1. 1. ABSTRACT Disturbance by geomorphic and anthropogenic processes affects riparian substrate, nutrient levels, canopy shading, and hydrology. As such, fluvial systems commonly serve as conduits for the dispersal of exotic plant species. This study involves spatial analysis of vascular plant distribution in the riparian understory of the Luckiamute River basin, central Oregon Coast Range. Preliminary results are used to decipher geomorphic and anthropogenic controls on adventive species propagation in an unregulated watershed. Over 1700 m 2 of riparian understory was surveyed using 1- by 100-m transects oriented perpendicular to the active channel, with 20 survey stations irregularly spaced (D avg = 5.1 km) along the lower 100 km of the drainage (A d = 815 km 2 ). Vascular plant species were identified in each transect with observations on distance from channel, cover area, frequency of occurrence, origin, canopy composition, and light intensity. The majority of survey stations were located on incised floodplain surfaces characterized by riparian tree cover, silty-clay loams, and slopes less than 10%. Survey results are summarized as follows: No. of Adventive Species = 55; No. of Native Species = 75; Adventive Cover = 26.7%; Native Cover = 12.8%; Native:Adventive Ratio = 2.1. The two most common adventives are Rubus armeniacus (Himalayan blackberry) and Phalaris arundinacea (Reed canarygrass). Polygonum cuspidatum (Japanese knotweed) has limited frequency, but ranks in the 95th percentile of total invasive area. The most abundant native species include Rubus leucodermis (Blackcap), Symphoricarpos albus (Snowberry), Urtica dioca (Stinging nettle), and Polystichum munitum (Sword fern). Distribution analysis provides a framework for positing mechanisms of adventive plant dispersion. Longitudinally, R. armeniacus and P. arundinacea are ubiquitously distributed throughout the lower watershed, while P. cuspidatum is restricted to upper reaches. Transverse to the floodplain, P. cuspidatum is limited in occurrence to less than 30 m from the channel, while P. arundinacea and R. armeniacus are common throughout. Results suggest that hydrochory is the primary dispersal mechanism for the former two species, while mixed modes apply to the latter. A combination of geomorphic (flooding) and anthropogenic disturbance (timber harvesting) processes result in substrate alteration and canopy gaps, thus diminishing barriers to exotic plant colonization. Geological Society of America Fall 2007 Meeting, Denver, Colorado Session No. 53 Geomorphology (Posters) Riparian Plant Distribution in the Luckiamute River Basin, Central Oregon Coast Range: Preliminary Analysis of Geomorphic and Anthropogenic Controls on Adventive Species Propagation in an Unregulated Watershed Taylor, Stephen B. 1, Dutton, Bryan E. 2, Noll, Katherine 1, and Pirot, Rachel 3, (1) Earth and Physical Science Dept, Western Oregon University, Monmouth, OR 97361, taylors@wou.edu, (2) Biology Dept, Western Oregon University, Monmouth, OR 97361, (3) Dept. of Geology, Portland State University, Portland, OR 97207 2. INTRODUCTION Invasive plant species in western Oregon are a pervasive problem that disrupt native habitats and create annual economic losses of millions of dollars for public and private landowners (Oregon Department of Agriculture, 2001). Nationwide, the United States experiences annual losses of over $130,000,000.00 due to non-native species (Pimentel and others, 2000). Vegetative disturbance of natural ecosystems by geomorphic and anthropogenic processes affect soil substrate conditions, nutrient availability, canopy shading (solar influx), and riparian hydrology. The most abundant concentrations of invasive species are typically associated with disturbed zones that have been altered by human activity. As such, disturbed zones on the landscape act as primary conduits for the dispersal of non-native species (Pabst and Spies, 1998). Understanding the controls on spatial distribution of invasive plants in the context of disturbance regime is critical for designing effective watershed conservation and restoration plans. The purpose of this research was to conduct a reconnaissance survey to delineate associations between geomorphic and anthropogenic disturbance regimes, and distribution patterns of invasive plant species in the Luckiamute Watershed of western Oregon (Figures 1 and 2). 4. METHODS Riparian plant surveys were conducted at 20 sites using a 1-m 2 sampling grid along survey lines oriented transverse to the active channel. Field and analytical procedures following those prescribed by Elzinga and others (1998) (Figures 4 and 5). Sample sites were confined to wooded riparian zones within a 100-m buffer along the 100-year floodplain of the channel system. Final site selection was determined on the basis of logistical access, property owner permission, and position in canopy-covered riparian zone. All understory and overstory species were identified along the transects and light measurements were collected in the 400 to 700 nm wavelength range using a Quantum light meter. GPS positions and general geomorphic observations were recorded as well. 5. RESULTS 5 A. Understory Vegetation Over 1700 m 2 of riparian understory was surveyed using 1- by 100-m transects oriented perpendicular to the active channel, with 20 survey stations irregularly spaced (D avg = 5.1 km) along the lower 100 km of the drainage (Ad = 815 km 2 ). 170 vascular plant species were identified in the understory (Table 1). The majority of survey stations were located on incised floodplain surfaces characterized by riparian tree cover, silty-clay loams, and slopes less than 10%. Survey results are summarized as follows: No. of Adventive Species = 55; No. of Native Species = 75; Adventive Cover = 26.7%; Native Cover = 12.8%; Native:Adventive Ratio = 2.1 (Table 2). The two most common adventives are Rubus armeniacus (Himalayan blackberry) and Phalaris arundinacea (Reed canarygrass). Polygonum cuspidatum (Japanese knotweed) has limited frequency, but ranks in the 95th percentile of total invasive area (Figure 6). The most abundant native species include Rubus leucodermis (Blackcap), Symphoricarpos albus (Snowberry), Urtica dioca (Stinging nettle), and Polystichum munitum (Sword fern) (Figure 7). 3. PHYSIOGRAPHIC SETTING 3 A. Geology and Geomorphology The Luckiamute River comprises a portion of the Willamette basin in west-central Oregon (Figure 1). This seventh-order watershed (sensu Strahler, 1957) drains eastward from the Coast Range into the Willamette River and occupies a total drainage area of 815 km 2. Land surface elevations range from 46 m (150 ft) at the confluence with the Willamette River to 1016 m (3333 ft) at Fanno Peak. The Luckiamute has an average gradient of 3 m/km, a total stream length of 90.7 km, and an average basin elevation of 277 m (910 ft) (Rhea, 1993; Slack and others, 1993). Lithostratigraphic units are grouped into four spatial domains in the Luckiamute, these include the Siletz River Volcanics domain (south), the Tyee domain (west-southwest), the Yamhill-Intrusive domain (north- northwest), and the Spencer-Valley Fill domain (east). Geomorphic systems are divided into a valley-floor regime to the east and hillslope-colluvial regime to the west (Figure 3). Hillslope landforms and colluvial processes dominate the Siletz River, Tyee, and Yamhill-Intrusive domains, while fluvial landforms and alluvial processes are characteristic of the Spencer-Valley Fill domain. The lower Luckiamute is characterized by a mix of alluvial stratigraphic units and geomorphic surfaces. Landforms include active channels, floodplains, fill terraces, and strath-pediment surfaces (McDowell, 1991). In addition to these fluvial landforms, the lower Luckiamute is also associated with swaths of low-relief colluvial hillslopes supported by the Spencer Formation (Figure 3). Pleistocene through Holocene terrace development records a complex history of base level fluctuation, internal erosion- deposition cycles, and glacial-outburst floods (Missoula Floods) from the Columbia River system. The active channel of the lower Luckiamute is incised 8 to 9 m below the floodplain, with higher level terrace surfaces at 12 to 15 m above mean annual stage (Reckendorf, 1993). The higher-level terrace surfaces are covered with rhymically-bedded, silty slack-water deposits of the Willamette Formation (Missoula Flood deposits; 13.5-12 Ka). These late Pleistocene surfaces are inset with lower terrace and floodplain deposits that are predominantly Holocene in age (post-Missoula Flood; <12 Ka) (Figure 3; O'Connor and others, 2001). 3 B. Vegetation and Land Use The Coast Range portion of the Luckiamute watershed lies in the Tsuga heterophylla Zone of Franklin and Dyrness (1988). Dominant forest species include Pseudotsuga menziesii (Douglas fir), Tsuga heterophylla (western hemlock), and Thuja plicata (western red cedar), with lesser occurrence of Abies grandis (grand fir). These species formed part of the classic old growth timber stands that were logged extensively in the Pacific Northwest during the early 1900's. Lower reaches of the Luckiamute watershed lie in agricultural crop and pasture land, with local patches of mixed Quercus garryana (Oregon white oak) and urban mosaic species. Since European settlement, the dominant economic activities in the Willamette Valley have centered on agriculture in the lowlands and timber harvesting in upland forests. Over the past several decades, industrialization and rapid population growth have resulted in significant impact to the habitat of the region. A large portion of the upper Luckiamute is owned by private timber companies and 67% of the watershed classified as forest. In contrast, the eastern valley section is comprised of a mix of agricultural lands (15% of total), native vegetation (3%), and urban development (1%) (Urich and Wentz, 1999). Primary commodities in the agricultural zones include grass seed, wheat, hay, oats, and mixed crops (clover, sweet corn, mint, alfalfa, filberts) (Wentz and others, 1998). From Luckiamute Watershed Council Figure 1. Location map of the Luckiamute Watershed, western Oregon. Figure 2. A. Oblique aerial photograph overlooking a portion of the Luckiamute Basin. View is to the west towards Coast Range. Note agricultural landuse in lowlands and forestry management in the uplands. B. Photo of the main stem of the Luckiamute River channel at bankfull stage Luckiamute River at Helmick State Park; 3800 cfs on March 27, 2005. Note riparian vegetation along channel-margin floodplains, the focus of this study. Figure 3. Generalized geomorphic map of the Luckiamute Watershed (after O’Connor and others, 2001). Figure 5. Photos showing 1-m by 100-m quadrat methodology used in plant surveys. Figure 4. Locations of plant-survey transects in the Luckiamute Watershed. Table 1. Summary of plant species encountered in the riparian understory, Luckiamute Watershed. AB C Figure 6. Field photos showing the most common invasive plant cover (by area) encountered in transects. A. Rubus armeniacus (Himalayan blackberry). B. Phalaris arundinacea (Reed canarygrass); C. Polygonum cuspidatum (Japanese knotweed). Western Oregon University A B

2. 5 B. Invasive Plant Distribution Distribution analysis provides a framework for positing mechanisms of adventive plant dispersion. Two spatial variables are considered: (1) longitudinal distribution of invasives along the channel network, and (2) transverse distribution across the riparian zone, perpendicular to the active channel. Figure 8 is a distribution map showing the occurrence of the top three invasive species in the watershed. Quadrat data are re-plotted in Figure 9 according to distance upstream from the river mouth (km) and transverse distance across the floodplain (m). Longitudinally, Rubus armeniacus (Himalayan blackberry) and Phalarus arundinacea (Reed canarygrass) are ubiquitously distributed throughout the lower watershed, while Polygonum cuspidatum (Japanese knotweed) is Figure 7. Field photos showing the most common native plant cover (by area) encountered in survey transects. A. Rubus leucodermis (Blackcap); B. Symphoricarpos albus (Snowberry); C. Urtica dioca (gracilis) (Stinging nettle); D. Polystichum munitum (Sword fern). Refer to Table 2 for summary of invasive and native plant distribution statistics in the Luckiamute Watershed. AB CD Table 2. Plant survey summary statistics, Luckiamute Watershed. 7. CONCLUSION Annual economic losses and habitat degradation by invasive plant species in the United States are well documented. Understanding the controls on spatial distribution of invasives is critical for designing effective watershed conservation and restoration plans. To this end, this study: (1) provides important baseline data on the dispersion of non-native invasive plant species in western Oregon, (2) provides a framework for discerning patterns of invasion, (3) provides a methodology for assessing susceptibility of land areas to invasion based on select comparator species, and (4) contributes to a more thorough understanding of the basic biology of a adventive plant species in the region. The results of this preliminary work will form the basis of more extensive studies in the region and have potential use for development of larger scale predictive models of invasive plant dispersion. 8. ACKNOWLEDGMENTS This project was generously funded by the The Oregon Community Foundation and the Western Oregon University Faculty Development Fund. The authors also thank the following research assistants for their diligent work over the past several years: Daniel Asakawa, Catherine Drury, Moriah LaChapell-Shalock, Benjamin Purkerson, Shannon Wineland. Gratitude is expressed to the numerous landowners that granted property access. Elzinga, C.L., Salzer, D.W., and Willoughby, J.W., 1998, Measuring and monitoring plant populations: Bureau of Land Management Technical Reference 1730-1, Denver, Colorado, 477 p. Franklin, J.F., and Dyrness, C.T., 1988, Vegetation of Oregon and Washington, second edition: Oregon State University Press, Corvallis, Oregon, 216 p. Merritt, D.M., and Wohl, E.E., 2006, Plant dispersal along rivers fragmented by dams: River Research and Applications, v. 22, p. 1-26. 9. SELECTED REFERENCES Pabst, R.J., and Spies, T.A., 1998, Distribution of herbs and shrubs in relation to landform and canopy cover in riparian forests of coastal Oregon: Canadian Journal of Botany, v. 76, p. 298-315. Pimentel, D., Lach, L., Zuniga, R., and Morrison, D., 2000, Environmental and economic costs of nonindigenous species in the United States: BioScience, v. 50, p. 53-65. Swanson, F.J., Franklin, J.F., and Sedell, J.R., 1990, Landscape patterns, disturbance, and management in the Pacific Northwest, USA, Changing landscapes: Springer-Verlag, p. 191- 213. 6. Discussion Merritt and Wohl (2006) provide an excellent summary of the mechanisms by which plants are dispersed along river corridors. Seeds and plant fragments may be transported by four fundamental mechanisms: (1) anemochore (wind dispersion), (2) barochore (gravity dispersion), (3) hydrochore (water dispersion), and (4) zoochore (animal transport, including humans). The mechanism of dispersal is species dependent and a function of numerous variables related to the life history traits of the plant in question. The spatial distribution of adventive species in the Luckiamute watershed in part reflect the life history characteristics and in part geomorphic and anthropogenic processes occurring in the basin. As discussed by Merrit and Wohl (2006), hydrochoric plants are typically limited in occurrence to within several meters of the stream channel, while wind and animal dispersion tend to disperse species more ubiquitously throughout the riparian zone and along the riparian corridor. Western Oregon University Figure 8. Map showing spatial distribution of the top three invasive cover plants in the riparian zone of the Luckiamute basin. Figure 10A. Plot of Himalayan blackberry (Rubus armeniacus) cover along survey traverses perpendicular to the active channel. Data are tallied from all 1-m 2 quadrats at the 20 survey locations shown in Figure 4. Figure 9. Plot of dominant invasive plant species distribution according to transverse distance across the riparian zone (m, X axis) and distance in the channel system upstream from the mouth of the basin (km, Y axis). Symbols are the same as those used in Figure 7. Figure 10B. Plot of Reed canarygrass (Phalaris arundinacea) cover along survey traverses perpendicular to the active channel. Data are tallied from all 1-m 2 quadrats at the 20 survey locations shown in Figure 4. Figure 10C. Plot of Japanese knotweed (Polygonum cuspidatum) cover along survey traverses perpendicular to the active channel. Data are tallied from all 1-m 2 quadrats at the 20 survey locations shown in Figure 4. Figure 11. Plot of species richness (total no. of invasive and native understory species per m 2 ) vs. percent cover of the top three invasives (Himalayan blackberry, Reed canarygrass, Japanese knotweed). The upper envelope line defines the association between the limit of species richness by percent cover of invasives. Data are tallied from all 1-m 2 quadrats at the 20 survey locations shown in Figure 4. restricted to upper reaches. Transverse to the floodplain, Polygonum cuspidatum (Japanese knotweed) is limited in occurrence to less than 30 m from the channel, while Phalarus arundinacea (Reed canarygrass) and Rubus armeniacus (Himalayan blackberry) are common throughout. It is clear from the survey data that Rubus armeniacus (Himalayan blackberry) is a generalist that is ubiquitously distributed in all portions of the watershed system. Phalarus arundinacea (Reed canarygrass) is more extensively distributed across the floodplain in downstream reaches, and becomes restricted nearer the channel in upstream sectors. Figures 10A, 10B, and 10C further demonstrate these transverse spatial patterns. (Discussion – Cont.) While preliminary in nature, results from the Luckiamute study suggest that hydrochory is the primary dispersal mechanism for Polygonum cuspidatum (Japanese knotweed), while mixed modes apply to Rubus armeniacus (Himalayan blackberry) and Phalarus arundinacea (Reed canarygrass). The working hypothesis is that a combination of geomorphic (flooding) and anthropogenic disturbance (timber harvesting) processes result in substrate alteration and canopy gaps, thus diminishing barriers to exotic plant colonization. Adventive plant species are problematic for both native and agricultural plant communities as they can compete for resources and displace competitors. Local extirpation of native plant species has obvious impacts on wildlife and natural habitats. Competition between plant species is a part of any habitat, but introduction of non-native species disrupts relationships evolved among native plants and their communities within those specific habitats. The ecological impacts of adventive vegetation on species diversity in the Luckiamute is notably demonstrated in Figure 11. This plot relates species richness (total no. of invasive and native understory species per m 2 ) to percent cover of the top three invasives (Himalayan blackberry, Reed canarygrass, Japanese knotweed). The upper envelope defines the limiting association of species richness by percent cover of invasives.