AUTHOR=Göransson Gunnel , Apler Anna , Dahlberg Anna-Karin , Löfroth Hjördis , Josefsson Sarah , Wiberg Karin , Frogner-Kockum Paul , Nylander Per , Hedfors Jim , Snowball Ian TITLE=Assessing the Risk of Contaminant Dispersion From Fibrous Sediments of Industrial Origin JOURNAL=Frontiers in Marine Science VOLUME=8 YEAR=2021 URL=https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2021.729243 DOI=10.3389/fmars.2021.729243 ISSN=2296-7745 ABSTRACT=

Unregulated discharges of wastewater from pulp and paper factories resulted in the formation of relatively thick organic (cellulose) rich sediments in shallow waters along the Swedish coast. These deposits are known as fiberbanks and are contaminated by persistent organic pollutants (POPs), metals and methylmercury, which can be dispersed by diffusion and advective processes coupled to propeller wash, high river discharges, strong wind waves and submarine landslides. Based on a case study of polychlorinated biphenyls (PCBs), one group of prevalent POPs in the fiberbanks, we present a probabilistic approach to estimate the potential risk of dispersion of fiberbank contaminants. The approach allows for estimation of the dispersal pathways that dominates the risk within a given time and provides more insight about the significance of various dispersion processes. We show that it is highly likely that chemical diffusion and advection triggered by ship-induced resuspension will disperse PCBs (sum of seven congeners; Σ7PCB) above a threshold level for environmental impact, while the likelihood of river and wind-wave generated resuspension dispersion pathways are lower (∼20%, respectively). We further show that there is approximately 5% likelihood that a submarine landslide will disperse Σ7PCB above the threshold level. The study implies that the governing parameters for risk assessment specifically should include reliable data on contaminant concentration, water depth above the fiberbank, estimation of concerned fiberbank areas, time duration of erosive fluid flows and measured diffusion. The approach provides insight into the importance of various dispersion processes. We suggest that it can be applied to support risk assessment, especially when there are limited available data and/or knowledge about the system under study.