|
<br />Study Area
<br />
<br />Cherry Creck has its headwaters on thc Palmcr Dividc at an elevation of about 2,350 m and flows
<br />northerly to its connuence wilh the South Plalle River (1.576 m) in Denver (tjg, I), In parI due 10
<br />lack of rainfall and flood dala, it has commonly been assumed that the Palmer Divide has little
<br />orographic effect on exlreme rainfall amounts and flooding (McCain and Jarrett, 1976; Livingston
<br />and Minges, 1987; Hansen et aI., 1988), A long-standing issue is whether the Palmer Divide, the
<br />ridge that separates the South Platte and Arkansas River basins (fig. I). has an orographic dreCl on
<br />rainfall and flooding in eastern Colorado. Major floods in eastern Colorado result from a southerly
<br />flow of low-level moisl air from the Gulf of MexicO, which often crosses over the Palmer Divide
<br />(Collins et aI., 1991). The Palmer Divide has a southern topographic relief as much as 1,035 m, thus,
<br />may be in the rain shadow of the Divide (tig. I),
<br />
<br />Cherry Creek basin primarily is underlain by Tertiary sedimentary rocks; sandstone,
<br />eonglomerale, siltstone, claystone, and shale (Tweto, 1979), Unconsolidaled surficial deposits
<br />(Pleistocene and Holocene) derived from eolian, colluvial, and fluvial processes cover much of the
<br />basin, Topography in the basin is moderately rolling except steep hills and ridges in the upper basin
<br />(U,S, Army Corps of Engineers, 1976), Soils primarily consist of well,drained, sandy loam, and
<br />some parts of the basin have exposed bedrock or shallow soils, Much of the basin has pasture and
<br />prairie grasses, agricultural crops, scrub oak, and ponderosa pine, Lower Cherry Creek basin has
<br />extensive residential, business, and industrial development.
<br />
<br />Methodology
<br />
<br />Paleoflood J n vestigations
<br />
<br />Paleoflood hydrology is the study of past or ancient floods (Baker, 1987), Floods leave
<br />distinctive deposits and landforms in and along stream channels, as well as botanic evidence (Baker,
<br />1987; Jarrett, 1990, 1991), Slack-water deposits of sand-sized particles, flood scars on trees,
<br />accumulation of woody-flood debris, erosion scars, and bouldery flood-bar deposits commonly used
<br />as indicators of past flood levels are called paleostage indicators (PSIs). When flows are large
<br />enough, streambed and bank materials are mobilized, transported, and deposited as new PSIs, In
<br />addition, non-inundation (NI) surfaces are geomorphic surfaces that have not been exceeded by
<br />floods, which would leave definitive erosional evidence, in a datable time span (Jarrell and Costa,
<br />1988; Ostenaa and Levish, 1995), The strategy of a paleoflood investigation is to visit the most likely
<br />places where evidence of out-of-bank flooding, if any, might be preserved, and to identify nOn-
<br />inundation (NI) surfaces. Ideally, the highest PSIs would correspond with the lowest NI-surfaces,
<br />thus, providing similar estimates of paleoflood discharge,
<br />
<br />Paleoflood discharge was detennined from estimates of flood width and depth corresponding to
<br />the top of the PSIs (or lowest NI-surfaces) and ch~nnel slope for each cross section obtained during
<br />onsite visits to streams, The slope-conveyance and critical-depth methods (Barnes and Davidian,
<br />1978) were used to estimate paleoflood discharge. Flow-resistance coefficients were estimated from
<br />methods derived from analyses of hydraulic data for ColOr.ldo rivers (Jarrett, 1985).
<br />
<br />Age estimates for paleoflood deposits in this study were detennined using relative-dating (RD)
<br />methods (Jarrett and Tomlinson, in press), RD methods used for this study primarily were degree of
<br />soil development, surface-rock weathering, surface morphology, and boulder burial. Surficial deposit
<br />age is based on post-depositional modifications that vary with age; immediately after deposition, the
<br />age is zero years, A composite relative age using several RD methods usually enables one to
<br />distinguish deposits of various ages, Although there are uncertainties with an individual RD ages, a
<br />composite age is more accurate, For the use here in attempting to identify the paleoflood record
<br />length for the largest flood during the Holocene, such uncertainties can be addressed in the f1ood-
<br />frequency analyses.
<br />
<br />Regional Analyses of Flood Data
<br />
<br />Predicting the upper limits of the magnitude of tloods for has been a long standing challenge in
<br />flood hydrology, Envelope curves encompassing maximum floods in a relatively homogeneous
<br />
<br />3
<br />
|