Skip to main content
ARS Home » Pacific West Area » Maricopa, Arizona » U.S. Arid Land Agricultural Research Center » Pest Management and Biocontrol Research » Research » Publications at this Location » Publication #295434

Title: Large scale, spatially-explicit test of the refuge strategy for delaying insecticide resistance

Author
item CARRIERE, YVES - University Of Arizona
item ELLERS, KIRK - University Of Arizona
item HARTHFIELD, K - University Of Arizona
item DEGAIN, B - University Of Arizona
item BARKLEY, CROWDER - University Of Arizona
item DENNEHY, T - University Of Arizona
item ELLSWORTH, PETER - University Of Arizona
item LI, X - University Of Arizona
item FOURNIER, A - University Of Arizona
item LAROCQUE, G - McGill University - Canada
item OUTILLEUL, P - McGill University - Canada
item ANTILLA, LARRY - Arizona Cotton Research And Protection Council
item Naranjo, Steven
item PALUMBO, J - University Of Arizona
item TABASHNIK, B - University Of Arizona

Submitted to: Proceedings of the National Academy of Sciences (PNAS)
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/8/2011
Publication Date: 6/10/2011
Citation: Carriere, Y., Ellers, K.C., Harthfield, K., Degain, B., Barkley, C.D., Dennehy, T., Ellsworth, P., Li, X., Fournier, A., Larocque, G., Outilleul, P., Antilla, L., Naranjo, S.E., Palumbo, J., Tabashnik, B. 2011. Large scale, spatially-explicit test of the refuge strategy for delaying insecticide resistance. Proceedings of the National Academy of Sciences. 109:775-780.

Interpretive Summary: The development of resistance in arthropod pests to pesticides is a major barrier to sustainable pest control in agricultural system. One strategy to slow or reverse resistance evolution is to provide crop areas for the pest where the pesticide in question is not used. This is called the refuge strategy, but there is little empirical data to show that such an approach actually works as theoretical models would predict. Here, we combined an eight-year, large-scale data set of resistance monitoring to the insect growth regulator pyriproxyfen in 84 populations of the sweetpotato whitefly sampled in cotton fields in Central AZ with spatial data on whitefly affected crops, to formulate an empirically-based statistical model linking the evolution of resistance to the distribution of refuges and evaluate the accuracy of predictions of this model. We found that the distribution of alfalfa and melon was not significantly associated with resistance, while treated and untreated cotton significantly influenced resistance. The model incorporating the distribution of treated and untreated cotton accounted for the highest proportion of variation in resistance at a scale of 2.75 km surrounding the sampled sites. This model predicted spatial variation in resistance, indicating that refuges of untreated cotton at < 2.75 km from treated fields will delay resistance. Our approach has the potential to improve refuge strategies for numerous pests and systems.

Technical Abstract: The refuge strategy used worldwide to delay the evolution of arthropod resistance to pesticides consists of leaving areas where a pesticide is not used near fields where the pesticide is used. Yet, empirical approaches are lacking to characterize effects of putative refuges on resistance evolution. Furthermore, predictive refuge strategy models have not been evaluated in the field, which is needed to improve credibility of this strategy for resistance management. Here, we combined an eight-year, large-scale data set of resistance to the insect growth regulator pyriproxyfen in 84 populations of the sweetpotato whitefly sampled in cotton fields in Central AZ with spatial data on whitefly affected crops, to formulate a statistical model linking the evolution of resistance to the distribution of refuges and treated fields and evaluate the accuracy of predictions of this model. We found that the distribution of alfalfa and melon was not significantly associated with resistance, while treated and untreated cotton significantly influenced resistance. The model incorporating the distribution of treated and untreated cotton accounted for the highest proportion of variation in resistance at a scale of 2.75 km surrounding the sampled sites. This model predicted spatial variation in resistance, indicating that refuges of untreated cotton at < 2.75 km from treated fields will delay resistance. Our approach has the potential to improve refuge strategies for numerous pests and systems.