Factors influencing detection and co-detection of ranavirus and batrachochytrium dendrobatidis in midwestern North American anuran populations

Amphibian populations are in decline worldwide as they face a barrage of challenges, including infectious diseases caused by ranaviruses and the amphibian chytrid fungus Batrachochytrium dendrobatidis (Bd). Here we describe seasonal dynamics of Bd and ranavirus detection in free-ranging post-metamorphic wood frogs Lithobates sylvaticus, boreal chorus frogs Pseudacris maculata/triseriata, and gray treefrogs Hyla versicolor/chrysoscelis, sampled over a 3 season gradient in Minnesota (USA) wetlands. We detected Bd in 36% (n = 259) of individuals sampled in 3 wetlands in 2014, and 33% (n = 255) of individuals sampled in 8 wetlands in 2015. We also detected ranavirus in 60% and 18% of individuals sampled in 2014 and 2015, respectively. Ranavirus and Bd were detected concurrently in 26% and 2% of animals sampled in 2014 and 2015, respectively. We report clinical signs and associated infection status of sampled frogs; of the clinical signs observed, skin discoloration was significantly associated with ranavirus infection. Using generalized estimating equations, we found that species, season, wetland, and a species × season interaction term were significant predictors of Bd detection, whereas test year approached significance as a predictor of ranavirus detection. The odds of detecting both pathogens concurrently was significantly influenced by species, season, a species × season interaction term, year, and environmental ammonia. We propose an amphibian health monitoring scheme that couples population size surveys with seasonal molecular surveys of pathogen presence. This information is crucial to monitoring the health of remaining strongholds of healthy amphibian populations, as they face an uncertain future of further anthropogenic change.