Climate variation mediates a host-parasite interaction: Field and laboratory evidence at multiple temporal scales

Background/Question/Methods

Climate change is expected to influence ecological communities dramatically, by changing species ranges and influencing species interactions.  Much research has focused on predicting ecological effects of mean climate parameters, such as temperature, precipitation, and CO₂ levels.  However, climate variability is also projected to change as the Earth warms, yet we know little about how climate variability influences species interactions, including host-parasite relationships.  Although theoretical studies have implicated temperature variability as a potentially important mediator of species interactions, to our knowledge there has been no previous experimental demonstration of temperature variation influencing a host-parasite system.   In this study, we used a combination of field data and laboratory experiments to investigate the impact of temperature variability, at different time scales, on the interaction between the fungal pathogen Batrachochytrium dendrobatidis (Bd) and its amphibian hosts.

Results/Conclusions

In our analysis of patterns of amphibian extinctions in the frog genus Atelopus, putatively caused by the fungal pathogen Batrachochytrium dendrobatidis (Bd), we found evidence that El Niño events raise extinction probabilities by increasing month-to-month temperature variability and the diurnal temperature range.  These two climate variability parameters were better predictors of extinctions than any of the mean climate parameters.  Laboratory experiments also revealed important biological effects of temperature variability that could not have been predicted from models parameterized with data from constant-temperature experiments.  In a replicated controlled-temperature experiment with Cuban treefrogs (Osteopilus septentrionalis), a tropical frog species, we found that frogs subjected to an unpredictable drop in temperature were less resistant to Bd infection than frogs already acclimated to the exposure temperature.  In a separate experiment, we found that both Bd and Cuban treefrogs are adapted to diurnal temperature fluctuations, with lower Bd loads on frogs with diurnal temperature fluctuations (15°C to 25°C) than on frogs with constant average (20°C) or randomly shifting (15°C or 25°C) temperatures, despite the opposite pattern for Bd growth in culture (higher for diurnal temperature fluctuations).  These results suggest that climate variability, on the monthly and daily time scales, might be as or more important than mean climatic variables in driving this host-parasite interaction.  We therefore urge biologists to consider the effects of climatic variability on species interactions, as well as of mean climate predictors, when investigating how climate change influences ecological systems.