Climate contributes to the evolution of pesticide resistance

Abstract

Aim The evolution of pesticide resistance through space and time is of great economic significance to modern agricultural production systems, and consequently, is often well documented. It can thus be used to dissect the evolutionary and ecological processes that underpin large‐scale evolutionary responses. There are now nearly 600 documented cases of pesticide resistance in arthropod pests. Although the evolution of resistance is often attributed to the persistent use of chemicals for pest suppression, the rate of development of resistance should also depend on other factors, including climatic conditions that influence population size and generation time. Here, we test whether climatic variables are linked to evolution of resistance by examining the spatial pattern of pyrethroid resistance in an important agricultural pest. Location Southern, agricultural regions of Australia. Time period 2007–2015. Major taxa studied The redlegged earth mite, Halotydeus destructor. Methods We quantified patterns of chemical usage based on paddock histories and collated long‐term climatic data. These data were then compared against presence–absence data on resistance using a boosted regression‐tree approach, applied here for the first time to the spatial categorization of pesticide resistance. Results Although chemical usage was a key driver of resistance, our analysis revealed climate‐based signals in the spatial distribution of resistance, linked to regional variation in aridity, temperature seasonality and precipitation patterns. Climatic regions supporting increased voltinism were positively correlated with resistance, in line with expectations that increased voltinism should accelerate evolutionary responses to selection pressures. Main conclusions Our findings suggest that the prediction of rapid evolutionary processes at continental scales, such as pesticide resistance, will be improved through methods that incorporate climate and ecology, in addition to more immediate selection pressures, such as chemical usage. Boosted regression trees present a powerful tool in the management of resistance issues that has hitherto not been used.

Climate change and the genetics of insecticide resistance

“…It would also be interesting to revisit the previous sample sites of a large comprehensive study involving D. melanogaster and repeat the experiments in a few decades to determine whether climate change has affected the allele frequency of these populations over time. Another method is to use long‐term climatic data and insecticide usage records to build models that may predict how these two variables interact in the development of field resistance, as discussed in the introduction …”

“…Evidence that climate can influence the evolution of resistance in the field is supported by modeling the association between long‐term climatic data and insecticide usage or resistance data in specific regions. An analysis of pyrethroid resistance in the redlegged earth mite, Halotydeus destructor , a major pest of Australia's grain and pastoral industries, across many regions in Western Australia indicated that regional variations in aridity, temperature seasonality, and precipitation patterns affected the spatial pattern of resistance . Another study modeling the development of deltamethrin resistance in Triatoma infestans , a Chagas disease vector, suggests that climate may influence the development of resistance in a region in Argentina …”

Escalating insecticide resistance in Australian grain pests: contributing factors, industry trends and management opportunities

“…A technique commonly applied in machine learning was recently used to successfully capture the current distribution of insecticide resistance of H. destructor within Australia from environmental and management factors hypothesized to increase resistance risk. This modeling highlighted geographic locations without resistance, but with similar properties to areas with resistance . Since this study, resistant field populations have been detected within regions identified as high risk (Fig.…”

“…Both help to bring additional value to the monitoring and management programs described earlier (Table ). For example, models can aid in identifying high‐risk areas or practices for pre‐emptive management . The evolution and management of insecticide resistance is multi‐dimensional, and computational approaches help in making this complexity more manageable and in identifying factors influencing resistance risk …”

Identifying critical research gaps that limit control options for invertebrate pests in Australian grain production systems

“…Few growers implement practices that mitigate the impact of insecticides on natural enemies (Roubos et al 2014). Given that there are only four mode-of-action pesticide groups in products registered to control H. destructor and that resistance against two has been demonstrated, the long-term viability of pesticide-based control options is uncertain (Maino et al 2018). However, grain growers will face new challenges in the future that may alter how and when they use insecticides.…”