on the beach crest as far asthe Visitor Centre. Sinaa Street continues east, landward of the VisitorCentre, the museum, the Amarok building, and two residences, which allhave subsistence infrastructure on their seaward side. Immediatelybeyond this is the small-craft basin at the foot of the main breakwater.Across Sinaa Street is the Grind and Brew cafe, with Pumping Station 1behind it. There are several residential properties in this area,including beachfront homes, multi-family structures, and a row oftownhouses across from the breakwater and boat launch. The road in thisarea has been flooded at extreme high tides (Fig. IB). Moving on to thesoutheast into the cemetery area, a small pocket beach backed by asingle-family home lies between the cemetery and the breakwater. Thecoast between the cemetery beach and Apex is composed of rock slopes andcliffs, with all structures located on high ground. The Apex Riverdischarges to the flats at the east end of Apex beach. The rest of theApex coast is a steep bedrock slope or terraced sand and gravel.

METHODS

Documenting coastal hazards in Iqaluit was essentially a mappingexercise enriched by analysis of relevant archival data sets. Fieldworkwas conducted between 2009 and 2011. Coastal surveys took place inAugust of all three years and were augmented by wave and water-levelmonitoring in 2010 and 2011, with measurement of currents in August2011. Sea ice observations and surveys were conducted in February andNovember 2011. For further details on the fieldwork, see Flatcher et al.(2014). The data used in this study can be classified into fivecategories: topography and bathymetry, infrastructure exposure, climateand weather, waves, and water levels.

Topography and Bathymetry

Topographic elevation points were collected using survey-gradereal-time kinematic (RTK) geographic positioning system (GPS) data. Thesystem used was an Ashtech Z-Extreme receiver with an Ashtech dual bandcarrier-phase antenna. Revisiting various control points established anestimated survey error of [+ or -] 0.05 m (Flatcher et al., 2014). Theexceptions were indicators of high water levels surveyed on outer InuitHead and the south side of Long Island, where real-time corrections werenot available and the data were post-processed to [+ or -] 0.15 m.Bathymetry was determined using GPS-positioned single-beam echosounding(Hatcher et al., 2014) and subsequently augmented by multibeam surveysusing the Government of Nunavut research vessel MV Nuliajuk (HughesClarke et al., 2015; Mate et al., 2015).

Elevations are always reported with reference to a vertical datum.In this study two were used: the Canadian Geodetic Vertical Datum of1928 (CGVD28) as orthometric datum (nominally equivalent to mean sealevel) and hydrographic Chart Datum. All GPS positions were recorded asellipsoidal elevations, which were subsequently converted to orthometricelevations using a separation value of 10.166 m. Elevations given inthis paper are reported with respect to CGVD28. Chart Datum (the tidegauge zero level) is derived from local tide gauge records andarbitrarily established at a level close to that of the lowest tide. InIqaluit, mean water level (MWL) is 5.9 m above Chart Datum (CHS, 2001).Lower Low Water Large Tide (LLWLT) is the lowest expected tide at 0.5 mChart Datum and Higher High Water Large Tide (HHWLT), the highestexpected tide, is 11.6 m above Chart Datum (CHS, 2001). Surveys of thetidal benchmark FBI-1968 established that Chart Datum has an elevationof -6.05 m CGVD28. Thus the orthometric elevations of the various tidelevels are as follows: -6.28'm (lowest recorded hourly WL), -5.55 m(LLWLT), -0.01 m (MWL), 5.55 m (HHWLT), and 6.04 m (highest recordedhourly WL).

A digital elevation model (DEM) provided by Natural ResourcesCanada was produced by stereo-pair photogrammetry using Worldview 2satellite imagery. This model excluded much of the intertidal flatsbecause they were partially underwater at the time the imagery wasacquired. In this study, coastal survey data were collected to build aseamless digital surface model across the tidal flats and into thenearshore. Elevations in the DEM were defined with respect to CGVD28.Shore-normal transects were surveyed with RTK-GPS at roughly 50 mspacing alongshore. Coincident points (where GPS points overlappedpixels of the DEM) were used to assess the accuracy of the elevationstaken from the DEM. Using the GPS points as reference, the standarderror was 0.4 m, but with some errors as large as 9 m. Larger errorsoccurred near the base of buildings as artifacts of the method employedin creating the DEM, but open-area elevations were much less prone toerror. The open-area accuracy of the DEM is assumed to be [+ or -] 0.5m. The resulting horizontal uncertainty in mapping of flood limits is afunction of slope ([+ or -] 5 m at a beach face slope of 6[degrees]; [+or -] 8 m at a backshore slope of 3.5[degrees]).

Infrastructure Exposure

Coastal infrastructure was classified into six categories:residential, commercial, municipal, cultural, federal, and subsistence.Residential includes housing within 75 m of the high-tide line.Commercial property within the 75 m planning zone includes the NorthMart (grocery store) and the Grind and Brew cafe. Municipalinfrastructure includes the two pumping stations, as well as the sewagedams, road and culvert elevations, and the old territorial courthouse(now owned by the city). Cultural infrastructure refers to municipalbuildings that are culturally significant, including the UnikkaarvikVisitor Centre and the Nunatta Sunakkutaangit Museum in downtown Iqaluitand the Hudson's Bay Company buildings on Apex beach. Federalproperty includes the Coast Guard and other Department of Fisheries andOceans (DFO) buildings. Finally, subsistence infrastructure describesthe sheds and sea cans (re-purposed shipping containers) used by localcountry-food harvesters, who are organized under the Amarok Hunters& Trappers Association.

Infrastructure elevations were acquired using RTK-GPS on keyinfrastructure as determined by this classification. Key infrastructurewas defined as all municipal, commercial, cultural, and federalbuildings found within the 30.5 m coastal planning zone. For residentialand subsistence infrastructure, we surveyed the foundation elevations ofrepresentative buildings. Where the building was raised above groundelevation on piles driven into permafrost, which is common in Iqaluit,we collected both ground and foundation (off-ground) elevations. Wherethe two elevations were equivalent, only one value was required (Fig.8A). For key infrastructure such as the courthouse or the pumpingstations, elevations were generally taken on the corner of the buildingfacing the coast, assuming a level foundation. Some categories, such asthe subsistence infrastructure or the roadbed elevations, include manypoints, covering the range of elevations for that type of infrastructurealong the length of the study area shoreline.

Climate and Sea Ice

Environment Canada meteorological records of temperature, wind, andprecipitation for Iqaluit have been collected since 1946, withquality-controlled hourly measurements at a continuously occupied sitesince 1953 (Table 1). Additional data have more recently been collectedby a climate station located between Iqaluit and Apex. Reliablemeteorological information is therefore available in Iqaluit for thelast 60 years.

The ice foot along the main Iqaluit waterfront was surveyed inFebruary 2011 using RTK-GPS to obtain elevations on the surface of theice. These points were directly over transects surveyed in the summer.This survey allowed an estimate of ice foot thickness and elevation forthe 2011 ice season. Direct observations of freeze-up by the authors inNovember 2011 included documentation of ice pile-up along thewaterfront.

We used two data sets to evaluate trends in the dates of freeze-upand breakup. These were the Canadian Ice Service Digital Archive (CISDA)(CIS, 2006) and the combined microwave sensor freeze-up/breakup analysisarchive from the NASA Goddard Space Flight Center (Markus et ah, 2009).The microwave sensor directly measures the onset and completion offreeze-up and breakup by detecting water on the surface. The CISDArecords report sea ice concentrations as a fraction of 10 (10 being 100%concentration). We defined the timing of freeze-up and breakup followingGagnon and Gough (2005) as the point at which ice concentration lastcrosses 7/10 for the season. The presence of a trend was determinedusing the non-parametric Mann-Kendall test for monotonic trends (Helseland Hirsch, 1992), as is done in sea ice trend analysis elsewhere(Gagnon and Gough, 2005; Laidler et ah, 2009). To add to the two datasets, we also considered ice thickness from a time series initiated byTransport Canada in 1959, but now maintained by the Canadian IceService. Weekly thickness surveys are conducted less than 1 km offshorewithin Koojesse Inlet between January and May of every year.

Waves and Run-Up

Instrument moorings in 2010 and 2011 contributed information onincident waves, currents, and water levels (Hatcher et ah, 2014). Wecollected wave data using seabed-mounted pressure sensors and anacoustic doppler current profiler (ADCP). The pressure sensors werelocated in the intertidal zone and recorded wave information every 30minutes. They were deployed from August to October of 2010 and 2011 invarious positions in the intertidal zone. A total of six deployments oftide and wave recorder (TWR) pressure sensors (RBR TWR-2050 instruments)and three deployments of a Nortek Aquadopp[c] 1.0 MHz ADCP provided dataon waves and currents. The ADCP recorded surface waves and currentvelocity profiles at one location on the flats and two in the harbourchannels.

The TWRs record simultaneous measurements of wave characteristicsand tidal water levels. The water-level measurements have a publishedprecision of [+ or -] 0.05% (equivalent to [+ or -] 0.005 m in theshallow water configuration used here), whereas the wave measurementuncertainty depends on the dominant wave frequency (Gibbons et al.,2005).

Wave hindcasting to estimate potential run-up heights used acombination of the archived wind records and the simple empiricalwind-wave relationships presented in Hurdle and Stive (1989), as revisedfrom the Shore Protection Manual (USACE, 1984). The wind-stress factorwas corrected for air-sea interface temperature difference andanemometer elevation according to USACE (1984), using the 10 manemometer winds reported at the Iqaluit weather station.

Water Levels and Sea Level Rise

Water-level records are available in two forms: the historical tidegauge records provided by the Canadian Hydrographic Service (CHS) andthe pressure sensor water levels recorded in 2010 and 2011 using the TWRinstruments described above. The CHS record is an irregular time seriesof hourly data with 28 198 hours of data between 1963 and 1977. Thismeans that over the 14-year span for which data exist, there were noobservations 77% of the time. The field data include 2670 hours of datain the open water seasons of 2010 and 2011.

To estimate past extreme water levels on the coast within the studyarea, we used RTK-GPS to survey two types of high water-level proxies:(1) water-level elevation, surveyed retroactively using a photo of aflooding event in 2003 (Fig. 1A, boulder surveyed is at centre of photodirectly below red shed), and (2) storm swash limit lines preserved atvarious places in the vicinity of Iqaluit. These lines were scatteredaround the outer limits of the inlet on undisturbed beaches (Fig. 4C).Surveying elevations on these swash limits provided undated estimates ofextreme high water levels combined with wave run-up.

In this study, a 70 cm rise in relative sea level over 90 years(2010-2100) was adopted as a precautionary estimate, initially based onearlier work by James et al. (2011). The most recent projections forIqaluit, based on the Fifth Assessment Report (AR5) of theIntergovernmental Panel on Climate Change (IPCC) (Church et al., 2013)and appropriate accounting for crustal motion, gravitational effects,dynamic oceanography, and other factors, indicate a rate of rise for thehighest rate of forcing, the so-called "representativeconcentration profile" 8.5 (RCP 8.5) close to zero, with a 95%confidence interval of about [+ or -] 40 cm (James et al., 2014). Theuse of RCP 8.5 is considered appropriate as a precautionary approach andalso recognizes that global C[O.sub.2] concentrations are tracking nearthe upper limits of IPCC projections (Friedlingstein et al., 2014),while observed global sea level rise has been similarly high (Rahmstorfet al., 2007; Church et al., 2013). The IPCC AR5 recognized thatadditional sea level rise from accelerated drawdown of the WestAntarctic Ice Sheet, for which the potential is poorly constrained,would not likely exceed several tenths of a metre during this century(Church et al., 2013). To allow for this scenario, James et al. (2014)provided an enhanced projection of +65 cm based on a number of publishedestimates of the likely effects of marine ice-sheet instability in WestAntarctica. Recent work suggests that increased oceanic melting andhydrofracturing of ice shelves could lead to collapse of the WestAntarctic Ice Sheet much sooner than previously thought and toaccelerated ice loss from the East Antarctic Ice Sheet (Pollard et al.,2015). The precautionary local sea level rise of 70 cm for 2010-2100adopted in this paper incorporates an upper 95% estimate of 40 cm forRCP 8.5 enhanced by a smaller 30 cm allowance for instability of theWest Antarctic Ice Sheet.

RESULTS

Iqaluit Topography and Waterfront Exposure

Beach crest elevations throughout the study area vary from beach tobeach, largely as a function of exposure. The lowest crest elevationsare found on the Iqaluit waterfront (5.1 m), with higher crest levels atthe cemetery beach (6.1 m elevation) and Apex beach (6.2 m elevation).Backshore slopes are fairly consistent throughout, except wherehigher-relief bedrock is exposed. The mean slope of all the backshoretransects surveyed (13 in total) is 3.5[degrees]. This value translatesto a 1:16 slope, where a 1 m rise in water level would floodapproximately 16 m horizontally into the backshore, which hasimplications for flood hazard projections, especially with a strongonshore wind that could drive additional setup and wave run-up.

Measured infrastructure foundation elevations in the waterfrontzone range between 4.25 m and 10.13 m elevation (Fig. 9). Subsistenceinfrastructure is predominantly located at the lowest elevations,closest to the water on the uppermost part of the beach. Residentialbuildings are, on average, at much higher elevations, although thelowest is a house at 5.6 m (10 cm above HHWLT).

Sea Ice Hazards

Results of the sea ice freeze-up and breakup timing analysis areshown in Table 2. The ice-free season has lengthened by 1.5 days/yearsince 1969 (99% confidence). The dates of breakup and freeze-up, asdefined in the NASA data, show comparable trends toward earlier breakup(-0.55 days/year) and later freeze-up (+0.48 days/ year). Using thedefinition of breakup and freeze-up for the CISDA data results in trendsof-0.95 days/year (breakup) and +0.54 days/year (freeze-up) (Fig. 10).Despite limitations imposed by a lack of satellite coverage prior to1979, as well as ambiguity in defining the onset of breakup orfreeze-up, this analysis suggests that Frobisher Bay is experiencing adecline in the length of the ice season (increase in the length of theopen water season). This result is consistent with the trend reported bylocal observers, including researchers at the Nunavut Research Institute(NRI), who have been monitoring freeze-up and breakup dates since 2002(R. Armstrong, NRI, pers. comm. 2011).

Examples of minor ice pile-up and ride-up were observed to bewidespread throughout the study area in November 2011, but the mostsubstantial occurrences were along a particular segment of shorelinenear the base of the breakwater, where a revetment (artificialsteepening of the shore) has been built. In this area, there wasevidence for both thicker floes and more significant pile-up duringspring-tide conditions (Fig. 5B). Along the beaches, the establishmentof the ice foot about halfway down the beach face restricted ice pile-upto the lower beach face, well seaward of any infrastructure. The icefoot seems to be established on a depth-dependent basis: in 2011 theseaward edge rested at a consistent seabed elevation of 3.5-4.0 m.

Flooding Hazards and Sea Level Rise

Evidence for extreme water levels is summarized in Table 3. The95th and 99th quantile water levels from the tide gauge record are 4.00m and 4.87 m elevation, respectively. The maximum level in the tidegauge record was 6.04 m on November 21, 1964. These values, being fromhourly records, may not capture the highest water levels, which may havepeaked up to 0.2 m higher, giving a possible extreme instantaneous highwater level of 6.25 m (12.3 m Chart Datum; CHS, 2001). Elevations ofsurveyed storm lines (which do not necessarily record still waterlevels) ranged from 5.16 m to 6.51 m. The surveyed high water levelapproximated from the October 2003 flood photograph (Fig. 1A) was 5.33m. A swash line found at the base of the Inuit Head pipeline had anelevation of 5.66 m, and two other swash lines found farther out InuitHead were at 6.19 m and 6.06 m ([+ or -] 0.15 m for these twoelevations). The highest swash lines on the outer shores of Long Islandwere at 6.48 m. The highest elevation swash line in the study area, at6.51 m, was surrounding the sewage lagoon at the head of the inlet.

The scenario of a 0.7 m rise in mean sea level combined with thehistorical high water limit was mapped onto the DEM to visualizepotential flood limits (Fig. 11). Unlike many coastal beach systemswhere a defined storm ridge or dune line protects against periodic highwater, the fairly even backshore slope in Iqaluit produces incrementallandward incursion of floodwater. A rise in sea level of 0.7 m with ahigh spring tide (0.7 m above HHWLT) would inundate 28% of the 30.5 mcoastal ("Open Space") zone and 14% of the coastal SijjaangaDistrict. The combination of this scenario for sea level rise and arepeat of the highest recorded water level would flood 46 of the 91coastal structures (50%), and five of the 61 municipal structures (8%)within the coastal district (Fig. 11).

Waves and Run-Up Hazards

For an open water fetch of 50 km, using meteorological records fromIqaluit and empirical wind-wave relations (Hurdle and Stive, 1989), thegreatest wave-producing winds on record (22 Sept 1960, 97 km/h for 3hours) give a hindcast significant wave height of [H.sub.s] = 1.6 m withpeak period [T.sub.p] = 5.9 s. The potential run-up from these waves ona beach slope of 5[degrees] typical of Koojesse Inlet is 0.6 m (Hunt,1959).

Observed [H.sub.s] reached 0.7 m over the flats and 1 m in deeperwater, with peak periods up to about 5 s. At a wave period of 5 s, thewaves begin shoaling well out over the tidal flats (in 9.8 m depth,based on the depth-to-wavelength ratio h/L < 0.25). Up to 80% energydissipation between two sensors placed along the path of incident waveswas observed in this study. The hindcast 5.9 s waves for the 22September 1960 event would begin to shoal in a depth of about 14 m andthus would suffer energy dissipation across the full width of the flatseven at high spring tide. The wave height at breaking and run-up heightson the beaches depend to a large extent on the incoming wavelength andtide level, as well as on the roughness of the shore face, the extent ofenergy dissipation during shoaling, and the slope of the beach. Thelargest waves observed during this study coming in at high spring tidewould suffer relatively little dissipation and would produce run-up of 1m or less on beach slopes ranging up to 6[degrees] (Hunt, 1959).

Overtopping and Erosion Hazards

Overtopping of one of the sewage lagoon dams could have highlynegative impacts on the health of the inlet ecosystem and do damage tothe subsistence fishery. Our surveys show crest elevations of 7.7 m onthe eastern dam and 7.3 m on the western dam. This is 1.3 m above thehighest recorded water level in the tide gauge record. Surveys of stormswash lines near the dams, however, show a run-up limit of 6.5 m, 0.5 mabove the highest recorded still water level. With an RMS survey errorof 0.05 m, there is 0.08 [+ or -] 0.1 m of freeboard (elevation aboverun-up level) to preserve the integrity of the dam and lagoon (Fig. 12).At the low end of this range, a freeboard of 0.7 m leaves little to noallowance for more extreme events or sea level rise. Apart fromdowncutting of the tidal flats, erosional retreat of the coast is not aserious concern in Iqaluit. Some parts of the shoreline are resistantbedrock, and in other places the beach sediments have formed at a levelconsistent with the highest swash run-up levels. However, a rise inrelative sea level would lead to movement of the beach system to adjustto the new mean water level and tidal limits, which would lead tolandward and upward movement of the beach sediments.

DISCUSSION

Results of this study investigating the major drivers of coastalhazards and the severity of hazard exposure along the Iqaluit waterfrontsuggest limited risk for much of the shorefront infrastructure.Nevertheless, some roads, structures, and other key facilities andresources are at risk from flooding, wave run-up, or ice impacts.Detailed mapping of coastal infrastructure shows that development hasbeen concentrated along the beachfront sections of the coast. In theseareas some critical infrastructure is found in the backshore, andnumerous subsistence-support resources (sheds, sea cans [shippingcontainers], boats, motors, skidoos, qamutiqs, and other equipment) areconcentrated on the uppermost part of the beach. The subsistenceinfrastructure is found primarily below the elevation of past extremewater levels, implying a tangible risk at the present time (Fig. 11).Much of the other waterfront infrastructure has a freeboard ranging from1.0 to 1.5 m or more above the highest observed historical water level.At the upper limit of projected local sea level rise over the nextcentury adopted in this study (0.7 m), this freeboard would be reducedto 0.3-0.8 m. Notwithstanding the 2003 flood event, community awarenessof the extent of exposure may be limited because the recorded extremehigh water level occurred more than 50 years ago, before development atthis site.

It is important to acknowledge the remaining uncertainty in the sealevel projections, for which the error bars at Iqaluit span a range fromfalling to rising sea level. This range reflects both the closeapproximation of the median projections to zero and uncertainties insome of the inputs, such as vertical crustal motion and glacial massbalance, both on Baffin Island and in Greenland (James et al., 2014).The sign of the sea level change at Iqaluit this century is highlysensitive to these variables, one of which (vertical motion) can beresolved by acquiring a longer time series of geodetic monitoring, whilethe other (ice mass reduction) not only requires more data, but alsowill be dominated by its response to the global human developmenttrajectory and greenhouse gas emissions (Friedlingstein et al., 2014;Fyke et al., 2014).

The spatial resolution of the two data sets used to determine seaice trends is somewhat different. The data from Markus et al. (2009) areat a 25 km grid resolution, and the CISDA data are 0.25[degrees]resolution (14 km in this area). In this analysis, trends werecalculated over the entirety of Frobisher Bay. It is therefore assumedthat long-term trends in sea ice breakup and freeze-up for Frobisher Bayas a whole are representative of the trend that would be observeddirectly off Iqaluit in the upper bay. Obviously, the complexity of seaice distribution in the bay makes this assumption problematic, but forthe purpose of determining the sign of the trend of freeze-up andbreakup (later or earlier in the year), it is considered valid. TheCISDA record shows high year-to-year variability in the length of theopen water season (Fig. 10), but statistically significant overalltrends toward longer open water seasons since 1979 were found in boththe CISDA and the NSIDC data sets (Table 2).

Where the beach extends out onto the tidal flat, the ice footestablished near the high-tide line extends far enough down the beachface to protect the upper crest area of the beach (and infrastructurelocated there) from sea ice pileup (Forbes and Hansom, 2011). Theelevation of the ice foot terrace may be lower initially, depending onthe phase of the tides, but it rises over time as successively highertides flood the surface and freeze (Fig. 13). Where the coast has beenartificially steepened by construction of revetments for shoreprotection, a narrower ice foot and deeper water close to shore favourhigher ice pile-up under appropriate ice, wind, and spring-tideconditions (Fig. 5B). Severe ice pile-up resulting from onshore windscombined with high tides has been widely documented elsewhere (Forbesand Taylor, 1994). Despite the protection offered by the ice foot, thepotential exists for damaging events at Iqaluit. Our field observationsdocumented numerous small pile-up ridges and ridging events (Fig. 5). Noaccounts of severe ice damage to infrastructure in Iqaluit have beenfound in discussions with local residents, but this is anotherphenomenon sensitive to the tide level in a macrotidal setting. As forflooding, the large tidal range reduces the probability of an extremeevent, which requires near coincidence with high water, but when thatlow-probability event occurs, its effects may be unprecedented.

Other research has shown the damaging effects of later freeze-up onthe subsistence food harvest (Statham et al., 2015). Discussions withresidents suggest that the longer the freeze-up remains dynamic andsusceptible to autumn storm effects, the harder it can be to transit theupper foreshore and beach (T. Tremblay, CNGO, pers. comm. 2011). Withprogressive delay of freeze-up into the fall storm season, there is anincreased likelihood of onshore winds acting on mobile ice (Hatcher,2014).

There is little published information on freeze-up in FrobisherBay, and its timing can be quite variable (Fox, 2003), with a range ofalmost 60 days (Hatcher, 2014). In 2011, when our on-site study offreeze-up took place, Koojesse Inlet became largely ice-covered over thecourse of a week (22-28 November). In the previous year, 2010, openwater persisted anomalously into January, and large waves developedduring a storm on November 27 (D. Mate, CNGO, Iqaluit, pers. comm.2011). Another late freeze-up, though not quite so extreme, occurred in1985. Progressively later freeze-up dates, as observed over the past fewdecades and expected with climate warming (Table 2), increase the riskof wave run-up events, irrespective of any change in storm climate, byincreasing the seasonal window for storms to occur over open water(Forbes and Hansom, 2011; Hatcher, 2014).

Some flooding at Iqaluit has occurred in the absence of storm winds(R. Armstrong, NRI, pers. comm. 2010). This raises the question ofwhether floods are attributable to high perigean spring tides. Long-termlunar cycles can add an extra 0.2 m on top of high tide levels (Haigh etal., 2011). No flooding occurred during the last two high periods ofthese long-term lunar cycles in Iqaluit, though a difference of 0.2 mcould be substantial here, given the small freeboard. Also, as observedin other large tidal embayments (e.g., Gehrels et al., 1995), a changein sea level (or ice conditions) may alter the tidal dynamics andamplitude at Iqaluit. Modeling of these changes is beyond the scope ofthis study, and the effects at Iqaluit will likely be minor.

We are unaware of any eyewitness accounts or traditional knowledgeof the 1964 flood event. It coincided with a moderate storm with minimumsea level air pressure of 98.6 hPa and easterly winds above 35 km/h,sustained for four hours. In an earlier preliminary report (Hatcher etal., 2011), we erroneously documented a 1.37 m storm surge associatedwith this event. However, subsequent more rigorous analysis of the tidaldata and predictions uncovered timing errors, and the actual offsetaveraged over that tidal cycle was 0.2 m. This suggests that upperFrobisher Bay, inside the band of islands in the mid-bay region, issomewhat protected from storm surges, although decimetre-scale windsetup, barometric, and ocean dynamic effects occur, as well as wavesetup amounting to less than 10% of deepwater wave height (Dean et al.,2005).

Iqaluit's coastline is a complex zone of physical and socialinteraction with a range of stakeholders and infrastructure types(critical municipal infrastructure, cultural resources, commercialproperties and residences, sealift freight handling facilities, and thesubsistence infrastructure belonging to hunters and fishers). Risk tothe subsistence infrastructure is rooted in the expansion of urbandevelopment into the backshore zone, which has left limited space forhunters and fishers, who need to locate on land with direct access tothe sea. Furthermore, the lack of undeveloped space landward of thepresent subsistence infrastructure prevents retreat in the face ofexisting and future hazards. This study shows the potential for futurehigher and more frequent floodwater and sea ice incursion into thesubsistence use zone. As visualized in Figure 12, an increase in sealevel not only raises the reach of extreme tide and surge events, butalso increases the frequency (probability) of flooding to levels rarelyflooded today. The potential impacts on subsistence infrastructure, inthe context of food security challenges in Iqaluit (Lardeau et al.,2011), represent a source of inequity and non-sustainability. The 2010General Plan outlines policies for coastal development based on tourism(City of Iqaluit, 2010). It would seem that co-planning of joint usebetween subsistence activities and tourism may be worth consideration toavoid an increase in coastal vulnerability along the city'sspatially constrained waterfront. A more holistic approach todevelopment, exemplified in the Sustainable Community Plan (City ofIqaluit, 2014), may be important to increase resilience on a number offronts.

Serious limitations are imposed by the scarcity of some key data inthis region. In particular, estimates of high water levels areconstrained by the short duration, sporadic coverage, and seasonal biasof the tide gauge data. The instrument moorings in 2010 and 2011,deployed as part of this study, provided the first measurements of wavesin the vicinity of Iqaluit, but unfortunately did not record a majorstorm event. The surveyed run-up levels, geomorphology, and localknowledge demonstrate that wave impacts on the coast do occur. Thisstudy has shown that the tidal flats play an important role in shoreprotection, dissipating a large proportion of incoming wave energy,except at the highest tides. The storm of concern is the rare wave eventcoinciding with the highest tide. The macrotidal regime makes thiscoincidence more critical than at sites with lower tidal range, reducingthe window of opportunity for an extreme water-level event. In theabsence of more complete water-level measurements, it is not possible tocompute the probability of such an event from empirical records. Inaddition, remaining ambiguities in the rate of crustal uplift are amajor limitation for projections of sea level change in the area.However, ongoing geodetic data collection is expected to provide morereliable rates of uplift, which will help to constrain the projectionsof local (relative) sea level change.

Sustainable development planning in Iqaluit would benefit fromfurther studies. In particular, it would be valuable to improve ourunderstanding of coastal ice mobility and projections of freeze-up andbreakup dates (and length of the open water season) as a function ofregional climate projections. The limited importance of coastal erosionremoves that issue from the monitoring agenda. Thus coastal monitoringin this area should focus primarily on sea level change, sea icedynamics, wave climate, wave shoaling, and run-up levels. Detailedsurveys of ice pile-up ridges during freeze-up, as well as theconditions that caused them, would help to better define this hazard inthe local context. Related monitoring of ice foot growth and dimensions,including year-to-year variability and trends, would complement thisanalysis.

Despite the urban context, high number of wage earners, and largeproportion of residents originally from elsewhere, the value oftraditional knowledge in planning should not be overlooked. Asubstantial proportion of Iqaluit's residents use the sea iceduring the winter for access to country food and have accumulatedknowledge on the characteristics of winter ice. During the summermonths, residents travel by water throughout the bay and thus arefamiliar with the patterns of wind and waves and the impacts of stormevents. Within the urban centre, hazard events of the past two decades,at least, remain in the memories of longtime residents. This is aknowledge source that can contribute to effective decision making andcommunity resilience. At the same time, instrumental monitoring of keyenvironmental variables, including vertical crustal motion, wind, waterlevels, waves, and ice, can play an important role in detecting andtracking change, validating and refining projections, and quantifyingevolving risks to the people and infrastructure of the city.

CONCLUSIONS

This study has identified three coastal hazards relevant toinfrastructure in Iqaluit. (1) Exposure to ride-up or pile-up of seaice. This hazard involves a variety of factors associated with freeze-upand breakup and an increase in risk associated with climate-inducedexpansion of the open water season. Ice also plays a protective role inthe form of the winter ice foot, which shelters the shore and nearbyinfrastructure from direct impacts of the mobile ice over the tidalflats. (2) Exposure to flooding of coastal infrastructure. The dominantrisk factor is tidal dynamics, combined with relatively minorcontributions from steric, barometric, and wind stress events (stormsurge). The documented flooding, at least in 2003, seems to haveoccurred as a result of an extreme tide, perhaps associated with aregional dynamic anomaly. A contribution from a minor storm surge cannotbe ruled out. (3) Wave run-up and associated setup. These events havethe potential to overtop the sewage-retaining berms and damage otherinfrastructure along the urban waterfront. A more detailed analysis ofthis hazard is warranted.

Interacting with all of these, the trend of relative sea level isthe dominant control on the vertical extent and landward reach ofspecific hazard processes. This study has evaluated the change inflooding extent that would result from a 0.7 m rise in local sea level.This scenario is close to the upper limit of plausible change over the90 years 2010-2100 and appropriate for a precautionary approach with lowrisk tolerance. It is important to recognize that the statisticaluncertainty in the sea level projections includes both rising andfalling relative sea level for all representative concentration profilesconsidered in the IPCC AR5 (James et al., 2014). Nevertheless, analysesof global trends in major climate variables, including temperature andglobal mean sea level, have shown that the world is tracking near theupper limits of the range of projections (Rahmstorf et ah, 2007; Churchet ah, 2013). Assuming that relative sea level is rising in Iqaluit,this change will dominate the rise in probability of waterfront floodingand extreme high water events.

This study points to a number of implications for adaptationplanning in Iqaluit:

* Steepening of the coastal profile through revetment or armouringmay protect against waves, but a steeper profile with a narrow ice footallows higher ice pile-up, increasing exposure of infrastructuredirectly landward of the revetment to potential ice impact.

* Accurate surveys of coastal infrastructure have allowed theestimation of waterfront elevations and freeboard under various sealevel rise, high water, and wave run-up scenarios. The maximum recordedwater level is 6.04 m above mean sea level (1964), and the highestsurveyed swash line is 6.51 m (date unknown). This study has shown thatfor an observed extreme high water event added to a plausible upperlimit of the most recent projections of sea level for 2100 (0.7 m abovethe 2010 mean sea level), 50% of the infrastructure within the coastal"open space" planning zone would be affected, and significantareas of land would be flooded in the developed backshore.

* Some shorefront infrastructure in Iqaluit is already at risk offlooding in extreme high water events, as demonstrated by the tide gaugerecord for 21 November 1964 and the anecdotal and photographic evidencefrom October 2003. The expanded flood risk from potential sea level risewithin the range of the latest projections warrants attention. This isparticularly the case for the coastal subsistence infrastructure, whichis an essential contributor to sustainability in Iqaluit, yet itsposition on the coast means that it is most exposed to any change inhazards arising from sea level rise or changing sea ice and waveregimes.

http://dx.doi.org/10.14430/arctic4526

ACKNOWLEDGEMENTS

This study was supported under the C-Change InternationalCommunity-University Research Alliance (funded by the Social Sciencesand Humanities Research Council and the International DevelopmentResearch Centre), and also by ArcticNet (funded by the tri-councilNetworks of Centres of Excellence Canada through the Natural Sciencesand Engineering Research Council), the Geological Survey of Canadathrough the Climate Change Geoscience Program of the Earth SciencesSector, Natural Resources Canada (NRCan), the Northern ScientificTraining Program of Aboriginal Affairs and Northern Development Canada(AANDC), and Memorial University of Newfoundland (MUN). Moral andlogistical support was provided by the Nunavut Research Institute (NRI),the Canada-Nunavut Geoscience Office (CNGO), the Government of Nunavut(GN) Climate Change Program, and the City of Iqaluit. In particular weacknowledge assistance from Meagan Leach and Robyn Campbell (City ofIqaluit), Colleen Healey (GN), David Mate and Tommy Tremblay (CNGO),Mary Ellen Thomas, Bruce Armstrong, Mosha Cote, Jamal Shirley, and AlexFlaherty (NRI), Michel Allard (Universite Laval), and Anne-Marie LeBlanc(NRCan). It is a pleasure to acknowledge project assistance from DanLane and Colleen Mercer Clarke (C-Change), Gavin Manson (NRCan), PaulBudkewitsch (AANDC/ formerly NRCan), Trevor Bell and Pam Murphy (MUN).Gavin Manson (NRCan), Dominique St-Hilaire,(MUN Marine Institute), TommyTremblay (CNGO), and Alex Flaherty (NRI) provided much appreciated fieldassistance. We thank Tom James (NRCan) for discussion and assistancewith sea level projections. This paper has benefited from reviews ofearlier drafts by Evan Edinger, Norm Catto, Ian Walker, Mike Lewis, AnnGibbs, and an anonymous journal reviewer, all of whom we thank for theirinsight and suggestions. This is contribution number 20140457 of theEarth Sciences Sector, Natural Resources Canada.

REFERENCES

Allard, M., Doyon, J., Mathon-Dufour, V., LeBlanc, A.-M.,L'Herault, E., Mate, D., Oldenborger, G.A., and Sladen, W.E. 2012.Surficial geology, Iqaluit, Nunavut. Canadian Geoscience Map 64(preliminary version), scale 1:15,000. Ottawa: Geological Survey ofCanada. http://dx.doi.org/10.4095/289503

Are, F., Reimnitz, E., Grigoriev, M., Hubberten, H.-W., andRachold, V. 2008. The influence of cryogenic processes on the erosionalArctic shoreface. Journal of Coastal Research 24(1):110-121.http://dx.doi.org/10.2112/05-0573.1

Arehart, M. 2012. Akiak declares erosion disaster. Alaska PublicMedia, October 4. http://www.alaskapublic.org/2012/10/04/akiak-declareserosion-disaster/

Catto, N.R., and Parewick, K. 2008. Hazard and vulnerabilityassessment and adaptive planning: Mutual and multilateralcommunity-research communication, Arctic Canada. In: Liverman, D.G.E.,Pereira, C.P.G., and Marker B., eds. Communicating environmentalgeoscience. London: Geological Society, Special Publications305:123-140. http://dx.doi.org/10.1144/SP305.12

CHS (Canadian Hydrographic Service). 2001. Canadian tide andcurrent tables. Volume 4: Arctic and Hudson Bay. Ottawa: Fisheries andOceans Canada.

Church, J.A., Clark, P.U., Cazenave, A., Gregory, J.M., Jevrejeva,S., Levermann, A., Merrifield, M.A., et al. 2013. Sea level change. In:Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K.,Boschung, J., Nauels, A., Xia, Y., Bex, V., and Midgley, P.M., eds.Climate change 2013: The physical science basis. Contribution of WorkingGroup I to the Fifth Assessment Report of the Intergovernmental Panel onClimate Change. Cambridge and New York: Cambridge University Press.1137-1216. http://www.ipcc.ch/pdf/assessment-report/ar5/wg1/WG1AR5_Chapter13_FINAL.pdf

CIS (Canadian Ice Service). 2006. Canadian Ice Service digitalarchive regional charts: History, accuracy, and caveats. CIS ArchiveDocumentation Series No. 1. Ottawa: Environment Canada.

City of Iqaluit. 2010. General plan: By-law 703.http://www.iqaluit.nu.ca/i18n/english/pdf/FINA LGeneralPlanDraftMay2010Eng.pdf

--. 2014. Iqaluit sustainable community plan.http://sustainableiqaluit.com/resources/iqaluit-sustainablecommunity-plan/

Dale, J.E., Leech, S., McCann, S.B., and Samuelson, G. 2002.Sedimentary characteristics, biological zonation and physical processesof the tidal flats of Iqaluit, Nunavut. In: Hewitt, K., Byrne, M.-L.,English, M., and Young, G., eds. Landscapes in transition: Landformassemblages and transformations in cold regions, Vol. 111. Dordrecht:Kluwer. 205-234.

Dean, R., Collins, I., Divoky, D., Hatheway, D., and Scheffner, N.2005. Wave setup: FEMA coastal flood hazard analysis and mappingguidelines. Focused Study Report.http://www.fema.gov/media-library-data/20130726-154120490-1234/frm_plwavel.pdf

Eno, R.V. 2003. Crystal Two: The origin of Iqaluit. Arctic56(1):63-75. http://dx.doi.org/10.14430/arctic603

Forbes, D.L., ed. 2011. State of the Arctic coast 2010: Scientificreview and outlook. International Arctic Science Committee, Land-OceanInteractions in the Coastal Zone, Arctic Monitoring and AssessmentProgramme, International Permafrost Association. Geesthacht, Germany:Helmholtz-Zentrum. 178 p. http://arcticcoasts.org

Forbes, D.L., and Hansom, J.D. 2011. Polar coasts. In: Flemming,B.W., and Hansom, J.D., eds. Estuarine and coastal geology andgeomorphology. Treatise on estuarine and coastal science, Vol. 3.Wolanski, E., and McLusky, D.S., editors-in-chief. Waltham: AcademicPress. 245-283.

Forbes, D.L., and Taylor, R.B. 1994. Ice in the shore zone and thegeomorphology of cold coasts. Progress in Physical Geography18(1):59-89. http://dx.doi.org/10.1177/030913339401800104

Forbes, D.L., Leach, M., Allard, M., Campbell, R., Hatcher, S.V.,LeBlanc, A.-M., and Manson, G.K. 2012. Implications of landscape hazardsand climate change for sustainability in an Arctic coastal municipality.Caribbean Conference 50-50: Surveying the Past, Mapping the Future.21-23 March 2012, Grand Cayman.http://coastalchange.ca/download_files/Caymans_Conference_Presentation_Forbes.pdf

Forbes, D.L., Bell, T., James, T.S., and Simon, K.M. 2014.Reconnaissance assessment of landscape hazards and potential impacts offuture climate change in Arviat, southern Nunavut. In: Summary ofactivities 2013. Iqaluit, Nunavut: Canada-Nunavut Geoscience Office.183-192. http://cngo.ca/summary-of-activities/2013/

Ford, J.D., Bell, T., and St-Hilaire-Gravel, D. 2010. Vulnerabilityof community infrastructure to climate change in Nunavut: A case studyfrom Arctic Bay. In: Hovelsrud, G.K., and Smit, B., eds. Communityadaptation and vulnerability in Arctic regions. Dordrecht: SpringerScience + Business Media. 107-130.http://dx.doi.org/10.1007/978-90-481-9174-l

Fox, S. 2003. When the weather is Uggianaqtuq: Inuit observationsof environmental change. Boulder: University of Colorado, GeographyDepartment, Cartography Lab. Distributed by National Snow and Ice DataCenter. CD-ROM.

Friedlingstein, P., Andrew, R.M., Rogelj, J., Peters, G.P.,Canadell, J.G., Knutti, R., Luderer, G., et al. 2014. Persistent growthof C[O.sub.2] emissions and implications for reaching climate targets.Nature Geoscience 7:709-715. http://dx.doi .org/10.1038/NGEO2248

Fyke, J.G., Vizcaino, M., and Lipscomb, W.H. 2014. The pattern ofanthropogenic signal emergence in Greenland Ice Sheet surface massbalance. Geophysical Research Letters 41(16):6002-6008.http://dx.doi.org/10.1002/2014GL060735

Gagnon, A.S., and Gough, W.A. 2005. Trends in the dates of icefreeze-up and breakup over Hudson Bay, Canada. Arctic 58(4):370-382.http://dx.doi.org/10.14430/arctic451

Gehrels, W.R., Belknap, D.F., Pearce, B.R., and Gong, B. 1995.Modeling the contribution of [M.sub.2] tidal amplification to theHolocene rise of mean high water in the Gulf of Maine and the Bay ofFundy. Marine Geology 124(1-4):71-85.http://dx.doi.org/10.1016/0025-3227(95)00033-U

Gibbons, D.T., Jones, G., Siegel, E., Hay, A., and Johnson, F.2005. Performance of a new submersible tide-wave recorder. In:Proceedings of MTS/IEEE Oceans Conference, Oceans 2005, 17-23 September2005, Washington, D.C. Vol. 2:1057-1060.http://dx.doi.org/10.1109/OCEANS.2005.1639895

Haigh, I.D., Eliot, M., and Pattiaratchi, C. 2011. Globalinfluences of the 18.61 year nodal cycle and 8.85 year cycle of lunarperigee on high tidal levels. Journal of Geophysical Research 116,C06025. http://dx.doi.org/10.1029/2010JC006645

Hatcher, S.V. 2014. People at the tidal flats: Coastal morphologyand hazards in Iqaluit, Nunavut. MSc thesis, Memorial University ofNewfoundland, St. John's, Newfoundland and Labrador.

Hatcher, S.V., Forbes, D.L., and Manson, G.K. 2011. Coastal hazardassessment for adaptation planning in an expanding Arctic municipality.In: Proceedings, Sustainable Development of Coastal Communities:Challenges and Solutions International Conference, 1-3 June 2011,Port-of-Spain, Trinidad & Tobago. C-Change ICURA Working Paper 23.Ottawa: University of Ottawa, C-Change Secretariat.http://www.coastalchange.ca/images/stories/Documents_Tab/workingpaper23_hatcherforbes_manson_2011.pdf

--. 2014. Coastal geoscience field work near Iqaluit, Nunavut,2009-2011. Open File 7653. Ottawa: Geological Survey of Canada.http://dx.doi.org/ 10.4095/295628

Helsel, D.R., and Hirsch, R.M. 1992. Chapter 12: Trend analysis.In: Statistical methods in water resources. Studies in EnvironmentalScience 49:323-355. http://dx.doi.org/10.1016/S0166-1116(08)71111-7

Hodgson, D.A. 2005. Quaternary geology of western Meta IncognitaPeninsula and Iqaluit area, Baffin Island, Nunavut. GSC Bulletin 582.Ottawa: Geological Survey of Canada.

Hughes Clarke, J.E., Muggah, J., Renoud, W., Bell, T., Forbes,D.L., Cowan, B., and Kennedy, J. 2015. Reconnaissance seabed mappingaround Hall and Cumberland Peninsulas, Nunavut: Opening up southeastBaffin Island to nearshore geological investigations. In: Summary ofactivities 2014. Iqaluit, Nunavut: Canada-Nunavut Geoscience Office.133-144. http://cngo.ca/summary-of-activities/2014/

Hunt, I. A., Jr. 1959. Design of seawalls and breakwaters. AmericanSociety of Civil Engineers, Journal of the Waterways and HarborsDivision 85(3):123-152.

Hurdle, D.P., and Stive, R.J.H. 1989. Revision of SPM 1984 wavehindcast model to avoid inconsistencies in engineering applications.Coastal Engineering 12(4):339-351.http://dx.doi.org/10.1016/0378-3839(89)90011-2

IPCC (Intergovernmental Panel on Climate Change). 2013. Summary forpolicymakers. In: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M.,Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., and Midgley,P.M., eds. Climate change 2013: The physical science basis. Contributionof Working Group 1 to the Fifth Assessment Report of theIntergovernmental Panel on Climate Change. Cambridge and New York:Cambridge University Press.http://www.climatechange2013.org/images/report/WG1AR5_ SPM FINAL.pdf

James, T.S., Simon, K.M., Forbes, D.L., Dyke, A.S., and Mate, D.J.2011. Sea level projections for five pilot communities of the NunavutClimate Change Partnership. Open File 6715. Ottawa: Geological Survey ofCanada. http://dx.doi.org/10.4095/288019

James, T.S., Henton, J.A., Leonard, L.J., Darlington, A., Forbes,D.L., and Craymer, M. 2014. Relative sea level projections in Canada andthe adjacent mainland United States. Open File 7737. Ottawa: GeologicalSurvey of Canada. http://dx.doi.org/10.4095/295574

Laidler, G.J., Ford, J.D., Gough, W.A., Ikummaq, T., Gagnon, A.S.,Kowal, S., Qrunnut, K., and Irngaut, C. 2009. Travelling and hunting ina changing Arctic: Assessing Inuit vulnerability to sea ice change inIgloolik, Nunavut. Climatic Change 94(3):363-397.http://dx.doi.org/10.1007/s10584-008-9512-z

Lane, D., Mercer Clarke, C., Forbes, D.L., and Watson, P. 2013. Thegathering storm: Managing adaptation to environmental change in coastalcommunities and small islands. Sustainability Science 8(3):469-489.http://dx.doi.org/10.1007/s11625-013-0213-9

Lardeau, M.-P., Healey, G., and Ford, J. 2011. The use ofPhotovoice to document and characterize the food security of users ofcommunity food programs in Iqaluit, Nunavut. Rural and Remote Health11(2): 1680. http://www.rrh.org.au

Larsen, P.H., Goldsmith, S., Smith, O., Wilson, M.L., Strzepek, K.,Chinowsky, P., and Saylor, B. 2008. Estimating future costs for Alaskapublic infrastructure at risk from climate change. Global EnvironmentalChange 18(3):442-457. http://dx.doi.org/10.1016/j.gloenvcha.2008.03.005

Leech, S.R. 1998. The transport of materials by ice in a subarcticmacrotidal environment, Koojesse Inlet, southeast Baffin Island. MScthesis, University of Regina, Regina, Saskatchewan.

Markus, T., Stroeve, J.C., and Miller, J. 2009. Recent changes inArctic sea ice melt onset, freezeup, and melt season length. Journal ofGeophysical Research 114, C12024. http://dx.doi.org/10.1029/2009JC005436

Mate, D.J., Campbell, D.C., Barrie, J.V., Hughes Clarke, J.E.,Muggah, J., Bell, T., and Forbes, D.L. 2015. Integrated seabed mappingof Frobisher Bay, Nunavut to support infrastructure development,exploration, and natural hazard assessment. In: Summary of activities2014. Iqaluit, Nunavut: Canada-Nunavut Geoscience Office. 145-152.http://cngo.ca/summary-of-activities/2014/

Maxwell, J.B. 1981. Climatic regions of the Canadian ArcticIslands. Arctic 34(3):225-240. http://dx.doi.org/10.14430/arctic2525

McCann, S.B., and Dale, J.E. 1986. Sea ice breakup and tidal flatprocesses, Frobisher Bay, Baffin Island. Physical Geography7(2):168-180.

McCann, S.B., Dale, J.E., and Hale, P.B. 1981. Subarctic tidalflats in areas of large tidal range, southern Baffin Island, easternCanada. Geographic physique et Quaternaire 35(2)483-204.http://dx.doi.org/10.7202/1000436ar

Miller, G.H., Locke, W.W., III, and Locke, C.W. 1980. Physicalcharacteristics of the southeastern Baffin Island coastal zone. In:McCann, S.B., ed., The coastline of Canada: Littoral processes and shoremorphology. GSC Paper 80-10. Ottawa: Geological Survey of Canada.251-265.

Mitrovica, J.X., Tamisiea, M.E., Davis, J.L., and Milne, G.A. 2001.Recent mass balance of polar ice sheets inferred from patterns of globalsea level change. Nature 409:1026-1029.http://dx.doi.org/10.1038/35059054

Nielsen, D. 2007. The City of Iqaluit's climate changeimpacts, infrastructure risks, and adaptive capacity project. Iqaluit,Nunavut: City of Iqaluit.

NRTEE (National Round Table on the Environment and the Economy).2009. True North: Adapting infrastructure to climate change in northernCanada. Ottawa: NRTEE'.http://collectionscanada.gc.ca/webarchives2/20130322143509/http://nrtee-trnee.ca/climate/true-north

Pollard, D., DeConto, R.M., and Alley, R.B. 2015. PotentialAntarctic Ice Sheet retreat driven by hydrofracturing and ice clifffailure. Earth and Planetary Science Letters 412:112-121.http://dx.doi.org/10.1016/j.eps1.2014.12.035

Rahmstorf, S., Cazenave, A., Church, J.A., Hansen, J.E., Keeling,R.F., Parker, D.E., and Somerville, R.C.J. 2007. Recent climateobservations compared to projections. Science 316(5825):709.http://dx.doi.org/10.1126/science.1136843

Serreze, M.C., and Barry, R.G. 2011. Processes and impacts ofArctic amplification: A research synthesis. Global and Planetary Change77(l-2):85-96. http://dx.doi.org/10.1016/j.gloplacha.2011.03.004

Shirley, J. 2005. Inuit Qaujimajatuqangit of climate change inNunavut. Iqaluit, Nunavut: Department of Environment, Government ofNunavut.

Short, N., LeBlanc, A.-M., Sladen, W.E., Allard, M., andMathon-Dufour, V. 2012. Seasonal surface displacement derived fromInSAR, Iqaluit, Nunavut. Canadian Geoscience Map 66 (preliminaryversion), 1:15,000. Ottawa: Geological Survey of Canada.http://dx.doi.org/10.4095/289606

Smith, I.R. 2014. Reconnaissance assessment of landscape hazardsand potential impacts of future climate change in Kugluktuk, westernNunavut. In: Summary of activities 2013. Iqaluit, Nunavut:Canada-Nunavut Geoscience Office. 149-158.http://cngo.ca/summary-of-activities/2013/

Smith, I.R., and Forbes, D.L. 2014. Reconnaissance assessment oflandscape hazards and potential impacts of future climate change inCambridge Bay, western Nunavut. In: Summary of activities 2013. Iqaluit,Nunavut: Canada-Nunavut Geoscience Office. 159-170.http://cngo.ca/summary-of-activities/2013/

Statham, S., Ford, J., Berrang-Ford, L., Lardeau, M.-P., Gough, W.,and Siewierski, R. 2015. Anomalous climatic conditions during winter2010-2011 and vulnerability of the traditional Inuit food system inIqaluit, Nunavut. Polar Record 51(3): 301-317.http://dx.doi.org/10.1017/S0032247414000151

Statistics Canada. 2014. Focus on geography series, 2011 census:Census subdivision of Iqaluit, CY--Nunavut.http://www12.statcan.gc.ca/census-recensem*nt/2011/as-sa/fogs-spg/Facts-csd-eng.cfm?Lang=eng&GK=CSD&GC= 6204003

Strzelecki, M. 2011. Cold shores in warming times - current stateand future challenges in High Arctic coastal geomorphological studies.Quaestiones Geographicae 30(3):101-113.http://dx.doi.org/10.2478/v10117-011-0030-0

USACE (U.S. Army Corps of Engineers). 1984. Shore protectionmanual. Vol. 1. Vicksburg, Mississippi: Coastal Engineering ResearchCenter, Department of the Army.

(Received 3 October 2014; accepted in revised form 15 July 2015)

Scott V. Hatcher (1,2) and Donald L. Forbes (1,3)

(1) Department of Geography, Memorial University of Newfoundland,St. John's, Newfoundland and Labrador A1B 3X9, Canada

(2) Corresponding author: [emailprotected]

(3) Geological Survey of Canada, Natural Resources Canada, BedfordInstitute of Oceanography, Dartmouth, Nova Scotia B2Y 4A2, Canada;[emailprotected]

[c] Her Majesty the Queen in Right of Canada. Administered by theArctic Institute of North America

TABLE 1. Available climate data for the Iqaluit area. Three of thestations (2402590, 2402591, and 2402594) are at 33.5 m elevation.The other station (2402592) is at 22 m elevation. All stations arelocated in the same position at 63[degrees]45' N, 68[degrees]33' W,near Iqaluit airport at the head of the inlet.Station ID Hourly data Daily data2402590 1953-01 -01 to 2011-05-11 1946-01-01 to 2008-07-312402591 2008-07-03 to 2011-05-11 2008-07-01 to 2011-05-312402592 2004-12-16 to 2012-01-01 2004-05-01 to 2011-05-312402594 NA 1997-04-01 to 2007-11-30TABLE 2. Trends in two data sets for Frobisher Bay. Significancelevels, taken from the Kendall tau rank correlation coefficient, areshown by *** (99%) and ** (95%). Negative values show breakupearlier in the year (negative Julian days), and positive values showfreeze-up later in the year (positive Julian days). Positive durationshows the lengthening of the ice-free open water season.Type CISDA trend NSIDC trendBreakup -0.95 ** -0.55 ***Freeze-up +0.54 ** +0.49Ice-free season 1.58 *** 1.05 ***TABLE 3. Description, elevation above mean sea level (CGVD28),and data source for high water levels in Iqaluit. DGPS surveyrefers to the differential geographic positioning system surveyperformed in this study.Description Elevation (m) Source95th quantile 4.00 tide gauge99th quantile 4.87 tide gaugemax recorded 6.04 tide gaugeOct 2003 flood 5.33 photograph (DGPS survey)Cemetery beach 5.31 swash line (DGPS survey)Inuit Head pipeline 5.66 swash line (DGPS survey)Inuit Head station 6.06 swash line (DGPS survey)Inuit Head station 6.19 swash line (DGPS survey)Long Island 6.48 swash line (DGPS survey)sewage lagoon 6.51 swash line (DGPS survey)