QUANTIFICATION OF ESCHERICHIA COLI AND ENTEROCOCCI LEVELS IN WET WEATHER AND DRY WEATHER FLOWS

Forty percent of rivers, 45% of lakes and 50% of estuaries assessed by the National Water Quality Inventory (2000) were not clean enough to support designated fishing and swimming uses. Pathogens were found to be one of the leading causes of impairments in these waters. Urban runoff is recognized as a leading source of organisms potentially indicating the presence of pathogens. Urban runoff can be defined as any discharge from a separate storm drainage system. Urban runoff traditionally had been defined to include precipitation and wash off from lawns and other landscaped areas, buildings, roadways and parking lots. However, other flows may enter the storm drainage system from such sources as infiltrating groundwater, leaking domestic water supplies and sewage, washwaters, and other inappropriate entries to the storm drainage system. This research was conducted to quantify the levels of indicator bacteria, and their sources, in urban areas. The main objective of this research was to identify possible sources of E. coli and enterococci bacteria in dry and wet-weather flows in storm drainage systems.An urban area consists of many different kinds of land uses such as residential, institutional, commercial, industrial open spaces, etc. Each type of land use consists of various types of source areas, such as roofs, parking lots, landscaped areas, playgrounds, driveways, undeveloped areas, sidewalks. Four representative source area types were sampled during this research; including rooftops, parking lots, open spaces, and streets. Two parallel sites were sampled for each source area type; one affected by birds and other animals, and another set with less influence from birds and other animals. A section of Cribbs Mill Creek in Tuscaloosa, Alabama, was also selected for dry weather sampling at outfalls. The section of the creek was selected such that the drainage areas contributing to outfalls had either commercial or residential land uses. Potential inappropriate discharge water samples were also obtained, including influent samples from the Tuscaloosa sewage treatment plant, local springs, irrigation runoff water, domestic water taps, car wash, and laundry water. Overall, total coliforms, E. coli and enterococci bacterial analyses were conducted on 202 wet weather and 278 dry weather flow water samples. All samples were analyzed using IDEXX Quantitray enumeration procedures.E. coli and enterococci levels larger than 2,400 and 24,000 MPN/100 mL, respectively, were observed in wet weather samples collected from various source areas which could not possibly be contaminated with sanitary sewage. The levels of indicator bacteria present in the urban runoff source area samples exceeded the EPA 1986 single sample maximum value water quality criteria in 31% of the samples for E. coli and in 74% of the samples for enterococci. The geometric mean criteria were exceeded in 100% of the source area samples. Since both the indicator organisms studied (E. coli. and enterococci) only originate in intestines of warm-blooded animals, birds and other urban animals can be considered important sources of bacteria in stormwater.This assumption was tested by conducted additional monitoring. Comparisons of samples collected from areas prone to urban animal use and those that are not, showed that large overlaps exist between the bacterial concentrations found from both types of areas. Bacterial levels from roofs prone to urban animal use (squirrels and birds) were significantly higher than from roofs not exposed to such use. The other source areas did not show any significant differences in bacterial levels between areas prone and not prone to urban animal use, except for some street areas. This could be the result of a combination of factors, such as the persistence of bacteria in soil, the inadvertent contamination by runoff from other areas frequented by animals, the mobility of small urban animals, or the ubiquitous presence of moderate levels of these organisms in most urban areas. Statistical analyses problems were also caused by periodic very high bacteria values that exceeded the range of the experiments.A further objective of this study was to find how E. coli and enterococci could be effectively used to identifying the presence of inappropriate sanitary sewage in storm drainage systems during dry weather. Many stormwater system managers believe that the presence of indicator bacteria exceeding regulatory levels indicates the likely presence of sanitary sewage. During this study, sewage samples were compared with wet weather and dry weather source area samples (from the project reference sample library). The probability of the sewage and source area sample bacteria levels being significantly different was determined using the Mann Whitney test. When the values of the probabilities were ≤0.05, the diluted sewage sample bacteria levels were determined to be significantly higher as compared to bacterial levels in other source area samples (with a 1 in 20 error level). It was found that the dry-weather outfall samples showing E. coli and enterococci levels higher than 12,000 MPN/100 mL and 5,000 MPN/100 mL respectively, are likely contaminated by sanitary sewage. Levels lower than this can be caused by other sources, such as irrigation runoff, carwash water, or laundry water.Other findings of this research included:Bacteria levels in urban areas are not source limited, i.e. measured bacteria levels did not decrease with increasing amounts of rain, or even with increasing rain intensities. The levels may increase, or decrease, somewhat with time, but stayed generally level.Seasons having low temperatures are associated with decreased bacterial levels.The ratio of E. coli /enterococci was not constant and varied greatly for all conditions.Wet weather samples had mostly higher enterococci levels than E. coli, while dry weather source area samples (such as springs and irrigation runoff) had higher E. coli levels than enterococci levels.Both the indicators followed the same general trend for every site; i.e. both E. coli and enterococci levels increased or decreased simultaneously, although by different amounts.Sewage samples need vigorous agitation before analyses to break up the lumps of fecal matter in which bacteria are present.Samples must be kept refrigerated and analyzed shortly after sample collection. Samples a day old and unrefrigerated can be expected to have decreased bacteria levels compared to chilled and fresh samples.This research was funded as part of a 104(b)3 grant from the U.S. Environmental Protection Agency (Bryan Rittenhouse was the project officer) to the Center for Watershed Protection (under the project management of Ted Brown and Tom Schueler) in 2001. The University of Alabama was a subcontractor to the Center. Sumandeep Shergill conducted much of the research reported in this paper, with the assistance of other graduate students at UA, and his master's thesis reporting this work was accepted by the University in May of 2004.