StreamStats in Illinois
The three reports below document (1) the implementation of StreamStats for Illinois and (2) development of the regression equations that are available in the Illinois StreamStats application for estimating flood-peak discharges of selected recurrence intervals for rural and urban streams in Illinois. The first report describes how StreamStats was implemented for Illinois, tests done to assess the accuracy of the application, and adjustments made to measurements of some basin characteristics to avoid bias in the flow estimates. The second and third reports describe the methods used to measure the basin characteristics and to develop the rural and urban peak-flow equations, respectively, references to GIS data layers used in the analysis, and the errors associated with the estimated discharges. Users of the StreamStats application should familiarize themselves with all three reports before using StreamStats to obtain estimates of streamflow statistics for ungaged sites.
- Ishii, A.L., Soong, D.T., and Sharpe, J.B., 2010, Implementation and evaluation of the Streamflow Statistics (StreamStats) Web application for computing basin characteristics and flood peaks in Illinois: U.S. Geological Survey Scientific Investigations Report 2009–5197, 25 p.
- Soong, D.T., Ishii, A.L., Sharpe, J.B., and Avery, C.F., 2004, Estimating flood-peak discharge magnitudes and frequencies for rural streams in Illinois: U.S. Geological Survey Scientific Investigations Report 2004-5103, 147 p.
- Over, T.M., Saito, R.J., Veilleux, A.G., Sharpe, J.B., Soong, D.T., and Ishii, A.L., 2017, Estimation of peak discharge quantiles for selected annual exceedance probabilities in northeastern Illinois (ver. 2.0, November 2017): U.S. Geological Survey Scientific Investigations Report 2016–5050, 50 p. with appendix.
Issues with implementation of urban peak-flow equations from Over and others (2016)
When attempting to obtain peak-flow estimates for user-selected sites in Illinois, users may choose whether to obtain estimates of rural or urban peak flows, or both. By default, if users request to obtain both rural and urban estimates, then StreamStats outputs will provide tables of the estimates from both sets of equations. For hydrologic region 2, in northeastern Illinois, the rural peak-flow equations from Soong and others (2004) were replaced by a single set of equations from Over and others (2016) that are applicable for rural and urban peak flows, but are presented in the urban peak-flow statistics tables in StreamStats outputs. As a result, if users choose to obtain rural peak-flow estimates from StreamStats for a site in hydrologic region 2, under the “Rural Peak Flow Basin Characteristics” table in the output they will see a message saying “The selected watershed is entirely in an area for which flow equations were not defined.” See figure 1, page 3 of Over and others (2016) for a map of region 2.
Over and others (2016) provided a regression equation for estimating the urban 200-year (0.05 percent probability) peak flow for hydrologic region 2, but it is not implemented in StreamStats. As the other hydrologic regions do not have equations for estimating that statistic, inclusion of the hydrologic region 2 equation in StreamStats would have created problems that would have resulted in StreamStats providing no urban peak-flow estimates when at least part of a delineated basin was in hydrologic region 2. Users who want estimates of the 200-year peak flow can solve the equation manually using basin characteristics that they can obtain from StreamStats.
Normally, when the delineated basin for a user-selected site has drainage area that is in more than one hydrologic region, StreamStats will provide estimates for the site by applying the equations for each hydrologic region as if the entire basin was in that region, and StreamStats also will provide area-averaged estimates. The area-averaged estimates are determined by multiplying the peak-flow estimates determined for each hydrologic region by the proportion of the delineated basin that is in each region and then summing to obtain the final weighted estimates. For Illinois, there are two issues that can occur with this process when a user-selected site has a drainage area that is partly in hydrologic region 2 and partly in other hydrologic regions:
1. Rural peak-flow estimates.-- If users choose to obtain rural peak-flow estimates for such sites, StreamStats outputs will provide tables of flow estimates using only the equations for the hydrologic regions outside hydrologic region 2. No area-averaged estimates will be provided. The rural peak-flow estimates provided using only the equations for the parts of the delineated basin that are outside of hydrologic region 2 may not be indicative of the rural peak flows that could be expected from the entire basin. Hydrologic judgement should be used when considering use of such estimates in decision making.
2. Urban peak-flow estimates.-- With urban peak-flow estimates, indicators of the errors associated with the estimates, including prediction errors and 90-percent prediction intervals, are available only for hydrologic region 2. In the output tables of urban peak-flow estimates from the equations for hydrologic regions other than hydrologic region 2, the columns that normally would contain the error indicators will be blank. This causes a problem when StreamStats attempts to compute error indicators for the area-averaged flow estimates. StreamStats erroneously uses zeros in the weighting scheme for the error indicators when no values are available. Consequently, the final error indicators for the area-averaged flow estimates are erroneous. The actual errors associated with the area-averaged flow estimates are unknown. This bug will be corrected in a future release of StreamStats.
Revision of Over and others (2016)
Subsequent to publication of the initial Illinois urban peak-flow regression report, SIR 2016–5050 version 1, in June, 2016, an error was found in the reported methodology. It was discovered that the at-site flood frequencies computed using so-called “weighted” skew coefficient values were not actually computed using the weighted skew; that is, as a weighted average of the station and regional skew values. Instead, the flood frequencies were computed using the regional skew only.
The preferred skew coefficients to use for computing at-site flood frequencies are the weighted skews; however it was determined that the flood frequencies obtained from using the regional skew alone have only modest differences from those obtained using the weighted skews. As a result, it was decided to correct the report on an interim basis, creating SIR 2016–5050 version 2, to reflect the methodology actually implemented. The revised report was released on November 28, 2017, at https://doi.org/10.3133/sir20165050.
An update of the analysis and report are planned for October, 2018, which will use the weighted skew values and include additional years of record.
Improved estimates by use of the drainage-area ratio method
Improved estimates of rural peak flows for an ungaged site can be obtained by use of the drainage-area ratio method when the drainage area for the ungaged site is within ± 50 percent of the drainage area for an upstream or downstream streamgaging station. Methods described on page 26 of Soong and others (2004) can be applied to obtain these estimates. Users can easily compute these estimates using StreamStats-generated drainage areas and the methods from Soong and others (2004), page 26, or with the National Streamflow Statistics program (NSS) which can be obtained from http://water.usgs.gov/software/NSS/.
The USGS has developed StreamStats for Illinois to provide up-to-date flood-frequency information for the State, in cooperation with the Illinois Department of Transportation, the Illinois Department of Natural Resources, Offices of Water Resources, the Illinois Center for Transportation, University of Illinois at Urbana-Champaign and the Federal Highway Administration.