My WebLink
|
Help
|
About
|
Sign Out
Home
Browse
Search
WMOD00282
CWCB
>
Weather Modification
>
DayForward
>
WMOD00282
Metadata
Thumbnails
Annotations
Entry Properties
Last modified
7/28/2009 2:32:29 PM
Creation date
1/8/2008 11:54:38 AM
Metadata
Fields
Template:
Weather Modification
Sponsor Name
USBR Technical Serivce Center, River Systems & Meteorology Group
Project Name
Snow Accumulation Algorithm for the WSR-88D Radar, Version 1
Title
Snow Accumulation Algorithm for the WSR-88D Radar, Version 1
Prepared For
USBR
Prepared By
Arlin B. Super and Edmond W. Holroyd
Date
6/1/1996
State
CO
Weather Modification - Doc Type
Scientific Study
There are no annotations on this page.
Document management portal powered by Laserfiche WebLink 9 © 1998-2015
Laserfiche.
All rights reserved.
/
115
PDF
Print
Pages to print
Enter page numbers and/or page ranges separated by commas. For example, 1,3,5-12.
After downloading, print the document using a PDF reader (e.g. Adobe Reader).
Show annotations
View images
View plain text
<br />/ <br /> <br />scheme discussed in section 11 been applied. The relatively high R values also suggest that <br />gage measurements were reasonably accurate. More variability might be expected with <br />unshielded gages in windy locations. <br /> <br />It is reasonable to assume that equation (8), derived from the gage nearest the radar, offers <br />the best available Ze-S relation for the Cleveland area. In order to provide additional insight <br />into range effects, figures 4 through 7 were prepared by applying equation (8) to each gage's <br />data and the vertical array radar data. (The radar array was 3 by 3 range bins for each of <br />these gage locations.) A regression line was also calculated for reference as shown by the <br />dashed lines. Gage values that appeared to be outliers on these and all similar plots, and <br />all hourly totals of 0.10 inch h-1 or more, were verified by checking yet again against the <br />original charts. <br /> <br />Table 9 summarizes the results of the various calculations. <br /> <br />Table 9. - Summary of applying equation (8) to the data set from each Cleveland area gage. Also <br />noted are the intercept (A) and slope (B) of regression lines (gage = A + B x radar), correlation <br />coefficients (R), standard error of estimate, average hourly gage-observed and radar-estimated snowfall <br />accumulations, and the gage/radar estimate ratios. The number of pairs is the same as given in <br />table 8. <br /> <br />Gage Distance A B R Standard Gage Radar ,;fh J 4..v-1 <br />No. from Radar (in) Error Observation Estimation Gagel <br /> (km) Estimate {in} (in) (in) Radar C~JL <br />1 36 0.001 0.985 0.73 0.0136 0.0196 0.0196 1.00 /.pv <br />2 61 0.000 1.187 0.80 0.0140 0.0204 0.0173 1.18 ' r'/ g <br />3 87 0.004 1.400 0.68 0.0224 0.0260 0.0160 1.63 .fc/5 <br />4 115 0.009 1.856 0.72 0.0177 0.0232 0.0074 3.14 '3/~ <br />5 146 0.013 1.607 0.55 0.0162 0.0196 0.0043 4.56 - -u9 <br /> <br />Reference to figures 4 through 7 and table 9 shows a number of interesting features. Most <br />important, the gage/radar ratio increases with range as anticipated. The ratios suggest that <br />radar estimates are 85 percent (111.18) of gage average snowfall by a 61-km range and 61 <br />percent by 87 km. Bya 115-to 146-km range, the radar predictions have fallen to only 32 to <br />22 percent of the gage amounts. As shown in table 8, the center height of the 0.50 tilt beam <br />increases from about 390 m above gage No. 1 to over 2500 m above gage No.5, so <br />underestimates with increasing range should be expected. <br /> <br />Although radar underestimation is serious beyond 60 km, these results are not as drastic as <br />those reported by Wilson (1975) for a nearby region of the country. Gage:radar ratios were <br />estimated from the curve on his figure 2 for the ranges of Cleveland gages No. 1 to 4. These <br />ratios, divided by the 36-km ratio, yielded results compatible with table 9 (all of Wilson's <br />ratios exceeded 3.0 because of use of an inappropriate Z~S relation). Resulting values for <br />ranges of 61, 87, and 115 km are 1.7, 3.7, and 18.0, much higher than the ratios of table 9. <br />At least three factors probably improve the results given here. The WSR-88D radar beam <br />is narrower (0.95 versus 1.70) and tilted nearer the ground (0.5 versus 0.90) than the radar <br />used by Wilson. In addition, the WSR-88D is more sensitive than radars available at the <br />time of Wilson's study. <br /> <br />27 <br />
The URL can be used to link to this page
Your browser does not support the video tag.