WMOD00283

• Sponsor Name: USBR, Technical Service Center, River Systems 7 Meteorology Group
• Project Name: Use of NEXRAD WSR-88D Radar Snow Accumulations Algorith in the GCIP LSA-NC
• Title: Use of NEXRAD WSR-88D Radar Snow Accumulations Algorith in the GCIP LSA-NC
• Prepared For: USBR Technical Service Center
• Prepared By: Arlin B. Super
• Date: 1/1/1998
• State: MN
• Weather Modification - Doc Type: Report

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

b. Determine the feasibility of using Level III reflectivity data with the BAA and
compare results with Level II data runs.

As will be discussed further, the SAA was designed to use high resolution
Level II equivalent reflectivity factor (hereafter simply reflectivity)
measurements. These data are available at each WSR-88D radar, but not
available in real time to non-NEXRAD agencies (a few research universities
are exceptions). However, a lesserresolution product called Level III base
reflectivity is available in near-real time. It was desirable to modify the SAA
to use the Level III product as input and to compare the results of Level II
and Level III runs. If not too much loss of accuracy results with Level III
data, the SAA could be run practically anywhere. This capability is
particularly important over the next few years until the SAA is added to the
standard suite ofNEXRAD algorithms (O'Bannon 1998).

2. USE OF THE VERTICAL PROFILE OF
REFLECTIVITY FOR RANGE CORRECTION

Reclamation is working toward improving its SAA by developing a range correction
scheme based on the vertical profile of reflectivity (VPR). It is well documented
that radars underestimate both stratiform rain and snowfall at mid to far ranges.
The primary reasons are understood. They are related to curvature of the earth,
radar beam spreading, and lack of beam filling. At sufficiently far ranges, even the
lowest available radar beam will completely overshoot shallow precipitation
producing clouds, resulting in negligible reflectivity values and lack of precipitation
detection. Radar underestimation increases with range, in part because of the
known VPR for snow and stratiform rain, which has maximum values near the
suiface. This phenomena has received increased attention in the scientific
literature in recent years. The work of J. Joss and colleagues in Switzerland is
noteworthy as having done much to illustrate this problem and to present correction
schemes. For examples, see Joss and Waldvogel (1989) and Joss and Lee (1995).
Only recently has VPR been recognized as the main source of error in estimating
widespread precipitation with radar as discussed by Joss and Waldvogel (1990) and
Koistinen (1991).

The developing SAA range correction scheme is based on the hypothesis that the
VPR will have similarities among snow storms and that these similarities can be
used to adjust mid to far range SWE estimates closer to reality, based on a seasonal
average correction. As pointed out by Joss and Waldvogel (1989), even a seasonal
adjustment (as opposed to daily or real-time adjustments) is preferable to totally
ignoring the known range underestimation effect as is done with the current
NEXRAD operational precipitation algorithm. Earlier work, especially in Europe,
has suggested that reflectivity will usually decrease with increased altitude for
snow and stratiform rain. Although short-term VPR variations in time and space

2