CleanZone Field Brief
The 40-Decibel Rule: Mapping Night Noise
The WHO guideline for night noise in Europe is clear: outdoor Lnight should not exceed 40 dB if sleep is to remain restorative. Most European cities exceed it. Most American cities far exceed it. Here is how the guideline maps to transport corridors, and what to look for at the parcel level.
At a glance
- WHO's 2009 Night Noise Guidelines set 40 dB Lnight outdoors as the long-term goal for undisturbed sleep; the 2018 update split it by source — road 45 dB, rail 44 dB, aircraft 40 dB (all "strong recommendations").
- Decibels are logarithmic: +3 dB is roughly double the sound energy; +10 dB is roughly a perceived doubling of loudness. A "small" 10 dB rise is an order-of-magnitude physical event.
- The body reacts to night noise long before it wakes you — heart rate, blood pressure, and stress hormones respond at levels well below conscious awakening.
- The EU's Environmental Noise Directive requires strategic noise maps every five years for agglomerations over 100,000 residents, major roads, rail lines, and airports.
- CleanZone's
noisefield places a cell relative to these public thresholds; it does not substitute for a metered reading at your specific window.
The lorry you never woke up for
Somewhere between 2 a.m. and 4 a.m., most people living within a few hundred metres of a motorway, a freight line, or a flight path have a night-noise event they will never remember. They don't wake up. They don't roll over. In the morning they report sleeping fine. Their heart rate, meanwhile, spiked, their blood pressure ticked up, and a small dose of stress hormone entered their bloodstream — a measurable autonomic response with no accompanying memory. This is the part of night noise that brochures don't show and residents don't self-report, which is precisely why public-health bodies had to go and measure it directly.
The World Health Organization's answer is a single number applied to the entire night, not the loudest moment in it: Lnight, the A-weighted equivalent continuous sound level averaged over the eight-hour night period. One truck passing at 80 dB does not breach a Lnight limit if it happens once; twenty trucks at 60 dB each, spaced through the night, can. It's an energy average, and averaging decibels is where most people's intuition about noise quietly falls apart.
Because the scale is logarithmic, the gap between "quiet" and "loud" compresses almost all of audible experience into a 130-decibel span — from leaves to a jet at 100 metres. That compression is a feature for engineers and a trap for residents: it makes a 10 dB rise feel like "a bit more noise" when it is physically ten times the sound energy hitting the window.
Why ten decibels isn't ten per cent
The decibel is defined as ΔdB = 10·log10(I/I0), where I is sound intensity relative to a reference I0. Run the arithmetic and two facts fall out that govern almost everything about how night noise is regulated and how people misjudge it: a rise of roughly 3 dB doubles the actual sound energy reaching a surface, while a rise of roughly 10 dB — ten times the energy — is what a human ear perceives as roughly a doubling of loudness. Energy and perception are not the same curve, and neither is remotely intuitive from the number alone.
That mismatch is why two properties both described as "a bit noisier than average" can differ by an order of magnitude in the raw physics, and why regulators anchor rules to Lnight in decibels rather than to any plain-language description of "busy" or "quiet."
What the guideline actually sets, source by source
The figure most people quote — "40 decibels" — comes from the WHO's 2009 Night Noise Guidelines for Europe, which set Lnight,outside 40 dB as the long-term goal to protect the public, including the most vulnerable groups, from the adverse health effects of night noise. The 2018 WHO Environmental Noise Guidelines for the European Region then split the recommendation by source, reflecting that different noise types produce different sleep-disturbance profiles at the same measured level:
Aircraft gets the strictest number despite usually producing a lower average Lnight than a motorway, because aircraft noise arrives as discrete high-amplitude events rather than a continuous hum — and event-driven noise disturbs sleep architecture more per decibel than steady-state noise does. Road traffic's 45 dB and rail's 44 dB reflect the opposite case: near-continuous or frequent exposure that the ear partially — never fully — adapts to.
| Source | Typical L_night at 50 m from source | Distance to reach 40 dB | Key parameter |
|---|---|---|---|
| Major airport approach | 50 – 65 dB | Typically 1 – 3 km from threshold, depending on flight-path density | Operations per night, runway-in-use frequency |
| Motorway (100 km/h, >25k AADT) | 55 – 65 dB | 200 – 400 m in flat ground; farther in a valley, closer on a cutting | Traffic volume, HGV %, road surface type |
| Main rail line (freight + passenger) | 45 – 55 dB | 100 – 250 m, depending on speed and number of night trains | Train speed, axle load, number of night services |
| Urban arterial | 50 – 60 dB | 150 – 300 m, but façade reflection and street-canyon effects raise levels | Geometric spreading, façade reflection, canyon ratio |
Illustrative — modelled mid-range distance estimates; local terrain and façade geometry can shift these figures significantly.
The stress you sleep through
Sleep is not a single state that noise either interrupts or doesn't. It is a cycle of stages — light, deep, REM — and the body's tolerance for disturbance varies through that cycle. Noise events too small to produce a conscious awakening still routinely produce a "sub-cortical arousal": a shift toward lighter sleep, a rise in heart rate, a measurable blood-pressure response, sometimes a release of stress hormones. None of this shows up in a sleep diary. All of it shows up on a polysomnograph, which is how the evidence base behind the WHO guidelines was built in the first place — the health-outcome literature on cardiovascular disease and night noise doesn't rely on people accurately remembering being woken up.
The EU's Environmental Noise Directive (2002/49/EC) requires member states to produce strategic noise maps — and action plans based on them — every five years for every agglomeration over 100,000 residents, every major road carrying more than three million vehicles a year, every railway with more than 30,000 train passages a year, and every major airport. It is the regulatory machinery that turns the WHO's health-based guideline values into something a planning authority actually has to look at.
Two systems, two philosophies
Europe and the United States regulate night noise from fundamentally different starting points, and the gap between them is wide enough to matter when comparing a European suburb to an American one on paper.
United States — DNL, event-threshold model
The FAA's Day-Night Average Sound Level (DNL) already applies a 10 dB penalty to nighttime operations before averaging, then sets 65 dB DNL as the federal threshold for a "significant" noise impact requiring mitigation. There is no equivalent federal mandate to map residential noise exposure block-by-block the way the EU requires.
European Union — L_night, mapping-first model
The END mandates population-level strategic noise maps on a fixed five-year cycle, and the WHO's source-specific Lnight values (40–45 dB) sit far below the US action threshold — meaning a level considered background in much of the US would already breach the EU's health-based guideline.
Transport-noise exposure is not fixed to a property the way a postcode is. Runway-in-use at an airport changes with wind direction, so a home quiet during westerly operations can sit directly under an approach path during easterly flow. Road and rail volumes shift with infrastructure projects and timetable changes. Any noise figure — including a modelled one — describes a period, not a permanent state.
How the grid encodes it
CleanZone's noise field places a cell against these public thresholds, drawing on the same category of inputs the regulators themselves use: strategic noise-map outputs reported under the END for Europe, published traffic and rail volumes, and aviation operational data for flight-path modelling. It does not claim to replicate a certified acoustic survey at a specific window, and it makes no claim about measurement precision beyond what the underlying public sources support — it exists to show, against the WHO's own numbers, which side of the line a location falls on.
Night noise is the most underestimated of the standard environmental stressors precisely because it produces no memory. A property that seems perfectly quiet at midday can sit above the WHO's guideline value every single night for years, with the only trace being an elevated resting heart rate nobody thought to link back to a bedroom window. The noise field exists so that the 40 dB line is checked against public data before the lease is signed, not diagnosed afterward.