- Earth is experiencing unprecedented biodiversity loss. Amphibians are at the forefront of this biodiversity loss, with species declines estimated to be more severe than those of birds and mammals. Amphibian population declines and extinctions are driven by a number of factors including climate change, habitat destruction, contaminants and disease but are often due to a combination of factors rather than any acting singularly. The fungal pathogen, Batrachochytrium dendrobatidis (hereafter, Bd) and a strain of Ranavirus, frog virus 3 (FV3) belonging to the family Iridoviridae, are considered important drivers of amphibian declines worldwide and are therefore of conservation concern. Despite the ecological importance of these two pathogens, we lack inventory and long-term monitoring to properly address pathogen-driven amphibian declines and an understanding of the role of each pathogen in a community context. Thus, this dissertation employs a combination of manipulative experiments and field observations to explore the effects of amphibian pathogens on aquatic communities and study the ability to detect amphibians and their pathogens using molecular-based tools.
Research on the amphibian-Bd system has primarily focused on the direct effects of the pathogen on its hosts, while few studies have investigated how the fungus affects dynamics of the aquatic community. In Chapter 2, I experimentally manipulated the presence of Bd and western toad (Anaxyrus boreas) larvae in outdoor mesocosms to explore the interactions among primary producers, zooplankton and Bd-exposed or unexposed larvae. Bd infections can cause larval mortality in some species, thus indirectly increasing periphyton biomass (a density-mediated indirect effect). However, Bd infections, while not always lethal, can cause oral deformities in amphibian larvae, reducing feeding rates and indirectly increasing periphyton biomass (a trait-mediated indirect effect). We found that Bd infections reduced amphibian larval densities, but this only weakly benefitted periphyton. After controlling for larval densities, mesocosms with Bd-exposed larvae had significantly more periphyton than mesocosms with unexposed larvae. These results are consistent with the trait-mediated indirect effect of Bd-induced mouthpart damage reducing larval feeding rates on attached algae. These results provide evidence that Bd has the potential to cause both trait- and density- mediated indirect effects that can alter community composition.
Although the number of amphibian populations experiencing disease-related declines and extinctions is increasing as studies accrue, we may actually be underestimating population declines given our lack of species inventory and long-term monitoring. Freshwater ecosystems, home to many amphibians, are considered the most imperiled habitats on Earth, thus are considered a conservation priority. However, traditional aquatic surveys present a variety of challenges as they can be costly, cause ecological disturbance or ineffective. Additionally, species with short survey windows or small, patchy distributions may be overlooked using traditional methods. The collection of environmental DNA (eDNA), genetic material released into the aquatic environment from the organism, has recently become an important biodiversity assessment tool. Although the use of eDNA surveys has promise, there are aspects of the technology that are poorly understood and warrant further investigation before field application. Interpretation of eDNA results for surveys of aquatic species occupancy, abundance or biomass is limited by our lack of knowledge regarding the drivers of eDNA production, diffusion, transport, persistence and distribution in aquatic systems. In Chapter 3, I experimentally manipulated the presence of western toad (A. boreas) eggs, larvae and juvenile amphibians in plastic containers, sampled for eDNA, then compared DNA concentrations across life stages and across time. The concentration of DNA was greatest in containers with amphibian eggs and larvae and lowest in containers with juveniles. Degradation of eDNA was exponential with less than twenty-five percent detection on day twenty-one across treatments. This experiment is the first to test eDNA detectability across amphibian ontogeny. These results highlight the importance of understanding the phenology of the target species for maximizing detectability by timing sampling efforts thoughtfully.
Therefore, in Chapter 4, I compared detection of two amphibian species, western toads (A. boreas) and wood frogs (Rana sylvatica) and two amphibian pathogens, Bd and Ranavirus using traditional visual-encounter surveys and eDNA sampling. If a site was occupied by A.boreas, detection probability was greatest using visual encounter surveys. However, if a site was occupied by R. sylvatica, Bd or Ranavirus, detection probability was greatest using eDNA sampling. The use of eDNA surveys for pathogens was particularly successful, likely because both pathogens can persist in the environment, thus could be underestimated by traditional sampling approaches. Our results illustrate the importance of pilot testing eDNA sampling for the specific species of interest as detection using eDNA compared to traditional survey methodologies differs across target organisms.
In Chapter 5, we present the first documentation of Bd and Ranavirus in the Chugach National Forest, Alaska, USA. Although this is the first documentation of Bd on the Chugach National Forest, Alaska, USA, it is not the first documentation of Bd in Alaska. In fact, several samples taken from wildlife refuges north of the Chugach National Forest tested positive for Bd. This is the first documentation Ranavirus in the state of Alaska. Although Ranavirus was recently documented at similar latitudes, our samples represent the detection the northernmost part of the pathogen’s range.