The development and use of models describing outbreaks of resistance to fungicides

The importance of resistance has increased in the last decade, together with the increasing use of site-specific fungicides. Mathematical models have been developed to deal with disruptive selection that occurs when the initial population of the target organism in terms of fungicide sensitivity consists of two or more distinct, non-overlapping distributions around widely different central values. Whether these models are static or dynamic, they lead to the same basic conclusions as to what factors would accelerate resistance build-up, namely the initial incidence of the resistant mutants, the efficacy of the fungicide at risk and the degree of resistance to it, and the degree of control pursued in the field. Dynamic models add to those factors the speed of the pathogen spread in the field, the fitness of the resistant mutants and the size of the pathogen population exposed to the fungicide. Resistance build-up can be delayed by the use of mixtures. In such a case, the higher the efficacy of the companion fungicide, the greater the delay. Alternation of fungicides will delay resistance build-up in calendar time: however, in the case of fit resistant mutants it will not extend the useful life of the fungicide at risk. No models have been developed to deal with the directional selection that occurs when, in terms of fungicide sensitivity, the population of the pathogen consists of one wide unimodal frequency distribution. Further research is recommended towards verifying the disruptive selection models, creating models to deal with directional selection and generating baseline sensitivity data for new fungicides before their commercial introduction.