Abstract

The U.S. Geological Survey, in cooperation with the City of Omaha, investigated the water quality of combined sewer overflows, stormwater, and streams in the Omaha, Nebraska, area by collecting and analyzing 1,175 water samples from August 2006 through October 2007. The study area included the drainage area of Papillion Creek at Capeheart Road near Bellevue, Nebraska, which encompasses the tributary drainages of the Big and Little Papillion Creeks and Cole Creek, along with the Missouri River reach that is adjacent to Omaha. Of the 101 constituents analyzed during the study, 100 were detected in at least 1 sample during the study. Spatial and seasonal comparisons were completed for environmental samples. Measured concentrations in stream samples were compared to water-quality criteria for pollutants of concern. Finally, the mass loads of water-quality constituents in the combined sewer overflow discharges, stormwater outfalls, and streams were computed and compared.

The results of the study indicate that combined sewer overflow and stormwater discharges are affecting the water quality of the streams in the Omaha area. At the Papillion Creek Basin sites, Escherichia coli densities were greater than 126 units per 100 milliliters in 99 percent of the samples (212 of 213 samples analyzed for Escherichia coli) collected during the recreational-use season from May through September (in 2006 and 2007). Escherichia coli densities in 76 percent of Missouri River samples (39 of 51 samples) were greater than 126 units per 100 milliliters in samples collected from May through September (in 2006 and 2007). None of the constituents with human health criteria for consumption of water, fish, and other aquatic organisms were detected at levels greater than the criteria in any of the samples collected during this study. Total phosphorus concentrations in water samples collected in the Papillion Creek Basin were in excess of the U.S. Environmental Protection Agency’s proposed criterion in all but four stream samples (266 of 270). Similarly, only 2 of 84 Missouri River samples had total phosphorus concentrations less than the proposed criterion. The proposed total nitrogen criterion for the Corn Belt and Northern Great Plains ecoregion was surpassed in 80 percent of the water samples collected from the stream sites. Samples with total nitrogen concentrations greater than the proposed criterion were most common at Papillion Creek and Big Papillion Creek sites, where the proposed criterion was surpassed in 90 and 96 percent of the samples collected, respectively. Elevated concentrations of total nitrogen were less common at the Missouri River sites, with 33 percent of the samples analyzed having concentrations that surpassed the proposed nutrient criterion for total nitrogen. The three constituents with measured concentrations greater than their respective health-based screening levels were nickel, zinc, and dichlorvos.

Differences in water quality during the beginning, middle, and end of the combined sewer overflow discharge and the stream hydrograph rise, peak, and recession were investigated. Concentrations from the ending part of the combined sewer overflow hydrograph were significantly different than those from the beginning and middle parts for 3 and 11 constituents, respectively. No constituents were significantly different between the beginning and middle parts of the combined sewer overflow discharge hydrograph. For the stream site upstream from combined sewer overflow outfalls on Cole Creek, the constituents with geometric mean values for the hydrograph rise that were at least twice those for the values of the peak and recession were specific conductance, magnesium, nitrite, N,N-diethyl-meta-toluamide (DEET), methyl salicylate, p-cresol, and Escherichia coli. Similarly, the constituents where the hydrograph peak was at least twice that for the rise and recession at the upstream Cole Creek site were total suspended solids, silver, and benzo[a]pyrene. At the Papillion Creek Basin stream sites downstream from combined sewer overflows, the only constituent with concentrations for the hydrograph rise that were significantly different than those for the peak and recession part was specific conductance, although this is a result of groundwater having larger specific conductance than stormwater runoff; therefore, the values decrease as there is more stormwater in the sample (at the peak). No constituents had significantly different concentrations when comparing the peak with the other two hydrograph parts and the recession with the other two hydrograph parts of the storm hydrograph at Papillion Creek Basin sites downstream from combined sewer overflows. At the Papillion Creek Basin stream sites downstream from combined sewer overflows, constituents with significantly larger concentrations in hydrograph-peak samples relative to those from one of the other two parts of the hydrograph were mostly determined to be derived more from stormwater and included benzo[a]pyrene, fluoranthene, phenanthrene, and pyrene. Mercury was the only compound with significantly larger concentrations in the hydrograph-peak samples compared to those from the hydrograph rise that was determined to be derived more from sewage.

The ratio of combined sewer overflow load to receiving stream load was computed for those samples where a concentration and discharge were measured. The higher this ratio, the more of an effect the combined sewer overflow discharge will have on the concentration in the receiving stream. The combined sewer overflow site at 64th and Dupont Street (CSO205) had 88 constituents that had the maximum combined sewer overflow load to receiving stream load ratio for a single sample. The combined sewer overflow site with the most constituents that had the largest geometric mean combined sewer overflow load to receiving stream load ratio also was site CSO205, with 78 constituents. The tributary stream load to the receiving stream load also was computed for those Papillion Creek Basin samples where the constituent of interest was detected in the downstream site of a receiving-stream reach and discharge was measured. The higher the value of this ratio, the larger the effect a tributary stream will have on the concentration in the receiving stream. Of the 38 constituents that were detected in streams in the samples used for these calculations, the maximum single-sample tributary stream load to the receiving stream load ratio for 22 constituents was found in samples from Big Papillion Creek. Big Papillion Creek had the largest geometric mean for this ratio for each of the 38 constituents for which the ratio could be calculated.

Based upon the data collected in this study and analysis using a custom weighting scenario for summarizing the results, the combined sewer overflow sites where implementation of additional controls or the best management practices could potentially achieve the largest water-quality effect for the greatest number of compounds in receiving streams are the combined sewer outfalls at 64th and Dupont Street (CSO205), 69th and Evans Street (CSO203), Monroe Street (CSO119), and the North Interceptor Combined Sewer Overflow (CSO106). Scores from applying a similar custom weighting scenario to stream data indicated that implementing best management practices potentially could have the largest water-quality effect on the greatest number of compounds in Big Papillion Creek, Papillion Creek, Cole Creek, and Little Papillion Creek (in that order).

Posted October 27, 2009