Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Multivariate analysis of terrestrial bryophyte-habitat relationships in a subalpine forest of coastal… Sadler, Kella Darleen 1999

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

Item Metadata

Download

Media
831-ubc_2000-0156.pdf [ 6.36MB ]
Metadata
JSON: 831-1.0089382.json
JSON-LD: 831-1.0089382-ld.json
RDF/XML (Pretty): 831-1.0089382-rdf.xml
RDF/JSON: 831-1.0089382-rdf.json
Turtle: 831-1.0089382-turtle.txt
N-Triples: 831-1.0089382-rdf-ntriples.txt
Original Record: 831-1.0089382-source.json
Full Text
831-1.0089382-fulltext.txt
Citation
831-1.0089382.ris

Full Text

MULTIVARIATE ANALYSIS OF TERRESTRIAL BRYOPHYTE - HABITAT RELATIONSHIPS IN A SUB ALPINE FOREST OF COASTAL BRITISH COLUMBIA by KELLA DARLEEN SADLER B.Sc, Simon Fraser University, 1997  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Botany We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA October 1999 © Kella Darleen Sadler, 1999  In presenting  this  degree at the  thesis  in  partial fulfilment  of  University of  British Columbia,  I agree  freely available for reference copying  of  this  department publication  or  and study.  of this  his  or  her  &n/-z?*n  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  A W .  /.  that the  representatives.  may be It  is  thesis for financial gain shall not be  permission.  Department of  requirements  I further agree  thesis for scholarly purposes by  the  m<?  that  for  an  advanced  Library shall make it  permission for extensive  granted  by the  understood  that  allowed without  head  of  my  copying  or  my written  11  ABSTRACT  This thesis had two general objectives: to describe the patterns of distribution and abundance of terrestrial bryophytes in old growth subalpine forests at two study areas (Cypress and Mt. Seymour Provincial Parks) in southern coastal British Columbia, and to determine the scale at which environmental factors influence these patterns. Using principal components analyses, it was demonstrated that regional scale ("biogeographical") plant associations correspond only weakly to the patterns of distribution and abundance of terrestrial bryophytes. Thus, bryophytes may "sense" the environment differently than is the case with vascular plants, i.e. their distribution may be more related to microclimate characteristics. Patterns of bryophyte diversity and abundance were examined at the scale of 20m x 20m study plots using canonical correlation analysis and Pearson correlation coefficients. At this scale, bryophyte diversity and coverage were found to be positively correlated with variables relating to increased light availability. Plots with a northern aspect, reduced canopy coverage, more abundant vascular vegetation, lower abundance of woody debris, and lower LFH depth and organic matter content were associated with higher bryophyte diversity and abundance. The microscale (i.e. 0.1m microplots) environmental preferences of individual 2  bryophyte species were examined using Pearson correlation coefficients, and by examining the range of microhabitat conditions in which species were noted to occur. Species in open habitats exhibited narrower ranges of soil surface layer (LFH) characteristics (pH, C:N ratios) than those in closed habitats, possibly because of greater flux in soil surface conditions (e.g. litterfall) in closed sites. The substratum affinities of bryophytes were examined using their microscale frequencies on available substrata. To investigate the combined influence of substratum and microhabitat, the species associations on different substratum types were investigated using principal components analysis. Microscale features (environment and/or substratum) were found to be closely related to the distributions of some bryophyte species. Assemblages on fine litter and woody debris in open, "stable" habitats with abundant bryophytes and vascular plants differed from those in closed-canopy, steeper, "unstable" habitats with lower  Ill  bryophyte and vegetation cover. Bryophyte assemblages occurring on exposed humus and creeping stems were most closely related to the availability of the substratum types. The relationship of bryophyte species to stable  and dynamic types of  microenvironments was investigated using logistic regression. Bryophytes associated with unstable environments (i.e. dynamic surface layer) had a more random distribution than species that were associated with open, stable habitats, or that displayed strong substratum affinities.  The physiological tolerance of species for microhabitat features (environment  and/or substratum), and the degree of microhabitat stability seem to play important roles in determining the structure and dynamics of terrestrial bryophyte vegetation.  iv TABLE OF CONTENTS  Abstract  ii  List of Tables  vi  List of Figures  ix  Acknowledgements  xi  CHAPTER I Introduction  1  1.1 Ecological Significance of Bryophytes in Subalpine Forests  1  1.2 Thesis Objectives  2  1.3 Literature review  2  1.3.1 Bryophyte Habitat  3  1.3.2 Bryophyte Ecophysiology  7  1.3.3 Bryophyte Community Structure: Competition and Dispersal  11  1.3.4 Bryophyte Sampling 1.4 Biogeography of the Study Areas  .....15 17  1.4.1 Location and Geological History  17  1.4.2 Climate  18  1.4.3 Vegetation and Soils  20  CHAPTER II Methods 2.1 Locations and Methods of Field Sampling  22 22  2.1.1 Study Area and Study Site  22  2.1.2 Plot Layout and Description  24  2.1.3 Microplot Layout and Description  25  2.2 Data Analysis CHAPTER III Results and Discussion 3.1 Influence of Study Site  26 30 30  3.1.1 Biogeoclimatic Characteristics  30  3.1.2 Distribution and Coverage of Bryophytes  40  3.2 Influence of Plot Characteristics on Bryophyte Distribution  46  3.2.1 Patterns of Coverage and Diversity  46  3.2.2 Bryophyte Species Assemblages  53  3.3 Influence of Microhabitat  55  3.3.1 Microscale Bryophyte - Environment Relationships  55  3.3.2 Microscale Bryophyte - Substratum Relationships  62  3.3.3 Combined Influence of Environment and Substratum  64  3.3.4  74  Predicting Bryophyte Assemblage Characteristics  CHAPTER IV Summary of Subalpine Terrestrial Bryophyte Ecology  79  Literature Cited  87  Appendix I  93  Summary of Species and Variable Abbreviations  Appendix II Environmental Variable Ranges of Bryophytes in Microplots Appendix III Literature Summary of Bryophyte Ecology  95 97  VI  LIST OF TABLES  Table 1.4.1. Twenty-five year average (1960-1985) of April 1 snow survey information (from B.C. Ministry of Environment 1999) 20 Table 3.1.1. Summary of landscape characteristics for the six study sites. Values shown are plot averages or microplot averages (mp) 32 Table 3.1.2. Summary of soil data for the six study sites. Values shown are plot averages .. 32 Table 3.1.3. Summary of substratum data for the six study sites. Values shown are microplot averages of fine litter (FLI), woody debris (WOOD), exposed humus (EXH), creeping stems (CRS), exposed roots (EXR), and rock (ROCK)  32  Table 3.1.4. Summary of vascular plant data for the six study sites. Values shown are microplot (mp) averages of percent cover, and plot (p) frequency, from presence or absence in plots 33 Table 3.1.5. Summary of stand stmcture data for the six study sites. Values shown are plot averages of tree seedlings (SE), saplings (SA), pole trees (PT), medium trees (MT), large trees (LT) and very large trees (VL) 34 Table 3.1.6 (a,b). Correlations (factor loadings) of variables with Axis 1 (a), and Axis 2 (b) produced from a PCA of landscape, soil, substratum, vascular plant, and stand structure data 35 Table 3.1.7. Summary of bryophyte abundance (averaged from microplots) and frequency of occurrence in plots at each site, and overall. Total numbers of plots per site are shown in brackets 41 Table 3.1.8. Diversity of all bryophytes, mosses, and liverworts across sites in terms of total number of species (y), average numbers of species within plots (p-cc), and average number of species in microplots (mp-a) are shown 42 Table 3.1.9 (a,b,c). Comparison of (a) # bryophyte, (b) # moss and (c) # liverwort species composition across sites using the Simpson similarity index 42 Table 3.1.10 (a,b). Correlations (factor loadings) of species with Axis 1 (a), and Axis 2 (b) produced from a PCA of bryophyte plot averages 45 Table 3.2.1. Pearson correlation coefficients between some environmental variables (plot averages); correlations greater than 0.3 are shown in bold 48  Vll  Table 3.2.2 (a-d). Results from canonical correlation analyses between environmental variable groups (landscape, substratum, and vegetation) and bryophyte coverage and diversity are summarized (a). Significance tests for the prediction o f independent variables (Sig. Pred), the Stewart-Love canonical redundancy index, and X significance for canonical variates are shown. The canonical loadings on significant canonical variates ( C V ) are shown for different groups o f variables: (b) landscape and soil (c) vegetation, and (d) substratum . . . 50 2  Table 3.2.3 (a,b). Correlations o f environmental variables with A x i s 1 (a), and A x i s 2 (b) produced from a P C A o f bryophyte plot averages 54 Table 3.3.1. Pearson correlation coefficients for bryophyte species abundance and environmental variables (quarter-plot averages)  56  Table 3.3.2 (a,b). Correlations (factor loadings) o f variables with A x i s 1 (a), and A x i s 2 (b) produced from a P C A o f microplot averages o f environmental variables where bryophyte species are present 60 Table 3.3.3. Relative frequency o f bryophyte species on substratum types. Relative frequency is calculated as the number o f microplots in which a species occurs on a substratum divided b y the number o f microplots i n which that substratum occurs  63  Table 3.3.4 (a,b). Correlations o f variables with A x i s 1 (a), and A x i s 2 (b) produced from a P C A o f quarter-plot averages o f bryophytes on fine litter (needles and twigs < l c m i n diameter) 65 Table 3.3.5 (a,b). Correlations o f variables with A x i s 1 (a), and A x i s 2 (b) produced from a P C A o f quarter-plot averages o f bryophytes on woody debris ( l - 1 0 c m i n diameter) 68 Table 3.3.6 (a,b). Correlations o f variables with A x i s 1 (a), and A x i s 2 (b) produced from a P C A o f quarter-plot averages o f bryophytes on exposed humus (compacted, decayed organic matter) 71 Table 3.3.7 (a,b). Correlations o f variables with A x i s 1 (a), and A x i s 2 (b) produced from a P C A o f quarter-plot averages o f bryophytes on creeping stems (living stems and branches <5cm from the ground) 74 Table 3.3.8. Summary o f logistic regression analyses relating presence/absence (response/reference) o f separate bryophyte species to environment and substratum variables. Values denote prediction probabilities (i.e. smaller values denote stronger prediction). Variables with significant effects (p<0.05) are shown i n bold. A l s o shown are Rho-squared, correct placement i n categories (correct res/ref), success gained over a random model (Succlnd res/ref), and overall predictive success (TotCorrect) 76 Table 4.1. Summary o f substratum affinity, habitat characteristics, and species associations (plot and microplot scale) for the terrestrial mosses identified i n this study 82  Summary of substratum affinity, habitat characteristics, and species associations (plot and microplot scale) for the terrestrial liverworts identified in this study 83 Table 4.2.  IX  LIST OF FIGURES  Figure 1.4.1. Distribution of the Mountain Hemlock Biogeoclimatic Zone (from Pojar et al., 1991) 19 Figure 1.4.2. Climate of the Mountain Hemlock Biogeoclimatic Zone. Bars represent precipitation (white=snow, grey=rain). Temperature is shown as a monthly average, maximum and minimum (circles), and as an annual average (solid line). Freezing level is shown as a dotted line (from Brooke et al., 1970) 19 Figure 2.1.1. Cypress study area - site locations: l=Strachan Mountain, 2=Hollybum Mountain, 3=Black Mountain. Study area is located at approximately 49°21'00' 'N, 123°12'00"W (from B.C. Ministry of Crown Lands: Surveys and Resource Mapping Branch 1992) 23 Figure 2.1.2. Seymour study area - site locations: 4=Mystery Lake, 5=Seymour Park Trail, 6=Dinkey Peak. Study area is located at approximately 49°22'30"N, 122°56'30"W (from B.C. Ministry of Crown Lands: Surveys and Resource Mapping Branch 1992) 23 Figure 2.1.3. Plot layout  25  Figure 3.1.1. Diagram showing ecosystem units in the Forest Subzone of the M H (from Brooke etal. 1970) 31 Figure 3.1.2. Positions of plots on PCA axes produced from landscape, soil, substratum, vascular plant, and stand structure data. Plots are grouped by site using 68.3% confidence ellipses (sites 1-3, Cypress study area; sites 4-6, Seymour study area) 35 Figure 3.1.3. Positions of plots on PCA axes produced from landscape, soil, substratum, vascular plant, and stand structure data. Plots are grouped by plant association using 68.3% confidence ellipses (1 = Amabilisfir-Mountain hemlock - Twistedstalh (AMT) association, 2 = Mountain hemlock - Amabilis fir - Blueberry (MAB) association, 3 = Mountain hemlock - Amabilis fir - Mountain-heather (MAM) association) 36 Figure 3.1.4. Positions of plots on PCA axes produced from bryophyte data (plot averages). Plots are grouped by site using 68.3% confidence ellipses 44 Figure 3.1.5. Positions of plots on PCA axes produced from bryophyte data (plot averages). Plots are grouped by plant association using 68.3% confidence ellipses (1 = Amabilis fir Mountain hemlock - Twistedstalk (AMT) association, 2 = Mountain hemlock - Amabilis fir Blueberry (MAB) association, 3 = Mountain hemlock - Amabilis fir - Mountain-heather (MAM) association) 44  X  Figure 3.2.1. Summary of interrelatedness of variables (plot averages). Pearson correlation coefficients (from Table 3.2.1) are shown. Inferred relationships that were not measured are indicated with dashed lines 49 Figure 3.2.2. Correlation of bryophytes with axes produced from a PCA of bryophyte plot average data (plot averages) 54 Figure 3.3.1. Positions of bryophytes on PCA axes produced from microplot averages of environmental variables where species are present 60 Figure 3.3.2. Positions of bryophytes (factor loadings) on PCA axes produced from quarterplot averages of bryophytes on fine litter (needles and twigs <lcm in diameter) 65 Figure 3.3.3. Positions of bryophytes (factor loadings) on PCA axes produced from quarterplot averages of bryophytes on woody debris (l-10cm in diameter) 68 Figure 3.3.4. Positions of bryophytes (factor loadings) on PCA axes produced from quarterplot averages of bryophytes on exposed humus (compacted, decayed organic matter) 71 Figure 3.3.5. Positions of bryophytes (factor loadings) on PCA axes produced from quarterplot averages of bryophytes on creeping stems (living stems and branches <5cm from the ground) 74  XI  ACKNOWLEDGEMENTS  I would like to thank my supervisor, Dr. Gary Bradfield for his assistance and guidance throughout this project.  Thanks also to my committee members; Dr. Wilf  Schofield, who generously offered his taxonomic expertise and enthusiasm for bryophytes, and Dr. Rob DeWreede, for his support and advice during this research. This project was partially supported by a grant to Dr. Schofield from the National Sciences and Engineering Council of Canada, and by a grant to Dr. Bradfield from Forest Renewal British Columbia. Dr. Les Lavkulich and Raphael Fernandez generously provided assistance, advice, and laboratory equipment for soils analysis. Thanks are also extended to lab members Lyn Baldwin for field collaboration, and to Patrick Williston and Dr. Wei Zhang for manuscript and seminar suggestions. Special thanks to Dr. Terry Mcintosh for academic advice and support. Thanks also to my friends, who have contributed advice, support, or inspiration in various ways: Aniko Varga, Keith Pardee, Jennifer, Jason, and Cassidy Goodman, Chelsea Eldstrom, family members (both immediate and extended), and, of course, Hamley. Special thanks to my parents, who taught me the value of listening to your heart, and following your bliss. And 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.  1 CHAPTER I Introduction 1.1 Ecological Significance of Bryophytes in Subalpine Forests The Pacific coast of North America may harbour the richest bryophyte flora of any area of equivalent size on the continent (Schofield 1984). Currently, however, information regarding bryophyte ecology is limited. Bryophytes have been neglected in ecological studies, usually as a result of the taxonomic challenge they present, the difficulties associated with sampling, and, at times, their quantitative insignificance (Lee & LaRoi 1979). Much of the information describing suitable substrata for species in coniferous forests of the Pacific Northwest is limited to floristic works, taxonomic keys (e.g. Lawton 1971), and anecdotal accounts in field guides (e.g. Vitt et al. 1988, Schofield 1992). Subalpine forests have been particularly underrepresented in bryophyte research. Even general ecological studies are rare from the mountainous Pacific Northwest region, although it is clear that bryophytes are a significant component of these forest ecosystems and often contribute substantially to biodiversity. Bryophytes play a major role in nutrient cycling, serving as sinks that absorb the bulk of their mineral nutrients from precipitation, and then slowly leach nutrients back into the ecosystem (Brown & Bates 1990). Bryophytes can also influence water dynamics, contribute significantly to stand biomass and production, and serve as food and habitat for a variety of species of vertebrates and invertebrates (Jonsson 1997). As lowland forests become depleted, the forest industry has been shifting its focus to montane and subalpine forest areas as a source for timber. At the same time, the importance of ecologically based forest management has been brought to the foreground in response to public pressure and the desire to maintain native biodiversity as well as a sustainable resource. Obtaining basic ecological information, such as the patterns of distribution and abundance of bryophyte species in these subalpine forests, is necessary for interpreting ecosystem structure and function, and is a requirement for ecologically sustainable forest management, as well as the application of any conservation strategy (Peck, et al. 1995). Bryophytes have considerable variability among species and much plasticity in morphology and tolerance; this is related to their unique physiology. The habits and habitats  2 of some species may vary from one geographic region to the next, and possibly from one microsite to the next (Peck et al. 1995).  Generalizations about bryophyte species  distributions are best made and utilized on a regional basis where geographic and climatic variables are relatively consistent. While a growing number of bryophytic studies exist that describe changes in bryophyte flora along habitat gradients at the forest stand level, relatively few of these studies have attempted to investigate the responses to within-habitat differences. Such local and micro-scale variation may explain the distribution of many of the less obvious species that show little apparent pattern in gradient and stand-scale studies (Carleton 1990). In an effort to increase understanding of subalpine bryophyte ecology, research was undertaken to describe distribution patterns, substratum affinities, and species assemblages of terrestrial bryophytes in old growth subalpine forests at two study areas in southern coastal British Columbia. Data obtained from laboratory and field identifications were used to describe the bryophyte component in terms of community structure and substratum preferences. The resulting characterizations are compared to existing ecological information, and discrepancies are discussed. 1.2 Thesis Objectives  The specific objectives of the thesis are: 1. To examine the relationship between study site characteristics and patterns of bryophyte distribution and abundance. 2. To investigate factors influencing bryophyte species richness and abundance at the plot level, and to describe plot-level patterns of distribution. 3. To describe the relationship between bryophytes and microhabitat, in terms of environmental factors and substratum availability. 1.3 Literature Review  The following sections review the literature that is relevant to the study of bryophyte patterns of distribution.  The examination of  current descriptive, theoretical, and  experimental information regarding bryophyte habitat, physiology, and community strucutre creates a context for the present study. It is evident that there is currently a much larger number of descriptive and comparative studies than those which are experimental. This skew  3 in the focus of research is related to the scarcity of bryophyte ecological studies in general. Before the function and dynamics of bryophyte communities are investigated through experimental studies, it is necessary to separate components of bryophyte communities (e.g. terrestrial, epiphytic, epixylic, aquatic, disturbed, and undisturbed habitats) and recognize how these communities are organized (i.e. patterns of distribution), and what kinds of environmental features are most related to these patterns.  1.3.1 Bryophyte Habitat Bryophytes thrive in humid environments and require water to grow and reproduce successfully. They often form an almost continuous carpet on the floor of humid forests (Schofield 1985); Pacific Northwest rainforests, in particular, are known for their richness in bryophyte abundance and diversity. Subalpine forests of this area appear to have decreased bryophyte abundance compared to communities at lower elevations.  The reduction in  bryophyte abundance in higher-elevation stands may be a function of reduced temperatures, greater snowpacks, and less free water during the growth season (Peck et al. 1995). Furness & Grime (1982) report slower growth rates at lower temperatures for many bryophytes, although in some cases assimilation and respiration occurred at very low temperatures, suggesting that some species may have thermal adaptations. As a group, bryophytes have a broad ecological tolerance and can occupy a variety of surfaces which vascular plants may be unable to colonize, including rocks, tree bark, decaying wood, soil and litter. However, their persistence in a plant community may be limited, reflecting their response to successional changes or impermanence of habitat; thus bryophyte dispersal and reproductive strategy is important for long term survival. These factors are discussed in a later section of this thesis. Bryophytes are intimately associated with vascular plants and are often confined to particular locations in vascular plant communities (Schofield 1985). Patterns of ecological specialization in bryophytes exhibit strong parallels with those recognized in vascular plants. However, physiological and morphological constraints have restricted their ability to adopt certain strategies. As a group, terricolous cryptogams possess no roots or lignified tissue; unsophisticated transport systems and water relations have limited their ability to capture resources and dominate vegetation in productive, undisturbed habitats. The inability of these organisms to compete successfully with vascular plants has resulted in their relegation to a  4  subordinate role in perennial vegetation (Grime et al. 1990). In other words, bryophytes tend to occupy niche space that is created or avoided by the vascular plants, which in turn may strongly influence many ecological factors, including microclimate (During & Van Tooren 1990). For example, Rincon (1988) noted significant differential effects of various types of herbaceous litter on the vigor of five bryophyte species. Microbial decay of the litter, coupled with the release of mineral nutrients might be expected to benefit the growth of bryophytes over a period of time when temperature and humidity favour moss growth. However, the possible benefit (from increased litter in autumn) coincides with a decrease in the amount of light available to under-canopy bryophytes. Previous bryological studies have demonstrated the sensitivity of bryophytes to substratum type, and that different assemblages develop on different substrata (e.g. Soderstrom 1993). Substratum affinity is pronounced among many bryophytes and species richness has been positively correlated with the number of substrata available (Lee & LaRoi 1979). Thus changes in abundance of substrata such as boulders, exposed soil, coarse woody debris, and deciduous trees can contribute to changes in bryophyterichnessand composition on different geomorphic surfaces (Jonsson 1997). Many species of lichens and bryophytes find optimum habitat in old-growth forests. Lesica et al. (1991) compared old-growth and second-growth forests of the Swan Valley, Montana and found that although depth of litter and duff did not differ between the two groups, the volume and size of snags and the volume of both sound and rotten logs was greater in the old growth. Greater amounts of water-holding woody debris and the deeper canopy of old growth generally result in higher humidity (Franklin et al. 1981, cited in Rambo & Muir 1998b). In consequence, old growth stands seem to provide an environment for bryophytes that is more heterogeneous, more humid, and with more coarse woody debris than do mature second-growth forests. In addition, the greater abundance of some mosses as epiphytes in old growth increases their likelihood of occurrence in incorporated litterfall and probably increases their establishment on coarse woody debris and humus in the ground layer (Rambo & Muir 1998a). Bryophyte vegetation on decaying wood depends on decay stage, climate, and moisture content of the wood, and species identity of the log; humidity is especially important for hepatics (Andersson & Hytteborn 1991). The sequence of succession of  I  5 species on decaying logs can be divided into four groups - facultative epiphytes, early and late epixylics, and ground flora species.  Soderstrom (1988) studied the succession of  bryophytes on decaying coniferous logs in Northern Sweden. Wood texture (degree of surface erosion) was found to be the most important variable in separating species. Facultative epiphytes and some early epixylics are confined mainly to hard wood, whereas ground flora species and some late epixylics are confined mainly to soft wood. The height of the log from the forest floor and its diameter are important determinants of whether or not it will be overgrown by ground flora species; with decreasing height and diameter, the probability that ground flora species can colonize increases.  However, larger logs have  greater water-holding capacity, which facilitates growth of many species sensitive to drought. Although the diversity of substratum types is an integral factor in the establishment of bryophyte communities, landscape characteristics and other abiotic factors are also significant. Interrelationships between soil and landscape variables are complex, reflecting simultaneous changes in microtopography, slope, and soil stability. Understanding the relative importance of abiotic variables in controlling the distribution of non-vascular flora is complicated by the high correlation between individual variables. For example, rainfall is probably a primary determinant of plant cover, which in turn affects light, nutrient availability, and litter levels, and, therefore, bryophyte distribution (Eldridge & Tozer 1997). In addition to the complication of high correlation between particular abiotic variables, apparently homogenous areas may be quite heterogeneous at the microscale level. Frego & Carleton (1995a) found that although the forest floor of boreal woodland was apparently uniform at the scale of hectares, it was heterogeneous in relation to temperature, vapor pressure deficit, incident precipitation, photosynthetically active radiation, and coniferous needle input over a six-month period. Most ecological studies of bryophytes have investigated stand-level relationships of bryophyte communities with environmental factors, usually in the form of gradient analysis (e.g. Muhle & LeBlanc 1975; Lee & LaRoi 1979; Carleton 1990; Vitt 1990). Results from a gradient study in Canadian boreal forests by Carleton (1990) implied that bryophytes are more responsive to environmental influences in terms of their distribution patterns than are many vascular plants. Lee & LaRoi (1979) found that the change in bryophyte species composition was greater along moisture gradients than along elevational gradients. It seems  6  contradictory that species that are often independent of the substratum for moisture and nutrient supply should be the best site indicators. Lesica et al. (1991) studied the distribution of lichens and bryophytes in old growth and managed forest stands of the Swan Valley, Montana.  They concluded that  environmental factors most likely controlling the non-vascular plant distributions in these stands are light, humidity, climatic equitability, quantity and quality of coarse woody debris, and long-term continuity of the woody vegetation; many of these factors depend on stand structure. Rambo & Muir (1998a) found differences in forest floor bryophyte diversity and community composition between two sites in old growth forests of western Oregon. They attributed these differences to disparities in climate, the availability of substratum, and overstory composition. They found that hardwoods influence their environment through variations in light, leachate nutrients, and pH of precipitation throughfall, potentially affecting forest floor bryophyte communities. A study of bryophytes in semi-arid eastern Australia by Eldridge & Tozer (1997) indicated that bryophytes are related to total annual rainfall, soil pH, calcium carbonate levels, plant cover, texture, organic carbon, and soil texture.  Pharo & Beattie (1997)  investigated the regional species richness, variation of species richness, and species turnover of bryophytes and lichens in lowland forests of southeastern Australia.  They explored  correlations between selected environmental variables and patterns of diversity. Lichens and bryophytes responded to the same three variables: vascular plant cover, time since last fire, and topographic position. The above studies suggest that substratum type as well as a wide range of biotic and abiotic factors affect bryophyte communities. It seems apparent that many environmental variables might be of significance when studying habitat preferences of bryophytes. Probably the most important (and feasible) variables to study are those which are related to physiological tolerance (e.g. substrate and light availability), and nutrient uptake (e.g. amount and type of vascular vegetation coverage, soil characteristics). Although bryophytes form an important element in many terrestrial ecosystems, interactions between bryophytes and other elements in the network of an ecosystem (such as vascular plants) have received little attention (During & Van Tooren, 1990). The following section discusses physiological attributes of bryophytes as well as their role in the forest ecosystem.  7 1.3.2 Bryophyte Ecophysiology Bryophytes play an important role in the function of many terrestrial ecosystems and interact with a wide range of organisms. Bryophytes have ecophysiological attributes that are quite different from vascular plants (Nakatsubo 1997). Through unsophisticated transport systems, lack of lignified tissue, and lack of roots, terricolous bryophytes compete poorly with vascular plants for resources such as light and space, and are susceptible to soil motility. These organisms 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). However, these attributes can allow the plants to grow in habitats where most vascular plants cannot colonize, depending on transient water and nutrient supply (Nakatsubo 1997). Bryophytes also interact with other organisms in a variety of ways from obligate symbioses to occasional epiphytism (During & Van Tooren 1990) of various fungi and lichens. In some situations, bryophytes can serve as forage material for some organisms; herbivory seems to be more common in cold environments (Prins 1981). Davidson et al. (1990) found that slugs showed a preference for immature capsules and protonema of all the moss species they tested.  However, the gametophytes of most conspicuous mosses are  perennial and appear to be relatively immune to grazing by slugs or other animals. Possible explanations for the poor utilization of bryophyte biomass include low nutrient value, immunity resulting from the cell wall being physically tough or resistant to digestive enzymes, or the presence of certain chemical compounds and/or elements (Davidson et al. 1990). Bryophytes are also used by vertebrates; birds sometimes use the gametophores and even moss sporophytes for nest-building material (Schofield 1985). Mosses and liverworts may have a substantial and distinctive influence on the functioning of ecosystems where they are abundant, such as boreal forests and the humid, mossy forests of the Pacific Northwest.  Bryophytes contribute moderately to primary  production and may play a major role in nutrient cycling, serving as sinks that absorb the bulk of their mineral nutrients from precipitation and then leach nutrients back into the ecosystem (Peck et al. 1995). In addition to aerial inputs, plants growing beneath vascular plant vegetation may acquire nutrients via throughfall. The substratum may also represent a substantial source of nutrients to overlying bryophytes. More research is needed to clarify the nutrient budgets of these plants. The efficiency of bryophytes in trapping and releasing  8 nutrients is still unknown, as is the speed at which different elements get transferred to other parts of the nutrient cycle (Brown & Bates 1990). It is possible that after achieving a certain biomass and element content, bryophytes become relatively independent of external nutrient sources. Nutrient uptake from external sources could be important during the establishment phase or when specific nutrients are limiting in particular habitats. Under these conditions, aerial inputs may be a major nutrient source (Brown & Bates 1990). If bryophytes are dependent on the substratum for nutrients during a particular phase of development, it may explain why they are such good site indicators. Fertilizer experiments have been used to increase the understanding of how nutrient transfers occur in bryophytes.  Fertilizer additions often cause reduction in bryophyte  biomass resulting from overgrowth by other plants. However, when this problem is avoided, fertilizers frequently fail to enhance the growth, photosynthetic rate or biomass of bryophytes (Skre & Occhel 1979, cited in Brown & Bates 1990). Additional supplies of nutrient elements, therefore, do not appear to be permanently retained within the new growth of mosses; rather, rapid transfer of elements to other parts of the nutrient cycle apparently occurs (Brown & Bates 1990). In terms of species diversity, field studies indicate that vascular plants and bryophytes display opposite responses to nutrient fertilization. For vascular plants, fertilization increases the coverage and diminishes the number of species, whereas for bryophytes it diminishes coverage and increases the number of species. In a study by Ingerpuu et al. (1998), forest species such as Hylocomium splendens and Pleurozium schreberi were the first to disappear in response to fertilizing. Bryophytes are typically small, and have a unique independence from their substratum in terms of water uptake; these characteristics have important influences on their environmental physiology. Many mosses and liverworts tolerate high levels of dessication; photosynthesis declines with water loss, and resumes with greater or lesser delay on remoistening (Proctor 1990). The completeness of recovery is dependent on the intensity and duration of desiccation, and on drought-hardening. The boundary layer dynamics of leaf surfaces are critically important in determining water loss; structures such as papillae decrease the speed at which moisture is evaporated. The time it takes for bryophytes to photosynthesize after rain is determined by storage capacity and rate of water loss; both of these factors are strongly related to growth form (Proctor 1990).  9 Depending on the growth forms of species comprising a community, the high waterholding capacity of many bryophytes may affect water dynamics in some forests. The low thermal conductivity of mosses and liverworts, combined with their relative immunity to grazing and their slow decomposition, often contributes to humus formation or results in their accumulation as a thick mat or as peat (Longton 1984). These extensive bryophyte mats can be significant in the water balance of the forest. At times when precipitation is low, all available moisture may be taken up by the bryophyte mat, restricting filtration through to the soil and roots of the seed plants. At other times, a loose, deep bryophyte carpet enclosing many capillary spaces can prevent rapid water loss from the upper soil layers (Schofield 1985). Ingerpuu et al. (1998) suggest that direct interactions between seed plants and bryophytes in communities should not be overlooked when comparing the dynamics of diversity in these two plant groups. Despite playing a characteristically subordinate role in perennial vegetation, some bryophytes are capable of dominating the plant biomass in various habitats. At least on a local scale, some bryophytes are often able to suppress the establishment and growth of vascular plants; Sphagnum peatland serves as a particularly impressive example.  Many bryophytes that dominate in unproductive habitats of low  biomass have not achieved their dominance through competitive interactions with seed plants. Rather, they have assumed dominance very slowly through the ability to retain and protect captured resources in conditions where stronger competitors suffer relatively greater losses through higher rates of turnover and ineffective defenses against herbivory (Grime et al. 1990). Bryophyte communities may play a role in the recruitment of seed plants. Ingerpuu et al. (1998) found that plots with more extensive bryophyte cover had a lower seedling frequency; the shading provided by bryophytes and the chemical compounds derived from them may be unfavourable to seedlings. Herbs with large, nutrient-rich seeds were found to be more tolerant to bryophyte cover in a species-rich meadow community. These results suggest that in some situations bryophytes can regulate the functioning and composition of plant communities.  However, vegetative reproduction does occur frequently in some  vascular plants, so that the regulating effects of bryophytes may not always be strong.  10 Zackrisson et al. (1997) determined that some moss species had the capacity to suppress tree seedling regeneration substantially in northern boreal forests, particularly in late successional stages. From their study and other indirect evidence they hypothesized that nutrient immobilization may be caused by a tight organic nutrient cycle restricted to the uppermost portion of the humus layer and ground vegetation. They proposed that a three part interaction among feathermosses, ericoid mycorrhizae, and ericaceous dwarf shrubs may block tree regeneration and immobilize nutrients that ultimately may lead to such long-term effects as lowered biomass production in the tree layer, and stand degradation in late successional forests. Conversely, bryophytes can be beneficial to coexisting vascular plants in some situations. For example, dispersed seeds on a bryophyte mat are less apparent to predators than those on bare ground (Verkarr et al. 1986, cited in During & Van Tooren 1990). In addition, a moist bryophyte carpet sometimes forms a suitable seed-bed, although it can prevent plant roots from penetrating through to the mineral layer if the mat is very thick (Schofield 1985). Bryophytes have been observed to equalize temperature fluctuations in the soil and to reduce rates of soil moisture loss (Longton 1984). In this way, mosses and liverworts help buffer the effects of extreme environmental conditions on vascular plants. Since the bryophyte layer in most ecosystems is rather patchy in distribution, structure, and species composition, this variation can increase the spatial heterogeneity of resource availability, which influences the possibilities for niche differentiation and co-existence of vascular plants (Tilman 1982, cited in During & Van Tooren 1990). Therefore, bryophytes may have a greater impact on nutrient cycling, on soil temperature and moisture regimes, and on the range of habitats available to other organisms than is generally recognized in ecosystem studies (Longton 1984). In addition to physiological tolerances and interactions with other organisms, the composition and distribution of bryophytes may be related to competitive interactions with other bryophytes. These latter processes would be expected to be influenced by life history strategies and dispersal abilities. Depending on the permanence of the substratum, competition may play an important role in determining bryophyte community structure. The following section discusses these aspects of bryophyte ecology in more detail.  11  1.3.3 Bryophyte Community Structure: Competition and Dispersal  Bryophytes are often colonial, occurring in patchy, temporary habitats such as rotting logs, shifting soil, dried pond depressions, and fragmented forests. As such, some bryophyte species assemblages can be temporary.  Slack (1982, 1990) has suggested that many  bryophytes, and perhaps especially hepatics, are largely opportunistic (or fugitive) species that disperse and are successful in one temporary habitat after another. Slack postulates that their primary role to species diversity is in their coexistence not indefinitely, but persistently in ultimately temporary habitats. In consequence, a community may consist of a set of dominant, or "core" species which interact strongly enough to reach a competitive equilibrium, but are surrounded by a larger set of coexisting, non-equilibrium species (Kimmerer & Allen 1982). These "fugitive" bryophytes can avoid inevitable local extinction resulting from competitive exclusion essentially by avoiding competition. Okland (1994) has suggested alternatives to competition in regulating bryophyte communities: static mechanisms for avoidance (such as morphological differences), mobility (the ability to move to more favourable microsites), and the importance of density-independent mortality (related to fine-scale disturbance). As might be expected, the persistence and abundance of such opportunistic species in an area results from the dynamic balance between habitat persistence and the ability of the species to disperse and colonize such habitats. Reducing habitat parameters (decreased area, fewer localities and/or substratum patches, and increased distance between patches) generally results in lower persistence of species inhabiting these temporary refuges.  The greater  sensitivity of fugitive species to reduction in habitat size or availability reflects their strong, short-term dependence on the process of establishment. The reproductive system operating within a plant population has been thought to have a profound influence on the pattern of variation exhibited by the population (Longton & Miles 1982). For many bryophytes, the crucial factor influencing distribution is spore transport capacity (Herben & Soderstrom, 1992). A study by Marino (1991) supports the suggestion that the composition and diversity of moss communities growing in short-lived habitats can be strongly influenced by dispersal and physiological tolerance.  12 Since most bryophyte populations are patchy in their distribution, the production, dispersal, and establishment of diaspores is important (Soderstrom 1994).  Bryophyte  diaspores are structures that have the ability to reproduce the plant; these include spores, vegetative fragments, and asexual reproductive units such as gemmae (Schofield 1985). Although the dispersal distances of many bryophyte gametes are often quite short, i.e. a few centimetres, distances traveled by spores are orders of magnitude greater, i.e. several kilometers (Wyatt 1982).  Very small diaspores that become airborne may be carried  considerable distances. Some bryophytes can leave resistant gemmae that lay dormant until the next favourable season, and many can survive poor conditions by virtue of their enormous regenerative capacities (Muller-Stoll 1965, cited in During 1979). Diaspores may accumulate in substrata over time, forming a diaspore bank. The diaspore bank of bryophytes could play a role similar to that of the seed bank in vascular plants: it allows species to survive unfavourable periods, it facilitates rapid colonization after disturbance, and it influences the post-disturbance species composition and diversity (Jonsson 1993). The appearance of bryophyte propagules at a newly exposed site may be achieved either by dispersal in space, or by being persistent at the site in a diaspore bank.  The fitness of most bryophytes is measured primarily by numbers and sizes of  propagules, and by longevity and resistance of shoots and propagules, depending on the life strategy adopted (During 1979). Jonsson (1993) found that the dominant life strategies of bryophytes in the diaspore bank of a Picea abies old growth forest were characterized by good opportunistic abilities, short life span, and high reproductive effort (monoicous species with frequent gemmae and spore production). Conversely, long-lived, perennial bryophytes dominated the undisturbed forest floor. Some studies have examined the community dynamics of bryophytes that occur on temporary habitats. Jonsson & Esseen (1990) compared patterns of bryophyte distribution and abundance in boreal forest floor patches that were undisturbed (100cm quadrats) with 2  those that had been disturbed via uprooting (minimum 60cm quadrats). They found that bryophyte diversity and species richness in treefalls were significantly higher than that of the corresponding samples in undisturbed forest floor (pO.OOl). explanations for this observation.  They proposed several  First, uprooting may create space for bryophyte  colonization that is free from potential competitors. Second, disturbed patches likely have  13 high habitat heterogeneity (hence accommodating more species).  Third, within-patch  disturbance may continue long after patch formation through erosion from the tip-up mound. Fourth, the small patch size implies a short distance to potential sources of bryophyte diaspores (which should increase the chance of establishment). The results of that study found that treefall disturbances are important for both the persistence of colonists and the maintenance of high bryophyte diversity in boreal forest ecosystems. Gaps produced by natural disturbances can liberate resources and provide colonization opportunities for species that cannot establish themselves in later successional stages. Among colonist species, differential establishment success can promote coexistence and increase species diversity (Kimmerer & Young 1996). Disturbance may also play an important role in the spatial patterning and reproductive ecology of epixylic bryophytes Kimmerer (1993, cited in Kimmerer & Young 1996). Kimmerer and Allen (1982) examined the role of disturbance in the pattern of a riparian bryophyte community. Analysis of species composition, abundance, and diversity showed spatial pattern at two scales: large scale elevational zonation of the dominant species, upon which was superimposed a smaller-scale, patchy distribution of species and open substratum. Their results suggested that small-scale pattern in the form of discrete patches was influenced significantly by the magnitude of individual disturbance events, whereas large-scale patterning was strongly influenced by disturbance frequency.  The study,  therefore, supports the suggestion that frequency of disturbance facilitates the coexistence of many species. In forests of the Pacific Northwest, it seems likely that undisturbed forest floor may be one of the habitats that might remain stable long enough for bryophyte species to interact and reach a competitive equilibrium. Slack (1977) considered forest floor habitats to be among the more stable sites and, thus, potentially suitable for "equilibrium" species of bryophytes. However, Watson (1980) concluded that forest floor microhabitats do not persist for sufficiently long periods of time for equilibrium processes to become an important component of population regulation. Some species share similar habitat responses and similar substratum affinities yet manage to coexist in what appear to be stable bryophyte assemblages. In humid, mossy forests with a more or less continuous bryophyte carpet composed of more than one species,  14 it is reasonable to assume that interspecific interactions occur. Further, such interactions may operate to partition spatially discrete resources associated with microsites. The spatial distributions of sedentary organisms such as bryophytes are more likely to be defined by a range of habitat conditions than by consumable resources. Therefore, if niche partitioning does occur in these organisms, it should result from competition for space containing those conditions. Frego & Carleton (1995a) hypothesized that the spatial pattern of four bryophyte species on the forest floor of boreal woodland represented habitat partitioning, corresponding to microhabitat heterogeneity in terms of temperature, light, precipitation, and litterfall. Their results, however, did not support habitat partitioning by the species on the basis of the variables examined; species occupied sites which overlapped in microhabitat characteristics. Frego & Carleton (1995b) performed a reciprocal transplant study of the same four bryophyte species.  Their results showed that spatial separation within the feathermoss  community is not a result of habitat partitioning, although they did not rule out the force of slow competitive exclusion. Furthermore, they found that growth rates were related to the relative abundance of the four species studied, rather than microsite. Lee & LaRoi (1979) studied three feathermosses in relation to substratum affinities, and observed that all species coexist under apparently homogenous conditions. They hypothesized that distributions might result from divergence along niche dimensions that they did not measure, e.g. nutrient gradients. Alternatively, they proposed that such coexistence might be locally temporary but globally stable, resulting from cyclical microsuccessional processes. Finally, they suggested that continually shifting environmental conditions could prevented competitive exclusion. Okland (1994) examined the patterns of associations between 36 bryophytes and their relationships with trends in diversity at five different scales (from 1 m to 1/256 m ) in Norwegian boreal spruce forest. He found that the number of positive associations was significantly higher than predicted from random distribution for all sample plot sizes except the smallest. He proposed that the excess of positive associations was related to the presence of a-diversity (total number of bryophyte species within a plot; this usually relates to environmental heterogeneity), P-diversity (compositional turnover along environmental gradients e.g. moisture or altitude), and possibly, facilitation (positive density-dependence caused by more favourable moisture conditions within dense stands). Conversely, he also  15 suggested that the low proportion of negative associations indicated that interspecific competition was not important in the vegetation examined. Okland (1994) suggested the existence of a micro-scale coenocline, running from the "normal forest floor" dominated by mosses and liverworts, to patches dominated by small liverworts. This micro-scale coenocline may be associated with recolonization after finescale disturbance and/or physiological tolerance of the species to a more extreme micro-site environment.  Rapid dynamics of the bottom layer, as has been documented for other  ecosystems (During & Van Tooren 1990), would support the former explanation. The lifestrategies of these two groups differ along a gradient as well. The mats of small liverworts, mostly growing closely appressed to the substratum, are represented by more stress-tolerant species, while those of the tall turfs, carpets and wefts of the forest floor species are more likely to be competitor species. The fact that both groups have the properties of perennial stayers - small spores, low reproductive effort, and long potential life span - suggest that a micro-environmental complex gradient is responsible for this differentiation. Okland (1994) suggests that forest mosses are excluded from these pockets of small liverworts by lack of physiological tolerance to low light, or by problems associated with establishment and maintenance of viable populations on steep slopes, under overhangs, or on chiefly inorganic soils. 1.3.4 Bryophyte Sampling Sampling bryophyte vegetation is often difficult, owing to problems with field collections and identifications. Often, bryophyte vegetation is composed of very diverse taxa which may be difficult to determine accurately from material lacking sporophytes or reproductive features. Cover and abundance measurements are often difficult as well, as species may be inter-tangled and/or too small to discriminate in the turf of a larger species. Another major problem with bryophyte sampling is ascertaining the appropriate size of sampling plots and method of sampling. Some methods of sampling that have been used for bryophytes include whole plot sampling, the use of belt transects or microplots, as well as various plotless methods (McCune & Lesica 1992). Because bryophyte distribution has a generally patchy nature, a completely random sampling scheme is likely to exclude some important aspect of the community structure  16 (Slack 1984).  Therefore, some form of stratified or systematic sampling is usually  undertaken. McCune & Lesica (1992) state that one "rule of thumb" in choosing microplot size is that when cover is estimated, plot size should be small enough so that the entire plot can be viewed at one time with individual organisms still discernible. Plot sizes from 0.01 to 0.1 m have been considered appropriate for studying bryophytes and small vascular plants (Daubenmire 1968).  Okland (1994) found that interspecific association structure  disintegrated towards finer scales of microplot sizes as a result of weakened relationships in response to the reduced number of species per sample plot. He noted that the critical limit for species representativeness was trespassed at about 0.01m . When measuring the coverage of terrestrial bryophytes, taking the average of many small quadrats will be more accurate for common species than a single estimate of a large area. However, this accuracy comes at the price of a poor representation of infrequent species (McCune & Lesica 1992).  When vegetation is sparse and most species are  infrequent, an incomplete species list results unless the number of microplots is increased greatly.  McCune & Lesica (1992) compared the efficiency and accuracy of studying  bryophytes using whole plots, belt transects, and microplots in forests" of Montana. In their study, microplots were found to be the most efficient, but had the poorest species capture. These authors suggest that if the number of microplots is increased greatly it might obtain comparable species capture, but then the method would likely be just as time consuming as studying the whole plot. It seems that if the microplot sample size is small and many species have low coverage, one is left with a poor estimate of the variance in cover among stands, the data being essentially presence or absence. However, studying the whole plot is only beneficial when one is interested solely in coverage and diversity of the bryophyte layer; it is not useful in terms of making specific associations with substratum or microclimatic conditions. In addition, if one wishes to study only a particular habitat within a plot (for example, undisturbed forest floor), avoiding atypical areas, a whole-plot survey may prove to be problematic in the sense of defining boundaries for each type of habitat. Furthermore, many bryophytes (most notably the small liverworts) are nearly microscopic and presence is discovered only during laboratory examination of samples. In consequence, depending on plot size and type of bryophyte vegetation, performing a whole-plot survey might be very time consuming and biased in  17 favour of moss species.  Depending on the vegetation, certain plots are much less  accommodating than others; plots with impenetrable vegetation are very difficult to examine thoroughly and/or define habitat boundaries. Nevertheless, if time allows, the best strategy is probably to record specific information in microplots, and then perform a whole-plot ocular survey to record the presence of any species missed in the microplot samples. 1.4 Biogeography of the Study Areas 1.4.1 Location and Geological History The two areas chosen for this study were Cypress Provincial Park and Seymour Provincial Park. The mountains in these parks are part of the southwestern extent of the Coast Range in British Columbia. The Coast Mountains dominate the relief along the mainland coast of British Columbia, forming a rugged, strongly articulated chain that rises from sea level to altitudes of up to four kilometres (Brooke et al. 1970). The Coast Range extends over 150km wide and about 1000km in length, merging with the St. Elias Mountains of the Yukon in nearly unbroken continuity. The topography of the area is considerably younger than the rocks themselves (Parrish 1982); the Coast Mountains are largely underlain by a complex of plutonic rocks, dominated by granodiorite and quartz diorite that range in age from the Cretaceous to the Tertiary (Culbert 1971). Fjords and sounds, which penetrate deeply inland and which once served as outlets for ice movement during the Pleistocene, interrupt the continuity of the zone.  The mountain chain functions as a climatic and  physiological barrier; there are few through-valleys that link coastal and interior sides. The North Shore Mountains, located 8-65km north of Greater Vancouver, represent the southwesterly extent of the Coast Range (Brooke et al. 1970). The topographical features of the Cypress and Seymour study areas were almost completely concealed by ice during the Pleistocene era.  This period of glaciation is  evidenced by the cirques, hanging valleys, and extensive mantle of glacial drift of varying thickness conspicuous in much of the area; pre-glacial lines of drainage were widened and deepened by the Pleistocene ice (Brooke et al. 1970). Soil development and colonization by plants in both of these study areas occurred after recession of the Pleistocene ice approximately 10,000-11,000 years ago.  Colonizing plants re-established from southern  areas, as well as from northern refuges; it seems likely that the contribution of species from  18 northern refuges would be insignificant for the Cypress and Seymour areas.  Glacier  resurgence, with general periods of climatic change, were most likely associated with corresponding changes in the flora and vegetation of present subalpine elevations in the Coast mountains. However, in the North Shore mountain area, no recent morainal ridges exist to indicate the presence of glaciers since the recession of the Pleistocene ice (Brooke et al. 1970). 1.4.2  Climate  Although there may be considerable variation in microclimate between sites depending on position, elevation, and topography, one can distinguish areas within the same geographic region that share the same macroclimate. Forested areas of the subalpine in the Coast Range share similar macroclimatic effects and, thus, fall within the same biogeoclimatic unit, as described by Krajina (1965) and Pojar et al. (1991). The Mountain Hemlock Biogeoclimatic Zone occurs all along the coast in British Columbia and extends north through south-eastern Alaska, and south into Washington and Oregon (Figure 1.4.1; Pojar et al. 1991). The zone occupies an elevational band from approximately 900-1675m on windward slopes, and 1100-1825m on leeward slopes in south-western British Columbia (Eekman 1976). The Mountain Hemlock (MH) zone is characterized by cool short summers, and long, cold, wet winters with heavy snow cover over unfrozen ground for several months (Pojar et al. 1991). The climatic pattern representing the MH zone is shown in Figure 1.4.2. Average annual temperature is approximately 3°C to 7°C, with a monthly average of -9°C to -1°C in January, and a monthly average of 11°C to 13°C in July. Average monthly temperature remains below 0°C for 1-6 months, and above 10°C for 2-3 months.  Total annual  precipitation ranges from 1780-4320mm, 20-70% of which is snow. The wettest month, which usually occurs in autumn or winter, receives approximately 305-410mm of precipitation. In contrast, spring and summer months are relatively dry; the driest month receives only 33-84mm of precipitation (Krajina 1969). At subalpine elevations in the North Shore mountains, cloudiness, high humidities, heavy precipitation and snow, coolness, and sudden changes in climatic pattern are common. Snow and freezing temperatures may occur in any month of the year (Eekman 1976).  19 Figure 1.4.1. Distribution of the Mountain Hemlock Biogeoclimatic Zone (from Pojar et al., 1991).  Figure 1.4.2. Climate of the Mountain Hemlock Biogeoclimatic Zone. Bars represent precipitation (white=snow, grey=rain). Temperature is shown as a monthly average, annual maximum and minimum (circles), and as an annual average (solid line). Freezing level is shown as a dotted line (from Brooke et al., 1970).  J  F  M  A  M  J  J  A  S  O  N  D  20  Within the Mountain Hemlock Biogeoclimatic Zone, the most important climatic feature is the large accumulation of snow. The depth and duration of snow is an excellent indicator of variations in the subalpine climate in the North Shore mountains as it is a sensitive integrant of temperature and precipitation patterns; it has a marked influence on the continuity of forest cover and other vegetation patterns (Brooke et al. 1970). In fact, most zonal features of the vegetation are related to the annual quantity and quality of snow (Krajina 1965). The data in Table 1.4.1 summarize the 25-year averages (1960-1985) of snow depth, water content, and snow/water equivalent ratios for the April 1 surveys of Hollyburn Mountain and Mount Seymour (B.C. Ministry of Environment 1999). The data indicate that, as expected, Hollyburn Mountain on Cypress receives a slightly higher amount of snow, with greater water content, on average, than Seymour. Twenty-five year average (1960-1985) of April 1 snow survey information (from B.C. Ministry of Environment 1999).  Table 1.4.1.  Seymour  Hollyburn  1070  1100  80-620  155-630  average snow depth (cm)  356.3  369.6  average water content (cm) snow/water equiv. ratio  158.9  168.8  2.24  2.19  elevation (m) snow depth range (cm)  2.1.4 Vegetation and Soils  The most dominant tree species in the Mountain Hemlock (MH) Zone are mountain hemlock (Tsuga mertensiana), amabilis fir (Abies amabilis),  and yellow cedar  (Chamaecyparis nootkatensis). The upper limit of the M H zone is the altitudinal tree-limit of Tsuga mertensiana, above which is the Alpine zone. Differences in exposure and snow duration can influence large-scale vegetation patterns (Brooke et al. 1970). The irregular topography of the North Shore mountains, therefore, results in an uneven upper altitudinal tree limit which falls anywhere between 900-1675m (Krajina 1965). The lower limit of the M H zone is the altitude at which T. mertensiana ceases to be a dominant component of forests; below this elevation it is replaced by western hemlock (Tsuga heterophylla) which dominates in the adjacent Coastal Western Hemlock zone.  21 Tree growth becomes progressively poorer with increasing elevation because of the shorter growing season, increased duration of snow cover, and cooler temperatures (Pojar et al. 1991).  Based on differences in forest structure, Krajina (1965) distinguished two  subzones within the M H zone. At the upper limit of the M H zone (and the lowest limit of the Alpine zone) is the "subalpine parkland subzone". Trees in this subzone rarely form closed forest stands - they occur in isolated clumps and irregular small patches, frequently interrupted by other plant communities. Depending on duration of snow, heather, meadow or fen plant communities may develop (Pojar et al. 1991). The "subalpine forest subzone" is characteristic of lower elevations. This subzone consists of closed forest stands, which are only rarely interrupted by other plant communities. In the southern Coast Mountains, this subzone occurs between 900-1100m on windward sides, and 1100-1300m on leeward sides (Krajina 1969). Snow duration within this subzone seldom exceeds 8 months, but is usually at least 6 months. The slightly warmer climate, longer growing periods, and shallower snow covering has favoured the establishment of complex forested ecosystem units (Brooke et al. 1970). Soil development in the M H zone is greatly influenced by prevailing climatic conditions, which characteristically produce a short vegetative season and snow-free period. Although soils in the M H zone have had relatively little time to develop, they possess distinct morphological features. Low temperatures and increased moisture content slow down the decomposition of plant remains resulting in a high accumulation of acid, snow-compacted organic matter on the forest floor. The instability of surface soil at high elevations in areas with little plant coverage is characterized by minimal profile differentiation (Brooke et al. 1970). The prolonged melt period from accumulated snow may influence soil development in different ways depending on drainage capacity. Leaching potential can be high in welldrained soils, whereas imperfectly-drained soils are subject to gleying (i.e. saturation with stagnant water). In most circumstances, soils are maintained in a moist to saturated state throughout the year. In summary, characteristic soil processes in the M H zone include high organic matter accumulation, mycelial mor humus formation, gleying, leaching, and eluviation. Podzols and Folisols are the predominant soil types (Pojar et al. 1991).  22 CHAPTER II Methods  2.1 Locations and Methods of Field Sampling 2.1.1 Study Area and Study Site  The two study areas (Cypress Provincial Park and Seymour Provincial Park) were chosen mainly for their accessibility; the prevalence of continuous forest adjacent to ski runs and hiking trails facilitated field sampling. In addition, the availability of aerial photos and area maps also improved field sampling efficiency. The field portion of this study took place during July and August of 1998, and was completed by myself and Lyn Baldwin, a Ph.D. candidate at UBC who examined bryophyte assemblages on rotting logs in a concurrent study. Within each study area, three different sites were chosen based primarily on accessibility within the desired elevation range (1100m to 1300m). The location of these sites are illustrated in Figure 2.1.1 and Figure 2.1.2. All of the sites chosen were Tsuga mertensiana - Abies amabilis - Chamaecyparis nootkatensis old growth stands which are not  thought to have been influenced by forest fire. The number of plots studied at each site are as follows: Cypress Provincial Park:  1. Strachan Mountain =12 plots 2. Hollyburn Mountain = 5 plots 3. Black Mountain = 8 plots Total plots (1100m to 1300m) = 25 Seymour Provincial Park:  4. Mystery Lake area = 7 plots 5. Seymour Park Trail area = 6 plots 6. Dinkey Peak - 7 plots Total plots (1100m to 1200m) = 20  23 Figure 2.1.1. Cypress study area - site locations: l=Strachan Mountain, 2=Hollyburn Mountain, 3=Black Mountain. Study area is located at approximately 49°21'00"N, 123°12'00"W (from B.C. Ministry of Crown Lands: Surveys and Resource Mapping Branch 1992).  Figure 2.1.2. Seymour study area - site locations: 4=Mystery Lake, 5=Seymour Park Trail, 6=Dinkey Peak. Study area is located at approximately 49°22'30"N, 122°56'30"W (from B.C. Ministry of Crown Lands: Surveys and Resource Mapping Branch 1992).  24 2.1.2 Plot Layout and Description All plots (20m x 20m) were located within homogeneous stands representative of the site studied. Topographical features such as rock outcrops, steep slopes (greater than 45 degrees), and creeks greater than lm in width were avoided by offsetting plots a minimum of 10m. Plots were established a minimum of 50m from forest edges, based on research by Matlack (1993), who detected significant edge effects in forest stands, in some cases affecting the forest microenvironment up to 50m from the edge. Plots were spaced at least 50m from each other. For most sites in Cypress Park we were able to put down plots more or less systematically, resulting in two or more series of plots that ran (approximately) along a contour. In Seymour Park, we had difficulty replicating this sampling strategy owing to frequent topographical variations (rock outcrops, cliffs, large creeks). As a consequence, the plots in Seymour Park were somewhat less systematically distributed, but still within the same elevational zone and forest type. Distance within and between the plots was measured using a hipchain, and a compass was used to aid in establishing plot .boundaries. In each plot, large scale environmental features were described. These included estimations of topographical position, aspect and slope. The structure of the plot was described by estimating the coverage of different strata (canopy, understory, and surface features) and by quantifying stand structure.  Stand  structure was described by counting the numbers of trees (by species) within diameter at breast height (DBH) categories: seedlings (<5 cm DBH), saplings (5-10 cm DBH), pole trees (10-20 cm DBH), medium trees (20-40 cm DBH), large trees (40-60 cm DBH), and very large trees (>60 cm DBH). At or near the center of each plot, a soil pit was dug to estimate the depths of the LFH layer and the inorganic soil horizons (A,B). For each plot, three soil samples (litter, A horizon, and B horizon) were collected for laboratory examination. Laboratory analysis of the soil samples consisted of pH testing (all three layers), quantification of organic matter content (i.e. carbon) through loss on ignition testing (LFH layer only), and total Kjeldahl nitrogen (LFH layer only). These estimates of carbon and nitrogen allowed for a comparison of plot productivity in terms of the C:N ratio of the LFH layer. The methods used to obtain pH and organic matter content followed Carter (1993). The procedure for determining total Kjeldahl nitrogen (TKN) followed guidelines from Lavkulich (1999, pers. comm.).  25 2.1.3 Microplot Layout and Description Within each plot, eight 0.1m microplots were systematically located and examined 2  (see Figure 2.1.3).  Although eight microplots may seem to be a small sample size  considering the recommendations by McCune & Lesica (1992), it was considered the maximum allowable, given time constraints in the field and the laboratory. Microplots were studied using a standard Daubenmire 0.2m x 0.5m metal frame with overlying plastic mesh divided into 2cm x 2.5cm squares (two squares represented approximately 1% of the microplot). Microplots were located at 4m and 8m from the center point in each of the four directions along the contour line and the fall line. Sampling was carried out both parallel and perpendicular to the slope. Figure 2.1.3. Plot layout. >  ?0m  gg?l Microplot (20cm x 50cm) ©  Plot Center (site of soil pit)  Fall Line  Canopy coverage was estimated above each of the eight microplots using a spherical densiometer;  four measurements (N,E,S,W)  were  recorded.  Following canopy  measurements, vascular plant coverage, height, and species composition were described at the scale of the microplot. The availabilities of seven types of substratum were estimated: fine litter (needles and woody debris less than lcm in width), medium litter (woody debris 15cm in width), coarse litter (woody debris 5-10cm in width), exposed humus, creeping stems, exposed roots, and rocks. Total available substratum was always equivalent to 100%. Total  26 bryophyte coverage in the microplot was estimated, followed by estimates of individual species coverage on each type of substratum.  2.2 Data Analysis Following laboratory identification of field samples, all plot and microplot data were entered onto an Excel spread sheet for statistical analysis. Combined and individual characteristics of study sites were described by summarizing the estimates for different environmental variables (landscape, stand structure, vegetation, soils, substrata). The overall and site-level composition of bryophytes encountered in this study were described using plot averages of coverage and frequency. Bryophyte a-diversity, defined as total number of species recorded at the plot or microplot level, was also calculated. The Simpson Similarity index (SI = (#shared A B ) / (#unique A + #unique B + # shared A B ) where A and B represent species compositions of different sites) was used to determine similarities in overall composition. The degree of compositional change in bryophytes among plots and microplots (i.e. P-diversity) was examined using detrended correspondence analysis (DCA). The units of D C A axes are measured in standard deviation units (SD) which represent average standard deviation of species turnover (Gauch 1982). D C A ordinations therefore allow for a direct assessment of P-diversity via gradient lengths associated with the axes produced; the longer the gradient (SD units), the more substantial the p-diversity. D C A estimates ecological distance optimally when gradient lengths exceed 3.0 SD (Okland 1986). If a D C A analysis produces only short gradient lengths (i.e. less than 3.0 SD) it is assumed that there is a lack of substantial P-diversity in the data set.  In absence of substantial P-diversity, the use of  principal components analysis (PCA) is considered to be justified. Because of the interrelatedness of many of the variables quantified in this study, principal components analysis (PCA) was an appropriate statistical technique to describe overall data structure and provide some insight into the underlying causes of the data structure. When used in an exploratory fashion, P C A does not involve some of the strict parametric assumptions of other tests, such as random sampling. Further, P C A does not make assumptions about the distributions of variables (which is useful in analyzing ecological data), although the analysis is degraded with decreasing normality. P C A does,  27 however, make the assumption that relationships among pairs of variables are linear (Tabachnick & Fidell 1996). The linearity assumption was considered reasonable in this study owing to the relatively homogeneous forest conditions examined. PCA (and other multivariate data techniques) have been used in previous studies to describe bryophyte distribution patterns (e.g. Kenkel & Bradfield 1986, Rambo & Muir 1998a) To investigate the influence of site characteristics, a PCA was performed on the plot averages of environmental variables (landscape, stand structure, vegetation, soils, substrata), which were grouped by site using 68.3% confidence ellipses (the default ellipse chosen by SYSTAT, approximately equal to one standard deviation). This environmental PCA was compared to a PCA using plot averages of microplot bryophyte cover estimates. For both PC As, variables or species which were recorded in fewer than 5% of plots were deleted from the data set to clarify relationships. The amount of variation among plots in sites was interpreted by the size and distinction of confidence ellipses on the PCA axes. The relative amounts of plot variation in sites demonstrated by the two (environmental and bryophyte) PCAs were used to speculate about the importance of site-level characteristics to bryophyte composition and abundance. To investigate plot-level patterns of bryophyte diversity and abundance, canonical correlation analysis was employed.  This analysis examined how different groups of  environmental variables (landscape features, stand structure, vegetation, soils, and substrata) correlated with overall patterns of bryophyte diversity and abundance. The strength of the correlations between groups (environment and bryophyte) could be compared, and interpreted using relationships inferred by the Pearson correlation coefficients between individual plot variables. To investigate patterns of distribution and abundance of individual bryophyte species at the plot level, a PCA of plot averages was performed. The summary axes of this PCA were correlated with environmental variables, and the positions of species in relation to these axes and to other species were used to interpret plot-level assemblages. To investigate microscale patterns of bryophyte distribution and abundance, environmental preferences, substratum affinities, and the combined influence of these factors were investigated. The environmental range of tolerance of each species was determined by calculating the average and standard deviation of all environmental variables from the miroplots where that species occurred. Species which showed less than 5% plot frequency  28 were excluded from this analysis; the apparent restriction to a narrow range of environmental conditions may have been misleading. To further investigate microscale relationships between species and environmental factors, the Pearson correlation coefficients between species coverages and environmental variables were determined. For these correlations, information for all of these variables were averaged for each quarter of the plot; of the eight microplots, the two laying on the same axis from the center point were grouped together. This effectively reduced the number of data by one half, while retaining more sensitivity than using plot averages would permit. This procedure therefore reduced the amount of variation present in the data set, while still allowing for the interpretation of microscale relationships. Bryophyte substratum affinities were determined by examining their relative frequency on different substratum types. The relative frequency of each species on each substratum was computed as the number of microplots a species occurred on a substratum divided by the total number of microplots containing that substratum. If a species had less than 5% relative frequency on a substratum (or if the substratum itself had less than 5% frequency in microplots), it was described as "infrequent" on that substratum, as it was unknown whether the species was generally rare, or if its primary substratum had not been sampled. To investigate the combined influence of environmental preference and substratum affinity on bryophyte assemblages, PCAs of different substratum types were performed. This analysis used quarter-plot averages of data (as previously described). The correlations of the PCA axes produced with environmental variables were calculated. The positions of species in relation to the summary axes, and in relation to other species, were used to investigate how bryophyte assemblages on the same substratum type may differ, depending on habitat characteristics. To clarify relationships, only species which had greater than 5% relative frequency within quarter-plots were considered in the PCA analyses of substratum types. In addition, substrata that had less than 5% frequency overall (exposed roots, rocks) were not analyzed because they were inadequately represented. Logistic regression (LR) was used to investigate the predictive relationship between environmental features and bryophyte species presence or absence at the microscale. LR is a technique which allows the prediction of a discrete outcome, such as group membership, from a set of variables that may be continuous, discrete, dichotomous, or a mixture. LR is a  29 more flexible test of predicting group outcomes than many other analytical techniques used for this purpose, such as discriminant function analysis. It has no assumptions about the distributions of predictor variables; they do not have to be normally distributed, linearly related, or of equal variance within each group (Tabachnick & Fidell 1996).  These  characteristics made LR an appropriate technique to use in analyzing this data set. For these analyses, quarter-plot averages for data (described previously) were used. Separate LRs were performed on each species, using landscape, vegetation, and substratum variables to predict presence or absence of individual bryophyte species in quarter-plots. Variables which had predictive significance in predicting species presence or absence were noted. For all of the LR analyses, the Rho-squared value (analogous to an R in a multiple 2  regression), the success gained over a random model for different categories, and the total correct placement achieved by the model were of particular interest.  30 CHAPTER III Results and Discussion 3.1 Influence of Study Site  r  3.1.1 Biogeoclimatic Characteristics All sites surveyed for this project were located within the subalpine forest subzone (described as the "forested Moist Maritime MH subzone" by Pojar et al. 1991). Brooke et al. (1970) have provided detailed descriptions of relationships and sequences of ecosystem units within this forest subzone, based on plant associations which reflect the integrated effects of climate, topography, and substratum characteristics. These associations are illustrated in Figure 3.1.1. Pojar et al. (1991) have condensed this information to distinguish four common associations within the forested M H subzone. Since vascular plant associations reflect the integrated effects of many environmental variables, these units may be related to bryophyte community structure. Although it was not the purpose of the present study to examine the bryological differences between such biogeoclimatic units, the classification of study sites into defined associations may be useful in characterizing the general ecological properties of sites.  An understanding of these  properties may contribute greater insight into the interpretation of patterns of distribution and abundance of bryophytes. The purpose of this section is, therefore, to characterize the differences between sites using stand structure, vascular plant species composition and coverage, and substratum data, and to integrate this information with that available from Brooke et al. (1970), and Pojar et al. (1991).  Subsequently, speculation is presented  concerning how these larger-scale differences seem to influence the bryophyte component of the vegetation. Summaries of plot landscape features, soil characteristics, substratum availability, stand structure, and vascular plant characteristics for each site are shown in Tables 3.1.13.1.5, respectively. Site 2 (Hollyburn) has been divided into two sub-sites because of the obvious heterogeneity among plots which was noted during field sampling. Abbreviations for all species and variables contained in these (and succeeding) tables are explained in Appendix 1. To summarize the patterns of correlation among landscape, soil, substratum, vascular plant, and stand structure variables, a principal components analysis was performed  o  o o  CQ o  o <u c o  N -O =3 00 »  oo  <u o 45  S3 3  co >^ on O O <U 00 S3  o  45  w CT3  Q  3 WO  32 Table 3.1.1. Summary of landscape characteristics for the six study sites. Values shown are plot averages or microplot averages (mp). 2b  3  4  5  6  1  2a  #plots Asp (deg-N)  12 130.4  2  3  8  7  6  7  133.0  80.0  136.6  23.6  28.5  22.7  23.4  87.0 28.2  62.9  Slope (deg)  155.6 18.2  mp - avgcan  82.0  81.2  64.5  81.0  81.8  76.6  71.2  mp - avgvas  17.8  20.1  48.3  30.1  49.8  41.7  58.4  mp - htvas(m)  0.35 4.92  0.41  0.88  0.62  0.57  0.73  8.00  11.00  5.13  0.68 9.71  8.17  8.00  2.4  7.5  4.7  3.3  2.4  6.7  1.0  %minsoi  0.8  0.0  0.0  0.6  0.0  0.0  %water  2.3  0.0  0.0 5.0  0.5  5.0  1.0  0.7  %logs %litter  11.2 69.2  10.0 85.0  4.0 36.7  6.9 79.4  12.3 90.7  5.2 85.8  1.9 71.4  vascdiversity %outcr  15.4  Table 3.1.2. Summary of soil data for the six study sites. Values shown are plot averages. 2b  3  4  5  6  133.0  80.0  155.6  28.5  22.7  18.2  136.6 23.4  87.0 28.2  62.9 15.4  0.104  0.101  0.106  0.102  0.102  0.097  0.094  95.19  96.27  91.45  94.12  94.49  93.37  88.96  1.229 78.87  1.308  0.928 102.94  1.309 76.72  1.244  74.03  1.477 62.34  78.36  1.050 87.29  1  2a  Asp (deg-N)  130.4  Slope (deg)  23.6  H20wt(g) %CLOI TKN C/N ratio pH-LFH  3.9  3.7  4.1  3.6  3.7  3.7  3.7  pH-A/H  3.8  3.9  4.2  3.9  3.5  3.8  3.6  pH-B  3.4  4.0  4.6  4.1  3.7  3.9  3.8  3.8  3.6  3.8  3.7  40.1 14%  27.2 67%  17.7 29%  AVGpH LFH(cm) defhor  3.9  3.9  4.3  46.1 0%  61.3  51.3  22.9  50%  0%  63%  Table 3.1.3. Summary of substratum data for the six study sites. Values shown are microplot averages of fine litter (FLI), woody debris (WOOD), exposed humus (EXH), creeping stems (CRS), exposed roots (EXR), and rock (ROCK). %FLI %WOOD  1  2a  2b  3  4  5  6  74.9  82.9  81.4  90.6  89.8  88.9  92.0  19.7  10.8  7.5  6.4  8.0  7.3  5.9  1.6  1.5  2.8  0.2  %EXH  4.6  6.3  7.8  %CRS  0.3  0.0  3.3  1.4  0.7  0.9  1.7  %EXR  0.3 0.2  0.1  0.0  0.0  0.0  0.0 0.1  0.0  0.0  0.0 0.0  %ROCK  0.1  0.1  -a c  (-; o o o o ~ o o o ' . o o o o o o o o o  > O o  \<n  O  O  O  O  O  O  O  O  O  VO  O  Nrdddd--(NTj c>dd(N :  c  o  CN > o  r-  C3 i-T  O  — O  O  ' d  d  d  Q  O  O  O  ^  J  O  o d d o ^  O  d  O  J  ^  d  o o o ^  °  o »c , o o o o o o o m ' . ^ o o d - ^ d d d d o © © f - i ~ i 2 d d o |  u o >-.  <u o o o o d d d d  OH  o <u  o o o o  5)  M>  ci  >  I C\ I Tf  O  ©  O  O  O  O  O  o o o o o o ^ r o o T j - o o c o  O  O  O  C  N  O  O  O  O  O  O  O  >  n  O  O  (  N  O  O  O  O  O  ^  O  O  O  O  O  r  n  .  O  O  d d d d o o o o o ^ d d o l  O  to  e o o ") o o " ! ^  "cl  m  f  "! "! ©  <-i  a  ^j— —  r-  > c  o  s  O T r r r r - o o r - 4 0 0 m o o  o o o  1-  —  o o o o o  O  O O  O O ^ O O O  —  i ~  O O O O m O O O O (  N  0  O  TT  _  O  O O l  c o o d  o o o o c d d © d  o d  >  ^ I  OT  O m  O O  O O  O O  O O  O O  v T  O T  O  "1  O  o  4)  O O O O O O T O O O O O O  r  "1 " V o o o o o o o  O  Q (N r2 K ° — OO  1  O O r ^ O O O O O O O  r  o o o d d d  I  n  ^  d  O O O O O O  p O  p p p r O O O  O  O  O  O  o o o o o  n  3  r- cn m r~- m l O  •5 J2  I TJI O  O  o o o o o r - r - r - c \ O T j - t j - o O  O  O  O  c  M  C  M  O  O  O  O  O  O  t  N  —  O  O  O  O  o r- o O  O  O  o  O  O  O  O  O O O M D T J - O O O O  o o o o  O O O V O ^ O T J - O  3  0  T  f  O  O  O  O  O  O  O  I T T O O O O O O O  d d „  '  \0 &  — J2 <  &  = JS k. t l < <  o  C> G  '  0  0  0  0  0  0  '  0  0  0  0  Q  d o d d d d d o o o o  o o o o o  m —  O  O O  O O  O O  O O  O G<z>  T  J  ro H  o  oo ?  rr\ <D  • O  O  o o o o o o  —  O  O  O  -  p  6  (N O  O  O  r-i q oo d  O  O  q d  O  O  0  O  O  O  '  O  o o o o o T  J  -  O  O  O  O  O  O  .  o o o o o  o  O  O  O* —  C\  U-l o o o o  CN  Tf'  d d  d  q o  o o w n o o o ^ o ^ f o o o l  O  £ §. 3. S. !  11! | H  w w  tfl  w H  0 I H > > - > > > 5  8  34 Table 3.1.5. Summary of stand structure data for the six study sites. Values shown are plot averages of tree seedlings (SE), saplings (SA), pole trees (PT), medium trees (MT), large trees (LT) and very large trees (VL). 1  2a  2b  3  4  5  6  A v g # plot S E  30.6  24.5  31.0  26.1  42.9  47.0  71.4  A v g # plot S A  22.3  21.5  46.7  42.9  28.0  26.8  18.9  A v g # plot P T  10.9  13.5  12.7  16.4  8.6  8.3  8.7  A v g # plot M T  7.8  6.0  3.7  8.6  4.1  3.7  4.1  A v g # plot L T  2.8  5.0  1.3  3.0  3.3  3.2  3.1  A v g # plot V L  4.3  2.5  1.3  2.6  3.7  1.8  2.1  % S E Abam  83.7  81.6  34.4  55.0  74.3  58.9  23.2  % S E Chno  4.4  14.3  22.6  35.4  8.7  17.4  35.6  % S E Tsme  12.0  4.1  43.0  9.6  17.0  23.8  41.2  % S A Abam  65.9  30.2  45.0  47.5  55.6  67.7  36.4  % S A Chno  18.0  37.2  27.1  44.3  10.2  10.6  26.5  % S A Tsme  16.1  32.6  27.9  8.2  34.2  21.7  37.1  % PT Abam  67.9  59.3  47.4  63.4  33.3  46.0  41.0  % P T Chno  9.9  29.6  5.3  25.2  5.0  4.0  16.4  % P T Tsme  22.1  11.1  47.4  11.5  61.7  50.0  42.6  % M T Abam  54.8  58.3  27.3  53.6  51.7  45.5  34.5  % M T Chno  17.2  8.3  0.0  5.8  3.4  4.5  0.0  % M T Tsme  28.0  33.3  72.7  40.6  44.8  50.0  65.5  % L T Abam  57.6  50.0  0.0  33.3  60.9  5.3  45.5  % L T Chno  15.2  10.0  0.0  12.5  8.7  5.3  0.0  % L T Tshe  2.9  0.0  0.0  0.0  0.0  0.0  0.0  % L T Tsme  27.3  40.0  100.0  54.2  30.4  89.5  54.5  % V L Abam  30.8  0.0  0.0  4.8  26.9  9.1  6.7  % V L Chno  26.9  20.0  0.0  14.3  23.1  18.2  6.7  % V L Tshe  1.9  0.0  0.0  0.0  0.0  0.0  0.0  % V L Tsme  42.3  80.0  100.0  81.0  50.0  72.7  86.7  using plot averages. Figure 3.1.2 illustrates the position of plots in relation to the summary axes produced; site is used as a grouping variable (68.3% confidence ellipses), and as a character label. It is evident that most of the plots within a site are more similar to each other than to plots from other sites in terms of stand structure, vegetation, and soil variables; however, considerable overlap among site characteristics also occurs. Site 2 (Hollyburn) seems to show the most variation; this is not surprising, as the site was noted to be distinctly heterogeneous during field sampling. The principal components produced from this analysis (Table 3.1.6 a,b) can be used to interpret what the first two PCA summary axes are describing. In general, the first axis seems to represent a transition from closed to open habitats; it is negatively correlated with canopy coverage and LFH organic matter content, and positively correlated with vascular plant abundance. The second axis seems to be more  35 related to substratum characteristics; it is negatively related to C:N ratio in the LFH layer, and positively related to LFH depth and pH. Figure 3.1.2. Positions of plots on PCA axes produced from landscape, soil, substratum, vascular plant, and stand structure data. Plots are grouped by site using 68.3% confidence ellipses (sites 1-3, Cypress study area; sites 4-6, Seymour study area).  -  2  -  1  0  1  2  FACTOR(1)  Table 3.1.6 (a,b). Correlations (factor loadings) of variables with Axis 1 (a), and Axis 2 (b) produced from a PCA of landscape, soil, substratum, vascular plant, and stand structure data. a) Axis 1 AVGVAS XPHYLEMP TMSE PHYLEMP CLADPYR CNSE XCASSMER LEUTPEC XGRASS XLEUTPEC CASSMER XRUBUPED CARESP XCLADPYR DVASC HTVASC  b) Axis 2 0.795 0.741 0.740 0.688 0.680 0.670 0.658 0.635 0.633 0.621 0.617 0.601 0.587 0.566 0.514 0.486  % Total Variance Explained  CRS XTSUGMER XVACCSP TMSA VACCMEM AAVL CNVL LFH AAPT CNMT H20WT AAMT WOOD TASP %CLOI AVGCAN  0.484 0.440 0.439 0.392 0.337 -0.309 -0.342 -0.388 -0.390 -0.426 -0.461 -0.528 -0.566 -0.613 -0.687 -0.778 15.0  TKN STRESP CLINBOR LFH PHLFH XCLINBOR XSTRESP XBLECSPI ATHYFEL VERAVIR XRUBUPED DVASC AASE AAVL %LOGS BLECSPI  0.790 0.617 0.560 0.547 0.545 0.487 0.486 0.449 0.447 0.442 0.438 0.427 0.427 0.414 0.411 0.397  % Total Variance Explained  NCNGERM SORBSP %WATER WOOD XABIEAMA AAPT TMMT AHOR XMENZFER TMLT CNSE XCHAMNOC CNSA RHODALB CNPT CNRATIO  0.357 0.332 0.327 0.325 0.315 -0.332 -0.374 -0.396 -0.410 -0.414 -0.427 -0.498 -0.566 -0.578 -0.599 -0.795 10.5  36 Figure 3.1.3 illustrates the position of plots in relation to the same PCA summary axes, but in this case the vascular plant associations of sites are used as a grouping variable (68.3% confidence ellipses), and as a character label. It is evident that while there is some overlapping in vegetation within the association groups, each seems to occupy a distinct position on this PCA graph. The following summarizes site characteristics - as determined from site averages and the PCA of plot information - in relation to their biogeoclimatic associations. Figure 3.1.3. Positions of plots on PCA axes produced from landscape, soil, substratum,  vascular plant, and stand structure data. Plots are grouped by plant association using 68.3% confidence ellipses (1 = Amabilis fir - Mountain hemlock - Twistedstalk (AMT) association, 2 = Mountain hemlock - Amabilis fir - Blueberry (MAB) association, 3 = Mountain hemlock - Amabilis fir - Mountain-heather (MAM) association).  01  -2  -  1 0 FACTOR(1)  1  2  1. Amabilis fir - Mountain hemlock - Twistedstalk (AMT) association  Site 1, Site 2a, and Site 4 seem characteristic of the Amabilis fir - Mountain hemlock - Twistedstalk association as described by Pojar et al. (1991). This association represents moist and nutrient-rich to nutrient-very rich sites. association occurs on lower and mid-slopes.  Topographically, this  The habitats have a hygric to mesic  moisture regime and a weak to strong temporary seepage influence. Habitats of this  37 association are often free of snow 2-3 weeks earlier than adjacent habitats. In many instances, this is related to the relatively warmer southern exposures that the unit occupies in combination with steepness of slope; these factors combine to decrease snow accumulation and increase the length of the snow-free period. Soils are deep and include Brunisols, Podzols, Regosols, and Folisols. Humus forms range from raw humus and fine-textured greasy mor to moders. Well-defined Ae horizons are uncommon in soils of this site association. The deep soils, and ample moisture and nutrient supply of this site association promote excellent tree growth. •  Site 1 - Strachan Mountain, Site 2a - Hollyburn Ridge Plots in these sites are characterized by a closed canopy dominated by Abies  amabilis, with codominants Tsuga mertensiana and (to a lesser extent) Chamaecyparis nootkatensis; Tsuga heterophylla is infrequent. The LFH (organic layer) is quite thick (average 53.9cm), and differentiation in the " A " soil horizon differentiation is very infrequent (7%). There are high amounts of woody debris (logs and coarse litter) present. Vascular plant coverage, height, and diversity are low. •  Site 4 - Mystery Peak Abies amabilis and Tsuga mertensiana dominate plots in this site almost equally.  All sizes of Chamaecyparis nootkatensis are relatively infrequent. Plots have a typically open canopy, although abundance of woody debris (both logs and coarse litter) is high. LFH depth is quite high (40.1cm) and differentiation of the A horizon is not common (17%). The vascular plant community (especially herbs) is more diverse, and attains moderately high abundance and height. Vascular species present include Streptopus spp., Clintonia uniflora, Rubus spectabilis, Athyrium filix-femina, Tiarella spp., Blechnum  spicant, and Veratrum viride.  These species seem to indicate moist, nutrient rich  conditions.  2. Mountain hemlock - Amabilis fir - Blueberry (MAB) association  Site 3 seems characteristic of the Mountain hemlock - Amabilis fir - Blueberry association as described by Pojar et al. (1991). This association represents nutrient-very poor to nutrient-medium sites. These sites occupy benches, upper slopes, and broad  38 ridges. The humus of these soils is an extremely acid, thick, mycelial root mor, greasy mor, or root mor, and contains much decayed wood. Pronounced humps of organic material are characteristic of the ground surface of the habitat. The humus has a high water-holding capacity and rather slow drainage.  Eluviated Ae horizons are well  developed and may be up to 15 cm thick. Dominant soils are Ferro-Humic Podzols. Some of the soils can be gleyedfromtemporary seepage during early summer. The soils remain wet during fall and winter reflecting the heavy rainfalls and wet winter snow pack. •  Site 3 - Black Mountain Plots in this site are characterized by a closed canopy dominated by Tsuga  mertensiana, with codominants Abies amabilis and (to a lesser extent) Chamaecyparis nootkatensis. C. nootkatensis seedlings and saplings are frequent.  The LFH layer is  relatively thin (average 22.9cm), and the A horizon is frequently differentiated (63%). Abundance of woody debris is low. Vascular plant coverage, height, and diversity are low, although the herb and shrub layer are more developed than Site 1. Typical species are Rhododendron albiflorum and Rubus pedatus.  3. Mountain hemlock - Amabilis fir - Mountain-heather (MAM) association  Sites 2b, 5, and 6 seem to fall within the Mountain hemlock - Amabilis fir Mountain-heather association as described by Pojar et al. (1991).  This association  represents slightly dry and nutrient-very poor to nutrient-medium sites that occur on and around rock outcrops of ridges and upper slopes, or on upper to lower slopes where bedrock is close to the surface. Soils are deep, very stony, generally well drained, and range from Folisols to Podzols. Organic accumulations are thick, often greasy in character, and have a high water-holding capacity. Canopy coverage is low, and there is a very dense shrub layer. Brooke et al. (1970) describe this as the Cladothamno (pyrolaeflori) - Tsugetum mertensianae association, and distinguish two variants.  The cladothamno-tsugosum  mertensianae variant (perhaps represented by Site 2b and Site 5) occurs in areas where bedrock is exposed near the surface. Podsols and weakly gleyed equivalents are  39 characteristic. Following heavy rains, temporary seepage may be present in the soils, and they remain moister throughout the year than the other variant.  The vacciniosum  alaskensis variant (perhaps represented by Site 6) occurs on exposed rock outcrop knolls and has the driest moisture regime of any forested unit in the forest subzone.  The  shallow soils dry out slightly earlier than other associations, which is related to earlier snowmelt. •  Site 2b - Hollyburn Ridge The plots at this site are conspicuously different from other Hollyburn Ridge plots  (Site 2a). Placement of these plots on the PCA scatterplot seems to suggest that they are more similar to Site 5 in terms of vegetation, aspect (mainly northern), and substratum characteristics. The stand structure is dominated by Tsuga mertensiana and (to a lesser extent) Abies amabilis.  Chamaecyparis nootkatensis seedlings and saplings are very  frequent. LFH depth is quite high (51.3 cm) and there is no differentiation in the A horizon. Woody debris is not abundant. Canopy cover at this site is quite open, with a well developed vascular plant community in terms of abundance and height. Vaccinium spp., Menziesia ferruginea, Phyllodoce empetriformis, Cladothamnus pyrolaeflorus, Sorbus sitchensis, and Luetkea pectinata are typical species present.  •  Site 5 - Seymour Park Trail Plots in this site are dominated by Tsuga mertensiana, with Abies amabilis and  Chamaecyparis nootkatensis as lesser codominants. Plots are open, and vascular plant coverage, height, and diversity are substantial. Woody debris (logs, coarse litter) is not abundant.  LFH depth is relatively low (27.2cm) and the A horizon is usually  differentiated (67%). The vascular plant community attains moderately high abundance, diversity, and height.  Typical species are Vaccinium spp., Menziesia ferruginea,  Phyllodoce empetriformis, Sorbus sitchensis, and Cladothamnus pyrolaeflorus.  •  Site 6 - Dinkey Peak This site is similar in most respects to Site 5.  differences.  However, there are a few  Site 6 has a more north facing aspect than Site 5 and a lesser slope.  40 Although the stand is dominated by the same species as Site 5, there is a much higher frequency of Chamaecyparis nootkatensis seedlings and saplings. The LFH layer is thin (17.7cm) although the A horizon is typically not differentiated (29%). Vascular plants attain high abundance and height. The species at this site are similar to Site 5, with increasedfrequencyof Cassiope mertensiana and Luetkea pectinata; Sorbus sitchensis is  less frequent. Plot 44 seemed to be an outlier, so it was not included in the principal components analysis; this plot obtained the highest vascular plant coverage of all plots, a high proportion of which was grass. 3.1.2 Distribution and Coverage of Bryophytes in Sites A total of 42 terrestrial bryophyte species were identified from the Cypress and Seymour study areas; 26 mosses and 16 liverworts. A list of abbreviations for species names is provided in Appendix 1. Coverage and plotfrequencyfor individual species (overall and by site) are summarized in Table 3.1.7. Table 3.1.8 illustrates bryophyte a-diversity at the plot level and at the microplot level. Site level diversity (y-diversity) is also shown. The Simpson similarity index shown in Table 3.1.9(a,b,c) reflects the number of species that sites have in common, based on presence/absence data. To investigate the existence of compositional turnover along environmental gradients (i.e. p-diversity) a detrended correspondence analysis (DCA) was performed on the bryophyte data, for both microplot cover estimates and plot averages of species. In both cases, the plot-level DCA results contained no gradients of sufficient length (over 3.0 standard deviation units) to demonstrate the occurrence of species turnover. In the absence of substantial P-diversity, the use of PCA with the bryophyte data was considered to be justified. It is evident (Table 3.1.8) that bryophytes are more diverse in sites of the Seymour study area (y-diversity). Plot and microplot bryophyte and moss a-diversity is also higher for Seymour study sites. Sites 5 and 6 (the M A M association) seem to be particularly diverse at the plot and microplot level for both mosses and liverworts. Site 5 has the highest liverwort plot and microplot a-diversity of any site, while Site 6 has the highest moss plot and microplot a-diversity. Site 1 has the lowest a-diversity of mosses and liverworts of any site at both the plot and microplot level. Site 4, which has been described as belonging to the  c o3  o  03  "cL _c < ou  _  04  OO — ON — CN —  o  o  o  o  •—  i v - i T T f O i n r N r o  o  TJ- ©  > o  11(7)  <D fc 3 O O  o ' o >> oc <u  O  o  O  -H  — O O o o o m  o  00 CN  C-J  —  O o  O O O O o o o o  1-  r—O f N o  o  o  O r ^ O r N O O O ^ f  o  o  o  O  o  o  © r j - v - > © © w - i  3 •« -  -  ro — o o o o o  O o  m  o  (N  o  o  —  o  r  o  o ^ n o  o  O  ©  O  o  CO  o  o  O  ^  O f N O o o ^ o o f N ^ r o o - ^ f o c N O o o v o ^ r o o o o o o o o o - ^ r o o o o o o o o o o o o o o o o o o o o o o o o o o o o o  fN » 0 O O CN O © © © fN o o o o  f N O CN *0 O r o O W) o o o — o o o o o o o  O —  — © r- © — o v i o o o  o o  o  o  o  o  O ^O  O  O  o o  o  —o ©  w-i r - o  o1  o  o  m o o o  o  o  o  3  a"  to  O  Tf  o o  o  ro  Tf  C 03 1/3 -4—»  n(7)  _o a, o o i—  O  1/3  •°  o  2 -°  c-  vi o  o  in o  o  o  o  CN  O  (N (N >C -  O M l  v~t  (  N  ^  o  o  o  o o o o o ^ o o m o o u O O O O O f N O O O O  D O o o O O O O o  o  r-  -rr  o  o  o  o  o  o  © — © © O u"t  o  o  - i o O O o  o  o  o o m — o  o  o  o  —  O  o o o ^ r o o N O O N O O o O O O P - O O O O N © © ©  o  o  o  O O —  o  o  —  o  o  o  o  O O fN p o o  o  o  o  o  o  o  i  i  o  o  —  r N o r - o o © c N © r - c N O  o  o I  O o  '  O O <3 o ' o -  Tf  u - i T f T j - c f N O — o o © r - o r - o o ^ r T f o o © c N O o ^ r — 0 \ © ^ 0 0 0 © W " ) © © 0 © o I o o o o o o o o O O O O O O O O O O O © © © © © v o o r - o o o m o T t © © © o o ' o ' o ' o - - o ' o ' o o o o o o © o o o o o o o o o © o © © © o o o o o r — © O © I  ©  -  ro —  T*  O 00  O  o  ©  o  o  m  O  c  N  m  o  o  ©  o  o  o  ©  o  o  o  O  P  -  f  N  O  ©  r  -  i  © © —  —  —  OO —  O  O  a m  c|  — O O  3  3  — o o o o  ro  oo  > •=!  03  Tf  O fi O  T  e  c  O  O  Ul  T3  (L)  Tf O O >0  O O O r o O «") — m O © — Oo O—O 00 O O O O p C N O p O O O p p p O r ^ O O O O O O O O f N O O O O O O O O O O  C N © 0 0 \ 0 0 0 0 p p O O O O O O O O O O O O O  0 t n  O  r o o o r - \ o o © o o « / - i ~ O O f N f ^ p p p © © © © © © © © o o ^ J  — o o © © © o o o  o 1/1 <D l03  O O r N O r O f N r o O O r o -  — O  O— — O  O— C  N  ©  ©  r  o  O  O  O  O  O  — — O © (N © © — © (N r o O — O  c •S © ©  p o o d  O d  p d  N d  O d  O o  O f  f N O O O O p p p p p f N O O ^ N o d d d o ' d c J o d d d d  p p —  O  p p c " ) — © O p p O T f p r N O o p — © G o o o o c D < z > < ^ < z i c S c ^ S S c i G ^ { c S c S < 6  (D  "S °O-  >. o _D  CL  S  — 0  0  0  <  N  ©  f  N  O  r  N  O  O  (  N  O  ©  ©  0  0  — —  O  O  f  N  O  O  O  O  O  f  N  O  O  ©  fN © © O N O O O O m O O O O O f N O O O o O O C T v O m © O O — O f N O O O O m O O O O p o O O — p r o © d o o o o d d d o d o o ' o o ' d d d — o o d  oo  ti—i U-i  Bs  O O f N O f N © ( N O O C N O  d  ©  O O O O O fN © O o o O © © © © r o o o o o o o o © O O O O O O O O © © © © © o d © © o o o o o d r o  © © © o o o © © ©  S  3  r-  o  CN  n(l  3  CO  f  0  0  0  0  0  _  ;  0  -  O — O ~  O fN  © © O O O r o O f N O O " 3 - © 0 0 0 0 0 \ 0 O O O O O O O O O - ^ f O O O O O f N O © d d d d d © o d < N d d d © o o o  o  r f  o  o  ©  ~  o  O  o  O  o o r - t N O O ON o o o o — o o o o o o o o © o o © © © o o o o o o o o  I  H  O fN r o  o. ^  a  «j «  2  o  © © o  o  r - o o o o o — © o o © O O O O ro o © o o o N O O o  o  ~  = 3  . •§  o  fi. .S"  u  -  -  5  2  g •& 5i 3 •= ™ S = S S t 3 g g SB s S,€€Q.Q.Q.^-  .a  j j J ? O a , b o . a . b a . a , b b a . o . E i : 3 : o : X B ! K v i « «  42  Table 3.1.8. Diversity of all bryophytes, mosses, and liverworts across sites in terms of total number of species (y), average numbers of species within plots (p-oc), and average numbers of species in microplots (mp-a). Cypress Study Area  y bryo y moss y Ivrt a p-bryo a p-moss a p-lvrt a mp-bryo a mp-moss a mp-lvrt  SE * * * 0.54 0.26 0.43 0.20 0.12 0.13  avg 21 11 10 8.42 4.42 4.00 2.57 1.60 0.97  avg 13 6 7 11.50 5.00 6.50 3.50 2.13 1.38  Site 3  Site 2b  Site 2a  Site 1  avg 19 9 10 11.33 5.33 6.00 3.38 2.08 1.29  SE * * 1.50 1.00 0.50 0.64 0.30 0.46  * * 0.33 0.33 0.58 0.47 0.26 0.32  avg 22 11 11 10.25 5.13 5.13 3.23 2.20 1.03  SE * * * 1.70 0.78 0.96 0.30 0.18 0.19  avg 42 26 16 11.07 5.80 5.27 3.60 2.26 1.34  SE  SE * * * 0.90 0.23 0.85 0.24 0.12 0.17  Seymour Study Area  y bryo y moss Y lvrt a p-bryo a p-moss a p-lvrt a mp-bryo a mp-moss a mp-lvrt  Site 6  Site 5  Site 4  avg 28 16 12 12.00 6.86 5.14 4.05 2.55 1.50  avg 25 13 12 14.17 6.67 7.50 4.81 2.75 2.06  SE *  * * 1.54 0.67 0.99 0.30 0.18 0.19  avg 25 16 9 12.71 7.57 5.14 4.39 2.86 1.54  SE * * * 1.33 0.72 1.12 0.42 0.19 0.35  All Plots  SE * * * 0.52 0.26 0.35 0.12 0.07 0.09  Table 3.1.9 (a,b,c). Comparison of (a) #bryophyte, (b) #moss and (c) #liverwort species composition across sites using the Simpson similarity index. a) Similarity Index - Total Bryophyte Species 1 2a 2b  3  4  5  6  *  2a 2b 3 4 5 6  0.62 0.48 0.65 0.63 0.53 0.48  +  0.60 0.59 0.46 0.52 0.46  b)Similarity Index - Moss Species 1 2a  * *  0.58 0.42 0.52 0.52  0.67 0.74 0.68  0.61 0.61  0.72  +  2b  3  4  5  6  0.54 0.32 0.47 0.39  0.50 0.60 0.59  0.45 0.52  0.71  3  4  5  * *  *  2a 2b 3 4 5 6  0.55 0.43 0.57 0.59 0.41 0.42  *  0.67 0.55 0.38 0.46 0.38  c) Similarity Index Liverwort Species 1 2a 2b  * * *  6  *  2a 2b 3 4 5 6  0.70 0.54 0.75 0.69 0.69 0.58  *  0.55 0.64 0.58 0.58 0.60  *  0.62 0.57 0.57 0.73  *  0.92 0.92 0.82  *  0.85 0.75  *  0.75  *  43 same association as Site 1 (AMT) has the lowest plot bryophyte a-diversity and the lowest moss and liverwort a-diversity of all of the Seymour study sites. It is interesting that Site 2, which was noted to differ drastically in terms of vegetation, does not seem to differ in terms of plot and microplot a-diversity, although site y-diversity is higher for Site 2b. The similarity indices shown in Table 3.1.9 demonstrate that the bryophyte species composition may differ to varying degrees between study sites. The similarity among sites in terms of liverwort composition is generally greater than moss composition, demonstrating that mosses are unique to particular sites. Sites 5 and 6 (of the M A M association) share the greatest number of total bryophyte and moss species.  Site 1 shares a relatively high  proportion of species with Site 2a and Site 4 (which all belong to the AMT association). Sites 2a and 2b, which were noted to differ in terms of vegetation and topography, do not seem to differ greatly in their bryological component.  Site 3 (MAB association) has a  relatively high similarity to all study sites when total bryophytes, moss species, and liverwort species are considered. Perhaps this association represents a kind of transitional environment between the closed forests of the AMT association, and the open, vegetation-abundant habitats of the M A M association. Such an environment may produce a mixture of species that is generally similar to both the AMT and M A M associations. To examine more adequately the relationship between site characteristics and bryophyte compositions, a PCA was performed using the eight microplot estimates for coverages of individual bryophyte species averaged to the plot level. To clarify patterns of correlations, species that were recorded in less than 5% of plots were deleted from the data set; this reduced the total number of bryophytes from 42 to 23. Figure 3.1.4 illustrates the position of plots in relation to PCA axis 1 and 2, using site as a grouping variable (68.3% confidence ellipses). In contrast with the previous PCA in which stand structure, vegetation, and substratum characteristics were analyzed (Figure 3.1.2), it is evident that the confidence ellipses overlap more extensively. Plots 21 and 44 were noted to be outliers in terms of bryophyte characteristics, and so they have not been included in the PCA. When sites are grouped by biogeoclimatic associations, differences in bryophyte characteristics become more apparent. Figure 3.1.5 illustrates that bryophyte assemblages of the AMT association and the M A M association differ the most, while the bryophyte characteristics of the M A B association are intermediate, sharing characteristics with both  44  Figure 3.1.5. Positions o f plots on P C A axes produced from bryophyte data (plot averages). Plots are grouped by plant association using 68.3% confidence ellipses (1 = Amabilis fir -  Mountain hemlock - Twistedstalk ( A M T ) association, 2 = Mountain hemlock - Amabilis fir Blueberry ( M A B ) association, 3 = Mountain hemlock - Amabilis fir - Mountain-heather ( M A M ) association).  2h  10  +3  +3 CM  af O 1-  0  LL  -1  o <  -2  -  +3  2  -  1 0 1 FACTOR(1)  2  3  45 groups. This result is in accordance with what was speculated using similarity indices (Table 3.1.9). It should be reiterated that since this study did not set out to examine the bryological differences between biogeoclimatic associations, the representation of association units may not be adequate. The small amount of variation along either axis of the MAB association may be simply due to the fact that the lowest number of plots were sampled for that association. On the basis of this analysis, it is evident that the separation of bryophyte assemblages at the association level seems to be fairly one-dimensional; the confidence ellipses only occupy different spaces along the first PCA axis. The environmental variables which correlate the most with this axis are LFH organic matter content and canopy coverage (Pearson correlations of -0.699, and -0.659, respectively), and average plot bryophyte coverage (Pearson correlation of +0.764).  Using the correlations of different bryophyte  species with the PCA axes (Table 3.1.10), it is evident that in the more open habitats of the M A M association, species such as Barbilophozia floerkei, Dicranum pallidisetum, Lophozia  guttulata, and Rhytidiadephus squarrosus may be more prominent. Conversely, in the closed AMT association, the most prominent species may be Lophocolea heterophylla. Table 3.1.10 (a,b). Correlations (factor loadings) of species with Axis 1 (a), and Axis 2 (b) produced from a PCA of bryophyte plot averages. b) Axis 2  a) Axis 1 DICRPAL  0.819  PLEUSCH  0.311  HYPNCIR  0.597  BRACLEI  -0.268  BAPvBFLO  0.811  DIPLTAX  0.303  LEPIREP  0.387  RHYTLOR  -0.270  RHYTSQU  0.803  HYPNCIR  -0.291  BLEPTRI  0.320  PLEUSCH  -0.288  LOPHGUT  0.759  LEPIREP  -0.304  LOPHGUT  0.315  RHYTROB  -0.311  RHYTROB  0.437  LOPHHET  -0.451  RHYTSQU  0.262  ANTICUR  -0.576  CALYMUE  0.314  PSEUELE  0.260  PSEUBAI  -0.746  DIPLTAX  0.238  % Total Variance Explained  15.5  % Total Variance Explained  9.9  Although bryophyte assemblages seem to differ slightly between these two plant associations, it is difficult to determine whether this difference is related to an increase in the frequency of appropriate microhabitats, a physiological preference for site characteristics, or to even larger-scale differences in biogeography (such as position of study area). Site 4 (AMT association) seems distinctly different than Sites 5 and 6 (MAM association) in terms of bryophytes, which suggests that these association differences may be influential. On the other hand, Site 2b (on Cypress), which had vascular vegetation of the M A M association, did  46  not possess a bryophyte community similar to the MAM association plots on Seymour. This discrepancy suggests that large-scale differences between study areas (e.g. topography, climate) may have some important effects on bryophyte distributions. The information that can be gained by simply studying site characteristics in the interpretation of bryophyte distribution and abundance is limited in this study. The large amount of variation in bryophyte frequency and abundance that exists among sites and among associations indicates that within-site variation could be more influential to distribution. Vascular plant communities have been noted to be sensitive integrants of topographical and climatic features, which can be organized into recongnizable units. In contrast, these results demonstrate that bryophyte assemblages do not respond to the same degree to such large-scale factors. Their diminished size and the ubiquitous presence of readily dispersed propagules in diaspore banks and asexual fragments may contribute to the increased importance of micro-environmental conditions. Thus, the relevant ecological conditions for bryophytes may have to be studied at a much finer scale than is usually recognized for vascular plants. 3.2 Influence of Plot Characteristics on Bryophyte Distribution 3.2.1 Patterns of Coverage and Diversity  In the preceding section, bryophyte assemblages were noted to vary within sites and within zonal vegetation associations. This observation suggests that patterns of bryophyte distribution and abundance may be strongly dependent on smaller-scale influences, such as plot and microplot characteristics. The present section examines the information which can be gained from studying plot-level environmental characteristics. The relationships between plot environmental features and bryophyte coverage and diversity are investigated, and plotlevel species assemblages are described. The correlations in Table 3.2.1 may be used to speculate about the directions and strengths of ecological relationships in terrestrial subalpine systems, and how different environmental factors may combine to influence bryophyte diversity (i.e. species richness) and coverage. The large number of indirect relationships, and interrelatedness among variables makes it difficult to pinpoint the features that are most influential. To make sense out of the complexity which has resulted in the current ecological setting for bryophytes, it is  47 assumed that the most influential factors are those which directly influence growth - light availability, water and nutrient regime, and the physiological tolerances of individual species for substratum type and stability. Figure 3.2.1 represents a purely subjective interpretation of the main relationships, based on the correlations in Table 3.2.1. To examine such trends statistically, canonical correlation (CANCOR) analysis was used to relate plot-level bryophyte coverage and diversity to different  groups of  environmental variables (landscape and soil characteristics, substratum type, and vegetation characteristics). Groups of variables were separated in this way to examine their individual influence, and also to avoid high levels of intercorrelation in the analysis. The results of the C A N C O R analyses summarized in Table 3.2.2(a). The correlations of variables (bryophyte and environmental sets) with the canonical variates produced are presented in Table 3.2.2 (bd). Landscape characteristics are thought to be important in the establishment of bryophyte assemblages. Interrelationships between soil and landscape factors are complex, reflecting simultaneous changes in factors such as microtopography, slope, and soil stability. Table 3.2.2(a) demonstrates that landscape and soil variables, as a group, have predictive significance in determining both bryophyte coverage and diversity. In the present analysis, both canonical variates produced were significant using the X test for significance. It is 2  noted that landscape and soil variables have the highest Stewart-Love redundancy (i.e. there is the highest variance overlap between environmental and bryophyte data sets), and the highest overall canonical correlation. Table 3.2.2(b) shows that for the first canonical variate, bryophyte diversity and coverage are negatively related to organic matter content in the L F H layer, southern aspect, and L F H depth. For the second canonical variate, bryophyte diversity is positively related to T K N and slope, and negatively related to southern aspect. Bryophyte abundance displays the opposite response to these variables for this variate.  This observation agrees with the  findings of Ingerpuu et al. (1998), that nutrient fertilization diminishes the coverage of bryophytes. However, this comparison may not be appropriate; in the present study, different forms of nitrogen compounds were not distinguished in laboratory analysis of L F H samples; usually only nitrate (NO3) is considered in the generation of C : N ratios. A large component of the T K N may include high proportions of unusable compounds (e.g. humic acid).  00  CD  Tf  c a  43  •a 1-  e o o o 00  <u 60 > 03  oo  I >  -*—»  > c  c (D  fl 'o (U  o o c _o *-4—»  o o c o  oo D  o c  OS  fl  o 43 oo  w "2  os x  O S  49 Figure 3.2.1. Summary of interrelatedness of variables (plot averages). Pearson correlation coefficients (from Table 3.2.1) are shown. Inferred relationships that were not measured are indicated with dashed lines.  Aspect (degrees from North)  Potential Solar Radiation Temperature Amount of Precipitation Duration o f Snow Pack Soil Productivity/Stability  Vascular Plant Cover  Canopy Cover  4  (-) 0.632  (+) 0.647 /(+) 0.469 Vascular Plant  ^+10376  Organic Matter in L F H Layer (%C  L O  i)  Diversity J - ) 0.389  H_ Total Nitrogen in L F H Layer (TKN,  (-) 0.489  <-) 0.496  50  Table 3.2.2 (a-d). Results from canonical correlation analyses between environmental variable groups (landscape, substratum, and vegetation) and bryophyte coverage and diversity are summarized (a). Significance tests for the prediction of independent variables (Sig. Pred), the Stewart-Love canonical redundancy index, and X significance for canonical variates are shown. The canonical loadings on significant canonical variates (CV) are shown for different groups of variables: (b) landscape and soil (c) vegetation, and (d) substratum. 2  a) Comparison of environmental variable groups. L a n d & Soil  Vegetation  Substrate  0.704  0.491  0.458  Prob  0.000  0.003  0.006  Sig pred. AVGBRYO  0.000  0.003  0.010  Sig pred. NUMBRYO  0.006  0.047  0.036  Stewart-Love  0.468  0.302  0.284 0.006  R  2  X  1-2  0.000  0.003  X  2-2  0.036  n.sig  n.sig  0.763 0.540  0.627  0.619 n.sig  2  2  Can. Corr. variate 1 Can. Corr. variate 2  n.sig  b) Bryophytes and landscape/soil variables.  cvi  CV 2  0.920 0.607  -0.391 0.795 0.392  Bryophyte set  AVGBRYO NUMBRYO percent of variance redundancy  0.607 0.353  0.115  Environmental set  H20WT  -0.415  %CLOI TKN  -0.638 -0.488  -0.300 -0.264 0.728  AVGPH  -0.254  0.072  LFH  -0.581 -0.347 -0.601  0.140 0.518 -0.492  0.224 0.141  0.213 0.051  SLOPE TASP percent of variance redundancy  c) Bryophytes and vegetation variables. CV 1  d) Bryophytes and substrate variables.  B r y o p h y t e set  Bryophyte set  CV 1  AVGBRYO  0.928  AVGBRYO  NUMBRYO  0.591  NUMBRYO  percent of variance  0.605  redundancy  0.238  0.855 0.714  percent of variance  0.620  redundancy  0.238  Environmental set  Environmental set  -0.743  EXH  -0.334  AVGVAS  0.883  WOOD  -0.930  HTVASC  0.573  CRS  0.508  NUMVASC  0.243  EXR  -0.475  NUMGERM  -0.345  NSUBST  -0.214  AVGCAN  percent of variance  0.430  percent of variance  0.365  redundancy  0.145  redundancy  0.116  51 The relationship between TKN and bryophyte abundance may, therefore, be indirect; the closed canopies that promote organic matter deposition and a build-up of nitrogen compounds may, coincidentally, affect bryophyte abundance. It is not surprising that slope is negatively correlated with bryophyte abundance; high levels of slope are likely to have less stable surface conditions and, therefore, the existence of permanent bryophyte assemblages may be diminished. The positive correlation between bryophyte diversity and slope suggests that the movement of propagules among plots may be facilitated by rapid dynamics of the ground layer. Stand structure and vascular plant coverage may have many direct and indirect influences on bryophyte diversity and abundance. Bryophytes tend to occupy niche space that is left by the vascular plants which, in turn, can strongly influence many ecological factors, including microclimate (During & Van Tooren 1990). For example, Rincon (1988) found significant effects of various types of herbaceous litter on the vigor of five moss species. Vascular plant abundance may be related to bryophyte abundance for several reasons: it may be a more sensitive indication of light availability at ground level; it may promote abundance of bryophytes directly through increased nutrient throughfall; it may indicate areas possessing increased surface stability, permitting the existence of permanent bryophyte assemblages. Stand structure also may be directly related to bryophyte diversity, whereby closed stands may have a higher number of species incorporated into the ground layer via litterfall. Table 3.2.2(a) demonstrates that vegetation variables, as a group, have predictive significance in determining both bryophyte coverage and diversity. In this analysis, only the first canonical variate was significant using the X test for significance. It is noted that 2  vegetation variables have an intermediate Stewart-Love redundancy, and an intermediate canonical correlation, when compared to the landscape and substratum groups. Factor loadings for variables (Table 3.2.2c) demonstrate that bryophyte abundance and diversity are negatively correlated with canopy coverage, and positively correlated with vascular plant abundance.  These results suggest that light availability is important to bryophytes,  particularly in terms of coverage. The trend toward increased species incorporation through litterfall is not apparent in this study; diversity is highest in open stands.  52 The relationship between substratum type and bryophyte diversity and abundance has been investigated by various researchers. It has been shown that the presence of woody debris can influence bryophytes favourably; Franklin et al. 1981 (cited in Rambo & Muir 1998b) inferred that the greater amounts of woody debris and the deeper canopy of old growth generally result in higher humidity. Previous bryological studies have demonstrated the sensitivity of bryophytes to substratum type, and that different assemblages develop on different substrata (e.g. Soderstrom 1993). For many bryological studies (e.g. Lee & LaRoi 1979), it has been demonstrated that speciesrichnessis positively correlated with the number of substratum types available. For terrestrial bryophytes, the amount of woody debris in a plot may not directly influence bryophyte distributions. Rather, the amount of woody debris appears to reflect stand structure density, and correspondingly, the aspect of the plot. This variable may be related indirectly to bryophyte coverage because it reflects a combination of the effects of the aforementioned variables, as well as unmeasured factors such as surface stability. Canonical correlation results suggest that substratum variables, as a group, have predictive significance for both bryophyte abundance and diversity (Table 3.2.2a). However, the Stewart-Love redundancy (the variance overlap between bryophyte and environment variable groups) and the overall canonical correlation are lower than for other groups. The results of the present study contrast with those of previous studies; plot bryophyte coverage and diversity were found to have a negative relationship with abundance of woody debris (Figure 3.2.1) and with the total number of substrate types available (Table 3.3.2d). The discrepancy between expected and actual results may be related to the fact that only terrestrial bryophytes were examined in this study; epiphytic and epixylic bryophytes were not considered. Understanding the relative importance of factors controlling the distribution of bryophytes is complicated by the high correlation between individual factors. Overall, it appears that factors influencing bryophyte diversity are more complex and, therefore, more difficult to characterize than those that influence bryophyte coverage. The suspected large amount of variation in physiological tolerance among bryophyte species for specific ecological situations (e.g. particular substratum types) may weaken the relationships between environmental variables and bryophyte diversity and coverage. To fully understand patterns  53 of bryophyte distribution and abundance it is, therefore, important to investigate the relationships between individual bryophyte species and environmental characteristics. 3.2.2 Bryophyte Species Assemblages To investigate plot-level patterns of bryophyte distribution and abundance in absence of the consideration of site characteristics, a PCA of plot bryophyte averages (Figure 3.2.2, factor loadings graph) was interpreted by investigating the correlations of environmental variables with PCA axes (Table 3.2.3a,b). It is evident that organic matter content in the LFH layer, canopy cover, and aspect (degrees from North) are highly negatively correlated with axis 1, while bryophyte coverage, plot counts of Tsuga mertensiana and Chamaecyparis nootkatensis seedlings, and vascular plant coverage are highly positively correlated with this axis. The correlations for axis 2 are not as pronounced. This axis is weakly positively correlated with liverwort, moss, and vascular plant diversity, substratum diversity, Rubus pedatus coverage, TKN in the LFH layer, and Menziesia ferruginea plot frequency. Conversely, it is negatively correlated with C:N ratio, moisture content in the LFH layer, Vaccinium membranaceum abundance, and bryophyte coverage in microplots. The factor loadings plot shown in Figure 3.2.2 therefore can be used to illustrate the species assemblages one would expect to find in plots with different environmental characteristics.  The quadrants shown are useful in summarizing the communities of  bryophytes one would expect to find in: Quadrant 1: open-canopy areas that have a low C:N ratio and are species-rich Quadrant 2: open-canopy areas that have a high C:N ratio and are species-poor Quadrant 3: closed-canopy areas that have a high C:N ratio and are species-poor Quadrant 4: closed-canopy areas that have a low C:N ratio and are species-rich Figure 3.2.2 can be used to infer that Hypnum circinale and Lepidozia reptans are prominent in closed-canopy habitats that have increased vascular plant and substratum diversity and low C:N ratios.  Conversely, Brachythecium leibergii, and Lophocolea  heterophylla are typically more prominent in closed-canopy areas with lower vascular plant and substratum diversity and high C:N ratios. As evidenced by their weak association with  54 Figure 3.2.2. Correlation of bryophytes with axes produced from a PCA of bryophyte plot average data (plot averages). 1.0  1 Quadrant 1  I  Quadrant 4 HYPNCIR Q  0.5  LEPIREP Q  LOPHGUT  \  PSEUEDEtJ 1 PTIL5MLC|  PLAGLAEG^ P S E U S T E ^  O <  LOPHHETe^^  i^oSCAPBOL /RF W L O ^ X P L E U S C H  BRACLEI0  -0.5  IP  —  >s RHYTROB  ©ANTICUR ©PSEUBAI  Quadrant 3  Quadrant 2 1  1  -1.0  -0.5  0.0 0.5 FACTOR (1)  1.0  Table 3.2.3 (a,b). Correlations of environmental variables with Axis 1 (a), and Axis 2 (b) produced from a PCA of bryophyte plot averages. b) Axis 2  a) Axis 1 AVGBRYO AVGMOSS %GRASS %CASSMER CNSE %LEUTPEC TMSE %PHYLEMP AVGLVRT fCASSMER fCLADPYR AVGVASC %CLADPYR NUMMOSS fPHYLEMP fCARESP NUMBRYO fLEUTPEC CRS %VACCSPP %RUBUPED HTVASC %VACCMEM %TSUGMER FL1  0.764 0.710 0.680 0.678 0.678 0.677 0.675 0.626 0.590 0.530 0.511 0.509 0.450 0.444 0.376 0.367  %MENZFER fVACCSPP TMMT %LOGS AALT %CLINBOR fVERAVIR AASA fBLECSPI NUMGERM AAMT NTMSE AAVL CNVL SLOPE fSTRESP  0.335 0.334 0.327 0.296 0.291 0.261 0.222 0.221 0.219  NUMSUBS TKN WOOD H20WT fCLINBOR LFH TASP AVGCAN %CL01  % Total Variance Explained  0.205 0.205 0.201 -0.210 -0.210 -0.237 -0.257 -0.266 -0.268 -0.293 -0.293 -0.301 -0.305 -0.305 -0.320 -0.343 -0.347 -0.386 -0.396 -0.409 -0.409 -0.487 -0.509 -0.659 -0.699 1S.S  AVGLVRT fMENZFER NUMLVRT %RUBUPED TKN NUMBRYO fRUBUPED NUMVASC fSTRESP %GRASS NUMSUBS fCARESP PHLFH %LEUTPEC AALT fSORBSP  0.319 0.317 0.317 0.296 0.276 0.269 0.261 0.260 0.243 0.228 0.227 0.225 0.223 0.223 0.220 0.217  WOOD AVGPH TMSE %CLOI AVGCAN %MENZFER AVGBRYO TMLT TASP AVGMOSS fRHODALB CNPT H20WT CNSA CNRATIO %VACCMEM  % Total Variance Explained  0.210 0.205 0.200 -0.210 -0.213 -0.218 -0.223 -0.282 -0.295 -0.300 -0.301 -0.303 -0.317 -0.375 -0.384 -0.388 9.9  55 both PCA axes, species such as Pseudotaxiphyllum elegans, Ptilidium Plagiothecium  laetum, Pseudoleskea stenophylla,  Cephalozia lunulifolia,  californicum, Cephalozia  bicuspidata, Calypogeia muelleriana, and Scapania bolanderi seem to be frequent and/or  abundant in more than one type of habitat. In open-canopy areas which have a low C:N ratio and high vascular plant and substratum diversity, Lophozia guttulata and Rhytidiadelphus squarrosus are common. Conversely, Rhytidiadelphus loreus, Pleurozium schreberi, and Rhytidiopsis robusta are  typically more prominent in open-canopy areas with low vascular plant and substratum diversity.  The species Barbilophozia floerkei and Dicranum pallidisetum seem to be  prominent in any type of open habitat, as is evidenced by their weak association with the second PCA axis.  The species Antitrichia curtipendula and Pseudoleskea baileyi are  abundant in areas with low vascular plant and substratum diversity and high C:N ratios, while Blepharostoma trichophyllum is prominent in habitats which show low C:N ratios and higher vascular plant and substratum diversity. This section has investigated the relationships between plot characteristics and bryophyte distribution and coverage.  Although this analysis may describe the habitat  relationships of some species adequately, it may not be adequate for the interpretation of others. In particular, a species that has a narrow range of physiological tolerance in terms of environment or substratum may be less related to plot characteristics than to microhabitat conditions. If such microhabitat specificity exists, it may diminish the interpretative value of plot-level bryophyte characteristics.  The following section analyzes the relationships  between bryophytes and environmental variables at the microscale, and investigates the information to be gained by studying relationships at this level.  3.3 Influence of Microhabitat 3.3.1 Microscale Bryophyte - Environment Relationships Table 3.3.1 illustrates the microscale correlations of bryophyte species abundance with environmental variables (quarter-plot averages).  This table is useful in that it  demonstrates how species relate differently to certain environmental features. In addition to studying factors that influence bryophyte abundance, it also seems useful to study the features that may influence their frequency. By studying the range in microhabitat conditions  NO  uo  CO IU  60 o3 s-i CD  > 03 "cl I u  oo CO  "S =  o i m o o  r- M o M o o  00 VO T J ^ 'J o o O o o o o o o o d o"  o  « 3 (U  S;  s  1 '  o  fl > J O c u  o  o  o  i - M O I ro o — q p ; ' © ©  9  o  Tf 2^  T f ; oo r » ;? v> ro rO N ^ ^ ^ v O ^ O j ^ ^ ro Tf £ £ £ J r - ^ ) _ o P P P — — i — 0 P P 0 P 0 P P 0 q ' © © d o d o © o i d d p P d P d P P d  VO M rq p P d d  9  —'  -  idd9d©ddd©99 ' '  T3  C o3  vo ^ r; vo o ot 9  A A 03  J  00  ; o  U  ON »  ON  Tf  j  d  ON  5  ON r - vo ' vo •—• 22 vo ro ! o o P o o — P o 1 d d c> d d d  9  o  °  9  1  t  0  0  rON "o 22 r o o ON vo "o _ r o 2 O 00 © o — : o o P o ro © d d o o d d d d  9  o ^ r O O O ^ ™ > . r > . O O T f ^ * f ~ | ' i ' * ' O f f i O ^ ' " " o o P P o o o P P P P d d 9 S l d d d 0 P P P U  r  r  c  99  9  1  1  )  r  ,  l  N  r  n  0  ; © o © o ©d  > o — © o ; d d o o  6  o d o  S S ^ ^ ^ ^ S  , uo O wo r - i — o P — " T O O  —^ |s  ^ U0  i O © © © — O © O O © j p p o p P o o o o o 9 9 d 9 9 ' d d d d d  9  ;  ^  1  -  r-  r- —  ON ON  VN  90909  © TTff N ( _  (N  i—1 ^  1  OvrSr^vouorovOTfr^Tf0Tfrj  0  Sr- _ r - v o r o O O v o T f o o r N u o Z © © — © —p —— —p o p ©  — _ ro— r--oo —  ro  .r^  n  1  t  r-  oor--roON_-worof>-  P —P P — o ™ - ; o o P o P P P _ © © o © © ' © o o © © ' © © © o o © '  — — 2 Id o P 009900009 ©9  •~~  vo(s r~-ONTf'^ r-- ^ * — T f r j O N r - ^ O N O U o r j o ^ r o r O T f r o — ro — ~ © © — P o P P P P P 9 O © 9 9 — — 0  M  N  -  ,  IJ  1  d d ° d d d 9 d 9 ° ° ° °  ro P  O  dd9  t r ) C T v  ^ o o o o N O O T f — 1 — — r: S  r  0  Tf <  =  r-. 00 00 o o" 0  ro — —  ^ J  •— OA 'O ^ N r~- ' O ^ * ^ N ON * — — o P P o o  >>  s  O  CD  o  m  t  . •— r O f - O N  l ~ CC O VD ~ roo — o ! o P o o o P d d d d  ° — d  O„ © O O P — 0 P 0 P — o — o uo- ON O O o 6 d o d d 9 9 9 d 6 9 d d 9d9 0 0 0 0 0 0 0 0 0 9 ° ? ? ° ° roP  oo  r  (S ro — d  - • 2 « o O N T r O o _ o o o _ v o .  I  9 9  'o  <u  O > O > | T f ^ T f ^ ^ ^ - , — I— i / i O i / ) f f > i o — q q o P o P o o o q q ; q d p d d d  r  >  ' © Tf £: Tf  <U  -a  r - ON JCJ ° ) ~> — o . — — q . v _ J _ _ v _ J 0 P ° . — °. i © © d d d 0 0 ; o o ' o d d °  — p  p — w ^ M r o r o o o ^ r ^ ^ O r ^ u o ^ c w r ^ ' ^ M O v o u - , - - — ° <•? • - - ~ £ v o r - v o w - i < N v o e N , rt^-O — — 0 p ^ — — — < — —p q q P P P q P P P P P P ~ ; ~ : o o o o o _ • o o : — o o — — — ~ ~ : o P—• — c 0 0 c : c d d © ' © o o o o , ^ — ^ o o O O O o O ® ^ ® ^ — — ' d d o d d d d d  <D  o fl  ' § 5CT\  9  i ©  1/-,  0  0  (N  . ..  — q q o P P P ,rOONTfTff»,OTff», — 00 vo — r-> q q q q o q q o P o —o p P P P P P O '  oooooqqoqddd9  1  X> 1-  •  'JOVOTfONrOtNvOk -  P n  J  r N M ^ O o — • -• o o -  °  P  P  P ,-.  9999 9  c  P  0  P  2^voU.ror-Tf~—  —  1  o  P  o  i r ^ ^ w f ^ O N W O v 7 v v o r - - o O N r o r - - O N V O _ _ r o v o ro ON r-. , MMn ^ - - : o O T f r - „ w o o O ( O w T r N — O «• q q o —p ! p _ o P o o P — P q P P P ~ P - - P P - p; — m  O w O .— O M P p P P P p P _•d 9 d d d 9 d 9 9 9 d °  0  i 9 d © 9 d © 9 9 9 d 9  0  ? ?  0  c  '  0  0  0  0  c  * d d  .2  'o  P  CN  oo T dfd d d 9 ° fl o q  p p p P p cO O rO  ?  t  ^  Q  '  O  Q  O  O  O  O  O  O  O  O  O  O  O  1000 — 0 0 0 0 i d o o o o d o d 9 9 °  fc! ' 3 , c, ir> c « '1tr, o O  ON — M o  « CO  I NO VQ I ~ o  •9 6  c  '  0  '  0  c  1 r - TT ro o : p o p p o d ; o o o o  ' °  0  0  c  '  c  O  O  • TH  NO 00 (N 00 q q q U d d d H Z  =>  •<  ro 3 d H  0  rs ro q q d d  999  00d0  • i  O o  M d  P °  P ^  P 0  P 0  d  M  P 0  d  M 0  P M O d  U0  —  ^ f O O O V O o ^ f i  q o p P P o ~ : ~ : P O  Q  O  ; o o ; d d o  . ^ r ^ c o r - O O N ^ r s v o T f © o d  1 t"- ON ; •* vo o uo jo ON 1 • vo — 1 r* O I o o P P P o • q q q d d d ' q* d d d 1  O O O O  d  !00 0  '  03 co x PH  ?  , . J 00 uo 1 o o o  i v o O N O i T r o T r r o r o " g O N T r .- — r c 00 c . — J ° r ^ v— oovo P P ~ o n P o o P P o o  8  CNvo^ON  O S p 9 9 p 9 9 9 p 9 P 9 — p*5 — 9 9  O  O  Q  O  O  O  O  P P P P O O O t S f N O  — 0 '  O  0  d  <  Q 06 & U  1  < = ' ° = > ® ' = > ' = ' ' !  (  Tf iCNTrrviTriTNvuiw'CNc-H^Trr'i  009  O ro , —I ON r' ro P O O O  P o  o  0  ' ON • NO  , O — O O O O O o , n ON O o C — ' O ; d d 0 ' ° ° o ° ° ° c > , ^ o c ' d  M  O r O O O O O N — O N V O r o O O T f o f j Q r - ^ i - ^ r O f N f N — O r O — VO<N — (— ^ O f N ^ r N O q p p p p p p q q q q p o o P o o P o o o o o o © o © o o © c S o ' O o d  d  0  ;9  • *i r-.'  d  0  i T f V O o N O O r O ^ O ^ T f r O — r - O O r o o o O O T f U O — ON !, P ^ P ^ , !2 P^ d r ^ q g q S S o P P o — P P r t r t r t ! d d \ d> <6 <6 <6 d> d  1  d  0  : d  00 —0 0 P o o d  —P  9  : P o 1  9  d  —  CQ  z  J >  z O  H b h < W j a. cfl « S S S S S H H H H i -  2  ^ > S H  10  '  O  O  • <6 <6  • C O N O o © r O , ~ , < N , + c-p — UO — NO — — r s O \ T r r O — — — — <NfN — — r f rn p O p P p — ~ P © q © w w © 0 w ° O -O _ O : O ~ O ~O O O O O O c , 0 o d d q o q o ' P P d d d ° P 0 0 S ^ d S d  i  0  o o o O - o - H q o o r i H N q o ' ' d o d o ' o o o o d ' o P ' ' ? ' ? ' N  0  0  0  T j y - j O r O r o P c N r N  ©099©?©©  N 00 —  «  *  o o „ o  w  © d o d o  — P C J O P © © © © ©  — ©  . _  ^ o  w  w  ,  9 © ~ © —  —• —  «  —. —. —  n  „  o o © S o o o o  f r N o N O ^ r M ^ r - ^ W ^ O ^ C C r N  1  0  NO r o —• r— fN NO U0  0  X  P  9 ° ° ° ° 9  0  o"  c  — —  M  P  P  M  ^  P  ^»o22J^°  P © O P  — O  0  O —  — ' 1  r f W O O r o o r - O N ™  i i ; © o © © o  n  v  O N  0  " " "  0  1  p P p q p P p P p P P P  *  i  '  d  d  o  o  d  o  O  o  O  O  ^  1  — r - (N r o , — r o ,  ©— ©^ c0o 0r f  — —  O N f N U O v O T f ^ ^ r - O O O - T i?) ro ro r -  9 9 0 © ©  © © 0 0 © u O r - - 0 ^  (  v  '  T  .  1  I S S « £ S ' > 9 © © P P 1 i 9 d d 9 9  , © © o — ©  ? r " - ( N  0  NO NO  r o 00 00 © r - rN — T f  „  ! © © © © o o •• 9' 9" ' J o ©  0 0  <  r  r  !  (  N  ,  ,  0 ( V  O  o o o o <  00 ON O ro • - JN  O  P  P  ( N r s l  — O  d d © 9 9 d ? 9 9 dd99dddd99dd  NO — r^ —  ONr-uorxirsiO  p p r n —p o p —  pv]-  9900999009  1—1  O N O N O M f O f N r ^ o o O O p s i T f r - T t o o o O N O w o - t . r o — ^ r -  £ £ £ * o 3 q q q q — p p o P P f N p ro O O P P — o d o c i ^ ^ © © © © ' © © ' d d o 9 9 9  o o o o  0999  '**i""^^ M  1—1  —  09000  l  P. T f OO NO o o o o , ©9 9 9 ,  r— \ o r o r O O N © O N r o o o O j f r o o o u o u o T f r o O N O o " ® — O — — O © " © © " © " © " © ' © " © © -  1  uo — O N O U O N T N  9000000'  ——  o n - o - 9 9 "  0009900900999 O H  S i £ft;£ S c2 o u. a j o o Z £z  o s fe w Q > 3 < < Z (-  o o o o o  c  t> o ro uo uo * . l i ro S 2o © ^ ^t" 00 ro O "™ ^ 9S — <*i oo r— t Ty — 55 — — * ^ q q q p p p q P P — p p P P — — ^ P p p - ^ P P p S ——P M — P P P P o ! ? 9 9 9 9 9 d 9 9 d © © 9 9 © © 9 9 © 9 9 9 9 I_I  o P P P © " ^ © © © '  CO. co,  o —  c  0  0  0090099999900  1*1  — O _ — O  t N < , - , w w . - . - ' , © r- S ^o — S 5 : ( N _ 2 2 T f O N < N S r o S r o — r n o T r S s o - S S l S T t S ^_ _ N©O \qO q © o q q q p p o >— © a q P q P P o q P q p P P P q p l N p P P P P o ^ o ® 9 © o o ~ > © 9 © d 9 d 9 9 © © 9 © © 9 ' 9 ' > ' "© 9 9 9 ' 9 ' 9 © 9 d © 9 > 9 9 ® 9 d 9 9 9 d 9 0 9 9 9 o 9 0  — 9 — P o p p p - o™; — o " t 9 f * } © ' . ~ d 9 © 9 d 9 9 P ' d © p ' © d 9 ' d © © 9 d  9 d 9 ©  ,  * d d d d © d o ' d d o ' d d d o o o  q p o o p © p o ^©^ ©'^do'o'o'do'o'd  , _* i N w - ^ T T 'N O O r -IN J ' 'J' — M ^ M N O w C T i w h - -  O  © O O  o  —  o  © d d © d © d d d d d d  j p p p o P P o o P o P P o o P o P o © ; o p ' o ' o ' © 9 o d o © o ' o " d © P © P d ©  rN r - -  ; o ^ £ £ ^ u o o ^ _ ' - ! . »  0  M  • . —  ,  ^ S t ^ ^ ^ D J S ^ o o o o T f o — o — pq o o o q ^ q o S q - S o d d d d d o o o p p o p O o ' 0 0 ' 0 ' '  M  — — © o o o o o , © o d d . © o o o o  o d o o  . — r- — -  w  f— r s [— ( N O ^  — — r-  i A N O — O r s f O w v O i o _ q q q p q p u o p 0 < 3 J O © c i © o © © 0 ,  -  **} <s °. © o o  0  P  ^  <  ;d©'dd©'©©©©©qdq999©9©99  CO — 0O TT <N o \ NO N o o o - u - v o o r - w o — * o o ^ n — c O f f t O N O ^ NO UO g © u o © o O N O r o r o T f M r o « 3 w o O O O O O O O o O O O O O O © — O o  d d  1  o o o o ^ q © q o © q © q o © ©  0  (  oo ^1  — oo O ON — NO r o tN O _ © p ©  00 © r o — TT CN N \ D V I ~ t O C4 O O p O d> o d S <6 S  0  © © ' d ° ° 9 o ° 9 9 9 d d d d d 9 o 9 ° ° 9 9 9 9 ? 9 ° " 9 - ' ' >  — CO cN ^ ro O O o o o o o d © © d o © o d d  NO  • Q V , r - ro CN Tf —' — ^ v O r S t O ^O TO rO J u w ot O * ^ O a O rO^ or oO ^ O O O ~  —  .J 0. < X  H U  x u  U  N c < i « c a 0 M » O O S 5 S " * S E ; - S ^ n o r j o o o o o p p — : P P — p—:o © d © d © © © © 9 9 9 9 9 © 9 9 ° '  u CO Z y < Id E 3 < y SS u >< g > D ^ H33 •  <N r -  G ES|2O££1^  0©© ©  r - r-. £  N  ddddd999  9  0  O l a . u - a a, a. S 2 «  < <zzzz z < <o uuuo  00 r o 00 o 9 P p p — p  990090  58 under which a species occurs, one may gain a better understanding of its range in physiological tolerance. The averages and standard deviations of environmental variables for each bryophyte species are listed in Appendix 2. For each species, all of the measurements were considered from the microplots where the species was present. This information allows one to consider the varying ranges of tolerance of different species.  For example, Pseudotaxiphyllum  elegans shows a relatively narrow range with respect to the C:N ratio and pH of the LFH layer. Hypnum circinale and Lepidozia reptans are found only within relatively small ranges of %CLOI in the LFH layer, and canopy coverage. Conversely, species such as Rhytidiopsis robusta, and Dicranum pallidisetum show a wide range of tolerance (shown by the large standard deviation, and the greater difference between maxima and minima) for most of the variables considered. One must be careful in interpreting these ranges in reference to the true range of tolerance of these species, as only terrestrial environments were examined. In particular, it should be noted that the presence of some of the less frequent species may have resulted from chance dispersal from the appropriate substratum (i.e. they may be more frequently epiphytic or epixylic). In this case, the apparent restriction to a very small range of environmental variables is potentially misleading. For this reason, microplot averages for the least frequent species (i.e. present in fewer than four microplots) have not been included in Appendix 2. The PCA analyses in previous sections have focused on how plots differ in terms of environmental variables and bryophyte compositions. It is likely that microplot variation has a large influence on species distributions. To investigate the relationships of bryophyte species to particular habitats, as well as overall species associations in these habitats, a PCA was performed on the microplot averages for the environmental variables where each bryophyte species was present.  This analysis is useful because it shifts the focus from  characterizing placement of sites, plots, or microplots on PCA axes to the direct placement of bryophyte species against an environmental backdrop. Although this PCA allows for more direct inferences regarding bryophyte preferences, it places more emphasis on the frequency at which a species occurs in a particular microhabitat rather than its overall prominence (abundance) in different ecological situations.  59 Figure 3.3.1 illustrates the position of bryophyte species on the first two axes produced from this PCA. Table 3.3.2(a,b) demonstrates the correlations of these two axes with environmental variables.  These axes seem to be associated with the following  transitions in environmental characteristics (from negative to positive): Axis 1: The transition from closed canopy, woody, wet areas with low vascular plant diversity and coverage, to open canopy areas with less wood, more fine litter, drier (more well-drained) soil, and higher vascular plant diversity and coverage. Axis 2: The transition from areas of high C:N ratios and more acidic LFH layers, to areas of low C:N ratios and less acidic LFH layers. It is evident that these axes are similar to those described in the previous analysis of plot-level bryophyte abundance. In both cases, the first axis is related to canopy coverage, and the second is associated with C:N ratios in the LFH layer. The fact that similar axes are produced is not surprising; environmental variables correlate similarly at the plot scale as at the microplot scale. It is the information gained about bryophytes that is of interest. The positions of bryophyte species on these axes may be used to provide insight into their breadth of ecological tolerance (frequency, rather than abundance, is emphasized). Those species that may be assumed to have a wider range of tolerance and occur in many environmental situations are those that have little association with one or both of the PCA axes. For example, Brachythecium leibergii and Dicranum pallidisetum are positioned  closest to the center of the PCA graph (low association with both axes), suggesting that they are likely to be frequent in a variety of ecological situations. A noticeable feature of this PCA graph is that most species negatively correlated with the first axis (canopy coverage) do not have a strong relationship with the second axis (LFH characteristics). In particular, species such as Hypnum circinale, Lophocolea heterophylla, Plagiothecium laetum, Ptilidium californicum, Blepharostoma trichophyllum, and Lepidozia  reptans show little correlation with the axis describing LFH characteristics. It seems that under closed canopies with abundant woody debris, bryophyte assemblages possess greater tolerance to unstable, shifting substratum conditions. If this is true, bryophytes in this environment may be able to persist only by clinging to conifer needles or woody debris, or by existing on small patches of exposed humus. If few bryophytes are able to persist in  60 Figure 3.3.1. Positions of bryophytes on PCA axes produced from microplot averages of environmental variables where species are present. 1  -  •  r,  A#icur8  0  tyRRfiiLtn topnnet  p|  a g  |  a e  ?yraclei J T le  3t  i  u o Lepjjg^gHR^Riical Scapb^  0 y t  r o b  1  o Pseuele  Cepl Calymu  Pseuste  ,  °o  o Barbflo rpal »ophgut o° Dipltax o Pleusch— oRhytsqu  D l c  ubai o Pse  o Rhytlor 1  1  -  2  -  1  1 0 1 FACTOR(1)  2  3  Table 3.3.2 (a,b). Correlations (factor loadings) of variables with Axis 1 (a), and Axis 2 (b) produced from a PCA of microplot averages of environmental variables where bryophyte species are present. a) Axis 1 0.920 TMSE 0.904 NUMVASC 0.886 AVGVASC 0.873 CNSE %PHYLEMP 0.804 0.800 TMMT 0.687 %RUBUPED 0.669 %FLI 0.667 %LEUTPEC 0.653 HTVASCM 0.643 TMSA 0.617 DEFHOR 0.569 TMPTL %CLADPYR 0.451 0.369 %VACCSPP %CHAMNOO 0.362 0.323 TMLT -0.305 A ALT -0.320 AASA -0.337 SLOPE % Total Variance Explained  CNLT AASE AAPT %CLINBOR CNMT NCNgerm TKN %WOOD AAMT AAVL CNVL NUMSUBS NUMSEED TASP NTMgerm %CLOI H20WT AVGCAN LFH  -0.375 -0.433 -0.465 -0.469 -0.548 -0.564 -0.692 -0.698 -0.752 -0.770 -0.771 -0.780 -0.787 -0.807 -0.810 -0.879 -0.899 -0.916 -0.921 34.9  b) Axis 2 0.895 PHLFH 0.791 AVGPH TMVL 0.703 0.692 %STRESP 0.659 %LEUTPEC 0.611 TKN %ABIEAMA 0.605 0.459 %RUBUPED 0.445 %CLADPYR 0.394 NAAgerm 0.363 %CRS 0.354 AASE % Total Variance Explained  TASP %CLOI %RHODALB %VACCMEM DEFHOR CNMT CNPT TMLT %MENZFER CNSAL %CHAMNOO CNRATIO  -0.309 -0.348 -0.359 -0.389 -0.513 -0.518 -0.588 -0.628 -0.655 -0.670 -0.767 -0.924 17.5  61 patches on fine litter, it is not surprising that LFH characteristics are not correlated with the distribution of this group. Among the closed-canopy group that seem to have a relationship with the second PCA axis, Antitrichia curtipendula and Pseudoleskea stenophylla are more frequent in  conifer stands dominated by Abies amabilis, with occasional large Tsuga mertensiana and Chamaecyparis nootkatensis. The LFH layers in these habitats are less acidic and have lower C:N ratios. In contrast, Cephalozia lunulifolia, Dicranum pallidisetum, Scapania bolanderi,  and Rhytidiopsis robusta seem to be more frequent in habitats dominated by Tsuga mertensiana with Chamaecyparis nootkatensis understory. In this habitat, C:N ratios in the LFH layers are higher, and pH is relatively low. In contrast with closed-canopy environments, open habitats seem to be associated with a greater amount of variation among bryophyte species along the second PCA axis. In such open canopy environments with minimal slope and higher vascular plant coverage, it seems likely that the stability of the terrestrial environment would increase. In consequence, patches of bryophytes may be able to form more permanent assemblages on fine litter. In such circumstances, one might expect that the quality of the humus layer (in terms of pH and productivity) would be more relevant to distribution. The quality of overlying vegetation also seems to be important (although this effect can be indirect). The species Rhytidiadelphus squarrosus and Barbilophozia floerkei show greater  frequency in any type of open habitat (correlated only with the first axis). Conversely, Cephalozia  bicuspidata,  Calypogeia muelleriana, Lophozia  guttulata,  Diplophyllum  taxifolium, and particularly Pseudotaxiphyllum elegans are positively correlated with both the first and second PCA axis. These species may be more frequent in areas dominated by only a few large Tsuga mertensiana or Abies amabilis. The LFH layer in these habitats is generally thin and well drained, with a low C:N ratio, low organic matter content, and relatively high pH. Bryophytes that are more frequent in open habitats and are negatively correlated with the second axis include Rhytidiadelphus loreus, Pleurozium schreben, and Pseudoleskea  baileyi.  These species may be more frequent in areas dominated by medium and large  Tsuga mertensiana. The LFH layer in this habitat is generally thin, well drained, with a higher C:N ratio, low organic matter content, low total nitrogen, and a relatively low pH.  62 It was noted earlier that bryophytes that are mostfrequentin closed communities do not seem to be as physiologically restricted by the characteristics of the LFH layer as those that are in open habitats. This observation leads to the suggestion that substratum stability plays a major role in influencing community structure. If this is the case, one might expect that bryophytes may use substrata quite differently, depending on the microenvironment. The following section will examine the substratum specificity of bryophytes within different types of microenvironments. This analysis will be useful in separating the effects of environmental preferences and substratum affinity; these effects will be used to suggest what other factors may influence community structure. 3.3.2 Microscale Bryophyte - Substratum Relationships Bryophytes may differ in their usage of substrata in different types of habitats; the presence or absence of an individual species may depend on the substratum affinity and the range of environmental tolerance. This section will investigate the nature of bryophytes on substratum types in an effort to understand the relative importance of substratum availability and microhabitat characteristics, and how they combine to influence frequency and abundance of different species. Many bryophytes in this study occurred on more than one substratum type. Although different assemblages of bryophytes are apparent on different substratum types, it is important to realize that thefrequencyof these assemblages are sometimes quite low. In the analysis of bryophyte assemblages on substratum types, it is necessary to eliminate species that occur infrequently on a substratum type so that true relationships can be interpreted more readily. Table 3.3.3 illustrates thefrequenciesof bryophyte species on individual substratum types. The relativefrequencyvalues shown in this table reflect the number of microplots in which a species is recorded on a particular substratum-type, divided by the total number of microplots containing that substratum type. Studying the relativefrequencyof a species on a substratum improves interpretation of its affinity by reducing the bias caused by differences in substratum availability. Substratum affinity is determined as follows: primary substrata are those on which a species occurs at least twice as often as the next most common substratum. If a species has less than 5% relative frequency on a substratum (or if the substratum itself has less than 5% frequency  63 in microplots), it is described as "infrequent" on that substratum. This classification is made because it is unknown whether the species is generally rare, or if its primary substratum has not been sampled.  Table 3.3.3. Relative percent frequency of bryophyte species on substratum types. Relative frequency is calculated as the number of microplots in which a species occurs on a substratum divided by the number of microplots in which that substratum occurs. # Micro-  # Plots  Fine Litter  plots  Woody  Exposed  Creeping  Exposed  Debris  Humus  Stems  Roots  Rocks  Anticur  4  4  1.1  0.0  0.0  0.0  0.0  0.0  Barbflo Bleptri  33  14  8.3  0.8  2.8  0.0  0.0  0.0  35  24  2.8  2.8  25.4  0.0  0.0  0.0  Braclei Calymue  25  15  5.8  0.6  5.6  2.0  0.0  16.7  47  28  7.5  2.2  22.5  0.0  0.0  0.0  Cephbic  16  11  1.1  1.9  8.5  0.0  0.0  0.0  Cephlun  49  29  4.4  4.2  31.0  2.0  0.0  0.0  Dicrhet  2  2  0.3  0.0  1.4  0.0  0.0  0.0  Oicrfus Dicrpal  4  2  0.8  0.0  1.4  0.0  0.0  0.0  254  45  62.2  8.1  39.4  4.0  0.0  16.7  Diplpli Dipltax  1  1  0.0  0.0  1.4  0.0  0.0  0.0  18  12  3.9  0.3  5.6  0.0  0.0  0.0  Dryppat Geocgrav  1  1  0.3  0.0  0.0  0.0  0.0  0.0  1  1  0.3  0.0  0.0  0.0  0.0  0.0  Hetepro  1  1  0.3  0.0  0.0  0.0  0.0  0.0  Hypncir Isotsto  74  28  12.8  5.8  23.9  2.0  28.6  0.0  1  1  0.3  0.0  0.0  0.0  0.0  0.0  Lepirep  12  11  1.4  0.6  11.3  0.0  0.0  0.0  Lophgut Lophhet  35  17  5.8  3.1  11.3  2.0  0.0  0.0  100  40  13.9  8.1  42.3  4.0  14.3  0.0  Mniuspi  2  2  0.6  0.0  0.0  0.0  0.0  0.0  Oligpar  1  1  0.3  0.0  0.0  0.0  0.0  0.0  Plaglae Plagund  94  39  15.0  6.7  33.8  2.0  0.0  16.7  1  1  0.3  0.0  0.0  0.0  0.0  0.0  Pleusch  8  4  2.2  0.0  0.0  0.0  0.0  0.0  Pohlnut  2  2  0.3  0.3  0.0  0.0  0.0  0.0  Polyalp Porecor  1  1  0.3  0.0  0.0  0.0  0.0  0.0  1  1  0.3  0.0  0.0  0.0  0.0  0.0  Pseubai  64  27  13.6  3.9  4.2  42.0  0.0  0.0  Pseuste  12  11  1.7  1.4  0.0  2.0  0.0  0.0  Pseuele  10  6  2.2  0.0  1.4  0.0  0.0  33.3  Pterffl  1  1  0.3  0.0  0.0  0.0  0.0  0.0  Ptilcal  125  40  13.3  19.2  19.7  28.0  0.0  0.0  Racoeri  1  1  0.3  0.0  0.0  0.0  0.0  0.0  Racosud Rhiznud  2  2  0.0  0.0  0.0  0.0  0.0  33.3  3  1  0.6  0.0  1.4  0.0  0.0  0.0  Rhytlor  13  9  3.3  0.6  2.8  0.0  0.0  0.0  Rhytsqu Rhytrob  16  10  4.2  0.8  1.4  0.0  0.0  0.0  217  44  59.7  5.8  16.9  12.0  0.0  33.3  Scapbol  5  5  0.0  1.1  1.4  0.0  0.0  0.0  Scapdip Scapsca  1  1  0.3  0.0  0.0  0.0  0.0  0.0  2  2  0.3  0.0  1.4  0.0  0.0  0.0  64 Barbilophozia floerkei, Dicranum pallidisetum, and Rhytidiopsis robusta seem to  have a high affinity for fine litter, occurring on it at least twice as often as on any other substratum.  Blepharostoma  trichophyllum,  Calypogeia  muelleriana,  Cephalozia  bicuspidata, Cephalozia lunulifolia, Diplophyllum taxifolium, Hypnum circinale, Lepidozia reptans, Lophozia guttulata, Lophocolea heterophylla, and Plagiothecium laetum all have  exposed humus as a primary substratum. It is interesting that all but two of these species are liverworts; this substratum type seems to be very important for liverworts in subalpine terrestrial ecosystems. Pseudoleskea baileyi is the only species that occurs primarily on creeping stems of vascular plants (Menziesia ferruginea, Rhododendron albiflorum, and Vaccinium spp. in particular).  Dicranum pallidisetum, Hypnum circinale, Lophocolea  heterophylla, Plagiothecium laetum, Ptilidium californicum, and Rhytidiopsis robusta are  species which seem capable of existing on many kinds of substrata. 3.3.3 Combined Influence of Environment and Substratum Affinity To investigate the combined effect of substratum and microhabitat on the abundance and associations of bryophyte species, a separate principal components analysis was performed on the abundance of bryophyte species on fine litter, woody debris, exposed humus, and creeping stem substrata. For these analyses, the quarter-plot data were used. To clarify relationships, only species that had greater than 5% relative frequency within quarterplots were considered in the following PCA analyses of substratum types. In addition, substrata with less than 5% frequency overall (exposed roots, rocks) were not analyzed because they were inadequately represented. •  Fine Litter The factor loadings for the PCA of bryophytes on fine litter substratum (Figure 3.3.2)  demonstrate the existence of four distinguishable groups: 1. Hypnum circinale, Ptilidium californicum, Plagiothecium laetum, Lophocolea heterophylla 2. Cephalozia lunulifolia, Blepharostoma trichophyllum, Rhytidiadelphus squarrosus  65 3. Lophozia guttulata, Rhytidiadelphus squarrosus, Calypogeia muelleriana, Diplophyllum taxifolium, Barbilophozia floerkei, Dicranum pallidisetum, Rhytidiopsis robusta 4. Pseudoleskea baileyi, Brachythecium leibergii  Figure 3.3.2. Positions of bryophytes (factor loadings) on PCA axes produced from quarterplot averages of bryophytes on fine litter (needles and twigs <lcm in diameter). 1.0  I  I  0.5 0 BARBFLO PSEUBAI Q  or  g o <  BRACLEI o \ ^ ^ ^ Q R C A L Y M U E  0.0  _  \ ^\>~-3LEPTRI I^^i^-~s RHYTSQU © LOPHHETsePRtfUN  LL  A PLAGLAE  -0.5 PTILCAL6 HYPNCIR o  •1.0 •1.0  I  I  -0.5  0.0 FACTOR(1)  1.0  0.5  Table 3.3.4 (a,b). Correlations of variables with Axis 1 (a), and Axis 2 (b) produced from a PCA of quarter-plot averages of bryophytes on fine litter (needles and twigs <lcm in diameter). b) A x i s 2  a) A x i s 1 AVGBRYO  0.477  CNPT  -0.108  CNSE  0.244  AAVL  -0.103  NUMBRYO  0.454  AALT  -0.114  AVGBRYO  0.238  PHLFH  -0.107  TMSE  0.419  CNSA  -0.123  CNRATIO  0.226  NUMGERM  -0.112  %LEUTPEC  0.369  SLOPE  -0.128  NUMVASC  0.172  AVGCAN  -0.122  CNSE  0.309  %LITTER  -0.140  %PHYLEMP  0.163  %CLOI  -0.124  NUMVASC  0.302  AAPT  -0.149  CNSA  0.159  NTMgerm  -0.126  AVGVASC  0.291  CNVL  -0.155  CNPT  0.157  NUMBRYO  -0.142  %RUBUPED  0.221  CNMT  -0.175  TMLT  0.157  %OUTCR  -0.155  NAAgerm  0.198  NTMgerm  -0.176  TMSE  0.142  AASA  -0.186  HTVASC  0.131  %LOGS  -0.183  %MENZFER  0.140  SLOPE  -0.220  %VACCSP  0.130  AAVL  -0.187  CNLT  0.126  %LOGS  -0.236  %FL1  0.126  %WOOD  -0.206  HTVASC  0.123  TKN  -0.261  %ABIEAMA  0.120  H20WT  -0.236  AVGVASC  0.116  %CLADPAR  0.115  AAMT  -0.241  % T o t a l V a r i a n c e Explained  %PHYLEMP  0.114  LFH  -0.303  TMSA  0.107  AVGCAN  -0.365  CNLT  -0.106  %CLOI  -0.443  NUMGERM  -0.108  TASP  -0.457  % T o t a l V a r i a n c e Explained  13.3  10.4  66 The correlations of these species with the two PCA axes, and the correlations of the axes themselves with environmental variables (Table 3.3.4a,b) suggest that the groups of species exist under quite different circumstances on fine litter. Group 1 exists on fine litter in nutrient-rich sites (e.g. stands of Abies amabilis) that have poor bryophyte coverage as a result of larger amounts of woody debris (logs in particular), increased slope, and generally reduced light levels. The fact that these species are mostfrequent/abundantin areas of low bryophyte abundance, high slope, and low light levels supports the qualitative field observation that these bryophytes may commonly persist, or continually recolonize, on mobile fine litter (i.e. on individual conifer needles or very small bits of debris) in relatively unstable habitats which are intolerable for other bryophytes. For example, the frequency and abundance of bryophytes capable of forming carpets (Rhytidiopsis robusta, Dicranum pallidisetum) are negatively correlated with this group. The second group of bryophytes on fine litter is most prominent in north-facing, open, well-drained areas that have relatively high bryophyte abundance, high vascular plant abundance and diversity, and high frequencies of Tsuga mertensiana and Chamaecyparis nootkatensis established seedlings. This assemblage is most common in areas with low levels of woody debris. However, the amount of wood (logs), canopy coverage, and slope is slightly greater than for the following assemblage, as shown by its negative association with the second PCA axis. The third assemblage of bryophytes is differentiated by having a positive association to the first and second PCA axis. This group represents the bryophytes that reach maximum abundance/frequency on fine litter in the most open, north-facing, well-drained habitats with high C:N ratios. These areas have the lowest frequency of logs, the least degree of slope, and the maximum total bryophyte coverage. This environment thus seems to be relatively stable and conducive to permanent, abundant populations of bryophytes. The fourth group of bryophytes on fine litter is mostfrequent/abundantin somewhat more south-facing, relatively closed habitats which have low levels of slope, woody debris (logs), and TKN build-up in the LFH layer. Of the south-facing environments, this habitat seems to be the most conducive to the development of permanent bryophyte assemblages. These species, therefore, may be characterized as being the most prominent in stable gaps within forest stands. This group is somewhat negatively correlated with the second group;  67 however, the low correlation of these species with both PCA axes suggests a relatively low correlation with the aforementioned variables, and an increased probability that this group of species may be frequent in other types of habitats (i.e. more variable in distribution). The origin of the bryophytes within each of these assemblages may differ. The bryophyte assemblages in the most closed, unstable, bryophyte-poor habitats (group 1) might be expected to derive a large majority of species from the large quantities of litterfall and persist by clinging to needles, small twigs, or small portions of bark or branches. Alternatively, many of these species may be more typically epixylic.  The increased  frequency of logs in areas with higher tree density would aid in the persistence of small groups of epixylic bryophytes, such as those that survive on portions of debris from decaying logs. Because these species probably do not persist in one particular locality on fine litter, it seems less likely that they are influenced directly by vascular plants or substratum features. Alternatively, bryophytes within the most open, stable, bryophyte-abundant habitats (group 3) might be expected to be derived from other forest floor communities. In this case, the proximity of a species to an exploitable patch of fine litter may influence its distribution. Because bryophytes in this type of habitat likely persist in more stable patches, it seems probable that types of vegetation coverage and soil characteristics will influence them. The absence of negative correlations between species in these open habitats, and the large amount of habitat that was observed to be unexploited, suggests that competition is not occurring. Instead, proximity to a source patch and variations in physiological tolerance are probably more influential. •  Woody Debris The factor loadings for the PCA of bryophytes on woody debris (Figure 3.3.3)  suggests the existence of three groups: 1. Plagiothecium laetum, Hypnum circinale, Ptilidium californicum, Lophocolea heterophylla 2. Blepharostoma trichophyllum, Dicranum pallidisetum, Cephalozia lunulifolia, Lophozia guttulata 3. Rhytidiopsis robusta, Pseudoleskea baileyi  68 The correlations of axes with environmental variables (Table 3.3.5a,b) suggest that groups 1 and 2 are most prominent in areas that have a high frequency of newly germinated conifer seedlings (particularly Chamaecyparis nootkatensis).  This variable is likely  associated with proximity to seed-bearing conifers; the abundance and species of newly germinated seedlings may, therefore, be indirectly related to quantity and quality of litterfall. Epiphytic origin of appropriate woody substratum (e.g. Chamaecyparis nootkatensis woody debris), therefore, may be important to many of the bryophyte species on woody debris. Figure 3.3.3. Positions of bryophytes (factor loadings) on PCA axes produced from quarterplot averages of bryophytes on woody debris (1-1 Ocm in diameter). 1  I  9> PLAGLAE  0.5  BPBZRhlCIR LOPHHET  CM  /  of  g  0.0  o <  gLEPTRI  affltBMpAL  LOPHGUT  -0.5  4  n  "To  -0.5  0.0 0.5 FACTOR(1)  1.0  Table 3.3.5 (a,b). Correlations of variables with Axis 1 (a), and Axis 2 (b) produced from a PCA of quarter-plot averages of bryophytes on woody debris (1-1 Ocm in diameter). b) A x i s 2  a) A x i s 1 NUMBRYO  0.439  AAMT  -0.110  AAVL  0.207  HTVASC  -0.114  NCNgerm  0.301  AAPT  -0.115  AVGCAN  0.180  TMMT  -0.118  NUMGERM  0.194  CNRATIO  -0.120  TKN  0.179  TMSE  -0.120  %WATER  0.145  CNMT  -0.130  NUMGERM  0.171  NUMVASC  -0.126  %LITTER  0.131  CNSE  -0.131  %WOOD  0.168  DEFHOR  -0.127  TKN  0.120  CNPT  -0.143  NTMgerm  0.146  %RUBUPED  -0.128  SLOPE  0.119  CNSA  -0.153  %CLOI  0.130  CNRATIO  -0.134  AASA  0.118  TMMT  -0.159  H20WT  0.115  %CRS  -0.146  AALT  0.105  %LEUTPEC  -0.156  %FLI  -0.101  NAAgerm  -0.160  AVGBRYO  -0.102  AVGVASC  -0.165  %VACCSP  -0.108  NUMBRYO  -0.167  % Total Variance Explained  21.5  % Total Variance Explained  13.8  69  Group 1 bryophytes are more prominent in areas that have a higher frequency of large Abies amabilis and reduced canopy cover. The high correlation of this group with the number of newly germinated seedlings suggests an increased reliance on epiphytic origin. This group has the same composition as the "unstable" assemblage noted to occur on fine litter in dense, Abies amabilis-dominated forest (group 1, previous section). In an unstable, severe environment that does not promote bryophyte abundance, this group of bryophytes seems to be the most persistent. The second group of bryophytes is more abundant/frequent in areas that have fewer large Abies amabilis, and less nitrogen build-up in the LFH layer. This group is more prominent in areas that are more open; they are slightly less correlated to the number of newly germinated seedlings than the previous group. These characteristics support the suggestion that this group of bryophytes prefers somewhat more stable, bryophyte-abundance promoting areas. In comparison with the fine litter groups of bryophytes, this assemblage is comprised of a mixture of bryophytes from the most open, "stable" habitats (groups 2 and 3, previous section). It seems likely that some of the species on woody debris in these habitats may be derivedfromthe bryophytes that predominate on fine litter, rather than from litterfall. Rhytidiadelphus  squarrosus,  Calypogeia  muelleriana,  Diplophyllum  taxifolium,  Barbilophozia floerkei, Rhytidiopsis robusta are species that achieve maximumfrequencyor abundance on fine litter in open habitats, but show decreased prominence (or are absent) on woody debris. Perhaps this discrepancy is a result of lower physiological tolerance for woody substrata. Group 3 has little association with the first axis and seems to be more related to the frequency of large Abies amabilis, and the build-up of nitrogen in the LFH layer. This group is somewhat negatively correlated with the first group. In comparison with the bryophyte assemblages identified on fine litter in this type of habitat (group 4, previous section), Pseudoleskea baileyi is similarly prominent, but Brachythecium leibergii is absent. It is  interesting that Rhytidiopsis robusta, which is most prominent on fine litter in open habitats, is more common on the woody debris in closed areas. However, the decreased vector-length of this group suggests a relatively low correlation with the aforementioned variables, an  70 increased probability that this group of species may be frequent in other types of habitats (i.e. more variable in distribution). To clarify bryophyte - substratum preferences, it might have been useful to examine the decay class and species of the woody debris encountered. This could have permitted clarification as to whether the species present were growing on recent (undecayed) litterfall, or decaying bits of wood which had broken off from logs. An attempt was made to distinguish between grades of decayed wood during field sampling, but it was noted that the frequency of decayed wood (of any kind) was very low. Most of the woody debris less than 10cm in diameter (the maximum diameter studied) seemed to be derived from relatively intact litterfall - branches, splinters of newly downed trees, and bark. These categories were, therefore, merged into one category because of inadequate representation of the separate decay class categories. In addition to the distinction between decay classes, different sizes of woody debris (medium=l-5cm, coarse=5-10cm) were studied separately.  These categories have been  merged for this analysis for two reasons. First, it became apparent during field sampling that the bryophyte compositions on these two substrata did not seem to differ substantially. Second, the quality of the two substrata did not seem to differ. The increased coverage of bryophytes (or increased affinity) may simply have resulted from the greater surface area (and, thus, an increased probability for occurrence) associated with larger pieces of wood. •  Exposed Humus The factor loadings for the PCA of bryophytes on exposed humus (Figure 3.3.4)  suggest the existence of two distinguishable groups: 1. Rhytidiopsis robusta, Plagiothecium laetum, Calypogeia muelleriana, Cephalozia bicuspidata, Brachythecium leibergii 2. Lophocolea heterophylla, Ptilidium californicum, Blepharostoma trichophyllum, Diplophyllum taxifolium, Lepidozia reptans, Lophozia guttulata, Dicranum pallidisetum, Hypnum circinale  The PCA axes correlations (Table 3.3.6a,b) demonstrate that both of these groups are most prominent in areas that have a high availability of exposed humus substratum. The  71 Figure 3.3.4. Positions of bryophytes (factor loadings) on PCA axes produced from quarterplot averages of bryophytes on exposed humus (compacted, decayed organic matter).  '-1.0  -0.5  0.0 FACTOR(1)  0.5  1.0  Table 3.3.6 (a,b). Correlations of variables with Axis 1 (a), and Axis 2 (b) produced from a PCA of quarter-plot averages of bryophytes on exposed humus (compacted, decayed organic matter). b) A x i s 2  a) A x i s 1 %EXH  0.515  H20WT  0.102  %L1TTER  0.261  PHAH  -0.142  NUMBRYO  0.388  TASP  0.101  CNMT  0.155  %CRS  -0.145  NUMSUBS  0.359  %PHYLEMP  -0.109  AAPT  0.151  NUMVASC  -0.149  NAAgerm  0.215  HTVASC  -0.112  AVGCAN  0.123  AVGBRYO  -0.153  LFH  0.167  CNSE  -0.121  TASP  0.119  AVGPH  -0.164  %CLOI  0.129  %FLI  -0.164  CNPT  0.116  NTMgerm  -0.177  SLOPE  0.127  AAMT  0.104  TMSA  -0.189  TMPT  -0.106  %LEUTPEC  -0.338  TMSE  -0.113  % Total V a r i a n c e Explained  18.7  % Total V a r i a n c e Explained  12.0  availability of exposed humus was found to be the most influential factor in terms of bryophyte diversity and coverage. If the substratum is available, it is preferentially used and exploited by a variety of species. The favourability of this substratum is probably linked to its stability. Exposed humus was often noted to occur on the sides of abrupt drops (shelves), or on the down-slope-facing side of tree humps (i.e. mounds of organic material collected around tree bases). As such, exposed humus may present islands of stability in relatively unstable environments. It may also be favourable in open environments because of its high  72 moisture-retention ability. The occurrence of exposed humus, however, is related to the frequency of Abies amabilis germinants, LFH depths, organic matter composition, southern aspects, and higher degrees of slope. In consequence, it is not as likely to occur in the most open, north-facing, well-drained sites that have little humus layer build-up. Because of its favourability, the composition of species on exposed humus is likely dependent on what is able to get to it first. Of the species associated with fine litter and woody debris, only  Barbilophozia  floerkei,  Pseudoleskea baileyi, Rhytidiadelphus  squarrosus and Rhytidiadelphus loreus do not occur on exposed humus with a substantial frequency. However, this substratum does promote the occurrence of the species Cephalozia bicuspidata and Lepidozia reptans.  If exposed humus is available, the groups of species that occur on it are most related to the variables that promote or diminish the amount of overall bryophyte coverage (axis 2). The first group is most prominent on exposed humus in open areas with abundant bryophyte coverage and higher coverage by the vascular plant species Rubus pedatus (a low-growing species capable of existing in open patches under relatively closed canopies). Because this substratum is not frequent in the most open (north-facing) sites, it is likely that group 1 occurs mainly on exposed humus in canopy gaps within variably dense stands.  It is  interesting that the prominence of Plagiothecium laetum is associated with relatively more open areas on exposed humus than was noted for fine litter or woody debris. The slight positive correlation of the second group with the second axis suggests an association with areas that do not promote bryophyte abundance. These species, therefore, are slightly more prominent on exposed humus in closed habitats with low bryophyte coverage and low coverage of Rubus pedatus. The prominence of the species Lophocolea heterophylla, Hypnum circinale and Ptilidium californicum in these closed habitats is not  surprising, as they were found to be common in this habitat on both fine litter and woody debris. However, the prominence of the species Blepharostoma trichophyllum, Diplophyllum taxifolium, Lophozia guttulata, and Dicranum pallidisetum on exposed humus in the most  bryophyte-poor environments is unexpected. These species were the most common in open, bryophyte-abundant habitats on fine litter and woody debris. The absence of exposed humus from the type of habitat these species favour on other substrata may be one factor contributing to this discrepancy. In addition, the diminished  73 correlation of these species with the second PCA axis suggests that this group is not as influenced by factors contributing to bryophyte abundance as they are by the availability of exposed humus (axis 1). In essence, they may be prominent on this substratum any place where it has high availability. The fact that these species are able to occur on exposed humus in the most closed (and, most probably, unstable) environments supports the idea that patches of exposed humus provide islands of stability. These "islands" permit the growth of species that may be favoured by more open, stable environments when growing on other substrata. Perhaps, for these species, incident radiation and aspect are not physiologically limiting factors; rather, their persistence may be dependent on the availability of a permanent substratum. A continual supply of bryophytes associated with litterfall may be less important to the structuring of communities on exposed humus than was suggested for other substrata. •  Creeping Stems The factor loadings for the PCA of bryophytes on creeping stems (Figure 3.3.5)  suggest the existence of two distinguishable groups: 1. Rhytidiopsis robusta, Ptilidium californicum 2. Pseudoleskea baileyi  The PCA axes correlations (Table 3.3.7) demonstrate that these assemblages are not negatively associated with each other, although they correlate differently with the two PCA axes. The first assemblage is prominent wherever there is a high availability of the creeping stem substratum and increased bryophyte coverage (Tsuga mertensiana forest with low organic matter build-up). Pseudoleskea baileyi is also more prominent where there is more creeping stem substratum, but it is more related to vascular plant height (particularly Vaccinium spp.) and abundance than to bryophyte coverage. This difference in distribution may be attributed to the fact that creeping stems are the primary substratum of Pseudoleskea baileyi; as a result, it is most prominent when such stems are large and abundant. The other group of bryophytes, on the other hand, may be more incidental on creeping stems, occurring on them only when ground coverage by bryophytes is maximal. In this case, the proximity of creeping stems to abundant bryophyte growth may be more important.  74 Figure 3.3.5. Positions of bryophytes (factor loadings) on PCA axes produced from quarterplot averages of bryophytes on creeping stems (living stems and branches <5cm from the ground). 1.0  I  I  I  0.5 DU  O  \— 0.0  o<  -  PTILCAL o -  -0.5  -1.0 -1.0  I  I  -0.5  © PSEUBAI  0.0 0.5 FACTOR(1)  1.0  Table 3.3.7 (a,b). Correlations of variables with Axis 1 (a), and Axis 2 (b) produced from a PCA of quarter-plot averages of bryophytes on creeping stems (living stems and branches <5cm from the ground). a) A x i s 1  %CRS AVGBRYO TMVL  b) A x i s 2  0.373 0.145 0.132  % Total Variance Explained  CNSE %VACCMEM TKN  0.122 0.106 -0.102 54.3  %LOGS AVGCAN %WOOD AALT CNLT CNSE CNRATIO NUMVASC TASP  0.177 0.134 0.129 0.106 0.104 -0.102 -0.106 -0.11 -0.119  % Total Variance Explained  VACCMEM TMVL CNPT %VASC VACCSPP HTVASC CNSA H20WT %CRS  -0.123 -0.127 -0.148 -0.154 -0.184 -0.203 -0.21 -0.237 -0.432 33.3  3.3.4 Predicting Bryophyte Assemblage Characteristics The previous analyses have demonstrated that bryophytes differ in their distribution, depending on their relationships to environmental factors, substratum affinities, or a combination of both. Because the assemblages of bryophytes on fine litter and woody debris seemed to differ, depending on microenviroment, I wanted to test statistically the randomness associated with species distributions at the microscale. For the groups of species prominent in environments that have been suggested to promote a dynamic surface layer, I expected that  75 their distribution at the microscale would be more stochastic than for species more prominent in stable habitats, or which had strong substratum affinities.  The technique used to  investigate the degree of randomness to which species were distributed at the microscale was logistic regression (LR). In this analysis, LR was used to characterize the predictive relationship between species presence/absence and habitat features (landscape, vegetation, and substratum variables). Quarter-plot averages were used for this analysis. Table 3.3.8 summarizes the significance of each variable in predicting the presence or absence of some bryophyte species when a direct LR is performed on a full model (adjusted for quasi-maximum likelihood covariance).  It is evident that for different species, different variables are important  predictors of presence or absence.  It is also noted that the predictive significance of  environmental variables in logistic regression may not agree with their separate correlations with individual species (cf. Table 3.3.1).  The discrepancy relates to the covariance  relationships among variables which are taken into account in logistic regression. The high total correct placement estimate in terms of predicting species presence may be misleading in some cases. Many species were relatively infrequent; therefore, there were often an abundance of "absents" in the data sets used. For very infrequent species, the high number of "absents" in the data set may have weighted the correct placement in the absent (reference) category. Results of this analysis show that it is much harder to predict the presence than the absence of an infrequent species. As a consequence, in determining the overall success of a logistic regression, one should consider the success gained over a random model for both response and reference categories. The strength of the LR models in this case lies in their ability to demonstrate the predictive relationship between environmental variables and species distributions. The Rho9  •  ••  squared value shown is the equivalent of an R value in a multiple regression; it is a measure of the strength of association between the outcome and the predictors (i.e. how much variation in the outcome can be explained by the predictors). Because Rho-squared is often under-estimated, Tabachnick & Fidell (1996) speculate that a Rho-squared of greater than 0.2 is satisfactory. The Rho-squared values in this study seem to be related to the degree of success gained over a purely random model for each species.  76 Table 3.3.8. Summary of logistic regression analyses relating presence/absence (response/reference) of separate bryophyte species to environment and substratum variables. Values denote prediction probabilities (i.e. smaller values denote stronger prediction). Variables with significant effects (p<0.05) are shown in bold. Also shown are Rho-squared, correct placement in categories (correct res/ref), success gained over a random model (Succlnd res/ref), and overall predictive success (TotCorrect).  CONSTANT %CLOI TKN PHLFH LFH DEFHOR SLOPE TASP AVGCAN AVGVAS HTVASC NUMVASC NUMGERM %FLI %WOOD %EXH %CRS NUMSUBS Rho-squ pred res pred ref corr res corr ref succ ind res succ ind ref TOT CORR  BARBFLO  BLEPTRI  0.098 0.106 0.762 0.614 0.839 0.025 0.245 -1.028 0.316 0.570 0.245 0.214 0.883 0.765 0.072 0.021 0.824 0.634 0.407 26 154 0.462 0.909 0.318 0.054 0.845  0.637 0.930 0.292 0.331 0.822 0.176 0.231 0.904 0.551 0.705 0.838 0.148 0.128 0.347 0.499 0.143 0.380 0.793 0.087 33 147 0.254 0.832 0.070 0.016 0.726  0.363 0.362 0.010 0.179 0.136 0.526 0.030 0.546 0.375 0.581 0.348 0.582 0.839 0.048 0.767 0.040 0.114 0.313 0.176 24 156 0.264 0.887 0.131 0.020 0.804  LOPHGUT LOPHHET PLAGLAE CONSTANT %CLOI TKN PHLFH LFH DEFHOR SLOPE TASP AVGCAN AVGVAS HTVASC NUMVASC NUMGERM %FLI %WOOD %EXH %CRS NUMSUBS Rho-squ pred res pred ref corr res corr ref succ ind res succ ind ref TOT CORR  0.159 0.119 0.732 0.966 0.337 0.764 0.517 0.124 0.087 0.925 0.945 0.904 0.997 0.230 0.551 0.824 0.267 0.223 0.243 32 148 0.390 0.868 0.212 0.046 0.783  0.295 0.375 0.153 0.237 0.481 0.786 0.147 0.018 0.179 0.826 0.820 0.437 0.073 0.169 0.175 0.746 0.021 0.085 0.133 78 102 0.525 0.637 0.092 0.070 0.588  CEPHBIC  CEPHLUN  DICRPAL  DIPLTAX  HYPNCIR  LEPIREP  0.568 0.198 0.558 0.603 0.812 0.349 0.640 0.143 0.480 0.125 0.475 0.149 0.213 0.122 0.656 0.002 0.251 0.036 0.257 43 137 0.454 0.829 0.215 0.068 0.739  0.073 0.002 0.295 0.101 0.086 0.125 0.459 0.900 0.220 0.786 0.738 0.049 0.896 0.766 0.598 0.161 0.171 0.448 0.270 16 164 0.319 0.934 0.230 0.022 0.879  0.566 0.639 0.728 0.915 0.367 0.112 0.213 0.495 0.132 0.776 0.188 0.708 0.333 0.852 0.879 0.025 0.278 0.071 0.128 44 136 0.352 0.790 0.107 0.035 0.683  0.107 0.792 0.831 0.213 0.369 0.705 0.895 0.403 0.025 0.107 0.080 0.620 0.188 0.507 0.236 0.288 0.294 0.811 0.365 161 19 0.932 0.421 0.037 0.315 0.878  0.421 0.011 0.294 0.762 0.792 0.309 0.222 0.607 0.743 0.921 0.333 0.518 0.363 0.305 0.021 0.105 0.265 0.351 0.396 18 162 0.443 0.937 0.333 0.037 0.887  0.363 0.203 0.017 0.171 0.453 0.831 0.550 0.485 0.879 0.644 0.627 0.093 0.337 0.809 0.544 0.937 0.225 0.034 0.221 56 124 0.488 0.769 0.177 0.080 0.682  0.051 0.331 0.413 0.148 0.477 0.218 0.575 0.436 0.006 0.188 0.370 0.002 0.025 0.398 0.035 0.067 0.516 0.491 0.357 12 168 0.325 0.952 0.259 0.018 0.910  PSEUBAI  PSEUSTE  PTILCAL  0.787 0.716 0.703 0.859 0.041 0.042 0.239 0.005 0.235 0.522 0.342 0.847 0.968 0.586 0.053 0.255 0.415 0.055 0.281 51 129 0.502 0.803 0.218 0.086 0.718  0.977 0.276 0.342 0.981 0.643 0.989 0.954 0.505 0.254 0.637 0.300 0.668 0.160 0.122 0.977 0.341 0.520 0.879 0.174 12 168 0.177 0.941 0.110 0.008 0.890  0.901 0.834 0.107 0.644 0.108 0.162 0.099 0.081 0.299 0.958 0.404 0.381 0.153 0.006 0.910 0.274 0.107 0.722 0.114 95 85 0.596 0.548 0.068 0.076 0.574  BRACLEI CALYMUE  0.366 0.633 0.258 0.523 0.718 0.812 0.343 0.204 0.584 0.154 0.983 0.431 0.055 0.032 0.013 0.963 0.683 0.772 0.110 74 106 0.492 0.646 0.081 0.057 0.583  RHYTLOR RHYTSQU RHYTROB SCAPBOL 0.182 0.062 0.857 0.105 0.900 0.863 0.002 0.383 0.053 0.766 0.045 0.252 0.727 0.568 0.933 0.136 0.548 0.399 0.403 12 168 0.378 0.956 0.311 0.022 0.917  0.394 0.066 0.132 0.482 0.522 0.084 0.140 0.000 0.284 0.896 0.149 0.015 0.466 0.303 0.333 0.613 0.449 0.889 0.512 15 165 0.504 0.955 0.421 0.038 0.917  0.458 0.990 0.173 0.622 0.213 0.464 0.870 0.252 0.372 0.409 0.665 0.629 0.115 0.117 0.000 0.726 0.336 0.480 0.258 143 37 0.854 0.434 0.059 0.229 0.768  0.989 0.677 0.615 0.332 0.816 0.605 0.097 0.005 0.598 0.078 0.023 0.564 0.006 0.537 0.593 0.990 0.241 0.997 0.202 5 175 0.104 0.974 0.077 0.002 0.950  77 The logistic regression results can be compared with the results from preceding analyses regarding microhabitat variability. Lophocolea heterophylla, Ptilidium californicum, and Plagiothecium laetum were associated with "dynamic surface layer" habitats on fine litter and woody debris. It is noted that the LR models for these species have generally lower Rho-squared values and predictive success than for other species. The reduced ability of statistical models to account for the presence or absence of these species adds credibility to the idea that their distribution is somewhat more stochastic than that of other species. In contrast, species such as Lophozia guttulata, Rhytidiadelphus squarrosus, and  Barbilophozia floerkei were associated with habitats that support permanent, "stable" patches of bryophytes. The logistic regression models for these species tend to have higher Rhosquared values, and higher success gained over random models. Species with specific substratum affinities, such as Pseudoleskea baileyi, which occurs predominantly on creeping stems, are also associated with relatively higher Rho-squared values and predictive success. Thus, the increased success of the statistical models gives credibility to the idea that the corresponding species distributions are less random than for other species with lower Rhosquared and success-gained values. It should be noted that there are many combinations of variables that may be used to produce a logistic regression model that does not differ significantly from a model containing all variables (i.e. a "full model") in terms of predictive success. For example, a model containing only subsets of variables (substratum, vegetation, and landscape groups), or combinations of variables among these groups may be able to predict the presence/absence of particular bryophyte species with success equal to that of a full model. One useful technique for achieving such a "reduced" model involves using LR in a stepwise fashion, in which the analysis will add or remove variables in order of their significance/importance to the model. However, Tabachnick & Fidell (1996) caution that it is very easy to misinterpret the exclusion of a predictor variable in a stepwise LR; a predictor may be highly correlated with the outcome, but not included in the equation because it was usurped by another predictor, or combination of predictors. It should be noted that as variables are added or removedfromLR models, the significance in effects of those predictors will change, i.e. the significance of predictors will depend on which predictors are already included in the model.  78 In this thesis, only the full model form of logistic regression was used, as my goal was to compare the predictability of bryophyte presence or absence from the same set of predictor variables. To design a model for practical use (e.g. use in a forest management strategy), it may be beneficial to select a smaller number of predictors, for example those indicated by a stepwise LR  79 CHAPTER IV Summary of Subalpine Terrestrial Bryophyte Ecology The geography, climate, and vegetation of the North Shore mountains have been investigated in an attempt to enhance the understanding of terrestrial bryophytes in these subalpine ecosystems.  Using principal components analyses, it was demonstrated that  biogeographical plant associations correspond only very generally to the patterns of distribution and abundance of terrestrial bryophytes.  The large amount of variation in  frequency and abundance of bryophytes that exists within sites and association groups seems to indicate that smaller scale variations in habitat (e.g. plot or microplot level) may be of importance. Patterns of bryophyte diversity and abundance were examined at the plot level using canonical correlation analysis and Pearson correlation coefficients.  This analysis  demonstrated that coarse scale patterns of diversity and abundance of bryophytes seem to be related to different sets of variables. Diversity and coverage were found to be positively related to northern aspect, and LFH depth and organic matter content. Bryophyte diversity and coverage were noted to differ in their response to some landscape and soil features (TKN and slope). Although steeper inclines may not promote the existence of permanent patches of terrestrial bryophytes, they may facilitate movement of propagules among plots. Bryophyte diversity and coverage were found to be significantly related to the vegetation characteristics of plots. Both diversity and coverage were negatively correlated with canopy coverage, and positively correlated with vascular plant abundance.  These  results suggest that light availability and, possibly, nutrient throughfall from overlying vegetation may be important to bryophyte distributions. Substratum variables were also found to have predictive significance for plot-level bryophyte coverage and diversity. Both bryophyte characteristics were negatively related to the abundance of woody debris, and positively correlated with the availability of creeping stems. It seems that these substratum characteristics are indirectly related to bryophyte diversity and coverage, and are more directly related to stand density characteristics that affect light availability.  The high  correlation that exists between individual factors obscures the interpretation of their relative importance. The large amount of variation in physiological tolerances among bryophyte  80 species also contributes to the difficulty in accounting for overall patterns of distribution at the plot scale. The environmental preferences of individual bryophyte species were examined using Pearson correlation coefficients, and by examining the range of microhabitat conditions in which species were noted to occur. Species prominent in open habitats exhibited much narrower ranges of LFH characteristics (pH, C:N ratios) than those of closed habitats. This observation may provide some insight regarding community structure: assemblages in closed habitats (with perhaps less surface stability) may not show particularity in terms of LFH characteristics because their position may be only temporary. To determine the substratum affinities of separate bryophyte species, their microscale frequencies on available substratum types were examined. To investigate the combined influence of substratum and microhabitat, the species associations on different substratum types were investigated using principal components analysis. It was found that microscale features (environment and/or substratum) are closely related to the distributions of some bryophyte species. Assemblages on fine litter seemed to be related to differences in light availability and nutrient dynamics. Bryophyte assemblages on woody debris, on the other hand, seemed to be related to stand characteristics and numbers of newly germinated seedlings {Chamaecyparis nootkatensis in particular). Bryophyte assemblages on fine litter and woody debris in open, "stable" habitats with abundant bryophytes and vascular plants differed from those in closed-canopy, steeper, "unstable" habitats that had lower bryophyte and vegetation cover. Assemblages on exposed humus and creeping stems were related to the availability of the substratum types. The relationship of bryophyte species to stable  and dynamic types of  microenvironments was investigated using logistic regression. As expected, bryophytes associated with unstable environments (i.e. dynamic surface layer) had a more random distribution than species associated with open, stable habitats, or which displayed strong substratum affinities.  The physiological tolerance of species for microhabitat features  (environment and/or substratum), and the degree of microhabitat stability seem to play important roles in determining the structure and dynamics of vegetation.  terrestrial bryophyte  81 Okland (1994) suggested the existence of a micro-scale coenocline running from the (more stable) "normal forest floor", dominated by mosses and liverworts, to (less stable) patches dominated by small liverworts. This micro-scale coenocline may be associated with recoIonization after fme-scale disturbance, and/or physiological tolerance of the species to a more extreme micro-site environment. During & Van Tooren (1990) noted that rapid dynamics of the bottom layer, as has been documented for other ecosystems, would support the former explanation. These findings are supported by the results of the present study. Species most frequent in closed habitats include a large proportion of small liverworts that show a broad range of tolerance for substratum conditions. The information gained from the analyses in this study can be used to summarize the environmental, substratum, and species relationships of the terrestrial mosses and liverworts encountered (Tables 4.1 and 4.2, respectively). The ecological information about bryophytes obtained by this study may be compared with that from existing literature (Lawton 1971; Schofield 1976, 1992; Vitt et al. 1988; Godfrey 1977; Schuster 1966, 1969, 1974, 1980, 1992a, 1992b; Conard 1979; Davison 1993). The total habitat information for mosses and liverworts encountered in this study and available from these sources, is summarized in Appendix 3. In comparing the information on habitat relationships documented in currently available literature with results from this study, a few discrepancies are evident. Before discussing particular discrepancies, it is important to point out that this study examined only terrestrial habitats, not including logs or tree bases. The information contained in Tables 4.1 and 4.2 is, therefore, only a subset of the total habitat information. In addition, this study examined only a narrow range of subalpine environments; a broader range of habitats would have provided information more comparable with that summarizedfromthe literature. This study is useful in its narrow range, however, in that a more detailed examination of terrestrial habitats is possible. Such information allows for the elucidation of species with narrower, or broader, substratum and habitat preferences than is documented in literature. For example, Plagiothecium undulatum, documented to be a lowland species, was found to occur (although on only one occasion) in the M H zone, demonstrating a potentially greater range of tolerance than literature sources suggest. Similarly, the habitat of Rhytidiopsis robusta as recorded in the literature - "not usually found in open sites" - contrasts with the  CN OO  c/3  <u C/J C/3  3" < ^y^  O £  0  13  £ Si i  t/1  I ' 3  cu  c  <u  h  o a,  uf H  <3 £  cd O  ~ ul  o  aS  o  Id 6 °  3 .  » a,  3  5 a' . 1  w  11  o  •s 5 CO  u. «  O  5  Z  £  CQ* eT  X  ^3  SS S 3  552  SI  a<  3  e  I2 l  of  cd cn cu  i  'o oj  ml  f  ft ,3 g• * < | -5 S * 8  PH  CO  • r i l l  - ^3  **£1 J .3 H3  CD O cd  3  3 s ft  Q.  S  fe  M  If  43  •2 §  II 5 B-  C4-H  cd CJ ed  8-S!  i-H  H—' C/3  X> 3 c/j  <4-H  if 5/3  —  3  S3  H  8.  1 -8  O  S  3 s  f J if ails 1g|  2 =5 3  oo  i  M  a  g- 5  £ .a  11  8.8  git.  £ 11 *l s | B-rl  H Q" -J  K  IPH3  c o  8  tail  Q II 3  ft.  T3  5*  ag&  C/3  g JS  ul O  ia  sal 3 d.  a  1 §• 8 S3  p,  tr  i  <u  asp •* Ri £  o  cn oo  cd  •c  -J" E—*  3 » #  to CD  lie  £  CD  3 8" < oc  D  sW oPL, I I 3  H—>  , —  CD  s  JS  •a £  ii  ,° S h — a. D x; T) 0 5  cd O  s a  l-H  CJ  o a q z JS £ d £. •s a H >z § o o. PC  •S u  -rat  E  w o 2 o H CL.I 5S3  a.  e  x y 3  £3  A  Ss Si  a "  cd  <  £ •sag SJ  D O X  I . CU.  i - >• a,  a. o  i  a  B-  ids  §53 a  _ <;  s § s! _  a  1^3  PQ ! ?  <§ § p fe gQct.g  S3 S 3 & J p 8 Q g a, a-  ill  D  ft, a] H  1'S < ! f S3 I « Bs i  5 < >S 8  S3 a a p X  Q."  * -l 33 a I  •a J3 ft  j  si aa a£ f& ^: a.  D  1 1 1  :  J  ll  S3 is  1*3  CQ  O  s  1  S  !3  <s x  fe" *  a. « so 3 s? fc  Is S &  CO  -5  ? a H B. o 3  <  (3 <£ a  IS  i  3  ~< ° 2 o  to « U f f l o J J B ! CO Z H z #• < 0I 5 P a •S o I i= fe S PL. £ O •S BQ n. X  3  8-  3; Bo  .£ §3gS ^•S£x x *2 a IIS. —r CQ ?? O  p 3 ? «• 3w£a >  _o "E. CO  o o  o </> cd  'o CD  & co T3  C cd </f cj  •c  SI  CJ  o  cd l-H  cd XI  ts  o  8 •"  5§  a S  •8 JS  2 " •8  ;  o £  cu  II  II 8 8  *  8  84 findings of the present study. Perhaps this discrepancy lies in the fact that only Mountain Hemlock Zone sites were studied here. In contrast to other zones, canopy cover in the M H zone is, generally, dense. The two sources of information can be reconciled by the statement that R. robusta has the highest abundance in open patches within closed-canopy sites. Some bryophytes were noted to occur on different substrata than the literature suggests. For example, Antitrichia curtipendula and Isothecium stoloniferum are sometimes  considered as primarily epiphytic or rock-affiliated species. However, these mosses were observed growing firmly attached to fine litter during the course of this study (four times for Antitrichia curtipendula, and once for Isothecium stoloniferum). Peck et al. (1995) similarly noted the occurrence of these two species, well established on the forest floor, in coniferous forests in the central western Cascades of Oregon. It is possible that, in forests, these species may persist in limited abundance as terrestrial species after falling from their epiphytic habitat.  Pterigynandrum filiforme and Porella cordaeana are also documented to be  primarily epiphytic or rock-affiliated species. Since only a single, impermanently attached strand of each of these bryophytes was found on the forest floor in this study, their description as epiphytes cannot be disputed. Pseudoleskea baileyi and P. stenophylla are described in the literature as occurring primarily on the living twigs and branches of trees or shrubs, and occasionally on rock. In accordance with this, these mosses were found to occur primarily on creeping stems and ' branches of shrubs in this study. However, they were also quite frequent on fine litter and wood substrata, thus demonstrating a broader range of tolerance than is currently indicated in literature. Brachythecium leibergii, Rhytidiadelphus squarrosus, and Ptilidium californicum  are other bryophytes that were noted to occur on more types of substrata than literature sources describe. Although rock outcrops were common in both study areas, bare rock was not a frequent terrestrial substratum in any of the plots sampled; almost all rock surfaces were covered by a layer of fine litter. Although fine litter depth over other substrata was not incorporated into this study, it may relate to the occurrence of bryophytes that are primarily lithophytic.  For example, the mosses Dryptodon patens, Heterocladium procurrens, and  Racomitrium ericoides were all noted to occur (although very infrequently) on fine litter substrata.  85 Variations in fine litter depth also may be related to the presence of other species, such as those described as mainly occurring on bare soil (a substratum that did not occur in any of the microplots). For example, the mosses Dicranella heteromalla, Oligotrichum parallelum, and Rhizomnium nudum, which are documented to occur primarily on soil, were observed growing on fine litter and/or exposed humus.  Similarly, the liverworts  BarbUophozia floerkei, Diplophyllum taxifolium, Scapania diplophylloides, and Scapania  scandica are thought to have rock or soil as their primary substratum, but in this study were noted to occur on exposed humus and/or fine litter. Bryophyte assemblages on logs were not examined in this study. As a result, the habitats available for primarily "woody" species were limited to woody debris on the forest floor. Exposed humus seemed to be very important to such species, particularly liverworts. For example, the liverworts Cephalozia bicuspidata and Cephalozia lunulifolia  are  documented to occur primarily on decaying logs and stumps; however, in this study, they were noted to occur most often on exposed humus. Scapania bolanderi, also a "woody" species, was observed to have a high affinity for this substratum as well. In conclusion, both of the major goals for this thesis have been met. The species and substratum affinities of terrestrial bryophytes in coastal subalpine forests have been described, and their patterns of distribution at different scales and on different substrata was accomplished using multivariate analyses. Such basic ecological information may serve as a reference for future descriptive and/or experimental ecological studies of bryophytes in subalpine forests.  The recognition of different patterns of distribution has led to the  formation of hypotheses about the causes of these patterns. For example, it has been suggested that in closed environments with a dynamic surface layer, the bryophytic input from litterfall and decayed logs may influence distribution patterns. In contrast, it was suggested that in open habitats with permanent patches of bryophytes, the input of species from litterfall and logs may not be as important as competitive and dispersal differences between species. Before experimentally testing any hypotheses about terrestrial bryophyte community structure, it would be desirable to integrate results from bryophyte studies on other substrates (e.g. epiphytes and epixylics) in a similar subalpine environment. Only by assimilating these different pieces of information may we achieve a cohesive understanding of bryophyte  86 patterns of distribution, and consequently hypothesize about their importance to the subalpine ecosystem.  Because the scope of this thesis included only one aspect of bryophyte  communities in subalpine forests (i.e. terrestrial species), inferring how these bryophytes may be incorporated into forest management strategies is premature. Adding to the difficulty in deriving strategies for bryophyte conservation in these forests is that it is unclear what the objective in conservation should be; the preservation of bryophyte species richness and diversity, or the preservation of bryophyte "function" in subalpine habitats.  If the goal of conservation is to preserve biodiversity, then the  preservation of a diverse number of microhabitats (i.e. increased heterogeneity at the microscale) may preserve the largest variety of terrestrial bryophyte assemblages. However, if the goal of conservation is to preserve the "function" of terrestrial bryophytes (i.e. nutrient cycling, water dynamics), then the preservation of large, stable patches of bryophytes, which are more likely to impact these processes, may be appropriate. As such, if the goal of conservation is to preserve bryophyte function, future studies must examine experimentally how bryophytes contribute to water and nutrient cycling, and how these bryophytes are influenced by environmental disturbance.  87 Literature Cited Andersson, L.I., & H. Hytteborn (1991). Bryophytes and decaying wood - a comparison between managed and natural forest. Holarctic Ecology, 14: 121-130. B.C. Ministry of Crown Lands: Surveys and Resource Mapping Branch (1992). 1:20 000 TRIM topographical maps. Produced by Digital Mapping Group, Ltd. B.C. Ministry of Environment (1999). History snow survey data for British Columbia: http://vvww.elp.gov.bc.ca/wat/snow bulletin/archive/historic.html  Brooke, R.C., E.B. Peterson, & V J . Krajina (1970). The subalpine mountain hemlock zone. Ecology of Western North America, 2(2): 153-349.  Brown, D.H. & J.W. Bates (1990). Bryophytes and nutrient cycling. Botanical Journal of the Linnean Society, 104: 129 l47'. L  Carleton, T.J. (1990). Variation in terricolous bryophyte and macrolichen vegitation along primary gradients in Canadian boreal forests. Journal of Vegetation Science, 1: 585-594. Carter, M.R. (1993). Soil Sampling and Methods of Analysis. Canadian Society of Soil Science, Lewis Publishers. Conard, H.S. (1979). How to Know the Mosses and Liverworts. W.C. Brown Co., Dubuque Iowa. Culbert, R.R. (1971). Study of tectonic processes and certain geochemical abnormalities in the Coast Mountains of British Columbia. Ph.D. Thesis, University of British Columbia. Daubenmire, R. (1968). Plant Communities. A Textbook of Plant Synecology. Harper Publishing, New York. Davidson, A.J., J.B. Harbome, & R.E. Longton (1990). The acceptibility of mosses as food for generalist herbivores, slugs in the Arionidae. Botanical Journal of the Linnean Society, 104: 99-113. Davison, P.G. (1993). Floristic and phytogeographic studies of the hepatic flora of the Aleutian Islands, Alaska. Ph.D. Thesis, University of Tennessee. During, H J . (1979). Life strategies of bryophytes: a preliminary review. Lindbergia, 5: 218. During, H.J. & B.F. Van Tooren (1990). Bryophyte interactions with other plants. Botanical Journal of the Linnean Society, 104: 79-98.  88 Eekman, G.C. (1976). Plant associations within the Subalpine Mountain Hemlock Zone as indicators of recreational land use capability. Ph.D. Thesis, University of British Columbia. Eldridge D J. & M.E. Tozer (1997). Environmental factors relating to the distribution of terricolous bryophytes and lichens in semi-arid eastern Australia. The Bryologist 100(1): 2839. Franklin, J., K. Cromack, W. Denison, A. McKee, C. Maser, J. Sedell, F. Swanson & G. Juday (1981). Ecological characteristics of old-growth Douglas-fir Forests. USDA Forest Service General Technical Report PNW-GTR-118. Frego, K.A. & Carleton, T.J. (1995a). Microsite conditions and spatial pattern in a boreal bryophyte community. Canadian Journal of Botany, 73: 544-551. Frego, K.A. & Carleton, T.J. (1995b). Microsite tolerance of four bryophytes in a mature black spruce stand: reciprocal transplants. The Bryologist, 98(4): 452-458. Furness, S.B. & J.P. Grime (1982). Growth rate and temperature responses in bryophytes. II. A comparative study of species of contrasted ecology. Journal of Ecology, 70: 525-536. Gauch, H.G. (1982). Multivariate Analysis in Community Ecology. Cambridge University Press, U.S.A. Godfrey, J.D. (1977). The hepaticae and anthocerotae of southwest British Columbia. Ph.D. Thesis, University of British Columbia. Grime, J.P., E.R. Rincon, & B.E. Wickerson (1990). Bryophytes and plant strategy theory. Botanical Journal or the Linnean Society, 104: 175-186.  Herben, T. & L. Soderstrom (1992). Which habitat parameters are most important for the persistance of a bryophyte species on patchy, temporary substrate? Biological Conservation, 59: 121-126. Ingerpuu, N., K. Kull & K. Vellak (1998). Bryophyte vegetation in a wooded meadow: relationship with phanerogram diversity and responses to fertilization. Plant Ecology, 134(2): 163-171. Jonsson, B.G. & P.A. Esseen (1990). Treefall disturbance maintains high bryophyte diversity in a boreal spruce forest. Journal of Ecology, 78: 94-936. Jonsson, B.G. (1993). The bryophyte diaspore bank and its role after small-scale disturbance in a boreal forest. Journal of Vegetation Science, 4: 819-826. Jonsson, B.G. (1997). Riparian bryophyte vegetation in the Cascade mountain range, Northwest USA: patterns at different spatial scales. Canadian Journal of Botany, 75: 744761.  89  Kenkel, N.C. & G.E. Bradfield (1986). Epiphytic vegetation on Acer macrophyllum: a multivariate study of species-habitat relationships. Vegetatio, 68: 43-53. Kimmerer, R.W. & T.F.H. Allen (1982). The role of disturbance in the pattern of a riparian bryophyte community. American Midland Naturalist, 107(2): 370-383. Kimmerer, R.W. (1993). Disturbance and dominance in Tetraphis pellucida: a model of disturbance frequency and reproductive mode. The Bryologist, 96: 73-79. Kimmerer, R.W. & C C . Young (1996). Effect of gap size and regeneration niche on species coexistence in bryophyte communities. Bulletin of the the Torrey Botanical Club, 123(1): 16-24. Krajina, V.J. (1965). Ecology of Western North America. University of British Columbia, Department of Botany. Lavkulich, L.M. Personal communication, June 1999. University of British Columbia, Vancouver, B.C. Lawton, E. (1971). Moss Flora of the Pacific Northwest. Hattori Botanical Laboratory, Nidiman, Japan. Lee, T.D. & G.H. LaRoi (1979). Bryophyte and understory vascular plant beta diversity in relation to moisture and elevation gradients. Vegetatio 40(1): 29-38. Lee, T.D. & G.H. LaRoi (1979). Gradient analysis of bryophytes in Jasper Natinal Park, Alberta. Canadian Journal of Botany, 57: 914-925.  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. Longton, R.E. & C.J. Miles (1982). Studies on the reproductive biology of mosses. Journal of the Hattori Botanical Laboratory, 52: 219-240. Longton, R.E. (1984). The role of bryophytes in terrestrial ecosystems. Jounal of the Hattori Botanical Laboratory, 55: 147-163.  Marino, P.C. (1991). Dispersal and coexistence of mosses (Splanchaceae) in patchy habitats. Journal of Ecology, 79: 1047-1060.  Matlack, G.R. (1993). Microenvironment variation within and among forest edge sites in the eastern United States. Biological Conservation, 65: 185-194.  90 McCune, B. & P. Lesica (1992). The trade-off between species capture and quantitative accuracy in ecological inventory of lichens and bryophytes in forests in Montana. The Bryologist, 95: 296-304. Muhle, H. & F. LeBlanc (1975). Bryophyte and lichen succession on decaying logs. I. Analysis along an evaporational gradient in eastern Canada. Jounal of the Hattori Botanical Laboratory, 39: 1-33.  Muller-Stoll, W.R. (1965). Regeneration bei niederen Pflanzen (in physiologischer Betrachtung). - In: Ruhland, W. (ed), Handbuch der Pflanzenphysiologie XV/2: 92-155. 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): 4354. Okland, R.H. (1994). Patterns of bryophyte associations at different scales in a Norwegian boreal spruce forest. Jounal of Vegetation Science, 5: 127-138. Okland, R.H. (1986). Rescaling of ecological gradients. I. Calculation of ecological distance between vegetation stands by means of their floristic composition. Nordic Journal of Botany, 6: 651-660. Parrish, R.R. (1982). Cenozoic thermal and tectonic history of the Coast Mountains of British Columbia : as revealed by fission track and geological data and quantitative thermal models. Ph.D. thesis, University of British Columbia. Peck, J.E., S.A. Acker & W.A. McKee (1995). Autecology of mosses in coniferous forests in the central western Cascades of Oregon. Northwest Science, 69: 184-190. Pharo, E.J. & A.J. Beattie (1997). Bryophyte and lichen diversity: a comparative study. Australian Journal of Ecology, 22: 151-162. '  Pojar, J., K. Klinka, & D.A. Demarchi (1991). Chapter 7: Mountain Hemlock Zone in Ecosystems of B.C., D. Meidinger & J. Pojar (Eds). B.C. Ministry of Forests, Victoria. Prins, H.H. (1981). Why are mosses eaten in cold environments only? Oikos 38: 374-380. Proctor, M.C.F. (1990). The physiological basis of bryophyte production. Botanical Jounal of the Linnean Society, 104: 61-77.  Rambo, T.R. & P.S. Muir (1998a). Forest floor bryophytes of Pseudotsuga menziesii Tsuga heterophylla stands in Oregon: influences of substrate and overstory. The Bryologist, 101: 116-130. Rambo, T. & P.S. Muir (1998b). Bryophyte species associations with coarse woody debris and stand ages in Oregon. The Bryologist, 101(3): 366-376.  91  Rincon, E. (1988). The effect of herbaceous litter on bryophyte growth. Journal of Bryology, 15: 209-217. Rougharden, J.D. (1977). Patchiness in the spatial distribution of a population caused by stochastic fluctuations in resources. Oikos, 29:52-59. Schofield, W.B. (1976). Bryophytes of British Columbia III: habitat and distributional information for selected mosses. Syesis, 9: 317-354. Schofield, W.B. (1984). Bryogeography of the Pacific coast of North America. Journal of the Hattori Botanical Laboratory, 55: 35-43.  Schofield, W.B. (1985). Introduction to Bryology. MacMillan Publishing Company, New York. Schofield, W.B. (1992).  Some Common Mosses of British Columbia.  Royal British  Columbia Museum, Canada. Schuster, R.M. (1966).  The Hepaticae and Anthocerotae of North America: Volume I.  Columbia University Press, New York. Schuster, R.M. (1969).  The Hepaticae and Anthocerotae of North America: Volume II.  Columbia University Press, New York. Schuster, R.M. (1974). The Hepaticae and Anthocerotae of North America: Volume III.  Columbia University Press, New York. Schuster, R.M. (1980). The Hepaticae and Anthocerotae of North America: Volume IV.  Field Museum of Natural History, Chicago. Schuster, R.M. (1992a). The Hepaticae and Anthocerotae of North America: Volume V.  Field Museum of Natural History, Chicago. Schuster, R.M.(1992b).  The Hepaticae and Anthocerotae of North America: Volume VI.  Field Museum of Natural History, Chicago. Skre, O. & W.C. Oechel (1979). Moss production in a black spruce Picea mariana forest with permafrost near Fairbanks, Alaska, as compared with two permafrost-free stands. Holarctic Ecology, 2: 249-254.  Slack, N.G. (1977). Species diversity and community structure in bryophytes: New York State studies. New York State Museum Bulletin 428. Slack, N.G. (1982). Bryophytes in relation to ecological niche theory. Journal of the Hattori Botanical Laboratory, No. 52: 199-217.  92  Slack, N.G. (1984). A new look at bryophyte community analysis: field and statistical methods. Journal of the Hattori Botanical Laboratory, 55: 113-132.  Slack, N.G. (1990). Bryophytes and ecological niche theory. Botanical Journal of the Linnean Society, 104: 187-213.  Soderstrom, L. (1988a). Sequence of bryophytes and lichens in relation to substrate variables of decaying coniferous wood in Northern Sweden. Nordic Jounal of Botany, 8(1): 89-97. Soderstrom, L. (1988b). The occurrence of epixylic bryophyte and lichen species in an old natural and a managed forest stand in northeast Sweden. Biological Conservation, 45: 169178. Soderstrom, L. (1993). Substrate preference in some forest bryophytes: a quantitative study. Lindbergia 18: 98-103. Soderstrom, L. (1994). Reproductive biology of bryophytes. Scope and significance of studies on reproductive biology of bryophytes. Journal of the Hattori Botanical Laboratory, 76: 97-103. Tabachnick, B.G. & L.S. Fidell (1996).  Using Multivariate Statistics - 3rd Edition.  California State University, Northridge. HarperCollins College Publishers, U.S.A. Tilman, D. (1982). Resource Competition and Community Structure. Princeton: Princeton University Press. Verkaar, H.J., A.J. Schenkeveld & C.L. Huurnink (1986). The fate of Scabiosa columbiana (Dipsacaceae) seeds in a chalk grassland. Oikos, 46: 159:162. Vitt, D.H., J.E. Marsh & R.B. Bovey (1988). Mosses and Lichens of Northwest North America. Lone Pine Publishing, Edmonton, Canada. Vitt, D.H. (1990). Growth and production dynamics of boreal mosses over climatic, chemical, and topographic gradients. Botanical Journal of the Linnean Society, 104: 35-59. Watson, M.A. (1980). Patterns of habitat occupation in mosses - relevance to considerations of the niche. Bulletin of the Torrey Botanical Club, 107(3): 346-372.  Wyatt, R. (1982). Population ecology of bryophytes. Journal of the Hattori Botanical Laboratory, 52: 179-198. Zackrisson, O., M.C. Nilsson, A. Dahlberg & A. Jaderlund (1997). Interference mechanisms in conifer-Ericaceae-feathermoss communities. Oikos, 78(2): 209-220.  93 Appendix 1. Summary of Species and Variable Abbreviations. V a s c u l a r Plants  Bryophytes Anticur Barbflo Bleptri Braclei Calymue Cephbic Cephlun Dicrhet Dicrfus Dicrpal Diplpli Dipltax Dryppat Geocgrav Hetepro Hypncir Isotsto Lepirep Lophgut Lophhet Mniuspi Oligpar Plaglae Plagund Pleusch Pohlnut Polyalp Porecor Pseubai Pseuste Pseuele Pterfil Ptilcal Racoeri Racosud Rhiznud Rhytlor Rhytsqu Rhytrob Scapbol Scapdip Scapsca  Antitrichia curtipendula Barbilophozia floerkei Blepharostoma trichophyllum Brachythecium leibergii Calypogiea muelleriana Cephalozia bicuspidata Cephalozia lunulifolia Dicranella heteromalla Dicranum fuscescens Dicranum pallidisetum Diplophyllum plicatum Diplophyllum taxifolium Dryptodon patens Geocalyx graveolens Heterocladium procurrens Hypnum circinale Isothecium stoloniferum Lepidozia reptans Lophozia guttulata Lophocolea heterophylla Mnium spinulosum Oligotrichum parallelum Plagiothecium laetum Plagiothecium undulatum Pleurozium schreberi Pohlia nutans Polytrichastrum alpinum Porella cordaeana Pseudoleskea baileyi Pseudoleskea stenophylla Pseudotaxiphyllum elegans Plerygynandrum jiliforme Ptilidium californicum Racomitrium ericoides Racomitrium sudeticum Rhizomnium nudum Rhytidiadelphus loreus Rhytidiadelphus squarrosus Rhytidiopsis robusta Scapania bolanderi Scapania diplophylloides Scapania scandica  % (prefix)  average percent cover (microplots)  f (prefix)  plot frequency  Arnilat  Arnica latifolia Athyrium filix-femina Blechnum spicant Carex sp. Cassiope mertensiana Cladothamnus pyrolaeflorus Clintonia borealis Fauria crista-galli Goodyera oblongifolia  Athyfel Blecspi Caresp. Cassmer Cladpyr Clinbor Faurcri Goodobl Grass  (Poaceae)  Leutpec  Valesit  Luetkea pectinata Listera cordata Menziesia ferruginea Mitella sp. Monotropa sp. Phyllodoce empetriformis Polystichum munitum Pyrola sp. Rhododendron albiflorum Rubus pedatus Rubus spectabilis Sambucus sp. Smilacina stellata Sorbus sitchensis Streptopus sp. Tiarella trifoliata Tiarella unifoliata Vaccinum membranaceum Vaccinium alaskense/V. ovalifolium Valeriana sitchensis  Veravir  Vpralmim viridp  Listcor Menzfer Mitesp. Monosp. Phylemp Polymun Pyrosp. Rhodalb Rubuped Rubuspe Sambsp. Smilste Sorbsp. Stresp. Tiartri Tiaruni Vaccmem Vaccala/ova  Violsp.  Viola sp.  numvasc  number o f vascular plant species  avgvas  average vascular plant coverage  htvasc  m a x i m u m vascular plant height  Substrate FLI  F i n e Litter (needles, w o o d < l c m )  avgbryo  average bryophyte coverage  WOOD  W o o d y debris (1-1 O c m in diameter)  numbryo  number o f bryophyte species  EXH  D e c a y e d , exposed humus  avgmoss  average moss cover  CRS  C r e e p i n g stems  nummoss  number o f moss species  EXR  E x p o s e d roots  avglvrt  average liverwort coverage  ROC  Rock  numlvrt  number o f liverwort species  numsubs  number o f substrate types  94 Stand Structure  Landscape characteristics  Avgcan  A v e r a g e canopy cover  Tasp  A A , Abam  Slope  S l o p e o f plot (degrees)  %outcr  % rock outcrop in plot  %minsoi  % mineral soil in plot  T M , Tsme  Abies amabilis Chamaecyparis nootkatensis Tsuga heterophylla Tsuga mertensiana  %water  % water in plot  SE  trees <5 c m in diameter  %logs  % logs in plot  SA  trees 5-10 c m in diameter  %litter  % litter ( w o o d y debris and needles)  PT  trees 10-20 c m in diameter  MT  trees 20-40 c m in diameter  Soil Features  LT  trees 4 0 - 6 0 c m in diameter  L F H (cm)  O r g a n i c layer (over soil)  VL  trees >60 c m in diameter  DEFHOR  A h o r i z o n definition (present/absent)  NUMGERM  total # o f germinants (microplot)  H 2 0 w t (g)  water content o f L F H layer  NAAGERM  # A. amabilis germinants # C. nootkatensis germinants # T. mertensiana germinants  %CLOI  C a r b o n lost o n ignition (organic matter)  TKN  total K j e l d a h l nitrogen in L F H layer  C:N RATIO  C a r b o n . N i t r o g e n ratio in L F H layer  PHLFH  p H o f the L F H layer  AVGPH  average p H ( f r o m L F H , A , B horizon)  C N , Chno T H , Tshe  NCNGERM NTMGERM  A s p e c t (degrees f r o m N ) .  UO  ON  GO vo O UO — • CO CN CN CN , m d CN © d  f-^  5.1-  oo m  ON NO ,  NO CO  ON NO UO  r—  m ON  E= Si  . ON o ri 6  O  « » ON . NO _ ;  ro  NO q ON d d  ro  oq r~NO  r- ro  NO r - uo uo ,  . CN T f q uo d O  ^  ON CN  X P £  uo ON NO |  TT  o  r—  r-  ,  0  NO ON r- T f ON T f . °. Tf' ^ O CM O — O O  . UO r o NO  M  ^  oo  00  5'  r-  o  o  ro  r-  .3  uo © O -  BJ  m «  ~". ^  ^ £  o d  ^  -  ™ °  (N ^ — UO NO CO —' - J — © T f ON — <-> — WO — CN  cc ~  Tf  oo  CN  CO  ON  ro  O  CO OO T f Tf (N  ro  O  NO O  S  ™ °°. °°. ON U0 U0  E > 3 w S Q •a H •a w  P-; CN O CN  ^  2  _ o oo Tf Q — ~ UO X  O ON  Tf' ^  «> . O  T f O N O T f C N O O O O r o ' U0 r o O — —  H  NO NO  o  S  ^  p - oo p - oo — ON m ' d i d CN ;  r-*  —  00  UO CN — T f r o  ON  fl  5 > =  r-  •S- >  — 00  C  CN T f r~~ r-~ d O ro Tf  r ^ r ^ a N T f o r ^ N o q u o c N d — — NOTfvo — O O C N !  ^ o  1  rn  q Tf  uo o  r-; T f r o uo  vo O U0  O  CN 00 T f ro • CN - r- oo , -H oo ro ro  00 (N  uo uo  1  . co  2  . °° " -  oo r » oo -  1  O  o  O CN  o b I3 00  uo uo  CO CO CN  O OO UO ro r o  r-» m  Q  X  2?  h- O  o. <  - 1 -O  H  <  5  X  i/l  tt!  > g3 j .gO X  *  sP X o-  «  o  „  . £ 3  fcfcLL.  o-  ^  : 3  C  U J C J LU  ro o  Tf  co uo — r*^ od d d r o  ^  \° 3t o"*  X  X  J  o  : X > X ^ r - O n . < _ J - a c / D r - < ^  > .fi  fc 3 S  c 3  C  UH ? u O ui s p s P sP sP sP oo"1  q  B E •> s  3 S=  c  ON  3>  5* U J -X Q  s oo oo m v©  CN  9  =5 >  "5  (N  SO  Q  OO  ^  oo  O  - o ^ t  vo  ON  \Q t>  r - — fN  >o n d  O W i O ON O r i CN 6 r i  it  in  ©  o o  :  o  oo  q  .  m co  so — © —  , -  cj,  o  od  ON  tv  "  n o\  ^ - \ 0 ^ (N vi « r i d  co n  n o n o o  ri t IT-,  ri  r>  w «  O  O  g  O  <7\  CM o r i ^ V>  >c -  - 2 <=>  -Si Q  >  oo ~ ; " 1 n ri ^ ^ d  >o — CN d —^ r i  CN  r- o\ i n C\  o \ — r - r -  oo  f i f i  r  ri O  cnhTr--ifloO'3;r] d d - ' t t r i d d d r i  rj  — oo m  1I pE5  o  O  ©  (N  TT  oi V r i d <*i od  O  oCT,a  o — o o  as  TT  f i  ro  O n  I vi t cd oi rn  -  2  od  _  Ch 0\ « rn ( N O O ri  K ri  m  r; ^ \ q 0\ 0 \ r ; wi X *o ^ — ° — f l — (N  oo  in  © Tt  _  o r<i  O O  O ri © © ri  5, m  a aI •3 w  ;  t  t  O  ~  ^  • TJ- o  M \ o >o  O  r - f i — —  t  n  O O Cf; O O ri  O  r - *o ^ —  m  f > m ^  °°. °° od S  o - cn CSo °-  n  ri- r-  ovo —« o *o  -  « h  s o - t f r - H ^ ^ I c M f i o o o d — ^ d ^ S ^ T r ' d d r J  TI t t t< ™ ON' — © o —  <*i  w rf  a  U J  O  is ° 5 H  n  r - vo  W *0  s>  ON  3  O —  M t  CM O O ^  r- m  VJD  o  K ^ ^ rf  —  ON  m n  f i ON O v cK ^ - 1 CN  n ^ oo  Tt  CN  —  g  — o o — °  = L—  fl  : V t <N d  9  Q  « C\ VI ^6 rS r i > o - t c «  ^  ON f l © O  6C  -  •= X X z  7  ^ S a  00 d  0\ ( N r n M K *o  M  LO  oo h  ^ qrn ri « 6 r n  Q  u  „ C J  K 5! >  I I f f 51 1 w  H <  LO  vi  J= C  2 >3 ?2  E w fli X X W U W  11  O  00 OO ^  -J  JIJ —  o ri x  © r t  ON  OO f i f i  O  in  f l  o - "S 2  x x LL,  CJ,  C J w i 55 or, «  I-  5  H  <  .  "  2 X C/l OH a> 3 S x PS x a BH ? w o w o 5? 5? S? OJ  \0 s o fi Tt ^ I  .a &  B CD  O  CO  %  tt  u  CQ  2  £  <J  tt  K  2  3  u  1  TI  a  tv  i  14 i 2  K  A, 1 K TJ O TJ  o  3 £ tv •3 a -S  a  8  u  W  R  I  K  1 1i  K  4  tt  3  K  i  i  8  1  &£ 1  i  H  11  u tt *£ o  1  1  TJ  a.  1  U tt  ©  "C  •a  % Z  1  1  tt  .a  i  K  &  n  &  it  •« a  I 1 •a- 1 tt S | | | i • a i •s 3 #  5  3  1§•  ts  ©  1  sea level to subalpine, rarely at alpine elevations  from sealevelto 1800m  from 1000 to 1500m or higher; lower elevations in the north  1  *  1o  g o  3  TJ 18 a ts  0  g sc i  -B tt  mineral soil banks, humus soil, decaying wood  rock o  e  tt  u  1  1  1  11  ts  E E 0  R  a  a  £  •a*  TB  •2  TJ  i  rocks, soil occasionally epiphytic up tree bases  1 a  TJ  •2 •3 •c  on soil and humus among litter  logs (rarely)  on logs, in swamps, on sandy soil, or in lawns seepages, along water courses among grass  rotten logs, stumps, humus, litter  O  sealevelto subalpine (to about 1700m)  acidic ro ck outcrop s  rocks and trunks of trees  soil among heaths  bases of trees, or rotten logs  noncalcareous rock, soil, humus, earth of banks 1  sometimes on twigs on the ground,rarely on rock  branches of living trees  gravelly or sandy soil, rock, boulders, outcrops (sterile, acidic substrates)  sometimes on rock  twigs and branches of trees or shrubs, especially in late snow areas  occasional on rotten logs  in rock crevices; Sphagnum hummocks  sterile litter, humus, soil, hummock tops, rock, occasionally ascending tree bases and in bogs cliff ledges  1  sealevelto subalpine (to about 1300m)  generally subalpine to alpine;fromsea level about 2000m  subalpine to alpine elevations  sea level to subalpine; up to about 3400m  arctic-alpine; in the mountains usually in coniferous woods, shaded  montane and subalpine coniferous forests; not usually found in op en site s  lawns, wet grassy slopes, swampy areas; usually somewhat shaded  in humid coastal lowland and montane coniferous forests  in the woods and along streams; moist  moist, shaded to open areas  roadsides, acidic glacial outwash; often on dry substrates in open areas  ] forests  humid montane forests, usually somewhat shaded  subalpine forest  coniferous forest  montane and boreal forests, often in barren places; variable moisture, light  boreal and montane forests, bogs; generally moist, shaded to open  often on rotten logs, stumps, or bases of trees sometimes on soil and humus overrock, or on humid coastal coniferous forests, shaded areas forest floor humus or boggy soil  in the PNW most often above 500m, to 2300m earth ofbanks, cliff crevices, overturned tree or higher (to alpine elevations) roots, forestfloor,mineral soil  from the lowlands to subalpine and alpine elevations  mainly at low elevations to 2000m or higher (occasionally into alpine)  low elevations  rotten logs and stumps, bases of trees  at high elevations, usually above 700m, or sometimes in the lowlands  humus or soil,frequentlyon soil over boulders and cliffs  | coniferous forest  |on rock, soil and rotten logs  |  from about 1000-2000m  J on humus or at tree bases  rock, soil, rotten logs, rooftops  epiphytic on tree trunks, branches, and on shrubs  sealevelto 1300m or higher (to subalpine)  from the lowlands to ab out 2200m  humid coniferous forests; shaded to open  occasionally in humus in subalpine forest, sometimes on rock  ]especially subalpine forest humid fore st  j  especially subalpine forest non-calcareous rock, sometimes on soil  | |  | subalpine forest and heath slopes  |rarely on rotting logs or decayed wood  forests; usually somewhat shaded  |  |  sometimes on humus over soil or rock  subalpine forest by roadsides and margins of woods; damp, shaded  on logs in early stages of decomposition, on tree trunks  1  1 rock  1humus or soil over rock (terrestrial)  1  humid coniferous forest; coastal  sealevelto about 1300m(to subalpine)  from the lowlands to about 1800m  from 600m to 2000m  usually at high elevations, as low as 500m  on decaying wood and humus; on living tree trunks  g -a o rS  robusta  o o o rs a  S  Rhytidiopsis  B  from the lowlands to 1000m or higher (to subalpine)  o  fitscescens  V) U  soil, rotten wood(e.g. logs)  %  banks of bare mineral soil, recently disturbed soil  vt  cliffs, boulders, rock (rarely); dry hummocks in peatlands  %  sea level to 2000m or higher  |  I  trees (trunks, branches), logs  primary substrata  S  Dicranum  1  sealevelto subalpine  1  Antitrichia  curtipendula  Appendix 3. Literature summary of bryophyte ecology: information for mosses and liverworts identified in this study.  r-  |  o o  CN  d  g D  .s  *  1a  l i  n  species  1  Cl  8  ?  i  Cl K  Q  £  Cl  E  !  a-  «J  1E  c JJ  a «  a *£  <3  .a fe  1  1  1  ,<J  fe i  a  1  fe  1  | *  *  13  JB  0 «  &  Cl  <5  ©  •S  S3  r  .CJ  §  5" B£  «-1  I'i  • tl  6  °  II &s  I. |gf *  s  •«! Q JJ  |  O  V  ©  8" •X  o> w  1  also growing on rock substrates, on humus and sandy soil over boulders, outcrops, and cliffs  open to partially shaded  substrate dry to mesic, moist sites substrates are dry to mesic mesic to moist areas; substrate moist  humid west coast rain forest  forest; near the coast at higher elevations  forest (especially lowland coastal rain forest)  £»  a  |  fi  js  „§  2-  .ft.  •S2  9  §•  usually growing on sandy soil (occasion- occasionally growing on decaying wood, as on sand over a decaying stump and on a ally with humus or among bryos and lichens) over rock substrates, on boulders decaying log on boulder-slopes, and on cliff faces  dry to mesic sites  can tolerate subacidic andbasic conditions moist areas (e.g. seepage)  in+/- exposed sites  on acidic soil with Lepirep, Blephtri, Cephbic; on humus over rocks with Lepirep, Geoc, Bleph, Cephbic  ' can grow in rather ins olated Bleph (basic sites) places  partially shaded sites  usually partially shaded  Cephlun partially shaded mesic substrates  forest infrequently growing on Utter over rock outcrops, tree bases with decayingbark, on moist soil near streams, on damp rocks, or in dried out bogs  on peaty humus/soil over basic rocks  * cr  usually occurs on bare or soil-covered rocks  3 =:  p in very humid sites, also growing on rocks, boulders, and outcrops; occasionally epiphytic on tree trunks  o  very common on decaying logs and stumps  jf^ also found on fallen and decaying logs, stumps, bark or wood; rarely growing over rock  °  usually epiphytic, occuring on several species of trees  -a  scandica  4  on organic substrates; decaying wood, living trees, and on humus and litter  with other Lophozia spp., also withGeocgra, Cephspp, Bleph, and occasionally Lepirep  primarily a forest species, occasionally found in sites adjacent to the forest e.g. fields, boulder-slopes, or subalpine parkland  occasionally with humus, mosses and litter; never over bare rock or inorganic substrates  generally with Blephtri, Dicrfus, Plaglae; with Dipltax on humus, Pohlnut on logs shade  sites are mesic to moist; usually in partial shade humid; unable to tolerate flooding  in moist, deeply shaded woods; inable to tolerate very basic or very acidic conditions; low toleration of light  also onbases of living trees; infrequently found on humus over rock outcrops  generally growing on organic substrates, especially on rotting wood(decaying logs, stumps, well into decay), and on humus and peat  Lepirep and Bleph  on logs with Lophgut, Blephtri, and peaty margins of acid ponds with Cephbic  damp or moist substrates  substrate mesic to moist; usually partially shaded constant supply of water (requires relatively diffuse light)  occurring in humid forests; tolerance for sub-calcareous conditions  also growing on rocks, cliff walls (with or without litter), and on mud and clayey sand  on decaying logs, rotting wood, roots of overturned trees, litter, humus, and peat  rock, i.e., often with humus.soil, or litter, over rocks, boulders, and outcrops  open to partially shaded; in substrate generally mesic; needs high atmo- xeric conditions, limitied to extremely sheltered sites spheric humidity  moderate shade to open sites  along trails and streams; calciphobe (pH 4.5-5.6); form extensive mats, competing with a few mosses  growing on on stream banks and near humid places waterfalls in cold, humid places along cold mountain streams  also growing on soil banks  on the W coast of VI, growing on decaying logs, trunks of trees, and peaty rotting stumps, in rainforest and Sphagnum bogs  substrate usually mesic partially shaded; without strong toleration for dire ct sunlight  primarily on decaying wood, especially on occasionally found on humus on the ground; predominantly in forests (coniferous and logs (persists until logs have decayed infrequently, on tree bases; soil banks, moist deciduous), also on cliffs on ocean beaches; wide pH tolerance (3.8-6.3) into humus) rocks (CO)  partially shaded to open;seems intolerant of desiccation  substrate mesic  sand or sandy humus over rocks (boulders, outcrops, rock walls)  Dipltax, Lophgut, Cephsp., Calymuell  Dicranum sp.  in the forest(JG); old logging roads/paths substrate mesic to moist; in partial shade; absence of on damp rocks with Bleptri, Lepirep reasonably constant direct sunlight (SHU) water supply  partially shaded to open; requires relatively diffuse Hght  primarily on decaying wood(especially also frequently on humus and sandy soil, on predominantly in forests; also on ocean beaches, in bogs, and in subalpine on logs, also on stumps and rotting wood the ground or over rock (acidic) meadows on the ground)  I5 To«  Scapania  i .tf a, B  bolanderi  P. JJ  Scapania  IA  'a  cordaeana  •f •rf  Porella  i o, o  guitulata  3 1  Lophozia  3  bicuspidata  1  on decaying logs and stumps (JG); often occasionally growing on humus and litter on thin layers of soil or humus, over damp (over soil) on the ground (JG); rare on decaying logs (SHU) rocks or over peat (SHU)  .1  usually mesic to moist, in the forest or along stream banks; as threads among other bryos, or occasionally or becoming dry when on rock; perfers (especially when on rock) in pure tufts; constant moisture avoids calcareous sites  U  with climax mosses e.g.  8"  occasionally on humus and litter on the ground; less frequently on trees; also on sand, soil, humus, or among mosses over inorganic substrates  |  a  in rather deep and permanent shade  light s .3 e  mesic abundant at high elevations, occuring in scattered or diffuse patches under conifers  u *>  organic substrates, i.e. decaying wood (logs, stumps); also on damp, shaded often vertical rock faces  1 vt  Cephalozia  other substrata .s  usually on rocks (acidic) or over boulders humic sites covered with needle humus  A I  j  oo  ON  g  a  

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.831.1-0089382/manifest

Comment

Related Items