|
<br />hydrometeorologic region have long been used in flood hydrology, Regional analysis provides
<br />improved estimates of streamflow characleristics by decreasing time-sampling errors of short
<br />gagerecords by substituting space (many flood estimates in a region) for lime (short gage record),
<br />Flood data for Cherry Creek basin streams (McKee and Doesken,1997) were compiled to develop
<br />envelope curves of peak discharge versus drainage area. Incorporating paleoflood datu provides an
<br />opportunity to add a new level of contidence to envelope curves (Jarrett and Tomlinson, in press),
<br />
<br />Flood-Freqllency Relatiolls
<br />
<br />Flood-frequency relations were developed from an analysis of annual peak flows lhrough 1997
<br />for Cherry Creek near Franktown (06712000) and near Melvin (06712500), which is located about 3
<br />km upslream from Cherry Creek Reservoir (fig, I), Flood-frequency relations were developed using
<br />a Log-Pearson Type III frequency distribution (IACWD, 1981), IACWD guidelines were established
<br />to provide consislency in federal flood-risk management such as for handling low- and high- oulliers.
<br />need for regionalized skew, and zero-flow adjustment, This analysis was done for various
<br />combinations of gage and paleoflood data available at each sileo To help facilitate risk assessments of
<br />rare floods (e,g.. defining the upper end of frequency curves), paleoflood data (magnitude and ages)
<br />were incorporated into the flood-frequency analysis to extend the gaged record,
<br />
<br />Results
<br />
<br />Paleoflood Investigations
<br />
<br />Paleoflood data were obtained at 99 sites on streams in Cherry Creek basin, PSI data are readily
<br />identifiable onsite by coarse-grained flood deposits (figs, 2 and 3), Recent flood deposits in Cherry
<br />Creek basin typically are coarse, sandy gravel (often with cobble and boulder-sized clasts) and have
<br />little or no soil-profile development. Deposits with increasing age have soil-profile development,
<br />increased surface-rock weathering, muted surface morphology, and increased surficial flood deposit
<br />burial. Thick, clay-rich, fine-grained, well-developed soils (labelled colluvium on figs, 2 and 3)
<br />generally occur in non-flooded areas in the valley, Because these colluvial soils have few particle
<br />sizes larger than sand, they were not deposited by main stream flooding; rather, the sediments
<br />primarily originated from hillslope (sheetflow) runoff and eolian (wind-blown) sediments (R.
<br />Madole, USGS, pel'S, commun" 1997), These colluvial organic-rich soils, termed Piney Creek
<br />alluvium, were dated from a minimum of about 1,000 years to greater than 5,000 years (Hunt, 1954)
<br />and were the primary NI-surface used, Thus, a lack of erosional and depositional features provided a
<br />minimum age since floods have inundated these surfaces. These colluvial soils are easily eroded by
<br />flood waters (Matthai, 1969), thus, they also provide physical constraints on the present channel
<br />geometry since originally deposited (figs, 2 and 3), The other RD methods provided additional
<br />supporting relative age constraints.
<br />
<br />In reaches where flood sediments were deposited as flood bars (PSIs) or use of NI-surfaces, peak
<br />discharges were computed with surveyed channel geometry and by subtracting the area of the
<br />estimated flood deposit (figs, 2 and 3; area below the surveyed channel geometry and dashed line),
<br />Paleoflood discharge (corresponding to the range of the PSIs or NI-surfaces such as shown in figs, 2
<br />and 3) reflects the larger (conservative) estimate in the sensitivity analysis of factors affecting
<br />discharge reconstruction (e,g" changes in channel geometry, water slope, uncertainty in n values,
<br />reliability of PSI and NI-surfaee heights),
<br />
<br />The largest paleoflood for Cherry Creek near Franktown (06712500) is about 1,050 m3/s, which
<br />resulted from the failure of Castle wood Canyon Dam during a rainstorm (75 to 230 mm in about 9
<br />hours) on August 3, 1933 (Matthai, 1969), Excellent paleoflood evidence (top of coarse bouldcry
<br />flood-deposited sediments) for the 1933 flood is preserved at sites downstream from Castlewood Dam
<br />(fig. 2). The maximum paleoflood upstream from Castlewood Dam is about 850 m3/s, The
<br />maximum paleoflood downstream from CastIewood Dam at the gage (06712000) near Franktown,
<br />
<br />For comparison, data for floods in other streams in eastern Colorado (Jarrett, 1990; McKee and
<br />Doesken, 1997) were used to develop an envelope curve (fig, 4; data not shown), Maximum
<br />contemporary flooding in eastern Colorado is about 2.3 times greater than maximum paleofloods in
<br />Cherry Creek basin,
<br />
<br />4
<br />
|