Sometimes the wind gust in the Wind Speed graphic on FreshWDL is less than the sustained wind speed, which should not be. Does anyone know why this happens in FreshWDL and how it can be mitigated? See attached example screen capture.
This actually occurs quite often in WD and hence FreshWDL. I presume it is because the wind is the “rolling average of the last 60 readings, updated every sec, unless station provides average” according to clientrawdescription.txt, while the gust is just the one snapshot reading.
If it’s any consolation, I don’t actually understand the explanation. . . I presume it means it’s the average speed over a minute.
Or can be set to 10 minutes if remember right…
Yeah, I should have said it happens in clientraw.txt and hence FreshWDL. . . I plot a 10-min average in WD graphs but I do not know what the “Average Wind Speed” shown in WD Current Conditions is.
Ah, that might be it, just on the graphs, i cant look in the advanced menu to double check, i think there may be something there on the program tab, entering the advanced menu breaks my main screen so i will let you check if ya dont mind and get the chance, please?
No, you’re right, there is the option to use 10-min average on the main screen Current Conditions too. So I suppose the default is 1-minute average, like clientraw.
Thanks for checking for me, i thouth i’d seen it somewhere, acutally, i think that is what i am using, both there and the graphs, just trying to install WD on to this Windows 11 machine, i think i need to install in compatability mode. so then i can check these things myself 
I was looking at my clientraw.txt using Ken’s WD clientraw parser (Saratoga-Weather.org - Weather-Display clientraw*.txt parser) and it shows:
The first value, 12345 is field 0 and is labeled “Beginning of file”. ,
The second value, field 1, is labelled “Average Windspeed (avg last 60 secs)”, and
The third value, field 2, is labelled “Current Windspeed”.
If FreshWDL is using fields 1 and 2, then maybe that could explain why the “gust” is sometimes lower than the sustained (average) wind speed if it’s plotting the current wind speed that might not be the gust.
Field 140 is labelled “Max Gust In Last Min”. But that’s not an instantaneous gust, but it looks like it might be a better field to use?
For my current clientraw.txt is shows:
Average Windspeed (avg last 60 secs) = 9.9 mph
Current Windspeed = 5.0 mph
Max Gust In Last Min = 19.6 mph
I’m sure it is.
then maybe that could explain why the "gust" is sometimes lower than the sustained (average) wind speed if it's plotting the current wind speed that might not be the gust.
I think it is just the current wind speed at the time clientraw.txt is updated.
Field 140 is labeled "Max Gust In Last Min". But that's not an instantaneous gust, but it looks like it might be a better field to use?
Interesting idea, covers the same time period as the average speed. But #140 is only integer knots. . . #1 and #2 are decimal knots. Does that matter?
See here for similar discussion re PWS Dashboard.
I think even though #140 is in integer knots, at least the gust should then always be higher than the one-minute average wind speed (#1).
gust is actually current speed
Hi Brian,
Then how can the gust, or current speed sometimes be less than the average speed?
Bill
very easily
because current windspeed includes lulls and gusts
e.g
lull is 5, gust is 15, average is 10
so if you happen to see a current/gust speed of 5 but the average is 10, then its a lull
Brian,
I have a recommendation and apologize in advance for such a long post, but I thought it would be a good idea to look at some official definitions of a wind gust.
Your statement that the current windspeed includes lulls and gusts is correct and within an average wind speed there are maxima (gust) and minima (lull) that results in an average. But the gust is always the maximum speed recorded within a period of time. So, while the current wind speed could be a lull or a gust (better called a min or max), the lull cannot be the gust. For years, I’ve wondered how the WD “gust” could be less than the average but never really dug into it. In the attached screen capture within the yellow circle, the “gust” is sometimes lower than the “average” which is incorrect if the red line is actually supposed to be the gust. But if the red line is the “current speed” and blue line is the “average speed”, then the graph is correct but it’s not showing average and gust, it’s showing average and current. Now that I’ve looked into the clientraw.txt fields, I understand. I also recommend that consideration be given to use the average and gust for reporting wind in WD instead of average and current. I respectfully make my case below.
I spent 15 of my 40 years as a meteorologist working with weather sensors at Kennedy Space Center. For space launch, Space Shuttle landings, and now SpaceX booster landings, the wind gust, also called peak, is defined in this NASA technical report (https://science.ksc.nasa.gov/amu/final-reports/windstats-phase4.pdf) that my group completed and on page 11 it states: “The raw wind speed and direction were sampled every second. The 5-min mean is the average of 600 1-sec observations in a 5-min period. The peak is the maximum 1-sec speed in the 5-min period.”
Looking at official wind gust definitions from several government publications also indicates the gust is always the maximum wind speed as follows:
The NOAA NWS Glossary (https://w1.weather.gov/glossary/index.php?word=wind+gust) defines a wind gust as “Rapid fluctuations in the wind speed with a variation of 10 knots or more between peaks and lulls. The speed of the gust will be the maximum instantaneous wind speed.”.
The UK Met Office definition (How we measure wind - Met Office) states: “The gust speed and direction are defined by the maximum three second average wind speed occurring in any period.”
The US Federal Meteorological Handbook No. 1, Surface Weather Observations and Reports (https://www.icams-portal.gov/resources/ofcm/fmh/FMH1/fmh1_2019.pdf) definition of a wind gust is “The maximum instantaneous speed in knots in the past 10 minutes is reported.”
The US National Institute of Health, which oversees the National Institute of Standards and Technology (NIST), published an article in 2018 titled “Wind Gust Measurement Techniques - From Traditional Anemometry to New Possibilities” (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5948875/). The authors, from the Finnish Meteorological Institute, state “The measurement of wind gust speed requires a high temporal resolution of the anemometer system, because the gust is defined as a short-duration (seconds) maximum of the fluctuating wind speed.”.
In the US, the Automated Surface Observing System (ASOS) network records wind gusts as the maximum 5-s average of wind speed during the past 10-min period preceding the time of observation. At the same time, the averaging time for the mean wind speed is 2 min.
Finally, Davis states a wind gust is “Highest measured wind speed over 2 or 10 minutes or another period.”
All this said, as a professional meteorologist, I recommend that we follow the definition of a wind gust as the maximum wind speed within a given period. Equating the “Current Speed” from Field 2 to a “Gust” is incorrect by accepted definitions. Therefore, it would be correct to report Field 1 from clientraw.txt as the “Average” defined in clientraw.txt as “Average Windspeed (avg last 60 secs)” and the “Gust” from Field 140 defined in clientraw.txt as “Max Gust In Last Min”. Doing so would be consistent with the accepted official reporting of winds in which over a 60-second period, the gust is the maximum speed recorded and the average is the mean of all the measured speeds in that same 60-second period.
Bill
I am a little confused, are you saying the gust cannot be lower than the average (in the graph), doesn’t the gust drop though? It’s not always up there, it has to come down again, therefore it will show lower than the average, the average is over x amount of time, the gusts are instantaneous and will drop lower than that, what you shouldn’t see and we dont, is the max gust lower than the max average.
That’s correct, a gust plotted versus average can never be lower than the average, based on the definition of a gust. The gust is not instantaneous, in clientraw.txt, gust is correctly defined as the maximum wind speed in a 60-second period. And the average is correctly defined as the average in a 60-second period. Mathematically, the gust must be higher than the average for that 60-second period. But current speed is the instantaneous speed, which can be the gust or the lull. That’s why I believe the WD graph is plotting current versus average and since the current can include a “lull”, it sometimes shows up lower than the average on the plot. After all, what is an average? It’s the mean values of all low and high values. I also think that this is why the WD graph sometimes doesn’t display the maximum gust in a time period, as several people have noticed and posted, including myself in this previous post from December: http://discourse.weather-watch.com/p/557401. WD plotted what I assumed were the gusts, yet the data showed higher wind speeds than the plot, which makes sense if the plot was based on current, or instantaneous, wind speeds and the actual maximum speed within a 60-second period. And when compared to WU, they must be actually plotting the gust, not the current speed as I show in several graphs in that post.
Brian, I agree completely with Bill. I know from personal experience that listening to him will always lead to a win-win!
What about a Davis wind, that’s just under every 3 seconds sent to wd
I would think the Davis 3-second send to WD would be the “current wind”, but again, by definition of a gust, Field 140 of clientraw.txt should be the maximum speed measured of the 20 3-second wind speeds sent by Davis during the 60-second period defined in Field 140. And likewise for other manufacturer’s wind sensors, depending on the anemometer send rate.
For example, say we had 20 3-second wind speeds measured by the wind sensor in a 60-second period as follows. The average speed is 11.2 and the maximum or “gust” is 19.0. Any one of the 20 measured speeds could be the “current” speed that varies from a “lull” of 3.0 to a “gust” of 19.0 but if we want to plot the gust every time, it has to be the maximum in that 60-second period, not just any “current speed”. Admittedly, I don’t know the inner workings of WD or exactly how the gust, speed, and lull are determined and written to clientraw.txt - that’s Brian’s area of expertise. But I do know how to determine all three mathematically and by standard definition the gust has to be the maximum speed recorded in that period.
Speed
12.0
13.0
7.0
3.0 (minimum)
9.0
15.0
17.0
13.0
11.0
8.0
13.0
19.0 (maximum)
17.0
7.0
4.0
8.0
11.0
14.0
9.0
13.0
Average
11.2
Max
19.0
the gust line on the graph in WD is the maximum current speed over the last minute
and so if there is a long lull, then it can be less than the rolling average