geomorphology and sea-level rise on one of canada's most

Geomorphology and Sea-level Rise on one of Canada's Most Sensitive Coasts:Northeast Graham Island, British Columbia

INTRODUCTION

The Queen Charlotte Islands (Haida Gwaii) is an archipelagolocated 80 km offshore from British Columbia's north coast(54°N, 132°W, Fig. 1). The northeastern region of GrahamIsland, known as the Argonaut Plain or Naikoon peninsula, iscomprised of unconsolidated Quaternary sediments ofglaciofluvial origin ., 1982). This region isdistinct in British Columbia in that it experienced limited to noice cover during the Late Wisconsinan glaciation (ca. 16000 yrBP) and may have served as a glacial refugium ( .,1982). Dramatic fluctuations in relative sea level have occurredin the northern Pacific margin of Canada over the lateQuaternary. In Hecate Strait, a rapid regression to -150 m

occurred between 14600 - 12400 C yr BP due to isostatic

rebound followed by transgression to +16 m by 8900 C yr BPdue to combined eustatic rise and subsidence of a glacioisostaticforebulge ( ., 1997, and

2002). Since 3800 C yr BP, sea level has regressed leaving aseries of relict shorelines and prograding beach ridges. Over the

20 century, relative sea level has been rising at a rate of +1.6

mm a and , 2005).The modern coastal landscape of NE Graham Island consists

of over 100 km of sandy shoreline that is host to a variety oflandforms including low gradient, dissipative beaches backedby prograding foredunes on North Beach; reflective cuspatecobble beaches with sandy low tide terraces on Rose Spit; andmultiple-barred beaches backed by migrating foredunes andparabolic dunes on East Beach. This coast experiences 'extreme'conditions including a macrotidal range, an energetic waveclimate with frequent storm surges and frequent, strong winds

exceeding gale force (65 km hr ). This creates a dynamiccoastline that is maintained by northward littoral transport ofsediments along East Beach by strong wave and tidal currents.

This coast is retreating by 1-3 m a and greater during extremeevents such as the 1997-98 El Niño that caused 0.4 m of regionalsea-level rise and 12 m of localized retreat ( and

2002). In contrast, the dissipative shores of North

Beach are prograding at 0.3 to 0.6 m a ( 1980). Onboth shores, much of the littoral sediment is moved onshore viaaeolian delivery in frequent transporting winds to active,migrating dune systems despite a moist maritime climate,extensive log jams and dense forest cover.

Given the tidal range, wave regime, erodible sediments andongoing rates of sea-level rise and erosion, the GeologicalSurvey of Canada (GSC) identifies NE Graham Island as one ofCanada's most sensitive coastlines to sea-level rise (

., 1998). In that study of past and modern responses to sea-level rise may portend future impacts, the purpose of this paperis to discuss the late Quaternary and modern geomorphology ofthis coast and identify potential responses to sea-level rise. Thisexamination provides a preliminary physical basis for a largerinterdisciplinary study of the vulnerability of this area to futuresea-level rise. A preliminary integrated conceptual frameworkfor this larger study appears in a companion paper ( and

, 2006).

Graham Island experiences a marine west coast cool (Cfb)

climate and receives 1398 mm a precipitation. EnvironmentCanada 30 yr climate normals for Sandspit Airport (150 km S)show that 69% of this precipitation falls as rain from Septemberto March though appreciable (300-500 mm), yet short-lived,snowfall occurs from December to March. The moderatinginfluence of the Pacific Ocean on average temperatures isseasonally pronounced with above freezing daily temperatures

(CLAGUE

WARNER

JOSENHANS BARRIE CONWAY,

(ABEYSIRIGUNAWARDENA WALKER

BARRIE

CONWAY,

HARPER,

SHAW

DOLAN

WALKER

et al

et al

et al

etal

14

14

14

th

-1

-1

-1

-1

-1

PHYSICAL SETTING

Climate

Journal of Coastal Research SI 39 220- 226 ICS 2004 (Proceedings) Brazil ISSN 0749-0208

I. J. Walker† and J. V. Barrie‡

†Department of Geography,University of VictoriaVictoria, British ColumbiaV8W 3P5 [email protected]

WALKER, I. J., and BARRIE, J.V., 2006. Geomorphology and sea-level rise on one of Canada's most 'sensitive'coasts: Northeast Graham Island, British Columbia. Journal of Coastal Research, SI 39,

220 - 226. Itajaí, SC, Brazil, ISSN 0749-0208.

This paper documents the geomorphology of one of Canada's most dynamic coastlines, northeast Graham Island,Queen Charlotte Islands (Haida Gwaii). This area has undergone major sea-level changes (-150 to +16 m) over the

Holocene and is currently rising at +1.6 mm a . Relict shorelines and recent progradational ridges tell of landscaperesponse to sea-level changes. Given a macrotidal range, energetic wave climate and ongoing erosion, theGeological Survey of Canada identifies this as one of Canada's most sensitive coasts to future sea-level rise. Retreat

of 1-3 m a and tens of metres in extreme years (e.g., El Niño 1997-98) occurs on East Beach, while the dissipative

shores of North Beach prograde at 0.3-0.6 m a . Over 100 km of sandy beach are maintained by strong littoraltransport by tidal currents and storm waves. Northward longshore transport on East Beach heads westward aroundRose Spit to North Beach, particularly during SE storms. Despite a moist climate and dense vegetation, aeolianactivity is high and moves sand onshore into foredunes and driftwood jams that stabilize the backshore against waveattack. Foredunes migrate on the order of metres per year while larger parabolic dunes migrate slower (metres perdecade). Migrating shore-attached bars feed beach-dune systems while their leading edge is a locus for beacherosion. Dendrochronological (tree-ring) evidence is explored to provide proxy evidence of past climate changesand geomorphic responses. Instead of landward erosion of the beach and deposition in the nearshore per the Bruunmodel, the response of this coast to ongoing sea-level rise involves high onshore sand transfer, accretion andforedune migration that provide a buffer against wave attack. This preliminary examination sets the geomorphicstage for a larger, interdisciplinary study on sea-level rise impacts that threaten communities, ecological reserves,cultural sites and critical infrastructure in the region.

(Proceedings of the 8thInternational Coastal Symposium),

-1

-1

-1

ADDITIONALINDEX WORDS: Climate change, sea-level rise, coastal erosion, Queen Charlotte Islands.

ABSTRACT

‡Geological Survey of Canada - Pacific,Institute of Ocean SciencesSidney, British ColumbiaV8L 4B2 [email protected]

Journal of Coastal Research Special Issue 39, 2006,

of 3.2°C in January (high 5.6°C, low 0.7°C) and mild summertemperatures of 15°C inAugust (high 17.9°C, low 12.1°C).

Coastal systems in this region of the NE Pacific experienceseasonally opposed winds that are strongest in winter months.In the fall, the Aleutian Low pressure system develops andintensifies causing a counter clockwise circulation of surfacewinds. This brings moisture-laden S-SE winds up Hecate Straitto Haida Gwaii from October through April. Come May, theAleutian Low declines and retreats to the NW while the NorthPacific High expands and intensifies. This shifts the windregime from a dominant SE mode (winter) to a W-NW mode in

summer (Fig. 1). Annual average wind speeds are 8.5 m s inHecate Strait with less than 1% calm conditions ( .,1993). For the period 1995-1999, 67% of winds recorded at theRose Spit meteorological station were above the accepted sand

transport threshold of 6 m s ( 1979). Maximum

gusts often exceed 160 km hr and a maximum wind of 113 km

hr (storm force) was recorded. These are some of the strongest,most persistent, and most competent recorded winds in Canada.

Hecate Strait is also known for its severe wave conditions.Annual significant wave height, H , for the area is 1.8 m. The

peak (most energetic) period is 10 s with waves of H <3 m being

the most frequent. The most active season is from November

(H =2.8 m) through January (H =2.6 m) with maximum

observed Hs of 14.3 m in December ( ., 1993). Highervalues (H > 3.5 m) occur in the shallower waters of Dogfish

Banks along East Beach and prevail for 20-30% of the timeduring winter months ( 1981). The dominant wavedirection in Hecate Strait is S-SW though SE winds and onshorerefraction in shallower waters cause SE waves on East Beach.On North Beach, NW swell and storm waves are comparativelyinfrequent and less energetic.

Currents in Hecate Strait and Dixon Entrance are drivenmainly by tidal forcing with coastal topographic effects onlylocally important ( ., 1995, . 2002).Tides are semi-diurnal mixed and range 5-7 m with HHWMTexceeding 7 m. Reversing flood-ebb tidal currents occur aswater floods into Hecate Strait around Rose Spit from DixonEntrance, then reverses at high tide. Typical flood current

speeds are 0.25-0.5 m s near Rose Spit and 0.15 m s on the ebbalong East Beach ( ., 1995). Two gyres exist: aclockwise gyre centred on Rose Spit and a counter-clockwisegyre centred near Cape Ball. Thus, a current divergence existson East Beach that shifts S on the ebb while northward currentsincrease on the northern 20-30 km of the beach. In DixonEntrance, the Rose Spit eddy rotates counter-clockwise andeastward on North Beach ( and 1987) at

speeds of 0.06-0.11 m s ( 1987). River discharge andwinds appear to have minimal effect on tidal currents exceptduring intense storms when wind-shear currents can approach

0.25 m s ( ., 2002). ., (1995) show thatstorm-enhanced currents transport significant amounts ofsediment from East Beach, around Rose Spit to North Beach.

Aregional sea level regression on the northern Pacific marginof Canada began after the late Wisconsinan glacial maximumca. 16000 yr BP. Crustal flexure and rebound due to aglacioisostatic forebulge caused a rapid sea level regression ca.

14600-12400 C yr BP to -150 m in southern Hecate Strait( . 1997) and to as much as -100 m in northernHecate Strait ( and 1999, 2002,

2003) (Fig. 2). At this time, much ofHecate Strait along East Beach was subaerially exposed and

Wind and Wave Regime

Tides and Currents

Sea Level History and Landscape Response

-1

-1

-1

-1

-1 -1

-1

-1

14

EID

FRYBERGER,

EID

THOMSON,

AMOS STRONG

AMOS

CRAWFORD GREISMAN,

FLATHER,

STRONG AMOS

JOSENHANS

BARRIE CONWAY,HETHERINGTON

et al

et al

al et al

et al

et al et al

et al

et al.,

s

s

s s

s

ET

Geomorphology and sea-level rise on one of Canada's most 'sensitive' coasts: Northeast Graham Island, British Columbia

Figure 1. Study region on northeastern Graham Island, HaidaGwaii (Queen Charlotte Islands), British Columbia, Canada.Wind rose and sand drift potential rose are shown.

Figure 2. Generalised sea-level curves for the Queen CharlotteIslands and for the Northern Hecate Strait modified from

(1982), and (2000),and (1999, 2002) and (2003).CLAGUE FEDJE JOSENHANS BARRIE

CONWAY HETHERINGTON

et al.et al.

Figure 3. 1980 airphoto (NAPL #A25613-39) showing relictprograding shorelines of Naikoon Peninsula over the Holocene.Recent, gradual progradation to the W shown by closer ridges onNorth Beach.Area of coverage for Figs. 4 and 5 also shown.


221

may have provided a corridor for early human migration( ., 1995). Following this, eustatic rise andsubsidence of the forebulge caused rapid sea level transgression

to +16 m by 8900 C yr BP ( and , 2000).Drowned shorelines, barrier islands and wave cut scarps exist asmuch as 100 m below modern sea level ( ., 1984,

and 2002). Since 3800 C yr BP sea level hasregressed, leaving a series of relict shorelines extending NNWfrom Argonaut Hill that demarcate distinct progradationalphases over the late Holocene (Fig. 3). The extent and spacingof the shorelines indicates that a significant amount of sedimentwas available for spit platform growth. Gradual isostaticadjustments, rather than abrupt tectonic motions are thought tohave controlled spacing of the earlier phases ( and

2002). Closer ridges to the W side of Rose Spit

indicate a more recent phase that may have began 1100 C yr BP( ., 1982) due to tectonic uplift ( 1980).However, a transect survey of 7 beach ridges extending 0.4 kminland at South Beach indicates foredune progradation at astable datum. This is an important distinction as the term 'beachridge' is poorly defined ( 2000) and the role of aeolianprocesses is often understated ( 1984, 2002). Opticaldating of relict shorelines is underway to ascertain the timingand extent of past sea level changes.

Over 100km of sedimentary shoreline exist on NE GrahamIsland that is host to a variety of landforms including:dissipative beaches backed by prograding foredunes on NorthBeach; reflective cuspate cobble beaches with a sandy low tideterrace on NW Rose Spit; and multiple-barred beaches backedby migrating foredunes and parabolic dunes on East Beach. TheEast Beach coastline is maintained by active littoral transport oferoded bluff sediments from Capes Ball and Fife by strong waveand tidal currents ( 1995). Figure 4 showsnearshore bars, a log drift line demarcating the berm crest, 5-10m foredunes with blowouts and parabolic dunes migrating NW

(Right) up to 1.2 km into muskeg on an old shore platform.

Currently, this coast is retreating by 1-3 m a ( and1994) and greater during extreme events (e.g., El Niño

1997-98, and 2002). However, erosion (i.e.,sand loss from the beach) appears to be localized to bluffsections with narrow backshores while accretion (though notprogradation) via aeolian transfer is widespread in areas withwide backshores and foredunes (Fig. 4b).

Nearshore sediments are stored and cycled in shore-parallelbar systems on East Beach that are maintained by intense swashaction and strong longshore tidal currents. In some areas,beaches are slightly rhythmic with rip currents. Bars arereworked periodically by storm waves ( . and

2002) and on northern East Beach are shore-attached.These bars provide enhanced sand supply to backshore dunesystems (Fig. 5) similar to that observed by (2000) onthe coast of France. Comparison of 1980 and 1997 airphotosshows migration of this bar complex about 5km N toward CapeFife. On the leading edge of these bars, erosion of beach andbluff systems occurs as wave energy is focused onshore and lesssand is available in the nearshore to dissipate this energy. This,coupled with erosion by high storm waves, may have led to thedrainage of Kumara Lake between 1980 and 1994 (and 1994). Interestingly, 1954 and 1966 photos alsoshow the lake drained without bars present suggesting that dunegrowth and subsequent erosion during extreme storms may be arecurring process in lake maintenance.

In contrast to the eroding shores of East Beach, the

dissipative shores of North Beach are prograding at 0.3-0.6 m a( 1980). Much of this sediment originates from EastBeach rather than offshore from Dixon Entrance ( .,1995). During SE storms, northward alongshore transport onEast Beach is funneled through a nearshore channel westwardaround Rose Spit. Sand is then worked onshore to North Beachby NW waves and winds.

The convergence of North Beach and East Beach forms one ofCanada's most spectacular coastal landscapes, Rose Spit (Fig.6), that is known for its ecological and cultural significance.Rose Point extends several kilometres north into DixonEntrance as a flat cobble-gravel plain capped by aeolian sands.It is bounded on East Beach by 1-5 m foredunes that grade towave cut scarps toward the end of the plain. Beyond this, RoseSpit proper extends another 3 km as a steep-faced intertidalcobble beach where sediment-rich currents from East Beachcollide with the waters of Dixon Entrance. During SE storms,currents flow westward to North Beach as reflected in thewestern curve of the spit. On the NW side, beaches are coarser

JOSENHANS

FEDJE JOSENHANS

WARNER

BARRIE CONWAY

BARRIE

CONWAY

CLAGUE HARPER,

OTVOS,HESP,

AMOS

CONWAY

BARRIE,BARRIE CONWAY,

ANTHONY

ORFORD,

ANTHONY

CONWAY

BARRIE

HARPER,AMOS

et al

et al

et al

et al.,

cf

et al

14

14

14

-1

-1

Modern Geomorphologyand SedimentDynamics

Figure 4. a) East Beach looking S from Kumara Lake/Cape Fiferegion. Treed ridge behind dunes (right) is a former shoreface. b)Migrating 5-10 m foredunes with blowouts and incipient dune attoe. Seaward, an incipient dune also exists on the driftwood jam.

Figure 5. Comparison of 1980 (a) and 1997 (b) airphotosshowing shore-attached bars on East Beach. Bar system hasmigrated northward 5 km and Kumara Lake has drained overthis period.

a)

b)

Walker and Barrie


222

and more reflective with a distinct sandy low tide terrace. Ascarped, cuspate cobble berm formed at spring tide is evidentwith one to two smaller terraces formed at lower tide stages.Terraces diminish westward on North Beach as the systembecomes more dissipative. This transition is distinct and maydemarcate the eastern extent of onshore sand delivery from EastBeach. There appears to be little alongshore transport on NorthBeach eastward to Rose Spit, which would explain the highaccretion rates and sediment coarsening toward the spit.

fer during wave attack. Foredune scarpsbehind drift jams indicate that complete removal could occurduring high storms. Once re-deposited however, accretionwould resume and logjams would fortify the beach betweensignificant storms. The role of such backshore storagecomponents and onshore aeolian delivery in general is oftenoverlooked in assessments of coastal sedimentary dynamics.

The second function of driftwood is as nuclei for incipientdune formation. Windblown sand from the beach is deposited asshadow dunes within the drift matrix and, provided depositioncontinues in the absence of wave erosion, the matrix fills andvegetation establishes. Dune establishment may take severalseasons in areas with wide, log-jammed backshores andrequires high onshore sand transport and low frequency ofdestructive wave surges. On North Beach where driftwood isless dense and the backshore is less exposed to wave attack,dune development is more continuous. Unlike vegetatedincipient foredunes ( 1984, 1989), sizable incipient dunescan develop in driftwood jams before vegetation colonizationoccurs. Figure 7b shows an incipient driftwood dune seaward ofan established foredune vegetated with dunegrass (

). Generations of Sitka spruce ( ) havecolonized older dunes and provide a chronology of beachprogradation. Dune development via this process is key to theprogradation and stabilization of this coast.

The third impact of driftwood is to dam river discharge and/orpreserve backshore swales. The resulting lakes are common atthe base of foredunes along East Beach (Fig. 4a). Aeolianaccretion appears to be key in the development of these featuresand, in some areas, sand has subsequently filled backshorelakes. Rapid drainage by channel incision or wave erosion andrupture of the drift jam is common. High onshore sand transport,

Coupled with shifting nearshore bars causes deranged drainageon East Beach.

Much of the littoral sediment transported along the shores ofGraham Island is moved landward by strong winds into activedune systems; this despite a wet climate and dense forest cover.The wind regime is highly energetic with transporting windsoccurring 67% of the time for the period 1995-99. To assesssand drift potential, the (1979) model was usedwith wind data for this period from the Rose Spitmeteorological station. Resulting DP values (in vector units,VU) are plotted as a drift rose (Fig. 1). The total drift potentialfor the region of 4566 VU is well above those documented fordesert regions (80-489, 1979) or for the Canadianprairies (300-1600, and 1999) and surpassesthose for the Netherlands coast (1700-4000, 2000). Theresultant drift potential vector (RDP) is 2967 VU to 316 (NW)which reflects the dominant SE winds in the regime. This isslightly less than an average dune alignment of 336 ± 8°(determined from airphoto analysis of 94 parabolic dunes on thecoast) which may reflect steering from a secondary (W-NW)mode in the regime from summer winds. Foredune migrationrates are on the order of metres per year while parabolic dunesmove slower (metres per decade).

and (1999) suggest that in high-energyprairie environments sand supply, not wind, is the limitingfactor for dune migration. Most prairie dunes are 'closed' to newinputs and rely largely on reworked sand eroded locally fromstabilized dunes for maintenance. On NE Graham Islandhowever, dunes receive high sand supply moved onshore viafrequent competent winds. Instead soil moisture, which limitssand transport at low concentrations ( and1995), and vegetation cover may be more important at limitingaeolian activity, especially beyond the backshore environment.

Dendrochronological (tree-ring) records provide high-resolution accounts of past climate. For instance, Mountainhemlock ( ) at treeline on southern Graham

Impacts of Driftwood on Beach-dune Systems

AeolianActivity

Dendroclimatology and dendrogeomorphology

Driftwood, mostly felled timber, litters backshoreenvironments along East and North Beach (Fig. 7). Byproviding a major store for sediment in the backshore, logjamsserve three important geomorphic functions. First, logjams actas 'accretion anchors' that fortify the backshore by storingsignificant amounts of aeolian sand. In some areas, rapidaccretion and/or complete burial of the drift jam have occurred.This stabilizes the beach by providing a store of sand and debristhat is reworked as a buf

HESP,

FRYBERGER,

FRYBERGER,WOLFE LEMMEN,

TSOAR,

WOLFE LEMMEN,

NAMIKAS SHERMAN,

Elymusmollis Picea sitchensis

Tsuga mertensiana

Figure 7. Driftwood jams act as accretion anchors and stabilizebackshore environments along East Beach near Rose Spit (a).On North Beach, logs act as nuclei for incipient foredunegrowth (b).

Figure 6. 1980 airphoto of Rose Spit (NAPL #A25558-116,inset) and oblique photo showing mixing of sediment-richwaters from East Beach (left) under NW swell. Rose Bar shownin forefront.

a)

b)



223

Island is sensitive to June-July temperature and coastal Shorepine ( ) is very sensitive to seasonal precipitation( 1999). This latter record shows notable droughts in thelate 1940s and early 1950s. These preliminary reconstructionsemphasize the dynamic nature of climate in Haida Gwaii and,while there is great variability in the controls of forest growth, itis likely that climate variations are reflected in existing foreststructure.

Next to temperature and precipitation, no other climaticfactor has a greater impact on tree growth than wind( 1996). Aeolian processes impact trees viaabrasion, burial or deflation and application of wind stress andsediment loads. This produces distinct physiological responsesthat include tilting, flagging, bark removal, adventitious roots,reduced and/or contorted growth, and death. Impacts arewidespread on NE Graham Island including: dwarf krummholzSitka spruce, tilted and buried trees on foredunes, 'ghost forests'killed by dune migration, exhumed adventitious roots formedby rapid accretion and subsequent deflation, prop rootbuttressing, and abraded trunks. Dendrogeomorphic techniquesare being applied to identify periods of dune activity andmigration ( and 1992). This, withdendroclimatological evidence, can provide proxy records ofcoastal landscape responses to climate variability that couldextend back several hundreds of years.

The Geological Survey of Canada (GSC) defines 'sensitivity'as the degree to which a rise in sea level would initiate oraccelerate coastal geomorphic changes given local conditionsand other climate change effects (e.g., increased storminess)( 1998). Admittedly, this definition overlooksecological and socio-economic factors such as the capacity ofimpacted communities to adjust and adapt ( .1992, and 2003). Identified impacts includeincreasing tidal inundation, storm surge flooding, coastalretreat, erosion and accretion. All of these affect coastal formand function and in some areas are changing at geologicallyrapid rates. Recognizing potential hazards to coastalcommunities and ecosystems, the GSC mapped a 'sensitivityindex' for the entire Canadian coastline ( 1998). Theindex is a function of: rock type, relief, landforms present, sealevel change, wave height, tidal range and shoreline change.Most of the Canadian coast (67%) has a low sensitivity, 30% ismoderately sensitive and 3% is highly sensitive. The coast ofBritish Columbia (10.5% of the Canadian coast) has amoderate-low sensitivity due to a prevalence of steep, rockyshores. Exceptions include the Fraser Delta, Vancouver and NEGraham Island, which has a mean sensitivity value of 13 (low3.2, high 25) and ranks among the most highly sensitive inCanada.

The northeast Pacific experiences dramatic variability inclimate driven, in part, by two large-scale modes of variability:the El Niño/Southern Oscillation (ENSO) and the PacificInterdecadal Oscillation (PDO). The warm phase of the PDO ischaracterized by an enhanced wintertime Aleutian Low andwarmer water along the west coast of NorthAmerica (

. 1997). The cool phase is approximately opposite andtransitions between phases are abrupt and occur approximatelyevery 25 years ( and 2001). The extratropicalexpression of ENSO is similar, with El Niño events resemblingthe warm phase of the PDO. Strong ENSO events recur every 3to 7 years and rarely last longer than one year ( and

, 1986). Recent ENSO events (1982-83, 1997-98)produced enhanced storm waves and winds that caused intenseerosion along the Pacific coast from California to Washington( ., 2000, and 2001, and

2002). To date, little is known about the impacts of

such modes of variability on the shores of Graham Island.During El Niño 1997-98, and (2002)observed a regional sea-level rise of 0.4 m, and enhanced stormwaves and winds that caused 12 m of retreat. Though stormswere more intense this season due to the enhanced AleutianLow, it is uncertain as to whether they occurred more frequently.Because the effects of PDO and ENSO are additive( and , 1998) and the PDO may have beenin the cold phase ( and , 2000), it is possible thatfuture events may have greater impact, especially ifsuperimposed on a rising sea level. Furthermore, some modelspredict future scenarios with a semi-permanent ENSO-likestate ( ., 1999), which would entail morefrequent impacts.

The dominant view on how coastlines will respond to sea-level rise has been framed by the (1962) model, whichsuggests that shorelines will undergo landward retreat due towave erosion of the upper beach and consequent deposition ofsediment in the nearshore to a depth that is equal to the increasein sea level (S).As such, shoreline retreat is simply a function ofbeach slope and rise in sea level and is in the range of 50-100Sfor most beaches ( 89, 1991).Davidson- (2005) challenges the assumptions of theBruun model and proposes a model that considers onshore sandtransfer to beach-dune systems with no net transfer to thenearshore. By way of sand delivery to the backshore, this modelshows preservation and landward migration of foredunes as aresponse to sea-level rise. This appears to be the case on East

Beach where, under sea-level rise of 1.6 mm a , periodic wave

scarping and net shoreline retreat of 1-3 m a , high onshore sandtransfer is able to rapidly replenish backshore storage, restoreand maintain actively accreting and migrating foredunes (Fig.4b). Thus, 'erosion' in the broadest sense per the Bruun model(i.e., coastline retreat and sand loss from the beach) does notseem to be the current response of this coastline to ongoing sea-level rise. Rather, erosion is localized to bluffs with narrowbackshores while sand accretion in log jammed backshores andmigrating foredunes is widespread. What remains to be seen ishow potential increases in storm frequency and magnitude andrates of sea-level rise would affect this response.

Despite a moist climate and dense vegetation, aeolian activityis high and sand is moved onshore by frequent transportingwinds into foredunes and driftwood dunes that stabilize thebackshore. Foredunes migrate on the order of metres per yearwhile large parabolic dunes (up to 1.2km long) move slower

Pinus contorta

cf.

et al.

cf et al.

et al.,

etal

et al

et al

SMITH

SCHWEINGRUBER,

MARIN FILION,

SHAW

LUITZEN

SMIT PILIFOSOVA,

SHAW

ZHANG

GEDALOF SMITH,

YARNAL

DIAZ

STORLAZZI DINGLER REISS, ALLAN

KOMAR,

BARRIE CONWAY,

GERSHUNOV BARNETT

HARE MANTUA

TIMMERMANN

BRUUN,

SCOR WORKING GROUP

ARNOTT,

A 'SENSITIVE' COAST

Climate Change and Potential Sea-level RiseImpacts

CONCLUSIONS

-1

-1

This paper documents the coastal geomorphology of one ofCanada's most dynamic coastlines: NE Graham Island, QueenCharlotte Islands (Haida Gwaii). This area has undergonesignificant sea-level changes (-150 to +16 m) over the Holocene

and is currently rising at +1.6 mm a . Relict shorelines andprogradational ridges indicate landscape response to former sealevels. Given a macrotidal range, energetic wave climate andongoing erosion, the Geological Survey of Canada identifiesthis as one of Canada's most sensitive coasts to future sea-levelrise.

Over 100 km of sedimentary beach systems are maintainedby strong littoral transport by waves and tidal currents.Northward longshore transport on East Beach heads westwardaround Rose Spit to North Beach, particularly during SE

storms. Retreat of 1-3 m a and tens of metres in extreme years(e.g., 1997-98 El Niño) occur on East Beach, which providessand to shore-attached bars that migrate northward and enhanceboth onshore sand supply to beach-dune systems at low tidestages, as well as beach erosion on their leading edge due to alack of available sand and focus of wave energy. Reflectivecuspate terraces on NW Rose Spit diminish westward on NorthBeach as the system becomes more dissipative and sandaccretion predominates. This transition is distinct and maydemarcate the extent of sand delivery from East Beach.Eastward longshore transport toward Rose Spit is minimal.

-1

-1

Walker and Barrie


224

(metres per decade). Dune alignment (336 8) is skewed to theresultant drift vector (316°) perhaps due to reworking bysummer winds. Dendrochronological geomorphic evidenceshows the impact of aeolian processes on trees including tilting,flagging, burial, trunk scarring, exhumed adventitious roots,dwarf growth and ghost forests killed by dune migration.Dendrochronological evidence is being used to provide insightinto climate variability and landscape responses over the pastseveral hundred years.

Contrary to the Bruun model (i.e., landward erosion of thebeach-dune system and nearshore deposition), this region isresponding to ongoing erosion and sea-level rise by retreatingon East Beach whilst maintaining high onshore aeolian sandtransfer and accretion in driftwood jams and migratingforedunes that stabilize the shore against wave erosion. Thedissipative shores of North beach continue to prograde rapidly

at 0.3-0.6 m a .The preliminary assessment provided in this paper sets the

geomorphic background for larger, interdisciplinary study ofthe vulnerability of this coast to climate change and sea-levelrise impacts that threatens communities, cultural sites,resources, ecological reserves, and critical infrastructure in theregion.

This project is facilitated with financial support fromNSERC and the Government of Canada's Climate ChangeAction Fund to IJW and contributes to Natural ResourcesCanada's Climate Change program. Impacts and AdaptationsField assistance from K. Pearce, J. Axelson, J. Anderson and W.Zantvoort is appreciated, as is logistical support from BC Parksand the Council of the Haida Nation. Contributions by Drs.Stephen Wolfe (GSC-Ottawa) and Dan J. Smith and Ze'evGedalof (UVic Tree Ring Laboratory) are also recognized.

D.S. and I.J. (2005).Meteorological and oceanographic responses to climatevariability and change on the North Coast of BC."Proceedings of the 12th Canadian Coastal Conference, 6-9November 2005, Dartmouth, Nova Scotia.

. and ., 2002. Extreme storms on thePacific northwest coast during the 1997-8 El Niño and 1998-9 La Niña. 18, 175-193.

C.L., J.V. and J.T., 1995. Storm-enhanced sand transport in a macrotidal setting, QueenCharlotte Islands, British Columbia, Canada. In:

B.W., and A. (eds.),. Special

Publication of the International Association ofSedimentologists 24, 53-68.

E.J., 2000. Marine sand supply and Holocenecoastal sedimentation in northern France between the Seineestuary and Belgium. In: K. and J.R.L. (eds.),

. Special Publicationof the Geological Society of London. London, pp. 87-97.

E.J. and J.D., 2002. Between Wave- andTide-Dominated Coasts: the Middle Ground Revisited.

SI 36, 8-15.J.V. and K.W., 1999. Late Quaternary

Glaciation and Postglacial Stratigraphy of the NorthernPacific Margin of Canada. , 51, 113-123.

J.V. and K., 2002. Rapid sea level changesand coastal evolution on the Pacific margin of Canada.

, 150, 171-183.P., 1962. Sea Level Rise as a Cause of Shore Erosion.

, 88, 116-130.J.J., R.W. and B.G., 1982. Late

Quaternary geology of eastern Graham Island, QueenCharlotte Islands, British Columbia.

, 19, 1786-1795.K.W. and J.V., 1994. Coastal erosion on the

east coast of Graham Island, Queen Charlotte Islands,British Columbia. , 53-58.

W.R. and P., 1987. Investigation ofpermanent eddies in Dixon Entrance, British Columbia.

, 7(8), 851-870.R.G.D., 2005. Conceptual Model of the

Effects of Sea Level Rise on Sandy Coasts. Journal osCoastal Research, 21, 1166-1172.

J.R. and T.E., 2001. Changes to Monterey Baybeaches from the end of the 1982-83 El Niño through the1997-98 El Niño. , 3029, 1-15.

A.H. and I.J., 2006. Understanding coastalvulnerability to climate change related risks.

, SI 39 (Proceedings of the 8 InternationalCoastal Symposium), March 14-17, 2004. Itajaí, SC, Brazil.

B., C., M. and B., 1993.Wind and wave climate atlas volume IV: The west coast ofCanada. , Halifax.

D.W. and H.W., 2000. Drowned forests andarchaeology on the continental shelf of British Columbia,Canada. , 28, 99-102.

R.A., 1987. A tidal model of the northeast Pacific., 25(1), 22-45.

S.G., 1979. Dune forms and wind regime. In:E.D. (ed.), . USGS

Professional Paper 1052, 137-169.Z. and D.J., 2001. Interdecadal climate

variability and regime-scale shifts in Pacific NorthAmerica., 28, 1515-1518.

A. and T.P., 1998. Interdecadalmodulation of teleconnections.

, 79, 2715-2725.S.R. and N.J., 2000. Empirical evidence for

North Pacific regime shifts in 1977 and 1989., 24, 691-699.

J.R., 1980. Coastal Processes on Graham Island,Queen Charlotte Islands, British Columbia,

, 13-18.P.A., 1984. The formation of sand "beach ridges" and

foredunes. , 15(9-10), 289-291.P.A., 1989.Areview of biological and geomorphological

processes involved in the initiation and development ofincipient foredunes.

, 96B, 181-201.P.A., 2002. Foredunes and blowouts: initiation,

geomorphology and dynamics. , 48, 245-268.

R., J.V., R.G.B., R.and Smith, D.J. (2003). Paleogeography, glacially-inducedcrustal displacement, and Late Quaternary coastlines on thecontinental shelf of British Columbia, Canada.

, in press., R. and J., 1997.

Early Humans and Rapidly Changing Holocene Sea Levelsin the Queen Charlotte Islands - Hecate Strait BritishColumbia, Canada. , 277, 71-74.

andJ.V., 1995. Post glacial sea levels on the western Canadiancontinental shelf: Evidence for rapid change, extensivesubaerial exposure, and early human habitation.

, 125, 73-94.P. and L., 1992. Recent dynamics of subarctic

dunes as determined by tree-ring analysis of white spruce,Hudson Bay, Quebec. , 38, 316-330.

S.L. and D.J., 1995. A review of theEffects of surface moisture content on aeolian sandtransport. In: V.P. (ed.),

. Chapman and Hall, London, pp. 269-293.E.G., 2000. Beach ridges - definitions and significance.

�

-1

th

ACKNOWLEDGEMENTS

LITERATURE CITED

ABEYSIRIGUNAWARDENA, WALKER,

ALLAN, KOMAR,

AMOS, BARRIE, JUDGE,

FLEMMING, BARTOLOMA,

ANTHONY,

PYE, ALLEN,

ANTHONY, ORFORD,

BARRIE, CONWAY,

BARRIE, CONWAY,

BRUUN,

CLAGUE, MATHEWES, WARNER,

CONWAY, BARRIE,

CRAWFORD, GREISMAN,

DAVIDSON-ARNOTT,

DINGLER, REISS,

DOLAN, WALKER,

EID, CALMAN, HENSCGEL, MCGRATH,

FEDJE, JOSENHANS,

FLATHER,

FRYBERGER,MCKEE,

GEDALOF, SMITH,

GERSHUNOV, BARNETT,ENSO

HARE, MANTUA,

HARPER,

HESP,

HESP,

HESP,

HETHERINGTON, BARRIE, REID, MACLEOD,

JOSENHANS, H., FEDJE, D., PIENITZ SOUTHON,

JOSENHANS, H.W., FEDJE, D.W., CONWAY, K.W. BARRIE,

MARIN, FILION,

NAMIKAS, SHERMAN,

TCHAKERIAN,

OTVOS,

J.C P.D

Journal of Coastal Research,

TidalSignatures in Modern and Ancient Sediments

Coastal and Estuarine Environments - Sedimentology,Geomorphology and Geoarchaeology

Journal of Coastal Research

Quaternary Research

Journal of Sedimentary Geology

ASCE Journal of Waterways and Harbors Div.

Canadian Journal ofEarth Sciences

GSC Current Research 1994-E

Continental Shelf Research

Marine Geology

Journal of

Coastal Research

Transport Canada Report TP10820E

Geology

Atmosphere-Ocean

A Study of Global Sand Seas

Geophysical Research Letters

Bulletin of theAmerican Meteorological Society

Progress inOceanography

GSC CurrentResearch 80-1A

Search

Proceedings of the Royal Society ofEdinburgh

Geomorphology

QuaternaryScience Reviews

Science

MarineGeology

Quaternary Research

Desert AeolianProcesses



225

Walker and Barrie


226

GeomorphologyTree Rings and Environment:

Dendroecology

Journal ofCoastal Research

GSC Report No. 505

Climate change,adaptive capacity and development

Journal of CoastalResearch

BritishColumbia Offshore Hydrocarbon Development: Report ofthe Scientific Review Panel (Volume I).

Institute of Ocean Sciences, Government of Canada, MarineSciences Directorate, Pacific Region, Report No. 81-10

Nature

Israel Journal ofEarth Sciences

Quaternary Research

Science

GSCBulletin 534

Journal ofClimatology

Journal of Climate

, 32, 83-108.F.H., 1996.. Haupt PaulAG Verlag, Berne. 609 p.

, 1991. The response of beaches tosea-level changes:Areview of predictive models.

, 7, 895-921and

S., 1998. Sensitivity of the coasts of Canada tosea-level rise. , Ottawa. 79 p.

B. and O., 2003. From adaptation toadaptive capacity and vulnerability reduction. In:J.B. R.J.T. and S. (eds.),

. World Scientific, RiverEdge, New Jersey, pp. 1-20.

D.J., 1999. Dendroclimatological Investigations in theQueen Charlotte Islands: A Preliminary Report of ResearchActivities, 1999. University of Victoria Tree RingLaboratory, Report No. 1999-08, Victoria. 3 p.

C.D., C.M., and G.B. 2000.Comparative impacts of the 19823 and 19978 El Niñowinters on the central California coast.

16, 10221036.D., P. and D., 2002.

BC Ministry ofEnergy and Mines Report, Victoria. 51 p.

R.E., 1981. An analysis of wind and currentobservations collected in the Queen Charlotte Sound-Hecate Strait-Dixon Entrance region during 1954 and 1955.

,Victoria. 84 p.

. and E., 1999. Increased El Niñofrequency in a climate model forced by future greenhousewarming. , 398, 694-696.

H., 2000. The geomorphological background andpaleogeography of the sand dunes that have formed thekurkar ridges in the coastal plain of Israel.

, 49, 189196., J.J. and R.W., 1984.

Geology and paleoecology of a mid-Wisconsin peat fromthe Queen Charlotte Islands, British Columbia, Canada.

, 21, 337-350.. and , J.J., 1982. Ice-

Free Conditions on the Queen Charlotte Islands, BritishColumbia, at the Height of Late Wisconsin Glaciation.

, 218, 675-677.S.A. and D.S., 1999. Monitoring dune

activity in the Great Sand Hills region, Saskatchewan., 199-210.

B. and H.F., 1986. Relationships betweenextremes of the Southern Oscillation and the winter climateof the Anglo-American Pacific coast.

, 6, 197-219.. and D.S., 1997. ENSO-

like interdecadal variability: 1900-93. ,10, 1004-1020.

SCHWEINGRUBER,

SCOR WORKING GROUP 89

SHAW, J., TAYLOR, R.B., FORBES, D.L., RUZ, M.H.SOLOMON,

SMIT, PILIFOSOVA,SMITH,

KLEIN, HUQ,

SMITH,

STORLAZZI, WILLIS, GRIGGS,

STRONG, GALLAGER, MUGGERIDGE,

THOMSON,

TIMMERMANN, A., OBERHUBER, J., BACHER, A., ESCH, M.,LATIF, M ROECKNER,

TSOAR,

WARNER, B.G., CLAGUE MATHEWES,

WARNER, B.G., MATHEWES, R.W CLAGUE

WOLFE, LEMMEN,

YARNAL, DIAZ,

ZHANG, Y., WALLACE, J.M BATTISTI,

geomorphology and sea-level rise on one of canada's most

Documents