FLOOD01375

• County: Larimer
• Title: Mesoscale and Radar Observations of the Fort Collins Flash Flood of 28 July 1997
• Date: 2/1/1999
• Prepared By: American Meteorological Society
• Floodplain - Doc Type: Flood Documentation Report

28 JULY 1997,2130 MDT: U-WIND (Z=I KM)

, , , I I JIIIIII-JIIIIII
-14 -10 ,6 -2 0 2 6 10 14
xo

60
4"
:;;
.
~
.
Q 20
"
;
,-
-; 0
;-
,E
or. -20 ,C , 0"
i ,
.
-
=
, -40
,00
-xo
..j" -20 "

.

~,
I. i:
'AIa .-->
9'7 / ....
'\~~"
"\ 'i,:"'",
~,,,

.~

20

40

0"

Distance E-\\' rrom Taft and Drake (km)

HO

One interesting characteristic of the
convection was the marked reduction in
lightning activity' as compared to that of
convection located 40-80 lan southeast
of FCL (e,g" the bow echo system;
Fig, II), For example, the peak in CG
lightning activity for the FCL event oc,
curred between 1930 and 2015 MDT
(minutes 120-165, Fig, 15), yet CG flash
rates were only 0,5 flashes per minute,
Conversely, peak CG flash rates of 3-4
per minute were typical of the cells 10,
cated southeast and east of FCL in
Fig, I L During the heaviest rainfall of
the evening (2125-2205 MDT, minutes
240-280 in Fig, 15), only 5 CG flashes
occurred over a 40,min time period, The
relative reduction in CG lightning activ,
ity, coincident heavy convective rainfall,
and observed thermodynamic structure
of the troposphere (Fig, 6; section 4c)
suggest the presence of a rainfall process
120 similar to that characteristic of tropical
environments (e,g" Rutledge et aL 1992;
Williams ct aL 1992; Zipser and Lutz
1994; Zipser 1994; Petersen et aL 1996;
Petersen et aL 1999),
Indeed, the mean vertical structure of
reflectivity associated with the flood con,
vection (Fig, 16) bears some resemblance to that ob,
served in tropical monsoon-oceanic convection (cL
Szoke and Zipser 1986; Williams et aL 1992; Zipser
and Lutz 1994; DeMott and Rutledge 1998), For ex-
ample, in Fig, 15, although echo tops olien exceeded
12 lan, mean reflectivities> 35 dBZ were generally
located near or below the height of the -IOOC level
(5,5 km), The decrease in the normalized vertical gra'
dient of radar reflectivity at temperatures lower than
OOC (Fig, 16) in the FCL case is similar to that reported
for convective cores over the tropical Atlantic Ocean
during the Global Atmospheric Research Program
(GARP) Atlantic Tropical Experiment (GATE) (e,g"
Szoke and Zipser 1986) and only slightly smaller than
that observed over the western Pacific during TOGA
COARE (DeMott and Rutledge 1998), As in the tropi,

,
,

..
L

>

}.
.
,
,
,~

100

FIG. 13. Dual-Dopplcr-denved [I-wind component (shaded) for lhe I-km height
level at 2130 MDT 28 July 1997. Positive (negative) values of u are contoured
every 2 m 5-1 using a solid (dashed) line. Grid origin as in Fig. 9.

several individual cells embedded within the broader
envelope of these pulses (e,g" minutes 25, 120,240)
can be summarized as follows,

I) t = 0 mine Development of radar echo bclow the
3,5-4,lan level (2 OOC),
2) t = 10-20 mine Peak in echo,top elevation between
9 and 13 km, In the minute 120 and 240 rainfall
pulses, CG lightning (albeit infrequent) occurred
with penetration of 30-35 dBZ mean reflectivities
(maximum reflectivities 2 40 dBZ) into regions of
the cloud colder than -lODe. The occurrence of
larger reflectivity in the cold region of the cloud is
indicati ve of a more vigorous updrali and a more
robust ice process, both of which are conducive to
enhanced electrification processes (e,g" Saunders
1995; Williams 1995),
3) t = 25-40 mine Descent of the echo "core," peak
in the surface rain mass flux, and continued CG
lightning (minutes 120,240),
4) t = 40-60 mine Rainfall begins to decrease, CG
lightning ceases,

206

.'In addition to the small number of CG flashes detected by the
NLDN, a multitude of observers in FCL noted a distinct lack of
lightning and thunder afler sunset during the period of heaviest
rainfall.

Vol. 80, No.2, February 1999