Dominant Wind Direction

[Nottub]

I have numerous entries on my 'Dominant Wind Direction' which are blank and would like to understand where these values are derived from?

My historic records from Weatherlink have been converted to files that CumulusMx read, it is here that, through the years, odd values are missing. Since I've had CumulusMX up and running any new values are fine.

I can add manually from my records extracted from Weatherlink but these differ from those created in Cumulus.

Is there a specific file in Cumulus that shows the values used to derive the 'Dominant Wind Direction'? If so could you let me know which file I can look at.

I presume that Weatherlink looks at purely the direction, whilst Cumulus looks at the speed as well, or am i missing the obvious?

Thanks

Martyn

[mcrossley]

You are correct that CMX uses a wind vector averaging - i.e. each direction is weighted by the current average speed.

The dominant for the day is calculated by averaging all the speed/directions sampled once a minute over the day.

[Nottub]

Thanks Mark,
I presume then that I cannot add anything to the dayfile (or anywhere else) to recover/ regenerate the missing values?

Or should I resign myself to sticking in the directions from Weatherlink which are pure directions with no consideration of the speed from that direction.

Thanks
Martyn

[mcrossley]

Do you have the monthly log files?

[Nottub]

Mark,

I have all the wlk files for Weatherlink back to yr2000, and the CumulusMX log. txt files in the data folder within CMX.

A good example on my website is Dec 2020 where the 5th, 11th and 16th to 29th wind directions are all missing. Other values are there.

https://www.calvertoncam.co.uk/weather/yadr.php

If there is something I can check/ update I'd be grateful for any pointers.

Thanks

Martyn

[mcrossley]

If you have the log files, then one option is to use the CreateMissing utility to recreate the dayfile. It will do the calculations for you.

I'd suggest making a copy of your install, then...
1. Delete the dayfile from your copy install
2. Run CreateMissing to recreate a new dayfile
3. Copy the direction values from this new dayfile back into your "real" dayfile

[Nottub]

Think I've identified the issue

My orginal Davis Weather Monitor 2 recorded true North as 0 degrees which is read as a null value by the Betel_Dayfile Reader.

Weatherlink happily recorded 0 degrees as North, 359 degrees as North, and 1 degree as North ( the nearest wind direction).

However the Betel_Dayfile Reader sees this 0 degrees as a null value, but not the 359 and 1 degree as these are valid, and correctly show as North.

So the errors are limied to those very odd occasions where the wind was recorded as a true North i.e. 0 degrees.

When I look at the original Monitor 2 Consol manual (P11) which identifies North as 0 degrees, whereas the Vantage Pro 2 Consol manual (p21) shows North as 360 degrees.

Just need to work out if there's an easy way to amend the Betel_Dayfile Reader.

Think Ill post in the Dayfile Reader posts.

Martyn

[freddie]

In the meteorological world, 0 is not a "real" wind direction - it is a reserved value used as the direction when the speed is 0. The bearing that should be used for north is 360 degrees. Anything else is incorrect.

[Nottub]

Agreed, but that's how the Davis Weather Monitor 2 recorded its North value.