Epidemiological modelling of amphibian disease :

Department of Zoology and Entomology, The University of Queensland, St Lucia Brisbane, Queensland, Australia.

Email: hmccallum@zoology.uq.edu.au

Mathematical models are important in assessing the possible impact of pathogens such as chytrid fungus on frog populations. Results of simple models are sometimes counterintuitive: for example, pathogens that have the greatest impact on populations are not those that are most pathogenic to individuals. The interaction between rainforest frogs and chytrid fungus has several features that limit the direct application of standard epidemiological models. In particular: there are evidently reservoirs within which the pathogen has less effect than on the frogs that disappeared; the system consists of a number of loosely connected habitat patches; and the fungus has the ability to reproduce saprophytically. I present a series of simple models including these features, and discuss the implications of their results, both in terms of understanding Australian frog declines, and in identifying directions for future experimental and field research.