<br />2974
<br />
<br />JARRETT AND TOMLINSON: REGIOI\AL INTERDISCIPLINARY PALEOFLOOD METIlOD
<br />
<br />\
<br />.~
<br />
<br />6
<br />
<br />5
<br />
<br />Colluvtll (Ilnlilu11r) ~
<br />
<br />/ /boUlder'~~II~~t:~ltfe~ ~i
<br />a :1~~i~le:~:~pe\:
<br />(NI) CJ
<br />............uiA.;;:i>sl... .....
<br />FB--' <)
<br />? ....,0 0000 _~
<br />__ 'DOc> _- 0
<br />~o 0-00";;" q,o"<;l -C1 000
<br />oo~PO~O~~oO &.o.}
<br />oooOou:;''tYqooooo
<br />Pleistocene-outwash gravels
<br />
<br />
<br />E 4
<br />"
<br />.2
<br />..
<br />>
<br />.. 3
<br />w
<br />
<br />
<br />2
<br />
<br />1
<br />o
<br />
<br />30
<br />
<br />40
<br />
<br />10
<br />
<br />20
<br />Stationing, m
<br />
<br />Figure 7. Graph of channel cross section for Eikhead Creek
<br />downstream from North Fork Elkhead Creek shown in Figure 6.
<br />
<br />During on-site visits, peak discharge using 1995 HWMs also
<br />was estimated for seven sites at streamflow-.gaging stations
<br />(Table 2, type is "gage"). These estimates using the critical-
<br />depth and slope-conveyance methods were compared to the
<br />peak discharge for 1995 obtained from gage records to help
<br />assess the accuracy of paleodischarge estimation methods. The
<br />estimated discharges generally are within about 10% of the
<br />gaged discharge (Table 2, "Q" and "Q"go Difference" col-
<br />umns) thus showing that the methods used to estimate paleo-
<br />flood discharge are reliable. However, additi.)nal sources of
<br />uncertainties (e.g., channel change) are more difficult to quan-
<br />tify. Attempts to estimate other uncertainties i.n discharge are
<br />reflected in terms of discharge uncertainty (Table 2, "Q, %"
<br />column) and age (Table 2, age).
<br />Paleoflood estimates in bedrock channels generally were
<br />assigned an uncertainty of 25%, whereas alluvial channels were
<br />assigned an uncertainty of 30%. Paleoflood estimates in allu.
<br />vial channels with minor arroyo development, though the age
<br />of NI surfaces are very old, were assigned an age of 100 years
<br />to account for historical arroyo development. Additional con-
<br />fidence in paleoflood estimates is exhibited when multiple sites
<br />are used and results are similar. For example, for Elkhead
<br />Creek downstream from Elkhead Reservoir, paleoflood esti-
<br />mates ranged from 127 m' s'[ to 135 m' s'[ (sites 47 and 48
<br />in Table 2), a difference of about 3%. Simihrly, paleoflood
<br />estimates along Elkhead Creek upstream from the reservoir
<br />increase consistently from 79 mJ S-l to 135 m3 S-l with in-
<br />creasing drainage area.
<br />Streams in northwestern Colorado have few coarse flood
<br />deposits on the floodplain. Where present, the deposits are
<br />either associated with the record snowmelt flooding in 1984, or
<br />the deposits are very old (Table 2). The fact that streams in
<br />northwestern Colorado have no substantial paleoflood evi-
<br />dence is important, because it indicates the lack of substantial
<br />flooding in Elkhead Creek basin is not due to chance. The
<br />paleoflood data then were used to help define the regional
<br />maximum flooding and for flood-frequency analyses in north-
<br />western Colorado.
<br />
<br />5.2. Regional Analyses of :vIaximum Rainfall and Flood Data
<br />
<br />5.2.1. Maximum rainfall. A relation betNeen maximum
<br />24-hour rainfall and elevation for the study area in northwest-
<br />ern Colorado is presented in Figure 8; this rdation was con-
<br />structed from documented rainstorms from about 1900
<br />
<br />through 1990 [Jam/!, 1990bJ and was updated through 1997
<br />with recent data [McKee and Doesken, 1997]. Although there
<br />has been an extensive program for documenting extreme rain-
<br />storms in Colorado, there have been few intense flood-
<br />producing rainstorms documented in northwestern Colorado
<br />(triangles on Figure 8). The maximum 24-hour rainfall data for
<br />181 stations and bucket surveys in southwestern Colorado also
<br />are shown on Figure 8 for comparison to help define the
<br />northwestern Colorado region. The maximum 24~hour amount
<br />for northwestern Colorado is 82 mm (Meeker). Maximum
<br />monthly values for northwestern Colorado only slightly exceed
<br />record maximum 24-hour amounts for southwestern Colorado
<br />(Figure 8), dramatic evidence of large relative difference in
<br />flood-producing rainfall from northwestern to southwestern
<br />Colorado. The maximum rainfall amount for southwestern
<br />Colorado is about 150 mm in a few hours for Sweetwater
<br />Creek (1976) and for Dove Creek in 24 hours (1972) [McKee
<br />and Doesken, 1997J; these areas are influenced by the flow of
<br />moist air from the southwest [Collins el aI., 1991J. Maximum
<br />24-hour rainfall in southwestern Colorado is somewhat larger
<br />than in northwestern Colorado but is substantially less than the
<br />maximum 6-hour rainfall amounts of up to 610 mm in eastern
<br />Colorado. Of particular interest in western Colorado, maxi.
<br />mum 24-hour precipitation amounts fell as snow and are pre.
<br />sented as snow-water equivalent (SWE) (Figure 8, open
<br />squares).
<br />A subjective, indirect indicator of the occurrence of intense
<br />rainfall and associated flooding is the development of rill
<br />(light) and or gully (deep) erosion on hillslopes [JarTert, 1990b;
<br />Ja"ett and Browning, 1999]. Generally, erosion potential in-
<br />creases as slope steepens; steeper slopes usually require com-
<br />paratively small amounts of rainfall before substantial erosion
<br />occurs, although erosiveness also depends on the type of soil,
<br />
<br /> . NW co 24-hr
<br />24-hr PMP . NW co month
<br /> . SWco 24-hr
<br /> , W CO 24-hr SWE
<br /> -10D-yr,24-hr
<br /> ......1D-yr.24-hr
<br /> 1976 Sweetwater
<br /> 1/ Dove Creek
<br />6-hr PMP
<br /> . .
<br /> . , ,
<br /> OOOOC>"'Boo~gow ~ ,
<br /> . . .~~-at! . .,. ~~,.......... ,
<br /> ~.~fJ'.., i.; o:lb eo ,
<br /> . . .
<br />
<br />600
<br />
<br /> SOD
<br /> 400
<br />E
<br />E
<br /> 300
<br />~
<br />c
<br />..
<br />0: 200
<br /> 100
<br />
<br />o
<br />1000
<br />
<br />2000 3000
<br />Elevation. m
<br />
<br />4000
<br />
<br />Figure 8. Ma-amum 24-hour and maximum monthly precip-
<br />itation for northwestern (NW) Colorado and maximum 24-
<br />hour precipitatioo for southwestern (SW) Colorado [JarTell,
<br />1987, 1990b; McKee and Doesken, 1997J. Two of the largest
<br />southwestern Colorado rainstorms (Sweetwater Creek and
<br />Dove Creek) are noted. It is important to note that numerous
<br />large snowstorms reported as snow-water equivalent (SWE)
<br />account for some of the largest 24-hour precipitation amounts
<br />in all of western (W) Colorado. The lO-year and 100-year,
<br />24-hour duration rainfall amounts [Miller el al., 1973J are
<br />shown to place contemporary rainfall data into a frequency
<br />context. The 6.hour and 24-hour duration probable maximum
<br />precipitation (PMP) values [Hansen et aI., 1977] are also shown
<br />for comparison.
<br />
|