Estimating Flood-Peak Magnitudes and Frequencies for Rural Streams in Illinois

Scientific Investigations Report (SIR) 2004-5103

This report presents the flood-frequency estimates for rural streams in Illinois. At-site results and regionalization techniques from the 2004 study can be used for estimating the magnitude and frequency of flood-peak discharges at U.S. Geological Survey (USGS) streamflow-gaging stations or at ungaged stream locations in Illinois. Knowledge of the magnitude and frequency of flood-peak discharge is essential for planning, managing, designing water-resources projects. To provide up-to-date flood-frequency information for the State of Illinois, the USGS conducted this study in cooperation with the Illinois Department of Natural Resources, Offices of Water Resources, Realty and Environmental Planning – Conservation 2000 Program, and Resources Conservation; and the Illinois Department of Transportation.

View the report in Adobe Acrobat PDF format or view the Table of Contents.

A regional flood estimate update is warranted when better data and/or analytical technologies become available. Updates from a 1987 study (Curtis, 1987), the 2004 study includes:

At-Site Analysis:

• Used flood-peak discharge data up to water year 1999

• Estimated flood-frequency curves based on the annual maximum series (AMS) for recurrence intervals between 2 and 500 years, and on partial duration series (PDS) for recurrence intervals between 0.8 and 5 years.

• Updated regional skew coefficients for weighted-skew approach in the Log-Pearson 3 analysis for the AMS model.

Regional Analysis:

• Delineated seven hydrologic regions, on the basis of physiographic and hydrologic characteristics, for AMS and PDS analyses.

• Developed 38 basin characteristics using the BASINSOFT program and digital datasets including: USGS topographic elevation model (DEM), State Soil Geographic (STATSGO) database (Natural Resources Conservation Service, 1993), and the USGS National Land Cover Data (NLCD) for 288 rural watersheds. Subgroups of the basin characteristics were used for developing the regional equations.

• Applied generalized least square technique to develop AMS regional equations.

Note that the AMS and PDS flood series have different data structures; and the definitions for common terms, such as recurrence intervals, are different.  Applications of flood quantiles estimated from each flood series should not be mixed. More detailed explanations of the data, techniques, and results can be found in the project report. Text for selected topics also can be viewed on the Table of Contents page.

Main results from the analyses include:

• At-site AMS flood-frequency estimates and regional equations (see below) for AMS flood-frequency estimates for recurrence intervals of 2, 5, 10, 25, 50, 100, and 500 years. The AMS flood estimates are suitable for flood protection and prevention analyses, similar to those presented in previous statewide flood-frequency analysis. The estimates of Q₅₀₀ are included for application in floodplain delineation and flood-insurance studies. Understandably, higher uncertainties are associated with estimates at larger recurrence intervals no matter which flood series is used. 

• At-site PDS flood-frequency estimates and regional equations (see below) for PDS flood-frequency estimates are developed for recurrence intervals of 0.8, 1.01, 1.5, 2, 3, and 5 years. The PDS flood estimates may be more suitable for environmental studies where damages caused by more frequent or repeated flood magnitudes in a year are of concern.

Regional equations are presented in the general form:

where:

QT is the estimated flood quantile, in cubic feet per second, for the designated recurrence interval T, in years.
a is the coefficient of the equation, b, c, d, e, and f are exponents for variables TDA, MCS, PermAvg, BL, and (%Water+5), respectively.
TDA is the total drainage area, in square miles.
MCS is the main channel slope, in feet per mile.
PermAvg is the averaged permeability of the watershed, in inches per hour.
BL is the basin length, in miles
(%Water +5) is the calculated percentage of open water and herbaceous wetland in the watershed plus a constant 5 percent (to avoid zero values).
(%Water +5) is in percent.
RF(N) is the regional factor for hydrologic region N.

A list of coefficients for AMS regional equations and PDS regional equations are presented using this generalized form (see pages 37 and 38 in the PDF version of the report). Methods for measuring these explanatory variables are explained. Shape files for determining PermAvg can be downloaded for GIS applications. The coordinate system is in NAD 1983 State Plan Illinois East FIPS 1201 (feet).

The flood frequencies estimated in 2004 are appreciably different at various locations from those estimated in 1987 (Curtis, 1987) because of the occurrence and distribution of major flood events during the 1990’s. The results presented represent in the current report the most updated flood-frequency estimates of streamflow information and should be used where the techniques are applicable. However, users are reminded that the accuracy of flood-frequency analysis depends on the representativeness of the flood data at streamflow stations, and on assumptions corresponding to the techniques used in the analysis. Results presented in this study are designated for areas where hydrologic characteristics have not appreciably altered by urbanization or channel regulations. For urban areas or regulated streams, other approaches such as rainfall-runoff modeling may be used for estimating the flood-frequency relations. The regional equations estimate the mean values of flood quantiles of different basins in the hydrologic region with the same set of explanatory variables. Therefore, if the study site is noticeably different in basin characteristics from those described in the regional equations, engineering judgment should be used in evaluating the results from the regional equations. The regional equations should not be extrapolated beyond the range of selected explanatory variables in each equation. The equations are not designed for evaluating the effects of land-use changes within a watershed.

References Cited:

Curtis, G.W., 1987, Technique for estimating flood-peak discharges and frequencies on rural streams in Illinois: U.S. Geological Survey Water-Resources Investigations Report 87-4207, 79 p.

Illinois Department of Natural Resources, 1996, Illinois land cover – an atlas: Springfield, Ill., Critical Trends Assessment Project Phase II: Illinois Department of Natural Resources, 157 p. and CD-ROM.

Natural Resources Conservation Service (formerly Soil Conservation Services), 1993, State Soil Geographic (STATSGO) database: U.S. Department of Agriculture, Miscellaneous Publication No. 1492, 110 p. and computer data.

U.S. Geological Survey, 2-Arc-Second (30-Minute Digital Elevation Model) Quadrangle; accessed October 2004 a URL: http://gcmd.gsfc.nasa.gov/servlets/md/getdif.py?entry_id=DEM_30&xsl=full_display.xsl&portal=gcmd.

U.S. Geological Survey, National land use and land cover data; accessed October 2004 a URL: http://landcover.usgs.gov/nationallandcover.asp.