An agent-based model of insect resistance management and mitigation for Bt maize: A social science perspective
Yuji Saikai,Paul D. Mitchell,Terrance M. Hurley
DOI: https://doi.org/10.1101/732776
2019-08-12
Abstract:Abstract Managing and mitigating agricultural pest resistance to control technologies is a complex system in which biological and social factors spatially and dynamically interact. We build a spatially explicit population genetics model for the evolution of pest resistance to Bt toxins by the insect Ostrinia nubilalis and an agent-based model of Bt maize adoption, emphasizing the importance of social factors. The farmer adoption model for Bt maize weighed both individual profitability and adoption decisions of neighboring farmers to mimic the effects of economic incentives and social networks. The model was calibrated using aggregate adoption data for Wisconsin. Simulation experiments with the model provide insights into mitigation policies for a high-dose Bt maize technology once resistance emerges in a pest population. Mitigation policies evaluated include increased refuge requirements for all farms, localized bans on Bt maize where resistance develops, areawide applications of insecticidal sprays on resistant populations, and taxes on Bt maize seed for all farms. Evaluation metrics include resistance allele frequency, pest population density, farmer adoption of Bt maize and economic surplus generated by Bt maize. Based on economic surplus, the results suggest that refuge requirements should remain the foundation of resistance management and mitigation for high-dose Bt maize technologies. For shorter planning horizons (< 16 years), resistance mitigation strategies did not improve economic surplus from Bt maize. Social networks accelerated the emergence of resistance, making the optimal policy intervention for longer planning horizons rely more on increased refuge requirements and less on insecticidal sprays targeting resistant pest populations. Overall, the importance social factors play in these results implies more social science research, including agent-based models, would contribute to developing better policies to address the evolution of pest resistance. Author Summary Bt maize has been a valuable technology used by farmers for more than two decades to control pest damage to crops. Using Bt maize, however, leads to pest populations evolving resistance to Bt toxins so that benefits decrease. As a result, managing and mitigating resistance has been a serious concern for policymakers balancing the current and future benefits for many stakeholders. While the evolution of insect resistance is a biological phenomenon, human activities also play key roles in agricultural landscapes with active pest management, yet social science research on resistance management and mitigation policies has generally lagged biological research. Hence, to evaluate policy options for resistance mitigation for this complex biological and social system, we build an agent-based model that integrates key social factors into insect ecology in a spatially and dynamically explicit way. We demonstrate the significance of social factors, particularly social networks. Based on an economic surplus criterion, our results suggest that refuge requirements should remain the foundation of resistance mitigation policies for high-dose Bt technologies, rather than localized bans, areawide insecticide sprays, or taxes on Bt maize seed.