Advancing the environmental risk assessment of non‐target arthropods for plant protection products by accounting for the impact on ecosystem services and on the ecological function. Experimental work on inter‐species sensitivity

Le résumé fourni par la source

Abstract This report presents the protocol designed to assess inter‐species sensitivity of non‐target arthropods (NTAs) to plant protection products (PPPs), along with the final experimental results. Sixteen species were selected to represent key NTA functional groups (predators, parasitoids and pollinators) considered most vulnerable to pesticide exposure. Three insecticides with different modes of action were tested as formulations: Karate Garden (a.i., lambda‐cyhalothrin), Sanium (a.i., flupyradifurone), and Mainspring (a.i., cyantraniliprole). For comparability, direct spray with the formulated products was chosen as the exposure route. In each test, a range of doses was applied using a Potter tower spray system to ensure precise and homogeneous overspray deposition on the test organisms. The application rates were validated using filter discs. Mortality was recorded for a minimum of 48 hours. Median lethal doses/rates (LD/LR50) were calculated at 24 and 48 hours where possible, and Species Sensitivity Distribution (SSD) curves were constructed for each PPP. Karate Garden (a.i., lambda‐cyhalothrin) was the most toxic formulation, followed by Sanium (a.i., flupyradifurone) and Mainspring (a.i., cyantraniliprole). Despite the observed differences in sensitivity across species and functional groups, consistent patterns could be identified across different dose metrics. Under the per‐individual metric (ng a.i./individual), small‐bodied taxa, particularly parasitoids and the predatory mite Typhlodromus pyri, consistently ranked among the most sensitive species, whereas bees and ground beetles were generally less sensitive. These patterns were partly influenced by body size, with shifts in relative sensitivity occurring after normalisation to body mass (μg a.i./g organism). Nonetheless, some species, including T. pyri and Coccinella septempunctata (larval stage), remained highly sensitive irrespective of the dose metric, indicating higher intrinsic susceptibility. Life‐stage differences were also observed, with larvae of C. septempunctata consistently more sensitive than adults. Although at least one Tier‐1 species appeared among the most sensitive taxa in each SSD, indicator species did not consistently provide protection for other species within their respective functional groups across PPPs. Overall, the study highlights pronounced interspecific variability in pesticide susceptibility driven by chemical properties, functional groups, and life‐stage differences. While the proposed overspray protocol offers a robust and reproducible method for assessing acute contact toxicity, the findings support broadening the range of test species (particularly including further species from NTA groups considered vulnerable) and products (e.g., different pesticide groups), as well as expanding the exposure conditions considered in risk assessments. Incorporating multiple representative taxa and accounting for life stage‐dependent and route‐specific sensitivity will enhance the ecological realism and protective capacity of pesticide risk assessment for non‐target arthropods.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.