FLOOD06859 - Laserfiche WebLink

Home Browse Search

To understand runoff processes. the causative factors, par- ticularly precipitation, need to be understood. Since before 1900, the U.S. Bureau of Reclamation has documented and compiled rainfall amounts of large and intense rainstorms in the western United States. Ninety-seven of these storms have occurred in Colorado. The distribution of 6-hour rainfall with elevation is shown in Figure 4 (Jarrett, 1987). Rainfall at lower elevations in the plains of eastern Colorado has exceeded 500 Mm. Six-hour rainfall decreases abruptly from about 500 mm to less than 50 mm, at eleva- tions above 2,440 m. '" ? --- . 125.9 SQUARE KILOMETER) S, HOUR RAINFALL ... TOTAL POINT RAINfALL '" 111 i " J i ~'" ..; J ~ z ii ~'" .. '" I; ~ ~2O!l II! ~ . . ~ !Ij ,.. . . 1.000 1.&00 2.000 2.500 3.000 uoo ElEVATION. IN MmRS Figure 4. Relation between precipitation and elevation. including envelope curve. for large storms in Colorado. 3.2 Paleohydrologic techniques Historic and prehistoric floods leave distinctive deposits and landforms in valleys and channels. Geomorphic evidence of large floods in steep mountain basins, such as in the Big Thompson River. is easy to recognize and long-lasting because of the volume and size of sediments deposited. Lack of paleoflood evidence from large floods is as important as tangible onsite evidence of such floods (Jarrett, 1987). ~ Paleoflood research has been conducted in the South Platte River basin (Jarrett, 1987), particularly in the Big Thompson River basin (Jarrett and Costa. 1988). The strategy of these studies 1s to ex- amine those sites where evidence of any large floods might be preserved. Sites studied include: (1) Locations of rapid energy dissipation', where coarse sediments would be deposited, such .. tributary junctions. abrupt decreases of stream slope, or abrupt valley expansions; (2) locations downstream from glacial morain.. across valley floors where floods would deposit sediments eroded from the moraines; and (3) locations along the sides of valley. In wide, expanding reaches where sediments would likely be deposited. These paleoflood investigations indicate that large but infrequent floods have occurred in all basins below 2,300 m. No evidence of water flows much higher than bankfull discharge was found in any stream above 2,300 m in Colorado (Jarrett, 1987). A paleoflood investigations was conducted in the Big Thompson River basin (Jarrett and Costa, 1988). A principal purpose of that study was to investigate whether there was any evidence of large post-glacial floods in the velleys draining into Lake Estes, which is formed by Olympus Dam. There is an absence of any paleoflood evidence of large floods upstream from 2,300 m in tpe Big Thompson River basin, Radiocarbon dating of paleoflood deposits in the Big Thompson River basin yielded estimates of relative frequency of the 1976 flood, This dating indicated that the 1976 flood was the largest since the occurrence of glacial melting 8,000 to 10,000 years ago. 3.3 Flood information-transfer techniques Flood characteristics commonly are needed at ungaged sites; the inJ formation also can increase the reliability of estimates of flood characterisitics at short-record gaged sites. Investigations of flood potential based on rainfall-runoff modeling techniques f.il to adequately describe the flood hydrology of foothill and mountain streams: generally these techniques significantly overestimate flood magnitudes (Jarrett. 1987). Investigators have assumed that the to- tal basin area contributes runoff during rainstorms. However, 1n the foothills and mountains of Colorado below 2,300 .. floods are caused by rainfall, generally intense ahort-duration thunder.to~ of limited areal extent. Therefore, regional flood estimating pro- cedures need to compute rainfall-generated flood estimates on that part of the basin below about 2,300 m in Colorado. One component of rainfall-runoff modeling to determine the ..go. nitude of floods is the use of storm transposition. Jarrett and Costa (1988) demonstrated that the concept of storm transposition from lower elevations to higher elevations in Colorado is not sup- -ported by meteorological, hydrological, and paleoflood information. Also, depth-area relations, used in rainfall-runoff methods to reduce point rainfall for the size of the watershed, applied to the foothills and mountains of Colorado were not developed with data from that area and result in large rainfall-runoff flood estimates.