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
I

5. SAA IMPLEMENTATION, TESTING, AND RESULTS

The SAA was run locally at Cleveland and Minneapolis on WDSS systems (1996-97).
SAA users indicated satisfaction after these field tests, with some need for improvement (Super
and Holroyd. 1997a). The simple range correction scheme was implemented before 1997-98
testing al Albany. Bright-band contamination of SAA output was frequent at this site.
Unexpectedly. however. the algorithm outperfonned the PPS for cold stratiform min in thai
region (Quinlan and Sinsabaugh. 1999).

I
I
II
II
I

Data from Super and Holroyd (1998) showed
thaI the power P in the Z-S relationship
varied linlc from 2.0 and that S was
insensitive 10 these variations. The
coefficient n, however. showed
significant spatial variation and a
corresponding strong influence on S.
Figure I illustrates this variation with
mdar range for several radar sites.
Invariancc of p and substantial variation
of a was also evident in a climatological
study of rain in t>.lississippi (Steiner and
Smith. 20(0). NOle that while all sites
show a decrease in a with range. this
decrease (slope) is highly variable
between them.

I

I
I
I
I
I
I
I
I
I
I
I

R~n OJ'' ~m
Figure 1.-Plols of individual relations for variation of
alpha with range. Thick curved line is VPR range
correction used for the 1998-99 season for the north-
central States.

300

'"
N
"
., 200
~
.
1Il

"
o
~

o "'0
<
o
<:

"

.

~.
~
~ ~
~~ "
. ~
~o---".~h'
.'Vt"...~.~
''4:...... ...........'"
'-."

.'""
.-,
.

.

.

',"

.

.

.

o

40

80

'20

Since the radar beam becomes
increasingly elevated above the earth's
surface with increasing range from the
radar, the depicted decrease in a also represents a decrease with elevation. Because a controls S,
which is calculated from~, then 'l.: is also diminishing with heam elevation. This change of7~
with height is referred to as the vertical profile of reflectivity. or VPR. The usual dccreasc in S
with radar range is caused by the radar's inability to sample precipitation near the surface at long
ranges. This situation is exacerbated with stratiform rain and snowfall, which are shallow and
lend to have maximum ~ values near the terrain (Joss and Lee. 1995). The range correction was
initially fomlUlatcd by constructing a VPR, then the vertical position was converted to an
equivalent range using standard beam refraction. This modification was related by Holroyd
(1999). The VPR profile assumed a linear decrease of about 20 percen! in Z. per kilometer
height above ground. It was realized that, in certain situations, a VPR rather than range
correction should be applied. Reclamation submitted a January 1997 proposal to the aSF. at
lheir request. to pursue this change. This proposed work was not funded. Reclamation
eventually developed a VPR correction for the SAA with GCIP and Reclamation's S&T Program
support. and this modification will be discussed in seclion 6.2.

13