Since 1996 I have been investigating patterns of species diversity in arctic tundra plant communities in northern Alaska. Initially as a postdoctoral researcher working with Dr. Gus Shaver (Ecosystems Center, Marine Biological Laboratory) and since 1999 with additional collaborators (in particular Dr. Sarah Hobbie, University of Minnesota, and Dr. John Moore, Colorado State University), I have tested hypotheses related to the influence of landscape age (time since deglaciation), soil pH, nutrient and light availability, temperature, and herbivore activity on species diversity. Much of our understanding of terrestrial tundra ecosystems comes from experiments conducted in the most common vegetation type in northern Alaska: moist acidic tundra. Hobbie and I focused our work on this community and the second most common: moist non-acidic tundra, which occurs on landscapes that have been more recently deglaciated or receive loess deposition. My work and others (e.g., Gough et al. 2000b) has shown that soil pH is highly correlated with species diversity and particular combinations of species across landscape gradients in the region, and that moist non-acidic tundra has higher pH, greater species diversity, more graminoids and fewer shrub species than moist acidic tundra. Working on experimental plots maintained by Shaver as part of the Arctic LTER (Long-Term Ecological Research), we have tested several hypotheses related to the causes of these diversity differences as well as their implications for carbon cycling. The LTER plots are located near Toolik Lake, Alaska; the non-acidic site is located on a younger (11-25 ka) Itkillik II glacial surface while the acidic site is located on older (50-125 ka) Itkillik I glacial surface. To date we have found distinct differences between these two sites, including lower net primary production, plant biomass, litter and soil organic matter decomposition rates, and net nitrogen mineralization rates at the non-acidic tundra (Gough et al. 2000b; Hobbie & Gough 2002, 2004).
I am particularly interested in the controls over the distribution of a deciduous shrub, Betula nana, which is a common species at the acidic site and virtually absent in non-acidic tundra. In long-term experiments (Shaver et al. 2001), Betula becomes dominant in nutrient-amended plots and has a slower but similar increase in warmed plots. Thus predictions about changes in carbon cycling under warmer temperatures and increased nutrient mineralization for this tundra type suggest more carbon will be sequestered by this woody species in the future. In my experiments, Betula seeds sown at the non-acidic site germinated, and Betula seedlings and adults survived at the non-acidic site after transplantation (Gough 2006). Germination was greater in plots where non-acidic mosses had been removed, suggesting a biotic barrier to increased establishment in non-acidic tundra. After five years of added nutrients, woody biomass has not increased in the non-acidic community (Gough & Hobbie 2003) as much as in the acidic community, perhaps because of the lack of shrubs (Hobbie, Gough & Shaver 2005). Thus the initial species composition and shrub abundance (particularly of Betula) may lead to very different carbon cycling patterns in these two vegetation types under warmed conditions.
I am currently exploring how plant species diversity is linked to soil faunal food web structure as well as activity of mammalian herbivores with John Moore. We have conducted our research on LTER plots in moist acidic tussock tundra and dry heath tundra. To date we have determined that mammalian herbivore presence does have an effect on growth (Gough et al. 2007) and relative abundance of particular plant species, and in some cases this is exacerbated by increased soil nutrient availability. In particular, it appears that mammals (both voles and caribou) facilitate the productivity of the better competitors in both communities when soil nutrients are increased. At dry heath tundra, mammals decrease abundance of soil lichens as well as change abundance of certain vascular plant species (Gough et al. 2008). Moore and I are in the process of combining our plant and soil data into his soil food web models to better understand the linkages between these two components of these terrestrial ecosystems.
Collaborators: Sarah Hobbie, John Moore, Gus Shaver
This material is based upon work supported by the National Science Foundation under Grant Nos. 9902721, 0137832, 0425827. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.
Most of the plant communities that I have worked in have been dominated by clonal plants (e.g., New England salt marsh, Louisiana coastal marsh, arctic tundra). In 1999 I began a collaboration with Dr. Deborah Goldberg (University of Michigan) to investigate how two aspects of clonal growth, physiological integration among ramets and aggregation of ramets in space, affect competitive ability.
We tested hypotheses about the importance of these traits at both the individual and community level using sedge species native to calcareous fens in southeastern Michigan. To adequately understand the importance of integration for growth and survival, we conducted a garden integration experiment and found that a species with closely packed ramets (clumper) was more integrated than a species with spread out ramets (runner). We created community mesocosms of different plant densities in which we manipulated integration and aggregation (described in Gough et al. 2002a), and followed plant abundance in these mesocosms over several years. We have seen distinct differences in growth patterns between runners and clumpers over time, with runners initially colonizing the low density mesocosms faster, but clumpers proving to be better competitors over time (Goldberg et al. in prep). A related field experiment showed that effects of integration varied across species and that generalizing about the importance of integration within growth forms may be unsuccessful (Pauliukonis & Gough 2004). Our data were also used to parameterize a model of competition among clonal plant species (Wildova et al. 2007).
Collaborators: Deborah Goldberg, Chad Hershock, Radka Wildova
This material is based upon work supported by the National Science Foundation under Grant No. 9974284. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
Much of my work has involved applying a perturbation to a plant community (such as excluding herbivores, adding nutrients, increasing temperature) and monitoring how that community responds over time. As part of the LTER network I have been involved with a group seeking to examine the patterns between productivity and diversity in plant communities along natural and artificial productivity gradients (Waide et al. 1999, Mittelbach et al. 2001). We conducted a meta-analysis to look at these patterns at seven LTER sites, including the Arctic site, and found that the shape of the relationship depended on the scale at which it was examined (Gross et al. 2000), and that all sites showed similar magnitude of productivity increase with added nitrogen, but the declines in species richness that occurred were more variable (Gough et al. 2000a). Since then I have been involved with a larger group of LTER scientists to analyze a database of the individual plants that either increase or decrease in response to nitrogen additions to determine if certain traits are predictive of response across communities and ecosystems. We have published multiple papers resulting from this work (Collins et al. accepted, Cleland et al. 2008, Clark et al. 2007, Pennings et al. 2005, Suding et al. 2005), and continue to expand and analyze the database.
Collaborators: Katie Nash Suding, Elsa Cleland, Scott Collins, Kay Gross, Daniel Milchunas, Chris Clark, Steve Pennings, Joe Fargione
The Ft. Worth Nature Center and Refuge (FWNCR) is a 3600 acre city-owned reserve near Arlington that encompasses several different plant communities including limestone prairie, post oak savannah, and riparian bottomlands. I am currently seeking funding to examine questions related to clonal growth form and invasive species in remnant prairie communities to determine how plant demography contributes to native prairie species response to various management techniques (e.g., burning, bison grazing, clearing, etc.). I am working with the Natural Resources Manager at the refuge to develop a research program to continue investigating basic ecological questions and simultaneously provide her with data she needs to improve restoration efforts. I currently have a graduate student beginning a separate project at the site to examine the allelopathic effects of Johnson grass (Sorghum halepense) on native prairie communities. The refuge is an underutilized resource in North Texas and provides many opportunities for future ecological and restoration research in several different ecosystems.