Page 1 of 9

Fire Weather, a new approach

Posted: Mon 12 Aug 2019 9:04 am
by HansR
Recently I restarted my weather station and as in the past, I wanted to include fire weather warning info. I had lost the software so I asked on the forum and got a reply about FWIcalc. That software seemed to have reached the end of its life-cycle and the author had stopped. The orphanaged software is still available and can be used but:
  1. it requires a PC to run, separate from the weather system (unless you run cumulus on a PC permanently on);
  2. it calculates as a black box, meaning no info on the theory behind it, is readily available.
Both points are an issue I think. Apparently fire weather warning is not of much interest to many owners of PWSs, but it is to me. The subject itself is interesting (having studied forestry) and the strong relation to local meteorology makes it a great combination to theoretically investigate. Besides, I would like to see some software to generate automatically the FWI and display it on the site. This would require a calculation on the Cumulus side or on (web)server side. Wherever it is done, it must be integrated in the system somehow.

So I began by studying and created a blog to track my findings (and also for other PWS related stuff). You can find it here and you can select the relevant FWI blogs by using the tag FWI.

I will indicate and update here with either relevant theory, software or otherwise.
Comments are welcome.

Update: I developed relevant software in the CumulusUtils (get software and see here).
You can read the blogs here.

Re: Fire Weather, a new approach

Posted: Mon 19 Aug 2019 11:47 am
by HansR
As promised I made a follow up on the fire weather index with a new blog : An effort for a simpler fire weather index. It describes a variation on a fire weather index published recently (see the blog) and discusses the theory. I will implement this theory in my add2cumulus program (additions to cumulus) through which it will become available as a separate procedure.
I will update the corresponding message accordingly with explanation when done.

In addition, for testing purposes, I would like to have some larger dayfile.txt (with station latitude/longitude and altitude), especially from the dryer regions, possibly semi-arid. But moderate, boreal and tropical data are very welcome as well of course. You can post the file in response to this message or by PM.

Best,

Re: Fire Weather, a new approach

Posted: Wed 21 Aug 2019 3:26 pm
by HansR
The first implementation is up and running (some hickups in the page now and then).
I do not need other dayfile.txt anymore. A discussion with beteljuice made things more clear to me. Thanks so far. Any progress, I'll let you know.
Best,

Re: Fire Weather, a new approach

Posted: Fri 23 Aug 2019 5:55 am
by meteosangonera
HansR wrote: Wed 21 Aug 2019 3:26 pm The first implementation is up and running (some hickups in the page now and than).

I am still looking for other dayfile.txt but with the remarks below in mind!

So far I received one dayfile.txt for testing purposes. This taught me a lot about the possible variations in dayfile.txt (and other logfiles). I was probably a bit optimistic in generic handling of the dayfile. I will postpone this until I find an easy way to deal with:
  1. different separation characters for dates and fields
  2. different units and the conversions between them
So, for the time being, anybody interested in this must must have dash and semicolon separators (-;) and must be metric (C, mm, km and hPa).

Best,
Hi, I can test, or I can borrow my dayfile.txt

Re: Fire Weather, a new approach

Posted: Fri 23 Aug 2019 1:53 pm
by HansR
meteosangonera wrote: Fri 23 Aug 2019 5:55 am Hi, I can test, or I can borrow my dayfile.txt
Hi, Thnx for the reply. See my edited message just before yours.

Best,

Re: Fire Weather, a new approach

Posted: Fri 30 Aug 2019 5:01 pm
by HansR
So, I took the next step for my new fire weather index. Made an implementation, and tested it against several 'dayfile'-instances from all over the world. That was programmatic an interesting exercise, I may come back on that sometime. For now I just wrote a blog on the pwsFWI evaluation.

Best,

Re: Fire Weather, a new approach

Posted: Fri 30 Aug 2019 8:53 pm
by beteljuice
Informative, well written blog Hans ..

Informatieve, goed geschreven blog Hans .. :clap:

Re: Fire Weather, a new approach

Posted: Sat 31 Aug 2019 6:16 am
by HansR
Thnx @beteljuice.

I just received an interesting reply from Phil (inverell location Australia) with two fire reports which I can relate to my calculation.

Code: Select all

The most recent major fire was Feb 16 2019.
https://www.inverelltimes.com.au/story/5903445/tingha-fire-2019-in-dozens-of-photos/ 

And another nearby Jan 30th.
https://www.inverelltimes.com.au/story/5878991/firefighters-keep-tingha-blaze-under-control/ 
My data are:
pwsFWI inverell 0219.JPG
(get the file to see the larger dataset)

What we see is an extremely strong relation between pwsFWI and actual fires in the area, meaning the warning code is very close (warning does not mean a 100% a fire will ignite). I am on the right road.

Thnx @Phill!

Re: Fire Weather, a new approach

Posted: Thu 12 Sep 2019 9:48 pm
by Phil23
Hi Hans,

Just ran your latest exe under windows & it produced the txt file as predicted.
I just renamed it to htm to see the data.
Capture.JPG
Interesting that we had huge outbreaks of fires on 06/09/19.

From memory, the warnings started going out on 03/09/19 from memory.

If I can get some guidance on embedding it in a page with my headers & footers, I'll put it up live.
Presume I'll need the SS from the Linux version;
I already run a few daily scripts, so can handle that part.

Just not good on the html understanding of how to embed in in a page.
I'd just like to give it it's own page with my header, footer & Nav Menus.

Cheers.

Edit,

Should add, the fires last Friday; I believe there were 3 or 4 within a 50km radius of my location.
The were not major out of control blazes.

The bigger fires fell inside a 250km radius, both roughly 200km away.

Re: Fire Weather, a new approach

Posted: Thu 12 Sep 2019 11:23 pm
by Phil23
Ok,

Sort of got a page up.

Very crude at this point but would appreciate suggestions on what needs to change.

Thanks

Phil.

Edit, I just used my http://weather.inverellit.com/datasummary.php & changed the include line at this point.

http://weather.inverellit.com/fwi.php

Code: Select all

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">

<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">

<head>

<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />

<meta name="description" content="Phil's Backyard weather data" />

<meta name="keywords" content="Cumulus, Phil's Backyard weather data, weather, data, weather station" />

<title>Phil's Backyard</title>

<link rel="shortcut icon" href="favicon.ico" type="image/x-icon" />

<link href="weatherstyle.css" rel="stylesheet" type="text/css" />

<link rel="stylesheet" type="text/css" media="screen" href="750betel_datasummary.css"  /> 

<script src="https://ajax.googleapis.com/ajax/libs/jquery/1.8.2/jquery.js" type="text/javascript"></script>

<META HTTP-EQUIV="Pragma" CONTENT="no-cache">

<META HTTP-EQUIV="Expires" CONTENT="-1">

</head>

<body>

<div id="content">

<h1>Phil's Backyard</h1>

<h2 style="text-align: left; text-transform: none;" >

&nbsp;Latitude&nbsp;S&nbsp;29&deg;&nbsp;46&#39;&nbsp;26&quot;&nbsp;&nbsp;&nbsp;&nbsp;Longitude&nbsp;E&nbsp;151&deg;&nbsp;07&#39;&nbsp;14&quot;&nbsp;&nbsp;&nbsp;&nbsp;Elevation&nbsp;586&nbsp;m</h2>

<table style="width:100%;border-collapse: collapse; border-spacing: 0;" >

    <td colspan="4" class="td_navigation_bar">:<a href="index.htm">now</a>: :<a href="today.htm">today</a>: :<a href="yesterday.htm">yesterday</a>: :<a href="datasummary.php">daily&nbsp;data</a>: :<a href="thismonth.htm">this&nbsp;month</a>: :<a href="thisyear.htm">this&nbsp;year</a>: :<a href="record.htm">records</a>: :<a href="monthlyrecord.htm">monthly&nbsp;records</a>:<br>:<a href="gauges.htm">gauges</a>: :<a href="graphs.htm">graphs</a>: :<a href="wz.htm">weatherzone</a>: :<a href="https://cumulus.hosiene.co.uk/" target="_blank">forum</a>: :<a href="https://weather.inverellit.com/pws/" target="_blank">dashboard</a>: :<a href="http://weather.inverellit.com/webcams.htm">webcam</a>:</td>

</table>

<div id="tableData">

<?php include('pwsFWI.txt');?>

</div>

<table style="width:100%;border-collapse: collapse; border-spacing: 0;" >

    <td colspan="4" class="td_navigation_bar">:<a href="index.htm">now</a>: :<a href="today.htm">today</a>: :<a href="yesterday.htm">yesterday</a>: :<a href="datasummary.php">daily&nbsp;data</a>: :<a href="thismonth.htm">this&nbsp;month</a>: :<a href="thisyear.htm">this&nbsp;year</a>: :<a href="record.htm">records</a>: :<a href="monthlyrecord.htm">monthly&nbsp;records</a>:<br>:<a href="gauges.htm">gauges</a>: :<a href="graphs.htm">graphs</a>: :<a href="wz.htm">weatherzone</a>: :<a href="https://cumulus.hosiene.co.uk/" target="_blank">forum</a>: :<a href="https://weather.inverellit.com/pws/" target="_blank">dashboard</a>: :<a href="http://weather.inverellit.com/webcams.htm">webcam</a>:<br>:<a href="http://weather.inverellit.com/historic/datasummary.php">historic&nbsp;data</a>:</td>

</table>

<p class="credits"><br />

Page updated 13/09/2019 9:10:00 AM<br />

powered by <a href="https://cumuluswiki.wxforum.net/a/Cumulus_MX" target="_blank">CumulusMX</a> v3.0.1 (3051)

<a href="http://validator.w3.org/check?uri=referer"> Valid HTML</a> | <a href="http://jigsaw.w3.org/css-validator/check/referer"> Valid CSS</a><br/>

dayfile reader by <a href="https://cumulus.hosiene.co.uk/viewtopic.php?f=14&t=11397" target="_blank">beteljuice</a> from the Cumulus Forum<br/><br/>

<a href="http://cumulussites.net/" target="_blank"><img src="http://cumulussites.net/button.php?u=Phil23" alt="Cumulus Top Weather Sites" border="0" width="162" height="55"></a></br>





</p>

</div>

</body>

</html>



Re: Fire Weather, a new approach

Posted: Fri 13 Sep 2019 12:18 am
by beteljuice
Looks fine to me.

.... although (navigation) you still insist on <table><td> instead of <table><tr><td>

You need a 'title' somewhere ...

something like:
</table>
<h1>Fire Weather Index (experimental)</h1>
<div id="tableData">

Re: Fire Weather, a new approach

Posted: Fri 13 Sep 2019 5:58 am
by HansR
Hi Phil,

Thanks for installing and be my first tester (others are welcome!).

Your technique of include is correct. And yes, I noticed some minor errors in the generated HTML. I will fix that with the new version and will implement a header as well, including version nr and some reference. New versions of the software will also include my top10 and status-page (for rPi at least). I will publish this in a new thread on the forum. Another thread which will focus on the software so this thread will be on the pwsFWI specifically: how does the pwsFWI relate to reality.

I am not happy with the fires your region saw lately, about which you informed me, but it relates positively with my pwsFWI. And that is very good!! 8-) (I don't :clap: for myself, but maybe I should, the good result is kind of better than expected!).

My idea is that the page title and <h1>-header are for the user to set. I just supply the pwsFWI table. I chose this technique because the variety of sites in the wild is so big that, if I add more , wishes will increase and I want to focus on meteo related info and not on site related info.

You say:
Presume I'll need the SS from the Linux version;
What do you mean by that?

You say:
I already run a few daily scripts, so can handle that part.
Personally I run it from CumulusMX from the external program section, daily frequency (don't forget the commandline parameter '-day'), but you can run it any way you want as long as it is in the CumulusMX-directory with the data-dir just below it.

BTW: anybody who is interested in this must be aware that, for the time being, your datefile.txt must contain metric data with windspeeds in km/h. In future I will make automatic conversions for imperial etc... , but that is not what I focus on now.

Best,

Re: Fire Weather, a new approach

Posted: Fri 13 Sep 2019 8:56 pm
by HansR
beteljuice wrote: Fri 13 Sep 2019 12:18 am Looks fine to me.

.... although (navigation) you still insist on <table><td> instead of <table><tr><td>
Thanks. I had forgotten about that error. Fixed it. Will appear in the next version. I do not see a difference though?
beteljuice wrote: Fri 13 Sep 2019 12:18 am You need a 'title' somewhere ...

something like:
</table>
<h1>Fire Weather Index (experimental)</h1>
<div id="tableData">
I answered it to Phil23 already, now I see it was you who raised it.:

My idea is that the page title and <h1>-header are for the user to set. I just supply the pwsFWI table. I chose this technique because the variety of sites in the wild is so big that, if I add more , wishes will increase and I want to focus on meteo related info and not on site related info.

But, it's just a line I draw.

Re: Fire Weather, a new approach

Posted: Fri 13 Sep 2019 9:47 pm
by Phil23
beteljuice wrote: Fri 13 Sep 2019 12:18 am Looks fine to me.
Just not sure what's going on here with the bars etc.
Capture1.JPG
.... although (navigation) you still insist on <table><td> instead of <table><tr><td>
Yes I've taken not of your comments on that. It's inherited from the original distributions, which don't have it either.

It's on the to do list; do one then replace it in all the others.
Global replaces scare me a bit....
You need a 'title' somewhere ...
Plan to make it look like my other pages & also change the credits.
Hans can suggest the most appropriate place in his Blog to point the credit link to.

At this point the main thing I wanted to do was get the data up & prove I could work out the "include" bit.
(Very much a beginner on page coding).

Cheers

Re: Fire Weather, a new approach

Posted: Fri 13 Sep 2019 9:51 pm
by beteljuice
Hans wrote:My idea is that the page title and <h1>-header are for the user to set. ...

But, it's just a line I draw.
No arguments from this codester :lol: