Population and disease dynamics of the amphibian chytrid fungus in the stream-associated frog Litoria rheocola

Infectious diseases pose a major threat to global biodiversity. Chytridiomycosis is an amphibian disease that is caused by the pathogenic chytrid fungus, Batrachochytrium dendrobatidis (Bd). Bd has caused declines in hundreds of species of amphibians and extinctions of dozens. Some species that suffered local extinctions when the disease first emerged have reappeared and seem to be coexisting with the disease. Multiple processes can regulate changes in infection dynamics of a disease including physical and chemical characteristics of the environment and interactions among intra- and interspecific individuals. For example, the infection dynamics of Bd is heavily influenced by season and elevation and Bd can infect multiple hosts. Some hosts act as reservoirs, which are less susceptible to the pathogen, and enable the pathogen to persist in populations even when host populations are small. To fully understand the effects of a disease on a population, it is important to consider the interactions between multiple hosts, as well as the influences of environmental factors such as season and elevation. Chytridiomycosis has many reservoir hosts including animals that are not amphibians, and adult and larval amphibians. The interactions between infection dynamics of Bd in adult amphibians and their larval life stage are poorly understood. To better understand infection dynamics of Bd, I examined six populations (adult and tadpole) of Litoria rheocola, the common mistfrog, in northern Queensland, Australia. I studied changes in prevalence of infection over time in adults and tadpoles and determined how prevalence of Bd was affected by season and elevation. To quantify elevational influences of infection I surveyed populations of adult and tadpole L. rheocola at three site types: 1) high elevation (> 400 m above sea level (ASL)), 2) low elevation (< 400 ASL) sites connected by stream flow to high elevations (i.e., contiguous low elevation sites), and 3) low elevation sites that lacked connectivity to high elevations (i.e., non-contiguous low elevations). I tested all frogs and tadpoles captured to determine if they were infected with Bd. I marked adult frogs with visible implant elastomer tags to quantify population dynamics of frogs. I determined the probability of survival and recapture, and compared these estimates among seasons and site type. Infection dynamics of Bd in contiguous low elevation sites could be influenced by two, non-exclusive processes: (i) the flow of cool water from higher elevations maintaining cooler water temperatures, making the site more hospitable to Bd, and (ii) downstream transport of Bd zoospores from high elevation. Prevalence of Bd in tadpoles fluctuated seasonally, and was high in winter and low in summer. Prevalence of Bd was also influenced by site type: L. rheocola tadpoles at all low elevation sites had lower maximum prevalences than those at high elevation sites. There was a significant interaction between the effects of season and site type on the prevalence of Bd in tadpoles. Seasonal changes were more prominent at high elevation sites than at low elevation sites, and the patterns of seasonal change differed among site types. It is possible that being connected to a high elevation site greatly influenced the infection dynamics of Bd at contiguous low elevation sites due to the flow of cool water from high elevations and/or the flow of Bd zoospores downstream. In adult L. rheocola populations, both season and site type influenced prevalence of Bd. Prevalence of Bd was highest in winter and lowest in summer. One population, each at both the contiguous and non-contiguous low elevation sites, had prevalences of zero in summer; however, infections reappeared in autumn, strongly suggesting that reservoirs maintain Bd in these sites. In comparison, infection persisted throughout summer and winter in populations at high elevations. In adult frogs, contiguous and non-contiguous low elevation populations had similar Bd infection dynamics, suggesting that connectivity to high elevation sites did not have a direct effect on infection dynamics of Bd in adults. This contrasts with the effect of site type on the dynamics of Bd in tadpole populations, where site type had a strong influence on dynamics. With high prevalence of Bd in summer, tadpoles seem to be maintaining disease in adult populations at contiguous low elevation sites. In comparison, tadpoles at the non-contiguous low elevation site were not infected with Bd in summer. Therefore, tadpoles may not be an important reservoir for Bd at non-contiguous low elevation sites. Other species may be more effective reservoirs in these non-contiguous populations, as the disease persists in these areas, in spite of occasional apparent complete disappearance of Bd in adults. The probability of survival of adult frogs, estimated using program MARK, was not influenced by chytrid infection, but did differ among seasons and site type. Recapture probabilities were influenced by site type only. Rates of incidence of infection were influenced by season. Recovery rate remained constant at 80.3% across all site types: high, contiguous low, and non-contiguous low elevations. These results suggest that instances of individual mortality caused by Bd do not translate into overall low survival probabilities at the population level; thus it appears that in the populations I studied, disease-induced mortality is compensatory rather than additive. Even high elevation L. rheocola populations are now coexisting with Bd, despite strong evidence that the fungus caused local extirpations in the past. The coexistence of L. rheocola and Bd suggests that either L. rheocola populations have evolved increased resistance to chytridiomycosis or that recent environmental conditions (or other factors operating more recently, e.g., population density) have not favoured the development of outbreaks of fatal disease. I found that site type (elevation and contiguity with infected upland sites) influences disease dynamics of Bd, especially in tadpoles. In addition, my research strongly suggests that infection dynamics of tadpoles are influencing infection dynamics of Bd in co-occurring adult amphibian populations. My data on re-established populations suggests that the host-pathogen relationship has changed, either temporarily or more permanently, to favour the host. My study emphasizes the importance of simultaneously investigating multiple processes that could affect infection dynamics of a disease (e.g., transmission among multiple hosts and environmental characteristics, such as topography and season). By looking at the effects of season, elevation, and multiple hosts I get a better understanding of disease dynamics and the effects of disease on populations; looking at multiple processes will become important in future studies in order to fully understand disease dynamics