Count population viability analysis finds that interacting local and regional threats affect the viability of a rare plant

Ensuring the best use of limited conservation resources requires comprehensively assessing the relative importance of multiple threats, some of which occur at local and some at regional spatial scales. Multiple threats are rarely modeled in traditional population viability analyses (PVAs) due to the high data requirements necessary to parameterize age or stage based population models. Count based PVAs have been shown to provide robust results, and count data are readily available from many monitoring programs. Despite this, we are not aware of any studies that have used count based PVAs to assess multiple threats for plant populations. To demonstrate the utility of count based PVAs of assessing multiple treats at multiple spatial scales, we use long-term monitoring data by the Chicago Botanic Garden's Plants of Concern program to assess the main and interactive effects of two local threats (woody invasive species, browsing by deer) and one regional threat (climate change) on the viability of the rare forb, Eurybia furcata. We found an interaction between local and regional threats, which suggests that management actions targeting local threats can improve the viability of E. furcata populations both by directly reducing the risk of extinction and indirectly by decreasing this species vulnerability to climate change. Therefore, we recommended that land managers prioritize local scale management, specifically woody invasive species encroachment, to reduce this species' overall risk of extinction. The threat of climate change will act in concert with other anthropogenic factors, but conservation planning has historically focused on local scale threats. Adapting management to consider the regional threat of climate change requires threat analysis from multiple populations and at regional spatial scales. This task may seem daunting, but our results provide an optimistic outlook that count data can be effectively utilized for this purpose. Applying this approach widely to count based monitoring data already in existence would result in robust recommendations to land managers for many species of concern.