The 2-, 5-, 10-, 25-, 50-, and 100-year recurrence interval floods are related to basin and climatic parameters for natural streams in Oklahoma by multiple regression techniques through the mathematical model,

Q_x=aA^bS^cP^d,

where Q_x is peak discharge for recurrence interval x, A is contributing drainage area, S is main channel slope, P is mean annual precipitation, and a, b, c, and d are regression constants and coefficients. One equation for each recurrence interval applies statewide for all natural streams of less than 2,500 mil (6,500 km²), except where manmade works, such as dams, flood-detention structures, levees, channelization, and urban development, appreciably affect flood runoff. The equations can be used to estimate flood frequency of a stream at an ungaged site if drainage area size, main channel slope, and mean annual precipitation are known. At or near gaged sites, a weighted average of the regression results and the gaging station data is recommended.

Individual relations of flood magnitude to contributing drainage area are given for all or parts of the main stems of the Arkansas, Salt Fork Arkansas, Cimarron, North Canadian, Canadian, Washita, North Fork Red, and Red Rivers. Parts of some of these streams, and all of the Neosho and Verdigris Rivers are not included because the effects of. major regulation from large reservoirs cannot be evaluated within the scope of the report.

Graphical relations of maximum floods of record for eastern and western Oklahoma provide a guide to maximum probable floods.

A random sampling of the seasonal occurrence of floods indicated about two-thirds of all annual floods in Oklahoma occur during. April through July. Less than one-half of one percent of annual floods occur in December.

A compilation of flood records at all gaging sites in Oklahoma and some selected sites in adjacent States is given in an appendix. Basin and climatic parameters and log-Pearson Type III frequency data and statistics are given for most station records. A second appendix gives a reprint of the U.S. Water Resources Council Bulletin 15 which describes procedures for fitting a log-Pearson Type III distribution to gaging station data.