Learn by Directing AI
All materials

data-dictionary.md

Data Dictionary

Air Quality Data

air-quality-data.csv

Field Name Type Description Example Value Notes
station_id string Unique monitoring station identifier STK-01 Format: city prefix (STK=Stockholm, GOT=Gothenburg, MAL=Malmo) + sequential number
date date Measurement date 2021-06-15 Format: YYYY-MM-DD. Daily averages computed from hourly readings.
pm25 float PM2.5 concentration in micrograms per cubic meter 12.4 Particulate matter 2.5 micrometers or smaller. The primary pollutant targeted by the vehicle emission regulation.
no2 float NO2 concentration in micrograms per cubic meter 28.7 Nitrogen dioxide. Also affected by vehicle emissions but not the primary regulatory target.
ozone float Ozone concentration in micrograms per cubic meter 64.2 Ground-level ozone. NOT targeted by the vehicle emission regulation. Ozone has an inverse seasonal pattern to PM2.5 (higher in summer).
so2 float SO2 concentration in micrograms per cubic meter 4.1 Sulfur dioxide. Primarily from industrial sources, not vehicles. Relatively stable over the study period.

Notes:

  • Data spans January 1, 2019 through December 31, 2025 (7 years).
  • The vehicle emission regulation took effect on January 1, 2022.
  • Readings are daily averages per station, aggregated from hourly monitoring data.
  • Some dates have missing readings due to equipment maintenance -- these are genuine gaps, not errors.
  • All 40 monitoring stations have been operating for the full study period.

Weather Data

weather-data.csv

Field Name Type Description Example Value Notes
station_id string Monitoring station identifier (matches air quality data) STK-01 Same station identifiers as air quality data.
date date Measurement date 2021-06-15 Format: YYYY-MM-DD. Same dates as air quality data.
temp_c float Daily mean temperature in Celsius 8.3 Nordic climate range: approximately -15 to 35 degrees Celsius.
wind_speed_ms float Daily mean wind speed in meters per second 4.2 Higher wind speeds disperse pollutants. Coastal cities (Gothenburg, Malmo) tend to have higher wind speeds.
precip_mm float Daily precipitation in millimeters 2.1 Precipitation can wash particulates from the air.

Notes:

  • Weather variables affect air quality readings substantially -- cold, calm days produce higher PM2.5 concentrations.
  • Weather data is station-matched to allow for local weather effects.

Station Metadata

station-metadata.csv

Field Name Type Description Example Value Notes
station_id string Unique monitoring station identifier STK-01 Same identifiers as in the air quality and weather data.
city string City where the station is located Stockholm One of: Stockholm, Gothenburg, Malmo.
station_name string Descriptive station name Hornsgatan Traffic Human-readable name indicating location and context.
station_type string Classification of monitoring environment traffic One of: traffic (near roads), urban background (city, away from roads), suburban (residential areas). Station type affects baseline pollutant levels.
latitude float Station latitude 59.3172 WGS84 coordinates.
longitude float Station longitude 18.0486 WGS84 coordinates.
start_date date Date station began operating 2015-03-01 All stations started before the study period (2019).

Notes:

  • Stockholm: 12 stations (4 traffic, 5 urban background, 3 suburban).
  • Gothenburg: 14 stations (5 traffic, 5 urban background, 4 suburban).
  • Malmo: 14 stations (5 traffic, 5 urban background, 4 suburban).
  • Total: 40 monitoring stations.