DISCRIMINANT ANALYSIS OF ANTIBIOTIC RESISTANCE PATTERNS OF FECAL STREPTOCOCCUS TO DETERMINE THE SOURCE OF FECAL POLLUTION IN LAKE LINGANORE, MD

Abstract

Analytical test methods such as the American Public Health Association Membrane Filter Technique and the Multiple Tube Fermentation Technique, are frequently used to estimate fecal coliform density, however these methods are unable to identify the source of the fecal contamination (e.g., humans or non-human). An alternate method, described by Wiggins (1996) isolates fecal streptococci from polluted surface water and potential sources for the characterization of their antibiotic resistances. Discriminant analysis categorizes the bacteria isolated from polluted water samples into one or more potential sources based on their patterns of antibiotic resistance. This new method was applied to the waters of Lake Linganore, MD where high fecal coliform levels have been documented since its creation in the 1970's. The lake is used for recreation, fishing, and as a drinking water source for neighboring communities. The design of Lake Linganore is unique in that the sanitary sewer lines serving the community are located both buried and elevated around the lake, generally following its perimeter. A combination of confirmed and suspected sewer leaks along with the periodic closing of the lake beaches due to poor water quality have concerned the community as to whether or not the origin of the fecal pollution was human in nature. Several sources are theorized to be contributors to the elevated fecal contamination. For this project, the potential sources were cattle, both beef and dairy, horses, wild birds (primarily geese), and humans. Samples were collected from all potential sources and characterized by their antibiotic resistance. Water samples were then collected from four locations around the lake for comparison to the source data. By using discriminant analysis, as suggested by Wiggins (1996) to categorize the antibiotic resistance of the isolates collected during this study, a large majority (52-100%) could be grouped into an animal-source category. Therefore, the fecal streptococcus isolates collected from the lake were primarily from animal sources rather than human waste.