Impacts of environmental factors on anuran body temperature and repercussions for disease dynamics

Background/Question/Methods

Researchers have attributed variable patterns in amphibian declines from Batrochochytrium dendrobatidis (Bd) to differences in environmental factors (e.g., temperature) and species ecology (e.g., behavior).  Previous researchers have assumed that ambient temperature drives disease dynamics because anurans are small ectotherms that should conform to ambient temperature.  We tested whether ambient temperature is a good estimate of body temperature (T_b) in two families of Illinois anurans (Hylidae and Ranidae).  We also determined which environmental and biological factors affected anuran T_b in the field and determined whether T_b is correlated with Bd infection status.  We quantified T_b of 176 individuals of 7 species (Acris crepitans, Hyla versicolor / chrysoscelis complex, Pseudacris crucifer, Rana catesbeiana, R. pipiens, R. areolata and R. sphenocephala) from 9 locations throughout Illinois during the 2009 breeding season (15 March – 5 June).  For each capture we measured the following variables expected to influence T_b: substrate temperature (T_sub), substrate type (terrestrial or aquatic), Julian day, time of day, species, mass and sex.  We swabbed each frog and used qPCR to quantify the number of Bd zoospores on each frog.  We used an analysis of covariance (ANCOVA) to determine which variables explained the variation of T_b within Ranidae (n=63) and Hylidae (n=113). 

Results/Conclusions Hylid T_b was more variable than ambient temperature suggesting that large-scale estimates of ambient temperature do not capture the variation in T_b present within a population.  T_b ranged from 9.8 – 26.2ºC (average=18.7ºC) and the difference from T_sub ranged from -2.1 – 9.3ºC (average=1.2ºC).  Only T_sub affected T_b in Ranidae, while T_sub, substrate type, species, time, and mass affected T_b in the smaller hylid tree frogs.  T_b was not correlated with Bd status in Ranidae but was correlated in Hylidae because hylids had greater temperature variability.  Infected hylids were cooler than uninfected hylids and this pattern was driven by A. crepitans which had the largest sample size (n=74).  Within A. crepitans, males had a higher prevalence of Bd (87%) than females (34%) and males were significantly cooler than females relative to their T_sub. The difference in Bd prevalence between sexes may be explained by either behavioral differences (e.g., males spend more time at breeding sites and are potentially exposed to Bd for longer periods of time) or physiological differences (e.g., females can raise their T_b and clear Bd infection).  Field studies should account for this variation in host ecology, behavior, and physiology when evaluating disease susceptibility.