Abstract​

​

     Brachypodium sylvaticum is an invasive Eurasian bunchgrass responsible for forming dense stands in coniferous and deciduous forests of Oregon’s Willamette Valley. Though B. sylvaticum can tolerate a range of conditions, its distribution is associated with disturbance and coniferous forests. To better comprehend the physical and ecological interactions between B. sylvaticum and forest leaf litter within the invaded range, an eight-week, open-air greenhouse experiment was conducted. This experiment assessed the effect of coniferous, mixed (50/50), and deciduous leaf litter depth on the emergence, establishment, survival, and growth of B. sylvaticum seedlings using these three litter types combined with different litter depths—high (5.6 cm), low (2.3 cm), and no litter (bare). In addition, seeds from six B. sylvaticum populations from Oregon were used.

      Coniferous litter facilitated the highest seedling emergence and establishment. Seedlings were more likely to emerge and establish in low litter depth than they were in high litter. Survival was significantly different among treatments when litter type and depth were combined. Growth and invasion potential, measured by biomass and total tillers, were greater in the coniferous litter, especially in the high litter treatment. Treatments that held more moisture aided an increased seedling biomass. Seedlings in deciduous litter preformed poorly in all measured variables and mixed litter was between coniferous and deciduous litter. Variation in emergence among populations was affected by seed weight and population, while seed weight was positively correlated with larger growth and invasion potential.

     This experiment examined B. sylvaticum seed recruitment in habitats with an intact litter layer. Disturbance in these habitats breaks up forest litter and allows B. sylvaticum roots to access litter and soil moisture. The observations and results of this study suggest that secondary invasion might be more widespread in coniferous forests, whereas deciduous forest litter provides an unsuitable microenvironment for root growth and penetration and may have implications in its ability to act as a mulch barrier to further restrict B. sylvaticum seedling colonization, growth, and spread.

​

 

Introduction​​​



     Leaf litter has long been recognized as a determining factor in plant community composition and structure (Facelli & Pickett 1991). The effect of leaf litter on individual plants and entire ecosystems varies significantly and includes both positive and negative effects. Litter can facilitate plant growth by leaching nutrients (Facelli & Pickett 1991), stabilizing soil temperature (Evens & Young 1970), and retaining soil moisture (Fowler 1986). However, a meta-analysis by Xiong & Nilsson (1999) concluded that litter has an overall negative effect on vegetation. Litter has been shown to alter pH (Facelli & Pickett 1991), exude allelopathic chemicals, (Ahlgren & Ahlgren 1981) and block critical light levels needed for seed germination (Sydes & Grime 1981b). In addition, the litter layer forms a mechanical barrier that can affect seed germination and seedling establishment. Seeds near the top of the litter produce roots that may have difficulty penetrating the litter to reach the soil surface (Borchert et al. 1989, Fowler 1986). Similarly, a thick litter mat may hinder shoot penetration. Smaller seeded species, which have fewer energy reserves, often are less successful in breaking through the litter layer (Winn 1985). The geometry of the shoot and the density of the litter can exert an influence on seedling success (Grime 1979). Litter can also affect total plant growth. Sydes & Grime (1981a) found a negative correlation between litter weight and herbaceous shoot biomass in a deciduous woodland.

     The introduction of exotic plant species has been shown to alter ecosystem dynamics including resource availability, nutrient cycling, and local microclimate (D’Antonio & Vitousek 1992). Invasive plants usually lack natural predators in the invaded range and can be more efficient competitors, allowing them to displace native species. Many published reports document that repeated introductions into disturbed environments aid an invasive’s establishment in the new habitat (Lockwood et al. 2005). Disturbance can destroy existing vegetation, making resources more available to invaders.

     Brachypodium sylvaticum is a perennial bunchgrass that has recently invaded at least 10,000 ha of forest understory in Oregon’s Willamette Valley (Kaye 2003). This grass is capable of forming thick monocultures that outcompete native vegetation, especially within disturbed sites (Holmes et al. 2008, Rosenthal et al. 2008). The success of B. sylvaticum can be attributed to its long awns which aid vector transport via wind, herbivores, and humans; leaves that are unpalatable to predators; its ability to grow in full sun and shade conditions and in a diversity of soil types and nutrient conditions (Rachael Workman personal communication). Even though B. sylvaticum can establish and survive in a variety of environments, it is commonly found in large patches in shaded forest understory. Its preference for these habitats could be due to the larger quantity of organic matter and nutrients typical under forest canopies (Bashan et al. 2000).

     Seedling density of B. sylvaticum in the invaded range is mostly driven by high propagule pressure—the repeated introductions needed for establishment (Taylor & Cruzan 2011). Taylor & Cruzan (2011) examined the role of disturbance, habitat, and leaf litter on B. sylvaticum seedling abundance and concluded that disturbed sites had a shallower and a less dense litter layer. These sites were more prone to primary invasion by this highly invasive grass, likely because the exposed soil facilitated seedling establishment. Moreover, B. sylvaticum was more abundant in coniferous forests than in deciduous forests. A separate study examined the effect of litter depth and coarseness (fine or course) on culm density of B. sylvaticum at Milo McIver State Park, Oregon (Capstone 2012). The researchers concluded that litter depth was negatively correlated with culm density—disturbance being the most likely reason behind the change in litter depth. B. sylvaticum was also found in higher densities in finer litter (coniferous litter).

     Litter depth is an important component in the early life and growth of seedlings, because a seed’s emergence success depends on its position within the litter (Facelli & Pickett 1991). Differences in litter texture may have a similar effect. Seeds germinating near the top are exposed to different challenges than seeds that germinate in other litter depths. The consequences include a higher probability of drying out, differential access to moisture, and increased distance of root penetration. Combinations of these factors will affect how fast and large a plant will grow. In sum, we perceive litter type and depth to play an important role in the growth and spread B. sylvaticum, but without a controlled experiment to separate the effects of litter and other ecological variables such as soil moisture and light levels, we can not be confident that litter type and depth are solitary contributing factors to the establishment, growth, and spread of this invasive grass.

     To my knowledge there have not been any published studies that have focused on the interaction between litter type and the success of B. sylvaticum. The purpose of this study was to investigate the effects leaf litter composition and depth on the early growth of the invasive bunchgrass, B. sylvaticum, under manipulative, open-air, greenhouse conditions. The experiment simulated the effects of B. sylvaticum colonization into various, intact forest communities. As a result, the seeds were sowed in the upper part of litter layer. Three different litter types were used (coniferous, deciduous, and a mix of coniferous and deciduous) each with a no litter, low, and high litter depth all nested in six B. sylvaticum populations. I hypothesized that 1) coniferous litter would have the highest emergence and establishment (after the no litter treatment) however, due to deciduous’ litter insulating effect and contrary to the results of Taylor & Cruzan, deciduous litter would allow the highest seedling survival. I also hypothesized that 2) high litter will not affect emergence (seeds will be placed near the top of the litter), but will have lower establishment and the highest survival. In addition to emergence, establishment, and survival, I tested the hypothesis that 3) low coniferous litter will allow the largest shoot biomass and facilitate the highest invasion potential (Ip). Increased soil-litter moisture was hypothesized to positively affect shoot biomass. Population differences in emergence, establishment, survival, biomass, and Ip were also compared.



​