Gauging invasive plant impacts under a multi-stressor framework

Jeremy Collings, Department of Biology, University of Oregon, Eugene, OR and Andrea Dávalos, Department of Biological Sciences, SUNY Cortland, Cortland, NY

Background/Question/Methods
Though land managers spend considerable resources on managing invasive species, management is seldom paired with empirical monitoring strategies that track success. Additionally, management of invasive species rarely considers other co-occurring stressors that may be driving ecological change. By incorporating a holistic monitoring strategy that evaluates the role of multiple co-occurring stressors, land managers can assess the efficacy of current monitoring strategies for ecosystem recovery as well as target future management according to relative impacts of various stressors on the local community. We have partnered with the New York State Parks to establish a monitoring program assessing the efficacy of their management of the invasive vine, Vinetoxicum rossicum, while taking into consideration potential impacts of co-stressors (i.e. invasive earthworms and deer) on native vegetation. In order to identify an ideal monitoring program, we concurrently employed to approaches to monitoring: a community-level approach using vegetation surveys and a sentinel approach using transplanted native species. We employed both approaches in research areas that were either managed by Parks or left unmanaged. Within areas, we established plots exclosed with fencing and plots left open to assess the impact of deer. Finally, we conducted earthworm sampling in all plots to assess impacts of invasive earthworms.
Results/Conclusions
We found that plant community composition varied by management area and earthworm abundance. We also found that higher earthworm abundance was associated with lower native plant diversity. We identified no community-level impacts of deer after two years of fencing. Using the sentinel approach, we found that the response of transplants to the stressors was species specific. The one woody species, Fraxinus americana, was unaffected by any of the stressors. Survival of the forb Eurybia divaricata was negatively correlated with earthworm abundance but was not affected by V. rossicum management or deer exclusion. Survival of a second forb species, Solidago flexicaulis, was higher in fenced plots than in open plots in managed areas (significant management area x fencing interaction) but was not correlated with earthworm abundance. Inconsistencies among metrics used for monitoring suggest that the choice of monitoring approach affects conclusions and influence management decisions. Land managers’ decisions about monitoring should be made by considering goals of management, ease of implementation, and ease of interpretation. Additionally, results of our study support the claim that with any ecological change, there are winners and losers. Dynamics between drivers of change, winning species, and losing species may result in shifting community composition and diversity.