Welcome to the Cumulus Support forum.

Latest Cumulus MX release 3.9.5 (build 3100) - 14 January 2021 (please see announcement regarding releases since 3.5.0)
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 the NOAA report and Snow Index drop-down menus to 2030)

Use this Wiki link to Download the Software (Cumulus MX / Cumulus 1 and other related items).

On Air Quality

Discussion of the Cumulusutils tool and website generator.

Moderator: HansR

Post Reply
User avatar
HansR
Posts: 1651
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 10 (Buster)
Location: Wagenborgen (NL)
Contact:

On Air Quality

Post by HansR »

Air quality is a complex matter. Not so much in terms of technology but because every country has it's own environmental organisation which produces its own research and creates its own Air Quality Indices (AQIs) for Particle Matter, which for the sake of simplicity are often combined with other pollutants like NO2 and O3 , deviate in normative periods (1 hr average, 3 hr average or 24 hr average). This makes it almost impossible to present a unified Air quality image from the AQI perspective. If one seeks compliance with the normative country one supports and/or chooses.

It is clear that devices like the AirLink, PurpleAir and organisations like AirNow are busy setting a de facto standard. My advise to the user would be if you are confused or your country does not really has great support for an Air quality standard, go for the US standard. It is clear, well worked out and supported worldwide.

Many sites show Air Quality worldwide. Look e.g. at PurpleAir , IQAir or Davis (although for the last I can't find the map - Anybody?). A special site is aqicn.org which not only shows a lot of info, has great FAQ but also has a great sensor page. It is a good place to start.

Some scientific articles on sensor comparison and AQI can be found here on the forum with this post. Scientists have looked into these sensors to see how they compare to the official measurements. The cheap sensors compare positively. Some articles on this subject are included as well.

The cheap sensors are often referred to as toys. They are not. Many environmental organisations set up a community of users using this type of sensors to get indicative measurements. They can be a bit compared to WOW but are more open. Take a look at Sensor.Cummunity, Luftdaten.info and the Dutch RIVM community page (leading to Sensor.Community). Look in your own country for organisations like this.

For meteorology most people know where to find the information or are educated in that field. For Air Quality that is less so and it is a good idea to read yourself into the matter. Knowing that an AQI is basically a unit related to the statistic of how many people are submitted to hospital with certain types of diseases and the fact that the AQI is apparently not a clear objective figure like temperature. That is kind of worrying in my opinion.

I will modify and extend this post when and if I find additional info or when required. You are invited to add information yourself in the comments. It is more or less independent of the AirQuality module.

[Edit: 29 November 2020]
While finishing my AirQuality module of CumulusUtils I added this info.

The numbers used by the Air Link (and stored each minute by CMX) are:
  • Minute values;
  • 1 hour running averages;
  • 3 hour running averages;
  • 24 hour running averages;
  • nowcast (see below, also a figure every minute)
The specs of the calculations of nowcast can be found here (the nowcast)) or from the nowcast wiki.

You can easily see from the graphs on CumulusUtils that the difference between the 1hr/3hr and the nowcast running averages is not that big. Nevertheless, the nowcast figure is a hard demand in the VS apparently. Just as the absolute concentrations may differ in percentages from the official figures, it is the trend which shows it all. If a cheap sensor shows 100 μg/m3 and somebody else says it is only 20 μg/m3 than you are in for an argument. And that is the real value of the cheap sensors. That is also where the European sensor.community is appreciated by (e.g.) the Dutch National Health Organisation.

BTW: These sensors become dirty by definition because of the constant airflow through the measurement channel where the laser and the lens are. That needs cleaning once in a while (let's say once a year): opening the sensor, carefully cleaning the dirt away.

[Edit: 13 december 2020]
As AQI is very different in different countries I will publish here the documentation and a summary sheet in the following comment.
You do not have the required permissions to view the files attached to this post.
https://meteo-wagenborgen.nl / Cumulus 3.9.4 (build: 3099)
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian Buster / 5.4.79-v7+ armv7l
Mono JIT compiler version 5.18.0.240 (Debian 5.18.0.240+dfsg-3 Sat Apr 20 05:16:08 UTC 2019)

User avatar
HansR
Posts: 1651
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 10 (Buster)
Location: Wagenborgen (NL)
Contact:

Re: On Air Quality

Post by HansR »

Here are the references for the AQI / concentration relation - these are about how CMX calculates the AQI.
The summary is given as an image and the spreadsheet is attached (Open Office format).

Note the huge differences between the AQI and the concentrations per country. It can not be that Belgians are more sensitive to particulate matter then the Dutch. A similar distinction holds for the Americans and the British. There will no doubt be political reasons as to why to do this. Economy and pollution are highly related.

Also note the time for the series 1hr, 3 hr, 24 hr which also determines the levels.

I will think about how to handle this in - specifically - the graphing of the sensor values.
In a later release of CumulusUtils, AQI zones will be displayed which will indicate the air quality alarm level.
The manual and the Help button in the graph will then show what you see and why.
  1. US: https://www.airnow.gov/sites/default/fi ... ay2016.pdf
  2. UK: https://assets.publishing.service.gov.u ... _index.pdf
  3. EU: AQI: http://www.airqualitynow.eu/about_indic ... nition.php
  4. EU: CAQI: https://www.airqualitynow.eu/download/C ... orders.pdf
  5. CA: https://en.wikipedia.org/wiki/Air_Quali ... x_(Canada)
  6. AU: https://www.environment.nsw.gov.au/topi ... lity-index
  7. NL: https://www.luchtmeetnet.nl/informatie/ ... ndex-(lki)
  8. BE: https://www.irceline.be/en/air-quality/ ... anguage=en
Schermafbeelding 2020-12-13 152010.png
You do not have the required permissions to view the files attached to this post.
https://meteo-wagenborgen.nl / Cumulus 3.9.4 (build: 3099)
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian Buster / 5.4.79-v7+ armv7l
Mono JIT compiler version 5.18.0.240 (Debian 5.18.0.240+dfsg-3 Sat Apr 20 05:16:08 UTC 2019)

User avatar
PaulMy
Posts: 2390
Joined: Sun 28 Sep 2008 11:54 pm
Weather Station: Davis VP2+ Cumulus1 / CummulusMX
Operating System: Windows8 / Windows10
Location: Komoka, ON Canada
Contact:

Re: On Air Quality

Post by PaulMy »

Hi Hans,
I have the update installed and have run website on both www.komokaweather.com/cumulusmxwll/index.html and test www.komokaweather.com/cumulusmxnew/index.html
In both there is an error at line 106 of the log (attached) but everything seems ok on the website

Here are the local AQ sites - we are on the outskirts of London, ON with no close by industry. One shows Low similar to my AirLink, and the other Moderate:
https://aqicn.org/city/canada/ontario/london/
https://weather.gc.ca/airquality/pages/onaq-011_e.html

Thanks,
Paul
Last edited by PaulMy on Thu 17 Dec 2020 9:32 pm, edited 1 time in total.

User avatar
HansR
Posts: 1651
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: Davis Vantage Pro 2+
Operating System: Raspbian GNU/Linux 10 (Buster)
Location: Wagenborgen (NL)
Contact:

Re: On Air Quality

Post by HansR »

Hi Paul,
PaulMy wrote:
Thu 17 Dec 2020 8:30 pm
I have the update installed and have run website on both www.komokaweather.com/cumulusmxwll/index.html and test www.komokaweather.com/cumulusmxnew/index.html
In both there is an error at line 106 of the log (attached) but everything seems ok on the website
The first link gives me the 4.8.4Patch2, the second link just 4.8.4 I don't know why you run both though much respect you can :clap:
They look good indeed and I will check how that CA AQI is working out because I dislike it is not done for all averages. That has to do with the way it is calculated by CMX.
I'll come back on that.
PaulMy wrote:
Thu 17 Dec 2020 8:30 pm
Here are the local AQ sites - we are on the outskirts of London, ON with no close by industry. One shows Low similar to my AirLink, and the other Moderate:
https://aqicn.org/city/canada/ontario/london/
https://weather.gc.ca/airquality/pages/onaq-011_e.html
Thanks, I'll study those.
And btw: the first link uses the US standards (converted to US EPA AQI standards), the second apparently the Canadian standards and most likely does not only incorporate particulate matter but also NO2 etc...
Nice to have comparison ;)

BTW: the error in the log is not disastrous but not good either. I don't get why he can't parse the time. Work :roll:
[Edit:] Ah... some locale issue. Too late for today.
https://meteo-wagenborgen.nl / Cumulus 3.9.4 (build: 3099)
Davis Vantage Pro 2+ / Raspberry Pi 3 Model B Plus Rev 1.3
Raspbian Buster / 5.4.79-v7+ armv7l
Mono JIT compiler version 5.18.0.240 (Debian 5.18.0.240+dfsg-3 Sat Apr 20 05:16:08 UTC 2019)

User avatar
PaulMy
Posts: 2390
Joined: Sun 28 Sep 2008 11:54 pm
Weather Station: Davis VP2+ Cumulus1 / CummulusMX
Operating System: Windows8 / Windows10
Location: Komoka, ON Canada
Contact:

Re: On Air Quality

Post by PaulMy »

The first link gives me the 4.8.4Patch2, the second link just 4.8.4 I don't know why you run both though much respect you can
Possibly a cache as both sites show the same 4.8.4 with patch for me.
The cumulusmxnew is a test page to run 3 templates on 1 CumulusMX and eliminate the separate CumulusMX installs. So far results are ok.

Enjoy,
Paul

Post Reply