Of all the information that forecast offices receive from spotters, wind speed estimates may be the most prone to error. Some reports are based on damage while others are based on spotter-derived estimates. These estimates are very subjective! While spotters do their best, they tend to overestimate win speeds (LaDue, 2003).
This issue can be a particular problem at marginally severe or sub-severe wind speeds since more attention has been paid to observing these events over the last several decades (Weiss et al., 2002).
In the cases where damage is reported, there is at least some objective result to the wind event. The problem is that wind damage is often not well reported (or not well documented before it gets to the warning forecaster).
From a small sample of events, we found that reports in phone logs generally had some flavor of wind damage listed in only 1 in 5 reports. The climatological record also lacks detailed records of damage
for ¾ of reports (Weiss and Vescio, 1996).
Even if damage is reported, much of the wind damage reports are characterized by tree damage. It’s doubtful that many spotters know all the factors that may impact how strong winds must be to damage a tree. In some areas of the country, significant tree damage can occur at wind speed below the severe criteria.
The point here is that it is very easy for spotters to misjudge the intensity of thunderstorm winds. How are warning forecasters supposed to know if spotters are getting it right?
While wind gust estimates may not always be accurate, they do have value. Much like hail reports, forecasters should be aware of people seeming to repeat back forecast wind speed values from your products.
In our review of some previous events, it was found that about 2/3 of wind reports included some kind of wind estimate.
Instead of focusing on the numerical value of the wind estimate, it might be better to look at where the estimate fits into a range of values relative to the severe threshold. Say something like 0-20 (light), 20-45 (strong, but definitely sub-severe), 45-65 (marginally sub-severe to marginally severe), 65-80 (definitely severe), and 80+ (take cover now).
The numbers will, and should, vary depending on your CWA. One key to such a system is to have a good idea at what wind speed tree and other wind damage occurs in your CWA. Why you ask? So you can use reports of wind damage to QC the wind estimates. Of course, you can’t use wind damage to quality control estimates if you don’t receive reports of damage.
When a report comes in, especially if you’re receiving a first hand report, ask if there was any damage. Make sure you document the answer! You’re likely to forget which damage occurred where in a few minutes if you don’t! Another good way to QC wind estimates, if possible, is to use measured wind speeds from METARs, mesonet sites, or spotters (via portable or hand-held anemometers) located near by.
It is helpful if peak gust values from these sites can be recorded on spotter log sheets as storms move through. Even if the data comes in minutes, or an hour later, it helps to get the information in your logs.
From:
http://www.wdtb.noaa.gov/courses/awoc/docu...sson3_print.pdf[/b]
