|
<br />a.
<br />
<br />KCYS: 28 July 19971730-2225 MDT
<br />I 1-1-, ,-I~Uil'~
<br />1 2 3 3,5 4 4,5 5 5.5
<br />
<br />10
<br />
<br />
<br />~
<br />~ 5
<br />~
<br />"
<br />.:.::
<br />"
<br />....
<br />Q
<br />"0 0
<br />=
<br />'"
<br /><:::
<br />'"
<br />Eo<
<br />~ -5
<br />rf)
<br />i
<br />
<br />/
<br />,
<br />,
<br />,
<br />I
<br />I
<br />I
<br />,
<br />I
<br />J.
<br />-,.
<br />I
<br />I
<br />I
<br />,
<br />\
<br />,
<br />I
<br />I
<br />I
<br />
<br />I
<br />I
<br />I
<br />/
<br />/
<br />
<br />I
<br />\
<br />,
<br />
<br />
<br />I
<br />I
<br />I
<br />
<br />I
<br />
<br />-10
<br />-10
<br />
<br />
<br />-5 0 5
<br />E-W of Taft and Drake (KM)
<br />
<br />b.
<br />
<br />CHILL: 28 July 1997 1730-2215 MDT
<br />1 [-'1-, , I<',r-r~ 1
<br />1 2 3 4 5 6 6,5 7 7.5 8
<br />
<br />10
<br />
<br />
<br />~
<br />::;:
<br />~
<br />~
<br />"
<br />.:.::
<br />'"
<br />....
<br />Q
<br />"0
<br />=
<br />'"
<br /><:::
<br />'"
<br />Eo<
<br />"'"
<br />o
<br />rf)
<br />i
<br />
<br />
<br />5
<br />
<br />~
<br />
<br />o
<br />
<br />,
<br />,
<br />I
<br />I
<br />I
<br />I
<br />.( ~
<br />--..
<br />1 (
<br />I
<br />l-
<br />I
<br />, I
<br />.,.' I
<br />/
<br />,
<br />
<br />,
<br />,\
<br />--
<br />
<br />-5
<br />
<br />10
<br />
<br />-10
<br />-10
<br />
<br />10
<br />
<br />-5 0 5
<br />E- W of Taft and Drake (KM)
<br />
<br />FIG, 19, Storm,total rainfall (in,), 1725-2215 MDT 28 July 1997, (a) NEXRAD Z-R estimate using KeyS reflectivity data, (b)
<br />CSU-CHILL blended R(Kop, ZOR) estimate. The contour interval in (a) and (b) is 1 in. The origin ("0") is located as in Fig. 9. FeL
<br />street grid and location of Spring Creek are also depicted as in Fig. 9. Note the change in scale for shading of rainfall amounts in (a)
<br />and (b),
<br />
<br />standard errors of the estimators reported in Bringi
<br />et aL (1996) for the CSU-CHlLL radar, The "tropi,
<br />cal" Z-R rela1ionships shown in Table 2 (Short et aL
<br />1997; Rosenfeld et aL 1993) were utilized in the
<br />blended product (Fig, 19b), Sensitivity tests suggest
<br />that the blended product was relatively insensitive to
<br />the tropical Z-R relationship used and also to small
<br />changes (e,g" 0,10 km-', 0,1 dB, I dBZ) in the Kop' ZOR'
<br />and reflectivity thresholds chosen, Relative to the
<br />R(Kop, ZVR) estimate, the blended rainfall product (Fig,
<br />19b) exhibited a slight decrease (7%) in the maximum
<br />STP, but smoother (spatially) instantaneous rain rates,
<br />In summary, the NEXRAD Z-R relationship ap'
<br />plied to both KCYS and CSU-CHILL radar reflectiv-
<br />ity data produced STP maxima that were only
<br />50%-65% of the gauge STP maximum (similar to un-
<br />derestimates of rainfall by the NEXRAD in the Madi-
<br />son County, Virginia, flash flood of 1995; Smith et al.
<br />1996), While good agreement was attained between
<br />gauge-measured STPs and STPs computed using the
<br />(Short et aL 1997) Rosenfeld et aL (1993) tropical
<br />Z-R relationship with the (CSU-CHILL) KCYS
<br />NEXRAD data, the relative degree of agreement be,
<br />tween the Z-R,computed STP and the gauges was
<br />clearly a function of both the radar and relationship
<br />
<br />212
<br />
<br />.
<br />
<br />used (Table 2), This is not surprising given the strong
<br />functional dependence of Z-R,computed rain rates on
<br />sampling error, radar calibration, and variations in the
<br />drop size distribution (cf, Doviak and Zmic 1993),
<br />A previously published R(K"p' ZOR) relationship
<br />applied to the data in both pure and blended forms
<br />(e,g" combined with the COARE tropical Z-R) pro,
<br />vided a reasonable estimate of the STP with minimal
<br />tuning (low by a factor of 0,8), It is hypothesized that
<br />the combined phase and power information contained
<br />in the R(Kop, ZOR) estimate better accounted for spa-
<br />tial and temporal variations in the raindrop size dis,
<br />tribution (e,g" Ryzhkov and Zmic 1995),
<br />
<br />8. Summary
<br />
<br />Extensive meteorological sampling of the environ,
<br />mem, convection, and heavy rainfall associated with
<br />the Fort Collins flash flood provided key insights into
<br />the physical processes responsible for the flood event.
<br />From a meteorological perspective the Fort Collins
<br />case occurred under the influence of many classic syn-
<br />optic signatures (e,g" Maddox et al. 1978; Maddox
<br />et al. 1980; Caracena et al. 1979; Doswell et aL 1996),
<br />
<br />Vol. 80, No.2, February 1999
<br />
|