Laserfiche WebLink
<br />~~, - <br /> <br />. <br /> <br />Flood-frequency relations expressed in this report may be used to <br />estimate magnitude and frequency of floods on most Minnesota streams. The <br />applicability and accuracy of these relationships depends on whether the basin <br />characteristics above the site under consideration are within the range of <br />characteristics used to define the frequency relations. The range in sampled <br />basin characteristics is large enough to allow use of the f,equency relations <br />at most sites where streamflow is not significantly affected by regulation, <br />diversion, or urbanization. The acceptable range for each of the physical <br />characteristics to be considered is tabulated in table 6 (table 6 is at end of <br />the report). Where runoff is included as an independent variable in an <br />equation, the sampling is complete enough to ensure that the entire range of <br />values may be used. <br /> <br />Corrections must be made at sites immediately below a lake or ponding <br />area where the storage capacity is large in relation to total drainage area <br />and could seriously alter flood characteristics. In such places, the <br />frequency relations may be used as an aid in developing a hydro graph of inflow <br />for use in routing flow through the storage area to the site. <br /> <br />SUMMARY AND CONCLUSIONS <br /> <br />This report contains an analysis of significant flood information for <br />Minnesota streams, except for miscellaneous flood measurements, and very short <br />or unpublished records (see tables. 2 to 5). Flood-frequency analyses of the <br />annual series peak data from 246 gaging stations were used to investigate <br />regional relations. A regression analysis of the regionalized data relates <br />peak flows to basin characteristics. The resulting regional equations can be <br />used to estimate flood flows at ungaged. unregulated sites for 2-, 5-, 10-, <br />25-, 50-, and 100-year recurrence intervals. Analyses of the standard errors <br />of estimate, regression coefficients, and residuals show that these equations <br />provide good estimates of selected frequency annual series peak flows subject <br />to certain limitations. <br /> <br />The use of weighting factors for stations in the regression analysis <br />provides an expedient technique for reduction of standard error. This tech- <br />nique increases the weight of data from longer-term stations and stations with <br />more accurately defined flood characteristics so that shorter-term records may <br />be incorporated into the analysis without adversely affecting the results. <br /> <br />18 <br />