Hanging out with the cool frogs: Operative and body temperature effects on population response to disease

Background/Question/Methods Batrachochytrium dendrobatidis (Bd) is a fungal pathogen causing amphibian population declines.  Bd has a narrow thermal tolerance and requires moisture to survive.  Differences in frog biology, pathogen biology and/or environment may determine population response to disease.  Population responses to Bd vary among sites, habitats, species and populations (Brem & Lips, 2008).  In the tropics, stream dwelling species decline to a greater degree compared to forest species, yet not all stream species decline to extirpation and not all forest species survive.  We hypothesized that variation in operative temperature (T_e) and/or body temperature (T_b) might explain differences in host population change documented among sites, among habitats and among species.  We sampled three montane Panamanian sites (elevation 375-1300 m) during 2.5 months of the 2008 wet season.  We collected T_e and T_b of anurans along stream and forest transects.  Additional environmental variables such as air temperature, time, height and substrate were collected concomitantly.  We used an analysis of covariance to determine whether these factors influenced T_e and T_b.  We compared frequency distributions of T_b and T_e to a Bd thermal growth curve to determine: 1) whether temperatures above Bd’s critical thermal maximum were available to frogs, and 2) whether populations of species that have declined occupied habitats more frequently in Bd’s optimal thermal range compared to species that have not. 

Results/Conclusions

Results indicate that T_e differed among sites with cooler temperatures at higher elevation.  Air temperature (F_3,2920=56.86, p<0.0001) and time of day (F_3,266=10.61_, p<0.0001) significantly predicted T_e, though their effects on T_e differed among sites and among habitats.  Within a site, T_e was not different between habitats.  T_b of frogs differed among sites (F_14,166=2.90, p=0.0048) but not among habitats (F_1,166=1.53, p=0.2180).  Within a site, T_b varied within and among species.  Average T_e and T_b for all sites fell within Bd’s thermal tolerance range, but the low elevation site had T_b ranges extending above Bd’s critical thermal maximum.  Although temperature may explain greater losses at higher elevations, we found no significant difference in operative temperatures between stream and forest habitats at any site which indicates that temperature alone does not explain greater losses of stream anurans.  Species that have declined to extirpation elsewhere did not consistently have cooler body temperatures compared to surviving species.  Future studies should determine how Bd is influenced by other environmental factors such as humidity or by lineage specific factors such as immunity to Bd.