Laserfiche WebLink
<br />GOI)39~ <br /> <br />THE DENVER FLASH FLOOD PREDICTION PROGRAM: <br />A JOINT HYDROMETEOROLOGICAL EFFORT <br />AND <br />OPERATIONAL PREDICTION OF THE DENVER SEVERE <br />HAIL/RAINSTORM OF 13 JUNE 1984 <br /> <br />by <br /> <br />John F. Henz and Ron A. Kelly <br />Henz Kelly &.Associates <br />Denver. Co.lorado 80210 <br /> <br />Introduction <br /> <br />Flash floods are more difficult to predict than to define. though <br />some may argue the point. Flash floods may be defined as a flood that <br />rises and falls quite rapidly with little or no advance warning usually <br />as the result of intense rainfall over a small area. Other possible <br />causes are the release of ice jams or dam failures. While this definition <br />is hardly quantitative it does support the contention that flash floods <br />occur quickly. <br /> <br />The speed of a flash flood occurrence negates the usual flood <br />prediction engineering approach of measuring rainfall using a gage network <br />to estimate volumetric rainfall, correlating this to stream gage <br />measurements and initializing the responses in a basin model. Unfortunately <br />the flash in the flash flood negates this standard approach because it <br />can rarely provide sufficient warning time for community response and <br />evacuation. Instead a premium is put on the prediction of the flash <br />flood before it happens and that means accurate rainfall prediction. But <br />can rainfall be accurately predicted before the fact, <br /> <br />National statistics argue that the current flash flood forecast <br />effort mounted by the National Weather Service has not been accurate and <br />that warnings come too late. A recent report (September 1983) by the <br />Inspector General. Department of Commerce, entitled. Thunderstorm Killers <br />Flash Floods and Li htnin Need to 1m rove Service Weather Forecastin <br />DEN-X-158-02-4800-83-008 . quotes the following national statistics <br />achieved by the National Weather Service. <br /> <br />1. In 1980 more than 40 percent of flash flood watches <br />issued were false alarms with no flash flood. . In <br />1981, the false alarm figure reached nearly 50 percent, <br /> <br />2. In 1980 almost half (49 percent) of all flash flood <br />warnings had no lead time for the public to prepare <br />for the flash flood. Another quarter of the flash <br />flood warnings had from a few minutes to an hour for <br />the public to prepare. <br />