Estimating the Impact of Disease in Species Threatened by Amphibian Chytrid Fungus: Comment on Murray et al.

Abstract

Chytridiomycosis is an emerging disease of amphibians caused by the chytrid fungus Batrachochytrium dendrobatidis (hereafter Bd) (Fisher et al. 2009). Chytridiomycosis is thought to be contributing significantly to global amphibian declines (Berger et al. 1998; Stuart et al. 2004; Skerratt et al. 2007). In the Neotropics the disease has spread in a wave-like fashion (Lips et al. 2008), and arrival at previously pathogen-free sites in this region results in mass mortalities and extinctions (Lips et al. 2008). Little is known, however, about the effects of Bd after population declines have occurred. In some places, Bd persists where it has invaded (Retallick et al. 2004). Thus, amphibians may evolve resistance to this novel pathogen such that the negative effects of Bd are attenuated over time. Murray et al. (2009) address this very important question. They used a multistate mark-recapture model to estimate survival of Bd-infected and Bd-free frogs (Litoria pearsoniana) in Queensland, where Bd has been present for 30 years. The use ofmark-recapturemethods is a clear improvement over earlier studies in which survival was estimated but did not account for imperfect detectability of animals in field studies (Faustino et al. 2004; Schmidt 2004). Murray et al. found that Bd causes a substantial reduction in survival. This matters because a reduction in adult survival can lead to amphibian population declines (Schmidt et al. 2005) and because it implies that Bd remains a dangerous pathogen, such that amphibian populations are unlikely to recover from population declines caused by Bd for many decades, if ever. Here, I point out that the apparent monthly survival estimates ofMurray et al. are likely confoundedwith perma- Paper submitted November 11, 2009; revised manuscript accepted December 14, 2009. nent emigration from the study site (Sandercock 2006). If infected and uninfected frogs have different emigration probabilities, then one would obtain different apparent survival estimates from the mark-recapture analysis for Bd-positive and Bd-negative frogs.