In lieu of an abstract, here is a brief excerpt of the content:

7 Estimating Flood Frequency David Eash Soon after the 2008 floods in Iowa reached their peak, the media labeled them as 100-year floods or 500-year floods. What exactly do these terms mean, and how do they apply to Iowa’s floods of 2008? These terms are expressions of the extreme magnitude of the 2008 floods and the frequency with which they are expected to occur. Flood frequency is expressed in several ways. One is as a probability, expressed as a percentage, that a flood of a specific level will be equaled or exceeded in any given year. Thus, a 1 percent chance flood has a 1 in 100 chance of occurring in any particular year, and a 0.2 percent chance flood has a 1 in 500 chance of occurring any year. Another common way to express flood frequency is as the time between floods of a certain magnitude (known as the recurrence interval). If a specific flood level has a 1 in 100 chance of occurring in a given year (a 1 percent chance flood), then the average recurrence interval between floods of this size is 100 years. This flood level is called the 100-year flood. Likewise, if a specific flood level has a 1 in 500 chance of occurring in a given year (a 0.2 percent chance flood), the average recurrence interval between floods is 500 years, and it is called a 500-year flood. By this definition, a 500-year flood is larger than a 100- 62 rising rivers, spreading waters year flood, which is larger than a 50-year flood, and so on. Larger floods occur less frequently than smaller floods, have a smaller chance of occurring in any particular year, and have a longer average recurrence interval. The terms “100-year flood” and “500-year flood” are often misunderstood: people tend to believe this level of flood should only occur once every 100 or 500 years (Dinicola 1997). However, the 100-year terminology does not imply that there is a regular or uniform time interval between flood events or that a given 100-year period will contain one and only one 100-year flood event (Subcommittee on Hydrology 2008). The truth is that there is a chance that an unusually large flood will occur any year—it can occur next year, or the year after, or even in two sequential years.1 Flood-frequency estimates for streamgage locations are based on measurements of a river’s annual maximum discharge—that is, the peak rate of streamflow that moves past a point in the river during the course of a one-year period. Discharge is expressed in cubic feet per second (cfs); one cfs is equal to 7.48 gallons of water per second. A record of annual maximum discharges is a collection of the major flood peaks measured over a historical period. But not all annual maximum discharges are floods, as a river may not overflow its banks every year. For most streamgage locations with a sufficient record of annual maximum discharges (at least 10 years), federal agencies estimate flood-frequency discharges using the statistical techniques outlined in Bulletin 17B (Interagency Advisory Committee on Water Data 1982). As additional years of data are collected at streamgages, Bulletin 17B estimates of flood-frequency discharges are updated and become more statistically reliable. For watersheds significantly affected by urbanization or river locations downstream from major floodcontrol structures—like the Coralville Reservoir and Dam, which regulates flows by storing floodwaters and controlling releases to reduce flood discharges downstream—more complex statistical methods are employed to estimate flood-frequency discharges. For ungaged river locations, regional equations are used to estimate flood-frequency discharges by relating Bulletin 17B floodfrequency estimates for streamgage sites to ungaged sites with similar watershed characteristics (Eash 2001). Thus, flood-frequency discharges can be estimated for any river location. What does flood-frequency estimation tell us about Iowa’s 2008 floods? Were they 100- or 500-year floods? The answer depends on the particular river location being considered. Rainfall and snowmelt, and resulting tributary streamflow, vary from one location to another, and so the level of flooding can [52.14.85.76] Project MUSE (2024-04-25 09:07 GMT) estimating flood frequency 63 vary significantly from one river to another and from one location to another along the same river. Consider, for example, the 2008 flood discharges and recurrence intervals for streamgages on the Iowa and Cedar Rivers listed in table 7...

Share