Water Quality Variations and Trends in Monument and Fountain Creeks

Home Browse Search

h Flow in Fountain Creek is perennial from Colorado Springs to Fountain. In this reach, the streamflow is augmented by discharge of wastewater effluent from Colorado Springs, Garden Valley, Security, Fort Carson, Widefield, and Fountain (Edelmann and Cain, 1985). The percentage of annual flow contributed by wastewater effluent does not clearly explain the effects of wastewater effluent on the flows in Fountain Creek (Edelmann and Cain, 1985). Short periods of heavy runoff in Fountain Creek contribute a large volume to its annual flow, while flows of wastewater effluent are more constant. Downstream from Fountain, no -flow periods may occur during the summer because of agricul- tural diversions. Most of the land use between Colorado Springs and Pueblo is agricultural. Mean annual streamflow increases downstream and ranges from 14.6 to 72.1 ft /s for the period of record at three continuous streamflow- gaging stations: Fountain Creek near Colorado Springs, Fountain Creek at Colorado Springs, and Fountain Creek at Pueblo. SUMMARY STATISTICS For the nine stations where water - quality data were collected for this study, seven statistical parameters that define the distribution of the data were determined for streamflow and all water - quality properties and constitu- ents that had a sufficient number of values. These parameters were the mean, standard deviation, minimum, lower quartile (25th percentile), median (50th percentile), upper quartile (75th percentile), and maximum. The criteria for a sufficient number of values were defined as follows: (1) At least 10 values equal to or exceeding the detection limit, and (2) no more than 75 percent of the values less than the detection limit (censored values). If either cri- terion was not satisfied, only the minimum and maximum values were reported to indicate the range of measured values. Because the data for many properties and constituents included censored (less -than) values (because of improvement in analytical methods over time, resulting in a lowering of the detection limits), the methods used to compute the summary statistics were selected based on their accuracy for estimating specific distributional.parameters for data sets containing censored values. Gilliom and Helsel (1986) determined that the log probability regression was the best method to estimate the mean and standard deviation of such data sets and that a lognormal maximum- likelihood method was best for estimating the median and quartile values. The software used in this study to compute the summary statistics incorporated these two methods, modified to accept data sets that had more than one detection limit (censored values) (Helsel and Cohn, 1988). Summary statistics for the data compiled for the nine stations in this study are listed in tables 4 through 12 in the "Hydrologic Data" section at the back of the report. Also included in the tables are the applicable State water - quality standards that were in effect in 1989 (Colorado Department of Health, 1989). The standards vary in different stream segments. The standards are maximum allowable concentrations, except for the pH standard, which is an allowable range, and the dissolved- oxygen standard, which is a minimum allow- able concentration. Where two values are listed for the dissolved - oxygen standard, the larger value applies during periods of spawning of cold -water fish. The standards for most trace metals were based on total recoverable concentration except for iron and manganese, which have standards for dis- solved and total recoverable concentrations. T