Suivi spatio-temporel de la pollution de MP dans la Garonne et potentiel transfert dans le réseau trophique

The omnipresence of microplastic (MP) represents a novel environmental pressure acting on freshwater ecosystems and a better understanding of the dynamic of this pollution is needed. Here, we investigated the spatial and temporal changes in MP pollution (size range 700 μm – 5 mm) in the Garonne catchment (Southwestern France) and the consumption of these particles by aquatic organisms. The composition of MP was verified through attenuated total reflectance Fourier transformed infrared spectroscopy (ATR-FTIR). First, a total of 14 sites located in the main river and several tributaries were sampled during four seasons. We found that MP concentration averaged 0.15 particles.m-3 (± 0.46 SD) and strongly varied both in space and in time, driven by urbanization and hydrological conditions. Higher MP concentrations and smaller particle sizes were observed in warm seasons with low discharge. Second, we analysed the changes in MP pollution caused by flooding. Two sites in the Garonne River, located upstream and downstream of Toulouse, were sampled during two flood episodes. We found a general increase in MP concentration during flood episodes. This was driven by river discharge, but this increase was greater in the downstream site. Regarding MP characteristics, a predominance of larger particles was observed. Using multivariate analysis of the infrared spectra, we quantified the changes in MP chemical profile during flooding. A higher oxidation profile, represented by an increased carbonyl spectral band, was found in particles collected during the flood. Third, a novel pathway of MP into freshwaters was assessed by quantifying MP pollution in angling baits. We analysed three different categories of industrially-produced baits (‘groundbait’, ‘boilies’ and ‘pellets’). From 160 bait samples, 28 MP were identified in groundbait and boilies. No MP within the studied size range were found in pellets. We revealed that MPs introduced accidentally during bait manufacturing and/or those originating from contaminated raw ingredients might be transferred into freshwaters. Fourth, the consumption of MPs by macroinvertebrates and fish was quantified. This consumption was linked to individual trophic niches, which were measured by stable isotope analyses (δ13C and δ15N). We demonstrated that the abundance of ingested MP differed between macroinvertebrates and fish and was not significantly related to MP pollution. We also found that MP characteristics differed between the abiotic (water and sediment) and biotic (macroinvertebrates and fish) compartments. The abundance of ingested MP increased with organism size in both fish and macroinvertebrates and tended to increase with the trophic position of macroinvertebrates. The origin of the resources consumed by fish significantly affected the abundance of MP ingested in fish. Altogether, these results suggested the absence of MP bioaccumulation in the studied size range in freshwater food webs and the dominance of direct consumption, most likely accidentally. In conclusion, we highlighted that MP pollution should be perceived as a multi-stressor due to the particulate behaviour and potential interactions with other environmental contaminants. The consequences of these interactions should be the focus of future research. This work contributes to improve our understanding to elucidate the drivers of the dynamic and consumption of MP, and further studies are needed to quantify the actual risks associated with this pollution in freshwater ecosystems.