Laserfiche WebLink
<br />**DRAFT** Supplementing :Flood Frequency Analysis in Eastern Colorado <br />using Rainfall-Runoff Modeling <br /> <br />STEVEN B. BARTON <br /> <br />Masters Candidate, Civil Engineering Department, Colorado State University, Fort Collins, Colorado <br /> <br />Submitted for fulfillment of requirements for the Master of Science degree <br /> <br />July 30, 1999 <br /> <br />ABSTRACT <br /> <br />Flooding is a major cause of damage in the United States. Much of the efforts of hydrologist are directed to <br />better understanding the uncertainties of flooding in order to better mitigate its effects. Rainfall-runoff <br />modeling is one method that can be used to estimate flood frequency relationships and construct them in <br />ungaged locations. Two models, HEC-HMS and SCS CN, are used in this study to show that calibrated <br />rainfall-runoff models can reasonably describe the flood frequency relationship of three small basins in <br />northeastern Colorado. The frequency curves are estimated using 5-minute rainfall-runoff data and <br />paleoflood data using Bulletin 17B guidelines. A multiple scenario modification to Bulletin 17B was used <br />to adjust for excessively censored data. The rainfall-runoff simulations are compared to the mean <br />frequency curve for validation. The HEC-HMS model proved to be valid beyond the 5-year flood. The <br />SCS CN model yielded questionable results attributed to tack of robustness. Frequency estimates using the <br />expected moments algorithm (EMA) are compared to the Bulletin 17B ensemble with favorable results and <br />less effort. Areas of continued research are suggested. <br /> <br />1. Introduction <br /> <br />Flooding is one of the most widespread and destructive natural disasters in the world (Costa, <br />1987). Flood damage in the United States between 1986 and 1995 averaged $3.5 billion <br />annually (Ingram, 1996). As development along a floodplain increases, the expected damage <br />resulting from flooding will increase. For the period only almost ten years prior, Jarrett (1991) <br />reported average annual damage of $2.4 billion between 1978 and 1988. Costa (1987) reported <br />an average annual death toll of 200 from flooding between 1968 and 1986. Engineers attempt to <br />minimize the economic impact and loss of life by designing structures to detain and convey flood <br />flows. <br /> <br />The design of hydraulic structures for this purpose requir~ estimates of the flood volume, peak <br />discharge and frequency of occurrence. The U,S. Georogical Survey (USGS) operates and <br />maintains an extensive network of hydrometeorological gages throughout the United States. 'In <br />the western half of the nation, records are generally short with many locations having systematic <br />records less than twenty years. The majority of catastrophic floods are caused by floods with <br />recurrence intervals greater than or equal to 100 years (Costa, 1987). Extrapolating statistical <br />characteristics from systematic records with periods of record much less than the recurrence of <br />the flood of interest imparts a significant level of uncertainty. <br /> <br />/ <br />