Evaluating the impacts of an urban catchment on water and sediment quality of a receiving river

The EU Water Framework Directive requires all water bodies to achieve good ecological and chemical status by 2027. To achieve this a range of measures to improve the quality of water, particularly in urban areas, are required. It is within this context that this thesis uses a combination of field, laboratory and desk-based studies to identify contaminant loadings and their distributions within water bodies located in the Lower Lee catchment (London, United Kingdom). Specifically, water and sediment samples were collected at 11 sites on the River Lee, its’ Navigation Channel and main tributaries, over a period of two years. Samples were analysed for a range of metals (cadmium, copper, lead, mercury, nickel, tin and zinc) and 11 polyaromatic hydrocarbons (PAHs) including anthracene, fluoranthene, benzo(b)fluoranthene and benzo(a)pyrene. Laboratory batch test experiments which focused on evaluating the release of metals from field sediments were undertaken to better understand the relationship between sediment and the overlying water column. Substance flow analysis (SFA) was then applied to predict mass loads of selected pollutants entering the receiving waters within 1 km of sampling sites to evaluate the potential use of the approach as a screening tool to identify pollutant hotspots in an urban river catchment. Field sampling data and substance flow analysis outputs were compared to evaluate the use of substance flow analysis as a desk-based approach to predict sediment pollutant hotspots in the field. Use of the approach as a tool to support catchment managers identify locations for interventions to improve water and sediment quality, as well inform the development of policies targeting environmental enhancement, are discussed.
The results show that mean cadmium (2.33 ± 2.79 μg/g), copper (141.07 ± 111.00 μg/g), mercury (0.53 ± 0.45 μg/g), lead (175.70 ± 82.96 μg/g) and zinc (499.92 ± 264.66 μg/g) concentrations in the sediment exceed selected Dutch (Esdat, 2000) and Canadian (CCME, 2001) sediment guidelines. Comparison of mean polyaromatic hydrocarbon concentrations against relevant Canadian and Dutch sediment guidelines also indicates exceedances. With regard to aqueous samples, results reported here refer to total metal concentrations whereas the United Kingdom Technical Advisory Group and European Union Environmental Quality Standard both refer to dissolved and/or bioavailable concentrations. Thus, the exact implications of comparison of results to these standards are unclear as the fraction of each metal in the dissolved phase was not determined. Batch test results indicated that the amount of metal released into the sediment varied between metals and sites with the level of variation generally within an order of magnitude, ranging from a minimum of 0.12 % (tin, site A) to a maximum of 6.12 % (cadmium, site E). Through the use of reported emission factors, the substance flow analysis results predicted that a total of 19,304 kg/year (sum of six metals) and 781 kg/year (sum of five polyaromatic hydrocarbons) were discharged from the identified activities into surface waters within 1 km of each sampling site , with Deephams Sewage Treatment Works associated with a total of 6,715 kg/year for metals and a total of 12 kg/year for PAHs, corresponding to 33.5% of the total discharges for all selected pollutants (metals and PAHs) by mass.
When evaluating trends in substance flow analysis predictions in relation to sediment field data, a very strong correlation (r ≥ 0.94 and p ≤ 0.05), was observed for the tributaries for cadmium, copper, mercury, lead and zinc, suggesting that substance flow analysis is a suitable tool to support catchment managers in identifying sediment metal hotspots in relatively smaller water bodies. However, the relationships between field and substance flow analysis data sets for metals at other sites and for PAHs at all sites (with the exception of anthracene) were not statistically significant. This indicates that all PAH predicted loads do not reflect those determined in river sediment. A range of substance flow analysis model limitations were identified, including the inability of the current approach to include emissions from combined sewer overflows and aerial deposition as well as account for in-sediment processes such as remobilisation, transport and degradation and further research in these areas is recommended.