Close to reality? Micro-/mesocosm communities do not represent natural macroinvertebrate communities

Abstract Background The European environmental risk assessment of plant protection products considers aquatic model ecosystem studies (microcosms/mesocosms, M/M) as suitable higher tier approach to assess treatment-related effects and derive regulatory acceptable concentrations (RAC). However, it is under debate what extent these artificial test systems reflect the risks pesticidal substances with potential harmful on natural macroinvertebrate communities, whether field communities are adequately protected by results M/M studies. We therefore compared composition, sensitivity vulnerability benthic macroinvertebrates established in control (untreated) groups 47 selected stream at 26 reference sites. Results Since 2013 number taxa present has increased 39% a mean 38 families per study. there only an average 4 study that comply recommendations provided EFSA (EFSA J 11:3290, 2013), i.e.: (i) allowing statistical identification least 70% according minimum detectable difference (here criteria slightly modified) (ii) belonging insects or crustaceans (potentially sensitive for substances). Applying criterion physiological SPEAR pesticides concept, decreases from 2.3 Conclusions Most recent do not suitably represent freshwater communities. First, because their abundances often sufficient detection order determine appropriate endpoint subsequent RAC. Recommendations given improve detectability such reliability. Second, especially vulnerable terms traits. uncertainties linked high considering representativity assemblages comparability factors determining recovery time. Thus deriving RAC (i.e., ERO-RAC) recommended. In addition, this paper discusses further concerns regarding broader context recommends development alternative tools shift towards new paradigm.