Five reports dealing with flood magnitude and frequency in urban areas in the southeastern United States have been published during the past 2 years by the us Geological Survey (USGS). These reports are based on data collected in Tampa and Tallahassee, Florida; Atlanta, Georgia; and several cities in Alabama and Tennessee. Each report contains regression equations useful for estimating flood peaks for selected recurrence intervals at ungauged urban sites in their respective study area. A nationwide study of urban flood characteristics by the USGS published in 1983 contains equations for estimating urban peak discharges for ungauged sites throughout the United States. At the t: ime that the nationwide study was conducted, data from only 35 sites in the southeastern United States were available. The five new reports contain data for 88 additional sites in the southeastern United States. These new data show that the sevenparameter estimating equations developed in the nationwide study are unbiased and have prediction errors less than those described in the nationwide report. On the other hand, the new data indicate that the three-parameter equations are biased and significantly underestimate flood discharge in four of the new study areas. The five new reports on the southeastern United States and the nation-wide report provide reliable methods for estimating design discharges.

Rapid expansion and development of urban areas in the United States bring many informational needs. An important one is flood data, which are necessary in the design of stream channels, canals, storm sewers, detention ponds, roadways, bridges, and culverts. Likewise, the magnitude, frequency, and boundaries of floods are required for zoning and insurance purposes. Gauging, or measuring, all streams and locations where data are needed is not feasible. Instead, flood data are collected at a few selected sites, and the information is transferred to ungauged sites by various regionalization procedures. To this end, the us Geological Survey (USGS) has collected urban flood data in many cities throughout the Uni tea States during the past 20 to 30 years. These data have been published for public use and analyzed for flood-frequency regionalization studies. Reports have been published describing flood characteristics for individual cities, metropolitan areas, or selected groups of cities. References to and brief abstracts of many of these reports as well as other urban flood-frequency procedures may be found in a literature review by Rawls et al.(!)• A nationwide regionalization of urban flood characteristics by Sauer et al.(2) describes techniques for estimating flood magnitude and frequency for cities throuqhout the United States, including Alaska. That report contains flood data and basin characteristics for 269 urban sites in 56 cities and 31 states. It also has an extensive list of references for both urban and rural flood-frequency procedures. After analysis of the nationwide regionalization (2), five new urban studies were prepared for citi-; s, metropolitan areas, and states in the southeastern region of the United States. These include a statewide study for Alabama (l); the Atlanta, Geor-