What speeds does the #wspddata webtag give?

[brassing]

I am looking at using the #wspddata array to calculate a windspeed average over different periods using php. The wiki says that the array is of wind speeds. I am using a Fine Offset station and I was expecting the array to contain the 48 s average speed values. Instead it seems to contain the gust values. Is this correct? If so is it intentional? The 48 s average values would be much more useful. I can't really see much point in having an array of gust values - you just need a peak gust value usually.

[For those unfamiliar with the Fine Offset, it outputs two speed values every 48 s. One is the average value over the 48 s, the other is the 2 s peak value over the 48 s]

[steve]

It contains the individual (non-averaged) wind speed readings (the 'latest' values), which for the Fine Offset station happen to be what the station calls 'gust'.

[brassing]

It contains the individual (non-averaged) wind speed readings (the 'latest' values), which for the Fine Offset station happen to be what the station calls 'gust'.

Well it isn't just what the station calls 'gust'. They are actually the gust values. i.e. the maximum 2 s value over the 48 s. It isn't an academic point. I have two Fine Offset stations now. The one at home is surrounded by trees and I find the gust values are double the average values. ( I am recording the raw average and gust values at 1 min itervals.) The other station is at my sailing club, in a more open location, and there the gust values are about 50% higher than the average ones. If I posted the gust values on our club website people would sometimes be put off coming to sail by apparently stronger winds than they were happy with. (On the other hand the windsurfers could be disapointed!).

For the Fine Offset the primary wind speed value is the 48 s average and it would be really useful to get access to those values. Should I put in an enhancement request?

I can see that the Fine Offset causes problems by its unusual way of reporting wind speeds. However a lot of people use it so it would be useful to cater for its quirks.

[steve]

Well it isn't just what the station calls 'gust'. They are actually the gust values. i.e. the maximum 2 s value over the 48 s. It isn't an academic point.

The definition of a gust is a tricky thing to pin down. This is what NOAA say: "According to U.S. weather observing practice, gusts are reported when the peak wind speed reaches at least 16 knots and the variation in wind speed between the peaks and lulls is at least 9 knots. The duration of a gust is usually less than 20 seconds."

In any case, many Fine Offset users find that using the values that the station calls 'gust' actually gives a better indication of the true wind speed than the 48-second average value, as illogical as that may appear.

Should I put in an enhancement request?

It would certainly be a reasonable option to output the average values instead of the 'gust' values, so please do.

[steve]

By the way, for the purpose you described, you might want to consider collecting the average speed values from the realtime.txt file instead.

[brassing]

I think most of the fine print is about whether or not you report a gust value or not. Cumulus doesn't have that option for any weather station and I wouldn't want it to. The only way the Fine Offset is different is that it uses a 2 s averaging period rather than the "official" 3 s. My point was that the Fine Offset gust values are what most people mean by that term and are certainly not average speed values. There is an interesting discussion on gust values here: http://www.ukweatherworld.co.uk/forum/f ... p?tid=7018 (the main topic is high end weather stations).

My guess is that the reason a lot of people like the gust values from the Fine Offset is that they tend get installed in non-optimum positions where the actual wind speeds are lower than than the clear air values. The gust values thus tend to be closer to the clear air values. There isn't a lot of validity in using the gust values though. I have done a rough calibration of a Fine Offset anemometer by poking out through my car sunroof and the average speed values seem to be correct within the experimental error (about 10% I would guess) so I don't think there is anything about the Fine Offset per se that would indicate using the gust values as a better measure of wind speed.

I was looking at using the realtime.txt average value as you suggest and may well end up going that route (at least until, or if, the option of having a webtag for an array of average values gets implemented). However that is just one value and my web skills aren't up to storing data on the server. I was wanting to produce say 10 min 30min and 1 hour averages and that requires access to the array of past values.

[steve]

Have you seen the (one-off) figures I posted elsewhere comparing the two different calculations with my Davis anemometer? The result using the 'gust' value (while admittedly slightly high) came out very close to the Davis; the result using the average value was significantly lower.

But anyway, this is all somewhat irrelevant; I've already indicated that offering the option of making the array of recent readings contain the 48-second average instead of the 'gust' would be a reasonable enhancement request. Bear in mind that the array of values is somewhat 'flawed' as it contains every 10-second reading made by Cumulus, and as you know the station only updates every 48 seconds.

[brassing]

That sounds very interesting. Where can I find it?

[nking]

Here you go https://cumulus.hosiene.co.uk/viewtopic.p ... 5&start=10