In Andhra Pradesh, India, pyrethroid resistance in the noctuid moth, Helicoverpa armigera, follows a seasonal cycle. Levels of resistance increase from August to March, and then decrease in the hot dry period between April and July. An hypothesis expressed in the form of a simple simulation model is presented to explain these fluctuations. Uncultivated hosts, upon which there is no selection for pyrethroid resistance, act as a reservoir of more susceptible moths. A movement of moths from the reservoir to cultivated hosts is postulated to explain the drop in resistance. The model incorporates the effects of seasonal variation in host availability and of constraints on the mixing of genotypes between cultivated and uncultivated hosts. The most critical factor is the rate at which the uncultivated refuges are contaminated by resistant genotypes. This is minimised if the seasonal cycles of cultivated and uncultivated hosts are in synchrony. Rate of refuge contamination is most rapid if uncultivated hosts are scarcest one to two moth generations before availability of cultivated hosts falls to a minimum. A small population of wild plants may have a marked impact on the rate at which resistance develops in the Helicoverpa population on crops. The relationship between selection pressure and rate of resistance development is exponential, so successive reductions in insecticide usage lead to larger and larger delays in resistance.
Ecological Modelling (1995) 82 (1) 61-74 [doi:10.1016/0304-3800(94)00071-O]