Using a dynamic mass balance model to predict fate and transport of PCBs in a polluted boreal lake in Sweden

Abstract

In 2013, a screening survey including fish (European perch, Perca fluviatilis) from 20 locations in the Stockholm region of Sweden indicated exceptionally high levels of PCBs (>450 ng ΣPCB7/g ww) in Lake Oxundasjön. An extensive sampling program was launched to define the magnitude and area of impact of PCBs. Moreover, a dynamic mass balance model approach was applied to identify and quantify key transport processes and predict the long-term turnover of PCBs given various remediation scenarios.Based on the dating of sediment profiles, primary emissions of PCBs to Lake Oxundasjön have likely occurred from the end of the 1940s until 1980, reaching the lake via one of its tributaries. Presently, the main source of PCBs is diffusion from the lake sediments. From the lake outlet, >400 g ΣPCB7/yr are transported to Lake Mälaren (the third largest lake in Sweden), supplying drinking water for parts of the Stockholm area. Remediation actions are necessary to reduce the PCB levels in fish below today's marketing limits and environmental quality standards. With natural recovery, our results indicate that the PCB levels in non-migratory fish from Lake Oxundasjön will be elevated for decades to come.The mass of PCBs stored in the lake sediments was estimated, and to our knowledge, Lake Oxundasjön is the most heavily PCB contaminated lake in Sweden. The system constitutes a unique opportunity to test and develop a mathematical mass balance model for PCBs, with substantial data acquired from different aquatic matrices. The model presented in the paper is applicable for risk assessments of PCBs, and the results contribute to the general understanding of the transport and turnover dynamics of PCBs in aquatic ecosystems.