Scientific Investigations Report 2005-5255
U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2005-5255
Elevated concentrations of fecal bacteria and nitrate in lowland streams have created significant environmental problems in the lower Nooksack River basin of Whatcom County, Washington. A study was conducted to evaluate if ground water was a pathway for contamination to enter surface water streams and to better understand the processes that affect attenuation and degradation of those contaminants at or near the ground water/surface water interface. Reconnaissance surveys were conducted to identify locations of ground-water discharge and four sites were selected in order to evaluate spatial and temporal variability of ground-water discharge in differing settings and scales within the lowland. The information learned at each of the study sites contribute to a better understanding of ground-water/surface water interactions and contaminant transport, which will assist water-resources managers in making better informed decisions.
The reconnaissance survey of the South Fork of the Nooksack River revealed that spatial variability of ground-water discharge is closely related to the geologic materials adjacent to the streambed. When the streambed encounters various geologic materials with differing aquifer properties, a greater degree of variability in ground-water discharge can be expected. However, small-scale variability also can be seen at a more local scale, even when geologic materials are more homogeneous. Ground-water discharge study sites on Fishtrap Creek and Fourmile Creek revealed small-scale variations from one field to the next or from one side of the stream to the other. Such differences may be due to local site conditions such as topography, tile drains, or other impacts like dredging or beaver dams.
Temporal variations in ground-water discharge were much more pronounced. At the mainstem of the Nooksack River near Everson and at Fishtrap Creek, ground-water discharge to surface water was nearly continuous throughout most of the year, except during winter storm events. High river stages caused temporary reversals in vertical hydraulic gradients that lasted for several hours during peak high water conditions. At all ground-water discharge study sites, upward vertical hydraulic gradients varied seasonally with generally more positive gradients in the winter and less positive gradients in the late summer. However, temporary reversals in vertical hydraulic gradients were observed at Fishtrap Creek and the tributary of Bertrand Creek. At these sites, the gradient reversed from positive (ground-water discharge) to negative (ground-water recharge) during the late summer, the driest time of the year. The return to a positive gradient (ground-water discharge) occurred during the autumn shortly after the first rains of the winter wet season. At each of the ground-water discharge study sites, ground-water levels closely mimicked surface-water levels while maintaining a nearly constant hydraulic head difference. This suggests that hydrostatic pressure effects resulting from changing river stage are rapidly transmitted through the streambed even though actual flow of water may be somewhat slower.
Water-quality samples were collected at each of the ground-water discharge study sites in order to better understand bacteria transport and denitrification processes. Although E. coli bacteria were detected in all samples of surface water, it was only detected in a few samples of ground water. This suggests that fecal coliform and E. coli bacteria are effectively removed during ground-water transport before being discharged into the streams. Increasing concentrations of bacteria at downstream sample sites suggests that there may be inputs of contamination between sampling sites. Alternative pathways for bacteria input include overland flow such as the flood event observed on October 19, 2003 and discharge from tile drains. The level of contaminants introduced from tile drains could not be adequately addressed in this study. Because tile drains are difficult to locate and their outlets are often submerged in the stream only three samples were collected, none of which had detectable bacteria concentrations.
The ability of streambed sediments to provide refuge for bacteria was addressed by conducting a laboratory microcosm experiment. The experiment showed that although some bacteria are able to survive for as many as 100 days, more than 99.8 percent died off during the first 65 days. The experiment suggests that overland flow events (floods) during winter months are not the likely cause of high bacteria concentrations in late summer months.
Nitrate contamination of ground water is a well documented problem in the Nooksack River lowland. Although median concentrations of nitrate in ground water are 3.8 mg/L, concentrations in discharging ground water were much lower or absent. Low levels of dissolved oxygen and elevated concentrations of ferrous iron indicate that reducing conditions adjacent to the streambed are sufficient for denitrification to occur. Concentrations of dissolved nitrogen and argon were measured to confirm the presence of excess nitrogen, the end-product of denitrification. In spite of gas-stripping, excess nitrogen gas appears to be present in most samples suggesting that nitrate is being degraded before discharging to surface water. Concentrations of dissolved argon and nitrogen gas measured in samples of ground water indicated that typically 4-8 mg/L and in come cases as much as 16 mg/L nitrogen as nitrate in ground water is denitrified prior to its discharge to surface water.
Data from this study will help water-resources managers focus on effective management strategies for reducing contaminant levels in surface water streams of the Nooksack River lowland. The lack of significant contaminants in discharging ground water and the fact that surface-water contaminant levels increase between some sampling points suggests that there are alternative pathways for contamination to enter streams.
Subsurface tile drains that have been installed beneath some agricultural fields could provide a network of preferential flow paths to surface water and have been shown to be a source of fecal contamination in surface waters in other areas. This study was unable to adequately address this question for the lowland streams of the Nooksack River basin. However, if drains are determined to be a significant source of bacteria in these streams, they also may provide a mechanism to mitigate the problem in areas where they are present. Engineering controls such as a gate valve on the outlet of the drain pipe may provide a mechanism to regulate the transport of bacteria and other contaminants. Controlling the rate of drainage could allow more time for natural die-off process to reduce bacteria concentrations while helping to provide additional ground-water discharge during late summer minimum flow conditions. In addition, controlled drainage also may allow more time for natural denitrification processes to occur thus, improving the quality of water entering the stream.
For more information about USGS activities in Washington, visit the USGS Washington Water Science Center home page .
U.S. Department of the Interior |
U.S. Geological Survey
URL: https://
Page Contact Information:
Page Last Modified:
