Fine-scale circulation as a component of gray whale (Eschrichtius robustus) habitat in Clayoquot Sound, British Columbia

Fine-scale Circulation as a Component of Gray Whale (Eschrichtius robusttis) Habitat in Clayoquot Sound, British Columbia.

Brian William Kopach

B.Sc., University of Saskatchewan, 1999 A Thesis Submitted in Partial Fulfillment of the

Requirements for the Degree of MASTER OF SCIENCE in the Department of Geography

d -

O Brian William Kopach, 2004 University of Victoria

All rights reserved. This thesis may not be reproduced in whole or in part by photocopy or other means, without permission of the author.

Supervisor: Dr. David A. Duffus

Abstract

Gray whale (Eschrichtius robustus) habitat structure has previously been described using fixed physical features including depth and sea floor slope. The role of the water column in structuring gray whale habitat is currently unknown and is the focus of this investigation. This study has two components, both using an Acoustic Doppler Current Profiler (ADCP) to collect fine-scale (1 to 1000 m) flow data. First, data were collected over repeated transects to describe the velocity, direction and variation of tidal currents (floodlebb) in an area encompassing three gray whale prey habitats (mysids (Family Mysidae), amphipods (Family Ampeliscidae), porcelain crab larvae (Family

Porcellanidae)). Second, time-series data were collected to determine if whales were selecting for specific current velocities when foraging for the hyper-benthic dwelling mysids, the whales' main prey item along the British Columbia coast. Direction of flow through the study area is controlled primarily by bathymetry, except along the western boundary of the study area where current vectors follow the progression of the tidal wave northward on a flood and southward on an ebb tide. Bays had the weakest mean current velocities (12.1 cm s-I), while headlands had the strongest mean velocities (18.5 cm s-I) and the most turbulent flow regimes (320 cm s-I), using velocity variance as a proxy measure for

turbulence. Flow through amphipod habitat was the weakest (10.0 cm s-I) and least turbulent (49 cm s-I) of the three prey habitat types. Mysid habitat had average flow velocities of 14.9 cm s-l, this is near the previously reported maximum mean velocity (13 cm s-') in which adult mysids are able to maintain their position in the water column (Clutter, 1969). Independent t-tests

determined whales were not selecting for specific current velocities, either above or below the mean velocity within mysid habitat. When viewed together these results highlight the complex nature of gray whale habitat structure that

influences the distribution of both predator and prey. This study represents a first look at the role of nearshore circulation in the formation of gray whale habitat and on fine-scale predator-prey distribution patterns.

Table of Contents

. .

...

Abstract 11

...

Table of Contents iv

...

List of Tables vi

...

...

List of Figures viii

...

Acknowledgments x

...

Chapter

1

General Introduction

1

...

References 13

Chapter

2

Advection patterns in a gray whale foraging area off

...

the coast of British Columbia

18

...

Introduction 18

...

Methods 20

...

Site Description 20

...

Large-Scale Oceanography off Vancouver Island 23

...

Data Collection 25

...

Sample Design 26

...

Data Analysis 28

...

Results 31

...

General Flow Patterns 32

...

Cow Bay 32

...

Siwash 36

...

Grassy Knoll 36 Rafael Bay

...

37

...

Rafael Point 37

Flow Conditions Related To Habitat

...

38

...

Discussion 40

...

Amphipod Habitat 46

...

Porcelain Crab Larvae Habitat 46

...

Conclusions 47

...

Acknowledgements 50

...

References 51

Chapter 3 Role of current velocity as a component of

...

gray whale habitat 56

...

Introduction 56

...

Methods 61

...

Study Area 61

...

Gray Whale Habitat Structure 65

...

Data Collection 68

...

Sampling Procedure 69

...

Data Analysis 70

...

Results 71

...

Discussion 75

...

Acknowledgements 81

...

References 82

...

Chapter 4 Final Discussion 88

...

Gray Whales and Nearshore Hydrodynamics 88

...

ADCP: A Scary Acronym for Ecologists? 93

...

Conclusion 98

...

List of Tables

Chapter 1

Table 1. NameISize definition of spatial scales referred to in this thesis, with an example of a physical process operating

on that spatial scale.

...

5

Chapter 2

Table 1. Sampling dates for each survey area over the course of the study period and the timing relative to the lunar tidal

stage and the difference in daily maximum tidal heights.

...

28 Table 2. Physical, biological and sampling features for

each of the 5 survey areas described in this study.

...

31

Table 3. Generalized descriptions of flow conditions within each of the sampling subdivisions used to describe

current flow within the study area.

...

31

Table 4. Average current velocities and descriptive statistics

classified by survey area.

...

33

Table 5. Average current velocities and descriptive statistics classified relative to shoreline proximity, coastline morphology

and habitat type.

...

39

Table 6. Generalized flow conditions within study area.

...

44

Chapter 3

Table 1. Flow velocity and variance, reported in the

previous chapter, classified into flow strength and turbulence. Flow strength is defined as Strong > 15 cm s-* < Weak

.

Turbulence was defined using variance as

Table 2. Mean current velocities and other descriptive statistics for mysid and amphipod habitats. This table

also reports the "no whale" habitat statistics.

....

.

..

.

..

.

.. . .

. .

.. .

..

.

..

. . .

. .

. . ..

.

..

. . .

.

.

72 Table 3. Habitat use by gray whales within the study area

between May 15 and September 12 classified by habitat type and survey areas. The values here include whales that were not included in the current sampling program, for a

List of Figures

Chapter 1

Figure 1. Diagrams of relationship between time and space

scales of upper ocean physics (A) and zooplankton aggregation (B)

(adapted from Haury and McGowan, 1998:164).

...

6 Figure 2. Location of study area in Clayoquot Sound, British

Columbia.

...

10

Chapter 2

Figure 1. The study area is located off the southwest coast of Flores Island in Clayoquot Sound, British Columbia, Canada. Ranging from Rafael Point in the north to the eastern edge

of Cow Bay in the southeast.

...

21 Figure 2. Gray whale foraging habitats off the coast of

Flores Island, containing three different prey species and diverse physical conditions, that were previously

identified by Dunham and Duffus (2001).

...

22 Figure 3. Survey area names and locations within the

Clayoquot Sound study site.

...

27 Figure 4. Scatterplot of wind direction vs. wind speed,

measured hourly during the course of the study, at the

LaPerouse Bank weather buoy.

...

33

Figure 5. Vector map of average ebb currents through

the study area.

...

34

Figure 6. Vector map of average flood currents through

Chapter 3

Figure 1. Location of the study area (49" 12'N, 126"

15'W)

off the coast of Flores Island.

...

62 Figure 2. Distribution of habitat for three spatially discrete

gray whale prey species within the study area. Each occupies a physically unique portion of the nearshore fringes. Mysid habitat is located on rocky, nearshore reefs, running

throughout the study area.

...

66 Figure 3. Location of feeding sites, for mysids (red), and

amphipods (black), sampled throughout the study period.

...

71 Figure 4. Mean velocities in each habitat type according to tidal state.

...

72 Figure 5. Scatterplots showing variance in current speed

and direction, sorted according to gray whale habitat type,

Acknowledg;ements

This study represents my first attempt at pulling off a research project of my own design. As with any project, it could not have come to be without the input and support of many, many people. Christina Tombach and Heather Patterson took in a prairie boy in the summer of 2000 and imparted the

knowledge required, and friendship, to make the west coast a comfortable place to a complete newby. All the members of the UVic Whale Research Lab accepted me in wholeheartedly when I started.

Special thanks have to go to the people that assisted me in the field, HP, Stephanie Olsen, and Kecia Kerr all assisted in the data collection process. The volunteers from SEACR also provided valuable help during this process. This project developed into something I never could have imagined when I started. The use of sophisticated data collection tools and analysis methods would have brought about my quick demise had it not been for the support of my committee members Dr. Richard Dewey and Dr. Dan Smith. Without Richard's generosity with gear and MATLAB tutorials this project would never have happened. For this assistance I am eternally grateful. Thanks to Dan, the subtle outside voice of editing prowess.

I

would also like to thank Dr. Jack Littlepage for his useful comments and understanding of my project.

Funny thing with doing graduate work is that a person gets extremely focused on the research, forgetting that life happens concurrently. As a result I will always consider myself lucky for the friends I made during my time in Victoria. Louise, Kecia, Andy, Charlie many experiences shared, some good, some bad, most remembered and all valuable. Thank you all for taking in a prairie boy and making him feel at home in this large, wet world. Thanks to Dr. Dave Duffus who saw something in me during a short meeting that convinced him I would be a valued contributor to his lab and research program. What do you say to a man of few words to display your appreciation? Innaresting times these have been, thanks Duff. Let's roll another number shall we. Finally that leaves my parents, without whom

I

would never have gotten this far. Their unwavering support as their son travelled to the wilds of the Northwest Territories and British Columbia helped me to chase the goals that made me happy. For that I will always be indebted.

Chapter

1

Introduction

Fluid motion creates habitat for numerous species, in a variety of environments. Wind and water motion shapes species-habitat interactions throughout terrestrial and marine food webs. Fluid effects can be direct (e.g. transport of individuals) (Garland et al., 2002), or indirect (e.g. creating prey patches for apex predators) (Hunt, 1997). The specific effect depends on factors ranging from body size and other physiological constraints, to prey selection and the ecological relationships that define a species niche. My study looks at the role of ocean motion as an element of gray whale (Eschrichtius robustus) habitat off the coast of British Columbia.

Habitat for any organism consists of a unique combination of physical and biological factors where it can successfully attain all the resources required for survival and reproduction (Hall et al., 1997). In marine environments habitat is a dynamic three-dimensional entity with the water column exerting considerable influence over the patchy distribution of species. Hamner (1988) drew a

distinction between substrate-oriented and pelagic ecosystems when he discussed patch formation. He noted that patchiness in the former usually means an absence of organisms, whereas in the latter system it means presence of organisms. In either case, a patch is recognized when compared to some

background density level. Patches of pelagic zooplankton can reach densities 100 to 1000 times the average density of the population (Omori & Hamner, 1982). For apex predators, which require high-densities of prey to meet their energetic requirements (Coyle et al., 1992), water currents that aggregate prey serve an important role in habitat formation.

Prey patchiness appears to be an important driver of predator distributions in the ocean. Piatt et al. (1989) showed that the distribution of capelin (Mallotus villosus) patches accounted for 63% of the variation in humpback whale (Megapteua novaeangliae) abundance off the coast of

Newfoundland. Environmental conditions (e.g. tidal range, water temperature, wind vectors) had an indirect affect on whale behaviour

in

that study. A clearer understanding of the interaction between environmental conditions and prey distribution will shed light on the spatio-temporal distribution of top predators, such as whales (Piatt et al., 1989; Gowans & Whitehead, 1995). To elucidate this relationship, fine-scale studies of habitat structure and prey distribution are required (Hunt & Schneider, 1987; Hastie et al., 2004).

Ocean currents have been recognized as a component of habitat for apex marine predators across taxa. Flow features, including langmuir cells (Hamner & Schneider, 1986), fronts (Franks, 1992), and general advection patterns

and serve as predictable habitat for predators. The importance of flow features and advection patterns in the formation of foraging habitat has been

demonstrated for a variety of sea bird species (Kinder et al., 1983; Coyle et al., 1992; Hunt et al., 1998). Schneider et al. (1987) showed that sea bird attendance at fronts depended on the strength of the topographically-induced flow gradient, there were more birds associated with stronger flow gradients. A flow gradient, or shear, as defined by Schneider et a1.(1987) is a change in water velocity in a horizontal or vertical direction. Not surprisingly, similar relationships have been noted for marine mammals. A study on bottlenose dolphins (Tursiops truncatus) illustrates the role a tidal front serves as a predictable foraging location (Mendes

et

al., 2002), and Zamon (2001) identified tidal advection as an important

component of harbour seal (Phoca vitulina richardsi) habitat.

Currents have indirect effects on the distribution of large-bodied

predators that are able to overcome the force of the water motion by swimming. As with many predator habitat parameters, they are more directly related to prey distribution than predator distribution. The difficulties inherent in observing most marine predator-prey interactions make the use of indirect habitat measures necessary to explain predator distributions.

Water currents are the dynamic abiotic portion of habitat for marine organisms. Other abiotic habitat components include oceanographic conditions

(e.g., sea-surface temperature (SST), salinity, density) and physical structure (e.g., depth, bathymetry). Life-history and ecological properties, including growth rates, death rates, fecundity, predator-prey interactions form the biotic portion of a species habitat (Hooker et al., 1999). With so many variables to consider, discerning the effects of biology from those of physics is a formidable task (Denman & Gargett, 1995). In dynamic environments, identifying the range of any single factor is required to understand its role in habitat structure and function.

The specific role of a single environmental variable in forming habitat is blurred by the variation of each factor on differing spatial and temporal scales. Scale is an important consideration when sorting out the issue of pattern in ecology (Levin, 1992). Species distribution patterns are embedded in an ever- changing set of environmental conditions varying on different time and space scales. Stommel(1963) was one of the first to discuss the importance of scale and illustrate the relationship between the spatial and temporal scales of physical processes in the ocean. Haury

et

al. (1978) took this "scale perspective" further (Fig. I), illustrating the relationships between physics, zooplankton aggregations and scale. As the spatial scale of physical processes increases there is a

corresponding lengthening of the temporal scale over which the process is observed (Fig 1A).

Figure 1. Diagrams of relationship between time and space scales of upper ocean

physics (A)

and zooplankton aggregation

(B)

(adapted from Haury and

For this study I adapted spatial scale definitions from Haury et al. (1978) and Hunt and Schneider (1987) (Table 1).

Table 1. Name/Size definition of spatial scales referred to in this thesis, with an example of a physical process operating on that spatial scale.

Name Spatial Scale Physical Process Example

Mega > 3000km Biogeographic regions linked to global circulation Macro 1000-3000km Equatorial Circulation Patterns

Meso 100-1000km California Current System

Coarse 1 to 100km Fronts and Eddies

Fine 1 to lOOOm Langmuir Cells and Waves

Micro 1 to lOOcm Turbulence and Mixing

Long-term fluctuation in ocean conditions is linked to mega-scale climatic variability (Miller & Schneider, 2000; Lluch-Cota et al., 2001). At the end of the last ice age the melting ice caps changed the position of the coastlines by raising sea levels. This altered the bathymetry and location of ecosystems tied to specific portions of the marine environment (e.g., intertidal vs. subtidal) (Graham et al., 2003). Variation in the intensity of the Aleutian Low Pressure System has been tied to oscillations in ocean conditions, on the scale of decades or longer, that can alter ecosystem function by creating conditions suitable for different suites of organisms (McFarlane & Beamish, 2001; Chavez et al., 2003). Flow conditions on long temporal scales are altered by sea level change as the water moves over different benthic features. Changes in wind conditions tied to long-term oscillations such as the El Nifio-Southern Oscillation (ENSO) and the Pacific Decadal Oscillation (PDO) alter flow velocities and can subsequently reduce

productivity by decreasing the level of upwelling (Miller & Schneider, 2000). Flow patterns on shorter temporal scales (4 year) include seasonal current reversals that create upwelling conditions, as seen along the British Columbia continental shelf break (Freeland

et

al., 1984), while zooplankton aggregating features such as langmuir cells, fronts and eddies occur on temporal scales of hours to days. Generally the larger the spatial scale of a process the longer its duration or variation (Stommel, 1963). Scale, spatial or temporal, represent fundamental components of our understanding of a species habitat structure and selection patterns.

Habitat selection by gray whales occurs on a number of scales. On the mega-scale, the eastern Pacific gray whales migrate seasonally between breeding grounds in Baja Mexico and their main foraging grounds in the Bering and Chukchi Seas (Pike, 1962; Kim & Oliver, 1989). On these northern foraging

grounds the whales' primary prey are benthic amphipods (Nerini & Oliver, 1983; Highsmith & Coyle, 1991). These amphipod beds cover 40,000 km2 and are one of the most productive benthic communities in the world (Highsmith & Coyle, 1992). They support the majority of eastern Pacific gray whales, a population that has recovered from the impacts of commercial whaling, to a current

1999). While this reported population estimate is 5 years old, subsequent stock assessments reported similar estimates in 2000 and again in 2002.

A small number of whales remain in southern waters to forage for the summer. The southern foraging grounds extend from northern California to the Aleutian Peninsula in Alaska (Oliver

et

al., 1984; Kim & Oliver, 1989). The waters off northern Washington and Vancouver Island appear to be the most heavily utilized portion of the southern foraging grounds, in terms of both whale presence and research effort (Calambokidis et al., 2002). Instead of benthic amphipods, hyper-benthic mysids serve as the primary prey for the

approximately 180 whales that utilize the waters off Vancouver Island (Dunham & Duffus, 2001; Calambokidis, 2002). Gray whales are known to forage for

amphipods in these waters, but they are the secondary prey choice for the whales in this area (Dunham & Duffus, 2001,2002). The amphipod beds are smaller than those in the Bering and Chuckhi Seas (Oliver et al., 1984; Kvitek & Oliver, 1986), and cover a smaller portion of the known coastal habitat of gray whales on the southern foraging grounds (Carruthers, 2000; Dunham & Duffus, 2002). Selecting between the northern and southern foraging grounds is gray whale habitat selection on the macro-scale.

Why gray whales select habitat on that scale is currently the source of debate. Moore et a!. (2003) speculate that changes in amphipod biomass on the

northern foraging grounds are forcing whales to look elsewhere for profitable foraging opportunities. Other speculation revolves around the notion that gray whales have reached carrying capacity and are being pushed into marginal foraging areas, where they are forced to utilize a wider variety of prey items to meet their metabolic requirements (Highsmith & Coyle, 1992; Dunham & Duffus, 2001). Better understanding of gray whale habitat structure will help to shed light on the importance of the southern foraging ground as a part of the eastern Pacific gray whales home range.

The fine-scale, daily habitat selection patterns of gray whales have been studied intensively for the past decade in British Columbia (Duffus, 1996; Bass, 2000; Dunham & Duffus, 2001,2002; Meier, 2003). The study area (Fig. 2), off the coast of Flores Island in Clayoquot Sound, British Columbia, contains habitat for mysids (Family Mysidae), benthic amphipods (Family Arnpeliscidae) and porcelain crab larvae (Family Porcellanidae); three known prey species of gray whales on the southern foraging grounds (Dunham & Duffus, 2001; 2002). Gray whales select between prey patches and species on fine-scales. Because of the whales increased reliance on prey within the water column, currents play a more direct role in habitat structure on the southern foraging grounds compared to the benthically dominated habitat in the Bering and Chukchi seas. This study is an

Island

Figure 2. Location of study area in Clayoquot Sound, British Columbia.

examination of nearshore advection as it relates to gray whale habitat structure. The purpose of my research was to describe current structure on the same scale as observed gray whale foraging patterns, and examine its role in

structuring prey habitats. Daily current variation off Vancouver Island is driven mainly by tides and winds, with tidal fluctuation dominating the current signal (Foreman & Freeland, 1991). Using an Acoustic Doppler Current Profiler

(ADCP) I collected fine-scale water current data. The ADCP presented unique deployment and analytical challenges by operating it in the turbulent nearshore waters where gray whales regularly forage. This thesis consists of 4 chapters; reference lists are included with each separate section of the thesis. The second chapter describes current magnitude, direction, and variation across the tidal cycle in relation to coastline morphology (headlands, bays, exposed coast) within the study area, and in relation to gray whale prey habitats (mysids, amphipods, porcelain crab larvae). This chapter represents a first look at fine-scale, nearshore hydrodynamics related directly to gray whale habitat structure. The third

chapter discusses the results of an experiment testing the relationship between foraging gray whales and current velocities. Velocities around foraging whales were compared to random current measures to determine if whales were

selecting for specific current velocities when mysid foraging. Foraging whales can serve as markers of adequate prey biomass, measuring conditions around

foraging whales provides direct information about flow conditions in mysid patches dense enough for whales to forage successfully. The fourth and final chapter is a general discussion tying the results of the previous chapters together to complete the description of nearshore water currents as a part of gray whale habitat. Included in the last chapter is a section discussing data collection and analysis issues to be considered by wildlife ecologists interested in using an ADCP as part of their own marine habitat studies.

References

Bass J. 2000. Variations in gray whale feeding behaviour in the presence of whale-watching vessels in Clayoquot Sound, 1993-1995. Ph.D., Department of Geography, University of Victoria, Victoria, British Columbia. 156pp. Calambokidis J, Darling JD, Deecke V, Gearin PI Gosho MI Megill W, Tombach

CM, Goley D, Toropova C and Gisborne B. 2002. Abundance, range and movements of a feeding aggregation of gray whales (Eschrichtius robustus) from California to southeastern Alaska in 1998. Journal of Cetacean Research and Management 4: 267-276.

Carruthers EH. 2000. Habitat, population structure and energy value of benthic amphipods, and implications for gray whale foraging in Clayoquot Sound, British Columbia. unpblished M.Sc., Department of Geography, Queen's University, Kingston, Ontario. 10lpp.

Chavez FP, Ryan J, Lluch-Cota SE and Niquen MC. 2003. From anchovies to sardines and back: multidecadal change in the Pacific Ocean. Science 299: 217- 221.

Coyle KO, Hunt GL, Decker MB and Weingartner TJ. 1992. Murre foraging, epibenthic sound scattering and tidal advection over a shoal near St. George Island, Bering Sea. Marine Ecology Progress Series 83: 1-14.

Denman

KL

and Gargett AE. 1995. Biological-physical interactions in the upper ocean: the role of vertical and small scale transport processes. Annual Review of Fluid Mechanics 27: 225-255.

Duffus DA. 1996. The recreational use of grey whales in southern Clayoqout Sound, Canada. Applied Geography 16: 179-190.

Dunham JS and Duffus DA. 2001. Foraging patterns of gray whales in

Clayoquot Sound, British Columbia. Marine Ecology Progress Series 223: 299- 310.

Dunham JS and Duffus DA. 2002. Diet of gray whales (Eschrichtius robustus) in Clayoquot Sound, British Columbia, Canada. Marine Mammal Science 18: 419- 437.

Foreman MGG and Freeland HJ. 1991. A comparison of techniques for tide removal from ship-mounted acoustic Doppler measurements along the southwest coast of Vancouver Island. Journal of Geophysical Research 96: 17,007-17,021.

Franks PJS. 1992. Sink or swim: accumulation of biomass at fronts. Marine Ecology Progress Series 82: 1-12.

Freeland HJ, Crawford WR and Thomson RE. 1984. Currents along the Pacific Coast of Canada. Atmosphere-Ocean 22: 151-172.

Garland ED, Zimmer CA and Lentz SJ. 2002. Larval distribution in inner-shelf waters: The roles of wind-driven cross-shelf currents and die1 vertical migration. Lirnnology and Oceanography 47: 803-817.

Gowans S and Whitehead H. 1995. Distribution and habitat partitioning by small odontocetes in the Gully, a submarine canyon on the Scotian Shelf. Canadian Journal of Zoology 73: 1599-1608.

Graham MH, Dayton PK and Erlandson JM. 2003. Ice ages and ecological transitions on temperate coasts. Trends

in

Ecology & Evolution 18: 33-40. Hall LS, Krausman PR and Morrison ML. 1997. The habitat concept and a plea

for standard terminology. Wildlife Society Bulletin 25: 173-182.

Hamner WM. 1988. Behavior of plankton and patch formation in pelagic ecosystems. Bulletin of Marine Science 43: 752-757.

Hamner WM and Schneider

D.

1986. Regularly spaced rows of medusae in the Bering Sea: role of langmuir circulation. Limnology and Oceanography 31: 171- 177.

Hastie

GD,

Wilson

B,

Wilson LJ, Parsons KM and Thompson PM. 2004.

Functional mechanisms underlying cetacean distribution patterns: hotspots for bottlenose dolphins are linked to foraging. Marine Biology 144: 397-403. Haury LR, McGowan

JA

and Wiebe PH. 1978. Patterns and processes in the

time-space scales of plankton distributions. In: Steele JH, editor. Spatial Pattern in Plankton Communities. Plenum Press, New York, USA. pp. 277-327.

Haury LR and McGowan JA. 1998. Time-space scales in marine biogeography. Proceedings of the 2nd International Conference. Final report of SCOR/IOC Working Group 93 'Pelagic Biogeography'. Noordwijkerhout, The

Netherlands, 9 July-14 July 1995. no. 142, pp. 163-170.

Highsmith RC and Coyle KO. 1992. Productivity of arctic amphipods relative to gray whale energy requirements. Marine Ecology Progress Series 83: 141-150. Hooker SK, Whitehead H and Gowans S. 1999. Marine protected area design

and the spatial and temporal distribution of cetaceans in a submarine canyon. Conservation Biology 13: 592-602.

Hunt GL. 1997. Physics, zooplankton, and the distribution of least auklets in the Bering Sea - a review. ICES Journal of Marine Science 54: 600-607.

Hunt GL and Schneider DC. 1987. Scale-dependent processes in the physical and biological environment of marine birds. In: Croxall JP, editor. Seabirds: Feeding Ecology and Role in Marine Ecosystems. Cambridge University Press, Cambridge. p 7-41.

Hunt GL, Russell RW, Coyle KO and Weingartner T. 1998. Comparative foraging ecology of planktivorous auklets in relation to ocean physics and prey availability. Marine Ecology Progress Series 167: 241-259.

Kim SL and Oliver JS. 1989. Swarming benthic crustaceans in the Bering and Chukchi seas and their relation to geographic patterns in gray whale feeding. Canadian Journal of Zoology 67: 1531-1542.

Kinder TH, Hunt GL, Schneider D and Schumacher JD. 1983. Correlations between seabirds and oceanic fronts around the Pribilof Islands, Alaska. Estuarine, Coastal and Shelf Science 16: 309-319.

Kvitek RG and Oliver JS. 1986. Side-scan sonar estimates of the utilization of gray whale feeding grounds along Vancouver Island, Canada. Continental Shelf Research 6: 639-654.

Levin SA. 1992. The problem of pattern and scale in ecology. Ecology 73: 1943- 1967.

Lluch-Cota DB, Wooster WS and Hare SR. 2001. Sea surface temperature variability in coastal areas of the Northeastern Pacific related to the El Nino- Southern Oscillation and the Pacific Decadal Oscillation. Geophysical Research Letters 28: 2029-2032.

McFarlane GA and Beamish RJ. 2001. The re-occurrence of sardines off British Columbia characterises the dynamic nature of regimes. Progress in

Oceanography 49: 151-165.

Meier SK. 2003. A multi-scale analysis of habitat use by gray whales

(Eschrichtius robustus) in Clayoquot Sound, British Columbia, 1997-99. M.Sc., University of Victoria, Department of Geography, Victoria, BC. 140pp. Mendes S, Turrell WR, Lutkebohle T and Thompson PM. 2002. Influence of the

tidal cycle and a tidal intrusion front on the spatio-temporal distribution of coastal bottlenose dolphins

.

Marine Ecology Progress Series 239: 221-229. Miller AJ and Schneider N. 2000. Interdecadal climate regime dynamics in the

North Pacific Ocean: theories, observations and ecosystem impacts. Progress in Oceanography 47: 355-379.

Moore SE, Grebmeier JM and Davies JR. 2003. Gray whale distribution relative to forage habitat in the northern Bering Sea: current conditions and

retrospective summary. Canadian Journal of Zoology 81: 734-742.

Nerini MK and Oliver JS. 1983. Gray whales and the structure of the Bering Sea benthos. Oecologia 59: 224-225.

Oliver JS, Slattery PN, Silberstein MA and O'Connor EF. 1984. Gray whale feeding on dense ampeliscid amphipod communities near Bamfield, British Columbia. Canadian Journal of Zoology 62: 41-49.

Omori M and Hamner WM. 1982. Patchy distribution of zooplankton: Behavior, population assessment and sampling problems. Marine Biology 72: 193-200. Piatt JF, Methven DA, Burger AE, McLagan RL, Mercer V and Creelman E. 1989.

Baleen whales and their prey in a coastal environment. Canadian Journal of Zoology 67: 1523-1530.

Pike GC. 1962. Migration and feeding of the gray whale. Journal of the Fisheries Research Board of Canada 19: 815-838.

Rugh DJ, Muto MM, Moore SE and DeMaster DP. 1999. Status review of the eastern North Pacific stock of gray whales. NOAA Technical Memorandum NMFS-AFSC-103. 96pp.

Schneider D, Harrison NM and Hunt GL. 1987. Variation in the occurrence of marine birds at fronts in the Bering Sea. Estuarine, Coastal and ShelfScience 25:

135-141.

Stommel

H.

1963. Varieties of oceanographic experience. Science 139: 572-576. Vermeer K, Szabo I and Greisman P. 1987. The relationship between plankton-

feeding Bonaparte's and mew gulls and tidal upwelling at Active Pass, British Columbia. Journal of Plankton Research 9: 483-501.

Zamon JE. 2001. Seal predation on salmon and forage fish schools as a function of tidal currents in the San Juan Islands, Washington, USA. Fisheries

Chapter

2

Advection patterns in a gray whale foraging area off the coast of

British Columbia

Introduction

Species distribution patterns result from a complex mosaic of physical and biological interactions. While this is true in any ecosystem, one major difference that distinguishes marine systems from their terrestrial counterparts is the constant three-dimensional motion of the water column (Hamner, 1988). Hydrodynamics act to structure marine communities by altering the strength of top-down (e.g., trophic interactions) or bottom-up forces (eg., nutrient, food delivery) (Menge, 1992; Leonard

et

al. 1998). Advection patterns can be a key element in determining the spatial and temporal distribution of marine species on all trophic levels.

The current regime at a single geographic location is driven by a variety of forces. Tides, wind, and freshwater input act independently, or in concert, to drive flow patterns (Thomson

et

al., 1989). In continental shelf waters, current patterns are further shaped by the interaction of the water with local bathymetry and coastline morphology (Archambault

et

al., 1998; Hunt

et

al., 1998). Together, these factors create a number of fine (1 to 1000

m)

to coarse (1 to 100 km) scale features (Haury

et

al., 1978), including langmuir cells, fronts, eddies, and

turbulent patches of water which can aggregate phytoplankton and zooplankton at densities high enough to satisfy the energetic requirements of large predators such as sea birds, fish and whales (Genin et al., 1988; Coyle et al., 1992; Chen et al., 1995). Flow patterns, on a number of scales, serve as the physical basis on which marine food webs are built, and maintained.

For predators, water currents form habitat by acting indirectly to

aggregate zooplankton or by transporting and depositing sediment and nutrients to create areas of suitable habitat for benthic dwelling species (Frechette et al., 1989; Natunewicz et al., 1991; Franks, 1992; Guichard & Bourget, 1998). Gray whales (Eschrichtius robustus) are an apex predator that forages on prey, within the water column, whose distribution may be influenced by flow conditions (Dunham & Duffus, 2001). While gray whales traditionally forage on benthic amphipods (Family Ampeliscidae) in the Bering and Chukchi seas, studies off the west coast of British Columbia (BC) show the gray whale as a predator capable of utilizing a wide variety of nearshore benthic and pelagic prey species (Highsmith & Coyle, 1992; Dunham & Duffus, 2002). The main prey for the whales in these waters are hyper-benthic mysids (Family Mysidae), a small shrimp-like

crustacean (Dunham & Duffus, 2001,2002).

The distribution of cetaceans have been related to oceanographic

scales Dunham and Duffus (2001) described patterns of daily habitat use by gray whales in the waters off Clayoquot Sound, BC. Little is known about the fine scale variation in the daily current regime in this area, or how that may affect gray whale habitat structure and use. Scale can be a confounding issue in any habitat study when physical or biological forces acting on different scales

interact. Therefore, it is important to consider physical drivers on the same time and space scales as the pattern in question (Levin, 1992).

The purpose of this chapter is to: 1) describe the fine-scale current patterns within a gray whale foraging area over the daily tidal cycle and 2) to discuss the flow characteristics within three different gray whale prey habitats

(mysids/amphipods/porcelain crab larvae). The measurements presented here will lead to a better understanding of the dynamic structure of gray whale habitat. This study will help fill an information void between coarse scale oceanographic studies and fine scale ecological studies (Menge, 1992).

Methods

Site Description

The

study

area (Fig. 1) is the site of a long-term investigation into gray whale foraging ecology along the British Columbia coast (Garner, 1993; Duffus,

Figure 1. The study area is located off the southwest coast of Flores Island in Clayoquot Sound, British Columbia, Canada. Ranging from Rafael Point in the north to the eastern edge of Cow Bay in the southeast.

1996; Bass, 2000; Malcolm & Duffus, 2001; Dunham & Duffus, 2001,2002). Encompassed within the study area, ranging from Dagger Point in the north to Red Rocks in the southeast, are three distinct prey habitats utilized at various times within

and

between seasons by the whales (Fig. 2). Benthic amphipod habitat ranges over -10 km2 of low relief, sandy substrates in the middle of Cow Bay, ranging in depth from 16 m to 30 m (Carruthers, 2000). Porcelain crab larvae (Family Porcellanidae) have been found periodically in the study area over boulder substrates in waters >1 km off Rafael Point. Mysids are found in patches 4 km from shore, 1 to 10 cm off the bottom over rock reefs and within kelp beds (Guererro, 1989; Dunham & Duffus, 2001). The benthic substrate, water depth, and relief each play a role in altering the local flow fields creating gray whale prey habitat.

Large-Scale Oceanography off Vancouver Island

The continental shelf waters along the BC coast are an area of higher than normal biomass and primary productivity (Mackas, 1992). Off southwestern Vancouver Island maximum average Chl a concentrations reach 5.5-8.5 mg mJ during the summer upwelling period, lasting from April to October. For

comparison, average Chl

a

concentrations seaward of the shelf-break peak at 0.8- 3 mg m-3 (McFarlane et al., 1997). Two key physical factors contribute to the overall productivity of this area. First, the reversal of prevailing wind directions

Figure 2. Gray whale foraging habitats off the coast of Flores Island, containing three different prey species and diverse physical conditions, that were previously identified by Dunham and Duffus (2001).

in the spring and fall, from southeast in winter to northwest in summer, creates the upwelling zone that extends from -25"-50•‹N latitude (Mackas, 1992). Second, outflow from Juan de Fuca Strait contributes nutrient-rich waters to the euphotic zone along southern Vancouver Island (Crawford & Dewey, 1989).

The physical oceanography of the continental shelf waters off Vancouver Island has been studied intensely over the last two decades, beginning with the Coastal Oceanic Dynamics Experiment (CODE) in 1979-80 (Crawford &

Thomson, 1991). As a result, the seasonal variation in current reversal is well known (Freeland & Denman, 1982; Freeland

et

al., 1984; Foreman

et

al., 1992; Pal & Holloway, 1996). Shoreward of the shelf break, driven in part by Fraser River outflow through Juan de Fuca Strait and freshwater buoyancy currents, is the Vancouver Island Coastal Current (VICC) (Hickey

et

al., 1991).

The VICC flows parallel to Vancouver Island, past Clayoquot Sound, year round. Maximum current speeds during the summer are 50 cm s-I near the surface to 15 cm s-l near the bottom (Thomson, 1981). The VICC flows against the prevailing northwesterly winds, which must reach velocities of 15 to 20 m s-I to reverse the flow direction of the VICC. During the winter the VICC flows with the prevailing southeasterly winds reaching speeds >75 cm s-I (Thomson, 1981).

It is meso-scale (100 to 1000 km) physical processes that create the

biological conditions necessary to support foraging gray whales every summer along the BC coast. Based on photo identification surveys Calambokidis

et

al. (2002) estimated 180 gray whales utilize the waters off Vancouver Island from March to November. Studies done by Duffus & Dunham (2001,2002), and Meier (2003), describe gray whale habitat selection on the scales of days and kilometres within this area. The daily foraging patterns of gray whales are operating on smaller spatio-temporal scales than the meso-scale VICC. The tides and local winds are likely the major drivers of currents at the same scales as gray whale foraging patterns.

Data Collection

A vessel mounted Acoustic Doppler Current Profiler (ADCP) was used to measure currents along repeated transects. Using a Workhorse 307kHz ADCP from RD Instruments (RD Instruments, 1996) I collected data between August 8 and September 11,2001. The small size of the vessel (7 m) increased the

susceptibility to pitch and roll effects during data collection. This made it important to collect data on relatively calm days, therefore sampling occurred when wind velocity was <15 knots.

The

ADCP

measures the Doppler shift of a sound signal reflected off particles, such as plankton >10 mm in length, to determine the speed and

direction of the water currents (RD Instruments, 1996). The particles are assumed to be floating passively in the water. Sound pings were transmitted every 2 seconds; the results were temporally averaged by the ADCP into 2 minute interval ensembles. One ensemble represents a two-dimensional area that is approximately 300 m in length, based on an average boat speed of 2.5 m secl or approximately 5 knots. The ADCP also measures flow velocities over user specified depth intervals, or bins. Each vertical bin was 50 cm in depth during data collection and further averaged during data analysis to reduce the level of noise in the data. The ADCP stored data on an internal memory chip and relayed it, real-time, to an onboard laptop computer running WMDAS from

RD

Instruments, used for raw data collection and storage.

Sample Design

For data collection purposes the study area was subdivided into 5 survey areas (Fig. 3). Depending on the size of each survey area between 3 and 5

transects were established to describe the fine-scale spatial variation in tidal currents. Coverage within each survey area consisted of a nearshore transect parallel to shore, all other transects ran perpendicular from shore out to approximately the 30 m depth contour. The depth was determined from the real-time ADCP data output to the onboard laptop computer.

Island

Figure 3. Survey area names and locations within the Clayoquot Sound study site.

Sampling in each survey area was carried out, weather permitting, over one tidal cycle. Two floodlebb surveys were completed in each of the areas. During the course of this study there were two neap (Aug. 8 & 25) and two spring tides (Aug. 18 and Sept. 2). Their occurrence is shown relative to sampling dates in Table 1.

Table 1. Sampling dates for each survey area over the course of the study period and the timing relative to the lunar tidal stage and the difference in daily

maximum tidal heights.

S a m ~ l i n ~ Date Surveu Area (PeriodlTide) MaxlMin Tidal Difference (m) Tide

August 3 3.1 Spring

August 8 Cow Bay 2.3

August 9 Rafael Bay (flood) 2.2

August 12 1.6 Neap

August 14 Rafael Bay (ebb) 1.6

August 16 Siwash 2.3

August 17 Rafael Point (ebb) 2.9

August 18 Rafael Point (flood) 3.6 Spring

August 19 Grassy Knoll 3.8

August 25 1.9 Neap

August 28 Grassy Knoll, Rafael Bay 1.8

August 30 Siwash 2.2

September 2 2.7 Spring

September 6 Cow Bay (flood) 2.2

September 7 Cow Bay (ebb), Rafael Point 2.2

September 10 1.6 Neap

Data Analysis

WinADCP, a data management program supplied by RD Instruments, was used in the first stage of analysis. Each transect was inspected to ensure the majority of bins were measured accurately, with percent good measurements

above 80%. Percent good is a measure of sample quality based on a variety of different rejection criteria (RD Instruments, 1996). Examples include large error velocities, which result from an inhomogeneity between the 4 acoustic beams of the ADCP, and fish detection, which is a large echo return in only one beam. Parameters could be set within WinADCP to automatically filter data.

Data for each transect were output in MATLAB language from WinADCP. MATLAB 5.3 was used for analysis and mapping of transect data. Programs were written to remove erroneous bins from each file and for vertically

averaging data from each ensemble along the individual transects. Bins from the top and bottom of each ensemble were removed to ensure data accuracy. Bins near the ocean surface and benthic layer had low percent good values (40%) and recorded "erroneous" flow velocities of 200 to 300 cm s-l. This is an order of magnitude larger than current speeds in the middle portion of the water column. The top four bins, representing 2 m total depth, were removed from every file. This removed interference from turbulence created by the boat hull as it travelled through the water. Bubbles would obscure the path of the ADCP signal as they travelled across the transducer face and upwardly skew the velocity

measurements. The ADCP transducer face was at a depth of 80 cm and with a blank distance of 1 m the first layer of bins began at 4 m depth. Due to side lobe contamination the lower 10% of each ensemble was removed (RD Instruments,

1996). Given the turbulent nature of nearshore environments, and the small vessel size, pitch values were as high as 16", rendering the vertical velocity measurements nearly useless. This discrepancy was due to beams sweeping across, as the vessel pitched in the waves or swell, measuring different water volumes within the same bin.

Four surveys of all transects were completed, two on each tide. The data for each transect were averaged together by tidal state to describe the direction, magnitude and relative differences between the flood and ebb tidal currents in the study area. Current vectors from the top, middle and bottom of each ensemble were mapped to look for vertical differences in flow magnitude and direction. There was no difference in current magnitude or direction between the top and bottom of the water column. Based on this analysis I averaged each ensemble through the entire water column, giving one measure of velocity and direction for each ensemble, that were then input into SPSS 10.0 for descriptive (mean, SE, Max., variance) statistical analysis. In order to relate the transect measurements to gray whale habitat features, data points along each transect were reclassified by coastline morphology, prey habitat (Table 2), and proximity to shore (nearshore(<20 m depth)/offshore(>20 m depth)) for further analysis.

Table 2. Physical, biological and sampling features for each of the 5 survey areas described in this study.

Survey Area # Transects Coastline Morphology Prey Habitat

Cow Bay ₄ Bay MysidsIAmphipods

Siwash Point 3 Headland Mysids

Grassy Knoll 3 Exposed Coast Mysids

Rafael Bay 5 Bay Mysids

Rafael Point 4 Headland Mysids/Porcelain Crab Larvae

Within each sampling classification (Table 2), current conditions were generalized using three different flow properties, defined in Table 3. These definitions provide a framework for readers when comparing flow conditions throughout the study area as they relate to gray whale habitat structure, and when discussing the flow effects on prey distribution.

Table 3. Generalized descriptions of flow conditions within each of the sampling subdivisions used to describe current flow within the study area.

Flow Property Categories Definition

Flow Strength StrongtWeak Mean Velocity: Strong > 15 cm s-I < Weak

Turbulence Turbulent/Laminar Variance: Turbulent > 100 cm s-I< Laminar

Results

Tides are the major driving force of daily current variation in the study. There is a south/southeasterly flow in the exposed western part of the study area during ebb tides (Fig. 4), with a corresponding north/northwesterly flow in those areas on a flood tide (Fig. 5). This pattern corresponds with the progression of the tidal wave north along Vancouver Island.

General Flow Patterns

Cow Bay

Situated at the southeastern end of the study area, Cow Bay is the largest (-15 km2) and most protected bay in the study area. Siwash Point, lying on its western margin, protects the bay from prevailing summer northwesterly winds, and a series of islands lying to the south provide some element of protection from southeasterly winds, often associated with low pressure systems. Those are the dominant wind directions along the coast of Vancouver Island. Wind data collected during the study period from a weather buoy located on La Perouse Bank (48" 84'N, 126" OO'W), 25 km south of the study area, reflects this (Fig. 4).

Current directions in Cow Bay appear constant regardless of tidal state in the offshore areas. Currents flowed east to west across the mouth of Cow Bay (Fig 5 & 6). There is an onshore movement of water along the 20 m depth

contour, through amphipod habitat, until it reaches the nearshore mysid habitat where the variable bathymetry and shoreline configuration exerts considerable influence on current directions. Along the 10 m contour there is a convergence zone in the middle of the bay as water flowing through amphipod habitat meets water flowing back offshore. Water flushes out of Cow Bay through the

Wind Speed (mlsec.)

Figure 4. Scatterplot of wind direction vs. wind speed, measured hourly, between May 15 and September 15,2001 at the LaPerouse Bank weather buoy.

Total mean current velocities in Cow Bay are the weakest of the five survey areas (Table 4). Velocities in the centre of the bay were consistently the weakest in the entire study area (Fig 5 & 6).

Table

4. Average current velocities and descriptive statistics classified by survey

area.

Sample Unit Name N Mean (cm s-I) Max Std Error Variance

Survey Area Cow Bay 138 11.2 47 0.68 63

Siwash 36 18.2 48 2.13 164

Grassy Knoll 55 16.6 46 1.33 97

Rafael Bay 88 13.6 56 0.89 69

Siwash Point

Siwash Point is a headland lying between Cow Bay and the Grassy Knoll. It is a large partially submerged reef that drops quickly to 30 m depth. The west side of the reef is exposed to northwesterly prevailing summer winds.

Flow directions around Siwash Point are heavily influenced by water coming across the mouth, and exiting the nearshore areas, of Cow Bay. All vectors were oriented in a west/southwest direction. The slopes around this headland cause water currents to speed up as they exit neighbouring Cow Bay. Grassy Knoll

Grassy Knoll is the local name given to the straight section of rocky coastline running between Siwash Point and Rafael Bay. With depths ranging between 10 to 20 m within 1 km from shore, bathymetry slopes quickly to 20 m

close to shore, before flattening out further offshore. In the offshore areas along the Grassy Knoll current flow follows the northwest/southeast trend of the tidal wave as it moves along Vancouver Island. In nearshore areas along the Grassy Knoll separate water masses coming around Siwash Point and out of Rafael Bay create a convergence zone. The location of the convergence zone depends on the tidal state. On a flood tide it will be at the north end of the Grassy Knoll, and moves south nearer to Siwash Point on an ebb tide (Fig. 5).

Rafael Bay

Rafael Bay, lying on the west coast of Flores Island, is more exposed than Cow Bay. Flow through this bay, on either tide, is dominated by the presence of Rafael Point to the north; which acts to funnel water into and around Rafael Bay. Once the water reaches the 10 m depth contour the direction of the water

entering Rafael Bay is controlled by the shoreline configuration. Regardless of the tidal state, water travels through the nearshore margin before exiting the south edge of the bay, where it is directed either back offshore or along Grassy Knoll. On a flood tide this meets water moving up the Grassy Knoll to

strengthen the convergence zone at this location. On an ebb tide, currents flowing southeasterly out of Rafael Bay continue in that direction travelling south along the Grassy Knoll. Currents are on average 2 cm s-l faster here than in Cow Bay, with a similar variance.

Rafael Point

Rafael Point is a prominent headland situated along the west coast of Flores Island. It is the westernmost extent of land in the study area, and as such is the most exposed of the five study areas to ocean swell. South of this area the shoreline runs in a southeasterly direction increasing the effect of Rafael Point as a barrier to water flowing up from Grassy Knoll into Rafael Bay. As a result, the flow past Rafael Point is heavily influenced by tidal flows (Fig. 5 & 6). Water

generally moves north on a flood tide and south on an ebb tide. Flow velocities at Rafael Point are the strongest and most turbulent of all the survey areas (Table

4).

Flow Conditions

Related to Habitat

Each of the survey areas shares biotic and abiotic features that provide the structural basis of gray whale foraging habitat. Flow conditions related to gray whale habitat features were examined by reclassifying data from each transect according to the physical and biological habitat features defined earlier (Table 2). The results revealed interesting, if not expected, trends (Table 5). The velocity and variance in bays, and amphipod habitat were the weakest, while headlands and mysid habitat had the strongest current velocities and largest variances measured in this study.

Table 5. Average current velocities and descriptive statistics classified relative to shoreline proximity, coastline morphology and habitat type. Sample Unit Type N Mean (cm s-1) Max Std Error Variance (cm ~-9~ Proximity to Shore Nearshore 221 15.4 115 0.94 196 Offshore 1 72 13.5 59 0.73 92 Coastline Morphology Bay 226 12.1 56 0.54 67 Exposed Shoreline 55 16.6 46 1.33 97 Headland 112 18.5 115 1.69 320 Prey Habitat Amphipod 52 10.0 34 0.97 49 Mysid 302 14.9 115 0.73 159 Porcelain Crab Larvae 35 17.8 58 2.40 202

Discussion

A consideration with this data set is the relationship between current velocities and the neaplspring cycle. During the first sampling period the weakest current area (Cow Bay) was sampled on a neap tide and the strongest area (Rafael Point) was sampled on a spring tide. By not accounting for the lunar tidal cycle, it is possible that the difference between average current strength in bays vs. headlands is simply a consequence of fortnightly tidal varation. To examine this possibility, neaplspring current samples taken inside Cow Bay were compared. They showed only a small difference (f lcm s-I) in velocities,

suggesting the effects of the lunar cycle may be small relative to the velocity differences between any of the survey areas, or habitat features, reported in this study.

Flow direction is controlled primarily by local bathymetry, leading to a large degree of directional variance, depending on the level of exposure and coastline features of the specific survey area. By classifying the study area according to coastline morphology (bay/headland/exposed coast), shoreline proximity (nearshore (<20 m)/offshore(>20 m)), or prey habitat type

(mysids/amphipods/porcelain crab larvae), a slight pattern is visible in mean velocities and variances, which can be a proxy measurement

of

turbulence. Overall the data described here presents a picture of gray whale habitat as a

turbulent environment where the different prey species are able to maintain spatial separation in the face of fine-scale current forcing.

Trends were visible in the data when classified by physical structure and prey habitats (Table 5). Currents are stronger around headlands than in bays, as are the variances, indicating headlands are more turbulent than other portions of the study area. Headlands are known to alter nearshore flow patterns, creating habitat for predators such as sea birds (Hunt & Schneider, 1987). They serve as important influences on nearshore processes such as zooplankton aggregation

(Archambault et al., 1998) and are often correlated with areas of localized upwelling (Thomson, 1981).

Bays had the lowest mean current velocities and variances. Flow direction in both bays was controlled largely by the local bathymetry, and shoreline

configuration, after an onshore push of water past the 20

m

depth contour. They both have a nearshore current that follows the coastline, until meeting with currents along the Grassy Knoll, in the case of Rafael Bay, or exiting Cow Bay past Siwash Point.

Flood currents appeared to converge at the north end of Grassy Knoll Fig. 5). This convergence moves farther south as ebb currents travel southeast meeting currents flowing west past Siwash Point (Fig. 4). Convergence zones, as seen along the Grassy Knoll, are features that aggregate prey for other cetacean

species and thereby influence predator distribution (Chen et al., 1995; Kenney & Wishner, 1995; Wishner et al., 1995; Beardsley et al., 1995). Relative to prey, nearshore mysid distribution has been related to current convergences off the coast of California (Clutter, 1967). Mysid presence in the study area is not directly tied to the presence of the convergence zone, but it may serve an important purpose in aggregating mysids at certain times into patches dense enough to meet the foraging requirements of a gray whale.

The convergence zone appears to be part of a nearshore circulation pattern driven by tidal current fluctuation. There is an alongshore current pattern visible from Siwash Point to Dagger Point, extending out to the 30 m depth contour. The direction is determined by tidal fluctuation and is evident in the more exposed portions of the study area. Currents flow north along the island on a flood tide and south on an ebb tide.

Cow Bay appears to have a more constant flow pattern, less dependant on the tidal cycle. Currents flow onshore past the 20 m depth contour, through sandy-bottomed amphipod habitat (Carruthers, 2000; Dunham & Duffus, 2001, 2002). It is an area with little relief, which is reflected in the low variance and mean velocities in central Cow Bay (Table 4).

The entire study area can be considered gray whale habitat, as spatially distinct habitats for three prey species is distributed throughout the 5 survey

areas. Amphipod habitat is found only in the middle of Cow Bay, while

porcelain crab larvae habitat is found in offshore areas by Rafael Point and Rafael Bay. Mysids are the only species that are found in a portion of each survey area. This exposes them to the widest variety of flow conditions. The specific flow patterns in each prey habitat are discussed in the following section and the flow property (Table 3) comparison between sampling and habitat features can be seen in Table 6.

Mysid Habitat

Mysids are hyper-benthic, lying in carpets that are 1 to 10 cm off the bottom (Guerrero, 1989). Currents in mysid habitat were stronger (+ 4.9 cm s-I) and more turbulent (+ 110 cm s-l) than amphipod habitat (Table 5). Clutter (1969) showed that adult mysids (Metamysidopsis elongata spp.) off California are able to maintain their position in water velocities up to 13 cm s-I, while juveniles are swept away at velocities much less than that. The maximum velocity at which adult mysids can maintain position is comparable to the mean current velocity in their habitat within the study area, a speed at which juveniles would be swept away. Differential transport of mysids by individual size or life stage may have significant impacts on apex predator distribution.

Size selective predation, predicted by foraging models based on energy intake (MacArthur and Pianka, 1966), has been observed in gray whales foraging

for mysids along the coast of British Columbia (Stelle, 2001). Since adults can hold position better than juveniles, the forces that adults can withstand become important factors in patch aggregation and ultimately predator distribution. It is energetically inefficient for mysids to swim at their maximum rate continuously, which is what they would have to do to maintain their position in mean flow velocities of 14.9 cm s-I (SE

+

.73). Flow velocities on either side of the mean may have a role in forming prey patches for gray whales, and thereby acting as a structural element of gray whale habitat. Velocities can vary in strength

temporally and spatially. In weaker currents, mysids may aggregate themselves relative to their own prey patches, or in order to stay out of energetically

draining stronger currents. Clutter (1967) hypothesised mysids were exploiting prey patches by aggregating in a rip current in nearshore areas along the

California coast. If the currents overcome the mysids swimming ability then direct individual transport can occur, leading to physical aggregation creating suitable prey patches for the whales. The next chapter tests the relationship between foraging whales and flow velocities in mysid habitat, to see if the whales are selecting for specific current velocities when foraging.

The mysids preference for hyper benthic locations may aid in location maintenance. Current velocities are weaker in the bottom layer, as they are reduced by friction along the bottom. These velocities could not be directly

measured due to side lobe contamination during data collection (RD Instruments, 1996).

Amphipod Habitat

Amphipods are found in sandy bottom areas, which have the slowest current velocities of the three prey species. The slower flow conditions aid in habitat formation by depositing sediment in which the amphipods can build their tubes (Butman, 1987). Strong currents are not required for dispersion as the amphipods do not have a pelagic phase; they disperse by crawling along the bottom (Carruthers, 2000). The slower currents allow for the settling of

phytoplankton, and other organic debris, that serves as the amphipods food base (Highsmith and Coyle, 1991; Carruthers, 2000). Other than speed, current

direction may only have an indirect role in amphipod population dynamics and spatial distribution.

Porcelain Crab Larvae Habitat

Porcelain crab larvae habitat is located in the offshore waters near Rafael Point. Fine-scale currents may not explain the ephemeral occurrences of

aggregations of porcelain crab larvae in the study area (Dunham and Duffus, 2001,2002). Currents flow mainly northlsouth through this area. Winds

stronger than those seen during this study (15 knots) can redirect flow in the top portion of the water column (Thomson, 1981). Regional scale factors may be

more important than smaller temporal and spatial scale drivers. Along the coast of California, larval dispersion has been related to relaxation in upwelling winds (Farrell et al., 1991; Wing et al., 1998). Regional current patterns and variation affect Dungeness crab (Cancer magister) off Vancouver Island (Jamieson et al., 1989). With the upwelling winds coming from the northwest, a relaxation occurs when wind comes from the southeast. Winds from that direction compliment the VICC, and tidal flow, each travelling in a northwesterly direction. Larvae would be carried along the Vancouver Island coast, reaching the inshore waters of Clayoquot Sound the larvae would be held in waters -30 m deep, against the longshore current that develops along Flores Island. The Hesquiat Peninsula, extending from Vancouver Island almost 20 km due west of Flores Island (Fig. I), may act to direct the crab larvae into the area, effectively trapping the northward flow of nearshore crab larvae.

Conclusions

The data presented here is the first description of fine-scale advection as it relates to physical and biological characteristics of gray whale habitat. Flow conditions varied in a predictable manner (Table 6) when placed in context with regional circulation patterns, including both oceanographic and atmospheric circulation. Amphipod habitat is characterised by weak, laminar flow

and food required by the benthic dwelling prey species. Fine-scale flow

conditions may have less of an impact on porcelain crab larvae aggregation than currents flowing on larger scales. Larval distribution is often linked to

relaxations in upwelling, which would accentuate the VICC given the circulation patterns along the BC coast. This would lead to increased flow velocities and downwelling currents would advect the crab larvae towards shore.

Mysids are the most important prey item for gray whales along the BC coast (Dunham and Duffus, 2001,2002; Stelle, 2001). Average flow conditions within the study area supported the evidence that mysids are able to maintain their position in the face of local flow conditions. The turbulent nature of their habitat is one of its most striking features. Variation in current velocities, either stronger or weaker than the average velocity, may have significant impacts on the distribution of individual mysids and subsequently gray whales.

Directional flow patterns are visible throughout the study area, driven by tidal movement and directed mainly by bathymetry. The small differences in mean current velocity throughout the study area represents ecologically

significant differences based on the known physical and biological characteristics of the study area. The role of turbulence in forming mysid habitat presents an area for future research regarding flow thresholds on mysid distribution patterns and patch dynamics. Unravelling questions regarding the fine-scale distribution

of mysids is a key to understanding resource selection in gray whales along the

BC

coast.

The difficulties inherent in examining hyper-benthic zooplankton make the use of indirect relationships necessary when defining habitat for many apex predators. This chapter is an examination of the water column as the dynamic third dimension of gray whale habitat. It provides the foundation for future studies of prey distribution in this area, and begins to illuminate the link

between large scale oceanographic processes and nearshore ecology. Gray whale prey species are able to maintain spatially segregated habitats despite varying flow conditions within and between species habitats. The significance of currents, specifically the role of current velocity as a habitat variable for gray whales, will be further examined in the next chapter.