Microplastics transport during Managed Aquifer Recharge – A potential cause of groundwater contamination?

Managed Aquifer Recharge (MAR) belong to the nature-based engineering approaches poised to play an increasingly positive role in climate change adaptation. In fact, the increased groundwater availability thank to MAR can buffer future temporal water shortages driven by forecast longer droughts. However, one concern of MAR regards the quality of the infiltrating water as it affects groundwater quality.There is growing concern about groundwater contamination by microplastics (MP) delivered by the infiltrated surface water. Indeed, MPs have been found in rivers globally. Groundwater contamination by MPs could then have direct detrimental consequences on groundwater management and availability for human uses.In this study, we measured MPs larger than 20 µm in the different important stages of a major MAR-water supply system in Switzerland. The MAR system has been diverting an average of 95,000 m3/day of Rhine water through channels and ponds since 1958. Samples of filtered water were taken in triplicates from the Rhine River near Basel, the treatment stages before the managed infiltration outlets, the pumped groundwater, before and after the activated carbon filters. The methodology involved the analysis of the filters by means of micro-Fourier Transform InfraRed (FTIR) spectroscopy and MPs identification by means of the machine-learning model developed by the Purency company. The possible contamination by MPs smaller than 20 µm could not be assessed due to practical challenges for water filtration in the field and quantification (reliable micro-FTIR measurements down to 20 µm). Before analyses, we added surrogate particles of polyethylene with an average size of 60 µm for quality assurance and quality control.The measurements revealed satisfactory MPs reduction along the MAR system. While the number of MPs in the raw Rhine River had to be quantified still, MPs average concentration decreased from about 9.75 particles/l in the treatment stage to about 1.3 particle/l after the activated carbon filter. MPs average concentration increased to 7 particles/l between the pumped groundwater and the activated carbon filters. The increase was driven by a high count of polypropilene MPs in one of the triplicates. Our study could not exclude a possible MPs contamination by construction materials used in the facility. The measured concentrations referred to a randomised scan of about 50% of the area of the filters encompassing the centre and the edges of the filters. Most of the surrogate particles accumulated along the edges. Yet, we only considered particles with a value of the Relevance and Similarity metrics greater than 0.3.Acknowledgements: We thank our colleague Reto Britt who supported with the sampling campaign and laboratory measurements and the drinking water supplier for allowing us to carry out the study within their premises.