Wind averaging

[MacDude]

Hi

Having just had a rather windy weekend it got me to checking our reported wind speed figures. I have an Oregon weather station and therefore turned on the 10 minute averaging on Cumulus. I assume this is a simple average of the values received from the station over the last 10 minutes. How does it work for gusts though? And does anyone happen to know what algorithm the WM928 uses to define a gust? A literature search has underwritten the 10 minute average for wind speed as a standard but I can't find any references to gusts.

We seem to be reporting average wind and gust speeds well below other local stations and, even given the less that ideal siting of our sensor, I am surprised by the differences.

All ideas welcome.

Thanks

Andy

[steve]

Andy,

The WMR-928 sends two wind speed values in each wind data packet (I think it's every 12 seconds). In the documents describing the protocol which are available on the 'net, and which I've used for Cumulus, the first value is usually described as 'gust' or sometimes just 'speed', and the second is the 'average speed'.

Cumulus displays the 'gust' as 'Latest' on the main screen, and uses the highest of those values over the previous 10 minutes for the 'Gust' on the main screen.

If you have averaging turned off in Cumulus, it just displays the average as received, and if you have averaging turned on, it does as you suspect - it takes a simple average of all the 'gust' values received in the previous 10 minutes and ignores the supplied 'average' value.

What I don't know, and have never been able to find out, is what the WMR-928 does to calculate the two figures it supplies.

Steve

[MacDude]

Wouldn't a better 10 minute average be an average of the averages rather than the average of the gusts? As you say it depends on what the WMR928 is really measuring, weather "gust" or "speed" means "instantaneous value now" and "average" means "average since last values sent" or if "gust" is actually the "peak value since last value sent".

Is the "average" ever higher than the "speed"? If so then "speed" <> gust.

Are there any good forums where WMR928 hardware is debated?

thx/andy

[steve]

Andy,

A few forums that I know of are at http://www.ukweatherworld.co.uk, netweather.tv, and at accuweather.com.

I've just done a bit of a search, and found Tim from Meteorologica posting in a few places about the WMR928 and its wind speed calculations. He says that the anemometer basically counts the number of rotations over a 14 second period, works out the wind speed over the 14-second period from that, and then sends that figure to the console, which reports this as 'gust'. So that's actually an average already. He says the console then uses those values to calculate an average over a longer period (he's not sure how long) which it reports as the 'average'.

If this is correct, it doesn't seem to matter which of the two values I use to calculate the 10-minute average. It also explains why the WMR928 will typically report lower gust speeds than other stations that sample more often (e.g. the Davis stations, which sample at 2.5 second intervals.). You would expect the average speeds to be roughly comparable, though.

How did your figures actually compare to the nearby stations? How far off the ground is your anemometer?

Steve

[MacDude]

Thanks for the explanation. It starts to make sense now.

My sensor is about 30 feet off the ground but is slightly shadowed from some directions. I am planning to move it to on top of my house which is higher and pretty clear of any obstructions. I am sure that will improve things.

I was typically getting readings about half the local RAF stations although about 75% of my nearest neighbour on Wunderground who has a WS2300 with WeatherDisplay.

thx/andy