Recently EFSA has published a scientific opinion as a precursor of the future guidance document on risk assessment on Non Target Arthropods (NTA) at EU level. Currently the risk assessment for NTA concerning the application of plant protection products is addressed separately for in-field and out-field assuming no interaction between the zones. However these two areas seem to be limited compared to the agricultural landscape. Therefore, the risk assessment has to be focused on local and landscape scales to take into account the effects not just on NTA species but also the impacts on biodiversity and on ecosystem services, such as food web and pollination. The proposed assessment scheme provides a local scale evaluation (for the in-field and off-fieldareas) followed by a landscape level to be performed only if the first one resulted in an acceptable risk. Product authorisation or substance approval will be granted when feedbacks from both the evaluations meet the acceptable criteria.
The local scale starting from laboratory studies as a first tier to semi-field and field studies as an higher tier option.
Due to the sensitivity of the taxonomic groups, the life stages involved, the relevant exposure pathways and specific modes of action of the current laboratory protocols, some concerns raised about their ability to detect unacceptable effects. For this purpose it is recommended to test as minimum four species including a leaf-dwelling herbivore, such as lepidopteran larvae, to represent the oral route of exposure. It is suggested to extend the duration using a long-term exposure to observe the effects in more different life stages in particular during the reproduction phase and to carefully consider the contact exposure, the oral consumption or the overspray in accordance with the application method.
The uncertainties of the laboratory tests can be overcome by performing higher tier studies to evaluate the effects in realistic conditions at community level. Field trials have some limitations due to their experimental design. The present protocols consist of several in-field plots in which the mobility of the tested species could be critical so that a combination of large and small plots is desirable.
The landscape assessment of the effects, including recovery, makes use of modelling. It has to be carried out to evaluate the risk for the population in its spatial distribution and time scale, currently the scenarios and the species modelled are quite limited and so much work has to be done…