Research

Research CV Publications People Webinars Links SEED ZONES Contact

Currently I am the Coordinator for the USDA Northwest Climate Hub serving farmers, ranchers, forest land owners and natural resource managers in Alaska, Idaho, Oregon and Washington manage their working landscapes in the face of the many challenges due to climate change.

My research focuses on understanding how the environment and other species affect wild plant populations. I study wild plants to comprehend the factors that affect how plant populations grow and change traits and life history events over time. I am interested in how microorganisms affect plant populations, the role of maternal effects in trait evolution, and how mating patterns affect trait evolution and maintain diversity in natural populations. I use this information to guide restoration efforts in the Interior Mountain West.

As a Research Geneticist (post-doc) at the USDA Forest Service, Pacific Northwest Research Station in Corvallis, OR. To guide post-fire restoration efforts in grasslands in the Interior Northwest, I am investigating the efficacy of seed-zones via a reciprocal transplant experiment with Bluebunch wheatgrass (Psuedoroegneria spicata). This project is in collaboration with J. Brad St. Clair of the Pacific Northwest Research Station and members of the Rocky Mountain Research Station in Boise, ID including Nancy Shaw and Francis Kilkenny.

As a Postdoctoral Research Associate collaborating with Laura F. Galloway in the Department of Biology at the University of Virginia, I focused on the American Bellflower (Campanulastrum americanum) to investigate the role of maternal effects in population adaptation. For my dissertation work, I examined the effects of virus on wild squash (Cucurbita pepo) populations and the ecological effects of transgenic squash (Cucurbita pepo) with virus-resistance. When investigating the effects of virus on wild plant populations, my research found that virus does not always have detrimental effects on population growth rate of wild plants.

Projects

Project descriptions

Effectiveness of seed zones for bluebunch wheatgrass (photos)

Populations are often adapted to local environments. For instance, populations of bluebunch wheatgrass (Pseudoroegneria spicata), an ecologically important restoration species differ in traits important for adaptation to precipitation and temperature. To yield successful outcomes in restoration efforts the use of plant material from areas similar to the restoration site is essential. Thus, seed transfer zones have been developed to guide restoration efforts. With J. Brad St. Clair and Francis F. Kilkenny we are testing the efficacy of seed zones specific to bluebunch wheatgrass for the Interior Northwest to ensure successful establishment and allow for long-term adaptation by maintaining genetic diversity. We are comparing differences in establishment, survival, and reproduction of bluebunch wheatgrass populations from local seed zones compared to non-local seed zones. We hypothesize that in the long-term populations from local seed zones will better establish, survive, and reproduce. To test this hypothesis, we established a reciprocal transplant study in two broad regions (i.e., transects) each with eight seed zones and each seed zone represented by four to five wild populations. Eight common gardens in each transect were established in fall 2014 (PHOTOS) with each garden representing the climate of the local seed zone. Each common garden was planted with bluebunch wheatgrass from each seed zone within a region (transect). In addition to bluebunch wheatgrass populations from each seed zone, one widely used variety and one selected germplasm were included at each common garden. Here is the seed zone map for bluebunch wheatgrass noting our experimental gardens, wild population site collections, level III ecoregions, and state boundaries (Seed zone map).

Seed zone map

Results of this work will help land managers to determine sources of bluebunch wheatgrass populations for post-fire restoration. Furthermore, this study will explore the consequences of changing climates for adaptation by substituting space for time to evaluate different populations in different climates. Long-term productivity and adaptation will be modeled to allow evaluation of trade-offs between different management options for current and future climates. Population movement guidelines and associated seed zones can be adjusted based on results from this study and management objectives.

Click here for additional information on seed zones.

For more information on using seed zones check out USFS Science Findings March 2015 or to find seed zones in the US.

Effects of virus on life history traits and population dynamics

Viruses are ubiquitous, but it is unclear how viruses affect wild plant populations. Recent studies have demonstrated that viruses are not consistently detrimental to plants. Many studies have focused on the effects of viruses on a single trait, such as fecundity, and assumed changes in that trait scale-up to affect the population. However, it is not always apparent if fecundity is the best trait to focus on as minute changes in an important trait to the life cycle of an organism can be overshadowed by large changes in less important traits. To understand how viruses affect a population, it is essential to evaluate traits across a life cycle. One way to integrate a broad collection of traits is with matrix models, which project population growth and dynamics.

In this study, I used data from common garden experiments with plants from three wild populations of Cucurbita pepo ssp. texana or ozarkana inoculated with one of three virus treatments. These data parameterized a deterministic matrix model that allowed us to estimate the effects of virus on components of fitness and population growth rate. Virus did not reduce fruit number, but population growth rates varied among virus treatments and wild C. pepo populations. The effect of virus on population growth rate was not always detrimental and it depended on the virus species and the wild C. pepo population. Contributions of life history transitions and traits to population growth rates varied among populations and virus treatments. However, this population-virus interaction was not evident when examining individual components of fitness. Thus, caution must be used when interpreting the effects of changes in individual traits, as single traits do not always accurately predict population-level change (Prendeville et al. 2014).

In a wild population, a novel trait can reduce the effects of limiting factors and consequently lead to range expansion or enhance competitive ability. For this reason, it is important to understand the ecological risks associated with using transgenic crops (also known as genetically modified crops) and the effects of transgenes in wild plant populations. An ecological risk of using transgenic crops is movement of transgene into wild populations via hybridization followed by introgression (Pilson and Prendeville 2004). Since virus occurs in wild C. pepo populations (Prendeville et al. 2012) and can affect population growth, then natural selection may favor a virus-resistant transgene in wild C. pepo populations. Using a field experiment, I parameterized a mathematical model to determine the effects of virus on wild C. pepo growth rate in experimental populations with and without transgenic virus-resistance. I found virus-resistant transgenes in wild C. pepo can increase population growth rate, but it depends upon the virus species and wild plant population.

Population growth rates of three wild C. pepo populations grown in a common garden experiment to estimate population growth with three virus treatments.

Effects of assortative mating on resistance evolution

Many models of character evolution assume random mating, although assortative mating is common. Assortative mating occurs when individuals with similar phenotypes are more likely to mate with one another than under random mating. Moreover, if the phenotype causing assortative mating is genetically determined, then such mating leads to an increase in genetic variance for the trait, which makes the trait more responsive to natural selection. Thus in a population with assortative mating, a trait favored by selection may be fixed more rapidly than expected under random mating.

Using population genetics model and transgenic squash with virus-resistance (C. pepo) as a model system, I am investigating the effects of virus infection on plant mating patterns and the evolution of virus resistance. Though there is an overlap in flowering of susceptible and resistant plants, differences in pollinator preferences (Prendeville and Pilson 2009) can lead to assortative mating. However the degree of assortative mating is diminished as differences in sex ratio increase between virus susceptible and resistant plants. Results from these data indicate that variation in mating patterns is one mechanism that can maintain polymorphisms in natural plant populations.

Role of maternal effects in population adaptation

The phenotype of an organism is due to genetic and environmental factors of the organism as well as maternal effects. Maternal effects are the contributions of the maternal environment and maternal phenotype on the offspring phenotype. Maternal effects are ubiquitous across plant taxa. In the American Bellflower (Campanulastrum americanum), the initiation of flowering in the maternal plant affects the timing to germination in the offspring. The timing of germination in the American Bellflower determines the life history of an individual (i.e. annual or a biennial). This phenotypic plasticity allows a plant to respond to environmental heterogeneity and may allow for population adaptation to local environments. To investigate the role of maternal effects on population adaptation, Laura F. Galloway and I are 1) comparing the contributions of annuals and biennials to population growth rates among wild populations of American Bellflower along a range-wide transect, 2) determining the amount of variation in life history frequency among three common garden experiments over a latitudinal gradient, and 3) evaluating if selection on life history schedule acts on maternal traits, individual traits or both.

Distribution map of C. americanum indicated by grey-shaded area with populations noted by purple dots, bumble bee visiting C. americanum in a natural population, and biennial rosettes indicated by red swizzle sticks in a wild population of C. americanum.

Phenology

Phenology is the timing of life cycle events of an organism and changes in phenology can affect populations and communities. The phenology of many organisms is altering due to global climate change. However, it is unclear if all organisms or even populations within a species will have equivalent responses to novel environments. We have examined the phenology a native wild flower, American Bellflower (Campanulastrum americanum) to understand the relationship between local climates and population differentiation in phenology to anticipate responses to novel selective environments caused by climate change. The phenology of C. americanum is variable and leads to profound effects on individual life history schedules (i.e. resulting in annuals or biennials). We investigated populations of C. americanum grown under uniform conditions and found geographical clines with earlier germination and fruit maturation in northern populations (Prendeville et al. 2013). This fits the expectation of potential adaptive differentiation with earlier phenology with shorter growing seasons However, flowering phenology among populations was idiosyncratic, which suggests population differentiation in flowering phenology is due to local factors such as pollinator abundance or edaphic conditions. Thus, an increase in temperature due to climate change will not result in the evolution of earlier flowering in C. americanum. This pattern may be common as most studies investigating climate change and phenology have focused on spring ephemeral plants. Understanding this association between phenology and climate highlights that environmental variation may not result in similar responses among all species or their component of phenology.

Phenological variation among populations of C. americanum grown under similar conditions.

Spatiotemporal variation in deer browse and tolerance in a woodland herb (Ecology)

White-tailed deer are a significant herbivore in North America that have been broadly documented to affect plant reproductive success. If variation in the frequency and impact of herbivory by deer correlates with a broad-scale latitudinal gradient, climactic effects may be important for shaping plant-herbivore interactions. Alternatively, the lack of broad-scale gradients suggests local or regional factors such as plant community composition and deer densities are affecting herbivory. To investigate broad-scale patterns of deer herbivory, we examined the frequency and reproductive consequences of deer browse over three years in 17 populations of Campanulastrum americanum spanning the latitudinal extent of its range. Even though deer are overabundant throughout the range of C. americanum, we found variation in deer browse frequency (0-0.96, mean 0.46) and its effects on plant reproductive success (Prendeville et al. 2015). The four southernmost populations experienced high levels of herbivory, and were responsible for generating a negative relationship between latitude and herbivory. In general, variation in the frequency and impact of herbivory across the entire latitudinal gradient pointed to the importance of local rather than broad-scale factors. Within a population, deer ate larger plants. Across many populations and years, average fitnesses of browsed and uneaten plants were similar, suggesting that plants are tolerant to browse. However, since large plants have greater reproductive success and are more likely to be browsed, tolerance may be influenced by plant size. When plant size was accounted for, most populations did not fully compensate for browsing. There was no relationship between browsing intensity and tolerance, suggesting that browsing may be too variable to consistently select for tolerance, or that increases in deer density are too recent for increased tolerance to evolve.