Water Quality Variations and Trends in Monument and Fountain Creeks

Home Browse Search

0137 stations 07105500 Fountain Creek at Colorado Springs and 07105905 Fountain Creek below Fountain. The median concentration of most constituents at station 07105500 Fountain Creek at Colorado Springs, downstream from the mouth of Monument Creek, is more similar to the median concentration of most constituents at station 07104905 Monument Creek at Bijou than to the median concentration of most constituents at station 07103700 Fountain Creek near Colorado Springs, indicating that the water quality of Monument Creek has a large effect on the water quality of Fountain Creek. To determine the effect of the Colorado Springs Wastewater- Treatment Plant on the water quality of Fountain Creek, station 07105500 Fountain Creek at Colorado Springs, upstream from the wastewater - treatment plant, and station 07105530 Fountain Creek below CSWWTP were compared. Tukey's test results indicated that there were no statistically significant differences in the water quality between the two stations for the following property and constituents: water temperature, nitrite plus nitrate as nitrogen, and fecal coliform bacteria. The following properties and constituents had statisti- cally significant increases downstream: instantaneous streamflow, specific conductance, total ammonia as nitrogen, un- ionized ammonia as nitrogen, total recoverable copper, total recoverable zinc, and 5 -day BOD. The following properties and constituents had statistically significant decreases down- stream: pH, dissolved oxygen, suspended solids, and total recoverable iron. Time - Series Trends Time - series trends at each station were analyzed to determine if changes in the values of water - quality properties and concentrations of constituents had occurred over time. Factors such as land use, water use, and climate in the basin can affect water quality. Detection of temporal trends in water quality can indicate changes in the factors that affect water quality. Trend analysis of time - series data for water - quality properties and constituents is complicated by several common characteristics of the data: nonnormality, seasonality, serial dependence, and censoring. Nonnormal data cannot be described by a symmetrical, unimodal, bell- shaped distribution. Seasonal data have a natural sequential order over time and vary, depending on the time of year. Water- quality data often are serially dependent; the constituent concentration at one point in time is dependent upon and related to prior data. Censored data contain less -than values due to the detection limits of the analytical methods. The seasonal Kendall test, which was used in this study to detect temporal trends in water quality, is based on methods developed by the U.S. Geological Survey (Hirsch and Slack, 1984). The seasonal Kendall test is a statistical technique unaffected by the problem characteristics described above (Hirsch and others, 1982; Hirsch and Slack, 1984). This technique is used to identify statistically significant monotonic changes (only increasing or only decreasing trends) in data over time. The technique also provides an estimate of the magnitude and direction of the change, which can be used to calculate the percent change in the median constituent concentration for the period of record. The seasonal Kendall test is a nonparametric statistical technique; the test statistic is determined by using ranks of the data rather than the actual data. Nonnormality of the data set and censored values are 13