Contribution of forested Watersheds and Beaver Ponds to Fecal Contamination of Surface Waters in Vermont

    Fecal indicators are used to predict the risk of human illness due to pathogens transmitted through fecal matter. Few studies have examined baseline levels of the fecal indicator E. coli coming from undeveloped, forested watersheds. Obtaining background levels of fecal indicators, such as E. coli, is an important component of understanding sources of fecal pollution and is critical to the assessment of proposed changes to water quality standards in Vermont. Evaluating potential sources of fecal contamination to undeveloped watersheds, such as beaver ponds, is also an important contribution to our current understanding.     The objectives of this study were to 1) quantify the baseline levels of E. coli in undeveloped, forested watersheds, 2) examine differences between non-storm and storm concentrations of E. coli, 3) evaluate the relationship between stream discharge and E. coli concentration, 4) examine the use of load as a metric for fecal pollution, and 5) evaluate the contribution of beaver ponds to fecal contamination of streams. A total of 17 undeveloped sub-watersheds within the Mad River Valley, VT, were sampled for E. coli during the summers of 2001 and 2002. Fifteen beaver ponds sites were also sampled during August – September of 2003.

    Results from this study verified that during non-storm events, E. coli concentrations were consistently low (95% C. I. = 3.8 – 7.0 organisms/100 mL). However, during storm events, E. coli concentrations were significantly higher (95% C.I. = 20.9 – 66.4 organisms/100 mL). Similarly, Vermont Water Quality Standard violations (> 77 organisms/100 mL) occurred 1% of the time for all non-storm events compared to 34% of the time for all storm events. The violation rate during storms was even higher (50%) for 2002, the wetter sampling season. Although E. coli concentrations were higher during storm events, a clear relationship between streamflow and E. coli concentrations was not found. Areal load calculations that combine concentration with streamflow measurements and watershed area, estimate that forested land in the Mad River Valley contributes 1.6 – 5.2% of the total fecal load during non-storm weather.

    Beaver ponds, a hypothesized source of E. coli, were shown to have consistently higher concentrations in the outlet water during non-storm events than those found in other streams draining undeveloped, forested watersheds. E. coli concentrations in beaver pond outlet water sampled during storm events were also significantly higher than non-storm concentrations. Data from a very large storm (56 mm) measured in 2002 indicate that concentrations can be 12 -15 higher than those measured in 2003. Additionally, water samples collected following disturbance of the pond sediments had higher concentrations than those found in the water column, indicating that sediment may constitute a large reservoir for fecal contamination.