BRENNEN

Industries / Energy, water and waste / Environmental services

Air quality monitoring

Air quality monitoring measures the gases and particles in outdoor air, in what leaves chimneys and smokestacks, and in the air inside buildings, so that people and regulators know what is being breathed and who is putting it there.

1852

On 4 May 1852 the chemist Robert Angus Smith read "On the Air and Rain of Manchester" to the Manchester Literary and Philosophical Society, reporting that rain "was found to contain sulphuric acid in proportion as it approached the town.".

archive.org

39

Milestones achieved across 6 domains.

9

To watch for: announced, not yet confirmed outside.

12

Open deliverables. Nobody has shipped them.

In one minute

Where it started

In 1852 Robert Angus Smith, unable to analyze Manchester's air directly, collected its rain and found more sulfuric acid the closer it fell to the town. A century later the work became routine: the UK began a national survey of smoke and sulfur dioxide in 1961, and in 1971 the United States wrote stack test methods into law alongside its first national air quality standards.

Where it stands

The World Health Organization says 99% of the world's people lived in 2019 where its air quality guideline levels were not met, and that outdoor air pollution caused 4.2 million early deaths that year. Measurement now comes from certified stations, cheap sensors, instruments in smokestacks and satellites that see some regions every hour, but it is uneven: OpenAQ's 2026 survey of 198 countries found that 33% have no government monitoring. Programs can also shrink, as when the US stopped publishing its embassy readings in 2025 and dropped a continuous dust-monitor rule for power plants in 2026.

What is open

Much of the checking is still done by hand: staff validate station data hour by hour, trained observers grade smoke by eye, and auditors measure station siting with a tape. Few cheap sensors and home monitors have been tested by anyone other than their makers, and satellites that can now see single plants have rarely been checked against those plants' own stack meters. Open data, phone cameras, small AI models and public satellite archives make each of these jobs possible for a small team.

Domains

AchievedWatch forOpen

Advances to build with

1 research note

Open in air quality monitoring

3 of 12 open deliverables

Open · Reference stations

A public test for validation

An open benchmark for the hand-checking of station data. Take hourly readings from European reference stations twice, as first reported in near real time and as later validated by the network's staff, and mark every hour the validators deleted or changed. Publish the paired data, a scoring script and a baseline model, so that any automatic checker can be scored against human decisions on a year it has not seen.

Built with what exists

The European Environment Agency's air quality download service, which serves station data as Parquet files; Python with pandas or DuckDB; LightGBM, the open-source library ENVEA's module uses, as a first baseline; Zenodo or Hugging Face to host the data.

Credit · unclaimed

Open · Low-cost and mobile sensors

Test NO2 sensors to targets

An independent, public test of at least five commercial nitrogen dioxide (NO2) sensor systems, run to EPA's February 2024 base-testing protocol beside a regulatory NO2 monitor, with results reported in EPA's fillable templates and all raw data published.

Built with what exists

Three units each of five or more NO2 sensor systems on sale today; space beside a state or local agency's NO2 reference analyzer; EPA's 2024 supplemental report for NO2, carbon monoxide and sulfur dioxide sensors, with its checklists and templates; Python or R for the metrics.

Credit · unclaimed

Open · Testing at the stack

Pass smoke school by software

An open-source program that reads plume opacity from ordinary video, frame by frame, and sits the same certification test a human smoke reader sits: 25 black and 25 white plumes from a smoke generator, scored against the generator's built-in opacity meter.

Built with what exists

A phone or action camera; Python, OpenCV and PyTorch; a small vision model fine-tuned on labeled frames from smoke-generator runs; the Method 9 text and EPA's Visible Emissions Field Manual for the scoring rules; a smoke school that will let the camera film a certification session.

Credit · unclaimed

Where it sits