2018 ESA Annual Meeting (August 5 -- 10)

COS 7-4 - Identifying optimum rewilding strategies for restoring seed dispersal to a defaunated tropical island

Monday, August 6, 2018: 2:30 PM
R07, New Orleans Ernest N. Morial Convention Center
Hugo Thierry, Department of Geography, McGill University, Montreal, QC, Canada, Julie Savidge, Department of Fish, Wildlife and Conservation Biology, Colorado State University, Ft Collins, CO, Evan C. Fricke, Department of Ecology, Evolution, and Organismal Biology, Iowa State University, IA, Evan Rehm, Ecology, Evolution, and Marine Biology, University of California, Santa Barbara, Santa Barbara, CA, Henry Pollock, Natural Resources and Environmental Sciences, University of Illinois at Urbana-Champaign, Urbana, IL and Haldre S. Rogers, Ecology, Evolution, and Organismal Biology, Iowa State, Ames, IA
Background/Question/Methods

Global change and anthropogenic pressures have led to the reduction or disappearance of many species around the world. Some of these species have been associated with key ecological functions, and thus should be the focus of rewilding efforts to restore function. We use the island of Guam as our case study, which has seen all its native seed dispersers functionally or completely extirpated by the invasive brown treesnake. The absence of seed dispersers is linked to reductions in species diversity and changes in native forest structure, as well as slowed regeneration of degraded forest. Our goal is to identify possible management scenarios to restore the native forest by reintroducing seed dispersal throughout the island. In previous work, we identified two bird species which are effective seed dispersers: the Micronesian Starling and the Mariana Fruit Dove. The Micronesian starling is still present on Guam in very small populations, whereas the fruit dove has been extirpated. We then built a model integrating spatial data such as land cover, topography, forest quality, and land ownership, along with data on existing Micronesian starling populations and efficacy of snake control tools. Combined with an optimization algorithm, we explore rewilding scenarios with a goal of maximizing forest restoration.

Results/Conclusions

We first modeled the current status of seed dispersal services throughout the island, and identified large areas across the island where the forest is likely experiencing slow degradation as a result of not receiving seed dispersal services. We then selected several key areas for the reintroduction of starling and fruit dove populations throughout the island, using an optimization algorithm. These areas tend to be primarily in urban or semi-urban areas adjacent to intact forest. Since any bird reintroduction effort would require snake control, we also identified the optimal snake control method for each landcover type- traps were best for urban areas, fences for semi-urban or accessible forest areas, and toxicant drops for larger remote areas. The approach we conducted on the reintroduction of seed dispersers on Guam could serve as a conceptual framework for other case studies looking at the reintroduction of key species in order to restore ecological functions to ecosystems.