UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Testing a nearshore biophysical classification system Morris, Mary 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1996-0254.pdf [ 4.84MB ]
JSON: 831-1.0086980.json
JSON-LD: 831-1.0086980-ld.json
RDF/XML (Pretty): 831-1.0086980-rdf.xml
RDF/JSON: 831-1.0086980-rdf.json
Turtle: 831-1.0086980-turtle.txt
N-Triples: 831-1.0086980-rdf-ntriples.txt
Original Record: 831-1.0086980-source.json
Full Text

Full Text

TESTING A NEARSHORE BIOPHYSICAL CLASSIFICATION  SYSTEM  by M A R Y CAROLINE MORRIS B S c . University of British Columbia, 1975 A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF THE REQUIREMENTS FOR THE D E G R E E OF M A S T E R OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Botany)  W e accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH C O L U M B I A April  1996  © Mary C . Morris, 1996  In presenting this thesis in partial fulfilment of the  requirements for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department  or by his  or  her representatives.  It is  understood  that  copying or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department of  fiOTPtrJ  y  The University of British Columbia Vancouver, Canada  Date  DE-6 (2788)  firpfVU  25 •  ABSTRACT  Habitat managers in B C have identified the need for a nearshore subtidal habitat classification system to inventory nearshore biophysical resources. The objective of this thesis was to develop a nearshore habitat model, as defined by algal assemblage, substrate, depth and wave exposure. The model for the Gabriola Island study area in the southern Strait of Georgia and representative of bedrock-dominated, semi-exposed coastline was developed from two existing datasets: 1) a systematically-quantified subtidal algal survey collected in the Gabriola area; and 2) a regional physical intertidal shore-zone mapping system. Algal assemblages were determined using T W I N S P A N multivariate cluster analyses of the algal dataset, and nine subtidal algal assemblages were identified. The algal assemblages were linked to four specific substrate definitions and to four nearshore depth intervals: 0-2 m, 2-5 m, 5-10 m and 10-20 m. The physical shore-zone dataset was used to extrapolate the algal assemblage results by substrate and wave exposure. Predictions from the model for nearshore substrate and algal assemblages by depth interval were compared to independendy-collected observations from eight subtidal transects at Saltery Bay Provincial Park, approximately 80 km north of the Gabriola area, in the Strait of Georgia. The predictions for the nearshore substrate descriptions from the physical shorezone data at least partly-matched the observed substrate descriptions in 86% of comparisons. Predictions for the nearshore algal assemblages by depth interval and substrate, at least partlymatched the observed algal species assemblages every time when the substrate descriptions from transect observations were used as the basis for the prediction. The complete matches were 45% of the comparisons. When the predicted substrate was the basis for the predicted  ii  algal assemblages, matches and partial-matches dropped to 89% of the comparisons, and complete matches were only 24% of the comparisons. From the nearshore biophysical habitat model, algal assemblages can be predicted with the most confidence for three general habitat descriptions: 1) shallow ( < 5 m depth), immobile bedrock/boulder substrate, 2) shallow (<7 m), sand/mud/pebble substrate, and 3) deeper than 5 m, with continuous to scattered boulder/bedrock. General estimates of 'standing crop', calculated from the biological database for the general types 1) and 3) can be used together with mapped nearshore depth intervals and predicted substrate to provide a first approximation of nearshore biomasses for these assemblages. Although these represent a 'best guess' of algal standing crop, there are currently no systematic summaries of the nearshore environment. The nearshore subtidal biophysical habitat classification model can be used as a basis for a general regional summary of the nearshore habitat in the higher wave exposure areas of the southern Strait of Georgia.  iii  T A B L E OF CONTENTS  ABSTRACT TABLE OF CONTENTS LIST O F T A B L E S LIST O F FIGURES ACKNOWLEDGMENTS  ii iv vi ix x  INTRODUCTION  1  METHODS Description of the Gabriola Study Area Description of the Saltery Bay Test Site Constructing the Datasets for the Gabriola Study Area The Foreman Database The Physical Databases Data Analysis using T W I N S P A N Wave energy/depth/substrate model Substrate data and interpretations for the Gabriola study area Linking the datasets Subtidal profiles observed at Saltery Bay test site  10 10 13 16 17 18 23 26 30 32 33  RESULTS Physical attributes in the Gabriola study area Analyses of the Foreman data The algal assemblages' affinity to depth A l g a l assemblage occurrence by year of collection The affinity of each algal assemblage type to substrate Wave exposure and algal assemblage The predictive nearshore biophysical subtidal habitat model Physical attributes at the Saltery B a y test site Subtidal profiles observed at Saltery Bay sites  35 35 40 51 53 54 55 56 59 61  DISCUSSION The predictive nearshore habitat model Analyses of the S P E C I E S dataset The predictive substrate model  72 72 78 81  CONCLUSIONS  84  BIBLIOGRAPHY  87  iv  APPENDIX A  List of data attributes and definitions for the Foreman database  91  APPENDIX B  List of data attributes and definitions for B C physical shore-zone mapping system 94  APPENDIX C  Predicted areas of nearshore polygons by depth and substrate within the study area  97  APPENDIX D  Substrate/depth observations from hydrographic charts and field sheets  98  APPENDIX E  Detail of count of collections by algal assemblage type by depth interval, all collections with recorded depths, all sites 99  LIST OF T A B L E S Table 1.  Physical and biological factors which affect shoreline community structure  Table 2.  Steps taken for selecting and formatting the two source datasets used i n this project  16  M o d e l for determining wave exposure category at a shore unit, using 'maximum fetch' and 'modified effective fetch'  27  Estimated wave heights and corresponding calculated depth of wave base for wave exposure  28  Wave-generated seabed disturbance at each exposure class and depth interval  29  M o d e l for predicting nearshore substrate from intertidal wave exposure and substrate.  31  Substrate codes from various data sources, 'interpreted' to fit substrate categories defined in Table 6  32  Summary of shoreline lengths by shoreline type and by wave exposure i n the study area  36  Summary of the areas of predicted substrates by depth interval of the nearshore polygons in the study area  36  Table 3.  Table 4.  Table 5.  Table 6.  Table 7.  Table 8.  Table 9.  6  Table 10. Summary of the substrate observed at the units sampled from the Foreman sites compared to the substrate predicted by the model in Table 6  38  Table 11. Exposure classes and fetch distances for the Foreman sites in the study area. Sites sorted by 'modified effective fetch'  40  Table 12. Count of collections at each of the Foreman sample sites  41  Table 13. Summary of the clustering of algal assemblages from the T W T N S P A N analyses ( T W N 1 and T W N 2 )  48  Table 14  A l g a l assemblages identified by T W I N S P A N clustering, with indicator species and associated species  50  Table 15. Number of collections of each algal assemblage type in each nearshore depth interval  51  vi  Table 16. Count of collections of each algal type by year for: 1) all sites, 2) BI01 only, 3) SI02 only  53  Table 17. Substrate classes and depth intervals assigned to the nine algal assemblages derived from the T W I N S P A N analyses  58  Table 18. The physical shore-zone mapped information for the units at the Saltery Bay test site  60  Table 19. Wave exposure categories for Saltery B a y profiles in test site  60  Table 20. The substrate model (from Table 6) predictions attached to the nearshore depth polygons and profiles observed at the Saltery B a y test sites  61  Table 21  Table 22  Comparison details for Saltery Bay profiles SP1 and SP2, for predicted substrate compared to observed substrate and for the algal assemblage predictions based on both the predicted substrate and on the observed substrate  68  Comparison details for Saltery Bay profiles SP3 and SP4, for predicted substrate compared to observed substrate and for the algal assemblage predictions based on both the predicted substrate and on the observed substrate  68  Table 23a. Comparison details for Saltery Bay profiles S C I , for predicted substrate compared to observed substrate and for the algal assemblage predictions based on both the predicted substrate and on the observed substrate  69  Table 23b. Comparison details for Saltery Bay profiles SC2, for predicted substrate compared to observed substrate and for the algal assemblage predictions based on both the predicted substrate and on the observed substrate  ..69  Table 24  Comparison details for Saltery B a y profiles SC5 and S C 6 , for predicted substrate compared to observed substrate and for the algal assemblage predictions based on both the predicted substrate and on the observed substrate  70  Table 25a. Comparison of the number of times the predicted substrate code matched the observed substrate for each Saltery B a y profile  71  Table 25b. Comparison of the number of matches by nearshore depth interval between algal assemblages predicted to occur on the predicted substrate and the observed algal assemblage of each Saltery B a y profiles  71  vii  Table 25c. Comparison of the number of matches between the predicted algal assemblage predicted for the observed substrate matching the observed algal assemblage for each Saltery B a y profile  71  Table 26. Comparison of the species assemblage analyses done previously on the Foreman database with the T W I N S P A N analyses for this thesis  74  viii  LIST OF FIGURES Figure 1.  Locations of the Gabriola study area and of the Saltery B a y test site  Figure 2.  Detail location of the Gabriola study area, showing Foreman database  11  collections sites  12  Figure 3.  Detail of the location of the Saltery Bay test site  14  Figure 4.  Sketch to explain concept of 'shore units, and across-shore physical mapping  20  Figure 5.  The nearshore subtidal depth polygons  21  Figure 6.  Schematic of databases and relevant attributes used in analyses  22  Figure 7.  Schematic of the T W I N S P A N analysis T W N 1  47  Figure 8.  Schematic of T W I N S P A N run T W N 2 for algal-rich group from T W N 1  49  Figure 9.  A l g a l assemblage 'types' determined from T W I N S P A N analyses  52  Figure 10.  A l g a l assemblages showing distribution by depth, by 'preferred' substrate  57  Figure 11  Observed substrate and observed algal species, with interpretation of substrate summary code and algal assemblage type numbers, Saltery B a y boat ramp, Profile SP2 Observed substrate and observed algal species, with interpretation of substrate summary code and algal assemblage type numbers, Saltery B a y boat ramp, Profile SP4  Figure 12.  Figure 13.  Figure 14.  64  65  Observed substrate and observed algal species, with interpretation of substrate summary code and algal assemblage type numbers, Saltery Bay mermaid profile, S C I  66  Observed substrate and observed algal species, with interpretation of substrate summary code and algal assemblage type numbers, Saltery B a y campsite profile, S C 6  67  ix  ACKNOWLEDGMENTS  I am most grateful to my committee, family and friends who helped me get through this thesis with their continuing support and encouragement. In particular, thanks to D r . R. Foreman for granting me access to his dataset which made it possible for me to pursue the original idea for this project. Thanks to D r . C . Levings of Fisheries and Oceans who supported my work through contract and encouragement. Thanks to Dr. Bradfield for getting me organized on T W I N S P A N and lending moral support. Thanks to D r . Harrison for the benthic ecology course and those assignments which helped with background for my thesis. Thanks to D o n Howes, at the B C Land Use C o ordination Office for advice and access to the provincial shore-zone mapping databases used i n my project. The Fisheries and Oceans funding was through the D F O Habitat Action Plan, Environmental Analysis Component. The Foreman database was developed with financial assistance to D r . Foreman from Environment Canada, Fisheries and Marine Service Grant-In-Aid of University Research, and National Research Council of Canada Operating Grant N o . A6241. Office space and technical support from Coastal and Ocean Resources is gratefully acknowledged. Special thanks to my mum for her patient ear and wisdom when things were getting a bit much and most special thanks to my partner and mentor, D r . J. Harper, without who's support I'd have never made it through.  INTRODUCTION Coastal habitat managers in the province of British Columbia ( B C ) have identified the need for a nearshore subtidal habitat classification system, based on measurable biophysical criteria which can be related to coastal resources (Levings & Thorn, 1994). Specifically such a system is needed for modeling of coastal marine resources and habitat usage, and ideally, will result in both an inventory of biophysical resources as well as a means of measuring and monitoring changes over time (Levings & Thorn, 1994). T o date there has been no systematic nearshore subtidal classification system developed or tested in B C waters at the scale appropriate for regional habitat characterization. The Strait of Georgia is "by far the most important marine region of British Columbia" (Thomson, 1981) and is the focus of concern of coastal resource managers  due to  development pressures. With the majority of the province's population living on the shoreline of the Strait of Georgia, this inland waterway is subject to increasing management concerns on changing nearshore habitats. The B C Ministry of Environment has developed a system for inventory and classification of biophysical attributes of the intertidal shoreline. ('Biophysical' is used here to include the combination of the biological assemblages of the shoreline with the physical form, geology and coastal processes.) Aerial video imagery collected during low tides is used as the data source for professional interpretation, creating digital coastline maps together with associated databases (Howes et al., 1994). The data are not strictly quantitative, as they are based on a remote observer's visual interpretation of parameters such as width, density, and materials, species identifications, rather than not on-site measurements or collections. In  1  particular, the biological mapping done for the Strait of Georgia (Howes et al., 1993) is for site-specific resources such as sea bird colonies and seal haulouts. Shore-zone biota are only addressed as: "marsh", "vegetated" or "non-vegetated". The coastal mapping for the Strait of Georgia was done prior to the development of the B C biological shore-zone mapping system (Searing & Frith, 1995) During the 1970s, D r . R o n Foreman at the University of British Columbia developed a quantitative database of information for intertidal and subtidal invertebrates and algae, where each species' abundance, biomass, and other details were measured. The species data are attached to specific physical parameters at the collection sites, enabling a link between the biological data and the physical site features. This database is georeferenced by location and covers many stations along the B C coast, primarily in the Strait of Georgia (Foreman, 1976 and 1979, Keen & Foreman, 1980). Although this database is twenty years old, the rigourous collection and documentation of the species information and the fact that the data are still accessible from D r . Foreman, make it reasonable to use this quantified biological dataset as the basis for analyses of species assemblages. The broad goals of this thesis are to develop a classification of subtidal algal species assemblages, and to develop and test a model of biophysical description of the nearshore subtidal habitat in the southern Strait of Georgia. The basic assumption of this project is that: Subtidal coastal algal assemblages in the Strait of Georgia occur within certain 'habitat types', determined and predictable from  intertidal and subtidal  substrate, wave exposure, and other measurable nearshore biological and physical processes.  2  Thus, the objectives of this project are:  1) to develop an integrated database of nearshore biophysical information for an area in the Strait of Georgia, using quantified biological data from the Foreman dataset; the physical coastal intertidal inventory system developed by the B C Ministry of Environment; and a new database of the nearshore subtidal biophysical descriptors  2) to analyze data for algal species assemblages, compare with earlier classifications from the Foreman data and relate to physical data, describing biophysical habitat types  3) to develop a model which uses physical parameters of the nearshore to predict the occurrences of habitat types for comparable shoreline types.  4) to test the confidence level in the model's predictions by comparison to nearshore habitat types observed at a test site outside the study area.  The hypothesis to be tested from this 'nearshore biophysical model' is:  That algal assemblages are directly related to physical habitat as defined by depth,  wave-energy,  and  substrate;  and  that the  occurrence  of  those  assemblages can be predicted from definition of physical habitat-types.  3  The test of the hypothesis will be to compare  'predicted habitat types/algal  assemblages' to 'observed habitat types/ algal assemblages' at a site elsewhere in the Strait of Georgia. This thesis combines existing databases, using new techniques, to apply and test the predictive power of a nearshore biophysical habitat classification. The concept for this thesis came from work on several coastal intertidal inventory contracts with both the B C Ministry of Environment, Lands & Parks and with Parks Canada. The technique of using aerial videos to map intertidal physical shore-zone features has been developed and applied over the past 10 years, while the integration with biological classification was developed recently for Parks Canada's G w a i i Haanas intertidal inventory and others (Harper et al., 1994a, 1994b, 1995, Searing & Frith, 1995). Community classification was preliminary and indicator species were checked for presence/absence during the field program, but systematic measurements of species biomass or abundance were not done; nor were collections made to confirm species identifications. These investigations did not extend into the nearshore subtidal, where the data for the biophysical processes and attributes are more difficult to collect. In coastal classification, as with other 'ecological mapping', the ideal classification system would be derived from the models of community structure, and from the interactions of physical and biological processes which shape the species assemblages, combined with practical considerations of data collection and scale of mapping. That is, the system would be based on the ecological principles governing the processes that structure communities, and those 'structuring forces' would be the basis of the classification used to develop a mappable  4  inventory system. Community descriptions would be based on the observed or measured physical and biological distinctions between community assemblages. Many biophysical attributes structure shoreline biotic assemblages as summarized in Table 1. None of these attributes works in isolation from the others and in fact, shoreline communities - that is, the habitat types and species assemblages that persist or are present on a nearshore site - are the result of the inter-connections and interactions of a number of these attributes. A good example of this is the related effects of depth, substrate and wave exposure ( see Methods section for further discussion). Some of these environmental parameters can be measured and used as an 'index' of several others. A n example of this is the physical attribute of 'depth'. Light, water motion from surface waves, temperature and nutrients may all vary along the depth gradient (Schiel & Foster, 1986). Ecological classification of coastal species assemblages has been the focus of many studies in ecology. M o s t studies have worked in the intertidal area (Schiel & Foster, 1986). Early summaries of zonation tended to be based on qualitative assessments and have been often quoted since. These early biophysical 'classification systems' have included work on the shore of Britain by Lewis (1964) and Ballantine (1961) and a summary of shoreline communities i n different areas of the world in Stephenson and Stephenson (1972). A l l of these systems observe and describe various species assemblages at different substrate and wave exposure regimes.  5  Table 1. Some physical and biological factors which affect shoreline community structure. Physical Factors substrate: consolidated or unconsolidated  •  wave exposure  •  tidal regime  •  •  Kffects shaping community ltiologicul Factors mobility of the substrate dispersal and determines the stability recruitment of the attachment; sand and moveable gravels cause scour or abrasion complex interaction of predation and herbivory shoreline topography, slope, orientation to incoming swell, nearshore depths, and wave fetch determine the amount of energy delivered to the shore primary structuring force inter- and intraspecific of vertical zonation by competition determining the basis of the inundation regime vertical zonation also affected by wave exposure complex chemistry of 'succession' estuaries; and rainfall or desiccation during air exposure create salinity gradients  salinity  •  nutrients & water quality  • cycles of nearshore life history upweling and nutrient input from land influence the food and nutrient available in nearshore community • patterns of water temporal scale movement determine distribution of nutrients and planktonic life stages • stratified by season, depth and inundation regime • attenuation by depth, seasonal variation • geographic variations, micro-habitat variability within site 'landscape' such as slope, roughness, permeability of substrate, shade, site dimensions  currents  temperature  light spatial scales & geomorpholgy  Effects shaping community • unpredictable numbers and patterns of dispersal of juvenile, planktonic life stages affect community structure • presence or absence of predator species may influence overall abundance of prey  • may be species-specific interactions between predators and prey, and between species assemblages competing for settlement space • related to the disturbance regime of the community, patterns of succession determine the community mosaic; may depend on speciesspecific recruitment • recruitment, reproductive schedule & life cycle strategy, growth form  • seasonal aspects of life cycles, disturbance patterns, inter-specific interactions of predation/herbivory  6  In a study of intertidal algal communities along the outer coast of California, Foster et al. (1990) examined algal species distribution and abundance  on rocky platforms of  comparable morphology. Algal species assemblages based on the most abundant species showed "considerable site-to-site variation in vertical position and abundance", but specific wave exposure or slope was not measured at the sample sites. From this work, Foster et al. (1990) suggested the importance of including physical characteristics of algal assemblages in a 'site type' descriptor to make algal communities comparable. That is, they recognized the need to develop a number of biophysical habitat descriptors to characterize coastal systems. Hily et al. (1992) sampled soft bottom subtidal algal communities in the Bay of Brest, France, determining that wave disturbance was the main ecological mechanism controlling community structure. The spatial patterns of the five algal communities on the various soft substrates were related to a wave energy/disturbance gradient, and the interaction of physical disturbance/substrate stability and biotic grazing/competition determined which algal species would persist (Hily et al., 1992). Fuller et al. (1991) classified intertidal communities along several hundred kilometres of the coast of Northern Ireland using the multivariate cluster analysis program T W I N S P A N and several other cluster analyses. The purpose of their work and the scale of application of the technique, were very comparable to the work being done in B C and to the objectives of this project. The Fuller et al. (1991) study was done to develop a classification of rocky intertidal shores which could be used for comparative evaluation of marine shorelines for conservation management. The measurement of wave exposure was not clearly quantified but seems to be similar to that suggested by Howes et al. (1994), which is based on wave fetch,  7  fetch window and shoreline topography. Biological data collected were 'semi-quantitative' with species present and general abundance recorded for all macrobiota. Fuller et al. (1991) found six different species assemblages which related to physical and biological attributes of different areas along the coast. In an area of the eastern coast of South Africa, Anderson and Stegenga (1989) distinguished three distinct subtidal algal communities on the basis of substrate, wave exposure and depth. Wave exposure was described and related to depth as a relative "amount of water movement". Deeper algal communities experienced the least disturbance from wave energy at the surface. Data were analysed using T W I N S P A N and changes in the algal assemblages due to urchin grazing were noted within the 'lower energy - medium depth' type. A n example of subtidal mapping for an area within B C is the work done in the Goose Islands, of the Hakai Pass Recreation Area (Emmett et al., 1994). 'Nearshore physical units' were defined using adjacent shore units classified using the B C Shore-zone mapping system (Howes et al., 1994), together with observations of intertidal and subtidal substrate. Because it was found during the preliminary analysis of the biological data, that the nearshore physical units did not alone predict the distribution of the biological communities, Emmett et al. (1994) subsequently defined 'Biophysical Units' as "a combination of the species most characteristic of a community and its physical environment". They also found that the species assemblages observed within these biophysical units appeared relatively consistently, suggesting a degree of predictability: certain indicator species and physical parameters defined a 'biophysical unit'. Recent work on rocky intertidal shores of San Juan Island (Schoch & Dethier, in press) has also demonstrated the predictability of algal assemblages based on classification of  8  the habitat type. Within the intertidal shoreline, low, medium and high elevation segments were defined and sampled quantitatively for species and abundance of macrobiota. A number of geomorphological features were measured in each shore unit in much finer detail than the Howes et al. (1994) technique. Only those units with bedrock substrate were included in further analysis and it was shown that the greatest geomorphic variation between units could be accounted for by segment slope. Schoch and Dethier (in press) concluded that "distinct distribution patterns in intertidal fauna and flora can be statistically associated  with  geomorphologically distinct beach characteristics". While the distributions of organisms within the same group of beach segments were the same, even when the beaches were several kilometers apart, the distribution of organisms on dissimilar beaches was significantly different, even when the geomorphic differences among these beaches were subtle. From reviews of other coastal algal community classifications, it seems that it is necessary to characterize physical attributes and processes into a 'habitat type' to attach to the algal communities observed. O n the basis of earlier intertidal studies, biophysical habitat description offers a potential method for characterizing subtidal habitats on a regional scale.  9  METHODS Description of the Gabriola study area The first steps to developing the nearshore biophysical model for testing were to select a portion of the Foreman algal dataset and a portion of the B C physical shorezone database for a defined area of the Strait of Georgia. The study area mapped in detail is near Nanaimo, B . C . in the central Strait of Georgia (Figures 1 and 2). The specific area was defined as lying between latitudes 4 9 ° 3' and 4 9 ° 10' and longitudes 123° 37' and 123° 45' and covers the east end of Gabriola Island, the north half of Valdes Island, and the Flat Top Islands complex near Silva Bay. The Gabriola study area was chosen to enclose as many of the Foreman collection sites as possible and to cover a variety of coastal morphologies and wave exposure. For the purposes of this project, the nearshore was defined to include the area from zero tide datum subtidally to the 20 metre isobath. The 20 m depth contour matches the definition of the nearshore subtidal used by the provincial coastal shore-zone mapping system (Howes et al., 1994) and encloses the majority of the 'photic zone' - the depth to which solar radiation is available for biota (Krebs, 1994). Within the study area, the tidal range is about 3 m and follows a mixed semidiurnal pattern (Anonymous, 1996). Tidal currents are present within the study area, both on the wave-sheltered west side of Valdes Island in Pylades Channel and through Gabriola Passage (Figure 2). In Gabriola Passage, current tables show maximum currents occasionally over 8 knots on flood, and over 9 knots on ebb (Anonymous, 1996). The underlying geology of the intertidal and nearshore of the study area is lateCretaceous sedimentary rock of the Nanaimo group (Yorath & Nesmith, 1995), with much of the shoreline being steep rock cliff or rock platforms. The more resistant bedrock is either  10  Figure 1. Locations of the Gabriola study area ( A ) , at the east end of Gabriola Island and of the Saltery B a y test site (B), Jervis Inlet, in the Strait of Georgia, B C .  Figure 2. Detail of the Gabriola study area, showing Foreman database collection sites and acronyms. 12  conglomerate or sandstone, interspaced with softer, more-erodable shale (Yorath & Nesmith, 1995). Nearshore sediment is generally 'hard' with pockets of sediment. N o major streams occur within the study area. The proximity of Gabriola study area to the Fraser River estuary influences the temperature, salinity and turbidity of the nearshore (Thomson, 1981). During late winter (February-March) the near-surface waters can be the coldest water in the Strait of Georgia, as low as 5 C due to the Fraser River runoff. In the early summer, the warm water of the Fraser 0  River freshet builds to peak in mid-summer, bringing some of the warmest water in the Strait of about 20 C (Thomson, 1981). 0  Salinity in the surface waters in the study area also varies with season depending on the Fraser River's discharge. During winter low flow, surface waters are generally 27 to 29 parts per thousand. During freshet, the salinity can drop to as low as about 15 parts per thousand (Thomson, 1981).  Description of the Saltery B a y test site The nearshore habitat classification model was developed from the biophysical data from the Gabriola study area and predictions from the model about the occurrence of algal assemblages were compared to subtidal observations from a nearshore site at Saltery Bay Provincial Park, on the north shore of Jervis Inlet (Figures 1 and 3). The conditions of coastal geomorphology and oceanography of the nearshore waters at this site are generally similar to the study area on the west side of the Strait of Georgia. In the nearshore 20 meters at Saltery Bay, the slope, substrate and wave exposure are comparable to the Gabriola site. However, at  13  Boat Launch  Campsite  Figure 3. Detail of the location of Saltery Bay Provincial Park, showing the separate areas of the boat launch and the campsite portions of the park, as well as the shore units and the profile locations. Profiles SP1 to SP4 are at shore units 2478 and 2479 and profiles S C I , 2, 5, and 6 are in units 2481 and 2482. The Saltery Bay ferry terminal is east of the campsite area, in unit 2485.  14  the Saltery Bay site, water conditions are probably considerably less estuarine than those at Gabriola Island, as there is less influence of the Fraser River plume. Also, the Saltery Bay shoreline is close to the abyssal depth conditions of the Jervis Inlet fjord, while the Gabriola shoreline is adjacent to shallow rock platforms. Other than a general description of B C fjord dynamics (Thomson, 1981), specifics of water temperature and salinity are not available for the offshore oceanography of the Saltery B a y site.  15  Constructing the data sets for the Gabriola study area A number of steps were taken to retrieve the databases and prepare for the analyses (Table 2).  Table 2. Steps taken for selecting and formatting two of the data subsets used in this project. The Biological Dataset (from the Foreman database) (Foreman 1976 and 1979) - obtain datasets from D r . Foreman, Botany Dept., U B C - determine geographic coordinates of Gabriola study area and select data subset  The Physical Shoreline Dataset (from the provincial physical shore-zone mapping for southern Strait of Georgia) (Harper & Reimer 1993) - obtain database from provincial shoreline data, southern Strait of Georgia (SSOG) - create subset of shoreline data for study area, using geographic coordinates and ARCVIEW  - select physical site data for each collection in the study area from Q U A D R A T database, using geographic coordinates of sites  - selection of the intertidal across-shore descriptions, combining south Strait of Georgia databases' U N I T , E X P , and X S H R datasets into one shoreline database for the study area  - select associated data from S P E C I E S database for collections within study area, attach by collection number  - create nearshore polygon mapping system for subtidal adjacent to linear shoreline dataset  - correct collection stations lats and longs from review of Arcview plots and review of old data sheets, ensure location on the coastline - create data entry for wave energy calculation for quadrat locations  - digitize nearshore polygons for each shore unit, from hydrographic charts for bathymetry:  - correct depths recorded to ' m below 0 datum' from 'feet below mean tide' - ensure correct formatting of data (text/numeric/date) and create verifications for other data entries to check for inconsistencies in interpretation of the database  - create wave energy/ substrate model and apply to study area  0 - 2 m, 2 - 5 m, 5 - 10 m, 10 - 20m - ensure correct formatting of data (text/numeric/date), correct entries for intertidal shore units split or combined in applying adjacent polygons  - create data entry for new subtidal polygons, using C H S field sheets, Foreman data for specific sites, aerial video, other data sources  * Note: data dictionaries for database description of fields attached as Appendix A & B .  16  The Foreman database The Foreman database formed the basis of the 'biological datasets' used to develop the habitat classification model for this thesis. Permission to use the data were acquired from its owner, D r . R . E . Foreman, at U B C Department of Botany. The data consists of several hundred files stored on three separate tapes, with the last updates stored in March 1981. Only a portion of the Foreman database was used in this study, consisting of the sites collected within the Gabriola study area (Figure 2) and only those quadrats which had species' weights recorded. Data collections were made from 1972 through 1977. Collection methods of the original data are outlined in Lindstrom (1973), Foreman (1976 & 1977) and Lindstrom & Foreman (1978). M o s t of the Foreman files are programs written for original analyses or are various subsets of the data. The files used for this project were:  1) Q U A D R A T (last updated Sept. 6,1978) The  QUADRAT  data file consists of physical details for each quadrat  collection site, and includes up to Collection Number 611 (August 1977) for sites throughout the Strait of Georgia, and elsewhere along the shoreline of Vancouver Island.  2) S P E C I E S (last updated Dec. 23, 1979) In the S P E C I E S database, weights and details about phenology, substrate and other attributes were recorded for each plant species collected at each quadrat.  17  The S P E C I E S data set contains collections, referenced by Collection Number to the Q U A D R A T sites, for algal species only, and has entries for all the sites in Q U A D R A T (up to collection number 611).  Appendix A lists the 'data dictionary' - the description of the fields in Q U A D R A T and SPECIES. Animal species were also collected, identified and weighed in 1972 and 1973. Because these invertebrate datasets matched only the first year-and-a-half of the algal data, the invertebrate data sets were not included in the analyses of the species assemblages in the study area. The tape archive data, stored as M T S format, was translated to text files available to P C , and translated again to .dbf files which were accessible to contemporary database programs. Because the original data include geographic coordinates (latitudes and longitudes for each collection) the .dbf files are accessible to digital mapping as in A R C V I E W . Considering the age of these data, it is fortunate that the original format and storage media are still accessible today.  The Physical Databases The  basis for the physical dataset used in this thesis is the B . C . Ministry of  Environment's shore-zone mapping. Steps taken for setting up the physical nearshore databases are outlined in Table 2. Digital coastal mapping has been completed by the B . C . Ministry of Environment for the northern and southern Strait of Georgia coastline to be used  18  for coastal zone planning and oil spill risk assessment (Howes et al., 1993, Harper & Reimer, 1993). These data are mappable linear intertidal shoreline features made up of longshore 'units'. Figure 4 shows a schematic representation of a shore unit, and gives an example of the definition of across-shore. Attached to each zone of each unit are a set of physical attributes describing the coastal form and substrate materials, recorded in a dataset and associated with the digitized shoreline map. (The data dictionary for the fields included in this shore-zone database are listed in Appendix B.) This linear shore-zone map and dataset were used as the framework to create a second physical database of the adjacent nearshore subtidal map polygons (Figure 5), defined by depth intervals. Each intertidal shore unit was attached to up to four subtidal polygons, mapping the nearshore depth intervals of: - 0 to 2 m , - 2 to 5 m , - 5 to 10 m and - 10 to 20 m The nearshore subtidal polygons were created by digitizing the marine chart. The associated nearshore database includes: a) a unique identifier number for each polygon, b) the associated shoreline unit I D number, c) depth range, d) substrate and e) calculated area. The relationships between the subsets of the Foreman datasets and the physical databases used i n the analyses in this thesis are shown schematically in Figure 6.  19  Higher high water line  ;ure 4 The upper diagram shows an along-shore mappable shore unit, with the across-shore components and zones. The lower figure is a sketch of a shore unit showing examples of acfoss-shore zones. A 1 is the backshore, and B 1,2,3 are intertidal zones across the beach face and bedrock platform.  20  =tfc u  a o •a  u o =3  o (J  <U v> trt X> d •M d  c C/J  w E O (J w C-i  Q H  00  a o  .5  a  1)  T3  o  CL,  a a  <u >-. 2  O u  ,y u « u Q,_5 ti  u O pq  <*  d  <+-, 2  *  5  Q  ^ a  x» C/5  > -5  <u d  i <u  o O  X) cJ  4-1  Q •— I  ir  a • rt  '5b o <U C O 'EL <+-  o 4-1  o  <*> .ti o  (X  3 ^  8  T3  'rt  u  no  •s 0  M  Q N  0 J3 O i-*  4-J  U  dw §  1  a aI g u w  8 <s a  22  Data analysis using T W I N S P A N A multivariate cluster analysis was applied to the S P E C I E S data subset for the Gabriola study area, to determine the algal species assemblages represented in the data. The program used, called T W I N S P A N (an acronym for two-way indicator species analysis), is a F O R T R A N program which was developed in the late 1970s by M . O . H i l l (Hill, 1979). The program was designed as a 'two-way' classification which means that species are classified as well as the samples, with the output tabular matrix similar to the Braun-Blanquet table (Hill, 1979 and Gauch, 1982). The program is a 'polythene divisive classification' which begins with all the samples together, and successively divides the group into a hierarchy of smaller and smaller 'dichotomized' clusters (Gauch, 1982). T W I N S P A N creates a classification of the samples first, then uses that grouping to classify the species according to their occurrence and relative abundance i n each sample (Hill, 1979). The combination of features in T W I N S P A N has made it one of the most popular programs in community ecology (Jongman et al., 1995). A s a divisive technique (rather than the less-rigourous agglomerative approach) it begins by examining the largest differences within the dataset which usually coincide with major community features related to differences in history and environment (Gauch, 1982).  23  Advantages of using T W I N S P A N for community classification in this project include:  1) the full extent of the data available for species abundance and sample sites are used to construct the classification (i.e. the 'polythetic' approach, where groups are distinguished on the basis of more than single attributes (Krebs, 1994)); 2) both species and sample sites are classified; 3) the tabulated data matrix output makes the interpretation of the clustering easier to understand and explain (Gauch, 1982): and, 4) indicator species are named and can be used to assign new, unclassified sites to one of the groups indicated by T W I N S P A N (Jongman et a l , 1995)  Further a hierarchical classification system is preferred for examination of coastal biotic assemblages (Searing & Frith, 1995) and T W I N S P A N , "being both polythetic and divisive, is recommended for hierarchical classification because of its effectiveness and robustness" (Gauch, 1982). In the T W I N S P A N program, each division level is arrived at through a series of three ordinations. The primary ordination is a form of 'reciprocal averaging'. The second step forms a 'refined ordination', identifying differential species from the primary ordination and is the one used to determine the dichotomy of sites in the output, as to the samples' preference to one side or other of the dichotomy. A third ordination is also performed which identifies the diagnostic species for the site groups identified (Hill, 1979).  24  Another feature of T W I N S P A N , integral to the creation of the classifications, is the treatment of quantitative species abundance in a transformed scale of 'pseudospecies' (Hill, 1979). The levels of abundance are specified in a scale defined in the data input to the program and are called 'pseudospecies cut levels'. The inclusion of species abundance adds discriminatory information to the classification and the 'pseudospecies' created from the quantitative data can be considered as indicators for the site groups identified (Hill, 1979). Further explanation of the technique appears in H i l l (1979) and Jongman et al. (1995). Several trials of T W I N S P A N were run on the S P E C I E S dataset for the study area, as follows:  1) all species with weights recorded 2) species with weights recorded which occurred in more than 10 collections 3) species with weights recorded which occurred i n more than 5 collections 4) species with weights recorded of more than 1 gram and which occurred in more than 5 samples.  The most successful output seemed to be for the fourth trial listed: 'species weights over 1 g and occurring in more than 5 samples'. The T W I N S P A N output was the most complete when the smallest samples and those rarely occurring species were not included i n the analysis.  25  The quantitative species data used were the 'dry weight' (g/m ). Dry weight was 2  selected to be most comparable to other studies which generally use dry weight to measure biomass. In the T W I N S P A N analysis of species weights over 1 gram and over 5 times occurring, the 'pseudospecies cut levels' used were: 1  0 - 4 . 9 g/m  2  5 - 24.9 g/m  3  25 - 100 g/m  4  > 100 g/m  2  2  2  2  These cut levels were selected to follow the suggestion by H i l l (1979) for: 'present, a little, a lot, and more-or-less dominant'.  A wave energy/depth/substrate model Determination of shoreline wave energy and how that energy is present at depth in the nearshore subtidal is integral to the prediction and understanding of nearshore substrate. Substrate, wave exposure and depth interact and contribute to the definition of nearshore biota. Measuring or estimating these nearshore physical parameters are important to the development of an integrated biophysical habitat model. Nearshore wave exposure can be indexed by shoreline 'fetch' measurements - the uninterrupted distance the wind travels over the water. Heights and periods of waves generated by wind are determined by the wind velocity, duration and by the fetch. The 'fetch distance' determines the time over which energy can be transferred from wind to waves  26  (Komar, 1976). A modified fetch measurement is then, an index for wave exposure by being in effect, an index of the amount of energy transferred from the wind to the waves created. For the purpose of this study, the method of determining a 'modified effective fetch' has been used together with a site's 'maximum fetch' to calculate a shoreline site-specific 'index of wave exposure', as described by Howes et al. (1994). W i n d speed and direction, together with an estimate of 'fetch window' (the open-water area offshore of the unit over which waves can be generated by wind) are used to determine 'modified effective fetch'. The wider the fetch window, the greater the wave exposure of the shore segment (Howes et al., 1994). Including the 'maximum fetch' in the exposure model calculations includes an indirect estimate of the maximum wave height which may affect the shore unit. Table 3 outlines the lower-energy portion of the wave exposure model, which is relevant to the fetches measured in the Gabriola study area (Howes et al., 1994).  Table 3. M o d e l for determining wave exposure category at a shore unit, using 'maximum fetch' and 'modified effective fetch' (from Howes et al., 1994). Maximum fetch (km) <1 <10 10-50 50 - 500  Modified Effective Fetch (km) <1 1 - 10 10-50 very protected n/a n/a protected protected n/a n/a semi-protected semi-protected n/a semi-exposed semi-exposed  50 - 500 n/a n/a n/a semi-exposed  A n index of wave exposure at the surface permits an estimation of the wave energy effects at depth in the nearshore subtidal. The surface wave energy required for moving material at depth is related to sediment size, wave height, water depth, wave length and wave period (Komar, 1976).  27  Fetch data were used together with engineering tables and formulae in Komar (1976) to develop a prediction for 'wave base' - the depth to which wave-generated energy will move sand-sized substrate, for the wave exposures and wind speeds in the Gabriola study area (Harper, 1995) (Table 4). Table 4. Estimated wave heights and corresponding calculated depth of wave base for wave exposures (after Harper, 1995). Wave exposure  Protected (P) < 10 k m  Semi-protected (SP) 10-50 km  'mean' fetch used  <10km  35 k m  wind speed (kn) 20 30 40 50  height (m) 0.7 0.7 1.5 1.9  est. depth of wave base (m) 3.4 5.3 10.3 14.1  height (m) 1.1 1.9 2.6 3.4  est. depth of wave base (m) 7.8 14.5 20.8 28.8  Semi-exposed (SE) 50 - 500 k m •  150 k m  1.8 3.1 4.4  est. depth of wave base (m) 14.7 27.9 42.2  5.9  57.8  height (m)  Comparison of that the above predictions of wave heights to the actual recorded wave heights in the Strait of Georgia suggest that the height calculations are realistic. O n Sturgeon Bank, in waters 139 m deep off the mouth of Fraser River a few kilometres east of the study area, wave measuring buoys have recorded wave heights up to 2.7 m (Thomson, 1981) and this is for deep, not nearshore waters. Further, 10% of the time, the measured wave heights exceeded 1.2 m at Sturgeon Bank, and more than 30% of the time, the measured waves exceeded 0.5 m (Thomson, 1981), providing a rough index of the frequency of wave-generated disturbance on semi-exposed shoreline. The maximum fetch at Sturgeon Bank is about 100 k m which is comparable to that of the semi-exposed shoreline i n the Gabriola study area (Table 11).  28  The wave base depth was further generalized by Harper (1995) to suggest the relative amount of seabed disturbance due to wind waves for each wave exposure and for each depth interval in the nearshore polygon database (Figure 5 and Table 5). Table 5. Wave-generated seabed disturbance at each exposure class and depth interval (from Harper 1995). *  Exposure Class  0-2  protected (P) semi-protected (SP) semi-exposed (SE)  high very high very high  Relative disturbance by De 3th Class (m) 2-5 5-10 10-20 moderate low low high moderate low very high high moderate  >20 low low low  * definitions of disturbance frequencies: very high - seabed almost constantly subject to wave-generated disturbance high - seabed frequently subject to wave - generated disturbance moderate - seabed occasionally subject to wave - generated disturbance low - seabed rarely subject to wave - generated disturbance Harper (1995) continued to develop the model for predicting nearshore substrate at depth for each nearshore depth interval. Empirical observations show that nearshore substrate tends to show a size shift, from coarser to finer substrate material, moving from the beach towards offshore - 'seaward-fining'. Nearshore subtidal substrates tend to be finer in protected environments, even though the intertidal beaches may be comprised of similar gravel material at all three exposure classes (Table 5) (Harper, 1995). "For known intertidal substrate and exposure, provided by the physical shorezone mapping database, the nearshore substrate conditions for the various depth categories are predicted. The predictions are based on consideration of the 'wave-base' criteria [as in Table 5]. The model used the known intertidal information as a starting point, then applied the 'seaward-fining' assumption to predict nearshore substrates." (Harper, 1995)  29  Using the major intertidal substrate and 'shore-types', Harper (1995) assigned a predicted nearshore substrate description to each exposure and depth class (Table 6). This substrate prediction model was developed as a 'best-guess' for application in Baynes Sound area of the northern Strait of Georgia to compare several techniques for remote sensing of benthic substrate.  Substrate data and interpretations for the Gabriola study area The Foreman datasets recorded substrate information in detail both in the  site  description for the sample in the Q U A D R A T database and in many of the species' attachment codes for the collections in the S P E C I E S database (Summary in Table 7 and detail in Appendix A ) . These substrate observations can be located in a number of the nearshore polygons in the Gabriola study area and used as a 'test' of the substrate predictions from Table 6. Hydrographic charts and the original field sheets collected by the Hydrographic Service for preparing the charts contain information about subtidal substrate. Generally the Hydrographic Service's substrate information is not collected or described systematically, and is best for distinguishing 'soft' bottom (sand/mud) from 'hard'  (bedrock/cobble/boulder).  Categories are summarized in Table 7.  30  Table 6. Substrate predicting model for estimating nearshore substrate description from intertidal wave exposure and geomorphology (after Harper, 1995). Intertidal Characteristics Major Substrate ROCK  Sediment * none [1 to 5] gravel [6 to 10]  Exposure **  0-2  2-5  5-10  10-20  SE  R  SP P and V P  RGS SR  RGS GSR SM  GS S MS  SG SM M  SE  RG RGS SGR  RGS GSR SMG  SG SG MS  SG SM M  SP Pand V P  ROCK  SE  &  S&G [11 to 15]  RGS RSG SGR  GSR SGR SG  SG SM MS  SM  SP Pand V P  RS SR SR  SR S SM  S SM MS  SM  sand [16 to 20]  SE SP Pand V P  MS M  SE  SEDIMENT  SEDIMENT  *  Predicted Substrate *** by Depth Class (m)  S M  gravel [21 to 23]  SP Pand V P  G GS SG  GS GS SMG  SG SG MS  SG SM M  GS SG SG  SG SG SG  SG SM  SM  S&G [24 to 26]  SE SP Pand V P  MS  SE sand/mud  SP  S S  S S  S SM  [27 to 31]  Pand V P  S  SM  MS  MS M SM MS M  numbers in brackets list the 'shore type' summary codes describing intertidal form and material (complete key in Appendix B ) ** S E = semi-exposed, SP = semi-protected, P = protected, V P = very protected (see Table 3) *** substrate definitions used (see Table 7 detailed definitions from Howes et al. (1994): R - rock (continuous bedrock) G - gravel (mixture of pebble, cobble, boulder > 2 mm) S - sand (0.06 to 2 mm) M - mud (< 0.06 mm) The predicted substrate descriptions are listed in descending order of occurrence from 'most abundant' to 'least abundant', i.e. R S G indicates bedrock dominant, with some pockets of sand and gravel  Table 7. Substrate codes from various data sources, 'interpreted' to fit substrate categories defined in Table 6.  Table 6 symbol SM no equivalent G(>2mm) G (> 2 mm) G (> 2 mm) R  Howes etal. (1994)* sand or silt (<2mm) shell (in 'Biogenic') pebble (0.2 to 6.5mm) cobble (6.5-25cm) boulder/block(>25cm) bedrock  Foreman data in QUADRAT sand/mud shell discontinuous rock discontinuous rock discontinuous rock continuous rock  Foreman data in SPECIES sand/mud shell pebble (<2in) rock (2-12in) boulder(>12in) bedrock  CHS chart symbol SorM Sh G G R R  * also includes 'gravel' - mixture of pebble, cobble, boulder  The majority of the Gabriola study area polygons lack any information regarding substrate. The substrate/depth/wave  exposure model (Table 6) substrate predictions were  applied in those cases. Tests of the Table 6 predictions against other substrate remote sensing methods (mainly subwater acoustic signaling such as R O X A N N ) have shown reasonable collaboration of the results (Harper, 1995). In the absence of direct observations of substrate in the nearshore subtidal, the model appears to give a reasonable 'best guess' of the nature of the substrate.  Linking the datasets The various databases used in this thesis - both the portions of existing datasets and the new data compiled for this project, were connected for the analyses through a series of common attributes. The data manipulations and summations were done using A C C E S S , a relational database management program. The datasets and the linkages between them are shown schematically in Figure 6. Definitions of the database attributes are listed in Appendix A and B .  32  The physical databases were linked by the 'unit key' which was the unique identifier for the shoreline units in the Gabriola study area, and also attached the shoreline map information to the new subtidal polygons. Unit key numbers are the same as listed in the southern Strait of Georgia oil spill response atlas (Howes et al., 1993). Biological databases (the Foreman data and the results of the T W I N S P A N analyses) were linked through 'collection number'. The physical databases were then linked to the biological databases through a 'key' dataset which positioned each of the Foreman collection 'sites' in the appropriate 'unit'.  Subtidal profiles observed at Saltery B a y test site Observations from eight subtidal transects, collected in early September 1995, from several across-shore profiles at Saltery Bay Provincial Park were used to compare to the substrate and the algal-assemblage predictions developed from the Gabriola study area. The Saltery Bay area was used to test the nearshore model because a) Saltery B a y is in the same general marine area of the Strait of Georgia, b) the substrate and species descriptions were collected using the same classification system as was used i n this thesis, and c) the opportunity to collect and use these new data for the test of the nearshore model in this thesis was available at no cost. Observations of the substrate, macroflora and fauna were recorded by S C U B A divers by depth and horizontal distance on the profile. The slope distance and depths were measured on each profile and the transects were laid out across two areas, both at the campsite and at the boat launch area of the park (Figures 1 and 3). Both areas of the park are within about 1  33  km along the shoreline, and profiles were spaced approximately 100 m apart. Substrate was described following definitions of Howes et al. (1994) and the observed flora was recorded to genus level when possible or by general description (such as, 'filamentous reds, Laminaria'). N o collections or measured weights were taken of the algae observed. Percent cover of the substrate types was estimated and the relative abundance of the species was noted as abundant, common, or few.  34  RESULTS Physical attributes i n the Gabriola study area From the intertidal shore-zone mapping, the wave energy calculations and the Table 6 nearshore sediment predictions, a description of the nearshore physical environment in the Gabriola study area was created. From the digitized mapping of the southern Strait of Georgia shore-zone classification (Howes et al., 1993), shoreline lengths of the different shoreline types and different wave exposures have been summarized for the Gabriola study area (Table 8). Nearly 87 k m of coastline is included in the study area. Each of the shoreline wave exposures (protected, semi-protected, and semi-exposed) are about evenly represented (Table 8), confirming that the study area includes the range of wave exposures found in the Strait of Georgia. The morphology and material of the shoreline is about two-thirds bedrock-dominated (Table 8). Only about 15% of the study area coast has been classified as gravel or sand beach. Shore-zone processes and morphologies will, therefore, not be a major source of sediment in the nearshore subtidal, and one would expect the dominant substrate in the nearshore to be bedrock. For each nearshore polygon (Figure 5), a measure of the area was calculated by the digital mapping. The 'nearshore substrate model' in Table 6 was then applied to the nearshore depth intervals, following the wave exposure for each unit. The substrates predicted for each depth-interval area are summarized for the Gabriola study area (Table 9). Full details of the predicted areas of each substrate type and depth interval are listed in Appendix C .  35  Table 8. Summary of shoreline lengths by shoreline type and by wave exposure in the study area. Shoreline length in each wave exposure category (km) shoreline types* 1 to 5 (rock) 6 to 20 (rock & sediment) 21 to 31 (sediment) 32 (man-made)  description bedrock cliff, ramps, platforms sand &/or gravel beaches, associated with bedrock cliff, ramp or platform gravel, sand or combination beaches, inch mudflat and estuary man-made shoreline, usually dyking or fill totals by wave exposure:  totals by shoreline type (km)  VP 6.3  P 20.6  SP 5.0  SE 22.2  54.1  2.4  2.6  9.0  4.2  18.2  1.8  3.0  7.7  1.0  13.5  0.9  -  -  -  0.9  11.4  26.2  21.7  27.4  86.7  * A complete list of the 'shoreline type' code descriptions are included in Appendix B . Note that 'type 31 - estuary' does not occur within the study area.  Table 9. Summary of the areas of predicted substrates by depth interval of the nearshore polygons in the study area. Nearshore polygon, depth interval (m) 0-2 2-5 5 - 10 10-20 0.7 1.4  -  3.8 2.4  4.1  7.1  13.5  3.1  4.1  7.1  19.7  2.8  1.0  rock & sediment  1.7 0.9 5.4  total (km ) 2  2  -  rock-dominated sediment  total (km )  36  The 'test' for the Table 6 substrate prediction model within the Gabriola study area was the comparison of the substrates predicted to the substrate observations for specific collection sites in the Foreman data (Table 10). Only 14 shoreline units and 38 nearshore polygons had direct information about substrate from the Foreman collections to use as a check against the predictions of the substrate/depth model. The Foreman data from the quadrat substrate and from the attachment codes of the samples (Table 7) were interpreted to assign a substrate code to compare to the model. Sixty-six percent of the time, the observed substrate in the study area agreed partially or completely with the substrate predicted by the Table 6 model (25 of 38 occasions, Table 10). The best agreement was in the two shallower intervals ( 0 to 2 m and 2 to 5 m) where 20 of the 25 substrate observations agreed partly or completely with the predictions. N o t only were fewer samples in the Foreman data set i n the deeper intervals (5 to 10 m and 10 to 20 m) but there were also fewer successful predictions - only 5 of 13 deeper water substrate observations matched the predictions (Table 10). The substrate model under-predicted the occurrence of bedrock in the 5 to 10 m and 10 to 20 m depth intervals. A s the study area is generally bedrock-dominated shoreline (Table 8), less-than-predicted sediment in deeper water is not surprising. Nearshore substrate  descriptions were located on the  marine charts  and  the  hydrographic service field sheets within the study area and were assigned to the appropriate nearshore polygon (Appendix D ) . The intent was to use these substrate observations as another source of ground-truth information which might be compared to the substrate/ wave exposure model predictions (Table 6) for the study area. Few substrate annotations were  37  Table 10. Comparison of the 'substrate observed' at the units sampled from the Foreman sites to the 'substrate predicted' by the substrate/depth/energy model in Table 6.  Site  Unit  52 GACL 75 BI01 BI02 75 BICH 76 78 BW01 78 BW02 79 RR01 159 RX01 PYL1 161 199 VI01 199 VI10 199 VIII 200 VI07 200 VI08 200 VI09 203 VI05 203 VI06 209 VI03 VI04 209 VI02 210 1913 SI02 GARF 1930 complete agreement: partial agreement: did not agree:  Nearshore polygon 2-5m 5-10m 10-20m substrate observed [substrate predicted] R[R] R G [RGS] R [RGS] R G , Sh [GS] R[R] R[R] R S [RGS] R[SG] S G [RGS] Sh [GSR] R S [RGS] S M [GS] Sh [SG] R[SR] R[M] R[SR] R[MS] R [SGR] R [RGS] R[GS] R[R] R, G [RGS] R[SG] R[R] R G [SG] R [RGS] 0-2m  R[R]  6 (67%) 3 (33%)  Sh [GSR] R [GSR] R [RGS] S h G [RGS] S h G [RGS] R G [RGS] R [GSR] 11 (69%) 5 (31%)  R G S h [GS]  R[SG]  -  -  G R S h [SG] G R [SM]  2 (40%) 3 (60%)  3 (37%) 5 (63%)  38  found in the nearshore on the charts and the substrate descriptions tend to be too general for a precise description of the substrate. However, a total of 27 nearshore polygons in 20 shore units contained substrate notes. Six of the eight substrate notations in the 0 to 2 m and 2 to 5 m depth ranges showed at least part-match to the predicted substrate. In deeper water, as with the Foreman substrate observations compared to the model, fewer observations matched only 9 of 19 (Appendix D ) . The shoreline wave exposure is an important physical influence on the structure of nearshore habitat and algal assemblages. Using the formula for calculation of 'modified effective fetch' from Howes et al (1994) and Table 3, the shoreline wave exposure category was measured for each of the Foreman data collection sites (Table 11). None of the collections sites was classified as 'very-protected'. The majority of the sites are relatively high energy (semi-exposed) and are in bedrock dominated shore types (types 1 through 5).  39  Table 11. Exposure classes and fetch distances for the Foreman sites in the study area. Sites sorted by 'modified effective fetch'.  Site  Unit key  Shoreline type  RR01* DXFT* RX01 PYL1 BI02 BW01 BI01 GARF BW02 GACL BICH VIOl VI02 VI03 VI04 VI05 VI06 VI07 VI08 VI09 VI10 VIII SI02  79 218 159 161 75 78 75 1930 78 52 76 199 210 209 209 203 203 200 200 200 199 199 1913  1 2 4 13 1 1 1 2 1 3 4 3 4 1 1 12 12 1 1 1 3 3 1  Maximum Fetch (km)  Azimuth of Max. Fetch (°)  15 6 5 30 28 83 83 83 83 82 . 83 80 80 80 80 80 80 80 80 80 80 80 83  135 165 359 133 32 120 120 125 120 320 120 120 120 120 120 120 120 120 120 120 120 120 125  Calc. Modified Effective Fetch (km) 1 2 3 5 ' 9 22 24 24 25 30 32 33 33 33 33 33 33 33 33 33 33 33 35  Azimuth of Shore normal ** o 100 242 44 120 345 105 112 110 135 7 130 55 55 55 55 55 55 55 55 55 55 55 125  Exposure Code P P P SP SP SE SE SE SE SE SE SE SE SE SE SE SE SE SE SE SE SE SE  * These sites are current-dominated, not wave-process dominated. ** 'shore-normal' is the perpendicular of the shore line.  Analyses of the Foreman data The subset of the Foreman data for the Gabriola study area includes information for 359 collections. Each 'collection' is a single, quantitative sample event, at one of the Foreman collection 'sites' (Figure 2). Most of the Foreman collections came from southwest Bath Island site (BI01) and from the nearby south end of Sear Island (SI02) (Table 12). These two sites account for 83% of the total collections.  40  Table 12. Count of subtidal collections at each of the Foreman sample sites. number of collections  site  BI01 BI02 BICH BW01 BW02 DXPT GACL GARF PYL1 RR01 RX01  168 22  VI01 VI02  8 2 2 1 3 2 2 2 2  VI03 VI04 VI05 VI06 VI07 VI08 VI09 VI10 VII1  SI02  129  total:  site  number of collections 4 1 1 1 1 1 1 1 1 2 359 collections  Both Bath Island and Sear Island (BI01 and SI02) are in the 'semi-exposed' wave energy category (Table 11) and are from bedrock-dominated shorelines. The number of collections which represent the higher energy, bedrock shoreline in the Gabriola study area is high (Tables 10, 11 and 12) while the lower energy, sediment-dominated shorelines in the study area are poorly represented by the Foreman data collections. D r . Foreman's data collections were originally designed to repetitively sample BI01 and SI02 over a number of years to monitor changes in algal assemblages due to sea urchin grazing (Foreman 1977) and were not intended to be a synoptic survey of wave exposure or shoreline types. The Foreman data can then be considered as an excellent source of detailed information on certain substrate/wave energies in the Gabriola area, where the most collections were made, and not as complete a database for sediment dominated, low energy shorelines.  41  Because so many of the Foreman collections are from BI01 and SI02 sites, a close consideration of the confidence in the shoreline wave exposure calculations at these sites was made to assist in the definition of the assumptions and biases in the nearshore habitat model being created from these data. The wave exposure index calculated for the SI02 site is in fact, somewhat anomalous. The maximum fetch measurement for that site happens to be the same as the 'shore-normal' which puts it into the semi-exposed category, although the fetch window is quite narrow. The wave exposure at SI02 is probably closer to the semi-protected category. The site from the north side of Bath Island (BI02) where 22 collections were made (Table 12) has the shore-normal orientation to the north, and has a narrow fetch-window similar to SI02, but the maximum fetch direction is less and it falls to the right-hand of the fetch window, not the shore normal. The wave exposure category for BI02 is then a lower category (SP - semi-protected) than the two southeast-facing sites on the other side of the island. To determine algal assemblages for relating to physical parameters used to develop the nearshore habitat classification model, a portion of the Foreman S P E C I E S database was analysed using T W I N S P A N . The results from two T W I N S P A N runs are discussed below, the first larger run, called T W N 1 , was from all of the collections in the study area which had species with recorded weights over 1 g/m , and included those species which were recorded in 2  5 or more collections. The T W N 1 analysis included 331 collection numbers with 66 species in the clustering. The results are shown schematically in Figure 7 and summarized in Tables 13 and 14.  42  The first level of division by T W I N S P A N separates the two most different groups from each other (the 0* group and the 1* group). The main environmental difference between these two groups (Figure 7) appears to be related to depth because all of the 1* group are intertidal collections, where the 0* are nearly all subtidal. A t level 3, three subtidal algal assemblages are named (Figure 7). These are all relatively 'algal-poor' groups (i.e., sparse collections by species diversity or by weight). Peyssonelia is an encrusting alga and dominates the 000* group, Type 2, showing only a few occurrences of other species (mostly small reds) at low weights. The Agarum group (001* and Type 1) has the deepest average depth at 9 m, and it was Peyssonelia which was the most often associated species, although a few small reds and Monostroma also occurred. The Agarum-type showed a high preference for continuous bedrock (over 60% bedrock, 30% boulder) while the Peyssonelia group had over 85% attachment to boulder/cobble (Table 13). The detail of the attachment code for both Agarum and Peyssonelia shows that both of these algal assemblages are strongly associated with immobile substrate, either bedrock or boulder. The Calliarthron group (011* and Type 5) is all from 1973 and 1974 at B I 0 1 , and represents the 'sea urchin grazed' algal type. Samples in subsequent years at this site measured recovery of the flora (Foreman 1977, Manson 1993) and Type 5, as an 'algal-sparse' assemblage by itself, disappears. The urchin-grazed Type 5 becomes a minor component of other types (Foreman, pers. comm., 1996). This grazed type is also strongly associated with bedrock substrate (84% attachment, (Table 13)). Four subtidal algal assemblages can be determined at the 6th level. The Zosteradominated group (010000 and Type 4) shows a wide depth range but is virtually all on  43  sand/shell substrate, showing a strong association with fine substrate. The Zostera type is the only group which showed significant amounts of pebble in the species attachment codes (Table 13), with preferred attachment as 50% pebble/cobble and 50% sand/mud/shell. Overall there are few records with the species Zostera marina occurring in the study area (only 21) but those where the attachment is specified shows the substrate as 'sand/mud' and all in the type 4 group. Other indicator and associated species of the soft-bottom type 4 group are:  Monostroma, Laminaria saccharina, Chondracanthus (Gigartina) exasperata and Gracilaria (Table 14). Not all of the collections i n the type 4 group have Zostera in the sample but nearly all have Monostroma. Monostroma by itself is not a good indicator of substrate, depth or algal-assemblage. A s a fast-growing, annual alga, it is not specific to any one substrate, depth or cluster group. The Laminaria.  saccharina/ Nereocystis-group (010001 and Type 3) shows a  relatively narrow depth range, averaging 4.4 m, and is found only on bedrock  or  boulder/cobble substrate (Table 13). The type 3 group shares a few species with the algal-rich group (Figure 7) but is distinguished by the co-occurrence of Nereocystis with L. saccharina. Amplisiphonia pacific a was also identified by the T W N 1 analysis as an indicator of this assemblage, although it also occurs in other 'algal-rich' groups (Table 14). The largest group in T W N 1 was the 'algal-rich' group which was run alone in a separate analysis ( T W N 2 ) to look for subtle trends within the group not detected by T W N 1 (Figure 8, Table 13 and 14). The group analysed consisted of 119 collections and 46 species. A l l of the algal-rich group are found mainly on continuous rock substrate at semi-exposed shore units, and are generally shallow subtidal (Table 13 and Figures 9 and 10). Clear  44  distinctions between these groups are obscure; however, four subgroups of the algal-rich type are detected by the T W N 2 analysis and some patterns can be distinguished. The first separation in T W N 2 is the Nereocystis - Laminaria groenlandica group (1* and Type 7). Comparing to the other Nereocystis group (Type 3) shows that not only are there different associated species but also the substrate preference of the group is different (Table 13). The Nereocystis/ L. groenlandica algal-rich group was found only on bedrock. The group site is relatively large (22 collections) and it is probably significant that all are found only on continuous rock. The overall depth ranges of the two groups with Nereocystis as indicators (Type 3 and 7) are not obviously different (Table 13). A small group of the algal-rich (type 9) were identified in both T W I N S P A N analyses, and are the shallowest-occurring group of the algal-rich - the mean depth is near zero datum (Table 13). The indicator species is Microcladia borealis. Sargassum muticum is also noted as an indicator species for this type. A few records for Sargassum are also at collections in the Zostera type 4, which has a similar shallow water depth range but is sand/mud substrate. The largest group in T W N 2 is ' O i l * ' ( T y p e 6, Figure 8) and is dominated by several small red algae (Table 14). This group may be further subdivided into two groups which include: 1) all of the collections which are not at BI01 or SI02 in algal-rich group 2) many of the 1977 winter sites  The most diverse group of T W N 2 (010* and Type 8) includes 33 sites in the shallow subtidal, showing a narrow depth range with an average depth of 1.1 m. One-third of this  45  group contains the greatest abundances and number of the species in T W I N S P A N analyses, and might be considered to be the 'true' algal-rich type. O f these, eight out of ten collections are from Sear Island (SI02) in 1972. Possibly the most-alike of the 'algal-rich' are those in type 8, which are all BI01 and SI02, and the rest of the 'algal-rich' collections are put together in type 6. The 'algal-richest' group 8 also shows the narrowest standard deviation of depth and is only found at shallow subtidal - less than 2 m and nearly all on bedrock substrate.  46  a  o •a  CO s o  •J  3' 3  -2 5  6  o«  hi  •IS-  00 O  H  m  H  o  s  ss  os  '*8 §«•>"  CO  B O  •a  § •a  Si  •a s  r»  -S *> o&2  s: o: •a:  i  ALF tICH  rill si  § •a  *  s 33  I  - 2l a °.  a  O  v>  1  1  o a  8 a  SI  CO  * o I—S U -  I  s  25?  a B O  •a  •Si  o 8  ,a CO  a.  Si  B O  -a  "S3  i  o •3  s  L&<3. 5 (N  8  3  SO  .3  .3  47  o  W  X  g u < u e  8  OH  oo a,  oo GO NO in  D.  CS CS  S os  a  0  &•  U  w o-  •a oo  ° os os ^  oo  9 os  o  oo  to  W  oo oo 00 r t  9  OH  oo oo CS CO  os  w 00  OS  CS  ag Se  OH  to «  <N CS CS  o  w * # ie S '•S.-S oo co oo  o  OH 00 OO O-i W  i—i NO  NO  8 SI 82  o 5 .2  CD  S 2 ^  #  =3  CD -  -3  oo o  8  2 CS TJ- CS co uoS s m T*  o  o 2  82  •s I<u o  co  I  jn HS  <L>, O  HO  1-8*8 S  -  00  cu .5 pj Q  oo oo  cs  5  oo  NO  CS  i ^5  CO  U  ON  a3 * s -8  CS  p  3  5 -  CO  CS  2  NO  2  CO  •n  ON  vd  o  HH  in ©  ON CS'  CO CS  ON  in S d +  c  o in  co co  o o  •a 00  CD  5  s c  •2  u 'ft  H  1  1 0,  •4 S2  1 3  ^  gs  3 a  3  s  5  III  3  NO  <V3 OS  49  Table 14. Algal assemblages identified by T W I N S P A N clustering, with indicator species and associated species. Type #  group name  FROM TWINSPAN RUN: TWN1: Agarum/ Peyssonelia 1  2  Peyssonelia  3  Laminaria saccharina/ Nereocystis  4  Zostera/Monostroma  Calliarthron 5 FROM TWINSPAN RUN TWN2 algal-rich one 6  indicator species  Agarum fimbriatum Monostroma fuscum Laurencia spectabilis Peyssonelia pacifica Amplisiphonia pacifica Costaria costata Antithamnion defectum L saccharina Nereocystis luetkeana Amplisiphonia pacifica Zostera marina Gracilaria verrucosa Gigartina (sic) exasperata Monostroma fuscum Calliarthron tuberculosum Amplisiphonia pacifica Callophyllis firma Constantinea subulifera Herposiphonia plumula Plocamium coccineum Pterosiphonia dendroidea  7  Nereocystis/ L groenlandica/ Calliarthron  Nereocystis luetkeana L groenlandica Calliarthron tuberculosum  8  algal-rich two - 'algalrichest'  9  Intertidal algal-rich  Amplisiphonia pacifica Bossiella spp. Callophyllis flabellulata Corallina officinalis Constantinea subulifera Cryptopleura ruprechtiana Herposiphonia plumula Laurencia spectabilis Plocamium coccineum Prionitis lanceolata Pterosiphonia dendroidea Microcladia borealis  commonly associated species Peyssonelia pacifica Pterosiphonia spp. Gelidium spp. Lithothamnion spp. Monostroma fuscum Pterosiphonia spp. L saccharina Neoagardiella bailyei Sargassum muticum  Constantinea subulifera Corallina officinalis Cryptopleura ruprechtiana Iridaea (sic) cordata Laurencia spectabilis Monostroma fuscum Odonthalia floccosa Prionitis pacifica Rhodymenia pertusa Bossiella spp. Constantinea subulifera Corallina officinalis Laurencia spectabilis Iridaea (sic) cordata Callophyllis spp. Odonthalia floccosa  Bossiella spp. Corallina officinalis Iridaea (sic) cordata Laurencia spectabilis Monostroma fuscum Odonthalia floccosa Sargassum muticum  50  The algal assemblages' affinity to depth The 'mean depth' and the 'standard deviation' of the depth for each algal assemblage type are summarized graphically in Figure 9. Several of the assemblages can be separated from each other and their occurrence predicted simply by consideration of the indicator species observed and the depth at which they occur. For the purposes of constructing the model for predicting the algal assemblage at each nearshore map polygon, the number of collections of each algal type at each nearshore depth interval are summarized (Table 15). This table shows each assemblage's affinity to each nearshore depth interval and is a subjective gauge of the confidence in the predictions of occurrences of the algal assemblages, based on knowing the depth alone.  Table 15. Number of collections of each algal assemblage type in each nearshore depth interval. Number of Collections in each Nearshore Depth Interval Assemblage Type 0-2m  2-5m  5 - 10 m  10 - 20 m  1 - Agarum  0  2 - Peyssonelia  0  0 5  29 15  14 4  3 - Nereocystis/ L. saccharina  3  4 - Zostera  5  7 12  10 2  0 2  5 - Calliarthron  6 15 1  12 20 14  4 6 4  0 0 0  21 4  5 0  0 0  0 0  6 - algal-rich one 7 - Nereocystis/Laminaria groenhmdica  8 - algal-richest 9 - intertidal algal-rich  There are a total of 328 collections at sites within the study area, all sites, all years. 32 collections have no depth recorded, and 71 collections are intertidal and are not included in these depth intervals. A l l depths are i n ' m below 0 datum'.  51  5  E 3  aj  •a a  <D  >  -5+ALGAL ASSEMBLAGE TYPES 1 Agarum 2 Peyssonelia 3 L. saixharuia/ Nereocystis 4 Zostera/Monostroma 5 Calliarthron 6 algal-rich one 7 Nereocystis/ L. groeolandica S algal-richest 9 intertidal algal-rich  a.  T3  -10 •  -15-1  _  1  2  3  4  5  6  _  7  .  8  9  algal assemblage type  Figure 9. Algal assemblage 'types' determined from T W I N S P A N analyses, showing mean depth and standard deviations of depth range. Details shown i n Table 13.  52  Algal assemblage occurrence by year of collection Patterns in occurrence of the nine algal types by year and by site are also apparent in the attributes of these cluster analyses (Table 16). The urchin outbreak at Bath Island in 1973 with the resulting changes in algal communities and abundances has been documented by Foreman (1977). These changes are also reflected in this cluster analysis of the data for the Gabriola study area as the occurrence/ non-occurrence of several of the types seems to be related to the year of the collection.  Table 16. Count of collections of each algal type by year for: 1) all sites, 2) BI01 only and 3) SI02 only. 1) AL1L SITES i n study area, including BI01 and SI02 algal assemblaj?e type number total year 1972 1973 1974 1975 1977  by year  1  121 59 101 27 20  22 2 16 5 6  2 7 3 14 0 0  3 12 2 2 6 2  4  5  6  7  8  9  6 7 11 1  1 12  16 0 1  18 1  6 3 4 1  0  0  16 12 11 5 2  1  8  6  12  10 2 2  5 0  2) site BI01 on. y: grazed by urchins i n 1973 9 0 52 3 3 1972 2 2 0 0 1973 46 2 2 1 7 1974 40 1 5 11 0 0 1975 3) site SI02 on) y: not grazed by urchins 2 1 3 1972 49 13 1 0 0 1 0 1973 1 12 6 0 1974 45 0 0 12 5 0 1975 2 0 6 0 1977 20  10 0  10 0  0 0 0 0 0  5 2 5  intertidal  0  17 16 22 3 4  5  9  0 0  8 1 2  3 2  16 11  0  0  1  2  6 0 1 1 2  9 0  2 0 2 0 0  8 0 11 1  3 2  9 3 1  7 3 1  4  53  A t BI01, the grazed Calliarthron (type 5) appears in 1973 and 1974, then disappears. Type 5 does not occur at SI02. Both of the Nereocystis types (3 and 7) are well represented in the 1972 collections then less-well represented in later collections. The algal-rich types (6, 8 and 9) follow a similar trend. The occurrence of the deeper communities (Agarum and Peyssonelia, types 1 and 2) do not seem to relate to the year of collection.  The affinity of each algal assemblage type to substrate Detail of the substrate 'preference' for each algal assemblage was determined from the attachment coding in the S P E C I E S database (Table 13). The coding of the quadrat substrate data was not specific about the size of the 'discontinuous rock' and did not include enough detail to describe the substrate specifically for each of the algal clusters revealed by TWINSPAN. Several of the algal types had strong affinity to specific substrate. The Agarum and Peyssonelia types (1 and 2) were all bedrock or boulder attachment, with the Agarum type showing the higher preference for bedrock. The two Nereocystis types (3 and 7) showed different substrate preferences: the type 7 was found only on bedrock, while type 3 had about half the attachment codes as boulder/cobble and half bedrock. The algal-rich types (6,8 and 9) also showed high preference for the solid bedrock attachments. The only type which showed strong affinity to the sand/shell/cobble attachments was the Zostera type 4. Close examination of the substrate detail of the S P E C I E S database confirms that the study area and Foreman collection sites are sediment deficient. O f the 6400 species records with the substrate coded for the attachment of the alga, 178 are cobble and only 32 are  54  'pebble'. M o s t of the 'discontinuous rock' codes are for 'boulder', which are sediment with diameters greater than 25 cm.  Wave exposure and algal assemblage A s has already been noted, most of the Foreman collection sites within the Gabriola study area are in the semi-exposed wave exposure category (Table 11 and 12). One would expect that the sites characterized by sand/mud/pebble substrate, supporting the Zostera type 4 community would be at the lower wave energies (Table 5 and 6) and in fact, a higher proportion of the semi-protected sites do occur in the Zostera-type than at the hard-substratepreferring other types (Table 13). Because so few of the algal collections are from lower wave energy sites, the algal types from the T W I N S P A N analyses (except for the Zostera type 4), must be considered to be representative of the semi-exposed wave exposure. Some of the Zostera type collections are from site B I C H , in the channel between Bath and Saturnina Island. Strictly applied, the wave exposure for this site places it in the 'semiexposed' category, due to the relatively large maximum and shore-normal fetches (Table 11). If the exposure at that site was ' S E ' , according to the model in Table 6, fine substrate would not dominate in the shallow nearshore, the '0 to 2 m ' and the '2 to 5 m ' depth intervals. The wave energy at the shore would preclude sand/mud from being the dominant substrate where the relative disturbance at the seabed due to wave energy is 'very high' (Table 5). In fact, the Zostera records at B I C H are from depths of 2 m and less, and those quadrats indicated 'sand/mud' or 'shell' as the dominant substrate. In the case of the B I C H site, the presence of  55  Zostera is a better index of the site-specific detail of the 'micro-wave energy climate' at the shoreline than the fetch index calculations in Table 3 are able to detect.  The predictive nearshore biophysical subtidal habitat model The co-occurrence of the indicator algal species, at the depth interval and substrate typical of the assemblage are used together to create the nearshore biophysical habitat model which assigns a 'predicted algal assemblage type' to each nearshore polygon, based on substrate, depth and wave exposure (Figure 10). The general substrate codes, equivalent to the nearshore sediment prediction model in Table 6, are shown for each substrate/ algal assemblage category on Figure 10. A n interpretation of the detailed substrate  attachment  codes for each type from Table 13 is also included on Figure 10. The detailed substrate description for each algal type is necessary to assign the codes describing the 'observed substrate' and the 'observed algal assemblages' from the Saltery Bay profiles for comparing the field observations to the predictions from the habitat model. The Figure 10 diagram of the nearshore habitat model is restated in Table 17, where the algal assemblages are assigned to the specific nearshore depth polygons used in the study. The illustration of the habitat model in Figure 10 and the detailed assignment of the algal types to each mapped nearshore polygon were used together to interpret the substrate and algal species observations at the Saltery Bay profiles.  56  5  E 3 «  •o o  2  <D >  -5  Q. T3  -10 + GR (bcR)  SG (Spc)  .  1 5  j  4  1  :—:  2  3  RG (Rb) 1  :  1 3 5 6 9 algal assemblage by substrate type  7  1  8  Figure 10. A l g a l assemblages showing distribution by depth and by 'preferred' substrate. The substrate detail codes: S - sand, G - gravel, R - bedrock, p - pebble, c - cobble, b boulder. K e y to algal-type numbers: 1 - Agarum 2 - Peyssonelia 3 - L saccharina/Nereocystis 4 - Zostera/ Monostroma 5 - Calliarthron  6 - algal-rich one 7 - Nereocystis/ L. groenlandica 8 - algal-richest 9 - intertidal algal-rich  57  The predictive nearshore habitat model is based on a subjective assessment of the affinity of each algal species group to: depth, substrate, wave exposure and a consideration of the presence/absence  of algal assemblages due to the disturbance by sea urchin grazing  (Tables 13, 15, 16, and Figures 9 and 10). The prediction of most of the algal types can be made by taking into account the site's depth, substrate description and wave energy. Although the grazed Calliarthron type 5 is included in Table 17, no type 5 assemblages are predicted in the Saltery B a y test because sea urchins are not currently present at 'outbreak' populations. The order of the abbreviations in the substrate code show the substrate preference of the  algal type.  For  example,  the  GR  code  means  that  'discontinuous  rock'  or  boulder/cobble/pebble dominate with patches of bedrock. The R G bedrock/gravel code means that bedrock is dominant with lesser coverage of boulder/cobble.  Table 17. Substrate classes and depth intervals assigned to the nine algal assemblages derived from the T W I N S P A N analyses.  depth interval (m)  sand/gravel SG Spc*  gravel/bedrock GR bcR*  bedrock/gravel RG Rb*  bedrock R  0-2  4  9  2-5 5-10  4(?2)  (9) 3, ?2 (6)  3, 5, 6 (?2)  7, 8 (9) 7(5)  4(1)(2)  2, ?3 (1)  10-20  (1)  (1)  1, ?3 (2) 1  (1) (1)  * detail of gravel substrate: p - pebble, c - cobble, b - boulder ? indicates the algal assemblage type depth range extends partially into depth interval () indicates the secondary substrate preference of the algal type Key to algal-type numbers: 1 - Agarum 2 - Peyssonelia 3 - L saccharina/Nereocystis 4 - Zostera/Monostroma 5 - Calliarthron  6 - algal-rich one 1 - Nereocystis/ L groenlandica 8 - algal-richest 9 - intertidal algal-rich  58  Physical attributes at the Saltery B a y test site Physical intertidal mapping for the area is included in the northern Strait of Georgia ( N S O G ) database from the provincial physical shore-zone mapping (Reimer & Harper, 1993). Profiles at the Saltery Bay site are from two different locations: the boat launch and the campsite/Mermaid site (Figure 3). The profiles fall into four different intertidal shoreline units with slightly different mapped data (Table 18). Keys to the shoreline-type and the form and material codes in Table 18 are listed in Appendix B . The intertidal physical descriptions are the basis for the shoreline type summary codes used in the nearshore substrate model (Table 6) to establish the 'predicted substrate' for the subtidal profiles. From the intertidal mapping for these four Saltery Bay units, one would expect that the SP profiles in units 2478 and 2479 will show finer substrate in the nearshore subtidal than the S C profiles. That is, the 'form' in unit 2479 is 'beach' with 'pebble/cobble/sand'. Nearby, the 2478 unit is also dominated by cobble/pebble/boulder with a secondary component of bedrock. A t the S C profiles in units 2481 and 2482, the intertidal is dominated throughout by bedrock ramp, platform and reef. These details of the predictions of the subtidal form and material are compared to the profile observations below. Wave exposures calculated for the Saltery Bay test site units put the profiles in the lower-side of the semi-protected (SP) wave exposure category (Table 19). If differences are noted in the algal assemblage/substrate predictions, one of the reasons could be the difference in wave exposure between the Gabriola study area where the model was prepared and the Saltery Bay profiles.  59  Table 18. The physical shore-zone mapped information for the units at the Saltery B a y test site.  unit key from NSOG 2478 2479  2481  2482  site name (profile numbers at each unit bracketed) w. of boat launch (SP3 + SP4) boat launch (SP1 + SP2)  shoreline type  22 24  4  west end campsite (SC5 + SC6) mermaid site & campsite bay (SCI + SC2)  7  mapped detail from database across-shore form zone  material  A B A B B A B  Clia;Bb Bf;Pir Bb; As Bf Bt Clir Pir  U;At/Cg Ccpb;vCcpb/Rm At/Cpc;At/Ac Cpcs Cps Rm Rm  A B B B  Clis Pir; Phi Pfi F  At/Rm Rm;vCg/Rm Rm Rm  Note: key to shoreline-type and other codes in Appendix B .  Table 19. Wave exposure categories for Saltery Bay profiles in test site.  mermaid site (SC profiles) boat ramp site (SP profiles)  maximum fetch (km) 15  max. fetch direction  15  shore normal bearing  225°  modified effective fetch (km) 7  180°  exposure category SP  210°  5  180°  SP  60  Application of the nearshore substrate prediction model from Table 6 to the Saltery Bay results in the predicted substrates for each profile as shown in Table 20.  Table 20. The nearshore substrate model predictions attached to the nearshore depth polygons and profiles observed at the Saltery B a y test sites. Substrate codes defined on Table 6. Unit 2478 2479 2481 2482  profile code SP3, SP4  shoreline type 22  Exposure code SP  SP1, SP2 SC5, SC6 SCI SC2  24 4 7 inside reef:  SP SP SP P  0-2m  2-5m  5 - 10m  10 - 20m  GS SG RGS RGS SGR  GS  SG SM S SG MS  SM  SG GSR GSR SMG  MS SM SM M  Subtidal profiles observed at Saltery Bay sites Eight across-shore subtidal profiles were measured at the Saltery Bay test site. The across-slope distances were recorded for each different substrate type observed and for each macroalgal assemblage observed. The observed substrate and observed algae for four example profiles (SP2, SP4, S C I , and S C 6) are presented (Figures 11, 12, 13 and 14). These profiles represent each of the four along-shore physical intertidal mapping units within the Saltery Bay test site (Tables 18 and 20). The profiles are plotted in a relative scale of horizontal distance by elevation above or below 0 chart datum. Notations on the profile comment on the substrate observed at depth and on the macroflora observed. A l l eight of the profiles are included in the 'observedpredicted' comparisons for substrate and algal types on the summary tables below. The substrate and algae observed in each of the four nearshore depth intervals are then 'interpreted' to the best summary descriptive code for the substrate (as in Table 6) and for the most-likely algal assemblage type number (Figure 10). Summarizing the observed substrate  61  and the observed algal assemblages in this way permits the comparisons to the nearshore substrate model predictions for each profile (Table 20) and for each algal type (Table 17 and Figure 10). The observed algal assemblage type is compared to the two different sets of predicted algal types for each of the eight profiles. 1) the 'predicted - predicted' algal assemblage where the nearshore habitat model from Table 17 is applied to the substrate predicted from Table 6 for each profile. 2) the 'predicted - observed' where the algal assemblage predicted is based on the substrate observations made by the divers along each profile. Both of the boat ramp sites (SP1 and SP2) are primarily sandy substrate with observed algal assemblage of the Zostera type 4 (Figure 11 and Table 21). The profiles further to the west (SP3 and SP4) at the boat ramp area are dominated by gravel/sands and have narrower intertidal beach adjacent (Figure 12 and Table 22). A t the campsite area of Saltery Bay, the S C I profile is across steep bedrock (Figure 13 and Table 23a) The S C 2 profile is shallow, (Table 23b) and dominated by sand substrate, and is in 'protected' wave exposure in the lee of an offshore reef. This S C 2 profile is more similar to the S P profiles (sand/gravel dominated) than the other three S C profiles (bedrock dominated). A t the west end of the campsite area, S C 5 and S C 6 cross a bedrock ramp and represent observations of bedrock-dominated profiles (Figure 14 and Table 24). S C 6 profile (Figure 14) has the greatest bedrock area in the under 5 meter depth range and should be the profile with the highest diversity of species (Figure 10).  62  A subjective summary of the number of matches between substrate model and the nearshore algal assemblage model predictions (Table 6 and Table 17) and the observations for the substrate and for the algal assemblages ('predicted - predicted' and 'observed - predicted') are listed below in Table 25. M o s t of the substrate predictions at least partly-matched the observed substrates - 22 out of 29 comparisons (Table 25a). Three were a complete match. For the 'predicted-predicted' comparison between the algal types predicted for the predicted substrates, in 19 of 29 depth intervals, the predicted algal type partly matched the observed algal type (Table 25b). Only three had 'no match' between the types predicted and the algal types observed and 7 comparisons were a complete match. When the comparisons between the predicted and actual observed algal types occurring on the observed substrates, the number of complete matches increases to 13 of 29, and no depth intervals had 'no match' (Table 25c).  63  64  8.1 • 00 em.S  c  jg  a, CO  00  £  •13  c/3  2  a,  2  _  * §  C3  a  so  a)  o  CO  •3 S ^ is £ ? XCOI  CQ  1/3 3  C/3  i a-s «s 03  CO  C/3  g  to ^ eg  111 .2 «2 *  1 -a 5 1)  «  O  S a. •S .2  * 2 -S  8"S3  O  ID  C  • 3 2 E  "8  Q,  O  £ 3e e 00  •8.8  3  c  e3  B ffl 5*  3 £ B 3  *a  13 ^ >  S to  (in) amrep o mag uouBAarg  u  x  D X  =  52 S  c  O . § rS CN*  8  3 DO  65  a x>  a co  J& vs 8 CL, " '32  .s § cn  *3  — -5 £ ^•S c 52,  §  c  U T3 O O  to  s  CL,  is  00 X>  E E  s  </3  •a c  3  1  00  CS CO  oo tU CL, >•>  oo X>  w  Cd  oo O DO 03  »—*  a So £  o 3 g e E * •a 8 "R OO  o c T3 £> O J= «1 VI VI  O  o  <  «- 1 Q,  eo  a *a .a (U  CL,  c -a pj T3 ±3 - -s £  6 .8  Id §  -  "  0 / 2  co  -3  « o S  • i8 f c £ " T3  u  oo X) O T3 C  aU  •a 8o c c  CS  C3  CL,  as a x> j= CO M  "O  (ra) U H U B D o uiojg uopjAajg  .  <D  g  u,  ^| § O  X5  -*  S9 £c -5 c  en  2  3 04  66  67  4) 00 03  T3 4)  co  B  o  CD  DH 00  £;  <%  1  % ^ -§ •§ V-i  _00  T3  -t->  CN  .2 « 1? 2  CO  -O  (D  1) co  Q,  0)  B  00 03 X  03 Ul -4-1  00 O X  X  8>-o H \ S ^ S  I-a -  oo-g ^ C  g g  "8  co  OH  is  co  C C H X  OH  o  3  5  f> 00 u> CO 4) 03 00 13 O  O0T3  M  T3  5  -if  o3  CN  PL,  O  CM - O  4> &  CN  C  CO CO  OH  4) ^  CO  § x, co o  4)  t3 is  e  O  00  CO CO  OH  CO  -S  O  g  _  ST> DH  oo CO  6 g PQ -2  a.a T3  CO  1  CO  CM  oo 4) S3 Xco O  4)  s  3 C ^4) CM  T3  O  T3  o3  "C  oj  CN  CO  CN  DH  CO  03  co  X  3  oo  t  CO  DH  4)  CO  O  co  o  CO  O  4) ^  CO  O co  DH  4)  oo  CO  •£  CO  O  CO  O CO  O  T3  i  is CO  0)  Si"  4) T3 *-» fl? 4)  03  |!  1  CN  W T3  N-H  1  O  ON  T3 4)  1  3  J  OH  X  co •£ oo O  03 Q , CO 4> u, QO . O o3  <N  CO  43 •  X  b-2  o  DH  00  is co  P  O  •§ e co  e  -a  X  DH  -S  is C oo 4) XI c« oo  CO  C M X 3  00  s % 3  *—I CL,  lM  £ .a  CO  DH  oo  CN  CO  oo  a g  M T3  3  6 CO cfl  o3 4> tJ  . T3  •X  w  B  © 2  *e3  is oo  CN  xi  "c3  O oo "O X c  On CO  •a -a  DH  CO  "c3  a *  CO  E •§  CO  00  CN  — 4)  5 2  43  co  oo  o3 _00  as  T—I  CO X 3  ^  T3  CO  DH  CO  E £  3  CN  CO  <§ *  03 Ui •*-» X 3 CO  1  CO  00  w  CM  o  4>  CN  T3  .a  *2 CO DH  CO  CH  DH  CO  CO CO  O  a  a  CO  4)  CN  4)  CN CN 4)  T3 O O 4)  CM 4) T3  O  E E oE  CN  ICNI  CN  I  68  Table 23a. Comparison details for Saltery Bay profile S C I , for predicted substrate compared to observed substrate and for the algal assemblage predictions based on both the predicted substrate and on the observed substrate. substrate predicted  substrate observed  algal assemblage observed *  SCI  SCI  SCI  RGS GSR SG SM  R  depth interval 0-2m 2-5m 5 - 10 m 10 - 20 m  RS R RS  9 9(4) 2or?3 1(?2)  algal assemblage predicted on predicted substrate * SCI  algal assemblage predicted on observed substrate * SCI  9(4)  7, 8 (9) 3, 6 (4) 7 1 (2) ?3 1  3 (4) (6) ?2 4 (1) (2) (1)  Table 23b. Comparison details for Saltery Bay profile S C 2 , for predicted substrate compared to observed substrate and for the algal assemblage predictions based on both the predicted substrate and on the observed substrate. substrate predicted  substrate observed  algal assemblage observed *  SC2  SC2  0-2m  SGR  2-5m  SMG MS M  SC2  algal assemblage predicted on predicted substrate * SC2  algal assemblage predicted on observed substrate * SC2  S  4  4(9)  4  S  4  4 ( 3 ) (6)  4  -  -  -  -  depth interval  5 - 10 m 10 - 20 m  * Bracketed algal type numbers are the 'less-represented' in the depth interval. Unbracketed types are those species assemblages interpreted as best typifying the depth interval  69  oo <U  o  00  T3  <U  •'g fi  03 00  X)  C  oo  NO  u  OS  oo  3  NO"  E  en  to CO  CO  03 CO  co  1)  CL, X i 0) § OT  T3  O <D  DO  C3 T3  OO  3 £  tu o  cd  O  oo  Ti  CH  3  8  a  oo  -2  o3  O  u  <U co O X!  ON  e  CO  03  "cd  co O  73  >N +->  00  I O O  £  CH  co  Xi  IS  03 co  ON  X3  c  co  o  a  1) + ->  3  cn  u, XI 3  D  •a  B  CH  3  4) — C  & Cw  co  Xi  TJ T3  U  0) 00  .3 * E 1) co oo  o3  "o3  ^ -3  NO  00  Xi  cS  O  S TJ 03 X>  PQ  co  c  £\2  13  CO In  CM  CO  0) 00  c2 1)  ca  3 E  co  co  oa  ^  O  J)  <D  <D co O  m U CO  X!  CM  T3  ON  B  c o  oo  CH  a £  NO  X)  <D co  CO  co  O  XS  CO  co  O  (D  ID  co  ^ -s  a00 <U£  ^ -e 03  8  U  CH  O  co oo  03  >  < u U co  O O co  co  S 3 C  ^ N O  CO  o  CO  CO CO  a  O  c/3  .S  J3  S3 THS 3 S2 X)  U co CO  2  (4 co  •i->  CH  00 u. .to ^> ^  CH  >-, >> HJ  ^  fll  3PX3  3  ^  03  t B  U -2  .s  Xi 03  T3  H  CH  SH  c < u o  ON  CO  4) >  "S co  u  CH  s s so  l<N  m o  l<N|  T—t  CO  T3  o3  W  fa  2 &  t  m  *  .S  7 0  Table 25a. Comparison of the number of times the predicted substrate code matched the observed substrate for each Saltery B a y profile. SPl  SP2  SP3  SP4  SCI  SC2  SC5  SC6  total  match part-match no match  -  -  1  1  -  -  -  3  4  3  3  2  -  -  -  -  3 1  -  3 1  1 1 2  3 22 4  totals:  3  4  4  4  4  2  4  4  29  Table 25b. Comparison of the number of matches by nearshore depth interval between algal assemblage predicted to occur on the predicted substrate and the observed algal assemblage for each Saltery Bay profile SPl  SP2  SP3  SP4  match part-match  1 2  3 1  2 2  -  no match  -  -  -  3 1  totals:  3  4  4  4  SCI  SC2  SC5  SC6  total  -  -  4  2  -  -  3 1  1 2 1  7 19 3  4  2  4  4  29  Table 25c. Comparison of the number of matches between the predicted algal assemblage predicted for the observed substrate matching the observed algal assemblage for each Saltery Bay profile. SPl  SP2  SP3  SP4  SCI  match part-match  2 1  4  -  1  4  3  no match  -  -  2 2  -  -  totals:  3  4  4  4  SC2  SC5  SC6  total  2  2  -  13  2  4  16  -  -  -  -  0  4  2  4  4  29  71  DISCUSSION  The predictive nearshore habitat model The overall objective of this thesis was to develop and test a predictive model of nearshore subtidal biophysical habitat by making use of existing physical and biological datasets. T o this end, the predictive model, as summarized in Figure 10 and in Table 17, was found to be at least partially successful in predicting the occurrence of algal assemblages at the nearshore depths and substrates at the Saltery Bay profiles. In the Table 25 comparisons of the algal assemblages predicted with those observed, the number of matches increased with increased confidence and detail in the description of the substrate at depth (Table 25c). Substrate is probably the most important determinant of the occurrence of a site's algal assemblage. The algal types determined in these analyses could be characterized in part by quite specific substrate information (Table 13) from the attachment codes of the collections. It seems that the species groups determined by the T W I N S P A N analyses did have different substrate attachment definitions, more subtle than just the distinction between 'sand' and 'rock'. Substrate difference - whether the substrate is mobile, fine material, or firm attachment of boulder or bedrock - is the most important substrate characteristic in determining which algal assemblage will establish at a site. For this reason, the general description of substrate using the summary codes as in the substrate prediction model works fairly well. It should be possible, however, to distinguish more subtle predictions of specific algal assemblages using the detailed attachment codes for each algal type when more detailed  72  substrate information for the nearshore subtidal is available. When substrate information is observed or predicted, it is important to gather as much specific information as is possible on the percent compositions of the 'gravel' components - the pebble/cobble/boulder makeup of the substrate. These ratios appear to be important to accurately describing the nearshore habitat and for increasing the accuracy of the predictions of occurrence of associated algal assemblages for the site. Disturbance and the patterns of recovery of the algal assemblages from  those  disturbance have not been included in the nearshore habitat model, but the algal assemblages determined in this thesis are comparable to those developed by Foreman (1977) and also reported in Levings et al. (1983) which did relate species assemblages to 'maturity' (Table 26). Foreman (1977) determined species 'importance value' by year, from frequency of occurrence and mean biomass for many of the algal indicator species for the groups found in the T W I N S P A N analyses (Figures 7 and 8 and Table 14). Annual algal species, in particular Nereocystis occurrence and abundance, appeared to be strongly related to the history of disturbance at the location studied. Nereocystis is an opportunistic annual plant which established rapidly in areas where the understory algal had been cleared by grazing of urchins (Foreman 1977) or by winter storms (Foreman 1984). Algal-rich indicator species {Laurencia and Constantinea, for example) were found at maximum species importance value three to six years after disturbance (Foreman 1977). A similar pattern was found with Agarum, the perennial kelp indicator of the deeper water Type 1. It seems then, that indicator species detail for the algal assemblages from the T W I N S P A N analyses could be used to provide information about the successional relationship between the assemblages determined.  73  Table 26. Comparison of the species assemblage analyses done previously on the Foreman database with the T W I N S P A N analysis for this thesis Lindstrom & Foreman 1978  Foreman 1979, and Levings et al. 1983  TWINSPAN analyses 1995  the entire Strait 359 sites from quadrats in the Bath of Georgia data set (all quadrats/ Island area, including east Gabriola species records geographicaly Island & northern half of Valdes located in the Strait - including Island, 123°37' to 123°45' and those from the northern regions) 49°3' to49°10'. basis of analysis 75 algal species (only those algal species, dominant species, algal species, with recorded weights species which occurred importance value: the mean dry over 1 g, and occurring more than 5 most often), wet weights In weight x frequency of occurrence times g/m2 •details of analyses or description of data used are not specified substrate substrate environmental substrate detail, depth, season depth - as description of gradients used to depth assemblage explain clustering IT (intertidal): Fucus, Corallina, intertidal (1*): Fucus, results of cluster from interpretation of six different cluster analyses, Monostroma; rock; above 0 datumMastocarpus, all above 0 datum, all analysis defined 6 groups: ULVA(intertidal): Ulva, (assemblage continuous rock, 62 sites Enteromorpha; mud; intertidal name, dominant intertidal: Fucus, 1. Agarum (001): Agarum Mastocarpus & other SHALO (shalow subtidal): spp & fimbriatum, average depth 9.2m, filamentous red algae Constantinea, Iridea, Laminaria discont or cont rock, 51 sites description of deep water: Agarum, sp; rock; 2 m mean depth environmental 2. Peyssonnelia (000): Peyssonnelia sparse understory of reds; IRCO: Iridea, Constantinea, parameters ) pacifica, Amplisiphonia pacifica, mean ~ 9 m ± 3.5 m depth Odonthalia; rock; 1.5 m mean average depth 7m, all cont. or sand: Zostera - Laminaria depth discont rock, 24 sites saccharina; - 3 m ± 0.5 m AGAR: Agarum; rock; 8.5 m 3. L. saccharina/ Nereocystis depth mean depth (010001): Nereocystis, Laminaria foliose reds: high % cover LAMI-1: Laminaria spp, saccharina; discont or cont rock, av of Iridea, Prionitis, Sargassum; rock; 3 m mean depthdepth 4m, 25 sites Odonthalia & other GRAZ* & CALLTU*: (grazed 4. Zostera (010000): Zostera filamentous reds; - 0 toby1urm chins) Calliarthron, marina, Monostroma fuscum, all depth Lithothamnion; rock; 3 m mean sand/mud or shell substrate, av depth kelp: Nereocystis, L. depth 3 m, 25 sites groenlandica, Costaria, 5. Calliarthron (011*). Calliathron KELP*: Nereocystis, with coralline & Constantinea, Iridea; rock, 4 m tuberculosum, av depth 3.0m, all filamentous reds; ~ 7 mm ±ea 2n depth 1973-74, all cont. rock, 23 sites m depth 6. 7, 8, 9 algal rich (01001*): Iridea LAMI-2: Laminaria spp; turf, sparse cover of cordata, Constantinea subulifera, sand/shel; 1.5 m mean depth filamentous & blade reds; ~ DEEP: Agarum; rock or sand; 8 Cryptopleura ruprechtiana, Odonthalia floccosa, Pterosiphonia m mean depth 6 m ± 2 m depth dendroidea, Plocamium coccinewn, * urchin - grazed areas look SAND: Zostera, Monostroma, like sparse kelp-type Laminaria; sand/shel; 3 m mean Laurencia spectabilis, Corallina officinalis, Amplisiphonia pacifica, understory depth Bossiella, Calliarthron tuberculosum, average depth 2.5m * assemblages are successional (range +1.7 to 9 m), all cont or stages discont rock, 119 sites area included in study  124 summer quadrats, 1972, Bath & Sear Island  74  Without including the determination of the successional relationship of the algal types in the nearshore habitat model, the nine species assemblages determined from the S P E C I E S abundances, can be generalized into three 'most confident' subjective algal types, separated by depth, substrate and indirectly by wave energy. These simplifications of the algal assemblages determined for the Gabriola study area are a way of bringing more confidence into the model's predictions. The general types are: 1. the shallow (less than 5 m), hard substrate types (all the algal-rich and kelp groups) 2. the shallow (less than 7 m), soft substrate types (the Zostera group), and 3. the deep (more than 5 m) types. A t present, the level of information available about the nearshore subtidal substrate is limited in most cases to generalizations from a substrate model like the one used for the predictions and comparisons in this thesis. Information about nearshore resources are also so limited that even broad generalities for key species or general algal assemblages w i l l be useful for coastal resource mangers' regional summaries. The three general nearshore biophysical habitat characterizations outlined above can be discussed in terms of the predictions and the observations at the Saltery Bay site. The shallow, hard substrate types developed in the nearshore habitat model (item (1) above) were the most difficult to assign to the observations at the Saltery Bay profiles. One reason that this happened could be that the Saltery Bay sites are representative of assemblages which have not recently been subject to disturbance - a factor which seems to be significant in structuring the algal species which were present in the Gabriola study area. Another possibility is that the dkgdl-xkfolNereocystis types are present in the semi-exposed (and presumably more nutrient-  75  rich) Foreman collections but are absent from the lower energy sites at Saltery Bay. The S C profiles at Saltery Bay were bedrock-dominated, the defined substrate of the algal-rich types, but the rich complex of species of type 6,7 and 8 algae were not strongly represented. Few collections included in the nearshore habitat model represent the sand - substrate community; however, the Zostera type 4 algal assemblage is so strongly associated with soft substrate that its occurrence can be predicted with some confidence in the shallow subtidal, sand substrate. The Zostera type often occurs as a pure stand of an unmistakable indicator species which means that the subtidal observations from Saltery Bay profiles can be classified as Zostera-type 4 without doubts about species identification, as might occur with 'mixed reds'. The correct prediction of the occurrence of the Zostera type on sand substrate at Saltery Bay accounted for many of the 'matches' to the predicted algal assemblages listed in Table 25. The wave exposure for the Saltery Bay profiles is semi-protected which also matches the 'preferred' conditions of the Zostera type, and likely also contributed to the number of successful predictions for this type at the Saltery B a y sites. The third general algal type that frequently matched the predictions for the Saltery Bay profiles and was confidently predicted for depths > 5m was the 'algal-sparse' Agarum type. The Agarum type is the only algal assemblage that is predicted to occur at depths over 10 m. From the video imagery of the transects at Saltery Bay profiles and the divers' observations, Agarum plants were seen at all but one of the transects' depth intervals greater than 5 m, even when the depth interval was dominated by sand substrate (Tables 21 to 24). Given the Agarum 'preferred' attachment of bedrock/boulder/cobble (Table 13), the importance to the algal species presence of even the patchy distribution of rocky attachments along a transect  76  which is primarily sand is noted below 5 metres on SP2 and SP4 (Figures 11 and 12). Agarum was observed on each, attached on the bedrock/cobble substrate. T w o of the general biophysical habitat characterizations described above are made up of single algal assemblage types, as determined in the nearshore habitat model. This means that from the substrate and depth interval definitions of those two types (Table 17), together with the calculation of nearshore habitat (depth interval and predicted substrate) from the nearshore subtidal polygon database created for the Gabriola study area, predictions about the occurrence of those algal assemblages in the study area may be made. These predicted occurrences must be considered as a 'first approximation' or 'best guess' of the nearshore subtidal habitat and are not suitable for site - specific detail, nor for consideration as being an exact measure of nearshore biomass. The substrate definition for the Zostera type 4 is S G (sand/pebble/cobble) and it is predicted to occur i n the three shallow nearshore depth intervals: 0 to 2, 2 to 5 and 5 to 10 m. The area of the S G substrate in the Gabriola study area in the < 10m depth range, is predicted to be a total of 1.28 k m (Appendix C ) . One can suggest then, ignoring the affects of wave 2  exposure as being implicit in model, that the type 4 Zostera assemblage will be found over about 1.3 k m of nearshore subtidal habitat. This is not considering that the Zostera type 2  might also occur in other substrate types where sand is present as a subcomponent, for example the S M (sand/mud) substrates. Further, the average biomass of the Zostera type collections can be calculated from the average of the dry weight for all of the species within type 4, from S P E C I E S data. The average biomass is calculated as 218.8 g/m . From just the Zostera records (18 in total), the 2  77  average dry weight is 77.7 g/m . Applying these estimates to the area of nearshore substrate 2  and depth suitable for the Zostera assemblage, a total biomass of 2.8 x 10 k g is predicted. 5  Considering the Zostera species average biomass alone, a biomass of at least 9.95 x 10 k g is 4  estimated for the Gabriola study area. Similar calculations could be made for the Agarum type. (The biomass estimates calculated could be more accurately called an estimate of the 'standing crop' from the years of the data collections i n the early 1970s.)  Analyses of the S P E C I E S dataset The T W I N S P A N multivariate analyses to determine the algal species assemblages in the Gabriola study area were integral to the development of the nearshore biophysical habitat model. Although the grouping of the algal species were examined for substrate, wave energy and depth, only a general assessment of the algal species assemblages over a temporal scale was considered, as mentioned above. The appearance and disappearance of the grazed type 5 assemblage (Table 16) has already been mentioned, as has the probable successional nature of the Nereocystis types (3 and 7). The occurrence or not of each assemblage type related specifically to the nearshore habitat model are difficult to compare year to year because the original collection strategy was not designed to sample each depth/substrate combination equally throughout each year of the study. For example, only a single collection was made at the 'undisturbed' SI02 site in 1973, the year of most marked difference in the algae at B I O L However, it is likely significant that over the five years of the collections, the algal-rich types 6, 7 8 and 9 as well as the Nereocystis type 3 did not seem to 'recover' to the representation that was found for each of these assemblages in the 1972, pre-grazing collections. The  78  Peyssonelia type 2 was not collected in 1975 or 1977 and it is not possible to say whether that is due to a community change or to sample gaps (Table 16). The species assemblages determined in this project similar to those found in earlier analyses of the Foreman data (Table 26). The species assemblages derived for this thesis are from a relatively small geographic area, while those for the Foreman study (Foreman, 1979, Levings et al., 1983) are for a wider area and include sites from northern Vancouver Island. Levings et al. (1983) suggested that the algal assemblages determined from the Foreman data could be described not only across a depth gradient, but also across a 'maturity' gradient related across a 'recovery - from - disturbance - gradient' over time. The grazed types ' G R A Z ' and ' C A L L T U ' (equivalent to the Type 5 Calliarthron  from Figure 7) are  replaced over time by ' S H A L O ' then T R C O ' (likely approximately equivalent to Type 6 'algal-rich one' and Type 8 'algal richest' from Figure 8) (Levings et al., 1983). The Levings et al. (1983) review also suggested that urchin-grazed types in deeper nearshore would lead to first the ' K E L P ' group then the ' L A M F . The Foreman data used in the analyses in this thesis represent the algal assemblages that were present in the Gabriola study area 20 year ago. Manson (1993) used an analysis of the Foreman 1970s' data to compare to several sets of recent samples of algae from Bath and Sear Island. Manson (1993) found probable long-term changes in the species present in the algal community, in particular increased abundance of warm-water preferring introduced algae. However, the overall type and structure of the algal assemblages is probably not very different.  79  Sargassum is an introduced species which has increased its abundance in the study area since the Foreman data were collected (Manson 1993). Manson also found that a 1970s data assemblage of 'Callophyllis-Cryptopleura-Nereocystis-Polyneura-Desmarestia'  is now  replaced by a group mcluding Sargassum and Lomentaria and other reds suggesting that the algal assemblages have changed over time. The groups of algal species determined by Manson (1993) are difficult to compare to the clustering of the Foreman data in this project; however, it is possible that the new Sargassum group which Manson refers to is now present in the substrate/depth niche which was part of the shallow Nereocystis type (algal-rich Nereocystis, type 7). The depth range of Sargassum in Gabriola data matches the depths where Sargassum was recorded from the Saltery B a y transects. In the T W I N S P A N analyses in this thesis, the largest group determined in T W N 1 was the algal-rich group. This group was subdivided into four slighdy different groups with T W N 2 . It is possible that the diverse-species collections in the algal-rich group might have been grouped by T W N 1 in the analyses for this thesis partly by the cutoff weights chosen for the pseudospecies. Larger species' weights were selected for by the analysis groups: 1 - 5 g, 5 - 25 g, 25 - 100 g and > 100 g. A s most of the diversity of the algal-rich group are i n small red algae, differences in species weights within the 5 to 25 g range would not be clustered separately in T W N 1 . In the T W N 2 analysis, a finer scale at the lower weights was defined in the cut levels: 0 - 2 g, 2 - 5 g, 5 - 10 g, 10 - 20 g and > 20 g. This may have helped to distinguish more subtle abundance differences between groups. The T W I N S P A N cluster analysis identifies each set of species from a collections site as either within an algal species assemblage or not in that group. That is, individual collections  80  are not classified as a mixture of assemblages. This might explain how apparently some of the groups are mixtures of several types and distinctions between the various 'algal-rich' types are not clearly defined. Another strength of the Foreman data that was not used specifically in this project were the details recorded in the dataset for individual species distributions and phenologies. These 20-year-old data could be compared to newly collected information for particular species, looking for a potential measure of long-term changes in algal characteristics in the Gabriola area.  The predictive substrate model The predictive model for nearshore substrate is an integral part of the nearshore biophysical habitat model for subtidal areas where more detailed substrate information is not available. A s discussed above, substrate is the most important environmental parameter for prediction of the algal assemblage at a site. The substrate predictions for the nearshore matched the substrate observations most often in the 0 to 2 m and the 2 to 5 m depth intervals (Table 10, and 21 to 24, and Appendix D ) . M o s t of the algal assemblages defined in the nearshore habitat model (7 out of 9) also occur in the 0 to 2 m and the 2 to 5 m depth intervals. Thus, for general predictions about nearshore substrate  and associated algal assemblages, the substrate  model provides a  reasonable estimate about two-thirds of the time. The substrate observations which did not agree with the predicted substrate were generally because the substrate model consistently under-predicted the occurrence of bedrock  81  in the nearshore subtidal. The substrate model was constructed for a trial in a lower-energy, sediment-rich nearshore area in Baynes Sound (Harper, 1995). One would therefore expect that the sediment predictions at the Saltery Bay sites might match the substrate model better than the semi-exposed areas in the Gabriola study site. A t Saltery Bay, the  sediment  predictions partly or completely matched the sediment observed in over 85% of the comparisons (Table 25a). In the Gabriola area, the part and complete match between the predictions  and the Foreman substrate observations  occurred  in about 66%  of  the  comparisons. Neither the Saltery Bay nearshore area nor the Gabriola study area are as sediment-rich, nor as low in energy and slope as the Baynes Sound area which could explain why the substrate model is over-predicting the accumulation of fine substrate in the over 10 m depths. Another way to improve the accuracy of the substrate model predictions is to refine the definition of the substrates for each of the 'shoreline-types' (Table 6), by making use of the details of form and materials for the intertidal shore units in the Strait of Georgia shoreline mapping database. The present substrate model in Table 6 predicts the substrate for the four nearshore polygons using only a generalization for a group of shoreline types. For the Saltery Bay units, the form and material descriptions are shown in Table 18. If the model were refined to consider not only the 'shoreline type' but the intertidal slope, shoreline form, and the relative amounts of sediment and sediment sizes, perhaps the predictive power of the subtidal substrate model would be improved. Additional parameters could be also measured for the database associated with the nearshore subtidal polygons from existing marine chart information, and included in the  82  nearshore substrate model conditions. A measure of mean slope from 0 datum to the 20 m depth contour would improve the forecast of subtidal material - steep slopes (for example, those greater than the 'angle of repose') would most likely be bedrock cliff, not accumulated sediment. A n estimate of nearshore slope, together with wave exposure estimates at depth, included in the nearshore substrate model predictions would be a more accurate reflection of the actual nearshore conditions. The subtidal wave-base calculations (Table 5) could be refined to perhaps make a better match between the substrate predicted at the deeper depths to that observed. According to the wave energy model in Table 5, both of the deep algal types (Agarum and Peyssonelia) with mean depths of 9 and 7 m, would be subject to 'high' disturbance at the S E sites where they occur. A l l of the other algal types would be subject to 'very-high' disturbance at the S E sites. A n indirect index of the wave disturbance at depth would be to compare the depth range of the Agarum and Peyssonelia types at shorelines with different wave exposures. One would expect to see that on lower wave energy shorelines, the upper limit of the type would be shallower (less disturbance) than at higher energy, deeper wave base sites. Unfortunately, with few examples of these types from low energy sites in the S P E C I E S database, testing the wave base model i n this way is not possible. Overall what is needed to better test the nearshore sediment model is a systematic substrate observation across a variety of physical shoreline types and nearshore depths. The comparisons from the model's predictions to the observed substrates in the Gabriola study area and at Saltery Bay can be considered a gauge of the potential for this predictive model, not a rigourous test.  83  CONCLUSIONS  1) The nearshore biophysical habitat model developed in this thesis can be used to generally characterize and estimate areas of the subtidal algal assemblages by depth and by predicted substrate within the Gabriola study area. The species assemblages identified in this are applicable to semi-exposed, bedrock dominated shorelines in the southern Strait of Georgia. The estimates of nearshore habitat from this model can be considered a reasonable 'best-guess', not a quantitative prediction, and are most useful for application at a regional scale.  2) Algal  assemblages  noted  in cluster analyses can be explained primarily  by  the  environmental variables of depth and substrate. Effects of wave exposure and disturbances are also noted but are secondary in the nearshore biophysical model developed here. Different algal species assemblages have different affinity to different substrates. Species attachment codes describe species assemblage substrate preferences.  3) Algal assemblage affinity to substrate is most clearly defined between soft substrate (i.e. sand/gravel/mud) and immobile rock (i.e. bedrock/boulder).  4)  The substrate model successfully predicts nearshore substrate occurrence at least partly correctly in over 60% of the comparisons in this thesis. The nearshore substrate model's approximations could be improved by the addition of slope designation; by consideration  84  of higher energy, sediment-deficient shores in the model's biases; and by using all of the available detail for the intertidal across-shore form and materials to predict the nearshore substrate. The present substrate model uses only the summary intertidal 'shoreline type'.  5) The predictions of algal assemblage show more matches to the actual algal assemblage at the Saltery Bay test site when the substrate is correcdy described. This is a demonstration of the importance of accurate substrate definition for successfully predicting algal assemblage occurrence. Algal assemblages can be generally predicted correctly, using the nearshore biophysical habitat model shown in Figure 10 and Table 17 for areas and conditions comparable to those for the Gabriola study area.  6) The Gabriola area shoreline is sediment poor, and most of the Foreman sites in the area represent relatively high energy shoreline in the Strait of Georgia. Algal  assemblages  developed from the species data reflect those physical conditions  7) Confidence in the predictions of algal assemblages from the model can be improved, and the application of the model can be generalized into three broader descriptions of < 5 m depth, hard substrate; < 7 m depth soft substrate nearshore; and, deep areas of more than 5 m. The second and third of these general nearshore habitat descriptions can be quantified on a regional scale and first-approximations of nearshore biophysical descriptions such as standing crop, and area of general substrate type can be drawn. These approximations,  85  though not strictly quantitative would be of value to coastal resource managers for use in areas where no specific information is available.  86  BIBLIOGRAPHY  Anderson, R J . and H . Stegenga. 1989. Subtidal algal communities at B i r d Island, Eastern Cape, South Africa. Botanica Marina. 32: 299-311. Anonymous. 1996. Canadian Tide and Current Tables V o l . 5 1996. Supply and Services Canada, Ottawa. 105 p. Ballantine, W . J . 1961. A biologically-defined exposure scale for the comparative description of rocky shores. Field Studies 1: 1-19. Emmett, B . , L . Burger and Y . Carolsfeld. 1994. A n inventory of mapping of subtidal biophysical features of the Goose Islands, Hakai Recreation Area, British Columbia. Unpub. rept. by Archipelago Marine Research, Victoria, B . C . for B C Parks, Prince George, B . C . 73p. Foreman, R . E . 1976. Nearshore Ecosystems Study, Annual report 1976. Unpub. rept. B E R P Report 76-3. University of B . C . , Vancouver. Foreman, R . E . 1977. Benthic community modification and recovery following intensive grazing by Strongylocentrotus droebachiensis. Helgolander wiss. Meeresunters. 30: 468-484. Foreman, R . E . 1979. Nearshore Biophysical Inventory Program, 1978-1979. Unpub rept. B E R P Report 79-1. University of B . C . , Vancouver. 33 pp. Foreman, R . E . 1984. Studies on Nereocystis growth i n British Columbia, Canada. Hydrobiologia 116/117, 325-332. Foster, M . S . 1990. Organization of macroalgal assemblages in the Northeast Pacific: the assumption of homogeneity and the illusion of generality. Hydrobiologia 192: 21-33. Fuller, I.A., T . C . Telfer, C . G . Moore, M . Wilkinson. 1991. The use of multivariate analytical techniques in conservation assessment of rocky seashores. Aquatic Conservation: Marine and Freshwater Ecosystems. V o l . 1: 103-122. Gauch, H . G . 1982. Multivariate Analysis in Community Ecology. Cambridge University Press. Cambridge. 298p. Harper, J.R. and P . D . Reimer. 1993. Physical Shore-zone mapping of the southern Strait of Georgia for oilspill sensitivity assessment, Final summary report. Contract report by Harper Environmental Services, Sidney, B . C . for B . C . Ministry of Environment, Victoria, B . C . 34 p.  87  Harper, J.R., W . C . Austin, M . Morris, P . D . Reimer and R. Reitmeier, 1994a. A biophysical inventory of the coastal resources in G w a i i Haanas. Contract Report by Coastal and Ocean Resources, Sidney, B C for Parks Canada, Calgary, A B , 114p. Harper, J.R., H . R . Frith and M . C . Morris. 1994b. The B C coastal biotic mapping system, Field verification of mapping procedures. Contract report by Coastal & Ocean Resources, Sidney, B . C . for B . C . Ministry of Environment, Lands & Parks, Victoria, B . C . 15p Harper, J.R. 1995. Seabed classification evaluation: Fisheries geomatics project. Contract report by Coastal & Ocean Resources, Sidney, B . C . for Terra Surveys, Sidney, B . C . 31p. Harper, J.R., M . Morris, P . D . Reimer, R . Frith. 1995 . Biophysical shore-zone mapping project for the west coast of Vancouver Island, 1993 -1995. Contract report by Coastal & Ocean Resources, Sidney, B . C . for B . C . Ministry of Environment, Lands & Parks, Victoria, B . C . H i l l , M . O. 1979. T W I N S P A N - a F O R T R A N program for arranging multivariate data i n an ordered two-way table by classification of the individuals and attributes. Section of Ecology and Systematics, Cornell U n i v . Ithaca, N e w York. 60p. H i l y , C , P Potin, J. F l o c ' h . 1992. Structure of subtidal algal assemblages on soft-bottom sediments: fauna/flora interactions and role of disturbances in the B a y of Brest, France. Marine Ecological Progress Series 85: 115-130. Howes, D . E . , P. Wainwright, J . M . Haggerty, J.R. Harper, E . H . Owens, P . D . Reimer and K . Summers, J. Cooper, L . Berg & R . Baird. 1993. Coastal Resources and O i l Spill Response Atlas for the Southern Strait of Georgia. Environmental Emergency Coordination Office, B . C . Ministry of Environment, Lands & Parks, Victoria, B . C . 317p. Howes, D . E . , J.R. Harper and E . H . Owens, 1994. British Columbia Physical Shore-Zone Mapping System. Resources Inventory Committee (RIC) report by the Coastal Task Force, R I C Secretariat, Victoria, B . C . 71p. Jongman, R . H . , C.J.F. ter Braak & O.F.R. van Tongeren. 1995. Data Analysis i n Community and Landscape Ecology. Pudoc, Wageningen. 299p. Keen, S. & R . E . Foreman. 1980. Nearshore Faunal Program, Final Report 1979-1980. Unpub. Rpt. B E R P Report 80-1. University of B . C . , Vancouver. 56 pp. Komar, P . D . 1976. Beach Processes and Sedimentation. Prentice H a l l , N . Jersey. 429p.  88  Krebs, C.J. 1994. Ecology: the experimental analysis of distribution and abundance. 4th ed. HarperCollins, N e w Y o r k . 801p. Levings, C D . , R . E . Foreman, and V . J . Tunnicliffe. 1983. Review of the benthos of the Strait of Georgia and contiguous fjords. Canadian Journal of Fisheries & Aquatic Sciences. 40: 1120-1141. Levings, C D . and R . M . Thorn. 1994. Habitat Changes in Georgia Basin: Implications for Resource Management and restoration. In: 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. Canadian Technical Report, Fisheries & Aquatic Sciences. N o . 1948. Lewis, J.R. 1964. The ecology of rocky shores. English Universities Press, London. Lindstrom, S . C 1973. Marine benthic algal communities in the Flat Top Islands area of Georgia Strait. M . S c . thesis. University of B . C . , Vancouver, B . C . 107 p. Lindstrom, S.C. and R . E . Foreman. 1978. Seaweed associations of the Flat Top Islands, British Columbia: A comparison of community methods. Syesis 11: 171-185. Manson, M . M . 1993. Changes in benthic marine macrophyte community structure in the Strait of Georgia: long-term and grazing responses. M . S c . thesis, Dept. of Botany, University of B . C . , Vancouver, B . C . 78 pp. Reimer, P . D . and J.R. Harper. 1993. Physical shore-zone mapping of the northern Strait of Georgia for o i l spill sensitivity assessment. Contract report by E M L Ltd. Victoria, for B C Ministry of Environment, Lands and Parks, Victoria. 56p. Searing, G . F . and H . R . Frith. 1995. British Columbia biological shore-zone mapping system. Contract report by L G L Ltd. of Sidney, B . C . for B . C . Ministry of Environment, Victoria, B . C . 46p. Schoch, G . C . and M . N . Dethier. in press. Scaling up: the statistical linkage between organismal abundance and geomorphology on rocky intertidal shorelines. Journal of Experimental Marine Biology & Ecology. 1996. Schiel, D . R . & M . S . Foster. 1986. The structure of subtidal algal stands in temperate waters. Oceanography & Marine Biology Annual Review. 24: 265-307. Stephenson, T . A . and A . Stephenson. 1972. Life between tidemarks on rocky shores. Freeman & Co., San Francisco. Thomson, R . E . 1981. Oceanography of the British Columbia coast. Canadian Special Publication Fisheries & Aquatic Sciences. 56: 291 p.  89  Yorath, C . J . & H . W . Nasmith. 1995. The geology of Southern Vancouver Island. Orca B o o k Pub. Victoria, B . C . 172 p.  90  APPENDIX A. List of data attributes and definitions for the Foreman database (from Foreman 1976) in the Q U A D R A T database attribute & definitions  column width  Collection number Tag number Date: Month Day Year Site (alphanumeric code) X-coordinate: m, locates quadrat Y-coordinate: m , transect distance from shore Quadrat size, converts to 'per m ' Sampling method A L - airlift SS - shore sample G C - general collection H A - hand H B - herbarium record M S - mussel study SB - succession block W S - weight sample Collection time Water depth, i n feet corrected to +,- mean water level Basal substrate 1 - sand/mud 2 - shell 3 - discontinuous rocky bottom 4 - continuous rocky bottom Bearing, degrees, for exposure direction Slope angle, degrees, average for quadrate Attenuation, cm max. vertical relief inside quadrat frame Derived quadrat values total number of species in quadrat total wet weight in g/ m total dry weight in g/ m total ash-free dry weight in g/ m Location: Latitude (degrees, minutes, decimal minutes) Longitude (degrees, minutes, decimal minutes) 2  1-4 5-6 7-8 9-10 11-12 13-16 17-19 20-22 23-26 27-28  29-32 33-35 36  37-39 40-41 42-44 45-69  2  2  2  70-76 77-83  in the S P E C I E S database attribute & definitions Collection number & tag number (same as Q U A D R A T ) Species code: first four letters of genus, first two letter of species and first letter of the variety Additional code blank or 0 - weight actual 1 - derived weight data Stratification code A - more than 1 m off bottom B - between 0.1 and 1 m off bottom C - between 0 and 10 cm off bottom D - encrusting E - epiphytic Attachment codes 0 - not attached 1 - sand/mud 2 - shell 3 - pebbles (less than 2 in diameter) 4 - small rocks (2 - 12 in diameter) 5 - boulders 6 - bedrock 7 - plants 8 - ariimals 9 - attached outside quadrat Growth form Exposure factor Orientation of attachment 1 - horizontal 2 - less than 4 5 ° slope 3 - 45 to 90 degree slope 4 - vertical or overhanging (>90°) 5 - crevices 6 - generalist Blank Cover value 0 - 0 . 1 % or less 1 - 0.5% or less 2 - 1% or less 3 - 5% or less 4 - 10% or less 5 - 20% or less 6 - 33% or less  column width 1-6 7-13 14  15-16  17  18 19 20  21 22  92  7 - 50% or less 8 - 75% or less 9 - 100% or less Sociability codes 1 - growing singly 2 - grouped, several individuals 3 - small patches 4 - extensive patches 5 - pure populations Vigor 1 - dead 2 - poor 3 - fair 4 - good 5 - excellent Phenology (AOO to A 9 9 for animals) P01 - vegetative gametophyte P02 - male gametophyte, reproductive, beta spore P03 - female gametophyte, reproductive, gamma-spore P04 - plurilocular, reproductive male and female P05 - isomorphic vegetative plant P06 - juvenile P07 - vegetative sporophyte P08 - reproductive sporophyte P09 - post-reproductive or perennial thallus P10 - fertile, unspecified Abundance number of individuals per quadrat Derived data density, number of individuals per m Wet weight, g/ m Dry weight, g/ m Ash-free dry weight, g/ m Weight scale for estimated wet weights 2  2  2  2  0 - minimum value 1 to 8 - for '< or = to O.OOlg to < or = to 10,000g Comments  23  24  25-27  28-30  31-34 35-40 41-47 48-53 54  55-80  93  A P P E N D I X B . List of Codes from Howes et al. (1994) used to describe physical processes i n the B . C . Ministry of Environment Shore-Zone Mapping. See also Figure 3. Detail for classification of 'shoreline type' for intertidal shore units (Howes et al 1994)  SUBSTRATE  ROCK  SEDIMENT  WIDTH  SLOPE  COASTAL CLASS Code & Description  WIDE (>30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  n/a (1) Rock Ramp, wide (2) Rock Platform, wide  NARROW (<30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  (3) Rock Cliff (4) Rock Ramp, narrow (5) Rock Platform, narrow  WIDE (>30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  n/a (6) Ramp w gravel beach, wide (7) Platform w gravel beach, wide  NARROW (<30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  (8) Cliff w gravel beach (9) Ramp w gravel beach (10) Platform with gravel beach  WIDE (>30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  n/a (11) Ramp w gravel & sand beach, wide (12) Platform w G&S beach, wide  NARROW (<30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  (13) Cliff w gravel/sand beach (14) Ramp w gravel/sand beach (15) Platform with gravel/sand beach  WIDE (>30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  n/a (16) Ramp w sand beach, wide (17) Platform w sand beach, wide  NARROW (<30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  (18) Cliff w sand beach (19) Ramp w sand beach, narrow (20) Platform w sand beach, narrow  WIDE (>30m)  FLAT(<5°)  (21) Gravel flat, wide  NARROW (<30ra)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  n/a (22) Gravel beach, narrow (23) Gravel flat or fan  WIDE(>30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  n/a n/a  NARROW (<30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  n/a  GRAVEL  ROCK +  SEDIMENT  SAND & GRAVEL  SAND  GRAVEL  SEDIMENT  SAND & GRAVEL  (24) Sand & gravel flat or fan n/a (25) Sand & gravel beach, narrow (26) Sand & gravel flat or fan  WIDE (>30m)  STEEP(>20°) INCLINED(5-20°) FLAT(<5°)  n/a (27) Sand beach (28) Sand flat (29) Mudflat  NARROW (<30m)  STEEP(>20°) INCLINED(5-20°) n/a  n/a (30) Sand beach  ORGANICS/FINES  n/a  n/a  (31) Estuaries  MAN-MADE  n/a  n/a  (32) Man-made, permeable (33) Man-made, impermeable  SAND/MUD  ANTHROPOGENIC  CURRENT-DOMINATED  (34) Channel  94  APPENDIX B - continued PHYSICAL SHORE-ZONE MAPPING SYSTEM FOR B.C. CODES OF 'FORM' - describes the form of a component, using a primary form descriptor, with or without a secondary form modifier (e.g. Ap, Bxfbu). Use of one primary form description indicates that it forms up to 75% of component. If two descriptors shown (separated by colon) then second is 25-50%. If three, each form covers 25% or more of component, (e.g. Bt:Plfi:Plfs). Where one form overlies another, codes are separated by slash (e.g. Bv/Ph).  A = Anthropogenic dolphin a breakwater b log dump c f float shell midden h jetty J dyke k marina m ferry terminal n port facility P seawall s wharf w outfall or intake X B = Beach berm b washover channel c f face I inclined multiple bars&troughs m n r s t w V  relic ridges single ridge/bar storm ridge low tide terrace washover fan veneer  C = Cliff eroding a passive P c cave fan,apron f t terraced i slopednclined (20-35°) slope: steep (>35°) s 1 height: low (<5m) m height: moderate (510m) height: high (>10m) h  D = Delta b bars f fan 1 levee m multiple channels s single channel E = Dune b blowouts i irregular n relic 0 ponds r ridge/swale parabolic P v veneer w vegetated F = Reef f horizontal i irregular r ramp s smooth L = Lagoon 0 open c closed M = Marsh h high 1 low c tidal creek e levee o pond  o = Offshore Island b barrier c chain of islets t table shaped pillar/stack P w whaleback 1 elevation: low (<5m) m elevation: moderate (510m) h elevation: high (>10m) P = Platform f horizontal h high tide platform i irregular 1 low tide pool r ramp t terraced s smooth tidepool P R = River Channel a perennial t intermittent m multiple channels a single channel T = Tidal Flat b bar,ridge c tidal channel e ebb tidal delta f flood tidal delta 1 levee s multiple tidal channels t flats tidepool P  95  APPENDIX B - continued MATERIALS CODES - The material descriptor consists of one primary term and associated modifiers (e.g. Cskb, Ad). Up to three descriptors may be written in order of importance to describe each layer. If only one descriptor is used, indicated material comprises 75% of the volume of the layer (e.g.Cs), if two, the first is 50-75% and second is 25-50% (e.g.Cs Be) and if three, each over 25% or more (e.g. Cs Be An). A surface layer of more than one layer thick can be described by prefix V for veneer (e.g. vCsk). - Forms with more than one layer or with more than one form descriptor are described by the order of the form and material descriptors (e.g. form = Bi;Ph, material = At/Cps;Rs indicates log material over pebble & sand beach berm, with platform of sedimentary rock. Beach is at least 75% of component, platform 25-50%).  a c d r t w  Anthropogenic metal concrete debris rubble logs wood  = Biogenic  c f 1 o p  coarse shell fine shell has trees organic little peat, organic sediment  = Clastic blocks (angular,>25cm) a boulders (round, subround,>25cm) b cobbles c diamicton (sand&>particles in matrix d silt&clay) fines (mix mud, silt, clay) f gravel (mix pebble, cobble, boulder >2mm) g k clay pebbles P rubble (boulders>lm) r sand s silt $ angular fragments (mix block & rubble) X  R = Bedrock rock type: i igneous m metamorphic s sedimentary v volcanic rock structure: 1 bedding 2 jointing 3 massive  DESCRIPTION OF SUBSTRATE Simplified from Wentworth scale GRAVELS boulder cobble pebble granule  > 25cm 6 to 25 cm 0.5 to 6 cm 0.2 to 0.5 cm  SAND from very coarse to very fine: all between .5mm to 2 mm MUD from silt to clay: smaller than .5mm also have: bedrock shell  96  APPENDIX C. Summary tables of substrates predicted for each nearshore depth interval, total areas from digitized mapping, substrates from model i n Table 6. Totals of areas i n 0 to 2 m interval nearshore polygons substrate area (m ) 266,505.4 GS R 1,684,904.4 107,802.1 RG RGS 661,579.5 332,096.2 RSG S 381,368.3 SG 201,205.6 343,874.1 SGR 1,317,109.2 SR 2  Totals for areas in 2 to 5 m interval nearshore polygons substrate area (m ) 148,633.2 GS 530,267.2 GSR RGS 1,004,399.6 S 59,853.5 159,990.4 SG SGR 152,764.9 SM 942,285.4 35,570.4 SMG 2  Totals for areas in 5 to 0 m interval nearshore polygons substrate area (m ) GS 1,160,356.6 MS 1,275,238.9 S 413,838.9 918,115.8 SG 356,744.2 SM 2  Totals for areas i n 10 to 20 m interval nearshore polygons substrate area (m ) M 1,259,801.5 MS 194,106.2 S 407,719.9 SG 2,209,180.5 SM 2,993,432.6 2  97  APPENDIX D. Substrate notes from Hydrographic Service field sheets and marine charts for nearshore polygons i n the study area. Unbracketed substrate code is for the substrate note from the Hydrographic Servide information, and the bracketed code is the 'substrate predicted' from the Table 6 substrate model. Unit  0-2m  from Field Sheet 1185-L 202 207 208 -  2-5m  5 - 10 m  10 - 20 m  -  -  R[SG]  R [RGS] R [RGS]  R[GS]  -  -  -  R, R S h [M]  from Field Sheet 1277 S and Chart 3475 R [GSR] 73 77 78 R[SM] 79 83 86 212 213 217 1911 1912 1916 1917 1919 1920 1921 1930 match part-match no match totals:  M[SG]  S [SR]  -  R S [SM]  -  -  R S h [RGS]  -  -  1 1 2  5 1 6  -  R[GS] R,GSh,R [MS]  S,R [M]  -  Sh [M]  R G [MS] R[MS]  S [MS] S[S]  Sh [S] R, R S h , Sh [GS1 1 3 5 9  SSh [SM] R S h [M]  -  R[SM] R S [SM] R [SG]  R, R S h , Sh [SG]  5 5 10  98  OO  CN  CO  co  e m o *—*  CN  NO  NO  1 s  CN Vi  CO  S 4) T3  co  CN  CN  CN  ON  00  CN  4>  CN CN  CO  00  >n  CN  CN  ve- co CN  NO  s  CN CN  in  CM  -a £ CN O  o  I*  4)  sX  X). NO  S  4)  3  Ml  •n  CO  co  o S o 4) on m  m  co  4)  CO  00  CN ON  CN  CO  CN  4)  CN  CN  CO  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items