ECOLOGICAL RESTORATION AND ECOSYSTEM MEMORY OF WILDLIFE FORAGE AND UNDERSTORY DIVERSITY IN A YOUNG PINE MONOCULTURE PLANTATION IN CENTRAL-INTERIOR BRITISH COLUMBIA by Julia Claire Bizon B.Sc., University of British Columbia, 2021 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN NATURAL RESOURCES AND ENVIRONMENTAL STUDIES UNIVERSITY OF NORTHERN BRITISH COLUMBIA April 2025 © Julia Claire Bizon, 2025 Abstract Ecological restoration has recently taken center stage in the rehabilitation of degraded forest ecosystems to improve multifunctionality, biodiversity, and wildlife habitat conservation. The research summarized herein assessed the efficacy of variable stand density thinning (200, 400, and 600 stems/ha) and artificial canopy gaps (0.2, 0.5, 1.0, 2.0 ha in size) as potential restoration treatments to enhance wildlife forage and native biodiversity in a young lodgepole pine (Pinus contorta var. latifolia Engelm. ex S. Wats.) forest in central-interior British Columbia. Field data collection was conducted from May-August, 2023 to assess the early response (1-3 years post-treatment) of the forest understory plant community to the treatments. The effects of the restoration treatments and time since treatment on the cover, richness, diversity, and composition of the shrub and herbaceous layer species were analyzed using linear mixed-effects models and permutational multivariate analysis of variance. Additionally, linear models were used to examine the influence of forest floor characteristics (leaf litter, bare ground, downed woody debris, and duff depth) on understory cover, richness, and diversity as well as the impact of the treatments on presence of wildlife activity (based on presence or absence of feces). Both restoration thinning and canopy gap treatments resulted in an increase in understory herbaceous species cover and richness compared to the untreated controls. Species cover, richness and diversity were generally higher in 3-years post-treatment plots compared to 1-year post-treatment plots. Furthermore, herbaceous species composition varied significantly between the restoration treatments versus the controls, with treated plots dominated by shade-intolerant, disturbance-dependent species. Forest floor characteristics, such as percent cover of leaf litter and bare ground, were ii associated with reduced herbaceous species cover and richness. There was evidence of increased wildlife presence in the thinning and canopy gap treatments (as compared with the untreated controls), where certain species of importance to moose, such as aspen (Populus tremuloides) and willow (Salix spp.), were found. Overall, these results suggest that reducing forest stand density through thinning and artificial canopy gaps may be effective for restoring native plant biodiversity and wildlife habitat in dense monoculture plantations. Soil samples were also collected from the study area for a seedling emergence trial to assess the viability of the soil seed bank. After 6 months of germination under favorable conditions in the greenhouse, predominantly graminoids and ruderals germinated from the seed bank. There was no difference found in the count or identities of the germinants between the restoration treatments, untreated controls, and old-growth reference stands when analyzed using permutational multivariate analysis of variance. Notably, the absence of seeds from target forage species, such as aspen, willow, and birch, from the soil seed bank suggested that we cannot rely solely on the seed bank for passive regeneration of native plant biodiversity in these young pine-dominated forests. Rather, active restoration methods, such as seeding and planting of native forage species, may be needed to rehabilitate and restore understory plant diversity and wildlife habitat in these stands. iii Preface This manuscript-based Master’s thesis consists of two data chapters. The research described in these chapters was primarily conducted by myself, Julia Claire Bizon, who planned and carried out the field data collection, analyzed the data, interpreted the research results, and led the writing of the manuscripts. Chapter 2: Effect of restoration treatments on understory vegetation and wildlife habitat use. Chapter 3: Assessment of the soil seed bank and ecosystem memory. iv Table of Contents Abstract ..................................................................................................................................... ii Preface ..................................................................................................................................... iv Table of Contents ...................................................................................................................... v List of Tables .......................................................................................................................... vii List of Figures .......................................................................................................................... ix Acknowledgment ..................................................................................................................... xi Chapter 1: Introduction ............................................................................................................. 1 Chapter 2: Effect of Restoration Treatments on Understory Vegetation and Wildlife Habitat Use ............................................................................................................................................ 5 Introduction .......................................................................................................................... 5 Materials and Methods ......................................................................................................... 8 Study Area ........................................................................................................................ 8 Restoration treatments ...................................................................................................... 9 Data collection ................................................................................................................ 10 Data Analysis .................................................................................................................. 12 Results ................................................................................................................................. 15 Stand structure characteristics ........................................................................................ 16 Effects of treatments on understory cover, richness, and diversity ................................ 17 Effect of treatments on understory species composition and indicator species.............. 21 Effect of forest floor characteristics on understory cover, richness, and diversity......... 24 Effects of understory cover, richness, and diversity on wildlife feces presence ............ 26 Discussion ........................................................................................................................... 27 Herb layer response to restoration treatments ................................................................ 27 Shrub layer response to restoration treatments ............................................................... 30 Effect of time since treatment ......................................................................................... 31 Influence of forest floor characteristics .......................................................................... 31 Wildlife presence in the stands ....................................................................................... 32 Limitations and Opportunities for Further Research ...................................................... 33 Conclusion ...................................................................................................................... 35 Chapter 3: Assessment of the Soil Seed Bank and Ecosystem Memory ................................ 37 Introduction ........................................................................................................................ 37 Methods............................................................................................................................... 40 Study Area ...................................................................................................................... 40 Sampling Method............................................................................................................ 41 Greenhouse Methods ...................................................................................................... 42 v Statistical Analysis.......................................................................................................... 43 Results ................................................................................................................................. 44 Effect of canopy treatments on species presence and richness....................................... 46 Comparing the above-ground vegetation to the soil seed bank ...................................... 48 Discussion ........................................................................................................................... 49 Species present in the seed bank ..................................................................................... 49 Impact of canopy treatments on the seed bank species composition.............................. 51 Comparison between above-ground vegetation and the seed bank ................................ 53 Limitations and Opportunities for Further Research ...................................................... 55 Conclusion ...................................................................................................................... 55 Chapter 4: Conclusion and Recommendations ....................................................................... 57 References............................................................................................................................... 59 Appendix................................................................................................................................. 73 vi List of Tables Table 1: Summary of stand structure variables (basal area, canopy closure, and observed stem density) at the three different thinning densities (200, 400, and 600 stems/ha) and the untreated control. Values represent average of the measurement for all replicates of the treatments with the standard error. ......................................................................................... 17 Table 2: Effect of thinning and gap treatments (trt), time since treatment (time), and their interaction (trt x time) on percent cover, species richness, and species diversity within the herbaceous (herb) and shrub layers.. ...................................................................................... 18 Table 3: Results of permutational multivariate analysis of variance (PERMANOVA) of effect of thinning treatment, gap treatment, time since treatment, and the interaction between treatment and time on species composition (61 species) in the herbaceous (herb) and shrub layers ....................................................................................................................................... 23 Table 4: Herbaceuous plants that were indicators of particular treatments (thinning density, gap size, or untreated control) after 1-year or 3-years post-treatment. A is the probability that the site will belong to the treatment if the species is found. B is the probability that the species will be found given the treatment. IndVal is the square root of the product of A and B ................................................................................................................................................ 24 Table 5: Effect of forest floor characteristics (bare ground, coarse wood, duff depth, fine wood, leaf litter, and rock) on understory plant cover, richness, and diversity in the herbaceous (herb) and shrub layers.. ...................................................................................... 25 Table 6: Results of the generalized linear model showing the effect of thinning treatments, gap treatments, time since treatment, and their interaction on wildlife (bear, ungulate, and grouse) feces presence ............................................................................................................ 26 Table 7: Detailed lists of the treatments (thinning density, canopy gap size, or reference forest) from which soil samples were collected that each species germinated from during the seedling emergence trial. ........................................................................................................ 47 Table 8: Results of permutational analysis of variation (PERMANOVA) of the effect of thinning treatment, gap treatment, time since treatment, and the interaction on plant species composition (9 species) of the seed bank. .............................................................................. 48 Table 9: Jaccard similarity coefficient (calculated value between 0 and 1) for binary data for the presence/absence of aboveground herbaceous species and the species present in the soil seedbank for each corresponding plot organized by treatment (thinning density, gap size, untreated control) and years since treatment. Only plots with at least 1 shared species are listed, all other plots had a Jaccard index of 0 (no shares species). Species shared between the aboveground vegetation and corresponding soil seed bank are listed. ................................... 49 vii Table A1 Number of replicates of each treatment sampled in 3 years post-treatment area and in 1 year post-treatment area, and untreated control forest. Total number of areas sampled is given. ...................................................................................................................................... 73 Table A2 All understory plant species observed in the study area (across all 51 plots) from June – August 2023, and their common names. If a species was mentioned in > 50% of the sources (3/6) the species was classified as being of high importance to moose. If a species was mentioned in < 50% of the sources the species was classified as medium importance for moose. If a species was not mentioned in any source, it was classified as low importance. . 74 Table A3: Results of linear models showing the effect of observed tree density and the square of observed tree density on herb and shrub cover, richness, and diversity ................. 77 Table A4 Estimated marginal means (EMMs) for the LMEs and GLMEs showing the effect of thinning treatment on herb layer cover and richness.......................................................... 77 Table A5 Estimated marginal means (EMMs) for the LMEs and GLMEs showing the effect of gap treatment and time on herbaceous (herb) layer cover and richness. ............................ 78 Table A6 Estimated marginal means (EMMs) for the LMEs showing the effect of thinning treatment and time since treatment on shrub layer diversity.. ................................................ 78 Table A7 Estimated marginal means (EMMs) for the LMEs showing the effect of gap treatment and time since treatment on shrub layer diversity. ................................................. 79 Table A8: Results of post-hoc test (pairwiseAdonis) of PERMANOVA showing effect of thinning treatments, time since treatment, and their interaction on plant species composition in the herbaceous layer ........................................................................................................... 79 Table A9 Results of post-hoc test (pairwiseAdonis) of PERMANOVA showing effect of gap treatments, time since treatment, and their interaction on plant species composition in the herbaceous layer ..................................................................................................................... 80 Table A10 Estimated marginal means (EMMs) for the general linear model showing the effect of thinning treatment on presence of feces in the herbaceous layer ............................. 81 Table A11 Estimated marginal means (EMMs) for the general linear model showing the effect of gap treatment on presence of feces in the herbaceous layer. ................................... 81 Table A12 List of the species that germinated during the greenhouse trial, with main method of dispersal, life form, and origin status (native or exotic). .................................................... 82 Table A13 The number of germinants of each species calculated per 1000 mL of soil for each treatment, separated by number of years post-treatment. ............................................... 83 viii List of Figures Figure 1: A diagram detailing the sampling design. One mapped hexagon of approximately 14 km2 is represented, containing several treated areas (represented by the shaded areas). Each treatment area represents a thinning density (200, 400, 600 stems/ha) or a canopy gap (0.2, 0.5, 1.0, 2.0 ha). .............................................................................................................. 12 Figure 2:Average herb (herbaceous) percent leaf cover within all thinning treatments and untreated control (ctrl) (A) and within all gap treatments and untreated control (ctrl) (B). Average herb percent leaf cover within thinning treatments one-year post-treatment vs threeyears post-treatment (C) and average herb percent leaf cover within gap treatments one-year post-treatment vs three-years post-treatment (D). .................................................................. 19 Figure 3: Average herbaceous (herb) species richness within all thinning treatments and untreated control (ctrl) (A) and within all gap treatments and untreated control (ctrl) (B). Average herb species richness within thinning treatments one-year post-treatment vs threeyears post-treatment (C) and average herb species richness within gap treatments one-year post-treatment vs three-years post-treatment (D). .................................................................. 20 Figure 4: Average shrub species diversity within all thinning treatments and untreated control (ctrl) (A) and within all gap treatments and untreated control (ctrl) (B). Average shrub species diversity within thinning treatments one-year post-treatment vs three-years post-treatment (C) and average shrub species diversity within gap treatments one-year posttreatment vs three-years post-treatment (D). .......................................................................... 21 Figure 5: Nonmetric multidimensional scaling (NMDS) plots of understory herbaceous (herb) plant composition grouped by thinning treatment (A) and gap treatment (B). After 20 iterations the final stress is 0.223 (A) and 0.236 (B). NMDS was run with 2 dimensions and Bray-Curtis distance. Each point represents one plot; a representatively coloured polygon groups the replicates of each thinning treatment (200, 400, 600 stems/ha, or untreated control) or gap treatment (0.2, 0.5, 1.0, 2.0, or untreated control). ........................................ 23 Figure 6: Coefficient plots (95% confidence interval) for linear models showing effect of forest floor characteristics for the herbaceous (herb) layer (A) and the shrub layer (B). Red represents models for cover, green represents models for richness, and blue represents models for diversity.. .............................................................................................................. 25 Figure 7: Number of plots where feces (distinguished as belonging to bear, grouse, or ungulate) was found in all replicates within each thinning treatment density (200, 400, 600 stems/ha) (A), gap treatment size (0.2, 0.5, 1.0, 2.0 ha) (B) and untreated control (ctrl) among both 1-year and 3-years post-treatment plots combined. ........................................................ 27 ix Figure 8: Images of trays (54 total) of soil samples in the Enhanced Forestry Lab at the University of Northern British Columbia, Prince George, undergoing the seedling emergence trial (left). Trays with humidity covers shown (centre). Initial germination of seeds in one sample 8 weeks after the start of the trial before germinants were removed (right). ............. 43 Figure 9: The number of germinants that emerged per 1000 mL of soil in each different gap treatment (0.2, 0.5, 1.0, 2.0 ha), thinning treatment (200, 400, 600 stems/ha), untreated control (ctrl), and reference forest (ref) during the seedling emergence trial. ........................ 46 Figure 10: Nonmetric multidimensional scaling (NMDS) plots of seed bank plant composition grouped by thinning treatment (200, 400, 600 stems/ha), untreated control (ctrl), and reference stand (ref) (A) and by gap treatment (0.2, 0.5, 1.0, 2.0 ha), untreated control (ctrl), and reference stand (ref) (B). Each point represents one plot. NMDS was run with 2 dimensions and Bray-Curtis distance. After 20 iterations the final stress is 0.047 (A) and 0.056 (B) ................................................................................................................................. 48 x Acknowledgments There are many people who got this thesis to the point where it is, and I am endlessly grateful for the support I have received over the last two and a half years. First, I would like to acknowledge the Society for Ecosystem Restoration in Northern British Columbia (SERNBC) for funding the field data collection and providing logistical support throughout this project. Thanks to Brandon Geldart and Dr. Jeffery Werner for providing field maps and sharing their knowledge of the area. I would also like to acknowledge the Habitat Conservation Trust Fund for awarding me the Together 4 Wildlife Scholarship: thank you for believing in my project and being a supportive resource. My thesis supervisor, Dr. Samuel Bartels, has my gratitude for believing in me and being the first line of defense when something went wrong: thank you for your calm presence, problem-solving intuition, and way with words. I want to thank my thesis committee members, Dr. Jeffery Werner and Dr. Roy Rea, for their guidance and suggestions during this process. I also want to give my appreciation to the curators Doug Thompson and Dr. Kennedy Boateng at the Enhanced Forestry Lab for their assistance in the greenhouse and for attending to my samples when I could not. I would also like to acknowledge my field assistants, Akshay, Akil, and Simran, for their tireless work in the face of wildfire smoke, thunderstorms, hail, and overcurious bears. In addition, I want to thank the Preston lab for adopting me in and making me feel welcome. A special thanks to Stephanie for always replying to my texts, curbing my anxieties, and providing puppy photos when needed. And finally, where would I be without the support of my family. Thank you for listening to my impassioned rants, boosting my selfesteem, and making me laugh. xi Chapter 1: Introduction Despite comprising less than 1% of overall biomass in most forest stands, the understory has many essential roles in a forest ecosystem. Due to high turnover and rapid decomposition, the understory can provide anywhere from 4% to 7% of the net primary productivity for a forest stand (Gilliam, 2007; Muller, 2003). Understory plants can impact the growth of overstory plants by affecting water and light availability, competition for soil nutrients, as well as by creating a physical barrier for germination with leaf litter layer (Dupuy & Chazdon, 2008; George & Bazzaz, 1999; Nilsson & Wardle, 2005). The understory is also an important source of forest diversity, supporting high levels of both flora and fauna biodiversity (Gilliam, 2007). Understory species are often used as indicator species of site conditions due to their sensitivity to light, nutrients, and soil moisture and acidity (G. Wang, 2000). Wildlife use the understory layer for forage and shelter; birds and small mammals nest in vegetation, and ungulates feed on plant material (Hagar, 2007; Simonetti et al., 2013). In the boreal forests of North America, moose are a fundamental part of the ecosystem, impacting plant composition, structure, diversity, productivity, succession, and nutrient cycling (Faison et al., 2016; Hobbs, 1996; Pastor et al., 1988). They are an important source of food and recreation for both hunters and non-hunters in Northern British Columbia, and also provide ceremonial, sustenance, and social purposes for Indigenous communities. However, over the last several decades, there has been a decline in moose populations, possibly driven by climate change, patterns of forest management, and other human disturbances (Kuzyk et al., 2019; Mumma & Gilligham, 2019; Murray et al., 2006). 1 The overstory has a significant impact on the understory by directly influencing soil nutrient concentrations and acidity, as well as by modulating the amount of light and water that penetrates the canopy (Augusto et al., 2002, 2003; Hart & Chen, 2008; Messier et al., 1998). The amount of light and water that reaches the forest floor is a function of canopy openness, with more closed canopies leading to diminished resources available below the canopy. Conversely, gaps in the canopy allow for water and light to penetrate the overstory layer, creating a heterogeneous environment on the forest floor, which can support a higher diversity of understory species (De Grandpré et al., 2011). Light, water, and nutrients are limited on the forest floor due to competition with overstory species and other understory species, and thus, understory structure is strongly influenced by resource availability (Hart & Chen, 2008; Lochhead & Comeau, 2012). There is debate over whether understory species are driven by habitat heterogeneity (such as variation in microsites and resource conditions) or resource quality (the amount and quality of the most limiting resources). The habitat heterogeneity hypothesis states that a more heterogeneous environment will result in higher species diversity, while the resource quantity hypothesis states that resource availability affects plant diversity (Ricklefs, 1977; Tilman, 1985). Ultimately, both resource quantity and heterogeneity drive understory vegetation dynamics (Bartels & Chen 2010), emphasizing the importance of the variation of their influence with stand age and disturbance. This suggests that the impact of stand structure on understory species may differ between stands of different successional stages and may be controlled by different mechanisms. Conversations around ecological restoration have become front and center of ecosystem management due to increasing societal awareness of the large-scale environmental 2 degradation that has occurred. Forest degradation has recently been defined as structural changes to forest cover, which includes the conversion of primary forests into plantation forests (Regulation 2023/1115, 2023). One restoration method in plantations is to diversify planted stands to improve floristic diversity, increase ecosystem services, and enhance resilience to disturbances such as disease and pests (Messier et al., 2022). Further, mixed stands comprising trees of varying ages and species create a more complex structure and heterogenous understory resource environment, which supports higher understory diversity and provides nesting and forage opportunities for birds and small mammals (Cavard et al., 2011; Hart & Chen, 2006). Another way to bring heterogeneity to the forest floor is to open the canopy by creating artificial canopy gaps or through selective stem thinning. These treatments increase the resources that reach the forest floor which may encourage the growth of different understory species, increasing vegetation cover and diversity (Davis & Puettmann, 2009; Hart & Chen, 2006). Understory vegetation is then usually left to passive restoration methods without human intervention, allowing the plants to re-establish from existing propagules in the soil. Conversely, active restoration relies on assistance from humans through planting and seeding. While there is evidence that active restoration methods may accelerate ecosystem restoration, outcomes vary with factors such as previous management history and natural resilience of the site (Meli et al., 2017). The outcomes of a restoration project are partially dependent on the ecosystem memory of a site. Ecosystem memory refers to the impact of pre-disturbance conditions on post-disturbance recovery. This includes propagules in the soil that the understory regenerates from after a major disturbance (Bergeron et al., 2017). The ecosystem memory of 3 a stand adapts over time to the natural disturbance cycles of a forest stand and is shaped by the stand's history. For example, stands that experience regular wildfire cycles may select for wildfire-adapted species. It is, therefore, important to have a solid understanding of the ecosystem memory of a specific stand when making management decisions. The ecosystem memory will impact the species that recover via passive restoration and may dictate whether active restoration methods should be considered. This thesis investigated the understory dynamics in a young monoculture lodgepole pine (Pinus contorta var. latifolia Engelm. ex S. Wats.) plantation located in central-interior British Columbia. A vast area of this plantation was recently subjected to a suite of habitat treatments with the goal of improving habitat for moose populations in the area. The first objective (Chapter 2) was to determine if there was an effect of varying stem densities, gap sizes, time since treatment, stand structure, and forest floor characteristics on understory plant abundance, richness, diversity, and composition. This chapter also investigated the impact of restoration treatments on wildlife presence based on presence/absence of feces. The second objective (Chapter 3) was to use a seedling emergence trial to identify the species in the soil seed bank which plants may germinate. These chapters specifically focused on plant species of importance to moose browsing. This research will assist with decisions about the future habitat management of this area and will inform restoration efforts in similar stands. 4 Chapter 2: Effect of Restoration Treatments on Understory Vegetation and Wildlife Habitat Use Introduction In an effort to combat forest loss, maintain production of wood and wood products, and restore ecosystem services such as carbon storage, biodiversity, and water and air purification, millions of trees have been planted worldwide over the last several decades (Holl & Brancalion, 2020; Paquette & Messier, 2010). Over 45% of those planted forests comprise monoculture plantations: single-species, even-aged stands with evenly spaced trees, which are usually intensely managed, also known as pure stands or monocultures (FAO, 2020). While monocultures allow for large amounts of uniform product, with sometimes intense management focused on maximizing growth rate and wood quality, they rarely resemble the natural stands they are replacing (Sedjo, 2001). However, forest management has been moving increasingly toward creating forests that more closely resemble natural unmanaged stands, with complex stand structures and diverse forest floors (Perring et al., 2015). This shift in management is due to mounting concern that monocultures may actually be detrimental to production and forest health over longer time frames (Kelty, 2006; Messier et al., 2022; Pretzsch & Schütze, 2016). Plantations decrease resistance to disturbances such as disease and pests and can negatively impact forest response to natural disasters such as drought and wildfires, which more recently have been exacerbated by climate change (Jactel et al., 2017, 2021; Messier et al., 2022). They are characterized by simplicity in stand structure and age, which leads to low levels of both functional and compositional biodiversity, particularly in the understory vegetation (Ampoorter et al., 2020; Felton et al., 2016; Messier et al., 2022). These 5 shortcomings of plantations can, however, be improved by planting trees that are a mix of species, refraining from intensive vegetation management and removal, and practicing selective logging to retain older trees and snags to preserve stand complexity (Aubin et al., 2008; Hartley, 2002; Paquette & Messier, 2010). The understory vegetation layer includes all vascular and non-vascular vegetation less than 1.3 meters in height (shrubs, forbs, seedlings, sedges and rushes, mosses, lichens, etc.); it contributes significantly to forest productivity and nutrient cycling, as well as providing shelter for birds and small mammals, while offering forage for many ungulate species, such as moose, deer, and elk (Gilliam, 2007; Hobson & Bayne, 2000; Hodder et al., 2013). The composition and abundance of understory vegetation is influenced by resource availability (moisture, light, nutrients), competition, site factors such as soil type, slope, and aspect, and the history of the stand (Brosofske et al., 2001; Gilliam, 2007; Hart & Chen, 2006). In stands characterized by low stand variation (even-aged, single-species stands) with closed canopies, it is common to see lower understory cover (Hart & Chen, 2006; Messier et al., 1998). Potential methods to enhance understory diversity and abundance include creating artificial gaps in the canopy or by selectively logging to reduce tree density, thereby increasing resource availability on the forest floor and encouraging understory growth. Thinning can enhance the functional diversity of the understory by improving heterogeneity and increasing water and light availability, with the potential for lasting positive impacts on understory growth (Ares et al., 2010; Davis & Puettmann, 2009; Willms et al., 2017). Similarly, tree gaps may promote heterogeneity by increasing resources locally, encouraging the growth of understory species that require different growing conditions and increasing herbaceous plant diversity at the stand-level (De Grandpré et al., 2011; Hart & Chen, 2006). 6 A variety of factors can impact understory regeneration following thinning or canopy manipulation. First, vegetation response to treatment could experience an initial lag due to high initial establishment of early succession and disturbance-dependent species and low recruitment of seeds from neighboring stands (Bartels & Macdonald, 2023; Hart & Chen, 2006; Rossman et al., 2018). There could also be a lack of or an excess of resources such as water, light, and nutrients which may be affected by stand density and canopy closure (Hart & Chen, 2006). Or, there may be an unsuitable environment on the forest floor including downed trees, leaf litter, topography, and edaphic conditions which are limiting plant growth (Brosofske et al., 2001). There have been few studies in northern boreal forests that have sought to assess all of these factors together with canopy treatment to get a full picture of post-treatment understory plant regeneration, and only a handful that have examined early regeneration within the first 3 years of treatment (Bartels & Macdonald, 2023; Haughian & Frego, 2016; Purdon et al., 2004; Rossman et al., 2018), with a majority of studies focusing on regeneration 10+ years after disturbance. While some studies have addressed the impacts of forest disturbance and management on understory plant communities (Ares et al., 2010; Bartels & Macdonald, 2023; Davis & Puettmann, 2009; Willms et al., 2017) or on wildlife use (Sullivan et al., 2010, 2002; Johnson & Rea, 2024; McLaren et al., 2000), the effect of experimentally altering overstory canopy cover to improve moose forage remains largely unexplored in central-interior British Columbia. The goal of this study was to examine the effects of restoration thinning and canopy gap treatments on understory plant community and wildlife forage provisionings in a young pine monoculture in central-interior British Columbia. The specific objectives were (1) Determine whether thinning and gap treatments affected understory cover, richness, 7 diversity, and species composition 1-year and 3-years post-treatment. (2) Assess the stand structure and forest floor to understand potential site limitations on understory regeneration beyond the treatments. (3) Evaluate if wildlife presence in an area was impacted by different thinning densities or gap sizes. We hypothesized that treated stands will have higher understory species richness and diversity compared to untreated controls, with treated stands more dominated by earlysuccessional species. Additionally, we hypothesized that time since treatment may impact understory vegetation metrics, with increased cover, richness, and diversity measured 3-years post treatment versus 1-year post-treatment (H1). We hypothesized that understory cover, richness, and diversity would be limited by mechanical and biological barriers in germination resulting from forest floor characteristics such as a thick duff layer and high percent cover of bare ground, coarse or fine woody debris, leaf litter, or rock (H2). We also hypothesized that evidence of wildlife would be found more frequently in treated stands than in control stands, with the goal that these results would highlight preferred stand densities and gap sizes (H3). Addressing these objectives will help to clarify the benefit of altering monoculture plantations for ungulates. Materials and Methods Study Area The study area was located approximately 60 km southeast of the district municipality of Vanderhoof in the Interior Plateau region of central-interior British Columbia. This area is characterized by dense coniferous forests dominated by even-aged, lodgepole pine (Pinus contorta var. latifolia Engelm. ex S. Wats.) plantations approximately 18-25 years old. The study sites are located in the Fraser Plateau ecoregion in the Nazko Upland ecosection, in the 8 sub-boreal spruce (SBS) biogeoclimatic zone (BEC) in the dry warm (dw3) subzone. The mean annual temperature in this area was 3.1° C and the mean annual precipitation was 536 mm from 2011-2020 (Wang et al., 2016). Historically, this area comprised lodgepole pine dominant stands with minor components of Douglas-fir (P. menziesii var. glauca (Beissn.) Franco) and hybrid white spruce (Picea glauca x engelmannii) (DeLong et al., 1993). Lodgepole pine in this area were affected by the mountain pine beetle (Dendroctonus ponderosae) over the last 25 years and were thus salvage logged and replanted (Lewis, 2009; Taylor et al., 2006). The topography of the area is primarily flat, with an elevation between 950 and 1200 m. The study area lies between two lakes (spanning approximately 150 ha and 480 ha, respectively) and contains many small unnamed lakes, creeks, and drainages. Soils have formed on predominantly morainal and lacustrine materials (DeLong et al., 1993). There is no known evidence of herbicide use in the area (Brandon Geldart, personal communication, December 3, 2024). Restoration treatments As part of broader efforts to address the loss of ungulate forage and a decrease in the supply of moose winter range, the BC Provincial Government recently initiated a large-scale ecological restoration experiment. A suite of restoration treatments, variable-density thinning and artificial canopy gaps, were created in the area to encourage understory regrowth and moose forage availability in this pine-dominated forested landscape. Several thinning treatments (200, 400, and 600 stems/ha) and four canopy gap sizes (0.2, 0.5, 1.0, and 2.0 ha) were created within three adjacent mapped hexagons, approximately 14 km2 each (Figure 1). Each hexagon containing the full suite of treatments covers the winter home range size of one female cow moose (Cederlund & Okarma, 1988). Each thinning treatment was 9 approximately 10 ± 5 ha in area. And in each gap treatment all existing stems were removed (stem density of 0 stems/ha). The first hexagon (Hex1) was treated in 2020 using heavy machinery, and the second (Hex2) and third (Hex3) in 2022 using manual brush saws for thinning treatments and full-sized bunchers for gap treatments. This corresponded to 3-years and 1-year post-treatments, respectively. Slash was left on site in piles and burned in fall 2023. Data collection Data collection was conducted during peak vegetation cover (June – August) in the summer of 2023. It is important to note that in 2023 the study area was experiencing drought conditions. This area saw a mean annual precipitation of 473 mm in 2023, compared to an average of 536 mm between the years 2011 and 2020 (Wang et al., 2016). Within the study area, three replicates of each thinning and canopy gap treatment were randomly selected for sampling using a random number generator in both 1-year and 3year post-treatment stands. In the untreated forest stands between the treatments, 3 separate areas (approximately 10 ± 5 ha in area) were sampled as the controls. One untreated control per hexagon was selected based on similarity to nearby treated stands (i.e., age) and convenience of access. A total of 47 plots were sampled across the study area (Table A1). In each selected thinning or gap treatment, a 400 m2 (11.28 m fixed radius) circular plot was established from the centre of the treated area, within which all vegetation measurements were taken. From the center of each circular plot, a 50 x 50 cm quadrat was placed at 2 m intervals along each transect in all cardinal directions (N, E, S, W), resulting in a total of twenty 50 x 50 cm quadrats sampled per plot (Figure 1). 10 Within each quadrat, the percent leaf coverage of all understory species in the herbaceous (herb) layer (including forbs, graminoids, and woody species < 1.3 m tall) was estimated (Anderson, 1986). Plants were identified in the field using a local field guide (Parish et al., 1996). The percent coverage of the following substrate characteristics was also captured for each quadrat: bare ground, rock, leaf litter, moss, lichens, fine wood (dead woody material < 7.2 cm diameter), coarse wood (dead woody material > 7.2 cm diameter), and duff depth. Leaf cover of woody species > 1.3 m tall (shrub layer) was estimated in each plot using four 5 x 5 m quadrats placed on each transect in each cardinal direction. Bias of cover estimates was minimized by using a single observer for every estimate. The entire plot area was surveyed for a complete census of all understory plant species present to include those not captured in quadrat surveys. Stand structure characteristics measured included the number of live stems and snags; canopy closure, which was estimated by using a convex spherical densiometer; and height and diameter-at-breastheight (DBH) of all trees > 5cm diameter, which was used to calculate basal area (π/4 × DBH²). Additionally, the plot was surveyed to capture evidence of wildlife; this was done by recording presence or absence and the representative species (bear, ungulate, grouse) of each feces pile present (identified using Halfpenny, 2001). Number of pellets was not recorded. 11 Figure 1: A diagram detailing the sampling design. One mapped hexagon of approximately 14 km2 is represented, containing several treated areas (represented by the shaded areas). Each treatment area represents a thinning density (200, 400, 600 stems/ha) or a canopy gap (0.2, 0.5, 1.0, 2.0 ha). In the centre of each treated area, a 400 m2 circular plot was established (fixed radius of 11.28 m). Within this plot, 4 transects in each cardinal direction were laid out, along which quadrat sampling occurred. Five 50 x 50 cm quadrats and one 5 x 5 m quadrat were sampled along each transect, for a total of 20 50 x 50 cm and 4 5 x 5 m plots sampled per plot. Data Analysis Understory vegetation response variables included (1) leaf area cover (calculated as the sum percent cover of all vascular plants averaged for all quadrats per plot), (2) species richness (total number of species per plot), (3) species diversity (calculated as inverse Simpson index (1/ ∑ 𝑃𝑖 2 ) where Pi is the proportional leaf area cover of each species in a quadrat, averaged for all quadrats per plot) and, (4) species composition (based on average percent cover per species for all quadrats per plot). Inverse Simpson index was chosen over other diversity 12 indices due to the emphasis on the species evenness component and for ease of interpretation (DeJong, 1975). All analyses were performed separately for the “herb layer,” which includes all cover data for 50 cm x 50 cm plots, and “shrub layer,” which includes all cover data for 5 m x 5 m plots. Forest structure attributes such as stem density, basal area, and canopy cover were averaged across the replicate plots for each treatment. The effects of the thinning and canopy gap treatments and their interaction with time since treatment on understory vegetation cover, richness, and diversity were analyzed using linear mixed-effects models (LMEs) and general linear mixed-effects models (GLMEs). In these models, the fixed effects were treatment (thinning density or canopy gap size), time since treatment, and the interaction between the treatment x time since treatment; plot was included as the random effect. The models were run separately for the thinning and canopy gap treatments. LMEs were performed using function lmer in package lme4 for the cover and diversity models. Species richness (count data) was analyzed using the glmer function in package lme4 (v1.1.34; Bates et al., 2015). When a significant interaction or main effect was obtained (α = 0.05), post-hoc tests were performed using estimated marginal means (emmeans) in the emmeans package (v1.10.0; Lenth, 2024). P-values were not adjusted for multiple comparisons throughout this analysis due to low sample size and to prevent underreporting of significant results (Schulz & Grimes, 2005) To capture any potential effects of observed stem density on understory cover, richness, and diversity, six linear models were also run. For each herb layer and shrub layer cover, richness, and diversity response variable, one model was run with tree density as the 13 explanatory variable, and one model was run with the square of tree density as the explanatory variable. These models included data from all control and thinning plots (n=23). The effects of the thinning and canopy gap treatments and their interaction with time since treatment on understory plant species composition were analyzed using permutational multivariate analysis of variance (PERMANOVA). Composition analysis was conducted separately on thinning treatments and gap treatments data. PERMANOVA analysis was conducted using the adonis2 function in vegan package (v2.6.4; Oksanen et al., 2022) using 999 permutations. Post-hoc tests were performed using pairwise.adonis function in the package pairwiseAdonis (v0.4.1; Martinez Arbizu, 2017) to determine which treatments had significantly different species composition. Nonmetric multidimentional scaling (NMDS) was performed with 2 dimensions and Bray-Curtis distance using the metaMDS function in the package vegan (Oksanen et al., 2022). Thinning treatments and gap treatments were run separately. After 20 iterations the final stress was 0.223 (thinning treatments) and 0.236 (gap treatments). Finally, to identify species that were strongly associated with specific treatments (thinning densities, gap sizes, or the untreated control), indicator species analysis (ISA) was conducted on the herb layer data using the function multipatt from the package indicspecies (v1.7.14; De Cáceres & Legendre, 2009). The ISA analysis was grouped by time since treatment and treatment and run separately for thinning and gap treatments with 999 permutations. The relationships between understory metrics (cover, richness, and diversity) and forest floor characteristics (average duff depth, percent cover of coarse wood, fine wood, leaf litter, rock, and bare ground) were analyzed using linear regression models. One model for 14 each cover, richness, and diversity was run which included all forest floor characteristics using the function lm in package stats (v4.3.1; R Core Team, 2023). Herb layer and shrub layer analysis were run separately. The presence of wildlife was measured by evidence of feces (presence or absence) found within the sample plots during the time of the survey. Generalized linear models (GLMs) were run by specifying the binomial family distribution to determine the relationship between wildlife feces presence/absence, thinning or canopy gap treatments, and time since treatment. GLMs were also run by specifying the binomial family distribution to determine the effect of treatment, and time since treatment on ungulate feces presence specifically. The significance of the model terms were examined using the function Anova from the package car (v3.1.2; Fox & Weisberg, 2019). The package emmeans was then used to run post-hoc tests on results with significant p-values; the function was adjusted to compare the probabilities of the effect of each treatment. Results Across all plots (all thinning treatments, gap treatments, and untreated controls), 63 understory vascular plant species were identified within both the herb and shrub layers (Table A2). Sixty-one species in the herb layer were identified across all plots, 12 different species in the shrub layer were identified across all plots, and only two species were found within the shrub layer that were not recorded within the herb layer. Thirty-six perennial forb species, 3 annual forb species, 22 deciduous shrub species and 2 evergreen shrub species were identified across all plots. A majority of the species present in the treatments were native (5 exotic species, listed in Table A2). Of the species identified in the study area, 5 are species that have been repeatedly cited as browsed by moose: Populus tremuloides, Rosa 15 acicularis, Rubus spp., Salix spp., and Viburnum edule (Breithaupt et al., 2024; Franzmann & Schwartz, 2007; Haeussler et al., 1990; Hodder et al., 2013; Poole & Stuart-Smith, 2005; Renecker & Hudson, 1992). Stand structure characteristics No significant differences were found in the average basal area (m2/ha), canopy closure (%), and observed stem density (stem/ha) between the different thinning treatment densities; differences in stand structure were only found between the treated plots and the untreated control plots (Table 1). The control plots had approximately 13 times higher basal area than the 200 stems/ha thinning treatment and almost 12 times higher basal area than the 600 stems/ha thinning treatment. Higher canopy closure was associated with higher thinning density, with the treatments having an average of 13.22% - 25.4 % canopy closure (open) and the control plots having almost complete canopy closure with approximately 98% closure. Finally, the stem densities in the treatments ranged from an average of 453 – 629 stem/ha, while the control plots were almost 6 times denser than the lowest thinning treatment (200 stems/ha) and over 4 times denser than the highest thinning treatment (600 stems/ha). There was no significant effect of tree density on herb layer or shrub layer cover, richness, or diversity (Table A3) likely due to the smaller number of replicate sample plots. Due to time and logistical restraints, only a portion of all the treatments performed in the area were randomly selected for sampling. It is possible that with more extensive sampling of a greater number of plots, more conclusive trends may emerge within the data. This could potentially provide information about which stem thinning density to prioritize in similar forested areas to benefit understory cover, richness, and diversity. 16 Table 1: Summary of stand structure variables (basal area, canopy closure, and observed stem density) at the three different thinning densities (200, 400, and 600 stems/ha) and the untreated control. Values represent average of the measurement for all replicates of the treatments with the standard error. Different letters represent significantly different values (at α = 0.05). Basal Area (m2/ha) Canopy Closure (%) Observed stem density (stems/ha) 200 stems/ha 2.14 ± 0.62a 13.22 ± 5.40a 453 ± 71a 400 stems/ha 2.07 ± 0.44a 15.51 ± 4.62a 721 ± 200a 600 stems/ha 2.38 ± 0.52a 25.43 ± 6.78a 629 ± 90a ctrl 28.20 ± 1.23b 98.11 ± 0.33b 2666 ± 28b Effects of treatments on understory cover, richness, and diversity The average percent cover of herb species did not vary between thinning treatments, but it varied significantly between the canopy gap treatments (Table 2). Post-hoc tests revealed that herb cover was significantly different between the 1.0 ha gap (x̄ = 25.9%) and the control treatment (x̄ = 16.4%), 0.2 ha gap (x̄ =22.6%) and the control (Fig. 2A-B). Moreover, herb species cover in both thinning and canopy gap treatments was significantly higher in year 3 than in year 1 post-treatment (Fig 2C-D). Shrub layer species cover, however, did not vary significantly among the thinning treatments nor the canopy gap treatments in either time periods (Table 2). Herb species richness varied significantly between the thinning treatments, with lower species richness in the control plots (x̄ =12) when compared to 200 stem/ha treatment (x̄ =17.6) and 600 stem/ha (x̄ =19). However, there was no significant difference in the herb richness between different thinning densities (Figure 3A). There was a significant difference in herb species richness between gap treatments, with significantly lower richness in the control plots (x̄ =12) when compared to 0.2 ha gap (x̄ =18.2), 0.5 ha gap (x̄ =17.2), 1.0 ha gap (x̄ =19.8), and 2.0 ha gap (x̄ =19) (Figure 3B). Herb species richness was significantly 17 higher in year 3 (x̄ = 18.4) than in year 1 (x̄ = 14.3) post-treatment within the thinning treatments (Fig 3C), but there was no significant difference with differences in time within the gap treatments (Figure 3D). Within the shrub layer, both thinning and canopy gap treatments had no significant effect on richness, nor did they vary with time since treatment (Table 2). There was no significant relationship between herb diversity and the thinning treatments or the gap treatments, nor was there a difference in herb diversity between the 1year and 3-year post treatment plots (Table 2). In the shrub layer, significant differences in diversity were only found between treatments in 3-years post treatment treatments (Figure 4A-B). Diversity was higher in the 3-years post treatment (x̄ =1.47) compared to 1-year posttreatment (x̄ =1.31) in the thinning treatments (Figure 4C-D). There was a significant interaction found between the terms in both the thinning and gap treatments (Table 2) Table 2: Effect of thinning and gap treatments (trt), time since treatment (time), and their interaction (trt x time) on percent cover, species richness, and species diversity within the herbaceous (herb) and shrub layers. Values are the F-value (p-value) of the linear mixedeffects models (cover and diversity) and the Chi-squared (p-value) for the general linear mixed-effect models (richness). Significant p-values (α = 0.05) are in bold text. Trt Time Trt x Time Trt Time Trt x Time Thinning Treatments Gap Treatments Cover Richness Diversity Cover Richness Herb layer 1.61 (0.221) 9.30 (0.026) 0.60 (0.626) 4.08 (0.014) 13.38 (0.0095) 7.71 (0.013) 5.02 (0.025) 0.33 (0.576) 31.48 (<0.0001) 0.0016 (0.969) 0.85 (0.487) 2.64 (0.451) 1.67 (0.219) 2.71 (0.06) 2.85 (0.583) 0.40 (0.758) 0.83 (0.380) 0.55 (0.658) 1.22 (0.748) 0.37 (0.543) 0.21 (0.976) Shrub layer 3.34 (0.078) 0.81 (0.538) 46.66 (<0.001) 2.81 (0.127) 8.57 (0.021) 2.09 (0.166) 0.74 (0.946) 0.0091 (0.924) 1.31 (0.859) Diversity 1.79 (0.174) 1.54 (0.230) 1.03 (0.420) 2.23 (0.185) 4.34 (0.064) 6.79 (0.027) 18 Figure 2:Average herb (herbaceous) percent leaf cover within all thinning treatments and untreated control (ctrl) (A) and within all gap treatments and untreated control (ctrl) (B). Average herb percent leaf cover within thinning treatments one-year post-treatment vs threeyears post-treatment (C) and average herb percent leaf cover within gap treatments one-year post-treatment vs three-years post-treatment (D). Error bars represent standard error. Different superscripted letters represent significantly different values (α=0.05). 19 Figure 3: Average herbaceous (herb) species richness within all thinning treatments and untreated control (ctrl) (A) and within all gap treatments and untreated control (ctrl) (B). Average herb species richness within thinning treatments one-year post-treatment vs threeyears post-treatment (C) and average herb species richness within gap treatments one-year post-treatment vs three-years post-treatment (D). Error bars represent standard error. Different superscripted letters represent significantly different values (α=0.05). 20 Figure 4: Average shrub species diversity within all thinning treatments and untreated control (ctrl) (A) and within all gap treatments and untreated control (ctrl) (B). Average shrub species diversity within thinning treatments one-year post-treatment vs three-years post-treatment (C) and average shrub species diversity within gap treatments one-year posttreatment vs three-years post-treatment (D). Error bars represent standard error. Different superscripted letters represent significantly different values (α=0.05). Uppercase and lowercase letters were used to separate year since treatment. Effect of treatments on understory species composition and indicator species Herb species composition was significantly influenced by thinning treatment, gap treatment, and time since gap treatment but not their interactions (Table 3). Post-hoc tests show a significant difference in herb species composition in each of the thinning treatments (200, 400, 600 stems/ha) versus the control, but no significant difference among the various thinning densities (Table A8). Similarly, there was a significant difference in herb species 21 composition between the gap treatments (0.2, 0.5, 1.0 ha) and the control, but no significant difference in the composition between different gap sizes (Table A9). This is represented visually using nonmetric multidimensional scaling (NMDS) plots (Figure 5), which shows similarity between herb species composition of each plot organized by treatment as represented by relative distance between points in space (points closer in space are more similar). These plots reiterate that the herb species composition of the control plots was different from the thinning treatment plots and the gap treatment plots. Further, PERMANOVA tests show a difference in herb vegetation composition with time since treatment, but only between the gap treatments (Table 3). No significant interaction was found between treatment and time since treatment. Within the shrub layer, no significant differences in composition were found between thinning treatments, gap treatments, time since treatment, or their interaction (Table 3). Indicator species analysis showed which species were associated with each treatment and indicated the following: Chamaenerion angustifolium was an indicator species in all the thinning and canopy gap treatments (not the control); Fragaria virginiana was an indicator of 600 stems/ha in the 1-year post-treatment; Rosa acicularis was an indicator of every gap treatment in both the 1-year post-treatment and 3-year post-treatment; Pyrola asarifolia was an indicator for 0.5 ha and 2.0 ha gap in the 1-year post-treatment; 1.0 ha in the 3-year posttreatment, and the controls; Rubus sp. was an indicator for 0.5 ha and 1.0 ha in the 1-year and 1.0 ha in the 3-year; Goodyera oblongifolia in the 2.0 ha 1-year and the controls; and Smilacina racemosa was an indicator for the controls (Table 4). 22 Table 3: Results of permutational multivariate analysis of variance (PERMANOVA) of effect of thinning treatment, gap treatment, time since treatment, and the interaction between treatment and time on species composition (61 species) in the herbaceous (herb) and shrub layers. Values represent F-value (p-value) with 999 permutations completed. Significant relationships (α = 0.05) are in bold text. Treatment Time Treatment x Time Herb layer Thinning treatment Gap treatment 2.31 (0.002) 1.83 (0.012) 1.05 (0.397) 2.46 (0.008) 0.63 (0.917) 1.13 (0.275) Shrub layer Thinning treatment Gap treatment 0.36 (0.981) 0.83 (0.685) 0.86 (0.498) 1.96 (0.068) 0.83 (0.626) 0.88 (0.587) Figure 5: Nonmetric multidimensional scaling (NMDS) plots of understory herbaceous (herb) plant composition grouped by thinning treatment (A) and gap treatment (B). After 20 iterations the final stress is 0.223 (A) and 0.236 (B). NMDS was run using Bray-Curtis distance. Each point represents one plot; a representatively coloured polygon groups the replicates of each thinning treatment (200, 400, 600 stems/ha, or untreated control) or gap treatment (0.2, 0.5, 1.0, 2.0, or untreated control). Similarity in species composition between plots is represented by distance between points in space. Overlapping polygons represent similar species composition between different treatments. 23 Table 4: Herbaceuous plants that were indicators of particular treatments (thinning density, gap size, or untreated control) after 1-year or 3-years post-treatment. A is the probability that the site will belong to the treatment if the species is found. B is the probability that the species will be found given the treatment. IndVal is the square root of the product of A and B. Probabilities are based on the model when run with 999 permutations. Only significant results (α = 0.05) are shown. Species Treatment A 1-year post-treatment B IndVal p-value Chamaenerion angustifolium 200, 400, 600 stems/ha 0.2, 0.5, 1.0, 2.0 ha 0.971 0.977 1.000 1.000 0.985 0.988 0.032 0.041 Fragaria virginiana Goodyera oblongifolia Pyrola asarifolia Rosa acicularis Rubus sp. Smilacina racemosa Spiraea sp. 600 stems/ha 0.787 control 1.000 0.5, 2.0, control 0.981 0.2, 0.5, 1.0, 2.0 0.996 0.5, 1.0 ha 0.986 control 0.999 0.2, 1.0, 2.0 0.959 3-year post-treatment 200, 400, 600 stems/ha 0.971 0.2, 0.5, 1.0, 2.0 ha 0.977 control 1.000 0.2, 1.0 ha 0.858 1.0 ha, control 0.981 0.2, 0.5, 1.0 0.996 1.0 ha 0.986 0.2, 0.5, 1.0, 2.0 0.959 1.000 0.667 0.667 0.857 0.889 0.667 0.857 0.887 0.816 0.809 0.924 0.936 0.816 0.907 0.012 0.017 0.048 0.013 0.014 0.027 0.044 1.000 1.000 0.667 1.000 0.667 0.857 0.889 0.857 0.985 0.988 0.816 0.926 0.809 0.924 0.936 0.907 0.032 0.041 0.017 0.005 0.048 0.013 0.014 0.044 Chamaenerion angustifolium Goodyera oblongifolia Hieracium umbellatum Pyrola asarifolia Rosa acicularis Rubus sp. Spiraea sp. Effect of forest floor characteristics on understory cover, richness, and diversity Understory cover, richness, and diversity were associated with forest floor characteristics such as duff depth, bare ground, coarse wood, fine wood, leaf litter, and rock (Table 5). For herb layer species, an increase in percent coverage of leaf litter was associated with a decrease in cover (p <0.001, R2=0.610), richness (p <0.001, R2 =0.320), and diversity (p = 0.018, R2=0.046). Increase in percent of bare ground was associated with a decrease in herb plant cover (p <0.001, R2=0.610) and richness (p = 0.029, R2=0.320). Duff depth, coarse wood, fine wood, and rock did not significantly affect herb vegetation values (Figure 6A). 24 An increase in the cover of fine woody debris was associated with a decrease in shrub layer diversity (p = 0.026, R2=0.152). No effect of duff depth, bare ground, coarse wood, leaf litter, or rock was found on shrub layer vegetation values (Figure 6B). Table 5: Effect of forest floor characteristics (bare ground, coarse wood, duff depth, fine wood, leaf litter, and rock) on understory plant cover, richness, and diversity in the herbaceous (herb) and shrub layers. Results of linear models (cover, diversity) and generalized linear model (richness) shown as model-coefficient (p-value). R2 value for each model is shown. Significant relationships (α = 0.05) are in bold text. Bare ground Coarse wood Duff depth Fine wood Leaf litter Rock R2 Cover -3.91 (<0.001) -0.79 (0.433) 0.12 (0.908) 0.27 (0.789) -7.41 (<0.0001) -1.00 (0.314) Herb layer Richness -2.19 (0.029) -0.78 (0.436) -0.34 (0.738) -0.11 (0.916) -4.44 (<0.0001) 0.06 (0.949) Diversity -0.04 (0.965) -0.69 (0.497) 0.47 (0.638) 0.64 (0.526) -2.46 (0.018) -0.64 (0.529) Cover -0.86 (0.395) -1.41 (0.165) -1.40 (0.169) -0.05 (0.959) 0.59 (0.555) -0.32 (0.753) Shrub layer Richness -0.27 (0.785) -0.54 (0.587) -1.27 (0.203) -0.59 (0.557) -1.18 (0.240) -0.45 (0.655) Diversity 1.45 (0.153) -0.51 (0.613) -0.03 (0.979) -2.30 (0.026) -0.93 (0.355) -1.77 (0.083) 0.610 0.320 0.046 0.025 0.217 0.152 Figure 6: Coefficient plots (95% confidence interval) for linear models showing effect of forest floor characteristics for the herbaceous (herb) layer (A) and the shrub layer (B). Red represents models for cover, green represents models for richness, and blue represents models for diversity. Significant variables are those that do not cross the 0.0 boundary. 25 Effects of understory cover, richness, and diversity on wildlife feces presence Presence/absence of wildlife feces (including ungulates, bears, and grouse) in the study plots was significantly related to thinning treatments but not canopy gap treatments (Table 6). Post-hoc tests suggested that thinning treatment was positively related to evidence of wildlife (ungulates, bears, and grouse) feces presence when compared to the control plots (Table A10). There was significant relationship between gap treatments and wildlife feces (Table A11). Twelve of the 20 thinned plots (60%) sampled and 11 of the 24 (46%) gap treatments sampled had feces present, compared to no presence (0%) in the untreated control plots (Figure 7). The treatments with the most plots where feces was found were the thinning treatments 600 stems/ha (4 plots) and 400 stems/ha (3 plots), and the 0.5 ha (3 plots), 1.0 ha (3 plots), and 2.0 ha (3 plots) gap treatments. Seven of the 20 (35%) thinned plots and 8 of the 24 (25%) gap treatments had instances of ungulate feces, specifically. However, there was no significant difference between treatments when looking specifically at ungulate feces. “Ungulate” feces in this area could include deer, moose, or elk, but due to limitations in the methodology they were not distinguished to species in this study. Table 6: Results of the generalized linear model showing the effect of thinning treatments, gap treatments, time since treatment, and their interaction on wildlife (bear, ungulate, and grouse) feces presence. Values represent Chi-squared value (p-value). Significant p-values (≤ 0.05) are in bold text. Treatment Time Treatment x Time Thinning treatments 10.3 (0.0162) 3.30 (0.0693) 1.96 (0.581) Gap treatments 7.98 (0.0922) 0.76 (0.385) 10.5 (0.0327) 26 Figure 7: Number of plots where feces (distinguished as belonging to bear, grouse, or ungulate) was found in all replicates within each thinning treatment density (200, 400, 600 stems/ha) (A), gap treatment size (0.2, 0.5, 1.0, 2.0 ha) (B) and untreated control (ctrl) among both 1-year and 3-years post-treatment plots combined. Discussion Herb layer response to restoration treatments No significant difference was found in observed stem density, basal area, or canopy closure between the different thinning treatments (200, 400, and 600 stems/ha). Despite this, analysis was still run to compare understory cover, richness, diversity, and composition of plots in each thinning density, gap size, and control stands. This was done to capture the reality of these treatments and to compare the realistic outcome of performing targeted stem thinning treatments in a lodgepole pine monoculture. However, for the purposes of this study, the differences between treatments (all thinning treatments or all gap treatments) and controls were considered in interpretation of results. 27 Both thinning treatments and canopy gap treatments appear to have affected the cover and richness of the understory vegetation in the herb layer. This supports our hypothesis (H1) and agrees with the findings of other studies that performed thinning and gap treatments in conifer stands (Ares et al., 2010; Davis & Puettmann, 2009). These results reinforce the theory that by opening the canopy, resources such as moisture and light were increased, and thus, a wider variety of species were able to regenerate (Bailey et al., 1998; Davis & Puettmann, 2009; Hart & Chen, 2006). The fact that many of the species recorded in the treated plots were shade-intolerant, disturbance-dependent species also supports this. The lack of significant difference in cover and richness between the different size gap treatments was expected because every gap treatment has the same tree density (0 stems/ha) with a 100% open canopy; therefore, there are not likely to be significant differences in the understory species between these treatments. It is more important to regard the richness and diversity these treatments add when considering these areas with the surrounding untreated stands (Davis & Puettmann, 2009; Fahey & Puettmann, 2008). Canopy gaps encourage early-successional species to grow adjacent to late-successional stands by increasing the local resource availability (De Grandpré et al., 2011). In this regard, the gap treatments have increased the overall richness of the stand. Herb layer diversity was unaffected by the treatments; this indicates that while the treatments allowed more species to regenerate (increased richness), many species did not regenerate in high abundance (low species evenness). This does not support our hypothesis (H1), as herb layer diversity was expected to increase with heterogeneity introduced by the treatments. In the treatments, the herb layer was dominated by several early-seral species (C. angustifolium, H. umbellatum, R. acicularis, Rubus sp., Spiraea sp.) and grasses, shade- 28 intolerant species that colonize open and recently disturbed areas and are often found in thinned areas (Leck & Schütz, 2005; Ringius & Sims, 1997). In contrast, a majority of the other species grew sparser and were often found in small patches. Species compositional comparisons between treatments and controls and indicator species analysis also indicated that early-successional species dominated treated plots. Species such as R. acicularis, Rubus sp., and Spiraea sp. found profusely throughout the treated plots are moose forage species (Breithaupt et al., 2024; Haeussler et al., 1990; Renecker & Hudson, 1992). However, these early-seral species may be inhibiting regeneration of more important species for moose (such as willow) by outcompeting them for resources (Bailey et al., 1998; Meier et al., 1995). This contrasts with G. oblongifolia and S. racemosa, indicator species of control plots only; G. oblongifolia is a shade-tolerant species and often found in mossy forests, while S. racemosa is tolerant of deep shade but grows best under open canopies and is often found in forest gaps (Haeussler et al., 1990; St. Hilaire, 2002). Neither of these species have been reported as important for moose forage. It is important to note that in 2023, when sampling occurred, the region where the stand is located was experiencing drought conditions. These dry conditions likely impacted the understory vegetation communities and how the understory vegetation responded to the treatments. Dry conditions may have limited growth of understory plants, with increased stress negatively impacting cover (Page et al., 2005). Further, it is possible that drought conditions impacted the species present, with species maladapted to low moisture conditions, such as isohydric species, having a lower likelihood of survival (McDowell et al., 2008). Therefore, potentially under non-drought conditions higher percent cover of species, as well 29 as increased understory species richness, may be seen. This could lead to more dramatic differences in the understory vegetation between treatments and controls. Shrub layer response to restoration treatments Our findings suggest that within the shrub layer, cover and richness, and diversity were unaffected by the treatments. We hypothesized that the cover, richness, and diversity of the shrub layer would be higher in plots with thinning treatments compared to controls, so these results do not support our hypothesis (H1). Possibly, this is because tall shrub species may require longer than herb species to recover fully after disturbance (Davis & Puettmann, 2009). Species such as Shepherdia canadensis, Populus tremuloides, and Sorbus scopulina were not present in the controls but had significant regrowth in the treated plots. By eliminating the upper and lower canopy layers, which included tall shrubs such as Alnus sp., existing stems, propagules, or seeds were able to thrive with the increased resources posttreatment. This has important implications for moose since those shrubs that were able to regenerate, such as Salix spp. and P. tremuloides, are species often browsed by moose (Breithaupt et al., 2024; Hodder et al., 2013; Poole & Stuart-Smith, 2005; Renecker & Hudson, 1992). Specifically, Salix spp. and P. tremuloides regenerate rapidly after cutting and are known to increase in abundance after overstory removal (Haeussler et al., 1990). There is current evidence that this may be the case. For instance, while there was no significant difference in shrub cover, there was a lower proportion of cover composed of Alnus sp. and a higher proportion of cover of other species, such as Salix spp. and Lonicera involucrata, in the treated plots compared to the controls. This suggests that if given more 30 time, a significant difference in shrub composition between treatments and controls may emerge. Effect of time since treatment Treatment had an effect on herb layer cover and richness and shrub layer diversity among both thinning and gap treatments, with higher cover, richness, and diversity found in 3-year post-treatment compared to 1-year post-treatment. This agrees with a study which found that the understory vegetation required 2 years after the creation of gap treatments to adjust to the new conditions (De Grandpré et al., 2011). This could be explained by a lag in regeneration; species that regenerate initially after the treatment will be early-seral, shade-intolerant species that maximize the increase in resources to recover quickly from propagules in the soil (Bartels et al., 2016; Hart & Chen, 2006; Meier et al., 1995). It is not until many years after the disturbance that the understory is able to adjust and react to the changes in the overstory (Bartels & Macdonald, 2023; Davis & Puettmann, 2009). As time since treatment increases, richness and diversity may continue to increase slowly; however, this may not be a viable solution for situations where the issue of wildlife use of the area is urgent. Influence of forest floor characteristics Beyond characteristics of the overstory, it is also important to focus on other factors that can impact understory vegetation, such as the forest floor. Leaf litter, bare ground and fine woody debris (3 of the 6 tested variables) were the only measured variables that impacted herb layer cover, richness, and diversity, somewhat supporting our hypothesis (H2). Leaf litter was primarily composed of pine needles left over from the treatments, which are slow to break down (Berg et al., 2010). Leaf litter may serve as a mechanical barrier through which new seedlings have difficulty penetrating, or it can block resources such as moisture 31 and light from reaching germinating seeds (Dupuy & Chazdon, 2008). Additionally, elemental fluxes from pine litter as a result of leeching caused by spring snowmelt may result in more acidic soils, impacting understory species composition (Augusto et al., 2003; Yavitt & Fahey, 1986). Bare ground may have been compacted by machinery, which has been shown to affect soil structure, nutrients, and moisture retention, negatively impacting understory forest vegetation growth (Demir et al., 2007; Godefroid & Koedam, 2004). Further, amount of fine wood had a negative impact on shrub layer diversity; piles of slash were left behind after treatments, which may prevent regeneration of the understory for similar reasons as leaf litter. Eventually, the large volume of aboveground woody debris will decompose, transferring carbon stores to the soil to assist understory growth; however, this may take decades (Herrmann & Prescott, 2008). Wildlife presence in the stands Both thinning treatments and gap treatments had a positive impact on the presence of wildlife (including ungulate, bear, and grouse) feces within the plots when compared to the untreated control. This agrees with our hypothesis (H3) and with Sullivan et al. (2007), who found a higher presence of wildlife fecal pellets in thinned stands (500 stems/ha) compared to stands with higher stem densities. They also found a positive relationship between wildlife presence and the abundance of forage. Further, it has been found that moose prefer stands dominated by more than two species of tree and tall shrub (Milligan & Koricheva, 2013). While this area is characterized as a lodgepole pine monoculture, in some of the treated plots, other species, such as P. tremuloides and Alnus sp., were regenerating to a degree that these areas are beginning to resemble mixed-species stands. 32 There were no significant effects of treatments on the presence of ungulate feces specifically despite an increased abundance of shrub species of interest for moose forage, such as Chamaenerion angustifolium, Lonicera involucrata, Populus tremuloides, Rubus sp., Salix spp., Vaccinium spp., and Viburnum edule in the treated plots. It is possible there were no significant findings for ungulates due to the complicated nature of the relationship between moose habitat preference and stem density; there are more factors to consider beyond stand structure, such as stressors like predators and road density, and presence of calves (Dussault et al., 2005; Kunkel & Pletscher, 2000; McLaren et al., 2000). Despite this, ungulate feces were found in thinning and gap treatments, indicating these species are present in these areas. Ungulates such as moose will spend time in areas that balance abundant forage with thermal cover and protection from predators (Dussault et al., 2004; Kunkel & Pletscher, 2000). However, it is difficult to ascertain if the treated areas are providing adequate forage and cover for ungulates; further studies would need to be performed to better understand how moose are using the area. Limitations and Opportunities for Further Research This study investigated the initial understory regeneration at 1- and 3-years postdisturbance. This information about early regeneration can be important for understanding the preliminary response of the understory to the treatments. However, limiting the study to such a recent time after the treatments limits our interpretation of the treatment impacts. For instance, it is expected that the primary understory vegetation to regenerate first will be early-seral, disturbance-dependent species (Bartels et al., 2016; Hart & Chen, 2006). As this stand progresses to a later seral stage with denser canopy cover, a higher proportion of shadeintolerant species may emerge (De Grandpré et al., 1993). Evidence from similar stands 33 already in the late-seral stage or examination of the species present in the soil seed bank may provide context to help predict the species that could germinate in the future. However longterm sampling at this site would be prudent to ensure that moose are well-supported in the future. In the absence of sampling the same area repeatedly over a number of years, adjacent areas that were treated a different number of years prior were used to represent differences in time since treatment. While these areas were close in distance to one another, and were similar in age, it is impossible to ensure that all abiotic conditions were the same between the areas. It is possible that the grouping of all the 3-year post-treatment plots in one hexagon and 1-year post-treatment plots in two other hexagons introduced a spatial effect on the understory vegetation. For the purposes of this study, evidence of wildlife presence in the stands was constrained only to observations of feces presence. These data served the purpose of giving a general idea of wildlife presence in the treated areas, but we acknowledge that it is not the most accurate method to estimate habitat use (Härkönen & Heikkilä, 1999). Future efforts could include size measurement of pellets to more accurately capture whether “ungulate” presence was deer, moose, or elk (Halfpenny, 2001). Also, counting the number of pellets can provide more information about habitat use, with higher number of pellets associated with the presence of more individuals or with longer time spent in the area (Månsson et al., 2011). Browse surveys could have been performed to understand wildlife use of the areas more accurately (Bergqvist et al., 2018). The precision of pellet surveys to understand populations can also be increased by using genetic analyses, estimating animal defecation rates, and calibrating observations with other density estimates (Moll et al., 2022; Rönnegård 34 et al., 2008). Further, the use of aerial surveys, hunter observations, and wildlife cameras could be other methods to more accurately detect the use of the treated plots not only by moose but by other ungulates and their predators (Crowley et al., 2025; Kenney et al., 2024; Moll et al., 2022; Rönnegård et al., 2008). Conclusion Within the study area, thinning and canopy gap treatments have increased understory cover and richness compared to nearby, untreated stands. This points to the fact that performing restoration treatments is beneficial to understory plant regeneration. Species composition differed significantly between treatments and controls, with treated plots dominated by shade-intolerant, early-seral species such as Chamaenerion angustifolium, Rosa acicularis, and Spiraea sp. Treatments have also shifted the woody shrub cover to decrease the proportional cover of Alnus sp. and increase the proportion of other species such as Shepherdia canadensis, Populus tremuloides, Sorbus scopulina, Lonicera involucrata, and Salix spp. However, competition from profusely abundant early-seral species may inhibit the regeneration of species of importance to moose. Further, large amounts of leaf litter, woody debris, and compact bare ground may limit germination due to blocking resources or mechanical barriers. Evidence of wildlife, as indicated by the presence of feces, was significantly higher in treated plots compared to untreated controls. While no significant relationship was found between the treatments and ungulate use specifically, ungulate feces was found in treated plots. This could be due to the increased abundance of certain forage species, such as P. tremuloides, Rubus spp., R. acicularis, and Salix spp., in some of the treated plots. While the presence of some browse species of interest seems promising for moose populations within 35 this area, regeneration of many of these species may not be strong or rapid enough to support moose populations without active restoration methods (seeding or planting of target species of interest). 36 Chapter 3: Assessment of the Soil Seed Bank and Ecosystem Memory Introduction Over the last half a century, forestry has been a major industry within Canada, with 700,000 – 800,000 hectares of forests harvested annually (Natural Resources Canada, 2023). This has resulted in large-scale establishment and management of planted forests to replace harvested stems (British Columbia Ministry of Forests, Lands and Natural Resources Operations, 20022011). Within these monoculture stands, clear-cut harvesting, selective logging, and salvage logging are all common harvesting practices, with clear-cut harvesting being the most popular in British Columbia (Berger, 2008; British Columbia Ministry of Forests, Lands and Natural Resources Operations, 2002). These methods involve the use of heavy machinery (such as tracked logging tractors, mobile yarders, bulldozers, etc.) to cut stems and can further disturb the forest floor, compacting the soil and sometimes destroying established understory vegetation (Berger, 2008). Following these disturbances, the understory is often left to regenerate via passive methods (i.e., left to naturally regenerate), with little invested in directing which species grow back. There are a variety of variables that can influence the establishment and regeneration of the understory, including but not limited to, resource availability, habitat heterogeneity, competition, and regenerative plant materials that remain in situ post-disturbance (Hart & Chen, 2006). Ecosystem memory refers to the materials and adaptations that endure after disturbance from which the ecosystem regenerates (Johnstone et al., 2016). This involves plant material that persists in the soil or forest floor, such as seeds, spores, roots, plant parts, and any regenerative material from which plants may propagate post-disturbance. This also includes material introduced from the surrounding environment, such as dispersing seeds 37 carried by wind or wildlife (Bergeron et al., 2017; Schweiger et al., 2019). Ecosystem memory also pertains to ways in which the composition of species present on site have adapted to disturbances, for instance, by shifting to a higher abundance of species welladapted to establish after disturbance (Johnstone et al., 2016). The ecosystem memory and soil seed bank can depend on disturbance severity, timing, and frequency, as dramatic changes in the disturbance regime can erase certain trait sets or taxa from the seed bank (Decocq et al., 2004; Keeley et al., 2011). The relationship between aboveground vegetation and the soil seed bank has been well-researched in grassland environments (Bekker et al., 1997; Kiss et al., 2016), with fewer studies investigating this relationship in forested ecosystems (Halpern et al., 1999; Plue et al., 2017; Stark et al., 2008; Zobel et al., 2007). Research into ecosystem memory following fires and major disturbances is also well-represented in the literature (Bergeron et al., 2017; McGee & Feller, 1993; Stark et al., 2006). However, relatively few studies have investigated the impact that harvesting may have on ecosystem memory and subsequent understory regeneration, especially in the context of intensive silviculture (Decocq et al., 2004; Qi & Scarratt, 1998; Stark et al., 2006). No studies have compared the soil seed bank between different stem densities or canopy treatments. Very few studies have also compared the seed bank after different amounts of time post-treatment, with one study investigating 1 and 2 years post-treatment (Qi & Scarratt, 1998) and one study investigating the differences between 1, 5, and 10 years post-treatment (Stark et al., 2006). Seed bank data can provide important information about the species composition, relative abundance, and potential distribution of regenerating vegetation needed to predict which species may populate the area in the coming years (Leck et al., 1989). 38 In one stand located about 60 km southeast of Vanderhoof in central-interior British Columbia, canopy restoration treatments were conducted in 2020 and 2022, providing an opportunity to study the impact of disturbance on the understory vegetation in a young 18-25 year-old conifer monoculture. Observations and analysis of the cover, richness, diversity, and composition in this stand highlight the potential recovery of aboveground species of interest for moose forage (refer to Chapter 2). However, the ecosystem memory and seed bank of these stands have not been studied. The species found in the seed bank are not often reflected by the species observed in the aboveground vegetation (Decocq et al., 2004; Qi & Scarratt, 1998), but an investigation into the seed bank of this stand can lend further insight into the potential future recovery of the stand and the mechanisms that may influence understory regeneration. The objectives of this study were (1) to examine the soil seed and spore bank of the stand to document the viable propagules in the soil from which vegetation may germinate in the future, (2) to compare the soil seed bank between treated plots (various thinning densities and gap sizes), untreated plots, and stands that had not been salvage logged with time since treatment (1-year and 3-years post-treatment) and (3) to compare the species found in the soil seed bank and those represented in the aboveground vegetation growing on-site. Two potential hypotheses can explain the limitations of understory regeneration at this site. First, it is possible that the treated stands have retained the pre-treatment seed bank and/or encouraged seed recruitment from nearby stands, but the target species are limited in their regeneration due to competition from early-seral species or inadequate resources (H1). Alternatively, the treated plots may have had poor recovery of the seed bank post-treatment; a difference in the seed bank composition between the treated stands and untreated stands 39 would support this (H2). Answers to these objectives will help determine if the soil seed bank can be relied on for regeneration of target species to benefit wildlife. Methods Study Area The study site was located approximately 60 km southeast of the district municipality of Vanderhoof in the Interior Plateau region of British Columbia. The study was conducted in the same study area and restoration treatment plots sampled in chapter 2 of this thesis (refer to Study Area in Chapter 2). This area falls within the sub-boreal spruce (SBS) biogeoclimatic zone (BEC) in the dry warm (dw3) subzone. It is characterized by secondgrowth forests comprised even-aged, lodgepole pine-dominated stands of approximately 1825 years old. The stand was divided into three adjacent hexagons, approximately 14 km2 each, for the purpose of performing canopy treatments. Three thinning treatments (200, 400, and 600 stems/ha) and four canopy gap sizes (0.2, 0.5, 1.0, 2.0 ha) were performed in this area in 2022 and in 2020 using manual brush saws for thinning treatments and full-sized bunchers for gap treatments. At the time of the study, the samples represented 1-year and 3year post-treatment respectively. Slash was left on site in piles and burned in fall 2023. The understory vegetation in the treated areas was primarily dominated by early-seral herb species, ruderals, and graminoids, with some plots containing Salix spp. and Populus tremuloides which likely regenerated from existing stems and suckers remaining from pretreatment. The untreated control stands located between the treated areas had an almost completely closed canopy and sparse understory layer with a thick layer of pine needle litter and high abundance of Alnus sp. Forest patches in the area that had not been salvage logged but were severely damaged by pine beetle were used as the reference forest to represent the 40 understory vegetation and seed bank without disturbance for over 30 years. These reference stands contained many downed trees, and the understory vegetation was more abundant than in treated or control plots and dominated by Vaccinium spp. and Spiraea sp. Sampling Method Sampling was conducted between May and August in the summer of 2023. Within the study area, three replicates of each canopy gap treatment and thinning treatment were randomly chosen for sampling using a random number generator; this was repeated in the 3-year posttreatment and 1-year post-treatment areas. Further, three patches of untreated stands (approximately 10 ± 5 ha in area) located between the treatments were sampled as the controls. Soil sampling occurred in the same plots as the aboveground vegetation sampling completed in Chapter 2 (refer to Data Collection in Chapter 2). Additionally, soil samples were collected in three nearby old growth patches (approximately 5 ha in area) located within the hexagons that had been badly damaged by bark beetle but not salvage logged (reference stands). One reference stand in each established hexagon was sampled, and areas were chosen based on convenience and availability; very few of these areas had been left standing. A total of 51 plots were sampled between all hexagons. In each selected replicate treatment within a hexagon, a circular plot area of 400 m2 (11.28 m fixed radius) was established from the plot center (see Figure 1 in Chapter 2). Within this plot, 3 soil samples were collected from random locations; one sample was collected close to the plot centre and the other two were collected at least 5 meters away from the plot centre and each other. The top 30 cm of soil were targeted using a Dutch auger, which is a sufficient depth to capture a majority of the viable seed bank (Decocq et al., 2004). All 3 samples were combined to create a single homogenized sample per plot, for a total of 51 samples. 41 Greenhouse Methods The ‘seedling emergence method’ was used to investigate the soil seed bank present at the field site, following the methods of Mackenzie & Naeth (2010). This method involves incubating soil samples under ideal conditions and recording the germinants that emerge. After collection from the field, the combined soil samples from each plot were stratified by being kept refrigerated at 3° C for 14 weeks. Soil samples were then sieved with a 3mm sieve to remove rocks and other debris, the volume of soil was recorded, and the sample was spread in a 1.5 cm thick layer on top of a layer of 2 cm thick sterilized sand in a 10 x 20 cm plastic tray. Each tray was covered with a humidity dome to prevent rapid drying out. Three control trays were filled with a layer of sterile potting soil on top of a layer of sterile sand and placed randomly among the treatments to determine whether cross-contamination between the trays occurred; no seeds germinated in the controls. Trays were randomly placed (using a random number generator to determine location) in a compartment at the Enhanced Forestry Lab at the University of Northern British Columbia Prince George campus that was kept between 15° C (nighttime temperature) and 25° C (peak daytime temperature). Grow lights ensured a 16-hour photoperiod. Samples were watered when necessary, approximately once every 2-3 days. Samples were placed in the greenhouse from mid-October to the end of November 2023 after the period of stratification was complete. They remained in the greenhouse for approximately 6 months, or until no new germinants emerged for 14 consecutive days. Samples were checked once a week, and any newly germinated plants were recorded; non-vascular species were not considered. Once strong enough, germinants were transplanted to potting soil to continue growing until they could be identified to genus 42 or species. Soil mixing occurred once a month to bring ungerminated seeds to the top of the soil. The emergence trial concluded in June 2024 (Figure 8). Figure 8: Images of trays (54 total) of soil samples in the Enhanced Forestry Lab at the University of Northern British Columbia, Prince George, undergoing the seedling emergence trial (left). Trays with humidity covers shown (centre). Initial germination of seeds in one sample 8 weeks after the start of the trial before germinants were removed (right). Photos taken by Julia Bizon, 2023. Statistical Analysis The number of germinants for each sample was calculated per 1000 mL of soil collected. This standardized the number of germinants per sample since samples had different soil volumes due to sieving and removal of debris. The life form, seed dispersal methods, and origin status (native or exotic) were detailed for each species that germinated (Table A9). Species composition of the soil seed bank were compared between treated plots, untreated control plots, and reference plots, as well as their interaction with time since treatment by using permutational multivariate analysis of variation (PERMANOVA) with the adonis2 function in vegan package (v2.6.4; Oksanen et al., 2022) using 999 permutations. The 43 analysis was run separately for thinning treatments and gap treatments. To determine which treatments had significantly different (α = 0.05) species composition, post-hoc tests were performed using pairwiseadonis2 function in the package pairwiseAdonis (v0.4.1; Martinez Arbizu, 2017). Nonmetric multidimentional scaling (NMDS) was performed with 2 dimensions and Bray-Curtis distance using the metaMDS function in the package vegan (Oksanen et al., 2022). Thinning treatments and gap treatments were run separately. After 20 iterations the final stress was 0.047 (thinning treatments) and 0.056 (gap treatments). The data collected from the field vegetation surveys conducted from May – August 2023 (refer to Methods and Results in Chapter 2) was used to compare the aboveground vegetation to the corresponding belowground soil seed bank for each plot. The Jaccard coefficient was calculated to determine similarity (based on presence/absence of species) between the aboveground vegetation and corresponding soil seed bank for each plot using function jaccard in package Jaccard (v0.1.0, Chung et al., 2018). Additionally, the difference in species richness between aboveground vegetation and corresponding seed bank was assessed for each plot using a paired t-test from package stats (v4.3.1; R Core Team 2023). All analysis was performed using R (v4.3.1; R Core Team 2023). Results Throughout the duration of the greenhouse trial, a total of 222 seeds successfully germinated in 36 of the 51 trays (Table A10; Figure 9). 98 of these individuals were graminoids, 74 of these individuals were forbs, 18 of these individuals were shrubs, and 6 individuals died before they could be identified (referred to as Unknown throughout this report). Apart from Unknowns, all other individuals that germinated were identified to either genus or species. Of the 92 germinants that were able to be identified to species, the following were identified: 44 Crepis tectorum (1), Epilobium ciliatum (49), Geranium bicknellii (3), Rubus idaeus (8), Stellaria crispa (7), Fragaria vesca (2). 22 germinants did not mature enough to identify further than the genus and were identified as Pilosella sp. (12) and Vaccinium sp. (10). 4 of the 8 emerged species were perennial forbs (E. ciliatum, S. crispa, Pilosella sp., and F. vesca), 2 were annual forbs (G. bicknellii and C. tectorum), and 2 were deciduous shrubs (R. idaeus and Vaccinimium sp.). Of the 8 species that germinated in the greenhouse, 2 were exotic (C. tectorum and Pilosella sp.). A majority of the seeds germinated within the first 8 weeks of the greenhouse trial; graminoids were the first germinants and appeared as quickly as 1 week after the start of the trial. All other species began germinating 4 weeks into the trial, except Vaccinium sp., which began germinating 7 weeks into the trial. No seeds germinated in the sterile potting soil controls, indicating no cross-contamination between the samples occurred. 45 Figure 9: The number of germinants that emerged per 1000 mL of soil in each different gap treatment (0.2, 0.5, 1.0, 2.0 ha), thinning treatment (200, 400, 600 stems/ha), untreated control (ctrl), and reference forest (ref) during the seedling emergence trial. Different colours represent different plant species (n=36). Effect of canopy treatments on species presence and richness The untreated plots only germinated graminoids, the samples from the reference forest only germinated graminoids and Vaccinium sp.; all other species that germinated were found in the treatment plots. E. ciliatum was found in every treatment except 200 stems/ha and 0.5 ha gap treatments. Pilosella sp. germinated in every treatment except 200 stems/ha and 0.2 ha gap treatments. All 8 species, as well as graminoids, germinated in the 0.1 ha gap; however, 6 of these species germinated in the same sample (Table 7). This sample was one of two samples that had a noticeably higher number of germinants than the other plots. One plot from a 1.0 ha gap treatment (3 years post-treatment) had a total of 41 germinants (33 46 individual graminoids), and one plot from a 400 stems/ha treatment (3 years post-treatment) had 52 germinants (49 individual graminoids). Permutational multivariate analysis of variance (PERMANOVA) tests did not find a significant difference in the species composition in treatments versus controls versus reference plots, a significant effect of time since treatment, or an interaction between treatment and time since treatment (Table 8). This is represented visually with nonmetric multidimensional scaling (NMDS) plots which show similarity between species composition of each plot as represented by relative distance of points in space (Figure 10). Points closer in distance in space represent plots with more similar species composition. Table 7: Detailed lists of the treatments (thinning density, canopy gap size, or reference forest) from which soil samples were collected that each species germinated from during the seedling emergence trial. Species germinated Crepis tectorum Epilobium ciliatum Fragaria vesca Geranium bicknellii Pilosella sp. Rubus idaeus Stellaria crispa Vaccinium sp. Treatment Gap treatment: 1.0 ha Thinning treatments: 400, 600 stems/ha Gap treatments: 0.2, 1.0, 2.0 ha Gap treatment: 1.0 ha Gap treatment: 1.0 ha Thinning treatments: 200, 400 stems/ha Gap treatments: 0.5, 1.0, 2.0 ha Thinning treatment: 200 stems/ha Gap treatments: 1.0, 2.0 ha Gap treatment: 1.0 ha Thinning treatment: 600 stems/ha Gap treatment: 1.0 ha Reference forest 47 Table 8: Results of permutational analysis of variation (PERMANOVA) of the effect of thinning treatment, gap treatment, time since treatment, and the interaction on plant species composition (9 species) of the seed bank. Values represent the F-value (p-value) with 999 permutations completed. Significant relationships (α = 0.05) are in bold text. Thinning treatments Gap treatments Treatment 0.93 (0.557) 1.52 (0.061) Time 0.88 (0.738) 0.54 (0.838) Treatment x Time 0.66 (0.885) 0.27 (0.999) Figure 10: Nonmetric multidimensional scaling (NMDS) plots of seed bank plant composition grouped by thinning treatment (200, 400, 600 stems/ha), untreated control (ctrl), and reference stand (ref) (A) and by gap treatment (0.2, 0.5, 1.0, 2.0 ha), untreated control (ctrl), and reference stand (ref) (B). Each point represents one plot. NMDS was run with 2 dimensions and Bray-Curtis distance. After 20 iterations the final stress is 0.047 (A) and 0.056 (B). Similarity in species composition between plots is represented by distance between points in space. Comparing the above-ground vegetation to the soil seed bank There was little similarity between species presence in the greenhouse plots and their corresponding aboveground vegetation when considering the Jaccard similarity coefficients for each plot. A coefficient of 0 indicates zero similarity between species presence in plots, while a coefficient of 1 indicates identical species presence; coefficients calculated were ≤ 0.1 (Table 9). Paired t-tests indicate there was a significant difference in species richness, 48 with a mean difference of 15 fewer species found in the greenhouse plots compared to their corresponding aboveground vegetation (p < 0.001). S. crispa and F. vesca were found in the soil seed bank but were not present in the aboveground vegetation. Fifty-five species were found in the aboveground herb layer that did not germinate from the seed bank. Seed banks from the 1.0 ha gap treatment plots shared E. ciliatum, R. idaeus, Pilosella sp., and Vaccinium sp. with their corresponding aboveground plots. Seed banks from plots from the 600 stems/ha treatment shared Vaccinium sp. with their aboveground plots (Table 9). Interestingly, G. bicknellii and C. tectorum were found in the seed bank of plots that did not have a corresponding presence in the aboveground vegetation; however, these species were found in the aboveground vegetation of nearby plots. Table 9: Jaccard similarity coefficient (calculated value between 0 and 1) for binary data for the presence/absence of aboveground herbaceous species and the species present in the soil seedbank for each corresponding plot organized by treatment (thinning density, gap size, untreated control) and years since treatment. Only plots with at least 1 shared species are listed, all other plots had a Jaccard index of 0 (no shares species). Species shared between the aboveground vegetation and corresponding soil seed bank are listed. Treatment 600 stems/ha 1.0 ha 2.0 ha 1.0 ha 1.0 ha Years since treatment 3 years 3 years 1 year 1 year 1 year Jaccard coefficient 0.0714 0.0417 0.0769 0.0417 0.1000 Shared species Vaccinium sp. Vaccinium sp. R. idaeus E. ciliatum R. idaeus Discussion Species present in the seed bank Most of the germinants that emerged were early-seral ruderals and graminoids, which agrees with other findings from similarly-aged, recently disturbed conifer stands (Halpern et al., 1999). Throughout the seedling emergence trial, 222 individuals successfully germinated. This is much fewer than expected based on similar studies, which have germinated over 1000 49 seedlings (Halpern et al., 1999; Qi & Scarratt, 1998). There are several potential reasons for this. First, it is well-recognized that seedling emergence trials tend to underestimate the number of viable seeds in the soil due to potential constraints by germination requirements; despite attempts to recreate ideal conditions for the maximum number of seeds of species expected to germinate, there were likely many that did not germinate due to unknown or imperfect germination or stratification conditions (Price et al., 2010). Additionally, some seeds may require more rigorous dormancy breaking techniques such as chemical or mechanical scarification (Penfield, 2017). However, it is also possible the site has a weak soil seedbank that may be limiting understory plant regeneration (Halpern et al., 1999), rejecting the hypothesis (H1) that poor aboveground regeneration is caused by unsuitable site conditions. Epilobium ciliatum and Pilosella sp. were the two most abundant species in the emergence trials aside from graminoids, both short-lived species with transient seed banks that do not persist long-term in the soil (French, 2021; Makepeace, 1985; Popay, 2014). Epilobium ciliatum is able to germinate under a wide range of conditions and has a short period between germination and flowering, lending it the ability to produce several generations within one growing season (Myerscough & Whitehead, 1966). This allows many seeds to be produced and may explain why this species was the most abundant in the trials. Since seeds from these species do not remain viable in the soil for long periods of time, it is unlikely these species germinated from dormant seeds in an established seed bank. Seeds that germinated during the greenhouse trial likely came from existing parent plants growing at the site; due to the short stature of these species, parent plants were probably located nearby, as wind dispersal likely occurred over short distances (Thomson et al., 2011). 50 Vaccinium sp., Rubus idaeus, C. tectorum, G. bicknellii, and graminoids were also common germinants during the trial. These species all have persistent seed banks and robust seeds that can survive for years in the soil until ideal germination requirements are met (S. Anderson, 1990; Granstrom, 1987; Haeussler et al., 1990; Hill & Kloet, 2005; Leck & Schütz, 2005; Reeves, 2007). It is possible that all of these species germinated from the established seed bank. Graminoids, in particular, are well-adapted to take advantage of increased resources and germinate rapidly with increased light and moisture (Leck & Schütz, 2005). This might explain why it was the most abundant seed to germinate during the trial. Alternatively, Vaccinium sp. and R. idaeus are deciduous shrubs that persist throughout the winter and recover vegetatively from dormancy via rhizomes or suckers or germinate from seeds in the spring (Haeussler et al., 1990). There is evidence these species may be moderately important for wildlife (Breithaupt et al., 2024; Haeussler et al., 1990) and are well-adapted to remain in the long-term seed bank. Impact of canopy treatments on the seed bank species composition Our results showed no significant difference in the seed bank between the treatments, controls, and reference forest, which agrees with other findings (Qi & Scarratt, 1998). Further, no difference in the seed bank composition was found between 1-year postdisturbance and 3-year post-disturbance. This may indicate that canopy treatments were not a significant enough disturbance to impact the established seed bank. This does not support the hypothesis (H2) that the treatments negatively impacted the soil seed bank. Existing vegetation in an area affects wind patterns and, therefore, can affect wind-dispersed seeds. While plants in heavily forested areas rely less on dispersal over long distances than in open environments, there is evidence that, in particular, heterogeneity in canopy structure, such as 51 introduced by canopy treatments, can heavily influence wind patterns (Damschen et al., 2014). Additionally, canopy treatments act as a major disturbance, dramatically changing the resource levels and physically perturbing the soil on the forest floor. Despite this, the soil seed bank was unchanged post-treatment. Pine stands located in the SBS dw3 subzone have a history of regular wildfire disturbances of varying sizes with a fire return interval of approximately 125 years (BC Environment & BC Ministry of Forests, 1995). Therefore, it would be expected that the ecosystem will be well-adapted to recover after disturbance (Johnstone et al., 2016). There was some evidence of a long-term seed bank in the greenhouse trials, with fire-adapted species such as G. bicknelli present in the seed bank (Reeves, 2007). However, a more robust long-term seed bank was expected due to the stand's history. It is possible that fire-adapted species required more robust dormancy-breaking mechanisms to germinate. It is also possible that this site is experiencing “resilience debt,” a term used by Johnstone et al. (2016) to capture a loss of ecosystem resilience over time due to changing disturbance regimes. This occurs when historical adaptations of an ecosystem become misaligned with the current disturbance cycle. This can include extensive harvesting, intensive silviculture, insect outbreaks, and increases in the frequency and intensity of wildfires seen in this area over the last several decades (Parisien et al., 2023; Taylor et al., 2006). If frequent disturbances occur before an ecosystem can recover, some species with long recovery periods may not have an opportunity to recover and can be lost from the seedbank (McRae et al., 2001). This could result in poor regeneration of the understory vegetation layer. 52 Almost all the germinants occurred in the treated plots, with only graminoids and Vaccinium sp. germinating in the untreated and reference forest plots. Treated plots had a significantly higher aboveground vegetation cover and species richness on-site compared to the untreated controls, which were dominated by early-seral, disturbance-dependent forbs (see Results in Chapter 2). This difference is due to an abundance of resources in the treated plots because of the less dense canopy, which likely resulted in profuse seed rain from parent plants. In contrast, sparse aboveground vegetation in control plots resulted in limited seed banks. This is further supported by the high number of germinants in one of the 1.0 ha treatment samples, whose corresponding aboveground plot on-site had the highest average percent cover and one of the highest aboveground species richness values. The potential for these early-seral species to continue to dominate the understory layer through rapid production of seeds and aggressive colonization methods is strong (Decocq et al., 2004; Halpern et al., 1999), though further sampling 5 years and 10 years post-treatment would provide more substantial evidence (Stark et al., 2006). Comparison between above-ground vegetation and the seed bank The Jaccard similarity coefficients and results of the paired t-test showed a significant difference in the species presence and richness of the aboveground vegetation on-site and the corresponding soil seed bank at each plot. This agrees with other findings that have found that the seed bank is a poor indicator for aboveground vegetation composition and that a high abundance of species in aboveground vegetation does not necessarily correspond with a high representation of these species in the seed bank (Decocq et al., 2004; Diemer & Prock, 1993; Zobel et al., 2007). This could be because the seed bank is dependent on proximity to parent 53 plants and environmental impacts such as wind speed and direction and soil properties rather than aboveground cover (Chambers, 1995). 2 of the 8 (25%) species that emerged during the greenhouse trial were not seen in the aboveground vegetation layer: F. vesca and S. crispa. Both of these plants are perennial herbs and have a transient seed bank; it is possible that the seeds remained dormant from the previous season and were unable to germinate in the field site due to competition or inadequate resources. It is also possible that seeds in situ germinated but were predated on, or died for other reasons, in the period of time between germination and when aboveground sampling occurred. Another explanation would be seed dispersal via wind from nearby stands that were not included in the sampled plots. These species could also have been transported via human or equipment, as seeds are evidenced to get stuck in the treads of boots or wheels (Decocq et al., 2004). The plots that germinated G. bicknellii and C. tectorum from the seed bank did not have these species present in the aboveground vegetation; however, these species were found in the aboveground vegetation in nearby plots. These species, too, could have been transported via wind or animal or could have germinated from banked seeds. The significance of the seed bank lies in what it may communicate about the future of a stand. The results show an overwhelming representation of perennial herbs and ruderals in the seed bank. Early successional forests have the highest seed abundance and diversity due to the high abundance of understory plants contributing to seed rain (Halpern et al., 1999). However, as communities mature and the understory layer becomes less abundant and diverse, seed abundance decreases with predation, viability, and lower seed production (Leck et al., 1989). There is concern that without a well-established seed bank, the understory will 54 only become less robust with time, negatively impacting both forest productivity and wildlife. Limitations and Opportunities for Further Research The soil at our site was dry and loose, and the structure of the soil core was not maintained during collection. Therefore, soil cores were not divided into different layers, impacting the data that could be collected. Many seedling emergence trials divided their cores into distinct layers to incubate for germination. This provides more information about the origin of the seeds; seeds from the top layer and the leaf litter are from seed rain, while those buried deep in the soil have been banked for longer periods (Halpern et al., 1999; Qi & Scarratt, 1998; Zobel et al., 2007). Since these layers were not divided during the trials, we were only able to make educated guesses about the origin of the seeds that germinated. Conclusion The species that germinated during the seedling emergence trial are dominantly ruderals or disturbance-dependent species (Crepis tectorum, Epilobium ciliatum, Fragaria vesca, Geranium bicknellii, Rubus idaeus, Pilosella sp., Stellaria crispa, and Vaccinium sp.). There was evidence of a weak soil seed bank for the study site, with a majority of the germinants composed of perennial herbs or graminoids. Notably, there was a lack of species of importance for moose forage found in the soil seedbank. This is likely due to the fact that species such as aspen and willow have limited seed longevity and viability and likely rely on vegetative regeneration. There was no significant difference in the species composition of the seed bank between the treated plots, untreated plots, and the reference forest; this indicates that the treatments did not impact the soil seed bank of the site. This suggests that possibly the seed 55 bank was weak pre-treatment and that the ecosystem memory of the site may have been negatively impacted by shifting disturbance regimes due to changes in land use and climate. This leads to concerns about the future of a stand that may have to rely on a weak soil seed bank for regeneration. 56 Chapter 4: Conclusion and Recommendations The restoration treatments increased understory herb cover and richness when compared to the untreated stands. However, the initial understory regeneration in the study area was found to be limited mostly to disturbance-dependent forbs and ruderals. There is evidence of presence of willow and aspen (important for moose forage) in some areas which have likely regenerated via vegetative regeneration from existing propagules. This result is mostly expected in the period of initial regeneration after disturbance. The seedling emergence trials conducted in the greenhouse can provide information about species present in the soil seed bank which may not be represented in the aboveground vegetation. However, results of these trials also showed a predominance of disturbance-dependent plant species. There is concern that plots with weak initial regeneration of target species and no evidence of regeneration from the seed bank will not be able to adequately support local wildlife populations in the future. Salix spp. and Populus tremuloides were only found in some of the treated plots, and regeneration of these species likely depended on a strong presence prior to treatment. In treated plots where these species are not already established, there is concern that regrowth will occur within the next several years due to poor evidence of seed recruitment from the soil seed bank or nearby stands. This is likely because Salix spp. and P. tremuloides have seeds that are only viable for a few weeks and rely primarily on methods other than seed banking to establish populations (Haeussler et al., 1990). Salix spp. depend on axillary buds to regenerate post-winter, while P. tremuloides regenerate vegetatively via suckers (Saska & Kuzovkina, 2010; Schier et al., 1985). Viburnum edule is another important species for moose browse (Franzmann & Schwartz, 2007; Haeussler et al., 1990; Hodder et al., 2013; 57 Poole & Stuart-Smith, 2005) observed at the field site but not in the seed bank. This is a seed-banking species with seeds that can remain viable for up to a decade. However, germination of this species is difficult due to the presence of a seed coat and embryo dormancy (Haeussler et al., 1990). This species, too, can regenerate from existing rootstocks and stem bases. To maintain the presence of these species (P. tremuloides, Salix spp., V. edule) in similar stands post-treatment, rather than relying on the seed bank for regeneration, it is recommended that damage to existing plants during treatments is minimized. For those plots without existing evidence of P. tremuloides, Salix spp., or V. edule, seeds will have to be naturally transported from nearby via wind or animal. Species that are important for moose forage but were not observed either in the aboveground vegetation or the seed bank, such as Cornus stolonifera, Betula spp., and Amelanchier alnifolia, have no current evidence based on this study of future germination from an existing persistent seed bank or seed recruitment from sites nearby in this area. Therefore, we suggest that active restoration methods, such as seeding and planting species of high importance to wildlife, complement the current passive restoration methods in this area to best support wildlife in the present and future. This can include bolstering the current population of Salix spp. and P. tremuloides by planting cuttings, or by establishing native species that are not currently growing, such as C. stolonifera, Betula spp., and A. alnifolia. 58 References Ampoorter, E., Barbaro, L., Jactel, H., Baeten, L., Boberg, J., Carnol, M., Castagneyrol, B., Charbonnier, Y., Dawud, S. M., Deconchat, M., Smedt, P. D., Wandeler, H. D., Guyot, V., Hättenschwiler, S., Joly, F.-X., Koricheva, J., Milligan, H., Muys, B., Nguyen, D., … Allan, E. (2020). Tree diversity is key for promoting the diversity and abundance of forest-associated taxa in Europe. Oikos, 129(2), 133–146. https://doi.org/10.1111/oik.06290 Anderson, E. W. (1986). A Guide for Estimating Cover. Rangelands, 8(5), 236–238. Anderson, S. (1990). The relationship between seed dormancy, seed size and weediness, in Crepis tectorum (Asteraceae). Oecologia, 83(2), 277–280. https://doi.org/10.1007/BF00317765 Ares, A., Neill, A. R., & Puettmann, K. J. (2010). Understory abundance, species diversity and functional attribute response to thinning in coniferous stands. Forest Ecology and Management, 260(7), 1104–1113. https://doi.org/10.1016/j.foreco.2010.06.023 Aubin, I., Messier, C., & Bouchard, A. (2008). Can plantations develop understory biological and physical attributes of naturally regenerated forests? Biological Conservation, 141(10), 2461–2476. https://doi.org/10.1016/j.biocon.2008.07.007 Augusto, L., Dupouey, J.-L., & Ranger, J. (2003). Effects of tree species on understory vegetation and environmental conditions in temperate forests. Annals of Forest Science, 60(8), 823–831. https://doi.org/10.1051/forest:2003077 Augusto, L., Ranger, J., Binkley, D., & Rothe, A. (2002). Impact of several common tree species of European temperate forests on soil fertility. Annals of Forest Science, 59(3), 233–253. https://doi.org/10.1051/forest:2002020 Bailey, J. D., Mayrsohn, C., Doescher, P. S., St. Pierre, E., & Tappeiner, J. C. (1998). Understory vegetation in old and young Douglas-fir forests of western Oregon1. Forest Ecology and Management, 112(3), 289–302. https://doi.org/10.1016/S03781127(98)00408-3 Barbier, S., Gosselin, F., & Balandier, P. (2008). Influence of tree species on understory vegetation diversity and mechanisms involved—A critical review for temperate and boreal forests. Forest Ecology and Management, 254(1), 1–15. https://doi.org/10.1016/j.foreco.2007.09.038 Bartels, S. F., & Chen, H. Y. H. (2010). Is understory plant species diversity driven by resource quantity or resource heterogeneity? Ecology, 91(7), 1931–1938. https://doi.org/10.1890/09-1376.1 59 Bartels, S. F., Chen, H. Y. H., Wulder, M. A., & White, J. C. (2016). Trends in postdisturbance recovery rates of Canada’s forests following wildfire and harvest. Forest Ecology and Management, 361, 194–207. https://doi.org/10.1016/j.foreco.2015.11.015 Bartels, S. F., & Macdonald, S. E. (2023). Dynamics and recovery of forest understory biodiversity over 17 years following varying levels of retention harvesting. Journal of Applied Ecology, 60(4), 725–736. https://doi.org/10.1111/1365-2664.14366 Bates, D., Machler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67(1), 1–48. https://doi.org/10.18637/jss.v067.i01 BC Environment, & BC Ministry of Forests. (1995). Forest Practices Code of British Columbia: Biodiversity Guidebook. BC Environment. British Columbia. Ministry of Forests. Bekker, R. M., Verweij, G. L., Smith, R. E. N., Reine, R., Bakker, J. P., & Schneider, S. (1997). Soil seed banks in European grasslands: Does land use affect regeneration perspectives? Journal of Applied Ecology, 34(5), 1293–1310. https://doi.org/10.2307/2405239 Berg, B., Davey, M. P., De Marco, A., Emmett, B., Faituri, M., Hobbie, S. E., Johansson, M.-B., Liu, C., McClaugherty, C., Norell, L., Rutigliano, F. A., Vesterdal, L., & Virzo De Santo, A. (2010). Factors influencing limit values for pine needle litter decomposition: A synthesis for boreal and temperate pine forest systems. Biogeochemistry, 100(1), 57–73. https://doi.org/10.1007/s10533-009-9404-y Berger, J. J. (2008). Forests Forever: Their Ecology, Restoration, and Protection. The Centre for American Places at Columbia College Chicago. Bergeron, J. A. C., Pinzon, J., Odsen, S., Bartels, S., Macdonald, S. E., & Spence, J. R. (2017). Ecosystem memory of wildfires affects resilience of boreal mixedwood biodiversity after retention harvest. Oikos, 126(12), 1738–1747. https://doi.org/10.1111/oik.04208 Bergqvist, G., Wallgren, M., Jernelid, H., & Bergström, R. (2018). Forage availability and moose winter browsing in forest landscapes. Forest Ecology and Management, 419– 420, 170–178. https://doi.org/10.1016/j.foreco.2018.03.049 Breithaupt, K., Rea, R. V., Gillingham, M. P., Aitken, D. A., & Hodder, D. P. (2024). Using winter diet composition and forage plant availability to determine browse selection and importance for moose (Alces alces) in a landscape modified by industrial forestry. Forestry: An International Journal of Forest Research, cpae019. https://doi.org/10.1093/forestry/cpae019 60 British Columbia Ministry of Forests, Lands and Natural Resources Operations. (2002). Annual reports. British Columbia Ministry of Forests, Lands and Natural Resources Operations, Victoria, BC. Brosofske, K. D., Chen, J., & Crow, T. R. (2001). Understory vegetation and site factors: Implications for a managed Wisconsin landscape. Forest Ecology and Management, 146(1), 75–87. https://doi.org/10.1016/S0378-1127(00)00447-3 Cavard, X., Macdonald, S. E., Bergeron, Y., & Chen, H. Y. H. (2011). Importance of mixedwoods for biodiversity conservation: Evidence for understory plants, songbirds, soil fauna, and ectomycorrhizae in northern forests. Environmental Reviews, 19, 142– 161. https://doi.org/10.1139/a11-004 Cederlund, G. N., & Okarma, H. (1988). Home range and habitat use of adult female moose. The Journal of Wildlife Management, 52(2), 336–343. https://doi.org/10.2307/3801246 Chambers, J. C. (1995). Disturbance, life history strategies, and seed fates in apine herbfield communities. American Journal of Botany, 82(3), 421–433. https://doi.org/10.2307/2445588 Chung, N. C., Miasojedow, B., Startek, M., & Gambin, A. (2018). Test Similarity Between Binary Data using Jaccard/Tanimoto Coefficients (Version 0.1.0) [Computer software]. Crowley, S. M., Johnson, C. J., & Hodder, D. P. (2025). Behaviour as an indicator of cyclic trends in abundance of Canada lynx (Lynx canadensis) and snowshoe hare (Lepus americanus). Ecological Indicators, 170, 113045. https://doi.org/10.1016/j.ecolind.2024.113045 Damschen, E. I., Baker, D. V., Bohrer, G., Nathan, R., Orrock, J. L., Turner, J. R., Brudvig, L. A., Haddad, N. M., Levey, D. J., & Tewksbury, J. J. (2014). How fragmentation and corridors affect wind dynamics and seed dispersal in open habitats. Proceedings of the National Academy of Sciences, 111(9), 3484–3489. https://doi.org/10.1073/pnas.1308968111 Davis, L. R., & Puettmann, K. J. (2009). Initial response of understory vegetation to three alternative thinning treatments. Journal of Sustainable Forestry, 28(8), 904–934. https://doi.org/10.1080/10549810903344611 De Cáceres, M., & Legendre, P. (2009). Associations between species and groups of sites: Indices and statistical inference. Ecology, 90, 3566–3574. https://doi.org/10.1890/081823.1 61 De Grandpré, L., Boucher, D., Bergeron, Y., & Gagnon, D. (2011). Effects of small canopy gaps on boreal mixedwood understory vegetation dynamics. Community Ecology, 12(1), 67–77. https://doi.org/10.1556/ComEc.12.2011.1.9 De Grandpré, L., Gagnon, D., & Bergeron, Y. (1993). Changes in the understory of Canadian southern boreal forest after fire. Journal of Vegetation Science, 4(6), 803–810. https://doi.org/10.2307/3235618 Decocq, G., Valentin, B., Toussaint, B., Hendoux, F., Saguez, R., & Bardat, J. (2004). Soil seed bank composition and diversity in a managed temperate deciduous forest. Biodiversity & Conservation, 13(13), 2485–2509. https://doi.org/10.1023/B:BIOC.0000048454.08438.c6 DeJong, T. M. (1975). A comparison of three diversity indices based on their components of richness and evenness. Oikos, 26(2), 222–227. https://doi.org/10.2307/3543712 DeLong, C., Tanner, D., & Jull, M. (1993). A Field Guide for Site Identification and Interpretation for the Southwest Portion of the Prince George Forest Region (Vol. 24). Ministry of Forests, Research Program. Demir, M., Makineci, E., & Yilmaz, E. (2007). Harvesting impact on herbaceous understory, forest floor and top soil properties on skid road in a beech (Fagus orientalis Lipsky) stand. Journal of Environmental Biology, 28(2), 427–432. Diemer, M., & Prock, S. (1993). Estimates of alpine seed bank size in two central European and one Scandinavian subarctic plant communities. Arctic and Alpine Research, 25(3), 194–200. https://doi.org/10.1080/00040851.1993.12003004 Dupuy, J. M., & Chazdon, R. L. (2008). Interacting effects of canopy gap, understory vegetation and leaf litter on tree seedling recruitment and composition in tropical secondary forests. Forest Ecology and Management, 255(11), 3716–3725. https://doi.org/10.1016/j.foreco.2008.03.021 Dussault, C., Ouellet, J.-P., Courtois, R., Huot, J., Breton, L., & Jolicoeur, H. (2005). Linking moose habitat selection to limiting factors. Ecography, 28(5), 619–628. https://doi.org/10.1111/j.2005.0906-7590.04263.x Dussault, C., Ouellet, J.-P., Courtois, R., Huot, J., Breton, L., & Larochelle, J. (2004). Behavioural responses of moose to thermal conditions in the boreal forest1. Écoscience, 11(3), 321–328. https://doi.org/10.1080/11956860.2004.11682839 Fahey, R. T., & Puettmann, K. J. (2008). Patterns in spatial extent of gap influence on understory plant communities. Forest Ecology and Management, 255(7), 2801–2810. https://doi.org/10.1016/j.foreco.2008.01.053 62 Faison, E. K., DeStefano, S., Foster, D. R., Motzkin, G., & Rapp, J. M. (2016). Ungulate browsers promote herbaceous layer diversity in logged temperate forests. Ecology and Evolution, 6(13), 4591–4602. https://doi.org/10.1002/ece3.2223 FAO. (2020). Global Forest Resources Assessment 2020—Key findings. FAO. https://doi.org/10.4060/ca8753en Felton, A., Nilsson, U., Sonesson, J., Felton, A. M., Roberge, J.-M., Ranius, T., Ahlström, M., Bergh, J., Björkman, C., Boberg, J., Drössler, L., Fahlvik, N., Gong, P., Holmström, E., Keskitalo, E. C. H., Klapwijk, M. J., Laudon, H., Lundmark, T., Niklasson, M., … Wallertz, K. (2016). Replacing monocultures with mixed-species stands: Ecosystem service implications of two production forest alternatives in Sweden. Ambio, 45(Suppl 2), 124. https://doi.org/10.1007/s13280-015-0749-2 Fox, J., & Weisberg, S. (2019). An R Companion to Applied Regression (Vol. 3). Sage. https://socialsciences.mcmaster.ca/jfox/Books/Companion Franzmann, A. W., & Schwartz, C. C. (2007). Ecology and management of the North American moose (2nd ed). University Press of Colorado. French, K. (2021). Invasion by hawkweeds. Biological Invasions, 23(12), 3641–3652. https://doi.org/10.1007/s10530-021-02604-6 George, L. O., & Bazzaz, F. A. (1999). The fern understory as an ecological filter: Emergence and Establishment of canopy-tree seedlings. Ecology, 80(3), 833–845. https://doi.org/10.1890/0012-9658(1999)080[0833:TFUAAE]2.0.CO;2 Gilliam, F. S. (2007). The ecological significance of the herbaceous layer in temperate forest ecosystems. BioScience, 57(10), 845–858. https://doi.org/10.1641/B571007 Godefroid, S., & Koedam, N. (2004). Interspecific variation in soil compaction sensitivity among forest floor species. Biological Conservation, 119(2), 207–217. https://doi.org/10.1016/j.biocon.2003.11.009 Granstrom, A. (1987). Seed viability of fourteen species during five years of storage in a forest soil. Journal of Ecology, 75(2), 321–331. https://doi.org/10.2307/2260421 Haeussler, S., Coates, D., & Mather, J. (1990). Autoecology of Common Plants in British Columbia: A Literature Review (FRDA Report 158; p. 272). Forestry Canada, British Columbia Ministry of Forests. Hagar, J. C. (2007). Wildlife species associated with non-coniferous vegetation in Pacific Northwest conifer forests: A review. Forest Ecology and Management, 246(1), 108– 122. https://doi.org/10.1016/j.foreco.2007.03.054 63 Halfpenny, J. C. (2001). Scats and Tracks of the Rocky Mountains: A Field Guide to the Signs of Seventy Wildlife Species (3rd ed.). The Globe Pequot Press. Halpern, C. B., Evans, S. A., & Nielson, S. (1999). Soil seed banks in young, closed-canopy forests of the Olympic Peninsula, Washington: Potential contributions to understory reinitiation. Canadian Journal of Botany, 77(7), 922–935. https://doi.org/10.1139/b99-085 Härkönen, S., & Heikkilä, R. (1999). Use of pellet group counts in determining density and habitat use of moose Alces alces in Finland. Wildlife Biology, 5(4), 233–239. https://doi.org/10.2981/wlb.1999.028 Hart, S. A., & Chen, H. Y. H. (2006). Understory vegetation dynamics of North American Boreal forests. Critical Reviews in Plant Sciences, 25(4), 381–397. https://doi.org/10.1080/07352680600819286 Hart, S. A., & Chen, H. Y. H. (2008). Fire, logging, and overstory affect understory abundance, diversity, and composition in boreal forest. Ecological Monographs, 78(1), 123–140. https://doi.org/10.1890/06-2140.1 Hartley, M. J. (2002). Rationale and methods for conserving biodiversity in plantation forests. Forest Ecology and Management, 155(1), 81–95. https://doi.org/10.1016/S0378-1127(01)00549-7 Haughian, S. R., & Frego, K. A. (2016). Short-term effects of three commercial thinning treatments on diversity of understory vascular plants in white spruce plantations of northern New Brunswick. Forest Ecology and Management, 370, 45–55. https://doi.org/10.1016/j.foreco.2016.03.055 Herrmann, S., & Prescott, C. E. (2008). Mass loss and nutrient dynamics of coarse woody debris in three Rocky Mountain coniferous forests: 21 year results. Canadian Journal of Forest Research, 38(1), 125–132. https://doi.org/10.1139/X07-144 Hill, N. M., & Kloet, S. P. V. (2005). Longevity of experimentally buried seed in Vaccinium: Relationship to climate, reproductive factors and natural seed banks. Journal of Ecology, 93(6), 1167–1176. https://doi.org/10.1111/j.1365-2745.2005.01034.x Hobbs, N. T. (1996). Modification of ecosystems by ungulates. The Journal of Wildlife Management, 60(4), 695–713. https://doi.org/10.2307/3802368 Hobson, K. A., & Bayne, E. (2000). Breeding bird communities in boreal forest of Western Canada: Consequences of “unmixing” the mixedwoods. The Condor, 102(4), 759– 769. https://doi.org/10.2307/1370303 64 Hodder, D. P., Rea, R. V., & Crowley, S. M. (2013). Diet content and overlap of sympatric mule deer (Odocoileus hemionus), moose (Alces alces), and elk (Cervus elaphus) during a deep snow winter in North- central British Columbia, Canada. Canadian Wildlife Biology & Management, 2(2), 43–50. Holl, K. D., & Brancalion, P. H. S. (2020). Tree planting is not a simple solution. Science. https://doi.org/10.1126/science.aba8232 Jactel, H., Bauhus, J., Boberg, J., Bonal, D., Castagneyrol, B., Gardiner, B., GonzalezOlabarria, J. R., Koricheva, J., Meurisse, N., & Brockerhoff, E. G. (2017). Tree diversity drives forest stand resistance to natural disturbances. Current Forestry Reports, 3(3), 223–243. https://doi.org/10.1007/s40725-017-0064-1 Jactel, H., Moreira, X., & Castagneyrol, B. (2021). Tree diversity and forest resistance to insect pests: Patterns, mechanisms, and prospects. Annual Review of Entomology, 66(1), 277–296. https://doi.org/10.1146/annurev-ento-041720-075234 Johnson, C. J., & Rea, R. V. (2024). Response of moose to forest harvest and management: A literature review. Canadian Journal of Forest Research, 54(4), 366–388. https://doi.org/10.1139/cjfr-2023-0158 Johnstone, J. F., Allen, C. D., Franklin, J. F., Frelich, L. E., Harvey, B. J., Higuera, P. E., Mack, M. C., Meentemeyer, R. K., Metz, M. R., Perry, G. L., Schoennagel, T., & Turner, M. G. (2016). Changing disturbance regimes, ecological memory, and forest resilience. Frontiers in Ecology and the Environment, 14(7), 369–378. https://doi.org/10.1002/fee.1311 Keeley, J. E., Pausas, J. G., Rundel, P. W., Bond, W. J., & Bradstock, R. A. (2011). Fire as an evolutionary pressure shaping plant traits. Trends in Plant Science, 16(8), 406– 411. https://doi.org/10.1016/j.tplants.2011.04.002 Kelty, M. J. (2006). The role of species mixtures in plantation forestry. Forest Ecology and Management, 233(2), 195–204. https://doi.org/10.1016/j.foreco.2006.05.011 Kenney, A. J., Boutin, S., Jung, T. S., Murray, D. L., Johnson, N., & Krebs, C. J. (2024). Motion-sensitive cameras track population abundance changes in a boreal mammal community in southwestern Yukon, Canada. The Journal of Wildlife Management, 88(4), e22564. https://doi.org/10.1002/jwmg.22564 Kiss, R., Valkó, O., Tóthmérész, B., & Török, P. (2016). Seed bank research in CentralEuropean grasslands—An overview (J. Murphy, Ed.; pp. 1–34). Nova Science Publisher. https://real.mtak.hu/99380/ Kunkel, K. E., & Pletscher, D. H. (2000). Habitat factors affecting vulnerability of moose to predation by wolves in southeastern British Columbia. Canadian Journal of Zoology, 78(1), 150–157. https://doi.org/10.1139/z99-181 65 Kuzyk, G., Procter, C., Marshall, S., Schindler, H., Schwantje, H., Scheideman, M., & Hodder, D. (2019). Factors Affecting Moose Population Declines in British Columbia 2019 Progress Report: February 2012–May 2019 [WR-127]. Ministry of Forests, Lands, Natural Resource Operations and Rural Development. Leck, M. A., Parker, V. T., & Simpson, R. L. (Eds.). (1989). Ecology of Soil Seed Banks. Academic Press, Inc. Leck, M. A., & Schütz, W. (2005). Regeneration of Cyperaceae, with particular reference to seed ecology and seed banks. Perspectives in Plant Ecology, Evolution and Systematics, 7(2), 95–133. https://doi.org/10.1016/j.ppees.2005.05.001 Lenth, R. (2024). emmeans: Estimated Marginal Means, aka Least-Squares Means (Version 1.10.0) [Computer software]. Lewis, D. (2009). Stand and landscape-level simulations of mountain pine beetle (Dendroctonus ponderosae) and salvage logging effects on live tree and deadwood habitats in south-central British Columbia, Canada. Forest Ecology and Management, 258, S24–S35. https://doi.org/10.1016/j.foreco.2009.08.031 Lochhead, K. D., & Comeau, P. G. (2012). Relationships between forest structure, understorey light and regeneration in complex Douglas-fir dominated stands in southeastern British Columbia. Forest Ecology and Management, 284, 12–22. https://doi.org/10.1016/j.foreco.2012.07.029 Mackenzie, D. D., & Naeth, M. A. (2010). The role of the forest soil propagule bank in assisted natural recovery after oil sands mining. Restoration Ecology, 18(4), 418–427. https://doi.org/10.1111/j.1526-100X.2008.00500.x Makepeace, W. (1985). Growth, reproduction, and production biology of mouse-ear and king devil hawkweed in eastern South Island, New Zealand. New Zealand Journal of Botany, 23(1), 65–78. https://doi.org/10.1080/0028825X.1985.10425309 Månsson, J., Andrén, H., & Sand, H. (2011). Can pellet counts be used to accurately describe winter habitat selection by moose Alces alces? European Journal of Wildlife Research, 57(5), 1017–1023. https://doi.org/10.1007/s10344-011-0512-3 Martinez Arbizu, P. (2017). pairwiseAdonis: Pairwise Multilevel Comparison using Adonis (Version 0.4.1) [Computer software]. McDowell, N., Pockman, W. T., Allen, C. D., Breshears, D. D., Cobb, N., Kolb, T., Plaut, J., Sperry, J., West, A., Williams, D. G., & Yepez, E. A. (2008). Mechanisms of plant survival and mortality during drought: Why do some plants survive while others succumb to drought? New Phytologist, 178(4), 719–739. https://doi.org/10.1111/j.1469-8137.2008.02436.x 66 McGee, A., & Feller, M. C. (1993). Seed banks of forested and disturbed soils in southwestern British Columbia. Canadian Journal of Botany, 71(12), 1574–1583. https://doi.org/10.1139/b93-192 McLaren, B. E., Mahoney, S. P., Porter, T. S., & Oosenbrug, S. M. (2000). Spatial and temporal patterns of use by moose of pre-commercially thinned, naturallyregenerating stands of balsam fir in central Newfoundland. Forest Ecology and Management, 133(3), 179–196. https://doi.org/10.1016/S0378-1127(99)00223-6 McRae, D. J., Duchesne, L. C., Freedman, B., Lynham, T. J., & Woodley, S. (2001). Comparisons between wildfire and forest harvesting and their implications in forest management. Environmental Reviews, 9(4), 223–260. https://doi.org/10.1139/a01010 Meier, A. J., Bratton, S. P., & Duffy, D. C. (1995). Possible ecological mechanisms for loss of vernal-herb diversity in logged Eastern deciduous forests. Ecological Applications, 5(4), 935–946. https://doi.org/10.2307/2269344 Meli, P., Holl, K. D., Benayas, J. M. R., Jones, H. P., Jones, P. C., Montoya, D., & Mateos, D. M. (2017). A global review of past land use, climate, and active vs. Passive restoration effects on forest recovery. PLOS ONE, 12(2), e0171368. https://doi.org/10.1371/journal.pone.0171368 Messier, C., Bauhus, J., Sousa-Silva, R., Auge, H., Baeten, L., Barsoum, N., Bruelheide, H., Caldwell, B., Cavender-Bares, J., Dhiedt, E., Eisenhauer, N., Ganade, G., Gravel, D., Guillemot, J., Hall, J. S., Hector, A., Hérault, B., Jactel, H., Koricheva, J., … Zemp, D. C. (2022). For the sake of resilience and multifunctionality, let’s diversify planted forests! Conservation Letters, 15(1), e12829. https://doi.org/10.1111/conl.12829 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(4), 511–520. https://doi.org/10.2307/3237266 Milligan, H. T., & Koricheva, J. (2013). Effects of tree species richness and composition on moose winter browsing damage and foraging selectivity: An experimental study. Journal of Animal Ecology, 82(4), 739–748. https://doi.org/10.1111/13652656.12049 Moll, R. J., Poisson, M. K. P., Heit, D. R., Jones, H., Pekins, P. J., & Kantar, L. (2022). A review of methods to estimate and monitor moose density and abundance. Alces, 58, 31–49. Muller, R. (2003). The Herbaceous Layer in Forests of Eastern North America (F. S. Gilliam & M. R. Roberts, Eds.). Oxford University Press, USA. 67 Mumma, M., & Gilligham, M. (2019). Determining factors that affect survival of moose in Central British Columbia. Habitat Conservation Trust Foundation. Munger, G. T. (2006). Fragaria vesca. In: Fire Effects Information System. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). Murray, D. L., Cox, E. W., Ballard, W. B., Whitlaw, H. A., Lenarz, M. S., Custer, T. W., Barnett, T., & Fuller, T. K. (2006). Pathogens, nutritional deficiency, and climate influences on a declining moose population. Wildlife Monographs, 166, 1–30. https://doi.org/10.2193/0084-0173(2006)166[1:PNDACI]2.0.CO;2 Myerscough, P. J., & Whitehead, F. H. (1966). Comparative biology of Tussilago Farfara L., Chamafnerion Angustifolium (l.) Scop., Epilobium Montanum L. and Fpilobium Adfnocaulon Hausskn. New Phytologist, 65(2), 192–210. https://doi.org/10.1111/j.1469-8137.1966.tb06352.x Najda, H. G., Darwent, A. L., & Hamilton, G. (1982). The Biology of Canadian Weeds. 54. Crepis tectorum L. Canadian Journal of Plant Science, 62(2), 473–481. https://doi.org/10.4141/cjps82-067 Natural Resources Canada. (2023). The State of Canada’s Forests ANNUAL REPORT 2023. Nilsson, M.-C., & Wardle, D. A. (2005). Understory vegetation as a forest ecosystem driver: Evidence from the northern Swedish boreal forest. Frontiers in Ecology and the Environment, 3(8), 421–428. https://doi.org/10.1890/15409295(2005)003[0421:UVAAFE]2.0.CO;2 Oksanen, J., Simpson, G., Blanchet, F. G., Kindt, R., Legendre, P., Minchin, P., O’Hara, R. . B., Solymos, P., Stevens, M. H. H., Szoecs, E., Wagner, H., & Barbour, M. (2022). vegan: Community Ecology Package (Version 2.6.4) [Computer software]. https://CRAN.R-project.org/package=vegan Page, H. N., Bork, E., & Newman, R. (2005). Understorey responses to mechanical restoration and drought within montane forests of British Columbia. Journal of Ecosystems and Management, 6(1), 8-21. https://doi.org/10.22230/jem.2005v6n1a304 Paquette, A., & Messier, C. (2010). The role of plantations in managing the world’s forests in the Anthropocene. Frontiers in Ecology and the Environment, 8(1), 27–34. Parish, R., Coupe, R., & Lloyd, D. (Eds.). (1996). Plants of Southern Interior British Columbia and the Inland Northwest. Lone Pine Publishing. 68 Parisien, M.-A., Barber, Q. E., Bourbonnais, M. L., Daniels, L. D., Flannigan, M. D., Gray, R. W., Hoffman, K. M., Jain, P., Stephens, S. L., Taylor, S. W., & Whitman, E. (2023). Abrupt, climate-induced increase in wildfires in British Columbia since the mid-2000s. Communications Earth & Environment, 4(1), 1–11. https://doi.org/10.1038/s43247-023-00977-1 Pastor, J., Naiman, R. J., Dewey, B., & McInnes, P. (1988). Moose, microbes, and the boreal forest. BioScience, 38(11), 770–777. https://doi.org/10.2307/1310786 Penfield, S. (2017). Seed dormancy and germination. Current Biology, 27(17), R874–R878. https://doi.org/10.1016/j.cub.2017.05.050 Perring, M. P., Standish, R. J., Price, J. N., Craig, M. D., Erickson, T. E., Ruthrof, K. X., Whiteley, A. S., Valentine, L. E., & Hobbs, R. J. (2015). Advances in restoration ecology: Rising to the challenges of the coming decades. Ecosphere, 6(8), art131. https://doi.org/10.1890/ES15-00121.1 Plue, J., De Frenne, P., Acharya, K., Brunet, J., Chabrerie, O., Decocq, G., Diekmann, M., Graae, B. J., Heinken, T., Hermy, M., Kolb, A., Lemke, I., Liira, J., Naaf, T., Verheyen, K., Wulf, M., & Cousins, S. A. O. (2017). Where does the community start, and where does it end? Including the seed bank to reassess forest herb layer responses to the environment. Journal of Vegetation Science, 28(2), 424–435. https://doi.org/10.1111/jvs.12493 Poole, K. G., & Stuart-Smith, K. (2005). Fine-scale winter habitat selection by moose in interior montane forests. Alces, 41, 1–8. Popay, I. (2014). Epilobium ciliatum (northern willowherb). CABI Compendium, CABI Compendium, 114114. https://doi.org/10.1079/cabicompendium.114114 Pretzsch, H., & Schütze, G. (2016). Effect of tree species mixing on the size structure, density, and yield of forest stands. European Journal of Forest Research, 135(1), 1– 22. https://doi.org/10.1007/s10342-015-0913-z Price, J. N., Wright, B. R., Gross, C. L., & Whalley, W. R. D. B. (2010). Comparison of seedling emergence and seed extraction techniques for estimating the composition of soil seed banks. Methods in Ecology and Evolution, 1(2), 151–157. https://doi.org/10.1111/j.2041-210X.2010.00011.x Purdon, M., Brais, S., & Bergeron, Y. (2004). Initial response of understorey vegetation to fire severity and salvage-logging in the Southern boreal forest of Québec. Applied Vegetation Science, 7(1), 49–60. https://doi.org/10.1111/j.1654-109X.2004.tb00595.x 69 Qi, M., & Scarratt, J. B. (1998). Effect of harvesting method on seed bank dynamics in a boreal mixedwood forest in northwestern Ontario. Canadian Journal of Botany, 76(5), 872–883. https://doi.org/10.1139/b98-061 R Core Team. (2023). R: A Language and Environment for Statistical Computing (Version 4.3.1). https://www.R-project.org/ Reeves, S. L. (2007). Geranium bicknellii. In: Fire Effects Information System. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). Regulation 2023/1115. The Making Available on the Union Market and the Export from the Union of Certain Commodities and Products Associated with Deforestation and Forest Degradation and Repealing Regulation (EU) No 995/2010. (2023). http://data.europa.eu/eli/reg/2023/1115/2024-12-26/eng Renecker, L. A., & Hudson, R. J. (1992). Habitat and forage selection of moose in the aspendominated boreal forest, Central Alberta. Alces, 28, 189–201. Ricklefs, R. E. (1977). Environmental heterogeneity and plant species diversity: A hypothesis. The American Naturalist, 111(978), 376–381. Ringius, G. S., & Sims, R. A. (1997). Indicator Plant Species in Canadian Forests. Canadian Forest Service, Natural Resources Canada. Rönnegård, L., Sand, H., Andrén, H., Månsson, J., & Pehrson, Å. (2008). Evaluation of four methods used to estimate population density of moose Alces alces. Wildlife Biology, 14(3), 358–371. https://doi.org/10.2981/0909-6396(2008)14[358:EOFMUT]2.0.CO;2 Rossman, A. K., Halpern, C. B., Harrod, R. J., Urgenson, L. S., Peterson, D. W., & Bakker, J. D. (2018). Benefits of thinning and burning for understory diversity vary with spatial scale and time since treatment. Forest Ecology and Management, 419–420, 58–78. https://doi.org/10.1016/j.foreco.2018.03.006 Saska, M. M., & Kuzovkina, Y. A. (2010). Phenological stages of willow (Salix). Annals of Applied Biology, 156(3), 431–437. https://doi.org/10.1111/j.1744-7348.2010.00400.x Schier, G. A., Jones, J. R., & Winokur, R. P. (1985). Vegetative regeneration in Aspen: Ecology and Management in the Western United States (General Technical Report RM-119; pp. 29–33). USDA Forest Service, Rocky Mountain Forest and Range Experiment Station. Schulz, K. F., & Grimes, D. A. (2005). Multiplicity in randomised trials I: Endpoints and treatments. The Lancet, 365(9470), 1591–1595. https://doi.org/10.1016/S01406736(05)66461-6 70 Schweiger, A. H., Boulangeat, I., Conradi, T., Davis, M., & Svenning, J.-C. (2019). The importance of ecological memory for trophic rewilding as an ecosystem restoration approach. Biological Reviews, 94(1), 1–15. https://doi.org/10.1111/brv.12432 Sedjo, R. A. (2001). The role of forest plantations in the world’s future timber supply. The Forestry Chronicle, 77(2), 221–225. https://doi.org/10.5558/tfc77221-2 Simonetti, J. A., Grez, A. A., & Estades, C. F. (2013). Providing habitat for native mammals through understory enhancement in forestry plantations. Conservation Biology, 27(5), 1117–1121. https://doi.org/10.1111/cobi.12129 Stark, K. E., Arsenault, A., & Bradfield, G. E. (2006). Soil seed banks and plant community assembly following disturbance by fire and logging in interior Douglas-fir forests of south-central British Columbia. Canadian Journal of Botany, 84(10), 1548–1560. https://doi.org/10.1139/b06-112 Stark, K. E., Arsenault, A., & Bradfield, G. E. (2008). Variation in soil seed bank species composition of a dry coniferous forest: Spatial scale and sampling considerations. Plant Ecology, 197(2), 173–181. https://doi.org/10.1007/s11258-007-9368-8 Sullivan, T. P., Sullivan, D. S., Lindgren, P. M. F., & Boateng, J. O. (2002). Influence of conventional and chemical thinning on stand structure and diversity of plant and mammal communities in young lodgepole pine forest. Forest Ecology and Management, 170(1), 173–187. https://doi.org/10.1016/S0378-1127(01)00775-7 Sullivan, T. P., Sullivan, D. S., Lindgren, P. M. F., & Ransome, D. B. (2007). Long-term responses of ecosystem components to stand thinning in young lodgepole pine forest: IV. Relative habitat use by mammalian herbivores. Forest Ecology and Management, 240(1), 32–41. https://doi.org/10.1016/j.foreco.2006.11.020 Sullivan, T. P., Sullivan, D. S., Lindgren, P. M. F., & Ransome, D. B. (2010). Long-term responses of mammalian herbivores to stand thinning and fertilization in young lodgepole pine ( Pinus contorta var. Latifolia ) forest. Canadian Journal of Forest Research, 40(12), 2302–2312. https://doi.org/10.1139/X10-173 Taylor, S. W., Carroll, A., Alfaro, R., & Safranyik, L. (2006). Forest, Climate and Mountain Pine Beetle Outbreak Dynamics in Western Canada. In The Mountain Pine Beetle – A Synthesis of Biology, Management, and Impacts in Lodgepole Pine (pp. 67–94). Thomson, F. J., Moles, A. T., Auld, T. D., & Kingsford, R. T. (2011). Seed dispersal distance is more strongly correlated with plant height than with seed mass. Journal of Ecology, 99(6), 1299–1307. https://doi.org/10.1111/j.1365-2745.2011.01867.x Tilman, D. (1985). The Resource-Ratio Hypothesis of Plant Succession. The American Naturalist, 125(6), 827–852. 71 Wang, G. (2000). Use of understory vegetation in classifying soil moisture and nutrient regimes. Forest Ecology and Management, 129(1), 93–100. https://doi.org/10.1016/S0378-1127(99)00142-5 Wang, T., Hamann, A., Spittlehouse, D., & Carroll, C. (2016). Locally downscaled and spatially customizable climate data for historical and future periods for North America. PLoS ONE, 11(6), e0156720. https://doi.org/doi:10.1371/journal.pone.0156720 Willms, J., Bartuszevige, A., Schwilk, D. W., & Kennedy, P. L. (2017). The effects of thinning and burning on understory vegetation in North America: A meta-analysis. Forest Ecology and Management, 392, 184–194. https://doi.org/10.1016/j.foreco.2017.03.010 Yavitt, J. B., & Fahey, T. J. (1986). Litter decay and leaching from the forest floor in Pinus Contorta (Lodgepole Pine) Ecosystems. Journal of Ecology, 74(2), 525–545. https://doi.org/10.2307/2260272 Zobel, M., Kalamees, R., Püssa, K., Roosaluste, E., & Moora, M. (2007). Soil seed bank and vegetation in mixed coniferous forest stands with different disturbance regimes. Forest Ecology and Management, 250(1), 71–76. https://doi.org/10.1016/j.foreco.2007.03.011 72 Appendix Table A1 Number of replicates of each treatment sampled in 3 years post-treatment area and in 1 year post-treatment area, and untreated control forest. Total number of areas sampled is given. Sampling Area Treatment Thinning 200 stems/ha 4 400 stems/ha 4 600 stems/ha 3 0.2 ha 3 0.5 ha 3 1.0 ha 3 2.0 ha 3 200 stems/ha 3 400 stems/ha 3 600 stems/ha 3 0.2 ha 3 0.5 ha 3 1.0 ha 3 2.0 ha 3 3 years posttreatment Gaps Thinning 1 year posttreatment Untreated # Replicates Sampled Gaps n/a 3 Total # Plots Sampled 23 21 3 TOTAL: 47 73 Table A2 All understory plant species observed in the study area (across all 51 plots) from June – August 2023, and their common names. If a species was mentioned in > 50% of the sources (3/6) the species was classified as being of high importance to moose. If a species was mentioned in < 50% of the sources the species was classified as medium importance for moose. If a species was not mentioned in any source, it was classified as low importance. Species Acer sp. Achillea millefolium Alnus sp. Anaphalis margaritacea Antennaria neglecta Aquilegia formosa Aralia nudicaulis Arnica sp. Castilleja miniata Chamaenerion angustifolium Cirsium vulgare Clintonia uniflora Cornus canadensis Crepis tectorum Epilobium ciliatum Fragaria virginiana Galium boreale Galium triflorum Geranium bicknelli Geum sp. Goodyera oblongifolia Hieracium albiflorum Hieracium aurantiacum Hieracium umbellatum Ledum groenlandicum Leucanthemum vulgare Linnaea borealis Lonicera involucrata Mitella nuda Oplopanax horridus Petasites frigidus Petasites sagittatus Common_Name Maple Yarrow Alder Pearly Everlasting Field pussytoes Red columbine Wild sasparilla Red paintbrush Fireweed Bull thistle Queen's Cup Canadian bunchberry Annual hawksbeard Purple-leaved willowherb Wild strawberry Northern bedstraw Sweet-scented bedstraw Bicknell's geranium Rattlesnake plantain White hawkweed Orange hawkweed Canadian hawkweed Labrador Tea Oxeye Daisy Twinflower Twinberry Common mitrewort Devil's Club Palmate coltsfoot Arrow leaf coltsfoot Native Importance to Moose Low Medium4,8 Low Low Low Low Low Low Low Perennial forb Perennial forb Perennial forb Native Exotic Native Native Medium4,5 Low Low Medium4 Annual forb Perennial forb Exotic Native Low Low Perennial forb Perennial forb Perennial forb Native Native Native Low Low Low Annual forb Perennial forb Perennial forb Perennial forb Perennial forb Perennial forb Native Native Native Exotic Native Low Low Low Low Low Low Evergreen shrub Perennial forb Evergreen shrub Deciduous shrub Perennial forb Deciduous shrub Perennial forb Perennial forb Native Exotic Native Native Native Native Native Native Low Low Low Medium5 Low Low Low Low Longevity1,2 Deciduous shrub Perennial forb Deciduous shrub Perennial forb Perennial forb Perennial forb Perennial forb Perennial forb Perennial forb Perennial forb Native/Exotic1,2 Native Native Native Native Native 74 Pinus sp. Plantago sp. Platanthera hyperborea Platanthera orbiculata Populus tremuloides Pyrola asarifolia Rhinanthus minor Rhododendron albiflorum Ribes hudsonianum Ribes lacustre Rosa acicularis Rubus idaeus Rubus parviflorus Rubus pedatus Rubus pubescens Rubus spectabilis Rubus spp. Salix spp. Sambucus spp. Sedum stenopetalum Senecio triangularis Senecio sp. Shepherdia canadensis Smilacina racemosa Solidago canadensis Sorbus scopulina Solidago sp. Spiraea betulifolia Spiraea douglasii Spiraea spp. Stellaria longipes Symphyotrichum spathulatum Taraxacum officinale Vaccinium membranaceum Vaccinium myrtilloides Pine Green-flowered bog orchid Round leaf orchid Trembling aspen Pink Wintergreen Yellow rattle White-flowered rhododendron Northern black currant Swamp currant Prickly rose Red raspberry Thimbleberry Five-leafed bramble Trailing raspberry Salmonberry Willow Elderberry Worm-leaved stonecrop Arrow leaved groundsel Buffalo berry False Solomon's Seal Canada goldenrod Western MountainAsh Birch-leaved spirea Pink spirea Long-stalked starwort Purple mountain aster Common dandelion Black huckleberry Velvet leaf blueberry Tree Forb Perennial forb Perennial forb Tree Perennial forb Annual forb Deciduous shrub Deciduous shrub Native Native Native Native Native Native Native Medium4,9 Low Low Low High3,5,6,7,8,9 Low Low Medium5 Medium5 Deciduous shrub Perennial forb Native Medium5 High3,5,8 Low Low Low Low Low High3,5,8 High3,4,5,6,7,8,9 Medium4,5 Low Perennial forb Native Low Native Native Low Low Low Native Native Low Low Deciduous shrub Deciduous shrub Deciduous shrub Deciduous shrub Perennial forb Perennial forb Deciduous shrub Deciduous shrub Deciduous shrub Forb Deciduous shrub Forb Perennial forb Deciduous shrub Forb Deciduous shrub Deciduous shrub Deciduous shrub Native Native Native Native Native Native Native Native Native Native Perennial forb Perennial forb Low Low Low Medium3 Low Low Perennial forb Deciduous shrub Native Exotic Native Low Low Deciduous shrub Native Low 75 Vaccinium ovalifolium Vaccinium parvifolium Vaccinium spp. Viburnum edule Viola sp. 1. 2. 3. 4. 5. 6. 7. 8. oval-leafed blueberry Red huckleberry Blueberry, Huckleberry Highbush cranberry Deciduous shrub Native Low Deciduous shrub Deciduous shrub Native Low Medium3,5 Deciduous shrub Native High4,5,6,8 Low Perennial forb In Klinkenberg, Brian. Ed. (2020) (Parish et al., 1996) (Breithaupt et al., 2024) (Franzmann & Schwartz, 2007) (Haeussler et al., 1990) (Hodder et al., 2013) (Poole & Stuart-Smith, 2005) (Renecker & Hudson, 1992) 76 Table A3: Results of linear models showing the effect of observed tree density and the square of observed tree density on herb and shrub cover, richness, and diversity. Values represent AIC value (p-value). Significant relationships (α = 0.05) are bolded. Stem density (Stem density)2 Cover 206.3 (0.145) 205.5 (0.093) Herb layer Richness Diversity 150.0 58.5 (0.205) (0.407) 148.4 58.1 (0.071) (0.308) Cover 227.0 (0.560 227.2 (0.681) Shrub layer Richness 93.6 (0.652) 93.4 (0.515) Diversity 24.2 (0.205) 25.0 (0.340) Table A4 Estimated marginal means (EMMs) for the LMEs and GLMEs showing the effect of thinning treatment on herb layer cover and richness. Values represent the F-value (pvalue) of the linear mixed-effects models (cover) and the Z-value (p-value) for the general linear mixed-effect models (richness). Significant p-values (α = 0.05) are bolded. contrast Herb cover Herb richness 200 – 400 0.29 (0.775) 0.46 (0.647) 200 – 600 -0.186 (0.859) -0.72 (0.471) 200 – ctrl 1.71 (0.132) 2.35 (0.019) 400 – 600 -0.468 (0.654) -1.09 (0.276) 400 – ctrl 1.39 (0.208) 1.70 (0.089) 600 – ctrl 1.90 (0.109) 2.97 (0.003) 77 Table A5 Estimated marginal means (EMMs) for the LMEs and GLMEs showing the effect of gap treatment and time on herbaceous (herb) layer cover and richness. Values represent the F-value (p-value) of the linear mixed-effects models (cover and diversity) and the Zvalue (p-value) for the general linear mixed-effect models (richness). Significant p-values (α = 0.05) are bolded. contrast 0.2 - 0.5 0.2 - 1.0 0.2 - 2.0 0.2 – ctrl 0.5 - 1.0 0.5 - 2.0 0.5 – ctrl 1.0 - 2.0 1.0 – ctrl 2.0 – ctrl Herb cover 0.53 (0.602) -1.10 (0.287) 1.39 (0.181) 2.31 (0.032) -1.62 (0.121) 0.85 (0.403) 1.76 (0.096) 2.49 (0.230) 3.32 (0.005) 0.95 (0.354) Herb richness 0.42 (0.678) -0.60 (0.549) -0.31 (0.760) 2.73 (0.006) -1.01 (0.311) -0.72 (0.472) 2.33 (0.020) 0.29 (0.769) 3.30 (0.001) 3.02 (0.003) Table A6 Estimated marginal means (EMMs) for the LMEs showing the effect of thinning treatment and time since treatment on shrub layer diversity. Values represent the F-value (pvalue) of the linear mixed-effects models (cover) and the Z-value (p-value) for the general linear mixed-effect models (richness). Significant p-values (α = 0.05) are bolded. Contrast 200 – 400 200 - 600 200 – ctrl 400 – 600 400 – ctrl 600 – ctrl estimate 0.230 0.328 -0.229 0.097 -0.46 -0.557 Contrast 200 – 400 200 – 600 200 – ctrl 400 – 600 400 – ctrl 600 – ctrl estimate 0.413 0.649 0.293 0.237 -0.120 -0.357 1-year post-treatment SE df 0.139 8.02 0.186 8.92 0.308 12.91 0.128 7.11 0.310 13.02 0.324 14.32 3-years post-treatment SE df 0.139 8.46 0.193 9.82 0.313 13.13 0.140 8.02 0.305 12.52 0.319 13.72 t.ratio 1.65 1.76 -0.74 0.76 -1.49 -1.72 p.value 0.136 0.112 0.470 0.471 0.161 0.108 t.ratio 2.967 3.363 0.935 1.692 -0.392 -1.119 p.value 0.017 0.007 0.367 0.129 0.702 0.282 78 Table A7 Estimated marginal means (EMMs) for the LMEs showing the effect of gap treatment and time since treatment on shrub layer diversity. Values represent the F-value (pvalue) of the linear mixed-effects models (cover) and the Z-value (p-value) for the general linear mixed-effect models (richness). Significant p-values (α = 0.05) are bolded. Contrast 0.2 - 0.5 0.2 - 1.0 0.2 - 2.0 0.2 – ctrl 0.5 - 1.0 0.5 - 2.0 0.5 – ctrl 1.0 - 2.0 1.0 – ctrl 2.0 – ctrl estimate -0.1426 0.1692 0.1932 0.2088 0.3118 0.3358 0.3514 0.0241 0.0397 0.0156 contrast 0.2 - 0.5 0.2 - 1.0 0.2 - 2.0 0.2 – ctrl 0.5 - 1.0 0.5 - 2.0 0.5 – ctrl 1.0 - 2.0 1.0 – ctrl 2.0 – ctrl estimate -0.4472 -0.9917 -0.8489 -0.6362 -0.5445 -0.4017 -0.1891 0.1428 0.3555 0.2127 1-year post-treatment SE df 0.294 19.98 0.301 19.95 0.168 4.73 0.290 20.00 0.214 5.58 0.294 19.98 0.171 5.02 0.301 19.95 0.216 5.85 0.290 20.00 3-years post-treatment SE df 0.264 7.35 0.312 19.88 0.233 6.43 0.184 5.44 0.301 19.95 0.214 5.58 0.216 5.85 0.294 19.98 0.290 20.00 0.171 5.02 t.ratio -0.485 0.563 1.152 0.721 1.454 1.142 2.054 0.080 0.184 0.054 p.value 0.633 0.580 0.304 0.478 0.200 0.267 0.095 0.937 0.860 0.958 t.ratio -1.693 -3.181 -3.641 -3.462 -1.812 -1.874 -0.877 0.485 1.228 1.243 p.value 0.132 0.005 0.010 0.015 0.085 0.114 0.415 0.633 0.234 0.269 Table A8: Results of post-hoc test (pairwiseAdonis) of PERMANOVA showing effect of thinning treatments, time since treatment, and their interaction on plant species composition in the herbaceous layer represented as F-value (p-value). Significant relationships (α = 0.05) are bolded. Contrast 200-400 200-600 200-ctrl 400-600 400-ctrl 600-ctrl R2 0.034 0.076 0.286 0.086 0.297 0.300 Adjusted p-value 0.945 0.506 0.001 0.426 0.005 0.012 79 Table A9 Results of post-hoc test (pairwiseAdonis) of PERMANOVA showing effect of gap treatments, time since treatment, and their interaction on plant species composition in the herbaceous layer represented as F-value (p-value). Significant relationships (α = 0.05) are bolded. Contrast 0.2 – ctrl 0.2 – 2.0 0.2 – 1.0 0.2 – 0.5 ctrl – 2.0 ctrl – 1.0 ctrl – 0.5 2.0 – 1.0 2.0 – 0.5 1.0– 0.5 R2 0.003 0.787 0.584 0.929 0.059 0.008 0.001 0.863 0.428 0.152 Adjusted p-value 0.004 0.800 0.544 0.945 0.052 0.013 0.002 0.857 0.425 0.158 80 Table A10 Estimated marginal means (EMMs) for the general linear model showing the effect of thinning treatment on presence of feces in the herbaceous layer. Values represent estimate, standard error, degrees of freedom, z-ratio, and p-values for all factor combinations. Significant p-values (α = 0.05) are bolded. Contrast 200 – 400 200 – 600 200 – ctrl 400 – 600 400 – ctrl 600 – ctrl estimate -0.29 -0.54 0.13 -0.25 0.42 0.67 SE 0.21 0.17 0.10 0.30 0.19 0.14 z.ratio -1.36 -3.12 1.15 -1.09 2.26 4.90 p.value 0.173 0.002 0.248 0.276 0.024 <.0001 Table A11 Estimated marginal means (EMMs) for the general linear model showing the effect of gap treatment on presence of feces in the herbaceous layer. Values represent estimate, standard error, degrees of freedom, z-ratio, and p-values for all factor combinations. Significant p-values (α = 0.05) are bolded. Contrast 0.2 - 0.5 0.2 - 1.0 0.2 - 2.0 0.2 – ctrl 0.5 - 1.0 0.5 - 2.0 0.5 – ctrl 1.0 - 2.0 1.0 – ctrl 2.0 - ctrl estimate -0.33 -0.33 -0.33 0.167 0.00 0.00 0.50 0.00 0.50 0.50 SE 0.24 0.14 0.24 0.14 0.19 0.27 0.19 0.19 0.00 0.19 z.ratio -1.41 -2.45 -1.41 1.23 0.00 0.00 2.60 0.00 25316.034 2.60 p.value 0.157 0.014 0.157 0.221 1.00 1.00 0.009 1.00 <0.0001 0.009 81 Table A12 List of the species that germinated during the greenhouse trial, with main method of dispersal, life form, and origin status (native or exotic). Scientific name Crepis tectorum1 Epilobium ciliatum2,6 Fragaria vesca3 Geranium bicknellii4 Pilosella sp5 Rubus idaeus6 Stellaria crispa Vaccinium spp.6 1. 2. 3. 4. 5. 6. Common name Narrowleaf hawksbeard Fringed willowherb Wood strawberry Bicknell's cranesbill Crisp starwort Blueberries and huckleberries Dispersal Life form Native/exotic anemochory anemochory zoochory zoochory anemochory zoochory annual forb perennial forb perennial forb annual forb perennial forb deciduous shrub perennial forb exotic native native native exotic native native zoochory deciduous shrub native (Najda et al., 1982) (Popay, 2014) (Munger, 2006) (Reeves, 2007) (Makepeace, 1985) (Haeussler et al., 1990) 82 83 Crepis tectorum 0 0 0 0 2.00 0 0 0 0 0 0 0 0 0 0 0 0 Epilobium ciliatum 2.22 0 0 13.33 32.49 6.15 0 0 0 2.22 0 15.00 0 0 0 0 0 Fragaria vesca 0 0 0 0 4.00 0 0 0 0 0 0 0 0 0 0 0 0 Geranium bicknellii 0 0 0 2.67 2.00 0 0 0 0 0 0 0 0 0 0 0 0 Pilosella sp 0 1.33 1.00 2.67 2.00 1.53 0 1.25 0 0 4.95 0 0 0 0 0 0 Rubus idaeus 0 0 0 0.80 0 7.40 0 0 1.81 0 0 0 0 0 0 0 0 graminoids 0 0 0 4.00 54.34 0.91 1.25 0 1.82 0 45.66 6.25 4.00 0 22.50 1.250 17.50 Stellaria crispa 0 0 0 0 14.00 0 0 0 0 0 0 0 0 0 0 0 0 Vaccinium sp 0 0 0 0 2.86 0 0 0 0 0 0 0 4.00 0 0 2.00 3.75 Unknown 0 0 0 0 0 0 0 0 4.66 0 0 0 1.33 1.82 0 0 0 Total germinants (1000 mL) 2.22 1.33 1.00 23.46 113.68 16.01 1.25 1.25 8.30 2.22 50.61 21.25 9.33 1.81 22.5 3.25 21.25 Table A13 The number of germinants of each species calculated per 1000 mL of soil for each treatment, separated by number of years post-treatment. Treatment 0.2 0.5 0.5 1.0 1.0 2.0 2.0 200 200 400 400 600 600 1800 ctrl ref ref Time Since Treatment 3 years 1 years 3 years 1 years 3 years 1 years 3 years 1 years 3 years 1 years 3 years 1 years 3 years 3 years 3 years 1 years 3 years