Sweep netting samples, but not sticky trap samples, indicate beneficial arthropod abundance is negatively associated with landscape wide insecticide use

• Published in: The Journal of applied ecology 2022-04, Vol.59 (4), p.942-952
• Main Authors: Bakker, Lieneke, Werf, Wopke, Bianchi, Felix J. J. A.
• Format: Article
• Language: English
• Subjects: Abundance; agricultural landscape; Agrochemicals; Arthropoda; Arthropods; Beneficial arthropods; Drift; ecosystem services; Indicators; insecticide use; Insecticides; Insects; Land management; parasitoids; pesticide; Pesticides; Pollinators; Populations; Predators; Risk assessment; Sampling; Source-sink relationships
• ISSN: 0021-8901; 1365-2664
• DOI: 10.1111/1365-2664.14106

Insecticide use and landscape context are major drivers for the abundance of beneficial arthropods, such as predators, parasitoids and pollinators. However, the relative importance of local and landscape‐wide insecticide use is not well understood, and it is unclear to what extent impacts of insecticides on beneficial arthropod populations are moderated by landscape composition. We collected and analysed data on beneficial arthropod abundance and local and landscape‐wide insecticide use across 38 Dutch landscapes. We used regression to study the associations between beneficial arthropod abundance (response), insecticide use and landscape factors. Insects were sampled by sweep netting and yellow sticky traps in field margins (‘sampling sites’) in the summer of 2017 and 2018. We paired organic and conventional sites in the same landscape setting. We used three indicators of insecticide use: management (organic vs. conventional), Treatment Frequency Index of insecticides applied in the field adjacent to the sampling site, and the estimated landscape‐wide average quantity of insecticides applied per ha within a 1 km radius around the sampling site. Abundances of beneficial arthropods in field margins, assessed by sweep netting, were negatively associated with the estimated quantity insecticide applied in the surrounding landscape, while arthropod sampling by sticky traps did not show this relationship. Pollinator abundance was negatively associated with conventional management and with the frequency of insecticide applications in the adjacent field, while predator and parasitoid abundance did not show significant associations with local insecticide use indicators. Negative impacts of insecticide use were not moderated in more diverse landscapes with higher proportions of semi‐natural habitats. Policy implications. Insecticide impacts on arthropod populations extend beyond the fields in which they are used. Awareness of the potential landscape‐wide effects of pesticides, either through drift of pesticides or insect movement, may encourage land managers to adopt practices to reduce insecticide drift and the frequency of insecticide applications, thereby reducing the impacts of landscape‐wide exposure on arthropods that provide services to agriculture. Risk assessments of insecticides and pesticide use regulations need to account for potential spill‐over effects and consider landscape‐scale consequences associated with the source–sink dynamics of arthropods. Same