UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Vegetation ecology of rock outcrop ecosystems of the Gulf Islands in the Coastal Douglas-fir zone, British… Sadler, Kella Darleen 2007

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

Item Metadata

Download

Media
831-ubc_2007-319192.pdf [ 18.54MB ]
Metadata
JSON: 831-1.0100752.json
JSON-LD: 831-1.0100752-ld.json
RDF/XML (Pretty): 831-1.0100752-rdf.xml
RDF/JSON: 831-1.0100752-rdf.json
Turtle: 831-1.0100752-turtle.txt
N-Triples: 831-1.0100752-rdf-ntriples.txt
Original Record: 831-1.0100752-source.json
Full Text
831-1.0100752-fulltext.txt
Citation
831-1.0100752.ris

Full Text

VEGETATION E C O L O G Y OF R O C K OUTCROP ECOSYSTEMS OF THE GULF ISLANDS IN THE C O A S T A L DOUGLAS-FIR ZONE, BRITISH C O L U M B I A by K E L L A D A R L E E N SADLER B.Sc , Simon Fraser University, 1997 M . S c , University of British Columbia, 1999 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Botany) THE UNIVERSITY OF BRITISH C O L U M B I A June 2007 © Kella Darleen Sadler, 2007 ABSTRACT Rock outcrop ecosystems of the Gulf Islands in the Coastal Douglas-fir (CDF) biogeoclimatic zone of British Columbia were investigated at multiple scales with the following objectives: (1) to refine distribution information for constituent species, (2) to investigate landscape (i.e. site-level) features that influence the patterning of native and introduced plant species of rock outcrop habitats, including (a) geographic position (latitude), (b) geology (rock type), and (c) grazing intensity, (3) to integrate vegetation patterns observed at each sampling scale (site, plot, microplot) to derive a classification scheme for rock outcrop vegetation, and (4) to interpret rock outcrop ecosystem dynamics and address conservation and management implications. A total of 311 plant species were identified from inventoried sites. The majority of plant taxa in rock outcrop ecosystems were herbs and bryophytes, and most of the rare species were mosses. Each landscape feature was associated with unique patterns of coverage and richness for different life form groups (studied by origin and rarity ranking). Uncommon bryophyte species richness was higher in southern sites where exotic vascular species coverage was highest. In contrast, the coverage and richness of native and uncommon graminoids was higher in northern sites. The overall richness of native herbs, and the richness of uncommon herbs was greatest in sedimentary rock sites, where exotic species also had greater coverage and richness. Ungrazed sites showed higher vascular plant species richness, whereas intensely grazed sites revealed higher bryophyte coverage, and greater richness of rare bryophytes. Sequential principal component analyses were used to classify vegetation and characterize scale-related vegetation-habitat relationships. Three major landscape categories were identified, based on the primary environmental gradients found to influence large-scale vegetation patterns: meta-igneous rock sites >49°N (META-N), meta-igneous rock sites <49°N (META-S), and moderately-grazed sedimentary rock sites (SED). Vegetation assemblages and fine-scale habitat relationships differed among categories, although there were some similar trends overall. The first gradient identified for each landscape category was related to microplot moisture. Within each category, the richness of native herbs (overall, and uncommon species) and the richness of bryophyte species was higher in microplots where seepage species had higher coverage. The second gradient identified for each landscape category was related to exposure and potential soil development. Within each category, bryophyte coverage and richness (and within M E T A - N and SED landscapes, the richness of rare bryophytes) was highest in microplots with the greatest exposure and lowest potential for soil development. Conversely, the richness of native vascular plant life form groups (particularly graminoids) was correlated negatively with this gradient. Rock outcrop ecosystems showed higher richness of native herbs and native mosses per unit area than did regenerating and mature forests (<5 years to >90 years old) in CDF zone landscapes. Rock outcrops also supported a greater proportion of uncommon herb and rare moss species per unit area than forested CDF habitats. Results showed a weak, negative relationship between distance of plots from forest edges and the number of native species "shared" with sampled CDF forests, indicating that edge proximity may be important for the perpetuation of some rock outcrop taxa, particularly woody plants and bryophytes. The richness of native outcrop "exclusive" species was correlated more to site-specific landscape factors (i.e. rock type, geographic position, and grazing regime) than to island size or rock outcrop ecosystem polygon size. Rock outcrop ecosystems are not equilibrial; instead they represent a collection of assemblages in different stages of continuous primary succession. Within each landscape category, unique fine-scale successional trajectories were identified in association with overall soil and vegetation development. Rock outcrop ecosystems, and their constituent microhabitats and species, are perpetuated by different types of disturbances that act at different scales and with varying frequencies, the effects of which depend on geographic position, rock type, and land use history. This thesis has shown that patterns of coverage, richness, and rarity may differ among life form groups. Whereas past management strategies for species recovery have focused predominantly on vascular plants, results presented here indicate that bryophytes also deserve our attention. Conservation objectives must be considered carefully, not only for the species and life form groups of primary concern, but also for the environmental variables that influence those species and groups. Successful management strategies will rely on multi-faceted and multi-scale approaches to ensure the perpetuation of diverse rock outcrop ecosystems within the CDF zone. i i i TABLE OF CONTENTS A B S T R A C T ii LIST OF T A B L E S vi LIST OF FIGURES ix A C K N O W L E D G E M E N T S xv CHAPTER I: INTRODUCTION 1 1.1 Ecological Significance of Vegetation in Rock Outcrop Ecosystems of the CDF Zone. 1 1.2 Literature Review of Vegetation Patterns on Rock Outcrops 3 1.2.1 Climate and Geography of the Study Area 3 1.2.2 Outcrop Morphology and Vegetation 6 1.2.3 Vegetation Assemblage Patterns on Rock Outcrops 8 1.3 Research Objectives 12 CHAPTER II: METHODS 15 2.1 Sampling Strategy 15 2.2 Data Analyses 19 CHAPTER III: RESULTS A N D DISCUSSION 25 3.1 Overview of Rock Outcrop Ecosystems of the Gulf Islands in the CDF Zone, BC 25 3.2 Rare Plants of Rock Outcrop Ecosystems of the Gulf Islands in the CDF Zone, BC...31 3.3. Landscape-scale Rock Outcrop Vegetation - Habitat Relationships on the Gulf Islands of the CDF Zone, BC 42 3.3.1 Site Geographic Position 42 3.3.2 Site Geology 50 3.3.3 Grazing on Meta-igneous Rock 58 3.4. Vegetation Assemblage Patterns in Rock Outcrop Ecosystems of the Gulf Islands in the CDF Zone, BC 64 3.4.1. Identification of Site- and Plot-level Vegetation-Habitat Relationships 65 3.4.2. Identification of Microscale Vegetation Patterns 73 iv 3.4.3 Synthesis of Multi-Scale Vegetation Patterns 85 3.5 Rock Outcrop Ecosystem Dynamics & Conservation Implications 89 3.5.1 Rock Outcrop Ecosystem - Vegetation Species Richness in Perspective 89 3.5.2 Rock Outcrop Ecosystems and Source vs. Sink Theory 91 3.5.3 Rock Outcrop Ecosystems and Island Biogeography Theory 93 3.5.4 Succession and Disturbance in Rock Outcrop Ecosystems 96 CHAPTER IV: CONCLUSIONS 101 LITERATURE CITED 104 APPENDIX 1. Specialized taxonomic keys for the study area: (a) Brachytheciaceae, (b) Bryum (c) Didymodon, (d) Grimmia and Schistidium, and (e) Racomitrium 113 APPENDIX 2. List of species identified at study sites, by life form group, origin, and provincial rarity status ranking. Average percent cover (Cover %) and overall site frequency in plot-sampled sites (Freq %) are also shown 129 APPENDIX 3. National and subnational conservation status definitions 140 APPENDIX 4. Summary of study site variables 142 v LIST OF TABLES Table 2.1.1. Summary of site information (ER=Ecological Reserve). Sampling strategy used noted as P (surveyed using plots), and I (unstructured bryophyte inventory only) 16 Table 3.1.1. Summary of site parent material (Rock Type) and grazing intensity (Grazing) 27 Table 3.2.1. Summary of species richness within life form groups, listed by categories of plant conservation status (see Appendix 3 for definitions of conservation categories) 32 Table 3.2.2. Vulnerable (S3-ranked), imperiled or critically-imperiled (SI - or S2-ranked), and unrankable (U), but possibly at-risk plant species recorded at study sites, by life form group (H=herb, M=moss, L=liverwort). Documented range and habitat information summarized from: Hitchcock and Cronquist (1973) and BC CDC (2006) for vascular plants; Lawton (1971), BC CDC (2006), Zander (2006), Spence (1988), Nyholm (1998), and Hastings and Greven (2006) for mosses, and Godfrey (1977) for liverworts 34 Table 3.3.1. Climatic stations (STN) used to characterize the trend in average annual rainfall across the study area. Station numbers (STN) are indicated in reference to Figure 3.3.1 43 Table 3.3.2. Sites used to compare effects of parent material 51 Table 3.3.3. Average site values for measured variables on meta-igneous (META) and sedimentary (SED) rock sites. Mean, standard error (SE), and /7-values (where /?<0.05) from two-sample t-tests comparing means are shown. Relationship direction is also shown, i.e. M E T A site means greater than (>) or less than (<) SED site means (N.S.=not significant, i.e. p>0.05) 52 Table 3.3.4. Sites used to compare effects of grazing intensity 58 Table 3.4.1. Pearson correlations of site-level environmental variables with PCA Axis 1 (Factor 1), generated from site-averaged species cover data 65 Table 3.4.2. Pearson correlations of meta-igneous rock environmental variables with PCA Axis 1 (Factor 1), generated from plot-averaged meta-igneous rock species cover data 66 Table 3.4.3. Pearson correlations of sedimentary rock environmental variables with PCA Axis 1 (Factor 1), generated from plot-averaged sedimentary rock species cover data 68 Table 3.4.4. Sites representing the four major rock outcrop ecosystem categories identified. Latitudinal position (Pos., i.e. north or south of 49°N), and grazing intensity or rock type shown 70 vi Table 3.4.5. Pearson correlations of northern meta-igneous rock environmental variables with the first two PC A axes 1 (Factor 1, Factor 2) generated from M E T A - N microplot species cover data (log-transformed) 73 Table 3.4.6. Summary of dominant habitat preferences of species with >5% frequency in M E T A - N microplots; NON-SPEC =habitat preference non-specific. Species shown by life form group (W=woody, G=graminoid, H=herb, B=bryophyte), and by origin for vascular plants (N=native, E=exotic). Species with >1% average cover within this data subset are indicated with larger, bold font. Species codes are listed in Appendix 2 75 Table 3.4.7. Pearson correlations of microplot species coverage and richness (by life form group and plant origin) with the first two axes of a PC A generated from M E T A - N microplot species cover data (log-transformed) 76 Table 3.4.8. Pearson correlations of microplot species coverage and richness (SI-S3-, and S4-ranked taxa) with the first two axes of a PCA generated from M E T A - N microplot species cover data (log-transformed) 76 Table 3.4.9. Pearson correlations of southern meta-igneous rock environmental variables with the first two PCA axes 1 (Factor 1, Factor 2) generated from META-S microplot species cover data (log-transformed) 77 Table 3.4.10. Summary of dominant habitat preferences of species with >5% frequency in META-S microplots; NON-SPEC.=habitat preference non-specific. Species shown by life form group (W=woody, G=graminoid, H=herb, B=bryophyte), and by origin for vascular plants (N=native, E=exotic). Species with >1% average cover within this data subset are indicated with larger, bold font. Species codes are listed in Appendix 2. Note PCA Axis 2 has been reversed to correspond with the direction of the exposure/soil development gradient observed for M E T A - N (Table 3.4.6). The (-) designation for Canopy-S (south-facing canopy cover) indicates an inverse relationship with the other variables shown 79 Table 3.4.11. Pearson correlations of microplot species coverage and richness (by life form group and plant origin) with the first two axes of a PCA generated from META-S microplot species cover data (log-transforrmed) 80 Table 3.4.12. Pearson correlations of microplot species coverage and richness (SI-S3-, and S4-ranked taxa) with the first two axes of a PCA generated from META-S microplot species cover data (log-transformed) 80 Table 3.4.13. Pearson correlations of sedimentary rock environmental variables with the first two PCA axes 1 (Factor 1, Factor 2), generated from SED microplot species cover data (log-transformed) 81 vii Table 3.4.14. Summary of dominant habitat preferences of species with >5% frequency in SED microplots; NON-SPEC.=habitat preference non-specific. Species shown by life form group (W=woody, G=graminoid, H=herb, B=bryophyte), and by origin for vascular plants (N=native, E=exotic). Species with >1% average cover within this data subset are indicated with larger, bold font. Species codes are listed in Appendix 2 83 Table 3.4.15. Pearson correlations of microplot species coverage and richness (by life form group and plant origin) with the first two axes of a PCA generated from SED microplot species cover data (log-transformed) : 84 Table 3.4.16. Pearson correlations of microplot species coverage and richness (SI-S3-, and S4-ranked taxa) with the first two axes of a PCA generated from SED microplot species cover data (log-transformed) 85 Table 3.5.1. Total numbers of native species identified from rock outcrops, by life form group, and corresponding numbers (and percent) shared with sampled regenerating and mature CDF zone forests (from Sadler 2004) 92 Table 3.5.2. Species with >1% average cover in microplots within each landscape category identified in Section 3.4 (META-N=meta-igneous rock >49°N, META-S=meta-igneous rock <49°N, SED=sedimentary rock); by life form groups (GRAM=graminoid, HERB=herb, BRYO=bryophyte, LCHN=lichen); origins (N=native, E=exotic) 98 viii LIST OF FIGURES Figure 1.2.1. Biogeoclimatic units of south-western BC. Dashed line indicates 49°N 4 Figure 1.2.2. Climate variation across eastern Vancouver Island. Graphs indicate average monthly temperature (lines) and average monthly precipitation (bars), 1961-1990 (from Warded al. 1998) 4 Figure 1.2.3. Simplified schematic diagram of topographic relationships among four common site associations of the CDF zone (from Nuszdorfer et al. 1991) 6 Figure 1.3.1. Conceptual diagram for this research: (a) the distribution patterns of native and exotic plant species (rare and common, overall and by life form group) were investigated in relation to landscape features (latitude, geology, grazing), and (b) the combined effects of landscape, local, and fine-scale environmental gradients were integrated to characterize the fine-scale patterns of constituent species groups 14 Figure 2.1.1. Study site locations. Sites where vascular plants and bryophytes were inventoried in sample plots are designated by black stars; sites where only bryophytes were inventoried are designated by white stars 15 Figure 2.1.2. Plot selection and layout. For each study site, a grid was placed over the rock outcrop polygon (shown in yellow), and three plot locations were randomly selected (Plot 1, Plot 2, Plot 3). Each 10m x 10m study plot contained eight nested lm x lm microplots and 25 25cm x 25cm microplots 17 Figure 3.1.1a,b. Summary of (a) vegetation and substratum coverage, and (b) plant species richness for plot-sampled sites, by life form groups and substratum-types: Woody=woody plants, Gram=graminoids, Herb=herbs, Bryo=bryophytes, Lchn=lichens, Bare PM=bare parent material, Org/Litter=organic matter (litter and soil) 25 Figure 3.1.2a,b. Typical rock outcrop ecosystem landscapes: (a) site view at Lasqueti Island, and (b) distant view of Saturna Island site (photo by Gary Lewis) 26 Figure 3.1.3a,b. Meta-igneous rock outcrop: (a) landscape view, and (b) fine-scale view...27 Figure 3.1.4a,b. Sandstone rock outcrop (a) landscape view, and (b) fine-scale view 28 Figure 3.1.5a,b. Conglomerate rock outcrop (a) landscape veiw, and (b) fine-scale view...28 Figure 3.1.6a-c. Shallow organic mats over (a) meta-igneous rock, (b) sandstone, and (c) conglomerate rock parent material 29 Figure 3.1.7a,b. Microhabitats characteristic to meta-igneous rock landscapes: (a) deep pockets of soil, and (b) ephemeral ponds 29 I X Figure 3.1.8a,b. (a) Ungrazed meta-igneous rock site (Winchelsea Island), and (b) sandstone rock site with relatively low grazing intensity (Drumbeg Park, Gabriola Island) -not sampled 30 Figure 3.1.9a,b. (a) Moderately grazed meta-igneous rock site (Texada Island), and (b) moderately grazed sandstone rock site (Channel Ridge, Saltspring Island) 30 Figure 3.1.10a,b. (a) Intensely grazed meta-igneous rock site (Sidney Island), and (b) intensely grazed sandstone rock site (Saturna Island) 31 Figure 3.2.1a-d. Pie graphs showing the proportions of exotic, native, and unrankable taxa, by life form group. Native taxa are depicted by provincial rarity status: rare (SI-S3), uncommon but not rare (S4), secure (S5), and unrankable plants identified within the study area 33 Figure 3.3.1. Rainfall trend across the study area, from Canadian Climate Station Averages (1961-1990) located throughout the CDF zone (Environment Canada 2006). Numbered labels refer to climatic stations listed in Table 3.3.1 42 Figure 3.3.2a-d. Relationships between latitude and soil properties: (a) average soil depth, (b) average standard deviation in soil depths, (c) average total carbon content in soil samples, (d) average total nitrogen content in soil samples 44 Figure 3.3.3a,b. Trends in vegetation (a) percent cover, and (b) species richness across a latitudinal gradient, by life form group (VASC=vascular plants, BRYO=bryophytes) 45 Figure 3.3.4a,b. Trends in vascular plant (a) coverage and (b) species richness across a latitudinal gradient, by origin (native, exotic) 46 Figure 3.3.5. Trends in exotic graminoid coverage along a latitudinal gradient 46 Figure 3.3.6a,b. Trends in native species richness along a latitudinal gradient: (a) graminoid species, and (b) woody plant species 47 Figure 3.3.7a,b. Trends in the percent coverage of (a) Sl-S3-ranked plants, and (b) S4-ranked plants along a latitudinal gradient 47 Figure 3.3.8a-c. Trends in the coverage of (a) Sl-S3-ranked bryophytes, (b) S4-ranked bryophytes, and (c) S4-ranked graminoids along a latitudinal gradient 48 Figure 3.3.9a,b. Trends in the richness of (a) S4-ranked bryophyte species, and (b) S4-ranked graminoid species along a latitudinal gradient 49 Figure 3.3.10. Average standard deviation in soil depth measurements within microplots, by parent material (CONG=conglomerate, META=meta-igneous, SED=sedimentary). Standard error bars shown 53 Figure 3.3.11. Average bryophyte coverage at sites, by parent material (CONG=conglomerate, META=meta-igneous, SED=sedimentary). Standard error bars shown 54 Figure 3.3.12a,b. Average coverage of (a) all exotic vascular plants, and (b) all exotic graminoid species in sites, by parent material (META=meta-igneous, SED=sedimentary). Standard error bars shown 55 Figure 3.3.13. Total number of native herb species at sites, by parent material (META=meta-igneous, SED=sedimentary). Standard error bars shown 55 Figure 3.3.14a-c. Total number of (a) exotic vascular plant species, (b) exotic graminoid species, and (c) exotic herb species at sites, by parent material (META=meta-igneous, SED=sedimentary). Standard error bars shown 56 Figure 3.3.15a,b. Total number of (a) S4-ranked plant species, and (b) S4-ranked herb species at sites, by parent material (META=meta-igneous, SED=sedimentary). Standard error bars shown 57 Figure 3.3.16. Total number of S4-ranked plant species at sites, by parent material (CONG=conglomerate, META=meta-igneous, SED=sedimentary). Standard error bars shown 57 Figure 3.3.17a,b. Average (a) percent canopy cover, and (b) plot vertical aspect (i.e. degrees from north) at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown 59 Figure 3.3.18a,b. Average (a) percent bare parent material, and (b) ratio of total percent carbon to total percent nitrogen in soil samples at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown 60 Figure 3.3.19a,b. Average percent cover of (a) total vegetation (vascular plants and bryophytes), and (b) bryophytes at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown 61 Figure 3.3.20a,b. Total number of (a) all vascular plant species, and (b) woody plant species at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown 61 xi Figure 3.3.21. Total number of native graminoid species at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown 62 Figure 3.3.22 a,b. Average percent cover of (a) all S4-ranked plants, and (b) S4-ranked bryophytes at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown 63 Figure 3.3.23. Total number of SI-S3-ranked bryophyte species at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH^high grazing intensity). Standard error bars shown 63 Figure 3.4.1. Ordination of sites in relation to the first two axes of a PCA performed on site-averaged species cover data. Plots are grouped by rock type (META=meta-igneous, SED=sedimentary) using 68.3% confidence ellipses 66 Figure 3.4.2. Ordination of plots in relation to the first two axes of a PCA performed on average species coverages within meta-igneous rock study plots. Plots are grouped by latitudinal position (North=>49°N, South=<49°N) using 68.3% confidence ellipses 67 Figure 3.4.3. Ordination of plots in relation to the first two axes of a PCA performed on average species coverages within sedimentary rock plots. Plots are grouped by grazing intensity using 68.3% confidence ellipses 68 Figure 3.4.4. Pie graphs showing average coverage of vegetation life form groups (Bryo=bryophytes, Lchn=lichens, Gram=graminoids, Herb=herbs, Woody=woody plants) and substratum components (Org/Litter=bare organic material and/or decayed plant litter, BarePM=bare parent material) within major rock outcrop ecosystem categories, organized in relation to soil development 69 Figure 3.4.5. Average minimum soil depth (in microplots) for landscape categories identified. Standard error bars shown 70 Figure 3.4.6. Average percent cover of exotic vascular plants in sites, by landscape category and life form group (VASC=overall, GRAM=graminoids, HERB=herbs). Standard error bars shown 71 Figure 3.4.7. Average species richness of exotic vascular plants in sites, by landscape category and life form group (VASC=overall, GRAM=graminoids, HERB=herbs). Standard error bars shown 71 Figure 3.4.8. Average richness of native herb species in sites, by landscape category. Standard error bars shown 72 xn Figure 3.4.9. Average richness of S4-ranked herb species in sites, by landscape category. Standard error bars shown 72 Figure 3.4.10. Ordination of microplots in relation to the first two axes of a PCA generated from M E T A - N microplot species cover data (log-transformed). Microplots are grouped by grazing intensity using 68.3% confidence ellipses 74 Figure 3.4.11. Ordination of microplots in relation to the first two axes of a PCA generated from META-S microplot species cover data (log-transformed). Microplots are grouped by plot edge class (EDGE=0-10m, OPEN=>10m from forest edges) using 68.3% confidence ellipses 78 Figure 3.4.12. Ordination of microplots in relation to the first two axes of a PCA generated from SED microplot species cover data (log-transformed). Microplots are grouped by rock type (CONG=conglomerate, SAND=sandstone) using 68.3% confidence ellipses 82 Figure 3.4.13. Summary of major fine-scale environmental gradients and vegetation properties across landscape categories (META-N, META-S, SED): correlations of PCA Axis 1 (moisture gradient) with percent coverage (%) and richness (#) of life form groups (Woody=Woody species, Gram=Graminoid, Bryo=Bryophyte) in microplots, by rarity status and species origin (Exo=Exotic, Nat=Native). Similar trends across categories are shown in bold font 88 Figure 3.4.14. Summary of major fine-scale environmental gradients and vegetation properties across landscape categories (META-N, META-S, and SED): correlations of PCA Axis 2 (soil and exposure gradient) with percent coverage (%) and richness (#) of life form groups (Woody=Woody species, Gram=Graminoid, Bryo=Bryophyte) in microplots, by rarity status and species origin (Exo=Exotic, Nat=Native). Similar trends across categories are shown in bold font 88 Figure 3.5.1a,b. Average number of (a) native herb species and (b) native moss species per 3000m2, in relation to CDF zone habitat type (RO=rock outcrop, CC=clearcut forest <5 years old, YNG=young forest 20-30 years old, MAT=mature forest 50-60 years old, and OLD=old forest >90 years old). Standard error bars shown for rock outcrop ecosystems (forested habitats represent the total area studied) 90 Figure 3.5.2a,b. Average number of SI-S3-ranked and S4-ranked species per 3000m , for (a) herbs, and (b) herbs and moss, in relation to CDF zone habitat type (RO=rock outcrop, CC=clearcut forest <5 years old, YNG=young forest 20-30 years old, MAT=mature forest 50-60 years old, and OLD=old forest >90 years old). Standard error bars shown for rock outcrop ecosystems (forested habitats represent the total area studied) 90 Figure 3.5.3a,b. Relationships between log number of rock outcrop species shared with CDF forested habitats and log distance to canopy (m), for (a) vascular plants, and (b) bryophytes 92 xiii Figure 3.5.4a,b. Relationship between log number of native rock outcrop "exclusive" vascular plant species and (a) log island size (ha), and (b) log polygon size 94 Figure 3.5.5a,b. Relationship between log number of native rock outcrop "exclusive" bryophyte species and (a) log island size (ha), and (b) log polygon size 95 Figure 4.0.1. A multi-scale approach to the study of vegetation in rock outcrop ecosystems permits the integration and reconciliation of prevailing theories in community ecology 102 xiv ACKNOWLEDGEMENTS I would like to thank my supervisor, Dr. Gary Bradfield for his assistance and guidance throughout this thesis. Thanks also to my committee members, Dr. Wil f Schofield, who generously offered his taxonomic expertise and enthusiasm for bryophytes, and Drs. Jeannette Whitton and Roy Turkington, for their support and advice during this research. I am grateful to the following sources of funding for my research: Natural Sciences and Engineering Council of Canada Post-Graduate Scholarship, University of British Columbia Graduate Fellowship (UGF), Bryology & Lichenology Fund (Vancouver Foundation) Bursary, Weyerhaeuser Ltd. research donation, and Islands Trust Fund research donation. In-kind support was provided by many generous agencies and individuals. Many thanks are extended to the following agencies for arranging permits, maps, and/or travel expenses: BC Parks (Bill Zinovich, Chris Kissinger), Canadian Wildlife Service (Peggy Ward), The Land Conservancy (Clint Abbott), The Galiano Club (Susan and Larry Friend), Proput Team Inc. (Brice Chapman), the Lyackson Band (Barbara Jimmy), and the Islands Trust Fund (Kathy Dunster). Numerous individuals provided invaluable support in the field, assisting with accommodation, data collection, and/or finding appropriate site locations: Kathy & Julian Dunster (Bowen Island), Richard Martin, Tony Law, and Lu Ackerson (Hornby Island), Karrie-Anne Friend & David Reid (Galiano Island), A l Gainsbauer (Lasqueti Island), Liz Webster (Savary Island), Charles Kahn (Channel Ridge on Saltspring Island), Rosalie Beech, Ralph & Mallory Pred (Reginald Hil l Strata on Saltspring Island), Ozzie Sexsmith (Sidney Island), Eric McLay (Valdes Island), Jan Kirkby (Pender Island), Harvey Jantzen (Saturna Island), and John Dove (Texada Island). I am most grateful to several individuals for assistance with difficult taxonomic groups of bryophytes: Drs. Roxanne Hastings (Grimmia), John Spence (Bryum), and Richard Zander (Didymodon). Special thanks to Dr. Terry Mcintosh for general assistance in this regard, as well as for his encouragement in the initial stages of this project. Many thanks to Dr. Les Lavkulich, Carol Dyck, and Keren Ferguson, who generously provided assistance, advice, and laboratory equipment for soils analysis. Thanks also to Dr. Mary-Lou Bevier for geological information and assistance. xv I would like to extend thanks to all of my friends and family members who have contributed advice, support, or inspiration in various ways, particularly Arleen Sadler, Derek Thomas, Elizabeth Heinz, Krista Thomas, Amanda Sadler, Paul Sadler, Beverley Reid, Peter Kellington, Kimberley-Anne Trueman, Robin Reid, Martha & Matthew Reid, Gord Singbeil, and Gabriola Island moms and dads (many thanks to Joni Jeffery, Vanessa Craig, and Steve Wilson for practical and technical support). Special thanks to Jane & Emma for their encouragement, project mascot Marley, and, of course, Hamley. Finally, a most special thank you to my husband, Thomas Reid, for his unfailing encouragement and support in all stages and aspects of this project; making himself available to act as field assistant, research assistant, photographer, editor, advisor, counsellor, public-relations liaison, funding scout, project promoter, chauffeur, mechanic, camp cook, and, of course, daddy. I cannot express how grateful I am to have such solid and unwavering support. xvi We shall not cease from exploration And the end of all our exploring will be To arrive at where we started And know the place for the first time. -T.S. Eliot CHAPTER I INTRODUCTION 1.1 Ecological Significance of Vegetation in Rock Outcrop Ecosystems of the CDF Zone The Coastal Douglas-fir (CDF) biogeoclimatic zone is confined to a small fraction of south-western British Columbia (ca 0.3% of the provincial land base). It is represented most extensively on the Gulf Islands and south-east Vancouver Island. Although small in total area, the CDF zone contains some of the rarest ecosystems and associated plant species in BC and Canada. Intense development pressure in the CDF landscape has resulted in the fragmentation and loss of much of the original habitat. Results from the Sensitive Ecosystems Inventory (SEI) of East Vancouver Island and Gulf Islands (initiated in 1993) indicate that less than 8% of the entire CDF zone can be considered relatively unmodified. Even within this 8%, however, many areas are degraded by fragmentation, human use, and introduced species (Ward et al. 1998). With so few of these rare and fragile ecosystems remaining, there is an urgent need to understand the factors that influence species diversity and composition in this region, and the relevant scale of conservation for species and habitats at risk. A critical information gap exists for bryophytes (mosses and liverworts), that are sensitive indicators of habitat change but are underrepresented in provincial and national floristic surveys. Crins (1997) noted that much less research has been done on rare habitats than on their constituent species. The focus on conserving habitats has resulted from the realization that there are too many rare species to deal with effectively, individually. Not all habitats in which rare species occur are themselves rare. Garry oak savannas and associated rock outcrop ecosystems in the CDF zone provide an example of a community type that is itself rare, and contains numerous provincially (and nationally) rare species. In the CDF zone, rock outcrops are locally frequent throughout the landscape. Previous research (Harpel 1997, Vitt & Belland 1997) has shown that rocky substrates support high taxonomic richness, particularly for non-vascular plants. As such, rock outcrop habitats present a natural focal point for ecological study in CDF ecosystems. Furthermore, the current scarcity of research 1 on rock outcrops in the Pacific Northwest necessitates the undertaking of a baseline descriptive study to support subsequent outcrop-related projects in this region. Many current conservation and management strategies are focused on preserving high biodiversity and/or the ecological integrity of rare and endangered ecosystem types. For example, Garry oak and associated ecosystems in BC are of high conservation interest (Fuchs 2001). This approach translates into attempting to maintain populations and proportions of a particular array of constituent native species within an ecosystem unit thought to promote high species richness and/or a number of rare taxa. Conservation efforts in coastal grasslands have involved the removal of encroaching exotic vegetation (MacDougall 2002), and the removal or limitation of other sources of external disturbances such as livestock grazing (McPhee et. al. 2000). There are some problems inherent in this approach. Often insufficient information is available to understand clearly (and try to replicate) all of the processes that operate within an ecosystem unit. Ecosystems are not static; some level of disturbance and fluctuation in constituent populations is natural, e.g. in response to climatic variations, fire, or other natural disturbance such as grazing. These fluctuations may occur with greater complexity, over a longer time period, and across a greater range of geographic scales than is accommodated by the prevailing perspective (i.e. maintaining a checklist of selected species within a landscape unit). A multi-scale approach to researching ecosystems and their constituent species affords the best chance at untangling some of this complexity and, therefore, long-term success in conserving overall ecosystem integrity. The ability to detect changes in pattern and make predictions at more than one scale (population, community, landscape) is a topic of fundamental importance in ecology. Considerations of scale provide a reconciliation of Clements' (1916) continental picture of "climax" plant communities constrained by macroclimate, and Gleason's (1926) view that plant distributions should be interpreted as individualistic responses to spatial gradients in the environment (Turner 1989, Peterson & Parker 1998). Dominant processes change when the focus of analysis changes from one scale to another. Unstable systems may appear stable, bottom-up control turns into top-down control, and competitive interactions become of less importance than climatic effects. Ultimately, patterns observed at a given focal level are a 2 function of constraints and processes operating at higher and lower organizational levels (Bedward 1995). The study of patterns and processes relating to scale is, therefore, vital to the progress of practical and theoretical aspects of ecological science. Studies integrating the effects of scale in species-environment relationships and ecosystem dynamics are vital to advances in fundamental and applied ecology, but such studies have only begun to appear in significant numbers in recent years. Nowhere is the need for multi-scale ecological research more essential than in the challenges facing the conservation of rare and endangered species and ecosystems. Cataloguing the elements of diversity that may be at risk (e.g. rare species or habitats), followed by studies of their ecological requirements, composition, structure, and function over different scales are prerequisite to generate credible conservation actions (Crins 1997). The recently proclaimed federal Species at Risk Act (Environment Canada 2000) further demonstrates the urgent need for a multi-scale research approach to conservation. 1.2 Literature Review of Vegetation Patterns on Rock Outcrops 1.2.1 Climate and Geography of the Study Area In British Columbia, the Coastal Douglas-fir (CDF) biogeoclimatic zone lies in the rainshadow of Vancouver Island and the Olympic mountains, a geographic position that results in a Mediterranean type climate characterized by warm, dry summers and mild, wet winters (Krajina 1965). This climatic region includes the southeastern portion of Vancouver Island, the Gulf Islands located in the Strait of Georgia (north to Savary Island), and along the mainland coast southward to the Fraser River Delta (Harpel 1997) (Figure 1.2.1). Mean annual temperature in the CDF zone ranges from 9.2 to 10.5°C, with the absolute minimum temperature ranging from -21.1 to -11.7°C (Figure 1.2.2). Mean annual precipitation varies from 647 to 1263mm; only approximately 5% of this falls as snow between November and April. Soil frost occurs only where mineral soil is exposed; it is extremely unlikely when the soil surface is protected by a forest floor (decomposing organic matter) or a cover of vegetation (Nuszdorfer et al. 1991). The Mediterranean type climate of the CDF zone is found in other areas along the Pacific Coast of North America; climatically similar areas occur in the Puget Trough and San Juan Islands of Washington state, and in 3 Oregon's Willamette Valley. The CDF zone represents the northern extent of this climatic region; this feature contributes to the occurrence of disjunct taxa and/or species at the northern limit of their distribution range. Figure 1.2.1. Biogeoclimatic units of south-western BC. Dashed line indicates 49°N. [Figure 1.2.2 has been removed owing to copyright restrictions. The information removed shows climatic variation across eastern Vancouver Island, demonstrating lower average monthly precipitation, and milder temperatures in southern sites. This figure was obtained from Ward, P., G. Radcliffe, J. Kirkby, J. Illingworth & C. Cadrin (1998). Sensitive ecosystems inventory: east Vancouver Island and Gulf Islands 1993-1997. Technical Report Series Number 320, Canadian Wildlife Service, Pacific and Yukon Region.]. Figure 1.2.2. Climate variation across eastern Vancouver Island. Graphs indicate average monthly temperature (lines) and average monthly precipitation (bars), 1961-1990 (from Ward et al. 1998). 4 The CDF zone is confined to elevations mostly below 150m. Geologically, the zone is included in the Nanaimo Lowland between the Georgia depression to the east, and the Vancouver Island mountains to the west. The area is underlain largely by sedimentary rocks of Upper Cretaceous age. The Nanaimo Lowland consists of undulating topography and sharp ridge-like crests separated by narrow valleys. Areas of moderate relief are characterized by deep, unconsolidated deposits from recent glaciations. Pudge and trough areas are the result of differential bedrock weathering - ridges are underlain by hard, relatively resistant sandstone and conglomerate rock; valleys are underlain by shales or follow along fault zones (BC Ministry of Environment 1978). Soils in the CDF zone are generally derived from morainal, colluvial, and marine deposits. The accumulation of organic materials in semi- to well-decomposed organic deposits is uncommon (Nuszdorfer et al. 1991). The dominant soil group in the CDF zone is Dystric Brunisol (BC Ministry of Environment 1978). With increasing precipitation, Brunisols grade to Humo-Ferric Podzols. The soils developing under Garry oak typically include a melanized (Ah) horizon and are Melanic Brunisols. Humus development is characterized by Moder to weak Mor formation (Nuszdorfer et al. 1991). Detailed information about soil groups on southern Vancouver Island and in the surrounding area is provided by the BC Ministry of Environment (1985). The unique geography and climate of the CDF zone is characterized by the distinctive vegetation apparent in the zone. Typical vegetation site associations are shown in Figure 1.2.3. Among these, the Douglas-fir - Salal (DS) and Douglas-fir - Shore pine - Arbutus (DPA) site associations are characterized by rock outcrops or pockets of shallow soil interrupting stands of trees with moderate to high frequency. These two associations are characterized by a dry to very dry soil moisture regime, and a very poor to medium soil nutrient regime. Mature stands are dominated by Pseudotsuga menziesii ssp. menziesii; Arbutus menziesii and Quercus garryana are present in DPA associations. In DS associations, the shrub layer is dominated by Gaultheria shallon and Mahonia nervosa; prominent understory plants include Pteridium aquilinum and Rubus ursinus. The shrub layer in DPA associations contains Holodiscus discolor, Mahonia nervosa, and Rosa gymnocarpa; the herb layer is usually diverse, including Bromus vulgaris, Lathyrus nevadensis, Lonicera ciliosa, Melica 5 subulata, and Trientalis latifolia. The mosses Kindbergia oregana and Rhytidiadelphus triquetrus are predominant in both associations (Nuszdorfer et al. 1991). Previous studies have surveyed different groups of plants within the study area, or within comparable biogeoclimatic units in BC. These studies include Ward et al. (1998) for vascular plants, Harpel (1997) for mosses, and Godfrey (1977) for liverworts. [Figure 1.2.3 has been removed owing to copyright restrictions. The information removed shows typical vegetation associations in the CDF zones, grading from hygric (Redcedar -Skunk cabbage, and Redcedar - Grand fir - Foamflower) to mesic (Douglas-fir - Salal) to xeric (Douglas-fir - Shore pine - Arbutus. This figure was obtained from Nuszdorfer, F.C., K. Klinka & D.A. Demarchi (1991). Chapter 5: Coastal Douglas-fir Zone, in Ecosystems of B.C., D. Meidinger & J. Pojar (Eds). B.C. Ministry of Forests, Victoria.]. Figure 1.2.3. Simplified schematic diagram of topographic relationships among four common site associations of the CDF zone (from Nuszdorfer et al. 1991). 1.2.2 Outcrop Morphology and Vegetation Different rock outcrop configurations have been associated with distinct plant communities (Main 1997). The number of taxa present on a rock outcrop will ultimately depend on the ranges of niches available for species. Microtopographical features (i.e. the presence of cracks, crevices, or depressions in the rock) combine with other abiotic factors to create microclimate; these features will have, in consequence, a large impact on the distributions of species, and hence biodiversity, within outcrop communities (Coates & Kirkpatrick 1992). Rock types (e.g. sedimentary vs. volcanic parent material) are characterized by their distinct chemical compositions and relative hardness. These features, in addition to the 6 strength of erosional forces weathering the rock, will determine microtopographical variation. As a result, rocks of different types provide different arrays of microhabitats for vegetation, influencing large-scale vegetation patterns on outcrops in terms of the composition and abundance of vegetation that develops. Although rock outcrops may differ in structural and biotic detail, these ecosystems are typically characterized by shallow, poorly developed soil, full insolation, high summer temperatures with wide day-night temperature differentials, extreme summer drought, and saturated winter soils in unsloping areas (Ware 1990). Consequently, vegetation associated with rock outcrops is comprised of plants with morphological or physiological adaptations that allow them to withstand severe conditions; this includes succulents (e.g. Opuntia), plants with thick or sclerophyllous leaves, plants with Crassulacean Acid Metabolism (e.g. Sedum), C4-plants (e.g. graminoids) to reduce photorespiration in hot dry environments (Ware 1990, Porembski et al. 1997), and cryptogamic taxa (lichens, mosses, and liverworts). In his categorization of plant strategies, Grime (1979) described these plants as "stress-tolerators", which have compromised vegetative and reproductive vigor to endure continuously unproductive environments. Owing to their unique physiology, bryophytes (particularly mosses) often dominate the very shallow soil or bare rock of outcrops. Bryophytes are non-vascular plants generally lacking organized transport systems, lignified tissue, and roots, instead having rhizoids that anchor them to their substrate. Bryophytes must obtain water and nutrients from atmospheric inputs or from the surfaces on which they grow rather than from a subsurface soil volume (Carleton 1990). These ecophysiological attributes make bryophytes poor competitors for resources such as light and space, and more susceptible to soil movement than vascular plants. However, the unique physiological attributes of bryophytes allow them to flourish in habitats that most vascular plants cannot tolerate, depending on transient water and nutrient supply (Nakatsubo 1997). In addition to being relatively independent from their substrate in terms of water uptake, many mosses and liverworts are tolerant of drying out. Photosynthesis declines with water loss, and resumes with greater or lesser delay on re-moistening. The completeness of recovery is dependent on the intensity and duration of desiccation, and on drought-hardening. The time it takes for bryophytes to photosynthesize 7 after rain is determined by storage capacity and rate of water loss. Both of these factors are strongly influenced by growth form (Proctor 1990); boundary layer resistance, influenced by features such as papillae or awns on leaves, is critically important in determining water loss. Outcrop plants that do not possess morphological adaptations to withstand water-loss often display life-history traits that permit them to flourish in favorable times. Common strategies include persisting underground as a tuber during drought, or surviving the extremely hot, dry summer months as dormant populations of drought-resistant seeds (Hopper et al. 1997, Main 1997), although seepage habitats may support year-round populations of some taxa. The overall abundances of different life forms on outcrops have been shown to include a lower proportion of trees and shrubs in relation to the proportion of annual herbs (Wyatt 1997). Seasonal species-turnover is significant in plant communities on shallow soil (Houle & Phillips 1989a). In the Pacific Northwest, a number of the unique vascular plant taxa associated with Garry oak meadows and rock outcrops are only prominent in spring months (i.e. April-June). 1.2.3 Vegetation Assemblage Patterns on Rock Outcrops It was predicted that climatic gradients associated with the CDF zone landscape (i.e. site geographical position) would influence patterns of species richness and composition. Harrison et al. (2000) reported that the diversity of serpentine plants of California (total flora and endemics) declined from north to south, and from the coast inland, in relation to decreasing rainfall. Similarly, Zechmeister et al. (2002) reported that, at the landscape-scale, only precipitation correlated with bryophyte species richness in Austrian agricultural landscapes, whereas at the site and habitat scale, species richness was influenced mainly by land-use intensity and substrate diversity. Belland (2005) found that climatic variables, in particular, temperature of the growing season, were the most important factors in determining moss species distribution patterns in the Gulf of St. Lawrence; edaphic factors, particularly the amount of calcareous rock outcrop, had a secondary influence and modified the patterns established by climate. Pharo et al. (2005) found that climatic variables, particularly those affecting humidity (temperature, precipitation) were important for predicting bryophyte diversity in grasslands of subhumid Tasmania. Similarly, Jonsgard and Birks (1993) found that measured microscale environmental variables accounted for little of the biological 8 variation in bryophyte distributions on rock outcrops; their ordination results suggested that macroclimate was the most important factor. Secondly, it was predicted that rock type would influence patterns of richness and composition within the CDF zone, particularly for bryophytes. Different rock types have contrasting chemistries, water retention, and erosional patterns; Aho and Weaver (2006) found that volcanic rocks had lower pH, absorbed more liquid and atmospheric water, and retained more water over time, than limestone rocks. Pentecost (1980) found that different rock types (pumice-tuff, rhyolite) possess distinct and contrasting lichen and bryophyte floras. Factors affecting the flora included nutrient-enrichment by birds, the degree of exposure, water seepage, and the physico-chemical characters (concentration of solutes in the water contacting rock surfaces). As such, the distribution of bryophytes may be affected by substrate characteristics, but, as Brown (1982) suggests, mosses may also alter the chemistry of the substrate. Bowe and Rayner (1993) studied bryophyte compositions of four 0.4ha outcrop types (gabbro, soapstone, granite, and diabase) in South Carolina, and found no obvious environmental factor to account for relationships between rock type and distribution of species. The authors suggest that site characteristics may partly explain the difference between species diversity of bryophytes on outcrops. Although soapstone outcrops (the softest rock type) were associated with greater species richness, soapstone was found only in mesic woods. The influence of site characteristics versus rock geology on bryophytes, therefore, could not be reliably distinguished in this case. Thirdly, it was predicted that historical and current disturbance levels within the CDF zone landscape would influence vegetation composition and richness patterns. Pueyo et al. (2006) found that historical elements of the landscape had a significant effect on current natural vegetation in tall arid brush and tall grass steppe of Spain. Similarly, Weiher (2003) found that disturbance history had the strongest effect on species richness in oak savannas of western Wisconsin, USA. The Gulf Islands within the CDF zone have been subject to contrasting human use histories. Human population densities are greater toward the south end of the zone, and concentrated on the larger islands. The longer history of human habitation and land use in southern areas results in a much more complex disturbance history (e.g. clearing, grazing, fire) than in northern sites. As such, historical disturbance within the 9 study area is linked with geographical position, whereas current grazing levels vary throughout the CDF zone. Disturbance by grazing (i.e. by deer or other large mammals) is known to affect soil compaction and modification, in addition to influencing constituent vegetation in Garry oak and associated grassland environments (e.g. Saenz & Sawyer 1986, Hatch et al. 1999, Kotanen 2004). Grazing by deer or livestock acts to remove or suppress the encroachment of woody plants, and creates opportunities for the establishment of colonizing taxa. As noted by Hierro et al. (2006), disturbance is one of the most important factors promoting exotic invasion. As such, within the CDF zone landscape, it was predicted that the coverage and richness of exotic species would be higher in areas with longer histories of human usage (correlating roughly with latitude, i.e. southern sites), and in sites which are currently experiencing more intense grazing pressure. In contrast, it was predicted that the coverage of native vegetation would be higher in areas with shorter human-use histories (i.e. northern sites), and in sites which are currently ungrazed. Although the particular array of taxa found on rock outcrops may vary depending on geographic position, geology, and patterns of current and historical disturbance, most studies have identified generally similar types of vegetation assemblages: these include cryptogams (bryophytes and lichens), lichen-annual, annual-perennial herb, herb-shrub, and herb-shrub-tree communities (Burbanck & Piatt 1964, Houle 1990, Porembski et al. 1997). As previously discussed, bryophytes are among the best adapted plants in habitats with extreme conditions, and consequently they often comprise the dominant assemblage type in rocky outcrop habitats. Since bryophytes are small, they are largely dependent on micro-scale environmental conditions. Microscale environmental factors observed to influence community patterns include aspect, slope, crevice depth, and frequency of flushing (Jonsgard & Birks 1993). These microtopographical features influence the capacity of a depression to accumulate organic matter and, therefore, ultimately determine vegetation composition. Collins et al. (1989) sampled in 17 soil-filled depressions ("islands") on a granite outcrop in southern Oklahoma and found that the number of species per island was significantly positively related to island size and maximum soil depth. Detrended Correspondence Analysis (DCA) produced community gradients associated with island size, soil texture, soil pH, soil depth, organic 10 matter, and distance to edge of the outcrop. Species composition differences among islands reflected variation in habitat quality (i.e. degree of soil development) as well as stochastic variation associated with seed dispersal and establishment. Similarly, Cox and Larson (1993) suggested that a complex soil gradient controls small-scale vegetation structure and composition on cliff-base talus in southern Ontario. Alpert (1985) related microtopography to microdistribution of bryophytes on granitic rocks in the inland chaparral of California (USA). The dominant species were found to grow on north, east, and west facing surfaces with slopes less than 60 degrees. Other bryophytes tended to grow on steep, concave, north and west facing surfaces - occurrence of these species was strongly associated with shade, as estimated from microtopography. Similarly, Ott et al (1996) found relative humidity to be important to cryptogam distributions in the rock-alvar of Gotland, Sweden (i.e. the northern-most distribution of chalk grassland plant communities which have their center of distribution in the Mediterranean region). Hedderson and Brassard (1990) evaluated the microhabitat relationships (15cm x 15xm quadrats) of populations of five mosses occurring on cliffs and scree slopes in eastern Newfoundland using discriminant function analysis (DFA) and multiple regression analysis (MRA). The first three discriminant functions accounted for 91.3% of the among-species variation in microhabitat characteristics, and were interpreted as representing species separation along water deficit, pH, and temperature gradients. The combined DFA and M R A results showed that the five moss species studied occupied distinct microhabitats within the cliff/scree-slope study area. In summary, it was predicted that prominent landscape features within rock outcrop ecosystems of the CDF zone, i.e. geographic position (latitude), geology (rock type), and grazing intensity, would be related to large-scale trends in vegetation abundance (i.e. percent coverage), plant species richness, and rarity (by native or exotic origin, and by life form group). Secondly, it was predicted that these landscape features would combine with local factors (e.g. slope, aspect, soil depth) to provide unique microhabitats relating to the fine-scale patterns of vegetation variation. 11 1.3 Research Objectives In this research, rock outcrop ecosystems in the Gulf Islands of the CDF zone were examined at multiple scales to address the following objectives: 1. To inventory vascular plants and bryophytes in rock outcrop ecosystems, refining distribution information for constituent species. It was predicted that several rare and/or previously undocumented taxa would be encountered, particularly bryophytes. A sub-component of this objective was the compilation of keys for the identification of taxonomically difficult groups of bryophytes occurring in the study area, i.e. Brachytheciaceae, Bryum, Didymodon, Grimmia, Schistidium, and Racomitrium. 2. To investigate how large-scale (i.e. landscape) features influence patterns of coverage and richness for plant species by origin, life form group, and rarity status, including: (a) Geographic position (i.e. latitude). It was predicted that climatic effects associated with rainfall and temperature gradients within the CDF zone would influence patterns of vegetation coverage and richness. Specifically, it was predicted that: (i) exotic species would have higher coverage and richness in the southern portion of the study area, which is more populated and has a longer history of human use (ii) rare and uncommon species would have higher richness in the southern portion of the study area, as certain taxa approach the northern limit of their range (particularly bryophytes, which are known to be more sensitive than vascular plants to subtle shifts in environmental conditions). (b) Geology (i.e. rock type). It was predicted that different rock types would influence patterns of vegetation coverage and richness. (c) Grazing intensity. It was predicted that grazing intensity would influence patterns of vegetation coverage and richness. Specifically, it was predicted that ungrazed sites would have higher coverage of native plant species, and that grazed sites would have higher coverage and richness of exotic plant species. 3. To use vegetation patterns observed at each sampling scale to derive a classification scheme for rock outcrop vegetation. It was predicted that environmental variables measured at each scale (site, plot, and microplot) would combine to produce unique 12 vegetation assemblages, with compositions that correlate with different fine-scale gradients. 4. To use integrated multi-scale habitat-vegetation data to interpret rock outcrop ecosystem dynamics, and address implications for conservation and management. Specifically, it was predicted that: (i) there would be higher richness of native species not exclusive to rock outcrops in plots closer to forest edges; suggesting that overall rock outcrop species richness is at least partially owing to the immigration of taxa from adjacent habitat types (ii) a higher number of rock outcrop-exclusive native taxa would occur on larger islands, and/or within larger units of rock outcrop ecosystems when sampling effort remained constant (iii) mechanisms acting to maintain rock outcrop ecosystems in a state of permanent primary succession would be unique to particular sites A conceptual diagram that illustrates the multi-scale approach used for studying rock outcrop ecosystems in this thesis is presented in Figure 1.3.1. Vegetation patterns were first examined at the large-scale to determine the individual effects of landscape features on coverage, richness and rarity trends. Subsequently, a multi-scale integration of landscape, local, and fine-scale features was used to link large-scale trends with fine-scale patterns of coverage, richness, and rarity. 13 Landscape Scale (Site) Fine Scale (Microplot) Exotic Native Rare Figure 1.3.1. Conceptual diagram for this research: (a) the distribution patterns of native and exotic plant species (rare and common, overall and by life form group) were investigated in relation to landscape features (latitude, geology, grazing), and (b) the combined effects of landscape, local, and fine-scale environmental gradients were integrated to characterize the fine-scale patterns of constituent species groups. 14 CHAPTER II METHODS 2.1 Sampling Strategy Data collection was completed over three field seasons (April-June 2000, 2001, 2002). Rock outcrop habitat was inventoried at eighteen sites located in the Gulf Islands of the CDF biogeoclimatic zone (Figure 2.1.1, Table 2.1.1). For the purpose of this study, rock outcrop habitats were defined as distinct, continuous topographical features greater than lha, characterized by bare parent material and/or shallow organic accumulations over parent material (little or no mineral soil), which had no directly overhanging canopy coverage (woody species >2m in height), and that was above the high-tide line. Figure 2.1.1. Study site locations. Sites where vascular plants and bryophytes were inventoried in sample plots are designated by black stars; sites where only bryophytes were inventoried are designated by white stars. 15 Table 2.1.1. Summary of site information (ER=Ecological Reserve). Sampling strategy used noted as P (surveyed using plots), and I (unstructured bryophyte inventory only). Site Code Site Location Island Latitude (N) Longitude (W) Notes BWN Cape Roger Curtis Bowen 49° 20' 07" 123° 24' 15" P CHR Channel Ridge Saltspring 48° 53' 40" 123° 33' 50" P DIS Discovery Island Prov. Pk. Discovery 48° 25' 22" 123° 14' 17" P G A L Bluffs Park & Mt. Galiano Galiano 48° 51'45" 123° 21' 35" P H B Y Helliwell Prov. Pk. Hornby 49° 31' 00" 124° 35'08" P LAS BC Parks ER - Jenkin's Cove Lasqueti 49° 27' 34" 124° 17'49" P M A X BC Parks ER - Mt. Maxwell Saltspring 48° 48'33" 123° 32' 10" P REG Reginald Hill Saltspring 48° 45' 37" 123° 25' 45" P SAT Mt. Warburton Pike Saturna 48° 46' 20" 123° 10' 14" P SID Wymond Point Sidney 48° 35'28" 123° 16' 22" P T E X Mouat Point Texada 49° 37' 11" 124° 25'52" P V A L N E coast of Valdes Island Valdes 49° 06' 30" 123° 40' 20" P WIN South Winchelsea Island Winchelsea 49° 17'28" 124° 04'41" P JED N E of Home Bay Jedediah 49° 30' 05" 124° 11' 40" I JER SW Jervis Island Jervis 49° 30' 16" 124° 13' 07" I PEN George Hill Pender 48° 48' 30" 123° 17'2I" I S A V Mace Point Savary 49° 57' 00" 124° 45'46" I T U A Mt. Tuam Saltspring 48° 43' 22" 123° 29' 10" I Sensitive Ecosystems Inventory (SEI) ecosystem maps (Ward et al. 1998) were used to locate rock outcrop sites. SEI habitat types considered for this study included Coastal Bluff (CB), Terrestrial Herbaceous (HT), and Open Woodland (WD) ecosystem units (represented as polygons). Sites outside the range of the SEI mapping territory were located using aerial photos. Vegetation types sampled ranged from bare rock/cryptogam-dominated communities to shallow grasslands intermittent in woodland habitats (e.g. mixed stands of Quercus garryana, Arbutus menziesii, and/or Pseudotsuga menziesii). Both volcanic (meta-igneous) and sedimentary (sandstone, conglomerate) rocks were sampled in relation to their occurrence within the study area. Grazing pressures varied among sites, with categories identified as (1) ungrazed (no current grazing by native deer, fallow deer, sheep, or goats), (2) moderately grazed (grazing by native deer populations only), and (3) intensely grazed (grazing by native deer populations, plus the presence of introduced feral deer, sheep, and/or goats). Thirty-nine 100m2 plots (10m x 10m) were sampled across a total of 13 study sites (three randomly placed plots per site) (Figure 2.1.2); an additional five sites were inventoried 16 for bryophytes, but without the use of plots. For each 10m x 10m sample plot, aspect and slope were noted. Plots were established to fall roughly in the same direction as the prevailing slope and aspect within the overall landscape (i.e. the site). Slope and aspect measured along the fall line of the plot were recorded as "vertical slope" and "vertical aspect", whereas slope and aspect measured perpendicular to the fall line of the plot were recorded as "horizontal slope" and "horizontal aspect". These measures were later combined to produce an estimate of "total slope" and "total aspect" (i.e. degrees from north). Land Plot I Plot 2 7 ^ V 7 ^ Plots randomly selected per site 10m 10m A! o c • • ] • • c , El • • T V ] • ft r Slope lm x lm microplot • 25cm x 25cm microplot Figure 2.1.2. Plot selection and layout. For each study site, a grid was placed over the rock outcrop polygon (shown in yellow), and three plot locations were randomly selected (Plot 1, Plot 2, Plot 3 ) . Each 10m x 10m study plot contained eight nested lm x lm microplots and 25 25cm x 25cm microplots (systematically placed, as shown at right). A sample of rock was collected from each plot for geological analysis. Four soil pits were excavated along a transect across the center line of each plot (corresponding with the 25cm x 25cm microplots on either side of the central microplot) to collect samples of the organic layer. Also recorded were total soil depth to parent material, and thickness of the organic (L) and humus (FH) layers. Each sample was tested for pH using a 1:5 ratio of 17 soil:water; organic matter (total carbon), and total nitrogen content were determined using a LECO® soil analyzer (UBC Department of Soil Science). The types of non-outcrop habitat associated with the study site were documented (e.g. Douglas-fir forest, Garry oak meadow, ocean), and the distance of each sample plot to contextual habitat types (i.e. the high-tide line, or the nearest tree >2m tall) was recorded. Two sizes of microplots nested within the 100m2 sample plots were used to describe vegetation patterns. Microplot sizes from 0.01 to 0.1 m 2 have been considered appropriate for studying bryophytes and small vascular plants (Daubenmire 1968), while larger plant communities may require considerably larger microplot sizes. Stohlgren et al. (1998) has shown that a multi-scale approach to sampling can enhance the detection and measurement of constituent species. As such, eight l m x lm microplots and twenty-five 25cm x 25cm microplots were systematically placed within each 100m plot to obtain a relevant scale of study for vascular plants, and bryophytes, respectively. For each microplot (both lm x l m and 25cm x 25cm sizes), the following variables were recorded: slope and aspect (horizontal and vertical estimates, corresponding with plot orientation), canopy coverage (as estimated from a convex spherical densiometer in north, east, south and west directions), coverage of plants by life form group (woody plants, graminoids, herbs, bryophytes, and lichen) and by individual species (for vascular plants and bryophytes), coverage of substratum-types (bare rock, bare soil and/or decayed organic matter and wood), and soil depths (as measured by 9 equi-distant points within each microplot). Plots were searched for additional vascular plant and bryophyte species that were not found in microplots. Areas surrounding plots (i.e. within approximately 1 lm of plot edges) were searched for additional plant species as well. This sampling approach provided a 2 2 uniform area to be inventoried in relation to each plot (1000m ) and site overall (3000m ). Voucher specimens were collected for all species encountered; selected samples were deposited at the UBC herbarium. Owing to the difficulty involved with identifying certain taxonomic groups of bryophytes, specialized keys for rock outcrop habitats in the study area were developed from existing keys and herbarium material for the following moss genera/families: Bryum, Brachytheciaceae, Didymodon, Grimmia & Schistidium, and Racomitrium (Appendix la-e). 18 In order to assess the status of rock outcrop ecosystems as hotpsots of richness and rarity within the Coastal Douglas-fir zone, results from Sadler's (2004) study were used to compare the vegetation of rock outcrops with that of CDF zone forests. Field work for that project took place during May-August 2003 in Weyerhaeuser Ltd.'s Northwest-Bay Operations area (Nanoose) in CDF and CDF/CWHxm transitional forest. Four age classes of regenerating and mature forests were investigated: 2-5 years, 20-30 years, 50-60 years, and >90 years. Each of the four age classes was represented by three study sites. Each study site was surveyed with three 15m x 22.2m (333m2) sample plots. In total, 12 terrestrial microplots were examined within each sample plot: six lm x lm microplots, and six 0.1m x 0.3m microplots. In addition, six tree bases and six logs (>10cm in diameter, nearest to each terrestrial microplot) were sampled using 0.1m x 0.3m microplots. Stumps were sampled in lieu of tree bases in newly harvested (i.e. 2-5 year-old) sites. Plots were surveyed for additional species not occurring in microplots. Overall, total areas of 1000m2 were sampled per study site, and 3000m2 were sampled per age class. 2.2 Data Analyses Following laboratory identification of field specimens, all plot and microplot data were entered onto an Excel spreadsheet for statistical analysis using SYSTAT (version 11). Species were organized by life form group (trees and shrubs were termed "woody plants", herbaceous species were termed "herbs", grasses, rushes, and sedges were termed "graminoids", and moss and liverwort species were termed "bryophytes". Each species was assigned an origin status ("native" or "exotic", i.e. non-native), and a rarity ranking based on provincial tracking lists (BC Conservation Data Centre 2006). The British Columbia Conservation Data Centre (BC CDC) records ecological and demographic information for all plant species in BC. Each native plant species and community type has been assigned a global (G) and provincial (S for "subnational") rarity rank according to an objective set of criteria established by the Nature Conservancy (USA), and a status on the provincial Red or Blue lists. Red listed species include any indigenous taxa considered to be extirpated, endangered, or threatened in BC. These species are segregated into two groups, SI and S2. SI species have 5 or fewer extant occurrences, and 19 are considered critically imperiled, whereas S2 species have 6-20 occurrences, and are considered imperiled. Blue-listed (S3) species include vulnerable indigenous taxa, i.e. species that could become candidates for the red list in the foreseeable future. The Blue list also features plants that are suspected of being vulnerable, but for which information is currently lacking (BC CDC 1996). Species with uncertain rankings (denoted "?") were assigned the next-lowest rarity ranking (e.g. "S3?" was treated as "S4" for the analysis), except in the case of Grimmia ovalis and Grimmia alpestris, where tentative S3 rankings were retained based on the advice of R.I. Hastings (2006, pers. comm.). Combination rankings were designated with the most "common" component of the ranking (e.g. "S3S4" was treated as "S4" for the analysis). Incomplete or uncertain identifications were not included in any analysis comparing rarity rankings, although they were included in analyses comparing overall vegetation richness, or richness of life form groups. Coverage estimates for substratum-coverage, species, and life form groups were recorded individually and, owing to overlap, could add up to greater than 100% coverage. Overall, site-level and plot-level coverages reported indicate averages as obtained from constituent microplots (vascular plant data averaged from lm x lm microplots, bryophyte data averaged from 25cm x 25cm microplots, and environmental data averaged from all microplots). Richness estimates for plots reflect the inclusion of all species found within the 10m x 10m sample unit (i.e. species identified within microplots, as well as any additional species recorded outside of microplots). Richness estimates for sites reflect the inclusion of all species recorded within plots, as well as any additional species recorded within 2 2 approximately 1 lm of plot edges (i.e. 1000m sample unit per plot; approximately 3000m sampled for each study site). Linear regression analysis was performed on untransformed data to investigate trends of plant coverage and richness (by origin status and rarity ranking) in relation to site geographic position (i.e. latitude). One site (Mt. Warburton Pike on Saturna Island) was excluded from this analysis to minimize the confounding effects of geology and grazing intensity; the remaining 12 sites had an even distribution of rock type (meta-igneous, sandstone, and conglomerate rock) and grazing intensity (none, moderate and high categories, as discussed) across the latitudinal gradient (i.e. north or south of 49°N). Weak 20 linear relationships (i.e. R2<0.1) are not discussed, and /^-values are reported only where significant (p<0.05). A l l species were included in this analysis, as described in the preceding paragraphs. Two-tailed Student's t-tests were used to compare the means of coverage and richness of species life form groups (by origin and rarity ranking) in relation to rock type. To minimize the confounding effects of contrasting levels of grazing intensity between rock types, only moderately-grazed sites were used in this analysis. Selected sites had an even representation of rock types (meta-igneous, sandstone, and conglomerate rock) across the latitudinal gradient (i.e. north or south of 49°N). The /^-values reported from t-tests refer to results obtained using separate (i.e. not pooled) sample variances. Rock types were compared by origin (i.e. meta-igneous rock vs. sedimentary rock, with conglomerate rock and sandstone rock sites pooled together). In addition, sedimentary rock types (i.e. conglomerate rock and sandstone rock) were compared separately. Only significant differences between groups (p<0.05) were reported. A l l species were included in this analysis, as previously described. Two-tailed Student's t-tests were used to compare the means of coverage and richness of species life form groups (by origin and rarity ranking) in relation to grazing intensity. To minimize the confounding effects of contrasting levels of grazing intensity between rock types (as well as insufficient representation of different grazing intensity categories on sedimentary rock), only meta-igneous rock sites were used in this analysis. Selected sites had an even representation of grazing intensity categories (none, moderate, and high, as described in Section 2.1) in relation to latitude (i.e. north or south of 49°N). The ^ -values reported from t-tests refer to results obtained using separate (i.e. not pooled) sample variances. Grazing categories were compared separately (low vs. moderate, low vs. high, and moderate vs. high). Only significant differences between groups (/?<0.05) were reported. A l l species were included in this analysis, as previously described. Vegetation within rock outcrop ecosystems was classified in three steps, using principal components analysis (PCA) to hierarchically categorize vegetation, based on the environmental variables that correlated most strongly with trends in vegetation patterning at each scale (i.e. the first PCA axes, as generated from species coverage data). Previous studies have used a similar hierarchical approach in describing vegetation assemblages, or in 21 identifying primary environmental gradients influencing vegetation patterning using ordination analysis (e.g. MacDougall et al. 2006, Gagnon & Bradfield 1987, Poore 1962). Site and plot-level PC As used vascular plant species cover estimates from l m x lm microplots, and bryophyte species estimates from 25cm x 25cm microplots. Microplot-level PC As used vascular plant and bryophyte estimates from lm x lm microplots; since the richness of microplots was directly compared, it was important to keep sample-area size constant. As recommended by Mead (1988, in Kenkel 2006), all biotic and physical variables were analyzed on a logarithmic scale to aid in meeting the assumptions of PCA (i.e. homogeneity, normality, and additivity). A l l microplot variables (excluding pH, which is already measured on a logarithmic scale) were log-transformed using a procedure recommended by McCune and Grace (2002), whereby the original order of magnitudes in the data are preserved, and whereby "zero" values result for records where the initial value was zero, i.e.: Given Min(x) is the smallest nonzero value in the data Int(;c) is a function that truncates x to an integer by dropping digits after the decimal point c = order of magnitude constant = Int(log(Min(x)) d- decimal constant = log"' (c), then the transformation is: bjj = log(x,y +d) -c, where Xy = the original value in row / and column j of the data matrix, and bjj = the adjusted value that replaces xu: For each PCA, axes were characterized and interpreted using Pearson correlation coefficients showing the correlative strengths of species, environmental variables, and measurements of coverage and richness by life form group (in terms of origin and rarity ranking). Site and plot-level PCAs were interpreted using environmental estimates from all microplots, whereas microplot-level PCAs were interpreted using environmental estimates from lm x lm microplots. For microplot PCAs, plot-level averages (as well as microplot estimates) were also correlated with axes to interpret the relative importance of large and fine-scale environmental features on fine-scale vegetation patterning. The first step in vegetation classification involved identifying the primary environmental gradient associated with vegetation patterns at the site level (i.e. in relation to the first PCA axis). For the site-level PCA, species that occurred in fewer than 3 sites were 22 excluded from the analysis. The second step in classifying rock outcrop vegetation involved identifying the primary gradient associated with vegetation patterns at the plot level (i.e. for each of the two categories determined from the site-level dichotomy, the gradients associated with the first PCA axis in each case). Plot-level averages were used for the second step to allow (a) local-scale (i.e. plot-level) as well as landscape-scale (i.e. site-level) features to be interpreted in relation to the primary gradients, and (b) to increase the sample size on which the analysis was based (a reduction in sample size for the level being studied was associated with each dichotomy). For each plot-level PCA, species with less than 5% frequency in plots and microplots were excluded. The third step in classifying rock outcrop vegetation involved characterizing the two most important environmental gradients associated with vegetation patterning at the fine scale (i.e. the first two PCA axes for each of the landscape categories identified using site-and plot-level PCAs); species with less than 5% frequency in lm x lm microplots were excluded. This hierarchical method of distinguishing vegetation assemblages was supported by the PCA results; in each case the total percent of variance explained by the primary axis was greater than the variance explained by the second PCA axis of the preceding analysis (i.e. from site- to plot- to microplot-level), even though sample size increased toward finer scales. As such, the first two PCA axes were used to interpret fine-scale vegetation patterning. To illustrate the relationship of categorical environmental variables with results generated using species cover, data points were grouped using 68.3% confidence ellipses. Species were grouped into assemblage types based on the strength of their correlations with the first two PCA axes from the microplot species coverage data. To broaden the comparison of rock outcrops to "contextual" CDF forests, results were compared to those in Sadler (2004). Average species richness in rock outcrop sites (3000m2 per site) was compared with total species richness in four age classes (5-year old, 25-year old, 55-year old, and 90+ year old) in CDF forests. Overall, a total area of 3000m was surveyed per age class. The species richness of native vegetation was compared between habitat types (rock outcrop vs. forest) overall, by life form group, and by rarity ranking. Species composition was compared by designating each native species occurring in rock outcrop ecosystems as either "shared" with forested ecosystems (i.e. recorded in any age class of regenerating or mature forest), or "exclusive" to rock ecosystems (i.e. not recorded in 23 any age class of regenerating or mature forest). Using linear regression analysis, the "shared" species richness of each plant life form group was further examined in relation to distance of rock outcrop plots from forest edges. The "exclusive" species richness of each plant life form group in sites was investigated in relation to rock outcrop polygon size, and island size. A l l values were log-transformed, as previously described. P-values are reported only where significant (p<0.05). 24 CHAPTER III RESULTS AND DISCUSSION 3.1 Overview of Rock Outcrop Ecosystems of the Gulf Islands in the CDF Zone, BC A total of 311 plant species were identified from the 18 inventoried sites; 204 vascular plants (26 woody plants, 39 graminoids, 139 herbs), and 107 bryophytes. The average percent cover and frequency of each species is summarized in Appendix 2, with origin and provincial conservation status and ranking information (definitions provided in Appendix 3). Rock outcrop ecosystems were dominated by bryophytes, graminoids, bare rock, and herbs (Figure 3.1.1a), with herbs and bryophytes comprising the majority of species (Figure 3.1.1b). Data collected at individual sites are summarized in Appendix 4. (a) (b) Figure 3.1.1a,b. Summary of (a) vegetation and substratum coverage, and (b) plant species richness for plot-sampled sites, by life form groups and substratum-types: Woody=woody plants, Gram=graminoids, Herb=herbs, Bryo=bryophytes, Lchn=lichens, Bare PM=bare parent material, Org/Litter=organic matter (litter and soil). Rock outcrop habitats were predominantly south-facing, with steep slopes, often directly intercepting prevailing winds, and typically occurred on bluffs overlooking the ocean (Figure 3.1.2a,b). Stands of Pseudotsuga menziesii, Quercus garryana, and/or Arbutus menziesii were scattered within the open landscape dominated by shallow grasslands and patches of bare parent material (i.e. exposed underlying bedrock). Owing to their position, these landscapes were highly exposed, with constituent plants having little or no protection 25 from solar radiation, or from wind and rain. Rock outcrop ecosystems in the CDF zone are associated with a unique array of plant species, many of which are rare and/or approaching the northern extent of their range. The distribution patterns of such species are discussed in detail in Section 3.3. In relation to the climatic gradients extending across the study area, it was predicted that latitude would influence the array of species (particularly rarities) encountered. Trends relating to geographical position are discussed in Section 3.3.1. (b) Figure 3.1.2a,b. Typical rock outcrop ecosystem landscapes: (a) site view at Lasqueti Island, and (b) distant view of Saturna Island site (photo by Gary Lewis). The site-level appearance of rock outcrop landscapes was found to be highly influenced by parent material and grazing intensity (Table 3.1.1). Three different types of parent material were encountered at study sites: meta-igneous rock, sandstone rock, and conglomerate rock. Meta-igneous rock is volcanic in origin, whereas sandstone and conglomerate rocks are sedimentary. Meta-igneous rock was the most frequent rock type 26 encountered in this study. Parent material type was associated with obvious differences in topography and vegetation patterning both at the landscape level (i.e. >100m2) and at finer scales (i.e. <100m2) (Figures 3.1.3a,b, 3.1.4a,b, and 3.1.5a,b). Based on visual observation in the field, meta-igneous rock had greater topographical variation at the landscape level, whereas conglomerate rock had greater variation at the fine scale. Sandstone had little topographical variation at either scale. Table 3.1.1. Summary of site parent material (Rock Type) and grazing intensity (Grazing). Site Code Location Island Rock Type Crazing BWN Cape Roger Curtis Bowen meta-igneous moderate CHR Channel Ridge Saltspring sandstone moderate DIS Discovery Island Prov. Pk. Discovery meta-igneous none G A L Bluffs Park & Mt. Galiano Galiano conglomerate moderate H B Y Helliwell Prov. Pk. Hornby conglomerate moderate L A S BC Parks ER - Jenkin's Cove Lasqueti meta-igneous high M A X BC Parks ER - Mt. Maxwell Saltspring meta-igneous moderate REG Reginald Hill Saltspring meta-igneous moderate SAT Mt. Warburton Pike Saturna sandstone high SID Wymond Point Sidney meta-igneous high T E X Mouat Point Texada meta-igneous moderate V A L N E coast of Valdes Island Valdes sandstone moderate WIN South Winchelsea Island Winchelsea meta-igneous none T U A Mt. Tuam Saltspring meta-igneous high PEN George Hill Pender conglomerate moderate JED N E of Home Bay Jedediah meta-igneous high JER SW Jervis Island Jervis meta-igneous moderate S A V Mace Point Savary meta-igneous moderate (a) (b) Figure 3.1.3a,b. Meta-igneous rock outcrop: (a) landscape view, and (b) fine-scale view. 27 Contrasting topographies result from differing combinations of physical and chemical properties of the parent material (i.e. erosional patterns). In CDF zone rock outcrop ecosystems, geological properties were observed to influence the quality and variety of habitat types within a site. For example, shallow soils on meta-igneous rock and sandstone consisted of varying levels of organic mats that were loosely tethered to the parent material (Figure 3.1.6a,b). In contrast, shallow soils on conglomerate rock were more gravelly, and organic mats were not as easily sloughed off (Figure 3.1.6c). Owing to the presence of crevices and deep pockets in meta-igneous rocks, there were often localized patches of relatively deep soil within meta-igneous sites (Figure 3.1.7a). Deeper soils over sandstone 28 and conglomerate rocks consisted of thicker organic mats. The low porosity of meta-igneous rock was associated with a higher frequency of seepage habitat, as well as ephemeral ponds (Figure 3.1.7b). Seepage habitats were less frequent in sandstone and particularly conglomerate rock sites owing to higher rock porosity, and the lack of large-scale topographic variation (crevices and pockets where water might accumulate were absent). The influence of site parent material on vegetation-habitat relationships is discussed in more detail in Section 3.3.2. (a) (b) (c) Figure 3.1.6a-c. Shallow organic mats over (a) meta-igneous rock, (b) sandstone, and (c) conglomerate rock parent material. (a) (b) Figure 3.1.7a,b. Microhabitats characteristic to meta-igneous rock landscapes: (a) deep pockets of soil, and (b) ephemeral ponds. Grazing intensity was also observed to impact the site-level appearance of rock outcrop ecosystems. Typically, there was a band of low-shrub vegetation extending out from surrounding canopies in ungrazed sites (Figure 3.1.8a,b). In contrast, sites with moderate 29 (Figure 3.1.9a,b) or high (Figure 3.1.10a,b) grazing intensity had reduced encroachment by shrubs, and rock outcrop edge boundaries were more distinct. No ungrazed sites on sedimentary rock were encountered in this study; all sedimentary sites with relatively lower grazing intensity were completely overgrown with encroaching broom (Cytisus scoparius). The influence of grazing intensity on vegetation-habitat relationships is discussed in more detail in Section 3.3.3. (a) (b) Figure 3.1.8a,b. (a) Ungrazed meta-igneous rock site (Winchelsea Island), and (b) sandstone rock site with relatively low grazing intensity (Drumbeg Park, Gabriola Island) -not sampled. (a) (b) Figure 3.1.9a,b. (a) Moderately grazed meta-igneous rock site (Texada Island), and (b) moderately grazed sandstone rock site (Channel Ridge, Saltspring Island). 30 (a) (b) Figure 3.1.10a,b. (a) Intensely grazed meta-igneous rock site (Sidney Island), and (b) intensely grazed sandstone rock site (Saturna Island). 3.2 Rare Plants of Rock Outcrop Ecosystems of the Gulf Islands in the CDF Zone, BC The BC conservation status rankings for all plants identified in this study are summarized in Table 3.2.1. For each life form group, most species were native, although a relatively high proportion of vascular plant taxa were of exotic origin (27% overall), as compared to bryophytes (<1%) (Figure 3.2.1a-d). The lack of sufficient demographic and ecological data for many bryophytes (i.e. all of the liverworts) has prohibited the determination of their conservation status; as such, there was a higher proportion of unrankable bryophyte species (of the 13 shown, 10 are liverworts). Several provincially rare taxa were identified; their documented and current distributions and habitat characteristics are summarized in Table 3.2.2. The vascular plant species Trifolium gracilentum will be a new addition to the flora of BC, pending confirmation of its identity. The majority (75%) of the 24 SI-S3 species identified were mosses, indicating that bryophytes comprise a significant portion of the rare species supported by rock outcrop ecosystems of the CDF zone. However, the information available for these taxa was largely lacking compared to vascular plants in the area, owing to confusion surrounding misidentification, the prevalence of synonyms, as well as absent or incomplete distribution information. As such, the nomenclature and current distributional information for bryophyte species designated as Red-listed (SI, S2) or Blue-listed (S3) is discussed in more detail below. 31 Table 3.2.1. Summary of species richness within life form groups, listed by categories plant conservation status (see Appendix 3 for definitions of conservation categories). Category Woody Graminoid Herb Bryophyte T O T A L Exotic 3 15 37 1 56 Native 22 24 101 106 253 SI 0 0 0 2 2 SI? 0 0 1 2 3 S1S3 0 0 0 1 1 S2 0 0 0 3 3 S2? 0 0 0 1 1 S2S3 0 0 1 7 8 S3 0 0 2 0 2 S3? 0 0 0 4 4 S3S4 1 2 19 20 42 S3S5 0 0 0 32 32 S4 5 9 34 17 65 S4S5 1 0 0 4 5 S5 15 13 42 0 70 Unrankable 1 0 3 13 17 T O T A L 26 39 139 107 311 Unrankable 4% Exotic 12% S5 61% (a) Woody Plants S1-S3 0% S4 23% (b) Graminoids Unrankable 12% Exotic 1% SI-S3 34% (c) Herbs (d) Bryophytes Figure 3.2.1a-d. Pie graphs showing the proportions of exotic, native, and unrankable taxa, by life form group. Native taxa are depicted by provincial rarity status: rare (SI-S3), uncommon but not rare (S4), secure (S5), and unrankable plants identified within the study area. 33 Table 3.2.2. Vulnerable (S3-ranked), imperiled or critically-imperiled (SI - or S2-ranked), and unrankable (U), but possibly at-risk plant species recorded at study sites, by life form group (H=herb, M=moss, L=liverwort). Documented range and habitat information summarized from: Hitchcock and Cronquist (1973) and BC CDC (2006) for vascular plants; Lawton (1971), BC CDC (2006), Zander (2006), Spence (1988), Nyholm (1998), and Hastings and Greven (2006) for mosses, and Godfrey (1977) for liverworts. Type Rank Species Documented Range in P N W Documented Habitat Current Observations Sites H S3 Isoetes nuttallii mostly west of the Cascades terrestrial, generally on wet ground at low elevations in shallow ephemeral pool over sandstone parent material V A L H S3 Sagina decumbens ssp. occidentalis west of the Cascades and Columbia River Gorge low elevations where moist conglomerate and meta-igneous parent material; 2/3 sites ungrazed DIS, H B Y , WIN H S2S3 Trifolium dichotomum Vancouver Island, western W A , southern O R to CA dry rocky or sandy slopes and fields, coastal prairie, mixed evergreen forest below 3500 feet over sandstone parent material; site intensely grazed SAT H SI? Trifolium Igracilentum west of the Cascades; occasional in W A , common south to Baja C A grassy knolls conglomerate and meta-igneous rock; low elevation, moderate to intense grazing H B Y , SID M S2S3 Bryum muehlenbeckii all provinces and states of the Pacific Northwest on wet soil or rock, predominantly coastal lowland, with a preference for Mediterranean climates, occasionally reaching subalpine sedimentary and meta-igneous parent material; mostly in southern part of the zone C H R , DIS, H B Y , L A S , M A X , P E N , R E G , S A T , SID, T U A , V A L M S2S3 Bryum torquescens B C , W A , OR, C A on dry soil in Mediterranean and lowland arid climates sedimentary and meta-igneous parent material; moderate to intense grazing C H R , H B Y , L A S , R E G M S2S3 Epipterygium tozeri west of the Cascade Mountains; B C , W A , OR, C A on wet clay and sandy banks in the lowlands meta-igneous rock; southernmost site; ungrazed DIS M S3? Grimmia alpestris all provinces and states of the Pacific Northwest; in B C south of SON exposed acidic granite and sandstone; 360— 3300 m meta-igneous rock; 2/3 sites intensely grazed; low elevation (to sea level) L A S , SID, T E X M S2S3 Grimmia longirostris all provinces and states of the Pacific Northwest exposed, dry, acidic granite and quartzite; 100-3050m meta-igneous rock; site intensely grazed, low elevation (sea level) SID M S3? Grimmia ovalis all provinces and states of the Pacific Northwest dry, exposed to partially shaded, acidic, sandstone and granite and basalt, montane to alpine; (30-) 1000-2450 m meta-igneous and sandstone parent material; moderate to intense grazing; low to moderate elevation B W N , S A T Type Rank Species Documented Range in P N W Documented Habitat Current Observations Sites M S2S3 Homalothecium arenarium B C , W A , O R , C A on sandy soil and rocks, near the coast in all but one sedimentary rock sites; on meta-igneous rock as well, but then at lower elev.; absent from northernmost sites (N of JED/JER) and from B W N DIS, G A L , H B Y , JED/JER, PEN, R E G , S A T , SID M S2S3 Racomitrium affine B C (incomplete data) on periodically moist, siliceous rocks variable habitat - more frequently found on meta-igneous rock, only in low elevation sites B W N , C H R , DIS, JED/JER, L A S , P E N , R E G , S A V , SID, T E X , V A L M S1S3 Scleropodium touretii var. colpophyllum B C , W A , O R , C A on banks, soil over rock, cliffs, or concrete walls, rarely on logs, from the lowlands to about 1000m variable habitat; moderate to intense grazing B W N , C H R , G A L , H B Y , L A S , R E G , SID, T E X , T U A , V A L M S2S3 Tortula papillosissiina all provinces and states of the Pacific Northwest on dry soil and or rock, often calcareous; montane (from 350m to about 2500m) 3/5 sites are sedimentary rock; low to moderate elevation; moderate to intense grazing G A L , H B Y , R E G , S A T , T U A M SI Bartramia stricta southwestern B C and northern C A on dry cliffs, outcrops and dry grassy areas; in B C at low elev., dry, warm, south-facing rocky slopes; on well humified soil that appears to be disturbed, or in crevices of rock outcrops meta-igneous rock, low elevation; both sites with intense grazing L A S , SID M S2 Bryum canariense B C , W A , O R , C A dry soil and soil over rock, predominantly coastal Mediterranean climates sedimentary and meta-igneous rock; absent from sites at fringes of C D F zone all except B W N , S A V M S2 Didymodon brachyphyllus all provinces and states of the Pacific Northwest soil, limestone, lava, mortar, steppe, road banks, near spring, streamside, arid grassland, soil over lava, sandstone cliffs; 80-2300 m conglomerate rock; sea level; moderate grazing H B Y M SI? Didymodon eckeliae C A (San Diego, Napa and Tehama counties) Trunk and bases of trees, soil over rock; moderate elevation conglomerate rock; moderate elevation, moderate grazing G A L , ( M A X ) * M SI Didymodon nicholsonii B C , OR, C A , M T wet rocks, quartzite, wet silty sand, stream bank, canyon walls, streamside, chaparral; 50--1830 m sedimentary and meta-igneous parent material; moderate elevation, intense grazing T U A , (SAT)** M SI? Didymodon norrisii C A (three localities in Lake Co.) and O R (one station in Jackson Co.) rock, outcrops, calcareous boulders, fields, cliffs, runoff areas, 400-1500 m sandstone parent material, moderate elevation, intense grazing; most basic site S A T Type Rank Species Documented Range in P N W Documented Habitat Current Observations Sites M S2 Entosthodon fascicularis B C , W A , ID on periodically humid or damp earth of terraces of exposed outcrop knobs in open stands of Arbutus menziesii and Quercus garryana on southern and eastern Vancouver Island; extremely local; on soil to 700m sedimentary (2/3 sites) or meta-igneous rock; low elevation, moderately grazed; all sites adjacent to open Garry oak woodland habitat C H R , H B Y , R E G M S2? Grimmia leibergii all provinces and states of the Pacific Northwest on dry acidic boulders; 500-1500 m over meta-igneous rock; low and low-moderate elev.; moderate to intense grazing JED/JER, L A S , R E G , T E X M U Bryum gemmascens B C , OR, C A on dry soil, rock, or rarely wood, at low elevations along the coast conglomerate and meta-igneous rock; low elevation; grazing variable DIS. H B Y , L A S L U Fossombronia longiseta W North America; in the study area, Dry Southern growing on soil (mud) on slopes or over rock outcrops, on knolls adjacent to Georgia Strait; substrate moist to wet during early spring, later drying out (along with the plants); open and sunny meta-igneous and conglomerate rock; low elevation; moderate grazing; sites with relatively more northern geographic position B W N , H B Y L U Lophozia obtusa W North America; in the study area, Dry Northern-Interior growing on needle-litter, mosses, soil, or humus on the ground or over rock in dry coniferous forest or on exposed rock outcrops; substrate moist to mesic to dry, open (or partially shaded) meta-igneous rock; low elevation; intensely grazed SID L u Riccia Ibeyrichiana N A - in the study area, Rare or Restricted, distribution poorly understood growing on mud at the sides of tire-track depressions in a flower-knoll near Georgia Strait, south of Nanaimo; on soil on a streamlet bank in subalpine meadows at Washington Pass, east of Ross Lake; substrate moist, open meta-igneous and conglomerate rock; low elevation; moderate grazing; sites with relatively more northern geographic position B W N , H B Y individual collection at Bayne's Peak on Mt. Maxwell (Saltspring Island); not associated with the study site (different type of parent material). **Collected on Mt. Warburton Pike, Saturna Island during a pre-sampling survey of the area. CO Grimmia alpestris is not currently ranked by the BC Conservation Data Centre, but based on its documented distribution in BC it should be considered at least S3 (Blue-listed), if not S2 (R.I. Hastings 2006, pers. comm.). This species has been incorrectly synonymized with G. tenerrima. Currently there are no authorities on Grimmia who recognize the validity of the name G. tenerrima; R.I. Hastings, H.C. Greven, and J. Munoz all agree that the real G. tenerrima is actually a synonym of another species, G. sessitana. Specimens named G. tenerrima will be either G. alpestris (most common) or G. sessitana (rarer). Some misidentified specimens can be G. caespiticia (R.I. Hastings 2006, pers. comm.). R.I. Hastings and H.C. Greven have 12 specimen records for Grimmia alpestris in BC (R.I. Hastings 2006, pers. comm.); all of them south of about 51°N latitude and stretching across the entire province. Hastings and Greven currently have 7 records for G. sessitana, all of them east of 120°W longitude in the southeast corner of BC. However, this species has been recorded in the Olympic Peninsula and around Seattle, so it would not be unexpected in southwestern BC. Hastings and Greven also have 7 records of G. caespiticia; all are south of 52°N and stretch across the province. Grimmia alpestris and G. caespiticia have both been found on Vancouver Island. Hastings and Greven (2006) describe the habitat and range of G. alpestris as occurring on exposed acidic granite and sandstone, 360-3300 m, in all states and provinces of the Pacific Northwest (excluding Alaska). In this study, G. alpestris was recorded at sites on Sidney Island, Lasqueti Island, and Texada Island. A l l of these sites are at low elevations, with meta-igneous rock parent material, and moderate to high grazing intensity. Grimmia longirostris has been synonymized with G. affinis (Munoz 1998), which is currently agreed upon by Grimmia authorities R.I. Hastings, H.C. Greven, and J. Munoz (R.I. Hastings 2006 pers. comm.). This species is currently blue-listed in BC (S2S3 ranking). It is most common in the eastern ranges of the Rocky Mountains (ranging from western Texas to Alaska), occurring on exposed, dry, acidic granite and quartzite, from 100-3050 m. It is rare in coastal areas, becoming more common inland (Hastings & Greven 2006). In this study, G. longirostris was recorded on Sidney Island, a low-elevation site with meta-igneous rock parent material and high grazing intensity. Grimmia ovalis is currently a yellow-listed (unthreatened) species in BC, with an uncertain S3 ranking. R.I. Hastings (2006, pers. comm.) suggests an S3 ranking for this 37 species, which would change its status to Blue-listed. Grimmia ovalis is common and widespread in high elevation sites in western North America from southern BC along the Rocky Mountain corridor to southern New Mexico and south central California. R.I. Hastings and H.C. Greven (2006) describe its habitat as dry, exposed to partially shaded, acidic rock, occurring on sandstone, granite and basalt, from montane to alpine elevations (30-) 1000-2450 m. In this study, G. ovalis was recorded at sites on Bowen Island and Saturna Island. Material from the site on Saturna (moderate elevation, sandstone parent material, high grazing intensity) is rather uncharacteristic of the species, requiring further investigation (R.I. Hastings 2006, pers. comm.). Bartramia stricta is a federally and provincially rare species. The Committee on the Status of Endangered Wildlife in Canada (COSEWIC) designates it as nationally endangered - i.e. facing imminent extirpation or extinction; in BC it has an SI ranking (Red list). Bartramia stricta's North American distribution is limited to southwestern British Columbia and northern California where it is restricted to low elevation, Mediterranean-like climates, occurring on dry cliffs, outcrops, and dry grassy areas (Belland 1997). According to Rene Belland's 1998 species status report for COSEWIC, in Canada, this species is known only from one site on Vancouver Island, on a low elevation rocky slope in open stands dominated by Quercus garryana (BC Conservation Data Centre 2006). In this study, Bartramia stricta was recorded at sites on Lasqueti Island and Sidney Island. These sites are both at low elevation, with meta-igneous rock parent material, and high grazing intensity. This corresponds with the habitat description by Belland (1998), i.e. that the species requires microhabitats free of competition from grasses and herbs. Bryum canariense is currently Red-listed (S2 ranking) in BC. It was, however, recorded at every site visited except for those located at the fringe of the CDF zone (i.e. Savary Island and Bowen Island). Its relative ubiquity suggests that it is locally frequent within rock outcrop ecosystems in CDF zone landscapes. Spence (1998) describes its habitat as dry soil and soil over rock, in predominantly coastal Mediterranean climates, ranging from BC to California. This species is morphologically similar to the more common Bryum capillare, which may cause some problems with misidentification (see Appendix lb), and which may have contributed to its perceived rarity. 38 Didymodon brachyphyllus (identification confirmed by Richard Zander) was recorded in Helliwell Park on Hornby Island, a low-elevation site with conglomerate rock parent material and moderate grazing intensity. This represents the first coastal collection of D. brachyphyllus in BC; previously it had been documented only from interior dry grassland habitats. Didymodon brachyphyullus is Red-listed provincially (S2 ranking). Zander (2006) describes its habitat as soil, limestone, lava, mortar, steppe, road banks, near springs, streamside, arid grassland, soil over lava, sandstone cliffs; 80-2300m. Didymodon brachyphyllus is a North American endemic, distributed sporadically from British Columbia to Mexico (Mcintosh 1989). Didymodon eckeliae was identified from an individual collection made on Bayne's Peak, Mt. Maxwell (Saltspring Island), as well as from the Mt. Galiano study site (Galiano Island). Identifications were confirmed by Richard Zander. Both locations had conglomerate rock parent material, and were at moderate elevations. To date, this endemic species is known only from a few collections in California (San Diego, Napa and Tehama counties). Zander (2006) describes its habitat as trunks and bases of trees, and soil over rock at moderate elevations. These collections represent the first record of this species in BC and Canada. As such, it should be designated as provincially rare (Red-list, SI), as well as nationally endangered (listed by COSEWIC). Didymodon nicholsonii was recorded in sites on Saturna Island (Mt. Warburton Pike), and Saltspring Island (Mt. Tuam). Identifications were confirmed by Richard Zander. Saturna and Saltspring (Mt. Tuam) sites were both at moderate elevations. Parent materials were conglomerate or meta-igneous rock, and grazing intensity was high at both sites. Didymodon nicholsonii is Red-listed (SI ranking) in BC. Zander (2006) describes its habitat as wet rocks, quartzite, wet silty sand, stream bank, canyon walls, streamside, chaparral; 50-1830 m. Its distribution in the Pacific Northwest includes BC, Oregon, Montana, and California. Didymodon norrisii was identified from Mt. Warburton Pike on Saturna Island. This site was at moderate elevation, with sandstone parent material and high grazing intensity. It was the most basic rock outcrop site (pH = 6.3) encountered in the survey. Identification was confirmed by Richard Zander. Interestingly, this species occurred in the same microplot as the uncharacteristic, aforementioned Grimmia ovalis. To date, this recently described 39 (Zander 1999) western species is known only from California (three localities in Lake County) and Oregon (one station in Jackson County) (BC Conservation Data Centre 2006). Zander (2006) describes its habitat as rock, outcrops, calcareous boulders, fields, cliffs, runoff areas; 400-1500 m. This collection is the first record of this species in BC and Canada, and thus it should be designated as provincially rare (Red-list, SI), as well as nationally endangered (listed by COSEWIC). Entosthodon fascicularis is Red-listed provincially (S2 ranking). Federally, COSEWIC lists it as of special concern - i.e. "a species that may become threatened or endangered because of a combination of biological characteristics and identified threats." In this study, Entosthodon fascicularis was recorded at sites on Hornby Island and Saltspring Island (Channel Ridge, Reginald Hill). Two of these sites had sedimentary rock parent material (Reginald Hi l l had meta-igneous rock). A l l were at low elevations, moderately grazed, and in or adjacent to open/scattered Garry oak woodlands. These occurrences correspond with previously documented habitat preferences, i.e. extremely local on periodically humid or damp earth of terraces of exposed outcrop knobs in open stands of Arbutus menziesii and Quercus garryana on southern and eastern Vancouver Island (Schofield 1976). In western North America, Entosthodon fascicularis occurs in BC, Washington, and Idaho (BC Conservation Data Centre, 2006). Grimmia leibergii is a western North American endemic. It has commonly been misidentified as either a variation of Racomitrium heterostichum (whose habit it resembles) or a variation of Grimmia trichophylla (R.I. Hastings 2006, pers. comm.). Grimmia leibergii is currently unranked by the BC Conservation Data Centre owing to confusion surrounding synonyms and misidentification. However, R.I. Hastings (2006, pers. comm.) suggests that it should currently have an S2 ranking (Red-list), as well as a significant global (G) ranking. Hastings and Greven (2006) describe Grimmia leibergii's habitat as dry acidic boulders; 500-1500 m. R.I. Hastings has 12 records for this species in BC, including southern Vancouver Island, Saltspring Island, near the city of Vancouver, and a northern outlier on Moresby Island in the Queen Charlottes (R.I. Hastings 2006, pers. comm.). It also occurs in Montana, Idaho, Oregon, and California. In this study, Grimmia leibergii was recorded at study sites on Jedediah and Jervis Islands, Lasqueti Island, Saltspring Island (Reginald Hill), and Texada Island. A l l of these sites had meta-igneous parent material, were 40 at relatively low elevations (on bluffs overlooking the ocean, although not at sea level), and had moderate to high grazing intensity. Bryum gemmascens is currently unrecognized (and unranked) by the BC Conservation Data Centre, but may be of provincial concern once distribution information is more complete. Bryum gemmascens is often considered to be a synonym or variation of Bryum capillare. However, using Spence's (1998) key to Bryum in the Pacific Northwest, this species was identified from Discovery Island, Hornby Island and Lasqueti Island sites. These sites were at low elevations, with conglomerate rock or meta-igneous rock parent material, and with grazing intensity ranging from none to high. Spence (1998) describes the habitat and range of Bryum gemmascens as dry soil, rock, or rarely wood, at low elevations along the coast in BC, Oregon, and California. This species is endemic to western North America. Currently there are no rarity status rankings for any liverwort species in BC. However, Godfrey's (1977) habitat and distribution descriptions (summarized in Table 3.2.2), suggest that the liverworts Fossombronia longiseta and Riccia Ibeyrichiana have a limited distribution in the province, and may be of conservation concern when provincial distribution information is more complete. In this study, both species were recorded at Hornby Island and on Bowen Island sites. Both of these sites were at low elevations, with moderate grazing intensity. In summary, the majority of rare species identified from rock outcrop ecosystems were bryophytes; 18 moss species had confirmed or tentative SI-S3 rankings, as compared to four herb species. This finding indicates that bryophytes (which are often overlooked or misidentified in ecological surveys) should receive significant attention in conservation and management strategies for rock outcrop ecosystems in the CDF zone landscape. In this regard, it was observed that the majority (13) of the rare mosses occurred only in moderate to intensely-grazed landscapes, suggesting that grazing may play an important role in providing microhabitats that support these taxa. 41 3.3. Landscape-scale Rock Outcrop Vegetation - Habitat Relationships on the Gulf Islands of the CDF Zone, BC This section examines the individual effects of three landscape (i.e. site-level) features that influence rock outcrop vegetation within the CDF zone, in relation to the coverage and richness of vegetation (overall, by life form groups, and by rarity status). The combined influence of these landscape features on fine-scale rock outcrop vegetation patterns is investigated in Section 3.3.4. 3.3.1 Site Geographic Position The geographic orientation of sites along a southeast-to-northwest gradient gave rise to a strong correlation between latitude and average annual rainfall throughout the study area (R2=0.53,^=0.007); (Figure 3.3.1). As such, study sites with a more northwestern position had higher annual rainfall than those located in the southeast. Temperature patterns among climate stations also were investigated, but trends associated with latitude/longitude were much weaker, and data were less complete. 1300 -g- 1200 s ^ 1100 •I IOOO Pi 1 900 e < 800 u uo 2 700 <u > ^ 600 500 -1 1 1 O 9 0 12 -o 11 - 50 / 10 --40 S 0 7 06 -- 30 -S 20 1 0 1 '0 8 1 1 -48.0 48.5 49.0 49.5 Latitude (Degrees N) 50.0 Figure 3.3.1. Rainfall trend across the study area, from Canadian climate station averages (1961-1990) located throughout the CDF zone (Environment Canada 2006). Numbered labels refer to climatic stations listed in Table 3.3.1. 42 Table 3.3.1. Climatic stations (STN) used to characterize the trend in average annual rainfall across the study area. Station numbers (STN) are indicated in reference to Figure 3.3.1. STN Climate Station Name/Location Latitude Longitude 1 VICTORIA G O N Z A L E S HTS 48°25' N 123° 19' W 2 VICTORIA G O R D O N H E A D 48°28" N 123°18' W 3 C E N T R A L SAANICH ISL VIEW 48°34' N 123°22' W 4 C O W I C H A N B A Y C H E R R Y PT 4 8 ° 4 3 ' N 123°33' W 5 D U N C A N FORESTRY 48°46' N 123°41' W 6 G A B R I O L A ISLAND 49°09' N 123°44' W 7 N A N A I M O D E P A R T U R E B A Y 4 9 ° 1 3 ' N 123°57' W 8 B A L L E N A S LIGHTSTATION CS 4 9 ° 2 1 ' N 124° 10' W 9 Q U A L I C U M R FISH R E S E A R C H 49°24' N 124°37' W 10 M E R R Y ISLAND 49°28' N 123°55' W 11 P O W E L L RIVER 49°52' N 124°33' W 12 O Y S T E R RIVER UBC 4 9 ° 5 3 ' N 125°08' W Site soil properties showed a weak but still noteworthy correlation with latitude (R >0.1) (Figure 3.3.2a-d). Toward the northwest, sites had lower average soil depth 2 o (R =0.28), lower standard deviation in average soil depth (R =0.23), higher % carbon 2 * 2 (R =0.29), and higher % nitrogen (R =0.23). In southeast sites, soils were deeper, had more fine-scale variability in soil depth, and higher proportions of inorganic compounds. Since the frequencies of rock types were comparable across the latitudinal gradient (i.e. north and south of 49°N), it is likely that the differences in soil qualities reflect contrasting abiotic influences, and a longer history of human usage in the southern part of the study area (e.g. as rangeland). It was predicted that climatic effects associated with rainfall and temperature gradients within the CDF zone would influence patterns of vegetation coverage and richness, by life form group, as well as by origin and rarity status, and that (i) exotic taxa would have higher coverage and richness in the southeastern portion of the study area, which is more populated and has a longer history of human use, and (ii) the number of rare and uncommon species encountered would also be higher in the southeastern portion of the study area, as certain taxa approach the northern limit of their range and respond to shifts away from the characteristically dry, mild Mediterranean-type climatic regime (i.e. the annual rainfall and temperature gradients associated with latitude). 43 48.0 (a) 48.5 49.0 49.5 Latitude (Degrees N) 50.0 48.0 48.5 49.0 49.5 Latitude (Degrees N) 50.0 (b) 48.5 49.0 49.5 Latitude (Degrees N) 50.0 48.0 48.5 49.0 49.5 Latitude (Degrees N) 50.0 (c) (d) Figure 3.3.2a-d. Relationships between latitude and soil properties: (a) average soil depth, (b) average standard deviation in soil depths, (c) average total carbon content in soil samples, (d) average total nitrogen content in soil samples. There were no clear trends for overall plant coverage and species richness in relation to latitude, perhaps owing to contrasting relationships between vascular plants and bryophytes (Figure 3.3.3a,b). There was lower vascular plant coverage in northern sites (R =0.13), whereas bryophyte coverage had no correlation with latitude. In contrast, 44 bryophyte species richness was lower in northern sites (R2=0.09), and vascular plant species richness increased weakly across the latitudinal gradient (R =0.05). 0 N V A S C x N B R Y O Latitude (Degrees N) Latitude (Degrees N) (a) (b) Figure 3.3.3a,b. Trends in vegetation (a) percent cover, and (b) species richness across a latitudinal gradient, by life form group (VASC=vascular plants, BRYO=bryophytes). The relationship between latitude and vascular plant coverage and species richness differed between plant origin groups (i.e. native vs. exotic) (Figure 3.3.4a,b). Only one bryophyte species (Campylopus introflexus, recorded at Texada Island) was exotic in origin, so bryophytes were excluded from this part of the analysis. Exotic vascular plant coverage decreased (R2=0.17), while native vascular plant coverage had a weak positive correlation (R =0.07) with latitude (Fig 3.3.4a). In contrast, the richness of native vascular plant species was higher in northern sites (R2=0.21, Figure 3.3.4b), whereas exotic vascular plant species richness was uncorrelated with latitude. As predicted, latitude was correlated with trends in decreasing exotic species coverage, and increasing native species richness, for vascular plants. 45 48.5 49.0 49.5 Latitude (Degrees N) (a) o Native x Exotic 50 0 0 48.0 48.5 49.0 49.5 Latitude (Degrees N) (b) o N Native x N Exotic 50.0 Figure 3.3.4a,b. Trends in vascular plant (a) coverage and (b) species richness across a latitudinal gradient, by origin (native, exotic). The decreased coverage of exotic vascular plants in northern sites reflects the decline in exotic graminoid cover (R2=0.20) (Figure 3.3.5). Native vascular plant richness was higher in northern sites owing to increasing numbers of native graminoids (R2=0.34, /?=0.048) and, to a lesser extent, increasing numbers of native woody plant species (R =0.12) (Figure 3.3.6a,b). 60 1 5 0 -c 48.0 48.5 49.0 49.5 50.0 Latitude (Degrees N) Figure 3.3.5. Trends in exotic graminoid coverage along a latitudinal gradient. 46 Latitude was positively correlated with the coverage of Red- and Blue-listed (S1-S3-ranked) plants (R2=0.32), and with the coverage of uncommon (S4-ranked) Yellow-listed plants (R2=0.32) (Figure 3.3.7a,b). This increase was reflective of higher coverage of SI-S3 bryophytes (R2=0.32) and the higher coverage of S4 bryophytes (R2=0.27) and S4-ranked graminoids (R2=0.33,/?=0.050) in northern sites (Figure 3.3.8a-c). 48.5 49.0 49.5 Latitude (Degrees N) 50.0 > o O £ 20 48.0 48.5 49.0 49.5 Latitude (Degrees N) 50.0 (a) (b) Figure 3.3.7a,b. Trends in the percent coverage of (a) Sl-S3-ranked plants, and (b) S4-ranked plants along a latitudinal gradient. 47 48.0 48.5 49.0 49.5 50.0 Latitude (Degrees N) (c) Figure 3.3.8a-c. Trends in the coverage of (a) Sl-S3-ranked bryophytes, (b) S4-ranked bryophytes, and (c) S4-ranked graminoids along a latitudinal gradient. Contrasting with trends in cover, the species richness of S4 bryophytes was negatively correlated with latitude (richness of SI-S3 bryophytes had no clear relationship with site geographic position). As predicted, higher numbers of "uncommon" bryophytes were found toward the southern end of the zone (Figure 3.3.9a; R2=0.20). Similar to the coverage trends, however, the richness of S4 graminoid species increased with latitude (Figure 3.3.9b; R2=0.20). 48 48.0 48.5 49.0 49.5 50.0 48.0 48.5 49.0 49.5 50.0 Latitude (Degrees N) Latitude (Degrees N) (a) (b) Figure 3.3.9a,b. Trends in the richness of (a) S4-ranked bryophyte species, and (b) S4-ranked graminoid species along a latitudinal gradient. In summary, patterns of coverage, richness, and rarity across the latitudinal gradient were found to be dependent on species origin and life form group. It was found that overall vascular plant cover declined with increasing latitude, as did bryophyte species richness. As predicted, there was higher coverage of exotic vascular plant species (particularly exotic graminoids) in southern sites. Concurrently, there was higher richness of native vascular plant species (particularly native graminoids and native shrubs) in northern sites. Also as predicted, the richness of uncommon bryophyte species declined with increasing latitude. In contrast with predictions, the coverage of rare and uncommon bryophytes, and the cover and richness of uncommon graminoids increased with latitude. It is likely that competitive exclusion by exotic graminoids is an important factor influencing the prominence (but for bryophytes, not necessarily the presence) of these species, particularly in southern sites, which have a longer history of human use. In this regard, MacDougall and Turkington (2004) found that exotic dominance in an invaded Garry oak savanna in BC was contingent upon the interaction of competitive strategies, resource availability and disturbance history rather than any single factor; their results highlight the importance of examining the historical context of invaded plant communities. Disturbance history and competitive influences may impact the species richness of life form groups differently; native graminoids and native woody plants appeared to be the 49 most sensitive in this regard. Despite their lower coverage, the richness of uncommon bryophytes was higher in the south end of the CDF zone, while herb species richness had no relationship with this gradient. This finding indicates that many bryophyte and herb species may persist despite exotic dominance. Similarly, MacDougall and Turkington (2006) observed that in a degraded Garry oak savanna in BC, the site-level concordance between regional and local diversity for native species was associated primarily with environmental influences (e.g. soil depth heterogeneity), not dispersal or competition. Remaining populations of native species appeared to be confined to optimal habitat, resisting competitive or stochastic displacement, possibly explaining why species loss is rare despite substantial habitat loss and invasion. Seabloom et al. (2006) found that weedy and invasive exotics were tightly linked to the distribution of imperiled species in California, USA, with the richest exotic flora occurring in low-lying (more populated) coastal sites that also harbour large numbers of imperiled species. In addition, Vanderpoorten and Engels (2003) found that even the least diverse, cultivated areas in central Belgium included a significant number of regionally rare species, thus lowering their ability to predict species richness and rarity from landscape features and soil conditions alone. 3.3.2 Site Geology Three main types of parent material were encountered: meta-igneous rock (volcanic in origin), and conglomerate and sandstone rock (both sedimentary in origin). The meta-igneous rocks collected in study sites ranged from mafic to felsic (Mary-Lou Bevier, pers. comm. 2003). The two most common types of cement in sedimentary rocks are calcium carbonate (CaCC^) and quartz (SiC^), the former being more basic and more subject to erosion (Monroe & Wicander 2001). A l l of the sedimentary rocks collected for this study had quartz cement material (Mary-Lou Bevier, pers. comm. 2003). The rock types of plot-sampled sites had contrasting levels of grazing intensity. To minimize confounding results in comparing parent materials, only moderately-grazed sites were used (the combined influence of geology and grazing intensity on vegetation patterns is investigated in Section 3.3.4). The eight plot-sampled, moderately-grazed sites were evenly distributed across the latitudinal gradient (i.e. above and below 49°N) (Table 3.3.2). Owing to the lower number of sedimentary rock sites, data for sandstone and conglomerate rock 50 sites were pooled for this analysis (measured environmental variables generally did not differ among these sites). Table 3.3.2. Sites used to compare effects of parent material. Site Code Rock Type Rock Origin Position Grazing BWN meta-igneous volcanic North Moderate T E X meta-igneous volcanic North Moderate M A X meta-igneous volcanic South Moderate REG meta-igneous volcanic South Moderate HBY conglomerate sedimentary North Moderate G A L conglomerate sedimentary South Moderate V A L sandstone sedimentary North Moderate CHR sandstone sedimentary South Moderate Characteristic large- and fine-scale topographies result from physical and chemical properties that are unique to rock type. Rock type is, therefore, related to the quality and variety of habitat types within a site and, thus, soil and vegetation development and constituent species distributions. Measured variables were averaged for plot-sampled sites sharing the same origin of parent material (volcanic or sedimentary). These data are shown in Table 3.3.3, withp-values resulting from two-sample t-tests comparing means. 51 Table 3.3.3. Average site values for measured variables on meta-igneous (META) and sedimentary (SED) rock sites. Mean, standard error (SE), and /rvalues (where /K0.05) from two-sample t-tests comparing means are shown. Relationship direction is also shown, i.e. M E T A site means greater than (>) or less than (<) SED site means (N.S.=not significant, i.e. /»0.05). M E T A SE SED SE p-value Number of Sites 4 - 4 - -Latitude (Degrees N) 49.1 0.21 49.1 0.15 N.S. Longitude (Degrees W) 123.7 0.25 123.8 0.27 N.S. Soil Depth (cm) 2.9 0.55 4.6 0.85 N.S. Maximum Soil Depth (cm) 6.1 0.98 7.7 1.58 N.S. Minimum Soil Depth (cm) 0.8 0.18 2.0 0.37 0.018 (<) Standard Deviation Soil Depth 1.9 0.27 1.9 0.44 N.S. Soil pH 4.7 0.10 5.1 0.16 0.043 (<) Soil % Total C 28.5 2.30 18.3 3.48 0.028 (>) Soil % Total N 1.7 0.25 1.1 0.18 N.S. C:N Ratio 17.9 1.20 15.4 0.96 N.S. Canopy Cover (N) 17.0 6.54 3.3 2.94 N.S. Canopy Cover(E) 11.6 5.73 4.2 2.59 N.S. Canopy Cover(S) 3.8 3.35 0.4 0.19 N.S. Canopy Cover (W) 2.3 1.55 4.1 1.67 N.S. Average Canopy Cover 8.7 3.34 3.0 1.57 N.S. Distance to Canopy (m) 12.9 3.27 14.4 4.70 N.S. Horizontal Slope (Degrees) 10.3 0.79 4.3 0.76 0.001 (>) Horizontal Aspect (Degrees-N) 73.0 7.63 42.9 9.01 0.022 (>) Vertical Slope (Degrees) 18.3 0.59 12.1 2.30 0.035 (>) Vertical Aspect (Degrees-N) 154.9 10.61 101.5 13.47 0.011 (>) Total Slope 22.2 0.24 13.8 2.07 0.013 (>) Total Aspect (Degrees-N) 131.4 8.90 98.8 12.04 0.038 (>) Maximum Veg. Height (cm) 18.0 4.78 24.3 5.22 N.S. % Cover Bare Parent Material 20.3 5.99 5.7 2.20 0.044 (>) % Cover Organic / Litter 6.7 0.89 11.7 1.98 0.042 (<) % Cover Wood 0.1 0.08 1.1 0.71 N.S. % Cover Total Vegetation 73.1 6.21 82.8 2.23 N.S. % Cover All Vascular Plants 25.1 4.04 46.3 5.96 0.015 (<) % Cover Woody Plants 0.1 0.04 0.6 0.61 N.S. % Cover Graminoids 12.5 3.77 33.7 4.37 0.005 (<) % Cover Herbs 13.3 1.56 14.0 1.77 N.S. % Cover Lichens 3.5 1.20 5.3 4.52 N.S. % Cover Bryophytes 49.6 7.30 44.6 8.28 N.S. N Total Vegetation 91.8 6.12 114.5 8.87 0.043 (<) N All Vascular Plants 53.5 4.48 74.0 7.82 0.037 (<) N Woody 6.5 2.40 6.0 1.08 N.S. N Graminoids 11.8 1.31 15.8 2.69 N.S. N Herbs 35.3 3.12 52.3 5.31 0.021 (<) N Bryophytes 38.3 3.42 40.5 1.85 N.S. 52 Meta-igneous rock sites had significantly higher bare rock coverage, shallower minimum soil depths in microplots, higher total % carbon in soil samples, and lower pH in soil samples than sedimentary rock sites. Microplot slopes (horizontal, vertical, and combined) were significantly higher in meta-igneous sites, reflecting greater fine-scale topographic variation on that rock type. Conglomerate rock sites had significantly greater standard deviation (i.e. variation) in soil depths within microplots (/?=0.048 for meta-igneous rock;/>=0.024 for sandstone rock), indicating patchy soil development (Figure 3.3.10). Average microplot aspects (horizontal, vertical, and combined) also were significantly higher in meta-igneous sites, reflecting a greater frequency of south-facing microplots (i.e. degrees from north). C O N G M E T A S A N D Parent Material Figure 3.3.10. Average standard deviation in soil depth measurements within microplots, by parent material (CONG=conglomerate, META=meta-igneous, SED=sedimentary). Standard error bars shown. Sedimentary rock sites had significantly higher coverage of vascular plants (particularly graminoids), as well as higher coverage of organic matter or decaying litter at ground-level. Overall species richness was significantly higher in sedimentary rock sites, reflecting greater numbers of vascular plant species (particularly herbs). Bryophyte coverage did not differ significantly between meta-igneous and sedimentary rock sites. However, sandstone rock sites had significantly higher bryophyte coverage than both the conglomerate rock sites (p=0.026) and meta-igneous rock sites (/?=0.039) (Figure 3.3.11). 53 70 u a a. 30 C O N G 3 M E T A S A N D Parent Material Figure 3.3.11. Average bryophyte coverage at sites, by parent material (CONG=conglomerate, META=meta-igneous, SED=sedimentary). Standard error bars shown. The overall coverage of native vascular plant species did not differ significantly between the rock types. In contrast, the overall coverage of exotic vascular plants was significantly higher in sedimentary rock sites (p=0.009), primarily owing to the higher exotic graminoid coverage (/?=0.006) (Figure 3.3.12a,b). Bryophyte species richness did not differ significantly between rock types, nor did the overall richness of vascular plant species, although the number of native herb species was significantly higher in sedimentary rock sites (p=0.020) (Figure 3.3.13). The overall species richness of exotic vascular plants also was significantly higher in sedimentary rock sites (/?=0.012), reflecting higher numbers of exotic graminoids (p=0.022), and higher numbers of exotic herbs (/?=0.044) (Figure 3.3.14a-c). 54 M E T A S E D Parent Material M E T A S E D Parent Material (a) (b) Figure 3.3.12a,b. Average coverage of (a) all exotic vascular plants, and (b) all exotic graminoid species in sites, by parent material (META=meta-igneous, SED=sedimentary). Standard error bars shown. M E T A S E D Parent Material Figure 3.3.13. Total number of native herb species at sites, by parent material (META=meta-igneous, SED=sedimentary). Standard error bars shown. 55 M E T A S E D M E T A S E D Parent Material Parent Material (a) (b) M E T A S E D Parent Material (C) Figure 3.3.14a-c. Total number of (a) exotic vascular plant species, (b) exotic graminoid species, and (c) exotic herb species at sites, by parent material (META=meta-igneous, SED=sedimentary). Standard error bars shown. The coverage of Sl-S3-ranked and uncommon S4-ranked species did not differ significantly between the two rock types, overall or by life form group. Similarly, the richness of Sl-S3-ranked taxa did not differ significantly between rock types. However, the richness of S4-ranked species was higher in sedimentary rock sites (/?=0.038), reflecting higher numbers of S4-ranked herbs (/?=0.021) (Figure 3.3.15a,b). Within sedimentary rock sites, the richness of S4-ranked species was significantly higher on conglomerate rock (p=0.035) (Figure 3.3.16). 56 M E T A S E D M E T A S E D Parent Material Parent Material (a) (b) Figure 3.3.15a,b. Total number of (a) S4-ranked plant species, and (b) S4-ranked herb species at sites, by parent material (META=meta-igneous, SED=sedimentary). Standard error bars shown. C O N G M E T A S A N D Parent Material Figure 3.3.16. Total number of S4-ranked plant species at sites, by parent material (CONG=conglomerate, META=meta-igneous, SED=sedimentary). Standard error bars shown. In summary, it was found that rock type influenced patterns of vegetation coverage and richness in rock outcrop ecosystems in the CDF zone. Results indicated that meta-igneous rock sites were associated with lower soil development (higher coverage of bare parent material, lower soil depth, lower soil pH), and higher fine-scale topographical 57 variation, as indicated by microplot slopes. Sedimentary rock sites had higher overall coverage of vascular plants (i.e. exotic graminoids), and richness of exotic vascular plants (i.e. exotic graminoids and exotic herbs). Conglomerate rock sites had significantly lower bryophyte coverage than meta-igneous rock and sandstone, although bryophyte species richness was not significantly different among rock types. Sedimentary rock sites had significantly higher richness of uncommon native herbs, particularly on conglomerate rock, which was associated with significantly higher variability in soil depth than other rock types. These results showed that variations in soil development, chemistry, and microtopography associated with rock types were related to differences in vegetation coverage and composition. Typically deeper, more basic soils over sedimentary rock sites had higher coverage and richness of exotic species, but also higher richness of uncommon herb species. This corresponds with vegetation patterns observed for bryophytes in relation to latitude; high coverage and richness of exotic species apparently does not preclude the presence of uncommon taxa, although competitive interactions may influence their coverage. 3.3.3 Grazing on Meta-igneous Rock The eight plot-sampled meta-igneous sites used to contrast categories of grazing intensity had an even distribution across the latitudinal gradient (i.e. above and below 49°N) (Table 3.3.4). Table 3.3.4. Sites used to compare effects of grazing intensity. Site Code Grazing Position Rock Type WIN none North meta-igneous DIS none South meta-igneous BWN moderate North meta-igneous T E X moderate North meta-igneous M A X moderate South meta-igneous REG moderate South meta-igneous LAS high North meta-igneous SID high South meta-igneous Moderately grazed sites had significantly higher canopy cover (p=0.044) than sites with no grazing, or sites with high grazing intensity (p=0.047) (Figure 3.3.17a). Both ungrazed sites were on relatively small, barren islands with intermittent and/or stunted canopy coverage. In contrast, all rock outcrop sites with moderate or high grazing were 58 adjacent to tall, continuous stands of Pseudotsuga menziesii on at least one side. In addition, the stunted tree/shrub layer extending out from adjacent canopies in ungrazed sites pushed the boundary of sampled habitat further from forest edges, hence canopy cover was lower. Sites with high grazing intensity were associated with more open, continuously bare landscapes with lower frequency of (or greater distance to) intermittent trees providing shade. The predominant aspect of plots, corresponding with the fall line (i.e. plot vertical aspect, summarized as degrees from north) was also significantly higher in moderately grazed sites as compared to ungrazed sites (/?=0.004) (Figure 3.3.17b). Ungrazed islands were significantly smaller (p=0.020), uninhabited, and more uniformly barren in nature. As such, randomly-located plots on ungrazed islands were less likely to be exclusively south- or southwest-facing (i.e. in the path of prevailing wind patterns). N O N E M O D HIGH N O N E M O D HIGH Grazing Intensity Grazing Intensity (a) (b) Figure 3.3.17a,b. Average (a) percent canopy cover, and (b) plot vertical aspect (i.e. degrees from north) at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown. Ungrazed sites had significantly higher bare parent material than either moderately grazed sites (/?=0.020), or sites with high grazing intensity (/7=0.046) (Figure 3.3.18a). Although ungrazed sites had more bare rock, soils were more productive as evidenced by the lower ratio of total percent carbon to total percent nitrogen (C:N ratio) in soil samples (Figure 3.3.18b). C:N ratios were highest in sites with high grazing intensity (/?<0.019). 59 N O N E MOD HIGH N O N E M O D HIGH Grazing Intensity Grazing Intensity (a) (b) Figure 3.3.18a,b. Average (a) percent bare parent material, and (b) ratio of total percent carbon to total percent nitrogen in soil samples at meta-igneous rock, sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown. It was predicted that grazing intensity would influence patterns of vegetation coverage and richness, by life form group, as well as by origin and rarity status. Specifically, it was predicted that ungrazed sites would have higher richness and coverage of native plant species, and that grazed sites would have a higher richness and coverage of exotic plant species. Overall vegetation cover (i.e. of vascular plants and bryophytes combined) was lowest in ungrazed sites (significantly so as compared to moderately-grazed sites (p=0.028)) (Figure 3.3.19a). Concurrently, bryophyte coverage was lowest in ungrazed sites (p<0.005) (Figure 3.3.19b). In contrast with coverage trends, the total number of vascular plant species was highest in ungrazed sites (/?<0.049), reflecting higher richness of woody plant species (p<0.043) (Figure 3.3.20a,b). 60 70-N O N E M O D HIGH Grazing Intensity NONE M O D HIGH Grazing Intensity (a) (b) Figure 3.3.19a,b. Average percent cover of (a) total vegetation (vascular plants and bryophytes), and (b) bryophytes at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown. N O N E M O D HIGH Grazing Intensity N O N E M O D HIGH Grazing Intensity (a) (b) Figure 3.3.20a,b. Total number of (a) all vascular plant species, and (b) woody plant species at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown. It was predicted that ungrazed sites would have higher richness and coverage of native plant species, and that grazed sites would have higher richness and coverage of exotic 61 plant species. In contrast with this prediction there were no significant differences in the overall coverage or richness of vascular plants among grazing levels when compared by origin (i.e. native vs. exotic), indicating that the higher number of species in ungrazed sites represented greater numbers of both native and exotic taxa. However, there were significantly fewer native graminoid species in highly-grazed sites than in moderately grazed sites (p=0.013) (Figure 3.3.21). NONE MOD HIGH Grazing Intensity Figure 3.3.21. Total number of native graminoid species at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown. The coverage of Sl-S3-ranked plants (in total, and by life form group) was not significantly different among grazing classes. The coverage of S4-ranked plants (Figure 3.3.22a) and the coverage of S4-ranked bryophytes (Figure 3.3.22b) were lowest in ungrazed sites (/?=0.038 and /?=0.048, respectively, when compared to moderately and grazed sites). The richness of S1 -S3 bryophytes was highest in sites with high grazing intensity (p=0.045 when compared to moderately grazed sites) (Figure 3.3.23). Although total vascular plant species richness was highest in ungrazed sites, the richness of rare taxa in rock outcrop ecosystems (being predominantly represented by rare bryophytes) was higher in sites with high grazing intensity. Grazed sites likely provide more favorable habitat for such species, i.e. greater soil coverage improves establishment, while grazing reduces competition from more aggressive vascular plants. 62 — 1 o ~ ~ -N O N E M O D HIGH N O N E M O D HIGH Grazing Intensity Grazing Intensity (a) (b) Figure 3.3.22 a,b. Average percent cover of (a) all S4-ranked plants, and (b) S4-ranked bryophytes at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown. N O N E M O D HIGH Grazing Intensity Figure 3.3.23. Total number of SI-S3-ranked bryophyte species at meta-igneous rock sites, by grazing intensity (none=no grazing, MOD=moderate grazing intensity, and HIGH=high grazing intensity). Standard error bars shown. 63 In summary, the preceding results indicate that the observed categories of grazing intensity are related to patterns of vegetation coverage and richness in rock outcrop ecosystems of the CDF zone, with contrasting trends among and within life form groups. Ungrazed sites had higher coverage of bare parent material, but also more productive soils (lower C:N ratios). This corresponds with the field observation of localized patches of deep, productive soils that occur in ungrazed environments compared to shallow, compacted mats that were typical of grazed sites. Ungrazed sites had lower overall vegetation cover (primarily reflecting lower bryophyte cover), but a higher number of total vascular plant species (woody plants in particular); sites with high grazing intensity had the lowest native graminoid species richness. In contrast with trends for vascular plant species richness, the coverage of common and S4-ranke'd bryophytes, and the richness of SI-S3-ranked bryophytes, was highest in intensely-grazed sites. The intermittently-moist, more uniformly shallow, compacted soils associated with intensely grazed sites is likely to prohibit the establishment of competitive vascular plants and promote bryophyte coverage and richness. This observation suggests that life form groups may vary in their response to disturbance regimes in rock outcrop habitats of the CDF zone landscape and, as such, contrasting levels of grazing will promote different arrays of component taxa. 3.4. Vegetation Assemblage Patterns in Rock Outcrop Ecosystems of the Gulf Islands in the CDF Zone, BC Section 3.3 investigated how individual landscape-level environmental gradients associated with geographical position (i.e. latitude), geology (i.e. rock type), and category of grazing intensity influenced patterns of coverage, richness, and rarities in rock outcrop habitats of the CDF zone (by life form group, and by origin). In this section, vegetation data at each scale (site, plot, and microplot) were analyzed to determine the relative importance, and integration of effects, of environmental gradients measured at corresponding scales. 64 3.4.1. Identification of Site- and Plot-level Vegetation-Habitat Relationships The first axis in a PCA on site-averaged species cover data explained 19.1% of the total variation in the data set (Table 3.4.1), and reflected a gradient of increasing soil development. Sites positively correlated with the first PCA axis had deeper soils, and were more basic; sites negatively correlated with this axis had higher coverage of bare parent material, and higher %C and % N in parent material (relative to inorganic matter). Meta-igneous rock and sedimentary rock sites were separated by an overall soil gradient summarized by PCA Axis 1 (Figure 3.4.1; PCA Axis 2 is shown for reference, but is not described here, as only the first axis was used to separate sites in this step). Table 3.4.1. Pearson correlations of site-level environmental variables with PCA Axis 1 (Factor 1), generated from site-averaged species cover data. Variable Factor (1) % Total Carbon -0.610 % Total Nitrogen -0.602 % Bare Parent Material -0.600 Horiz. Slope -0.567 % Organic/Litter 0.593 pH 0.629 Std. Dev. Soil Depth 0.673 Min. Soil Depth 0.760 Max. Soil Depth 0.791 Avg. Soil Depth 0.820 % Total Variance Expl. \9A 65 ROCK TYPE o META x SED - 2 - 1 0 1 2 3 Factor (1) Figure 3.4.1. Ordination of sites in relation to the first two axes of a PCA performed on site-averaged species cover data. Plots are grouped by rock type (META=meta-igneous, SED=sedimentary) using 68.3% confidence ellipses. Using the subset of meta-igneous rock sites, the first axis of a PCA on plot-averaged species cover data explained 18.0% of the total variation in the data set (Table 3.4.2). This axis correlated most strongly with geographic position (latitude) and soil development. Sites with a southern geographic position had higher average soil depth, as summarized by PCA Axis 1 in Figure 3.4.2 (PCA Axis 2 is shown for reference, but is not described here). This corresponds with the overall gradient in soil development correlating with the first axis in the site-level PCA. Table 3.4.2. Pearson correlations of meta-igneous rock environmental variables with PCA Axis 1 (Factor 1), generated from plot-averaged meta-igneous rock species cover data. Factor (1) Latitude (Site) -0.667 % Total Carbon -0.591 Longitude (Site) -0.553 Min. Soil Depth 0.516 pH 0.537 Avg. Soil Depth 0.556 % Total Variance Expl. 18.0 66 L A T I T U D E NORTH SOUTH - 2 - 1 0 1 2 3 FACTOR(l) Figure 3.4.2. Ordination of plots in relation to the first two axes of a PCA performed on average species coverages within meta-igneous rock study plots. Plots are grouped by latitudinal position (North=>49°N, South=<49°N) using 68.3% confidence ellipses. Using the subset of sedimentary rock sites, the first axis in a PCA on plot-averaged species cover data explained 21.8% of the total variation in the data set, and was correlated most strongly with grazing (Table 3.4.3). Vegetation assemblage patterns on sedimentary rock differed depending on grazing intensity (moderate vs. high), as represented by PCA Axis 1 (Figure 3.4.3; PCA Axis 2 is shown for reference, but is not described here). Owing to the presence of only one site with high grazing intensity (Mt. Warburton Pike on Saturna Island), this site was omitted from the remainder of the analysis. 6 7 Table 3.4.3. Pearson correlations of sedimentary rock environmental variables with PCA Axis 1 (Factor 1), generated from plot-averaged sedimentary rock species cover data. Factor (1) % Total Nitrogen -0.679 % Total Carbon -0.611 Canopy Cover - N 0.600 Total Aspect (Deg-N) 0.618 Vertical Slope 0.801 PH 0.813 Total Slope 0.850 Grazing Class 0.936 % Total Variance Expl. 21.8 GRAZING o HIGH x MOD ~-2 -1 0 1 2 3 Factor (1) Figure 3.4.3. Ordination of plots in relation to the first two axes of a PCA performed on average species coverages within sedimentary rock plots. Plots are grouped by grazing intensity using 68.3% confidence ellipses. In summary, this analysis resulted in three major categories within rock outcrop ecosystems (Figure 3.4.4), generally correlating with an overall gradient reflecting soil development (Figure 3.4.5); meta-igneous north (META-N) sites had significantly lower minimum soil depths than meta-igneous south (META-S) and sedimentary rock (SED) sites (/K0.022). Each of the three categories was represented by four study sites (Table 3.4.4). 6 8 M E T A - N Org/Litter Woody 3% Herb 12% Gram 9% Lchn 10% Bryo 34% M E T A - S Org/Litter W o o d y 0 % , 4 o / o _ 7 Herb 12% BarePM 23% f Gram 27% Bryo 39% SED Org/Litter Woody 1% 7 % Herb 15% BarePM 6% Bryo 43% 1 Gram 41% Lchn 5% Figure 3.4.4. Pie graphs showing average coverage of vegetation life form groups (Bryo=bryophytes, Lchn=lichens, Gram=graminoids, Herb=herbs, Woody=woody plants) and substratum components (Org/Litter=bare organic material and/or decayed plant litter, BarePM=bare parent material) within major rock outcrop ecosystem categories, organized relation to soil development. s i > < >- > -META-N META-S SED CATEGORY Figure 3.4.5. Average minimum soil depth (in microplots) for landscape categories identified. Standard error bars shown. Table 3.4.4. Sites representing the four major rock outcrop ecosystem categories identified. Latitudinal position (Pos., i.e. north or south of 49°N), rock type, and grazing intensity shown. Site Pos. Rock Grazing M E T A - N WIN N M E T A N O N E B W N N M E T A M O D T E X N M E T A M O D LAS N M E T A HIGH M E T A - S DIS S M E T A N O N E M A X S M E T A M O D REG S M E T A M O D SID s M E T A HIGH SED G A L s C O N G M O D H B Y N C O N G M O D CHR s SAND M O D V A L N SAND M O D Overall exotic vascular plant coverage increased from M E T A - N to META-S to SED sites (p<0.035 among all categories) (Figure 3.4.6). This increase reflected higher coverage of exotic graminoids in SED sites (p<0.027 among all categories), and higher coverage of exotic herbs in META-S and SED sites (p<0.034 as compared to M E T A - N sites). Exotic 70 vascular plant species richness was highest in SED sites (/?<0.026 as compared to M E T A categories) (Figure 3.4.7). This increase reflected higher richness of exotic graminoids (/?<0.031 as compared to M E T A categories) and higher richness of exotic herbs (p=0.016 as compared to M E T A - N sites). META-N META-S Category SED VASC G R A M HERB Figure 3.4.6. Average percent cover of exotic vascular plants in sites, by landscape category and life form group (VASC=overall, GRAM=graminoids, HERB=herbs). Standard error bars shown. • N V A S C • N G R A M • N HERB META-N META-S Category SED Figure 3.4.7. Average species richness of exotic vascular plants in sites, by landscape category and life form group (VASC=overall, GRAM=graminoids, HERB=herbs). Standard error bars shown. 71 Although exotic species were more prominent in SED sites (both in coverage and in richness), there was also higher richness of native herb species in SED sites (Figure 3.4.8), as compared to both M E T A - N and META-S sites (p<0.029). Concordantly, the only significant difference among life form group categories in SI-S3- or S4-ranked species was for S4-ranked herbs (Figure 3.4.9). The richness of S4-ranked herb species in M E T A - N sites was significantly lower than in META-S and SED sites (p<0.049). META-N META-S SED Category Figure 3.4.8. Average richness of native herb species in sites, by landscape category. Standard error bars shown. META-N META-S SED Category Figure 3.4.9. Average richness of S4-ranked herb species in sites, by landscape category. Standard error bars shown. 72 3.4.2. Identification of Microscale Vegetation Patterns This section investigates the primary environmental gradients influencing microplot-level vegetation patterning within the large-scale categories identified from preceding analyses: (1) Meta-igneous Rock - North Position (>49°N), (2) Meta-igneous Rock - South Position (<49°N), and (3) Sedimentary Rock (moderately grazed). (1) Meta-igneous Rock - North Position (META-N): The first two axes of a PCA performed on vegetation in meta-igneous rock microplots with a northern geographic position explained 14.9% and 10.1% of the total variation in the data set, respectively (Table 3.4.5). The first axis correlated relatively weakly with measured environmental variables (r<0.41). Examination of factor correlations for individual species indicated that the coverage of several well-known "seepage" species, e.g. Bryum miniatum, Mimulus spp. were correlated with this axis (r=0.43 and r=0.50, respectively). The first PCA axis was therefore interpreted to represent a moisture gradient, relating to the increasing influence of seepage within microplots. Table 3.4.5. Pearson correlations of northern meta-igneous rock environmental variables with the first two PCA axes 1 (Factor 1, Factor 2) generated from M E T A - N microplot species cover data (log-transformed). Factor (1) Factor (2) Canopy Cover - S (Plot) -0.340 Std. Dev. Soil Depth (Plot) -0.641 % Cover Org./Litter (Micro) 0.304 Dist. to Can. (Plot) -0.616 % Cover Org./Litter (Plot) 0.329 Max. Soil Depth (Plot) -0.604 pH (Plot) 0.413 Latitude (Site) 0.650 % Total Variance Expl. 14.9 Total Aspect (Plot) 0.654 Grazing Class (Site) 0.714 Vert. Slope (Plot) 0.727 % Total Variance Expl. 10.1 The second PCA axis was most strongly correlated with grazing intensity and other variables associated with exposure and potential soil development (increasing plot slope, increasing southern aspect, decreasing maximum soil depth, and decreased variation in soil depth). Distance to canopy was negatively correlated with increasing grazing intensity owing to the broad band of low-lying shrubs that separated ungrazed rock outcrop habitats 73 from adjacent forest. The position of microplots in relation to the first two axes of a PCA performed on microplot species coverages in northern meta-igneous rock sites is shown in Figure 3.4.10. 3 2 GRAZING -2 HIGH MOD NONE -3 -2 0 2 3 4 FACTOR(l ) Figure 3.4.10. Ordination of microplots in relation to the first two axes of a PCA generated from M E T A - N microplot species cover data (log-transformed). Microplots are grouped by grazing intensity using 68.3% confidence ellipses. Species habitat preferences were summarized in relation to the two major environmental gradients interpreted to influence microscale vegetation patterning within northern meta-igneous rock sites (Table 3.4.6). Species were divided into assemblages according to their correlations with the aforementioned axes (PCA Axis 1 was correlated with a microhabitat moisture gradient, and PCA Axis 2 was correlated with grazing intensity and other variables relating to exposure). Pearson correlations of r>0.3 or r<-0.3 were used to divide species into habitat groupings. 74 Table 3.4.6. Summary of dominant habitat preferences of species with >5% frequency in M E T A - N microplots; NON-SPEC.=habitat preference non-specific. Species shown by life form group (W=woody, G=graminoid, H=herb, B=bryophyte), and by origin for vascular plants (N=native, E=exotic). Species with >1% average cover within this data subset are indicated with larger, bold font. Species codes are listed in Appendix 2. PCA Axis 1 (Moisture) a. an < v O a. o£ c ' N o fN yi < < U Q. G R A Z E D N O N - S P E C . U N G R A Z E D •4 • DRY N O N - S P E C . M O I S T R A C O L A N (B) B R O M H O R (GE) V U L P M Y U (GE) P O L Y J U N (B) V U L P B R O (GE) M A D I G R A (HN) P O L Y P I L ( B ) S E L A W A L (HN) G R I M T R I ( B ) R A C O A F F (B) H Y P O R A D (HE) C E P H D I V ( B ) H E D W C I L (B) R A C O E L O (B) P S E U M E N (WN) A I R A S P P (GE) D A N T S P P (GN) C O L L I P A R (HN) A C H I M I L ( H N ) M I M U G U T (HN) L O T U S P P ( H N ) B R Y U C A N (B) P O L Y G O N U M (HU) B R Y U C A P ( B ) R A C O H E T (B) B R Y U M I N (B) C L A D O N I A (L) B R Y U P S E ( B ) C E R A P U R (B) C L A D I N A ( L ) F E S T R U B ( G N ) B R O D C O R (HN) H O L C L A N (GE) C E R A A R V (HN) S P I R R O M (HN) G E R A M O L (HN) E R I O L A N (HN) D I C R S C O (B) Microplot coverage and richness of plant life form groups (by origin) in M E T A - N sites were correlated with the first two axes of a PCA performed on microplot species cover data (Table 3.4.7); these patterns can be related to species assemblages described in Table 3.4.6. Overall coverage of woody species was greater in ungrazed microplots, but native herbs had higher coverage and richness in moist, grazed microplots. Native graminoid and herb richness was highest in ungrazed microplots (particularly in periodically moist microplots for native herb species). The coverage of exotic graminoids was highest in moist, grazed microplots. The richness of exotic graminoids and exotic herbs was highest in moist microplots (particularly in grazed microplots for exotic herbs). Bryophyte coverage was higher in dry, grazed microplots, but bryophyte richness was higher in moist, grazed microplots. 75 Table 3.4.7. Pearson correlations of microplot species coverage and richness (by life form group and plant origin) with the first two axes of a PCA generated from M E T A - N microplot species cover data (log-transformed). F A C T O R (1) F A C T O R (2) H A B I T A T % B R Y O -0.30 0.47 DRY, G R A Z E D % N A T V A S C 0.34 -0.05 MOIST % N A T WOODY -0.06 -0.25 U N G R A Z E D % N A T G R A M -0.08 -0.23 -% N A T HERB 0.51 0.28 MOIST, G R A Z E D % E X O V A S C 0.49 0.31 MOIST, G R A Z E D % E X O G R A M 0.61 0.25 MOIST, G R A Z E D % E X O HERB -0.05 0.23 # B R Y O 0.45 0.63 MOIST, G R A Z E D # N A T V A S C 0.55 -0.42 MOIST, U N G R A Z E D # N A T WOODY 0.05 -0.12 -# N A T G R A M -0.07 -0.42 U N G R A Z E D # N A T HERB 0.69 -0.30 MOIST, U N G R A Z E D # E X O V A S C 0.63 0.19 MOIST # E X O G R A M 0.50 0.06 MOIST # E X O HERB 0.59 0.29 MOIST, G R A Z E D In correspondence with overall native herb species trends, S4-ranked herb species cover was highest in moist, grazed microplots (Table 3.4.8). S4-ranked herb richness was highest in moist microplots (not grazing-specific). The coverage of S4-ranked bryophyte species was highest in dry, grazed microplots, whereas the coverage and richness of S1-S3-ranked bryophyte species, and the richness of S4 bryophyte species was highest in grazed microplots (not moisture-specific). Table 3.4.8. Pearson correlations of microplot species coverage and richness (SI-S3-, and S4-ranked taxa) with the first two axes of a PCA generated from M E T A - N microplot species cover data (log-transformed). F A C T O R (1) F A C T O R (2) H A B I T A T % S4 V A S C 0.39 0.30 MOIST, G R A Z E D % S4 G R A M -0.13 -0.01 -% S4 HERB 0.46 0.31 MOIST, G R A Z E D % S4 B R Y O -0.49 0.38 DRY, G R A Z E D %Sl-3 B R Y O -0.22 0.31 G R A Z E D # S4 V A S C 0.54 -0.09 MOIST # S4 G R A M -0.15 -0.15 -# S4 HERB 0.61 -0.05 MOIST # S4 B R Y O -0.05 0.53 G R A Z E D # Sl-3 B R Y O 0.09 0.48 G R A Z E D 76 (2) Meta-igneous Rock - South Position (META-S): The first two axes of the PCA performed on vegetation in meta-igneous rock microplots with a southern geographic position explained 13.9% and 11.3% of the total variation in the data set, respectively (Table 3.4.9). The first axis correlated relatively weakly with measured environmental variables (r<0.47). As was the case for northern meta-igneous microplots, several well-known "seepage" species (e.g. Bryum miniatum, Mimulus spp.) were correlated with this axis (r=0.52 and r=0.42, respectively). The first PCA axis was therefore interpreted to represent a moisture gradient, relating to the increasing influence of seepage within microplots. Table 3.4.9. Pearson correlations of southern meta-igneous rock environmental variables with the first two PCA axes 1 (Factor 1, Factor 2) generated from META-S microplot species cover data (log-transformed). Variable Factor (1) Variable Factor (2) % Total Carbon (Plot) -0.466 Canopy Cover - S (Plot) -0.580 Horiz. Slope (Plot) -0.378 Canopy Cover - S (Micro) -0.568 C:N Ratio (Plot) -0.368 Edge Class (Plot) 0.506 % Total Nitrogen (Plot) -0.327 % Cover Bare PM (Micro) 0.581 % Cover Org./Litter (Micro) 0.362 % Cover Bare PM (Plot) 0.765 pH (Plot) 0.388 % Total Variance Expl. 11.3 % Total Variance Expl. 13.9 Also in correspondence with the M E T A - N results, the second PCA axis was most strongly correlated with variables relating to exposure and potential soil development. With increasing distance from canopy edges, microplots had more bare parent material, and lower canopy cover (Figure 3.4.11). In contrast with M E T A - N , however, edge class (i.e. closest proximity of a plot edge to a tree providing canopy cover) was more influential to microscale patterning than grazing intensity in the META-S group. Presumably, this difference is related to the longer history of grazing and land use in southern sites which degrades relationships with current grazing pressures. 77 C O N T E X T EDGE OPEN - 2 - 1 0 1 2 3 4 FACTOR(l ) Figure 3.4.11. Ordination of microplots in relation to the first two axes of a PCA generated from META-S microplot species cover data (log-transformed). Microplots are grouped by plot edge class (EDGE=0-10m, OPEN=>10m from forest edges) using 68.3% confidence ellipses. Species habitat preferences were summarized in relation to the two major environmental gradients interpreted to influence microscale vegetation patterning within southern meta-igneous rock sites (Table 3.4.10). Species were divided into assemblages according to their correlations with the aforementioned axes (PCA Axis 1 was correlated with a microhabitat moisture gradient, and PCA Axis 2 was correlated with variables relating to exposure, i.e. edge proximity, canopy cover, and bare parent material coverage). Pearson correlations of r>0.3 or r<-0.3 were used to divide species into habitat groupings. Microplot coverage and richness of life form groups (by origin) in META-S sites were correlated with the first two axes of a PCA performed on microplot species cover data (Table 3.4.11); these patterns can be related to species assemblages described in Table 3.4.10. The richness of native woody and graminoid species was highest in microplots closer to adjacent forest edges; i.e. with increased canopy coverage. As was found for M E T A - N , native herb species richness was highest in microplots characterized as being periodically moist. The coverage of exotic graminoids and exotic herbs was highest in moist, edge microplots, as was the richness of exotic graminoids. Exotic herb richness was highest in 78 moist microplots (not edge-specific). Bryophytes had higher cover in dry, open microplots, but higher richness in moist, open microplots; this contrast in coverage and richness in relation to moisture was also noted for M E T A - N . Table 3.4.10. Summary of dominant habitat preferences of species with >5% frequency in META-S microplots; NON-SPEC.=habitat preference non-specific. Species shown by life form group (W=woody, G=graminoid, H=herb, B=bryophyte), and by origin for vascular plants (N=native, E=exotic). Species with >1% average cover within this data subset are indicated with larger, bold font. Species codes are listed in Appendix 2. Note PCA Axis 2 has been reversed to correspond with the direction of the exposure/soil development gradient observed for M E T A - N (Table 3.4.6). The (-) designation for Canopy-S (south-facing canopy cover) indicates an inverse relationship with the other variables shown. PCA Axis 1 (Moisture) O P E N N O N - S P E C . E D G E D R Y N O N - S P E C . M O I S T S E L A W A L (HN) M I M U G U T ( H N ) B R Y U M I N (B) C E R A P U R (B) G R I M T R I ( B ) D I C R S C O ( B ) F E S T R U B (GN) A I R A S P P (GE) P O L Y P I L ( B ) B R O M S T E (GE) C Y N O E C H (GE) R A C O L A N (B) H O L C L A N (GE) B R O D C O R ( H N ) C L A D I N A (L) P O A S P P (GE) M O N T F O N (HN) C E R A A R V (HN) P L A N E L O (HN) L O T U S P P ( H N ) T R I F O L I G (HN) M A D I G R A (HN) A L C H S U B (HE) VICISAT (HE) C E R A G L O (HE) H E D W C I L ( B ) G E R A M O L (HE) H O M A A R E (B) H Y P O R A D (HE) ISOTSPP(B) P L A N L A N (HE) KINDSPP(B) B R Y U C A N (B) P O L Y J U N (B) B R Y U C A P (B) R A C O A F F (B) B R Y U P S E ( B ) R A C O H E T (B) C E P H D I V ( B ) C L A D O N I A (L) DIDYV1N (B) T I M M C R A (B) R U M E A C E (HE) P S E U M E N (WN) LINABIC (HN) D A N T C A L (GN) V E R O A R V (HE) V U L P B R O (GN) H O M A P I N ( B ) A N T H O D O (GE) T O R T R U R (B) B R O M H O R (GE) L E P T O G I U M (L) C O L U P A R (HN) R A C O E L O (B) 79 Table 3.4.11. Pearson correlations of microplot species coverage and richness (by life form group and plant origin) with the first two axes of a PCA generated from META-S microplot species cover data (log-transformed). F A C T O R (1) F A C T O R (2) H A B I T A T % B R Y O -0.29 -0.29 DRY, OPEN % N A T V A S C 0.11 0.08 -% N A T W O O D Y 0.00 0.01 -% N A T G R A M -0.07 0.01 -% N A T HERB 0.23 0.11 -% E X O V A S C 0.36 -0.36 MOIST, E D G E % E X O G R A M 0.32 -0.31 MOIST, E D G E % E X O HERB 0.31 -0.34 MOIST, E D G E # B R Y O 0.54 0.34 MOIST, OPEN # NAT. V A S C 0.65 -0.37 MOIST, E D G E # N A T W O O D Y -0.09 -0.40 E D G E # N A T G R A M 0.04 -0.40 E D G E # N A T HERB 0.69 -0.24 MOIST # E X O V A S C 0.74 -0.23 MOIST # E X O G R A M 0.31 -0.40 MOIST, E D G E # E X O HERB 0.77 -0.08 MOIST In correspondence with M E T A - N trends, S4-ranked herb species richness was highest in moist microplots within META-S sites (Table 3.4.12). As was also found in M E T A - N , S4-ranked bryophytes had higher coverage in dry microplots. Richness of S4 bryophytes was highest in microplots away from canopy edges, whereas SI-S3 bryophyte richness was highest in moist microplots. Table 3.4.12. Pearson correlations of microplot species coverage and richness (SI-S3-, and S4-ranked taxa) with the first two axes of a PCA generated from META-S microplot species cover data (log-transformed). F A C T O R (1) F A C T O R (2) H A B I T A T % S4 V A S C 0.23 0.18 -% S4 G R A M 0.08 -0.01 -% S4 HERB 0.23 0.18 -% S4 B R Y O -0.36 0.22 DRY %Sl-3 BRYO 0.13 0.17 -# S4 V A S C 0.67 0.06 MOIST # S4 G R A M 0.03 0.01 -# S4 HERB 0.67 0.06 MOIST # S4 B R Y O -0.21 0.39 OPEN # Sl-3 B R Y O 0.47 0.18 MOIST 80 (3) Sedimentary Rock - Moderate Grazing (SED): The first two axes of the PCA performed on vegetation in moderately-grazed sedimentary rock, microplots explained 9.2% and 8.6% of the total variation in the data set, respectively (Table 3.4.13). The relatively low percent of variance explained by the PCA axes (as compared to the M E T A categories) suggests increased complexity of microscale vegetation-habitat relationships. It is likely that sandstone and conglomerate rock have contrasting fine-scale relationships, resulting in less pronounced patterns when they are grouped together. Unfortunately, the data collected for this study were insufficient to support classifying vegetation of the two rock types separately. Table 3.4.13. Pearson correlations of sedimentary rock environmental variables with the first two PCA axes 1 (Factor 1, Factor 2), generated from SED microplot species cover data (log-transformed). Variable Factor (1) Variable Factor (2) Std. Dev. Soil Depth (Plot) -0.500 % Bare PM (Plot) 0.415 Max. Soil Depth (Plot) -0.497 Vertical Slope (Plot) 0.462 Max. Soil Depth (Micro) -0.480 Total Slope (Plot) 0.490 , Std. Dev. Soil Depth (Micro) -0.459 pH (Plot) 0.546 Micro Vert. Slope (Deg) 0.451 % Total Variance Expl. 8.6 % Total Variance Expl. 9.2 The first PCA axis correlated relatively weakly with measured environmental variables (r<0.50), reflecting a gradient of increasing slope and decreasing soil depth. Overall, PCA axis 1 was associated with a separation of conglomerate rock and sandstone microplots, and was therefore interpreted to represent the strength of rock type preference among constituent species (Figure 3.4.12). As was found with the M E T A categories, the first PCA axis was also correlated with the coverage of seepage species (Philonotis fontana r=0.45, Bryum miniatum r=0.37); this corresponded with the lower frequency of seepage habitats observed in conglomerate rock sites (relating to its more rapid drainage). The second PCA axis correlated most strongly with plot pH and (as was found in M E T A categories) variables relating to exposure and soil development (i.e. plots that were more basic also had steeper slopes and higher percent coverage of bare parent material). 81 - 2 - 1 0 1 F A C T O R( 1) ROCK TYPE o CONG x SAND Figure 3.4.12. Ordination of microplots in relation to the first two axes of a PCA generated from SED microplot species cover data (log-transformed). Microplots are grouped by rock type (CONG=conglomerate, SAND=sandstone) using 68.3% confidence ellipses. Species habitat preferences were summarized in relation to the two major environmental gradients interpreted to influence microscale vegetation patterning within sedimentary rock sites (Table 3.4.14). Species were divided into assemblages according to their correlations with the aforementioned axes (PCA Axis 1 was correlated with rock preference, and PCA Axis 2 was correlated with plot pH and other variables relating to exposure). Pearson correlations of r>0.3 or r<-0.3 were used to divide species into habitat groupings. 82 Table 3.4.14. Summary of dominant habitat preferences of species with >5% frequency in SED microplots; NON-SPEC.=habitat preference non-specific. Species shown by life form group (W=woody, G=graminoid, H=herb, B=bryophyte), and by origin for vascular plants (N=native, E=exotic). Species with >1% average cover within this data subset are indicated with larger, bold font. Species codes are listed in Appendix 2. PCA Axis 1 (Rock Type, Moisture) B A S I C N O N - S P E C . w C O N G N O N - S P E C . S A N D P L E C C O N (HN) V U L P B R O ( G E ) R A C O E L O (B) G E R A M O L (HE) T R I F W I L L (HN) M Y O S D I S (HE) T R I P P U S ( H N ) P L A N L A N (HE) A L C H S U B (HE) T R I F D U B (HE) E R O D C I C (HE) H O M A A R E (B) S I L E G A L (HE) T O R T R U R (B) DIDYSPP(B) L E P T O G I U M (L) C Y T I S C O ( W E ) P S E U M E N (WN) A R B M E N (WN) B R O M S T E ( G E ) D A N T C A L (GN) B R O M H O R (GE) C E R A A R V (HN) F E S T R U B (GN) H O L C L A N (GE) G A L I A P A (HN) A I R A S P P (GE) B R O D C O R ( H N ) L O M A U T R (HN) B R O M T E C (GE) M A D I G R A ( H N ) S A N I C R A (HN) C Y N O E C H (GE) M O N T F O N (HN) V I C I S A T ( H E ) V U L P M Y U (GE) M O N T P A R (HN) B R Y U C A N (B) A L L I A C U (HN) P H L O G R A (HN) R H Y T T R I ( B ) C O L L I P A R (HN) S A X I I N T ( H N ) L O T U S P P (HN) S E L A V V A L (HN) LUPIBIC (HN) T R I F O L I G (HN) P L A N E L O (HN) T R I F W O R (HN) P O L Y G O N U M (HU) T E E S N U D (HE) H Y P O R A D (HE) B R Y U C A P ( B ) R U M E A C E (HE) B R Y U M I N (B) V E R O A R V (HE) B R Y U M U E ( B ) A U L A P A L (B) C A M P F R A ( B ) B R A C A L B (B) G R I M S P P ( B ) B R Y U P S E (B) P H I L F O N (B) C E P H D I V ( B ) P L E U S U B ( B ) C E R A P U R (B) P O L Y J U N (B) G R I M T R I ( B ) POLYPIL (B) H O M A P I N (B) R A C O V A R ( B ) H Y P N S U B ( B ) R A C O A C I (B) R A C O A F F (B) R A C O L A N (B) C L A D I N A (L) C A R E I N O (GN) C A R E B R E (GN) A N T H O D O ( G E ) A U L A A N D ( B ) P O A P R A T (GE) D I C R S C O ( B ) C A M A S P P (HN) C L A D O N I A (L) L U Z U M U L (HN) K I N D O R E (B) 83 Microplot coverage and richness of life form groups (by origin) in SED sites were correlated with the first two axes of a PCA performed on microplot species cover data (Table 3.4.15); these patterns can be related to species assemblages described in Table 3.4.14. The coverage and richness of native graminoid species was highest in acidic, conglomerate microplots, whereas the coverage and richness of native herb species was highest in basic, sandstone microplots. Recalling the higher frequency of seepage habitats in sandstone rock sites (i.e. compared to conglomerate rock), this finding corresponds with the higher native herb species richness found in periodically moist M E T A - N and META-S microplots. The richness of native woody plant species was also higher in sandstone microplots, as was the coverage and richness of bryophytes; it is recalled that bryophyte richness was also higher in moist microplots for both M E T A categories. Exotic coverage was higher in conglomerate rock microplots for all vascular plant life form groups (particularly in basic conglomerate microplots for exotic herbs), whereas the species richness of exotic graminoids and herbs was higher in basic microplots (irrespective of rock type). Table 3.4.15. Pearson correlations of microplot species coverage and richness (by life form group and plant origin) with the first two axes of a PCA generated from SED microplot species cover data (log-transformed). F A C T O R ( l ) FACTOR(2) H A B I T A T % B R Y O 0.42 -0.18 SAND % N A T V A S C 0.17 -0.09 -% N A T WOODY 0.01 -0.16 -% N A T G R A M -0.31 -0.41 C O N G , ACIDIC % N A T HERB 0.47 0.25 SAND, BASIC % E X O V A S C -0.44 0.15 C O N G % E X O W O O D Y -0.33 -0.06 C O N G % E X O G R A M -0.34 0.04 C O N G % E X O HERB -0.29 0.41 C O N G , BASIC # B R Y O 0.53 0.01 SAND # N A T V A S C 0.31 0.18 SAND # N A T W O O D Y 0.30 -0.13 SAND # N A T G R A M -0.27 -0.37 C O N G , ACIDIC # N A T HERB 0.38 0.35 SAND, BASIC # E X O V A S C -0.17 0.72 BASIC # E X O W O O D Y -0.21 0.01 -# E X O G R A M 0.02 0.36 BASIC # E X O HERB -0.19 0.74 BASIC 84 In correspondence with trends for native graminoids as a group, the coverage and richness of S4-ranked graminoids was higher in conglomerate microplots (Table 3.4.6). The coverage of S4-ranked herbs was, however, higher in sandstone microplots; as was the trend for native herbs, the richness of S4 herbs was higher in basic, sandstone microplots. The higher S4-ranked herb species richness in sandstone microplots corresponds with the higher S4-ranked herb species richness found in moist microplots within M E T A - N and META-S sites (sandstone rock had a higher incidence of seepage habitat). S4-ranked bryophytes had higher cover in acidic microplots (irrespective of rock type), and higher richness in acidic, sandstone microplots. In contrast, Sl-S3-ranked bryophytes had higher richness in basic microplots. Table 3.4.16. Pearson correlations of microplot species coverage and richness (SI-S3-, and S4-ranked taxa) with the first two axes of a PCA generated from SED microplot species cover data (log-transformed). F A C T O R ( l ) FACTOR(2) H A B I T A T % S4 V A S C 0.18 -0.16 -% S4 G R A M -0.41 -0.43 C O N G , ACIDIC % S4 HERB 0.56 0.18 SAND % S4 B R Y O -0.12 -0.44 ACIDIC %Sl-3 B R Y O 0.15 0.21 -# S4 V A S C 0.39 0.15 SAND # S4 G R A M -0.39 -0.41 C O N G , ACIDIC # S4 HERB 0.52 0.27 SAND, BASIC # S4 B R Y O 0.32 -0.34 SAND, ACIDIC # Sl-3 B R Y O 0.00 0.28 BASIC 3.4.3 Synthesis of Multi-Scale Vegetation Patterns Fine-scale vegetation-habitat relationships were found to differ among landscape categories, owing to the unique effects associated with site geology and latitude, although there were some similar trends overall. The first gradient identified for each landscape category could be related to microplot moisture (Figure 3.4.13). Within each category, the richness of native herbs (overall, and S4 species) and the richness of bryophyte species was higher in microplots where seepage species had higher coverage. For M E T A sites, this gradient was correlated positively with increased coverage and richness of exotic taxa, and correlated negatively with bryophyte coverage. This indicates that in M E T A sites, in response to the observed moisture gradient, (a) the increased coverage 85 and richness of exotic vascular plants did not preclude the presence of native herb and bryophyte species, although (b) the coverage of bryophytes was lower where exotic vascular plants were more prominent. In SED sites, the opposite trend was noted perhaps indicating more competitive relationships within this category (i.e. where there was increased overall soil development). Native herb richness (overall, and S4 species) and bryophyte richness were higher where exotic vascular plants had lower coverage and richness. In addition, in SED sites, the coverage and richness of bryophytes followed similar trends (i.e. both were highest within seepage habitats). The second gradient identified for each landscape category could be related to gradients in exposure and potential soil development (Figure 3.4.14). In M E T A - N sites, the second PCA axis was most strongly correlated with grazing intensity and other variables associated with exposure and soil development (increasing plot slope, increasing southern aspect, increasing distance from adjacent canopy, decreasing maximum soil depth, and decreased variation in soil depth). In META-S sites, the second PCA axis was similarly correlated with exposure and soil development variables, i.e. percent coverage of bare parent material, and percent canopy cover (south-facing). In SED sites, the second PCA axis correlated most strongly with plot pH, and (as was found in M E T A categories), variables relating to exposure and soil development, i.e. plot slope and percent coverage of bare parent material. Within each category, bryophyte coverage and richness (and within M E T A - N and SED landscapes, the richness of SI-S3 bryophytes) were highest in microplots that showed the greatest exposure, and lowest potential for soil development. Conversely, the richness of particular native vascular plant life form groups (particularly graminoids) was correlated negatively with this gradient. 86 Moisture (Fine Scale) u GO -ea -J a S p. ii Q '3 M E T A - N M E T A - S D R Y MOIST -Inc. %Bryo -Inc. % Exo Vase (Gram) -Inc. # Exo Vase (Gram, Herb) -Inc. % , # Nat Vase (Herb) -Inc. % S4 Bryo • w -Inc. % , # S4 Vase (Herb) -Inc. # Bryo D R Y MOIST -Inc. % Bryo -Inc. % S4 Bryo -Inc. % Exo Vase (Gram. Herb) -Inc. # Exo Vase (Gram. Herb) w -Inc. # Nat. Vase (Herb) -Inc. # S4 Vase (Herb) -Inc. # Bryo. # S1-3 Bryo S E D D R Y MOIST -Inc. % Exo Vase. # Exo Vase -Inc. # Nat Herb -Inc. # S4 Herb -Inc. %. # Nat Gram -Inc % Exo Vase (Woody. Gram. Herb) -Inc. %. # S4 Gram C O N G vs. S A N D ^ -Inc. % Nat Herb -Inc. # Nat Vase (Woody. Herb) -Inc. % S4 Herb, # S4 Vase (Herb) -Inc. %, # Bryo -Inc. # S4 Bryo - J Figure 3.4.13. Summary of major fine-scale environmental gradients and vegetation properties across landscape categories (META-N , META-S , SED): correlations of PCA Axis 1 (moisture gradient) with percent coverage (%) and richness (#) of life form groups (Woody=Woody species, Gram=Graminoid, Bryo=Bryophyte) in microplots, by rarity status and species origin (Exo=Exotic, Nat=Native). Similar trends across categories are shown in bold font. Soil Development (Large Scale) -Inc. %, # Exo Vase -Inc. # Nat Herb -Inc. # S4 Herb META-N META-S SED C5 s s £ a > a > a c -Inc. % Exo Vase (Gram) -Inc. # Exo Herb -Inc. % Nat Herb -Inc. % S4 Vase (S4 Herb) -Inc % Bryo, # Bryo -Inc. % S4 Bryo, # S4 Bryo -Inc. % Sl-3 Bryo, # Sl-3 Bryo -Inc. % Bryo, # Bryo -Inc. # S4 Bryo -Inc. % Nat Woody -Inc. # Nat Vase (Gram, Herb) -Inc. % Exo Herb -Inc. # Exo Vase (Gram, Herb) -Inc. % Nat Herb, # Nat Herb -Inc. # S4 Herb -Inc. # Sl-3 Bryo I c i S3 -Inc. % Exo Vase (Gram, Herb) -Inc. # Exo Gram -Inc. # Nat Vase (Woody) -Inc. % Nat Gram, # Nat Gram -Inc. % S4 Gram, #S4 Gram -Inc. % S4 Bryo, # S4 Bryo Figure 3.4.14. Summary of major fine-scale environmental gradients and vegetation properties across landscape categories (META-N , META-S, and SED): correlations of PCA Axis 2 (soil and exposure gradient) with percent coverage (%) and richness (#) of life form groups (Woody=Woody species, Gram=Graminoid, Bryo=Bryophyte) in microplots, by rarity status and species origin (Exo=Exotic, Nat=Native). Similar trends across categories are shown in bold font. 3.5 Rock Outcrop Ecosystem Dynamics & Conservation Implications Studies of rock outcrops have played a valuable role in investigating pathways and mechanisms of vegetation development owing to their relatively rapid dynamics, their lower structural complexity compared to forest ecosystems, and their insular nature (e.g. Whitehouse 1933, Oosting & Anderson 1939, Connell & Slatyer 1977, Shure & Ragsdale 1977, Noble & Slatyer 1980, De Angelis et al. 1986, Pickett et al. 1987a,b, Houle & Phillips 1989b, Houle 1990, Bowden 1991, McCook 1994, Fastie 1995, Ott et al. 1996, Wyatt 1997, Mitchell & Arthur 1998). In previous sections I have discussed vegetation-habitat relationships at various scales for different plant life form groups (by origin and rarity ranking), as well as for individual species. In this section, ecosystem dynamics are discussed using vegetation pattern data to address processes important for perpetuating constituent species (i.e. in relation to source vs. sink theory and island biogeography theory), as well as their associated microhabitats (i.e. in relation to succession, and the role of disturbance). 3.5.1 Rock Outcrop Ecosystem - Vegetation Species Richness in Perspective Within a forested landscape, the presence of rock outcrops can have important influences on local species composition and diversity (Arsenault & Bradfield 1995), especially for bryophytes and lichens (Lesica et al. 1991, Ohlson et al. 1997, Vitt and Belland 1997). In her study of the mosses of the San Juan Islands, Harpel (1997) noted that rock substrates alone support 71.4% of the total moss flora. Rock outcrop ecosystems in this study had higher richness of native herb species, as well as native moss species (Figure 3.5.1) per unit area than regenerating and mature forests (5 years to over 90 years old) in CDF landscapes (Sadler 2004). Rock outcrops also support a greater number of S4-ranked herbs and Sl-S3-ranked mosses per unit area than forested CDF habitats (Figure 3.5.2), indicating that they are logical targets for conservation efforts. 89 (a) (b) Figure 3.5.1a,b. Average number of (a) native herb species and (b) native moss species per 3000m , in relation to CDF zone habitat type (RO=rock outcrop, CC=clearcut forest <5 years old, YNG=young forest 20-30 years old, MAT=mature forest 50-60 years old, and OLD=old forest >90 years old). Standard error bars shown for rock outcrop ecosystems (forested habitats represent the total area studied). C ^ ^ HERB MOSS A G E A G E (a) (b) Figure 3.5.2a,b. Average number of Sl-S3-ranked and S4-ranked species per 3000m2, for (a) herbs, and (b) herbs and moss, in relation to CDF zone habitat type (RO=rock outcrop, CC=clearcut forest <5 years old, YNG=young forest 20-30 years old, MAT=mature forest 50-60 years old, and OLD=old forest >90 years old). Standard error bars shown for rock outcrop ecosystems (forested habitats represent the total area studied). 90 3.5.2 Rock Outcrop Ecosystems and Source vs. Sink Theory Since rock outcrops are associated with higher numbers of native herb and moss species (including uncommon and rare taxa), the relatively unproductive nature of such ecosystems (i.e. from a forestry perspective) may make them particularly well suited for integral conservation in forested areas (as suggested by Gagnon 1985). The high levels of biodiversity associated with outcrops may allow them to be identified as concrete attributes of forest remnants that both conservationists and forest managers can use to sustain species diversity within fragmented CDF landscapes without compromising silvicultural objectives. The issue central to conservation involves determining i f rock outcrop ecosystems are self-perpetuating, or if they are dependent on the surrounding habitats for long-term integrity. Connell (1978) noted that areas of high biodiversity may not have adequate populations of all species to be totally self-perpetuating. "Source and sink" theory (Pulliam 1988) contends that the conservation of high biodiversity areas will not always satisfy overall conservation objectives. The insular nature of rock outcrops, and the mosaic of fine-scale vegetation patterns apparent (e.g. within landscape categories) suggests that the proximity of adequate "source" populations may be important. Sources are areas that contribute disproportionately large quantities of recruits to future generations, while sinks receive recruits but contribute little (Roberts 1998). Species unique to rock outcrops may depend on other outcrop communities as "sources" during community development. However, the majority of species found on rock outcrops may be incidental within the ecosystem, rather than restricted to it. For example, Escudero (1996) found that the highest frequency plants in rock outcrop ecosystems were not rock specialists. This leads to the speculation that habitats surrounding outcrops may present important "sources" for many species. Approximately 25% of native plants identified from plot-sampled rock outcrop ecosystem sites were also identified from regenerating (5-55 year-old) and old (>90 years) CDF forests (Table 3.5.1). 91 Table 3.5.1. Total numbers of native species identified from rock outcrops, by life form group, and corresponding numbers (and percent) shared with sampled regenerating and mature CDF zone forests (from Sadler 2004). Rock Outcrops Shared Percent Vascular Plants 157 31 19.7% Woody 23 15 65.2% Herbs 108 9 8.3% Graminoids 26 7 26.9% Bryophytes 97 32 33.0% All Plant Species 254 63 24.8% It was predicted that the number of native species shared with forested habitats would be higher near forest edges and/or with the influence of canopy cover, indicating that rock outcrop ecosystems represent a "sink" habitat for a portion of the constituent taxa. There was a weak, negative, relationship between edge proximity and "shared" native species richness for vascular plants (r=-0.38) and bryophytes (r=-0.23) when plots with moderate grazing intensity were compared (Figure 3.5.3a,b). As such, edge proximity may be important for the perpetuation of a small proportion of rock outcrop taxa (woody plants and bryophytes in particular, as indicated by the percentage of "shared" species in Table 3.5.1). O OCO ! i . i L .0.51 1 1 1 0.4' ! ! 1 0.5 1.0 1.5 2.0 0.5 1.0 1.5 2.0 Log Distance to Canopy (m) Log Distance to Canopy (m) (a) (b) Figure 3.5.3a,b. Relationships between log number of rock outcrop species shared with CDF forested habitats and log distance to canopy (m), for (a) vascular plants, and (b) bryophytes. 92 1 Although the majority (75%) of rock outcrop ecosystem taxa were exclusive to that habitat type, suggesting these ecosystems represent a "source" rather than a "sink" for the majority of taxa, the strength of this interpretation is complicated by other factors. Exclusivity to rock outcrop ecosystems does not necessarily imply ambivalence to contextual habitat influences (e.g. proximity to adjacent forests). For example, disruption of the habitat surrounding an outcrop (e.g. logging surrounding areas, increased trampling or grazing pressures, introduction of non-native, highly invasive plants) may alter outcrop conditions (e.g. disturbance regimes, organic inputs, light levels, humidity, seepage flow), ultimately reducing or eliminating the microhabitats required for the perpetuation of constituent taxa. A l l of these changes, particularly alterations in seepage flow (shown to be important to diversity patterns for all rock outcrop categories in Section 3.4) would result in lower species richness within the ecosystem. 3.5.3 Rock Outcrop Ecosystems and Island Biogeography Theory If the majority of species in rock outcrop ecosystems depend on propagules generated internally, or on immigration from similar habitats, the size and isolation of rock outcrop ecosystem units may be important for the perpetuation of constituent taxa, and the level of species richness observed. The biota of rock outcrops has been interpreted using island biogeography theory; the "island" concept of outcrops stimulated by observing their biotic distinctiveness from their immediate surroundings (Main 1997). An application of classic island biogeography theory (MacArthur & Wilson 1967) might lead one to predict the richness of native "rock exclusive" species on outcrops to be a function of their size and isolation from each other; i.e. a larger island will support more species than a smaller one, and will have higher immigration rates and lower extinction rates. In the present study, rock outcrop ecosystem "islands" were studied as discrete units (represented as polygons) within the CDF biogeoclimatic zone landscape. The fact that these ecosystem "islands" occur on continental islands (i.e. the Gulf Islands of southwestern BC) compounds their insularity. Applying classical island biogeography theory, it was predicted that larger islands, and larger polygons, would have greater taxonomic richness (particularly for species exclusive to rock outcrop ecosystems) when sampling effort was constant. 93 The richness of outcrop-exclusive native vascular plant species had a weak negative relationship with island size and polygon size (r=-0.25, r=-0.26; Figure 3.5.4a,b), whereas the richness of native terrestrial bryophytes showed a weak positive relationship with island size and polygon size (r=0.32, r=0.32; Figure 3.5.5a,b). This result was likely influenced by the relationship between island size and grazing (i.e. ungrazed islands were significantly smaller than grazed islands;/?=0.020). Previous results (Section 3.3) have shown that there was higher bryophyte species richness in grazed habitats, and higher native vascular plant richness in ungrazed habitats. The two ungrazed islands used in this study were small, uninhabited, barren, windswept islets, whereas the larger Gulf Islands had a longer history of human habitation and grazing. When only moderately-grazed sites were examined, there was virtually no correlation between island size and richness for any life form group, or overall. 1 2 3 4 5 0.0 0.5 1.0 1.5 2.0 Log Island Size (ha) Log Avg. Polygon Size (ha) (a) (b) Figure 3.5.4a,b. Relationship between log number of native rock outcrop "exclusive" vascular plant species and (a) log island size (ha), and (b) log polygon size. 94 I 2 3 4 5 0.0 0.5 1.0 1.5 2.0 Log Island Size (ha) Log Avg. Polygon Size (ha) (a) (b) Figure 3.5.5a,b. Relationship between log number of native rock outcrop "exclusive" bryophyte species and (a) log island size (ha), and (b) log polygon size. Within islands and rock outcrop ecosystems, there may be complex interactions of factors influencing dispersal and establishment of species. Each island and polygon is unique in relation to other islands, major source pools, and placement in terms of dispersal routes (e.g. bird migration pathways, ocean currents, trade winds). Therefore, any single geographic isolation distance will imply a different degree of ecological isolation for each island and polygon. Islands also differ in their complement of habitats and in the outcome of species interactions, factors which may diminish the correlation between the frequency of immigration and isolation (Connor & Simberloff 1978). Owing to the weak richness:area relationships observed, and the complexity in determining isolation status for islands and polygons (particularly considering the intermittent and non-distinct nature of many rock outcrop ecosystem "boundaries"), the effects of ecosystem isolation were not examined here. The weakness of the relationships between island/polygon size and the number of native "outcrop-exclusive" plant species suggest that vascular plant and bryophyte richness and rarity levels were more related to site-specific landscape factors (i.e. rock type, geographic position, and grazing regime). Since bryophytes are characterized by high dispersal abilities, it is likely that the diversity/quality of constituent microhabitat conditions associated with site-specific landscape factors more commonly restricts moss and liverwort 95 establishment than propagule availability. As has been discussed, current and historical processes also exhibit non-uniform influences on the floristic compositions observed. Furthermore, the interpretive success of the MacArthur-Wilson model as applied to rock outcrop vegetation may be limited, owing to the fact that various biotic and abiotic factors act to maintain these ecosystems in a state of early succession. Biological equilibrium within the system (which the model implies) is unlikely to be achieved other than in very small patches, and only after long periods of time, because of varying forms of perpetual, stochastic environmental and biotic influences. The longer the period prior to attaining equilibrium, the more likely it is that an intermediate to high magnitude event will divert the system from the equilibrium, such that it is "permanently" non-equilibrial (Bush & Whittaker 1993). 3.5.4 Succession and Disturbance in Rock Outcrop Ecosystems In rock outcrop ecosystems, the most advanced successional stage is another transient stage of community development, not a stable climax stage (Farrell 1991). As Connell (1978) noted, frequent disturbances and/or gradual climatic changes can maintain diversity in non-equilibrial ecosystems by preventing the elimination of inferior competitors; in absence of disturbance, it is predicted there will be a progression toward a lower-diversity equilibrium community. As such, in attempting to preserve and perpetuate rock outcrop ecosystems and their constituent taxa, it is important to consider what biotic and abiotic factors act individually (or in combination) at various scales to influence vegetation development, ultimately holding each system in a state of early primary succession. Within rock outcrops, community development can be cyclic, owing to perpetual disturbance, which affects the diversity and quality of constituent microhabitat conditions. Disturbance events may be internally or externally driven. For example, at the fine scale, developing vegetation assemblages may reach some critical mass where they can no longer be supported (e.g. on steeper slopes; Escudero 1996). Alternatively (and/or concurrently), developing assemblages may be dislodged by an external disturbance event. In such cases, the process of colonization and successional development will begin again on the denuded underlying rock (Mitchell & Arthur 1998). Porembski and Barthlott (1997) suggested that stochastic disturbance events are of particular importance to the outcrop ephemeral 96 community, which must re-establish every year. As such, habitats in rock outcrop ecosystems may be patchy and temporary for many species and assemblages. Disturbance events that are prominent within CDF zone rock outcrop ecosystems can be related to grazing intensity (discussed in Section 3.3), and various abiotic factors (e.g. prevailing climatic weather patterns). The majority of rock outcrop ecosystems in the CDF zone are south-west facing, positioned to receive maximum radiation, and in the path of prevailing wind patterns, suggesting that abiotic factors have a strong influence on these ecosystems. In this regard, there are two major sources of peak winds that occur annually along the BC coast: Arctic Outbreaks and Pacific Lows. Arctic Outbreaks occur when high pressure systems with cold arctic air move down over the BC interior and spill out onto the coast through coastal inlets. The Fraser Valley, Howe Sound, and Bute Inlet are well-known for strong, cold, northeasterly outflow winds, which move across the Strait of Georgia and occasionally travel as far as Vancouver Island (Mitchell 1998). In this study, rock outcrop ecosystems on islands closer to the mainland (e.g. Bowen Island, situated at the mouth of Howe Sound) appear to be influenced by Arctic Outflow winds. Pacific Lows are low pressure systems which move in off the Pacific Ocean (Mitchell 1998). Well-known local Pacific Low patterns include the "Qualicum Winds", which likely affect northern Gulf Islands (Lange 2005), and the "Gap Winds of Juan de Fuca", which likely affect southeastern Gulf Islands (Yeaton 2004, Bach 2006). These wind patterns, in addition to local wind effects associated with the multitude of passes, narrows, and channels which separate Gulf Island land masses, likely have a strong influence on the development of CDF rock outcrop vegetation in different parts of the zone; i.e. wind intensity is increased wherever it is channeled through narrow topography. Previous research suggests that successional development on rock outcrop habitats is often limited as a result of environmental constraints imposed by a limited capacity for soil development (Collins et al. 1989), and that the stages of succession found on an outcrop will be influenced by its configuration and microtopography, e.g. the presence of crevices and depressions in the rock surface (Oosting & Anderson 1937). The importance of soil development in vegetation composition was observed at the large scale, i.e. increasing soil development among landscape categories (Figure 3.4.4, Figure 3.4.5), corresponding with decreasing bare parent material, and increasing plant cover. The prominent species within 97 each landscape category (>1% average cover in microplots within each data subset) reflected an overall successional trend in relation to increasing soil development (Table 3.5.2). Within each landscape category, the primary gradient influencing fine-scale vegetation patterning was the presence of seepage habitat. Also within each landscape category, the second gradient influencing fine-scale vegetation patterning could be related to variables reflecting soil development (maximum soil depth, percent coverage of bare parent material, and/or pH), and exposure (vulnerability to disturbance, i.e. slope, aspect, canopy coverage, and/or distance from canopy edges). Assemblages identified in Section 3.4.2 for M E T A - N (Table 3.4.6), META-S (Table 3.4.10), and SED (Table 3.4.14) can therefore be used to interpret fine-scale successional trends within each landscape category (i.e. in relation to the second PCA axis, in each case). Table 3.5.2. Species with >1% average cover in microplots within each landscape category identified in Section 3.4 (META-N=meta-igneous rock >49°N, META-S=meta-igneous rock <49°N, SED=sedimentary rock); by life form groups (GRAM=graminoid, HERB=herb, BRYO=bryophyte, LCHN=lichen); origins (N=native, E=exotic). G R A M H E R B B R Y O L C H N M E T A - N M E T A - S SED DANTSPP (N) AIRASPP (E) AIRASPP (E) FESTRUB (N) A N T H O D O (E) A N T H O D O (E) FESTRUB (N) BROMHOR (E) H O L C L A N (E) B R O M S T E (E) POASPP (E) C Y N O E C H (E) H O L C L A N (E) V U L P B R O (E) V U L P M Y U (E) H Y P O R A D (E) B R O D C O R (N) BRODCOR (N) S E L A W A L (N) H Y P O R A D (E) H Y P O R A D (E) R U M E A C E (E) S E L A W A L (N) S E L A W A L (N) HEDWCIL DICRSCO A U L A P A L POLYJUN POLYJUN DICRSCO POLYPIL POLYPIL R A C O E L O R A C O E L O R A C O E L O R A C O L A N R A C O L A N R A C O L A N C L A D I N A C L A D I N A C L A D I N A 98 In summary, the particular combination of stress and disturbance regimes (i.e. grazing, and geographic position in relation to prevailing wind patterns and other climatic effects) in combination with local and fine-scale landscape features affecting soil development (i.e. slope, aspect, canopy cover, edge proximity, microtopography) is complex, and unique to each site. The cyclical nature of disturbance and regeneration results in a collection of communities in different states of continuous succession (Ott et al. 1996). Therefore, although a rock outcrop ecosystem may be viewed as a relatively homogeneous landscape unit, there may be much spatial and temporal heterogeneity, pointing to the importance of disturbance and patch dynamics in overall ecosystem function and landscape pattern. The apparent stability of the ecosystem as a whole may be perpetuated by disturbances that operate at different spatial scales and with varying frequency (Connell 1978), resulting in different microhabitats, successional trajectories and constituent species. As such, ecosystem responses to changes in disturbance regimes such as alterations in climatic patterns (e.g. in association with global warming), or grazing exclusion, will be largely site-dependent. Owing to the scarcity of bryophyte surveys in Garry oak and associated rock outcrop ecosystems within the study area, few bryophyte taxa have been characterized as species at risk. As such, recovery strategies have focused predominantly on vascular plants; i.e. the coverage and richness of rare and uncommon herbs (particularly bulbous taxa such as Camassia spp.), and the presence of Garry oak seedlings. Previous results (Section 3.3) showed that native vascular plant richness was highest in ungrazed sites within rock outcrop ecosystems of the CDF zone. This observation would support the recommendation put forward by the Garry Oak Ecosystems Recovery Team (2005), i.e. that livestock grazing is "likely to result in destruction of critical habitat" for species in Garry oak and associated ecosystems. However, it is notable that the richness of bryophytes, and the richness of rare taxa (almost entirely represented by bryophytes) was highest in intensely grazed sites when all meta-igneous rock sites were pooled. Similarly, within M E T A - N and META-S landscape categories, bryophyte richness was highest in microplots that had higher levels of exposure (i.e. in relation to grazing, or edge proximity, respectively), in contrast with trends in native 99 vascular plant species richness within the same landscape categories. The promotion of one group of taxa may therefore be at the expense of another; i.e. the complete exclusion of grazers from an area may facilitate soil development and the establishment of graminoids, reducing the available habitat for less competitive taxa such as bryophytes, particularly in M E T A - N sites. Consequently, devising a unilateral management strategy to accommodate high species richness for all life form groups may not be possible. Conservation objectives (i.e. life form groups and/or species of concern) must be carefully considered before implementing any strategy that would influence or alter disturbance patterns within rock outcrop ecosystems. 100 CHAPTER IV CONCLUSIONS A persistent goal of ecological research has been to investigate the effects of biodiversity, and to understand the mechanisms that cause communities to change (e.g. Hooper et. al 2004, Kahmen et al. 2005). Understanding patterns of diversity and their causes remains a challenge in ecology because of the immense effort required to collect and analyze the relevant data, and the inherent complexity of causes and effects within ecosystems. Pieces of the puzzle exist on different temporal and spatial scales; the challenge has been to develop an approach to community ecology that allows us to conceptualize these connections as unambiguously as possible. The deterministic approach to describing vegetation patterns (i.e. that communities are classifiable, and exist as discrete vegetation units or associations) has been criticized as being less likely to reveal forces operating to determine distribution and abundance of plants (Harper 1982). Conversely, the reductionist approach (i.e. that communities can be understood via experimental and mechanistic descriptions of interactions among species), has been criticized because it is impossible to identify all the mechanisms that are responsible for the persistence and dynamics of a community. Since processes occurring at larger spatial and temporal scales influence local vegetation patterns, the information that can be gained from small-scale experimental studies will be limited. Explanations of why a community maintains a certain species diversity must go beyond the scale of local manipulations. The most favorable approach, as has been attempted in this thesis, is to expand the concept of community ecology to include processes occurring on larger spatial and temporal scales. This thesis has shown that by using a multi-scale approach to studying rock outcrop ecosystems, vegetation patterns at different scales can be interpreted and integrated. Furthermore, this approach allows for the reconciliation of several prevailing theories in ecology (Figure 4.0.1). Vegetation patterns can be viewed as deterministic at larger scales, e.g. allowing for the discrimination of distinct landscape categories as described in this study. At finer scales, vegetation patterns can be viewed as individualistic, e.g. resulting from 101 species interactions and individual niche tolerances occurring at the population level, as shown by microplot gradient relationships described in this study. Landscape Ecology - regional patterns - y-diversity - immigration, speciation, ecological equivalency e.g. Clements (1916), MacArthur-Wilson (1967) Mechanisms: e.g. climate, geography, geology, history L A R G E S C A L E (Holistic description, determinism) Community Ecology Each - across-community patterns - (3-diversity - species interactions, spatial and temporal heterogeneity e.g. Grime (1979), Tilman (1982), Gleason (1926), Hanski (1982), Connell(1978) Mechanisms: e.g. dispersal, competition, facilitation Mediated by change (i.e. d stresses relev; Population Ecology - within-community patterns - a-diversity - species requirements, niche relations e.g. Harper (1981), Grime (1979), Grubb (1977) Mechanisms: e.g. genotype, tolerance, phenology, life form, breeding systems scale: environmental isturbance), and nt to that scale FINE S C A L E (Mechanistic description, reduction ism) Figure 4.0.1. A multi-scale approach to the study of vegetation in rock outcrop ecosystems permits the integration and reconciliation of prevailing theories in community ecology. The research described here provides a context for future conservation and research in sensitive rock outcrop ecosystems. Within large-scale landscape categories, constituent taxa are associated with unique fine-scale gradients (i.e. in relation to moisture, and soil development), which are governed by different types of disturbances, operating at different scales. As such, the implementation of management strategies, and mechanistic descriptions 102 of vegetation-habitat relationships will be most relevant within the context of these categories. Characterization of fine-scale habitat relationships within the context of a classification scheme that is hierarchical at larger scales also improved knowledge of the niche requirements of individual rare species (e.g. rare taxa), and provides further context for conservation. In conclusion, large-scale patterns of species richness in rock outcrop ecosystems were linked to gradients in land use history, climate, geology, and current grazing intensity. At finer scales, richness patterns could be related to factors that permitted different life form groups to exist (i.e. a variety of niches, such as microplots that were exposed vs. protected, in relation to soil development). As noted by Harper (1981), there is a distinction between conserving diversity and conserving species. Rules for maintaining species diversity may be quite different from those appropriate for safeguarding particular species, and may defy generalization. Furthermore, our concept of what is rare will depend on the scale of our experience and the range and/or narrowness of our interests. Bryophytes comprise the largest portion of rare taxa in rock outcrop ecosystems examined in this study. The under-representation of bryophytes in surveys of the study area have resulted in fewer of them being characterized as species at risk. As such, recovery strategies have focused predominantly on vascular plants. However, results presented in this thesis have shown that patterns of coverage, richness, and rarity vary among life form groups, and that these relationships are unique to landscape categories, e.g. rare bryophyte species richness and native vascular plant species richness were associated with opposite microhabitats in relation to exposure and reduced soil development. Therefore, devising a unilateral management strategy to accommodate high species richness for all life form groups may not be possible. Conservation objectives (i.e. life form groups and/or species of primary concern) must be carefully considered with respect to variables influencing vegetation at each scale (as observed in this study) in order to promote the success of management strategies implemented within rock outcrop ecosystems of the CDF zone. 103 LITERATURE CITED Aho, K. & T. Weaver (2006). Measuring water relations and pH of cryptogam rock-surface environments. The Bryologist, 109(3): 348-357. Alpert, P. (1985). Distribution quantified by microtopography in an assemblage of saxicolous mosses. Vegetatio, 64: 131-139. Arsenault, A. , and G.E. Bradfield (1995). Structural-compositional variation in three age-classes of temperate rainforests in southern coastal British Columbia. Canadian Journal of Botany, 73: 54-64. Bach, A. (2006). Ozette Research Program website: www.ac.wwu.edu/~andybach/Ozette/index.html. Western Washington University, USA. BC Conservation Data Centre (1996). Provincial Tracking List for Vascular Plants. BC Ministry of Environment, Lands and Parks, Victoria. BC Conservation Data Centre (2006). Species Summary. BC Ministry of Environment. Available: http://srmapps.gov.bc.ca/apps/eswp/ (accessed Oct 13, 2006). BC Ministry of Environment (1978). The soil landscapes of British Columbia (ed. K.W.G Valentine). Resource Analysis Branch, BC Ministry of Environment, Victoria. BC Ministry of Environment (1985). Soils of southern Vancouver Island (ed. J.R. Jungen). Surveys and Resource Mapping Branch, BC Ministry of Environment, Victoria. Technical Report No. 17. Bedward, M . (1995). Simple models of pattern and process. Proceedings of the Linnean Society of New South Wales, 115: 17-23. Belland, R.J. (1998). The rare mosses of Canada. Committee on the Status of Endangered Wildlife in Canada, Canadian Wildlife Service, Environment Canada, Ottawa, Ont. Belland, R.J. (1997). Status Report on the Apple Moss Bartramia stricta Brid. in Canada. Rep. submitted to the Committee on the Status of Endangered Wildlife in Canada, Ottawa. Belland, R. J. (2005). A multivariate study of moss distributions in relation to environment in the Gulf of St. Lawrence region, Canada. Canadian Journal of Botany, 83(3):243-263. Bowden, R.D. (1991). Inputs, outputs, and accumulation of nitrogen in an early successional moss (Polytrichum) ecosystem. Ecological Monographs, 6(2): 207-223. 104 Bowe, L . M . & D.A. Rayner (1993). A floristic survey of bryophytes of four boulder outcrop communities in South Carolina. Castanea, 58(1): 10-17. Brown, D.H. (1982). Mineral nutrition, p. 383-444. In: Smith, A.J . (ed.). Bryophyte ecology. Chapman and Hall, New York. Burbanck, M.P. & Piatt, R.B. (1964). Granite outcrop communities of the Piedmont Plateau in Georgia. Ecology, 45: 292-306. Bush, M.B. & Whittaker, R.J. (1993). Non-equilibration in island theory of Krakatau. Journal of Biogeography, 20: 453-458. Carleton, T.J. (1990). Variation in terricolous bryophyte and macrolichen vegetation along primary gradients in Canadian boreal forests. Journal of Vegetation Science, 1: 585-594. Clements F.E. (1916) Plant Succession. Carnegie Institute Publication 242. Carnegie Institute, Washington, D.C. Coates, F. & J.B. Kirkpatrick (1992). Environmental relations and ecological responses of some higher plant species on rock cliffs in northern Tasmania. Australian Journal of Ecology, 17:441-449. Collins, S.L., G.S. Mitchell & S.C. Klahr (1989). Vegetation-environment relationships in a rock outcrop community in Southern Oklahoma. American Midland Naturalist, 122: 339-348. Connell, J.H. (1978). Diversity in tropical rain forests and coastal reefs. Science, 199: 1302-1310. Connell, J.H. & R.O. Slatyer (1977). Mechanisms of succession in natural communities and their role in community stability and organization. American Naturalist 111: 1119-1144. Connor, E.F. & D. Simberloff (1978). Species number and compositional similarity of the Galapagos flora and Avifauna. Ecological Monographs 48: 219-248. Cox, J.E. & D.W. Larson (1993). Spatial heterogeneity of vegetation and environmental factors on talus slopes of the Niagara Escarpment. Canadian Journal of Botany, 71: 323-332. Crins, W. (1997). Rare and endangered plants and their habitats in Canada. Canadian Field-Naturalist, 111(3): 506-519. Daubenmire, R. (1968). Plant Communities. A Textbook of Plant Synecology. Harper Publishing, New York. De Angelis, D.L. W.M. Post & C.C. Travis (1986). Positive feedback in natural systems. Springer-Verlag, Berlin. 105 Environment Canada (2000). The Species at Risk Act (SARA) : a guide. Ottawa. Environment Canada (2006). Canadian Climate Normals or Averages 1961-1990. National Climate Data and Information Archive: www.climate.weatheroffice.ec.gc.ca. Escudero, A. (1996). Community patterns on exposed cliffs in a Mediterranean calcareous mountain. Vegetatio, 125: 99-110. Farrell, T .M. (1991). Models and mechanisms of succession: an example from a rocky intertidal community. Ecological Monographs, 61: 95-113. Fastie, C.L. (1995). Causes and ecosystem consequences of multiple pathways of primary succession at Glacier Bay, Alaska. Ecology, 76: 1899-1916. Frisvoll, A . A . (1983). A taxonomic revision of the Racomitrium canescens group (Bryophyta, Grimmiales). Gunneria 41: 1-181. Frisvoll, A . A . (1988). A taxonomic revision of the Racomitrium heterostichum group (Bryophyta, Grimmiales) in N . and C. America, N . Africa, Europe and Asia. Gunneria 59: 1-289. Fuchs, M . A . (2001). Towards a Recovery Strategy for Garry Oak and Associated Ecosystems in Canada: Ecological Assessment and Literature Review. Technical Report GBEI/EC-00-030. Environment Canada, Canadian Wildlife Service, Pacific and Yukon Region. Gagnon, D. (1985). Forest vegetation of west-central Vancouver Island, British Columbia. Ph.D. Thesis, University of British Columbia. Gagnon, D. & G.E. Bradfield (1987). Gradient analysis of west central Vancouver Island forests. Canadian Journal of Botany, 65: 822-833. Garry Oak Ecosystems Recovery Team (2005). National multi-species recovery strategy for species at risk in maritime meadows associated with Garry oak ecosystems. Recovery of Nationally Endangered Wildlife Committee, Ottawa, ON. Gleason, H.A. (1926) The individualistic concept of the plant association. Bulletin of the Torrey Botanical Club, 53:7-26. Godfrey, J.D. (1977). The hepaticae and anthocerotae of southwest British Columbia. Ph.D. Thesis, University of British Columbia. Greven, H.C. (2003). Grimmias of the world. Leiden, Backhuys Publishers. Grime, J.P. (1979). Plant strategies and vegetation processes. John Wiley & Sons, Chichester. 106 Grubb, P.J. (1977). The maintenance of species-richness in plant communities: the importance of the regeneration niche. Biological Reviews, 52: 107-145. Hanski, I. (1982). Dynamics of regional distribution: the core and satellite species hypothesis. Oikos, 38: 210-221. Harpel, J. (1997). The phytogeography and ecology of mosses within the San Juan Islands, Washington State. Ph.D. Thesis, UBC. Harper, J.L. (1981). The meanings of rarity. In: The biological aspects of rare plant conservation (Ed. by H. Synge). John Wiley & Sons, New York, pp. 189-203. Harper, J.L. (1982). After description. In: The plant community as a working mechanism (Ed. by E.I. Newman). Blackwell Scientific Publications, Oxford, pp. 11-25. Harrison, S., J.H. Viers & J.F. Quinn (2000). Climatic and spatial patterns of diversity in the serpentine plants of California. Diversity and Distributions, 6: 153-161. Hastings, R.I. & H.C. Greven (2006). Grimmia. Bryophyte Flora of North America, Provisional Publication (Version 5). Missouri Botanical Garden, USA. Hatch, D.A., J.W. Bartolome, J.S. Fehmi & D.S. Hillyard (1999). Effects of burning and grazing on a coastal California grassland. Restoration Ecology 7:376-381. Hedderson T.A. & G.R. Brassard (1990). Microhabitat relationships of five co-occurring saxicolous mosses on cliffs and scree slopes in eastern Newfoundland Canada. Holarctic Ecology, 13(2): 134-142. Hierro, J.L., D. Villarreal, O. Eren, J .M. Graham & R.M. Callaway (2006). Disturbance facilitates invasion: The effects are stronger abroad than at home. American Naturalist, 168(2): 144-156. Hitchcock, C. L. & Cronquist, A. (1973). Flora of the Pacific Northwest. University of Washington Press, USA. Hooper, D.U., F.S. Chapin, J.J. Ewel, A . Hector, P. Inchausti, S. Lavorel, J.H. Lawton, D.M. Lodge, M . Loreau, S. Naeem, B. Schmid, H. Setala, A.J . Symstad, J. Vandermeer & D.A. Wardle (2005). Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecological Monographs, 75(1): 3-35. Hopper, S.D., A.P. Brown, & N.G. Marchant (1997). Plants of Western Australian granite outcrops. Journal of the Royal Society of Western Australia, 80(3): 141-158. 107 Houle G. (1990). Species-area relationship during primary succession in granite outcrop plant communities. American Journal of Botany, 77 (11): 1433-1439 Houle, G. & D.L. Phillips (1989a). Seasonal variation and annual fluctuation in granite outcrop plant communities. Vegetatio, 80: 25-35. Houle, G. & D.L. Phillips (1989b). Seed availability and biotic interactions in granite outcrop communities. Ecology, 70(5): 1307-1316. Jonsgard, B. & H.J. Birks (1993). Quantitative studies on saxicolous bryophyte-environment relationships in western Norway. Journal of Bryology, 17: 579-611. Kahmen, A. , J. Perner, V . Andorff, W. Weisser, N . Buchmann (2005). Effects of plant diversity, community composition, and environmental parameters on productivity in montane European grasslands. Oecologia, 142:606-615. Kenkel, N.C. (2006). On selecting an appropriate multivariate analysis. Canadian Journal of Plant Science, 86(3): 663-676. Kotanen P .M. (2004). Revegetation following soil disturbance and invasion in a California meadow: a 10-year history of recovery. Biological Invasions 6(2): 245-254. Krajina, V . J. (1965). Biogeoclimatic zones and biogeocoenoses of British Columbia. Ecology of Western North America, 1: 1-17. Lange, O.L. (2005). Wind patterns in the Georgia Basin - the Salish Sea. Proceedings of the 2005 Puget Sound Georgia Basin Research Conference (March 29-31, 2005). Seattle, Washington, USA. Lawton, E. (1971). Moss flora of the Pacific Northwest. The Hattori Botanical Laboratory, Japan. Lesica, P., B. McCune, S.V. Cooper & W.S. Hong (1991). Differences in lichens and bryophyte communities between old-growth and managed second-growth forests in the Swan Valley, Montana. Canadian Journal of Botany, 69: 1745-1755. MacArthur, R.H. & E.O. Wilson (1967). The Theory of Island Biogeography. Princeton University Press, New Jersey. MacDougall, A . (2002). Invasive Perennial Grasses in Quercus garryana meadows of southwestern British Columbia: Prospects for Restoration. USDA Forest Service Gen. Tech. Rep. PSW-GTR-184. MacDougall, A.S., J. Boucher, R. Turkington & G.E. Bradfield (2006). Patterns of plant invasion along an environmental stress gradient. Journal of Vegetation Science, 17: 47-56. 108 MacDougall, A.S. & R. Turkington (2004). Relative importance of suppression-based and tolerance-based competition in an invaded oak savanna. Journal of Ecology 92(3): 422-434. MacDougall, A.S. & R. Turkington (2006). Dispersal, competition, and shifting patterns of diversity in a degraded oak savanna. Ecology, 87(2): 1831-1843. Main, B .Y. (1997). Granite outcrops: A collective ecosystem. Journal of the Royal Society of Western Australia, 80(3): 113-122. McCook, L.J. (1994). Understanding ecological community succession: causal models and theories, a review. Vegetatio, 110: 115-147. McCune, B. & J.B. Grace (2002). Analysis of ecological communities. M j M Software Design, Oregon, USA. Mcintosh, T.T. (1989). Bryophyte records from the semiarid steppe of northwestern North America, including four species new to North America. The Bryologist, 92: 356-362. McPhee, M . , P. Ward, J. Kirkby, L. Wolfe, N . Page, K. Dunster, N .K. Dawe, & I. Nykwist (2000). Sensitive Ecosystems Inventory: East Vancouver Island and Gulf Islands (1993-1997). Volume 2: Conservation Manual. Technical Report Series No. 345, Canadian Wildlife Service, Pacific and Yukon Region, BC. Mead, R. (1988). The design of experiments: statistical principles and practical applications. Cambridge University Press, Cambridge, UK. Mitchell, S.J. (1998). Review of January 1, 1997 windthrow in MacMillan Park. Technical Report prepared for BC Ministry of Environment, Lands and Parks, BC. Mitchell, G. & W. Arthur (1998). Population interactions in primary succession: an example of contramensalism involving rock-colonizing bryophytes. Lindbergia, 23: 81-85. Monroe, J.S. & W. Wicander (2001). Physical geology: exploring the earth. 4 t h ed. Brooks/Cole, USA. Munoz, J. (1998). Materials toward a revision of Grimmia (Musci; Grimmiaceae): nomenclature and taxonomy of Grimmia longirostris. Annals of the Missouri Botanical Garden, 85: 352-363. Nakatsubo T. (1997). The role of bryophytes in terrestrial ecosystems with special reference to forests and volcanic deserts. [Japanese] Japanese Journal of Ecology (Tokyo) 47(1): 43-54. Noble, I.R. & R.O. Slatyer (1980). The use of vital attributes to predict successional changes in plant communities subject to recurrent disturbances. Vegetatio, 43: 5-21. 109 Nuszdorfer, F.C., K. Klinka & D.A. Demarchi (1991). Chapter 5: Coastal Douglas-fir Zone, in Ecosystems of B.C., D. Meidinger & J. Pojar (Eds). B.C. Ministry of Forests, Victoria. Nyholm, E. (1998). Illustrated Flora of Nordic Mosses. Fasc. 4. - Nord. Bryol. Soc, Copenhagen and Lund. Ohlson, M . , L. Soderstrom, G. Hornber, O. Zackrisson, and J. Hermansson (1997). Habitat qualities versus long-term continuity as determinants of biodiversity in boreal old-growth swamp forests. Biological Conservation, 81: 221 -231. Oosting, H.J. & L.E. Anderson (1937). The vegetation of a barefaced cliff in western North Carolina. Ecology, 18: 280-292. Oosting, H.J. & L.E. Anderson (1939). Plant succession on granite rock in eastern North Carolina. Botanical Gazette, 100: 750-768. Ott, S., U . Elders & H . M . Jahns (1996). Vegetation of the rock-alvar of Gotland: I. Microhabitats and succession. Nova Hedwigia, 63(3-4): 433-470. Pentecost, A . (1980). The lichens and bryophytes of rhyolite and pumice-tuff rock outcrops in Snowdonia, and some factors affecting their distribution. Journal of Ecology, 68: 251-267. Peterson, D.L. & Parker, V.T., Eds. (1998) Ecological Scale: Theory and Applications. Columbia University Press, New York. Pharo, E.J., J.B. Kirkpatrick, L. Gilfedder, L. Mendel., & P . A . M . Turner (2005). Predicting bryophyte diversity in grassland and eucalypt-dominated remnants in subhumid Tasmania. Journal of Biogeography, 32(11): 2015-2024 Pickett, S.T.A., S.L. Collins & J.J. Armesto (1987a). Models, mechanisms and pathways of succession. Botanical Review, 53: 335-371. Pickett, S.T.A., S.L. Collins & J.J. Armesto (1987b). A hierarchical consideration of causes and mechanisms of succession. Vegetatio, 69: 109-114. Poore, M.E.D. (1962). The method of successive approximation in descriptive ecology. Advances in Ecological Research, 1: 35-68. Porembski, S.S.R & W. Barthlott (1997). Inselberg vegetation and the biodiversity of granite outcrops. Journal of the Royal Society of Western Australia, 80(3): 193-199. Proctor, M.C.F. (1990). The physiological basis of bryophyte production. Botanical Jounal of the Linnean Society, 104: 61-77. 110 Pueyo. Y. , C.L. Alados & O. Barrantes (2006). Determinants of land degradation and fragmentation in semiarid vegetation at landscape scale. Biodiversity and Conservation, 15(3): 939-956. Pulliam, H.R. (1988). Sources, sinks, and population regulation. The American Naturalist, 132: 652-661. Roberts, C M . (1998). Sources, sinks, and the design of marine reserve networks. Fisheries, 23(7): 16-19. Sadler, K.D. (2004). Vegetation monitoring in Coastal Douglas-fir Zone forests of Vancouver Island: influence of age class and edge proximity on vascular plant and bryophyte distributions. Technical report prepared for Weyerhaeuser Ltd., Nanaimo BC. Saenz, L. & J.O. Sawyer (1986). Grasslands as compared to adjacent Quercus garryana woodland understories exposed to different grazing regimes. Madrono, 33: 40-46. Schofield, W.B. (1976). Bryophytes of British Columbia III: habitat and distributional information for selected mosses. Syesis, 9: 341-354. Schofield, W.B. (1992). Some common mosses of British Columbia: Royal British Columbia Museum Handbook. Royal British Columbia Museum, Victoria BC. Seabloom, E.W., J.W. Williams, D. Slayback, D .M. Stoms, J.H. Viers, & A.P. Dobson (2006). Human impacts, plant invasion, and imperiled species in California. Ecological Applications, 16: 1338-1350. Shure, D.J. & H.L. Ragsdale (1977). Patterns of primary succession on granite outcrop surfaces. Ecology, 58: 993-1006. Smith, A.J.E. (1978). The moss flora of Britain and Ireland. Cambridge University Press, Great Britain. Spence, J.R. (1988). Bryum Hedw. (Bryaceae) in Western North America. The Bryologist, 91(2): 73-85. Stohlgren, T.J., K . A . Bull & Y . Otsuki (1998). Comparison of rangeland vegetation sampling techniques in the Central Grasslands. Journal of Range Management, 51: 164-172. Syed, K. (1973). A taxonomic study of Bryum capillare Hedw. and related species. Journal of Bryology, 7: 265-326. Tilman, D. (1982). Resource competition and community structure. Monographs in Population Biology, Princeton University Press. I l l Turner, M.G. (1989). Landscape ecology: the effect of pattern on process. Annual Review of Ecology and Systematics, 20: 171-197. Vanderpoorten, A. & P. Engels (2003). Patterns of bryophyte diversity and rarity at a regional scale. Biodiversity and Conservation, 12(3): 545-553. Vitt, D. & R. Belland (1997). Attributes of rarity among Alberta mosses: Patterns and predictions of diversity. The Bryologist, 100(1): 1-12. Ward, P., G. Radcliffe, J. Kirkby, J. Illingworth & C. Cadrin (1998). Sensitive ecosystems inventory: east Vancouver Island and Gulf Islands 1993-1997. Technical Report Series Number 320, Canadian Wildlife Service, Pacific and Yukon Region. Ware, S. (1990). Adaptation to substrate - and lack of it - in rock outcrop plants: Sedum and Arenaria. American Journal of Botany, 77(8): 1095-1100. Weiher, E. (2003). Species richness along multiple gradients: testing a general multivariate model in oak savannas. Oikos, 101(2): 311-316. Whitehouse, E. (1933). Plant succession on central Texas granite. Ecology, 14(4): 391-405. Wyatt, R. (1997). Reproductive ecology of granite outcrop plants from the south-eastern United States. Journal of the Royal Society of Western Australia, 80(3): 123-129. Yeaton, B. (2004). Gap winds of Juan de Fuca. The Weather Notebook (National Radio Show, Oct. 21 2004 episode). Produced by the Mount Washington Observatory, USA. Zander, R.H. (2006). Didymodon. Bryophyte Flora of North America, Provisional Publication (Version 2b). Missouri Botanical Garden, USA Zander, R.H. (1999). A new species of Didymodon (Bryopsida) in western North America and a regional key to the taxa. The Bryologist, 102: 112-115. Zechmeister, H.G., A. Tribsch, D. Moser & T. Wrbka (2002). Distribution of endangered bryophytes in Austrian agricultural landscapes. Biological Conservation, 103: 173-182. 112 APPENDIX 1. Specialized taxonomic keys for the study area: (a) Brachytheciaceae, (b) Bryum (c) Didymodon, (d) Grimmia and Schistidium, and (e) Racomitrium. (a) Key to Brachytheciaceae in rock outcrop and shallow grassland habitats of the CDF Zone: adapted from Lawton (1971) and Schofield (1992). la. Leaves deeply plicate, linear cells extending nearly to the base, the basal cells narrow, 5-8(10)u wide...7 (Homalothecium group) lb. Leaves not deeply plicate, if so the basal cells not narrow or linear to the base.. .2 2a. Alar cells differentiated, dense, quadrate to rounded or transversely elongate, the walls thick; leaves coarsely serrate in the upper third; seta smooth.... 12 {Isothecium) 2b. Alar cells not differentiated, or i f differentiated and quadrate, the walls scarcely thickened... 3 3a. Branch leaves strongly serrate, the teeth extending to the base or nearly so; stem leaves triangular and with long acuminate point; operculum rostrate, seta rough...4 (Kindbergia) 3b. Branch leaves not strongly serrate and the stem leaves not broadly triangular; operculum not rostrate...5 4a. Plants regularly pinnate, with coarse branches (main stem simple or with few divisions), usually terrestrial, large; branch leaves 1-2 x 0.6-1.2mm, usually >lmm long.. .Kindbergia oregana 4b. Plants multipinnate, with slender and/or irregular branches, often in moist sites; branch leaves 0.6-1.1 x 0.3-0.64mm, usually <lmm long...Kindbergiapraelonga 5a. Plants branching +/- regularly, the branches in one plane, julaceous; leaves deeply concave, not plicate, ovate to ovate-oblong, apex abruptly acuminate; stem leaves 1.8-2.8 x 1.2-1.8mm, costa to mid-leaf or shorter.. .Pseudoscleropodium purum 5b. Plants not branching regularly, or if so, with one or more of the following characteristics: leaves not concave, leaves plicate, leaf-shape other than ovate to ovate-oblong, apex not abruptly acuminate, stem leaves <2mm long, or costa to mid-leaf or longer... 6 6a. Branches commonly irregular and julaceous; leaves neither falcate nor plicate, usually distinctly concave, to about 2mm long; margins entire or slightly serrulate at apex; dioicous. ..13 (Scleropodiunt) 6b. Plants variable, the leaves straight or falcate, smooth or plicate; rarely julaceous, if julaceous either plicate or autoicous or both, to >2mm long; leaves usually serrate, at least at apex... 15 (Brachythecium) 7a. Leaves mainly entire, but dentate at base and often at apex, the teeth sharp, usually recurved; usually epiphytic, also on cliffs, frequently on perpendicular surfaces; main shoot affixed to substratum, thus branches curl.. .Homalothecium nuttallii 113 7b. Leaves entire at base or with a few inconspicuous teeth.. .8 8a. Plants large and coarse, regularly pinnate, the branches in one plane, branches l-3cm long; alar and basal cells elongate (25-35 x 8-1 Ou), the walls thick and pitted; on litter in coniferous forests... Trachybryum megaptilum 8b. Plants smaller, not regularly pinnate or i f so, the branches not in one plane or less than lcm long; alar cells quadrate, rounded, or irregular, usually short, walls usually not pitted, the other basal cells often with pitted walls.. .9 9a. Alar cells clear, quadrate, stem leaves with 10 or more quadrate alar cells on the entire margin, the quadrate cells not extending higher within the margin; capsule curved and zygomorphic, operculum conic, cilia well-developed; plants regularly pinnate, branches about 5mm long; on rock, or soil in rocky places; not firmly attached to substratum, sometimes the apical portion of the plant (including its branches) coils upward when dry.. .Homalothecium pinnatifldum 9b. Alar cells not clear, quadrate to rounded or irregular, usually less than 10 cells on the entire or serrulate margin, about 5-6, the differentiated alar cells usually extending higher within (i.e. inside?) the margin; on sandy soil or rock... 10 10a. Four to six rows of basal cells similar to alar cells, short, broad, about 17x8u, walls thick and pitted; branch leaves often papillose near apex; stem leaves 1.3-2mm long; on sandy soil or rock.. .Homalothecium arenarium 10b. Basal leaf cells not similar to alar cells; stem leaves often more than 2mm long' on trees, rotten logs, rock, rarely on soil... 11 11a. Plants coarse, irregularly branched, predominantly epiphytic (especially broad-leafed maple), sometimes on rock; branch leaves 2.5-3mm long, with narrow points; sporangina somewhat curved, elongate (4-5:1), urn 2.4-2.8mm long; operculum short-rostrate, 1-1.5mm long; cilia +/- rudimentary...Homalotheciumfulgescens l ib. Plants yellow green, glossy, irregularly branched, dried shoots curling upward from substratum; usually on rock or soil over rock, rarely on trees or logs; branch leaves 1.3-2mm long, always with broad points, sporangia short-cylindric (2-3:1), urn 1.8-2.6mm long; operculum conic, 0.5-0.75mm long; cilia 2-3, well-developed, sometimes +/-appendicular.. .Homalothecium aeneum (NOTE: Brachythecium albicans is superficially similar, but dried shoots never curl upward, and are generally pale yellow-green) 12a. Quadrate alar cells numerous, extending at least lA the way up the leaf margins; leaves widest near the middle; branches julaceous; plants generally rusty green, strongly worm-like when dry, marginal teeth on leaves...Isothecium cristatum * i f stem leaves >2mm, and margins revolute above, see Antitrichia californica 12b. Quadrate alar cells few; leaves widest just above the base; branches infrequently julaceous, often stoloniferous; plants glossy, pale to dark green, with sharply toothed leaf-apices that are not narrowly pointed, obvious midrib...Isothecium stoloniferum 114 13a. Plants yellow or golden to pale green, branches strongly julaceous, tumid; leaf apex commonly obtuse to rounded, sometimes with a very short straight apiculus; alar cells inflated; plants +/- robust, in mats or tufts; costa without spines at the end; leaves imbricate, 1.4-1.6 x 0.8-1.3mm; annulus wanting...Sclewpodium obtusifolium 13b. Leaves without inflated alar cells, with spines at the end of the costa, with a recurved apiculus, or leaves less than -0.7mm wide..14 14a. Plants on soil, usually +/- robust; branches strongly julaceous and tumid to somewhat flagelliform; leaves relatively broad, on well developed branches 1.2-2 x 0.7-1.2mm, often smaller on attenuate branches; basal leaf cells +/- rectangular in two rows, rarely 3 rows across the base; capsule horizontal and somewhat zygomorphic...Sclewpodium tourettii (branches plainly julaceous, leaves deeply concave, apiculus recurved, areas that are periodically dry = var. tourettii; leaves and branches variable, apiculus rarely recurved, wetter places = var. colpophyllum) 14b. Plants epiphytic or on rock, smaller, the leaves narrower; branches julaceous but not tumid or turgid; well-developed branch leaves 0.8-1.8 x 0.3-0.8mm, usually less than 0.7mm wide; basal leaf cells quadrate to short-rectangular in 2-6 rows; capsule erect to somewhat inclined...Sclewpodium cespitans (seta rough throughout = var. cespitans; seta smooth or nearly so = var. sublaeve) 15a. Costa extending to the leaf apex or nearly so; seta rough throughout; leaves strongly decurrent; stem leaves 0.8-1.2mm long; rare in W N A . . .Brachythecium reflexum 15b. Plants otherwise; costa shorter, not extending to leaf apex.. .16 16a. Stem leaves plainly plicate wet or dry, branch leaves usually plicate.. .17 16b. Leaves not or only slightly plicate; leaves concave... 19 17a. Leaves strongly falcate; plants small, the stem leaves 1-1.8mm long; branch leaves with one or more spines at end of costa; mtoicous.. .Brachythecium velutinum (seta rough throughout, on soil or rotten wood = var. velutinum; seta smooth, or slightly rough at base, on soil or soil over rock = var. venustum) 17b. Leaves not or only slightly falcate, and/or stem leaves >2mm long... 18 18a. Leaves with long slender apices, entire or nearly so; branches sometimes julaceous; on rock, sandy soil, or grassy places, often in periodically dry places; dioicous.. .Brachythecium albicans 18b. Stem leaves usually +/- deltoid-ovate, sometimes with auricles, usually a row of +/-inflated cells across the base attaching the leaf to the stem; plants in wide loose mats, often stoloniferous; on wet soil, rotten logs, or tree trunks; dioicous.. .Brachythecium asperrimum * i f stem leaves <1.5mm long or branch leaves <lmm long, see Kindbergia praelonga (4b) 19a. Plants small, with concave imbricate leaves and julaceous branches; leaves abruptly acuminate, often somewhat secund; stem leaves 0.7-1.5mm x 0.35-0.7mm; on soil or soil over rock, 1000-3000m; autoicous; seta smooth or nearly so...Brachythecium collinum 115 19b. Plants larger, leaves less strongly concave and branches not usually julaceous; leaves never abruptly acuminate, usually +/- plicate; stem leaves 1.8-3 x 0.6-1.5mm; seta rough throughout.. .Brachythecium asperrimum * i f stem leaves <1.5mm long or branch leaves <lmm long, see Kindbergia praelonga (4b) 116 (b) Key to Bryum in rock outcrop and shallow grassland habitats of the CDF Zone: adapted from Spence (1998), Lawton (1971), Syed (1973), and Smith (1978). A. Median leaf cells 5:1 or longer, if shorter, the capsule erect and straight; leaves never bordered; costa ending before the apex to percurrent, never excurrent.. .PohiHa B. Median leaf cells 4:1 or often shorter, if longer, with at least one of the following characters: (a) leaves broadly ovate, oblong, or oblong-ovate; (b) apex obtuse to rounded; (c) costa excurrent; (d) leaf margins bordered by longer cells; (e) cilia of endostome plainly appendiculate... 1 (Bryum) la. Leaf apex hyaline; shoot silver-white when dry, julaceous, with small imbricate leaves, bulbils often present in leaf axils; costa weak, not reaching apex to percurrent.. .Bryum argenteum lb. Leaf apex rarely hyaline, shoot not or silver-white, rarely julaceous, leaves small to robust, imbricate or contorted; bulbils present or absent...2 2a. Plants lacking specialized vegetative propagules (filiform or bulbiform gemmae in leaf axils, rhizoidal tubers present, or both).. .3 2b. Plants with specialized vegetative propagules (filiform or bulbiform gemmae in leaf axils or rhizoidal tubers present, or both)... 16 3a. Leaves ovate to obovate, often strongly spirally twisted or contorted; stems more or less rosulate; leaf base reddish, cells long-rectangular (3:1+) towards base; rhizoids often with tubers...4 3b. Leaves not obovate, rarely spirally twisted when dry, stems not rosulate; leaf base green or reddish, cells long-rectangular or quadrate towards base.. .7 4a. Leaves large (3-4mm long), with short recurved hair point, often in 2 or more interrupted comal tufts along the stem; leaf border weak or lacking at apex; leaf cells with thick and pitted walls; tubers brown .. .Bryum canariense 4b. Leaves small to medium, with straight hairpoint, comal tuft usually single, leaf border strong above, of 2 or more rows; leaf cells not thick/pitted; tubers brown, red, or crimson...5 (Bryum capillare complex) *Bryum gemmascens can look similar, but leaves are never simultaneously obovate and spirally twisted when dry 5a. Leaves decurrent, hairpoint short or lacking; leaves often reddish-green, ovate to obovate; reddish leaf border; elongate stems - only weakly rosulate; tubers dark red to red-brown; plants d'\o\co\xs...Bryum erythroloma 5b. Leaves not decurrent; leaves and border greenish, hairpoint long/conspicuous; short, strongly rosulate stems; tubers brown or crimson/orange-red...6 6a. Tubers brown; dioicous.. .Bryum capillare 6b. Tubers crimson to orange-red; usually synoicous (also costa stronger, upper cells longly hexagonal and rather narrower than in Bryum capillare).. .Bryum torquescens 117 7a. Leaves ovate-lanceolate, long acuminate, with costa long-excurrent into a conspicuous hairpoint (upper leaves); leaf cells generally thin, not or only scarcely pitted; margins revolute/recurved; border unistratose; cells elongate at base.. .8 7b. Leaves with costa ending before apex, percurrent, or merely short-excurrent (upper leaves), shape variable; leaf cells thin or thick, pitted or not; margins recurved or not; margins bistratose or unistratose; cells quadrate or short-rectangular at base (if elongate then with bistratose margins or dense tomentum on stem)... 10 8a. Median leaf cells long, 6:7:1, border definite or not; plants dioicous.. .Bryum caespiticium 8b. Median leaf cells shorter, 2-3:1, leaf margins with a well-defined, unistratose border, plants to 4cm high, leaves slightly decurrent, concave, red at base; plants synoicous or autoicous...9 9a. Leaves imbricate to slightly contorted when dry; plants not in dense cushions, plants to lcm high; capsule not zygomorphic when mature, spores 12-16um; endostome processes with perforations 1:1 length to width; synoicous .. .Bryum creberrimum (=Bryum lisae) * Bryum intermedium overlaps with Bryum lisae except capsule mouth oblique, spores 15-25um; endostome processes narrowly perforated 9b. Leaves somewhat contorted when dry; plants in dense cushions, to 4cm high; capsule not zygomorphic when mature, spores 18-20um; endostome processes somewhat longer than wide, about 1.5-2:1 length to width; autoicous.. .Bryum pallescens *very close to Brym lisae, Bryum intermedium - sporophytes needed for positive ID 10a. Leaf border strong, bistratose (at least in part - check older leaves); plants with elongate stems bearing equidistant leaves; areolation lax, cells mostly thin-walled, short and broad (2-3:1), often >20um wide, gradually becoming elongate rectangular at base (3:1+); leaves ovate/oblong-lanceolate...ll 10b. Leaf border lacking to strong, unistratose; plants and leaves variable; areolation lax to dense, upper and middle cells thin walled or incrassate, usually 3-4:1 or more, generally <20um wide, becoming more or less quadrate or short-rectangular below, often abruptly so (if cells otherwise then plants with dense tomentum on stem)... 12 11a. Leaves with margins not or little recurved, leaves green (4-5mm long), hardly decurrent; autoicous...Bryum uliginosum l ib. Leaves with margins recurved nearly to the apex, leaves ovate, usually pinkish-tinged (2-3mm long), slightly decurrent; occasionally with filiform gemmae in leaf axils; dioicous, capsules <4mm long, exothecial cells short and broad, spores 19-28um.. .Bryum pallens 12a. Leaves usually contorted when dry, decurrent, narrowly ovate, not or slightly concave, upper leaves to 4mm long, apex acute; costa percurrent to short-excurrent as a short point; leaf border strong; upper leaf cells short and broad, 2-3:1, >20um, incrassate; dense tomentum usually present on stem.. .Bryum pseudotriquetrum 12b. Leaves imbricate, rarely contorted when dry, apex sometimes obtuse; costa usually not reaching apex or percurrent, rarely short-excurrent; leaf border mostly lacking; upper 118 and middle leaf cells usually elongate, 4:1 or more, generally <20um wide, becoming more or less quadrate or short-rectangular below, often abruptly so... 13 13a. Leaves red or purple-red, dull or with metallic sheen; stems short, usually <10mm long... 14 13b. Leaves small, green or brown, often with red tints but not red throughout... 15 14a. Plants tiny and budlike, stems l-5mm long; leaves often in two or more interrupted buds along stem; upper leaf often clear, hyaline hairpoint present; leaf border plane; bulbils lacking... Bryum gemmascens 14b. Stems usually <10mm; leaves small and crowded at tip in bud; upper leaves green or reddish, costa of sterile leaves not reaching apex or percurrent (upper leaves excurrent into hairpoint), apex acute; bulbils green, 200-500um long, leaf primordia about lA bulbil length.. .Bryum dichotomum 15a. Leaves concave, generally >2mm long, purple-red, occasionally green, margins plane or recurved only at base, leaf apex obtuse, often cucullate, upper and middle leaf cells 4-6:1, middle leaf cells mostly <20um wide... Bryum miniatum 15b. Leaves usually flat, usually <2mm long, dull red or red-green, margins recurved to midleaf or above, leaf apex broadly acute to obtuse, not cucullate; upper and middle leaf cells 3-4:1; middle leaf cells usually >20um wide.. .Bryum muehlenbeckii 16a. Filiform gemmae present in leaf axils; leaf apex acute to acuminate, leaves ovate, ovate-lanceolate or obovate; leaf border unistratose; leaves medium to large, arranged along an elongate stem, contorted when dry, ovate, strongly decurrent, apex denticulate; tubers lacking; dioicous or synoicous....5ry«/Mpseudotriquetrum 16b. Bulbiform gemmae present in axils or rhizoidal tubers present on rhizoids, filiform gemmae lacking; plants otherwise... 17 17a. Bulbiform gemmae present in leaf axils, rarely tubers present in rhizoids; leaves small, reddish at base, apex acute; stems short (5-15mm); bulbils green, 200-500um long, leaf primordia about yA bulbil length; costa of sterile leaves not reaching apex or percurrent.. .Bryum dichotomum 17b. Bulbiform gemmae lacking, tubers present on rhizoids; plants otherwise.. .18 18a. Plants rosulate or leaves in interrupted comose tufts along stem; leaves contorted to spirally twisted when dry, mostly obovate, often serrate at apex; upper leaf cells short and broad (3:1), becoming gradually elongate-rectangular below...4 18b. Plants with elongate stems, not rosulate or comose; leaves imbricate, not or slightly contorted when dry, ovate, ovate-lanceolate or triangular, apex smooth or denticulate; upper leaf cells elongate (usually 4:1 or more, often 6:1), becoming quadrate to short-rectangular at base, often abruptly so... 19 19a. Plants with stiff, imbricate leaves arranged along elongate stem, dark red or red-green to tips; leaves closely imbricate, somewhat concave, apex broadly acute to obtuse, costa 119 not reaching apex to percurrent, leaf base reddish; tubers occasionally in dense tomentum on stem, (upper leaf cells 3-5:1); bulbils lacking... Bryum muehlenbeckii 19b. Plants with soft somewhat twisted leaves, crowded towards apex of stem, rarely reddish tinged; costa in at least upper leaves short to long-excurrent into hairpoint; tubers common on rhizoids in soil or other substratum; tubers small (60-120um wide), rhizoids violet...Bryum violaceum 120 (c) Key to Didymodon in rock outcrop and shallow grassland habitats of the CDF Zone: adapted from Zander (2006) and Lawton (1971). l a . Stem hyalodermis present (sometimes indistinct); superficial walls of distal laminal cells (x-s) flat or very weakly convex on both sides; proximal cells often enlarged, hyaline, and thin-walled; axillary hairs usually of all cells clear; leaves distinctly papillose; usually on calcareous rock/soil.. .Bryoetythrophyllum lb. Stem hyalodermis usually absent; superficial walls of distal laminal cells (x-s) strongly convex to bulging on both sides; proximal cells usually not well differentiated; axillary hairs with a brownish proximal cell; leaves often only weakly papillose; habitat calcareous or not.. .3 3a. Leaves oblong-lanceolate or linguate, never long acuminate, apex +/- obtuse; leaf margins often revolute/recurved to above the middle but not to the apex...Barbula 3b. Leaves lanceolate to long-lanceolate, acuminate; axillary hairs with one or more brown proximal cells; basal laminal cells usually little differentiated from the distal; abaxial costal cells usually quadrate; laminal papillae absent or simple or only occasionally multiplex; gemmae composed of only 1-10 cells...4 (Didymodon) 4a. Leaf apices caducous or very fragile; leaves 2—3 mm, leaf margins broadly crenate in distal 2/3-3/'4.. .Didymodon eckeliae 4b. Leaf apices intact or only occasionally broken, or leaves less than 2mm, or leaf margins entire... 5 5a. Leaves not keeled when moist, not highly recurved, margins finely crenulate by bulging cell walls; costa thick, 6-10 cells wide above midleaf.. .Didymodon norrisii 5b. Leaves sometimes keeled or highly recurved when moist, margins usually entire or occasionally dentate or scalloped-crenate but not minutely crenulate; or costa thinner, less than 6 cells wide above midleaf.. .6 6a. Leaves deltoid to short-lanceolate or ovate, to 1.5 or rarely to 2.0mm; margins recurved or revolute to near apex, propagula sometimes present.. .7 6b. Leaves short to long-lanceolate or long-triangular, to 4.0mm; margins recurved near base or up to proximal 2/3 of leaf, propagula rare...8 7a. Costal section showing adaxial epidermal cells thin-walled, remainder of costal thick-walled; costa blunt apically, costa wider at midleaf than below, with a budging adaxial surface forming a long-elliptic unistratose pad of cells, guide cells in 2(-3) layers, leaf margins loosely revolute, tubers occasional on proximal rhizoids; gypsiferous deposits, limestone boulders, sandy soil.. .Didymodon nevadensis (note: the irregular red coloration distinguishes this from Bryoerythrophyllum species, which are evenly colored red) 7b. A l l cells of costal section about equally thickened; costa percurrent or very weakly excurrent from an obtuse or acute apex in 1 -3 cells; costa gradually narrowing distally, adaxial surface nearly flat and not forming a wide pad of cells, guide cells usually in 1 layer; leaf margins weakly recurved, lamina red in nature; arid habitats... Didymodon 121 brachyphyllus (note: small forms of Didymodon nicholsonii have the same leaf shape, but the lamina is 2-stratose; sterile Grimmia species may be confused with this species, but a small hyaline apex is commonly found on at least some leaves; small forms of Didymodon vinealis may be confused with Didymodon brachyphyllus, but the latter never has lanceolate leaves, and its perichaetial leaves are also short and rather deltoid). 8b. Leaves 1-stratose, or 2-stratose in small areas or patches marginally...9 (Didymodon vinealis) *sporophyte required for adequate ID (note: this species is often difficult to distinguish from sterile forms of Didymodon rigidulus, which has elongate cells on adaxial surface of costa near the boat-shaped leaf apex, usually strongly papillose laminal cells, and the usual presence of a distinct groove down the adaxial surface of the leaf along the costa; Bryoerythrophyllum recurvirostrum is distinguished by the clear, enlarged proximal cells) 8a. Leaves 2-stratose marginally or medially or both...10 9a. Peristome present, well developed, cells of operculum twisted (leaves lanceolate, <2.5mm long, recurved to above mid-leaf; apices little thickened or fragile, basal cells short- to long-rectangular)...Didymodon vinealis var. vinealis (Didymodon vinealis var. flaccidus, a.k.a. Didymodon insulans has been synonymized, but may be distinguished i f needed by leaves longer than 2.5mm, crisped when dry, distal margins plane) 9b. Peristome absent, cells of operculum straight or nearly so (leaves short lanceolate to triangular, apices often fragile and bistratose at least in patches, basal cells quadrate to short-rectangular)...Didymodon vinealis var. rubiginosus (syn. Didymodon occidentalis, Barbula rubiginosa) 10a. Leaves long-lanceolate, apex narrowly acute, margins evenly and broadly crenulate above leaf base, 2-stratose in 1-2 rows to near base... Didymodon eckeliae 10b. Leaves long-ovate to broadly lanceolate, seldom narrowly acute, apex usually blunt to broadly acute, margins smooth and 2-stratose marginally in 1 -several rows or rarely across leaf in distal leaf half or occasionally only in patches.. .Didymodon nicholsonii (intergrades +/- with Didymodon vinealis, but the ovate-lanceolate leaves and partially/completely 2-stratose distal lamina are distinctive) 122 (d) Key to Grimmia and Schistidium in rock outcrop and shallow grassland habitats of the CDF Zone: adapted from Hastings and Greven (2006), Greven (2003), Nyholm (1998), and Lawton (1971). la. Leaves with inner basal cells (or all leaf cells) highly sinuous, like those of the outer and medial basal cells; seta straight.. .Racomitrium *see Appendix 1(e) (note: i f leaves with a pair of lamellae or wings on the dorsal surface of the costa; leaf margin bistratose in the upper part... Grimmia ramondii) lb. Leaves with inner basal cells not sinuose; seta straight or arcuate.. .2 2a. Seta shorter or as long as the capsule, straight; stem from unbranched to strongly branched, with or without a central strand; leaves papillose or not (check dorsal costa and margins); costa fairly homogeneous in x-s, or with median guide cells separating two stereid bands; plants maritime or usually on base-rich rocks... 10 (Schistidium) 2b. Seta longer than capsule, straight or curved; stem +/- dichotomously branched, in x-s usually with a +/- well developed central strand; leaves not papillose (or if so with multistratose lamina and margins); costa without median guide cells separating two stereid bands; usually on siliceous rocks or boulders, more rarely on calcareous substrata...3 (Grimmia) 3a. Leaves linear lanceolate, strongly crisped and contorted when dry (spirally twisted around stem); without hairpoints or hairpoints short; usually multicelled gemmae on back of leaf; leaves 0.8-2.5mm long, keeled, lamina unistratose, upper cells strongly incrassate; margins partially reflexed, uni/bistratose upper; stems 0.5-4cm high; moist habitats (e.g. sheer cliffs)...Grimmia torquata 3b. Plants otherwise; leaves not strongly crisped and contorted when dry...4 4a. Leaves concave or keeled, margins plane to incurved; lamina with bistratose upper.. .5 4b. Leaves keeled, at least in upper part of vegetative leaf; margins recurved on one or both sides; upper lamina unistratose or bistratose.. .6 5a. Leaves concave, costa not prominent, ovate-lanceolate from an ovate base, 1.7-4mm long; basal juxtacostal leaf cells usually elongate (4-8:1), usually sinuose, and usually with thick lateral walls... Grimmia ovalis 5b. Leaves keeled, costa prominent, narrowly lanceolate to ovate-lanceolate, 1-1.8mm long; basal juxtacostal laminal cells quadrate to short-rectangular, straight, thick-walled. .. Grimmia alpestris 6a. Distal laminal cells 2-stratose with thick, prominent multistratose bands, margins multistratose and thick, occasionally papillose; stem leaves lanceolate to ovate-lanceolate, 2-3mm long; basal juxstacostal laminal cells short- to long-rectangular, sinuose-nodose, thick-walled; stems with central strand absent; dioicous, seta arcuate; usually >500m elev....Grimmia elatior 6b. Plants otherwise; lacking papillae and with multistratose bands/margins; basal juxtacostal marginal cells short-rectangular to linear...7 123 7a. Leaves 1-1.7mm, upper part transparent, unistratose; basal juxtacostal cells short-rectangular, thin-walled, not sinuose; upper leaves abruptly contracted into short to long, smooth to denticulate awn; autoicous, usually with capsules, arcuate seta...Grimmia pulvinata 7b. Leaves to >1.7mm (to 4mm); basal juxtacostal cells long-rectangular to linear, thick-walled, nodulose to sinuose; leaf gradually contracted into awn; dioicous or cladautoicous, seta straight or arcuate...8 8a. Medial laminal cells strongly sinuose, extremely thick lateral walls; basal juxtacostal cells elongate to linear, nodulose (inner basal cells vermicular, very long and thin lumina, with extreme thick cell walls); habit of Racomitrium heterostichum; stems with central strand absent (stems +/- crispy, sections with a snap); dioicous, seta arcuate... Grimmia ieibergii 8b. Medial laminal cells quadrate, slightly to moderately sinuose, thick-walled; basal juxtacostal cells long-rectangular to linear, nodulose to sinuose; stems with central stem present (stems +/- mushy, difficult section efficiently)...9 9a. Lamina unistratose (with only bistratose margins); leaves with one or two recurved leaf margins; dioicous, seta arcuate...Grimmia trichophylla 9b. Lamina 2-stratose; leaves with only one recurved leaf margin; cladautoicous, seta straight when wet...Grimmia longirostris 10a. Maritime plants; leaves muticous, bistratose or thicker to below middle; costa with median guide cells separating 2 stereid bands; leaves lanceolate, to 3.5mm long, margins/lamina +/- papillose in upper part, narrowly recurved (base to middle); plants in dense rigid tufts/cushions (stems 0.5-2cm high), dark green to yellow green.. .Schistidium maritimum 10b. Plants not maritime (out of reach of salt spray), or leaves not muticous and bistratose or thicker to below middle; costa without 2 stereid bands separated by median guide cells... 11 11a. Upper leaf lamina distantly to densely papillose with low or high papillae on dorsal side (smooth or slightly papillose on ventral side); plants with reddish secondary colours; upper/central lamina unistratose or with bistratose spots; central leaf cells predominantly oblong, sinuose; hair point often flexuose in pilose plants (rarely stiff and coarse); stem central strand absent or indistinct (1-6 cells in x-s); siliceous or calcareous rock. ..12 l ib. Lamina smooth (NB: upper leaf margins and dorsal part of costa may be smooth, papillose, or papillose-denticulate); upper/central lamina partly or completely bistratose; central leaf cells isodiametric, esinuose, or slightly sinuose; hair points +/- coarse and stiff, mostly +/- flattened in lower part; stem central strand distinct; calcareous rock.. .Schistidium crassipilum 12a. Leaves in distinct spiral rows, shoots mostly julaceous and slender, leaves rapidly contracted from an ovate base; lamina unistratose, dorsally densely papillose and 124 ventrally smooth with few, scattered papillae; capsule 1.1-1.3 times longer than wide.. .Schistidium strictum 12b. Leaves not in distinct spiral rows, shoots not julaceous, leaves gradually narrowed from base, often with distinct red patches; upper lamina unistratose or with bistratose patches, cells +/- densely and highly papillose; capsule 1.5-2.3 times longer than wide.. .Schistidium papillosum (also in area but less common = Schistidium vancouverense - lamina with fewer/lower papillae on dorsal lamina, cells very sinuose and incrassate; thick reddish cell walls) 125 (e) Key to Racomitrium in rock outcrop and shallow grassland habitats of the CDF Zone: adapted from Lawton (1971), Frisvoll (1983,1988), and Nyholm (1998). la. Leaves with a pair of lamellae or wings on the dorsal surface of the costa; leaf margin bistratose in the upper part; seta twisted counter-clockwise...Grimmia ramondii lb. Leaves without lamellae or wings on the dorsal surface of the costa, or i f so, then leaves with hair point...2 2a. Leaves papillose, i.e. with high or low papillae over lumens of lamina and/or hair point, or with the lower more irregular papillae on cell walls; margins unistratose; leaf apex broad or narrow.. .3 2b. Leaves not papillose but sometimes with bulging cell walls (x-s); margins unistratose or bistratose; leaf apex narrow... 9 3a. Hyaline point absent; leaves with low regular papillae on cell walls (x-s upper lamina); seta twisted clockwise throughout or (in Racomitrium fasciculare) counter-clockwise above... 4 3b. Hair point usually present; leaves with (high) papillae over lumens of lamina and/or hair point; seta twisted counter-clockwise...6 4a. Leaves +/- oblong; the apex broad and usually denticulate (and papillose-crenulate); on stones in/near streams...Racomitrium aciculare 4b. Leaves ovate-lanceolate to lanceolate; narrow towards apex, usually acuminate (and papillose-crenulate).. .5 5a. Stem usually with numerous branchlets (like Racomitrium canascens); upper leaf cells long (>3:1), papillae large (in x-s as rounded thickenings on cross walls); nerve thin, predominantly 2(-3) stratose and biconvex; usually in the mountains...Racomitrium fasciculare 5b. Stem without or with few short branches; upper leaf cells shorter, often isodiametric, papillae smaller (in x-s not as thickenings); nerve strong, predominantly 3-4 stratose, not biconvex; on wet rocks.. .Racomitrium ryszardii (formerly Racomitrium aquaticum) 6a. Cells of lamina not papillose; hyaline hair points rough with conspicuous papillae, usually erose-dentate and long-decurrent down the flat margin of the lamina; alar cells not inflated, with a row of +/- hyaline cells extending up margins.. .Racomitrium lanuginosum 6b. Cells of lamina papillose... 7 7a. Hair points decurrent and denticulate; plants with long branches; leaves 2.5-3.5mm long,, muticous and +/- obtuse, or often with upper or all leaves piliferous; leaf cells with several small papillae per cell, the papillae distinct or sometimes faint; alar cells +/-differentiated, with thinner nonsinuose walls; with a row of +/- hyaline cells extending up margins (as in Racomitrium lanuginosum)...Racomitrium varium 7b. Cells of hair points strongly papillose; plants with short, tuft-like branches; leaves 1.8-3 mm long, rarely muticous; leaf cells strongly papillose with several large, simple 126 papillae per cell; alar cells differentiated, short, hyaline or yellow, often slightly inflated, the walls thin; costa in x-s +/- ventrally convex.. .8 8a. Supra-alar marginal cells (usually) elongate and always thin-walled and not (or faintly) sinuose; hair point usually not decurrent, not or faintly denticulate and spinulose, and not/lowly papillose at base; hair point erect-flexulose when dry.. .Racomitrium ericoides 8b. Supra-alar marginal cells (usually) short and always thick-walled and sinuose; hair point usually decurrent down margin of lamina, strongly denticulate and spinulose, in lower part papillose and in upper part less papillose (rarely without papillae); hair point usually squarrose or recurved when dry.. .Racomitrium elongatum 9a. A l l leaves with chlorophyllous apex, without any trace of a hyaline or hyaline hair point... 10 9b. Some (or at least one) leaves with hair point; point long or short, hyaline or subhyaline...ll 10a. Apex usually broadly rounded and crenulate; margin shortly recurved and often nearly flat on one side, 1-stratose; alar group well-defined and sometimes auriculate, of short, thick-walled and porose cells; lowlands... Racomitrium pacificum (note: if leaves concave, >3xlmm; alar group not well defined, of large, usually thin-walled cells then see Racomitrium depressum - usually a mountainous species) 10b. Apex less broadly rounded and never crenulate; margin recurved towards the apex; margin smooth, 1- (or in part 2-) stratose in upper part; costa medium broad below, and there (3-) 4-stratose...Racomitrium affine 11a. Costa at least in part with low dorsal wings and/or furrows, strongly dorsally convex with 3-4 ventral cells; leaf apex narrow; hair point strongly spinulose, terete and not flexuose; margin uneven, 2-stratose for 1-3 cell rows, usually green and slightly branched plant; short upper cells (to 2:1)...Racomitrium occidentale l ib. Costa without such dorsal wings or furrows; margin unistratose or bistratose; the combination of the other characteristics different.. .12 12a. Leaf with basal cells esinuose, thick-walled and porose; with a differentiated basal marginal border of 10-20 usually esinuose (sometimes slightly sinuose) and hyaline or sometimes more thick-walled and pellucid cells; costa below 2- or 3- stratose, and narrow with 3-4 ventral cells (leaves <3mm).. .13 {Racomitrium microcarpon) 12b. Leaf with basal cells sinuose and usually less thick-walled (+/- porose); without such a basal marginal border; basal part of costa at least in some x-s with 5 or more ventral cells... 14 13a. Cells of the basal marginal border usually short, wide and hyaline.. .Racomitrium microcarpon f. microcarpon 13b. Basal marginal cells usually elongate, narrow and more or less sinuose and/or thick-walled... Racomitrium microcarpon f. afoninae 127 14a. Costa broadly canaliculate in mid-leaf, and there with many (4-8) ventral cells, narrower above (2-4 ventral cells); moderately dorsally convex... 15 (note: if hair point coarsely and acutely spinulose and denticulate; costa very broad above (5-8 ventral cells), lamina strongly pseudopapillose; innermost perichaetial leaves only slightly differentiated, pilose then see Racomitrium brevipes - usually a high elevation species) 14b. Costa not or less obviously canaliculate in mid-leaf, and there with few (3-4) ventral cells, strongly dorsally convex.. .16 15a. Leaf margin 2-stratose for 1-3 cell rows in its upper part, uneven; lamina distinctly narrowed at the connection with the hair point, which is stiff and not flexuose; seta short (3-4.5mm); coarse, slightly branched plant.. .Racomitrium obesum 15b. Leaf margin 1-stratose or less 2-stratose, smooth; lamina not much narrowed at the connection with the hair point, which is soft and (usually) flexuose; seta long (4-8mm); moderately robust, usually (much) branched plant...Racomitrium heterostichum 16a. Margin recurved to about l/2(-3/4) the leaf length on one side and shorter or almost flat on the other side; hair point frequently squarrose when dry (broad and long, usually 0.5-1.5mm, decurrent down margin of lamina); leaf long (>3mm); robust plant.. .Racomitrium lawtonae 16b. Margin recurved toward hair point or (especially on one side) somewhat shorter; hair point frequently not squarrose when dry... 17 17a. Leaf margin uneven, usually 2-stratose for 1-2(3) cell rows (sometimes with 1- or 3-stratose spots) in the upper part; hair point stout, not flexuose, strongly spinulose, lamina strongly contracted at the connection with the hair point; costa broad, with 4-9 ventral cells below; leaf long and broad (>3.2mm long); robust plant.. .Racomitrium obesum 17b. Leaf margin not uneven, usually 1-stratose or sporadically 2-stratose for l(-2) cell rows in the upper part; hair point soft, usually flexuose and less spinulose... 18 18a. Costa canaliculate and predominantly bistratose in its middle and lower upper part; leaves generally long-piliferous, +/- hoary-looking...Racomitrium heterostichum 18b. Costa not canaliculate and predominantly 3-stratose in its middle and lower upper part; leaves generally shorter-piliferous, +/- greenish.. .Racomitrium affine 128 APPENDIX 2. List of species identified at study sites, arranged by life form group: woody species (trees and shrubs), graminoids (grasses, rushes, and sedges), herbs, and bryophytes (mosses and liverworts). Also shown are species code names, as well as provincial rarity status (Yellow-, Red-, or Blue-Listed) (BC CDC 2006), and for native plants, provincial (S) conservation status rankings (see Appendix 3 for definitions). The average percent cover (Cover %) and overall site frequency in plot-sampled sites (Freq %) are also shown (inventory-only sites excluded). Values for species with uncertain status, ranking, cover, or frequency (owing to incomplete identification or presence in inventory-only sites) denoted with Species Prov Cover Freq Woody Plant Species Code Status Rank (%) (%) Abies grandis (Dougl. ex D. Don) Lindl. Abiegra Yellow S4 0.00 7.7 Acer macrophyllum Pursh Acermac Yellow S5 0.00 30.8 Amelanchier alnifolia Nutt. Amelaln Yellow S4S5 0.02 15.4 Arbutus menziesii Pursh Arbmen Yellow S5 0.00 53.8 Arctostaphylos uva-ursi (L.) Spreng. Arctuva Yellow S5 0.47 7.7 Crataegus douglasii Lindl. Cratdou Yellow S4 0.00 7.7 Cytisus scoparius (L.) Link Cytisco Exotic SNA 0.19 30.8 Holodiscus discolor (Pursh) Maxim. Holodis Yellow S5 0.00 7.7 Juniperus communis L. Junicom Yellow S5 0.00 23.1 Juniperus scopulorum Sarg. Junisco Yellow S3S4 0.00 30.8 Lonicera hispidula (Lindl.) Dougl. Lonihis Yellow S5 0.00 53.8 Mahonia nervosa (Pursh) Nutt. Mahoner Yellow S5 0.00 7.7 Physocarpus capitatus (Pursh) Kuntze Physcap Yellow S4 0.00 15.4 Pinus contorta Dougl. ex Loud. Pinucon Yellow S5 0.00 46.2 Prunus sp. (unidentified) Prunsp - - 0.00 7.7 Pseudotsuga menziesii (Mirbel) Franco Pseumen Yellow S5 0.03 100.0 Quercus garryana Dougl. Quergar Yellow S5 0.30 38.5 Ribes sanguineum Pursh Ribesan Yellow S4 0.00 7.7 Rosa gymnocarpa Nutt. Rosagym Yellow S5 0.00 7.7 Rosa nutkana Presl Rosanut Yellow S5 0.00 23.1 Rubus discolor Weihe & Nees Rubudis Exotic SNA 0.00 15.4 Rubus laciniatus Willd. Rubulac Exotic SNA 0.00 15.4 Rubus leucodermis Dougl. ex T. & G. Rubuleu Yellow S5 0.00 15.4 Rubus ursinus Cham. & Schlecht. Rubuurs Yellow S5 0.01 46.2 Species Prov Cover Freq Woody Plant Species (cont'd) Code Status Rank (%) (%) Symphoricarpos albus (L.) Blake Sympalb Yellow S5 0.00 30.8 Vaccinium ovatum Pursh Vaccova Yellow S4 0.00 15.4 Species Prov Cover Freq Graminoid Species Code Status Rank (%) (%) Achnathemm lemmonii (Vasey) Barkw. Achnlem Yellow S3S4 0.01 23.1 Agrostis exarata Trin. Agroexa Yellow S5 0.04 7.7 Agrostis microphylla Steud. Agromic Yellow S4 0.00 7.7 Agrostis scabra Willd. Agrosca Yellow S5 0.00 7.7 Aira spp. {A. caryophyllea L. and A. praecox L.) Airaspp Exotic SNA 5.07 100.0 Anthoxanthum odor a turn L. Anthodo Exotic SNA 6.60 76.9 Bromus carinatus Hook. & Arn. Bromcar Yellow S5 0.01 7.7 Bromus hordeaceus L . Bromhor Exotic SNA 1.76 84.6 Bromus rigidus Roth Bromrig Exotic SNA 1.76 15.4 Bromus sitchensis Trin. Bromsit Yellow S4 0.00 15.4 Bromus sp. (unidentified) Bromsp - - 0.10 23.1 Bromus sterilis L. Bromste Exotic SNA 0.89 30.8 Bromus tectorum L. Bromtec Exotic SNA 0.02 38.5 Carex brevicaulis Mack. Carebre Yellow S4 0.12 15.4 Carex inops Bailey Careino Yellow S3S4 0.36 23.1 Carex lyngbyei Hornem. Carelyn Yellow S5 0.00 7.7 Carex sp. (unidentified) Caresp - - 0.00 7.7 Carex viridula Michx. Carevir Yellow S5 0.00 7.7 Cynosurus echinatus L. Cynoech Exotic SNA 1.40 53.8 Dactylis glomerata L. Dactglo Exotic SNA 0.00 15.4 Danthonia californica Boland. Dantcal Yellow S5 0.38 53.8 Danthonia sp. (unidentified) Dantsp - - 0.92 23.1 Danthonia spicata (L.) Beauv. ex Roem. & J.A. Schult. Dantspi Yellow S4 0.30 7.7 Dichanthelium acuminatum (Swartz) Gould & Clark var.fasciculatum (Torr.) Freckman Dichacu Yellow S4 0.00 15.4 Dichanthelium oligosanthes (J.A. Schult.) Gould var. scribnerianum (Nash) Gould Dicholi Yellow S4 0.09 38.5 Distichlis spicata (L.) Greene Distspi Yellow S4 0.00 7.7 Species Prov Cover Freq Graminoid Species (cont'd) Code Status Rank (%) (%) Elymus glaucus Buckl. Elymgla Yellow S5 0.00 23.1 Festuca occidentalis Hook. Festocc Yellow S4 0.00 7.7 Festuca rubra L. Festrub Yellow S5 1.57 76.9 Holcus lanatus L. Holclan Exotic SNA 1.78 69.2 Juncus bufonius L. Juncbuf Yellow S5 0.00 7.7 Juncus effusus L. Junceff Yellow S5 0.00 7.7 Koeleria macrantha (Ledeb.) J. A. Schult. f. Koelmac Yellow S5 0.00 7.7 Leymus mollis (Trin.) Pilger Leymmoll Yellow S5 0.00 7.7 Lolium perenne L. Loliper Exotic SNA 0.00 15.4 Luzula multijlora (Ehrh.) Lej. Luzumul Yellow S4 0.17 61.5 Melica subulata (Griseb.) Scribn. Melisub Yellow S5 0.00 15.4 Poa annua L. Poaannu Exotic SNA 0.00 38.5 Poa pratensis L. Poaprat Exotic SNA 0.87 53.8 Vulpia bromoides (L.) S.F. Gray Vulpbro Exotic SNA 1.76 69.2 Vulpia myuros (L.) Gmel. Vulpmyo Exotic SNA 1.44 76.9 Species Prov Cover Freq Herb Species Code Status Rank (%) (%) Achillea millefolium L. Achimil Yellow S5 0.12 76.9 Agoseris grandiflora (Nutt.) Greene Agrogra Yellow S3S4 0.01 23.1 Alchemilla subcrenata Buser Alchsub Exotic SNA 0.04 76.9 Allium acuminatum Hook. Alliacu Yellow S3S4 0.01 46.2 Allium cernuum Roth Allicer Yellow S5 0.01 7.7 Allium sp. (unidentified) Allisp - - 0.00 15.4 Ambrosia chamissonis (Less.) Greene Ambrcha Yellow S5 0.00 7.7 Anaphalis margaritacea (L.) Benth. & Hook. f. ex C.B. Clarke Anapmar Yellow S5 0.00 7.7 Antennaria microphylla Rydb. Antemic Yellow S3S4 0.00 7.7 Anthriscus sp. (unidentified) Anthsp Exotic SNA 0.00 7.7 Arenaria serpyllifolia L. Arenser Exotic SNA 0.00 15.4 Armeria maritima (Mill.) Willd. Armemar Yellow S3S4 0.00 7.7 Artemisia campestris L. Artecam Yellow S5 0.00 7.7 Aspidotis densa (Brackenr.) Lellinger Aspiden Yellow S3S4 0.03 23.1 Species Prov Cover Freq Herb Species (cont'd) Code Status Rank (%) (%) Asplenium trichomanes L. Aspltri Yellow S4 0.00 7.7 Barbarea orthoceras Ledeb. Barbort Yellow S5 0.00 7.7 Bellis perennis L. Bellper Exotic SNA 0.00 23.1 Brodiaea coronaria (Salisb.) Engl. Brodcor Yellow S3S4 2.59 100.0 Calandrinia ciliata (Ruiz & Pavon) DC. Calacil Yellow S4 0.00 23.1 Camassia leichtlinii (Baker) S. Wats ssp. suksdorfii (Greenm.) Gould Camalec Yellow S4 0.00 15.4 Camassia quamash (Pursh) Greene Camaqua Yellow S4 0.00 23.1 Camassia sp. (unidentified) Camasp (Native) - 0.09 46.2 Cerastium arvense L. Ceraarv Yellow S4 0.14 92.3 Cerastium fontanum Baumg. ssp. triviale (Link) Jalas Cerafon Exotic SNA 0.00 15.4 Cerastium glomeratum Thuill. Ceraglo Exotic SNA 0.03 76.9 Chenopodium album L. Chenalb Exotic SNA 0.00 7.7 Claytonia exigua T. & G. Clayexi Yellow S3S4 0.00 7.7 Claytonia perfoliata Donn ex Willd. Clayper Yellow S4 0.01 69.2 Claytonia sibirica L. Claysib Yellow S5 0.00 7.7 Clinopodium douglasii (Benth.) Kuntze Clindou Yellow S5 0.00 23.1 Collinsia grandiflora Dougl. ex Lindl. Collgra Yellow S4 0.00 15.4 Collinsia parviflora Dougl. ex Lindl. Collpar Yellow S5 0.04 92.3 Cryptogramma acrostichoides R. Br. Crypacr Yellow S4 0.00 15.4 Daucus pusillus Michx. Daucpus Yellow S5 0.00 7.7 Delphinium menziesii DC. Delpmen Yellow S5 0.00 15.4 Digitalis purpurea L. Digipur Exotic SNA 0.00 46.2 Dodecatheon hendersonii A. Gray Dodehen Yellow S5 0.00 7.7 Dodecatheon pulchellum (Raf.) Merr. Dodepul Yellow S5 0.00 15.4 Epilobium minutum Lindl. ex Lehm. Epilmin Yellow S4 0.00 53.8 Eriophyllum lanalum (Pursh) Forbes Eriolan Yellow S5 0.02 30.8 Erodium cicutarium (L.) L'Her. Erode ic Exotic SNA 0.01 61.5 Erysimum cheiranthoides L. Erysche Yellow S4 0.00 7.7 Erythronium sp. (unidentified) Erythsp (Native) - 0.00 15.4 Fragaria vesca L. Fragves Yellow S5 0.00 15.4 Fritillaria affinis (Schult.) Sealy Fritlaff Yellow S5 0.00 23.1 Galium aparine L. Galiapa Yellow S5 0.03 92.3 Galium parisiense L. Galipar Exotic SNA 0.00 7.7 Species Prov Cover Freq Herb Species (cont'd) Code Status Rank (%) (%) Geranium carolinianum L. Geracar Yellow S4 0.00 7.7 Geranium molle L. Geramol Exotic SNA 0.09 84.6 Geranium robertianum L. Gerarob Exotic SNA 0.00 7.7 Gnaphalium purpureum L. Gnappur Exotic SNA 0.00 53.8 Grindelia integrifolia DC. Grinint Yellow S5 0.14 46.2 Heuchera micrantha Dougl. ex Lindl. var. diversifolia (Rydb.) Rosend., Butters & Lake la Heucmic Yellow S4 0.04 7.7 Hypochaeris radicata L. Hyporad Exotic SNA 1.85 100.0 Isoetes nuttallii A . Br. Isoenut Blue S3 0.00 7.7 Lactuca muralis (L.) Fresn. Lactmur Exotic SNA 0.00 7.7 Lathyrus sp. (unidentified) Lathsp - - 0.00 7.7 Lathyrus japonicus Willd. var. maritimus (L.) Kartesz & Gandhi Lathjap Yellow S5 0.00 7.7 Lathyrus nevadensis S. Wats. var. pilosellus (Peck) C.L. Hitchc. Lathnev Yellow S5 0.00 7.7 Leontodon sp. Leonsp Exotic SNA 0.00 7.7 Lepidium virginicum L. Lepivir Yellow S3S4 0.00 7.7 Linanthus bicolor (Nutt.) Greene Linabic Yellow S4 0.02 38.5 Lithophragma parviflorum (Hook.) Nutt. ex Torr. Lithpar Yellow S4 0.01 23.1 Lomatium sp. (unidentified) Lomasp - - 0.00 23.1 Lomatium utriculatum (Nutt. ex T. & G.) Coult. & Rose Lomautr Yellow S5 0.15 23.1 Lotus spp. (Lotus denticulatus (Drew) Greene and Lotus micranthus Benth.) Lotusp Yellow S5,S4 0.42 84.6 Lupinus bicolor Lindl. Lupibic Yellow S3S4 0.04 23.1 Lychnis coronaria (L.) Desr. Lychcor Exotic SNA 0.00 23.1 Lythrum hyssopifolia L. Lythhys Exotic SNA 0.00 7.7 Madia gracilis (J.E. Smith) Keck & J. Clausen ex Applegate Madigra Yellow S3S4 0.08 61.5 Mimulus alsinoides Dougl. ex Benth. Mimuals Yellow S3S4 0.00 30.8 Mimulus guttatus DC. Mimugut Yellow S5 0.02 69.2 Minuartia tenella (Nutt.) Mattf. Minuten Yellow S5 0.00 38.5 Moehringia macrophylla (Hook.) Fenzl Moehmac Yellow S5 0.00 7.7 Montia font ana L. Montfon Yellow S3S4 0.01 46.2 Montia howellii S. Wats. Monthow Yellow S3S4 0.00 7.7 Montia parvifolia (Moc. ex DC.) Greene Montpar Yellow S4 0.03 30.8 Myosotis discolor Pers. Myosdis Exotic SNA 0.02 76.9 Myosurus minimus L. Myosmin Yellow S3S4 0.05 30.8 Herb Species (cont'd) Nemophilapedunculata Dougl. ex Benth. Opuntia fragilis (Nutt.) Haw. Osmorhiza berteroi DC. Osmorhiza purpurea (Coult. & Rose) Suksd. Pentagramma triangularis (Kaulf.) Yatskievych, Windham & Wollenweber Phlox gracilis (Hook.) Greene Plagiobothrys scouleri (H. & A.) I.M. Johnst. Plant ago elongata Pursh Plant ago lanceolata L. Plantago maritima L. ssp. juncoides (Lam.) Hult. Plectritis congesta (Lindl.) DC. Polygonum sp. (unidentified) Polypodium glycyrrhiza D.C. Eaton Polystichum munitum (Kaulf.) K.B. Presl Prunella vulgaris L. Pteridium aquilinum (L.) Kuhn Ranunculus occidentalis Nutt. Rumex acetosella L. Rumex crispus L. Rumex salicifolius Weinm. Sagina decumbens ssp. occidentalis (S. Wats.) Crow Sagina procumbens L. Sanicula crassicaulis Poepp. ex DC. Saxifraga caespitosa L. sens. lat. Saxifraga ferruginea R.C. Grah. Saxifraga integrifolia Hook. Saxifraga rufidula (Small) Macoun Sedum oreganum Nutt. Sedum spathulifolium Hook. Selaginella wallacei Hieron. Senecio sylvaticus L. Sherardia arvensis L. — Silene antirrhina L. CP Species Code Status Prov Rank Cover (%) Freq (%) Nemoped Yellow S3S4 0.00 15.4 Opunfra Yellow S5 0.00 23.1 Osmober Yellow S5 0.00 7.7 Osmopur Yellow S5 0.00 7.7 Penttri Yellow S3S4 0.00 30.8 Phlogra Yellow S5 0.00 15.4 Plagsco Yellow S4 0.00 30.8 Planelo Yellow S4 0.17 61.5 Planlan Exotic SNA 0.25 61.5 Planmar Yellow S5 0.00 15.4 Pleccon Yellow S5 0.03 30.8 Polysp - - 0.03 46.2 Polygly Yellow S5 0.00 7.7 Polymun Yellow S5 0.00 7.7 Prunvul Yellow S5 0.03 30.8 Pteraqu Yellow S5 0.06 30.8 Ranuocc Yellow S5 0.01 53.8 Rumeace Exotic SNA 0.68 84.6 Rumecri Exotic SNA 0.00 7.7 Rumesal Yellow S3S4 0.00 7.7 Sagidec Blue S3 0.00 23.1 Sagipro Yellow SU 0.00 30.8 Sanicra Yellow S4 0.01 38.5 Saxicae Yellow S4 0.00 7.7 Saxifer Yellow S4 0.00 7.7 Saxiint Yellow S4 0.02 53.8 Saxiruf Yellow S4 0.00 23.1 Seduore Yellow S5 0.00 7.7 Seduspa Yellow S5 0.03 23.1 Selawal Yellow S4 4.44 92.3 Senesyl Exotic SNA 0.00 15.4 Sherarv Exotic SNA 0.11 46.2 Sileant Yellow S4 0.00 15.4 Species Prov Cover Freq Herb Species (cont'd) Code Status Rank (%) (%) Silene gallica L. Silegal Exotic SNA 0.02 30.8 Silene vulgaris (Moench) Garcke Silevul Exotic SNA 0.00 7.7 Sisyrinchium idahoense Bickn. Sisyida Yellow S3S4 0.00 7.7 Sonchus spp. (Sonchus arvensis L. and Sonchus asper (L.) Hill) Soncspp Exotic SNA 0.00 23.1 Spergularia rubra (L.) J.& K. Presl Sperrub Exotic SNA 0.00 7.7 Spiranthes romanzoffiana Cham. Spirrom Yellow S4 0.06 15.4 Stellaria calycantha (Ledeb.) Bong. Stelcal Yellow S3S4 0.00 7.7 Stellaria media (L.) Vill . Stelmed Exotic SNA 0.00 38.5 Taraxacum officinale G .H. Weber ex Wiggers Taraoff Exotic SNA 0.00 23.1 Teesdalia nudicaulis (L.) Ait. f. Teesnud Exotic SNA 0.05 30.8 Trifolium dichotomum H . & A. Trifdic Blue S2S3 0.00 7.7 Trifolium dubium Sibth. Trifdub Exotic SNA 0.38 46.2 Trifolium gracilentum T. & G. Trifgra Red? SI? 0.00 15.4 Trifolium microcephalum Pursh Trifmice Yellow S4 0.07 38.5 Trifolium microdon H. & A. Trirrnicd Yellow S4 0.08 46.2 Trifolium oliganthum Steud. Trifolig Yellow S4 0.19 61.5 Trifolium pratense L. Trifpra Exotic SNA 0.00 7.7 Trifolium variegatum Nutt. Trifvar Yellow S4 0.00 7.7 Trifolium willdenowii Sprengel Trifwil Yellow S4 0.05 69.2 Trifolium wormskioldii Lehm. Trifwor Yellow S5 0.10 38.5 Triphysaria pusilla (Benth.) Chuang & Heckard Trippus Yellow S4 0.03 30.8 Triteleia hyacinthina (Lindl.) Greene Trithya Yellow S4 0.00 23.1 Verbascum thapsus L. Verbtha Exotic SNA 0.00 15.4 Veronica arvensis L. Veroarv Exotic SNA 0.03 76.9 Veronica serpyllifolia L. Veroser Yellow S4 0.00 15.4 Vicia americana Muhl. ex Willd. Viciame Yellow S5 0.00 7.7 Vicia saliva L. Vicisat Exotic SNA 0.07 53.8 Viola adunca J.E. Smith Violadu Yellow S5 0.00 15.4 Zigadenus venenosus S. Wats. Zigaven Yellow S5 0.01 30.8 Bryophyte Species Amphidium californicum (Hampe ex C. Mull.) Broth. Anacolia menziesii (Turn.) Par. Antitrichia californica Sull. Antitrichia curtipendula (Hedw.) Brid. Asterella gracilis (Web.) Underw. Aulacomnium androgynum (Hedw.) Schwaegr. Aulacomnium palustre (Hedw.) Schwaegr. Bartramia stricta Brid. Brachythecium albicans (Hedw.) Schimp. Brachythecium asperrimum (Mitt.) Sull. Brachythecium sp. (unidentified) Bryum argenteum Hedw. Bryum caespiticium Hedw. Bryum canariense Schimp. Bryum capillare Hedw. Bryum dichotomum Dicks. Bryum gemmascens Kindb. Bryum miniatum Lesq. Bryum muehlenbeckii B.S.G. Bryum pallescens Schleich. ex Schwaegr. Bryum pseudotriquetrum (Hedw.) Gaertn., Meyer & Scherb. Bryum torquescens Bruch. Campylopus fragilis (Brid.) Bruch & Schimp. Campylopus introflexus (Hedw.) Brid. Cephalozia bicuspidata (L.) Dum. Cephaloziella divaricata (Sm.) Schiffn. Ceratodon purpureus (Hedw.) Brid. Claopodium crispifolium (Hook.) Ren. & Card. Claopodium whippleanum (Sull.) Ren. & Card. Cynodontium jenneri (Schimp.) Stirt. Dicranella heteromalla (Hedw.) Schimp. Dicranoweisia cirrata (Hedw.) Lindb. ex Milde Dicranum fuseescens Turn. Species Code Status Prov Rank Cover (%) Freq (%) Amphcal Yellow S3S5 0.00 30.8 Anacmen Yellow S3S5 0.00 7.7 Antical Yellow S3S5 0.00 46.2 Anticur Yellow S4 0.00 15.4 Astegra Yellow? - 0.00 23.1 Aulaand Yellow S4 0.01 76.9 Aulapal Yellow S4 1.18 30.8 Bartstr Red SI 0.00 15.4 Bracalb Yellow S3S5 0.03 61.5 Bracasp Yellow S3S5 0.00 15.4 Bracsp - - 0.00 15.4 Bryuarg Yellow S3S4 0.00 7.7 Bryucae Yellow S3S5 0.00 23.1 Bryucan Red S2 0.06 92.3 Bryucap Yellow S3? 0.09 100.0 Bryudic Yellow S3S4 0.00 23.1 Bryugem Red/Blue? - 0.00 23.1 Bryumin Yellow S3S4 0.14 69.2 Bryumue Blue S2S3 0.08 69.2 Bryupall Yellow S3S5 0.01 69.2 Bryupse Yellow S3S5 0.11 84.6 Bryutor Blue S2S3 0.00 30.8 Campffa Yellow S3S5 0.11 46.2 Campint Exotic SNA 0.00 7.7 Cephbic Yellow? - 0.00 7.7 Cephdiv Yellow? - 0.17 100.0 Cerapur Yellow S4S5 0.29 100.0 Claocri Yellow S3S4 0.00 15.4 Claowhi Yellow S3S5 - -Cynojen Yellow S3S5 0.00 15.4 Dicrhet Yellow S3S4 0.00 15.4 Dicrcir Yellow S3S4 0.00 84.6 Dicrfus Yellow S4 0.00 30.8 Bryophyte Species (cont'd) Dicranum scoparium Hedw. Dicranum tauricum Sapeh. Didymodon brachyphyllus (Sull.) Zand. Didymodon eckeliae Zand. Didymodon nicholsonii Culm. Didymodon norrisii Zand. Didymodon vinealis (Brid.) Zand. - all varieties Didymodon vinealis (Brid.) Zand, \ar.flaccidus (Bruch & Schimp.) Zand. Didymodon vinealis (Brid.) Zand. var. rubiginosus Zand. Didymodon vinealis (Brid.) Zand. var. vinealis Encalypta vulgaris Hedw. Entosthodon fascicularis (Hedw.) C. Mull. Epipterygium tozeri (Grev.) Lindb. Fissidens adianthoides Hedw. Fissidens limbatus Sull. Fossombronia longiseta Aust. Funaria hygrometrica Hedw. Grimmia alpestris (Web. & Mohr) Schleich. Grimmia leibergii Par. Grimmia longirostris Hook. Grimmia ovalis (Hedw.) Lindb. Grimmia pulvinata (Hedw.) Sm. Grimmia torquata Hornsch. Grimmia trichophylla Grev. Hedwigia ciliata (Hedw.) P. Beauv. Homalothecium arenarium (Lesq.) Lawt. Homalothecium fulgescens (Mitt, ex C. Mull.) Lawt. Homalothecium nuttallii (Wils.) Jaeg. Homalothecium pinnatifidum (Sull. & Lesq.) Lawt. Hypnum cupressiforme Hedw. Hypnum subimponens Lesq. Isothecium cristatum (Hampe) Robins. Isothecium stoloniferum Brid. Species Code Status Prov Rank Cover (%) Freq (%) Dicrsco Yellow S4 3.11 92.3 Dicrtau Yellow S4 0.00 15.4 Didybra Red S2 0.00 7.7 Didyeck Red? SI? 0.00 7.7 Didynic Red SI 0.00 15.4 Didynor Red? SI? 0.01 7.7 Didyvin Yellow - 0.03 84.6 Didyins Yellow S4 0.00 53.8 Didyrub Yellow S3S5 0.00 30.8 Didyvvi Yellow S3S5 0.02 69.2 Encavul Yellow S3? 0.00 7.7 Entofas Red S2 0.00 23.1 Epittoz Blue S2S3 0.00 7.7 Fissadi Yellow S3S4 0.00 7.7 Fisslim Yellow S3S4 0.00 15.4 Fosslon Red/Blue? - 0.00 15.4 Funahyg Yellow S4 0.00 7.7 Grimalp Blue? S3? 0.00 23.1 Grimlei Red? S2? 0.01 23.1 Grimlon Blue S2S3 0.00 7.7 Grimova Blue? S3? 0.01 15.4 Grimpul Yellow S3S4 0.00 30.8 Grimtor Yellow S3S5 0.00 7.7 Grimtri Yellow S3S5 0.41 100.0 Hedwcil Yellow S3S4 0.42 61.5 Homaare Blue S2S3 0.28 46.2 Homaful Yellow S3S4 0.00 7.7 Homanut Yellow S3S4 0.00 7.7 Homapin Yellow S3S5 0.29 76.9 Hypncup Yellow S3S5 - -Hypnsub Yellow S4 0.00 38.5 Isotcri Yellow S3S5 0.01 46.2 Isotsto Yellow S4S5 0.01 76.9 Bryophyte Species (cont'd) Kindbergia oregana (Sull.) Ochyra Kindbergia praelonga (Hedw.) Ochyra Leucolepis acanthoneuron (Schwaegr.) Lindb. Lophocolea bidentata (L.) Dum. Lophozia obtusa (Lindb.) Evans Lophozia ventricosa (Dicks.) Dum. Orthotrichum Irupestre Schleich. ex Schwaegr. Orthotrichum consimile Mitt. Orthotrichum lyellii Hook. & Tayl. Orthotrichum speciosum Nees Philonotis fontana (Hedw.) Brid. Plagiomnium venustum (Mitt.) T. Kop. Pleuridium subulatum (Hedw.) Rabenh. Pleurozium schreberi (Brid.) Mitt. Pohlia sp. (unidentified) Polytrichum commune Hedw. Polytrichum juniperinum Hedw. Polytrichum piliferum Hedw. Porella navicularis (Lehm. & Lindenb.) Lindb. Pseudobraunia californica (Lesq.) Broth. Pseudotaxiphyllum elegans (Brid.) Iwats. Pterogonium gracile (Hedw.) Sm. Racomitrium aciculare (Hedw.) Brid. Racomitrium afftne (Schleich. ex Web. & Mohr) Lindb. Racomitrium elongatum Ehrh. ex Frisv. Racomitrium heterostichum (Hedw.) Brid. Racomitrium lanuginosum (Hedw.) Brid. Racomitrium varium (Mitt.) Jaeg. Rhytidiadelphus triquetrus (Hedw.) Warnst. Riccia beyrichiana Hampe ex Lehm. Scapania americana Schistidium maritimum (Turn.) Bruch & Schimp. Schistidium sp. (unidentified; likely Schistidium papillosum Culm.) Species Code Status Prov Rank Cover (%) Freq (%) Kindore Yellow S4 0.11 61.5 Kindpra Yellow S3S4 0.01 38.5 Leucaca Yellow S4 - -Lophbid Yellow? - 0.00 7.7 Lophobt Yellow? - 0.00 7.7 Lophven Yellow? - 0.00 7.7 Orthrup Yellow S3S4 0.00 15.4 Orthcon Yellow S3S5 0.00 7.7 Orthlye Yellow S3S4 0.00 7.7 Orthspe Yellow S3S5 0.00 7.7 Philfon Yellow S3S5 0.05 61.5 Plagven Yellow S3S5 0.00 15.4 Pleusub Yellow S3S5 0.01 69.2 Pleursch Yellow S4S5 0.00 7.7 Pohlsp - - 0.00 76.9 Polycom Yellow S3S4 0.31 38.5 Polyjun Yellow S4 2.71 100.0 Polypi I Yellow S4 1.79 100.0 Porenav Yellow? - 0.00 7.7 Pseucal Yellow S3S5 0.15 30.8 Pseuele Yellow S4 0.00 7.7 Pterogra Yellow S3S4 0.00 7.7 Racoaci Yellow S4 0.10 38.5 Racoaff Blue S2S3 0.55 61.5 Racoelo Yellow S4S5 17.53 92.3 Racohet Yellow S3S5 0.25 84.6 Racolan Yellow S3S4 6.63 61.5 Racovar Yellow S3S5 0.16 69.2 Rhyttri Yellow S4 0.26 53.8 Riccbey Red/Blue? - 0.02 15.4 Scapame Yellow? - - -Schimar Yellow S3S5 0.00 23.1 Schisp - - 0.00 15.4 Species Prov Cover Freq Bryophyte Species (cont'd) Code Status Rank (%) (%) Scleropodium touretii (Brid.) L. Koch var. colpophyllum (Sull.) Lawt. ex Crum Sclecol Blue S1S3 0.00 69.2 Timmiella crassinervis (Hampe) L. Koch Timmcra Yellow S3S5 0.05 76.9 Tortella tortuosa (Hedw.) Limpr. Torttor Yellow S4 0.00 7.7 Tortula papillosissima (Copp.) Broth. Tortpap Blue S2S3 0.01 30.8 Tortula princeps De Not. Tortpri Yellow S3S4 0.00 15.4 Tortula ruralis (Hedw.) Gaertn. et al. Tortrur Yellow S3S5 0.11 69.2 Trachybryum megaptilum (Sull.) Schof. Tracmeg Yellow S3S5 0.02 23.1 Ulota phyllantha Brid. Ulotphy Yellow S3S5 - -Weissia controversa Hedw. Weiscon Yellow S3S5 0.01 61.5 Zygodon viridissimus (Dicks.) Brid. var. rupestris Lindb. ex Hartm. Zygovir Yellow S3S4 0.00 15.4 APPENDIX 3. National and Subnational Conservation Status Definitions. Listed below are definitions for interpreting conservation status ranks at the subnational (S-rank) level (i.e. within British Columbia) (BC Ministry of Environment - BC Species and Ecosystems Explorer online: http://srmapps.gov.be.ca/apps/eswp/search.do). Subnational (S) Conservation Status Ranks: Status Definition SX Presumed Extirpated—Species or community is believed to be extirpated from the nation or state/province. Not located despite intensive searches of historical sites and other appropriate habitat, and virtually no likelihood that it will be rediscovered. SH Possibly Extirpated (Historical)—Species or community occurred historically in the nation or state/province, and there is some possibility that it may be rediscovered. Its presence may not have been verified in the past 20-40 years. A species or community could become N H or SH without such a 20-40 year delay if the only known occurrences in a nation or state/province were destroyed or if it had been extensively and unsuccessfully looked for. The N H or SH rank is reserved for species or communities for which some effort has been made to relocate occurrences, rather than simply using this status for all elements not known from verified extant occurrences. SI Critically Imperiled—Critically imperiled in the nation or state/province because of extreme rarity (often 5 or fewer occurrences) or because of some factor(s) such as very steep declines making it especially vulnerable to extirpation from the state/province. S2 Imperiled—Imperiled in the nation or state/province because of rarity due to very restricted range, very few populations (often 20 or fewer), steep declines, or other factors making it very vulnerable to extirpation from the nation or state/province. S3 Vulnerable—Vulnerable in the nation or state/province due to a restricted range, relatively few populations (often 80 or fewer), recent and widespread declines, or other factors making it vulnerable to extirpation. S4 Apparently Secure—Uncommon but not rare; some cause for long-term concern due to declines or other factors. S5 Secure—Common, widespread, and abundant in the nation or state/province. SNR Unranked—Nation or state/province conservation status not yet assessed. 140 su Unrankable—Currently unrankable due to lack of information or due to substantially conflicting information about status or trends. SNA Not Applicable —A conservation status rank is not applicable because the species is not a suitable target for conservation activities. s#s# Range Rank —A numeric range rank (e.g., S2S3) is used to indicate any range of uncertainty about the status of the species or community. Ranges cannot skip more than one rank (e.g., SU is used rather than S1S4). Not Provided Species is known to occur in this nation or state/province. Contact the relevant natural heritage program for assigned conservation status. Other Qualifiers Rank Definition ? Inexact or Uncertain—Denotes inexact or uncertain numeric rank. (The ? qualifies the character immediately preceding it in the S-rank.) 141 APPENDIX 4. Summary of study site variables. Appendix 4(a). Total plant species richness, by life form group, for sites surveyed using plots (All Veg=all plant vegetation, Bryos=bryophytes). Site A H Veg # Total # Native Vacular Plants # Exotic # Total # Native Woody Plants # Exotic # Total # Native Graminoids # Exotic # Total # Native Herbs # Exotic # Total Bryos # Total B W N 80 33 9 42 7 0 7 5 3 8 21 6 27 38 C H R 105 35 27 62 3 1 4 4 10 14 29 15 44 43 DIS 99 41 22 63 11 2 13 2 8 10 28 12 40 36 G A L 108 44 25 69 4 1 5 7 12 19 33 12 45 39 H B Y 141 62 35 97 8 1 9 10 11 21 45 22 67 44 L A S 93 32 21 53 4 0 4 2 7 9 26 14 40 40 M A X 84 31 20 51 2 0 2 5 7 12 24 13 37 33 R E G 107 37 22 59 4 0 4 6 7 13 27 15 42 48 S A T 78 26 22 48 2 0 2 I 9 10 23 13 36 30 SID 100 34 19 53 4 I 5 3 6 9 27 12 39 47 T E X 96 47 15 62 13 0 13 9 5 14 26 9 35 34 V A L 104 43 25 68 6 0 6 3 6 9 35 18 53 36 WIN 84 49 16 65 10 2 12 7 8 15 32 6 38 19 Average 98.4 39.5 21.4 60.9 6.0 0.6 6.6 4.9 7.6 / 2 5 28.9 12.8 41.8 37.5 Appendix 4(b). Average percent cover of vascular plants, by life form group, for sites surveyed using plots (All Veg-all plant vegetation). Site A l l Veg % Total Vacu lar Plants % Native % Exotic % Total % Native Woody Plants % Exotic % Total % Native Graminoids % Exotic % Total % Native Herbs % Exotic % Total B W N 75.8 19.8 2.3 21.4 0.0 0.0 0.0 8.1 1.1 9.2 11.7 1.3 13.0 C H R 84.1 14.1 33.0 43.5 0.0 0.1 0.1 1.1 29.0 30.0 13.0 4.0 17.0 DIS 54.7 15.8 28.5 41.5 0.0 0.0 0.0 9.2 23.9 33.5 6.6 4.6 11.2 G A L 82.3 13.7 47.0 58.6 0.3 2.2 2.5 2.9 39.5 42.4 10.5 5.4 15.9 H B Y 77.0 12.7 39.8 52.2 0.0 0.0 0.0 8.0 30.4 39.3 4.8 9.4 14.3 L A S 63.5 15.2 3.8 18.4 0.0 0.0 0.0 0.0 3.4 3.4 15.2 0.4 15.6 M A X 88.7 5.5 21.5 26.7 0.2 0.0 0.2 1.8 15.5 17.3 3.5 6.0 9.6 R E G 68.4 14.7 22.5 35.6 0.0 0.0 0.0 1.6 18.4 20.0 13.1 4.1 17.2 S A T 67.2 14.3 46.8 59.4 0.0 0.0 0.0 0.3 41.6 41.9 14.0 5.2 19.3 SID 78.7 4.7 27.2 30.9 0.0 0.0 0.0 0.0 22.4 22.4 4.7 4.9 9.5 T E X . 59.3 13.1 3.8 16.7 0.1 0.0 0.1 2.5 1.0 3.5 10.5 2.8 13.2 V A L 87.7 7.5 24.8 31.0 0.0 0.0 0.0 0.1 23.2 23.3 7.4 1.6 9.0 WIN 43.5 22.1 1.9 22.9 11.3 0.0 11.3 6.2 1.5 7.7 4.7 0.4 5.1 Average 71.6 13.3 23.3 35.3 0.9 0.2 1.1 3.2 19.3 22.6 9.2 3.8 13.1 Appendix 4(c). Average percent cover of ground layer in study sites, by life form group and substratum type (Bryo=bryophytes, Lchn=lichens, Rock=bare parent material, Org=organic matter and decayed litter; Wood=woody debris). Site % B r y o % L c h n % R o c k % O r g % W o o d BWN 52.0 5.8 17.1 7.1 0.3 CHR 61.2 1.4 5.4 12.7 0.3 DIS 4.3 11.4 41.1 6.4 0.0 G A L 30.5 0.2 0.4 15.2 3.2 H B Y 30.2 0.6 11.2 12.8 0.2 LAS 44.3 3.4 28.5 8.8 0.1 M A X 69.4 2.0 6.2 5.2 0.0 REG 38.3 1.0 22.7 9.1 0.0 SAT 8.5 0.2 13.5 22.2 0.2 SID 47.2 7.4 17.4 4.0 0.0 T E X 38.9 5.3 35.0 5.6 0.3 V A L 56.5 18.8 5.7 6.0 0.8 WIN 3.1 21.5 51.9 9.6 0.0 Average 37.2 6.1 19.7 9.6 0.4 Appendix 4(d). Environmental measurements in study sites: AvgCan=average percent canopy cover (AvgCan), %C=total percent carbon contained in soil sample (average), %N=total percent nitrogen contained in soil sample (average), TotSlope=combined total slope, as estimated from horizontal slopes and vertical slopes measured in microplots, TotAsp=combined total aspect, calculated as degrees from North (maximum is 180°). MVH=maximum vegetation height measured (average, in cm), and AvgSD=average soil depth (in cm), as measured in microplots. Site A v g C a n p H % C % N C N ratio TotS lope T o t A s p M V H ( c m ) A v g S D ( c m ) BWN 13.8 4.4 31.2 2.4 17.0 22.8 132.7 11.0 2.5 CHR 2.7 4.9 8.9 0.7 13.3 14.6 125.7 17.3 3.8 DIS 0.7 5.0 18.7 1.3 14.2 13.9 96.5 17.6 5.7 G A L 7.4 4.9 18.2 1.1 16.5 19.1 99.1 39.8 7.0 HBY 0.0 5.6 20.3 1.4 14.3 9.3 103.4 18.7 4.7 LAS 0.6 4.6 ' 32.3 1.6 24.1 23.2 137.6 8.4 1.8 M A X 14.6 4.8 27.5 1.4 19.0 22.4 105.9 32.1 3.9 REG 5.7 4.8 22.5 1.4 15.0 21.8 142.3 14.1 3.7 SAT 12.1 6.3 5.3 0.3 14.7 35.8 160.8 33.3 11.3 SID 0.5 4.7 22.9 1.2 22.4 17.4 110.3 8.6 2.7 T E X 0.7 4.8 32.8 1.6 20.6 21.9 144.8 14.7 1.5 V A L 1.9 5.0 25.6 1.4 17.5 12.2 67.2 21.7 3.1 WIN 0.1 4.6 24.7 1.5 16.5 20.4 102.6 25.3 2.5 Average 4.7 5.0 22.4 1.3 17.3 19.6 117.6 20.2 4.2 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0100752/manifest

Comment

Related Items