|
Data Collection and Analysis 7
<br />Data Collection
<br />Ground-water samples were collected from the 11 wells
<br />semiannually during the spring and fall, following procedures
<br />outlined in Koterba and others (1995). Well water was sampled
<br />using either a submersible pump (no pressure tanks) or through
<br />an outside household spigot (all downstream from pressure
<br />tanks). During each sampling visit, field properties (water
<br />temperature, specific conductance, dissolved-oxygen concen-
<br />tration, pH, turbidity, and acid-neutralizing capacity) were
<br />measured, and ground-water samples were collected for
<br />laboratory analysis.
<br />Surface-water samples from site 12 (Fraser River) were
<br />collected bimonthly following procedures outlined in the USGS
<br />National Field Manual (U.S. Geological Survey, 1997 to
<br />present). Discharge and field properties (water temperature,
<br />specific conductance, dissolved-oxygen concentration, and pH)
<br />were measured, and surface-water samples were collected for
<br />laboratory analysis using the equal-width-increment method.
<br />Ground- and surface-water samples collected for analysis
<br />of dissolved solids, major ions, trace elements, and nutrients
<br />were filtered onsite through a 0.45-gm cellulose filter. Samples
<br />collected for analysis of dissolved organic carbon were filtered
<br />through a 0.45-gm silver filter. The silver filter was saved for
<br />analysis of particulate organic carbon. Samples collected for the
<br />analysis of suspended sediment and total recoverable phospho-
<br />rus were not filtered. Samples collected for nutrient and dis-
<br />solved and particulate organic carbon analysis were chilled to
<br />4°C, and all samples except suspended sediment were trans-
<br />ported to and analyzed at the USGS National Water Quality
<br />Laboratory (NWQL) in Lakewood, Colo. Suspended-sediment
<br />samples were analyzed for suspended-sediment concentration
<br />at the USGS Iowa Sediment Laboratory in Iowa City, Iowa. The
<br />procedures used for processing and preservation of the water-
<br />quality samples are described in Koterba and others (1995) and
<br />in the USGS National Field Manual (U.S. Geological Survey,
<br />1997 to present).
<br />Unfiltered samples for the analysis of bacteria (total
<br />coliform, Escherichia coli [E. coh], and fecal coliform) were
<br />collected in the field using sterilized equipment. Bacteria sam-
<br />ples were collected last, and the samples were processed within
<br />6 hours of collection. Collection and processing were done
<br />using m-ENDO and NA-MUG media for ground water and m-
<br />FC media for surface water (Britton and Greeson, 1989; U.S.
<br />Environmental Protection Agency, 1991; American Public
<br />Health Association, American Water Works Association, and
<br />Water Pollution Control Federation, 1999).
<br />For MBAS, unfiltered water samples were collected and
<br />chilled until analyzed. Samples for the analysis of radon-222
<br />were collected by withdrawing 10 mL of water into a glass
<br />syringe and injecting this water into a glass scintillation vial
<br />beneath mineral oil (U.S. Geological Survey, 1997 to present).
<br />After sampling at each site, all equipment except that for
<br />bacteria sampling was cleaned with a 0.1-percent nonphosphate
<br />detergent solution and then rinsed with tap water and distilled
<br />water. Tubing was stored in clean, sealable plastic bags, and
<br />organic carbon and radon equipment was reassembled, wrapped
<br />in aluminum foil, and stored in clean containers. Bacteria sam-
<br />pling equipment was washed, wrapped in craft paper, and then
<br />sterilized in the laboratory by autoclaving at 121 °C at 15 lb/in2
<br />for 20 minutes.
<br />Data Analysis
<br />Analysis of ground- and surface-water data collected for
<br />this study was conducted in a two-step process. First, ground-
<br />and surface-water samples were analyzed at the USGS NWQL.
<br />Then these results, along with the ground- and surface-water
<br />field properties, were analyzed using various statistical
<br />methods.
<br />Laboratory Analysis
<br />At the USGS NWQL, the ground- and surface-water sam-
<br />ples were analyzed for concentrations of dissolved solids, major
<br />ions, trace elements (iron and manganese), nutrients (ammonia
<br />[un-ionized ammonia (NH3) plus ammonium (NH4')]), total
<br />and(or) dissolved ammonia plus organic nitrogen, nitrite, nitrate
<br />plus nitrite (hereinafter nitrate), total or dissolved phosphorus,
<br />and orthophosphate), and dissolved organic carbon. The nutri-
<br />ents represent concentrations as either nitrogen or phosphorus.
<br />Ground-water samples also were analyzed for MBAS and
<br />radon-222, and surface-water samples also were analyzed for
<br />particulate organic carbon. Except for these latter three constit-
<br />uents plus bacteria and constituents as total concentrations, all
<br />samples were in the dissolved form. Samples for major ions and
<br />nutrients were extracted and analyzed by various methods
<br />(Fishman and Friedman, 1989; Fishman, 1993; U.S. Environ-
<br />mental Protection Agency, 1993). Samples for dissolved
<br />organic carbon were analyzed by UV-promoted persulfate oxi-
<br />dation and infrared spectrometry (Brenton and Arnett, 1993),
<br />and samples for particulate organic carbon were analyzed by
<br />wet-chemical oxidation and infrared spectrometry (Burkhardt
<br />and others, 1997). Samples for MBAS were analyzed using pro-
<br />cedures described by Burkhardt and others (1995), and samples
<br />for radon-222 were analyzed by liquid scintillation counting
<br />(American Society for Testing and Materials, 1999).
<br />The analytical results from the NWQL were reported as
<br />either the minimum reporting level (MRL) or laboratory report-
<br />ing level (LRL), along with the long-term method detection
<br />level (LT-MDL). The minimum reporting level is defined by
<br />the NWQL as the smallest measured concentration of a constit-
<br />uent that may be reliably measured using a given analytical
<br />method (Timme, 1995). For some constituents, the reporting
<br />level was changed at the beginning of a water year from the
<br />MRL to the LT-MDL and LRL (Childress and others, 1999).
<br />The LT-MDL is the lowest concentration of a constituent that is
<br />reported by the NWQL and represents that value at which the
<br />probability of a false positive is statistically limited to less than
<br />or equal to 1 percent. The LRL is calculated as 2 times the
<br />LT-MDL and represents the value at which the probability of a
|