Date of Award

2017

Document Type

Open Access Dissertation

Department

Environmental Health Sciences

Sub-Department

Earth and Environmental Resources Management

First Advisor

Gwen Geidel

Abstract

Fish tissue from various sampling locations in South Carolina rivers has shown elevated levels of mercury (Hg). A number of variables may impact this Hg in fish tissue and this study evaluated the impact of these variables using a predictive model. Over 341 variables were assembled from 6 different databases describing land use, water chemistry, point source pollutants, watershed characteristics and more. One dataset describes the location of historical gold mines throughout the sample site, which have been shown to increase Hg in fish tissue in other locations. The variables were assembled in ArcGIS to evaluate variable significance and to create a model inclusive of Hg in fish data from 2005 to 2015. The final version of the model (r2=0.84) showed predictable variation in mercury in fish tissue based on longitude, precipitation, watershed area within a municipal storm water sewage system, total kjeldahl nitrogen, and dam height, if any dams were present. A better model could not be made with the addition of any variables describing historical gold mines (r2=0.77) and therefore, this study does not find an interaction between mercury in fish tissue and historical gold mines in select watersheds of South Carolina.

Share

COinS