Scientific Investigations Report 2006-5073
U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2006-5073
Seven sites on the Union River were sampled during summer base-flow conditions from June 28–30, 2004 (fig. 1). Base-flow samples also were collected at eight sites on the Tahuya River from July 1 to 6, 2004, and eight sites on the Skokomish River from July 9 to August 4, 2004. Samples were collected from Union River, Mission Creek, Tahuya River, and three smaller drainage basins in the spring (March 18 and 25, 2004).
Streamflow was measured, whenever possible, following established protocols and procedures (Rantz and others, 1982). Streamflow was measured infrequently in channels in which wetted widths were minimal, requiring less than optimal sectioning; depths and velocity may have been below proper operating range for the meter available, resulting in reasonable departures from established protocols.
Water temperature, dissolved-oxygen (DO) concentrations, specific conductance, pH, alkalinity, and bicarbonate concentrations were measured on-site following procedures described by Wilde (variously dated). Water temperature and DO were measured directly in the stream and specific conductance and pH were measured using unfiltered, composited water; and alkalinity and bicarbonate were measured using a filtered aliquot of composited water. All samples for analyses of nutrients and suspended sediment, and base-flow samples for the analyses of major ions and δ15N were collected using a US DH-81 sampler (Wilde and others, 1999). The sampler holds a 1-liter polypropylene sample bottle, and all parts of the sampler that come into contact with sample water were made of polypropylene. All equipment used to collect and process samples was cleaned with a 0.2‑percent non-phosphate detergent, rinsed with deionized water, stored in a dust-free environment, and then rinsed thoroughly with stream water before sample collection and processing (Wilde, 2004; Wilde and others, 2004).
Samples were collected using the equal-width-increment (EWI) method, in which transects were established across the width of each stream/river. Water was collected at approximately 10 equally spaced intervals along the transect by lowering and raising the sampler vertically through the water column. The collected water from each interval was then composited into a polypropylene churn splitter, and then split into individual samples (Wilde and others, 2004). The pH and specific conductance were measured using unfiltered water from the churn splitter. Alkalinity and bicarbonate concentrations during base-flow sampling were measured with aliquots of filtered water using incremental titration with a digital titrator and sulfuric acid. Stream-water samples for organic carbon were collected at a single vertical column in the center flow of the stream using baked, amber glass bottles, which were lowered beneath the water surface until filled.
Samples collected for inorganic chemistry and δ15N chemical analyses were filtered through a 0.45-µm pore‑size, polypropylene-encapsulated filter. The filtrate for cation analyses was preserved with nitric acid. Samples for δ15N analyses were preserved by freezing, and samples for analyses of nutrients were filtered into opaque polyethylene bottles and preserved at less than 4oC. For organic carbon analyses, three aliquots of water were filtered through 25-mm diameter, 0.45‑µm, baked, glass-fiber filters using a Teflon™ filter holder. The filtrate was preserved with sulfuric acid for analysis of dissolved organic carbon (DOC) (Wilde and others, 2004), and the material retained on the three filters was used for particulate carbon and nitrogen analyses.
All samples for chemical analyses were shipped overnight on ice to the USGS National Water Quality Laboratory (NWQL) in Lakewood, Colorado. Samples for δ15N analyses were shipped frozen to the USGS Menlo Park Stable Isotope Laboratory in Menlo Park, California, and samples for analysis of suspended-sediment concentration were shipped to the USGS Cascades Volcano Observatory (CVO) Sediment Laboratory in Vancouver, Washington.
Samples of filtered orthophosphate, ammonia, and nitrate-plus-nitrite nitrogen were analyzed at the NWQL by colorimetry (Fishman, 1993). Ammonia and orthophosphate concentrations generally were less than the detection limit of 0.04 and 0.02 mg/L, respectively, and will not be addressed in the section “Surface-Water Quality in Rivers and Drainage Basins Discharging to the Southern Part of Hood Canal.” The concentration of a filtered sample of total nitrogen, the sum of inorganic and organic nitrogen, was analyzed by an alkaline persulfate digestion followed by colorimetry (Patton and Kryskalla, 2003). Major ions, iron, and manganese were analyzed by ion chromatography and inductively coupled plasma (Fishman and Friedman, 1989; Fishman, 1993). Particulate and dissolved organic carbon were analyzed by ultra violet (UV)-promoted persulfate oxidation and infrared spectrometry using the methods of U.S. Environmental Protection Agency (1997) and Brenton and Arnett (1993). Samples were analyzed for sediment concentrations at CVO following the method of Guy (1969). The nitrogen isotope samples were analyzed for δ15N of nitrate using a method in which bacterial cultures reduce dissolved nitrate to N2O (Sigman and others, 2001; Casciotti and others, 2002). Nitrogen isotopic compositions are expressed in per mil (percent) relative to atmospheric air:
,
(1)
where
 |
is |
the sample, and |
 |
is |
atmospheric nitrogen. |
The δ15N of nitrate can provide information about nitrogen sources. A δ15N greater than 10 per mil indicates possible manure or septic waste contamination. However, most of the δ15N data that were collected during base-flow conditions fall within the range covered by most sources of nitrate (1 to 5 per mil), thus precluding the identification of the nitrogen source (Kendall, 1998).
Quality-control samples included field blanks and replicates (table 5, at back of report). Low concentrations of calcium, suspended sediment, and DOC were detected in some of the blanks. Calcium detections in the blank samples were far less than the concentrations in the environmental samples. But the highest DOC and suspended-sediment concentrations in the blank samples were 0.4 and 1 mg/L, respectively, similar to the lowest concentrations in the environmental samples. Variability in the chemical data was measured with two pairs of replicate samples. The amount of variability between samples was expressed as the relative percentage of difference in the concentrations. Relative percentage of differences for the major‑ion samples ranged from 0 to 7 percent. Relative percentage of differences between replicates of nutrient samples tended to be larger than major ions, ranging from 0 to 22 percent. Relative percentage of differences between replicates of DOC ranged from 0 to 15 percent and 40 percent for suspended organic carbon. Changes were not made to any of the inorganic or organic chemical data sets on the basis of these replicate data.
The relative percentage of differences between replicates of δ15N ranged from 3 to 13 percent. Although δ15N had good reproducibility, some of the data are highly uncertain because instrument response was less than detection limits. These data will not be included in the discussion in the section, “Surface‑Water Quality in Rivers and Drainage Basins Discharging to the Southern Part of Hood Canal.”
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
Persistent URL: https://pubs.water.usgs.gov/sir20065073
Page Contact Information: Publications Team
Page Last Modified: Thursday, 01-Dec-2016 19:04:07 EST
