Skip to content

UK Networks

The UK operates several regional air quality monitoring networks that provide high-quality, reference-grade data.

AURN (Automatic Urban and Rural Network)

The UK's primary national air quality monitoring network, operated by the Environment Agency.

Coverage

  • ~150 monitoring sites across England, Wales, Scotland, and Northern Ireland
  • Mix of urban background, roadside, and rural sites
  • Continuous monitoring since 1973

Available Pollutants

  • Nitrogen dioxide (NO2)
  • Ozone (O3)
  • Particulate matter (PM2.5, PM10)
  • Sulphur dioxide (SO2)
  • Carbon monoxide (CO)

Usage

import aeolus
from datetime import datetime

# Get all AURN sites
sites = aeolus.networks.get_metadata("AURN")
print(sites[['site_code', 'site_name', 'location_type', 'latitude', 'longitude']])

# Download data
data = aeolus.download(
    sources="AURN",
    sites=["MY1", "KC1"],  # Marylebone Road, Kensington
    start_date=datetime(2024, 1, 1),
    end_date=datetime(2024, 1, 31)
)

Location Types

Type Description
Urban Background Representative of city-wide exposure
Urban Traffic Near major roads
Suburban Residential areas
Rural Away from direct sources

SAQN (Scottish Air Quality Network)

Additional monitoring sites operated by the Scottish Environment Protection Agency (SEPA).

Note: SAQD is also accepted as an alias for SAQN for backwards compatibility.

import aeolus
from datetime import datetime

# Get SAQN sites
sites = aeolus.networks.get_metadata("SAQN")

# Download
data = aeolus.download(
    sources="SAQN",
    sites=["ED3"],
    start_date=datetime(2024, 1, 1),
    end_date=datetime(2024, 1, 31)
)

WAQN (Welsh Air Quality Network)

Welsh monitoring network operated by Natural Resources Wales.

import aeolus
from datetime import datetime

data = aeolus.download(
    sources="WAQN",
    sites=["CARD"],  # Cardiff
    start_date=datetime(2024, 1, 1),
    end_date=datetime(2024, 1, 31)
)

NI (Northern Ireland Network)

Northern Ireland network operated by the Department of Agriculture, Environment and Rural Affairs.

import aeolus
from datetime import datetime

data = aeolus.download(
    sources="NI",
    sites=["BEL1"],  # Belfast
    start_date=datetime(2024, 1, 1),
    end_date=datetime(2024, 1, 31)
)

AQE (Air Quality England)

Local authority monitoring sites across England, providing additional coverage beyond the national AURN network.

import aeolus
from datetime import datetime

sites = aeolus.networks.get_metadata("AQE")
data = aeolus.download(
    sources="AQE",
    sites=sites["site_code"].head(3).tolist(),
    start_date=datetime(2024, 1, 1),
    end_date=datetime(2024, 1, 31)
)

LAQN (London Air Quality Network)

London's main regulatory monitoring network, managed by the Environmental Research Group (ERG) at Imperial College London. Provides hourly data from ~250 monitoring sites across Greater London, covering all 33 London boroughs.

Data is fetched from the openair RData feed at londonair.org.uk/r_data/ (one file per site/year, same format as AURN). Site metadata is fetched separately from the ERG JSON API at api.erg.ic.ac.uk/AirQuality/. No API key is required.

Units

The londonair RData files store gases in volume units (ppb; CO in ppm) — they are an interchange format that R's openair::importImperial() converts on read. Aeolus does the same, using Defra's published 20 °C factors (NO2 and NOx 1.9125, O3 1.9957, SO2 2.6609, CO 1.1642), so LAQN values are in µg/m³ (CO in mg/m³) and agree with LAQN-ERG, with the network's own API, and — for sites that are also in AURN, such as Marylebone Road — with Defra's files exactly for ratified years. Particulates are already in µg/m³ and are not converted.

North Kensington (KC1), January–March 2023

The londonair record for KC1 is stamped one hour late from 1 January to 26 March 2023 (the logger appears to have stayed on summer time). The AURN record for the same analyser is correct; prefer AURN for KC1 over that period.

Versions before 0.5.0

Earlier versions did not convert, and labelled the ppb values ug/m3. LAQN gas values from aeolus.download("LAQN", ...) were therefore too low by the factors above (NO2 and NOx by 1.91×, O3 by 2.00×). Anything stored from those versions should be re-downloaded.

Available Pollutants

  • Nitrogen dioxide (NO2)
  • Oxides of nitrogen (NOXasNO2)
  • Ozone (O3)
  • Particulate matter (PM2.5, PM10)
  • Sulphur dioxide (SO2)
  • Carbon monoxide (CO)

Usage

import aeolus
from datetime import datetime

# Find sites near a location
sites = aeolus.find_sites("LAQN", near=(51.5074, -0.1278), radius_km=5)

# Download data
data = aeolus.download(
    sources="LAQN",
    sites=["MY1", "KC1"],
    start_date=datetime(2024, 1, 1),
    end_date=datetime(2024, 1, 31)
)

Data quality

UK regulatory networks provide ratified data:

  • Quality assured by trained personnel
  • Calibrated reference-grade instruments
  • Regular maintenance and audits
  • Traceable to national standards

Data is typically ratified within 6–12 months of collection; until then rows carry qa_code = unverified (AURN) or Provisional (SAQN, WAQN, NI, AQE) and qa_tier = reference_provisional.

For AURN, SAQN, WAQN, NI and AQE, qa_code says whether each row has been ratified: verified/unverified for AURN, Ratified/Provisional for the others. Aeolus derives it per site and pollutant from the ratified_to date in each network's openair metadata (downloaded at most once a day per process, and shared with find_sites; if it cannot be read, Aeolus warns and qa_code is null for that download), so qa_tier is reference_full_qc for ratified rows and reference_provisional otherwise; ratification_stage is ratified/unratified. Near-real-time rows from the SOS sources are always unratified. A pollutant the metadata does not list gets a null qa_code and qa_tier = unknown. LAQN and LMAM publish no ratification field, so their rows stay unknown.

Combining UK Networks

Download from multiple UK networks simultaneously:

data = aeolus.download(
    sources={
        "AURN": ["MY1", "KC1"],
        "SAQN": ["ED3", "GLA4"],
        "WAQN": ["CARD"],
    },
    start_date=datetime(2024, 1, 1),
    end_date=datetime(2024, 1, 31)
)