Understanding wind measurements

[broadstairs]

Steve I am trying to get my head around the wind displayed on the main screen in particular the High Wind value. I have two installs of Cumulus both working off the same VirtualVP data stream on different PC's. My main install on the weather PC runs 24x7 (except for the odd update) the other test install gets restarted each day and does catch up from the archive data. Now I would expect both to display pretty much equivalent values, however the test install now shows a High Wind of 24.7 at 03:12 having downloaded the overnight data from the logger. The 'live' install currently shows High Wind of 18.7 at 03:32. I have ticked for Cumulus to calculate the 10 min average BTW.

Apart from the discrepancy of data what exactly is High Wind - I am presuming it the high or max average wind speed, is this correct?

Also neither Cumulus agrees anywhere near with WD's max average (calculated) of 26.3 at 03:31 although one nearly agrees on timing and the other is closer on speed!

Stuart

[steve]

Yes, the 'high wind' is the maximum value of the 10-minute average wind speeds. During the processing of the data from the logger, the 10-minute wind calculation probably doesn't have enough data points to work with. I suspect this is why you see the discrepancy between the two Cumulus values. I do realise that this means I am saying that the 18.7 figure is the more correct of the two! How did the two highest gust figures compare for value and time (and what were they)? What is your VP logging interval?

For stations that supply a 10-minute average themselves, like the VP and VP2, you're probably better off using that anyway, rather than getting Cumulus to calculate it. It would be interesting to compare the two Cumulus figures in that case. The timestamps will never agree all the time to the minute, as they will depend on the logging intervals, etc, but you would expect them to be within a few minutes of each other.

Steve

[Bob]

on this subject (almost)

Can anyone give guidance on wind calibration.

Now we have the ablility to adjust calibration in Cumulus presumably there is a factor that one uses todo with Height Above ground.

Mine is 10foot up...... in clear, I think they should be up higher for accurate readings so should I add a bit (a calibration factor)

regards

Bob

[broadstairs]

Yes, the 'high wind' is the maximum value of the 10-minute average wind speeds. During the processing of the data from the logger, the 10-minute wind calculation probably doesn't have enough data points to work with. I suspect this is why you see the discrepancy between the two Cumulus values. I do realise that this means I am saying that the 18.7 figure is the more correct of the two! How did the two highest gust figures compare for value and time (and what were they)? What is your VP logging interval?

For stations that supply a 10-minute average themselves, like the VP and VP2, you're probably better off using that anyway, rather than getting Cumulus to calculate it. It would be interesting to compare the two Cumulus figures in that case. The timestamps will never agree all the time to the minute, as they will depend on the logging intervals, etc, but you would expect them to be within a few minutes of each other.

Steve

Steve my high gusts match exactly with the test and 24x7 Cumulus and WD, its just the averages which dont match. I'll set Cumulus to use the VP measurement and see what difference that makes, however I'm still somewhat puzzled by the difference with WD. My VP logs at 1 minute intervals, so the average would be over ten data points. I agree that timestamps are not likely to match exactly but I wold expect them to be within a minute for both averages and gusts since all the data stream is the same via VirtualVP. I guess that since I have a 1.10 multiplication factor for gusts then I need to add that for the average as well if I use the VP average.

on this subject (almost)

Can anyone give guidance on wind calibration.

Now we have the ablility to adjust calibration in Cumulus presumably there is a factor that one uses todo with Height Above ground.

Mine is 10foot up...... in clear, I think they should be up higher for accurate readings so should I add a bit (a calibration factor)

regards

Bob

Bob all the stuff I've read on home anemometer installs say that most are likely to under read by at least 10% and probably more, some folk suggest 20%. Reading up on best install practice says it should be mounted 10 metres above the ground in an open aspect, try doing that in the average back garden and you will see the problem. I would suggest that the minimum addition to wind speed should be 10% and perhaps if you have a good and very close by reference site you could then compare over a period of time and see if it needs increasing.

Stuart

[steve]

Now we have the ablility to adjust calibration in Cumulus presumably there is a factor that one uses todo with Height Above ground.

Mine is 10foot up...... in clear, I think they should be up higher for accurate readings so should I add a bit (a calibration factor)

The formula for correcting wind speed from height h to height 10 metres (the 'standard' height) is:

Code: Select all

Vh/V10 = 0.233 + 0.656 * log10(h+4.75)

where h is in metres, Vh is the speed at height h, and V10 is the speed at 10 metres (source: UK Met Office Observers handbook).

I suppose I could put this calculation into Cumulus, so all you would have to do is supply your anemometer height.

Steve

[steve]

Steve my high gusts match exactly with the test and 24x7 Cumulus and WD, its just the averages which dont match.

Yes, but what were the actual gust figures?

timestamps are not likely to match exactly but I wold expect them to be within a minute for both averages and gusts since all the data stream is the same via VirtualVP.

Yes, with a 1-minute logging interval, you'd expect all the timestamps to be within a minute.

I guess that since I have a 1.10 multiplication factor for gusts then I need to add that for the average as well if I use the VP average.

The released version of 1.8.1 doesn't apply the gust calibration to the calculated average wind speed (as I mentioned elsewhere), so using the VP average or not shouldn't affect your decision to use a wind calibration.

Steve

[broadstairs]

Now we have the ablility to adjust calibration in Cumulus presumably there is a factor that one uses todo with Height Above ground.

Mine is 10foot up...... in clear, I think they should be up higher for accurate readings so should I add a bit (a calibration factor)

The formula for correcting wind speed from height h to height 10 metres (the 'standard' height) is:

Code: Select all

Vh/V10 = 0.233 + 0.656 * log10(h+4.75)

where h is in metres, Vh is the speed at height h, and V10 is the speed at 10 metres (source: UK Met Office Observers handbook).

I suppose I could put this calculation into Cumulus, so all you would have to do is supply your anemometer height.

Steve

Steve that still assumes that your anemometer is not in any wind shadow which is highly unlikely in a domestic installation plus its all very well saying it at 10 metres but if that is on a roof and then its only 1 metre above the roof line that throws out the whole calculation anyway.

I still feel that leaving it up to the individual to adjust with some guidance perhaps in the help file is the best way forward.

Stuart

[Bob]

blimey, heavy maths!!!

So mine is at 3m...it should be 10m's so my calibration tweak is?

Bob

[broadstairs]

Steve my high gusts match exactly with the test and 24x7 Cumulus and WD, its just the averages which dont match.

Yes, but what were the actual gust figures?

Steve

Sorry forgot to say that the max gust today was 35.2mph at 03:18 I mentioned the max average values in the original post.

Currently the test Cumulus and the 'live' one are tracking very close on current average wind speeds test is using VP and live is calculating. Both are a bit under WD's current average calculation.

Stuart

[broadstairs]

blimey, heavy maths!!!

So mine is at 3m...it should be 10m's so my calibration tweak is?

Bob

Bob in my view that calculation is only worth doing if your installation is (apart from height) ideal that is with no obstructions for quite a considerable distance all around and mounted on top of a 3 metre pole or tower (not 3 metres up on a roof and only half a metre from the roof line). I'd start with at least 10% (1.10 in cumulus parlance).

Stuart

[Bob]

its on a pole 3m up in the clear.I could put it aloft a 110ft tower LOL.

[broadstairs]

its on a pole 3m up in the clear.I could put it aloft a 110ft tower LOL.

How clear is clear? How far away is the nearest obstruction?

Stuart

[Bob]

about 25-30 foot

[broadstairs]

about 25-30 foot

I doubt that would be called clear enough although much better that I can achieve in my small garden.

Although it applied to wind turbines the recommendation is to place the turbine (or in our case anemometer) 9 metres above the tallest obstruction within 100 metres of the installation, so if your tallest obstruction (including trees not just buildings) within 100 metres radius is say 8 metres you tower should be 17 metres tall. This all starts to get very silly in terms or a home weather station, unless you live on a farm.

I'd start by trying to find a reference station reasonably close (like a METAR on an airfield or a Met Office reporting site) and see what sort of comparison it makes with your station. This will only a a rough comparison because we all know wind varies greatly between points. If your max wind is say within 10% of that station then maybe 10% is the right figure, if you are reading considerably differently the adjust the figure up, it is very unlikely you would be reading over.

Just my ten pence worth, as always YMMV....

Stuart

[Bob]

Hi Stuart,

I do live on a farm...well have some land, and have three towers 50ft, 65, 100ft...but all have rotating antennas at the top so would have to put sensors off to the side to avoid north moving with antenna LOL.

Anyway it agrees pretty well with RAF shawbury so I'll leave alone, just that Steve gives us new features, feel that you want to play!!!

cheers Bob

Amateur Radio Station G4UJS