Patterns exhibited in animal’s life history traits, can often be described by temperature. Temperature is also an important factor in many disease systems by influence traits like, growth rate or reproductive rate. However, the role of temperature in these systems is often studied in an artificial way, not allowing the temperature to change. Relating this data back to a more natural temperature scenario is difficult and relies on inaccurate methods. Temperature is especially important in the amphibian chytrid fungus system (Batrachochytrium dendrobatidis), where frogs that live in cooler ponds experience higher infection rates. With temperature being a main factor in many disease systems, it is important to develop accurate methods to relate constant temperature experiments to a varying temperature environment.
Results/Conclusions
One of the most common methods used to relate constant temperature experiments to predict traits in a varying temperature regime is rate summation. This method has not been widely tested to see when and where this technique works. Further, when this technique is tested the literature reports the method either over or under predicts the growth that an organism would experience. To test this method in the amphibian chytrid fungus system we fit a logistic growth model to optical density data, using Bayesian inference. The parameters of the model were estimated using JAGS. To determine temperature dependency, samples (n = 100) of the logistic growth rate (r) from the posterior distribution of the logistic growth model were taken. For each strain we plotted these samples of the logistic growth rate against temperature and fit the Briere curve. Using this the Briere curve, we can then use rate summation to make predictions about Bd growth in the varying temperature conditions. Data from the varying temperature experiments was then contrasted with the predictions. We found that rate summation over predicted the growth of Bd using constant temperature.