Welcome to the Cumulus Support forum.

Latest Cumulus MX V3 release 3.28.6 (build 3283) - 21 March 2024

Cumulus MX V4 beta test release 4.0.0 (build 4018) - 28 March 2024

Legacy Cumulus 1 release v1.9.4 (build 1099) - 28 November 2014 (a patch is available for 1.9.4 build 1099 that extends the date range of drop-down menus to 2030)

Download the Software (Cumulus MX / Cumulus 1 and other related items) from the Wiki

How To: the relation between the Windgraphs

Discussion of the Cumulusutils tool and website generator.

Moderator: HansR

Post Reply
User avatar
HansR
Posts: 5871
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

How To: the relation between the Windgraphs

Post by HansR »

While fixing an issue with the imperial units and unit conversions I ran into problems with the YADR report for the Windrun i.e. the legend at the bottom. All other legends for YADR reports are based on the metric default units used by CUtils. Then during startup - if the user uses imperial units - those legend values are converted to their imperial values and that way the same intervals are used.

However I did not do so for the Windrun (don't ask why, I forgot).

The Wind has basically three important charts/reports: The WindRose graph, The WindRun graph and the YADR WindRun report (with the legend).
All three have a legend for which the intervals are configurable with parameters:

Code: Select all

WindRoseNrOfWindforceClasses=6 
WindRoseMaxWindSpeed=60 
WindrunClassWidth=75
The charts then are calculated and configured with these two parameters which is too complex to just simply grasp by the user. I will add the calculation to the Windchart wiki for those who wish to understand what is happening (and post here when done). The other way can be trial and error to get a correct class width and a correct nr of classes to use (Windrun and Windrose) and have charts which suit the users need.

Especially the windrun requires configuring when the user uses imperial units (mi, kts) and the user is free to choose the class width which will differ depending on the region of the station. The default value uses a 75 km windrun class width which fits Dutch environment.

A related item is the nr of windrose compass sectors. Those are taken from Cumulus.ini, section [Display], parameter NumWindRosePoints.
However to get the compass letters right read the wiki.
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
User avatar
HansR
Posts: 5871
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: How To: the relation between the Windgraphs

Post by HansR »

OK, I updated the wiki page for the Wind Charts. I hope it is clear, if not I would appreciate some comment.

Having written the above, I think that my version 6.8.1 with corrections for the use of imperial units is finished. I'll give it a few more days for checks and then I will release it. Thanks for testing the beta @Andy and @Mike (I see you found your way through the windrose compass sectors ;) ).
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
AndyKF650
Posts: 704
Joined: Thu 20 Aug 2020 8:35 am
Weather Station: Davis VP2 Plus
Operating System: RPiv4 64bit Linux 12 Bookworm
Location: Jersey Channel Islands
Contact:

Re: How To: the relation between the Windgraphs

Post by AndyKF650 »

Hi Hans

This reminds me of an issue I have with the windrun chart (the "Chart") in the graphs section and trying to reconcile this to the numbers recorded for example in the YADR windrun report (the "Report").

The Chart for 2021, which correctly shows the distance is measured in miles in the title, suggests that for 7 months of the year I had a windrun of 200 miles or greater but the Report shows the maximum windrun of 196.4 miles in May and 7 months with a windrun above 155.4 miles.

I suspect that there is a conversion issue with the Chart data which is trying to measure the distance in kM, the easy solution would be to make the heading kM but is easy the most appropriate solution. ;)
AndyKF650
Posts: 704
Joined: Thu 20 Aug 2020 8:35 am
Weather Station: Davis VP2 Plus
Operating System: RPiv4 64bit Linux 12 Bookworm
Location: Jersey Channel Islands
Contact:

Re: How To: the relation between the Windgraphs

Post by AndyKF650 »

An extra for this topic, I have just fine tuned the windrose parameters for my station. The max speed, which is average speed, is now set to 25 mph and the wind force classes set to 10.

This really improves the granularity of the windrose and with the multi colours is much more interesting to look at. Previously I had set the wind speed to 50 mph with only 4 force classes which produced a boring chart.

I must have too much time on my hands!
User avatar
HansR
Posts: 5871
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: How To: the relation between the Windgraphs

Post by HansR »

AndyKF650 wrote: Sat 02 Apr 2022 5:08 pm An extra for this topic, I have just fine tuned the windrose parameters for my station. The max speed, which is average speed, is now set to 25 mph and the wind force classes set to 10.

This really improves the granularity of the windrose and with the multi colours is much more interesting to look at. Previously I had set the wind speed to 50 mph with only 4 force classes which produced a boring chart.
Well, I was just going to answer your previous question but I see that you got the hang of it. I think that 25 mph is OK, 10 classes may be a bit high (you might try 5 or 6) because the high winds (> 25) are rare on Jersey apparently.

For the windrun: your highest windrun until now is 178 mi (not so very high) so I would choose a classwidth of 30 (if it ever gets really high winds you might always increase that). 30 would be a bit less than what the default in km is. I assume your chart the n becomes reasonable.

Remember one thing (from you previous question) the windrun chart does not display miles but it displays the procent distribution of the winds in a class. And also, in both charts it is not the average speed which is charted but all entries in the standard logfiles are taken as samples and averaged over the number of samples to give an estimator of how often a certain wind (speed/direction) occurs.

The charts are improving :clap:
AndyKF650 wrote: Sat 02 Apr 2022 5:08 pm I must have too much time on my hands!
;) 8-)


NOTE: strange that there are rarely really high winds on the island. I would expect more of that. Is the anemometer sheltered or spider infested?
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
User avatar
HansR
Posts: 5871
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: How To: the relation between the Windgraphs

Post by HansR »

And btw, the tooltip you see in the Windrun chart (see screenshot) denote the month, the class width end point (so in this case it is the class of 0 - 75), the percentage of observations in the class and the Totals number of days on which the stacked bar is based. I may have to improve in clarity here.
Schermafbeelding 2022-04-02 195441.jpg
You do not have the required permissions to view the files attached to this post.
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
AndyKF650
Posts: 704
Joined: Thu 20 Aug 2020 8:35 am
Weather Station: Davis VP2 Plus
Operating System: RPiv4 64bit Linux 12 Bookworm
Location: Jersey Channel Islands
Contact:

Re: How To: the relation between the Windgraphs

Post by AndyKF650 »

Hi Hans

Thanks for the info on these graphs. Now I understand how they work and show the various changes I have amended the .ini file parameters to reflect a more appropriate display.

For the wind rose the max speed is 25 mph and the WindForce classes is 10;
For the Windrun the ClassWidth is 25 mph;

As you mention the wind speed and also windrun I measure is relatively low, I live on the south eastern side of a hill and tucked well in so the bulk of the westerlies blow over the top of me. Jersey Airport regularly gets Force 8 /9 winds recorded as it is on the top of the island in a very exposed place, ideal for an airport but not that you would want to live there. I suppose my wind recording is not representative of an open exposed situation but it does show just what our house is experiencing.

Thinking about the tool tip should the speed be shown as <75 km: 48 days. This then makes me wonder if you have a very different way of counting days in Holland, we use 31 days in August ;) That said on my charts each month of stacked blocks totals the number of days in the month rather than percentages.

I do think that it is good to revisit these charts from time to time especially in my case they were set up when I first started in CUtils and did not have a particularly long data run in CMX/Cutils.

Keep up the great work. :D
User avatar
HansR
Posts: 5871
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: How To: the relation between the Windgraphs

Post by HansR »

AndyKF650 wrote: Sun 03 Apr 2022 7:07 am Thanks for the info on these graphs. Now I understand how they work and show the various changes I have amended the .ini file parameters to reflect a more appropriate display.

For the wind rose the max speed is 25 mph and the WindForce classes is 10;
For the Windrun the ClassWidth is 25 mph;
:D And I think your figures are OK.
AndyKF650 wrote: Sun 03 Apr 2022 7:07 am As you mention the wind speed and also windrun I measure is relatively low, I live on the south eastern side of a hill and tucked well in so the bulk of the westerlies blow over the top of me. Jersey Airport regularly gets Force 8 /9 winds recorded as it is on the top of the island in a very exposed place, ideal for an airport but not that you would want to live there. I suppose my wind recording is not representative of an open exposed situation but it does show just what our house is experiencing.
Thought so ;)
AndyKF650 wrote: Sun 03 Apr 2022 7:07 am Thinking about the tool tip should the speed be shown as <75 km: 48 days. This then makes me wonder if you have a very different way of counting days in Holland, we use 31 days in August ;) That said on my charts each month of stacked blocks totals the number of days in the month rather than percentages.
Indeed, my bad. What you see is a percentage on the axis, the nr of days that windrun class was recorded, and the total number of days the statistics is based on (so for fulll months measured this will be a multiple of the number of days in that month).
And yes, I will update the tooltip to clarify this better.
AndyKF650 wrote: Sun 03 Apr 2022 7:07 am I do think that it is good to revisit these charts from time to time especially in my case they were set up when I first started in CUtils and did not have a particularly long data run in CMX/Cutils.
Yes, I assume after three years the averages and totals will more or less be representative so 1) setup, 2) first revision after 1 year 3) third revision after 3 or 4 years. It would surprise me if it would be needed more often.
AndyKF650 wrote: Sun 03 Apr 2022 7:07 am Keep up the great work. :D
Classic techique of stepwise improvement ;)
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
Post Reply