Wind gust

[dane]

Something has changed recently - I can't say how recently.
My wind gust graph used to show gusts as spikes - now they are more like plateaus.
How come? has the code changed?
On the main Cumulus display, the "Latest" seems to change in step with changes in the unit, but the "Gust" field does not change as frequently - actually don't know what makes it change?
What's the difference between the 2 fields? and which one is graphed ("gust" I guess...)?

2008-12-11_212324.png

[steve]

The 'gust' field is the peak value in the last 10-minutes, I think I changed to plotting this some time ago instead of the individual 'latest' readings, because those can be lower than the mean wind speed (I'm talking generally here, I'm not sure this applies to Fine Offset stations), so the graphs of 'gust' vs 'speed' look a bit odd and people ask how can the gust be lower than the average speed.

Of course, this does come back to the question we were asking some time ago regarding what exactly is it that the Fine Offset stations report as 'gust'.

Steve

[dane]

As far as I can tell from observations of the station, "gust" is the highest wind speed during the latest 48 secs, while "average" is the mean speed over the latest 48 secs. (As reported in Cumulus in "Latest" and "Average" respectively). And "gust" is never less than "average" from a Fine Offset station.

So it bothers me to see a graph showing a "gust" lasting for many minutes, well knowing that it was a spike in a 48-secs period.

Any chance you could go back to charting "Latest" as "gust" for Fine Offset stations?

[Frank02]

The "Latest" (which is the latest gust on Fine Offset stations) can be lower than the Cumulus average wind speed when using the Cumulus "Calculate 10-min wind average" option which is normal since Latest is an "instant" (last 48sec) value as oppose to a value calculated over a 10-min period. As Dane says, the Fine Offset average (not the Cumulus calculated one) will never be higher than the fine offset gust value since both represents a figure over the last 48 seconds.

I think that, maybe, if someone is using the Cumulus "Calculate 10-min wind average" option you should use the "Gust" value but if not you should use the "Latest" value on the chart. This way, one will compare figures for the same period of time (instant 48sec gust and average versus 10-min peak gust value and calculated average).

I hope this is clear.

[steve]

As far as I can tell from observations of the station, "gust" is the highest wind speed during the latest 48 secs, while "average" is the mean speed over the latest 48 secs. (As reported in Cumulus in "Latest" and "Average" respectively).

Well, yes, but what exactly does "highest wind speed during the last 48 seconds" actually mean. Think about it. These devices don't measure speed directly, they count rotations over a given period of time. So how does it calculate "highest wind speed during the last 48 seconds". Anyway, I guess it doesn't matter how it does it, as:

And "gust" is never less than "average" from a Fine Offset station.

Any chance you could go back to charting "Latest" as "gust" for Fine Offset stations?

See next message

Steve

[steve]

I think that, maybe, if someone is using the Cumulus "Calculate 10-min wind average" option you should use the "Gust" value but if not you should use the "Latest" value on the chart. This way, one will compare figures for the same period of time (instant 48sec gust and average versus 10-min peak gust value and calculated average).

Yes, this is a good idea, I'll implement that.

It will mean a change to the monthly data log file, as I don't currently log the 'gust' as read from the station, and the graphs are drawn at startup from those files. Well, possibly - I need to check something.

Steve

[steve]

I've realised that I will have to make this change only for Fine Offset stations, as the Davis (for example) 'latest/gust' figure is not a 'peak' figure as we assume the Fine Offset one is, so could still be below the mean speed even if Cumulus isn't calculating it.

Something else occurred to me - wasn't there an issue where people were seeing gusts lower than speed in banner.php, and this was because the banner was using the 'latest' gust figure? Or were those people also using the calculated 10-minute average?

Steve

[dane]

Something else occurred to me - wasn't there an issue where people were seeing gusts lower than speed in banner.php, and this was because the banner was using the 'latest' gust figure? Or were those people also using the calculated 10-minute average?

I remember seeing that in your own banner (when you displayed that...).
So maybe it's a Davis issue - or a non-optimal mix of options as mentioned by Frank02 above.

[steve]

I remember seeing that in your own banner (when you displayed that...).
So maybe it's a Davis issue - or a non-optimal mix of options as mentioned by Frank02 above.

It would certainly have affected Davis users, for the reason I described above. But I thought Fine Offset users had seen the same thing - yourself included, in fact. I must be mistaken. Anyway, no matter, this is definitely a sensible change to do for Fine Offset users, so I'll go ahead and do it.

[harrym1byt]

I would hazard a guess that the gust speed is based on the shortest interval between two pulses from the anemometer, in the previous 48 second period. So only two pulses are needed or used. I have never seen the gust show as being a lower speed than the average.