Entomologists have often used computational modeling to study the dynamics of insects in agricultural landscapes. Recently, important issues such as the movement of adults and immatures associated with insect resistance to GMO (genetically modified organism) crops have been addressed using computational models. We developed an individual-based model using the cellular automata approach (CA) to investigate how an intercropping system composed of maize engineered with Bacillus thuringiensis (Bt) gene, refuge areas (non-Bt maize), and grasses combined with off-season periods might influence the evolution of resistance in Spodoptera frugiperda (Lepidoptera: Noctuidae), one of the leading agricultural pests targeted by GMOs.

We designed the Bt and non-Bt plants in two different arrangements: (a) a seed mixture and (b) strips rows, adding grasses in areas adjacent to the field. We added the seasonal planting dynamics (crop season and off-season), to evaluate a total of six agricultural scenarios. We followed a crop calendar from the United States to create simulations close to agricultural practice.

The results showed that the frequency of the resistance allele was strongly related to the landscape arrangements and their dynamics. Since the adult insects are mobile, the seed-mixture scenario increased the frequency of the resistance the most (95.86%), followed by strips (82.10%), without grass fields. The maize harvest made it possible to reduce the frequency of resistance allele below 1%. Based on our results, we can expect that the maintenance of pasture areas, for instance next to the corn crops, will act as a reservoir of susceptible insects during off-season periods.