Aquatic Mesocosm Strategies for the Environmental Fate and Risk Assessment of Engineered Nanomaterials.

Abstract

In the past decade, mesocosms have emerged as a useful tool for the environmental study of engineered nanomaterials (ENMs) as they can mimic the relevant exposure scenario of contamination. Herein, we analyzed the scientific outcomes of aquatic mesocosm experiments, with regard to their designs, the ENMs tested, and the end points investigated. Several mesocosm designs were consistently applied in the past decade to virtually mimic various contamination scenarios with regard to ecosystem setting as well as ENMs class, dose, and dosing. Statistical analyses were carried out with the literature data to identify the main parameters driving ENM distribution in the mesocosms and the potential risk posed to benthic and planktonic communities as well as global ecosystem responses. These analyses showed that at the end of the exposure, mesocosm size (water volume), experiment duration, and location indoor/outdoor had major roles in defining the ENMs/metal partitioning. Moreover, a higher exposure of the benthic communities is often observed but did not necessarily translate to a higher risk due to the lower hazard posed by transformed ENMs in the sediments (e.g., aggregated, sulfidized). However, planktonic organisms were generally exposed to lower concentrations of potentially more reactive and toxic ENM species. Hence, mesocosms can be complementary tools to existing standard operational procedures for regulatory purposes and environmental fate and risk assessment of ENMs. To date, the research was markedly unbalanced toward the investigation of metal-based ENMs compared to metalloid- and carbon-based ENMs but also nanoenabled products. Future studies are expected to fill this gap, with special regard to high production volume and potentially hazardous ENMs. Finally, to take full advantage of mesocosms, future studies must be carefully planned to incorporate interdisciplinary approaches and ensure that the large data sets produced are fully exploited.