Krystal Siebert
Broad-Scale Assessement of the Influence of Wetland-to-Lake Connectivity on Lake Chemistry and Fish Mercury Levels ©2018
It is well-known that wetlands are favourable environments for the conversion of inorganic mercury into more bioavailable mercury forms such as methylmercury (MeHg). Furthermore, wetlands that are hydrologically connected to downstream receiving waters can act as sources of dissolved organic carbon (DOC) and toxic methyl mercury (MeHg). The focus of this thesis is twofold, to identify gaps in the wetland sediment Hg literature and to evaluate the influences of wetland-to-lake connectivity metrics on Lake DOC and fish THg concentrations. First, to examine broad-scale patterns in representation of Hg in wetland sediments a meta-analysis was conducted to build a province wide wetland sediment Hg dataset. A search of Hg studies conducted in Ontario between 1980 and present was carried out and information on wetland locations, size, type, and mercury species reported were extracted and compared to similar wetland information extracted from the provincial land cover layer. In terms of size and type, the meta-analysis revealed that previous studies (n=14; 190 individual wetlands or wetland complexes) do not reflect the actual distribution of wetland in Ontario. Second, to better understand the influence of watershed characteristics, such as wetland coverage, on the fate and transport of MeHg in freshwater systems, this study assesses the extent to which inclusion of wetland-to-lake connectivity metrics improves predictive models of lake DOC and Fish THg concentrations. Correlation and Partial Least Squares (PLS) regression was used to understand the influence of individual and groups of metrics to predict lake DOC and fish Hg concentrations for a set of 66 headwater lakes were used from the Broad-scale Monitoring (BsM) system. Average slope of the catchment, lake volume (normalized by catchment area), wetland area and lake adjacent wetland area as a proportion of the catchment, and average slope of the wetland-to-lake flow path were identified as important predictors.