The U.S. Geological Survey (USGS) operates a network of streamflow-gaging stations in Texas that provides discharge data used for water-management decisions and various other purposes. Operating stations at all locations where discharge data are needed is not feasible, but the statistical characteristics of the network station data can be used to estimate discharge characteristics at ungaged sites. Regionalization techniques such as regression analyses relate discharge-frequency characteristics to selected physical and climatic characteristics of drainage basins. A particular discharge-frequency characteristic that can be regionalized is the minimum 7-day, 2-year discharge1 (7Q2). In Texas, the 7Q2 is used at stream sites to analyze permit applications for water allocation, water-supply planning, aquatic maintenance (instream flow) requirements, and waste-load allocation for point and nonpoint source discharges.
In 1997 and 1998 the USGS, in cooperation with the City of Houston and the Houston-Galveston Area Council, conducted a study to estimate low-flow discharge characteristics for selected streams in southeast Texas. The objectives of this study were to estimate the 7Q2 for 17 ungaged stream sites (fig. 1) where such data were unavailable and to summarize selected water-quality properties and constituents at each of the ungaged sites during low-flow conditions. Discharge measurements were made and water-quality data were collected using USGS methods as described by Buchanan and Somers (1969), Rantz and others (1982), and Wells and others (1990).
This fact sheet presents estimated 7Q2 for the 17 ungaged sites and statistical summaries of four water-quality properties and three water-quality constituents for the sites. The ungaged sites are located on streams in or near the Houston metropolitan area in southeast Texas (fig. 1). For this study, 85 low-flow discharge measurements made at the 17 ungaged sites were used in conjunction with continuous-record discharge data from 10 USGS streamflow-gaging stations to estimate the 7Q2 for the ungaged sites.
Estimation of Minimum 7-Day, 2-Year Discharge
One technique to estimate low-flow discharge characteristics (such as 7Q2) in a region uses data from one or more index stations (Riggs, 1972; Stedinger and Thomas, 1985). These typically are USGS stations with a continuous record of discharge for a length of time sufficient (minimum 8 years) to provide reliable low-flow discharge characteristics. For an index station to be used in the analysis, the low-flow data should not have any significant temporal trends, which would skew the frequency characteristics. To test for a trend in the data, the Mann-Kendall test (Helsel and Hirsch, 1992) was used to determine the significance of Kendall’s tau. In this application, a Kendall’s tau p-value less than or equal to 0.10 indicated a statistically significant trend in the data. If a trend was detected in the data set, the record was truncated to a shorter period without a trend.
Initially, the 7-day minimum flow was determined for each year the index station has been operated (or for the shorter period if the data set had a trend). The 7Q2 for the index station was determined by computing the median discharge for the station time series. The median is the 50th percentile and by definition is the discharge with a 2-year recurrence interval, on the basis of station data and assuming no specified statistical distribution of the data set.
To transfer the 7Q2 from the index station to the ungaged site, a relation was developed between low-flow discharge at the station and at the ungaged site. Periodic discharge measurements were made during various low-flow conditions at the ungaged sites where the 7Q2 was to be estimated. At least five discharge measurements were made at each site. The low-flow discharge measurements then were used with concurrent discharge values determined at the appropriate index station(s) to develop the relation between low-flow discharge characteristics for the ungaged site and for the index station(s). An example of such a relation is shown in figure 2.
In figure 2, discharge measured at an ungaged site (Peach Creek at FM 1485; site 4 in fig. 1, table 1) is plotted on the ordinate (y-axis), and daily mean discharge for an index station (08070500 Caney Creek near Splendora, Tex.; station 103 in fig. 1) corresponding to the date of the discharge measured at the ungaged site is plotted on the abscissa (x-axis). Two relations are developed using these data pairs; a line of ordinary least squares and a line of organic correlation. The line of ordinary least squares minimizes errors in the y direction only and defines the best-fit straight line as the line that minimizes the sum of the squares of the distances of the data points from the best-fit straight line (Helsel and Hirsch, 1992, p. 275). The line of organic correlation minimizes errors in both the x and y directions and defines the best-fit straight line as the line that minimizes the sum of the areas of right triangles formed by horizontal and vertical lines extending from observations to the fitted line (Helsel and Hirsch, 1992, p. 276). In most instances, the line of organic correlation is the preferable statistical tool for 7Q2 estimation because the line of ordinary least squares tends to produce biased, in some cases highly biased, 7Q2 estimates (W.H. Asquith, U.S. Geological Survey, written commun., 1999).
The computed 7Q2 for the index station (fig. 2) is shown as a dashed line that perpendicularly intersects the abscissa at the 7Q2 value. To estimate the 7Q2 for the ungaged site, the intersection of the line of organic correlation and the index station 7Q2 line is located. A horizontal line is projected from this intersection to the ordinate, and the estimated 7Q2 for the ungaged site is determined from the scale. Although the line of organic correlation is preferable, the ordinary least-squares line can be used if, from observation, the least-squares line clearly and unequivocally fits the data better in the region of the graph for which a 7Q2 estimation is required.
This technique was used to estimate the 7Q2 for each of the 17 ungaged sites. If more than one index station was identified for the analysis for a particular ungaged site, multiple 7Q2 estimates were determined. Multiple 7Q2 estimates for an ungaged site were averaged to determine a final 7Q2 estimate. The final 7Q2 estimate for each ungaged site is listed in table 1. The index stations used for these analyses are listed in table 2.
Statistical Summary of Water-Quality Data
Selected water-quality data also were determined at each of the 17 ungaged sites. During site visits for low-flow discharge measurements, USGS personnel measured water temperature, specific conductance, pH, and dissolved oxygen with field meters and collected water-quality samples from the centroid of flow. These samples were analyzed by the City of Houston Wastewater Quality Control Laboratory for concentrations of total suspended solids, hardness, and dissolved chloride. Statistical summaries of these water-quality data are listed in table 3.
These water-quality data (specifically total suspended solids, hardness, and dissolved chloride) are used in trace metal limit determinations as part of the waste-load allocation process. For instance, the limiting criteria for several trace metals, such as cadmium, copper, and lead, are defined as a function of hardness. These data typically are collected during various flow conditions, which can affect concentrations.
Buchanan, T.J., and Somers, W.P., 1969, Discharge measurements at gaging stations: U.S. Geological Survey Techniques of Water-Resources Investigations, book 3, chap. A8, 65 p.
Helsel, D.R., and Hirsch, R.M., 1992, Studies in environmental science 49—Statistical methods in water resources: Amsterdam, Elsevier, 522 p.
Rantz, S.E., and others, 1982, Measurement and computation of streamflow—Volume 1. Measurement of stage and discharge: U.S. Geological Survey Water-Supply Paper 2175, 284 p.
Riggs, H.C., 1972, Low-flow investigations: U.S. Geological Survey Techniques of Water-Resources Investigations, book 4, chap. B1, 18 p.
Stedinger, J.R., and Thomas, W.O., Jr., 1985, Low-flow frequency estimation using base-flow measurements: U.S. Geological Survey Water-Resources Investigations Report 85–95, 83 p.
Wells, F.C., Gibbons, W.J., and Dorsey, M.E., 1990, Guidelines for collection and field analysis of water-quality samples from streams in Texas: U.S. Geological Survey Open-File Report 90–127, 79 p.