- Email: [email protected]
Large-scale sedimentary bedforms and sediment dynamics on a glaciated tectonic continental shelf: Examples from the Pacific margin of Canada
ARTICLE IN PRESS Continental Shelf Research 29 (2009) 796–806
Contents lists available at ScienceDirect
Continental Shelf Research journal homepage: www.elsevier.com/locate/csr
Large-scale sedimentary bedforms and sediment dynamics on a glaciated tectonic continental shelf: Examples from the Pacific margin of Canada J. Vaughn Barrie a,, Kim W. Conway a, Kim Picard a, H. Gary Greene b a b
Geological Survey of Canada–Pacific, Institute of Ocean Sciences, PO Box 6000, Sidney, British Columbia, Canada V8L 4B2 Tombolo, 2267 Deer Harbor Road, Eastsound, WA 98245, USA
a r t i c l e in f o
a b s t r a c t
Available online 24 December 2008
The Pacific margin of Canada has been subjected to tectonism, dramatic sea level change and vigorous storm and tidal energy since glacial times resulting in a complex seafloor. Extensive multibeam mapping of this shelf has provided an opportunity to understand how these processes have impacted sedimentology and morphology. Bathymetric restriction of the tidally dominated flow between the inland seas and the open Pacific has resulted in the development of very large subaqueous dune fields and terrace moats. For example, in the southern Strait of Georgia nearly symmetrical dunes with wavelengths between 100 and 300 m, dune heights up to 28 m, cover the seafloor in 170–210 m water depth. In northern Hecate Strait a 72 km2 area of large 2D dunes occurs at the transition with Dixon Entrance which opens to the Pacific Ocean and steep (4101) wave-cut terraces and drowned spits, a result of sea level changes during the Holocene, are now being undercut to generate moats 7 m deep, in a narrowing shelf trough. Currents, with velocities ranging between 0.2 and 2.2 m s1, are dominated by semi-diurnal tidal streams that are continually modified by wind and estuarine circulation. There appears to be a clear association of grain size, water depth and flow velocity controlling the size of the subaqueous dunes. Crown Copyright & 2009 Published by Elsevier Ltd. All rights reserved.
Keywords: Subaqueous dunes Wave-cut terraces Multibeam bathymetry Sediment dynamics British Columbia
1. Introduction The seafloor of the northwest Canadian Pacific continental shelf (Fig. 1) has developed under a complex interplay of tectonism, glaciation, sea level change and ocean energy. Each process has a significant impact on the development of the seafloor morphology, but it is the active tidal, storm and estuarine flow energy acting on the seafloor, that in many areas were once sub-aerial, that defines the dynamic morphology of the present continental shelf. Rapid crustal displacement brought on by the advance and retreat of continental and alpine glaciers during the Late Quaternary drove the rapid sea level change (Clague, 1983; Hetherington et al., 2004). Consequently, the position of the coastline changed with respect to the thickness and position of the ice, the elastic thickness of the lithosphere, and local mantle viscosity (Hetherington and Barrie, 2004). Mapping using multibeam bathymetry and high-resolution sub-bottom profiling allows for the continuous 3D imagery of sea level lowstand features such as wave-cut terraces, drowned lakes, river channels, spits and deltas. For example, wave-cut terraces are a dominant legacy feature of the rapid sea level changes and are now found in
Corresponding author. Tel.: +1 250 363 6424; fax: +1 250 363 6739.
E-mail address: [email protected] (J.V. Barrie).
water depths from 40 m to greater than 200 m (Barrie and Conway, 2002a). The same imagery highlights areas of significant sedimentary bedforms and sediment transport. Consequently it is now possible to visualize and model the primary sedimentary processes modifying the seafloor. Very large subaqueous dunes (45 m in height and 100 m in wavelength (Ashley, 1990)) have been found on continental shelves of southeast Africa (Flemming, 1978), South China Sea (Keller and Richards, 1967; Kubicki, 2008), NE Mediterranean Sea, (Lykousis, 2001), Argentina (Aliotta and Perillo, 1987), US Atlantic (Jordan, 1962; Fenster et al., 1990, 2006) and Pacific coasts (Bouma et al., 1980; Barnard et al., 2006), and the western Canadian continental shelf. A dynamic seafloor with the presence of large and mobile sedimentary bedforms can directly impact the design and feasibility of locating engineering structures on the seabed. For example, in 2003 a plan to bring natural gas from mainland of British Columbia to Vancouver Island for power generation involved the placement of a pipeline across the Strait of Georgia. To access the onshore facilities the proposed pipeline was to transit Boundary Pass, a narrow pass separating Canada from the USA (Fig. 2). During surveys to determine a pipeline route, a field of very large dunes was encountered (Barrie et al., 2005) requiring a significant change in the overall routing. In Queen Charlotte Basin, the proposed generation of wind power on Dogfish Bank (Fig. 1b) will require the laying of electrical transmission lines
0278-4343/$ - see front matter Crown Copyright & 2009 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.csr.2008.12.007
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
797
Fig. 1. (a) The regional and tectonic setting of Georgia Basin. The heavy line shows the maximum extent of the late Wisconsin ice (after James et al., 2000). Location of Fig. 2 is shown. (b) Queen Charlotte Basin showing the extent and flow direction of the late Wisconsin ice (Cordilleran ice is the solid line and Queen Charlotte Island ice is the dashed line) during the late Wisconsin maximum (modified from Barrie and Conway, 2002b). Locations of Figs. 6 and 8 are shown.
across Hecate Strait and such cables are very sensitive to burial by sediments because of transmission efficiency issues and thermal insulation of the cables. Consequently, the cables cannot be buried by sediment dunes. Subaqueous dunes can migrate over sig-
nificant distances when the ebb and flood velocities are unequal, due to a superimposed residual current or an M4-related asymmetry (Knaapen et al., 2005). In the case of a residual current, the flow velocities in the direction of the residual current
ARTICLE IN PRESS 798
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Fig. 2. Regional map of the southern portion of Georgia Basin showing the location of the interconnected waterways (Boundary Pass and Haro Strait) between the Strait of Georgia and Juan de Fuca Strait (Fig. 1a). The location of the Boundary Pass dune field, the current meters in Boundary Pass and Fig. 5 are shown.
are greater than the velocities in the opposite direction, resulting in sediment transport and dune migration (Nemeth et al., 2002). In addition, bathymetric steering both controls the direction of sediment transport and enhances the current. Our objective here is to demonstrate how the seafloor morphology, the product of tectonism, glaciation and sea level change, controls the sediment dynamics, and development and migration of large sedimentary bedforms on the Pacific margin of Canada. By using multibeam bathymetry imagery over a broad area of the western Canadian continental shelf it is possible to visualize the morphology of the marine basin, equivalent to that provided by high-resolution topographic maps on land.
2. Physical setting 2.1. Bathymetric and tectonic setting The inner continental shelf of the Pacific coast of Canada is composed of two sedimentary basins, Georgia Basin (GB) and Queen Charlotte Basin (QCB). Georgia Basin, which consists primarily of the Strait of Georgia, is an enclosed basin that stretches for approximately 220 km in a northwest-southeast direction between southern British Columbia, Vancouver Island and Washington State (Fig. 1a). The strait connects with the open sea in the south, first through the Gulf Islands and San Juan Islands, and then through Juan de Fuca Strait (Fig. 1a). Bottom
topography in the Gulf/San Juan Islands area (Fig. 2) is complex but mostly shallower than 100 m, except for two narrow, channels. Subsidence of the basin began in the late Cretaceous (85 million years ago). The tectonic regime over the last 40 million years has been dominated by subduction of the Juan de Fuca plate. The North American plate is presently overriding the oceanic Juan de Fuca plate at a rate of about 45 mm/yr1 (Riddihough and Hyndman, 1991). Queen Charlotte Basin is a semi-enclosed basin bordered to the east by the British Columbia mainland and to the west by the Queen Charlotte Islands (Fig. 1b). Hecate Strait opens to the Pacific in the south and southwest through Queen Charlotte Sound and to the north through Dixon Entrance (Fig. 1b). Tectonism gave rise to the existence and location of the Queen Charlotte Islands and adjacent basin. The underlying Pacific plate slides northward and slightly into North America at an average rate of 50–60 mm a1 along the seismically active Queen Charlotte Fault (Riddihough, 1988; Rohr et al., 2000). At the junction of the two plates this motion is accompanied by frequent earthquakes that vary in intensity along the fault (Fig. 1b).
2.2. Glaciation Glaciation affected the Pacific margin of Canada many times, although extensive evidence has been found for only the youngest glacial episode over much of the area (Barrie and Conway, 2002b).
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
The Fraser Glaciation began approximately 25,000–30,000 years ago (Clague, 1989). Ice moving south from the Coast Mountains of the Canadian Cordillera and Vancouver Island progressively coalesced, joined ice coming out of the Fraser Valley, and then divided into two tongues, one following Juan de Fuca Strait toward the ocean and the other going southward into Puget Sound (Fig. 1a). This large glacier reached the south end of the Puget Lowland and the western Juan de Fuca Strait at its maximum extent (Waitt and Thorson, 1983; Hewitt and Mosher, 2001), about 14,000 C14 BP (Porter and Swanson, 1998). In Queen Charlotte Basin, ice from the massive Cordilleran sheet extended westward across northern Hecate Strait and through Dixon Entrance and coalesced with ice from the Queen Charlotte Islands, deflecting it westward along Dixon Entrance (Sutherland-Brown, 1968; Hicock and Fuller, 1995; Barrie and Conway, 1999). Ice also moved south down the central trough in Hecate Strait (Barrie and Bornhold, 1989) coalescing with ice coming through the troughs of Queen Charlotte Sound to the edge of the shelf and met the ice in northern GB (Luternauer and Murray, 1983; Luternauer et al., 1989; Josenhans et al., 1995). Glaciation in QCB reached its maximum extent sometime after 21,000 14C yr B.P. (Blaise et al., 1990). 2.3. Sea level history Deglaciation was quite different in both basins. Most of GB was ice-free by 11,300 C14 BP (Barrie and Conway, 2002b). Deglaciation was very rapid with regional downwasting and widespread stagnation (Guilbault et al., 2003). In QCB rapid glacial retreat began sometime after 15,000 14C yr BP and ice had largely left the offshore by 13,500–13,000 14C yr BP (Barrie and Conway, 1999). With deglaciation in QCB a regional lowering of sea level occurred with sea levels falling in central Hecate Strait to a lowstand of 150 m by 13,000 14C years BP (Josenhans et al., 1995; Barrie and Conway, 2002a; Hetherington et al., 2004). Subsequent transgression was rapid (5 cm/yr (Josenhans et al., 1997)) with sea levels reaching the present shoreline on the Queen Charlotte Islands by about 9100 14C yr BP (Clague et al., 1982; Josenhans et al., 1995, 1997; Fedje and Josenhans, 2000). For GB sea levels remained high after deglaciation and fell to near or just below present between 12,000 and 10,000 14C years BP, before returning to present levels (Barrie and Conway, 2002b). 2.4. Oceanography The near and offshore of British Columbia is mesotidal to macrotidal with superimposed storm and estuarine circulation. Georgia Basin is characteristic of a partially mixed estuary with moderately strong tidal currents (2.6–3.4 m tidal range), seasonally varying stratification and late summer and late winter deepwater density intrusions (LeBlond, 1983; Crean and Ages, 1971; Thomson, 1994; Masson, 2002). The Fraser River discharge reaches a maximum of 10,000 m3/s during the spring freshet and a minimum of around 1,000 m3/s in late winter. The Fraser River accounts for about 73% of the mean annual freshwater discharge of 158 109 m3 into the Strait of Georgia (Johannessen et al., 2003). This freshwater influx then forces estuarine circulation in the southern strait that is characterized by a net outflow of low-salinity water towards Juan de Fuca Strait in the upper layer (o50 m depth) and a net northward inflow of highsalinity water in the lower part of the water column that reaches the Strait of Georgia in late summer (Mosher and Thomson, 2000). The strong currents in Boundary Pass and Haro Strait mix the water column and reduce stratification (Masson and Cummins, 2004) and the turbulence associated with super-critical flow, that
799
has been observed, contributes to particle resuspension (Johannessen et al., 2006). Two current meters were installed 10 km to the southwest of the Boundary Pass (Fig. 2) by Fisheries and Oceans Canada for two years at 110 and 125 m water depth. Maximum current velocities of 1.94 ms1 occurred in a northeast flood direction with the highest recorded ebb velocities of 1.29 ms1. A second mooring put in for less than 2 months just northeast of the Fisheries and Oceans mooring measured velocities up to 2.23 ms1 within the water column (ASL Environmental Services, 2001). In QCB, strong rectilinear tidal currents, usually in the order of 15–25 ms1, are classified as mixed, mainly semi-diurnal, with a 4.5 m tidal range (macrotidal) in the central portion of Hecate Strait (Thomson, 1981). The tidal crest enters Queen Charlotte Sound and spreads northward into Hecate Strait where it encounters the opposing crest that entered eastward through Dixon Entrance (Fig. 1b). Normally, tidal currents flow along the orientation of the Strait at nearly uniform speed at all depths, but where abrupt changes in the seafloor bathymetry occur currents are enhanced (Sinclair et al., 2005). In addition to tidal forcing, circulation in Hecate Strait is also affected by winds and river runoff. The influence of these two factors are generally out of phase, with river runoff (Skeena River (Fig. 1b)) dominating circulation in late spring through early summer and wind forcing having its greatest influence during the fall and winter months. Consequently, winter circulation is both tidal and wind driven with a net northward component resulting from the prevailing southeasterly storm winds (Crawford and Thomson, 1991). Current velocities exceeding 0.80 ms1, measured 15 m above the seabed, occur during winter storms and spring tides (Barrie and Bornhold, 1989). Most of the Pacific coast continental shelf of Canada is sediment starved with sediment capture within the coastal fjords and inlets. The one exception is in the southern Strait of Georgia where sedimentation from the Fraser River dominates the surficial geology with Holocene sediment thicknesses varying from zero on Pleistocene ridges to greater than 300 m within the basin (Mosher and Hamilton, 1998). Present day sedimentation rates vary from 10 cm/a near the river mouth to less than 3 cm/a in the distal parts of the prodelta (Hart et al., 1998).
3. Methods Some 12,000 km2 of multibeam swath bathymetry coverage was obtained on the Pacific coast of Canada since 2000. The multibeam coverage was undertaken to obtain contiguous full seafloor coverage of the deeper water areas of GB and areas of highest interest in QCB. The surveys were carried out from the CCGS Vector, and Otter Bay by the Canadian Hydrographic Service, in cooperation with the Geological Survey of Canada, using a hullmounted Kongsberg-Simrad EM1002 system, which operates at a frequency of 95 kHz utilizing 127 beams, for the deeper regions of the basin (450 m water depth) and a hull-mounted KongsbergSimrad EM3002 system, which operates at a frequency of 300 kHz utilizing 121–135 beams, for the shallow parts of the basin (o50 m water depth). In most of the areas, the tracks were positioned so as to insonify 100% of the seafloor with a 100% overlap. Positioning was by broadcast differential DGPS and the multibeam data were corrected for sound speed variations in the stratified water column using frequent sound speed casts. The data were corrected for tides and then edited for spurious bathymetric and navigational points using CARIS software. The gridded data were exported as ASCII files and imported into ArcInfo software for processing and image production. Analyses of repetitive dune field surveys utilized ArcInfo software routines.
ARTICLE IN PRESS 800
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Positioning using DGPS has a positional error of up to 5 m and vertical errors in the order of 1% water depth (International Hydrographic Organization, 1998). To improve positional accuracy, distinctive points of bedrock where used to assess accuracy where repeat surveys were undertaken. Regardless, the overall accuracy that can be used to interpret dynamic changes of the dunes was 5 m horizontal and 0.2 m vertical. The multibeam swath images formed the preliminary interpretive framework for delineation of sedimentary bedforms, drowned coastal features and modified seafloor morphology. Based on this interpretation several areas were identified for further investigation using a Huntec DTS high-resolution subbottom profiler (Mosher and Simpkin, 1999), sidescan sonar, and a Phantom ROV. Sediment samples were collected using a Shipek grab sampler within the subaqueous dune and wave-cut terrace areas. These were subsequently dried, split and then analysed for texture using a settling tube for the sand fraction (o2.0 mm to 40.063 mm) and a Sedigraph for the mud fraction (o0.063 mm).
crested to sinuous morphologies. The largest of the dunes measures 28 m in height and 760 m in ridge crest length (Figs. 3 and 4). Bedform steepness is about 1/12 (height/wavelength) for each of the larger dunes, resulting in slopes of 121. Similar submarine sand dunes with wave heights of up to 25 m have been identified in the eastern Juan de Fuca Strait (Mosher and Thomson, 2000), again located at the transition between a narrow
4. Results 4.1. Subaqueous dunes Large sedimentary bedforms are commonly observed within the multibeam imagery but several areas stand out due to the shear size of the bedforms and the extensive nature of the bedform fields. Two significant areas are the region between the southern Strait of Georgia and Juan de Fuca Strait in GB and the northern Hecate Strait region of QCB. Both areas represent narrow channels entering Straits which are open to the Pacific Ocean. In Boundary Pass, along the Canada/US boundary (Fig. 2), there is a series of nearly symmetrical, very large subaqueous dunes with wavelengths between 100 and 300 m, covering an area of 2.5 km2 in 170–210 m water depth (Figs. 3 and 4). Dune heights vary between 5 and 28 m. Based on ROV observations, smaller bedforms (0.25–1 m high) are superimposed on the much larger dunes. They consist of well-sorted coarse sand and gravel and the sediments have a complete absence of epibenthic organisms. The grain size varies from very coarse sand on the slopes and crest and fine gravel in the troughs (Wentworth Scale). These features are hydrodynamically formed of an unknown age and rank amongst the largest ever observed globally. The dunes display straight
Fig. 4. Difference map between November 2001 and September 2006 of dune movement and crest line change of the Boundary Pass very large dunes with an inset showing the detailed multibeam image of the dune field. Black line indicates location of Fig. 3.
Fig. 3. Huntec DTS sub-bottom profile through the subaqueous dune field in Boundary Pass (location of line shown in Fig. 4). Notice the reflectors indicating clinoform bedding on the northern most two dunes suggesting migration to the southwest.
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
801
Fig. 5. Multibeam image of the southern portion of Haro Strait and eastern Juan de Fuca Strait (Fig. 2) illustrating the two very large subaqueous dune fields, either side of the narrow passage between the two water bodies. Notice the similarity of the overall geometry between these dune fields and that shown in Fig. 4.
channel and inlet that opens to the Pacific (Fig. 5). In both cases, the dunes are classified as linear, slightly asymmetrical to symmetrical, stoss-erosional, lee-depositional, two dimensional, very large dunes (classification nomenclature of Ashley (1990)). The two dune fields also have a very similar distribution. Both fields have a tear drop or elliptical distribution with the largest dunes at the widest point of the ellipse and decreasing dune size in all dimensions to the narrowing point of the tear drop shape. However, the Juan de Fuca dune field contains two very large dunes, whereas the Boundary Pass field contains only one. Yet another tear shaped dune field of similar aerial size as the Boundary Pass and Juan de Fuca dune fields occurs just north of the entrance of Haro Strait into Juan de Fuca Strait (Fig. 5). These dunes have heights of 2–7 m and wavelengths from 60 to 80 m. In northern Hecate Strait a 72 km2 bedform field of large subaqueous dunes occurs in 88 to 96 m water depth. Though the bedform field appears to be located within a fairly open area of Hecate Strait, it occurs within a north-south oriented trough in east-central Hecate Strait (Figs. 1b and 6). The 80 m contour essentially outlines the field which is bordered on the west by the very shallow (20–40 m average depth) Dogfish Bank, to the east by a significant slope to the nearshore zone and to the north by a 70 m sill separating Hecate Strait and Dixon Entrance to the north
(Figs. 1b and 6). To the south the trough widens and deepens into Hecate Strait (Fig. 6). The field of bedforms appears to occur in a shallow erosional basin that contains two southward extending drowned, progradational spits (Fig. 6). Progradational spits are a common feature of Hecate Strait (Josenhans et al., 1995) and developed as sea level rose just after deglaciation, when an ample supply of sediment existed (Barrie and Conway, 2002a). These relict sediments provide the source for the development of the observed dune field. The dunes range in wavelength from 40 to 175 m, but the majority of the dune field have wavelengths of 50–70 m and lengths to 3 km (Fig. 7). Dune height is consistently between 2 and 3 m (Fig. 7). All but a small area of 3D linguoid dunes with wavelengths of 100 m are 2D large dunes (classification nomenclature of Ashley (1990)). The sands making up the dunes are primarily well-sorted medium sands.
4.2. Repetitive surveys Two geophysical surveys were completed for the Boundary Pass dune field in December 2001 and April 2003. Little difference in the overall morphology can be seen between the surveys, though the lines were not coincident. In both surveys the dunes in
ARTICLE IN PRESS 802
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Fig. 6. Multibeam image of the extensive subaqueous dune field in northern Hecate Strait (Fig. 1b). Notice the two drowned paleo-spits extending to the south, formed when sea levels were lower.
2006. The larger dunes showed minimal clockwise rotation, but the total migration was less than the error level for multibeam surveys, based on DGPS. However, the smaller dunes on the western end of the dune field showed up to 25 m of eastward migration toward the larger dunes (Fig. 5).
4.3. Wave-cut terraces
Fig. 7. A 3D image of the north Hecate Strait dunes shown in relation to the Huntec DTS sub-bottom profile. The multibeam has a 10 exaggeration. Location of the figure is shown by the black line in Fig. 6.
the northern part of the field appear to be slightly asymmetrical with movement to the southwest, while the majority of the dunes are symmetrical. On the western side of the dune field several of the dunes show a slight asymmetry suggesting movement to the northeast. The dunes rest on a highly reflective surface under which there is little sub-bottom penetration. The dunes at the north end of the field show sub-bottom clinoforms suggesting progradation to the southwest (Fig. 3). The dune field was initially surveyed in November 2001 and has been resurveyed by multibeam sonar three times since (October 2003, September 2004 and September 2006). The overall change between surveys is a clockwise rotation with up to 50 m of movement along the outer portions of the dunes (Fig. 4). The dunes in eastern Juan de Fuca were initially surveyed in 1998 (Mosher and Thomson, 2000) and resurveyed again in September
Within central Hecate Strait the multibeam imagery clearly identifies several wave-cut terraces in water depths from 37 m to greater than 160 m (Fig. 8). The dominant terrace is oriented almost directly north-south extending for over 120 km, before it arches around to the northwest terminating in a drowned re-curved paleo-spit (Barrie and Conway, 2009). The break in slope at the top of the terrace is defined by the 126 m contour, and the base of the terrace along most of its length terminates in a moat in water depths of between 140 and 164 m. The moat that parallels the base of the terrace is a dominant feature of the terrace morphology (Fig. 9). The moat along the main north south terrace varies in depth from 2 to 7 m, measured from the berm of the outside wall across to the base of the terrace slope, and is from 100 to 350 m wide. The outer berm roughly represents the volume of exhumed material. Slopes are as high as 121 on the terrace slope into the moat and 151 on the inside of the moat berm wall. Sediments at the base of the moat consist of fine to medium sands, whereas the terrace slope and top consist of lag gravel over medium to coarse sand. The moat berm sediments are more variable, but primarily consist of medium well-sorted sands. The moat around the drowned spit is continuous suggesting bathymetric control of seafloor currents (Fig. 8). In many areas along the berm, evidence exists for instability both on the inside and outside slopes (Barrie and Conway, 2009). For example the end of the drowned spit, at the northern end of the terrace, appears to have failed in the recent past, since failure splays are clearly seen extending east of the failure (Fig. 8). These are now being modified into hydrodynamic bedforms as the moat is again reforming. Subaqueous dunes are found adjacent to the base of the terrace slope in several areas and direction of
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
803
Fig. 8. Multibeam image of the extensive wave-cut terraces in central Hecate Strait (Fig. 1b) with the distinctive moats. At the northern end of the set of terraces is a paleospit with a surrounding moat.
Fig. 9. A Huntec DTS sub-bottom profile illustrating the stratigraphic cross-section of a typical wave-cut terrace with adjacent moat in central Hecate Strait (Fig. 1b). The cross section is located in Fig. 8 as A to A0 .
movement is to the north. The dunes are characterised by heights of 2–6 m and wavelengths of 40–80 m.
5. Discussion 5.1. Seabed morphology and sediment dynamics Three examples of sediment dynamics modifying the continental shelf of British Columbia have been discussed in detail. Each area has been modified as a result of the sediment transport
dynamics in a meso to macrotidal environment with superimposed wave and estuarine-induced current flow. The other primary characteristic controlling the environments is the overall bathymetric morphology that restricts flow within the basins. All three areas occur within narrowing basins that close at shallow sills. The British Columbia continental shelf is characterised by marine basins that are interconnected (Fig. 1) and open to the Pacific through geologically controlled straits, a result of modification of the seafloor by glaciers and terrestrial processes. Tidal exchange into the basin in many cases has to take place through narrow restricted passages, often with shallow sills, resulting in
ARTICLE IN PRESS 804
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
tidal reinforcement. The two areas that host significant bedforms occur just south of a sill that separates two straits (Strait of Georgia and Haro Strait in GB and Hecate Strait and Dixon Entrance in QCB) and the terrace/moat system occurs in a narrowing basin, with the north Hecate Strait sill at the head of the basin. In addition, the very large dunes in Juan de Fuca Strait (Mosher and Thomson, 2000) occur just seaward of the narrow opening that divides Haro Strait and Juan de Fuca Strait and another tear shaped dune field occurs at the southern end of Haro Strait where it meets Juan de Fuca Strait, close to its narrowest point (Figs. 2 and 5). The bedform fields form in depressions that likely result from erosion of the underlying Quaternary sediments. The proposed interpretation is that morphological restriction enhances current flow resulting in the erosion of the substrate providing the sediment for the development of subaqueous dunes. This usually occurs where the flow is coming over or into a sill or within a rapidly narrowing channel or trough. If enough sediment within the sand to fine gravel size exists, subaqueous dunes develop. Where bathymetric control of current flow is pronounced, extensive dune fields commonly develop as observed in Bungo Channel, Japan (Ikehara, 1998), Torres Strait off northern Australia (Harris, 1988), Malacca Strait in Southeast Asia (Keller and Richards, 1967), and at the mouth of San Francisco Bay (Barnard et al., 2006). Similarly, the development of moats adjacent to wave-cut terraces appears to occur within the narrowing central trough of Hecate Strait. Based on the observed surficial geology (Barrie and Conway, 2009), evidence of slope instability, and the known tidal and storm driven environment, the moats appear to be actively eroding. Although, no current measurements are available from within the moat system, interpretations of sediment transport processes (sediment texture and the development of subaqueous dunes) from morphological observations strongly support this interpretation. The flood tide, reinforced by the southeasterly storms, is perhaps the dominant sediment transport process. It is interesting to note that a moat has developed on all sides of the spit feature suggesting that contour forcing of the currents is a dominant process modifying the surficial geology. Whether the system has reached some level of equilibrium in sediment transport is not clear, though the evidence of failure along the wave-cut terrace and spit strongly suggests that erosion is ongoing. In particular, the obvious reworking of a failure at the end of the drowned spit appears to be reforming a moat quite rapidly (Barrie and Conway, 2009).
5.2. Large subaqueous dune migration Using the currents observed from the two moorings in Boundary Pass (Fig. 2) and the mean sediment grain size (0.7–2.0 mm) as inputs into the Ashley (1990) classification of large-scale subaqueous bedforms, it is clear that these dunes are mobile. Current velocities over the dunes would be expected to be even higher than those recorded by these two moorings (Fig. 2) because of topographic constriction of the flow to the east and because the moorings sites were off to one side of the main channel axis. It is likely that flood tides and estuarine circulation, particularly during summer, construct the dunes. If we assume a maximum velocity at 100 cm above the seafloor of 1.94 m1 (current meter data) then sediment sizes up to about 12–15 mm could be mobilized (Miller et al., 1977). Furthermore, the absence of any epibenthic organisms in the area of the bedforms strongly suggests that these sediments are frequently in motion. The critical threshold value of sediment transport of these coarse sediments would likely be reduced considering the bed roughness and the bathymetric restriction to flow.
Over the six years of repetitive multibeam measurement there is no clear net direction of movement, crestal flexing and rotation is the long term feature of wave dynamics, similar to the giant dune field in eastern Long Island Sound (Fenster et al., 1990, 2006). Plots of superimposed crest lines over the six year period suggest field-wide clockwise rotation of crest lines (Fig. 4). In essence, the large dunes have moved as non-rigid rotational bodies (Fenster et al., 2006) and maintain the same geographical location. The clockwise rotation was approximately 51 and up to 101 on some dunes with no apparent overall migration. The dune field appears to maintain the same overall geometry and volume. As these dunes appear to be largely stationary, with minor crest flexure, a limited volume of sand and gravel must move within a system that apparently gains or loses little sediment. This suggests that the sediments that form such large bedforms are partially derived through in situ erosion into the underlying and adjacent seabed. The scoured and hollowed morphology of the areas where the dune fields occur would indicate that erosion is ongoing. The lack of observed migration of the larger bedforms would also indicate that sediments introduced to the system are for the most part trapped in a tidally driven, oscillatory system that has allowed little net transport out of the area. The symmetrical nature of the dunes would support this conjecture. There appears to be a positive correlation between the mean grain size and the position across the very large flow-transverse bedforms, from moderately sorted fine gravel (42.0 mm) in the troughs to moderately sorted very coarse sand (0.71–1.4 mm) mid-dune to well-sorted coarse sand (0.71–1.0 mm) near the crest. A similar grain size trend was shown for a dune field in the eastern North Sea off the Danish west coast (Anthony and Leth, 2002). There also seems to be a strong correlation between mean grain size and dune height. The influence of grain size on dune shape is complex with dune height generally increasing as the grain size increases (Bartholdy et al., 2005), however previous studies suggest that this only occurs until about 0.5 mm, after which dune height decreases as grain size continues to increase (Allen, 1982; Dalrymple and Rhodes, 1995). In addition, dune height and wavelength generally increase in size with an increase in water depth (Allen, 1982; Ashley, 1990) and the grain size/wavelength relation in deep-water dunes favours coarse sediments (Flemming, 2000b). Flemming (2000a) suggests that in deep-water conditions, dunes will continue to grow as the critical threshold for sediment movement increases with increasing flow velocity. As critical threshold for sediment movement increases so will the grain size, consequently, very large dunes should be composed of very coarse sediments. He suggests that dunes with mean grain size greater than 0.5 mm can have dune heights greater than 24 m and wavelengths of approximately 380 m. This clearly corresponds with the Boundary Pass dunes where heights exceed 24 m, the mean grain size is 0.7–2.0 mm and the wavelengths reach over 300 m, slightly less than predicted (Flemming, 2000a). The north Hecate Strait dune field, that is composed of medium sands, do not exceed 3 m height and 75 m wavelength, again closely fitting the model proposed by Flemming (2000a). For the Boundary Pass Dunes the maximum current is approximately 1.90 ms1 allowing for initiation of sediment transport of fine gravels, while for the north Hecate Strait dune field the approximate maximum currents of 0.40–0.80 ms1 allow for the initiation of sediment transport of medium sands.
6. Conclusions Very large subaqueous dunes on the Pacific margin of Canada have formed as a result of a unique combination factors.
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Diurnal tides, estuarine circulation and bathymetric restriction have resulted in enhanced current flow, particularly where the seafloor morphology changes from a narrowing restricted channel into a body of water open to the Pacific. Grain size, water depth and flow velocity control the size of the dunes which have heights up to 28 m. In addition, sea level lowstand wave-cut terraces that formed in unconsolidated glacial and postglacial sediments are presently being eroded by these same enhanced currents to form moats up to 7 m deep in a narrowing shelf trough. The unique seafloor morphology that controls the sediment dynamics is a result of the tectonic, glacial and sea level history over the late Quaternary. Acknowledgements The multibeam data set would not exist without the field program management, data collection, and processing capability of the Canadian Hydrographic Service, Pacific. In particular, we would like to acknowledge the valuable support of the Rob Hare, Ernest Sargent and Peter Milner. Alex Shaw, Sean Mullan, and Robert Kung provided invaluable assistance in the development of the multibeam graphics. The manuscript was improved by the critical revision of Chuck Nittrrouer and one anonymous reviewer. Support for this work was through the Geoscience for Ocean Management Program of the Geological Survey of Canada, SeaDoc Society and Moss Landing Marine Laboratories. References Aliotta, S., Perillo, G.M.E., 1987. A sand wave field in the entrance to Bahia Blanca Estuary, Argentina. Marine Geology 76, 1–14. Allen, J.R.L., 1982. Sedimentary Structures—Their Characteristics and Physical Basis, Vol. 1. Development in Sedimentology, Vol. 30A. Elsevier, New York, 593 p. Anthony, D., Leth, J.O., 2002. Large-scale bedforms, sediment distribution and sand mobility in the eastern North Sea off the Danish west coast. Marine Geology 182, 247–263. Ashley, G.M., 1990. Classification of large-scale subaqueous bedforms: a new look at an old problem. SEPM Bedforms and Bedding Structures Research Symposium. Journal of Sedimentary Petrology 60, 160–172. ASL Environmental Services, 2001. Data Report: Oceanographic Current Measurements; Boundary Passage, BC, Canada; June 21–August 11, 2000. Report submitted to Aker Engineering and Williams Gas Pipeline, 83 p. Barnard, P.L., Hanes, D.M., Rubin, D.M., Kvitek, R.G., 2006. Giant sand waves at the mouth of San Francisco Bay. Eos 87 (29), 285–289. Barrie, J.V., Bornhold, B.D., 1989. Surficial geology of Hecate Strait, British Columbia continental shelf. Canadian Journal of Earth Sciences 26, 1241-1254. Barrie, J.V., Conway, K.W., 1999. Late Quaternary glaciation and postglacial stratigraphy of the northern Pacific margin of Canada. Quaternary Research 51, 113–123. Barrie, J.V., Conway, K.W., 2002a. Rapid sea level changes and coastal evolution on the Pacific margin of Canada. Sedimentary Geology 150, 171–183. Barrie J.V., Conway K.W., 2002b. Contrasting glacial sedimentation processes and sea-level changes in two adjacent basins on the Pacific margin of Canada. In: Dowdeswell, J., O’Cofaigh, C., (Eds.), Glacier-influenced sedimentation on highlatitude continental margins. Geological Society of London, Special Publication 203, pp. 181–194. Barrie, J.V., Conway, K.W., 2009. Paleogeographic reconstruction of Hecate Strait British Columbia: Changing sea levels and sedimentary processes reshape a glaciated shelf. International Association of Sedimentologists Special Publication, in press. Barrie, J.V., Hill, P.R., Conway, K.W., Iwanowska, K., Picard, K., 2005. Georgia Basin: seabed features and marine geohazards. Geoscience Canada 32, 145–156. Bartholdy, J., Flemming, B.W., Bartholoma, A., Ernstsen, V.B., 2005. Flow and grain size control of depth-independent simple subaqueous dunes. Journal of Geophysical Research 110 (F04S16), 12 p. Blaise, B., Clague, J.J., Mathewes, R.W., 1990. Time of maximum Late Wisconsin glaciation, west coast of Canada. Quaternary Research 34, 282–295. Bouma, A.H., Rappeport, M.L., Orlando, R.C., Hampton, M.A., 1980. Identification of bedforms in lower Cook Inlet, Alaska. Sedimentary Geology 26, 157–177. Clague, J.J., 1983. Glacio-isostatic effects of the Cordilleran Ice Sheet, British Columbia, Canada. In: Smith, D.E., Dawson, A.G. (Eds.), Shorelines and Isostasy. Institute of British Geographers Special Publication 16. pp. 321–343. Clague, J.J., 1989. Quaternary geology of the Canadian Cordillera. In: Fulton R.J. (Ed.), Chapter 1, Quaternary Geology of Canada and Greenland. Geological Survey of Canada, Geology of Canada No. 1, pp. 17–95.
805
Clague, J.J., Harper, J.R., Hebda, R.J., Howes, D.E., 1982. Late Quaternary sea levels and crustal movements, coastal British Columbia. Canadian Journal of Earth Sciences 19, 597–618. Crawford, W.R., Thomson, R.E., 1991. Physical oceanography of the western Canadian continental shelf. Continental Shelf Research 11, 669–683. Crean, P.B., Ages, A., 1971. Oceanographic records from twelve cruises in the Strait of Georgia and Juan de Fuca Strait. Department of Energy, Mines and Resources, Canada 1, 55 p. Dalrymple, R.W., Rhodes, R.N., 1995. Estuarine dunes and bars. In: Perillo, G.M.E. (Ed.), Geomorphology and Sedimentology on Estuaries, Developments in Sedimentology 53. Elsevier Science, Amsterdam, pp. 359–422. Fedje, D.W., Josenhans, H., 2000. Drowned forests and archaeology on the continental shelf of British Columbia, Canada. Geology 28, 99–102. Fenster, M.S., FitzGerald, D.M., Bohlen, W.F., Lewis, R.S., Baldwin, C.T., 1990. Stability of giant sand waves in eastern Long Island Sound, USA. Marine Geology 91, 207–225. Fenster, M.S., FitzGerald, D.M., Moore, M.S., 2006. Assessing decadal-scale changes to a giant sand wave field in eastern Long Island sound. Geology 34, 89–92. Flemming, B.W., 1978. Underwater sand dunes along the southeast African continental margin—observations and implications. Marine Geology 26, 177–198. Flemming, B.W., 2000a. The role of grain size, water depth and flow velocity as scaling factors controlling the size of subaqueous dunes. In: Trentesaux, A., Garlan, T. (Eds.), Marine Sandwave Dynamics, International Workshop, March 2000, University of Lille, France, pp. 55–60. Flemming, B.W., 2000b. On the dimensional adjustment of subaqueous dunes in response to changing flow conditions: a conceptual process model. In: Trentesaux, A., Garlan, T. (Eds.), Marine Sandwave Dynamics, International Workshop, March 2000, University of Lille, France, pp. 61–67. Guilbault, J.-P., Barrie, J.V., Conway, K.W., Lapointe, M., Radi, T., 2003. Paleoenvironments associated with the deglaciation process in the Strait of Georgia off British Columbia: microfaunal and microfloral evidence. Quaternary Science Reviews 22, 839–857. Harris, P.T., 1988. Sediments, bedforms and bedload transport pathways on the continental shelf adjacent to Torres Strait, Australia-Papua New Guinea. Continental Shelf Research 8, 979–1003. Hart, B.S., Hamilton, T.S., Barrie, J.V., 1998. Sedimentation on the Fraser Delta slope and prodelta, Canada, based on high-resolution seismic stratigraphy, lithofacies and 137Cs fallout stratigraphy. Journal of Sedimentary Research 68, 556–568. Hetherington, R., Barrie, J.V., 2004. Interaction between local tectonics and glacial unloading on the Pacific margin of Canada. Quaternary International 120, 65–77. Hetherington, R., Barrie, J.V., Reid, R.G.B., MacLeod, R., Smith, D.J., 2004. Paleogeography, glacially-induced displacement, and late Quaternary coastlines on the continental shelf of British Columbia, Canada. Quaternary Science Reviews 23, 295–318. Hewitt, A.T., Mosher, D.C., 2001. Late Quaternary stratigraphy and seafloor geology of the eastern Juan de Fuca Strait, British Columbia and Washington. Marine Geology 177, 295–316. Hicock, S.R., Fuller, E.A., 1995. Lobal interactions, rheologic superposition, and implications for a Pleistocene ice stream on the continental shelf of British Columbia. Geomorphology 14, 167–184. Ikehara, K., 1998. Sequence stratigraphy of tidal sand bodies in the Bungo Channel, southwest Japan. Sedimentary Geology 122, 233–244. International Hydrographic Organization, 1998. IHO standards for hydrographic surveys. International Hydrographic Bureau, Monaco, International Hydrographic Organization Special Publication no. 44. James, T.S., Clague, J.J., Wang, K., Hutchinson, I., 2000. Postglacial rebound at the northern Cascadia subduction zone. Quaternary Science Review 19, 1527–1541. Johannessen, S.C., Macdonald, R.W., Paton, D.W., 2003. A sediment and organic carbon budget for the greater Strait of Georgia. Estuarine Coastal and Shelf Science 56, 845–860. Johannessen, S.C., Masson, D., Macdonad, R.W., 2006. Distribution and cycling of suspended particles inferred from transmissivity in the Strait of Georgia, Haro Strait and Juan de Fuca Strait. Atmosphere-Ocean 44, 17–27. Jordan, G.F., 1962. Large submarine sand waves. Science 136, 839–848. Josenhans, H.W., Fedje, D.W., Conway, K.W., Barrie, J.V., 1995. Post glacial sea-levels on the western Canadian continental shelf: Evidence for rapid change, extensive subaerial exposure, and early human habitation. Marine Geology 125, 73–94. Josenhans, H., Fedje, D., Pienitz, R., Southon, J., 1997. Early humans and rapidly changing sea levels in the Queen Charlotte Islands-Hecate Strait, British Columbia. Science 277, 71–74. Keller, G.H., Richards, A.F., 1967. Sediments of Malacca Strait, Southeast Asia. Journal of Sedimentary Petrology 37, 102–127. Knaapen, M.A.F., van Bergen Henegouw, C.N., Hu, Y.Y., 2005. Quantifying bedform migration using multi-beam sonar. Geo-Marine Letters 25, 306–314. Kubicki, A., 2008. Large and very large subaqueous dunes on the continental shelf off southern Vietnam, South China Sea. Geo-Marine Letters 28, 229–238. LeBlond, P.H., 1983. The Strait of Georgia: functional anatomy of a coastal sea. Canadian Journal of Fisheries and Aquatic Sciences 40, 1033–1063. Luternauer, J.L., Murray, J.W., 1983. Late Quaternary morphologic development and sedimentation, central British Columbia continental shelf. Geological Survey of Canada, Paper 83–21.
ARTICLE IN PRESS 806
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Luternauer, J.L., Clague, J.J., Conway, K.W., Barrie, J.V., Blaise, B., Mathewes, R.W., 1989. Late Pleistocene terrestrial deposits on the continental shelf of western Canada: evidence for rapid sea-level change at the end of the last glaciation. Geology 17, 357–360. Lykousis, V., 2001. Subaqueous bedforms on the Cyclades Plateau (NE Mediterranean)-evidence on Cretan deep water formation. Continental Shelf Research 21, 495–507. Masson, D., 2002. Deep water renewal in the Strait of Georgia. Estuarine, Coastal and Shelf Science 54, 15–126. Masson, D., Cummins, P.F., 2004. Fortnightly modulation of the estuarine circulation in Juan de Fuca Strait. Journal of Marine Research 58, 439–463. Miller, M.C., McCave, I.N., Komar, P.D., 1977. Threshold of sediment motion under unidirectional currents. Sedimentology 24, 507–527. Mosher, D.C., Hamilton, T.S., 1998. Morphology, structure and stratigraphy of the offshore Fraser delta and adjacent Strait of Georgia. In: Clague, J.J., Luternauer, J.L., Mosher, D.C. (Eds.), Geology and natural hazards of the Fraser River Delta, British Columbia. Geological Survey of Canada Bulletin 525, pp. 147–160. Mosher, D.C., Simpkin, P.G., 1999. Status and trends on marine high-resolution seismic reflection profiling: data acquisition. Geoscience Canada 25, 174–188. Mosher, D.C., Thomson, R.E., 2000. Massive submarine sand dunes in the eastern Juan de Fuca Strait, British Columbia. In: Trentesaux, A., Garlan, T. (Eds.), Marine Sandwave Dynamics, International Workshop, March 2000, University of Lille, France, pp. 131–142. Nemeth, A.A., Hulscher, S.J.M.H., de Vriend, H.J., 2002. Modeling sand wave migration in shallow shelf seas. Continental Shelf Research 22, 2739–2762. Porter, S.C., Swanson, T.W., 1998. Radiocarbon age constraints on rates of advance and retreat of the Puget lobe of the Cordilleran ice sheet during the last glaciation. Quaternary Research 50, 205–213.
Riddihough, R.P., 1988. The northeast Pacific Ocean and margin. In: Nairn, A.E.M., Stehli, F.W., Uyeda, S. (Eds.), The Ocean Basins and Margins, The Pacific Ocean, Vol. 7B. Plenum, New York, pp. 85–118. Riddihough, R.P., Hyndman, R.D., 1991. Modern plate tectonic regime of the continental margin of western Canada. In: Gabrielse, H., Yorath, C.J. (Eds.), Geology of the Cordilleran Orogen in Canada. Geological Survey of Canada, Geology of Canada No 4, Chapter 13, pp. 435–455. Rohr, K.M., Scheidhauer, M., Trehu, A.M., 2000. Transpression between two warm mafic plates: The Queen Charlotte Fault revisited. Journal of Geophysical Research 105, 8147–8172. Sinclair, A.F., Conway, K.W., Crawford, W.R., 2005. Associations between bathymetric, geologic and oceanographic features and the distribution of the British Columbia trawl fishery. ICES Cm 2005/L:25, 31p. Sutherland-Brown, A., 1968. Geology of Queen Charlotte Islands, British Columbia. British Columbia Department of Mines and Petroleum Resources Bulletin 54, Victoria, British Columbia. Thomson, R.E., 1981. Oceanography of the British Columbia coast. Canadian Special Publications, Canadian Fisheries and Aquatic Sciences, 56. Thomson, R.R., 1994. Physical oceanography of the Strait of Georgia-Puget Sound-Juan de Fuca Strait system. In: Wilson, R.H., Beamish, R.J., Aitkens, F., Bell, J. (Eds.), Review of the marine environment and biota of Strait of Georgia, Puget Sound and Juan de Fuca Strait. Proceedings of the BC/Washington Symposium on the Marine Environment, January 1994, pp. 36–100. Waitt, R.B., Thorson, R.M., 1983. The Cordilleran ice sheet in Washington, Idaho, and Montana. In: Porter, S.C. (Ed.), Late-Quaternary Environments of the United States Vol. 1, The Late Pleistocene. University of Minnesota Press, Minneapolis, Minnesota, pp. 53–70.
Contents lists available at ScienceDirect
Continental Shelf Research journal homepage: www.elsevier.com/locate/csr
Large-scale sedimentary bedforms and sediment dynamics on a glaciated tectonic continental shelf: Examples from the Pacific margin of Canada J. Vaughn Barrie a,, Kim W. Conway a, Kim Picard a, H. Gary Greene b a b
Geological Survey of Canada–Pacific, Institute of Ocean Sciences, PO Box 6000, Sidney, British Columbia, Canada V8L 4B2 Tombolo, 2267 Deer Harbor Road, Eastsound, WA 98245, USA
a r t i c l e in f o
a b s t r a c t
Available online 24 December 2008
The Pacific margin of Canada has been subjected to tectonism, dramatic sea level change and vigorous storm and tidal energy since glacial times resulting in a complex seafloor. Extensive multibeam mapping of this shelf has provided an opportunity to understand how these processes have impacted sedimentology and morphology. Bathymetric restriction of the tidally dominated flow between the inland seas and the open Pacific has resulted in the development of very large subaqueous dune fields and terrace moats. For example, in the southern Strait of Georgia nearly symmetrical dunes with wavelengths between 100 and 300 m, dune heights up to 28 m, cover the seafloor in 170–210 m water depth. In northern Hecate Strait a 72 km2 area of large 2D dunes occurs at the transition with Dixon Entrance which opens to the Pacific Ocean and steep (4101) wave-cut terraces and drowned spits, a result of sea level changes during the Holocene, are now being undercut to generate moats 7 m deep, in a narrowing shelf trough. Currents, with velocities ranging between 0.2 and 2.2 m s1, are dominated by semi-diurnal tidal streams that are continually modified by wind and estuarine circulation. There appears to be a clear association of grain size, water depth and flow velocity controlling the size of the subaqueous dunes. Crown Copyright & 2009 Published by Elsevier Ltd. All rights reserved.
Keywords: Subaqueous dunes Wave-cut terraces Multibeam bathymetry Sediment dynamics British Columbia
1. Introduction The seafloor of the northwest Canadian Pacific continental shelf (Fig. 1) has developed under a complex interplay of tectonism, glaciation, sea level change and ocean energy. Each process has a significant impact on the development of the seafloor morphology, but it is the active tidal, storm and estuarine flow energy acting on the seafloor, that in many areas were once sub-aerial, that defines the dynamic morphology of the present continental shelf. Rapid crustal displacement brought on by the advance and retreat of continental and alpine glaciers during the Late Quaternary drove the rapid sea level change (Clague, 1983; Hetherington et al., 2004). Consequently, the position of the coastline changed with respect to the thickness and position of the ice, the elastic thickness of the lithosphere, and local mantle viscosity (Hetherington and Barrie, 2004). Mapping using multibeam bathymetry and high-resolution sub-bottom profiling allows for the continuous 3D imagery of sea level lowstand features such as wave-cut terraces, drowned lakes, river channels, spits and deltas. For example, wave-cut terraces are a dominant legacy feature of the rapid sea level changes and are now found in
Corresponding author. Tel.: +1 250 363 6424; fax: +1 250 363 6739.
E-mail address: [email protected] (J.V. Barrie).
water depths from 40 m to greater than 200 m (Barrie and Conway, 2002a). The same imagery highlights areas of significant sedimentary bedforms and sediment transport. Consequently it is now possible to visualize and model the primary sedimentary processes modifying the seafloor. Very large subaqueous dunes (45 m in height and 100 m in wavelength (Ashley, 1990)) have been found on continental shelves of southeast Africa (Flemming, 1978), South China Sea (Keller and Richards, 1967; Kubicki, 2008), NE Mediterranean Sea, (Lykousis, 2001), Argentina (Aliotta and Perillo, 1987), US Atlantic (Jordan, 1962; Fenster et al., 1990, 2006) and Pacific coasts (Bouma et al., 1980; Barnard et al., 2006), and the western Canadian continental shelf. A dynamic seafloor with the presence of large and mobile sedimentary bedforms can directly impact the design and feasibility of locating engineering structures on the seabed. For example, in 2003 a plan to bring natural gas from mainland of British Columbia to Vancouver Island for power generation involved the placement of a pipeline across the Strait of Georgia. To access the onshore facilities the proposed pipeline was to transit Boundary Pass, a narrow pass separating Canada from the USA (Fig. 2). During surveys to determine a pipeline route, a field of very large dunes was encountered (Barrie et al., 2005) requiring a significant change in the overall routing. In Queen Charlotte Basin, the proposed generation of wind power on Dogfish Bank (Fig. 1b) will require the laying of electrical transmission lines
0278-4343/$ - see front matter Crown Copyright & 2009 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.csr.2008.12.007
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
797
Fig. 1. (a) The regional and tectonic setting of Georgia Basin. The heavy line shows the maximum extent of the late Wisconsin ice (after James et al., 2000). Location of Fig. 2 is shown. (b) Queen Charlotte Basin showing the extent and flow direction of the late Wisconsin ice (Cordilleran ice is the solid line and Queen Charlotte Island ice is the dashed line) during the late Wisconsin maximum (modified from Barrie and Conway, 2002b). Locations of Figs. 6 and 8 are shown.
across Hecate Strait and such cables are very sensitive to burial by sediments because of transmission efficiency issues and thermal insulation of the cables. Consequently, the cables cannot be buried by sediment dunes. Subaqueous dunes can migrate over sig-
nificant distances when the ebb and flood velocities are unequal, due to a superimposed residual current or an M4-related asymmetry (Knaapen et al., 2005). In the case of a residual current, the flow velocities in the direction of the residual current
ARTICLE IN PRESS 798
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Fig. 2. Regional map of the southern portion of Georgia Basin showing the location of the interconnected waterways (Boundary Pass and Haro Strait) between the Strait of Georgia and Juan de Fuca Strait (Fig. 1a). The location of the Boundary Pass dune field, the current meters in Boundary Pass and Fig. 5 are shown.
are greater than the velocities in the opposite direction, resulting in sediment transport and dune migration (Nemeth et al., 2002). In addition, bathymetric steering both controls the direction of sediment transport and enhances the current. Our objective here is to demonstrate how the seafloor morphology, the product of tectonism, glaciation and sea level change, controls the sediment dynamics, and development and migration of large sedimentary bedforms on the Pacific margin of Canada. By using multibeam bathymetry imagery over a broad area of the western Canadian continental shelf it is possible to visualize the morphology of the marine basin, equivalent to that provided by high-resolution topographic maps on land.
2. Physical setting 2.1. Bathymetric and tectonic setting The inner continental shelf of the Pacific coast of Canada is composed of two sedimentary basins, Georgia Basin (GB) and Queen Charlotte Basin (QCB). Georgia Basin, which consists primarily of the Strait of Georgia, is an enclosed basin that stretches for approximately 220 km in a northwest-southeast direction between southern British Columbia, Vancouver Island and Washington State (Fig. 1a). The strait connects with the open sea in the south, first through the Gulf Islands and San Juan Islands, and then through Juan de Fuca Strait (Fig. 1a). Bottom
topography in the Gulf/San Juan Islands area (Fig. 2) is complex but mostly shallower than 100 m, except for two narrow, channels. Subsidence of the basin began in the late Cretaceous (85 million years ago). The tectonic regime over the last 40 million years has been dominated by subduction of the Juan de Fuca plate. The North American plate is presently overriding the oceanic Juan de Fuca plate at a rate of about 45 mm/yr1 (Riddihough and Hyndman, 1991). Queen Charlotte Basin is a semi-enclosed basin bordered to the east by the British Columbia mainland and to the west by the Queen Charlotte Islands (Fig. 1b). Hecate Strait opens to the Pacific in the south and southwest through Queen Charlotte Sound and to the north through Dixon Entrance (Fig. 1b). Tectonism gave rise to the existence and location of the Queen Charlotte Islands and adjacent basin. The underlying Pacific plate slides northward and slightly into North America at an average rate of 50–60 mm a1 along the seismically active Queen Charlotte Fault (Riddihough, 1988; Rohr et al., 2000). At the junction of the two plates this motion is accompanied by frequent earthquakes that vary in intensity along the fault (Fig. 1b).
2.2. Glaciation Glaciation affected the Pacific margin of Canada many times, although extensive evidence has been found for only the youngest glacial episode over much of the area (Barrie and Conway, 2002b).
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
The Fraser Glaciation began approximately 25,000–30,000 years ago (Clague, 1989). Ice moving south from the Coast Mountains of the Canadian Cordillera and Vancouver Island progressively coalesced, joined ice coming out of the Fraser Valley, and then divided into two tongues, one following Juan de Fuca Strait toward the ocean and the other going southward into Puget Sound (Fig. 1a). This large glacier reached the south end of the Puget Lowland and the western Juan de Fuca Strait at its maximum extent (Waitt and Thorson, 1983; Hewitt and Mosher, 2001), about 14,000 C14 BP (Porter and Swanson, 1998). In Queen Charlotte Basin, ice from the massive Cordilleran sheet extended westward across northern Hecate Strait and through Dixon Entrance and coalesced with ice from the Queen Charlotte Islands, deflecting it westward along Dixon Entrance (Sutherland-Brown, 1968; Hicock and Fuller, 1995; Barrie and Conway, 1999). Ice also moved south down the central trough in Hecate Strait (Barrie and Bornhold, 1989) coalescing with ice coming through the troughs of Queen Charlotte Sound to the edge of the shelf and met the ice in northern GB (Luternauer and Murray, 1983; Luternauer et al., 1989; Josenhans et al., 1995). Glaciation in QCB reached its maximum extent sometime after 21,000 14C yr B.P. (Blaise et al., 1990). 2.3. Sea level history Deglaciation was quite different in both basins. Most of GB was ice-free by 11,300 C14 BP (Barrie and Conway, 2002b). Deglaciation was very rapid with regional downwasting and widespread stagnation (Guilbault et al., 2003). In QCB rapid glacial retreat began sometime after 15,000 14C yr BP and ice had largely left the offshore by 13,500–13,000 14C yr BP (Barrie and Conway, 1999). With deglaciation in QCB a regional lowering of sea level occurred with sea levels falling in central Hecate Strait to a lowstand of 150 m by 13,000 14C years BP (Josenhans et al., 1995; Barrie and Conway, 2002a; Hetherington et al., 2004). Subsequent transgression was rapid (5 cm/yr (Josenhans et al., 1997)) with sea levels reaching the present shoreline on the Queen Charlotte Islands by about 9100 14C yr BP (Clague et al., 1982; Josenhans et al., 1995, 1997; Fedje and Josenhans, 2000). For GB sea levels remained high after deglaciation and fell to near or just below present between 12,000 and 10,000 14C years BP, before returning to present levels (Barrie and Conway, 2002b). 2.4. Oceanography The near and offshore of British Columbia is mesotidal to macrotidal with superimposed storm and estuarine circulation. Georgia Basin is characteristic of a partially mixed estuary with moderately strong tidal currents (2.6–3.4 m tidal range), seasonally varying stratification and late summer and late winter deepwater density intrusions (LeBlond, 1983; Crean and Ages, 1971; Thomson, 1994; Masson, 2002). The Fraser River discharge reaches a maximum of 10,000 m3/s during the spring freshet and a minimum of around 1,000 m3/s in late winter. The Fraser River accounts for about 73% of the mean annual freshwater discharge of 158 109 m3 into the Strait of Georgia (Johannessen et al., 2003). This freshwater influx then forces estuarine circulation in the southern strait that is characterized by a net outflow of low-salinity water towards Juan de Fuca Strait in the upper layer (o50 m depth) and a net northward inflow of highsalinity water in the lower part of the water column that reaches the Strait of Georgia in late summer (Mosher and Thomson, 2000). The strong currents in Boundary Pass and Haro Strait mix the water column and reduce stratification (Masson and Cummins, 2004) and the turbulence associated with super-critical flow, that
799
has been observed, contributes to particle resuspension (Johannessen et al., 2006). Two current meters were installed 10 km to the southwest of the Boundary Pass (Fig. 2) by Fisheries and Oceans Canada for two years at 110 and 125 m water depth. Maximum current velocities of 1.94 ms1 occurred in a northeast flood direction with the highest recorded ebb velocities of 1.29 ms1. A second mooring put in for less than 2 months just northeast of the Fisheries and Oceans mooring measured velocities up to 2.23 ms1 within the water column (ASL Environmental Services, 2001). In QCB, strong rectilinear tidal currents, usually in the order of 15–25 ms1, are classified as mixed, mainly semi-diurnal, with a 4.5 m tidal range (macrotidal) in the central portion of Hecate Strait (Thomson, 1981). The tidal crest enters Queen Charlotte Sound and spreads northward into Hecate Strait where it encounters the opposing crest that entered eastward through Dixon Entrance (Fig. 1b). Normally, tidal currents flow along the orientation of the Strait at nearly uniform speed at all depths, but where abrupt changes in the seafloor bathymetry occur currents are enhanced (Sinclair et al., 2005). In addition to tidal forcing, circulation in Hecate Strait is also affected by winds and river runoff. The influence of these two factors are generally out of phase, with river runoff (Skeena River (Fig. 1b)) dominating circulation in late spring through early summer and wind forcing having its greatest influence during the fall and winter months. Consequently, winter circulation is both tidal and wind driven with a net northward component resulting from the prevailing southeasterly storm winds (Crawford and Thomson, 1991). Current velocities exceeding 0.80 ms1, measured 15 m above the seabed, occur during winter storms and spring tides (Barrie and Bornhold, 1989). Most of the Pacific coast continental shelf of Canada is sediment starved with sediment capture within the coastal fjords and inlets. The one exception is in the southern Strait of Georgia where sedimentation from the Fraser River dominates the surficial geology with Holocene sediment thicknesses varying from zero on Pleistocene ridges to greater than 300 m within the basin (Mosher and Hamilton, 1998). Present day sedimentation rates vary from 10 cm/a near the river mouth to less than 3 cm/a in the distal parts of the prodelta (Hart et al., 1998).
3. Methods Some 12,000 km2 of multibeam swath bathymetry coverage was obtained on the Pacific coast of Canada since 2000. The multibeam coverage was undertaken to obtain contiguous full seafloor coverage of the deeper water areas of GB and areas of highest interest in QCB. The surveys were carried out from the CCGS Vector, and Otter Bay by the Canadian Hydrographic Service, in cooperation with the Geological Survey of Canada, using a hullmounted Kongsberg-Simrad EM1002 system, which operates at a frequency of 95 kHz utilizing 127 beams, for the deeper regions of the basin (450 m water depth) and a hull-mounted KongsbergSimrad EM3002 system, which operates at a frequency of 300 kHz utilizing 121–135 beams, for the shallow parts of the basin (o50 m water depth). In most of the areas, the tracks were positioned so as to insonify 100% of the seafloor with a 100% overlap. Positioning was by broadcast differential DGPS and the multibeam data were corrected for sound speed variations in the stratified water column using frequent sound speed casts. The data were corrected for tides and then edited for spurious bathymetric and navigational points using CARIS software. The gridded data were exported as ASCII files and imported into ArcInfo software for processing and image production. Analyses of repetitive dune field surveys utilized ArcInfo software routines.
ARTICLE IN PRESS 800
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Positioning using DGPS has a positional error of up to 5 m and vertical errors in the order of 1% water depth (International Hydrographic Organization, 1998). To improve positional accuracy, distinctive points of bedrock where used to assess accuracy where repeat surveys were undertaken. Regardless, the overall accuracy that can be used to interpret dynamic changes of the dunes was 5 m horizontal and 0.2 m vertical. The multibeam swath images formed the preliminary interpretive framework for delineation of sedimentary bedforms, drowned coastal features and modified seafloor morphology. Based on this interpretation several areas were identified for further investigation using a Huntec DTS high-resolution subbottom profiler (Mosher and Simpkin, 1999), sidescan sonar, and a Phantom ROV. Sediment samples were collected using a Shipek grab sampler within the subaqueous dune and wave-cut terrace areas. These were subsequently dried, split and then analysed for texture using a settling tube for the sand fraction (o2.0 mm to 40.063 mm) and a Sedigraph for the mud fraction (o0.063 mm).
crested to sinuous morphologies. The largest of the dunes measures 28 m in height and 760 m in ridge crest length (Figs. 3 and 4). Bedform steepness is about 1/12 (height/wavelength) for each of the larger dunes, resulting in slopes of 121. Similar submarine sand dunes with wave heights of up to 25 m have been identified in the eastern Juan de Fuca Strait (Mosher and Thomson, 2000), again located at the transition between a narrow
4. Results 4.1. Subaqueous dunes Large sedimentary bedforms are commonly observed within the multibeam imagery but several areas stand out due to the shear size of the bedforms and the extensive nature of the bedform fields. Two significant areas are the region between the southern Strait of Georgia and Juan de Fuca Strait in GB and the northern Hecate Strait region of QCB. Both areas represent narrow channels entering Straits which are open to the Pacific Ocean. In Boundary Pass, along the Canada/US boundary (Fig. 2), there is a series of nearly symmetrical, very large subaqueous dunes with wavelengths between 100 and 300 m, covering an area of 2.5 km2 in 170–210 m water depth (Figs. 3 and 4). Dune heights vary between 5 and 28 m. Based on ROV observations, smaller bedforms (0.25–1 m high) are superimposed on the much larger dunes. They consist of well-sorted coarse sand and gravel and the sediments have a complete absence of epibenthic organisms. The grain size varies from very coarse sand on the slopes and crest and fine gravel in the troughs (Wentworth Scale). These features are hydrodynamically formed of an unknown age and rank amongst the largest ever observed globally. The dunes display straight
Fig. 4. Difference map between November 2001 and September 2006 of dune movement and crest line change of the Boundary Pass very large dunes with an inset showing the detailed multibeam image of the dune field. Black line indicates location of Fig. 3.
Fig. 3. Huntec DTS sub-bottom profile through the subaqueous dune field in Boundary Pass (location of line shown in Fig. 4). Notice the reflectors indicating clinoform bedding on the northern most two dunes suggesting migration to the southwest.
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
801
Fig. 5. Multibeam image of the southern portion of Haro Strait and eastern Juan de Fuca Strait (Fig. 2) illustrating the two very large subaqueous dune fields, either side of the narrow passage between the two water bodies. Notice the similarity of the overall geometry between these dune fields and that shown in Fig. 4.
channel and inlet that opens to the Pacific (Fig. 5). In both cases, the dunes are classified as linear, slightly asymmetrical to symmetrical, stoss-erosional, lee-depositional, two dimensional, very large dunes (classification nomenclature of Ashley (1990)). The two dune fields also have a very similar distribution. Both fields have a tear drop or elliptical distribution with the largest dunes at the widest point of the ellipse and decreasing dune size in all dimensions to the narrowing point of the tear drop shape. However, the Juan de Fuca dune field contains two very large dunes, whereas the Boundary Pass field contains only one. Yet another tear shaped dune field of similar aerial size as the Boundary Pass and Juan de Fuca dune fields occurs just north of the entrance of Haro Strait into Juan de Fuca Strait (Fig. 5). These dunes have heights of 2–7 m and wavelengths from 60 to 80 m. In northern Hecate Strait a 72 km2 bedform field of large subaqueous dunes occurs in 88 to 96 m water depth. Though the bedform field appears to be located within a fairly open area of Hecate Strait, it occurs within a north-south oriented trough in east-central Hecate Strait (Figs. 1b and 6). The 80 m contour essentially outlines the field which is bordered on the west by the very shallow (20–40 m average depth) Dogfish Bank, to the east by a significant slope to the nearshore zone and to the north by a 70 m sill separating Hecate Strait and Dixon Entrance to the north
(Figs. 1b and 6). To the south the trough widens and deepens into Hecate Strait (Fig. 6). The field of bedforms appears to occur in a shallow erosional basin that contains two southward extending drowned, progradational spits (Fig. 6). Progradational spits are a common feature of Hecate Strait (Josenhans et al., 1995) and developed as sea level rose just after deglaciation, when an ample supply of sediment existed (Barrie and Conway, 2002a). These relict sediments provide the source for the development of the observed dune field. The dunes range in wavelength from 40 to 175 m, but the majority of the dune field have wavelengths of 50–70 m and lengths to 3 km (Fig. 7). Dune height is consistently between 2 and 3 m (Fig. 7). All but a small area of 3D linguoid dunes with wavelengths of 100 m are 2D large dunes (classification nomenclature of Ashley (1990)). The sands making up the dunes are primarily well-sorted medium sands.
4.2. Repetitive surveys Two geophysical surveys were completed for the Boundary Pass dune field in December 2001 and April 2003. Little difference in the overall morphology can be seen between the surveys, though the lines were not coincident. In both surveys the dunes in
ARTICLE IN PRESS 802
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Fig. 6. Multibeam image of the extensive subaqueous dune field in northern Hecate Strait (Fig. 1b). Notice the two drowned paleo-spits extending to the south, formed when sea levels were lower.
2006. The larger dunes showed minimal clockwise rotation, but the total migration was less than the error level for multibeam surveys, based on DGPS. However, the smaller dunes on the western end of the dune field showed up to 25 m of eastward migration toward the larger dunes (Fig. 5).
4.3. Wave-cut terraces
Fig. 7. A 3D image of the north Hecate Strait dunes shown in relation to the Huntec DTS sub-bottom profile. The multibeam has a 10 exaggeration. Location of the figure is shown by the black line in Fig. 6.
the northern part of the field appear to be slightly asymmetrical with movement to the southwest, while the majority of the dunes are symmetrical. On the western side of the dune field several of the dunes show a slight asymmetry suggesting movement to the northeast. The dunes rest on a highly reflective surface under which there is little sub-bottom penetration. The dunes at the north end of the field show sub-bottom clinoforms suggesting progradation to the southwest (Fig. 3). The dune field was initially surveyed in November 2001 and has been resurveyed by multibeam sonar three times since (October 2003, September 2004 and September 2006). The overall change between surveys is a clockwise rotation with up to 50 m of movement along the outer portions of the dunes (Fig. 4). The dunes in eastern Juan de Fuca were initially surveyed in 1998 (Mosher and Thomson, 2000) and resurveyed again in September
Within central Hecate Strait the multibeam imagery clearly identifies several wave-cut terraces in water depths from 37 m to greater than 160 m (Fig. 8). The dominant terrace is oriented almost directly north-south extending for over 120 km, before it arches around to the northwest terminating in a drowned re-curved paleo-spit (Barrie and Conway, 2009). The break in slope at the top of the terrace is defined by the 126 m contour, and the base of the terrace along most of its length terminates in a moat in water depths of between 140 and 164 m. The moat that parallels the base of the terrace is a dominant feature of the terrace morphology (Fig. 9). The moat along the main north south terrace varies in depth from 2 to 7 m, measured from the berm of the outside wall across to the base of the terrace slope, and is from 100 to 350 m wide. The outer berm roughly represents the volume of exhumed material. Slopes are as high as 121 on the terrace slope into the moat and 151 on the inside of the moat berm wall. Sediments at the base of the moat consist of fine to medium sands, whereas the terrace slope and top consist of lag gravel over medium to coarse sand. The moat berm sediments are more variable, but primarily consist of medium well-sorted sands. The moat around the drowned spit is continuous suggesting bathymetric control of seafloor currents (Fig. 8). In many areas along the berm, evidence exists for instability both on the inside and outside slopes (Barrie and Conway, 2009). For example the end of the drowned spit, at the northern end of the terrace, appears to have failed in the recent past, since failure splays are clearly seen extending east of the failure (Fig. 8). These are now being modified into hydrodynamic bedforms as the moat is again reforming. Subaqueous dunes are found adjacent to the base of the terrace slope in several areas and direction of
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
803
Fig. 8. Multibeam image of the extensive wave-cut terraces in central Hecate Strait (Fig. 1b) with the distinctive moats. At the northern end of the set of terraces is a paleospit with a surrounding moat.
Fig. 9. A Huntec DTS sub-bottom profile illustrating the stratigraphic cross-section of a typical wave-cut terrace with adjacent moat in central Hecate Strait (Fig. 1b). The cross section is located in Fig. 8 as A to A0 .
movement is to the north. The dunes are characterised by heights of 2–6 m and wavelengths of 40–80 m.
5. Discussion 5.1. Seabed morphology and sediment dynamics Three examples of sediment dynamics modifying the continental shelf of British Columbia have been discussed in detail. Each area has been modified as a result of the sediment transport
dynamics in a meso to macrotidal environment with superimposed wave and estuarine-induced current flow. The other primary characteristic controlling the environments is the overall bathymetric morphology that restricts flow within the basins. All three areas occur within narrowing basins that close at shallow sills. The British Columbia continental shelf is characterised by marine basins that are interconnected (Fig. 1) and open to the Pacific through geologically controlled straits, a result of modification of the seafloor by glaciers and terrestrial processes. Tidal exchange into the basin in many cases has to take place through narrow restricted passages, often with shallow sills, resulting in
ARTICLE IN PRESS 804
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
tidal reinforcement. The two areas that host significant bedforms occur just south of a sill that separates two straits (Strait of Georgia and Haro Strait in GB and Hecate Strait and Dixon Entrance in QCB) and the terrace/moat system occurs in a narrowing basin, with the north Hecate Strait sill at the head of the basin. In addition, the very large dunes in Juan de Fuca Strait (Mosher and Thomson, 2000) occur just seaward of the narrow opening that divides Haro Strait and Juan de Fuca Strait and another tear shaped dune field occurs at the southern end of Haro Strait where it meets Juan de Fuca Strait, close to its narrowest point (Figs. 2 and 5). The bedform fields form in depressions that likely result from erosion of the underlying Quaternary sediments. The proposed interpretation is that morphological restriction enhances current flow resulting in the erosion of the substrate providing the sediment for the development of subaqueous dunes. This usually occurs where the flow is coming over or into a sill or within a rapidly narrowing channel or trough. If enough sediment within the sand to fine gravel size exists, subaqueous dunes develop. Where bathymetric control of current flow is pronounced, extensive dune fields commonly develop as observed in Bungo Channel, Japan (Ikehara, 1998), Torres Strait off northern Australia (Harris, 1988), Malacca Strait in Southeast Asia (Keller and Richards, 1967), and at the mouth of San Francisco Bay (Barnard et al., 2006). Similarly, the development of moats adjacent to wave-cut terraces appears to occur within the narrowing central trough of Hecate Strait. Based on the observed surficial geology (Barrie and Conway, 2009), evidence of slope instability, and the known tidal and storm driven environment, the moats appear to be actively eroding. Although, no current measurements are available from within the moat system, interpretations of sediment transport processes (sediment texture and the development of subaqueous dunes) from morphological observations strongly support this interpretation. The flood tide, reinforced by the southeasterly storms, is perhaps the dominant sediment transport process. It is interesting to note that a moat has developed on all sides of the spit feature suggesting that contour forcing of the currents is a dominant process modifying the surficial geology. Whether the system has reached some level of equilibrium in sediment transport is not clear, though the evidence of failure along the wave-cut terrace and spit strongly suggests that erosion is ongoing. In particular, the obvious reworking of a failure at the end of the drowned spit appears to be reforming a moat quite rapidly (Barrie and Conway, 2009).
5.2. Large subaqueous dune migration Using the currents observed from the two moorings in Boundary Pass (Fig. 2) and the mean sediment grain size (0.7–2.0 mm) as inputs into the Ashley (1990) classification of large-scale subaqueous bedforms, it is clear that these dunes are mobile. Current velocities over the dunes would be expected to be even higher than those recorded by these two moorings (Fig. 2) because of topographic constriction of the flow to the east and because the moorings sites were off to one side of the main channel axis. It is likely that flood tides and estuarine circulation, particularly during summer, construct the dunes. If we assume a maximum velocity at 100 cm above the seafloor of 1.94 m1 (current meter data) then sediment sizes up to about 12–15 mm could be mobilized (Miller et al., 1977). Furthermore, the absence of any epibenthic organisms in the area of the bedforms strongly suggests that these sediments are frequently in motion. The critical threshold value of sediment transport of these coarse sediments would likely be reduced considering the bed roughness and the bathymetric restriction to flow.
Over the six years of repetitive multibeam measurement there is no clear net direction of movement, crestal flexing and rotation is the long term feature of wave dynamics, similar to the giant dune field in eastern Long Island Sound (Fenster et al., 1990, 2006). Plots of superimposed crest lines over the six year period suggest field-wide clockwise rotation of crest lines (Fig. 4). In essence, the large dunes have moved as non-rigid rotational bodies (Fenster et al., 2006) and maintain the same geographical location. The clockwise rotation was approximately 51 and up to 101 on some dunes with no apparent overall migration. The dune field appears to maintain the same overall geometry and volume. As these dunes appear to be largely stationary, with minor crest flexure, a limited volume of sand and gravel must move within a system that apparently gains or loses little sediment. This suggests that the sediments that form such large bedforms are partially derived through in situ erosion into the underlying and adjacent seabed. The scoured and hollowed morphology of the areas where the dune fields occur would indicate that erosion is ongoing. The lack of observed migration of the larger bedforms would also indicate that sediments introduced to the system are for the most part trapped in a tidally driven, oscillatory system that has allowed little net transport out of the area. The symmetrical nature of the dunes would support this conjecture. There appears to be a positive correlation between the mean grain size and the position across the very large flow-transverse bedforms, from moderately sorted fine gravel (42.0 mm) in the troughs to moderately sorted very coarse sand (0.71–1.4 mm) mid-dune to well-sorted coarse sand (0.71–1.0 mm) near the crest. A similar grain size trend was shown for a dune field in the eastern North Sea off the Danish west coast (Anthony and Leth, 2002). There also seems to be a strong correlation between mean grain size and dune height. The influence of grain size on dune shape is complex with dune height generally increasing as the grain size increases (Bartholdy et al., 2005), however previous studies suggest that this only occurs until about 0.5 mm, after which dune height decreases as grain size continues to increase (Allen, 1982; Dalrymple and Rhodes, 1995). In addition, dune height and wavelength generally increase in size with an increase in water depth (Allen, 1982; Ashley, 1990) and the grain size/wavelength relation in deep-water dunes favours coarse sediments (Flemming, 2000b). Flemming (2000a) suggests that in deep-water conditions, dunes will continue to grow as the critical threshold for sediment movement increases with increasing flow velocity. As critical threshold for sediment movement increases so will the grain size, consequently, very large dunes should be composed of very coarse sediments. He suggests that dunes with mean grain size greater than 0.5 mm can have dune heights greater than 24 m and wavelengths of approximately 380 m. This clearly corresponds with the Boundary Pass dunes where heights exceed 24 m, the mean grain size is 0.7–2.0 mm and the wavelengths reach over 300 m, slightly less than predicted (Flemming, 2000a). The north Hecate Strait dune field, that is composed of medium sands, do not exceed 3 m height and 75 m wavelength, again closely fitting the model proposed by Flemming (2000a). For the Boundary Pass Dunes the maximum current is approximately 1.90 ms1 allowing for initiation of sediment transport of fine gravels, while for the north Hecate Strait dune field the approximate maximum currents of 0.40–0.80 ms1 allow for the initiation of sediment transport of medium sands.
6. Conclusions Very large subaqueous dunes on the Pacific margin of Canada have formed as a result of a unique combination factors.
ARTICLE IN PRESS J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Diurnal tides, estuarine circulation and bathymetric restriction have resulted in enhanced current flow, particularly where the seafloor morphology changes from a narrowing restricted channel into a body of water open to the Pacific. Grain size, water depth and flow velocity control the size of the dunes which have heights up to 28 m. In addition, sea level lowstand wave-cut terraces that formed in unconsolidated glacial and postglacial sediments are presently being eroded by these same enhanced currents to form moats up to 7 m deep in a narrowing shelf trough. The unique seafloor morphology that controls the sediment dynamics is a result of the tectonic, glacial and sea level history over the late Quaternary. Acknowledgements The multibeam data set would not exist without the field program management, data collection, and processing capability of the Canadian Hydrographic Service, Pacific. In particular, we would like to acknowledge the valuable support of the Rob Hare, Ernest Sargent and Peter Milner. Alex Shaw, Sean Mullan, and Robert Kung provided invaluable assistance in the development of the multibeam graphics. The manuscript was improved by the critical revision of Chuck Nittrrouer and one anonymous reviewer. Support for this work was through the Geoscience for Ocean Management Program of the Geological Survey of Canada, SeaDoc Society and Moss Landing Marine Laboratories. References Aliotta, S., Perillo, G.M.E., 1987. A sand wave field in the entrance to Bahia Blanca Estuary, Argentina. Marine Geology 76, 1–14. Allen, J.R.L., 1982. Sedimentary Structures—Their Characteristics and Physical Basis, Vol. 1. Development in Sedimentology, Vol. 30A. Elsevier, New York, 593 p. Anthony, D., Leth, J.O., 2002. Large-scale bedforms, sediment distribution and sand mobility in the eastern North Sea off the Danish west coast. Marine Geology 182, 247–263. Ashley, G.M., 1990. Classification of large-scale subaqueous bedforms: a new look at an old problem. SEPM Bedforms and Bedding Structures Research Symposium. Journal of Sedimentary Petrology 60, 160–172. ASL Environmental Services, 2001. Data Report: Oceanographic Current Measurements; Boundary Passage, BC, Canada; June 21–August 11, 2000. Report submitted to Aker Engineering and Williams Gas Pipeline, 83 p. Barnard, P.L., Hanes, D.M., Rubin, D.M., Kvitek, R.G., 2006. Giant sand waves at the mouth of San Francisco Bay. Eos 87 (29), 285–289. Barrie, J.V., Bornhold, B.D., 1989. Surficial geology of Hecate Strait, British Columbia continental shelf. Canadian Journal of Earth Sciences 26, 1241-1254. Barrie, J.V., Conway, K.W., 1999. Late Quaternary glaciation and postglacial stratigraphy of the northern Pacific margin of Canada. Quaternary Research 51, 113–123. Barrie, J.V., Conway, K.W., 2002a. Rapid sea level changes and coastal evolution on the Pacific margin of Canada. Sedimentary Geology 150, 171–183. Barrie J.V., Conway K.W., 2002b. Contrasting glacial sedimentation processes and sea-level changes in two adjacent basins on the Pacific margin of Canada. In: Dowdeswell, J., O’Cofaigh, C., (Eds.), Glacier-influenced sedimentation on highlatitude continental margins. Geological Society of London, Special Publication 203, pp. 181–194. Barrie, J.V., Conway, K.W., 2009. Paleogeographic reconstruction of Hecate Strait British Columbia: Changing sea levels and sedimentary processes reshape a glaciated shelf. International Association of Sedimentologists Special Publication, in press. Barrie, J.V., Hill, P.R., Conway, K.W., Iwanowska, K., Picard, K., 2005. Georgia Basin: seabed features and marine geohazards. Geoscience Canada 32, 145–156. Bartholdy, J., Flemming, B.W., Bartholoma, A., Ernstsen, V.B., 2005. Flow and grain size control of depth-independent simple subaqueous dunes. Journal of Geophysical Research 110 (F04S16), 12 p. Blaise, B., Clague, J.J., Mathewes, R.W., 1990. Time of maximum Late Wisconsin glaciation, west coast of Canada. Quaternary Research 34, 282–295. Bouma, A.H., Rappeport, M.L., Orlando, R.C., Hampton, M.A., 1980. Identification of bedforms in lower Cook Inlet, Alaska. Sedimentary Geology 26, 157–177. Clague, J.J., 1983. Glacio-isostatic effects of the Cordilleran Ice Sheet, British Columbia, Canada. In: Smith, D.E., Dawson, A.G. (Eds.), Shorelines and Isostasy. Institute of British Geographers Special Publication 16. pp. 321–343. Clague, J.J., 1989. Quaternary geology of the Canadian Cordillera. In: Fulton R.J. (Ed.), Chapter 1, Quaternary Geology of Canada and Greenland. Geological Survey of Canada, Geology of Canada No. 1, pp. 17–95.
805
Clague, J.J., Harper, J.R., Hebda, R.J., Howes, D.E., 1982. Late Quaternary sea levels and crustal movements, coastal British Columbia. Canadian Journal of Earth Sciences 19, 597–618. Crawford, W.R., Thomson, R.E., 1991. Physical oceanography of the western Canadian continental shelf. Continental Shelf Research 11, 669–683. Crean, P.B., Ages, A., 1971. Oceanographic records from twelve cruises in the Strait of Georgia and Juan de Fuca Strait. Department of Energy, Mines and Resources, Canada 1, 55 p. Dalrymple, R.W., Rhodes, R.N., 1995. Estuarine dunes and bars. In: Perillo, G.M.E. (Ed.), Geomorphology and Sedimentology on Estuaries, Developments in Sedimentology 53. Elsevier Science, Amsterdam, pp. 359–422. Fedje, D.W., Josenhans, H., 2000. Drowned forests and archaeology on the continental shelf of British Columbia, Canada. Geology 28, 99–102. Fenster, M.S., FitzGerald, D.M., Bohlen, W.F., Lewis, R.S., Baldwin, C.T., 1990. Stability of giant sand waves in eastern Long Island Sound, USA. Marine Geology 91, 207–225. Fenster, M.S., FitzGerald, D.M., Moore, M.S., 2006. Assessing decadal-scale changes to a giant sand wave field in eastern Long Island sound. Geology 34, 89–92. Flemming, B.W., 1978. Underwater sand dunes along the southeast African continental margin—observations and implications. Marine Geology 26, 177–198. Flemming, B.W., 2000a. The role of grain size, water depth and flow velocity as scaling factors controlling the size of subaqueous dunes. In: Trentesaux, A., Garlan, T. (Eds.), Marine Sandwave Dynamics, International Workshop, March 2000, University of Lille, France, pp. 55–60. Flemming, B.W., 2000b. On the dimensional adjustment of subaqueous dunes in response to changing flow conditions: a conceptual process model. In: Trentesaux, A., Garlan, T. (Eds.), Marine Sandwave Dynamics, International Workshop, March 2000, University of Lille, France, pp. 61–67. Guilbault, J.-P., Barrie, J.V., Conway, K.W., Lapointe, M., Radi, T., 2003. Paleoenvironments associated with the deglaciation process in the Strait of Georgia off British Columbia: microfaunal and microfloral evidence. Quaternary Science Reviews 22, 839–857. Harris, P.T., 1988. Sediments, bedforms and bedload transport pathways on the continental shelf adjacent to Torres Strait, Australia-Papua New Guinea. Continental Shelf Research 8, 979–1003. Hart, B.S., Hamilton, T.S., Barrie, J.V., 1998. Sedimentation on the Fraser Delta slope and prodelta, Canada, based on high-resolution seismic stratigraphy, lithofacies and 137Cs fallout stratigraphy. Journal of Sedimentary Research 68, 556–568. Hetherington, R., Barrie, J.V., 2004. Interaction between local tectonics and glacial unloading on the Pacific margin of Canada. Quaternary International 120, 65–77. Hetherington, R., Barrie, J.V., Reid, R.G.B., MacLeod, R., Smith, D.J., 2004. Paleogeography, glacially-induced displacement, and late Quaternary coastlines on the continental shelf of British Columbia, Canada. Quaternary Science Reviews 23, 295–318. Hewitt, A.T., Mosher, D.C., 2001. Late Quaternary stratigraphy and seafloor geology of the eastern Juan de Fuca Strait, British Columbia and Washington. Marine Geology 177, 295–316. Hicock, S.R., Fuller, E.A., 1995. Lobal interactions, rheologic superposition, and implications for a Pleistocene ice stream on the continental shelf of British Columbia. Geomorphology 14, 167–184. Ikehara, K., 1998. Sequence stratigraphy of tidal sand bodies in the Bungo Channel, southwest Japan. Sedimentary Geology 122, 233–244. International Hydrographic Organization, 1998. IHO standards for hydrographic surveys. International Hydrographic Bureau, Monaco, International Hydrographic Organization Special Publication no. 44. James, T.S., Clague, J.J., Wang, K., Hutchinson, I., 2000. Postglacial rebound at the northern Cascadia subduction zone. Quaternary Science Review 19, 1527–1541. Johannessen, S.C., Macdonald, R.W., Paton, D.W., 2003. A sediment and organic carbon budget for the greater Strait of Georgia. Estuarine Coastal and Shelf Science 56, 845–860. Johannessen, S.C., Masson, D., Macdonad, R.W., 2006. Distribution and cycling of suspended particles inferred from transmissivity in the Strait of Georgia, Haro Strait and Juan de Fuca Strait. Atmosphere-Ocean 44, 17–27. Jordan, G.F., 1962. Large submarine sand waves. Science 136, 839–848. Josenhans, H.W., Fedje, D.W., Conway, K.W., Barrie, J.V., 1995. Post glacial sea-levels on the western Canadian continental shelf: Evidence for rapid change, extensive subaerial exposure, and early human habitation. Marine Geology 125, 73–94. Josenhans, H., Fedje, D., Pienitz, R., Southon, J., 1997. Early humans and rapidly changing sea levels in the Queen Charlotte Islands-Hecate Strait, British Columbia. Science 277, 71–74. Keller, G.H., Richards, A.F., 1967. Sediments of Malacca Strait, Southeast Asia. Journal of Sedimentary Petrology 37, 102–127. Knaapen, M.A.F., van Bergen Henegouw, C.N., Hu, Y.Y., 2005. Quantifying bedform migration using multi-beam sonar. Geo-Marine Letters 25, 306–314. Kubicki, A., 2008. Large and very large subaqueous dunes on the continental shelf off southern Vietnam, South China Sea. Geo-Marine Letters 28, 229–238. LeBlond, P.H., 1983. The Strait of Georgia: functional anatomy of a coastal sea. Canadian Journal of Fisheries and Aquatic Sciences 40, 1033–1063. Luternauer, J.L., Murray, J.W., 1983. Late Quaternary morphologic development and sedimentation, central British Columbia continental shelf. Geological Survey of Canada, Paper 83–21.
ARTICLE IN PRESS 806
J.V. Barrie et al. / Continental Shelf Research 29 (2009) 796–806
Luternauer, J.L., Clague, J.J., Conway, K.W., Barrie, J.V., Blaise, B., Mathewes, R.W., 1989. Late Pleistocene terrestrial deposits on the continental shelf of western Canada: evidence for rapid sea-level change at the end of the last glaciation. Geology 17, 357–360. Lykousis, V., 2001. Subaqueous bedforms on the Cyclades Plateau (NE Mediterranean)-evidence on Cretan deep water formation. Continental Shelf Research 21, 495–507. Masson, D., 2002. Deep water renewal in the Strait of Georgia. Estuarine, Coastal and Shelf Science 54, 15–126. Masson, D., Cummins, P.F., 2004. Fortnightly modulation of the estuarine circulation in Juan de Fuca Strait. Journal of Marine Research 58, 439–463. Miller, M.C., McCave, I.N., Komar, P.D., 1977. Threshold of sediment motion under unidirectional currents. Sedimentology 24, 507–527. Mosher, D.C., Hamilton, T.S., 1998. Morphology, structure and stratigraphy of the offshore Fraser delta and adjacent Strait of Georgia. In: Clague, J.J., Luternauer, J.L., Mosher, D.C. (Eds.), Geology and natural hazards of the Fraser River Delta, British Columbia. Geological Survey of Canada Bulletin 525, pp. 147–160. Mosher, D.C., Simpkin, P.G., 1999. Status and trends on marine high-resolution seismic reflection profiling: data acquisition. Geoscience Canada 25, 174–188. Mosher, D.C., Thomson, R.E., 2000. Massive submarine sand dunes in the eastern Juan de Fuca Strait, British Columbia. In: Trentesaux, A., Garlan, T. (Eds.), Marine Sandwave Dynamics, International Workshop, March 2000, University of Lille, France, pp. 131–142. Nemeth, A.A., Hulscher, S.J.M.H., de Vriend, H.J., 2002. Modeling sand wave migration in shallow shelf seas. Continental Shelf Research 22, 2739–2762. Porter, S.C., Swanson, T.W., 1998. Radiocarbon age constraints on rates of advance and retreat of the Puget lobe of the Cordilleran ice sheet during the last glaciation. Quaternary Research 50, 205–213.
Riddihough, R.P., 1988. The northeast Pacific Ocean and margin. In: Nairn, A.E.M., Stehli, F.W., Uyeda, S. (Eds.), The Ocean Basins and Margins, The Pacific Ocean, Vol. 7B. Plenum, New York, pp. 85–118. Riddihough, R.P., Hyndman, R.D., 1991. Modern plate tectonic regime of the continental margin of western Canada. In: Gabrielse, H., Yorath, C.J. (Eds.), Geology of the Cordilleran Orogen in Canada. Geological Survey of Canada, Geology of Canada No 4, Chapter 13, pp. 435–455. Rohr, K.M., Scheidhauer, M., Trehu, A.M., 2000. Transpression between two warm mafic plates: The Queen Charlotte Fault revisited. Journal of Geophysical Research 105, 8147–8172. Sinclair, A.F., Conway, K.W., Crawford, W.R., 2005. Associations between bathymetric, geologic and oceanographic features and the distribution of the British Columbia trawl fishery. ICES Cm 2005/L:25, 31p. Sutherland-Brown, A., 1968. Geology of Queen Charlotte Islands, British Columbia. British Columbia Department of Mines and Petroleum Resources Bulletin 54, Victoria, British Columbia. Thomson, R.E., 1981. Oceanography of the British Columbia coast. Canadian Special Publications, Canadian Fisheries and Aquatic Sciences, 56. Thomson, R.R., 1994. Physical oceanography of the Strait of Georgia-Puget Sound-Juan de Fuca Strait system. In: Wilson, R.H., Beamish, R.J., Aitkens, F., Bell, J. (Eds.), Review of the marine environment and biota of Strait of Georgia, Puget Sound and Juan de Fuca Strait. Proceedings of the BC/Washington Symposium on the Marine Environment, January 1994, pp. 36–100. Waitt, R.B., Thorson, R.M., 1983. The Cordilleran ice sheet in Washington, Idaho, and Montana. In: Porter, S.C. (Ed.), Late-Quaternary Environments of the United States Vol. 1, The Late Pleistocene. University of Minnesota Press, Minneapolis, Minnesota, pp. 53–70.