WMOD00425

• Title: Demonstration of Improved Operational Water Resources Management Through Use of Better Snow Water Equivalent Information
• Date: 9/1/2002
• Weather Modification - Doc Type: Report

I

""
.~.. -
HU-
mt
. ""
.
t
. ""
,
.
""
nil

AZlliUtH:172

ELEVATION ~NCLE.

SUI
H..
4111
ISII
II".
.
t
Hit .
oc ,
,
7111
'4C
HII
1111
III

I

--/
./

I
I
I

I

III

klll_.lul It.. "hI

I
I
I

Figure g.-Hypothetical diagram of precipitation. typing logic of PAA. Altitudes of threshold
temperatures are given by dotted red lines and terrain by solid brown line. Extent of 0.50 and 1.50
beams are solid black lines, and the centers of the beams are dashed black lines. Green. red, and
blue areas of the beam represent beam samples producing rain, melting snow (bright band), and
snow, respectively. Snow samples. in this case, are laken from 105" beam, as prescribed by the
hybrid scan. Other samples are Irom 0.50 beam.

We continuously download soundings from the National Centers for Environmental Prediction
(NCEP) Eta model (12-km horizontal resolution) for the (XXX>, 0600. 1200, and 1800 UTC runs.
Since these data become available 2~3 hours after lhe run cycles, soundings from the 3. 10 9~hour
forecasllimes arc nominally used (at 9 hours. the next run becomes available), From each hourly
sounding, Ihe highest 0 oc levels and lowest +4 oC levels are calculated for the precipitation type
dClennination within the algorithm. If such levels are below the ground at the sounding site, they
are approxim~ltely determined by extrapolating the temperature downward at the dry adiahatic
lapse rate.

I

Since the Rapid Update Cycle (RUC2, Benjamin et aI., 1998) model is updated hourly. I-hour
forecast soundings from Ihis model were compared to the Eta 3-hour forecast soundings for a few
snowslonns in the 20<X>-2001 cool season. The Eta produced slightly beller forecasts of low-
level temperatures in these storms, so we continued to use soundings from the Eta model.

I
I
I
I

The precipitation-typing procedure is subject to numerical model errors (c.g.. Evans and Grumm,
2()(x)) that can adversely affect the altitude of relevant bright band elevalioos. For purposes of
PAA output into the NWS river models, however, the new output will be fed into a more
rigorous precipitation typing system that lessens the need to funher refine our procedure at this
time (SNODAS. see section 7).

24

I
I
I
I
I

There are other efforts to deal with the bright band prohlem. NSSL's Worldwide Integrated
Sensor Hydrometeorology (WISH) group is developing a bright band identification scheme
(Calvert ;'1I1d Gourley. 20(0) thaI uses only radar data, updated every volume scan. The VPR
adjustments of Seo et al. (2(X)0) correct Ze values for precipitation estimation within the bright