Prevalence and Seasonality of the Amphibian Chytrid Fungus Batrachochytrium dendrobatidis Along Widely Separated Longitudes Across the United States

The chytrid fungus Batrachochytrium dendrobatidis (Bd) has been implicated in amphibian declines on almost all continents. We report on prevalence and intensity of Bd in the United States amphibian populations across three longitudinally separated north-to-south transects conducted at 15 Department of Defense installations during two sampling periods (late-spring/early summer and mid to late summer). Such a standardized approach minimizes the effects of sampling and analytical bias, as well as human disturbance (by sampling restricted military bases), and therefore permits a cleaner interpretation of environmental variables known to affect chytrid dynamics such as season, temperature, rainfall, latitude, and longitude. Our prevalence of positive samples was 20.4% (137/670), and our mean intensity was 3.21 zoospore equivalents (SE = 1.03; range 0.001-103.59). Of the 28 amphibian species sampled, 15 tested positive. Three sites had no evidence of Bd infection; across the remaining 12 Bd-positive sites, neither infection prevalence nor intensity varied systematically. We found a more complicated pattern of Bd prevalence than anticipated. Early season samples showed no trend associated with increasing temperature and precipitation and decreasing (more southerly) latitudes; while in late season samples, the proportion of infected individuals decreased with increasing temperature and precipitation and decreasing latitudes. A similar pattern held for the east-west gradient, with the highest prevalence associated with more easterly/recently warmer sites in the early season then shifting to more westerly/recently cooler sites in the later season. Bd intensity across bases and sampling periods was comparatively low. Some of the trends in our data have been seen in previous studies, and our results offer further continental-level Bd sampling over which more concentrated local sampling efforts can be overlaid.