Megan K. Gahl

Date of Award


Level of Access Assigned by Author

Campus-Only Dissertation

Degree Name

Doctor of Philosophy (PhD)


Ecology and Environmental Sciences


Aram J.K. Calhoun

Second Committee Member

William E. Glanz

Third Committee Member

David E. Green


Amphibian mortality events in protected and relatively pristine settings have increased dramatically over the past two decades. We investigated larval die-off events in Acadia National Park (ANP), Maine, USA, to determine disease ecology and incidence on three scales: landscape, within a single breeding pond, and species-specific responses. For the landscape approach, we examined 60 amphibian breeding wetlands of varying disease history to determine landscape features specific to wetlands that sustained mortality events caused by ranavirus. We used logistic regression and model selection criterion (QAICc) to identify landscape features key to predicting ranavirus incidence. High catchment position was the most important landscape feature associated with larval mortality events. No spatial autocorrelation of disease events was observed, suggesting that within our study area, epizootics are not clustered and landscape and within-pond stressors may be more influential in ranavirus occurrence than vector movement. At the breeding pond scale, we used field measurements of biological, chemical, and physical stressors in wetland sites that were concurrently being monitored for die-off events to identify stressors associated with disease incidence that merit more detailed investigation. Few sublethal stressors were significantly associated with ranavirus or the Perkinsus-like organism in our statistical results. From our work, we suggest that for ranavirus, the Perkinsus-like organism, and Ichthyophonus, disease ecology is more important than environmental conditions, with the potential exception of aluminum and temperature stressors. However, for chytridiomycosis (Bd) and Saprolegnia, environmental conditions may be important in controlling and instigating outbreaks. To approach amphibian species level responses, we used three years of comprehensive disease screenings of free-living amphibians combined with in-field health screenings in 26 wetlands to determine potential amphibian reservoir hosts, vectors, and amplifying hosts for four emerging amphibian diseases. We confirmed five major amphibian diseases within ANP: ranavirus, chytridiomycosis (Bd), Ichthyophonus, Saprolegnia, and a Perkinsus-like organism, but did not confirm suspected occurrences of Riberoia. Ichthyophonus, Bd, and Saprolegnia were relatively benign in our study, and may have natural controls in ANP. Although some amphibian populations in ANP experienced extensive die-off events caused by ranavirus and the Perkinsus-like organism, disease events do not seem to exacerbate natural breeding population fluctuations.