Do Ecological Succession and Invasive Plants Pose a Threat to Biodiversity in Iceberg Point, Lopez Island, WA?

DO ECOLOGICAL SUCCESSION AND INVASIVE PLANTS POSE A THREAT TO BIODIVERSITY IN ICEBERG POINT, LOPEZ ISLAND, WA? A PAPER SUBMITTED FOR COMPLETION OF SENIOR RESEARCH FOR THE COLLEGE OF ARTS AND SCIENCES STETSON UNIVERSITY BY Benjamin Chase IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE(S) OF BACHELOR OF SCIENCE ENVIRONMENTAL SCIENCE ADVISORS Dr. Wendy Anderson, Ph. D. Dr. J. Anthony Abbott, Ph.D. MAY 2016 i Table of Contents List of Illustrations……………………………………………………………………………...…ii List of Tables……………………………………………………………………………….….…iii Acknowledgements…………………………………………………………………………….…iv Abstract………………………………………………………………………………………..…..v Introduction……………………………………………………………………………………..…1 Literature Review……………………….…………………….……………………………….…..1 Methods……………………………………………………………………………………….…..6 Results……………………………………………………………………………………………..9 Discussion………………………………………………………………………………………..10 Conclusion………………………………………………………………………………….........13 Tables and Figures……………………………………………………………………………….14 Works Cited……………………………………………………………………………………..19 ii List of Illustrations Figure 1. San Juan Islands National Conservation Area………………………………..7 Figure 2. Map of Iceberg Point………………………………………………………….7 Figure 3. Hotspots of shrub species at Iceberg Point…………………………………..16 Figure 4. Hotspots of invasive species at Iceberg Point………………………………..17 Figure 5. Hotspots of charismatic species at Iceberg Point…………………………….18 iii List of Tables Table 1: Proportion of a species total occurrence within each habitat…………………14 Table 2: Proportion of total habitat in which a species of interest is found……………15iv Acknowledgements Thank you to Dr. Wendy Anderson for the use of her data from the vegetation survey, for including me in her survey team, and for her guidance through the entire process of my senior research. Thank you to Dr. J. Anthony Abbott for his advice on the research process, paper writing, and presentation. Finally, thank you to Tabitha Petri, Kathryn Nathenson, Sarah Coffey, and Morgen Holt. v Abstract Preserving ecosystems in their natural state can pose a challenge to land managers. On Lopez Island, two of the major threats to biodiversity of plants are shrub encroachment into grasslands and the destructive effect of invasive species. We conducted a botanical survey at Iceberg Point, an Area of Critical Environmental Concern that is managed by the Bureau of Land Management. We evaluated the spatial distribution of key species to determine if shrub encroachment was occurring, where it might be occurring, and what charismatic species might be at risk from habitat loss. We found locations of encroachment within grasslands, concurrent with or progressing towards the hotspots of charismatic species. We also found hotspots of invasive species that occurred in both these encroachment zones, and in hotspots of charismatic species. If encroachment continues, charismatic species will compete for the remaining habitat with invasive species and be at risk of local extinction. Combining this analysis with the historic land use patterns will help inform the developers of a new Resource Management Plan for Iceberg Point.1 Introduction Conservation practices are important for maintaining lands for future generations. One of the focuses of conservation efforts is on sustaining the native biodiversity and the ecosystem functions. In order to manage for a natural ecosystem, it is important to understand exactly what the native ecosystem is and the history of the land that may have influenced the native species and the processes that occur there. On conservation lands, the local biodiversity can be threatened by invasive species and ecological succession. Iceberg Point on Lopez Island, Washington is an area protected by the Bureau of Land Management, yet it may face a reduction in biodiversity due to these problems. Ascertaining the degree to which both invasive species and ecological succession could be threatening the overall biodiversity of this area will aid in the development of management plans for this site. Literature Review Government managed conservation lands are areas that are set aside for use of the public. Public use can incorporate leasing rights of the land to commercial interests for the harvest of materials, education of the public, and public recreation. The degree to which each kind of activity occurs on a parcel of land varies depending on how the land has been designated. This is part of the ultimate mission of the Forest Service and the Bureau of Land Management, to preserve ecosystems and their resources for future generations while ensuring that each generation has its own needs met (Culhane, 2013). The governmental agency that determines the use of these lands balances the needs of the present with the needs of the future. They might decide when, how, or if logging and recreation should take place at all. In all land use designations, an additional goal to the utilitarian purpose of the land is to manage for a functional and speciose ecosystem. Ideally, this results in future generations having the same access to land 2 as our current generation does, including the same aesthetic enjoyment. When logging occurs to meet the commercial needs of the current generation, there are lasting impacts on the ecosystem that has been harvested (Summerville and Crist, 2002; Petranka, Eldridge, and Haley, 1993; Holtby, 1988). Because recreation and commercial use of land leave ecosystems changed or degraded, there is a need to set aside some pristine areas that maintain biodiversity for future generations to enjoy. Conservation efforts are often focused on maintaining a natural ecosystem. Typically a natural ecosystem is defined as the community that lived in an area before humans began to influence it. When it comes to applying this definition to individual plants and animals in an ecosystem or what the “natural” state of an ecosystem is, the functional definition reverts to the first recorded state of the ecosystem by Europeans. By studying the historical ecology of a site, people can piece together what sorts of organisms inhabited a site and some of the processes that occurred there (Swetnam, Allen, and Betancourt, 1999). The historical ecology can stretch back over thousands of years, providing a large amount of information on the community composition at a given time as well as how the ecosystems changed over time. There are a few considerations to understand with this approach, however. The organisms present just after the most recent ice age may not be the organisms that would be considered natural in the current ecosystem because they are adapted to a different climate, but exist in refugia. Additionally, humans arrived and began altering the land a few thousand years after the most recent ice age (Rogers et al., 1991). Therefore, both the land that Europeans first encountered and the functional biota derived from historical ecology studies are in fact influenced by humans. The Salish people were settled in the Pacific Northwest when the Europeans arrived (Shebitz, 2015). The Coast Salish people were the first humans to live in the area, establishing 3 themselves in 8000 B.C. E. Traditionally the Coast Salish people lived in semi-permanent villages and moved with the seasons (Derr, 2014). A substantial portion of their diet was derived from fishing, but they also ate a large variety of plants. When Europeans first arrived and began settling the Island in the early 1800’s, they saw the Salish as nomads who moved from place to place searching for food. However, the Salish people had agricultural practices that increased the food supply in these places. They burned the land and weeded nettle patches to encourage further growth of useful plants (White, 1975). Using fire for agricultural purposes requires a comprehensive understanding of ecosystem functioning and growing seasons. An early settler reports watching the indigenous peoples near Vancouver Island set areas of woodland on fire in order to remove competition of their food crops and to make harvesting easier. The Salish would burn grasslands and woodlands after harvesting crops and before the next growing season, then move to a different area. They would then return to the original area they burned for the next harvest. (Derr, 2014). This system improved their overall crop yield from an ecosystem and in doing so in multiple areas, changed the landscape of the Pacific Northwest coast greatly. It can be difficult to separate natural fires from anthropogenic fires when looking through the charcoal records at a site. If the goal is to encourage an entirely natural ecosystem, one method is to study the current fire patterns in a similar climate and compare the timing and frequency of burning to the historic fire records (Slocum, et al., 2007). This can give land managers an idea of how a fire regime should look or if there should be one at all. After the first Europeans arrived, logging began in the area in the 1900’s along with the grazing of livestock (Shebitz, 2015). Like a fire regime, these practices also created and maintained grasslands. Since ecosystems vary over time naturally with disturbances such as fire or storms, the ecosystem that is recorded upon European arrival may not be what the natural dynamic state of a 4 landscape should be decades later (Sprugel, 1991). This can result in land managers fighting against natural processes to maintain a static image of what the ecosystem should be. Ecological succession is a natural process in which pioneer species colonize a recently disturbed area and are replaced by other species until a climax community is reached. Where conservation of human altered lands is concerned, it can be unclear whether ecological succession is due to the management practice of inaction, like fire suppression, or due to changes in climate that allow some plants to expand their local coverage (League and Veblen, 2006; Mast, Veblen, and Hodgson, 1997). In an ecosystem that is frequently disturbed, Ecological succession can threaten the local biodiversity of that ecosystem when the disturbance is halted. Often woody species of trees or shrubs will encroach on grasslands, excluding grasses and other shade intolerant forbs (League and Veblen, 2006). Fire suppression or neglecting to enforce a burn schedule in grasslands allows these woody plants to take over and the grasslands to be lost (Müller, et al, 2007). When viewed on a local scale, encroachment into grasslands can heavily impact the biodiversity of a conservation area (Ratajczak, Nippert, and Collins, 2012). Invasive species are non-native species that can have a substantial negative effect on the biodiversity of an ecosystem. New species can invade an ecosystem from seeds being dispersed by wind, water, or animals, or through human facilitation (Brown and Sax, 2004). Many of the newly arrived species may live harmlessly within an ecosystem without disrupting it, but a lack of specificity of language can imply that these species are “invasive” and a greater problem than they are (Colautti and MacIsaac, 2004). Since the removal of every non-native species can be costly and potentially cause more harm than good, it is important to identify the species that are causing ecological damage (Pimentel, Zuniga, and Morrison, 2005). Typically, species that are considered invasive will invade an ecosystem in a single point and overtake an area by 5 outcompeting native species (Muellerova et al., 2005). This can reduce the overall biodiversity of an area and push some species into local extinction (Hejda, Pyšek, and V. Jarošík, 2009; Colautti and MacIsaac, 2004). Communities that have high levels of disturbance, especially human disturbance, typically have higher invasion rates than those without disturbance (Sakai et al., 2001). Before management practices are enacted to conserve lands, it is important to determine which species are invasive and damaging, and what areas of the land are most at risk of invasion (Dunwiddie, 2002). The San Juan Islands National Monument (SJINM) is managed by the BLM and encompasses 75 parcels of land in the archipelago, Iceberg Point being one of those parcels. Iceberg Point has been managed as an Area of Critical Environmental Concern (ACEC) for the past 19 years, giving it a high level of protection, but its recent inclusion into the SJINM has brought its Resource Management Plan (RMP) up for review under the umbrella RMP of the rest of the monument. The purpose of this study is to inform the developers of the new RMP of the patterns and processes occurring at Iceberg Point that may pose a threat to the species distribution and composition of local and regional biodiversity. Specifically, we will assess, analyze, and map the distribution of key, invasive, non-native forbs to evaluate the patterns of invasive species. We will also assess, analyze, and map the successional shifts among the boundaries between grassland, shrub and forest habitats to determine which areas are undergoing the most rapid succession in the absence of historic human disturbance and the risks of local species loss if those areas are allowed to shift. 6 Methods Field Site Description We surveyed plants across the entire parcel of Iceberg Point, an “Area of Critical Environmental Concern” protected by limited access and services within the larger San Juan Islands National Monument. The entire Monument, which spans 75 separate parcels and small islands, including Iceberg Point, is managed by the Bureau of Land Management, which is based on Lopez Island, WA. Lopez Island is part of the San Juan Islands, which are located southeast of Vancouver Island and north of the Strait of Juan de Fuca (Figure 1). The archipelago includes 472 islands covering about 172 square miles with Lopez Island being the third largest at 29.4 square miles. The islands, which fall within the Olympic mountain rain shadow, receive between 26 and 150 inches of precipitation annually depending on the location. Iceberg Point (Figure 2) is located on the southwestern tip of Lopez Island and receives the least rainfall of the islands, 26 inches annually (Office of the Washington State Climatologist, 2014). Iceberg Point has a southern facing slope of grasslands that ends in a rocky shoreline, and forests to the north of the grasslands. For the analysis of how charismatic species might be affected by invasive species and encroachment, I focused on the grassland, shrub, and forest edge habitat types. This focus allows me to use the temporally limited data we collected to see how ecological succession has progressed so far and how it might continue. The grasslands habitats are dominated by grasses with forbs intermixed as well. Shrub habitats are characterized by low growing woody plants between two and six feet tall, with herbaceous, shade tolerant plants below or between shrubs. The forest edge habitats contain some herbaceous plants as well as trees and other woody plants, but do not form a continuous canopy, allowing in some light. The substrate of the site consists of glacial deposition on top of bedrock. (Jones, 1999). 7 Figure 1. San Juan Islands National Conservation Area Figure 2. Map of Iceberg Point8 Field Methods Our data collection took place over six weeks in May and June of 2015. We collected data by running transects from south to north 20 meters apart, using GPS coordinates to determine each transect and sample point. Areas too densely overgrown to access were deemed to be impassable and the dominant species in the impassable areas were visually assessed. We recorded the species found, the habitat type, and the percentage of the types of ground cover in a one square meter grid at each locus every 20 meters along each transect line. At each point we also made a general assessment of the species present in the 20 by 20 meter zone around the focal sampling point. Data Analysis I separated the species we encountered into exotic/ invasive species or native species using the Washington Native Plant Society as a reference (Knoke, 2004). To determine the distribution of plants between forest edge, shrub, and grassland habitats, I calculated the abundance of a species within each habitat as the proportion of loci with a particular species that were encountered in a particular habitat type. For instance, a species that was encountered five times in grasslands out of ten total encounters was said to have 50% of its distribution, or 50% of its local range, in grasslands. Species with a proportional occurrence over 70% in one habitat were considered to have a high affinity for that habitat. Species with less than 40% in a habitat were considered widely distributed with a low affinity. I then found the percentage of each habitat that contains a given species to find how common a species is within a habitat type. I used the results from each of those processes to select exotic, uncommon native, and potentially encroaching shrub species that showed unexpected trends in their occurrence to perform spatial analysis on. Using ArcGis, I performed a hotspot analysis on each species of interest. The 9 hotspot analysis calculates the Getis-Ord Gi* statistic, which assigns significance to points based on the values of all the other points with greater weight given to closer points. Therefore, a point surrounded by points of a similar value will have a greater significance than a point surrounded by points of random values. The GiZ score produced by the test represents the significance of clustering at a given point. Higher values indicate greater presence, lower values indicate greater avoidance, and values near zero indicate random distribution. While the affinity and range analyses showed how individual species related to habitat designations, the hotspot analysis gave spatial patterns of the groups of plants with possible clustering of shrubs in grasslands forming the encroachment line. Results Three shrub species, Rosa nutkana, Rubus ursinus, and Symphoricarpos albus, were widely distributed throughout the three habitats, while also occurring in a large proportion of grassland habitats. 48% of the grassland plots had one of the three shrubs occurring in them. Rosa nutkana was the most abundant shrub encountered in each of the three habitats analyzed, occurring in grassland plots twice as often as either of the other two shrubs. Syphoricarpos albus occurred more often in forest edge than in grasslands. Of the charismatic species we encountered at Iceberg Point, six were shown to have a high relative occurrence within grassland habitats, while also only existing within a small number of grassland habitats. Frittilaria lanceolata and Brodiaea coronaria are the most locally abundant species, occurring in 35% and 20% of the grassland habitat type, respectively. With the exception of Brodiaea coronaria, which also occurs in habitats other than the three analyzed, all of the selected charismatic species have 40% or more of their range in grasslands. All six charismatic grassland species have 10% or less of their range in shrub habitat and most have less10 than 5% of their range in forest edge. Similarly, the charismatic species were found in half as many shrub plots as grassland plots. Frittilaria lanceolata occurs in two to four times as many shrub plots as any other charismatic species. Eight of the invasive species we encountered show patterns of either being restricted to grasslands or of occurring in most of the grassland plots (Tables 1 and 2). Plantago lanceolata, Hypochaeris radicata, and Rumex acetosella occurred in over 50% of the grassland plots surveyed. Cirsium arvense, Medicago lupulina, Plantago lanceolata, Ranunculus arvense, and Vicia sativa had over 40% of their range in grasslands. From grassland to shrub habitats, invasive species decreased by a factor of 0.2-5. Cirsium arvense stayed near 20% occurrence between grassland and shrub habitats. Hypochaeris radicata, and Rumex acetosella both are also present in around 20% of shrub habitat, and Plantago lanceolata was found in 43% of the shrub habitat. Hotspot analysis of the shrubs shows a significant clustering of shrubs in parts of the southern grasslands in the middle of the parcel (Figure 3). Charismatic species and invasive species showed significant clustering in the southern central grasslands and southeastern grasslands (Figures 4 and 5). Invasive species also had a hotspot in the extreme east grasslands (Figure 4). Discussion The three shrub species selected are found in a large portion of grassland plots that were sampled. Rosa nutkana occurs in over a third of all grassland plots, while Rubus ursinus and Symphoricarpos albus were encountered in nearly a fifth of grassland plots. Historically fire and grazing maintained the grasslands by routinely destroying the slower growing woody plants (Derr, 2014). The removal of these controlling factors allows shrubs to continue to grow and shade out the low growing grassland species. The proportional occurrence of shrubs within the 11 grasslands shows that succession is occurring in these plots, and they will be overtaken by shrubs if not managed. Based on the hotspot analysis, most of the shrubs found in the grasslands are concentrated near existing shrub habitat already, again supporting the hypothesis that shrubs are expanding their range into the grasslands. Many of the charismatic grassland species we encountered are not only found mostly within the grasslands, but also occur in a low proportion of that habitat (Tables 1 and 2). This indicates that grassland conditions are important to the survival of these species in Iceberg Point, and that they are locally rare species. Therefore, losing even a small area of habitat to shrubs can have a large impact on the population of a species. Comparing the hotspot analysis of charismatic species (Figure 5) and the shrubs (Figure 3), we see an overlap in clustering of these groups in the southern central grasslands. This represents the area at greatest risk of habitat loss for these species due to shrub encroachment. Some species, such as Fritillaria lanceolata, was also encountered in 17% of the shrub plots, while Camassia quamash and Sisyrinchium idahoensi have 10% of their occurrence within shrub habitat. These species show us where encroachment might currently be happening. Their co-occurrence with shrubs means that initially some of these species are not choked out by a shrub habitat, but they will be greatly reduced in number and may eventually dwindle to extinction after a few seasons of being shaded out by shrubs. The eight invasive species that occur in a large proportion of the grasslands and were often found in plots with charismatic species. The invasive species are also widely distributed throughout shrub habitats and can compete successfully enough to coexist with shrub species. The greater danger of invasive species is not in competing with native species in the grasslands, but in competing with them in the restricted niche space left by ecological succession. Aside 12 from Medicago lupulina and Ranunculus arvense, the grassland invasive species were found in a greater percentage of shrub habitat than the charismatic species were. This indicates that they are better competitors with the woody shrubs and are not as easily shaded out. Cirsium arvense was found nearly as often in shrubs as it was in grasslands, so is barely affected by the encroachment of shrubs. Many of the invasive species, including Cirsium arvense and Plantago lanceolata, also have a large basal rosette, which can take up large amounts of space and crowd out charismatic species that could have remained following shrubs taking over a section of grasslands. Invasive species have less significant clustering in the southern central grasslands where charismatic species are most at risk, but they do share a hotspot in the southeastern grasslands with charismatic species. While the shared hotspot is not overlapped by a shrub hotspot, there is a significant clustering of shrubs in the grasslands to the north of the shared hotspot. If no steps are taken to stop shrub encroachment from taking place, the encroachment line will eventually reach the invasive-charismatic species hotspot. Charismatic species play an important role in encouraging people to conserve natural areas. Iceberg Point is home to many charismatic plants that can be found in only a few areas throughout the islands. The limited public access and services, regular monitoring by volunteers, and constant oversight by BLM staff at Iceberg Point have saved these species from development and destructive grazing practices, but it has also halted some of the processes that maintain the habitat itself. As shrub encroachment continues, the populations of these charismatic species are likely to shrink into unsustainably small populations. Shrubs need to be controlled at the encroachment hotspots in order to maintain enough grassland habitat for other species to exist. 13 Conclusion Ecosystems are not static over time. Often disturbance is part of a regular cycle and supports a high level of biodiversity and ecosystem function. Historically, Iceberg Point has been burned on a regular basis, which has maintained the grasslands and the species within them. Encroachment can be seen happening now and the impact on charismatic species is severe in areas where encroachment is happening. Without fire to destroy the slower growing woody species, encroachment will continue into the grasslands, leaving very little of this open slope left to some of the more remarkable plants found there. Conservation practices are usually focused at preserving areas for future generations to enjoy. As an Area of Critical Environmental Concern, the San Juan Islands National Monument should be preserved not through the complete removal of human activities, but as the dynamic system that it is. For Iceberg Point, this means reinstating a fire regime to keep a regular shifting of habitats in the short term and a diverse landscape in the long term. If encroachment is kept at bay, the negative impact of invasive species will be significantly reduced. Charismatic species will not be competing with invasive species for the limited amount of sunlight and nutrients that the shrubs don’t take up. Invasive species should still be controlled, but efforts to control them will be greatly helped by a routine burning of the landscape. After a fire knocks them back, it will be easier to locate the greatest trouble spots and remove invasive species to allow more room for charismatic species. Iceberg Point is a uniquely beautiful site that has a large support network throughout the community. The charismatic species garner a lot of attention, but they are only part of what makes Iceberg Point so special. What makes it so special is that it contains a diversity of life and landscapes and it is allowed exist for its own intrinsic beauty. It captures the imagination, and visitors can lay aside the stress of the modern world and see this land as it has been for thousands of years. The new 14 management plan should reflect the value that visitors derive from this site and preserve the biodiversity that this land supported historically. Tables and Figures Table 1. Proportion of a species total occurrence within each habitat Forest Edge Shrub Grass Other Shrubs Rosa nutkana 0.15 0.12 0.18 54.0 Rubus ursinus 0.16 0.14 0.16 54.2 Symphoricarpos albus 0.26 0.16 0.18 40.7 Charismatic Species Allium cernuum 0.20† 0.00 0.40 40.0 Brodiaea coronaria 0.01 0.02 0.33 63.4 Camassia quamash 0.05 0.10 0.48 38.1 Frittilaria lanceolata 0.05 0.09 0.59 28.0 Piperia maritima 0.20† 0.00 0.80 0.0 Sisyrinchium idahoensi 0.00 0.10 0.57 33.3 Invasive Species Cirsium arvense 0.04 0.13 0.45 37.3 Hypochaeris radicata 0.13 0.03 0.31 52.3 Medicago lupulina 0.07 0.02 0.52 39.1 Plantago lanceolata 0.09 0.07 0.44 39.9 Ranunculus arvense 0.10 0.04 0.55 31.4 Rumex acetosella 0.08 0.04 0.35 52.5 Vicia hirsuta 0.18 0.09 0.35 37.8 Vicia sativa 0.15 0.08 0.53 24.4 † The relatively high values are an artifact of low local abundance. Both species were encountered beside the path in forest edge, so their occurrence is likely a product of disturbance allowing more light in than typical of forest edge habitat. 15 Table 2. Proportion of total habitat in which a species of interest is found Forest Edge Shrub Grass Shrubs Rosa nutkana 0.49 0.76 0.36 Rubus ursinus 0.31 0.50 0.18 Symphoricarpos albus 0.43 0.50 0.18 Charismatic Species Allium cernuum 0.01 0.00 0.01 Brodiaea coronaria 0.01 0.05 0.20 Camassia quamash 0.01 0.05 0.07* Frittilaria lanceolata 0.05 0.17 0.36 Piperia maritima 0.01 0.00 0.03 Sisyrinchium idahoensi 0.00 0.07 0.13 Invasive Species Cirsium arvense 0.04 0.21 0.22 Hypochaeris radicata 0.42 0.21 0.62 Medicago lupulina 0.04 0.02 0.18 Plantago lanceolata 0.26 0.43 0.80 Ranunculus arvense 0.06 0.05 0.21 Rumex acetosella 0.21 0.19 0.54 Vicia hirsuta 0.16 0.17 0.19 Vicia sativa 0.22 0.24 0.49 *Sampling occurred after peak season for Camassia quamash resulting in a lower recorded occurrence. Actual presence is expected to be much higher throughout the grasslands while in season.16 Figure 3. Hotspots of shrub species at Iceberg Point Habitats types are displayed using the base colors with the results of the cluster analysis for shrubs overlaid in points. Red values show significant clustering of shrubs. Blue values show significant avoidance by shrubs. Red circles signal significant hotspots of shrubs in grassland habitats. 17 Figure 4. Hotspots of invasive species at Iceberg Point Habitats types are displayed using the base colors with the results of the cluster analysis for invasive speices overlaid in points. Red values show significant clustering of invasive species. Blue values show significant avoidance by invasive species. Black circles signal significant clusters of invasive species in grassland habitats.18 Figure 5. Hotspots of charismatic species at Iceberg Point Habitats types are displayed using the base colors with the results of the cluster analysis for charismatic species overlaid in points. Red values show significant clustering of charismatic species. Blue values show significant avoidance by charismatic species. Red circles represent shrub hotspots in grasslands from Figure 3. Black circles represent invasive species hotspots in Figure 4. 19 Works Cited Brown, J. H., and D. F. Sax. 2004. An essay on some topics concerning invasive species. Austral Ecology (29) 5: 530-536. Colautti, R. I., and H. J. MacIsaac. 2004. A neutral terminology to define ‘invasive’ species. Diversity and Distributions (10) 2: 135-141. Culhane, P. J. 2013. Public Participation. In Public lands politics: Interest group influence on the Forest Service and the Bureau of Land Management, ed. Anonymous 225-280. Routledge. Derr, K. M. 2014. Anthropogenic fire and landscape management on Valdes Island, Southwestern BC. Canadian Journal of Archaeology (38) 1: 250-279. Dunwiddie, P. 2002. Management and restoration of grasslands on Yellow Island, San Juan Islands, Washington, USA. Paper presented at: garry oak Ecosystem Restoration: Progress and Prognosis. Proceedings of the Third annual Meeting of the Bc chapter of the society for Ecological Restoration, pp. 78-87. Eldridge, D. J., M. A. Bowker, F. T. Maestre, E. Roger, J. F. Reynolds, and W. G. Whitford. 2011. Impacts of shrub encroachment on ecosystem structure and functioning: towards a global synthesis. Ecology Letters (14) 7: 709-722. Hejda, M., P. Pyšek, and V. Jarošík. 2009. Impact of invasive plants on the species richness, diversity and composition of invaded communities. Journal of Ecology (97) 3: 393-403.20 Holtby, L. B. 1988. Effects of logging on stream temperatures in Carnation Creek British Columbia, and associated impacts on the coho salmon (Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic Sciences (45) 3: 502-515. Jones, M. A. 1999. Geologic framework of the Puget Sound aquifer system, Washington and British Columbia. Knoke, D. 2004. Vascular Plant List of San Juan County. In Washington Native Plant Society [database online] http://www.wnps.org/plant_lists/counties/san_juan/san_juan_county.html (Accessed 15 September 2015). League, K., and T. Veblen. 2006. Climatic variability and episodic Pinus ponderosa establishment along the forest-grassland ecotones of Colorado. Forest Ecology and Management (228) 1: 98-107. Mast, J. N., T. T. Veblen, and M. E. Hodgson. 1997. Tree invasion within a pine/grassland ecotone: an approach with historic aerial photography and GIS modeling. Forest Ecology and Management (93) 3: 181-194. Muellerova, J., P. Pyšek, V. Jarošík, and J. Pergl. 2005. Aerial photographs as a tool for assessing the regional dynamics of the invasive plant species Heracleum mantegazzianum. Journal of Applied Ecology (42) 6: 1042-1053. Müller, S. C., G. E. Overbeck, J. Pfadenhauer, and V. D. Pillar. 2007. Plant functional types of woody species related to fire disturbance in forest–grassland ecotones. Plant Ecology (189) 1: 1-14. 21 Office of the Washington State Climatologist. 2014. N.W. Temperature, Precipitation, and SWE trend analysis. http://www.climate.washington.edu/trendanalysis/ (Accessed 28 January 2016). Petranka, J. W., M. E. Eldridge, and K. E. Haley. 1993. Effects of timber harvesting on southern Appalachian salamanders. Conservation Biology 363-370. Pimentel, D., R. Zuniga, and D. Morrison. 2005. Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecological Economics (52) 3: 273-288. Ratajczak, Z., J. B. Nippert, and S. L. Collins. 2012. Woody encroachment decreases diversity across North American grasslands and savannas. Ecology (93) 4: 697-703. Rogers, R. A., L. A. Rogers, R. Hoffmann, and L. D. Martin. 1991. Native American biological diversity and the biogeographic influence of Ice Age refugia. Journal of Biogeography 623-630. Sakai, A. K., F. W. Allendorf, J. S. Holt, D. M. Lodge, J. Molofsky, K. A. With, S. Baughman, R. J. Cabin, J. E. Cohen, and N. C. Ellstrand. 2001. The population biology of invasive specie. Annual Review of Ecology and Systematics 305-332. Shebitz, D. 2015. Puget lowland forests. Salem Press Encyclopedia of Science. Slocum, M. G., W. J. Platt, B. Beckage, B. Panko, and J. B. Lushine. 2007. Decoupling natural and anthropogenic fire regimes: a case study in Everglades National Park, Florida. Natural Areas Journal (27) 1: 41-55.22 Sprugel, D. G. 1991. Disturbance, equilibrium, and environmental variability: what is ‘natural’vegetation in a changing environment? Biological Conservation (58) 1: 1-18. Summerville, K. S., and T. O. Crist. 2002. Effects of timber harvest on forest Lepidoptera: community, guild, and species responses. Ecological Applications (12) 3: 820-835. Swetnam, T. W., C. D. Allen, and J. L. Betancourt. 1999. Applied historical ecology: using the past to manage for the future. Ecological Applications (9) 4: 1189-1206. White, R. 1975. Indian land use and environmental change: Island County, Washington: A case study. Arizona and the West 327-338.