Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ecosystem responses to climate variability, disturbances and environmental factors in southwest Yukon… Paudel, Shyam Krishna 2013

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

Item Metadata

Download

Media
24-ubc_2014_spring_paudel_shyam.pdf [ 3.2MB ]
Metadata
JSON: 24-1.0165664.json
JSON-LD: 24-1.0165664-ld.json
RDF/XML (Pretty): 24-1.0165664-rdf.xml
RDF/JSON: 24-1.0165664-rdf.json
Turtle: 24-1.0165664-turtle.txt
N-Triples: 24-1.0165664-rdf-ntriples.txt
Original Record: 24-1.0165664-source.json
Full Text
24-1.0165664-fulltext.txt
Citation
24-1.0165664.ris

Full Text

 ECOSYSTEM RESPONSES TO CLIMATE VARIABILITY, DISTURBANCES AND ENVIRONMENTAL FACTORS IN SOUTHWEST YUKON OF CANADA  by SHYAM KRISHNA PAUDEL M.Sc. (Distinction), International Institute for Geo-information Science and Earth Observation (ITC), Twente University, The Netherlands, 2000  A THESIS SUBMITTED IN PARTIAL FULLFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Forestry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) November, 2013  ? Shyam Krishna Paudel, 2013ii  Abstract  The south-western part of the Yukon Territory of Canada has experienced an unprecedented spruce bark beetle outbreak and frequent forest fires beyond the historical trends. Accumulating evidence also suggests that the southwest Yukon has experienced the impacts of recent climate change: warmer winters and warmer and drier summers over the past 15 years have contributed to the severe spruce bark beetle infestation, affecting white spruce on approximately 340,000 hectares of the traditional territory of the Champagne and Aishihik First Nations in southwest Yukon. The mortality caused by the bark beetle outbreak has also increased the risk of wildfire severity and frequency in the region.  I studied the impacts of climate variability, disturbances and environmental factors on stand structure, forest regeneration and vegetation diversity within the CATT. The research was conducted in close collaboration with the Ministry of Energy, Mines and Resources of the Yukon Government and the Champagne and Aishihik First Nations Government. Data were collected in the summer of 2008 from the forested landscape of the Champagne and Aishihik Traditional Territory.  I found that stand structure varied significantly by edaphic and topographical factors with higher forest productivity at lower elevation and on lower slope positions where there was a higher soil moisture regime. Overall stand productivity and vegetation diversity were higher on warmer aspects and in mixed stands. Although regeneration of all tree species was higher in burned areas, broadleaved species prevailed in these areas, indicating that persistent disturbances associated with the predicted increase in temperatures in the region may promote broadleaved species. A higher diversity was found in moderately disturbed open areas with higher mean temperature and precipitation. Salvage harvested areas had the highest diversity and highest composition of iii  broealeaved trees. The undisturbed mixed stands had the ecosystem characteristics that would closely meet the ecological goals of the Strategic Forest Management Plan (2004) in the region. The vegetation distribution were closely linked with topographical, climatic and disturbance regime in the study area, which could be a basis for vegetation classification, which is still lacking in the study area.  iv  Preface  Disturbance is an integral part of the boreal forest ecosystem, playing an important role in forest succession and helping to maintain ecosystem functioning when it occurs within historical limits. However, this is not the case of the south-western part of the Yukon Territory of Canada, where the forests have been affected by disturbances beyond the historical range of natural variability. The region has experienced a severe spruce bark beetle infestation since early 1990s, affecting white spruce on approximately 340,000 hectares of the traditional territory of the Champagne and Aishihik First Nations (CAFN), the study area of this research. The mortality caused by the bark beetle outbreak has increased the risk of wildfire severity and frequency. Between 1946 and 2003, 23 large fires and many small fires were recorded, burning approximately 100,100 ha of forest in the planning region. In 1998, there were two large forest fires near Haines Junction, affecting about 7000 ha of the forests. In the absence of active fire management, the average annual frequency of fire and total area burned in the Yukon Territory are projected to be doubled by the 2070s.  Climate change has been considered as the dominant factor for the increased forest disturbances in the Champagne and Aishihik Traditional Territories (CATT). An increase in mean annual temperature (MAT) of 0.6 ?C, with the greatest increase occurring in January (2.4 ?C) were recorded in the years between 1971?2000 compared to 1951?1980. The model projections indicated that the MAT in the region will be increased by 2 ?C by 2040 and 7 ?C by 2100, which poses greater uncertainty of future climate and forest disturbances in the region. The CAFN did not have prior forest management experiences except for limited forestry activities resulting in the widespread occurrence of old white spruce stands, which is also another reason for increased forest disturbances in the region.   The majority of the forestland is presently inaccessible due to the lack of a road network. Moreover, the local communities have very small populations (approximately 1200 people in the region), and there is no history of involvement in commercial forestry. Their economy is mainly based on traditional land-use practices. Forests provide significant socio-economic and cultural values to the local people; traditional practices include hunting, fishing, trapping and the collection of non-timber forest products (NTFPs). The major sources of income for local people are employment in service sector, wilderness-based tourism and small service businesses.  v   Forest disturbances have caused significant impacts on forest ecosystems in the region, specifically on forest composition, structure and vegetation diversity. About 80% of the dominant white spruce trees are dead, turning the landscape into a mosaic of dead conifer stands mixed with early successional broadleaved species. This situation may also significantly affect the local people?s traditional use of the forests. In response to the forest disturbances, a Strategic Forest Management Plan was prepared and and has been implemented since 2004. The plan has four goals and 33 objectives aiming to balance ecological, economic and social values from the affected forests. However, the plan lacked baseline information on existing ecosystems and their variations in the landscape, a clear roadmap for future development and silvicultural systems to achieve its goals and objectives. The plan did recognize salvage harvesting as an important forest intervention in response to forest disturbances. However, questions have been raised over the effectiveness of salvage harvesting in mitigating forest disturbances and reducing ecosystem impacts.  This research was triggered due to pressing needs to fill the information and knowledge gap on existing ecosystem components, specifically on vegetation composition and diversity, forest productivity, and natural regeneration dynamics and their relationship with biogeoclimatic and disturbance factors. In addition, information was sought on how these natural phenomena would interact with the cultural and socio-economic systems within the area. This knowledge and information would lay the foundation for devising appropriate silvicultural systems in support of the SFMP.  Accordingly, this research was conceptualized as per the agreement between Ministry of Energy, Mines and Resources, Government of Yukon and Faculty of Forestry, UBC, in order to assist the long-term management of the forests in the region. Over the course of the research, many changes occurred in the staff of the Ministry, and priorities changed. In addition, it soon became evident that the appetite amongst local people for more socio-economic surveys was very limited, and I was advised not to proceed with this part of the project. The following account reflects the research that emerged as a result of these changing circumstances.     vi  Table of Contents Abstract.. ........................................................................................................................................... ii Preface.... .......................................................................................................................................... iv Table of Contents .............................................................................................................................. vi List of Tables ..................................................................................................................................... x List of Figures .................................................................................................................................. xii List of Abbreviations ...................................................................................................................... xiv Acknowledgements .......................................................................................................................... xv 1. Introduction ......................................................................................................................... 1 1.1 Setting: the boreal ecosystem .............................................................................................. 1 1.2 Context: boreal forests in the Champagne and Aishihik Traditional Territory (CATT), southwest Yukon ........................................................................................................................... 2 1.2.1 Forest ecosystems in the CATT ................................................................................... 2 1.3 Research rationale: climate change and forest disturbances in the CATT .......................... 4 1.3.1 Climate change and its impacts in the CATT .............................................................. 4 1.3.2 Spruce bark beetle outbreak ......................................................................................... 5 1.3.3 Forest fire hazards ........................................................................................................ 6 1.3.4 Salvage harvesting ....................................................................................................... 7 1.4 Current understanding of boreal forest disturbances ........................................................... 8 1.4.1 Disturbance ecology ..................................................................................................... 8 1.5 Research objectives and questions .................................................................................... 15 1.5.1 Research questions ..................................................................................................... 16 1.6 Research framework and thesis structure .......................................................................... 17 2. Methodology ..................................................................................................................... 19 2.1 Study area justification: Why was the research important in the forest of CATT? .......... 19 2.2 Study area description ....................................................................................................... 20 2.3 Data collection ................................................................................................................... 22 2.3.1 Sampling methods ...................................................................................................... 22 2.3.2 Variables .................................................................................................................... 24 2.4 Data analysis ...................................................................................................................... 27 2.4.1 Stand structure analysis (Chapter 3) .......................................................................... 27 2.4.2 Regeneration analysis (Chapter 4) ............................................................................. 27 2.4.3 Vegetation diversity analysis (Chapter 5) .................................................................. 28 3. Stand structure variability in response to bio-geoclimatic and disturbance factors in southwest Yukon ............................................................................................................................. 35 3.1 Introduction ....................................................................................................................... 35 vii  3.2 Research problems, hypotheses and questions .................................................................. 38 3.3 Data analysis ...................................................................................................................... 40 3.3.1 Calculation of average diameter and height ............................................................... 40 3.3.2 Calculation of stand structure and composition ......................................................... 40 3.3.3 Density index ............................................................................................................. 41 3.3.4 Calculation of variation of structural diversity index across disturbance regime ...... 41 3.3.5 Calculation of correlation coefficients and scatter plots ............................................ 42 3.3.6 Multiple co-inertia analysis (MCoA) ......................................................................... 42 3.3.7 Geographical information system (GIS) .................................................................... 43 3.4 Results ............................................................................................................................... 44 3.4.1 Stand composition ...................................................................................................... 44 3.4.2 Average height, diameter and basal area of species by height class .......................... 44 3.4.3 Shannon-Weiner structural diversity Index across disturbance regime ..................... 45 3.4.4 Analysis of stand structure variations using scatter plots and ANOVA plots  .......... 45 3.4.5 Impacts of forest disturbances on stand structures ..................................................... 47 3.4.6 Multivariate analysis of stand structure variables using MCoA ................................ 48 3.5 Discussion ......................................................................................................................... 49 3.5.1 Relationship between stand density, basal area, age and tree size ............................. 49 3.5.2 Variation of stand structures by disturbance regime .................................................. 53 3.5.3 Variation of stand structures with climatic factors .................................................... 54 4. Tree regeneration in response to bio-geoclimatic and disturbance factors in the southwest Yukon of Canada ............................................................................................................................. 74 4.1 Introduction ....................................................................................................................... 74 4.1.1 Tree species and their regeneration in the southwest Yukon ..................................... 74 4.1.2 Climatic and disturbance factors and tree species regeneration in SWY................... 75 4.1.3 Research problems, hypotheses and questions .......................................................... 77 4.2 Data analysis ...................................................................................................................... 78 4.2.1 Multivariate canonical correlation (MCA) ................................................................ 79 4.2.2 Negative binomial generalized model ........................................................................ 79 4.2.3 The most nearest neighbor imputation model (MSN Imputation) ............................. 80 4.3 Results ............................................................................................................................... 82 4.3.1 Regeneration in response to disturbances, and climatic and environmental factors .. 82 4.3.2 Negative binomial regeneration abundance model .................................................... 83 4.3.3 MSN imputation model .............................................................................................. 84 4.4 Discussion ......................................................................................................................... 85 4.4.1 Disturbances vs. regeneration .................................................................................... 85 viii  4.4.2 Climate variability vs. regeneration ........................................................................... 87 4.4.3 Edaphic/topographic factors vs. regeneration ............................................................ 88 4.5 Chapter conclusions .......................................................................................................... 90 5. Vegetation diversity in response to landscape variability, disturbances and community dynamics in Southwest Yukon ...................................................................................................... 101 5.1 Introduction ..................................................................................................................... 101 5.1.1 Biodiversity and its importance ............................................................................... 101 5.1.2 Importance of studying vegetation diversity in SWY .............................................. 103 5.1.3 Research Issues: impacts of landscape variability, climate and disturbances on vegetation diversity ................................................................................................................ 105 5.1.4 Research problems, hypotheses and questions ........................................................ 107 5.2 Data collection and analysis ............................................................................................ 108 5.2.1 Richness, abundance and evenness analysis ............................................................ 109 5.2.2 Diversity indices ...................................................................................................... 109 Shannon-Wiener index, Simpson index and Renyi?s profiles were calculated; these are briefly described in the following sections. ....................................................................................... 109 5.2.3 Species spatial distribution and range ...................................................................... 113 5.2.4 Canonical correspondence analysis (CCA) .............................................................. 113 5.2.5 Diversity index mapping .......................................................................................... 114 5.3 Results ............................................................................................................................. 115 5.3.1 Characteristics and severity of disturbed plots ........................................................ 115 5.3.2 Species richness, abundance and spatial distribution ............................................... 116 5.3.3 Vegetation diversity in response bio-geoclimatic and disturbance factors .............. 118 5.3.4 Canonical correspondence analysis (CCA) .............................................................. 120 5.4 Discussion ................................................................................................................ 121 5.4.1 Variation in vegetation diversity by bio-geoclimatic and disturbance factors ......... 121 5.4.2 Overall plant diversity and distribution in the study area ........................................ 125 5.5 Chapter conclusions ........................................................................................................ 127 6. Forest Management in response to climate variability and disturbances in the CATT .. 137 6.1 Introduction ..................................................................................................................... 137 6.1.1 Current status of forest management in the CATT .................................................. 138 6.2 Key research findings and implications for forest management in the CATT ................ 139 6.2.1 Projected ecosystems of the CATT in the absence of forest management .............. 140 6.2.2 Suggested forest management strategies in the CATT ............................................ 143 6.2.3 Management of disturbed stands: salvage harvesting t ............................................ 143 ix  6.2.4 Pro-active forest management to maintain forest productivity, biodiversity and species regeneration ............................................................................................................... 145 6.3 Chapter conclusions ........................................................................................................ 152 7. Conclusions and recommendations ................................................................................ 153 7.1 Major findings ................................................................................................................. 153 7.2 Socio-economic implications .......................................................................................... 159 7.3 Forest management implications ..................................................................................... 160 7.4 Applicability of the research to other boreal regions ...................................................... 163 7.5 Research limitations ........................................................................................................ 164 7.6 Recommendations and future research ............................................................................ 165 Postscript ........................................................................................................................................ 166 References ...................................................................................................................................... 169 Appendix A: Scientific and common names of vegetation, their code names and type ................ 199 Appendix B: Descriptive analysis of collected variables .............................................................. 201 Appendix C: Multiple Co-inertia analysis outputs ........................................................................ 203 Appendix D: Canonical correlation outputs .................................................................................. 205 Appendix E: Full imputed and observed values ............................................................................ 208 Appendix F: Species richness and abundance (%) by sites ........................................................... 210 Appendix G: Abundance rank and spatial distribution of species in the study area ..................... 211 Appendix H: Species richness, abundance, Shannon and 1/Simpson indices .............................. 214 Appendix I: Canonical Correspondence Analysis outputs ............................................................ 219 Appendix J: List of vegetation locally used in Yukon ................................................................... 223     x  List of Tables  Table 2.1 Characteristics of disturbed plots ..................................................................................... 29 Table 2.2 Research blocks and their characteristics ......................................................................... 29 Table 2.3 Plot sizes, types of data collected and brief methodology ............................................... 30 Table 2.4 Variables (with full and code names) collected and generated for analysis .................... 31 Table 2.5 Categorized variables ....................................................................................................... 32 Table 3.1 Total number of trees, seedlings and saplings of the species sampled ............................ 59 Table 3.2 Stand composition index .................................................................................................. 59 Table 3.3 Average height, diameter and BA of the sampled tree species by dominance class ....... 60 Table 3.4 Shannon-Weiner structural diversity indices by disturbance types (Bold fonts indicate the highest values by species) .......................................................................................................... 60 Table 3.5 Stand structure variables that were significant with environmental variables ................. 61 Table 3.6 R2 values between stand structure variables and other ecological variables ................... 62 Table 3.7 Comparing density index by disturbance type ................................................................. 62 Table 3.8 Ratio of tree regeneration to parent trees by disturbance types ....................................... 63 Table 4.1 Mean density (number/ha) of trees and their regeneration by disturbance types ............ 92 Table 4.2 Axes and their association with environmental data (MCA) (correlation coefficient) .... 92 Table 4.3 Synthesis of canonical correlation analysis (correlations between the tree regeneration and the canonical axes of environmental data .................................................................................. 93 Table 4.4 Criteria for assessing goodness of fit ............................................................................... 93 xi  Table 4.5 Analysis of maximum likelihood parameter estimates of white spruce .......................... 94 Table 4.6 Analysis of maximum likelihood parameter estimates of trembling aspen ..................... 94 Table 4.7 Analysis of maximum likelihood parameter estimates of balsam poplar ........................ 95 Table 4.8 AIC and variables dropped in the models ........................................................................ 95 Table 4.9 Chi-square tests of the models ......................................................................................... 95 Table 4.10 Imputed and observed regeneration abundance for the first and last six plots .............. 96 Table 5.1 Characteristics of disturbed plots ................................................................................... 129 Table 5.2 Species richness and abundance in the study area ......................................................... 129 Table 5.3 Distribution and abundance of dominant vegetation by disturbance type ..................... 130 Table 6.1  Key research findings and forest management implications and considerations .......... 149 Table 7.1: Summary table: responses of stand structure variables, regeneration and vegetation diversity to bio-geoclimatic and disturbance factors...................................................................... 158    xii  List of Figures  Figure 1.1 Research framework ....................................................................................................... 18 Figure 2.1 Map of the study area showing blocks and plots where the data were collected. .......... 33 Figure 2.2 Plot design....................................................................................................................... 34 Figure 3.1 Species distribution in study area ................................................................................... 63 Figure 3.2 Distribution of average height (m) in the landscape ....................................................... 64 Figure 3.3 Stand diameter Vs environmental factors ....................................................................... 65 Figure 3.4 Stand heights vs. environmental variables ...................................................................... 66 Figure 3.5 Stand productivity vs. environmental variables .............................................................. 67 Figure 3.6 Stand basal area vs. environmental variables ................................................................. 68 Figure 3.7 White spruce average density vs. environmental variables ............................................ 69 Figure 3.8 Trembling aspen and balsam poplar average density vs. environmental variables ........ 70 Figure 3.9 Variation of stand density and basal area by disturbance type ....................................... 71 Figure 3.10 Axis I - Environmental/structure variables ................................................................... 72 Figure 3.11 Axis II - Species environmental axis ............................................................................ 73 Figure 4.1 Responses of white spruce regeneration (number/ha) to environmental factors ............ 97 Figure 4.2 Responses of trembling aspen regeneration to environmental factors ........................... 98 Figure 4.3 Responses of balsam poplar regeneration (number/hectare) to environmental factors .. 99 Figure 4.4 Relationship between imputed and observed values .................................................... 100 Figure 5.1 Vegetation diversity in the landscape (Shannon-Wiener index) .................................. 131 xiii  Figure 5.2 Species accumulation curve .......................................................................................... 132 Figure 5.3 Renyi?s profile showing vegetation diversity by site variables and disturbance types 133 Figure 5.4 Renyi?s profile showing vegetation diversity variation by climatic factors and stand types ............................................................................................................................................... 134 Figure 5.5 Distribution of species and environmental variables in the research sites ................... 135 Figure 5.6 Sites position by environmental variables combinations .............................................. 136 Figure 6.1 Variation of ecosystem attributes by disturbance and stand types in the CATT based on the findings of the research. ........................................................................................................... 142 Figure 6.2 Suggested forest management strategies in response to variation of ecosystem attributes by disturbance and stand type in order to reach desired ecosystem states ..................................... 148    xiv  List of Abbreviations  ACIA: Arctic Climate Impact Assessment ANOVA: Analysis of Variance AIC: Akaike Information Criterion CAFN: Champagne and Aishihik First Nations CCA: Canonical Correspondence Analysis CM = Centimeter GIS: Geographical Information System GLM: Generalized Model  IDH: Intermediate Disturbance Hypothesis KNPR: Kluane National Park and Reserve KM: Kilometers MCA: Multivariate Canonical Correlation MCoA: Multiple Co-inertia Analysis MEMR: Ministry of Energy, Mines and Resources (Yukon Government) MSL: Mean Sea Level NCE: Northern Climate Exchange PoC: Point of Commencement SFMP: Strategic Forest Management Plan SWY: Southwest Yukon YG: Yukon Government    xv  Acknowledgements  Completion of this dissertation would not have been possible without the continuous support and encouragement from a number of people who deserve sincere appreciation and gratitude.   Research committee: First of all, I would like to express my sincere gratitude to my research supervisor John Innes and to my two advisory committee members Craig Nitschke and Suzanne Simard who guided me throughout the research process, from conceptualization to final dissertation. I enjoyed working with Prof. Innes who provided total freedom to explore my research ideas and helped me to focus on the target. My committee members helped me to frame the research idea and guided me whenever I needed their support. I would like to express my very special thanks to Craig Nitschke, who guided me from the beginning of my research, included me in his Yukon research team, helped me to design and collect the data from Yukon, helped me to use and interpret multivariate analyses and to frame my research chapters. Thank you very much to all of you.    Yukon and the Yukoners: I would like thank Aynslie Ogden (YG) and Roger Brown (CAFN) for providing local support and guidance for data collection in Yukon. I thank the Champagne and Aishihik First Nations government and all the community members for allowing us to work in their Territory. I would also like to thank Brent and Linda for providing us with a feeling of being at home during my stay in Haines Junction for data collection. Yukon is deservedly known for its astonishing beauty and doing research in such a place is a life-time experience.   The Yukon research team: The Yukon research team received much support from many colleagues, including international interns. I would like to thank Patrick Waeber, Finella Pescott, Jan Willem xvi  Klaassen, Jean-Sebastien Jacquet, Corentin Cl?ment, Ambre Le Ferrec, Alyson McHugh and Raphael Chavardes for support in data collection, tabulation and analysis.   My lab mates: We have a very diverse and multi-cultural lab with students from all over the world.  My lab mates are always supporting, helping and sharing information for academic and professional growth. I would like to thank all of you for inspiring me during my work in the lab and for being my friends during my period at UBC: Judi Krzyzanowski, Angeline Gough, Reem Hajjar, Ajith Chandran, Monika Singh, Pano Skrivanos, Howie Harshaw; Suzi Malan, David Perez, Juan Chen, Tomoko Yoshida, Rebecca Klady, Patrick Waeber, Harris Gilani, Futao Guo, and Jos? Arias Bustamante, Guangyu Wang, Sang Seop Lim and Lianzhen Xu. Special thanks to Ajith and Monika Singh for their great helps during my final submission.  Statistical support: I would like to extend special thanks to Valerie LeMay, Tony Kozak and Indra Kshhetry for guiding me in statistics and research methods.  Funding agencies: This research would not have been possible without funding support from NSERC, PICS, the Yukon Territorial Government, and the UBC University Fellowship and Tuition Fee Awards    Family and friends: My thanks and appreciation to all my family members in Nepal and abroad, my friends and well-wishers who have been encouraging and motivating me in my study all the time. Finally, last but not least, special thanks to my wife Sunita and my kids Sunaina and Shishir for being part of my life and giving me hope and motivation for this study and for being proud of me and making me proud of  them.   1  1. Introduction 1.1 Setting: the boreal ecosystem  This research was undertaken in the landscape of the Champagne and Aishihik Traditional Territory (CATT), which is a part of Southwest Yukon?s (SWY) boreal forest. The ecosystems of the region are currently facing the impacts of climate change and related disturbances. This triggered the current research, which aims to provide ecological knowledge and information to policy makers and forest managers in order to assist the long-term management of the forests in the region.  Boreal forests cover 1.2 billion hectares of land, extending across about 17% of earth?s surface and comprising 30% of the global phytomass (Wein and MacLean, 1983; Stocks and Lynham, 1996). They play an important role in global climate, carbon and water cycles and the world economy (Johnson et al., 2003). About one third of the world?s boreal forest lies in Canada and Alaska (Hare and Ritchie, 1972).  The boreal forest varies significantly in climate and disturbance regimes (Chen and Popadiouk, 2002), with climate in the Canadian boreal ranging from dry and cold, with a mean annual temperature of ?8 ?C and a mean annual precipitation of 300 millimetres (mm), to warm and moist, with a mean annual temperature of 6 ?C and a mean annual precipitation of 1350 mm (Rowe, 1972). The characteristics and composition of the vegetation vary with climate, aspect, slope, soil type and moisture gradient (Meideniger and Pojar, 1991). The diverse physiographic and climatic factors have created diverse community types (Douglas, 1974). The forest composition is also strongly influenced by interactions between species regeneration capacity, longevity and the frequency and types of disturbance (Bergeron, 2000). 2  The overstory of the Canadian boreal forest is mainly dominated by trembling aspen (Populus tremuloides Michx.) or white birch (Betula papyrifera Marshall) in early successional stages, black spruce (Picea mariana (Mill.) B.S.P.) or white spruce (Picea glauca (Moench) Voss) in mid-successional stages and by balsam fir (Abies balsamea (L.) Mill.) in late successional stages (MacDonald, 1995). Understory vegetation plays an important role in ecosystem functioning by affecting canopy succession (Messier et al., 1998), and nutrient cycling (Weber and Van Cleve, 1981), by providing wildlife habitat (Okland and Eilertsen,1996) and by providing many valuable foods and medicines to human society (Penafiel et al., 2011).   1.2 Context: boreal forests in the Champagne and Aishihik Traditional Territory (CATT), southwest Yukon 1.2.1 Forest ecosystems in the CATT The CATT is located in the southwest Yukon (SWY), covering 272,220 hectares (ha) of forest lands excluding the forest area in Kluane National Park and Reserve (SFMP, 2004). The region has 20% forested land suitable for forestry activities, 28.8% forested land unsuitable for timber production, 1.8% poorly regenerated, 42% alpine shrubs and 7.4% of remaining lands classified as rivers, settlements, lakes and wetlands (Ogden, 2006). The vegetation in the Kluane region has been divided into three ecological regions based on elevation, namely montane valley bottom forest (780?1080 m), subalpine forest (1080?1370 m) and alpine tundra (above 1370 m) (Krebs and Boonstra, 2001). Forests are mostly confined to below 1100 m in the river valleys. Due to soil, topographic and climatic constraints, the forests in the region are characterized as slow- growing with low productivity (NCE, 2006a). 3   Forest ecosystems in the southwest Yukon include a small number of tree and understory species. According to SFMP (2004), the major tree species are white spruce (Picea glauca (Moench) Voss, 235,000 ha), trembling aspen (Populus tremuloides Michx., 36,300 ha) and balsam poplar (Populus balsamifera L., 920 ha). Less common tree species include lodgepole pine (Pinus contorta Dougl. ex Loud.), black spruce (Picea mariana (Mill.) B.S.P.) and subalpine fir (Abies lasiocarpa (Hook.) Nutt.). Trembling aspen is generally found in pure stands on well-drained sites at moderate elevations and on south-facing slopes. The forest understory is mainly composed of deciduous shrubs such as ground birch (Betula glandulosa Michx.) and willow (Salix spp.).  The understory also includes a variety of herbs such as arctic lupine (Lupinus arcticus Wats.) and fireweed (Epilobium angustifolium L. s.l.) (NCE, 2006a).   Disturbances play a dominant role in forest succession and community dynamics in southwest Yukon forests (Ogden, 2006). Prior to the spruce bark beetle infestation (before 1994), 80% of the forests were classed as mature stands between 80 and 120 years old (SFMP, 2004). However, the spruce bark beetle (Dendroctonus rufipennis) infestation has affected the forest age class and stand height distributions in recent years. The infestation is expected to affect species composition over time in a large area of the planning region (SFMP, 2004). The current forest composition in the affected areas comprises an overstorey of dead white spruce with a variety of regenerating plant species. In general, the forest in the region is characterized by low productivity (with over 80% of the forest classified as poor) and slow growth rate (with an average of 10 to 14 m height attainment at 100 years of age) (SFMP, 2004).   4  1.3 Research rationale: climate change and forest disturbances in the CATT 1.3.1 Climate change and its impacts in the CATT The IPCC 4th assessment report summarized that mean temperatures have increased globally; the warming is greater at higher latitudes where temperatures have increased at almost twice the global average rate in the past 100 years. Frugal and Prowse (2008) state that climate in the Canadian north is changing significantly, with increased temperature and precipitation being reported over the last 50 years. Model projections suggest that the increase in temperature will be higher in northern regions than in the rest of the world (Solomon et al., 2007), with the greatest winter warming (from 2 to 9 ?C) being projected in Alaska and northern Canada (Christensen et al., 2007; Anisimov et al., 2007). Annual winter precipitation is also expected to increase at higher latitudes (Christensen et al., 2007).  Climate change and its effects are apparent in the CATT (Ogden, 2006). The mean annual temperature in the region rose by 3 ?C between 1955 and 2005 (Field et al., 2007). The meteorological record from the Burwash Landing weather station reveals an increase in mean annual temperature of 0.6 ?C, with the greatest increase occurring in January (2.4 ?C) in the years between 1971?2000 compared to 1951?1980 (Ogden, 2006). In the future, the mean annual temperature in the region is projected to increase by 2 ?C by 2040 and 7 ?C by 2100 (ACIA, 2004). Moreover, increasing precipitation and shorter and warmer winters will decrease snowfall and snow cover in the region substantially through delayed snowfall and earlier spring snow melting (ACIA, 2004; Christensen et al., 2007). About a 10% decline in snow cover has been estimated in the past 30 years and a 10 to 20% decline has been projected by the 2070s (ACIA, 2004). 5  The impacts of climate change on natural and socio-economic systems are already apparent in southwest Yukon. Climate change has affected the extent of glaciers and permafrost, the frequency of floods and landslides, and has affected forests, wildlife and local communities. Permafrost is highly sensitive to disturbance and global warming (Smith and Burgess, 1999) especially during the winter season. Floods and fluvial erosion are apparent in the region due to snow and glacier ice melting precipitated by warmer weather (Ogden, 2006).   1.3.2 Spruce bark beetle outbreak  One of the most obvious problems related to climate change in the region is the spruce bark beetle outbreak (ACIA, 2004; NCE, 2006b). ACIA (2004) concluded that ?the region is experiencing the largest and most intense infestation of spruce bark beetle ever recorded in Canada?, which is an apparent example of ecosystem response to recent warming (Frugal and Prowse, 2008). Beetles attained epidemic levels in the region after 1992 (SFMP, 2004), although the infestation began to subside in 2007. Approximately 340,000 hectares of forest were affected by the beetle outbreak (NCE, 2006; Government of Yukon, 2010), which is the second largest infestation after the infestation in south central Alaska, where white spruce has died over about 1.2 million ha (Berg et al., 2006). Using aerial mapping, it was determined that white spruce forests affected by spruce beetle in 2007 covered only 10,286 ha, compared to 41,170 ha in 2006 and 82,620 ha in 2005 (Garbutt, 2007; Garbutt, 2008). The magnitude, intensity and severity of this infestation have been linked to recent climate change together with the presence of large areas of unmanaged, mature stands of white spruce (Barber et al., 2000; Berg and Henry, 2003; ACIA, 2004). As cold winters generally 6  suppress the spruce bark beetle population, the recent mild winters are thought to be the main reason for the major increase in the beetle population (SFMP, 2004). In addition, already susceptible mature and weaker trees have lost their resistance to beetle attack due to warmer summers and moisture stress in the region (ACIA, 2004; Juday et al., 2005). If the climate trends continue, white spruce mortality is expected to continue because of the higher population of spruce bark beetle (Berg and Henry, 2003; Berg et al., 2006).  1.3.3 Forest fire hazards Climate variability is considered to be one of the dominant factors affecting the occurrence of forest fires (Rosenzweig et al., 2007), and has contributed to an increase in forest fire hazard in the region. Because of weather patterns and fuel structures, fire is an important disturbance agent in the boreal forest (Johnson, 1992). Between 1946 and 2003, 23 large fires and many small fires were recorded, burning approximately 100,100 ha of forest in the planning region, excluding the Kluane National Park and Reserve (KNPR) (SFMP, 2004). Although, historically the region has been considered as an area of low wildfire probability compared to other areas in the boreal forest (NCE, 2006b), the situation has changed over the last decade due to the increased fire hazard in the forest caused by changes in climate, spruce mortality (thus increasing fuel loads) and changing forest structure (SFMP, 2004).  The white spruce mortality associated with the bark beetle infestation has contributed to an increase in the quantity and extent of fuel (Frugal and Prowse, 2008), which in turn has increased the possible severity of future forest fires (Flannigan et al., 2005). In addition, warmer and drier conditions have changed the characteristics of the fire regime (annual fire incidences, area 7  burned and seasonal fire severity rating) in the region (McCoy and Burn, 2005). In the absence of active fire management, the average annual frequency of fire and total area burned in the Yukon Territory may double by the 2070s (McCoy and Burn, 2005). Moreover, the possibility of lightning strikes may increase, elevating lightning-caused fire incidents by as much as 44% (Price and Rind, 1994; NCE, 2006b). As a result, overall fire danger, fire intensity, fire severity, fire occurrence, length of the fire season, and annual areas burnt are all likely to increase (Stocks et al., 1998; Ogden, 2007), with significant ecological, social and economic impacts.  1.3.4 Salvage harvesting The SFMP (2004) recognized salvage harvesting as an important forest intervention in response to the spruce bark beetle infestation and increased fire severity in the region. The plan recommended that although the sustainable harvesting of timber1 should be the focus of SFM, salvage harvesting should be carried out to promote ecological recovery (SFMP, 2004), focusing especially on the beetle-infested area as timber quality will degrade over time. As part of the plan, the CAFN and Government of Yukon (GY) have accepted a recommendation for the salvage harvesting of up to 1 million m3 of beetle-affected timber between 2006 and 2015. However, questions have been raised over the effectiveness of salvage harvesting in mitigating forest disturbances and reducing ecosystem impacts. Specifically, research is needed to understand how salvage harvesting affects forest dynamics such as the development of stand structure and regeneration and its effect on biodiversity. A part of the research presented in this                                                  1Sustainable harvesting refers to harvesting of timber based on growth, yield and allowable cut.  8  thesis therefore focuses on assessing the potential effects of salvage harvest on the potential development of stand structure, biodiversity and regeneration.   1.4 Current understanding of boreal forest disturbances 1.4.1 Disturbance ecology Disturbance ecology has become an important concept in forest ecology. White and Pickett (1985) define disturbance ecology as ?...any relatively discrete event in time that disrupts ecosystems, community or population structure and changes resources, substrate availability or the physical environment?. However, Rogers (1996) suggests that the definition of disturbance ecology should not be bound by size and time, as these factors are relative to the systems under consideration. White and Pickett (1985) have also categorized disturbances as perturbations (small events with specific alteration in the system) and catastrophes (rare and large destructive events). Forest disturbances have also been discussed in terms of the temporal and spatial scale of equilibrium or non-equilibrium states that the disturbances may bring to the forest. It has been argued that when large-scale disturbances occur frequently, it is unlikely that the affected forest will reach equilibrium or a steady state (Bergeron and Dansereau, 1993; Romme et al., 1986). However, if disturbances are smaller in scale and occur rarely then a forest may reach a steady state (Bormann and Likens, 1979; Lorimer and Frelich, 1994). The recent theory of complex adaptive systems suggests however that non-steady and non-equilibrium conditions characterize forests (Levin, 2005), even when disturbance regimes are small-scale. This suggests that forest management needs to be based on heterogeneity, unpredictability and adaptability (Messier and Puettmann, 2011).  9  Disturbances can be explained in terms of cause, scale (temporal and spatial) and intensity (intense or weak) (Rogers, 1996).  The causes of forest disturbances may be biotic (such as insects, disease, animals and human interventions) or abiotic (fire, wind, floods, drought, etc). According to Angelstam (1996), fire is the most important cause of stand-changing disturbances. However, forest harvesting has been considered a very important disturbance factor in the boreal forest of Europe and many parts of North America (Niemela, 1999; Loope, 1991). The heterogeneity and mosaics created by harvesting operations often differ significantly from those associated with natural disturbances such as fire (Niemela, 1999).  Disturbance affects forest ecosystems in various ways. One of the important implications of disturbances for forest ecosystems is gap dynamics and thereby changes in resource availability. This can significantly change the succession pattern, forest structure and biodiversity (McCarthy and Weetman, 2006; Lorimer, 1989). Gaps can be smaller, created by the mortality of a single tree or patches formed by the mortality of groups of trees, or they can be as large as entire stands (Rogers, 1996). Smaller gaps or patches generally create uneven-aged stands (Oliver, 1980), while stand replacement disturbances generally create even-aged stands (Romme and Despain, 1989). However, Lorimer (1989) has stated that the effects of small-scale disturbances will be limited or insignificant if large catastrophic disturbances occur at intervals shorter than the potential lifetime of the trees. Shifting from the conventional paradigm, which has considered disturbance as a negative factor for ecosystem functioning and productivity, natural disturbance regimes have been recognized as an essential process to promote healthy and dynamic ecosystems. Consequently, suppressing them may be deleterious to the long-term healthy functioning of forest ecosystems (Rogers, 1996). An understanding of the specific ecosystem process is necessary if disturbance 10  ecology is to be implemented in forest management. Disturbances should be monitored at both community and regional levels so that their extent, frequency, rotation period, intensity and severity can be better understood (Noss, 1990).   1.4.2 Impacts of disturbances in boreal forests Disturbance is an integral part of the boreal ecosystem. Both natural and anthropogenic disturbances have played an important role in the succession of boreal forests (Kuusela, 1990), as the combination of these disturbances governs the dynamics of boreal forests (Engelmark, 1999). Fire has been considered as a paramount disturbing factor in the region, although other disturbances such as wind, floods, insects, climate variability and their interactions are equally important (Engelmark, 1999). Similarly, it is important to understand the variation in the disturbance regime through time as this can have a significant impact on forest composition. The impacts of climate variability and disturbances on regeneration, stand structure and biodiversity are reviewed separately in the following sections.  1.4.2.1 Impacts on tree regeneration  As my study area is dominated by three tree species, namely Picea glauca (Moench) Voss (white spruce), Populus tremuloides Michx. (trembling aspen) and Populus balsamifera, L. (balsam poplar), I will focus mainly on the regeneration of these species. Impacts of climate change and disturbances on white spruce regeneration have been reported by several authors. Climate change may have significant effects on regeneration and 11  growth of white spruce (Oswald and Brown, 1990; Barber et al., 2000; Wilmking et al., 2004; Nistchke, 2009). Variation in precipitation may reduce moisture availability or increase the frequency of drought, which would significantly affect the spruce regeneration of white spruce (Wang and Kemball, 2005; Johnstone and Chapin, 2006a).  Similarly, increased disturbances could also reduce seed sources through the extensive mortality of parent trees, which would affect total seed produced (Jasinski and Payette, 2005). Although fire is considered to be a very important factor for the establishment of white spruce seedlings as interactions between masting years and fire can result in abundant regeneration (Peters et al., 2005), intense and severe fire may kill seeds and reduce the probability of germination (Neary et al., 2005). In areas of the northern boreal forest, fire may play an important role for white spruce regeneration by removing organic layers, removing permafrost and increasing soil temperature (Yoshikawa et al., 2003).  Like white spruce, the regeneration of trembling aspen is also sensitive to dry climatic conditions (Hogg and Wein, 2005). Regeneration by suckering is the principal method of aspen regeneration following a major disturbance (Bartos, 2001; Frey et al., 2003). Its regeneration responds positively to increasing fire severity (Johnstone and Chapin, 2006a). However, both seed germination and vegetative reproduction are very sensitive to soil moisture stress (Perala, 1990). Moreover, severe fire can damage roots and suckers and significantly affect the density, height and diameter of the regeneration (Wang, 2003). Increased fire frequency resulting from warmer temperatures may favour the establishment of trembling aspen and other herbaceous woody species in the boreal forest (Flannigan and Van Wagner, 1991). Post-fire regeneration of aspen can be prolific and rapid where moisture is not limiting and the disturbance is followed by 12  suitable climate conditions (Hogg et al., 2002; Hogg and Wein, 2005; Johnstone and Chapin, 2006a).  Balsam poplar has similar environmental requirements to trembling aspen for its regeneration. It requires adequate soil moisture and prefers higher temperatures (> 15 oC) for its germination (Zasada and Densmore, 1980; Krasny et al., 1988; Zasada and Phipps, 1990). Soil moisture deficits can reduce regeneration following disturbance, leading to a reduction in species abundance over time (Yarie, 2008). Fire is an important disturbance factor for its germination, provided that adequate moisture is available (Klinka et al., 2000).    1.4.2.2 Impacts on forest structure and composition The magnitude and intensity of spatial and temporal disturbances shape the structural development of the landscape (McCarthy and Weetman, 2006) by creating periodic gaps (Lorimer, 1989) and by developing a mosaic of patches of varying age and size (McCarthy and Weetman, 2006). Variations in disturbance characteristics such as frequency, size and severity strongly influence forest structure and function (Johnstone and Chapin, 2006b). However, Johnson (1984) suggests that smaller gaps created by scattered disturbances are not of long-term significance. A gap of 500 m2 or more will have a significant impact on stand structure (Lorimer, 1989).  Fire affects stand structure in various ways. The space and time variability of fire cycles (Payette et al., 1989; Bergeron, 1991) creates gap dynamics in such a complex way that stand mosaics are formed ranging from post-fire early successional deciduous species to mixed and pure coniferous stands in the older post-fire stands (Bergeron & Dansereau, 1993). Similarly, longer fire intervals increase the chance of a spruce budworm outbreak due to the increased 13  proportion of late successional species such as spruce (Bergeron and Leduc, 1998). The increased mortality due to insect outbreaks alters the structure and composition of stands (Bergeron et al., 1995; Harvey et al., 2002). Hence, heterogeneity increases with increasing stand age (Cumming et al., 2000; Brassard and Chen, 2006). Frequent fires and longer fire intervals, would all have similar impacts on stand structure, namely the elimination of pure deciduous stands and the development of mixed stands (Bergeron and Dansereau, 1993). Climate plays an important role in stand structure and tree growth variability (Lloyd and Fastie, 2002). There are many examples of how climate can affect stand structure variability. Climate directly affects the growth and behavior of individuals, modifies population size and structure, and affects ecosystem structure and function by affecting decomposition rates, nutrient cycling, water flows, and the distribution of ecosystems within landscapes (Gitay et al., 2002). One example of the impact of climate is the growth reduction associated with higher temperatures and drought stress (Barber et al., 2000). Higher temperatures can significantly reduce tree growth (Lloyd and Fastie, 2002). It was found that white spruce growth was negatively correlated when mean July temperature was more than 16 oC (Wilmking et al., 2004)). The mean July temperature in the study area was 13.2 oC which implies that increase in 3 oC temperature as predicted may affect white spurce growth in the study area.  Species composition is also affected by species tolerance to shade and moisture stress. For example, aspen is a light-preferring species that occupies the early stage of succession when enough light is available for its growth, but it is eventually replaced by white spruce or other shade-tolerant species (Bergeron and Dubuc, 1989). Drought plays a direct role in shaping and limiting species distributions (Aber et al., 2001; Hannah et al., 2002) and is a key determinant of the distribution of boreal tree species (Chhin and Wang, 2008). Climate also affects forest structure and forest 14  carbon by influencing fire regimes and the spread of insects and pathogens (Dale et al., 2001; Ma et al., 2012).  1.4.2.3 Impacts on biodiversity Disturbances can have major implications for biodiversity (Nilsson and Ericson, 1997) by creating a mosaic of structures and processes that influence the variety and diversity of species present in a forest (Palik et al., 2002). Studies in the boreal forest suggest that the frequency and severity of disturbances can affect forest ecosystems by influencing vegetation dynamics, diversity patterns and ecosystem processes (Wardle et al., 1997). As most forestry operations such as logging involve some kind of disturbance, affecting both diversity and ecosystem functions, an understanding of the effects of natural disturbances and disturbance dynamics on diversity has become important for forest managers (Haila et al., 1994; Angelstam, 1997; Bengtsson et al., 2000). It is also important to understand the relationships between disturbance dynamics and biodiversity (Hughes et al., 2007). Climate change influences forest biodiversity by affecting species interactions and ecosystem processes (Spittlehouse, 1996). It may change biodiversity, stand structure and species dominance by creating open stands that encourage a dense shrub cover (Hebda, 1998). Similarly, wildlife distributions and populations are expected to be affected due to changes in the quantity and quality of food availability, habitat structures and interactions between predators and prey (Ogden, 2006).  Community dynamics and their relationships with biodiversity are important elements that need to be considered in the study.  Variation in diversity could be both a cause and a consequence of variation in community dynamics (Loreau et al., 2001). It is therefore important 15  to understand how community properties contribute to the maintenance of biodiversity (Huston, 1979; Hughes et al., 2007). Similarly, the amount of course woody debris could also play a significant role in determining the presence or absence of several species, especially in boreal forests where many species are dependent on the density of dying and dead trees (Esseen et al., 1997; Nilsson et al., 2001).  1.5 Research objectives and questions  This research arose from the Strategic Forest Management Plan (2004) for the Champagne and Aishihik First Nations traditional territory of southwest Yukon. It was intended to support the plan?s goals and objectives by providing scientific information as agreed between the University of British Columbia and the Government of Yukon in November, 2007. In this connection, the research was designed and conducted to support one of the goals of the SFMP (2004), namely ?to maintain ecosystem functioning? with the objectives to ?maintain or restore or enhance forest regeneration and succession, species and ecosystem diversity?.  Specifically, the research would contribute to one of the key indicators of the SFMP for monitoring its goal, which is ?to assess the variation in pattern, composition and structure of the ecosystem? (SFMP, 2004). This would require the development of a sound understanding of the existing ecosystem structures and composition in the region. To achieve this, more research had been suggested to obtain a better understanding of the linkages between forest disturbances, ecological complexities/uncertainties and the ecological responses of the forests in the region. The NCE report (2006) also suggested that an understanding of the relationship between climate 16  change and forest disturbances was essential for the better understanding of future ecosystems under future disturbance regimes.  The main objective of this research is to understand the recent responses of forest ecosystems of the CATT to bio-geoclimatic and disturbance variability.  This is needed in order to understand better their variation and to develop possible scenarios of future ecosystem dynamics. Such an understanding would be extremely useful in developing forest management tools for sustainably managing the forests in the region in the context of predicted climate change and associated future uncertainties. Specifically the research;  I. studied the impacts and variation in forest ecosystems in response to bio-geoclimatic and disturbance factors.  II. assessed the implications of the research findings in relation to possible forest management strategies and activities to meet the ecological goals of SFMP (2004).   1.5.1 Research questions The general research questions associated with these objectives were as follows: ? How does stand structure vary with environmental factors, climate and disturbance types? ? How does landscape topography interact with climate variability to affect stand structures and composition?  ? What are the most important variables that explain the distribution and abundance of tree regeneration in the study area? 17  ? What are the overall species richness, evenness and abundance in the study area and how do they vary by landscape and ecological variability? ? What are the forest management implications and strategies in response to the above ecosystem variability in the CATT?   1.6 Research framework and thesis structure The research framework, linking the research issues, objectives and specific questions, is provided in Figure 1.1. The dissertation has seven chapters. The first chapter includes an introduction, literature review, justification of the research, research objectives and research questions. The 2nd chapter contains the research methodology, including the methods of data collection and analysis.  The 3rd, 4th and 5th chapters contain the main research sections. The third chapter deals with the variation in stand structure in response to climate variability, disturbances and geographical factors. The fourth chapter examines the distribution and abundance of regeneration of tree species in response to climate variability, disturbances, geographical variation and stand structure. The fifth chapter examines variation in vegetation diversity in response to bio-geoclimatic and disturbance factors. Chapter 6 is a synthesis of the previous chapters to assess forest management implications and to recommend appropriate management strategies and activities.  A final chapter provides conclusions and recommendations.  18   Figure 1.1 Research framework  [Note: The figure indicates the research background that justified the need for the research. The research scope corresponds to specific research chapters in the thesis. The last section in the figure highlights the research objectives that each chapter deals with] Research background Study scope Specific objectives Impacts on SWY Forest Ecosystems Warming temperature in the Canadian North Disturbances e.g. forest fire, spruce bark beetle, salvage harvesting   Impacts and variation in stand structure (Chapter III) Impacts on tree regeneration   (Chapter IV)   Impacts and variation in vegetation diversity (Chapter V)  Ecosystem variation Environmental factors Variation in average diameter, height and basal area Variation in stand composition and density by environmental gradient Distribution and abundance of regeneration in response to environmental gradient, climate and ecological variability.  Overall species richness and abundance Species richness and abundance by landscape variability  19  2. Methodology 2.1 Study area justification: Why was the research important in the forest of CATT? This research was undertaken in the Champagne and Aishihik Traditional Territory (CATT) within the southwest Yukon (SWY).  The main reasons for carrying out this research in the CATT were;  ? Issues of climate change: Model projections suggested that the increase in winter temperature in the region will be from 2 ?C by 2040 to 7 ?C by 2100 (ACIA, 2004). As the variation in mean annual temperature in the study area is 1.2 oC (Annex B-2), any variation in ecosystem attributes within 1.2 oC would suggest that there would be significant variation of ecosystems in the projected climate scenarios. ? Multiple forest disturbances: The forests of the CATT have been affected by multiple disturbances such as spruce bark beetle, fire and salvage harvesting.  The region is suffering from the largest and most intense infestation of spruce bark beetle ever recorded in Canada, and with an increase in fire frequency in recent decades (SFMP, 2004). In the absence of active fire management, the average annual fire frequency and total area burned in the Yukon Territory is projected to double by the 2070s (McCoy and Burn, 2005). Although salvage harvesting has been initiated as a short-term management solution, the possible consequences of the salvage harvesting on the forest ecosystem have not been assessed yet.  ? Value of forests to the First Nations: The forest has great socio-economic and cultural values to the Champagne and Aishihik First Nations who officially acquired their traditional territory in 1995 after signing land claim agreement. However, the increasing 20  forest disturbances have threatened the sustainability of those values derived from the forests.  ? Forest management issues and initiatives: To manage forest disturbances and to achieve long-term forest sustainability, a forest management initiative was started with the approval of the SFMP in 2004. A detailed ecosystem assessment in response to bio-geoclimatic and disturbance factors was suggested by SFMP (2004), NCE (2006) and Ogden (2007) in order to support forest management decisions in the region. Accordingly, this research was conducted as a part of the research agreement made between the University of British Columbia and the Government of Yukon in November, 2007.  2.2 Study area description The CATT extends across four eco-regions of Yukon: the St. Elias Mountains, the Ruby Range, the Yukon Southern Lakes and the Yukon Stikine Plateau; these characterize the landscape diversity and variability (SFMP, 2004). The study area occupies a latitudinal range between 60.63 and 61.20 ?N and a longitudinal range between -137.35 and -136.97 ?W (Figure 2.1). The climate of the study area is characterized by cold winters and warm/dry summers, with mean annual temperature -1.65  C and annual precipitation of 319.5 mm. The elevation range of the study area is 651 to 1158 meters above Mean Sea Level (MSL). The soil texture varies from mixtures of sandy and silty fluvial deposits beside rivers and lakes in lowland areas to mixtures of sandy and silty clays in the remaining area. The study area is dominated by white spurce stands (about 87%), with majority of them age class between 80 to 120 years (SFMP, 2004).  21  Approximately 380,000 hectares of forest were affected by the recent beetle outbreak (SFMP, 2004). Most of the beetle-affected forests that were not harvested or affected by fire had the highest stand densities (1400 stem/ha) and 63.93% average crown cover (Table 2.1).  Evidence of past fire is common in the study area. Between 1946 and 2003, the region experienced 23 large fires and numerous small fires (SFMP, 2004). Most of the areas in the region burned in the 1950s. Past fires were identified by the presence of burnt logs. These old fire sites mostly had mixed stands of white spruce and trembling aspen, with an average of 48% crown cover and 8.21 cm soil organic depth (Table 2.1), and had no evidence of spruce bark beetle activity. The most recent large fire occurred in 1998 about 15 kilometres (KM) east of Haines Junction along the Alaska Highway (research block D, Figure 2.1), severely affecting both the forest crown and the soil. Prior to the fire, these forests had been affected by spruce bark beetle. Plots2 in the recent fire indicated low average crown cover (12%), organic depth (2.75 cm) and soil moisture (19%), indicating a higher fire severity (Table 2.1).  Salvage harvesting of beetle-affected trees started in 2006 in research blocks 5 and 6 (Figure 2.1). Dead trees were removed, with severe impacts to both the soil and understory vegetation due to the operation of heavy equipment. The harvested plots also had lower average organic depth (5.19 cm) and soil moisture (18.59%). Based on the types and intensity of disturbances, the disturbed plots could broadly be categorized as (i) high severity: plots with a combination of either spruce bark beetle and fire (fire plots) or spruce bark beetle and salvage harvesting (harvested plots), (ii) medium severity:  plots which had more than 50% spruce bark beetle infestation without fire and salvage harvest (beetle plots), and (iii) low severity plots: plots with traces of old fires and  unaffected by spruce bark beetle (old fire plots).                                                   2 Explanation on plots and blocks are provided in the next section. 22  2.3 Data collection 2.3.1 Sampling methods  The selection of the most appropriate sampling method depends upon various factors, as there is no ideal technique for vegetation monitoring (Korb et al., 2003). The selected technique should have an acceptable level of precision, and be replicable, feasible, realistic and inexpensive (Sutter, 1996). Stratified systematic sampling (Kindt and Coe, 2005) was applied for the collection of data in this research. Stratified sampling has been suggested for landscape-level vegetation assessment when the vegetation has heterogeneous characteristics (Kindt and Coe, 2005; Cooper et al., 2006; Rolecek et al., 2007). The stratified sampling technique captures the representative heterogeneity of landscape (Cooper et al., 2006), captures the major sources of variability and is useful for gradient analysis and comparing the results (Kindt and Coe, 2005). Similarly, a systematic lay-out of the plots has a comparative advantage over random sampling as it is easier, saves time, avoids the chance of too short or too great distances between the plots and is useful for assessing environmental gradients (Kindt and Coe, 2005). Data were collected from 90 sample plots from six blocks (with five distinct strata types) in the forest landscape of the CATT (Figure 2.1, Table 2.2). The plots were classified into five strata based on the disturbance type i.e. (I) undisturbed forest (control), (II) an old fire area (>50 years old), (III) a recent fire area (<20 years old), (IV) a spruce bark beetle affected area, and (V) a salvage-harvested area. The plots were identified and mapped using the local knowledge of the Yukon Government?s Ministry of Energy, Mines and Resources (MEMR) and Champagne and Aishihik First Nations (CAFN) government in the summer of 2008.  23  Vegetation, soil and fuel data were collected in a nested system based on 20 m x 20 m sample plots. The number of sample plots in each stratum depended on the size of the respective research area. A summary of the plot design is given below (Figure 2.2). I. The landscape was divided into six blocks (with five strata as shown in table 2.2) based on discussion with YG and the CAFN government. II. In each block, a point of commencement (POC) was chosen, which was generally a landmark located along the road. III. A random bearing was selected using random numbers. The first plot was selected 100 meters (m) from the POC towards the direction generated by the random bearing selection. The subsequent plots were selected accordingly 100 m apart systematically following the dominant topographic gradient. IV.  A plot centre and four corners of the plots (20 m x 20 m) were marked with the help of compass and tape. GPS points and other various geographical parameters such as altitude, slope, aspect, vegetation types, and vegetation conditions were recorded. V. With each 400 m2 plot, three 5 m x 5 m (25 m2) nested subplots were randomly selected and used to collect seedling, sapling, and shrub data.  VI. Regeneration density and herb/shrub/bryophyte cover were collected from three randomly selected 1 m x 1 m subplots located within the 25 m2 subplots.  VII. One soil pit was dug in a randomly selected location within the 400 m2 plot to determine soil characteristics.   VIII. There were a total of 90 plots, 270 subplots and 270 mini plots. The locations of the plots were recorded by GPS for future monitoring. 24  IX. Data collected from the 20m x 20m, 5m x 5m and 1m x 1m plots are given in Table 2.3.  2.3.2 Variables The data were classified broadly into topographic, stand structure, edaphic, climatic and disturbance regime categories (Table 2.4). Basal area, slope, elevation, organic depth and average moisture were converted into categorical variables for the purpose of the analysis (Table 2.5).  2.3.2.1 Topographic variables Topographic variables represent macro-scale factors and for this study consist of latitude, longitude, elevation, slope and aspect. Aspect was transformed into Cosine of aspect (COS? = northness) and Sine of aspect (SIN? = eastness). The northness value ranges from 1 to -1 (1 = northward, -1 = southward, 0 = either eastward or westward). Similarly a SIN? value close to 1 represents east-facing slopes (1 = eastward, -1 = westward, 0 = either northward or southward) (Robert, 1986).   2.3.2.2 Stand structure variables  At each plot, tree3 density (number of trees per ha), diameter at breast height (DBH) and height                                                  3Tree: Vascular plant more than 1.3 meter height and 7 cm diameter.  25  (Ht) of trees (based on height dominance classes), crown cover (%), shrub4 cover (%), herb5 cover (%), seedling6 density (number per ha) and sapling7 density (number per ha) were recorded (Table 2.3), which represent the meso-scale factors. The definitions for tree, seedling, sapling, herb, and shrub followed the Yukon Monitoring Guideline (Ogden, 2008). Total and species-specific basal area (BA) per hectare were calculated. Average BA, DBH and Ht were calculated for each block to assess block effects. Tree and regeneration density were converted into a per hectare basis to standardize the data from plots and sub plots.   2.3.2.3 Ecological variables Vegetation data were collected as percent cover in 270 nested mini plots (1 m x 1 m). The vegetation included all bryophytes, lichens, herbs, shrubs and tree regeneration by species within the mini plots. Due to the different height layers of the vegetation, in many cases the total percent cover for all vegetation was more than 100%. Species were identified using local flora (Cody, 2000; Johnson et al., 1995). Photographs were taken and specimens were collected to verify the species.  The vernacular and scientific names of the species and their code names are provided in Appendix A. The edaphic variables of soil depth, soil moisture and soil texture were used as micro-scale variables. Soil moisture was collected from different layers of the soil using a Vernier soil moisture sensor (accuracy +/- 4%), which measures the dielectric permittivity of soil to                                                  4Shrub: Woody plants higher than 15 cm height 5Herb: Any herbaceous species (irrespective of height) and any plant under 15 cm height 6Seedling: Tree species less than 1.3 meter tall 7Sapling: Tree species with more than 1.3 meter tall but less than 7 cm diameter 26  determine the volumetric water content. Soil moisture was taken in single snapshot within two months perod during field study.  Multiple soil moisture readings were taken from each soil pit, and these were averaged for the purpose of the statistical analysis. Soil texture was segregated into percentage of sand and clay based on the classification of each sampled soil layer, and converted into soil texture classes in the field using a standard soil texture diagram.    2.3.2.4 Climatic variables  Climatic variables were used to represent large-scale factors. The ClimateBC  model (Wang et al., 2006) was used to project mean annual temperature, mean annual precipitation, mean annual maximum temperature, mean annual minimum temperature, number of frost free days, number of days below 0 ?C and number of days above 5 ?C for each plot (Table 2.4). Climate normals were based on the years 1970 to 2000. The ClimateBC model uses longitude, latitude and elevation to project climate variables. The standard error of predictions for ClimateBC monthly temperatures and precipitation varies from 0.8?1.3 ?C for temperature and 8 to 24 mm for precipitation (Wang et al., 2006). The variations in mean annual temperature and precipitation of the study area were 1.2 oC (range = from -2.4 oC to ? 1.1 oC) and 80 mm (range = 290 mm to 370 mm), respectively.    2.3.2.5 Disturbance regime  Disturbance was considered as an explanatory variable and represented large-scale variation. The four types of disturbances considered in this study were old fire (> 50 years), new fire (<20 years), spruce bark beetle and salvage harvest.  27  2.4 Data analysis Various simple and multivariate statistics were used to achieve the objectives of the research. Descriptive statistical analyses were undertaken for all the dependent and independent variables; these are presented in Appendix B. They helped to visualize the nature of data, their distribution and central tendency. The necessary data conversions were undertaken as per the requirement of each research question. This section provides only the names of the tools used in each research chapter. The details of each analytical tool are provided in the respective research chapters.   2.4.1 Stand structure analysis (Chapter 3)  Correlation coefficients, analysis of variance (ANOVA), proportionate ratio of regeneration versus parent trees, basal area ratio, density index, Shannon-Wiener structural diversity index,  multiple co-inertia analysis, geographical information system (GIS) are used in chapter 3 to assess the stand structure variation in the landscape.   2.4.2 Regeneration analysis (Chapter 4)  Multivariate canonical correlation analysis, negative binomial generalized model and the most nearest neighbour imputation model are used in chapter 4 to analyse the distribution and abundance of regeneration in the study area.     28  2.4.3 Vegetation diversity analysis (Chapter 5)  Simpson?s index, Shannon-Wiener index, richness, abundance and evenness analysis, Renyi?s profile, species spatial distribution and range calculation, canonical corresspondence analysis (CCA), species accumulation curves and geographical information system (GIS) are used in chapter 5 to describe and analyse the vegetation diversity variation in the study area.     29  Table 2.1 Characteristics of disturbed plots  Disturbance Type Soil  moisture (%) p < 0.05 Crown cover (%) p < .001 Stand Density (tree/ha) p < .001 Organic  depth (cm) p < 0.05 Shrub cover p < .001 Spruce bark beetle plots 25.96  63.93 1400 11.29  40.71 Fire plots 19.82 11.98 247 2.75 36.23 Harvested plots 18.59   29.00 591 5.19 40.08 Non-disturbed 19.42 44.48 789 8.98 61.11 Old fire plots 25.14 48.21 1039 8.21 55.59 Mean values 21.78 39.52 813.2 7.24 46.74  [Note: The above table shows the variation of soil moisture, organic depth, crown cover, stand density and shrub cover by disturbance types. The mean variation of above variables were significant at p <0.05]   Table 2.2 Research blocks and their characteristics  Blocks No of plots Elevation range (m) Slope range (%) Average white spruce age (years) Major disturbances  Major species A 13 933-967 3-24 60 Mostly undisturbed with the traces of old fires White spruce (dominant), aspen and willow B* 20 887-1158 3-65 102.8 Traces of old fire and beetle White spruce (dominant) mixed with willow and aspen C 10 754-786 4-20 121.8 Old fire White spruce dominant with the scattered patches of aspen D* 25 651-693 0-28 123 Recent fire and beetle White spruce dominant with many patches of willow E 13 778-791 0-13 93.13 Salvage harvested White spruce dominant  F  9  773-781  0-22   118.5  Salvage harvested  White spruce and aspen mixed forest  [Note: The table provides the details of each research blocks including number of plots in each blocks, elevation range, slope range, the major disturbances observed and major tree species. The number of plots depends upon the size of the each research block. The block D had the highest number of plots and F had the lowest number. *Blocks B and D are divided into two sub-blocks (B, B1, D and D1) for the purpose of vegetation diversity study.]    30  Table 2.3 Plot sizes, types of data collected and brief methodology  Variable plot sizes Data (variables) Indicators Methods 20 m x 20 m Tree counts By species By dominant class By dead or live Total enumeration   20 m x 20 m Crown cover Total crown cover Crown cover by species Ocular estimation. An average was considered after three field crewmembers estimated the cover  20 m x 20 m Tree measurements Tree height Tree diameter at breast height Height and diameter of sampled trees of each dominant class were measured 20 m x 20 m Tree age Growth ring Using tree ring borer 5 m x 5 m Tree counts By species By dominance class By  status (dead or alive) Total enumeration within the sub plot 5 m x 5 m Sapling counts By species By dominance class By status (dead or alive) Total enumeration 1m x 1m Shrubs estimation By species By percentage of area cover in the subplot Observation and ocular estimation 1 m x 1 m (mid-point) Soil profile Depth and texture of organic and mineral horizons (Ogden, 2008) Rooting depth Total depth Soil moisture Soil temperature  A soil pit was dug to the depth of the hard pan, and profiling the horizons using key. A moisture meter was used to measure soil moisture and temperature 1m x 1m  Regeneration count By species  Total regeneration count for each species 1m x 1m Herbs estimation By species By % of area cover in the mini plots Ocular estimation with proportionate area covered by each species    31  Table 2.4 Variables (with full and code names) collected and generated for analysis Geographical variables and disturbance regime Ecological Variables Stand structure variables Climate variables (1971-2000 mean) generated from BC climate model Aspect  {cosasp, sinasp}  Slope (slope) Latitude (lat) Longitude (long) Elevation (ele)   Disturbances Old Fire (OLDFR) New Fire (FR) Spruce bark beetle (BT)  Salvage harvest (SALV)  No disturbances  Density of white spruce (STT) Density of trembling aspen (ATT) Density of balsam poplar (PTT) Number of white spruce seedlings (SpsedTot) Number of trembling aspen seedlings (AtsedTot) Number of balsam poplar seedlings (PosedTot) Average soil moisture (AM) % of clay (CL) % of sand (SA) % of shrub (shrubcov) % of herb (herbcov) % of bryophytes % tree regeneration Total crown cover (TCC) Total basal area (ToBA) White spruce basal area (SpBA) Trembling aspen basal area (AtBA) Balsam poplar basal area (PoBA) White spruce average height (Spavht) Trembling aspen average height (Atavht) Balsam poplar average height (Poavht) White spruce average diameter (at breast height)  (Spavdia) Trembling aspen average diameter (at breast height) (Atavdia) Balsam poplar average diameter (at breast height) (Poavdia) Mean annual temperature (MAT) Mean annual precipitation (MAP) Maximum summer temperature (TMXsm) Minimum  summer temperature (TMNsm) Maximum winter temperature (TMXwt) Minimum winter temperature (TMNwt) Number of frost free days (NFFD) Number of days above 50C (NDD>5) Number of days below 00C (NDD<0)      32  Table 2.5 Categorized variables  Variables Very High (E) High (D) Medium (C) Low (B) Very Low (A) Trembling aspen basal area (m2/ha) > 8 6.1 to 8 4.1 to 6 2.1 to 4 < 2 Balsam poplar basal area (m2/ha) 1.6 to 2 1.2 to 1.6 0.8 to 1.2 0.4 to 0.8 <.4 White spruce basal area (m2/ha) > 40 30.1 to 40 20.1 to 30 10 to 20 <10 Elevation (m) >1050 951 to 1050 851 to 950 751 to 850 650 to 750 Slope (%) - >30 15.1 to 30 1 to 15 0 Soil organic depth (cm) >24 18.1 to 24 12.1 to 18 6.1 to 12 0 to 6 Average moisture (%) >40 >30 to 40 >20 to 30 >10 to 20 0 to 10  [Note: The variables were categorized for the purpose of ANOVA test and plotting the mean variation.]   33    Figure 2.1 Map of the study area showing blocks and plots where the data were collected.   [Note: Black A is mostly undisturbed forest predominantly consists of white spruce with small patches of Aspen and Willow; Block B has traces of old fires in 50% of its plots and consists of white spruce dominancy with scattered patches of Willow bushes;  Blocak C has old burnt logs in the most of the plots and consists of White spruce and Aspen; Block D has been affected by spruce bark beetle and was gone through recent fire and has White spruce with scattered patches of Willow; Block E has beetle and harvesting and consists mainly White spruce trees; Block F is a harvested plot and has mixed of White spruce and Aspen trees.]   34      Figure 2.2 Plot design  [Note: The square plot with dimension of 20 m length and 20 m breadth was designed for the purpose of data collection.  The 20 m by 20 m plot was divided into 16 sub-plots of 5 m x 5 m. Three sub-plots were further divided into 25 micro plots, each 1 m by 1 m in size for the purpose of micro site assessment including soil texture analysis, regeneration and herbs count]      20m x 20m plot    (total 90 plots) 5m x 5m plots    (total 270 plots) 1m x 1m plots    (total 270 plots) 35  3. Stand structure variability in response to bio-geoclimatic and disturbance factors in southwest Yukon   3.1 Introduction Stand structure has important implications for forest ecosystem health, biodiversity conservation and forest management (Sturtevant et al., 1997; Kimmins, 1997; Kuuluvainen, 2002; Brassard and Chen, 2006). Although, forest composition and structure are generally affected by bio-geoclimatic factors and the temporal and spatial scales of disturbances (Bonan and Shugart, 1989), the resulting ecosystems may vary depending upon the extent and magnitude of disturbances and geo-climatic variations. Forest managers need to understand disturbance dynamics and their effects on ecosystem structure so that better management practices can be used, yet such an understanding is still lacking for many forest types in the boreal forest including in SWY (Martin and Gower, 2006). This chapter deals with the variation in stand structure and composition in response to bio-geoclimatic factors across multiple disturbance regimes in southwest Yukon.  Stand structure is defined as the ?physical and temporal distribution of trees within a stand? (Oliver and Larson, 1996), including species dynamics such as spatial pattern, horizontal and vertical distribution, and size and age of trees (Stone and Porter, 1998). The change in stand structure over time refers to stand dynamics (Oliver and Larson, 1996). Climate, disturbance, permafrost, soil moisture, soil temperature, oceanic influence and complex interactions among physiographic patterns determine structure and composition in the boreal forest (Bonan and Shugart, 1989; Larsen, 1980). Bonan and Shugart (1989), Lawrence and Oechel (1983) and 36  Tryon and Chapin (1983) consider soil moisture and temperature as important factors for soil nutrition and tree growth as they can slow decomposition and lead to the high accumulation of organic matter. In Alaska and the Yukon, climatic and site factors influence the forest composition and future stand development by affecting post-fire recruitment (Johnstone et al., 2004). Stand heterogeneity typically increases with increasing stand age indicating that the stand structure and composition are a result of gaps caused by disturbances (Cumming et al., 2000; Brassard and Chen, 2006). The magnitude and intensity of spatial and temporal disturbances shape landscape structural development by creating periodic gaps (Lorimer, 1989) and by developing a mosaic of development patches of varying age and size structure (McCarthy and Weetman, 2006). The variations in disturbance characteristics such as frequency, size and severity strongly influence ecosystem properties and processes (Johnstone and Chapin, 2006b). In the boreal forest, fire is considered to be the most important disturbance agent (Amiro et al., 2001). However, other disturbances such as insect and disease outbreaks, wind throw and salvage harvest are also important in shaping forest structure and composition (Blais, 1981; Bergeron et al., 1995).  Fire may affect stand structure in various ways. The spatial and temporal variability of fire cycles (Payette et al., 1989; Bergeron, 1991) creates a mosaic of stand compositions ranging from post-fire early successional deciduous species to mixed and pure coniferous stands in older post-fire stands (Bergeron & Dansereau, 1993). In eastern Canada, longer fire intervals increase the chance of a spruce budworm (Choristoneura fumiferana Clemens) outbreak due to increased proportion of late successional species such as spruce and fir (Bergeron and Leduc, 1998). Stand mortality due to insect outbreaks can alter stand structure and composition (Bergeron et al., 37  1995; Harvey et al., 2002). The absence of disturbance can also affect stand structure through the processes of competition and succession; these gradually change pure deciduous stands to mixed stands and then to coniferous stands in the boreal forest (Bergeron & Dansereau, 1993). As stand composition changes from pure to mixed stands, the forest productivity may also change significantly (Linder, 1998 & Brassard and Chen, 2006).  Climate can play an important role in stand structure and tree growth (Lloyd and Fastie, 2002). Climate directly affects the growth and behavior of individuals, modifies population size and structure, affects ecosystem structure and function by affecting decomposition rates, nutrient cycle, water flows, and distributing ecosystems within landscapes (Gitay et al., 2002). An example of a climate impact on growth is the observed growth reduction of trees associated with higher temperatures and drought stress as a result of recent climate change (Barber et al., 2000). Higher temperatures can significantly reduce tree growth at both altitudinal and latitudinal limits (Lloyd and Fastie, 2002). Species composition is also highly affected by species tolerance to shade and moisture stress. For example, as a light-requiring species, trembling aspen typically occupies the early stages of succession when light is sufficient for their growth; over time aspen is replaced by white spruce and/or other shade-tolerant species resulting in changes in stand structure and composition (Bergeron and Dubuc, 1989). Drought plays a key role in limiting species distributions (Aber et al., 2001; Hannah et al., 2002). Climate also affects forest structure indirectly by influencing fire regimes and the spread of insects and pests (Dale et al., 2001). Topographic factors are also important in shaping vegetation structure and composition as they affect the availability of solar radiation, soil temperature, soil moisture and nutrient cycles (Van Cleve et al., 1983). Viereck et al. (1983) and Van Cleve et al. (1991) state that 38  topography together with soil parent materials and disturbance regimes contribute to tree species composition. Similarly, site conditions also determine the species composition and structure.  Site variability can also influence stand structure (Brassad and Chen, 2006). Soil, water and nutrient availability influence tree growth in boreal and temperate forests (Stoeckeler, 1960; Lambers et al., 1998). Harper et al. (2005) found structural differences according to site type and moisture regime in the western boreal forest of Qu?bec. Environmental variables such as slope angle, elevation, organic matter depth, and humus type affect site productivity, which in turn influences the diameter and height of stands (Boucher et al., 2006). Site type also influences stand composition, which in turn can determine the density and basal area of a stand (Chen and Popadiouk, 2002; Popadiouk et al., 2003).  3.2 Research problems, hypotheses and questions The literature indicates that stand structure of boreal forests is affected by disturbances, climatic factors and site variability; however, as the boreal forest is a complex mixture of uneven-aged stands (Martin and Gower, 2006), its structure and composition can vary significantly from region to region depending upon specific disturbance regimes, site characteristics and climatic variability. As an increase in fire frequency is predicted to occur due to climate change (Bergeron and Flannigan, 1995; Flannigan et al., 1998), an understanding of  how stand structure varies according to disturbance and climatic factors has been suggested as being important for improving management practices (Brassard and Chen, 2006). Specifically, it is important to understand which factors play a dominant role in shaping stand structure and how they affect the overall forest productivity.  According to Martin and Gower (2006), such an understanding is still lacking for many forest types within the boreal forest including that of Southwest Yukon 39  (SWY).  The main purpose of this chapter is therefore to assess how the structure and composition of the boreal forests of the CATT varies by bio-geoclimatic factors and disturbance regime. Specifically, this chapter will deal with the following hypotheses;   (i) I hypothesized that relationships among stand structure variables (diameter, density in stems per hectare, and basal area) vary significantly by stand type and are affected in my study area by edaphic factors. Specifically, I expect that mean stand diameter will decrease with increasing stand density, and this will be more pronounced in pure stand types and on sites with more fertile soils. I also expect higher stand productivity will occur in pure stands compared to mixed stands. (ii) I hypothesized that forest composition, structural diversity, and forest productivity will vary significantly by disturbance type in the study area. Specifically, I expect that the density of white spruce would increase with any disturbance but that the proportion of broadleaved trees and regeneration would increase with disturbance severity, particularly following fire.  (iii) Climatic factors are expected to have significant impacts on current and future stand structure and composition. I hypothesized that stand structure and composition will vary significantly within climatic variability.  I expect to observe higher mean diameter, height and stand productivity in warmer and wetter locations. In addition, the research addressed the following research questions, which may be important to support the forest management decision making in the CATT.  ? What are the overall stand structures and composition in the study area? How are the structural variables related? ? To what extend do stand structure variables such as diameter, height and density and their relationships significantly vary along environmental gradients?   40  3.3 Data analysis The required data were collected from Southwest Yukon and are described in Chapter 2. This section deals only with the data analysis specific to this chapter.   3.3.1 Calculation of average diameter and height Average height and diameter for each species were calculated. As DBH and Ht were measured for each dominance class, the average DBH and Ht values were normalized as follows; Normalized average DBH = (DBH of Dominant tree *number of dominant tree + DBH of Co-dominant tree * number of co-dominant tree + DBH of intermediate tree x*number of intermediate tree + DBH of suppressed tree* number of suppressed tree)/total number of tree. A similar formula was used for the normalization of height. Based on average diameter, average basal area was calculated using the formula          where,        and r = radius of the tree at breast height.   3.3.2 Calculation of stand structure and composition  Stand structure and composition were assessed using three different indices. The proportions (%) of the density (number of stems/ha) of tree, seedling, and sapling of each species with respect to the total density for that species were calculated. Similarly, average height and diameter of each species by dominance class were calculated to assess the species composition in the study area. Species composition was assessed using basal area ratio of a species with the total basal area (Huang and Titus, 1995), which can be calculated as ?species composition of specific species = 41  basal area of the species/total basal area?. The values range from 0 to 1 where a value of 0 indicates the absence of the species and 1 indicates a fully stocked species.   3.3.3 Density index A density index was developed to compare the density of trees, saplings and seedlings of white spruce, trembling aspen and balsam poplar by disturbances regime with respect to average density on all 90 plots. To aid comparison, the average density of 90 plots was converted to 100% ?as an index?. The percentages of density of trees, saplings and seedlings with respect to total density by disturbance regimes were calculated as follows;                ..................................Equation 3-1 where, Di is the density index, Dd is the average density of  tree, saplings or seedlings in a particular disturbance regime and Dt is the average density of the trees, saplings or seedlings in the 90 plots.  3.3.4 Calculation of variation of structural diversity index across disturbance regime The Shannon-Wiener index (Shannon and Weaver, 1949) has been frequently used to assess structural diversity. The Shannon-Wiener index has also been applied to species size diversity by grouping the DBH and Ht values into classes (Staudhammer and LeMay, 2001; Lexerod and Eid, 2006). The diameter and height diversities of white spruce and aspen were calculated across disturbance regime by using the following formula (3.2):                  ) ....................Equation 3-2   42  where, H? is the Shannon-Wiener Diversity Index and Di, is the proportion of each diameter class (N). The value ranges from 0 to Ln8 (N) where N equals the number of diameter classes (Lexerod and Eid, 2006). The Shannon-Wiener diversity index is useful for assessing stand diameter/height structure when classifying forest structural types (Boucher et al., 2006). The index has previously been used in coniferous forests (Sarkkola et al., 2003).  3.3.5 Calculation of correlation coefficients and scatter plots  R2 values were calculated for the variables related to stand structure. Scatter plots with fitted regression lines were constructed when both dependent and independent variables were numerical and plots of means were calculated for the nominal independent variables. The significance of the plots was tested using regression goodness-of-fit tests for scatter plots and ANOVA for the means of plots. R statistical software, version 2.12.0 (R Core Development Team, 2005) and MS Excel, version 2007 (Microsoft, 2007), were used to calculate ANOVA, R2 values and to plot the relationship among variables.  3.3.6 Multiple co-inertia analysis (MCoA) In addition to the above statistical tools, MCoA was used to assess the relationship between environmental variables and stand structure. MCoA has been recommended for the analysis of the response of community composition to environmental conditions (Dole?dec and Chessel,                                                  8 Natural Log 43  1994). It is a multivariate tool that analyzes the common structures of pairs of data tables and is more flexible than some other multivariate techniques, as it can utilize both qualitative and quantitative variables (Dray et al., 2003). It is a symmetric coupling method, which uses the co-inertia criterion on a set of orthogonal vectors (Chevenet et al., 1994). It aims to find vectors in species and environmental spaces with maximal co-inertia (Dray et al., 2003) and works on the co-variance matrix (Chevenet et al., 1994). Similarly, it is linked to partial least-squares regression (Dray et al., 2003).  MCoA is able to carry out the simultaneous ordination of several tables (Bady et al., 2004) and can simultaneously analyse the spatial and temporal variations in data (Dray et al., 2003). It optimises the variance within each individual and optimises the covariance between several individual ordinations (Bady et al., 2004). The method uses a pair of tables; if the number of tables is equal to one then the method corresponds to Principal Component Analysis (PCA) (Bady et al., 2004).   3.3.7 Geographical information system (GIS) The GIS database developed by the Yukon Ministry of Energy, Mines and Resources (MEMR) was used to illustrate the species distribution, crown cover, forest types and average height in the study area. A shape file of the study area landscape was used as a mask to clip the database from the MEMR GIS database using ArcGIS software (ESRI, 2009).   44  3.4 Results 3.4.1 Stand composition In this study, 2944 trees were sampled from 90 plots, each 400 m2 in area (the total area sampled was 3.6 hectares). As expected, white spruce dominates the landscape with 87.5% stem cover followed by aspen and balsam poplar, which cover 12.2% and 0.3% respectively (Table 3.1). Mixed stands had higher stand density (1063 stems/ha) than pure aspen stands (300 stems/ha) or pure white spruce stands (848 stems/ha). Similarly, white spruce had the highest sapling cover with 67.8%, followed by aspen (31.4%) and balsam poplar (2.6%) (Table 3.1). Stand composition index analysis (Huang and Titus, 1995) indicated that white spruce has the highest index (0.907), indicating the highest stocks on the landscape, followed by aspen (0.09) and balsam poplar (.001) (Table 3.2). The species distribution is shown in Figure 3.1.  3.4.2 Average height, diameter and basal area of species by height class An average height, diameter and basal area of white spruce and trembling aspen are provided in Table 3.3. The average stand height is plotted in Figure 3.2. The results indicated that the white spruce intermediate tree class was the most frequent on the landscape with 24.3% of the total cover. Its co-dominants, however, had the highest basal area, with 42.64% of the total in the landscape followed by white spruce co-dominant (20.6%) and dominant trees (18.3%). Similarly, trembling aspen had the highest number of co-dominant trees, covering 7.5% of the total cover, followed by intermediate trees (6.7%). Mixed aspen and white spruce stands had higher basal areas per plot (= 13.65 m2/400 m2) compared to pure white spruce stands (= 13.27 m2/400 m2).   45  3.4.3 Shannon-Weiner structural diversity Index across disturbance regime Shannon-Weiner structural diversity indices for white spruce and trembling aspen across disturbance regime (Table 3.4) indicated that white spruce heights and diameters were more diverse in undisturbed plots. Similarly, trembling aspen heights and diameters were more diverse in undisturbed and old fire plots respectively. The results indicated that structural diversities were higher in undisturbed plots.   3.4.4 Analysis of stand structure variations using regression fit line (scatter plots) and ANOVA plots (plots of means) 3.4.4.1 Variation in stand diameters and heights by geo-climatic factors The scatter plots with fitted regression lines (Figure 3.3 and Figure 3.4) showed an increasing trend of average diameter and height of white spruce with decreasing slope percentage and elevation, which indicated that overall white spruce productivity decreases with increasing elevation and slope. The average spruce height also increased with increasing soil moisture on northeast aspects (Figure 3.4). The variation in white spruce height and diameter by climatic factors was not significant. Similarly, the average white spruce diameter was positively correlated with average height (r2 = 0.89, p<0.05) and tree density (r2 = 0.23, p<0.05); however, average diameter decreased when the average density exceeded 500 trees per hectare (Figure 3.5). Figures 3.3 and 3.4 indicate that white spruce average heights and aspen average diameters were significantly higher at lower elevation, in plots with loamy sandy soil and on 46  eastern aspects. White spruce average diameter was positively correlated with stand density (r2 = 0.2, p<.05) and basal area (r2 = 0.3, p<.05) (Figure 3.5; Tables 3.5 and 3.6). The relationships between the average diameter and height of balsam poplar and the various environmental variables were not significant.    3.4.4.2 Variation in stand density and basal area by geo-climatic factors The average stem density and basal area of white spruce varied significantly by climatic, topographical and edaphic factors in the study area (Figures 3.6 and 3.7). Average densities were higher on sites occupying lower slopes, with higher mean annual temperature, with higher moisture levels, and with greater organic depth. As with diameter, white spruce average basal area (BA) decreased with increasing elevation and slope but increased with increasing number of growing degree days > 5 oC (Figure 3.6). A higher BA was found on northeastern and southern aspects. Productivity graphs (Figure 3.5) and r2 values (Table 3.6) indicate that average white spruce basal area was correlated with tree density and diameter, with r2 values of 0.73 (p<.01), and 0.43 (p<.05), respectively. White spruce age did not show a strong correlation with diameter (r2 = .04, P>.05) or height (r2 = 0.1, P>.05). The relationships between stem densities and basal areas of trembling aspen and balsam poplar and climatic and edaphic variables were not significant. However, plots of means (Figure 3.9) indicated that they were higher on NE and SW aspects and on sandy to loamy sandy soils. Similarly, the plots of means (Figure 3.8) indicated that balsam poplar is typically found on either southern aspects, sandy soils or a combination of the two. Productivity graphs (Figure 3.5) and r2 values (Table 3.6) indicated that aspen basal area was loosely correlated with its diameter 47  (r2 = 0.3, p<.05) and height (r2 = 0.25, p<.05); but was strongly related with its stand density (r2 = 0.76, p<.01).   3.4.5 Impacts of forest disturbances on stand structures  The density index results (Table 3.7) revealed that the density of white spruce tree was higher than average (mean index value for 90 plots = 100) in beetle-affected plots (196) and old fire plots (120) and much lower in the recent fire plots (32). However, the number of white spruce seedlings was higher in fire plots (205), followed by beetle plots (157). Similarly, the density of aspen trees was higher than average in the old fire (197) and undisturbed plots (137) and less than average in recent fire plots (20). Aspen saplings and seedlings had higher than average densities in the harvested and recent fire plots. The densities of balsam poplar saplings and seedlings were higher than average in harvested plots (257 and 418, respectively), and its seedlings had higher than average densities in the fire plots (370). The ratios of tree regeneration to parent trees across the disturbance regime showed an interesting trend (Table 3.8). With the exception of the fire and harvest plots, the ratios were less than one, indicating fewer seedlings and saplings than parent trees. However, with aspen, the ratio was higher than one in all the disturbance regimes, indicating greater numbers of regenerating trees than mature trees. The ratio was very high in fire plots (1:6 for white spruce and 1:45 for aspen). Although, the number of balsam poplar trees was negligible in the plots, regeneration (consisting of a few individuals) was found in both the fire and harvested sites. Similarly, Figure 3.9 provides the plots of means that show the relationship between the types of disturbances and stand structure variables. The plots that were significant with 95% 48  confident level (p<.05) are discussed. The plots indicated that the density and basal area of white spruce were higher in beetle-affected plots than other plots. Similarly, both density and basal area of aspen were significantly higher in old fire plots. Density and basal area of balsam poplar were not significantly correlated with disturbance type.   3.4.6 Multivariate analysis of stand structure variables using multiple co-inertia analysis (MCoA) MCoA was used to analyse the relationship between structural variables and geo-climatic factors. The output tables are provided in Appendix C. The output graphs provided a clear indication of the relationship between stand structure and environmental variables. In Figure 3.10, the average height of white spruce clearly increases9 with increasing summer maximum mean temperature, eastern aspect (Sin aspect) and decreases with increasing slope, elevation and summer precipitation. Similarly, trembling aspen average diameter increases with increasing degree-days > 5 ?C but decreases with increasing depth of organic soil, elevation and slope. Balsam poplar diameter and height increase with increasing mean annual temperature, number of frost-free days and minimum average summer temperature.   The species/environment axis (Figure 3.11) shows a similar trend. However, it shows the relationship between environmental factors and white spruce average diameter and trembling aspen average height, which were unclear in Figure 3.10. White spruce average diameter                                                  9 It is assumed that the mean age of white spruce is not significantly different in the study area as no management interventions were carried out in the past.  49  increases with increasing mean annual precipitation and trembling aspen average height decreases with increasing elevation and slope.    3.5 Discussion 3.5.1 Relationship between stand density, basal area, age and tree size 3.5.1.1 Stem density vs. tree sizes I found an inverse relationship between tree sizes and stand density for spruce and trembling aspen, in agreement with my first hypothesis that diameter declines with increasing density. Although, the density of co-dominant white spruce was 4% less than that of the intermediate class, the basal area was 22% higher (Table 3.3), indicating that there is an inverse relationship between tree size and density of the stand and that the average diameter plays a greater role than density in increased forest productivity (basal area). The results also indicated that a decreasing trend in the average diameter of white spruce occurs when tree density exceeds 500 stems per hectare (SPH). Generally, stand density has an inverse relationship with tree size (Lee et al., 1997; Delong and Kessler, 2000; Martin and Gower, 2006), resulting in old growth stands having larger trees and lower densities due to self-thinning processes (Delong and Kessler, 2000). My results suggest that this relationship holds true below 500 SPH; at greater stand densities tree size may become limited by inter-specific competition, whereas at lower densities other factors are at play (Huang and Titus, 1995). The results suggest that thinning and selective harvesting could play an important role in increasing forest productivity in the region. As with white spruce, the average diameter of aspen was higher when the density was less than 500 stems per hectare. Similar findings have also been reported by Perala (1991), 50  Weingartner (1991) and Kabzem et al. (2007). I found a positive relationship between aspen density and basal area, contrasting with Perala (1984), who reported that basal area and biomass of aspen increase as stand density decreases due to self-thinning processes.  3.5.1.2 White spruce age vs. size The ages of white spruce were poorly correlated with height and diameter, despite diameter being strongly correlated with height, density and basal area. McCarthy and Weetman (2006) obtained similar results in their study in Newfoundland, reporting that the age difference of trees with equal height and diameter could be more than 120 years. In my study, trees of similar size were found to be > 120 years apart, indicating that tree size is a poor indicator for defining old growth stands. The result suggests that size is more important than age when assessing stand dynamics and productivity in the region, a finding also supported by White (1980).  3.5.1.3 Stand type vs. species basal area and density  I found that the average density in mixedwood stand of trembling aspen and white spruce was 20% higher than in pure white spruce stands.  This result contrasts with my expectation that productivity would be higher in pure than mixed stands. For example, Linder (1998) and Brassard and Chen (2006) reported higher densities in pure coniferous stands than in deciduous and mixedwood stands. In my study, however, the average basal area was 11.6% higher in mixed aspen and white spruce stands (13.65 m2/400 m2) than in pure white spruce stands (13.27m2/400 m2). Linder (1998) also reported similar findings in his study in central Sweden, namely that the BA was more than 21% higher in mixedwood stands compared to pure coniferous stands. In this 51  regard, Kabzem et al. (2007) suggest that the presence of a small amount of aspen has a large effect on spruce productivity but that this effect diminishes when the aspen density is greater than 1000 stems per hectare. However, Stewart et al. (2003) reported that the basal area of pure coniferous stands was higher compared to mixedwood stands. Similarly, H?ly et al. (2000) reported higher basal areas in deciduous forests followed by coniferous and mixed stands, but Linder (1998) and Stewart et al. (2003) found lower BA in deciduous stands than old mixedwoods or coniferous stands. The reason for the higher basal areas of mixedwood stands in this study may be due to their higher stand densities compared to pure stands in the study area.   3.5.1.4 Variation of stand structures by edaphic and topographic factors (i) Stand density The density of white spruce is generally affected by the interaction of a number of factors and it is therefore difficult to generalize (Wang et al., 1994). In my study, white spruce was more abundant and productive on lower slopes, on sites with higher moisture levels, on S to NE aspects and on soils with a thick organic layer. These findings agree with my expectation that relationships between stand structural attributes vary with edaphic factors, and that productivity increases with site fertility. This finding agrees with studies showing greater productivity of white spruce on silty, alluvial soils (Timoney and Robinson, 1996). White spruce was also found growing well on moderately well-drained clay loams in Saskatchewan (Kabzems, 1971; Qualtierem, 2008) and on decomposed organic matter or a shallow humus layer over mineral soil in Alaska (Purdy et al., 2002).  Similarly, trembling aspen was abundant in sandy to loamy sand soils on southern aspects within the fire plots, which were warmer sites and had the lowest moisture regime, although 52  aspen generally colonizes mostly mesic sites (Gauthier et al., 2000; Brassard and Chen, 2006). Aside from soil moisture, slope position has also been reported as an important factor determining the distribution of aspen (Chen et al., 1998). However, this relationship was not significant in my study. Balsam poplar occurred under similar environmental conditions, mostly in harvested plots. However, the environmental factors affecting the distribution and density of balsam poplar are poorly understood (Bockheim et al., 2003). Floodplain locations with ample moisture have been reported as conducive to poplar growth (Viereck and Foote, 1970; Murray, 1980). Bockheim et al. (2003) reported that balsam poplar was mostly found in areas adjacent to rivers where favourable moisture conditions exist for the species.   (ii) Stand height and diameter  Both multivariate and univariate analyses confirmed that the average heights and diameters of  both trembling aspen and white spruce were higher at sites at lower elevation, and on lower slope positions, agreeing with the expectations in my first hypothesis. Site productivity has been suggested as one of the main factors influencing the diameter structure of coniferous stands (Boucher et al., 2006). As soil properties can significantly affect the height growth of boreal trees (Martin and Gower, 2006), lower rates of decomposition and high rates of leaching can cause lower nutrient availability at higher elevations and on steep slopes (Vitousek et al., 1988; Cole, 1995), and these factors may have affected lower height and diameter growth in this study.  White spruce average heights responded positively to soil moisture, and this has been reported to be one of the important factors affecting the height growth and site index of white spruce (Wang, 1995), which also controls the southern limit of white spruce in western Canada (Chhin and Wang, 2008). Similarly, higher average spruce diameters and heights were found on 53  NE and south aspects and the lowest on N or SE aspects. The reduced diameter growth on the N aspect can be explained by lower temperature, reduced availability of light and thereby shorter growing seasons. Conversely, the reduced diameter on SE aspects could be due to temperature-induced drought (Barber et al., 2000; Juday et al., 2003). Generally hot, dry, summer conditions are exacerbated on S and SE aspects, and may reduce radial growth (Chapin et al., 2004; Barber et al., 2004). Temperature induced summer droughts in the western boreal region were recorded by Barber et al. (2000).  (iii) Stand productivity I found that forest productivity (basal area per hectare) varied significantly by geographical position with higher basal area at lower elevation and on lower slopes. This could have important, implications for deciding silvicultural system in the study area. A similar trend was found in interior and northern British Columbia (Klinka et al., 1996), and northwestern Quebec (Chen et al., 1998 and Boucher et al., 2006). However, white spruce basal area was not correlated with stand age, which differs from the findings of Popadiouk et al. (2003), who reported increasing basal area with stand age in eastern boreal mixedwoods.  3.5.2 Variation of stand structures by disturbance regime I found that stand structure and composition varied significantly by disturbance type, leading me to accept my second hypothesis. Disturbances had significant effects on forest composition 54  including species densities and their regeneration, and promoted the development of a broadleaved forest in the region.  A higher density of white spruce trees and their regeneration was found in plots with some kind of past disturbance such as fire or spruce bark beetle, meeting my expectation that white spruce density would increase with any type of disturbance. The higher density of white spruce in old fire plots is indicative of the later stages of succession, which is generally dominated by shade-tolerant species such as white spruce (Brassard and Chen, 2006). The higher density of trembling aspen in old fire and undisturbed plots indicated that these plots are at an intermediate stage of secondary succession (Chapin et al., 2004). The higher regeneration of trembling aspen in recently harvested fire plots indicated that fire promotes aspen regeneration and that the gap dynamics created by these disturbances (harvested plots had 29% mean crown coverage) were sufficient to promote the regeneration of this light-demanding species (Cumming et al., 2000; Kuusela, 1990). Similarly, balsam poplar and its regeneration were also higher in disturbed forests, particularly in harvested plots. Balsam poplar, like trembling aspen, is a shade-intolerant species and prefers open areas to regenerate and grow (Alexander et al., 1990).  The increased proportion of broadleaves following disturbance met my expectation that proportion of broadleaved trees and regeneration would increase with disturbance.  3.5.3 Variation of stand structures with climatic factors I partially accept the third research hypothesis, that one or more climatic variables (e.g. MAT, MAP, TMXsm and NFFD) would significantly affect stand structure of all three species. I only partially accept the hypothesis, as some of the climatic variables were not significantly correlated 55  with the densities of white spruce and balsam poplar, or variation of height, basal area and density of trembling aspen.  I found higher basal area of white spruce on sites with a higher accumulation of growing degree-days and higher soil moisture. Similarly, the results of the multivariate analysis suggested that average diameter increases with increasing mean annual precipitation.  These two results agree with my expectation that stand productivity would be greater on warmer and wetter locations.  This expands the understanding of Wang and Klinka (1996) in their study in British Columbia, who suggested that the site productivity of white spruce was mostly influenced by edaphic factors rather than climate. In contrast, Chavardes et al. (2012), who undertook their study in the same research plots as my study, found that precipitation was important determining the growth of white spruce, although the relationship varied month to month, with a negative relation found with April precipitation.  Barber et al. (2000) reported that the radial growth of white spruce has decreased with increasing temperature in their study sites in Alaska, although this could be connected to soil moisture. Chavardes et al. (2012) also indicated that although white spruce growth was negatively related to maximum temperatures in general, increased precipitation accompanied by warmer temperature were correlated with increased tree growth in the last 30 years. It seems feasible that precipitation is a key factor affecting tree growth in the area but it may interact negatively with increasing temperature. The mean annual precipitation in the study area is of 319.5 mm (ClimateBC, 2006), which is projected to be increased by 20% (ACIA, 2004), which may favor spruce growth if temperature is not a limiting factor.  The multivariate analysis used in this study suggested that the average height of white spruce increased with increasing summer maximum mean temperature. Climatic factors are considered more important than edaphic factors for white spruce height growth as height 56  variations have been found in different climatic regions with the same edaphic site index (Wang et al., 1994). However, it has also been reported that summer drought caused by increased temperature and hence evapo-transpiration may limit the growth of white spruce (Barber et al., 2000). A similar result was found by Wilmking et al. (2004) in their study in Alaska.  The relationships between trembling aspen/balsam poplar structural variables and climatic factors were not significant in this study, contrary to my expectations.  Climatic factors are frequently reported to be important in determining the productivity and height growth of aspen (Hogg et al., 2002). However, the multivariate analysis indicated that the average diameter of aspen increased with increasing number of degree-days above 5 oC. This may explain why warmer temperatures favour the growth of this species (Fralish and Loucks, 1975), if no drought condition exits.  3.5.4 Overall stand composition and structural in the study area As expected, white spruce dominated the study area with the highest composition index (0.9) and stem density. Generally, the old growth sere is dominated by shade-tolerant species in the boreal forest succession unless there is a catastrophic disturbance (Bergeron and Dubuc, 1989; Brassard and Chen, 2006), as it can establish under shade in the absence of major disturbance (Takahashi et al., 2001). The continuous loss of light-demanding species such as aspen has been reported in the absence of major periodic disturbances (Rogers, 2002).  The density of white spruce were greater within the intermediate height class (Delong and Kessler, 2000), as generally stem density is generally higher in younger stands than old growths. 57  I found the maximum density of aspen within pure aspen plots to be 1575 trees per hectare, very close to what Bates (1989) has suggested as the ideal density of aspen within mature stands after natural thinning (1000 to 1730 per hectare). Generally, aspen density is lower in mature stands due to natural thinning processes (Bates, 1989). I found that the average regeneration density of aspen was more than 40,000 stems per hectare, enough to fully stock stands of aspen after natural thinning (Bates, 1989), given that a regeneration density of about 2500 to 3000 stems per hectare has been suggested as sufficient to produce fully stocked aspen stands at maturity (Ek and Brodie, 1975; Sorensen, 1968; Bates, 1989).  The maximum number of balsam popular recorded in the research plots was 100 trees per hectare (in mixed stands), lower than the ideal density of 1127 stem/ha reported by Douglas (1974) in balsam poplar dominant stands. The Shannon-Wiener diversity indices indicated that there was higher structural diversity in undisturbed or old fire plots. Structural diversity is one of the characteristics of old growth boreal forest (Brassard and Chen, 2006). Diversity generally differs with tree species dominance, stand age and with changing proportions of shade-tolerant and intolerant species (Varga et al., 2005).   3.5.5 Chapter conclusions The increasing forest disturbances and the changing climate have important implications for ecosystem structures and forest management in SWY. I hypothesized that the forest structures and composition were affected by bio-geoclimatic and disturbance factors in the region. My results showed that although white spruce dominates the landscape in terms of its density, basal 58  area, average height and DBH, its regeneration was proportionately less than that of trembling aspen, especially in disturbed plots, suggesting broadleaved species may become more common if disturbance persists in the region. I also found that mixed stands of aspen and white spruce had higher basal areas per hectare than pure stands of trembling aspen or white spruce and that average stand diameter was positively correlated with tree density up to 500 trees per hectare, beyond which the average diameter decreased. Average density and basal area of white spruce increased with decreasing elevation and were higher in lower slope positions. These findings suggest that forest management should promote mixed stands in boreal region for higher productivity with regular thinning or selective harvesting to enhance regeneration and forest productivity.   59  Table 3.1 Total number of trees, seedlings and saplings of the species sampled  Species Number of trees (400 m2) % of total Number of saplings  (25 m2) % of total Number of seedlings (1 m2) % of total White spruce 2577 87.5% 645 67.8% 2426 65.8% Trembling aspen 360 12.2% 292 30.7% 1173 31.4% Balsam poplar 7 0.3% 14 1.5% 95 2.6% Total 2944 100% 951 100% 3694 100%  [Note: The above table shows that white spruce has the highest density of tree, seedling and sapling. However, compared to tree density, the % of its seedling and sapling are lower. In case of trembling aspen, tree stem cover is 12.2% only but it has more than 30% seedling and sapling cover.]  Table 3.2 Stand composition index  Species Basal area (m2)     Index  White spruce  1253.4  0.907932  Trembling aspen  124.74  0.090356  Balsam poplar  2.34  0.001698  Total  1380.5   [Note: The above table shows the stand composition index for each species (Huang and Titus, 1995). White spruce has the highest composition followed by trembling aspen and balsam poplar]    60  Table 3.3 Average height, diameter and basal area of the sampled tree species by dominance class  Species Dominance class (height class) Total trees sampled (no) % of total sampled Average height (m) Average diameter (cm) Total basal area (m2) % of total Basal Area White spruce Dominant 63 18.3 15.9 25.7 12.8 23.2 Co-dominant 71 20.6 12.5 19 23.6 42.6 Intermediate 84 24.3 7.3 9.9 11.3 20.4 Suppressed 45 13 6.6 8.4 2.5 4.6 Trembling Aspen Dominant 19 5.5 11.4 19 1.2 2.1 Co-dominant 26 7.5 8.6 14.1 3.2 5.7 Intermediate 23 6.7 5.9 7.4 0.7 1.2 Suppressed 7 2.0 3.3 4.3 0.1 0.1 Balsam Poplar Dominant 2 0.6 7.4 15.1 0.1 0.1 Co-dominant 0 0 0 0 0 0 Intermediate 5 1.4 4.6 6.7 0.02 0.1 Suppressed 0 0 0 0 0 0 [Note: The above table shows that white spruce intermediate height class has the highest density (24.3%) but its co-dominant height class has significantly highest basal area (42.64%)]    Table 3.4 Shannon-Weiner structural diversity indices by disturbance types (Bold fonts indicate the highest values by species)  Shannon-Weiner Indices White spruce height  White spruce diameter (cm) Trembling aspen height (m) Trembling aspen diameter (cm) Beetle affected plots 1.5 1.7 1 1 Recent fire plots 1.6 1.6 0.6 0.6 Old fire plots 1.5 1.7 1.1 1.3 Undisturbed plots 1.9 2.2 1.2 1  61  Table 3.5 Stand structure variables that were significant with environmental variables  White spruce Trembling aspen Balsam poplar Height Elevation (F = 7.4; P < .001) Slope (F = 4.6; P =0.037) Soil moisture (F = 4.1; P =0.33) Disturbance type ( F = 5; P<0.001) Elevation (F = 4.6; P =.02) Aspect (F =3.8; P <.05) Soil texture (F = 2.7; P <.001) Mean annual prep (F = 6.5; P<.01) Not significant Diameter Elevation ( F = 3.7;  P = 0.046) Slope (F = 3.7; P <0.05) Aspect (F = 3.2; P<0.05) Elevation (F = 4; P = 0.039) Soil texture (F = 2.4; P <.001) Aspect (F = 4.6; P<0.05) Mean annual prep (F = 5.7; P<.01) Not significant Density Aspect (F = 2.2; P = 0.035) Slope (F = 3.4; P = 0.44) Mean annual temp. (F = 3.6; P <.05) Mean annual prep. (F = 4.5; P =0.03) Soil moisture (F = 3.6; P<.05) Soil texture (F = 4.3; P < 0.05) Disturbance type (F = 6.1; P<0.0001) Aspect (F = 4.6; P<.05) Soil texture (F =3.8; P<.05) Aspect (F=3.4; P<.05) Soil texture (F =2.4; P<.05) Basal area Aspect (F = 4.7; P<.05) Elevation (F = 3.1; P <.05) Slope (F = 5.4; P = 0.03) Number of days>5 0C (F =3.4; P = 0.042) Disturbance type (F = 10.5; P <0.001) Aspect (F = 4.2; P<0.05) Soil texture (F =3.8; P<.05) Not significant  [The above table shows the relationship of structural variables of white spruce, trembling aspen and balsam poplar with climatic, topographical, edaphic and disturbance variables. The categorical independent variables were tested using ANOVA (with means of plots) and numerical variables were tested by fitting regression lines. The only variables which were significant at p<0.05 are listed above. The table indicated that white spruce structures varied significantly with several factors. Trembling aspen diameter and height also varied significantly with edaphic and topographic factors.     62  Table 3.6 R2 values between stand structure variables and other ecological variables  STT ATT PTT TT TCC AtBA PoBA SpBA ToBA Spavht Spavdia Atavht Atavdia Poavht ATT 0.01 1             PTT 0.01 0.01 1            PTT 0.89 0.07 0.01 1           TCC 0.40 0.02 0.01 0.44 1          AtBA 0.01 0.75 0.01 0.05 0.03 1         PoBA 0.01 0.02 0.91 0.05 0.07 0 1        SpBA 0.72 0.03 0.02 0.63 0.36 0.01 0 1       ToBA 0.67 0.08 0.02 0.68 0.44 0.03 0 0.93 1      Spavht 0.29 0.01 0.07 0.24 0.24 0 0.01 0.48 0.47 1     Spavdi 0.28 0.07 0.09 0.19 0.19 0 0.02 0.43 0.42 0.89 1    Atavht 0.02 0.18 0.02 0.06 0.06 0.25 0.01 0.01 0.05 0.05 0.03 1   Atavdi 0.02 0.18 0.02 0.04 0.06 0.29 0.01 0.01 0.06 0.06 0.04 0.96 1  Poavht 0.06 0.04 0.98 0.07 0.06 0 0.9 0.01 0.01 0.01 0.02 0.01 0.01 1 Poavdia 0.08 0.04  0.99 0.008 0.01 0 0.95 0.01 0.01 0.01 0.02 0.01 0.01 0.98  [Note: R2 values greater than 0.2 have been highlighted. Abbreviation of names: STT = spruce density, ATT = aspen density, PTT = poplar density, TCC = total crown cover; AtBA = aspen basal area (m2), PoBA = poplar basal area (m2), SpBA = spruce basal area (m2), ToBA = total basal area (m2), Spavht = spruce average height, Spavdia = spruce average diameter, Atavht = aspen average height, Atavdia = aspen average diameter, Poavht = poplar average height, Poavdia = poplar average diameter]  Table 3.7 Comparing density index by disturbance type  Species  Mean index for  90 plots   Beetle Plots   Fire plots  Harvested plots  Old fire plots  Undisturbed plots  White spruce tree 100 192 32 71 120 92 White spruce sapling 100 141 70 74 157 83 White spruce seedling 100 157 205 73 67 70 Trembling aspen tree 100 30 20 80 179 137 Trembling aspen sapling 100 2 104 177 59 111 Trembling aspen seedling 100 18 322 94 96 64 Balsam poplar tree 100 0 0 257 92 83 Balsam poplar sapling 100 0 58 418 0 0 Balsam poplar seedling 100 7 370 204 0 24         [Note: The above table shows index values of trees, saplings and seedlings in each disturbance type compared to the mean values for 90 plots. It indicates that white spruce tree density was higher in beetle-affected and old fire plots. Both trembling aspen and balsam poplar regeneration were higher in fire and harvested plots]   63  Table 3.8 Ratio of tree regeneration to parent trees by disturbance types (values are averaged per plot)  STT Spsed STT:Spsed ATT Atsed ATT:Atsed PTT Posed PTT:Posed Beetle (16 plots) 55 52 1:0.95 1 2 1:2 0 0 0 Fire (10 plots) 9 59 1:6.6 1 45 1:45 0 4 0:4 Harvest (20 plots) 20 25 1:1.25 3 18 1:6 2 3 1:1.5 Old fire (16 plots) 34 29 1:0.85 7 15 1:2.1 0 0 0 Undisturbed (28 plots) 26 25 1: 0.96 5 12 1:2.4 0 0 0  [Note: The table shows that trembling aspen had much higher ratio of trees to regeneration in disturbed plots than that white spruce. Abbreviation of names:  STT = white spruce tree density; Spsed = white spruce seedling density; ATT = apsen tree density; Atsed = aspen seedling density; PTT = balsam poplar tree density; Poseed = balsam poplar seedling density.]   Figure 3.1 Species distribution in study area Note: White spruce data were received from Ministry of Eneery, Mines and Resources, Government of Yukon. Abbreviation:  A = Aspen; B = Balsam poplar, SW = White spruce, Blank = no species recorded.] 64    Figure 3.2 Distribution of average height (m) in the landscape [Note: The average height was higher in research blocks D, E and F, which were characterized by lower elevation, lower slope position or flat and exposed sites. Data were received from MEMR, Yukon Government]   65   Figure 3.3 Stand diameter Vs environmental factors  [Note: The figures indicate that white spruce diameter varied significantly with elevation, slope and aspect. Similarly trembling aspen diameter varied with slope, soil texture and aspect. The variations were significant at p<0.05]  66   Figure 3.4 Stand heights vs. environmental variables [Note: The figures indicate that white spruce average height decreased with increasing elevation and slope, but increased with increasing moisture level. Trembling aspen height also decreased with increasing elevation. Average heights of aspen were higher in loamy soil and on eastern aspect.]  67   Figure 3.5 Stand productivity vs. environmental variables [Note: Stand productivity indicates basal area per hectare. The figures indicated that diameters of both white spruce and trembling aspen were positively correlated to its density and basal area, though average diameter decreased once density reached 1000/ha. Relationship between white average age and diameter was not significant. The colours of the lines are the default outputs from R. Red colour lines indicate the general distribution of the data and green and black colours straight lines are the regression fit lines.] 68   Figure 3.6 Stand basal area vs. environmental variables {Note: White spruce basal area was greater at lower elevations and slope positions with higher growing number of days > 5 oC. Trembling aspen basal area and density followed similar trends]  69   Figure 3.7 White spruce average density vs. environmental variables [Note: The average densities of white spruce were higher in lower slope positions with higher mean annual temperature and higher soil moisture. Organic soils were present in dense spruce stands]  70   Figure 3.8 Trembling aspen and balsam poplar average density vs. environmental variables [Note: The densities of trembling aspen were higher in exposed sites on NE and SW slopes. The species was abundant on sandy and loamy sand soils]  71   Figure 3.9 Variation of stand density and basal area by disturbance type  [Note: White spruce density and basal area were higher in beetle affected plots whereas for trembling aspen, density and basal area were higher in old fire plots] 72   Figure 3.10 Axis I - Environmental/structure variables  [Note: Figure 3.10 was the ouput of the Multiple Co-intertia analysis. It shows that the average height of white spruce increases with increasing summer maximum mean temperature, eastern aspect and decreases with increasing slope, elevation and summer precipitation. Similarly, trembling aspen average diameter increases with increasing degree-days more than 50C but decreases with elevation and slope. Abbreviation of names: NFFD = number of frost free days; mat = mean annual temperature; TMNwt = minimum winter temperature; TMXwt = maximum winter temperature; AbDBH = average DBH of balsam poplar;  PPTsm = average precipitation; ele = elevation;  Org_D = organic depth; Soil_D = soil depth; DD5 =  degree days above 5;  AtDBH =  Average DBH of aspen;  TMXsm =  Maximum summer temperature; map = mean annual precipitation;  SwHt = white spruce average height; SwDBH = white spruce average diameter; Sinasp = Sine of aspect; Cosasp = Cosine of aspect; DD0 = degree days< 0 oC; PPTwt = Winter precipitation].   73   Figure 3.11 Axis II - Species environmental axis (Note: Figure 3.11 (output of the Multiple Co-inertia Analysis) provides additional information. It indicates that white spruce average diameter increases with increasing mean annual precipitation and trembling aspen average height decreases with increasing elevation and slope. Abbreviation of names: same as Figure 3.10)   74  4. Tree regeneration in response to bio-geoclimatic and disturbance factors in the southwest Yukon of Canada  4.1 Introduction The boreal forests of southwest Yukon (SWY) have been experiencing the consequences of climate change, particularly through increases in disturbance. These perturbations strongly affect the ecosystem properties and processes that shape community structure and composition (Johnstone and Chapin, 2006b). They are particularly important in determining the recruitment of boreal tree species and their abundance and distribution (Johnson et al., 2003) and have significantly affected species regeneration in the region. This chapter deals with the distribution and abundance of tree species regeneration in SWY in response to bio-geoclimatic and disturbance factors.   4.1.1 Tree species and their regeneration in the southwest Yukon The southwest Yukon has only three major tree species i.e. white spruce, trembling aspen and balsam poplar.  The regeneration potential of these species depends upon various factors including seed sources and viability, the condition of substrates, climatic condition and disturbances. White spruce regenerates from seed (Qualtierem, 2008); and therefore, its regeneration is primarily dependent on annual seed fall and masting events (Nienstaedt and Zasada, 1990). The condition of the soil substrate is very important for white spruce regeneration. Burned substrates with rotten logs and exposed mineral soils following the 75  disturbances provide better environments for recruitment (Simard et al., 2003; Stewart et al., 2001; Peters et al., 2005; Wirth et al., 2008), provided that there is sufficient moisture availability within the mineral substrates (Wang and Kemball, 2005; Johnstone and Chapin, 2006a). However, white spruce may also establish on an organic substrates provided that moisture is not a limiting factor (Wirth et al., 2008). In addition, this species typically prefers coarse-textured, well-drained sandy to loamy textured soils rather than clay-dominated soils (Galipeau et al., 1997; Lindbladh et al., 2007).   Both trembling aspen and balsam poplar require similar geoclimatic conditions for their regeneration, but with aspen more prevalent on upland sites and balsam poplar on lowland sites. They both regenerate vegetatively and by seeds (Perala, 1990; Zasada and Phipps, 1990). Suckering is the principal mode of regeneration however for both species following a major disturbance (Frey et al., 2003; Bartos, 2001; Zasada and Phipps, 1990). The seeds of both species are small and light and can be dispersed by wind over many kilometres (Perala, 1990; Zasada and Phipps, 1990). Once dispersed, the seeds remain viable for 2 to 4 weeks for aspen (Perala, 1990) and 4 to 6 weeks for balsam poplar (Zasada and Phipps, 1990). Soil moisture is very important factor for their germination of both species (Zasada and Densmore, 1980; Krasny et al., 1988; Perala, 1990). Both prefer exposed sites with high light conditions and mineral substrates, typically provided following major disturbances, (Zasada et al., 1981; Klinka et al., 2000; Johnstone and Chapin 2006a).  4.1.2 Climatic and disturbance factors and tree species regeneration in SWY Climate variability and disturbances can have a significant impact on the recruitment and survival of all three species. White spruce is very sensitive to fire because of its thin bark and 76  lack of an aerial seed bank (Kuusela, 1990; Messaoud et al., 2007). Intense and severe fire may kill seeds and reduce the probability of germination due to loss of nutrients by vaporization and combustion of organic matter (Neary et al., 2005). Increases in temperature and changes in soil moisture following fire may prevent or delay regeneration of white spruce (Oswald and Brown, 1990). Fire, however, a very important factor for white spruce seedling establishment and the interaction between masting years and fire can result in abundant regeneration of white spruce (Peters et al., 2005). In areas of the northern boreal forest that are dominated by cold, wet, organic soils and/or permafrost, fire plays an important role by removing organic layers and causing the degradation of permafrost (Yoshikawa et al., 2003).  Similarly, fire is important for aspen regeneration (Wang, 2003). Severe fire can however damage roots and suckers, which can significantly affect the density, height and diameter of aspen regeneration (Wang, 2003). Increased fire frequency resulting from warmer temperatures could favour the establishment of aspen and other broadleaved woody species in the boreal forest (Flannigan and Van Wagner, 1991). Pests and diseases are important disturbance agents that can affect the ability of species to regenerate. A reduction in seed availability due to the extensive mortality of maternal trees can affect the regeneration and establishment of white spruce (Jasinski and Payette, 2005). Annual variation in precipitation is an important factor for the regeneration of both balsam poplar and trembling aspen as these species are sensitive to dry climatic condition and moisture stress, which can limit their regeneration leading to species reductions over time (Perala, 1990; Hogg and Wein, 2005; Yarie, 2008).    77  4.1.3 Research problems, hypotheses and questions Disturbances affect community structure and composition, which in turn can affect ecosystem properties and processes (Johnstone and Chapin, 2006b). Post-disturbance recruitment plays an important role in shaping the landscape patterns of boreal forest vegetation, providing opportunities for the recruitment of boreal tree species, which in turn impacts upon species abundance and distribution (Johnson et al., 2003). In the Champagne and Aishihik Traditional Territory (CATT) of the Southwest Yukon (SWY), >90% of the forest has been affected by the recent spruce bark beetle (SBB) outbreak in addition to the 100,100 ha of forest that have burned in the region between 1946 and 2003.  Given the documented changes in climate over the last 50 years and the large areas of forest affected by the SBB outbreak and fires over the last five decades, I sought to identify what relationships exist between tree recruitment, climate and disturbance. In response to the recent spruce bark beetle epidemic, salvage harvesting has been proposed within the study region to generate economic growth opportunities for local communities, with forest re-establishment to be dependent upon natural regeneration (SFMP, 2004). However, factors that have determined past recruitment success in the region have not been quantified, which the SFMP (2004) recognized as an important research issue in the study area. The following research hypotheses were developed in connection to disturbances and climate variability in the region and their effects on tree regeneration;   i. Post disturbance recruitment is important in shaping boreal forest composition. As the CATT has experienced four types of disturbance and has three major tree species, the regeneration response of each species was expected to differ according to type and severity of disturbances. I hypothesized that the extent and scale of regeneration of each 78  species to vary significantly by disturbance type.  Specifically, I expected higher regeneration abundance of all three species following recent fire than the other disturbance types. I also expected that any disturbances would favor the regeneration of broadleaved species over coniferous species.   ii. Similarly, climatic factors, specifically temperature, have been reported to have very important effects on the regeneration success of boreal species. I hypothesized that the distribution and abundance of tree recruitment should be associated with climatic factors. I specifically expect that the regeneration of all three species would respond positively to higher temperature and precipitation. Besides reporting on my tests of the above hypotheses, this chapter also deals with following research questions that could have important implications for forest management decision making. ? What is the role of disturbance in the success of tree regeneration in SWY? ? How do climate variability and environmental factors affect the density and distribution of regeneration in SWY?  ? What edaphic factors limit the distribution and density of regeneration in the landscape in SWY?  4.2 Data analysis I used both univariate and multivariate statistical tools to assess the responses of tree regeneration to environmental and disturbance factors. Analysis of Variance (ANOVA) was performed to examine the mean variation of regeneration by disturbance type. The ratio of species regeneration to their parent trees was calculated by disturbance type to see the effect of 79  disturbance on the regeneration of each species. The analytical tools used in this chapter are described separately below.   4.2.1 Multivariate canonical correlation (MCA) MCA was used to investigate the multivariate correlation between dependent (regeneration abundance) and independent variables (climate, topography, stand structure, soils, and disturbance). MCA defines a multivariate relationship between two or more sets of variables and can accommodate multiple dependent variables. It identifies the sets of weights for the dependent and independent variables that provide the highest possible correlation between composite variates (McGarigal et al., 2000).     4.2.2 Negative binomial generalized model  The regeneration abundance of white spruce, trembling aspen and balsam poplar were modelled using a Negative Binomial Generalized Model (PROC Genmod), which uses the Maximum Likelihood method for estimating parameters (SAS Institute Inc, 2008; McCullagh and Nelder, 1989). Poisson and negative binomial probability distributions are generally used for count data modelling. A negative binomial distribution is generally preferred to Poisson when the variance is higher than the mean value; this was the case in this study so the former was utilised. Negative binomial distributions have frequently been used for modelling forest regeneration (Venables and Ripley, 2002; Fyllas et al., 2008). The overall model to estimate regeneration density was: 80  RD = f (Topographical variables +ecological variables +climatic variables +disturbance regime) + residual error           Log (RD) = b0 + b1*X1 +b2*X2 + ....+ bn*Xn, RD = exp(b.....+ b1*X1 +b2*X2 + ....+ bn*Xn)...............................Equation (4-1)   where RD is the regeneration count for each plot, b0 is the intercept of the slope of the model and X1 to Xn are explanatory variables that represent topographic, ecological and climatic variables, as well as the disturbance regime. I used a stepwise modelling process to remove insignificant variables. I tested the goodness-of-fit of the model by comparing the Akaike Information Criterion (AIC) values of full models to the final reduced model. The smaller the AIC value, the better I judged the model to be. Similarly, the model?s significance was tested using a Chi square test.  F-calculated = 2*(log likelihood of full model ? log likelihood of reduced model) If the calculated F value was larger than the critical F value at an alpha of 0.05, then I considered the model to be significant. I used SAS Inc. software to develop the models (SAS, 2008).   4.2.3 The most nearest neighbor imputation model (MSN Imputation) The most nearest neighbourhood imputation model was as a complementary model for estimating regeneration abundance in the study area. This model has frequently been used for estimating regeneration abundance (Ek et al., 1996; Hassani et al., 2002; Froese, 2003; LeMay et al., 2007). Imputation is defined as ?replacing missing or non-sampled measurements for any 81  unit in the population with measurements from another unit with similar characteristics? (Ek et al., 1997). Imputation methods include tabular imputation, most similar neighbour, nearest neighbour, k-nearest neighbour, and geo-statistical estimation (Hassani, et al., 2002).  The imputation method splits original data into ?reference? and ?target? variables. Data that have Y variables are classified as reference data and without Y variables are termed as target data (Ek et al., 1997). Programme ?yImpute? in R statistical (R Core Development Team, 2010) was used for this purpose.  The imputation method has several benefits. It gives the result as estimates within the bounds of biological reality as it provides an estimate from existing Y variables (Moeur and Stage, 1995; Haara et al., 1997). It also allows any form of data, so no assumptions or data transformations are required for the variables (Haara et al., 1997; LeMay et al., 2007). In addition, the method can estimate multiple Y variables simultaneously (Katila and Tomppo, 2002). The root mean square error (RMSE) and correlation (R) between imputed and observed values of each species were calculated. The lower the RMSE the better the predictive power of the model, and when RMSE equalled zero the model was judged perfect (Tabachnick and Fidell, 1996). RMSE =                ?............................Equation (4-2) Where    = observed variables;   = predicated variables; and n = number of sampling plots.    82  4.3 Results 4.3.1 Regeneration in response to disturbances, and climatic and environmental factors Regeneration of white spruce, trembling aspen and balsam poplar varied significantly (p< 0.05) by disturbance type (Table 4.1). Although the regeneration of all three species was highest in the recent fire plots, trembling aspen and balsam poplar had higher ratios of regeneration to their parent trees than spruce. The fire plots were characterized by thin organic humus layers (= 2.75 cm; sd = +/- 2.09), the lowest crown covers (= 12%; sd = +/- 25.5), lower soil moistures (= 19.82%; sd = +/- 11) and the least shrub cover (36.23%) compared with the other disturbances (Table 2.1).  Both non-disturbed and old fire plots had the lowest regeneration density for all tree species. These plots were characterized by about 50% crown cover, thick organic humus layers (>8 cm) and higher shrub covers (> 50%) (Table 2.1). The multivariate canonical correlation (MCA) categorized the total landscape into four axes (gradient). In MCA, each axis was correlated to several environmental variables with maximum variance (Table 4.2, Appendix, D). Axis 1 was defined by the variables balsam poplar basal area (PoBA) and total tree density (PTT), which were found to describe the occurrence of balsam poplar regeneration (Table 4.3). Axis 2 correlated positively with fire and negatively with total crown cover (TCC), organic depth, white spruce tree density and soil moisture. This habitat type was suitable for trembling aspen regeneration but was negative for white spruce saplings. Axis 3 was mostly influenced by lower elevations (650 m to 750 m), lower slopes (less than 15%), higher mean annual temperatures, higher growing degree days, and the occurrence of fire, which favoured spruce regeneration (both seedling and sapling). Axis 4 was influenced negatively by elevation and positively by slope, summer maximum temperature and the occurrence of spruce bark beetle, which also favoured the occurrence of both seedlings and 83  saplings of white spruce. The main differences between habitat types 3 and 4 were the slope (%) and disturbance type, indicating that white spruce was present in fire plots on lower slopes and beetle-affected plots on higher slopes.  The plots of regeneration density against specific environmental factors (Figure 4.1, Figure 4.2 and Figure 4.3) all showed similar results to that of the MCA. Figure 4.1 indicated that white spruce exhibited a positive relationship with sites at lower elevations and lower slope positions that have higher growing degree-days and higher mean summer temperature and with soils that had lower soil moisture and had been subject to disturbance. Both trembling aspen and balsam poplar showed similar responses to environmental factors. I found regeneration of these two species at sites affected by fire and harvesting, indicating their preference for disturbed sites. They also responded positively to sites with exposed aspects (flat, east to south to west), indicating a preference for sites with higher solar radiation inputs. Trembling aspen regeneration responded negatively to increasing mean annual temperature, crown cover, organic depth and responded positively to the presence of loamy to silty loam soil texture.  Regeneration of balsam poplar was greater in sandy and silty clay soil.   4.3.2 Negative binomial regeneration abundance model Altogether, 38 explanatory variables were used to predict the regeneration abundance of white spruce, trembling aspen and balsam poplar (Table 2.3).  The negative binomial dispersion parameters were estimated by the maximum likelihood method.  The stepwise modelling method resulted in many variables being dropped (p>0.05) with eight, 14 and nine variables retained for white spruce, trembling aspen and balsam poplar, respectively. Table 4.4 provides the criteria for 84  assessing goodness of fit of the models including AIC and log likelihood. The algorithms of all three models were converged and the results are provided for white spruce, trembling aspen and balsam poplar in Tables 4.5, 4.6 and 4.7 respectively. The dispersion values were higher than zero in all the models, justifying the application of the negative binomial model. The majority of the climate variables and the total basal area were significantly correlated with regeneration for all three species. Variables related to disturbance were significant for trembling aspen and balsam poplar only. The AIC values of full models of white spruce, trembling aspen and balsam poplar were 2207.32; 1302.42 and 346.73, respectively, which were reduced to 2164.98, 1273.94 and 340.68, respectively, by the stepwise method (Table 4.8). The reduced AIC signifies the model fitness. F-tests were performed for all three models using log likelihood values of full models and null models. The calculated F values for models were larger than the F table values at p ? 0.05 (Table 4.9). The F-tests demonstrated that all three models were statistically significant.  4.3.3 MSN imputation model An imputation model was used on all explanatory variables to impute the results from reference to target variables. The results of the heads and tails of imputed and observed values of the models are shown in Table 4.10. The full results have been provided in Appendix E. The model significances were tested using the correlation between observed and imputed values and the RMSE of the imputed values. The correlation between observed and imputed regeneration abundance of white spruce, trembling aspen and balsam poplar were 0.618, 0.656, and 0.874, respectively. The relationships between the imputed and observed regeneration abundance for 85  each species are illustrated in Figure 4.4. The RMSE of the models for the three species were 89014, 79974 and 10114, respectively. The correlation and RMSE values indicate that the abundance of balsam poplar regeneration was better predicted than the other two species.   4.4 Discussion 4.4.1 Disturbances vs. regeneration  I found that disturbance had positive effects on the successful regeneration of all three species in the study area, supporting my first hypothesis that this factor plays important role in tree recruitment by creating a favourable environment for regeneration. As expected, regeneration was particularly evident in recent fire plots, which had a thin layer of organic matter and low moisture content compared to other plots. Fire helps white spruce regeneration by removing cold wet organic soils and degrading permafrost, creating a suitable seedbed by exposing mineral soil (Peters et al., 2005; Yoshikawa et al., 2003; Purdy et al., 2002). However, time elapsed since last fire was an important factor as the old fire plots had significantly less regeneration than recent fire plots. In other studies, white spruce regeneration was greatest within first seven years after a fire but declined after that, with the decline becoming significant after 10 to 14 years (Johnstone et al. 2004; Purdy et al., 2002; Macdonald et al., 2001). The changes in micro-habitat conditions in the old fire plots were the main reason for lower regeneration; these plots were characterized by thicker organic layers, higher crown cover and the presence of a dense shrub layer. However, regeneration success after fire also depends upon the co-occurrence of masting events with fire events (Peters et al., 2005). I found the highest abundance of white spruce regeneration in beetle-affected plots, although the micro-sites within these plots were quite different to recent fire plots, 86  with a thicker layer of organic matter (= 11.29 cm; sd = +/- 9.5), higher crown cover (63.93%; sd = +/- 17.5), higher average moisture (= 25.96%; sd = +/- 11.8), lower shrub cover (40%), and the presence of abundant decayed logs, suggesting that spruce bark beetle disturbance had little effect on understory vegetation. The abundant presence of decayed logs may be the reason for the higher regeneration in these plots, as regenerations were densely concentrated on and around the decayed logs, which provide a favourable seedbed for white spruce regeneration (Vernon et al., 2006; Purdy et al., 2002). The regeneration of white spruce in intact forest or old fire forests was generally less than elsewhere since the mineral seedbed deteriorates with time since disturbance (Simard et al., 1998; Nienstaedt and Zasada, 1990), restricting regeneration to decayed logs with sufficient moisture availability (Coates et al. 1994).  Open light conditions created by disturbances had a positive impact on trembling aspen and balsam poplar regeneration and their regeneration was more abundant in fire plots than for white spruce, partially supporting my second expectation. Both species are shade-intolerant and unable to regenerate in the understory of mature stands (Landh?usser and Lieffers, 2001). Johnstone et al. (2004) obtained similar results in their study in the southeast Yukon, finding that seedling establishment of trembling aspen was highest after fire. Due to its ability to sucker, trembling aspen has a high tolerance to fire, making it better than white spruce at re-colonizing fire-disturbed areas (Purdon et al. 2002). Fire can benefit regeneration of balsam poplar by promoting vegetative reproduction (Rood et al., 2007). Regeneration of both broadleaved species was also higher in salvage-harvested plots. Harvested plots had similar micro-site conditions to the recent fire sites, with thinner layers of organic matter, lower moisture content, and lower crown and shrub cover. Similar results were obtained by Kurulok and Macdonald (2004; 2007) in their study in Alberta, although they concluded that salvage logging might negatively affect 87  aspen regeneration, as the salvaged stands tend to have higher abundances of shrubs and grasses.  The higher regeneration of both broadleaved and conifer species on disturbed plots therefore contradicted my expectation that any disturbance would favor the regeneration of broadleaved species over coniferous species.  4.4.2 Climate variability vs. regeneration This research identified a strong relationship between climatic variables and white spruce and trembling aspen regeneration in the SWY, supporting my hypothesis that climatic factors significantly affect boreal tree recruitment and influence its abundance and spatial distribution. White spruce was found to prefer sites with higher summer temperatures and a higher number of growing degree-days, partially agreeing with my expectations. This is supported by the work of Andalo et al. (2005) in Quebec, who reported that temperature is the key factor for the germination of white spruce. Contrary to my expectations, precipitation did not significantly affect white spruce regeneration in my models. According to Archibold et al. (2000), the base germination temperature of white spruce is significantly correlated with average June precipitation as the optimal temperature for germination of white spruce is 20 ?C, which can occur in June. However, drought and moisture limitations induced by higher temperatures and evapo-transpiration can significantly affect white spruce regeneration (Barber et al., 2000; Hogg and Wein, 2005).  The results of both the regeneration model and the MCA identified that temperature was negatively and precipitation positively correlated with aspen regeneration, suggesting that available moisture is an important factor for regeneration of aspen. It also suggests that this 88  species is sensitive to the drought conditions that can be associated with higher temperatures (Frey et al., 2003). These finding are consistent with those of Hogg et al. (2005), working in northeast British Columbia, who found that aspen is sensitive to drought conditions. Dry soil conditions and/or drought can also significantly affect regeneration and growth of aspen (Frey et al., 2003).  Thus, climatic factors favouring aspen regeneration contrasted strongly with those for spruce.  The model coefficients also suggested that balsam poplar was sensitive to high temperature and soil moisture. The species was significantly affected by changes in soil moisture. According to Tyree et al. (1994), this species is very sensitive to drought stress, which can be induced by high temperatures and low soil moisture regimes. I found that the number of growing degree-days > 5 ?C had a positive influence in the model, indicating that the species prefers a warm growing season with sufficient soil moisture for regeneration. This is supported by Kalischuk et al. (2001), who found seedling growth was positively correlated with the accumulation of growing degree-days in their studies in southeastern BC and southern Alaska.  4.4.3 Edaphic/topographic factors vs. regeneration  Besides climate variability and disturbances, all three species distinctly responded to diverse edaphic and topographic factors and their abundance and distribution varied significantly with those factors. I found that trembling aspen regeneration prefers loamy to silty loam soils, a finding supported by Perala (1990) and Hogg et al. (2008), who found that aspen prefers well- drained loamy soils. Balsam poplar regeneration was significantly higher on sandy or silty clay soils, which are the major soil types found on alluvial and flooded sites where this species is 89  commonly observed (Viereck et al., 1993; Cordes et al., 1997). Variation of white spruce regeneration abundance with soil texture was not significant in this study, although it does grow well on loamy to clay loam soils (Kabzems, 1971).  The abundance of white spruce regeneration decreased with increasing soil moisture content, with a greater abundance found on soils with moisture content between 10 to 20%. Purdy et al. (2002), working in mixed boreal forest of Alberta, also found that increasing soil moisture can limit white spruce regeneration, although the threshold for an impact of soil moisture was not mentioned. Similarly, trembling aspen regeneration decreased with increasing organic depth.  These findings are in accordance with their higher abundance in fire plots, which had no or thin organic layers that contribute to lower soil moisture content (Johnstone and Chapin, 2006a). The negative relationship between organic depth and regeneration was explained by Jonhstone and Chapin (2006); it correlates with the positive response to fire that removes the organic layer, exposes the mineral soil and enables warmer soil temperatures (Purdy et al., 2006).  Exposed mineral soil is generally warmer than soil with a high organic content (Kirschbaum, 1995) and it is considered as a good seedbed for white spruce regeneration (Youngblood and Zasada, 1991; Lieffers et al., 1996). The reduced aspen regeneration on soils with a thicker organic layer and higher moisture is supported by Johnstone et al. (2004) in their study in southwest Yukon. The abundance of aspen regeneration on sites with low soil moisture indicates that it can tolerate moisture stress, something also reported by Lieffers et al. (2001).  The results indicated that white spruce regeneration is poor on north-facing slopes. This is supported by Danby and Hik (2007) who found that the growth of seedlings on north-facing slopes is limited by low soil temperature in southwest Yukon. Similarly, in my study area, both broadleaved species and white spruce occupied sites at medium to lower elevation with gentle 90  slopes and east to southeast aspects or level ground, confirming that they are light and temperature-demanding species (Frey et al., 2003). However, Elliott and Baker (2004) found aspen growing at higher elevations on south-facing slopes, suggesting that this species can survive at higher elevations on south-facing slopes because of the longer growing season and warmer temperatures on these sites. Balsam poplar was found in riparian areas in the study area, confirming its preference for moist sites close to permanent water sources and floodplain areas (Bockheim et al., 2003; Viereck et. al., 1993). Floods can promote balsam poplar regeneration by scarifying shallow roots, thereby promoting suckering (Rood et al., 2007). In this study, the density of mature trees and poplar basal area were important factors for balsam poplar regeneration, which suggests that vegetative reproduction may be the dominant mechanism for regeneration within the study region.   4.5 Chapter conclusions Both multivariate analysis and maximum likelihood models suggested that the distribution and abundance of regenerations of white spruce, trembling aspen and balsam poplar could be explained by disturbances, climatic and edaphic factors supporting my hypotheses that these trends existed in the region.  I found that the regeneration of all three species benefitted from some form of disturbance. Spruce regeneration was favoured by salvage logging following bark beetle infestations. However, regeneration of trembling aspen and balsam poplar were higher than white spruce regeneration in fire-disturbed plots, supporting the hypothesis that broadleaved forest may dominate and prevail for a long time if fire disturbances occur persistently. This will be particularly important if the region faces the drier summers predicted by the IPCC (Christensen, et al., 2007), which would contribute significantly to increasing fire disturbances in 91  the region. Although applying forest management practices under the uncertain climate and disturbance circumstances is a challenging task, applying both pro-active and reactive management practices have been suggested to maintain the regeneration composition of both white spruce and broadleaved species. I would suggest carrying out selection or group harvesting to create gaps in undisturbed plots to enhance regeneration, as these plots had the lowest abundance of regeneration. I would also suggest to carry out salvage harvesting in phases with high retention (more than 30%) to maintain or enhance white spruce regeneration in beetle affected plots.  Similarly, prescribed fire after salvage would be useful to create favourable substrate for the regeneration of all three species.    92  Table 4.1 Mean density (number/ha) of trees and their regeneration by disturbance types  Disturbance Type White spruce regeneration p < 0.001 White spruce tree density  Ratio Trembling aspen regeneration p = .020 Trembling aspen tree density Ratio Balsam poplar regeneration p =.056 Balsam poplar tree density Ratio Beetle plots 142295 1373 1:11 7629 27 1:29 238 0 0:24 Fire plots 163394 230 1:71 106812 18 1:59 13042 1 1:13 Harvested plots 66337 511 1:13 43600 75 1:58 7253 5 1:14 Undisturbed 63273 656 1:96 28331 131 1;22 860 2 1:43 Old fire plots 61743 861 1:72 41924 177 1:24 0 2 2:0 Mean density 99408 726 1:14 45659 85.6 1:53 4278 2 1:21  [Note: The variations of mean regeneration density were significant at p<0.05. The ratio columns indicate the ratio of density of regeneration to density of their parent trees. The table indicates that broadleaved trees had significantly higher ratios compared to white spruce.]   Table 4.2 Axes and their association with environmental data (MCA) (correlation coefficient)     [Note: The table wasgenerated by multivariate canonical correlation analysis. The analysis has classified landscape into four habitats based on disturbance, edaphic and climatic factors.  Abbreviation of names: PTT= Balsam poplar density; PoBA = Balsam poplar basal area, Lat = Latitude. SALV = Salvage harvest, STT = White spruce density, FR = Fire, AM = Average soil moisture, TCC = Total crown cover, Ele = Elevation, TMXsm = Mean summer maximum temperature, DD5= Degree-days above 50C, SA = Sand, OD = Organic layer depth, BT = Beetle affected forests.]    Axis 1 Axis 2 Axis 3 Axis 4 PTT (.9156) STT (-.4883) Ele (-.4580) Ele (-.3133) PoBA (.9923) FR (0.4852) Slope (-.3031) Slope (.3895) Lat (-.1514) AM (-.4142) TMXsm (.4073) DD5 (.3213) SALV (.231) TCC (-.3114) DD5 (.4812) TMXsm (.2968) SA (.134) OD (-.3870) FR (.5213) BT (.2810) 93  Table 4.3 Synthesis of canonical correlation analysis (correlations between the tree regeneration and the canonical axes of environmental data  Axis 1 Axis 2     Axis 3     Axis 4                  Spsed 0.0418         0.1954         0.3562 0.3734 AtSed -0.0362         0.7464 0.1697        -0.0184         Popsed 0.8748 0.0504         0.2375        -0.1172        Atsap -0.0291         0.2757        0.1991         -0.3689 Spsap -0.0208        -0.4111 0.3798   0.3600 Popsap 0.9975 -0.0001        -0.0124        -0.0008         [Note: The table shows the correlations between tree regeneration and habitat factors. It indicates that Habitat 1 and 2 are related to balsam poplar and trembling aspen, respectively, and that Habitat 3 and 4 are related to white spruce regeneration. Abbreviation of names: Spsed = White spruce seedling, Atsed= Aspen seedling, Popsed = Balsam poplar seedling, Atsap = Aspen sapling, Spsap = white spruce sapling, Popsap = Balsam poplar sapling]  Table 4.4 Criteria for assessing goodness of fit Criterion  Balsam poplar Trembling aspen White spruce DF Value DF Value DF Value Deviance 80 30.0744           75 75.4884           81 109.6360           Scaled Deviance 80 30.0744           75 75.4884           81 109.6360           Pearson Chi-Square             80 27.4311           75 27.0187           81 43.1213           Scaled Pearson X2              80 27.4311           75 27.0187           81 43.1213           Log Likelihood                       3157610.9  44002776.3  90016345.7 Full Log Likelihood                      -159.3446  -620.9703  -1072.4911 AIC(smaller is better)                  340.6891  1273.9405  2164.9823      94  Table 4.5 Analysis of maximum likelihood parameter estimates of white spruce Parameter DF Estimate        Standard error Wald 95% confidence  limits Wald  Chi-square  Pr>Chi-sq Intercept 1     -35.8779      21.5492     -78.1135       6.3576           2.77        0.0959 Ele 1       0.0168       0.0051       0.0068       0.0268          10.93         0.0009 Slope 1      -0.0705       0.0127      -0.0954      -0.0456          30.83         <.0001 mat    1 9.5482       0.5310       4.5876      14.5088          14.23         0.0002 Map 1 -0.0222       0.0100      -0.0418      -0.0027           4.96         0.0260 TMXsm 1 5.0724       1.0646       2.9858       7.1591          22.70         <.0001 TMNsm 1      -6.8105       2.3985     -11.5115      -2.1096           8.06         0.0045 Spavdia 1       0.0629       0.0294       0.0053       0.1205           4.58         0.0323 ToBA 1      -0.0421       0.0140      -0.0697      -0.0146           9.02         0.0027 Dispersion 1       1.4336       0.1907       1.0599       1.8073   [Note: Algorithm converged. Stepwise modelling process retained only eight variables that were significant at p<0.05. Abbreviation of names: Ele = Elevation, mat = Mean annual temperature,Map = Mean annual precipitation. TMXsm = Mean summer maximum temperature, TMNsm = Mean summer minimum temperature, Spavdia = Average white spruce diameter, ToBA = Total basal area.]   Table 4.6 Analysis of maximum likelihood parameter estimates of trembling aspen Parameter DF Estimate        Standard error Wald 95%  confidence  limits Wald  Chi-square  Pr> chi-sq Intercept 1 1615.75     350.29     929.1772     2302.32          21.28         <.0001 Map   1 -7.29       1.49     -10.2264      -4.35          23.65         <.0001 DD0 1      -0.98       0.22      -1.4014      -0.56          20.81         <.0001 DD5   1 -0.63       0.13      -0.8903      -0.37          22.72         <.0001 TMXwt 1 -108.07      23.39     -153.904     -62.23          21.36         <.0001 PPTwt 1 15.51       3.16       9.3140      21.69          24.11         <.0001 PPTsm 1 9.72       2.06       5.6829      13.75          22.29         <.0001 BT              1 4.14       2.06       0.0940       8.18           4.02         0.0449 FR 1 5.87       2.41       1.1354      10.59           5.91         0.0151 TD              1 0.12       0.03       0.0600       0.18          15.38         <.0001 OD 1 -0.58       0.15      -0.8803      -0.29          14.75         0.0001 CL              1 -0.16       0.06      -0.2764      -0.05           7.55         0.0060 SA 1 -0.06       0.03      -0.1111      -0.0033           4.33         0.0375 ATT 1 0.01       0.001       0.0020       .0091           09.43         0.0021 ToBA 1 -0.14       0.05      -0.2388      -0.05           9.21         0.0024 Dispersion 1       6.53       1.08       4.4062       8.65   [Note: Algorithm converged. Stepwise modelling process retained only 14 variables that were significant at p<0.05. Abbreviation of names: Map = Mean annual precipitation, DD0 = Number of degree day below 0, DD5= Number of degree above 5, TMXwt = Mean maximum temperature in winter, PPTwt = Mean winter precipitation, PPTsm = Mean summer precipitation, BT = Beetle, FR = Fire, TD = Total soil depth, OD = Organic Depth, CL = % of clay, SA = % of sand, ATT = Aspen density (per ha), ToBA=Total Basal Area per hectare.] 95  Table 4.7 Analysis of maximum likelihood parameter estimates of balsam poplar Parameter DF Estimate        Standard error Wald 95% confidence  limits Wald  Chi-square       Pr> Chi-sq Intercept 1 2320.86    634.83     1076.63     3565.10          13.37         0.0003 Mat 1 -104.54      23.31     -150.24     -58.85          20.11         <.0001 TMXsm 1     -182.00      42.50     -265.30     -98.71          18.34         <.0001 DD5 1 2.11       0.50       1.13       3.09          17.75         <.0001 NFFD 1 -6.86       2.51     -11.77      -1.95           7.50         0.0062 BT 1 17.19       4.63       8.12      26.26          13.80         0.0002 FR 1 22.02       4.86      12.49      31.55          20.49         <.0001 AM 1 0.53       0.13       0.28       0.77          17.74         <.0001 PoBA 1 5.44       1.92       1.68       9.21           8.04         0.0046 ToBA 1 -0.44       0.12      -0.67      -0.21          13.92         0.0002 Dispersion 1 11.06       2.87        5.43      16.69   [Note: algorithm converged. Stepwise modelling process retained only nine variables that were significant at p<0.05 Abbreviation of names: Mat=Mean annual temperature, TMXsm = Mean summer maximum temperature, DD5= Number of degree above 5, NFFD = Number of frost-free days, BT = Beetle, FR = Fire, AM = Average Moisture, PoBA = Balsam poplar basal area, ToBA = Total basal area.]   Table 4.8 AIC and variables dropped in the models Species Full model   AIC values Reduced model  AIC Values Number of insignificant  variables dropped (P >.05) White spruce 2207.3221 2164.9823 29 Trembling aspen 1302.4169 1273.9405 23 Balsam poplar 346.7387 340.6891 28  [The AIC values of reduced models are smaller than full models, which signifies the goodness of fitness of the models]  Table 4.9 Chi-square tests of the models Species Log likelihood of  final model (l) Log likelihood null  model (l0) F calculted value  2*(l-l0) F table value  at p< .05 White spruce 90016345.79 90016322.49 46.59 15.51 (df =8) Trembling aspen 44002776.31 44002740.60 71.42 23.69 (df = 14) Balsam poplar  3157610.9 3157585.21 25.7 16.92 (df = 9)  [The tests showed that the calculated F value is higher than the tabulated value for all the species indicating that the models are statistically significant at p<0.05]  96  Table 4.10 Imputed and observed regeneration abundance for the first and last six plots  Observation White spruce seedling imputed Aspen seedling imputed Poplar seedling imputed White spruce seedling observed Aspen seedling observed Poplar seedling observed 1 933        0 0 21733 0 0 2 40133 15867 0 14000 13467 0 3 106667 0 0 6800 166267 0 4 248400     113200 0 84000 260133 0 5 6933 13600 0 17200 24933 0 6 13333 147733 0 40133 15867 0 85 17200 24933 0 10667 23600 0 86 17200 24933 0 6933 13600 0 87 133 0 0 20000 13333 0 88 6800 10000 0 60133 16667 0 89 40133 15867 0 73333 14733 0 90 3333 0 0 13333 36667 0 [Note: the table shows the top six and bottom six observed values and imputed values. The correlation between observed and imputed regeneration abundance of white spruce, trembling aspen and balsam poplar were 0.618, 0.656, and 0.874, respectively; suggesting imputed and observed values were highly co-related.]  97   Figure 4.1 Responses of white spruce regeneration (number/ha) to environmental factors  [Note: Scatter plots indicating that white spruce responded negatively to increasing slope, elevation and soil moisture and positively to increasing summer temperature and number of growing days at significance level p<0.05. Means of plots (ANOVA; p<0.05) indicated regeneration variation by disturbance regime with its higher abundance in fire and beetle plots. Abbreviation of names:  Beetle affected forests (B) Recent fire F), Salvage harvest (S); No disturbances (N) and Old fire (OF)]   98   Figure 4.2 Responses of trembling aspen regeneration (number/hectare) to environmental factors [Note: Scatter plots indicating that trembling aspen responded negatively to increasing mean annual temperature, total crown cover and organic depth layer at significance level p<0.05. Means of plots (ANOVA; p <.05) indicated higher abundance of regeneration in loamy to silty loam soil on eastern aspects and in fire plots. Abbreviation of names:  Beetle affected forests (B), Recent fire (F), Salvage harvest (S), No disturbances (N) and Old fire (OF). Clay (CL), Loamy (L),  Loamy sand (LS), No texture (NT), Organic (O), Sandy (S),  Silty clay (SiC),  Sandy loam (SL).Silty loam (SiL), Beetle affected forests (B) Recent fire F), Salvage harvest (S);  No disturbances (N), Old fire (OF), East (E), Level (L), North (N), Northeast (NE), Northwest (NW), South (S), Southeast (SE), Southwest (SW) and West (W).]   99   Figure 4.3 Responses of balsam poplar regeneration (number/hectare) to environmental factors  [Means of plots (ANOVA; p<.05) indicated balsam poplar?s higher abundance in level or south-eastern facing sites, in sandy to silty clay soils and in fire plots. Abbreviation of names: Beetle affected forests (B), Recent fire (F), Salvage harvest (S), No disturbances (N) and Old fire (OF). Abbreviation of names:  Clay (CL), Loamy (L),  Loamy sand (LS), No texture (NT), Organic (O), Sandy (S),  Silty clay (SiC),  Sandy loam (SL).Silty loam (SiL), Beetle affected forests (B) Recent fire F), Salvage harvest (S);  No disturbances (N), Old fire (OF), East (E), Level (L), North (N), Northeast (NE), Northwest (NW), South (S), Southeast (SE), Southwest (SW) and West (W).]   100   Figure 4.4 Relationship between imputed and observed values  [Note: The straight line in the figures indicates a correlation between imputed and observed values. For correlation coefficients close to 1, all the values would be along the line. Abbreviation of names: AtsedTot = Aspen seedling total; SpsedTot = White spruce seedling total; PosedTot = Balsam poplar seedling total.]    0e+00 2e+05 4e+050e+003e+056e+05ImputedObservedSpsedTot0e+00 2e+05 4e+050e+004e+058e+05ImputedObservedAtsedTot0 20000 40000 60000040000100000ImputedObservedPosedTotya1101  5. Vegetation diversity in response to landscape variability, disturbances and community dynamics in Southwest Yukon  5.1 Introduction The southwestern region of the Yukon Territory, Canada, has experienced an unprecedented spruce bark beetle outbreak and an increase in forest fires; both are greater than historical trends. Such disturbances can cause significant impacts on ecosystems by shaping plant community structure and composition (Johnstone and Chapin, 2006b), and can have major implications for biodiversity and sustainable forest management (Nilsson and Ericson, 1997). A Strategic Forest Management Plan (SFMP, 2004) for the Champagne and Aishihik Traditional Territory (CATT) has been implemented since 2004 in response to the spruce bark beetle infestation in the region. The plan promoted the use of salvage harvesting as a short-term measure to achieve long-term goal of sustainable forest management. One of the goals of the SFMP is to ?maintain ecosystem functioning? in the region, with the specific objectives being to maintain, restore or enhance forest succession, species and ecosystem diversity. This requires an understanding of the plant communities in the region. Such an understanding is still lacking, and this chapter presents the research that was undertaken in an attempt to fill this gap in knowledge.   5.1.1 Biodiversity and its importance  Hebda (1998; p. 195) defines biodiversity as ?...a living legacy with a unique history developed in situ; not simply the number of species, not simply a collection of genes kept in a zoo, or 102  frozen in a bottle, and not simply rarity or strangeness?. According to Chapin et al. (2000) species diversity can be described by the number of species present (species richness), their relative abundances (species evenness), species composition (proportion of each species) and temporal and spatial variations in the above properties. Haila and Kouki (1994) refer to biodiversity as the overall heterogeneity in nature that is a necessary property of an ecological system. Similarly, Palik et al. (2002, p. 384) define ?biodiversity as the variety and spatial patterns of physical structures, processes, species, and genotypes in a forest?. Both species and genetic diversity generally depend on structural and process diversity (Franklin, 1988).  Species diversity influences ecosystem processes and functions and thus has important implications for the long-term sustainability of forests (Chapin et al., 2000). It plays an important role in ecosystem functions such as primary productivity, carbon storage, nutrient cycling and water retention, all of which are vital for ecological stability, determining for example, resistance or resilience after disturbances (Peterson et al., 1998; Naeem, 1998; Tilman et al., 1996; Bengtsson et al., 2000).  Increased species richness also increases the efficiency and stability of ecosystem functions (Tilman, 1996; Peterson et al., 1998). It is particularly important during succession, as the higher diversity ensures that enough species survive or persist to maintain ecological functions if a species disappears from the system due to a large-scale disturbance and may as well prevent successional retrogression (Folke et al., 1996; Chapin et al., 2000). Rich diversity increases the probability of including more disturbance-resistant species and can compensate for the loss of vulnerable species (Tilman, 1996; Hughes et al., 2007). Species diversity also reduces the probability of insect outbreaks by diluting the availability of their host species (Chapin et al., 2000).  103  Biodiversity also affects the distribution and abundance of wildlife in boreal forests as much of the vegetation provides important habitat and food for animals such as moose (Alces alces), wood bison (Bison bison ssp. athabascae), grizzly bear (Ursus arctos ssp. horribilis), American marten (Martes americana), Canadian lynx (Lynx lynx), snowshoe hare (Lepus americanus), woodland caribou (Rangifer tarandus ssp. caribou), red fox (Vulpes vulpes), and muskrat (Ondatra zibethicus). Some of these species have important socio-economic significance for northern communities (Mulder et al., 2000).    5.1.2 Importance of studying vegetation diversity in SWY Several studies have been undertaken on the distribution and ecology of vegetation in the Canadian boreal forest and within Yukon Territory. Some of the more comprehensive studies include La Roi (1967), Hult?n (1968), Murray (1971), Douglas (1974), Wali & Krajina (1973), Douglas and Vitt (1976), Stanek et al. (1981), Bornor and Oswald (1985), Oswald and Brown (1986), Looman (1987), Meideniger and Pojar (1991), Johnson et al. (1995), Scudder (1997) and Cody (2000). The distribution of vegetation in Yukon typically follows temperature gradients along both latitudinal and elevation gradients, creating a complex mosaic of vegetation types (Hughes et al., 1983; Douglas and Vitt, 1976). The characteristics and composition of the vegetation vary with climate, aspect, slope, soil type and moisture gradient (Meideniger and Pojar, 1991). The dominant vegetation types in the Yukon are arctic tundra, alpine tundra, taiga or subarctic forest, boreal forest and subalpine-shrub forest (Scudder, 1997). Arctic tundra is the dominant vegetation type in the low arctic Yukon Coastal Plain, which is treeless but with a continuous 104  vegetative cover (Scudder, 1997). The subalpine zone is dominated mainly by the spruce-willow-birch ecosystem (Meidinger and Pojar, 1991). Deciduous communities dominated by species such as Scouler?s willow (Salix scouleriana Barratt), trembling aspen and common bearberry (Arctostaphylos uva-ursi (L.) Spreng.) are the results of disturbances such as fire and occur throughout the region (Douglas and Vitt, 1976).  The boreal understory vegetation plays an important role in ecosystem functioning by influencing canopy succession (Messier et al., 1998), nutrient cycling (Weber and Van Cleve, 1981), providing wildlife habitat (Okland and Eilertsen,1996) and providing many valuable foods and medicines to human society (Penafiel et al., 2011). However, forest management generally considers the overstory composition and structure, giving least priority to understory vegetation diversity (Hart and Chen, 2006). Besides maintaining ecological function, the biodiversity of the boreal forest provides many social, economic and cultural services for local people (Chen and Popadiouk, 2002). Although timber harvesting is considered to be economically profitable in a quarter of the boreal forest in Canada (CCFM, 2006), the shrub vegetation (non-timber forest products, NTFPs) provides many socio-economic and cultural benefits to the people. Working in Yukon, Holloway and Alexander (1990) found that 48 species of native plants were used by local communities as medicines (40%) and as food or beverage (56%). Wild fruits and mushrooms comprised the greatest proportion of edible plants. According to Holloway and Alexander (1990), mountain cranberry (Vaccinium vitis-idaea L.) accounts for 90% of the berries and mushrooms account for 70% of the food harvested in the Porcupine River region of Yukon. Duchesne et al. (2000) estimated the total value of NTFPs to be equivalent to $240 million (Canadian dollar) in 1997, although this figure did not include locally traded and used products. Wild berries and other wild 105  food accounted for $21 million annually. It has been estimated that the potential for NTFPs harvest in Canada is about $1 billion, which could provide viable economic opportunities for many rural communities (Mohammed, 1999).  5.1.3 Research Issues: impacts of landscape variability, climate and disturbances on vegetation diversity A landscape is an ecological unit with a unique or distinguishable character and structure where its components or units interact with each other as an ecological function (Forman and Godron, 1981). Landscapes are constantly changing due to geomorphologic processes, climatic change, human interference and disturbances (Noss, 1983; Bengtsson et al., 2000), which is one of the main reasons for biodiversity change (Noss, 1983). Change may influence diversity both negatively and positively. Bengtsson et al. (2000) suggested however that there is little or limited research on the effects of large-scale landscape changes on diversity, as most research has been conducted at relatively small scales, which may not be relevant for predicting large-scale dynamics of ecosystems. Community dynamics resulting from disturbances are natural features of the boreal forest (Niemela, 1999). Disturbances can have major implications for biodiversity and sustainable forestry (Nilsson and Ericson, 1997). Disturbances create a mosaic of structures and processes that influence the variety and diversity of species (Palik et al., 2002). Studies in the boreal forest also suggest that the frequency and severity of disturbances can affect forest ecosystems by influencing vegetation dynamics, diversity patterns and ecosystem processes (Wardle et al., 1997). As most forestry operations such as logging involve some kind of disturbance that can 106  affect both diversity and ecosystem function, the understanding of natural disturbances and disturbance dynamics on diversity has become an increasingly salient issue for forest managers (Haila et al., 1994; Angelstam, 1997; Bengtsson et al., 2000). Therefore, it is important to understand the relationship between disturbance dynamics and biodiversity if forest management practices are to conserve and sustain biodiversity (Hughes et al., 2007). According to the intermediate disturbance hypothesis (IDH), local diversity is maximized when disturbances are neither too frequent nor too rare. In other words, higher diversity is maintained when the intensity and frequency of disturbances are intermediate (Grime, 1973; Connell, 1978). The hypothesis states that species diversity is low when there are frequent disturbances as only a few tolerant species can survive, or when there is no disturbance so that the ecosystem is dominated by a few, highly competitive species. Frequent disturbances can lead to the creation of bare patches, such that competitive exclusion does not occur in the landscape (Connell, 1978; Wilson, 1994). Climate change may impact forest biodiversity by affecting species interactions and ecosystem processes (Spittlehouse, 1996). It may change both the biodiversity of an area and the stand structure and species composition (Hebda, 1998). Climate change may influence biodiversity directly by affecting individual organism, population, species distribution and ecosystem composition and function, and/or indirectly by the modification, loss or fragmentation of species habitats, increased frequency and severity of disturbances and introduction of non-native species (IPCC, 2002). More than 40% of Canada?s protected areas are projected to be affected by biome change as a result of climate change (Lemieux and Scott, 2005). Similarly, community dynamics and their relationship with biodiversity are also an important element. Variation in species diversity can be both a cause and a consequence of 107  variation in community dynamics (Loreau et al., 2001). It is therefore important to understand how community properties contribute to the maintenance of biodiversity (Huston, 1979; Hughes et al., 2007). Similarly, the amount of coarse woody debris may play a significant role in determining the presence or absence of some species, especially in boreal forests where many species are dependent on the density of dying and dead trees (Esseen et al., 1997; Nilsson et al., 2001).  5.1.4 Research problems, hypotheses and questions Plant diversity is an important aspect of forest management, both for maintaining, restoring or enhancing forest succession, species and ecosystem diversity and to fulfil the socio-economic needs of people. This has been recognized in the CATT (SFMP, 2004). However, a full understanding of the processes involved is still lacking in the region, which sets the scene for the major research questions asked in this chapter. I examined the variation in vegetation diversity in response to bio-geoclimatic factors and the disturbance regime in the CATT. My aim was to increase the knowledge of vegetation diversity in the region in order to help forest managers make more informed management decisions, thereby enabling the conservation and management of biodiversity. I hypothesize that:  (i) Canopy gaps created by harvesting and forest disturbances will increase vegetation diversity.  (ii) Vegetation diversity will increase with increasing stand structure and decrease with increasing stand density. 108  (iii) The highest diversity will occur at intermediate levels of disturbance (in terms of severity) sensu the IDH (Connell, 1978). To address the above hypotheses and to understand the overall diversity pattern in the study area, this chapter deals with the following broad research questions; ? What is the overall diversity and distribution of vegetation in the study area? ? What stand structure and disturbance factors affect the variation of vegetation diversity in the study area? ? How do vegetation patterns vary with edaphic and climatic factors? ? Are the results of different diversity indices consistent?  5.2 Data collection and analysis Vegetation data were collected from 90 randomly selected plots as explained in Chapter 2. Species were identified using local flora (Cody, 2000; Johnson et al., 1995). Specimens of each species were collected and photographs were taken to verify genus and species at the lab.  However, despite my best efforts, some plants could not be identified to the species level and are therefore listed only at the genus level.  For data analysis, as a single index of diversity or evenness may result in an incorrect interpretation as some communities cannot be ranked by diversity or evenness (Taillie, 1979; Ricotta, 2003), complementary techniques have been suggested to produce diversity and evenness profiles (Kindt et al., 2006). In this case, vegetation diversity has been assessed using both alpha and beta diversity indices. Species richness, abundance and evenness were calculated. Similarly, Simpson and Shannon-Wiener indices were calculated for each of the 90 sites (alpha 109  diversity) as well as for the variation along the environmental gradient (beta diversity). Species ranges and distribution were also calculated. Detailed descriptions of each method are provided in the following section.   5.2.1 Richness, abundance and evenness analysis Species richness, abundance and evenness were calculated since these are common indicators of diversity (Kindt et al., 2006). Richness is a measure of the total number of different species in a particular area or in a community. Abundance is the total number of all the species in an area, and is expressed as a total count or percentage of cover. Evenness is a measure of the relative abundance of the different species in proportion to the total species assemblage in an area, which gives homogeneity or relative diversity (Kindt and Coe, 2005). Evenness is considered as important as species richness. Evenness has a more direct relationship with total forest productivity than species richness (Naeem et al., 1994) and a reduction in evenness may negatively affect plant productivity (Wilsay and Potvin, 2000).  5.2.2 Diversity indices Shannon-Wiener index, Simpson index and Renyi?s profiles were calculated; these are briefly described in the following sections.    110  5.2.2.1 Shannon-Wiener and Simpson indices Shannon-Wiener and Simpson indices have been frequently used for assessing species richness (Shannon and Weaver, 1949; Simpson, 1949). The Shannon-Wiener index has been widely used for comparing diversity between different habitats (Clarke and Warwick, 2001) and has been used for both categorical and continuous variables (McRoberts et al., 2008). The Simpson index is the preferred method for comparing diversity between different species assemblages because of its ability to rank assemblages consistently when sample size varies (Magurran, 2004). It is also preferred to the Shannon-Wiener index due to its better performance as a general purpose of diversity index and as a measure of evenness (Magurran, 2004). The Simpson index is considered a dominance index as it is weighted towards the abundance of the most common species while being less sensitive to species richness (Whittaker, 1972; Smith and Grassle, 1977).  Although the Shannon-Wiener index is less dependent on the dominant species than the Simpson index (see below), it gives little weight to rare species (Zahl, 1977). The Shannon-Wiener diversity index (H') is calculated as (Shannon and Weaver, 1949):                      ) ???..??Equation (5-1)   where, H? is the Shannon-Wiener diversity index and Pi is the proportion of each species (N). The value ranges from 0 to Ln (N). The value of the Shannon-Wiener diversity index is usually between 1.5 and 3.5 and only rarely surpasses 4.5.  The Simpson index can be calculated by (Simpson, 1949):                     ??????.???.Equation (5-2)  111  where H is the Simpson index, n is the numbers of individuals in each species, and N is the total number of individuals. The Simpson index is expressed either as H (where higher value means lower diversity) or 1-H or 1/H (where higher values means higher diversity) (Kindt and Coe, 2005).  5.2.2.2 Renyi?s profile A single diversity index such as the Shannon-Wiener Index or the Simpson Index can provide misleading results (Kindt et al., 2006). As a result, Renyi?s profiling was used as a complementary method. This is a direct method for comparing diversity and evenness (Ricotta, 2003). Renyi?s profile provides many kinds of indices for species richness and diversity including modified Simpson and Shannon-Wiener indices (Kindt and Coe, 2005). Tothm?r?sz (1995) suggests that the Renyi?s profile is one of the most useful methods for diversity ordering. In addition, as the sample size significantly affects the diversity index (Kindt et al., 2006); Renyi?s values can reduce the sensitivity to sample size (Magurran, 1988). Renyi?s profile is calculated as:                     ??????Equation (5-3)   where    is Renyi?s diversity profile and Pi is the proportion of each species. It provides eight different   values (varying from 0 to infinity); each gives different index. For example alpha = 0 provides the logarithm of species richness, alpha = 1 provides Shannon-Wiener diversity index and alpha = infinity provides the information on the proportion of the most abundance species (Kindt and Coe, 2005). 112  The evenness profile can be derived from the diversity profile and orders communities in evenness (Smith and Wilson, 1996; Kindt et al., 2006). The evenness profile is calculated as:               ????????..?.Equation (5-4)  Renyi?s profile was used in this study to assess the variation of species richness and evenness of the communities along the disturbance, environmental (slope, elevation, climatic variations, moisture etc.) and ecological gradients (tree density, crown cover and regeneration density).  5.2.2.3 Diversity curves  Species accumulation curves and Renyi?s profiling curves were used in this study to show variation in diversity at the landscape-level. Species accumulation curves show the species richness for combinations of sites or pooled species richness (Kindt and Coe, 2005). The curves indicate how many new species are added when one new site is added to the analysis. It also provides the average standard deviation of species richness with each new site added. Species accumulation curves are very useful for comparing species richness for subsets in the data when sample sizes of subsets are different.  Renyi?s profiling quantifies the variation of diversity and species richness along an environmental gradient. Each assessment provides H-alpha and E-alpha metrics for showing diversity and evenness of the communities, respectively. The curves that start and end at higher points indicate higher diversity or evenness of the communities (Kindt and Coe, 2005). One of the advantages of Renyi?s profile is that sites can be ordered from higher to lower diversity.  113  As suggested by Kindt and Coe (2005) and Kindt (2004), Biodiversity-R (R, 2005) and the Vegan package (Oksanen et al., 2005) software programme were used for the data analysis. As required by the Biodiversity-R programme, the community and environmental data were separately prepared in Excel sheets as community, environmental and community stacked data.   5.2.3 Species spatial distribution and range The vegetation range in the landscape was analyzed by showing the presence or absence of particular species in each block. The range table also shows the abundance of each species in each block. This value indicates the type of habitat that each particular species may prefer. Similarly, distribution and abundance of understory vegetation by disturbance type were also assessed in order to identify the dominant and unique species related to specific disturbance type.    5.2.4 Canonical correspondence analysis (CCA) As environmental variables are often highly correlated, it is better to see the combined effects than their independent effects (Ter Braak and Verdonschot, 1995). When the number of influential environmental variables is more than two or three, it becomes increasingly difficult to infer results for several species together. Canonical correspondence analysis (CCA) is designed to fulfil this need by analysing and visualizing the relationships between many species and many environmental variables (Ter Braak and Verdonschot, 1995). Canonical ordination techniques were developed to detect the patterns in variation in species data that can be best described by a combination of environmental variables (Jongman et 114  al., 1995). This method was introduced by Ter Braak (1986) and has been widely used in ecological analyses (Birks et al., 1996). CCA is considered as a simple ordination method for arranging species along environmental variables, constructing linear combinations of environmental variables along which the distributions of the species are maximally separated (Ter Braak, 1986). Its application is mainly for detecting species-environment relations and for investigating specific questions about the response of species to environmental variables (Ter Braak, 1986).  CCA was used to detect the relationship between species distribution and combinations of environmental variables. Out of 69 species recorded, the 25 most widely distributed species (species covering at least 1% in average in each plot or more than 90% in 90 plots) were considered for the analysis. CCA has also been used to detect the relationship between environmental variables and major diversity indices such as the Shannon-Wiener and Simpson indices, and species richness and abundance, as it may provide information on how those indices are affected by or related to environmental variables.   5.2.5 Diversity index mapping A Geographical Information System (ArcGIS, 2009) was used to map Shannon-Weiner indices in the study area. Fisher Discriminant Analysis (Fisher, 1936) was used to develop a model for Shannon-Weiner indices against independent variables; slope, elevation and aspect. The model output was used to extrapolate the indices over the raster pixels of the landscape using ArcGIS (2009).  115  5.3 Results 5.3.1 Characteristics and severity of disturbed plots The disturbed plots varied significantly in terms of moisture content, depth of organic layer and crown cover (Table 5.1). The analyses of these variations were important as they indicated the disturbance characteristics and severity.  With some exceptions, the beetle-affected forests that were not harvested or affected by fire had the highest crown cover (= 63.93%; sd = +/-17.5), soil moisture (= 26%; sd = +/-11.8) and soil organic depth (= 11.29 cm; sd= +/- 9.5). The old fire plots were mainly comprised of mixed stands of white spruce and trembling aspen with an average of 48% (sd = +/- 24.4) crown cover, an average soil organic layer depth of 8.2 cm (sd = +/-6.2)  and 25% (sd = +/-11.8) soil moisture. The most recent large fire occurred in 1998 in a portion of the research area, severely impacting both forest crowns and soils. These plots were affected by spruce bark beetles prior to the occurrence of fire.  These recent fire plots had low average crown cover (=12%; sd = +/-25.5), organic depth (= 2.75 cm; sd = +/-2.09) and soil moisture (= 19%; sd = +/-11).  In response to these disturbances, the salvage harvesting of beetle-affected stands commenced in 2006 and has had noticeable impacts on both soil and understory vegetation, with harvested plots having lower average organic depth (= 5.2 cm; sd = +/-4) and soil moisture (= 18.6%;  sd = +/-5.3). Undisturbed plots had the thickest organic depth (= 8.98 cm; sd = +/-7.8) and moisture content (= 19.4%; sd = +/-10.8).   116  5.3.2 Species richness, abundance and spatial distribution In the study area, 63 genera and a minimum of 69 species were recorded (Table 5.2) of which four species were trees, 47 species were shrubs and herbs, 10 species were bryophytes and eight species lichens (Appendix A).  The average richness in each plot was 13.8 (~ 14), with maximum and minimum richness values of 21 and 4, respectively. Similarly, the average cover in the subplots was 114.7% (~115%), with maximum and minimum abundance of 18% and 269%, respectively. The abundance and richness data for each site are provided in Appendix F. Plot numbers 404, 106, 409, 502, 503 and 613 had the highest numbers for richness, with more than 20 species recorded in each plot. Similarly, 56 plots had more than 100% cover, with more than 200% cover in plot numbers 216, 217, 220 and 504 (appendix F). With the exception of research blocks C and F, which had an average abundance of about 80%; the blocks had more than 100% cover (Appendix H-1). The Shannon-Weiner indices have been plotted in the landscape using GIS (Figure 5.1) which shows higher diversity in research blocks four (recent fire), five and six (salvage harvested). However, the correlation analysis indicates that the richness and abundance data are weakly correlated (r = 0.25). Species distribution and abundance comparisons among the research blocks are provided in Appendix G. The five most abundant higher plants (trees, shrubs and herbs) were willow (Salix spp.), white spruce (Picea glauca (Moench) Voss), bearberry (Arctostaphylos uva-ursi (L.) Spreng. s.l.), and soapberry (Shepherdia canadensis (L.) Nutt.), trembling aspen (Populus tremuloides, Michx.) with total proportionate covers of 9.86%, 8.17%, 7.3% 6.54% and 4.5% respectively. Similarly, common dandelion (Taraxacum officinale Weber ex Wiggers), common juniper (Juniperus communis, L.s.l), black snakeroot (Sanicula marilandica L.), yellow monkey 117  flower (Mimulus guttatus DC.), hook-spur violet (Viola adunca J.E. Smith) and pink-flowered wintergreen (Pyrola asarifolia Michx.) were among the least abundant herbs and shrubs, with less than 0.05% proportionate cover.   Among the bryophytes and lichens; red stemmed feather moss (Pleurozium schreberi (Brid.) Mitt.) was highly abundant with more 12% coverage and broom moss (Dicranum scoparium Hedw.) and glow moss (Aulacomnium palustre (Hedw.) Schwagr) and sugary beard lichen (Usnea hirta (L.) F.H. Wigg.) were the least abundant in the study area. In terms of the spatial distribution of the species, red stemmed feather moss (Pleurozium schreberi (Brid.) Mitt.), Salix spp., Picea glauca (Moench) Voss, Arctostaphylos uva-ursi (L.) Spreng. s.l., Shepherdia canadensis (L.) Nutt., pixie cup lichen (Cladonia spp.), curled snow lichen (Flavocetraria cucullata (Bellardi) Karnefelt & A. Thell), freckled lichen (Peltigera aphthosa, (L.) Willd.), prickly rose (Rosa acicularis Lindl. s.l.), and Indian paint-brush (Castilleja spp.) were recorded in all blocks, indicating their wider distribution in the research area. In contrast, meadow barley (Hordeum brachyantherum Nevski), clear moss (Hookeria lucens Hedw. Sm.), fescue (Festuca spp.), cranberry (Oxycoccus microcarpus Turcz.), Taraxacum officinale Weber ex, broom moss (Dicranum scoparium Hedw), hook-spur violet (Viola adunca J.E. Smith), glow moss (Aulacomnium palustre (Hedw.)Schwagr.), sugary beard lichen (Usnea hirta (L.) F.H.Wigg.) and common juniper (Juniperus communis L.s.l.) were recorded from only one research block, indicating their limited distribution in the research area.  The species accumulation curve (Figure 5.2) rises steeply up to pool of 40 sites that includes almost 60 plant species while the remaining 50 sites added only 12 more species, flattening the curve out. This indicates that about 50% of sampling plots could accommodate 80% of the vegetation diversity in the study area.  118  5.3.3 Vegetation diversity in response bio-geoclimatic and disturbance factors The mean richness, mean abundance, Shannon-Wiener and Simpson indices were calculated and  the results of the indices are provided in the tables within Appendix H. Variation in vegetation diversity was assessed using Renyi?s profile independently by eco-geographical gradient (blocks, slope, aspect, elevation and soil moisture), disturbance type, stand dynamics (stand density, stand type and crown cover) and climatic factors (mean annual temperature, mean annual precipitation and number of growing degree-days > 5 ?C) Renyi?s profile (Figure 5.3) provides variation in diversity by eco-geographical gradient and disturbance types. Figure 5.3 does not reveal a clear variation in vegetation diversity with aspect, although the Shannon-Wiener index, which gives more weight to species richness, was higher on eastern aspects and, conversely, the Simpson Index was higher on northern aspects (Appendix H-4). Vegetation diversity was higher at low elevations and on flat ground and lowest on southwest aspects and steep slopes. The Shannon-Wiener and Simpson indices provided similar results. Indices were higher at lower elevations, on flat areas and on northwest, north and east aspects.   Renyi?s profile also illustrates the variation in species diversity by research blocks, disturbances and moisture gradient. The results suggested that diversity was higher in the harvested Block E, which has the driest soil moisture regimes, and lower in beetle-affected stands, which had higher soil moisture. The blocks with the lowest diversity were D and A, which were not differentiable. The evenness profile also showed a similar result to that of the diversity analysis.  The Renyi?s profile did not reveal a clear variation in vegetation diversity with disturbance types as index lines cross each other. However the Shannon-Weaver and Simpson 119  indices (Appendix H-5) indicated higher diversity in harvested plots (Shannon-Weaver Index = 2.16; Simpson Index = 7.19). The median diversity was found in fire plots (Shannon-Weaver Index = 196; Simpson Index = 5.5). The least diversity was found in old fire plots (Shannon-Weaver Index = 2.16; Simpson Index = 7.19), in spruce bark beetle plots (Shannon-Weaver Index = 1.84; Simpson Index = 4.67) and undisturbed plots (Shannon-Weaver Index = 1.79; Simpson Index = 4.85). Both harvested and fire plots were characterized by low crown cover, lower moisture content, thinner organic depth layer and the lowest shrub cover (Table 5.1). These findings indicate that medium soil disturbances (harvested plots in this study as indicated in section 5.3.1) favour higher vegetation diversity compared to highly disturbed (fire plots) or less/non-disturbed areas, suggesting that the diversity pattern is consistent with that expected under the Intermediate Disturbance Hypothesis (IDH).    Figure 5.4 provides the variation in diversity by climatic factors and stand dynamics. Diversity was higher in stands with lower stand density, lower crown cover and in mixed stands. The lowest diversity was recorded in stands with high stand density, high crown cover and in stands dominated by trembling aspen. Renyi?s profile matched with Shannon-Wiener and Simpson indices for crown cover. The variations among some of the climatic factors were mostly unclear as in most cases the curves intersect each other. Diversity was higher with increasing mean summer maximum temperature, higher mean annual temperature and precipitation and lowest in plots with colder temperatures and lower annual precipitation. The Shannon-Wiener and Simpson indices provided the similar results.   Table 5.3 shows the top 10 most abundant plants. Distributions of these species varied significantly with disturbance type.  Nine species were restricted to present in specific types of disturbance. Arctostaphylos rubra (Rehd. & Wils.) Fern. and Peltigera canina (L.) Willd. were 120  only present in beetle-affected plots. Similarly, Betula glandulosa Michx., Dicranum scoparium Hedw., Epilobium angustifolium L.s.l., and Equisetum arvense L. were only present in recent fire plots. Festuca spp. and Alnus incana (L.) Moench were only present in harvested and undisturbed plots. In contrast, Cladina mitis (Sandst.) Hustich, Pleurozium schreberi (Brid.) Mitt. and Shepherdia canadensis (L.) Nutt. were absent from specific disturbance types namely old fire, recent fire and beetle affected plots respectively. Out of these 10 species, six of them are locally used species (Table 5.3); implying disturbances may affect availability of these species.   5.3.4 Canonical correspondence analysis (CCA) The CCA results were similar to the results described in the previous section, although the analysis did provide some new information. It indicated how the independent variables are related in terms of vegetation distribution and diversity indices. The full results are provided in Appendix I. Most of the species were clustered around the centre of two axes (Figure 5.5), indicating no major influences of particular environmental variables. However, a few species showed a strong relationship to sandy soil, increasing density of trees, increasing slope and increasing degree days below and over 5 ?C. The species distribution in Figure 5.6 indicates that willow and white spruce are clustered in the centre indicating they are widespread at most of the sites. White spruce showed a similar trend in the multiple co-inertia analysis in Chapter 3. trembling aspen, Michx., Shepherdia canadensis (L.) Nutt., and Arctostaphylos uva-ursi (L.) Spreng. s.l. were centered in one cluster, indicating that they prefer similar environmental conditions. Similarly, Labrador-tea (Ledum groenlandicum Oeder), dwarf blueberry (Vaccinium caespitosum Michx.), crowberry (Empetrum 121  nigrum L.) and western rye grass (Elymus glaucus Buckl.) are on the same axis, indicating their similar environmental requirements. By overlapping the information in Figure 5.5 and 5.6, it can be inferred that these species mostly prefer sites that have a higher slope position, higher mean annual temperature and higher number of frost-free days.  5.4 Discussion 5.4.1 Variation in vegetation diversity by bio-geoclimatic and disturbance factors 5.4.1.1 Canopy gaps vs. vegetation diversity The findings support my hypothesis that higher diversity is associated with canopy gaps in the study area. Diversity was higher in harvested and fire plots, which had 29% and 12% canopy cover, respectively, the lowest of any of the disturbance types. Higher understory diversity in harvested areas was also observed by Whittle et al. (1997) and Bergeron and Dubuc (1989). The higher diversity in these blocks (with larger canopy gaps compared to other blocks) is likely due to exposure to sunlight as light is considered to be the most limiting resource affecting understory vegetation establishment and considered to be one of the most important environmental factors (Chen et al., 1996; Legare et al., 2002). However, the understory vegetation may change over time as the overstory vegetation changes, and light availability to the understory can decline to less than 15% (Okland et al., 2003), causing a shift in understory vegetation from shade-intolerant to shade-tolerant species (De Grandpr? et al., 1993). The increased abundance of white spruce seedlings under the closed canopy of aspen stands is one of the examples of the type of vegetation shift described in Chapter 3. Disturbances that create canopy gaps will likely increase diversity but will be followed by reduced diversity once the 122  canopy cover increases. As a result, the highest diversity can be expected to decline after crown closure, which has been shown to occur within 40 years of a disturbance (Hart and Chen, 2006).  My findings that vegetation diversity was higher in stands with in low stand density support my second hypothesis. The low diversity measured in beetle-affected sites may be due to the higher crown cover, tree density and presence of Pleurozium schreberi (about 32% coverage, Table 5.2) in these plots. The average stand density in all plots was 813 stems per hectare while it was 1306 stems per hectare in beetle-affected plots. Similarly, the crown cover was 60% in beetle-affected stands, which was higher than the average in all other plots (41%). The relationship between stand characteristics and vegetation diversity has been recognized by various researchers. Diversity can vary by stand dynamics and vice versa (Loreau et al., 2001; Schmid, 2002), such as when the addition of a new species in a community increases species richness by one, but finally leads to decreases in local diversity due to competition (Bruno et al., 2005; Hughes et al., 2007).  Vegetation diversity also varied by stand type as higher understory diversity was found in mixed stands of aspen and white spruce. Mixed overstory stands may support higher understory diversity than pure conifer and hardwood stands because these stands are structurally more diverse due to tree species-specific growth habits; this creates a more heterogeneous light environment, higher resource heterogeneity, and allow resource-demanding vascular plants to coexist with shade-tolerant species (Hart and Chen, 2006 & Saetre et al., 1997).  This finding also supports my second hypothesis that understory diversity increases with stand structural diversity. By contrast, Qian et al. (2003) found no difference in understory diversity between aspen and black spruce stands in British Columbia. They did not provide any specific reason for this finding indicating that the results may vary from one site to another.  Reich et al. (2001) 123  reported a higher diversity in aspen stands than in spruce stands in southern boreal forest of Minnesota, USA. The aspen stands in those sites were established after a logging operation, which was the main reason for higher understory vegetation.    5.4.1.2 Disturbances vs. vegetation diversity The highest diversity in harvested plots and the lower diversity in highly disturbed plots (fire plots) and non-disturbed or spruce bark beetle plots can be explained by the IDH (Connell, 1978), which states that higher diversity should occur at intermediate levels of disturbance and low diversity in non-disturbed or frequently disturbed areas.  The thickness of the organic layer has been considered as an indicator of soil disturbance severity after fire, as found by Johnstone and Chapin (2006a) in their study in interior Alaska and central Yukon. A lower organic thickness indicates a greater effect of the disturbance. Accordingly, I infer that the recent fire plots had the highest disturbance severity (organic depth = 2.75 cm10) followed by salvage harvested plots (organic depth = 5.2 cm) and old fire plots (organic depth = 8.21 cm). The spruce bark beetle (organic depth = 11.29 cm) and undisturbed plots (organic depth = 8.98 cm) had the thickest organic layers. Thus, the general trends in diversity among the different disturbance types in my study support my third hypothesis.  Below I discuss several caveats to this finding. The reciprocal relationship between disturbance and diversity has been a matter of debate; namely whether disturbance determines diversity or diversity determines the severity of the effects of a disturbance (Hughes et al., 2007). The disturbance hypothesis predicts that the frequency or intensity of disturbances can affect species diversity, which in turn affects                                                  10 The variation in organic depths by disturbance type was significant at p<.05 as provided in chapter two. 124  ecosystem function (Loreau et al., 2002). In the long term, it may also result in the addition of disturbance-resistant species to the community (Tilman, 1996). Chapin et al. (1997) and Vitousek et al. (1997) state that land-use changes, habitat fragmentation, nutrient enrichment by decomposition or fire, and environmental stress may cause a reduction in plant diversity. Hughes et al. (2007) found that disturbances had a negative effect on diversity.  Hart and Chen (2006) suggested however that fire has a positive effect by increasing understory vegetation diversity, with the highest diversity occurring within 40 years of a disturbance and gradually declining afterwards, which is consistent with the IDH (Connell, 1978).    The type of understory vegetation may be different by disturbance type as found in this study with recent fire plots having the highest number of unique species, which were not present in other type of disturbances. The disturbances also had socio-economic implications, as some of the locally used species were present or absent in a particular disturbance type. For example Ledum groenlandicum Oeder and Alnus incana (L.) Moench were present only in undisturbed or old fire plots indicating disturbances may reduce their abundance. Similarly, Epilobium angustifolium L.s.l. was present only in recent fire plot.  Shepherdia canadensis (L.) Nutt. was present in all type of disturbances except in beetle-affected plots. As mentioned above, soil texture, moisture content and canopy gaps may have played role in the variation of these species among disturbance type.     5.4.1.3 Impacts of geo-climatic variation on vegetation diversity Beside disturbances and canopy gap, variation in vegetation diversity was also related to geographical and climatic factors. The results indicated that higher diversity occurs at sites with higher mean summer maximum temperature and higher mean annual precipitation. The CCA 125  also indicated that higher Shannon-Wiener and richness values were positively correlated with mean maximum summer temperature. Generally, nonvascular plants are more affected by limited precipitation and by the amount of available light (Busby et al., 1978). Higher temperatures support higher diversity as there is less permafrost and higher nutrient availability due to the higher decomposition rate (Van Cleve et al., 1981; Bonan & Shugart, 1989). Similarly, the higher diversity at lower elevations and on level ground can also be related to climatic factors as the lower elevation sites had higher mean temperatures and a shorter snow cover period than at higher elevations. However, Chipman and Johnson (2002) found higher species diversity on upper slopes than on lower slopes in western Canada. The higher diversity on upper slopes in their studies was related to time elapsed since occurance of last fire, as a short fire cycles increase the diversity.   5.4.2 Overall plant diversity and distribution in the study area 5.4.2.1 Diversity measurement: effectiveness and applicability Sampling intensity is an important factor in determining diversity of any community. My species accumulation analysis indicated that 45 plots captured almost 90 % of the sampled species in the landscape. Diversity patterns are generally site-specific (Condit et al., 2002; Reyers et al., 2002; Wiersma and Urban, 2005). The minimum number of sites required to represent all species within a given area depends on the degree of species heterogeneity (beta diversity) among sites (Noss, 1996; Nekola & White, 1999; Condit et al., 2002), and within-site diversity (alpha diversity), which provides important information for biodiversity conservation and management. Similarly, I assessed variation in vegetation diversity along environmental gradients using the Shannon-Wiener Index, Simpson Index and Renyi?s alpha variation. Renyi?s profile uses many 126  indices for species richness and diversity including modified Simpson and Shannon-Wiener indices (Kindt and Coe, 2005). I found this method to be consistent and comprehensive. The results of the Shannon-Wiener and Simpson indices were not always consistent. For example, in the elevation category the Shannon-Wiener index indicates that lower elevations had higher diversity (H = 2.07) but the Simpson index indicates both higher and lower elevation categories having higher diversity (H= 6.71). Renyi?s profile was able to overcome these inconsistencies.   5.4.2.2 Overall diversity and species distribution The overall diversity measured in this study was comparable to other studies done in the region. Douglas (1974) reported that the flora of southwestern Yukon, as with other northern floras, has fewer species than further south. He recorded 58 species of low shrub and herb stratum in his floristic study in the Alsek river region, which is close to what I recorded in this study (47 species of shrub and herb11). The higher abundance of herbs and shrubs in this study, with an average cover of more than 100%, indicates that they are densely distributed in most of the plots.  Pleurozium schreberi (Brid.) Mitt. and Dicranum polysetum Sw. were abundantly distributed in blocks B and D. Both blocks had the highest density of white spruce, had moist soils and sporadic occurrence of permafrost. The dominant presence of feather mosses can be explained by the presence of moist and shady conditions (Cody, 2000). La Roi and Stringer (1976) found that moist areas with high precipitation are generally covered by bryophytes. However, bryophytes were recorded mostly in white spruce stands in this study as deciduous stands inhibit the growth of lichens and mosses (Hart and Chen, 2006). The higher abundance of                                                  11 At least because four genera of herb and shrub were not identified at species level.  127  these mosses may be one of the reasons for the lower species richness of understory species in beetle-affected plots as species richness generally decreases with increasing total cover of bryophytes (Vellak et al., 2003). Pleurozium schreberi (Brid.) Mitt. was absent in fire plots where both vegetation diversity and species regeneration were higher (see Chapter 4).   The high frequency of soapberry and willows in this study is consistent with Douglas (1973), who also found high cover (mean covers of 23% and 14%, respectively) in his study in Washington and British Columbia. Douglas also observed the frequent occurrence of mountain cranberry, which is also one of the species with the highest cover in this study. Among regenerating tree species, I found the highest cover for white spruce (8.17%) and trembling aspen (4.45%). Balsam poplar had with lowest cover (0.29%), with Douglas (1974) also reporting it as one of the infrequent species within the region. Douglas and Vitt (1976) reported that white spruce and trembling aspen occur frequently with soapberry (Shepherdia canadensis (L.) Nutt.) and twinflower (Linnaea borealis L.), which is consistent with the findings of my study.   5.5 Chapter conclusions Climate change and increasing forest disturbances have posed a great challenge and uncertainty for sustainable forest management in SWY, as there has been a poor understanding of ecosystem vulnerability in the area. I hypothesized that disturbance factors significantly affected the vegetation diversity in region, and expected plant distribution to vary with climatic factors.  Results indicated that higher diversities prevailed in fire and harvested plots, which were characterized by low crown cover and low stand density, consistent with the idea that the gap dynamics are an important element for vegetation diversity.  The diversity pattern of the study area could also be explained by the IDH, as the highest diversity was found at intermediate levels 128  of disturbance. Similarly, six culturally important species were uniquely present to specific type of disturbance, implying that disturbance heterogeneity is also important factor not only for higher species but also for the availability of culturally important species. As higher diversities were found in disturbed plots and in plots with higher mean temperature and precipitation, the predicted increase in temperature in the region (Christensen, et al., 2007) may promote overall vegetation diversity in the region in terms of its richness and abundance. However, increased frequency and severity of disturbances may negatively affect diversity, particularly if the region faces the drier summers predicted by the IPCC, which may increase the frequency of fires and bark beetles in the area.     129  Table 5.1 Characteristics of disturbed plots  Disturbance Type Soil moisture  (%) p < 0.05 Crown Cover  (%) p < .001 Stand Density (tree/ha) p < .001 Organic depth (cm) p < 0.05 Shrub cover p < .001 Spruce bark beetle plots 25.96 63.93 1400 11.29 40.71 Fire plots 19.82 11.98 247 2.75 36.23 Harvested plots 18.59 29.00 591 5.19 40.08 Non disturbed 19.42 44.48 789 8.98 61.11 Old fire plots 25.14 48.21 1039 8.21 55.59 Average of all plots 21.78 39.52 813.2 7.24 46.744 [Note: The above table shows the mean values of soil moisture, organic depth, crown cover, stand density and shrub cover by disturbance types. P-values indicate significant differences among disturbance types for each variable.]    Table 5.2 Species richness and abundance in the study area Richness   Abundance   Total species 69 Total abundance 10125 Mean per plot 13.8 Mean 114.5 Standard Error 0.4 Standard Error 4.73 Median 14 Median 107.3 Mode 15 Mode 100.9 Standard Deviation 3.7 Standard Deviation 44.9 Range 17 Range 250.4 Minimum 4 Minimum 18.3 Maximum 21 Maximum 268.7          130  Table 5.3 Distribution and abundance (% of cover) of dominant vegetation by disturbance type   Species Undisturbed Old fire Beetle affected Recent fire Harvested Arctostaphylos uva-ursi (L.) Spreng. s.l. 11.1 9.9 12.9 3.0 4.3 Arctostaphylos rubra (Rehd. & Wils.) Fern.     6.6     Populus tremuloides Michx. 6.8     6.1 7.1 Betula glandulosa Michx.       4.8   Carex spp.         4.7 Cladina mitis (Sandst.) Hustich 4.4   3.9 6.1 5.3 Cladonia spp.   3.4 3.4     Dicranum polysetum Sw.   3.9 7.2     Dicranum scoparium Hedw..       24.7   Empetrum nigrum L. 4.9 5.4       Epilobium angustifolium L.s.l.       6.2   Equisetum arvense L.       12.5   Festuca spp.         6.5 Ledum groenlandicum Oeder  9.8 8.0       Linnaea borealis L.     10.0   7.9 Alnus incana (L.) Moench 8.7         Peltigera canina (L.) Willd.     3.7     Pleurozium schreberi (Brid.) Mitt.  12.8 9.5 31.9   11.0 Shepherdia canadensis (L.) Nutt. 8.7 8.1   4.5 6.9 Picea glauca (Moench) Voss 8.6 14.9 11.1 8.3 7.4 Vaccinium vitis-idaea L.   5.5       Salix spp. 15.7 18.3 6.1 11.4 4.6 Total coverage 91.5 86.8 96.7 87.5 65.5 Note: species on bold fonts are locally used species Symbols          Species that uniquely occurred in a specific disturbance type (9 species)    Species occurred in all types of disturbances (3 species)   Species that appread in all sites except in  a specific disturbance type (3 species)    131   Figure 5.1 Vegetation diversity in the landscape (Shannon-Wiener index) [Note: diversity maps indicated that higher diversity in research blocks four (fire plots), five and six (salvage harvest)]    132   Figure 5.2 Species accumulation curve  [Note: The species accumulation curve indicates that 60 plots (sites) were sufficient to include about 90% of the species in the study area]      0 20 40 60 800204060sitesspecies richness133   Figure 5.3 Renyi?s profile showing vegetation diversity by site variables and disturbance types [Note: The values in Y axis (H alpha) provides diversity ranking from 0 to higher level and the values X axis (Alpha) indicates different indices such as  0 = the logarithm of species richness, 1 = Shannon-Wiener diversity index, 2 = logarithm of the reciprocal of Simpson diversity Index and inf. = proportion of the most abundant species. The graph that starts and ends at higher level shows higher diversity than that of lower one. The figures indicated the higher diversity in lower elevation, lower soil moisture regime and flat sites. The variation by aspects and disturbance regime were not clear due to overlapping lines.]  [Abbreviation of names and labels: Disturbances type (B = Beetle, F = Fire, OF = Old fire, H = Harvest, N = Not disturbed), Aspect (N = North, S = South, E = East, W = West, NE = Northeast, NW = Northwest, SE = Southeast, SW = Southwest, L = Level) slope (S = steep, M = Medium, L = Low, N = No slope) moisture regime/elevation (VH = Very high, H = High, M = Medium, L = Low, VL = Very low).] 134    Figure 5.4 Renyi?s profile showing vegetation diversity variation by climatic factors and stand types [Note: The figures indicated higher diversity in the sites with higher precipitation and higher mean maximum summer temperature. The higher diversities were also found in sites with the lowest stand density and the lowest diversity was found in aspen stands. Abbreviation of names and labels: Mean annual temperature, precipitation and summer maximum mean temperature (H = High, M = Medium, L = Low)]Tree density; VH = Very high, H = High, M = Medium, L = Low, VL = Very low; Stand type (A = Aspen, S = white spruce, M = Mixed, N = no tree); Crown cover/tree density (VH = Very high, H = High, M = Medium, L = Low, VL = Very low).]   135   Figure 5.5 Distribution of species and environmental variables in the research sites [Note: Figure 5.5 was the output of CCA. Round symbols indicate the 90 research plots; red + indicates species distribution. and abbreviated names are environmental variables. Abbreviation of names: DD0 = degree days below 0; DD5 = degree days above 5; CL = % soil; TT = total tree density; TCC = total crown cover; OD = Organic depth; AM = average moisture; MAP = mean annual precipitation; MAT = mean annual temperature; NFFD = number of frost free days; Ele = elevation; SA = % of sand]  -3 -2 -1 0 1 2 3-3-2-1012CS1CS2++++ +++++++++++++++++++++EleSlopeMATMAPDD0DD5NFFDODAMCLSATTTCC-101136   Figure 5.6 Sites position by environmental variables combinations [Note: Figure 5.6 is another output of CCA. Black dots are research plots and abbreviated names are species names.  The species clustered in one axis indicated that they required similar environmental conditions. By comparing figures 5.6 and 5.7, inferences can be made about which species prefer what kind of environmental conditions. For example Ledum groenlandicum, Vaccinium caespitosum, Empetrum nigrum, Elymus glaucus mostly prefer higher slope position, higher mean annual temperature and higher number of frost free days. Abbreviation of names as provided in Appendix A. Please also note that species distribution in the right below axis in Figure 5.7 matches with environmental variables of left upper axis in figure 5.6. and right upper axis in figure 5.7 matches with right below axis of 5.7]    d = 2  spcov  atcov  Wil_cov  So_cov  Bi_cov  Pel_aph  Car_spp  Dic_pal  Fes_spp  Clad_spp  Arc_uva  Ely_gla  Equ_spp  Ple_sch  Epi_aug  Lin_bor  Cla_spp  Dic_spp  Arc_rub  Ely_inn  Vac_vit  Emp_nig  Ros_aci  Fla_cuc  Lad_gro 137  6. Forest Management in response to climate variability and disturbances in the CATT  6.1 Introduction   Sustainable Forest Management (SFM) is defined as management ?to maintain and enhance the long-term health of forest ecosystems, while providing ecological, economic, social, and cultural opportunities for the benefit of present and future generations? (CSA, 2002). Forest management activity did not exist in the CATT prior to the implementation of the Strategic Forest Management Plan (SFMP, 2004) with the exception of fire suppression activities, resulting in the widespread occurrence of old and disturbed forest stands. The plan provides strategic guideline to achieve the long-term goal of sustainable forest management in the region. However, in the short-term, the plan focuses on salvage harvesting due to the high intensity of spruce mortality. Although the SFMP has been in effect since 2004, there are many challenges needed for its effective implementation, which are briefly highlighted as follows; - Uncertainty posed by climate change and forest disturbances: The growing uncertainty posed by predicted climate change, the spruce bark beetle outbreak and the increased fire hazards have created difficulties in the  implementation of appropriate management interventions in the territory that could cope with the uncertainty while still meeting the social, economic and ecological goals of SFM.   - Forest area and accessibility: The majority of the forestland is presently inaccessible due to the lack of a road network. - Lack of past experiences of forest management: The CATT did not have prior forest management experiences except for limited forestry activities. Therefore, any suggestion 138  about forest management and its impacts on forest ecosystem are largely dependent upon what has been learnt in other parts of boreal forests. The local communities have very small populations, and there is no history of involvement in commercial forestry. - Uncertain market: Although a harvest plan for up to 1 million m3 of beetle affected timber over the following ten years was approved in 2004, only 108,000 m3 dead trees had been harvested by the end of 2012. The main reasons are lack of a market for timber in the region, lack of a timber harvest plan and lack of capacity of local industries. Moreover, the costs and benefits of forest management have yet to be analyzed.   6.1.1 Current status of forest management in the CATT In July 2013, I contacted forestry officials in Yukon to get an update on forest management in the CATT. The following information is relevant to the management recommendations made in this chapter. - As mentioned above, about 108,000 m3 of wood had been salvaged by the end of 2012. It is expected that the quantity of salvage harvesting will increase due to the addition of a few small industries in the region and the growing capacity of the existing fuel wood company. Currently, a timber harvest plan is being prepared for the Burwash areas. Fuel abatement activities were carried out around Haines Junction areas in 2012. - A forest management implementation agreement between Champagne and Aishihik First Nations and Government of Yukon represented by the Ministry of Energy, Mines and Resources was signed on 16th August 2012. According to this new agreement, the Yukon Government will carry out forest inventory in the CATT. Based on the forest inventory, both governments will collaborate to determine a new annual allowable cut in the CATT 139  by 2015 to ensure the continuous supply of timber and fuel wood for the next twenty years. The new annual allowable cut includes both salvage harvesting and a separate harvest rate for green trees (unaffected by beetles). Both governments also agreed to collaborate in developing and implementing community fuel abatement plans. - The CAFN, in close collaboration with village of Haines Junction and Yukon Energy, has a plan to develop a biomass-based energy plant. This plant would require a continuous supply of biomass.  Accordding to MEMR, salvage harvesting will continue for the next two to three years with variable retentions. No decisions have been reached over the choice of the silvicultural systems that will be used after the completion of salvage harvesting. The research findings presented in this chapter are therefore of interest.  6.2 Key research findings and implications for forest management in the CATT As social and economic studies were outside the scope of this research, the management recommendations must primarily be based on the ecological perspectives, informed by this research. Specifically, this chapter will consider (i) the current situations of ecosystems in the CATT based on the key findings of the study (synthesis of the results), (ii) predicted ecosystem scenarios with or without forest management, (iii) forest management strategies suggested by other researchers in the CATT (e.g. Waeber, 2012; Ogden 2007; and SFMP, 2004) and; (iv) suggested management activities in other boreal forests with similar ecosystems by other researchers.  140  6.2.1 Projected ecosystems of the CATT in the absence of forest management The findings of this research and similar studies in the other boreal ecosystems have provided the basis for the projection of future structure and composition of forests in the CATT. As the average basal area was higher in the warmer plots, the overall forest productivity is expected to increase in the study area, as higher temperatures will generally result in a longer growing season (Johnston et al., 2006).  Both understory and overstory vegetation compositions may, however, change with increased coverage of broadleaved specie and a greater abundance of some species in disturbed forests, as I found that a few species were restricted to disturbed plots. Waeber (2012)12 also predicted a 20% reduction in white spruce forest and increase in aspen forests within a time frame of 50 to 150 years.  As higher structural diversity and understory vegetation diversity were found in mixed stands, salvage harvesting and even-aged management may undermine such ecosystem benefits (Montes et al., 2005; Wimberly and Spies, 2001).  The spruce bark beetle outbreaks could continue to plague the region due to the predicted increase in temperature (Goodman and Hungate, 2006), especially in pure and mature white spruce stands as found in this study. Similarly, mortality caused by bark beetle and increased summer temperature may increase the frequency of forest fires in the CATT.  Waeber (2012) predicted, however, that the risk of spruce bark beetle damage would decrease for the next 150 years due to the large amount of aspen replacing white spruce in the landscape. As more than 80% of trees in the beetle-affected plots were dead, there are significant implications for forest ecosystem management, with effects on both forest productivity and understory vegetation. My studies showed there was a lower diversity of understory vegetation in beetle-affected plots, with                                                  12 Waeber (2012) carried out his research in same research areas using same set of data as I have used for this research.   141  more than 30% cover of Pleurozium schreberi (red stemmed feather moss). However, beetle- affected plots had a higher abundance of white spruce regeneration, which was mostly clustered around the decaying woods. In other studies in Alaska, a lower abundance of white spruce regeneration was found in beetle-affected forests (Boucher and Mead, 2006).  Due to presence of differing vegetation under beetle-affected forests, forest management should be adopted to specific stands in response to large-scale beetle infestations (Boucher and Mead, 2006). For example, if the beetle affected dead trees are not salvaged, there is a high probability that large-scale fires will occur, as has already happened in the study area. In either case (salvaged or fire), forest ecosystems would be affected as the ecosystem attributes of the salvaged and the recent fire plots were different (Figure 6.1).   142                                Figure 6.1 Variation of ecosystem attributes by disturbance and stand types in the CATT based on the findings of the research. [ Note: These are the current state of ecosystems in the study area. The succession trajectory of each disturbance type will largely depends upon future climate, disturbances and management interventions (suggestions are made in the next section/figure)].Old fire (> 50 yrs old)  Forest Ecosystems of the CATT  Disturbed forests Undisturbed forests Beetle affected  Salvaged  Recent fire  (< 20 yrs old) Pure stands (overstory) Mixed stands ? Mixed stands with the second highest stand density (after beetle affected plots). ? Density of aspen tree was the highest but had highest abundance of white spruce regeneration than broadleaved species. ? Higher structural diversity of trembling aspen. ? Second highest soil moisture and crown coverage (after beetle affected plots)   ? High organic thickness with the highest  moisture content compared to other plots. ? Highest density and basal area of white spruce. ? Highest density of spruce regeneration ? Highest abundance of red stemmed feather moss. ? Lowest vegetation diversity. ? Lowest structural diversity. ? Lowest organic thickness and soil moisture.  ? Higher regeneration of broadleaved species compared to white spruce. ? Higher vegetation diversity than in old fire or beetle affected plots. ? Four herb species were uniquely present. ? Plots were previously affected by beetle.  ? Lowest organic thickness and soil moisture. ? Higher regeneration of broadleaved species.  ? Higher vegetation diversity.  ? Two shrub species were uniquely present. ? Highest average basal area. ? Average basal area decreased when aspen density was > 1000/ha. ? Higher diversity of understory vegetation. ?  No traces of spruce bark beetle.  ? Low regeneration abundance of both broadleaved and conifer species compared to disturbed plots.  ? Two locally used plants present, which were absent in disturbed plots. ? Vegetation diversity was lower than the disturbed forests. ? Lower regeneration of both white spruce and aspen compared to disturbed forests. ? Highest structural diversity of white spruce and trembling aspen diameters.  ? Average diameter decreased with increasing tree density. ? Understory vegetation diversity was lower than in mixed stands. ? Higher abundance of white spruce regeneration. 143  6.2.2 Suggested forest management strategies in the CATT  My results indicate that mixed stands had the ecosystem attributes (Figure 6.1) that would closely meet the ecological goals of SFMP (2004), with the highest average basal area, the highest vegetation and structural diversities with no traces of spruce bark beetle; these can be considered to be the preferred state of the forest. Forest management strategies and activities should focus on achieving the desired state by applying multiple management systems within the landscape.  Reactive and proactive management approaches have been suggested by Ogden (2007) to cope with the current disturbances and to prevent drastic ecosystem changes in the CATT under possible climate change scenarios. Under reactive management, specific management activities are carried out in response to changes that have already occurred. Proactive management is suggested to minimize anticipated vulnerabilities and to avoid the undesirable situations (Ogden, 2007).    6.2.3 Management of disturbed stands: salvage harvesting as a reactive forest management  The CATT SFMP is a reactive plan (Ogden and Innes, 2007) as it identifies salvage harvesting as a short-term solution to achieve the long term management goal. However, salvage harvesting has been considered as one of the most controversial practices after clear-cutting; it is generally applied to capture economic values from dead and decaying trees and may not be an effective tool to promote ecological recovery (Lindenmayer et al., 2008). Salvage harvesting can have important implications for the forest ecosystem and biodiversity (Purdon et al., 2002). As found 144  in this study, salvaged plots had a higher abundance of early successional species and understory vegetation with the presence of some unique species, which may compete with regenerating spruce (Goodman and Hungate, 2006). Although, higher diversity was found in salvaged plots in this study, the salvage harvesting may also create the opportunity for invasive ericaceous species (Jett? et al., 2009) and the process would also affect white spruce regeneration because of the mechanical impacts on the soil (Greene et al., 2006) and removal of decaying logs which provides an important substrate for white spruce recruitment (Lindenmayer et al., 2008). As the salvage harvesting affects overall stand structure, it would also affect overall species composition and species associated with key forest structures (Lindenmayer et al., 2008).  Salvage harvesting may be indispensable, however, when the areas and intensities of disturbances are quite large (such as in case of CATT) and can be both economically and ecologically feasible if the practice is carried out with proper planning. Retention of a minimum number of trees (even if they are dead) and decaying logs is suggested in order to maintain the structural mosaic of forest and to provide regeneration environment for the desired species and to minimize the effects on biodiversity (Fries et al., 1997;  Lindenmayer et al., 2008).  Prescribed burning followed by salvage harvesting would enhance white spruce regeneration and reduce competition from invasive species (Goodman and Hungate, 2006). Salvage harvesting followed by reforestation has been suggested as a way to limit the loss of forest cover, especially in response to climate change, as natural regeneration may decline due to reductions in seed germination (Le Goff et al., 2009). Beside salvage harvesting, Boucher and Mead (2006) have recommended reforestation, wildlife habitat improvement, and removal of hazardous fuel in the post-infestation management of the beetle-killed forests of Kenai Peninsula. Such a strategy could also be applied in the forest 145  of CATT. According to Ogden (2007), although the local forest practitioners? in SWY placed higher priority in some management activities such as fire suppression, enhancement of forest recovery by assisting tree regeneration, and minimization of habitat fragmentation and maintain connectivity, these management activities have not been implemented or realized yet.   6.2.4 Pro-active forest management to maintain forest productivity, biodiversity and species regeneration  Intensive and pro-active management activity is required to maintain healthy and productive stands, to counter the impacts of climate change and to reduce the frequency and intensity of disturbances (Stewart et al., 1998). One of the primary management activities suggested by some researchers is to divide the forest landscape into protection and intensive management zones (Le Goff et al., 2009) based on ecosystem variability, productivity and vegetation diversity.  Based on the findings from Chapter 3, the stands at lower slope and elevation positions should be allocated as production zone, which has productive stands whereas stands on upper slope and elevation should be protected due to higher abundance of useful shrubs and herbs to northern communities ( see Chapter 5).    As mentioned above, I suggest implementing silvicultural techniques to promote mixed stands both in currently disturbed and undisturbed stands. Bergeron and Harvey (1997) also argued that mixed stands are the most productive in terms of standing biomass. Herbaceous plant communities are more diversified in this type of forest (De Grandpre? and Bergeron, 1997),  As the canopy gaps have been found to be important for vegetation diversity and species regeneration in the study area, uneven aged management with frequent group or selection cuts could emulate the gap dynamics and maintain the mosaic of forest structures and composition 146  (Bouchard, 2009). Removal of  mature and old-growth trees through thinning and selection harvesting would reduce or prevent further bark beetle attack in the undisturbed stands and promote regeneration and biodiversity (Jett? et al., 2009). However, as old growth forests are highly dynamics in both space and time, the treatment should be designed in such a way as to represents the spatial pattern and the associated attributes of old growth forests  and to maintain the habitat of important species and the biodiversity (Grandpr? et al., 2009; Gauthier et al., 2009b).  The Three-Cohort model has been suggested as a way to maintain the mosaic of forest composition and structures in the landscape; this would ensure that the various development stages of the forest have the characteristics of both mixed forest and old-growth stands (Bergeron and Harvey 1997). Under this model, the first cohort is equivalent to even-aged management with low retention canopy occupied by shade-intolerant species (such as trembling aspen) and with an understory of shade tolerant species. The second cohort has an overstory of mixed stands and the third cohort canopy is mostly dominated by the shade-tolerant species. Such a model can also be achieved by applying varying rotation length to maintain different cohorts of stands (Epp et al., 2009; Seymour and Hunter, 1999).   Prescribed burning has frequently been suggested the best approach to achieve rapid regeneration of spruce forests (Goodman and Hungate, 2006).  Prescribed burning would also reduce the risk of uncontrolled fire, would enhance the fire-dependent ecosystems (Wade et al., 1989) and might control invasive species that would otherwise compete with desirable species (Le Goff et al., 2009).  However, as white spruce regeneration generally depends upon an ample supply of seed trees (Timoney and Peterson, 1996) the scarcity of live seed trees in beetle-infested and salvaged-harvested plots might limit spruce regeneration. However, advanced 147  regeneration of white spruce was abundant in the beetle-affected plots in my study area, and was mostly concentrated around decaying logs. Future salvage harvesting should maintain such decaying logs and shade trees to maintain white spruce regeneration in the study area. Shelter-wood felling might also be an option to promote regeneration especially for shade tolerant species both to provide seeds and shades (Montes et al., 2005).  As the ecosystems that were present in the CATT varied according to disturbance and stand type, as shown in Figure 6.1, specific forest management activities have been prepared for each disturbance and stand type (Figure 6.2). Similarly, forest management implications and suggestions in response to variation of stand structures, regeneration and vegetation diversity are provided in Table 6.1.    .  148              Figure 6.2 Suggested forest management strategies in response to variation of ecosystem attributes by disturbance and stand type in order to reach desired ecosystem statesForest Management Options in the CATT Disturbed stands Undisturbed stands Beetle affected  Salvaged stands Recent fire (< 20 yrs old)  Pure white spruce stands Mixed stands ? Maintain existing mixed stands.   ? Carry out thinning of trembling aspen to maintain density below 1000 stem/ha but avoid harvesting impacts on white spruce regeneration.    ? Apply salvage harvesting with variable retentions (both live and dead trees).  ? Apply harvesting in phases if feasible. ? Maintain existing decaying logs to promote white spruce regeneration. ? Apply prescribed burning in low regeneration areas. ? May require afforestation in the areas with low regeneration.   ? Apply salvage harvesting with variable retention.   ? Remove existing piles of wood debris. ? Prescribed burning may be required to reduce the higher abundance of moss and Festuca spp. and to promote the regeneration of the desired species.  ? This provides  relatively desirable state of forest ecosystem in the CATT having higher basal area, vegetation diversity and no traces of spruce bark beetle. ? Promote this state by applying selective harvesting and prescribed burning (to promote regeneration). ? Apply thinning of trembling aspen to maintain density below 1000/ha in order to maintain higher forest productivity. ? Apply selection or group harvestings and thinning to remove mature trees and to maintain stand density below 500/ha and to maintain the mosaic of stand structures. ? Apply group felling to maintain representative stands. Three-Cohort Model could be applied for the purpose. ? Create canopy gaps to promote vegetation diversity, regeneration and mixed stands.    Old fire (> 50 yrs old) 149  Table 6.1  Key research findings and forest management implications and considerations Findings related to stand structure and composition (Chapter 3) Forest management implications and considerations in the CATT i. Higher density and basal area were found in mixed stands than in pure stands of white spruce and aspen. However, average basal area decreased when aspen density was above 1000 stem/ha in mixed stands. These findings signify that mixed stands are important for higher stand productivity.  ii. The intermediate height class of white spruce had higher stand density whereas the co-dominant height class had the highest average basal area.  Similarly average diameter of white spruce decreased with increasing stand density. Both findings indicate that the overall productivity of white spruce decreases with increasing stand density.  iii. Structural diversities were found higher in undisturbed and old fire plots, which had mixed stands with higher vegetation diversity and had no traces of spruce bark beetle.   iv. Average diameter, height, basal area and density of white spruce increased with decreasing elevations and they were higher in lower slope positions. This finding indicates that white spruce productivity varies significantly by topographic positions and could help to prioritize forest management areas. v. Diameter and basal area of white spruce increased with increasing soil moisture and on northeast and southern aspects (which have longer exposure to sunlight and warmer than north aspects) which implicate both soil moisture and temperature are important factors for white spruce growth. vi. No strong relationship was found between age and diameter growths of white spruce, although I expected higher average diameters in older trees. vii. Disturbed plots were occupied by broadleaved species, except in forests affected by spruce bark beetle, where densities and basal areas of white spruce were significantly higher.  i. Promote mixed stands of hard and soft wood where appropriate (Biringer, 2003; Bergeron and Harvey, 1997).  Apply thinning operation to aspen stands to maintain their density below 1000 stem/ha.  ii. Carry out selection harvesting. Select dominant and co-dominant trees for harvesting. Carry out regular thinning operation of intermediate trees to promote their height and diameter growth of the remaining trees. Remove suppressed, diseased and poor quality individuals to increase resources availability for the remaining trees (Smith et al., 1997).  iii. As higher structural diversity promote biodiversity and reduce the insect/pest vulnerability, maintain higher structural diversity of the forest to promote biodiversity. Maintain representative forest types along environmental gradient (Holling, 2001). Apply partial and selective harvesting to maintain the structural component of the forest (Harvey et al. 2002). iv. Commercial harvesting should focus on lower elevation and lower slope position where higher basal area and density of species are available. These areas are easily accessible and would generate higher production with lower operational costs.   v. Drought may affect growth of boreal species. Singh and Wheaton (1991) suggested prioritizing management activities giving more focus to moist areas. Introduction of modified drought resistant varieties are suggested (Gitay et al., 2001), although there is no such plan in Yukon in the immediate future. vi. White spruce age is not a good indicator of tree size in the study area. Diversification and application of silvicultural practices both for even-aged and uneven-aged stands are suggested (Harvey et al., 2002). vii. The productive white spruce forests were affected by bark beetles. Salvaging them may promote broadleaved species. As stated in the first point, promoting mixed stands through selective harvesting may provide a long-term solution.   150  Findings related to tree regeneration (Chapter 4) Forest management implications and considerations in the CATT  i. The regeneration abundances of all three species were higher in fire plots, signifying fire as an important disturbance factor promoting regeneration in the study area. Regeneration of trembling aspen was significantly higher than white spruce in fire plots.  ii. White spruce regeneration was positively correlated with lower slope, elevation, presence of fire, higher mean annual temperatures and higher crown coverage. A higher density of white spruce regeneration was found in pure aspen stands. This species seems to benefit from disturbances that expose mineral soils, raise the soil temperature and have less impact on the forest canopy.  iii. Both trembling aspen and balsam poplar regeneration were strongly affected by the presence of parent trees. Their regeneration was also higher in harvesting plots indicating their preference for open habitat. iv. Regeneration of trembling aspen and balsam poplar exceeded the ratio compared to their parent trees compared to white spruce suggesting that broadleaved forest may dominate and prevail for long time periods if disturbance occurs persistently .    i. Selection or group harvesting with regeneration protection have been suggested as disturbances that have minimum effects on regeneration; clear cutting over small to large areas followed by seeding or planting has been suggested for severely disturbed forest (Bergeron et al., 1999). A selection system with variable retention is recommended for spruce bark beetle affected areas to maintain some shade that will promote white spruce regeneration (Goodman and Hungate, 2006). ii. Apply prescribed burning after selection harvesting to promote natural regeneration of both white spruce and trembling aspen.  iii. Promote natural regeneration (Kellomaki et al. 2005; Ogden and Innes, 2008). Regeneration models (Chapter 4) would help to assess the possibility of natural regeneration of white spruce, trembling aspen and balsam poplar.  iv. In areas where natural regeneration may be limited, the planting of alternative tree species or ecotypes may be an option (Singh and Wheaton, 1991). Spittlehouse and Stewart (2003) suggested identifying and planting more suitable genotypes through provenance trials, although this option did not receive high priority from local forestry practitioners (Ogden, 2007).           151  Findings related to vegetation diversity (Chapter 5) Forest management implications and considerations in the CATT i. Populus tremuloides Michx., Shepherdia canadensis (L.) Nutt. and Arctostaphylos uva-ursi (Rehd. & Wils) Fern. were highly centered in one cluster in the analysis, indicating that they prefer similar environmental conditions. Similarly, Ledum groenlandicum Oeder, Vaccinium caespitosum Michx., Empetrum nigrum L., and Elymus glaucus Buckl. were on the same axis indicating their similar environmental requirements.  ii. Vegetation diversity was higher in open areas, implying that gaps are important for promoting vegetation diversity. Understory diversity was also higher in mixed stands than in pure stands.  iii. Vegetation diversity was higher in moderately disturbed plots (in terms of severity which was assessed by measuring thickness of the organic layer), supporting the hypothesis that moderately disturbed stands have higher diversity than less or severely disturbed stands. iv. Diversity was higher in plots with higher mean summer temperature. Predicted higher mean temperature in the region may promote both richness and abundance of the vegetation diversity. v. Of the 21 species that are used locally, eleven species had below 1% cover (Annex J). Some of the locally important berry plants such as Rubus arcticus L. spp. had covers of less than 0.1% and had limited distribution (relative to total coverage by all 69 species). vi. Research blocks A and B had the highest percentage cover with 70%, and 96%, respectively. The highest herb cover (109%) was found in highest elevation in the research areas (1005 to 1158 meters above msl). These blocks also have the highest cover of Ledum groenlandicum (which is widely used as tea) with an average of 31% in each plot.  i. Species that require a similar environment can be managed by maintaining similar habitat and providing similar growth environment. Silvicultural practices should focus on maintaining species site relationships (Spittlehouse, 1997; Spittlehouse and Stewart, 2003) by creating favourable environments as required by the species such as creating gaps to maintain light-demanding species.    ii. Patches of canopy gaps and higher structural diversity should be maintained to enhance higher biodiversity. As biodiversity was higher in mixed stands, I recommend maintaining mixed stands by selective and partial harvesting and promoting other species (Harvey et al. 2002). iii. Apply ecosystem-based management by emulating natural disturbances (Gauthier et al., 2009a) to promote vegetation diversity.  iv. To facilitate forest ecosystem to adapt to climate change, apply aggressive selection logging in uneven-aged stands, reforestation with multiple species with the retention of climax stands in harvesting areas (Harding and McCullum, 1997) v. Focus should be given to conserving and managing locally important species listed in Annex J. Based on management objectives and important of particular species, the habitat of rare species should be protected or managed to avoid their extinction. Control undesired invasive species to reduce competition with the desirable species. If necessary, develop gene management and create artificial reserve to preserve these rare species (Noss, 2001). vi. Higher elevation areas have a higher abundance of important locally used species; allocate and prioritize these areas for biodiversity conservation. 152  6.3 Chapter conclusions The findings indicated that mixed stands are the most desirable situation to meet the ecological objectives of SFMP (2004). However, the ecological attributes of stands varied significantly by disturbance and stand types. Multiple management activities should be implemented specifically designed for each type of disturbance and stand type to reach the most desirable ecosystem states. For example, salvage harvesting with variable retention followed by prescribed burning is suggested for beetled-affected plots, which had lower vegetation diversity and higher stand productivity. Similarly, selective to group harvesting with thinning of mature stands is suggested in less perturbed plots to maintain the mosaic of stand structures and to emulate the gap dynamics to promote mixed stand in the long term.    153  7. Conclusions and recommendations 7.1 Major findings The forests of the southwest Yukon are facing the impacts of climate change, particularly through increases in the occurrence and severity of disturbances. A Strategic Forest Management Plan (SFMP, 2004) for the Champagne and Aishihik Traditional Territory (CATT) was implemented in 2004 in response to forest disturbances in the region with the objectives to maintain, restore or enhance forest succession, species and ecosystem diversity. To support the implementation of the recommendations in the 2004 SFMP, a study on the impacts of disturbances on the forest ecosystems of the Champagne and Aishihik Traditional Territory (CATT) was suggested by Ministry of Energy, Mines and Resources (MEMR), Government of Yukon. I assessed the impacts of bio-geoclimatic and disturbance factors on three attributes of the forest ecosystems in the CATT, namely (i) impacts on stand structure and composition, (ii) impacts on distribution and abundance of tree species regeneration and (iii) impacts on vegetation diversity. Five disturbance types and more than 30 bio-geoclimatic variables were considered as independent variables for assessing the ecosystem responses. I found that all the geographical variables (slope, elevation and aspect) were significant in explaining the variation in stand structure, regeneration and vegetation diversity in the region. Edaphic variables (soil texture and moisture) had impacts on tree species regeneration and structural variability (average height, diameter and basal area). However, soil texture did not have a significant impact on vegetation diversity. While the majority of climatic variables were significant in explaining species regeneration, mean annual temperature (MAT), mean annual precipitation (MAP) and 154  mean summer maximum temperature (TMXsm) were important for the variation stand structures and vegetation diversity.  Recent fires have had significant positive impacts on the regeneration of all three dominant tree species; white spruce, trembling aspen and balsam poplar. In addition to fire plots, aspen and balsam poplar regeneration was also higher in harvested plots. White spruce regeneration was more prolific in beetle-affected plots. Disturbances had significant impacts on stand composition as higher densities of broadleaved trees were found in fire and harvested plots. The study did not find any direct impact of the spruce bark beetle on species regeneration or stand structure. However, stand densities and basal areas of white spruce, which were mostly dead, were significantly higher on spruce bark beetle plots, indicating that beetle-affected areas were more productive compared to other plots. According to the original plan of the Government of Yukon, the majority of these beetle-affected forests will probably be salvaged. This may significantly change the forest composition in the future, either with or without salvage harvesting, towards broadleaved stands, as found in the salvaged plots. The key findings and the summary of responses of stand structure variables, regeneration and vegetation diversity to bio-geoclimatic and disturbance factors are highlighted in the bullet points below and are synthesized in Table 7.1.  Ecosystem variations in response to disturbances - Disturbed plots varied significantly by organic depth, soil moisture, crown coverage and shrub coverage. Beetled-affected and undisturbed plots had the highest soil moisture, organic thickness and crown cover. In contrast, the recent fire and salvaged-plots had the lowest organic thickness and soil moisture.  155  - Although white spruce dominates the study area with the highest stand composition index, its regeneration was proportionately less than that of trembling aspen, especially in disturbed plots.  Similarly, the regeneration abundances for broadleaved species was higher in exposed and disturbed sites especially in fire and harvesting plots indicating their strong preference for disturbed sites.  - Plots affected by the spruce bark beetle generally had high densities and basal areas of white spruce. Trembling aspen stand density was higher than average in old fire plots with more co-dominant and intermediate height class trees. - The diversity pattern in the study area could also be explained by the intermediate disturbance hypothesis (IDH) as the highest diversity was found at intermediate levels of disturbance. The depth of organic matter was considered as a criterion to classify the disturbance severity.  - The distribution and abundance of understory vegetation were affected by disturbance type, as nine understory vegetation species were uniquely present in specific types of disturbance.  Ecosystem variations in response to climate variability - Higher stand basal area and vegetation diversity were found in warmer plots signifying that the projected increase in temperature (2 oC by 2040) may favour these two variables. - The regeneration of white spruce was strongly influenced by climatic factors, especially temperature variables, which varied significantly by geographical positions with more regeneration abundance at lower elevation and slope positions. Spruce regeneration was favoured by higher mean annual temperatures, higher growing degree-days and higher 156  summer temperatures. However, regeneration of broadleaved species was negatively correlated higher temperatures.   Ecosystem variation in response to other environmental variables - Stand density and basal area varied significantly with geographical positions, temperature and soil texture.  Productivity of white spruce was higher at lower elevation and slope position.  Intermediate height class white spruce dominated the landscape, with 24.3% of the total cover, followed by co-dominant (20.6%) and dominant trees (18.3%). - The average DBH was not related with environmental variation although variations in diameter among soil texture, elevation and aspect were statistically significant.  - The mixed stands of aspen and white spruce had higher basal area per plot  than the pure white spruce stands. However, average basal area decreased when aspen density was above 1000 stem/ha in mixed stands. The average stand diameter was positively correlated with tree density up to 500 per hectare, but beyond that, the average diameter decreased signifying that density more than 500 per hectare will have negative impact on total aspen productivity.  - White spruce regeneration was more abundant with increasing crown coverage and soil moisture signifying that regeneration of this species will be affected by increasing temperature if drought condition prevails in the study area. - Vegetation diversity was significantly greater in plots that were characterized by low crown cover and low stand density (in open areas), at lower elevations and on flat sites. The results indicated the importance of gap dynamics in determining the vegetation diversity of the area.  Diversity in mixed stands was higher than in pure stands.  157  - Of the 21 species with documented local use (Johnson et al., 1995; Holloway and Alexander, 1990), 11 species had just less than 1% cover (Annex J). Some of the important berry plants such as arctic raspberry (Rubus arcticus L.) had covers of less than 0.1% and had limited distribution.  158  Table 7-1: Summary table: responses of stand structure variables, regeneration and vegetation diversity to bio-geoclimatic and disturbance factors Variables Species regeneration White spruce structure Trembling Aspen structure Balsam poplar Density Vegetation diversity White Spruce Aspen Balsam Poplar DBH Height Density Basal Area DBH Height Density Basal Area Geographical variables              Slope - - - - - - - - - x x x - Elevation - - - - - x - - - x x x - Aspect +L/NE +E/SE +L/SE +NE/S -N/+NE +S/NE +S/NE +E +E +NE/SW +SW +S +NW Disturbance regime              Fire + + + x x - - x x + x - + Harvesting - + + x x - - x x x x + + Old fire - - - + x + - x x + x x - Spruce bark beetle + - - + + + + x x - x - - No disturbances - - - x x + - x x + x x - Climatic variables              MAT + - - x x + X - - - x x + MAP - - x + - x - - - - x x + TMXsm + - - x + x x x + x x x + TMNwt - x x x x x x x X x x x x DD>5 + - x x + + + + + x x x x NFFD x x x n/a n/a n/a n/a n/a n/a n/a n/a n/a  Ecological variables              Soil texture L/SL +SL/L S/SiC x x X x LS LS/SiL S/LS S S x Organic soil depth - - x x x + x - - x x - x Soil moisture - - x + + + + x x x x x - Total crown coverage + - - n/a n/a n/a n/a n/a n/a n/a x x - Total density n/a - x x x x x x x x x x - Species density n/a + + - x n/a + + x x + x x Species basal area - x + + x + n/a + x + n/a n/a n/a  [Note: + = positive response; - = negative responses; x = no responses; N/A = not applicable or not assessed. Abbreviation of labels: L= level; NE = northeast; E = East; S = south; SW = southwest; NW = Northwest] 159  7.2 Socio-economic implications Although a detailed forest socio-economic implication of the results is outside the scope of this research, some brief highlights are provided here, especially results related to vegetation diversity, as there is the potential for some useful follow-up research. The information on distribution and abundance of locally used plants in the CATT is useful both for CATT community and forest managers to prioritize management activity in the future to conserve and maintain these plants. The study found 21 species have traditionally been used for food, medicines, fuel or timber in Yukon Territory and as such have socio-economic value in the region (Annex J).  These 21 locally used species cover about 45% of the total ground cover, of which about 30% is shrubs and herbs. Eleven species had less than 1% cover. Research blocks A and B had the highest percentage of cover with 70%, 83% and 109%, respectively. The shrub coverage was the highest (109%) in the higher elevation plots (1005 to 1158 meters above MSL) of block B. The block also has the highest cover of Ledum groenlandicum Oeder with an average of 31% in each plot (within the Block B) and covers 4% of total herbs abundance in the study area. Shepherdia canadensis, (L.) had the highest percent cover in the shrub category (8%) and was also widely distributed in the research areas. However, more than 50% of locally used species had less than 1% cover in the study area (Annex J).   The disturbances also had socio-economic implications, as some of the locally used species were present or absent in a particular disturbance type. For example, Ledum groenlandicum Oeder and Alnus incana (L.) Moench were present only in undisturbed or old fire plots indicating disturbances may reduce their abundance. Similarly, Epilobium angustifolium L.s.l. was present only in recent fire plots.  Shepherdia canadensis (L.) Nutt. was present in all 160  type of disturbances except in beetle affected plots. These findings indicated that the abundance and distribution of some locally used species would be significantly affected by change in the magnitude and frequency of disturbances. Forest management may require emulating natural disturbance to promote or reduce the abundance of species of interests.   7.3 Forest management implications Forest management involves balancing ecological, economic and social values and objectives associated with forest resources (Hickey, 2008). The health and vitality of the forest ecosystem are important in achieving the goal of sustainable forest management. However, climate change and natural disturbances pose major uncertainty for future forest condition, growth, and uses, and an understanding of such uncertainties is important in making forest management decisions (Juday et al., 2005; CCFM, 2006).  Although forest managers are already dealing with fire and insects outbreaks within the available institutional and financial capacity, new challenges and surprises are expected that will require adaptive forest management planning (Stewart et al., 1998). Intensive and proactive management activities may be required to counter the impacts of climate change and to reduce the frequency and intensity of disturbances.  Ecosystem-based adaptive management approaches may help to cope with the uncertainty posed by climate change and disturbances (Spittlehouse and Stewart, 2003; Schlaepfer, 1997). Ecosystem-based management (EBM) aims to maintain an ecosystem within its range of natural variability and to fulfill societal needs by reducing the differences between natural and managed forest landscapes (Gauthier et al., 2009; Schlaepfer, 1997).  EBM acknowledges the disturbance dynamic as a part of an ecosystem, and therefore, considers mimicking natural disturbance as an important approach for sustainable forestry (Noss 161  and Cooperrider, 1994). Similarly,  an adaptive approach has been considered an important tool for tackling challenges and uncertainties posed by climate change and disturbances that emphasizes adapting the forest management practices based on feedback provided by regular forest monitoring  (Spittlehouse and Stewart, 2003; Ogden and Innes, 2008).   The scope of this study was to assess our understanding of some important ecosystem elements of the CATT and their responses to bio-geoclimatic and disturbances factors, which would provide forest managers with a basic understanding of local ecosystems that they could use in management planning (Kimmins, 1996). The results shed some light on the implications of climate variability and disturbances on regional ecosystems. Although, the variability of mean annual temperature and precipitation of the study area were only 1.3 oC (-1.1 oC  to -2.4 oC) and 80 mm ( 270 mm to 390 mm), respectively, stand structure, regeneration and vegetation diversity varied significantly within this temperature range.  Overall forest productivity was higher in the warmer plots. Similarly, the abundance of white spruce regeneration was greater in warmer plots but the regenerations of broadleaved species were significantly lower. Vegetation diversity was higher in plots with higher mean summer temperature and mean annual precipitation. As stated in section 1.3.1, under predicted climate change scenario, the mean annual temperature in the region is projected to increase by 2 ?C by 2040 and 7 ?C by 2100 (ACIA, 2004), which is considerably higher than the variability of mean annual temperature in the study area (1.3 oC). As a result, climate change may significantly affect the forest ecosystems in the region. Higher temperature may also be associated with higher precipitation in the region (Christensen et al., 2007), which would affect the regeneration of white spruce by increasing soil moisture (see Chapter 4).  162  Although increasing mean temperature may benefit some attributes of the CATT ecosystem, it may negatively impact forest ecosystems by increasing forest disturbances. For example, regeneration of white spruce was proportionately less than that of trembling aspen in disturbed plots suggesting broadleaved species may prevail if disturbance persists in the region in the future. As projected by other researchers, the findings also indicated that there is a possibility of a shift in the general pattern of forest vegetation to warm and dry climate areas (Hebda, 1998), leading to more open stands at lower elevation that may be converted into dense shrubs (Hebda, 1997). Climate change may also result in the loss or reduction of alpine and subalpine ecosystems, affecting the unique species of plants and animals found there (Krannitz and Kesting, 1997).  One of the goals of the SFMP (2004) is to maintain the ecosystem functioning. The findings indicated that undisturbed mixed stands are the most desirable situation to meet the ecological objectives of SFMP (2004). However, the ecological attributes of stands varied significantly by disturbance and stand types. Multiple management activities should be implemented specifically tailored to each stand and disturbance type. The findings on stand structure variations (Chapter 3) would help forest managers to plan management activities such as delineating the locations for thinning, prioritizing possible harvest locations or changing stand structure to make the forest less vulnerable to fires (Lindenmayer and Franklin, 1997). The information and knowledge on regeneration distribution and its abundance in the landscape in response to bio-geoclimatic and disturbance factors (Chapter 4) would help forest managers to understand regeneration potential with or without disturbances and climate change scenarios and select appropriate management activities. For example, as the open areas created by disturbances are being occupied by light-demanding early successional species such as trembling aspen, 163  selective harvesting with prescribed burning could be one possible management intervention in such areas to facilitate white spruce regeneration (Goodman and Hungate, 2006). Similarly, knowledge of the distribution and abundance of useful plants (Chapter 5) is important for forest managers when prioritizing areas for special consideration for management and conservation for useful and/or rare species. This will provide information on the existing habitat condition of these species, which would help to guide managers to improve their habitat condition. Establishing biodiversity priority areas, applying multiple conservation strategies and adopting an adaptive management approach are some of the strategies suggested for biodiversity conservation by Lindenmayer et al. (2000).   7.4 Applicability of the research to other boreal regions Boreal forests vary significantly in terms of species composition, stand structure and biodiversity in North America and Eurasia. For example the species that occur in succession sequences are different in North America and Eurasia (Kuusela, 1990). Moreover, disturbance regimes and ecosystem resiliency to disturbances also differ from one region to another. For example natural disturbances are considered as major issues in North American boreal forests while forestry has been considered as an important issue in European boreal forests (Niemela, 1999).   This research was undertaken in the context of disturbances and climate variability in the forest landscape of the Champagne and Aishihik Traditional Territory (CATT) of the southwest Yukon of Canada. As explained in Chapter 1, the research need was identified by Ministry of Energy, Mines and Resources Government to support a management goal and objectives of the Strategic Forest Management Plan (2004) of CATT.  The research results were based on the 164  local ecosystems in the study area with respect to bio-geoclimatic and disturbance factors and the results may not apply to boreal forests in other regions or countries. Specifically, the results related to stand structure and biodiversity are generally site specific. However, the research results have been compared to similar research in eastern Canada, Alaska and Europe, with detailed literature citations in the discussion sections of each chapter and results were generally found to be consistent with studies done in other regions.   7.5 Research limitations Although I have tried my best to include as many variables as possible and used various complementary methods to assess the impacts of bio-geoclimatic disturbance factors on forest ecosystems in the region, there are still some limitations and gaps in the research.  One of the limitations of the research was that due to time and limited research scope, I was able to assess only three elements of the ecosystems (stand structure, regeneration and vegetation diversity). Ecosystems are complex phenomena and have numerous variables that could be assessed, including impacts on soil nutrients, their cycles and effects, microorganisms, wildlife habitat and populations, etc. However, the three elements that I included here are also important factors. Knowledge and information about these factors could directly affect management decisions. Similarly, I was unable to include socio-economic aspects of the ecosystem changes in the region. This would be of interest to local government and policy makers, especially how changes in the ecosystems due to climate change and disturbances will or could affect the socio-economic and cultural dynamics of the local residents. However, no research was undertaken in this area as the community had recently been the subject of detailed 165  consultative research (Ogden, 2007) and I was advised that community members were beginning to show signs of having been exposed to too many surveys and questionnaires.   7.6 Recommendations and future research As indicated in the previous section, ecosystem impacts have been analysed using current climatic scenarios. It would be interesting to see how future predicted climate would affect the ecosystems in the region. One of the recommendations for future research would be to include both current and future climate scenarios to see their impacts and implications for forest management.  Similarly, impacts on the socio-economic system is lacking in this research. Potential follow-up research would be to examine how the ecosystem variations caused by climate variability and disturbances would affect the socio-economic and cultural systems of local residents. This would be the interest to local forest managers in making appropriate forest management decisions. In addition, based on the findings in this thesis, it would be interesting to develop forest management scenarios to minimize the possible ecosystem effects of disturbances in the region.      166  Postscript  This research began with a broad scope of integrating social and ecological variables to develop an adaptive forest management tool in response to climate change and forest disturbances in the CATT. However, the social component of the research had to be dropped as we were informed that local people had been exhausted by the numerous surveys undertaken in the area in recent years. As a consequence, the research scope was narrowed to the ecological component in order to understand existing ecosystem attributes in the region with respect to forest disturbances and environmental gradients.    The research had to cover a relatively large forest landscape (approximately 53,000 ha) in a short period of time (3 months) and involved collecting information on more than 30 ecological variables.  This was achieved by a team of researchers from the Sustainable Forest Management Laboratory of the Faculty of Forestry at UBC. The advantages of conducting joint and collaborative research is one of the important lessons learned from this research, as it made possible for us to collect a large data set from 90 plots, 270 sub-plots and 270 mini plots. The team approach also made it possible to understand different ecological dynamics within the CATT forests, as different researchers focused their research on different aspects using the same set of data. Besides collecting ecological data, the research team also installed two weather stations in the region in order to collect long-term weather data. Due to time limitations, we were able to collect data on spatial variability but not on temporal variability. Temporal variability for indices such as soil moisture would provide important extra information, and this therefore reflects one of the limitations of the study.   Based on the ecological findings about current climate variability and disturbance regimes, I have recommended several silvicultural options, which would contribute to closely meeting the ecological goals of the region. However, these recommendations were largely based on ecological perspectives and learning from other boreal regions faced with similar situations. Therefore, the proposed silvicultural options should be considered only as preliminary recommendations; local people were not involved in developing these, and have not agreed with the suggested options. I believe that forest management has to be based on local perspectives and wishes and should consider local knowledge of forest management. It should also consider how 167  people would like to see their forests in the future. Although, I did not conduct formal interviews with local community members, I did receive some informal suggestions that the majority of local people are not in the favour of large-scale salvage harvesting, which may significantly alter the landscape in terms of vegetation composition and the visual aspects of the forest. The experiences from the salvage harvested areas of pine beetle affected forests in British Columbia have revealed that such practices may not actually result in desirable forest condition. Previous work in the area has also suggested that local are not in favour in introducing new tree species to the area, and would like to see the landscape returned to the condition that it was in prior to the bark beetle outbreak. Given the extent of climate change in the area, this may not be possible.  Based on my experience, I suggest that there is a need for an action research approach involving local communities during the whole research process, from the development of research questions to the finalization of research results. Such an approach would provide an opportunity for local communities to understand their ecosystems, the changes occurring in them and the factors responsible for the changes. In addition, they would also be able to share their local knowledge and experience in coping with forest disturbances and could suggest management tools that would be acceptable for them. Similarly, action research would also motivate local communities to take ownership of research findings and to implement them locally. For example, we established weather stations without the involvement of local people, which resulted lack of ownership and management of these stations, which have remained idle for the past three years.  Some of the findings of this research have direct value to local communities, such as the spatial variation of forest productivity, and the spatial distribution and abundance of locally important non-timber forest products. Action research would have provided opportunities for local people to learn the status of these locally important species, and to take decisions accordingly on management and utilization options for these resources. Similarly, I would consider the research finding regarding the factors associated with vegetation distribution and abundance in the study area as a significant scientific contribution, as these findings could form the basis for a vegetation classification that is still lacking in the region.  Personally I learned a lot from this research. The subject matter, context and study areas were completely new for me, as I came from another part of the world (Nepal) with completely different types of forests, forestry and people. My former experience mainly included working 168  for community forest management. However, my PhD mainly focused on ecological science using a variety of multivariate tools. Although, I went through a very difficult path during my research, it also provided me with a great learning opportunity, expanded my thinking process and built my confidence in forest ecology and management.    169  References  Aber, J., Neilson, R.P., McNulty, S., Lenihan, J.M., Bachelet, D., & Drapek, R.J. (2001). Forest processes and global environmental change: predicting the effects of individual land multiple stressors. Bioscience, 51: 735?51. [ACIA] Arctic Climate Impact Assessment (2004). Impacts of a Warming Arctic: Arctic Climate Impact Assessment. Cambridge, UK: Cambridge University Press.  Alexander, R.R., Shearer R.C., & Sheppard W.D. (1990). Subalpine Fir. In R.M. Burns, & B.H. Honkala (Eds), Silvics of North America, Volume 1: Conifers. Agriculture Handbook 654. United States Department of Agriculture and Forest Service, Washington, D.C., USA. Amiro, B.D., Stocks, B.J., Alexander, M.E., Flannigan, M.D., & Wotton, B.M. (2001). Fire, climate change, carbon and fuel management in the Canadian boreal forest. International Journal of Wildland Fire, 10: 405?413. Andalo, C., Beaulieu, J., & Bousquet, J. (2005). The impact of climate change on growth of local white spruce populations in Quebec, Canada. Forest Ecology and Management, 205: 169?182. Angelstam, P. (1997). Landscape analysis as a tool for the scientific management of biodiversity. Ecological Bulletin, 46: 140-170. Angelstam, P., (1996). The ghost of forest past natural disturbance regimes as a basis for reconstruction of biologically diverse forests in Europe. In DeGraaf, R.M., & R.I. Miller (Eds.), Conservation of Faunal Diversity in Forested Landscapes. London: Chapman and Hall.  Anisimov, O.A., Vaughan, D.G., Callaghan, T.V. Furgal, C., Marchant, H., Prowse, T.D., Vilhj?lmsson H.A., & Walsh, J.E. (2007). Polar Regions (Arctic and Antarctic). In M.L. Parry, O.F. Manziani, J.P.  Palutikof, P.J.  van der Linden, & C.E.Hanson, (Eds), Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press.  Archibold, O.A., Acton, C., & Riple, E.A. (2000). Effect of site preparation on soil properties and vegetation cover, and the growth and survival of white spruce (Picea glauca) seedlings, in Saskatchewan. Forest Ecology and Management, 131: 127-141.  Bady, P., Doledec, S., Dumont, B., & Fruget, J.F. (2004). Multiple co-inertia analysis: a tool for assessing synchrony in the temporal variability of aquatic communities. Ecology, 327: 29-36. 170  Barber, V.A., Juday, G.P., & Finney, B.P. (2000). Reduced growth of Alaskan white spruce in the twentieth century from temperature induced drought stress. Nature, 405: 668?673. Barber, V.A., Juday, G.P., Finney, B.P., & Wilmking, M. (2004). Reconstruction of summer temperature in interior Alaska from tree ring proxies: Evidence for changing synoptic climatic regime. Climate change, 63: 91-120. Bartos, D.L. (2001). Landscape Dynamics of Aspen and Conifer Forests. USDA Forest Service Proceedings, Rocky Mountain Research Station.  Bates, P.C. (1989). Aspen stand development following Harvest in the Lake States Region. Northern Journal of Applied Forestry, 6: 178-183. Bengtsson, J. Nilsson, S.G., Franc, A., & Menozzi, P. (2000). Biodiversity, disturbances, ecosystem function and management of European forests. Forest Ecology and Management, 132: 39-50. Berg, E.E, Henry J.D., Christopher, L. Fastie, C.L, Volder, A.D., & Matsuoka, S.M. (2006).  National Park and Reserve, Yukon Territory: Relationship to summer temperatures and regional differences in disturbance regimes. Forest Ecology and Management, 227: 219?232. Berg, E.E., & Henry, J.D. (2003). The history of spruce bark beetle outbreak in the Kluane region as determined from the dendro-chronology of selected forest stands. Parks Canada Report. Bergeron, Y. (1991). The influence of island and mainland lakeshore landscapes on boreal forest fire regimes. Ecology, 72: 1980?1992. Bergeron, Y. (2000).  Species and stand dynamics in the mixed woods of Quebec?s southern boreal forest. Ecology, 81: 1500?1516. Bergeron, Y., & Dansereau, P.R. (1993). Predicting the composition of Canadian southern boreal forest in different fire cycles. Journal of Vegetation Science, 4: 827-832. Bergeron, Y., & Flannigan, M.D. (1995). Predicting the effects of climate change on fire frequency in the south-eastern Canadian boreal forest. Water Air Soil Pollution, 82: 437?444. Bergeron, Y., & Harvey, B. (1997). Basing silviculture on natural ecosystem dynamics: an approach applied to the southern boreal mixed-wood forest of Quebec. Forest Ecology and Management, 92: 235-242.  171  Bergeron, Y., & Harvey, B. (1997). Basing silviculture on natural ecosystem dynamics: an approach applied to the sourthern boreal mixedwood forest of Quebec. Forest Ecology and Management, 92: 235-242. Bergeron, Y., & Leduc, A. (1998). Relationships between change in fire frequency and mortality due to spruce budworm outbreak in the southeastern Canadian boreal forest. Journal of Vegetation Science, 9: 493?500. Bergeron, Y., Harvey, B., Leduc, A., & Gauthier, S. (1999). Forest management guidelines based on natural disturbance dynamics: Stand and forest level considerations. The Forestry Chronicle, 75: 49-54. Bergeron, Y., Leduc, A., Morin, H., & Joyal, C. (1995). Balsam fir mortality following the last spruce budworm outbreak in north-western Quebec. Canadian Journal of Forest Research, 25: 1375?1384. Bergeron, Y., & Dubuc, M. (1989). Succession in southern part of the Canadian boreal forest. Vegetation, 79: 51-63. Biringer, J. (2003). Forest ecosystem threatened by climate change: promoting long term forest resilience. In L.J. Hansen, J. Biringer, & J.R. Hoffman (Eds), Buying time: A user?s manual for building resistance and resilience to climate change in natural systems (pp 41-69). World Wide Fund for Nature, Gland, Switzerland.  Birks, H.J.B., Peglar, S.M., & Austin, H.A. (1996). An annotated bibliography of Canonical Correspondence Analysis and related constrained ordination methods 1986-1993. Abstracta Botanica, 20: 17-36. Blais, J.R. (1981). Mortality of balsam fir and white spruce following a spruce budworm outbreak in the Ottawa River watershed in Quebec. Canadian Journal of Forest Research, 11: 620?629. Bockheim, J.G., O?Brien, J.D., Munroe, J.S., & Hinkel, K.M. (2003). Factors Affecting the Distribution of Populus balsamifera on the North Slope of Alaska, U.S.A. Arctic, Antarctic, and Alpine Research, 35: 331?340. Bonan, G.B., & Shugart, H.H. (1989). Environmental factors and ecological processes in boreal forests. Annual Review of Ecology and Systematic, 20: 1?28. Bormann, F.H., & Likens, G.E. (1979). Catastrophic disturbance and steady state in Northern hardwood forest. American scientist, 67: 660-669.  Bornor, M.G., & Oswald, E.T. (1985).The Yukon reconnaissance inventory. Information Report, BC-315, Pacific and Yukon Region. Pacific Forestry Centre. 172  Bouchard, M. (2009). Silviculture in a context of forest ecosystem management in boreal and southern boreal forests. In S. Gauthier, M. Vaillancourt, A. Leduc, L. De Grandpr?, D. Kneeshaw, H. Morin, P. Drapeau, and Y. Bergeron (Eds.), Ecosystem management in the boreal forest. Quebec, Canada: Presses de l?Universite du Quebec.  Boucher, D., Gauthier, S., & De Grandpr?, L. (2006).  Structural changes in coniferous stands along a chronosequence and a productivity gradient in the northeastern boreal forest of Qu?bec. Ecoscience, 13: 172-180.  Boucher, T.V., & Mead, B.R. (2006). Vegetation change and forest regeneration on the Kenai Peninsula, Alaska following a spruce beetle outbreak, 1987?2000. Journal of Forest Management and Ecology, 227: 233-246.  Brassard, B.W., & Chen, H.Y.H.  (2006). Stand structural dynamics of North American boreal forests. Critical Reviews in Plant Sciences, 25: 115?137. Bruno, J.F., Fridley, J.D., Bromberg, K.D. & Bertness, M.D. (2005). Insights into biotic interactions from studies of species invasions. In D.F. Sax, J.J. Stachowicz, & S.D. Gaines, (Eds), Species Invasions: Insights into Ecology, Evolution, and Biogeography (pp. 13?40). Sunderland, MA: Sinauer Associates, Inc.  Busby, J.R., Bliss, L.C., & Hamilton, C.D. (1978). Microclimatic control of growth rates and habitats of the boreal forest mosses, Tomenthypnum nitens and Hylocomium splendens. Ecological Monograph, 48: 95?110. Chapin, F.S.,  Zavaleta, E.S.,  Eviner, V.T., Naylor, R. L.,  Vitousek, P.M.,   Reynolds, H. L., Hooper, D.U., Lavorel, S.,  Sala, O.S.,  Hobbie, S.E.,  C. Mack, M.C., & D?az, S. (2000). Consequences of changing biodiversity. Nature, 405: 234-242. Chapin, F.S., Callaghan, T.V., Bergeron, Y., Fukuda, M., Johnstone, J.F., Juday, G., & Zimov, S.A. (2004).  Global change and the boreal forest: thresholds, shifting States or gradual Change?  Ambio, 33: 361-365. Chapin, F.S., Walker, B.H., Hobbs, R.J., Hooper, D.U., Lawton, J.H., Sala, D., & Tilman, D. (1997). Biotic control over the functioning of ecosystems. Science, 277: 500?504. [CCFM] Canadian Council of Forest Ministers (2006). Criteria and indicators of sustainable forest management in Canada: National Status Report 2005. Natural Resources Canada, Canadian Forest Service, Ottawa, Ontario, Canada. Chavard?s, R.D., Daniels, L.D.,  Waeber, P.O, Innes J.L., & Nitschke, C.R. (2012). Unstable climate?growth relations for white spruce in southwest Yukon, Canada. Climate Change, Online publication. DOI 10.1007/s10584-012-0503-8 173  Chen, H.Y.H, Klinka, K., & Kabzems, R.D. (1998). Site index, site quality, and foliar nutrients of trembling aspen: relationships and predictions. Canadian Journal of Forest Research, 28: 1743?1755 Chen, H.Y.H., & Popadiouk, R.V. (2002). Dynamics of North American boreal mixedwoods. Environmental Review, 10: 137?166. Chen, H.Y.H., Klinka, K., & Kayahara, G.J. (1996). Effects of light on growth, crown architecture, and specific leaf area for naturally established Pinus contorta var latifolia and Pseudotsuga menziesii var glauca saplings. Canadian Journal of Forest Research, 26: 1149?1157. Chevenet, F., Doledec, S., & Chessel, D. (1994). A fuzzy coding approach for the analysis of long-term ecological data. Freshwater Biology, 31: 295?309. Chhin, S., & Wang, G.G. (2008).Climatic response of Picea glauca seedlings in a forest-prairie ecotone of western Canada. Annals of Forest Science, 65: 207-214. Chipman, S.J., & Johnson, E.A. (2002). Understory vascular plant species diversity in the mixedwood boreal forest of western Canada. Journal of Applied Ecology, 12: 588?601. Christensen, J.H., Hewitson, B.,  Busuioc, A., Chen, A.,  Gao, X., Held, I., Jones, R., Kolli, R.K.,  Kwon, W.T.,  Laprise, R., Maga?a Rueda, V., Mearns, L., Men?ndez, C.G., R?is?nen, J.,  Rinke, A. Sarr, A.  & P. Whetton, (2007). Regional Climate Projections. In  S. Solomon, D. Qin, M. Manning, Z. Chen, M.  Marquis, K.B. Averyt,  M. Tignor,  & H.L. Miller (Eds), Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom: Cambridge University Press. Clarke, K.R., & Warwick, R.M. (2001). Changes in marine communities: an approach to statistical analysis and interpretation, (2nd Ed.). PRIME: Plymouth.  Coates, K.D., Haeussler, S., Lindeburgh, S., Pojar, R. & Stock, A.J. (1994). Ecology and silviculture of interior spruce in British Columbia. Forestry Canada and British Columbia Ministry of Forests, FRDA Report 220. Cody, W.J. (2000). Flora of the Yukon Territory (2nd Ed). Ottawa, Ontario, Canada: NRC Research Press.  Cole, W.D. (1995). Soil nutrient supply in natural and managed forests. Plant and Soil, 168-169: 43-53. 174  Condit, R., Pitman, N., Leigh, E.G., Chave, J., Terborgh, J., Foster, R.B., Nunezv, P., Aguilar, S., Valencia, R., Villa, G., Muller- Landau, H.C., Losos, E., & Hubbell, S.P. (2002). Beta diversity in tropical forest trees. Science, 295: 666?669. Connell, J.H. (1978). Diversity in tropical rain forests and coral reefs. Science, 199:1302-1310. Cooper, A., Mccan, T., & Bunce, R.G.H. (2006). The influence of sampling intensity on vegetation classification and the implications for environmental management. Environmental Conservation, 33: 118?127. Cordes, L.D., Hughes, F M.R., & Getty, M. (1997). Factors affecting the regeneration and distribution of riparian woodlands along a northern prairie river: the Red Deer River, Alberta, Canada. Journal of Biogeography, 24: 675-695. [CSA] Canadian Standard Associations (2002). Sustainable forest management: requirements and guidance. Canadian Standard Association. Mississauga, Ontario. Available online http://www.csa-international.org/product_areas/forest_products_marking/program_documents/CAN_CSA_Z809-02O_English.pdf Cumming, S.G., Schmiegelow, F.K.A., & Burton, P.J. (2000). Gap dynamics in boreal aspen stands: is the forest older than we think? Ecological Application, 10: 744-759. Dale, V.H., Joyce, L.A., McNulty, S.,  Neilson, R.P.,  Ayres, M.P., Flannigan, M.D., Hanson, P.J., Irland, L.C., Lugo, A.E.,  Peterson, C.J., Simberloff, D.,  Swanson, F.J.,  Stocks, B.J., & Wotton, B.M. (2001). Climate change and forest disturbances. Bioscience, 51: 723?734. Danby, Y., & Hik, D. (2007). Responses of white spruce (Picea glauca) to experimental warming at a subarctic alpine treeline. Global Change Biology, 13: 437?451.  De Grandpr?, L. and Bergeron, Y. (1997). Diversity and stability of understory communities following disturbance in the southern boreal forest. Journal of Ecology, 85: 777-784 De Grandpr?, L., Gagnon, D., & Bergeron, Y. (1993). Changes in the understory of Canadian southern boreal forest after fire. Journal of Vegetation Science, 4: 803?810. De Grandpre?, L., & Bergeron, Y. (1997). Diversity and stability of understory communities following disturbance in the southern boreal forest. Journal of Ecology. 85: 777-784.  Delong, S.C., & Kessler, W.B. (2000). Ecological characteristics of mature forest remnants left by wildfire. Forest Ecology and Management, 131: 93?106. Dole?dec, S., & Chessel, D. (1994). Co-inertia analysis: an alternative method for studying species?environment relationships. Freshwater Biology. 31:277?294. 175  Douglas, G.W. & Vitt, D.H. (1976). Moss Lichen Flora of St Elais Klaune Ranges Southwestern Yukon.  The Bryologist, 79: 437-456. Douglas, G.W. (1973). Alpine plant communities of the North Cascades Range, Washington and British Columbia. Ph.D. Thesis, University of Alberta, Edmonton. Douglas, G.W. (1974). Montane zone vegetation of the Alsek River region, southwestern Yukon. Canadian Journal of Botany, 52: 2505-2532. Dray, S., Chessel, D., & Thioulouse, J. (2003). Co-inertia analysis and the linking of ecological data tables. Ecology, 84: 3087-3089. Duchesne, L.C., Zasada, J.C., & Davidson-Hunt, I. (2000). Non-timber forest product industry in Canada: Scope and research needs. Forestry Chronicle, 76:743?746. Ek, A. R., Robinson, A.P., Radtke, P.J., & Walters, D.K. (1997). Development and testing of regeneration imputation models for forests in Minnesota. Forest Ecology and Management, 94: 129-140. Ek, A.R., & Brodie, J.D. (1975). A preliminary analysis of short-rotation aspen management. Canadian Journal of Forest Research, 5: 245-258. Ek, A.R., Robinson, A.P., Radtke, P.J., & Walters, D.K. (1996). Regeneration imputation models and analysis of forests in Minnesota. Staff paper series No. 115. Department of Forest Resources, University of Minnesota. Elliot, G.P., & Baker W.L. (2004). Quaking aspen (Populus tremuloides Michx.) at treeline: a century of change in the San Juan Mountains, Colorado, USA. Journal of Biogeography, 31: 733?745. Engelmark, O. (1999). Boreal forest disturbance. In L.R. Walker. (ed.) Ecosystem of disturbed ground. Amsterdam: Elsevier Science, B.V. Epp, B., Taradif, J.C., Kenkel, N., & De Grandpr?, L. (2009) Forest dynamics of the Duck Mountain Pronvicial Forest, Manitoba and the mmplications for the forest management. In S. Gauthier, M. Vaillancourt, A. Leduc, L. De Grandpr?, D. Kneeshaw, H. Morin, P. Drapeau, & Y. Bergeron (Eds.), Ecosystem management in the boreal forest. Quebec, Canada: Presses de l?Universite du Quebec.  [ESRI] Environmental Systems Resource Institute. (2009). ArcGIS 9.2. ESRI, Redlands, California. Esseen, P.A., Ehnstroem, B., Ericson, L., & Sjoeberg, K. (1997). Boreal forests. Ecological Bulletin, 46: 16-47. 176  Field, C.B., Mortsch L.D., Brklacich M., Forbes D L., Kovacs P., Patz J.A., Running S.W., & Scott, M. J. (2007). North America. Climate Change 2007: Impacts, Adaptation and Vulnerability. In M.L. Parry,   O.F. Canziani, J.P.  Palutikof, P.J. van der Linden, C.E.  Hanson, C.E. (Eds.), Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (pp. 617?652). Cambridge, UK: Cambridge University Press.  Fisher, R.A. (1936). The use of multiple measurements in taxonomic problems. Annals of Eugenics, 7: 179-188. Flannigan, M.D., & Van Wagner, C.E. (1991). Climate change and wildfire in Canada. Canadian Journal of Forest Research, 21:66?72. Flannigan, M.D., Bergeron, Y., Engelmark, O., & Wotton, B.M. (1998). Future wildfire in circum boreal forests in relation to global warming. Journal of Vegetation Science, 9: 469?476. Flannigan, M.D., Logan, K.A., Amiro, B.D., Skinner W.R., & Stocks, B J. (2005). Future area burned in Canada. Climate Change, 72: 1-16. Folke, C., Holling, C.S., & Perrings, C. (1996). Biological diversity, ecosystems and the human scale. Ecological Applications, 6:1018-1024. Forman, R.T.T., & Godron, M. (1981). Patches and Structural components for a landscape ecology. BioScience, 31: 733-740. Fralish, J.S., & Loucks, O.L. (1975).  Site quality evaluation models for aspen (Populus tremuloides Michx.). Canadian Journal of Forest Research, 5: 23-528. Franklin, J.F. (1988). Structural and functional diversity in temperate forests. In E.O. Wilson, (Ed.), Biodiversity (pp 166-175). Washington, DC: National Academy Press.  Frey, B.R., Victor, J.V.J., Landh?usser, S.M., Comeau, P.G., & Greenway, K.J. (2003).  An analysis of sucker regeneration of trembling aspen. Canadian Journal of Forest Research, 33: 1169?1179. Fries, C., Johansson, O., Pettersson, B., & Simonsson, P. (1997). Silvicultural models to maintain and restore natural stand structures in Swedish boreal forests. Forest Ecology and Management, 94: 89?103. Froese, K. (2003). Regeneration and small tree height increment in the interior Douglas-fir zone of the rocky mountain trench. A thesis submitted in partial fulfilment of the requirements for the degree of Master of Science. Department of Forestry Resource, University of British Columbia. 177  Frugal, C., & Prowse, T.D. (2008). Northern Canada. In D.S. Lemmen, F.J. Warren, J. Lacroix and E. Bush (Eds.), From Impacts to Adaptation: Canada in a Changing Climate (2007) (pp.57-118). Government of Canada, Ottawa, ON. Fyllas, N.M., Panayiotis, G.D., & Troumbis, AY. (2008). Regeneration dynamics of a mixed Mediterranean pine forest in the absence of fire. Forest Ecology and Management 256:  1552-1559.    Galipeau, C., Kneeshaw, D., & Bergeron, Y. (1997). White spruce and balsam fir colonization of a site in the southeastern boreal forest as observed 68 years after fire. Canadian Journal of Forest Research, 27:139-147. Garbutt, R. (2007). Yukon forest health report. Canadian Forest Service, Pacific Forestry Centre, Victoria, B.C.  Garbutt, R. (2008). Yukon Forest Health Report 2007. Government of Yukon, Ministry of Energy, Mines and Resources, Forest Management Branch. Available online: http://www.emr.gov.yk.ca/forestry/pdf/forest_health_report07.pdf Gauch, H.G. (1973). A Quantitative Evaluation of the Bray-Curtis Ordination. Ecology, 54: 829-836. Gauthier, S., De Grandpre, L., & Bergeron, Y. (2000). Differences in forest composition in two boreal forest ecoregions of Quebec. Journal of Vegetation Science, 11: 781?790. Gauthier, S., Luduc, A., Bergeron, Y., & Le Goff, H. (2009b). Fire frequency and forest management based on natural disturbances. In S. Gauthier, M. Vaillancourt, A. Leduc, L. De Grandpr?, D. Kneeshaw, H. Morin, P. Drapeau, & Y. Bergeron (Eds.), Ecosystem management in the boreal forest. Quebec, Canada: Presses de l?Universite du Quebec.  Gauthier, S., Vaillancourt, M., Kneeshaw, D., De Grandpr?, L., Claveau, Y., & Par?, D. (2009a). Forest ecosystem management: origin and foundations. In S. Gauthier, M. Vaillancourt, A. Leduc, L. De Grandpr?, D. Kneeshaw, H. Morin, P. Drapeau, &Y. Bergeron (Eds.), Ecosystem management in the boreal forest. Quebec, Canada: Presses de l?Universite du Quebec.  Gitay, H., Brown, S., Easterling, W., & Jallow, B. (2001). Ecosystems and their goods and services. In J.J. McCarthy, O.F. Canziani, N.A. Leary, D.J. Dokken, & K.S. White (Eds.) Climate Change 2001: Impacts, adaptation and vulnerability. Contribution of working group II to the Third Assessment Report of the Intergovenrmental Panel on Climate Change. Cambridge, UK: Cambridge Univesity Press.  178  Gitay, H., Su?rez, A., & Wilson, R. (2002). Climate Change and Biodiversity. Technical Paper V. Intergovernmental Panel on Climate Change. Geneva. Goodman, L.F., & Hungate, B.A. (2006). Managing forests infested by spruce beetles in south-central Alaska: Effects on nitrogen availability, understory biomass, and spruce regeneration. Forest Ecology and Management, 227: 267?274. Greene, D.F., Gauthier, S., Noel, J, Rousseau M., and Y. Bergeron. (2006). A field experiment to determine the effect of post-fire salvage on seedbeds and tree regeneration. Frontiers in Ecology and the Environment 4: 69-74. Grime, J.P. (1973). Competitive exclusion in herbaceous vegetation. Nature: 242: 344?347. Haara, A., Maltamo, M., & Tokola, T. (1997). The k-nearest neighbor method for estimating basal area diameter distribution. Scandevian Journal of Forest Research, 12: 200?208. Haila, Y., & Kouki, J., (1994).The phenomenon of biodiversity in conservation biology. Annales Zoologici Fennici, 31: 5-18. Haila, Y., Hanski, I.K., Niemela, J., Punttila, P., Raivio, S., & Tukia, H. (1994). Forestry and boreal fauna: Matching management with natural forest dynamics. Annales Zoologici Fennici, 31: 187- 202. Hannah, L., Midgley, G.F., & Millar, D. (2002). Climate change-integrated conservation strategies. Global Ecology & Biogeography, 11: 485?495. Harding, L.E., & McCullum, E. (1997). Ecosystem response to climate Change in British Columbia and Yukon: threats and opportunities for biodiversity In E. Taylor, & B. Taylor, (Eds.), Responding to global climate change in British Columbia and Yukon. Volume I of the Canada Country Study: Climate Impacts and Adaptation. Vancouver, BC. Hare, F.K., & Ritchie, J.C. (1972). The Boreal Bioclimates. Geographical Review, 62: 333-365. Harper, K. A., Bergeron, Y., Drapeau, P., Gauthier, S., & De Grandpr?, L. (2005). Structural development following fire in black spruce boreal forest. Forest Ecology & Management, 206: 293?306. Hart, S.A., & Chen, H.Y. H. (2006). Understory vegetation dynamics of North American boreal forests. Critical Reviews in Plant Sciences, 25: 381?397.  Harvey, B.D., Leduc, A., Gauthier, S., & Bergeron, Y. (2002). Stand landscape integration in natural disturbance based management of the southern boreal forest. Forest Ecology and Management, 155: 369-385. 179  Hassani, B., Marshall, P., LeMay V., Temesgen, H., & Zumrawi, A.A. (2002). Development of Regeneration Imputation Models for the ICHmw2 in the Vicinity of Nelson. Report prepared for Forest Renewal BC. Forest Resources Management Department, University of British Columbia Hebda, R.J. (1997). Impact of Climate Change on Biogeoclimatic Zones of British Columbia and Yukon. In E. Taylor and B. Taylor (Eds.), Future Climate Change in British Columbia and Yukon (pp. 13:1?15). Victoria, B.C: Environment Canada and Ministry of Environment, Lands, and Parks, Province of British Columbia. Hebda, R.J. (1998). Atmospheric change, forests and biodiversity. Environmental Monitoring and Assessment, 49: 195?212. Heinselman, M.L. (1973). Fire in the virgin forests of the Boundary Waters Canoe Area. Quaternary Research. 3: 329-382. H?ly, C., Bergeron,Y., & Flannigan, M.D. (2000). Effects of stand composition on fire hazard in mixed-wood Canadian boreal forest. Journal of Vegetation Science, 11: 813?824. Hickey, G.M. (2008). Evaluating sustainable forest management. Ecological Indicators, 8: 109-114. Hogg, E.H., & Wein, R.W. (2005). Impacts of drought on forest growth and regeneration following fire in southwestern Yukon, Canada. Canadian Journal of Forest Research, 35: 2141?2150.  Hogg, E.H., Brandt, J.P., & Kochtubajda, B. (2002). Growth and dieback of aspen forests in northwestern Alberta, Canada in relation to climate and insects. Canadian Journal of Forest Research, 32:  823-832. Hogg, E.H., Brandt, J.P., & Kochtubajda, B. (2005). Factors affecting inter-annual variation in growth of western Canadian aspen forests during 1951?2000. Canadian Journal of Forest Research, 35: 610?622. Hogg, E.H., Brandt, J.P., & Michaelian, M. (2008). Impacts of a regional drought on the productivity, dieback, and biomass of western aspen forests. Canadian Journal of Forest Research, 38: 1373?1384. Hogg, E.H., Saugier, B., Pontailler, J.Y., Black, T.A, Chen, W., Hurdle, P.A., & Wu, A. (2000). Responses of trembling aspen and hazelnut to vapour pressure deficit in a boreal deciduous forest. Tree Physiology, 20: 725?734. Holling, C.S. (2001). Understanding the complexity of economic, ecological and social systems. Ecosystems, 4: 390-405 180  Holloway, P.S., & Alexander, G. (1990). Ethnobotany of the Fort Yukon Region, Alaska. Economic Botany, 44: 214-225. Huang, S., & Titus, S.J. (1995). An individual tree diameter increment model for white spruce in Alberta. Canadian Journal of Forest Research, 25: 1455-1465. Hughes, A., Byrnes, J.E., Kimbro, D.L., & Stachowicz, J.J. (2007). Reciprocal relationships and potential feedbacks between biodiversity and disturbance. Ecology Letters, 10: 849?864. Hughes, O.L., van Everdingen, R.O., & Tarnocai, C. (1983). Regional setting-physiography and geology. In H.M. French & J.A. Heginbottom (Eds.), Guidebook to Permafrost and Related Features of the Northern Yukon Territory and Mackenzie Delta, Canada (pp.5 ?34). Fairbanks, Alaska: Division of Geological and Geophysical Surveys, Department of Natural Resources.  Hult?n, E. (1968). Flora of Alaska and neighbouring territories. Stanford, USA: Stanford University Press. Huston, M. (1979). A general hypothesis of species diversity. The American Naturalist, 113: 81?101. [IPCC] Intergovernmental Panel on Climate Change (2002). Climate change and biodiversity. Technical paper V. Cambridge, UK: Cambridge University press.  Jasinski, J.P.P. & Payette, S. (2005). The creation of alternative stable states in the southern boreal forest, Quebec, Canada. Ecological Monographs, 75: 561-583. Jett?, J., Vaillancourt, M., Leduc, A., & Gauthier, S. (2009). Ecological issues related to forest management. In S. Gauthier, M. Vaillancourt, A. Leduc, L. De Grandpr?, D. Kneeshaw, H. Morin, P. Drapeau, & Y. Bergeron (Eds.), Ecosystem management in the boreal forest. Quebec, Canada: Presses de l?Universite du Quebec.  Jongman, R.H.G., Ter Braak, C.J.F., & van Tongeren, O.F.R. (1995). Data Analysis in Community and Landscape Ecology. Cambridge UK: Cambridge University Press.  Johnson, D., Kershaw, L., Mackinnon, A., & Pojar, J. (1995). Plants of the Western Boreal Forest and Aspen Parkland. Alberta, Canada: Lone Pine Publishing.  Johnson, E.A. (1992). Fire and vegetation dynamics. Cambridge UK: Cambridge University Press.  Johnson, E.A., Morin, H., Miyanishi, K., Gagnon R. & Greene D. F. (2003). A process approach to understanding disturbance and forest dynamics. In P.J. Burton, C. Messier, D.W. Smith 181  and W.L. Adamowicz (Eds.), Towards Sustainable Management of the Boreal forest. National Research Council of Canada Research Press. Ottawa, Ontario. Johnson, J.A.  (1984). Small wood lot management by single tree selection: 21 years results. Northern Journal of Applied Forestry, 1: 69-7.  Johnston, M., Williamson, T., Price, D., Spittlehouse, D., Wellstead, A., Gray, P., Scott, D., Askew, S., & Webber, S. (2006).  Adapting Forest Management to the Impacts of Climate Change in Canada. BIOCAP Research Integration Program Synthesis Paper.  Johnstone J.F., & Chapin, F.S. (2006b). Fire interval effects on successional trajectory in boreal forests of northwest Canada. Ecosystems, 9: 268?277.  Johnstone, J.F., & Chapin, F.S. (2006a). Effects of soil burn severity on post-fire tree recruitment in boreal forest. Ecosystems, 9: 14?31.  Johnstone, J.F., Chapin, F.S., Foote, J., Kemmett, S., Price, K., & Viereck, L. (2004).  Decadal observations of tree regeneration following fire in boreal forests. Canadian Journal of Forest Research, 34: 267?273. Juday, G.P., Barber, V., Duffy, P., Linderhorm, H., Rupp, S., Sparrow, S., Vaganov, E., & Yarie, J. (2005). Forests, Land Management and Agriculture. Chapter 14. Arctic Climate Impact Assessment: Scientific Report. Cambridge, UK: Cambridge University Press. Juday, G.P., Barber, V., Rupp, S., Zasada, J., & Wilmking, M. (2003). A 200-year perspective of climate variability and the response of white spruce in interior Alaska. In D. Greenland, D.G. Goodin & R.C. Smith (Eds.), Climate Variability and Ecosystem Response at Long-Term Ecological Research Sites (pp. 226?250). New York: Oxford University Press.  Kabzem, R., Nemec, A.M., & Farnden, C. (2007).   Growing trembling aspen and white spruce intimate mixtures: Early results (13?17 years) and future projections.  BC Journal of Ecosystems and Management, 8: 1?14. Kabzems, A. (1971). The growth and yield of well stocked white spruce stands in the  mixedwood section of Saskatchewan. Saskatchewan Department of Natural Resources Forestry Branch, Prince Albert. Kalischuk, A.R., Rood S.B., & Mahoney, J.M. (2001). Environmental influences on seedling growth of cottonwood species following a major flood. Forest Ecology and Management, 144: 75 -89. Katila, M., & Tomppo, E. (2002). Stratification by ancillary data in multisource forest inventories employing k-nearest neighbour estimation. Canadian Journal of Forest Research, 32: 1548 ?1561. 182  Kellomaki, S., Petola, H., Bauwens, B., Dekker, M., Mohreh, F., Badeck, F.W., Gracia, C., Sanchez, A., Pla, E., Sebate. S. Linder, M., & Pussinen, A. (2005). European Mitigation adn Adaptation Potentials: Conclusions and Recommendations. In S. Kellomaki & S. Leinonen (Eds.), Silvistart: Management of European Forests under Changing Climatic Conditions. Tiedonantoja Research Notes No. 163. University of Joensuu, Faculty of Forestry.  Kimmins, J.P. (1996). Forest ecology. Foundation for sustainable management (2nd Ed.) New Jersey, USA: Prentice- Hall, Inc.    Kimmins, J.P. (1997). Biodiversity and its relationship to ecosystem health and integrity. Forestry Chronicle, 73: 229?232. Kindt R. (2004). Biodiversity Analysis Functions for R (Biodiversity.R). Nairobi, Kenya: World Agroforestry Center. Kindt, R., & Coe, R. (2005). Tree diversity analysis.manual and software for common statistical methods for ecological and biodiversity studies. Nairobi, Kenya: World Agroforestry Centre (ICRAF). Kindt, R., Damme, P.V., & Simons, A.J. (2006). Tree diversity in western Kenya: using profiles to characterise richness and evenness. Biodiversity and Conservation, 15: 1253?1270. Kirschbaum, M.U.F. (1995). The temperature dependence of soil organic matter decomposition and effect of global warming on soil organic matter decomposition, and the effect of global warming on soil organic C storage. Soil Biology and Biochemistry, 27: 753-760. Klinka, K., Wang, Q., Carter R.E., & Chen, H.Y.H. (1996). Height growth-elevation relationships in subalpine forests of interior British Columbia. Forestry Chronicle, 72: 193?198. Klinka, K., Worrall, J., Skoda, L., & Varga, P. (2000). The distribution and synopsis of ecological and silvical characteristics of tree species of British Columbia?s forests. Canadian Cartographics Ltd, Coquitlam, British Columbia, Canada. Korb, J.E., Covington, W.W., & Fule, P.Z. (2003). Sampling techniques influence understory plant trajectories after restoration: An example from ponderosa pine restoration. Restoration ecology, 11: 504:515.  Krannitz, P., & Kesting, S. (1997). Impacts of climate change on the plant communities of alpine ecosystems. In E. Taylor & B. Taylor (Eds.), Responding to Global Climate Change in British Columbia and the Yukon. Vancouver, B.C. Krasny, M.E.,Vogt, K.A., & Zasada J.C. (1988). Establishment of four Salicaceae species on river bars in interior Alaska. Holarctic Ecology, 11: 210?219. 183  Krebs, C.J., & Boonstra, R. (2001). The Kluane region.  In C.J. Krebs, S. Boutin, & R. Boonstra, (Eds.), Ecosystem dynamics of the boreal forest (pp. 9?24). New York: USA: Oxford University Press. Kurulok, S.E., & Macdonald, S.E. (2004). Impacts of post-fire salvage logging on tree regeneration and plant communities in the mixedwood boreal forest of Alberta. Project reports 2003-2004. Sustainable forest management network. Department of Renewable Resources, University of Alberta. Kurulok, S.E., & Macdonald, S.E. (2007).  Impacts of post-fire salvage logging on understory plant communities of the boreal mixedwood forest 2 and 34 years after disturbances. Canadian Journal of Forest, 37: 2637-2651.   Kuuluvainen, T. (2002). Natural variability of forests as a reference for restoring and managing biological diversity in boreal Fennoscandia. Silva Fennica, 36: 97?125. Kuusela, K. (1990).  Dynamics of Boreal Coniferous Forests. Helsinki: SITRA 112. La Roi, G.H. (1967). Ecological studies in the boreal spruce-fir forests of the North American taiga. I. Analysis of the vascular flora. Ecological Monographs, 37: 229?253. La Roi, G.H., & Stringer, M.H.L. (1976). Ecological studies in the boreal spruce-fir forests of the North American taiga. II. Analysis of the bryophyte flora. Canadian Journal of Forest Research, 54: 621-643. Lambers, H., Chapin, F.S., & Pons, T.L. (1998). Plant physiological ecology. New York: Springer.  Landh?usser, S.M., & Lieffers V.J. (2001). Photosynthesis and carbon allocation of six boreal tree species grown in understory and open conditions. Tree Physiology, 21: 243?250. Larsen, J.A. (1980). The Boreal Ecosystem. New York: Academic Press.  Lawrence, W.T., & Oechel, W.C. (1983). Effects of soil temperature on the carbon exchange of taiga seedlings root respiration. Canadian Journal of Forest Research, 13: 840-49. Le Goff, H., Flannigan, M.D., Bergeron, Y., Leduc, A., Leduc, A., & Logan, K. (2009). Management solutions to face climate change: The example of forest fires. In S. Gauthier, M. Vaillancourt, A. Leduc, L. De Grandpr?, D. Kneeshaw, H. Morin, P. Drapeau, & Y. Bergeron (Eds.), Ecosystem management in the boreal forest. Quebec, Canada: Presses de l?Universite du Quebec.  184  Lee, O.C., Crites, S., Nietfeld, M., Nguyan, V.H., & Stelfox, J.B. (1997). Characteristics and origins of deadwood material in aspen-dominated boreal forests. Ecological Applications, 7: 691?701. Legare, S., Bergeron, Y., & Pare, D. (2002). Influence of forest composition on understory cover of boreal mixedwood forests of western Quebec. Silva Fenn, 36: 353?366. LeMay, V., Taehee Lee, T., Sattler, D., Marshall, P., Robinson, D., & Zumrawi A.A. (2007). Modeling natural regeneration following mountain pine beetle attacks in the southern and central interior of British Columbia.  Mountain Pine Beetle Working Paper 2007-16. Mountain Pine Beetle Initiative PO # 8.35 Natural Resources Canada. Lemieux, C.J., & Scott, D.J. (2005). Climate change, biodiversity conservation and protected areas planning in Canada. The Canadian Geographer, 49: 384-399. Levin, S.A. (2005) Self-organization and the emergence of complexity in ecological systems. Bioscience, 55: 1075-1079.  Lexer?d, N.L., & Eid, T. (2006). An evaluation of different diameter diversity indices based on criteria related to forest management planning. Forest Ecology and Management, 222: 17?28. Lieffers V.J., Landh?usser S.M., & Hogg E. H. (2001). Is the Wide Distribution of Aspen a Result of Its Stress Tolerance? USDA Forest Service Proceedings RMRS-P-18. Lieffers, V.J., Stadt, K.J., & Navratil, S. (1996). Age structure and growth of understory white spruce under aspen. Canadian Journal of Forest Research, 26: 1002?1007. Lindbladh, M., Oswald, W.W., Foster D.R., Faison, E.K., Hou J., & Huang Y. (2007). A late-glacial transition from Picea glauca to Picea mariana in southern New England. Quaternary Research, 67: 502?508. Lindenmayer, D.B., & Franklin, J.F. (1997). Managing Stand Structure as Part of Ecologically Sustainable Forest Management in Australian Mountain Ash Forests. Conservation Biology, 11: 1053?1068. Lindenmayer, D.B., & Noss, R.F.  (2006). Salvage logging, ecosystem processes, and biodiversity conservation. Conservation Biology, 20: 949?958.  Lindenmayer, D.B., Margules, C.R., &   Botkin, D.B. (2000).  Indicators of Biodiversity for Ecologically Sustainable Forest Management. Conservation Biology, 14: 941-950. Linder, P. (1998). Structural changes in two virgin boreal forest stands in central Sweden over 72 years. Scandevian Journal of Forest Research, 13: 451?461. 185  Lloyd, A.H., & Fastie, C.L. (2002). Spatial and temporal variability in the growth and climate response of treeline trees in Aalaska. Climatic Change, 52: 481?509. Looman, J. (1987). The vegetation of the Canadian Prairie Provinces. IV. The woody vegetation. Part 4. Coniferous forests. Phytocoenologia, 15: 289?327. Loope, W.L. (1991). Interrelationship of fire history, land-use history, and landscape pattern within Pictured Rocks National Lakeshore, Michigan. Canadian Field Naturalist, 105:18-28. Loreau, M., Naeem, S., & Inchausti, P. (2002). Biodiversity and Ecosystem Functioning: Synthesis and Perspectives. Oxford: Oxford University Press,  Loreau, M., Naeem, S., Inchausti, P., Bengtsson, J., Grime, J.P., & Hector, A. (2001). Biodiversity and ecosystem functioning: current knowledge and future challenges. Science, 294: 804?808. Lorimer, C.G. (1989). Relative effects of small and large disturbances on temperate hardwood forest structure. Ecology, 70: 565-567. Lorimer, C.G., & Frelish, L.E. (1994). Natural disturbance regimes in old growth northern hardwoods: implication for restoration efforts. Journal of Forestry, 92:33-38. Ma, Z., Penga, C., Zhu, Q., Chen, H., Yu, G., Li, W., Zhou, X., Wang, W., & Zhang, W. (2012). Regional drought-induced reduction in the biomass carbon sink of Canada's boreal forests. Proceeding of National Academyof Science. 109: 2423-2427. www.pnas.org/cgi/doi/10.1073/pnas.1111576109. MacDonald, G.B. (1995). The case for boreal mixedwood management: an Ontario perspective. Forestry Chronicle, 71: 725?734. Macdonald, S.E., Peters, V., Purdy B.G., & Dale, M., (2001). Natural regeneration of white spruce following natural disturbance in the western boreal forest. Project report 2001-13. Sustainable Forest Management Network, University of Alberta. Magurran, A.E. (1988). Ecological diversity and its measurement. London: Croom and Helm.  Magurran, A.E. (2004). Measuring biological diversity. African Journal of Aquatic Science, 29: 285?286. Martin, J.L., & Gower, S.T. (2006). Boreal mixedwood tree growth on contrasting soils and disturbance types. Canadian Journal of Forest Research, 36: 986?995. 186  McCarthy, J.W., & Weetman, G. (2006).  Age and size structure of gap-dynamic, old-growth boreal forest stands in Newfoundland. Silva Fennica, 40: 209?230. McCoy, V.M., & Burn, C.R. (2005). Potential Alteration by Climate Change of the Forest-Fire Regime in the Boreal Forest of Central Yukon Territory.  Arctic, 58: 276? 285. McCullagh, P. & Nelder, J.A. (1989). Generalized Linear Models. London: Chapman and Hall. McGarigal, K., Cushman, S., & Stafford, S. (2000). Multivariate Statistics for Wildlife and Ecology Research. New York: Springer.  McRoberts, R.E., Winter, S., Chirici, G., Hauk, E., Pelz, D.R., Moser, W. K., & Hatfield, M.A. (2008). Large Scale spatial patterns of forest structural diversity. Canadian Journal of Forest Research, 38: 429-438. Meidinger, D., & Pojar, J. (Eds.) (1991). Ecosystems of British Columbia, Research Branch, Ministry of Forests, Province of British Columbia. Victoria, B.C. Messaoud, Y., Bergeron, Y., & Asselin, H. (2007). Reproductive potential of balsam fir (Abies balsamea), white spruce (Picea glauca), and black spruce (P. mariana) at the ecotone between mixedwood and coniferous forests in the Boreal zone of western Quebec. American Journal of Botany, 94: 746?754. Messier, C. & Puettmann, K.J. (2011). Forests as complex adaptive systems: Implications for forest management and modelling. Italian Journal of Forest and Mountain Environments, 66: 249-258. Messier, C., Parent, S., & Bergeron, Y. (1998). Effects of overstory and understory vegetation on the understory light environment in mixed boreal forests. Journal of Vegetation Science, 9: 511?520. Microsoft (2007). Office Excel 2007 for Windows (version 12). Microsoft Corporation. Moeur, M., & Stage, A.R. (1995). Most similar neighbour: an improved sampling inference for natural resources planning. Forest Science, 41: 337-359. Mohammed, G.H. (1999). Non-timber forest products in Ontario: An overview. Forest Research Information Paper No. 145, Ontario Forest Research Institute, Ontario Ministry of Natural Resources, Sault Ste. Marie, Ontario. Montes, F., S?nchez, M., del R?o, M., & Ca?ellas, I. (2005). Using historic management records to characterize the effects of management on the structural diversity of forests. Forest Ecology and Management, 207: 279-292.  187  Mulder, R, Rosie, R., Majisky, J., Peepre, J. & Jenkins, L. (2000). Frances Lake Wildlands  study: A Description of a Candidate Protected Area in the Southeast Yukon. CPAWS-Yukon Research Report # 7. Whitehorse, Yukon. Murray, D.F.  (1971). Notes on the alpine flora of the St. Elias Mountains. Arctic, 24: 301-304. Murray, D.F. (1980). Balsam poplar in Arctic Alaska. In N.W. Rutter, & C.E. Schweger  (eds.), The Ice-free Corridor and Peopling the New World. Canadian Journal of Anthropology, 1: 29-32. Naeem, S. (1998). Species redundancy and ecosystem reliability. Conservation Biology, 12: 39-45. Naeem, S., Thompson, L.J., Lawler, S.P., Lawton J.H., & Woodfin, R.M. (1994). Declining biodiversity can alter the performance of ecosystems. Nature, 368: 734?737. [NCE] Northern Climate ExChange (2006a). Forest Management in a Changing Climate: Building the Environmental Information Base for Southwest Yukon.  Climate Change and Ecosystem Dynamics in the Southwest Yukon (Backgrounder). Northern Climate ExChange, Yukon Research Centre, Yukon College. [NCE] Northern Climate ExChange (2006b). Forest Management in a Changing Climate: Building the Environmental Information Base for Southwest Yukon.  Climate Change and Major Forest Disturbance in the Southwest Yukon (Backgrounder). Northern Climate ExChange, Yukon Research Centre, Yukon College. Neary, D.G., Ryan, K.C., & DeBano, L.F. (Eds.) (2005). Wildland fire in ecosystems: effects of fire on soils and water. General Techical Report. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station.  Nekola, J.C. & White, P.S. (1999). The distance decay of similarity in biogeography and ecology. Journal of Biogeography, 26: 867-878. Niemela, J. (1999). Management in relation to disturbance in the boreal forest. Forest Ecology and Management, 115: 127-134. Nienstaedt, H., & Zasada, J.C. (1990). White Spruce. In R.M. Burns & B. H. Honkala (Eds), Silvics of North America, Volume 1: Conifers (pp. 204?226). Agriculture Handbook 654. Washington, D.C., USA:  United States Department of Agriculture and Forest Service. Nilsson, S.G., & Ericson, L. (1997). Conservation of plant and animal populations in theory and practice. Ecological Bulletin, 46:117-139. Nilsson, S.G., Hedin, J., & Niklasson, M. (2001). Biodiversity and its assessment in boreal and nemoral Forests. Scandinavian Journal of Forest Research, Supplementum, 3: 10?26. 188  Nitschke, C.R. (2009).Vulnerability and Adaptive Capacity of Yukon Tree Species Climate Change: Technical Report. Prepared for the: The University of Northern British Columbia and Forest Management Branch Department of Energy, Mines, Resources, Government of Yukon Noss, R.F (1990). Indicators for monitoring biodiversity: a hierarchical approach. Conservation Biology, 4: 355-364 Noss, R.F. (1983). A Regional Landscape Approach to Maintain Diversity. BioScience, 33: 700-706. Noss, R.F. (1996). Protected areas: how much is enough? In R.G. Wright (Ed.), National parks and protected areas: their role in environmental protection (pp 91?120). Cambridge, UK: Blackwell Science,  Noss, R.F. (2001). Beyond Kyoto: forest management in a time of rapid climate change. Conservation Biology, 15: 578-590. Noss, R.F., & Cooperrider, A.Y. (1994). Saving nature's legacy: protecting and restoring biodiversity. California: Island Press.  Ogden, A. E., & Innes, J.L. (2008). Climate change adaptation and regional forest planning in southern Yukon, Canada. Mitigation and Adaptation Strategy Global Change, 13: 833?861. Ogden, A.E. (2006). Forest Management in a Changing Climate: Building the Environmental Information Base for Southwest Yukon. Overview Report (Appendix C). Northern Climate ExChange, Yukon Research Centre, Yukon College. Ogden, A.E. (2007). Climate change adaptation and sustainable forest management in the Boreal forest. A thesis submitted in partial fulfilment of the requirement for degree of doctor of philosophy. University of British Columbia. Ogden, A.E. (2008). Yukon Forestry Monitoring Programme. Field Manual and Monitoring Protocols. Ministry of Energy, Mines and Resources, Government of Yukon.  Ogden, A.E., & Innes, J.L. (2007). Incorporating climate change adaptation considerations into forest management and planning in the boreal forest. International Forestry Review, 9: 713?733. Okland, R.H., & Eilertsen, O. (1996). Dynamics of understory vegetation in an old-growth boreal coniferous forest, 1988?1993. Journal of Vegetation Science, 7: 747?762. Okland, T., Rydgren, K., Okland, R.H., Storaunet, K.O., & Rolstad, J. (2003). Variation in environmental conditions, understorey species number, abundance and composition among 189  natural and managed Picea-abies forest stands. Forest Ecology and Management, 177: 17?37. Oksanen, J., Kindt, R., & O?Hara, R.B. (2005). Vegan: Community Ecology Package version 1.6-9. http://cc.oulu.fi/~jarioksa/. Oliver, C.D. (1980). Forest development in North America following major disturbances. Forest Ecology and Management, 3: 153-168. Oliver, C.D., & Larson, B.C. (1996). Forest Stand Dynamics. New York, USA: John Wiley & Sons, Inc. Oswald, E.T., & Brown, B.N. (1986). Forest communities in Lake Laberge Ecoregion, Yukon Territory. Canadian Forestry Service, Pacific Forestry Center, Victoria, BC. Oswald, E.T., & Brown, B.N. (1990). Vegetation establishment during 5 years following wildfire in northern British Columbiaand southern Yukon Territory. Forestry Canada, Pacific and Yukon Region, BC-X-320, Victoria, B. C. Palik, B.J., Mitchel, R.J., & Hiers, J.K. (2002).   Modeling silviculture after natural disturbance to sustain biodiversity in the longleaf pine (Pinus palustris) ecosystem: balancing complexity and implementation. Forest Ecology and Management, 155: 347?356. Payette, S., Morneau, C., Sirois, L., & Desponts, M. (1989). Recent fire history of the northern Quebec biomes. Ecology, 70: 656-673. Penafiel, D., Lachat, C., Espinel R., Damme, P.V. & Kolsteren, P. (2011). Systematic Review on the Contributions of Edible Plant and Animal Biodiversity to Human Diets. Ecohealth, 8: 381-399. Perala, D.A. (1984). How endemic injuries affect early growth of aspen suckers. Canadian Journal of Forest Research, 14: 755-762. Perala, D.A. (1990). Quaking Aspen. In R.M. Burns & B.H. Honkala (Eds.), Silvics of North America, Volume 2: Hardwoods (pp.555-569). Agriculture Handbook 654. United States Department of Agriculture and Forest Service, Wanshington DC, USA. Perala, D.A. (1991). ?Does it pay to thin young aspen?? In V.F. Haavisto, C.R. Smith, and C. Mason (Eds.), Space to grow: Spacing and thinning in northern Ontario (pp. 139?144).  Proceedings of a symposium sponsored by Forestry Canada, Ontario Region, and Ontario Ministry of Natural Resources. Joint Report No. 15.  Peters, V.S., Macdonald, S.E., & Dale, M.R.T. (2005). The interaction between masting and fire is key to the white spruce regeneration. Ecology, 86: 1744?175. 190  Peterson, G., Allen, C.R., & Holling, C.S. (1998). Ecological Resilience, Biodiversity, and Scale. Ecosystems, 1: 6?18. Popadiouk, R.V., Chen, H.Y.H., Bowling, C., & Vasiliauskas, S.A. (2003). Compositional and structural characteristics of Ontario?s boreal mixedwoods. Ontario Ministry of Natural Resources, NESI TR-045. Toronto, Ontario: Queen?s Printer for Ontario.  Price, C. & Rind, D. (1994). Possible implications of global climate change on global lightning distributions and frequencies. Journal of Geophysical Research, 99: 10823-10831. Purdon, M., No?l J., Nappi, A., Drapeau, P., Harvey, B., Brais, S., Bergeron ,Y., Gauthier, S. & Greene, D. (2002). The Impact of Salvage-logging after Wildfire in the Boreal Forest: Lessons from the Abitibi. NSERC-UQAT-UQAM Industrial Chair in Sustainable Forest Management, 4th Research Note. Purdy, B.G., Macdonald S. E., & Dale, M. R.T. (2002). The regeneration niche of white spruce following fire in the mixedwood boreal forest. Silva Fennica, 36: 289?306. Qian, H., Klinka, K., Okland, R.H., Krest` 1ov, P., & Kayahara, G.J. (2003). Understorey vegetation in boreal Picea mariana and Populus tremuloides stands in British Columbia. Journal of Vegetation Science, 14: 173?184. Qualtierem, E.J. (2008). Variation in germination response to temperature among collections of three confiners from the mixed wood forest. Thesis Submitted to the College of Graduate Studies and Research in Partial Fulfilment of the Requirements for the Degree of Masters of Science in the Department of Plant Sciences University of Saskatchewan Saskatoon. R Development Core Team (2005). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL  http://www.R-project.org. Reich, P.B., Bakken, P., Carlson, D., Frelich, L.E., Friedman, S.K., & Grigal, D.F. (2001). Influence of logging, fire and forest type on biodiversity and productivity in southern boreal forests. Ecology, 82: 2731?2748. Reyers, B.D., Fairbanks, H.K., Wessles, K.J., & Van Jaarsveld, A.S. (2002). A multi-criteria approach to reserve selection: addressing long-term biodiversity maintenance. Biodiversity and Conservation, 11: 769?793. Ricotta, C. (2003). On parametric evenness measures. Journal of Theoretical Biology, 222:189?197. Roberts, M.R. (2004). Response of the herbaceous layer to natural disturbance in North American forests. Canadian Journal of Botany, 82: 1273?1283. 191  Rogers, P. (1996). Disturbance ecology and forest management: A review of the literature. General Technical Report 336. United States Department of Agriculture.  Forest Ecology and Management, 155: 223?236. Rogers, P. (2002). Using forest health monitoring to assess aspen forest cover change in the southern Rockies ecoregion. Forest Ecology and Management, 155: 223?236. Rolecek, J., Chytry, M., Hajek, M., Lvoncik, S., & Tichy, L. (2007). Sampling design in large-scale vegetation studies: do not sacrifice ecological thinking to statistical purism. Folia Geobotanica, 42: 199-208. Romme, W.H., & Despain, D.G. (1989). The Yellowstone fires. Scientific American, 261: 37-46. Romme, W.H., Knight, D.H., & Yavit, J.B. (1986). Mountain pine beetle outbreaks in the Rocky Mountains: regulators of primary productivity?. The American Naturalist, 127: 484-494.  Rood, S.B., Goater, L.A., Mahoney, J.M., Pearce, C. M., & Smith, D.G. (2007). Floods, fire and ice: disturbance ecology of riparian cottonwoods. Canadian Journal of Botany, 85: 1019-1032. Rosenzweig, C., Casassa, G., Karoly, D.J., Imeson, A., Liu, C., Menzel, A., Rawlins, S., Root, T.L., Seguin, B., Tryjanowski P. (2007). Assessment of observed changes and responses in natural and managed systems In M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden, & C.E. Hanson, (Eds.) Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (pp.79-131).  Cambridge, UK: Cambridge University Press. Rowe, J.S. (1972). Forest regions of Canada. Department of Fisheries and Environment, Canadian Forest Service, Publication 1300.  Saetre, P., Saetre, L.S., Brandtberg, P.O., Lundkvist, H., & Bengtsson, J. (1997). Ground vegetation composition and heterogeneity in pure Norway spruce and mixed Norway spruce - birch stands. Canadian Journal of Forest Research, 27: 2034?2042. Sarkkola, S., Alenius, V., H?kk?, H., Laiho, R., P?iv?nen,  J., & Penttil?, T. (2003). Changes in structural inequality in Norway spruce stands on peatland sites after water-level drawdown. Canadian Journal of Forest Research, 33: 222?231. SAS Institute Inc. 2008.SAS/STAT? 9.2 User?s Guide. Cary, NC: SAS Institute Inc Schlaepfer, R. (1997). Ecosystem-Based Management of Natural Resources: A step towards sustainable development. IUFRO Occasional Paper 6. 192  Schmid, B. (2002). The species richness productivity controversy. Trends in Ecology and  Evolution, 17: 113?114. Scudder, G.G.E. (1997). Environment of the Yukon.  In  H.V. Danks & J.A. Downes (Eds.), Insects of the Yukon (pp.13?57). Biological Survey of Canada (Terrestrial Arthropods) Ottawa. Seymour, R.S., & Hunter, M.L. Jr. (1999).  Principles of ecological  forestry. In  M.L. Hunter (Ed.), Maintaining biodiversity in forest ecosystems. Cambridge, UK: Cambridge University Press. [SFMP] Strategic Forest Management Plan (2004).  Strategic Forest Management Plan for Champagne and Aishihik Territory. Government of Yukon and Champagne and Aishihik First Nation Government. Available at http://www.emr.gov.yk.ca/forestry/ fmp_champagne_aishihik_traditional_territory.htm Shannon, E., & Weaver, W. (1949). The mathematical theory of communication. Urbana: University of Illinois Press. Simard, M.J., Bergeron, Y. & Sirois, L. (1998). Conifer seedling recruitment in a southeastern Canadian boreal forest: the importance of substrate. Journal of Vegetation Science 9: 575?582. Simard, M.J., Bergeron, Y., & Sirois, L. (2003). Substrate and litterfall effects on conifer seedling survivorship in southern boreal stands of Canada. Canadian Journal of Forest Research, 33: 672?681. Simpson, E.H. (1949). Measurement of Diversity. Nature, 163: 688. Singh, T., & Wheaton, E.E. (1991). Boreal forest sensitivity to global warming: implications for forest management in western interior Canada. The Forestry Chronicle, 67: 342-248. Smith, B., & Wilson, J. B. 1996. A consumer's guide to evenness indices. Oikos, 76: 70-82. Smith, D.M., Larson, B.C., Kelty, M.J., & Ashton, P.M.S. (1997). The practice of silviculture: applied forest ecology (9th Ed.). New York, USA: John Wiley and Sons.  Smith, S.L., & Burgess, M.M. (1999). Mapping the response of permafrost in Canada to climate warming. In M. Tranter, R. Armstrong, R.E. Burn, G. Jones, M. Sharp & M. William (Eds.), Interaction between the Cryosphere, Climate and Greenhouse Gases (pp.71-78). Wallingford, Oxforshire: International Association of Hydrological Sciences Publication.   Smith,W., & Grassle, J.F. (1977). Sampling Properties of a family of diversity measures. Biometrics, 33: 283-292. 193  Solomon, S., Qin, D., Manning, M., Alley, R.B., Berntsen, T., Bindoff, N.L, Chen, Z., Chidthaisong, A., Gregory, J.M., Hegerl, G.C., Heimann, M., Hewitson, B., Hoskins, B.J., Joos, F., Jouzel, J., Kattsov, V., Lohmann, U., Matsuno, T., Molina, M., Nicholls, N.,  Overpeck, J., Raga, G., Ramaswamy, V.,  Ren, J., Rusticucci, M., Somerville, R., Stocker, T.F., Whetton, P., Wood, R.A.and Wratt, D. (2007): Technical Summary. In S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, B.Averyt, M. Tignor and H.L.Miller (Eds.), Climate Change 2007: The Physical Science Basis. Contribution of WorkingGroup I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press. Sorensen., W. (1968). Size of aspen crop trees little affected by initial sucker density. United States Department of Agriculture, Forest Service Research Note. Spittlehouse, D.L. (1996). Assessing and responding to the effects of Climate Change on Forest Ecosystems. In R.G. Lawford, P.B. Alaback & E. Fuentes (Eds.), High latitude Rain forests and Associated Ecosystems of the West Coast of the Americas (pp.306-319). New York:  Springer Verlag.  Spittlehouse, D.L. (1997). Forest management and climate change. In E. Taylor & B. Taylor (Eds), Responding to global climate change in British Columbia and Yukon (pp. 24: 1-8). Environment Canada, Vancouver, BC. Spittlehouse, D.L., & Stewart, R.B. (2003). Adapting to climate change in forest management. Journal of Ecosystems and Management, 4: 7-17. Stanek, W., Alexander, K., & Simmons, C.S. (1981). Reconnaissance of Vegetation and Soils along the Dempster Highway, Yukon Territory: I. Vegetation Types. Canadian Forestry Service, Pacific Forest Research Centre Information Report BC-X-217. Staudhammer, C.L., & LeMay, V.M. (2001).  Introduction and evaluation of possible indices of stand structural diversity. Canadian Journal of Forest Research, 31: 1105?1115. Stewart, B.J., Neily, P.D., Quigley, E.J., Duke, A.P., & Benjamin, L.K. (2003). Selected Nova Scotia old-growth forests: Age, ecology, structure, scoring. Forestry Chronicle, 79: 632?644. Stewart, J.D., Landh?usser, S.M., Stadt, K.J., & Lieffers, V.J. (2001). Predicting natural regeneration of white spruce in boreal mixedwood understories. The Forestry Chronicle, 77: 1006?1013. Stewart, R.B, Wheaton, E., & Spittlehouse, D. (1998). Climate change: Implications for the boreal forest. In A.H. Legge & L.L. Jones, (Eds.), Emerging Air Issues for 21st Century.  194  The need for multidisciplinary management. Alberta Canada; Proceedings of an International Speciality Conference.  Stocks, B.J., & Lynham, T.J. (1996). Fire weather climatology in Canada and Russia. In: J.G. Goldammer and V.V. Furyaev, (Eds.), Fire in Ecosystems of Boreal Eurasia (pp. 481?494). Boston: Kluwer Academic Publishers.  Stocks, B.J., Fosberg, M.A., Lynham, T.J., Mearns, L., Wotton, B.M., Yang, Q., Jin, J.Z., Lawrence, K., Hartley, G.R., Mason, J.A., & McKenney, D.W. (1998). Climate change and forest fire potential in Russian and Canadian boreal forests. Climatic Change, 38: 1?13. Stoeckeler, J.H. (1960). Soil factors affecting the growth of quaking aspen forests in the United States. University of Minnesota:  Agricultural Statistics Technical Bullelin 233. Stone J.N., & Porter J.L. (1998).  What is forest stand structure and how to measure it? Northwest Science, 72: 25-26. Sturtevant, B.R., Bissonette, J.A., Long, J.N., & Roberts, D.W. (1997). Coarse woody debris as a function of age, stand structure, and disturbance in boreal Newfoundland. Ecological Application, 7: 702?712. Sutter, R. D. (1996). Monitoring. In D. A. Falk, C. I. Millar, & M. Olwell (Eds.), Restoring diversity?strategies for reintroduction of endangered plants (pp.235?264). Washington, D.C: Island Press. Tabachnick, B.G., & Fidell, L.S. (1996). Using Multivariate Statistics (5thed.). Boston, USA: Person International Edition. Taillie, C. (1979). Species equitability: a comparative approach. In J.F. Grassle, G.P. Patil G.K. Smith, & C. Taillie (Eds.), Ecological Diversity in Theory and Practice (pp. 51?62). Fairland: International Cooperative Publishing House.  Takahashi, K., Homma, K., Vetrova, V.P., Florenzev, S., & Hara, T. (2001). Stand structure and regeneration in a Kamchatka mixed boreal forest. Journal of Vegetation Science, 12: 627?634. Ter Braak, C.J.F. (1986). Canonical correspondence analysis: A new eigenvector technique for multivariate direct gradient analysis. Ecology, 67: 1167-1179. Ter Braak, C.J.F., & Verdonschot, P.F.M. (1995). Canonical correspondence analysis and related multivariate methods in aquatic ecology. Aquatic sciences, 57: 255-289. Tilman, D. (1996). Biodiversity: population versus ecosystem stability. Ecology, 77: 350?363. 195  Tilman, D., Wedin, D., & Knops, J. (1996). Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature, 379: 718-720. Timoney, K.P. & Peterson, G. (1996). Failure of natural regeneration after clear-cut logging in Wood Buffalo National Park, Canada. Forest Ecology and Management, 87: 89-105. Timoney, K.P. & Robinson, A.L. (1996).  Old-growth white spruce and balsam poplar forests of the Peace River Lowlands, Wood Buffalo National Park, Canada: development, structure, and diversity. Forest Ecology and Management, 81: 179-196. Tothm?r?sz, B. (1995). Comparison of different methods for diversity ordering. Journal of Vegetation Science, 6: 283-290. Tryon, P.R., & Chapin, F.S. (1983). Temperature control over root growth and root biomass in taiga forest trees. Canadian Journal of Forest Research., 13: 827-33. Tyree, M.T., Kolb, K.J., Rood, S.B., & Pati?o, S. (1994). Vulnerability to drought-induced cavitation of riparian cottonwoods in Alberta: a possible factor in the decline of the ecosystem? Tree Physiology, 14: 455?466. Van Cleve, K., Barney, R., & Schlentner, R. (1981). Evidence of temperature control of production and nutrient cycling in two interior Alaska black spruce ecosystems. Canadian Journal of Forest Research, 11: 258?273. Van Cleve, K., Chapin, F.S., Dyrness, C.T., & Viereck, L.A. (1991).  Element Cycling in Taiga Forests State-Factor Control. Bioscience, 41: 78? 88. Van Cleve, K., Oliver, L., Schlenter, R., Viereck, L.A., & Dryness, C.L. (1983). Productivity and nutrient cycling in taiga forest ecosystems. Canadian Journal of Forest Research, 13: 747-766. Varga, P., Chen, H.Y.H., & Klinka, K. (2005). Tree size diversity between single and mixed species stands in three forest types in western Canada. Canadian Journal of Forest Research, 35: 593?601. Vellak, K., Paal, J., & Liira, J. (2003). Diversity and distribution pattern of bryophytes and vascular plants in a boreal spruce forest. Silva Fennica, 37: 3?13. Venables, W.N., & Ripley, B.D. (2002). Modern Applied Statistics with S (4th ed). New York, USA: Springer. Viereck, L.A., & Foote, J.M., (1970). The status of Populus balsamifera and P. trichocarpa in Alaska. The Canadian Field Naturalist, 84: 169?173. 196  Viereck, L.A., Dryness, C.T., & Foote, M.J. (1993). An overview of vegetation and soils and forest productivity in selected forest types in interior Alaska. Canadian Journal of Forest Research, 13: 703-720.  Viereck, L.A., Dryness, C.T., VanCleve, K., & Foote, M.J. (1983).Vegetation, soils and forest of the floodplain ecosystems of the Tanana River, interior Alaska. Canadian Journal of Forest Research, 23: 889-898. Vitousek, P.M., Matson, P.A., & Turner, D.R. (1988). Elevational and age gradients in Hawaiian montane rain forest: Foliar and soil nutrients.  Oecologia, 77: 565-570. Vitousek, P.M., Mooney, H.A., Lubchenko, J., & Melillo, J.M. (1997). Human domination of Earth?s ecosystems. Science, 277: 494?499. Wade, D.D., Lunsford, J.D., Dixon, M.J., Mobley, & Hugh, E. (1989). A guide for prescribed fire in southern forests. Technical Publication no. 11. Atlanta: U.S.: Department of Agriculture, Forest Service, Southern Region. Waeber, P. (2012). Integrated dual filter framework for forest management planning: a case study in the Southwestern Yukon.  A thesis submitted in partial fulfilment of the requirement for degree of doctor of philosophy. University of British Columbia Wali, M.K., & Krajina, V.J. (1973). Vegetation-environment relationships of some sub-boreal spruce zone ecosystems in British Columbia. Vegetation, 26: 237?281. Wang G.G., & Klinka, K. (1996). Use of synoptic variables in predicting white spruce site index. Forest Ecology and Management, 80: 95-105. Wang, C.G. (1995). White spruce site index in relation to soils, understory vegetation and foliar nutrients. Canadian Journal of Forest Research, 25: 29-38. Wang, G.G. (2003).  Early regeneration and growth dynamics of Populus tremuloides suckers in relation to fire severity. Canadian Journal of Forest Research, 33: 1998?2006. Wang, G.G., & Kemball, K.J. (2005). Balsam fir and white spruce seedling recruitment in response to understory release, seedbed type, and litter exclusion in trembling aspen stands. Canadian Journal of Forest Research, 35: 667?673. Wang, G.G., Marshal, P.L., & Klinka, K., (1994). Height growth pattern of white spruce in relation to site quality?  Forest Ecology and Management, 68: 134?144. Wang, T., Hamann, A., Spittlehouse, D., & Aitken, S.N. (2006). Development of scale-free climate data for western Canada for use in resource management. International Journal of Climatology, 26: 383-397. 197  Wardle, D.A., Bonner, K.I., & Nicholson. K.S. (1997). Biodiversity and plant litter: experimental evidence which does not support the view that enhanced species richness improves ecosystem function. Oikos, 79: 247?258. Weber, M.G., & Van Cleve, K. (1981). Nitrogen Dynamics in the Forest Floor of Interior Alaska Black Spruce Ecosystems. Canadian Journal of Forest Research, 11: 743?751. Wein, R.W., & MacLean, D.A. (1983).  The role of fire in northern circumpolar ecosystems.  Chichester, UK:  John Wiley and Sons. Weingartner, D.H. (1991). Spacing and thinning in aspen and mixedwood: Some thoughts, theories and observations?. In V.F. Haavisto, C.R. Smith, and C. Mason (Eds), Space to grow: Spacing and thinning in northern Ontario (pp. 145?149). Proceedings of a symposium sponsored by Forestry Canada. Ontario, Canada: Ontario Region and Ontario Ministry of Natural Resources Joint Report No. 15.   White, J. (1980). Demographic factors in populations of plants. In O.T. Solbrig, (ed.), Demography and evolution in plant populations, Vol. 15 (pp. 21?48). Oxford, UK: Blackwell Scientific Publications.  White, P.S. & Pickett, S.T.A. (1985). Natural disturbances and patch dynamics: an introduction. In White, P.S. & Pickett, S.T.A (Eds.), The ecology of natural disturbance and patch dynamics. Orlando FL.: Academic Press. Whittaker, R.H. (1972). Evolution and measurement of species diversity. Taxon, 21: 213-251. Whittle, C.A., Duchesne, L.C., & Needham, T. (1997). The impact of broadcast burning and fire severity on species composition and abundance of surface vegetation in a jack pine (Pinus banksiana) clear-cut. Forest Ecology and Management, 94: 141?148. Wiersma, Y.E., & Urban, D.L. (2005). Beta diversity and nature reserve system design in the Yukon, Canada. Conservation Biology, 19: 1262?1272.  Wilmking, M., Juday, G.P., Barber, V.A., & Zald, H.S. (2004). Recent climate warming forces contrasting growth responses of white spruce at treeline in Alaska through temperature thresholds. Global Change Biology, 10: 724?1736. Wilsay, B.J., & Potvin, P. (2000). Biodiversity and ecosystem functioning: importance of species evenness in an old field. Ecology, 81: 887?892. Wilson, J.B. (1994). The ?intermediate disturbance hypothesis? of species coexistence is based on patch dynamics.  New Zealand Journal of Ecology, 18: 176-181.  198  Wimberly, M.C., & Spies, T.A. (2001). Influences of environment and disturbance on forest patterns in coastal Oregon watersheds. Ecological Society of America, 82: 1443-1459. Wirth, C., Lichstein, J.W., Dushoff, J., Chen A., & Chapin F.S. (2008). White spruce meets black spruce: dispersal, post-fire establishment, and growth in a warming climate. Ecological Monographs, 78: 489?505. Yarie, J. (2008). Effects of moisture limitation on tree growth in upland and floodplain forest ecosystems in interior Alaska. Forest Ecology and Management, 256: 1055?1063. Yoshikawa, K., Bolton, W.R., Romanovsky, V.E., Fukuda, M., & Hinzman, L.D. (2003). Impacts of wildfire on the permafrost in the boreal forests of Interior Alaska. Journal of Geophysical Research, 108: 1-14.  Zahl, S, (1977).  Jackknifing an index of biodiversity. Ecology, 58: 907-913. Zasada, J.C., & Densmore, R. (1980). Alaska willow and balsam poplar seed viability after three years. Tree Planters' Notes, 31: 9?10. Zasada, J.C., Viereck L.A., Foote M.J., Parkenson R.H., Wolff J.O., & Lankford L.A. (1981). Natural regeneration of balsam poplar following harvesting in the Susitna Valley, Alaska. The Forestry Chronicle, 57: 57?65. Zasada, J.S., & Phipps, H.N. (1990). Balsam Poplar. In R.M. Burns & B.H. Honkala (Eds.), Silvics of North America, Volume 2: Hardwoods. Agriculture Handbook 654. United States Department of Agriculture and Forest Service, Washington D.C., USA.   199  Appendix A: Scientific and common names of vegetation, their code names and type13  Scientific name Code Type14  Common name Achillea millefolium L.s.l. Ach_mil H  Common yarrow Aconitum delphinifolium DC. Aco_del H  Monkshood Alnus incana (L.) Moench Aln_inc S  Grey alder Aquilegia formosa Fisch. Ex DC. Aqu_for H  Red Columbine Arctostaphylos rubra (Rehd. & Wils.) Fern. Arc_rub H  Red Berry Arctostaphylos uva-ursi (L.) Spreng. s.l. Arc_uva H  Kinnikinnick (bearberry) Arnica cordifolia Hook. Arn_cor H  Heart-leaved arnica Aster alpines L.ssp. Ast_alp H  Aster Aulacomnium palustre (Hedw.) Schwagr. Aul_pal B  Big moss or glow moss Betula glandulosa Michx. Bet_gla S  Ground birch Carex spp.  Car_spp H  Sedge Castilleja spp. Cas_spp H  Indian paint-brush Cetraria pinastri (Scop.) A. Gray Cet_pin L  Moonshine cetraria Cladina mitis (Sandst.) Hustich Clad_mit L  Green reindeer lichen Cladonia spp. Cla_spp L  Pixie cup lichen Dichodontium pellucidum (Hedw.) Schimp. Dic_pel B  Wet rock moss Dicranum polysetum Sw. Dic_pal B  Electric eels moss Dicranum scoparium Hedw.. Dic_sco B  Broom moss Elymus glaucus Buckl. Ely_gla H  Blue wild rye Elymus innovates (Beal) Pilg. Ely_inn H  Fuzzy spike wild rye Empetrum nigrum L. Emp_nig H  Black crow berry Epilobium angustifolium L.s.l. Epi_aug H  Fireweed Equisetum arvense L. Equ_arv H  Common horsetail Erigeron peregrines (Pursh) Eri_per H  Subalpine daisy Festuca spp. Fes_spp H  Fescue Flavocetraria cucullata (Bellardi) Karnefelt & A. Thell  Fla_cuc L  Curled snow lichen Geocaulon lividum (Richards.) Geo_liv H  Bastard toad flax Hedysarum mackenzii (Richards) Hed_mac H  Wild sweet pea                                                  13 Species were identified and verified using local flora books (Cody, 2000; Johnson et al., 1995) and online USDA database (plants.usda.gov). However some species could not be identified at species level due to lack of flowering during data collection.  14  Note: T= Trees; H = Herbs; S = Shrubs; L = Lichen; B = Bryophytes  200  Scientific name Code Type* English name Hookeria lucens (Hedw.) Sm. Hoo_luc B  Clear moss Hordeum brachyantherum Nevski Hor_bra H  Meadow barley Hyacinthoides non-scripta (L.) Chouard ex Rothm. Hya_non H  English blue bell Hypogymnia physodes (L.) Nyl. Hyp_phy L  Hooded tube lichen Juniperus communis L.s.l. Jun_com S  Common Juniper Ledum groenlandicum Oeder  Led_gro S  Labrador tea Linnaea borealis, L. Lin_bor H  Twin flower Lupinus arcticus Wats. Lup_arc H Arctic lupin Mimulus guttatus DC. Mim_gut H Monkey flower Oxycoccus microcarpus Turcz. Oxy_mic S Small bog cranberry Oxytropis campestris (L.) DC. Oxy_can H Field locoweed Peltigera aphthosa (L.) Willd. Pel_aph L Freckled lichen Peltigera canina (L.) Willd. pel_can L Dog lichen Petastites frigidus (L.) Fries Pet_fri H Palmate leaved Coltsfoot Picea glauca (Moench) Voss Pic_gla T White spruce Pleurozium schreberi (Brid.) Mitt.  Ple_sch B Red stemmed feather moss Polytrichum commune (Hedw.) Pol_com B Common hair cap moss Populus balsamifera L. Pop_bal T Balsam poplar Populus tremuloides Michx. Pop_tre T Trembling aspen Potentilla anserine L.s.l.  Pot_ans S Silverweed Potentilla fruiticosa L. Pot_fru H Shrubby Cinquefoil Pulsatilla patens (L.) Mill. Pul_pat H     Pasque flower Pyrola asarifolia Michx. Pyr_asa H Pink wintergreen Pyrola picta Sm. Pyr_pic H Whiteveined wintergreen Rhytidiopsis robusta (Hook.) Broth. Rhy_rob B Pipe cleaner moss Rosa acicularis  Lindl. s.l. Ros_aci H Prickly rose Rubus arcticus L. ssp. Rub_arc H Dwarf nagoonberry Salix spp.  Sal_spp T Willow Sanicula marilandica L. San_mar H Black snakeroot Saxifraga tricuspidata Rottb. Sax_tri H Three toothed saxifrage Sedum lanceolatum Torr. Sed_lan H Stonecrop Shepherdia canadensis (L.) Nutt. She_can H Soapberry, Soopolallie Solidago multiradiata Ait.  Sol_mul H Northern goldenrod Sphagnum papillosum Lindb. Sph_pap B Fat bog moss Taraxacum officinale Weber ex Tar_off H Common dandelion Tortula ruralis (Hedw.) G. Gaertn. Tor_rur B Sidewalk moss Trillium grandiflorum (Michx.) Salisb. Tri_gra H Trillium Usnea hirta (L.) F.H. Wigg. Usn_hir L Sugary beard Vaccinium caespitosum Michx. Vac_cae S Dwarf blueberry Vaccinium vitis-idaea L. Vac_vit S Lingonberry Viola adunca J.E. Smith Vio_adu S Hook-spur violet     201  Appendix B: Descriptive analysis of collected variables Table B-1: Descriptive statistics of environmental variables  Elevation (m) Slope (%) Total soil  depth (cm) Organic  depth (cm) Average  soil moisture (%) Mean 817.72 12.32 48.53 7.62 21.19 Standard Error 13.48 1.50 2.26 0.80 1.10 Median 781 8.75 44.5 4.75 19.66 Mode 781 0 23 6 19.66 Standard Deviation 127.90 14.24 21.46 7.58 10.44 Skewness 0.58 2.17 0.88 1.58 0.98 Range 507 74.75 95 30 47.47 Minimum 651 0 15 0 4.29 Maximum 1158 74.75 110 30 51.76 Sum 73595 1108.52 4368 685.5 1907.39 Count 90 90 90 90 90  Table B-2: Descriptive statistics of climate variables  MAT (OC) MAP (mm) TMXwt (OC) TMXsm (OC) PPTwt (mm) PPTsm (mm) DD0 (OC) DD5 (OC) NFFD (no) Mean -1.7 319.6 -9.7 18.0 60.5 129.9 2098.5 743.3 117.0 Standard Error 0.0 2.0 0.2 0.1 1.4 1.8 17.3 4.0 0.5 Median -1.9 316.5 -10.89 18.235 69.15 122.6 2214 747.5 115 Mode -2 346 -10.94 17.3 44.4 111.5 1894 790 114 Standard Deviation 0.42 19.12 1.51 0.66 13.55 17.52 164.13 38.29 4.54 Skewness -0.09 0.30 0.21 -0.42 -0.03 0.52 -0.11 -0.40 -0.20 Range 1.3 80 3.96 2.34 34.5 64.3 424 161 15 Minimum -2.4 290 -11.44 16.5 44 109.4 1892 632 108 Maximum -1.1 370 -7.48 18.84 78.5 173.7 2316 793 123 Sum - 148.5 28761 -874.7 1618.96 5440.6 11687.7 188868 66895 10530 Count 90 90 90 90 90 90 90 90 90 [Note: MAT = mean annual temperature; MAP = Mean annual precipitation; TMXsm = Mean summer maximum temperature; TMXwt = Mean winter maximum temperature; PPTsm = summer precipitation; PPTwt = Winter precipitation; DD0 = degree days below 0 C; DD5 = degree days above 5 C; NFFD = Number of frost free days]   202  Table B-3: Descriptive statistics of stand structure variables  Spavht (m) Spavdia (cm) Atavht (m) Atavdia (cm) Poavht (m) Poavdia (cm) AtBA (m2) PoBA (m2) SpBA (m2) ToBA (m2) Mean 9.01 13.37 4.68 7.24 2.35 3.98 3.90 0.59 16.07 17.26 Standard Error 0.46 0.60 0.39 0.64 1.05 1.98 0.97 0.46 1.49 1.48 Median 9.33 14.36 4.55 7.21 1.54 2.11 2.06 0.14 12.76 15.04 Standard Deviation 4.07 5.28 2.20 3.65 2.11 3.95 5.47 0.92 13.19 13.24 Skewness -0.18 -0.58 0.34 0.39 1.69 1.99 3.72 2.00 1.08 0.91 Range 16.06 22 7.98 13.65 4.525 8.125 29.84 1.87 52.19 52.19 Minimum 0.56 0.63 1.33 2 0.9 1.775 0.13 0.10 0.01 0.01 Maximum 16.61 22.63 9.3 15.65 5.425 9.9 29.96 1.96 52.20 52.20 Sum 702.53 1043.23 149.68 231.58 9.4 15.9 124.74 2.34 1253.44 1380.52 Count 78 78 32 32 4 4 32 4 78 80 [Note: Spavht = white sprue average height; Spavdia = White spruce average diameter; Atavht = aspen average height; Poavht = balsam poplar average height; Poavdia = balsam poplar average diameter; AtBA = aspen basal area; PoBA = balsam poplar basal area; SpBA = white spruce basal area; ToBA = total basal area]   203  Appendix C: Multiple Co-inertia analysis outputs15 pseudoeig: 20 pseudo eigenvalues 9.923, 2.795, 1.382, 1.13 0.9023 ... call: mcoa(X = KTABLAS, option = "uniform")    data.frame  nrow  ncol  content                                  1  $SynVar     90    2     synthetic scores                         2  $axis       24    2     co-inertia axis                          3  $Tli        360   2     co-inertia coordinates                   4  $Tl1        360   2     co-inertia normed scores                 5  $Tax        16    2     inertia axes onto co-inertia axis        6  $Tco        24    2     columns onto synthetic scores            7  $TL         360   2     factors for Tli Tl1                      8  $TC        24    2     factors for Tco                          9  $T4         16    2    factors for Tax                          10 $lambda     4     2     eigen values (separate analysis)         11 $cov2       4     2    pseudo eigen values (synthetic analysis) [1] 9.922991e+00 2.795277e+00 1.382351e+00 1.129755e+00 9.022824e-01 6.741350e-01 4.529459e-01 4.277802e-01 3.246419e-01 5.597977e-02 2.220464e-02  [12] 5.840040e-03 2.665416e-03 1.176611e-03 3.327083e-04 3.347987e-05 2.058054e-05 1.336260e-05 1.163953e-27 1.651378e-31 Array number 1 Enviro Rows 90 Cols 18    Iner  Iner+  Var   Var+   cos2  cov2  1 9.824 9.824  9.824 9.824  0.997 9.799 2 2.794 12.618 2.794 12.618 1     2.794 Array number 2 Sw Rows 90 Cols 2    Iner  Iner+ Var   Var+  cos2  cov2  1 1.891 1.891 1.207 1.207 0.035 0.042 2 0.109 2     0.793 2     0     0                                                      15 The fonts in the appendixes are same as the software outputs.  204  Array number 3 At Rows 90 Cols 2    Iner  Iner+ Var   Var+  cos2  cov2  1 1.976 1.976 1.969 1.969 0.024 0.048 2 0.024 2     0.031 2     0.033 0.001  Array number 4 Ab Rows 90 Cols 2    Iner  Iner+ Var   Var+  cos2  cov2  1 1.961 1.961 1.956 1.956 0.017 0.034 2 0.039 2     0.044 2     0.002 0      MCOA and Axis                  Axis1         Axis2 ele        -0.27201273  -0.233031651 slope      -0.16339181  -0.224227584 Cosasp      0.06937639  -0.371011447 Sinasp      0.18241733   0.072623398 mat        -0.30332398   0.138265752 TMXwt     -0.31811243  -0.008857122 TMXsm      0.29497003   0.165895527 TMNwt     -0.31757522   0.031986812 TMNsm     -0.29325288   0.199443093 PPTwt       0.29119171  -0.205323731 PPTsm      -0.29989281  -0.144754754 DD0         0.31745294  -0.034924514 DD5         0.16882948   0.391733215 NFFD       -0.28979661   0.199601497 Soil_D      0.01681583   0.359387913 Org_D      -0.04475665  -0.145826098 SoilMoist -0.04699352   0.028980809 SwHt        0.99309842   0.117283967 SwDBH      0.11728397  -0.993098420 AtHt        0.66402320   0.747711968 AtDBH       0.74771197  -0.664023202 AbHt       -0.74113007  -0.671361470 AbDBH     -0.67136147  0.741130068   205  Appendix D: Canonical correlation outputs Table D-1: Relationship among site characteristics and tree regeneration in Southwest Yukon                       The CANCORR Procedure Canonical Correlation Analysis  Adjusted     Approximate    Squared       Canonical       Canonical       Standard      Canonical        Correlation    Correlation      Error     Correlation  1       0.997833       0.996933       0.000478       0.995671 2       0.827347       0.735080       0.034841       0.684503 3       0.753143       0.594060       0.047792       0.567225 4       0.727287        .             0.052019       0.528946 5       0.687891        .             0.058176       0.473195 6       0.614102       0.517309       0.068785       0.377121   ? Test of H0: The canonical correlations in the current row and all that follow are zero   ? Eigenvalues of Inv(E)* = CanRsq/(1-CanRsq)                     Likelihood    Approximate Eigenvalue    Difference    Proportion    Cumulative    Ratio    F Value    Num DF    Den DF    Pr > F  1      230.0264      227.8568     0.9741        0.9741    0.00009135      5.77       186    280.03    <.0001 2        2.1696        0.8589     0.0092        0.9833    0.02110485      1.87       150    237.53    <.0001 3        1.3107        0.1878     0.0056        0.9889    0.06689397      1.63       116    193.33    0.0014 4        1.1229        0.2247     0.0048        0.9936    0.15456970      1.52        84    147.47    0.0133 5        0.8982        0.2928     0.0038        0.9974    0.32813575      1.38        54       100    0.0823 6                0.6054                   0.0026        1.0000    0.62287859      1.19        26        51    0.2937    Table D-2: Multivariate statistics and F approximations    S=6    M=12    N=22  Statistic                        Value    F Value    Num DF    Den DF    Pr > F  Wilks' Lambda               0.00009135       5.77       186    280.03    <.0001 Pillai's Trace              3.62666098       2.51       186       306    <.0001 Hotelling-Lawley Trace    236.13325553      56.44       186    195.03    <.0001 Roy's Greatest Root       230.02640465     378.43        31        51    <.0001 206  Table D-3: Correlations between the tree regeneration and their canonical variables reg1            reg2         reg3              reg4           reg5          reg6  Spsed         0.0419          0.2362       0.4730           0.5134       -0.6663        -0.1059 AtSed        -0.0363          0.9022       0.2253          -0.0253        0.3121         0.1895 Popsed        0.8767           0.0609       0.3153          -0.1612        -0.1006         -0.3037 Atsap        -0.0291          0.3333      -0.2644           0.5072        0.4259        -0.6161 Spsap        -0.0208           -0.4968       0.5042            0.4950         0.4242           0.2712 Popsap        0.9997         -0.0001      -0.0165          -0.0011       -0.0070         0.0158    Table D-4: Correlations between the site characteristics and their canonical variables                 site1         site2         site3         site4         site5         site6  lat           -0.1514       -0.1311       -0.1865       -0.0902        0.2108       -0.1359 long          -0.1228       -0.1836       -0.3133       -0.2356        0.2286        0.0469 ele           -0.0424       -0.0281       -0.4580       -0.3307        0.1746        0.0277 slope         -0.0886       -0.1028       -0.3031       -0.3895        0.1588       -0.1017 aspcon        -0.0492       -0.1146       -0.0733       -0.1197       -0.0540       -0.1168 mat           -0.1395       -0.2103       -0.1650       -0.1318        0.1476        0.0619 map            0.1372        0.0488       -0.1557       -0.2280       -0.1722        0.2453 TMXwt         -0.1073       -0.1700       -0.2972       -0.2309        0.1835        0.0346 TMXsm          0.0711        0.1063        0.4073        0.2968       -0.1788       -0.0223 DD0            0.1142        0.1987        0.2571        0.2105       -0.1773       -0.0568 DD5            0.0159        0.0117        0.4812            0.3213       -0.1059        0.0163 NFFD          -0.1206       -0.1919       -0.1457       -0.1235        0.2005        0.0777 BT            -0.0592       -0.2810        0.1855        0.2810       -0.1207        0.3076 FR            -0.0097        0.4852        0.5213       -0.1893       -0.1978       -0.2882 OLDFR         -0.0523       -0.0789       -0.0476        0.1359        0.3656       -0.0052 SALV           0.2310        0.0204       -0.1751        0.1323        0.1372       -0.1197 TD            -0.0794        0.2300        0.1348        0.0303        0.2725       -0.2550 OD            -0.0820       -0.3870        0.0115        0.1924       -0.0314        0.2279 T1M           -0.0951       -0.3748        0.1941       -0.1305        0.1132       -0.1001 AM            -0.0757       -0.4142        0.0724       -0.1229        0.0954       -0.0086 CL            -0.0596       -0.1310        0.3632       -0.1462        0.0046       -0.1091 SA             0.1347        0.2030       -0.1773        0.1223        0.2253       -0.1417 STT           -0.0747       -0.4883        0.1873        0.1165        0.1085        0.4819 ATT           -0.0452        0.0300       -0.2425       -0.0386        0.2119       -0.1484 PTT            0.9156       -0.0056       -0.0385        0.0317       -0.1224        0.0406 TCC           -0.0871       -0.3114        0.1672        0.1641        0.3244        0.4388 TshrubC        0.1262        0.1475       -0.1427        0.1066        0.3551        0.0884 TherbC        -0.0399       -0.0846        0.1690       -0.0613       -0.1435        0.1683 PoBA           0.9923       -0.0069       -0.0247        0.0156       -0.0441        0.0291 SpBA          -0.0184       -0.2797        0.1050       -0.0101        0.0345        0.5542 AtBA          -0.0416       -0.0331       -0.1798       -0.0886        0.1092       -0.0537 ToBA          -0.0141       -0.2848        0.0538       -0.0340        0.0633        0.5320 207    Table D-5: Correlations between the site characteristics and the canonical variables of the tree regeneration     reg1                          reg2                          reg3                        reg4                         reg5                     reg6 lat              -0.1510          -0.1085          -0.1404          -0.0656           0.1450          -0.0835 long             -0.1225          -0.1519          -0.2360          -0.1713           0.1572           0.0288 ele              -0.0423          -0.0232          -0.3449          -0.2405           0.1201           0.0170 slope            -0.0884          -0.0850          -0.2283          -0.2833           0.1093          -0.0624 aspcon           -0.0491          -0.0948          -0.0552          -0.0871          -0.0372          -0.0717 mat              -0.1392          -0.1740          -0.1242          -0.0959           0.1015           0.0380 map               0.1369           0.0404          -0.1173          -0.1658          -0.1185           0.1506 TMXwt            -0.1071          -0.1406          -0.2239          -0.1679           0.1262           0.0213 TMXsm             0.0709           0.0880           0.3068           0.2159          -0.1230          -0.0137 DD0               0.1140           0.1644           0.1936           0.1531          -0.1220          -0.0349 DD5               0.0158           0.0097           0.3624           0.2336          -0.0729           0.0100 NFFD             -0.1203          -0.1588          -0.1098          -0.0898           0.1379           0.0477 BT               -0.0591          -0.2325           0.1397           0.2043          -0.0830           0.1889 FR               -0.0097           0.4014                         0.3926          -0.1377          -0.1360          -0.1770 OLDFR            -0.0521          -0.0653          -0.0358           0.0989           0.2515          -0.0032 SALV              0.2305           0.0169          -0.1319           0.0962           0.0944          -0.0735 TD               -0.0792           0.1903           0.1015           0.0221           0.1875          -0.1566 OD               -0.0818          -0.3202           0.0086           0.1399          -0.0216           0.1400 T1M              -0.0949          -0.3101           0.1462          -0.0949           0.0779          -0.0615 AM               -0.0755          -0.3427           0.0545          -0.0894           0.0656          -0.0053 CL               -0.0595          -0.1084           0.2735          -0.1063           0.0032          -0.0670 SA                0.1344           0.1680          -0.1335           0.0889           0.1550          -0.0870 STT              -0.0745          -0.4040           0.1411           0.0847           0.0746           0.2959 ATT              -0.0451           0.0248          -0.1827          -0.0281           0.1458          -0.0911 PTT               0.9136          -0.0046          -0.0290           0.0230          -0.0842           0.0250 TCC              -0.0869          -0.2576           0.1259           0.1194           0.2232           0.2695 TshrubC           0.1259           0.1221          -0.1075           0.0775           0.2443           0.0543 TherbC           -0.0398          -0.0700           0.1273          -0.0446          -0.0987           0.1034 PoBA              0.9902          -0.0057          -0.0186           0.0113          -0.0303           0.0179 SpBA             -0.0184          -0.2314           0.0791          -0.0073           0.0237           0.3404 AtBA             -0.0415          -0.0274          -0.1354          -0.0644           0.0751          -0.0330 ToBA             -0.0141          -0.2357           0.0405          -0.0247           0.0435           0.3267   208  Appendix E: Full imputed and observed values SpsedTot AtsedTot PosedTot SpsedTot.o AtsedTot.o PosedTot.o 1       933        0        0      21733          0          0 2     40133    15867        0      14000      13467          0 3    106667        0        0       6800     166267          0 4    248400   113200        0      84000     260133          0 5      6933    13600        0      17200      24933          0 6     13333   147733        0      40133      15867          0 7         0        0        0        533      82267          0 8     20400   120000        0      10000          0          0 9     79600    56933        0     190800      13333      23333 10    30133    10133        0      36933          0          0 11        0        0        0      70000          0       3333 12     6933        0        0      80267      73600          0 13   248400   113200        0      90267     400000          0 14     8800        0        0      16800          0          0 15    91467        0        0      60133          0          0 16    53867   117733        0     103200          0          0 17   137467        0        0     126000          0          0 18     6800   166267        0     106667          0          0 19   251200        0        0     263867          0          0 20    30133    10133        0       3333          0          0 21    92400    50133        0       6667          0          0 22    70800    23333        0      13333          0          0 23    23333        0        0       3467          0          0 24   250267    13333        0     417733      33333          0 25    23333        0        0      72800          0          0 26   107200        0        0      44933          0          0 27    80267    73600        0       6933          0          0 28     3467        0        0        133          0          0 29    21733        0        0        933          0          0 30    30133    10133        0        133          0          0 31    10000        0        0          0          0          0 32     3467        0        0      23333          0          0 33     8800        0        0       3333          0          0 34   113333    13333        0     171067          0          0 35     6933    13600        0      45200          0          0 36        0        0        0      92400      50133          0 37    52133     6667        0      57333          0          0 38    60133        0        0      91467          0          0 39    13333        0        0      70800      23333          0 40    57333        0        0      52133       6667          0 41    13333        0        0        533          0          0 42    10000        0        0      20400     120000          0 43      133        0        0      30133      10133          0 209  SpsedTot AtsedTot PosedTot SpsedTot.o AtsedTot.o PosedTot.o 44   350000    83333        0     660000     110400       3333 46   126667   103733    30000     406667      96667      63333 47   133733        0        0      84533          0          0 48   190800    13333    23333      80000      53600          0 49    83333        0     3333      50000      43867       3333 50    50000    43867     3333      83333          0       3333 51     3333        0        0      70400          0          0 52    83333        0     3333     273333     136133      26667 53    80000    53600        0     100000     123600      13467 54   197333        0        0     201867      10000          0 55   251200        0        0     135333          0       3333 56    53867   117733        0      62000     129067          0 57   190800    13333    23333      79600      56933          0 58   106667        0        0      56667      46800          0 59   133733        0        0     250267      13333          0 60   113333    13333        0     168267       3333          0 61    16800        0        0       8800          0          0 62   107200        0        0     131067          0          0 63   106667        0        0     308133          0          0 64   126000        0        0     137467          0          0 65     8800        0        0      16800          0          0 66   113333    13333        0     133733          0          0 67   126000        0        0     197333          0          0 68   135333        0     3333     251200          0          0 69      533        0        0          0          0          0 70   168267     3333        0     120400     171733          0 71   248400   113200        0     125733      33333          0 72    90267   400000        0      46667     816667          0 73   131067        0        0     107200          0          0 74      933        0        0      36667      23333       6667 75   171067        0        0     113333      13333          0 76   406667    96667    63333     130667          0     138400 77   168267     3333        0     350000      83333          0 78      533        0        0          0          0          0 79    92400    50133        0          0          0          0 80   103200        0        0      53867     117733          0 81    84000   260133        0     248400     113200          0 82    60133    16667        0       6800      10000          0 83        0        0        0      43733      23333          0 84    10000        0        0      33600     107067          0 85    17200    24933        0      10667      23600          0 86    17200    24933        0       6933      13600          0 87      133        0        0      20000      13333          0 88     6800    10000        0      60133      16667          0 89    40133    15867        0      13333     147733          0 90     3333        0        0      13333      36667          0 210  Appendix F: Species richness and abundance (%) by sites sites richness abundance sites richness abundance sites richness abundance 101 7 62.3 218 14 142.1 418 11 140.4 102 5 83.3 219 15 80.6 419 21 125 103 10 61.8 220 17 268.7 420 16 159.6 104 7 101.5 301 13 61.1 421 11 113.7 105 16 100.9 302 13 73.5 422 9 105.5 106 10 146.3 303 12 89.4 423 13 168.1 107 12 172.3 304 9 85.7 424 14 159.2 108 10 95.6 305 15 57.5 425 16 144.3 109 16 122 306 12 67.9 501 9 113.2 110 13 128.1 307 13 70.7 502 16 180.4 111 14 101.6 308 11 55.2 503 20 161.7 112 15 117.9 309 12 100.9 504 17 209.4 113 17 93.8 310 9 57.1 505 15 90 201 11 100.6 401 10 72.8 506 16 18.3 202 16 95 402 16 90 507 14 94.3 203 11 117.4 403 19 125.2 508 18 116.6 204 11 143.4 404 21 106.5 509 19 173.1 205 13 109.8 405 12 102.9 510 7 101.6 206 15 157.7 406 13 99.5 511 4 86.7 207 11 107 407 14 117.1 512 21 139.7 208 10 170.4 408 11 133.7 513 20 155.9 209 14 162.3 409 11 139.8 601 19 74 210 13 50.3 410 15 106.1 602 15 39.4 211 15 77.5 411 15 142.2 603 21 153.2 212 17 153.3 412 13 107.6 604 17 112.3 213 15 176.7 413 14 99.8 605 15 75.5 214 13 110.3 414 11 49.7 606 18 90.4 215 15 119.9 415 16 116.8 607 19 84 216 11 201 416 21 153.2 608 19 63.1 217 14 263.5 417 14 128.4 609 11 75.7    211  Appendix G: Abundance rank and spatial distribution of species in the study area Rank Scientific name Type Block A Block B Block C Block D Block E Block F Total % of total no  of sites  Trees           1 Salix spp. T 208.4 410.1 88.2 230.1 87.4 22.8 1047 9.86 8 2 Picea glauca (Moench) Voss T 74.6 221 161.2 259.2 89.8 61 866.7 8.17 8 3 Populus tremuloides Michx. T 218.2 1.9 19.4 106.8 85.1 41.1 472.6 4.45 7 4 Populus balsamifera L. T 2.3   5.2 23.6  31.2 0.29 3  su total  503.5 633 268.8 601.3 285.9 124.9 2417.5 22.77   Average per plot  38.73 31.65 26.88 24.05 21.90 13.88 26.86                            Herbs and Shrubs           1 Arctostaphylos uva-ursi (L.) Spreng. s.l. H 366.6 26.5 57 172.9 132.2 20.2 775.3 7.3 8 2 Shepherdia canadensis (L.) Nutt. H 356 53.6 37.5 115.5 36.4 95.5 694.6 6.54 8 3 Ledum groenlandicum Oeder S  419.1  0.3 6.7  426.1 4.01 4 4 Linnaea borealis L. H  11.3  170.1 28.6 127.3 337.3 3.18 6 5 Alnus incana (L.) Moench S  299   3.6 0.3 302.9 2.85 4 6 Elymus glaucus Buckl. H    296.7   296.7 2.8 2 7 Epilobium angustifolium L.s.l H 6.7 224.4  14.9 3.3  249.3 2.35 5 8 Betula glandulosa Michx. S  92.2   140  232.2 2.19 3 9 Vaccinium vitis-idaea L. S  183.4 23.6 17.5   224.5 2.11 5 10 Arctostaphylos rubra (Rehd. & Wils)  H  50 12.6 125.6   188.2 1.77 5 11 Elymus innovates (Beal) Pilg. H  107.8 1.6 15.5 53.6  178.5 1.68 6 12 Empetrum nigrum L. H 7.5  28.3 22.2 97.2 22.1 177.3 1.67 6 13 Equisetum arvense L.  H 7.9 0.9 2.9 87 31.1 27.4 157.2 1.48 7 14 Rosa acicularis Lindl. s.l. H 6.2 16.5 8.4 45.7 10.5 69.2 156.5 1.47 8 15 Castilleja spp. H 2.6 3.8 0.9 18 95.3 1.9 122.5 1.15 8 16 Erigeron peregrines (Pursh) H  0.6  106   106.6 1 3 17 Pulsatilla  patens (L.) Mill. H 1.7 42.2  38.6 20  102.5 0.97 6 18 Carex spp. H 0.3 27.5 40 30.8 0.6  99.2 0.93 7 19 Aquilega formosa Fisch. Ex DC. H 23.6 1.2  10.9 28.8 1.2 65.6 0.62 5 20 Geocaulon lividum (Richards.) H 1.6   39 6.7 17.1 64.4 0.61 5 21 Pyrola picta Sm. H 0.3 46.5  8.1  7.8 62.7 0.59 5 22 Saxifraga tricuspidata Rottb. H  40.3     40.3 0.38 2 23 Hedysarum mackenzii (Richards.) H 2.3 7.1 0.9 21.1 6.9  38.4 0.36 6 24 Solidago multiradiata (Ait.) H 0.3  0.3 5.1 30.3  36 0.34 5 25 Potentilla anserine L.s.l. S  6.7  26.7   33.3 0.31 2 26 Petastites frigidus (L.) Fries H  27.3  3.3   30.6 0.29 3 27 Achillea millefolium L.s.l. H 1.2   2.4 22 3.3 28.9 0.27 5 212  28 Arnica cordifolia Hook. H 5.4 1.2 3.2 16.5 1.6  27.9 0.26 7 29 Hordeum brachyantherum Nevski H    25   25 0.24 1 30 Aconitum delphinifolium  DC. H     11.3 10.8 22.1 0.21 2 31 Lupinus arcticus Wats.  H 0.3 21.7     22 0.21 2 32 Aster alpines L.ssp. H     14.1 2.6 16.7 0.16 2 33 Vaccinium caespitosum Michx. S 10 2     11.9 0.11 3 34 Festuca spp. H     11.7  11.7 0.11 1 35 Hyacinthoides non-scripta (L.)  H 0.3 0.6 0.6 9.9   11.5 0.11 5 36 Potentilla fruticosa L. H 1.9 8.6     10.5 0.1 3 37 Rubus arcticus L. ssp. H  10  0.3   10.3 0.1 2 38 Sedum lanceolatum Torr. H 0.3 8.3     8.6 0.08 3 39 Trillium grandiflorum  (Michx.) Salisb. H 1.9  5.6    7.5 0.07 2 40 Oxytropis campestris L. (DC.) H 2.8 0.6 1.9 0.9 0.6  6.9 0.06 5 41 Oxycoccus microcarpus Turcz. S 5.3      5.3 0.05 1 42 Taraxacum officinale Weber ex H  1.7  0.3 1.7  3.6 0.03 3 43 Juniperus communis L.s.l. S     1.7 1.7 3.3 0.03 2 44 Viola adunca J.E. Smith S     1.7  1.7 0.02 1 45 Pyrola asarifolia Michx. H 0.6      0.6 0.01 1 46 Mimulus guttatus DC. H    0.3   0.3 0 2 47 Sanicula marilandica L. H    0.3   0.3 0 2  su total  813.6 1742.6 225.3 1447.4 798.2 408.4 5435.3 51.18   Average per plot  62.58 87.13 22.53 57.90 61.40 45.38 60.39                            Brayophytes           1 Pleurozium schreberi (Brid.) Mitt. B 8.9 484.2 98.9 598.3 86.5 80.2 1357.1 12.79 8 2 Dicranum polysetum Sw. B  54.2 60.3 73.6 65 18.8 271.9 2.56 7 3 Dichodontium pellucidum (Hedw.) Schimp. B   0.6 39.5 7.8  47.9 0.45 3 4 Rhytidiopsis robusta (Hook.) Broth. B 0.3 8.9  24.9 1.6  35.7 0.34 6 5 Tortula ruralis (Hedw.) G. Gaertn. B     28.6  28.6 0.27 1 6 Sphagnum papillosum Lindb. B 0.3 2.2 5.5 16.6  1.7 26.4 0.25 6 7 Hookeria lucens (Hedw.) Sm. B     21.6  21.6 0.2 1 8 Polytrichum commune (Hedw.) B 0.3 3.3   5  8.6 0.08 3 9 Dicranum scoparium Hedw. B  3.6     3.6 0.03 1 10 Aulacomnium palustre (Hedw.) Schwagr B  0.6     0.6 0.01 1  subtotal  9.8 557 165.3 752.9 216.1 100.7 1802 16.98   Average per plot  0.75 27.85 16.53 30.12 16.62 11.19 20.02                                 213      Lichens 1 Cladonia spp.  L 17.6 79.2 24.8 73.9 179.9 25.5 400.9 3.78 8      2 Flavocetraria cucullata (Bellardi) K.&A. Thell L 8.6 0.6 0 51.5 165.6 21.5 247.8 2.33 8 3 Peltigera aphthosa (L.) Willd. L 2.5 84.9 26.1 39.5 3.3 29.9 186.3 1.75 8 4 Peltigera canina (L.) Wildd. L 0.3   40.3 6.6 3.3 50.5 0.48 4 5 Cladina mittis (Sandst.) Hustich L 20.4  8.6    29 0.27 2 6 Cetraria pinastry (Scop.) A. Gray L     11.6 11.6 23.2 0.22 2 7 Hypogymnia physodes (L.) Nyl. L 7 2 0.3    9.2 0.09 3 8 Usnea hirta (L.) F.H. Wigg. L  0.9    0.9 1.8 0.02 2  sub total  56.4 167.6 59.8 205.2 367 92.7 948.7 8.94   Average per plot  4.34 16.76 2.99 20.52 14.68 7.13 105.41    Total abundance per plot   106.7 155 71.9 120 128.5 80.4 117.8 100  214  Appendix H: Species richness, abundance, Shannon and 1/Simpson indices by environmental gradient  Table H-1: Variation by research sites Blocks No of sites Richness Mean richness Total abundance Mean  abundance % Mean 1/Simpson Mean Shannon A 13 41 12 1387 107 3.5 1.51 B 10 35 13 1214 121 5.4 1.88 B1 10 42 15 1594 159 5.28 1.92 C 10 27 12 719 72 4.27 1.70 D 11 39 14 1236 112 5.51 1.96 D1 14 41 14 1771 127 4.6 1.85 E 13 43 15 1667 129 7.14 2.14 F 9 28 17 768 85 8.43 2.39   Table H-2: Indices by elevation category Elevation  No of sites Richness Mean richness Total abundance Mean  abundance % Mean 1/Simpson Mean Shannon Very high 4 29 15 542 135 4.41 1.78 High 14 46 12 1844 132 6.71 1.62 Medium 15 43 13 1810 121 5.34 1.85 Low 32 53 15 3154 99 6.71 2.07 Very low 25 46 14 3007 120 5 1.9     215  Table H-3: Indices by slope category Slope No of sites Richness Mean richness Total abundance Mean  abundance % Mean 1/Simpson Mean Shannon Low 47 62 14 5285 112 5.23 1.88 Medium 21 58 14 2210 105 5.55 1.92 Level 16 46 14 1854 116 6.31 2.02 Steep 6 32 14 1006 168 4.27 1.76  Table H-4: Indices by aspects Aspect No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon Eest 9 44 17 956 106 6.22 2.16 Level 20 53 15 2392 120 6.36 2.04 North 1 17 17 269 269 6.39 2.12 Northeast 5 39 15 757 151 4.89 1.85 Northwest 18 51 14 2203 122 6.45 2.03 South 3 27 14 334 111 4.43 1.75 Southeast 13 41 13 1273 98 4.38 1.71 Southwest 12 38 12 1212 101 3.9 1.64 West 9 42 13 960 107 4.64 1.79   Table H-5 Indices by disturbance types Disturbance No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon Beetle 14 46 15 1877 134 4.67 1.84 Fire 11 40 15 1237 112 5.5 1.96 Harvested 20 47 15 1949 98 7.19 2.16 Undisturbed 31 63 13 3822 123 4.85 1.79 Oldfire 14 43 13 1470 105 4.91 1.81  216  Table H-6: Indices by tree density variation Tree density No of sites Richness Mean richness Total abundance Mean  abundance % Mean 1/Simpson Shannon Very high 6 36 13 578 96 3.18 1.45 High 11 42 14 1205 110 6.30 2.06 Medium 14 42 14 1662 119 5.83 1.98 Low 24 56 14 2394 100 5.79 1.99 Very low 35 68 14 4517 129 5.24 1.85  Table H-7: Indices by crown cover variation Crown cover No of sites Cumulative richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon Very high 16 52 14 1857 116 4.68 1.73 High 19 47 14 1905 100 5.49 1.91 Medium 16 54 14 1889 118 5.42 1.94 Low 13 47 16 1365 105 6.79 2.17 Very low 26 66 13 3339 128 5.19 1.86   Table H-8: Indices by stand types Stand No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon Aspen 4 19 10 438 109 3.42 1.49 Mixed 28 60 15 2653 95 6.22 2.04 No trees 13 51 13 1553 119 5.83 1.92 Spruce 45 65 14 5712 127 5.01 1.86    217  Table H-9: Indices by soil moisture variation Soil moisture No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon Very high 5 34 13 685 137 4.43 1.66 High 16 51 15 1930 121 5.44 1.89 Medium 14 47 13 1552 111 4.66 1.8 Low 46 68 14 4809 105 5.67 1.95 Very low 9 68 14 1379 153 5.97 2.01   Table H-10:  Indices by Degree days below 00C DD0 No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon Very high 9 28 17 768 85.3 8.43 2.39 High 38 57 15 4674 123 5.73 1.98 Low 23 46 12 2106 91.6 3.85 1.59 Very low 20 47 14 2807 140.4 5.34 1.9   Table H-10: Indices by degree days above 50C (DD>5 oC) DD5 No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon Very high 36 50 13 3813 106 1.82 4.7 High 14 47 15 1824 130 2.14 6.98 Medium 32 60 14 3482 109 1.93 5.8 Low 5 27 13 767 153 1.76 4.36 Very low 3 29 16 469 156 1.9 4.78   218  Table H-11: Indices by number of frost free days NFFD No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon High 6 34 14 1076 179.3 4.35 1.77 Low 31 55 14 3925 126.6 6.01 2.02 Medium 7 31 12 749 107 4.52 1.79 Very high 37 56 12 3838 103.7 4.57 1.73 Very low 9 28 17 768 85.3 8.43 2.39    Table H-12: Indices by variation of mean annual temperature MAT No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon High 32 53 16 3736 117 6.7 2.12 Medium 15 44 14 1705 114 5.29 1.94 Low 43 60 13 4914 114 4.54 1.74  H-13: Indices by variation in mean annual precipitation MAP No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon High 19 55 15 2743 144 6.26 2.02 Medium 42 61 15 4675 111 6.02 2.03 Low 29 51 12 2938 101 4.05 1.65   H-14: Indices by variation in mean annual TMXsm MAP No of sites Richness Mean richness Total abundance Mean  abundance % 1/Simpson Shannon High 48 59 15 5531 115 6.25 2.06 Medium 18 42 11.6 1579 88 4.41 1.68 Low 24 54 13.4 3245 135 4.56 1.76 219  Appendix I: Canonical Correspondence Analysis outputs   Partitioning of mean squared contingency coefficient:               Inertia Proportion Total           5.766     1.0000 Constrained     1.808     0.3136 Unconstrained   3.958     0.6864  Eigenvalues, and their contribution to the mean squared contingency coefficient   Importance of components:                         CCA1    CCA2    CCA3    CCA4    CCA5   CCA6    CCA7 Eigenvalue            0.4371 0.33820 0.22238 0.17688 0.11250 0.1050 0.09551 Proportion Explained  0.0758 0.05865 0.03857 0.03068 0.01951 0.0182 0.01656 Cumulative Proportion 0.0758 0.13445 0.17302 0.20370 0.22321 0.2414 0.25798                          CCA8    CCA9   CCA10   CCA11   CCA12   CCA13   CCA14 Eigenvalue            0.07159 0.06545 0.05270 0.04302 0.03727 0.03368 0.01705 Proportion Explained  0.01241 0.01135 0.00914 0.00746 0.00646 0.00584 0.00296 Cumulative Proportion 0.27039 0.28174 0.29088 0.29834 0.30481 0.31065 0.31360                           CA1     CA2     CA3     CA4     CA5     CA6     CA7 Eigenvalue            0.43221 0.31869 0.28003 0.22996 0.20666 0.19005 0.15515 Proportion Explained  0.07496 0.05527 0.04856 0.03988 0.03584 0.03296 0.02691 Cumulative Proportion 0.38856 0.44383 0.49240 0.53228 0.56812 0.60108 0.62798                           CA8    CA9    CA10    CA11    CA12    CA13    CA14 Eigenvalue            0.14564 0.1395 0.12665 0.12311 0.12028 0.11271 0.09959 Proportion Explained  0.02526 0.0242 0.02196 0.02135 0.02086 0.01955 0.01727 Cumulative Proportion 0.65324 0.6774 0.69941 0.72076 0.74161 0.76116 0.77843                          CA15    CA16    CA17    CA18    CA19    CA20    CA21 Eigenvalue            0.08932 0.08161 0.07909 0.07213 0.06782 0.06159 0.06081 Proportion Explained  0.01549 0.01415 0.01372 0.01251 0.01176 0.01068 0.01055 Cumulative Proportion 0.79392 0.80808 0.82179 0.83430 0.84607 0.85675 0.86729                          CA22    CA23    CA24    CA25    CA26    CA27    CA28 Eigenvalue            0.05625 0.05120 0.04882 0.04463 0.04249 0.03811 0.03641 Proportion Explained  0.00976 0.00888 0.00847 0.00774 0.00737 0.00661 0.00631 Cumulative Proportion 0.87705 0.88593 0.89440 0.90214 0.90950 0.91611 0.92243                          CA29    CA30    CA31    CA32    CA33    CA34    CA35 Eigenvalue            0.03538 0.03380 0.03174 0.02700 0.02518 0.02392 0.02294 Proportion Explained  0.00614 0.00586 0.00550 0.00468 0.00437 0.00415 0.00398 Cumulative Proportion 0.92856 0.93442 0.93993 0.94461 0.94898 0.95313 0.95710                          CA36    CA37    CA38    CA39    CA40    CA41    CA42 Eigenvalue            0.02221 0.01988 0.01892 0.01738 0.01666 0.01502 0.01417 Proportion Explained  0.00385 0.00345 0.00328 0.00301 0.00289 0.00261 0.00246 Cumulative Proportion 0.96096 0.96440 0.96769 0.97070 0.97359 0.97620 0.97865                          CA43    CA44    CA45     CA46     CA47     CA48 Eigenvalue            0.01371 0.01228 0.01094 0.009794 0.008358 0.007403 Proportion Explained  0.00238 0.00213 0.00190 0.001700 0.001450 0.001280 Cumulative Proportion 0.98103 0.98316 0.98506 0.986760 0.988210 0.989490                           CA49     CA50     CA51     CA52     CA53     CA54 Eigenvalue            0.006916 0.006548 0.006211 0.005444 0.004938 0.004291 Proportion Explained  0.001200 0.001140 0.001080 0.000940 0.000860 0.000740 Cumulative Proportion 0.990690 0.991820 0.992900 0.993850 0.994700 0.995450                           CA55     CA56     CA57     CA58     CA59     CA60 Eigenvalue            0.003882 0.003208 0.002983 0.002697 0.002543 0.002153 Proportion Explained  0.000670 0.000560 0.000520 0.000470 0.000440 0.000370 Cumulative Proportion 0.996120 0.996680 0.997190 0.997660 0.998100 0.998480                           CA61     CA62     CA63     CA64      CA65      CA66 Eigenvalue            0.001801 0.001446 0.001276 0.001010 0.0008993 0.0005748 Proportion Explained  0.000310 0.000250 0.000220 0.000180 0.0001600 0.0001000 Cumulative Proportion 0.998790 0.999040 0.999260 0.999440 0.9995900 0.9996900                           CA67      CA68      CA69      CA70      CA71 Eigenvalue            0.000525 0.0004096 0.0002893 0.0001915 0.0001589 Proportion Explained  0.000090 0.0000700 0.0000500 0.0000300 0.0000300 Cumulative Proportion 0.999780 0.9998500 0.9999000 0.9999400 0.9999600                            CA72      CA73      CA74      CA75 Eigenvalue            0.0001022 5.086e-05 3.629e-05 1.766e-05 Proportion Explained  0.0000200 1.000e-05 1.000e-05 0.000e+00 Cumulative Proportion 0.9999800 1.000e+00 1.000e+00 1.000e+00  Accumulated constrained eigenvalues Importance of components: 220                          CCA1   CCA2   CCA3    CCA4    CCA5    CCA6    CCA7 Eigenvalue            0.4371 0.3382 0.2224 0.17688 0.11250 0.10497 0.09551 Proportion Explained  0.2417 0.1870 0.1230 0.09782 0.06221 0.05805 0.05282 Cumulative Proportion 0.2417 0.4287 0.5517 0.64953 0.71175 0.76980 0.82262                          CCA8    CCA9   CCA10   CCA11   CCA12   CCA13   CCA14 Eigenvalue            0.07159 0.06545 0.05270 0.04302 0.03727 0.03368 0.01705 Proportion Explained  0.03959 0.03619 0.02915 0.02379 0.02061 0.01862 0.00943 Cumulative Proportion 0.86220 0.89840 0.92754 0.95134 0.97195 0.99057 1.00000  Scaling 2 for species and site scores * Species are scaled proportional to eigenvalues * Sites are unscaled: weighted dispersion equal on all dimensions Species scores                CCA1      CCA2      CCA3     CCA4      CCA5      CCA6 STT      -0.004614 -0.814222  0.027358 -0.22589  0.041493 -0.045213 ATT       0.431965 -0.402952  0.961227 -0.26304  0.481020  0.346846 PTT       0.502891  0.793442  0.007846 -0.50741 -0.415324 -0.635293 spsap     0.044695 -0.586156 -0.200193  0.13300  0.136240 -0.166468 atsap     0.726728  0.593519  0.811439  0.28266  0.160341  0.201202 spsed     0.397170  1.031961 -0.917202 -0.96652 -0.116312  0.076696 spsed1    0.376649  0.117924 -0.293695  0.26146  0.116475 -0.141355 atsed     0.573977  0.674147  0.240252 -0.17967  0.011537  0.127179 posed     0.476036  0.797480 -0.811485  0.20769  0.178852  0.058837 Wil_cov  -0.286559  0.202185  0.264844  0.43125 -0.271542 -0.363299 So_cov    0.511751  0.113305  0.934134  0.15429 -0.423159 -0.109783 Ma_cov   -2.193401  0.810660  0.509416  0.48692 -0.155888  0.303306 Bi_cov   -0.708995  1.218729 -0.429631 -1.35180  0.511582  0.100375 Pr_cov    0.047981  0.051839  0.091372  0.14844 -0.613289  1.571930 Ju_cov    0.345079  0.490701 -0.083384 -0.88814 -1.343729  1.086863 Cr_cov    0.292593  0.728931  2.554662  0.58679  0.006104  0.309300 Cf_cov   -0.369726 -0.540319  0.758129  0.12275 -0.996165 -1.106395 Rc_cov   -1.528442  0.930650  1.185210 -0.02920  0.523632  0.843384 Cet_pin   0.654844 -1.130291  1.541046 -1.02874  1.737026 -0.003829 Oxy_can   0.520234 -0.509171  0.865905 -0.48714  1.004906 -0.040066 Pel_aph  -0.239484 -0.651464 -0.051646 -0.23028  0.076643  0.126256 Sol_mul   0.556731  1.246970 -0.889012 -0.97016 -0.310076 -0.575706 Tor_rur   0.542959 -0.462129  0.915140 -0.12593  0.942071  0.027451 Car_spp   0.388309  1.075530 -1.105226 -0.77965 -0.315427 -0.598149 Hed_mac   0.293600 -0.118380 -0.355589  0.17933  0.061790 -0.092990 Hyp_phy   0.392333 -0.097335  1.440271  0.43865 -0.256802 -0.925974 Dic_pal  -0.019360 -0.459936 -0.393705 -0.48410  0.340262  0.072458 Fes_spp   0.497567  0.912610 -0.730997 -0.71498 -0.247812 -0.140044 Pyr_pic  -0.254238 -1.407806 -0.040659  0.02665 -0.447458  0.805262 Pyr_asa   0.638472  0.602292  1.320666 -0.02237 -1.342978 -2.036503 Usn_spp  -0.135392 -1.300251  0.037704 -0.66108 -0.846812  0.661597 Clad_spp -0.488554  0.161877 -0.435434 -0.35751  0.729457  0.103286 Rhy_rob  -0.514492 -0.398464 -0.720017  0.69690 -0.313682 -0.202909 Arc_uva   0.524902  0.273238  0.853961 -0.04230  0.141796 -0.144990 Vac_cae   0.341470 -1.086216  1.775151 -1.67783  2.376412  0.266009 She_can   0.511148 -0.078507  1.535966 -0.09930  0.410694 -0.844481 Ely_gla  -1.195688  1.117092 -0.165314 -0.09979 -0.047212 -0.084657 Equ_spp   0.269576  0.149740 -0.837681  1.23309  0.294939  0.406577 Ple_sch  -0.477598 -0.587606 -0.323824  0.03996 -0.144629  0.061676 Pet_spp  -0.657452  0.280908 -0.054244  0.47227 -1.318584 -1.844063 Aul_spp  -2.089818  0.600094  0.603545 -0.04772  0.651589  1.063222 Sed_spp  -2.401677  0.832688  0.659584 -0.29074  0.849915  1.374136 Sax_spp  -1.008974 -0.405803  0.525847  0.38380 -0.524923 -0.963391 Dic_pel  -2.115774  1.031247  1.455621  2.32337 -0.988475 -0.240796 Dic_spp   0.729415  0.822568 -0.715508  1.34760  0.778950  0.447269 San_spp   0.506092  0.410399 -1.591414  1.47640  0.959019  0.119508 Hor_bra   1.176981  1.012647 -0.187908  1.19101  0.475160  0.369803 Epi_aug   0.638096  0.390133 -0.362593  0.26026 -0.146062  0.473120 Mim_spp   1.176981  1.012647 -0.187908  1.19101  0.475160  0.369803 Sph_pap  -0.890509  1.249926 -0.275665 -1.31397  1.412831  0.644005 Ely_gla1  0.382221 -0.562520 -0.749428  0.68409  0.157819  0.315870 Pot_spp  -0.096336 -0.410336 -0.796551  1.03630 -0.486156 -0.333251 Hoo_spp   0.462287  1.486113 -1.212597 -1.22559 -0.687318 -0.966530 Aqu_for   0.422337  0.633297  0.411920 -0.35794 -0.096566 -0.079189 Ast_spp   0.490884  1.162871 -0.761139 -1.21012 -0.722723 -0.259865 Aco_del   0.501379  0.491237 -0.472531 -0.90489 -1.305169  0.584188 221  Pol_spp  -0.801215  1.268133 -0.272326 -1.31527  1.011950  0.645706 Eri_per   0.502207  0.902510 -0.651207 -1.60941 -0.551964 -0.471607 Cas_spp   0.632879  0.178449  0.069714 -1.17381 -1.126684  1.111812 Pyr_spp  -1.598747  0.337675  0.390746 -0.52537  0.007611  0.772102 Vio_adu   0.614229  1.436370 -0.901506 -1.32668 -0.456793 -0.757793 Ach_mil   0.431529  0.907575 -0.636182 -0.99965 -0.620787 -0.171203 Rub_arc  -2.485164  1.571198  0.134819 -1.21863  2.805217  1.588117 Lin_bor   0.317969 -0.386572 -0.221526 -0.02159 -0.770776  0.984332 Cla_spp   0.230810  0.407863 -0.280529 -0.48442 -0.128449 -0.109832 Dic_spp1  0.687132  0.746409 -0.808102  1.42737  0.574433  0.379115 Arc_rub  -0.125444 -0.713951 -0.455681  0.42267 -0.090542 -0.188457 Ely_inn   0.474334  0.605233 -0.459258 -0.68306 -0.245819 -0.053581 Geo_liv   0.493579 -0.389876 -0.083904 -0.18113 -0.144267  0.951918 Vac_vit  -1.282782  0.052848  0.022495  0.19156  0.074747 -0.265936 Emp_nig  -1.253147 -0.032620  0.016244 -0.22082  0.190477 -0.494772 Arn_cor   0.432821 -0.222015 -0.144138  0.44106 -0.230890 -0.332894 Hya_non   0.820922  0.745086 -0.155558  1.11925  0.532946  0.403718 Pel_can   0.646291  0.142588 -0.589920  0.48397  0.153587  0.393273 Ros_aci   0.613124  0.041883  0.264534 -0.37252 -0.496474  0.781960 Spa       0.089278 -0.664123 -0.139090  0.24579  0.096640  0.432005 Fla_cuc  -0.348798  0.005604 -0.255249 -0.01968  1.369542  0.412063 Cro_spp   0.245634 -0.717035 -0.848125  0.10191  0.315231  0.285554 Tar_off   0.496569  1.395483 -1.159409 -1.31002 -0.691541 -0.967881 Lad_gro  -2.176354  0.654318  0.219314  0.05543  0.082674  0.205222 Pot_nan   0.753166 -1.492972  2.033768 -1.61974  2.605104  0.264608 Lup_arc   0.049439 -0.247893  0.213009 -0.25145 -1.783506  0.649235 Tri_gra   0.227644  0.047205 -0.585279 -0.49339 -1.837313  1.099197   Site scores (weighted averages of species scores)          CCA1    CCA2     CCA3      CCA4      CCA5     CCA6 X101  0.28224 -1.3228  0.96455 -0.417186  0.351738 -1.01165 X102  0.40594 -0.6902  2.50927 -0.022206  0.437962 -2.81663 X103  1.26666  1.1947  2.45927 -0.301913  1.018023  0.98868 X104  1.03054  0.6509  2.08359 -0.092054 -0.299673 -0.05381 X105  0.66752 -1.2948  2.75451 -1.661889  3.655019  0.71937 X106  1.19129  0.6562  3.09491 -0.011967  1.561929  0.39317 X107  1.05059  1.0086  3.02822  0.012085 -0.021805 -0.37723 X108  0.74177 -0.1712  2.57555 -0.056318 -0.587758 -0.99784 X109  0.78893  0.4095  1.36319  0.720901 -0.162561 -1.30022 X110  0.32085  0.2368  2.08437  1.305523 -2.311269 -2.02310 X111  0.52482  0.3108  1.90453  1.085963 -1.622305 -1.74843 X112  0.72204  0.3000  1.60220  0.256986 -1.012646 -1.31715 X113  1.01309  1.4125  1.28020 -0.288331  0.360103  0.40882 X201 -0.50525 -1.9548 -0.41890 -0.763778 -0.004967 -0.12841 X202 -0.13312 -1.7304 -0.39962 -0.469718  0.295489 -0.51403 X203 -0.71794 -1.0619 -0.34989 -0.155385  0.499891 -1.15651 X204 -0.57296 -1.1404 -0.53181 -0.287751  0.911995 -1.40092 X205 -1.18124 -0.7231 -0.08308 -0.042220  0.384197 -1.40328 X206 -0.34582  0.1291 -0.10969  0.700232 -2.485267 -2.89856 X207 -1.61313 -0.3957  0.08207  0.474064 -0.361147 -0.37918 X208 -2.84804  0.6993  0.48601  0.977851 -0.515944 -0.01792 X209 -2.59882  0.5125  0.23438  0.538204 -0.233757 -0.87509 X210 -2.88241  1.4768  0.92247  0.448476 -0.406952  2.34989 X211  0.56490 -0.1286 -0.34140  0.551362  0.220292 -1.09337 X212 -0.18067 -0.9376 -0.14025  0.230145 -1.035076 -1.94537 X213 -0.83460 -0.8258 -0.09262  0.183788 -0.206400 -1.22395 X214 -0.27697 -0.5109  0.95072 -0.052600 -0.329461 -0.82756 X215 -2.22826  0.9229  1.53373  1.546214 -1.272557  0.02509 X216 -2.80948  0.6064  0.53300  0.913946 -0.472101 -0.03514 X217 -2.31136  0.6590  0.37824  0.431278 -0.372845  0.01753 X218 -4.24979  1.8021  1.41821  0.565252 -0.339567  3.43477 X219 -1.86603  0.2896 -0.30352 -1.714541  2.622094 -1.84330 X220 -2.24559  1.0348 -0.83472 -2.214441  3.522095  1.70943 X301  0.39676 -0.5138 -0.77777  0.549380  0.755183 -1.29738 X302 -0.09988 -1.3040 -0.62567 -0.739606  2.292532  0.11648 X303  0.29312 -0.3295 -0.12789 -0.234579  1.788472  0.02914 X304 -0.15621 -1.8432 -0.49220 -0.847820  0.554655 -0.17836 X305  0.24315 -1.3064  0.39880 -0.424322  1.263980 -0.70847 X306  0.55837 -0.7950  0.62077 -0.279397  1.154125 -0.16401 X307  0.02097 -0.9785 -0.22253  0.443911  0.130122 -1.16301 222  X308 -0.35006 -1.9887 -0.30917 -0.607032 -0.096538 -0.26698 X309  0.19663  0.1156  1.59141  0.040103  0.263006 -0.52388 X310  1.03519 -0.2251  2.65489 -1.065495  1.337049  1.47903 X401  0.84708  0.6896 -0.86877  1.243482  0.649354 -0.81286 X402  0.92355  1.1198 -0.76459  0.109655 -0.992240  1.62616 X403  0.89248  0.9310 -1.66558  2.528295  1.424586  0.25629 X404  0.32943 -0.2240 -0.55626  1.721109  0.122763 -0.47734 X405  1.02031  1.3121 -1.71894  3.493436  2.069116  1.41126 X406  0.70101  0.6688  0.56551  2.039836 -0.973613  0.28254 X407  1.44957  1.6220 -2.12214  5.241437  2.904215  2.43758 X408 -0.18752 -1.0980 -1.30398  1.258141 -0.348826 -0.33215 X409  0.96690  1.2200 -0.99872  2.404212  1.714016  0.73356 X410  0.90948  0.9971 -0.90883  2.009675  0.373138  2.36008 X411  0.16166 -0.4669 -0.04443 -0.314099  2.281392 -0.19232 X412  0.29433 -0.8103  0.14691 -0.151508  0.241708  0.48005 X413  1.04525  1.1665  1.44384  0.827315  0.547465  1.54111 X414  0.75479  0.2129 -0.53205  0.694341  0.260726  0.36064 X415 -0.04791 -1.2312 -0.67585 -0.314186 -0.075759  1.13727 X416  0.68698  0.2923  0.23947 -0.009725  0.696557 -0.08799 X417  0.63183 -0.2334  0.69922  0.736586  0.138123 -0.53116 X418 -0.40242 -1.8851 -0.69345 -0.347472 -0.814591  0.89195 X419  0.41101 -0.7177 -1.08009 -0.295237  0.497354 -0.30463 X420  0.05860 -1.1417 -1.05504  0.430561  0.183718 -0.25613 X421 -0.09719 -0.9446 -0.82930  0.571524 -0.736884 -0.23731 X422 -0.22525 -1.5858 -0.90832  0.217227 -0.487409  0.34210 X423 -0.18810 -1.0813 -1.49657  1.308965 -1.612823  0.23981 X424  0.02483 -1.2017 -1.02241  0.512844 -0.687635  0.16141 X425  0.08657 -1.1440 -0.93136  0.259465 -0.497768  0.44069 X501  0.45432  3.1009 -3.09102 -4.362047 -3.378264 -4.32658 X502  0.74347  1.4604 -0.09914 -1.574153 -0.408121 -0.34460 X503 -0.05183 -0.1705 -0.88063 -1.559712  1.043809  0.21861 X504  0.77224  1.8307  0.07799 -2.201158  0.439967  0.36553 X505  0.45036 -0.1468 -0.98961 -0.035320 -0.396070  0.23593 X506  0.51377 -0.4111  0.35600 -0.422826  0.369415  0.52219 X507  0.14308 -0.3056 -0.16942 -0.409051 -1.325327 -0.13382 X508  0.81624  1.0028 -0.79740 -0.208576 -0.139975 -0.60257 X509  0.46406  1.1950 -1.21275 -1.018165 -0.230117 -1.19197 X510  0.20513  2.7052 -2.82623 -3.431674 -2.293798 -2.14929 X511  0.58337  3.1972 -4.36492 -4.134339 -2.684365 -4.84289 X512  0.60118  1.8485 -0.71023 -2.644156 -0.617208  0.27876 X513  0.82319  0.7686 -0.03248 -0.403221  0.021095  0.26098 X601  0.41786 -0.5242  0.16021 -1.297318 -1.856202  2.71206 X602  0.30996 -0.8950  0.23695 -0.428498 -0.659197  0.64439 X603  0.61926 -0.1445  0.93051 -0.432468 -2.069540  2.18539 X604  0.26878 -0.5179 -0.04602 -0.456604 -2.731442  2.28151 X605  0.15595 -1.3548  0.17018 -0.824673 -1.326003  1.67226 X606  0.43960 -0.7018  0.09320 -1.095841 -2.938624  2.98291 X607  0.50245 -0.5983 -0.02388 -0.422636 -2.073485  2.91854 X608  0.70364  0.2094  0.44954 -0.975046 -1.244170  2.04804 X609  0.73261 -0.6723  1.06146 -0.570076 -2.605540  5.05148  Model: cca(formula = CommunityDataset ~ lat + Ele + Slope + MAT + MAP + DD0 + DD5 + NFFD + OD + AM + CL + SA + TT + TCC, data = EnvironmentalDataset) Df  Chisq      F N.Perm Pr(>F)    lat       1 0.1994 3.7785     99   0.01 ** Ele       1 0.2680 5.0786     99   0.01 ** Slope     1 0.1700 3.2222     99   0.01 ** MAT       1 0.1373 2.6026     99   0.01 ** MAP       1 0.1486 2.8164     99   0.01 ** DD0       1 0.1423 2.6961     99   0.01 ** DD5       1 0.0697 1.3209     99   0.09 . NFFD      1 0.1279 2.4234     99   0.01 ** OD        1 0.0964 1.8262     99   0.01 ** AM        1 0.0589 1.1157     99   0.25    CL        1 0.0543 1.0285     99   0.46    SA        1 0.0571 1.0820     99   0.40    TT        1 0.1931 3.6591     99   0.01 ** TCC       1 0.0853 1.6164     99   0.02 *  Residual 75 3.9578                         --- Signif.codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 223  Appendix J: List of vegetation locally used in Yukon (sources: Johnson et al., 1995; and Holloway and Alexander, 1990)  SN Scientific name Total cover % Proportionate Cover % Available  blocks Local uses 1 Salix spp. 1047.0 9.86 A,B,C,D,E & F Pain soothe 2 Picea glauca (Moench) Voss 866.7 8.17 A,B,C,D,E & F Timber, medicine 3 Shepherdia Canadensis (L.) Nutt. 694.6 6.54 A,B,C,D,E & F Food, dessert 4 Populus tremuloides Michx. 472.6 4.45 A,B,C,D,E & F Fuel, mosquito repellent 5 Ledum groenlandicum Oeder 426.1 4.01 B,C & D Tea 6 Alnus incana L. (Moench) 302.9 2.85 B,E & F As dye and medicines 7 Epilobium augustifolium L.s.l. 249.3 2.35 A,B,D & E Food, Jam 8 Vaccinium vitis-idaea L. 224.5 2.11 B,C & D Food 9 Arctostaphylos rubra (Rehd.&Wils.) Fern. 188.2 1.77 B,C & D Food flavour 10 Aquilegia Formosa Fisch. Ex. DC. 65.6 0.62 A,B,D,E & F Medicine 11 Hedysarum mackenzii (Richards.) 38.4 0.36 A,B,C,D & E Food, Fodder, Medicine 12 Potentilla anserine L.s.l. 33.3 0.31 B & D Food, medicine 13 Populus balsamifera L. 31.2 0.29 A,D & E Fuel, tea, medicine 14 Petastites frigidus (L.) Fries 30.6 0.29 B & D Used as seasoning 15 Achillea millefolium L.s.l. 28.9 0.27 A,D,E&F Medicine; tobacco 16 Arnica cordifolia Hook. 27.9 0.26 A,B,C,D &E Medicine 17 Aconitum delphinifolium DC. 22.1 0.21 E & F Pest control agent 18 Rubus arcticus L. ssp. 10.3 0.10 B & D Food 19 Vaccinium oxycoccus 5.3 0.05 A & B Food 20 Juniperus communis L.s.l. 3.3 0.03 E & F Medicine 21 Pyrola asarifolia Michx. 0.6 0.01 A Medicine  Total 4778.2 45.02     

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.24.1-0165664/manifest

Comment

Related Items