Welcome to the Cumulus Support forum.

Latest Cumulus MX release v3.1.2 build 3055 - 4 November 2019.
Legacy Cumulus 1 release v1.9.4 (build 1099) - 28 November 2014

See the Wiki for downloading either version.

Fire Weather, a new approach

Talk about anything that doesn't fit elsewhere - PLEASE don't put Cumulus queries in here!
freddie
Posts: 508
Joined: Wed 08 Jun 2011 11:19 am
Weather Station: Davis Vantage Pro 2
Operating System: GNU/Linux (Ubuntu 18.04)
Location: Dorrington, Shropshire
Contact:

Re: Fire Weather, a new approach

Post by freddie » Sat 21 Sep 2019 10:56 am

Interesting stuff. I would have thought it would be easier to use the station pressure of the PWS (not the sea level pressure). That should be sufficient to calculate saturated vapour pressure without involving altitude.
Freddie
Image

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Sat 21 Sep 2019 3:41 pm

Ehmm.. OK, how would you calculate the saturated water vapour pressure from the station pressure (that is: the atmospheric pressure at station level)?
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

freddie
Posts: 508
Joined: Wed 08 Jun 2011 11:19 am
Weather Station: Davis Vantage Pro 2
Operating System: GNU/Linux (Ubuntu 18.04)
Location: Dorrington, Shropshire
Contact:

Re: Fire Weather, a new approach

Post by freddie » Sat 21 Sep 2019 4:17 pm

As you said in your blog, you already have the station relative humidity. You also have the station temperature.
So, with station-level pressure, do you not have everything you need - and have no need to know altitude?
Freddie
Image

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Sat 21 Sep 2019 5:12 pm

I am not sure about what you are saying, but even if I measure RH and temperature, I still do not know my saturation pressure of water at that temperature so I have to calculate that. Atmospheric pressure is not in that equation. The relation - approximation of the P(sat) of water - is not described by atmospheric pressure, only by temperature. And that approximation is at sealevel so I need to correct for the altitude. These are not simple relationships. I may be wrong, but than I need some good convincing ;)
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

freddie
Posts: 508
Joined: Wed 08 Jun 2011 11:19 am
Weather Station: Davis Vantage Pro 2
Operating System: GNU/Linux (Ubuntu 18.04)
Location: Dorrington, Shropshire
Contact:

Re: Fire Weather, a new approach

Post by freddie » Sat 21 Sep 2019 6:18 pm

Yes, I understand that the relationship is complex. Okay, I'll rephrase what I am asking: if there is no dependence on pressure, then why does altitude matter?
Freddie
Image

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Mon 23 Sep 2019 8:12 am

freddie wrote:
Sat 21 Sep 2019 6:18 pm
[...] I'll rephrase what I am asking: if there is no dependence on pressure, then why does altitude matter?
Sorry I thought I had answered, but apparently forgot to press submit :(

[edit:] Wrong note, so I better remove it.

[edit:] And to continue on this: the solution probably lies in the Psychrometric constant.

to be cont.
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Thu 26 Sep 2019 10:14 am

I guess you are right @freddie, I made an error in thinking.
Nevertheless, I had to change and add some other things. So here is the blog.
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

freddie
Posts: 508
Joined: Wed 08 Jun 2011 11:19 am
Weather Station: Davis Vantage Pro 2
Operating System: GNU/Linux (Ubuntu 18.04)
Location: Dorrington, Shropshire
Contact:

Re: Fire Weather, a new approach

Post by freddie » Thu 26 Sep 2019 4:26 pm

HansR wrote:
Thu 26 Sep 2019 10:14 am
I guess you are right @freddie, I made an error in thinking.
Vapour pressure is a difficult concept to get your head around.
You have opened up a can of worms with your blog. You have made the assumption that water is solid when its temperature is below zero. This is not actually true, as water can exist in liquid form at temperatures as low as -40 Celsius. This fact is very apparent in the atmosphere - especially when it comes to precipitation growth within clouds. If you Google "Bergeron Findeisen" you can read about this yourself. If I was you, I would use a linear "sliding scale" with your August-Roche-Magnus coefficients, such that at 0 they are the values for SVP over water, and at -15 they are the values for SVP over ice. That should capture the most likely values for SVP during those transition temperatures.
Freddie
Image

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Thu 26 Sep 2019 6:08 pm

freddie wrote:
Thu 26 Sep 2019 4:26 pm
You have opened up a can of worms with your blog. You have made the assumption that water is solid when its temperature is below zero. [...]
No I have not, that is too easy a conclusion. I am well aware of the existence of undercooled water, but it is not purely the theoretical subject I try to solve. I am working on the subject of meteorology and Wild vegetations (forests, grasslands, bogs etc...) and their fires. And undercooled water does not exist too much in and around forests. It may exist on high altitudes, large surfaces, but not around in vegetation situations. On the contrary I would say. If you walk a forest on a winters day, you will see white ice crystal covered leaves on the ground and grass. Almost the physical view of the frozen wet bulb. And yes, vegetation has its own effect on humidity. But that's in a way the idea behind what I am building.

So in general, as long as the wet bulb thermometer will not freeze I am good with the assumption of the above zero ARM coefficients. Below freezing I use the other set. I do not see why I should use a sliding value of coefficient on temp. Never seen that in literature anywhere and I think a linear sliding scale does not match the exponential character of the equations.

And above that, I would like to stick to what is accepted science in meteorology. More or less. There is a load of equations (I found around 20 I think), all dealing with the estimation of the Saturated Vapour Pressure of water (I include one paper below). That is very nice and good for the scientists who earn their money. But look at the variations. Within the domain of where I write my software for, for a certain goal with a the temperature range of let's say -40 up to +50 degrees, all equations suffice. So I chose to use a more or less standardized equation. the ARM equation, which is apparently adopted by the WMO. That makes me happy, decision taken. I will not slide away in the confusion of choice. I make my choice and I publish what I do. I don't think my blog opens a can of worms, meteorology is just physical chemistry on a large scale.

Attachment:
190926 Saturation vapor pressure formulations.pdf
You do not have the required permissions to view the files attached to this post.
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Sun 06 Oct 2019 9:40 am

A discussion on the pwsFWI will follow up here (if there is follow up ;)) so that other thread can be on the software.
Several discussions on fire weather have already been made in the other thread. The last one initiated this move.
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Thu 17 Oct 2019 9:31 am

Now pwsFWI has reached a maturity stage it is interesting to look at how it behaves. And that gives some surprising results half October. Especially in San Sebastian where despite a lot of rainfall, strong winds and high temperatures drive the warning level to Moderate, which must be regarded as pretty high (without reaching real danger levels), given the time of year. I keep getting surprised by the result of my own invention!

And actually, I think the relation with reality is correct given the map of a more global Spanish fireweather system. pwsFWI is known (has a built in) delay on weather changes because wood does not dry that fast. If the weather situation in Basque country continues like this I presume the warning level will go to yellow. And that I guess is interesting and not very common over there with 1300 mm rain per year!

Knipsel.JPG
Knipsel2.JPG
You do not have the required permissions to view the files attached to this post.
Last edited by HansR on Fri 18 Oct 2019 8:16 am, edited 1 time in total.
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

User avatar
kocher
Posts: 85
Joined: Sat 19 Apr 2014 7:57 pm
Weather Station: Davis Vantage Pro2+
Operating System: Windows 10
Location: San Sebastian - Spain
Contact:

Re: Fire Weather, a new approach

Post by kocher » Fri 18 Oct 2019 7:55 am

Good morning Hans and good morning everyone

I am very happy with your creation pwsFWI.
Especially effective is the calculation of the FWI with smooth changes after episodes of rain.

http://kocher.es/fwi-hans.php

http://kocher.es/NOAAsql/NOAA.php

Indeed there has been a curious situation during the last 5 days in San Sebastián (Basque Country, Spain), due to an episode of high temperatures and strong South wind (wind of African origin), but this is quite usual in San Sebastián, even in the month of December.
It usually corresponds to isolated episodes, without being able to consider it a trend.

The fires in our area are really rare although, cyclically, we can see some year with repeated fires, especially in the summer months coinciding again with times of heat and strong South wind.

I am very grateful for your interest in this very important issue and for your intelligent way of dealing with it.

Saludos desde San Sebastian

Javier
Image

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Fri 18 Oct 2019 8:14 am

Thanks for your reply and good info.

And yes, I check all sites daily for anomalies and if I don't understand I get curious :)
Curiosity is the basis for understanding this complex issue. I have seen many fires in my years in the South Of France (only 400 km of San Sebastian) and I became aware of how local meteorological conditions influence the susceptibility for fire more than anything else (and of course the availability of wood/fuel). That's where this program originated and that's why I sometimes ask clarification on the relation between what the program tells me and the actual situation around the station.
kocher wrote:
Fri 18 Oct 2019 7:55 am
I am very grateful for your interest in this very important issue and for your intelligent way of dealing with it.

Saludos desde San Sebastian

Javier
Thnx for the compliment!

Groeten vanuit Wagenborgen :)
Hans
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Wed 23 Oct 2019 5:16 pm

Recently the pwsFWI developed interestingly in San Sebastian, today it is beginning to be very interesting in Inverell Australia where around Phil's Backyard the pwsFWI went through all colours of warning levels from Extreme to Low (because of rain) to extreme the last fortnight. That is a fast development which I would not expect. And the regular fire warning system displayed via the local Australian Weatherzone remains blue (high) at the moment. That seems a big difference to me but somehow I think that the two systems will meet again. My pwsFWI is more local and follows direct meteo. The other system is McArthur-like (maybe it IS McArthur). Anyway, While meteosangonera just finished the fire season, Phil already had his first taste and now is preparing for summer I guess. This is the official fire danger map for thursday 24/10 (the left one for the bush)

Phils backyard 1.JPG
Phils backyard 2.JPG
You do not have the required permissions to view the files attached to this post.
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

User avatar
HansR
Posts: 366
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 9 (stretch)
Location: Wagenborgen (NL)
Contact:

Re: Fire Weather, a new approach

Post by HansR » Sat 26 Oct 2019 7:34 am

And indeed, both warning systems now are on level orange (pwsFWI : Very High; Weather Zone: Severe). Find the legend of Weatherzone here btw - bottom of page. Don't know when it reached that level though.

[Edit 28/10:] As of today pwsFWI again went to warning level extreme while the WeatherZone level remains at Yellow (Very High). AN interesting 'lagging behind' while the behaviour up to this point was that pwsFWI was more pre-cautious in increasing the level. There seems, above all, to be a difference in the speed of reaction to changing meteorology, WeatherZone being the quickest to react and increase the warning level. At the moment it is pwsFWI which is changing quickest and increasing. Interesting! Under stable circumstances, the warning level seems the same, they converge. I do not draw conclusions yet.
https://meteo-wagenborgen.nl / Cumulus MX 3.1.0-b3053
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian GNU/Linux 10 (buster) / Linux 4.19.75-v7+ armv7l
Mono JIT compiler version 6.4.0.198 (tarball Tue Sep 24 01:45:47 UTC 2019)

Post Reply