Bayesian Analysis for Inference of an Emerging Epidemic: Citrus Canker in Urban Landscapes

Neri, Franco M.; Cook, Alex R.; Gibson, Gavin J.; Gottwald, Tim R.; Gilligan, Christopher A.

This was due in part to the (largely unknown) time-lags in the effect of weather events on the pathogen and the host. It is also reasonable to assume that environmental, weather-related forcing was just one, if the most important, of the factors affecting the behaviour of b. Factors intrinsic to the host population might also have played an important role: tree age, cultivar, and horticultural care can affect the susceptibility to the disease [28] . In a population of residential trees, the distribution of such individual factors is extremely heterogeneous in space at several scales, and also fluctuates over time. In the present case, as a result, there was a high degree of spatio-temporal variability in the response of hosts to weather drivers. Fitting models with explicit individual factors is unfeasible in such a highly heterogeneous scenario; however, such a class of models might be very useful in future analyses of outbreaks within commercial citrus plantations, where host properties are more consistently distributed.