How we measure aircraft noise
Flight Path Noise publishes overflight counts and modelled noise levels built from recorded ADS-B flight tracks. This page explains exactly how those numbers are made, how often they refresh, and where they fall short.
Where the flight data comes from
Every statistic on this site is derived from ADS-B flight tracks — the position broadcasts that most powered aircraft transmit continuously. We record these tracks around each airport we cover, storing each aircraft's position, barometric altitude, and heading over time. Sampling density varies by airport: our busiest coverage records positions every few seconds, while some airports are sampled at a lower cadence and their tracks are densified by interpolation between recorded points.
Aircraft without transponders — some training, vintage, and private aircraft — are invisible to this method, so light-aircraft counts are a floor, not a ceiling.
How noise levels are estimated
Noise figures are modelled estimates, not microphone measurements. Each aircraft type is assigned a source noise level by category (heavy jet, medium jet, turboprop, light piston, helicopter), and ground-level values are computed from the aircraft's altitude and lateral distance using standard distance-attenuation, giving an estimated peak level for a single overflight and an energy-averaged level (Leq) over time.
Modelled values are most useful for comparison — which streets, suburbs, or hours are louder than others — rather than as absolute readings. Real-world levels vary with weather, engine variant, and aircraft weight on the day.
How suburb and area statistics are computed
Each suburb page blends every tracked airport in its metro area plus an "other" bucket for helicopters and light aircraft from untracked airfields. A flight is counted for a suburb when its track crosses the suburb's official boundary below 10,000 ft; per-day figures divide by the days of data in the rolling window (up to 90 days), which is stated on every page. Flights are attributed to an airport by proximity and altitude at closest approach, or by the flight's own departure/arrival coding; traffic serving an airport outside the metro area is excluded rather than mislabelled.
Suburb pages exist only where the data clears a threshold — roughly 20 overflights a day below 10,000 ft, or 5 a day below 5,000 ft with meaningful modelled noise. We publish what the data supports, not a page for every place name.
Boundaries: Australia — Australian Bureau of Statistics, Suburbs and Localities 2021 (CC BY 4.0), postcodes via GeoNames (CC BY 4.0). United Kingdom — ONS Built-Up Areas 2022 and Code-Point Open, Open Government Licence v3.0; contains OS data © Crown copyright and database right.
How often the data updates
Live airport maps update continuously as aircraft move. Airport statistics pages use a rolling 30-day window recomputed from the live database on every visit. Suburb and area pages are rebuilt from the full track archive on a monthly cycle, and every page states the exact data window it was computed from.
Limitations worth knowing
Altitudes are barometric and can differ from true height by a few hundred feet in unusual pressure conditions. Sparse sampling at some airports means individual flight paths are partly interpolated. Modelled noise is not a certified noise contour and should not be used as one — airport operators' official monitoring remains the authoritative source for regulatory purposes. Where our numbers and an official source disagree, trust the official source and tell us.
Corrections
Spotted something wrong? Email hello@flightpathnoise.com — corrections ship in the next data refresh.
