Noise mapping is the process of creating detailed visual representations of sound levels across a geographic area, typically using computer models that combine traffic data, building geometry, and the physics of how sound travels through air. These maps display estimated decibel levels at every point in a neighborhood or city, usually color-coded from green (quiet) to red (loud), and they have become a standard tool in urban planning and public health policy. The European Union’s Environmental Noise Directive made strategic noise mapping mandatory for large cities and major transport corridors over two decades ago, and the practice has since spread worldwide, evolving from static government reports into real-time sensor networks and crowdsourced smartphone apps.
How a Noise Map Gets Built
A noise map is not a recording. It is a calculation. The process starts with a noise emission model that estimates how loud a given source is, whether that source is a stream of cars on a highway, a railway line, or an airport flight path. For road traffic, the model needs to know the volume of vehicles, their speed, the mix of cars and trucks, and the road surface type. These inputs feed into a propagation model that calculates how sound spreads outward and weakens as it travels. The propagation model accounts for geometric spreading (sound gets quieter with distance), absorption by the ground surface, reflections off building facades, and diffraction around obstacles like noise barriers and building edges.
In Europe, the standard framework for these calculations is called CNOSSOS-EU. Its sound propagation module traces the path sound takes from source to receiver, considering both calm atmospheric conditions and conditions where temperature gradients bend sound waves downward toward the ground. For each path between a noise source and a point on the map, it calculates attenuation from the ground surface, geometric divergence, building reflections, and diffraction corrections.1Department Research areas Environmental Noise CNOSSOS-EU. CNOSSOS-EU The result is an estimated sound pressure level at every grid point across the mapped area, typically calculated at a receiver height of about 1.5 meters or 4 meters above the ground, depending on the directive’s requirements.
Countries that do not have their own standardized noise prediction model often adapt existing European or North American ones and then calibrate them against local measurements. A systematic review of these adaptations found that most calibration campaigns involve 15-minute sound pressure level measurements taken at 1.5 meters above ground level, and that many maps are calculated at this same height to make the comparison straightforward.2PubMed Central. Use of noise prediction models for road noise mapping in locations that do not have a standardized model: a short systematic review The challenge is that input data quality varies enormously. In Guangzhou, researchers used GPS data from floating cars to estimate traffic flow through speed-density relationships, then combined that with GIS-exported road and building geometry to calculate day and night noise levels across the city.3Applied Acoustics. Technical Note Road traffic noise mapping in Guangzhou using GIS and GPS In dense neighborhoods like Copacabana in Rio de Janeiro, researchers compared simulation results against field measurements at multiple locations and times of day, finding that the gap between modeled and measured levels had to be carefully analyzed before the maps could be trusted for a city of that size.4PubMed. Noise mapping of densely populated neighborhoods–example of Copacabana, Rio de Janeiro-Brazil
Why Governments Require Them
The single biggest driver of noise mapping worldwide has been the European Union’s Environmental Noise Directive (END), adopted in 2002 as Directive 2002/49/EC. The directive requires member states to produce strategic noise maps for all major roads, railways, airports, and cities above a certain population threshold, and then to develop action plans to reduce noise where levels are highest. Ireland’s first phase of mapping, for example, covered one agglomeration, one airport, and roughly 600 kilometers of major roads, describing noise exposure for about 1.25 million people. That work involved five separate noise mapping bodies and 26 action planning authorities, and the second phase lowered the thresholds, requiring even more local governments to participate.5PubMed. Implementation of the EU environmental noise directive: lessons from the first phase of strategic noise mapping and action planning in Ireland
The directive’s influence extends beyond Europe. Its mapping requirements created a standardized methodology that other countries reference when developing their own noise regulations. The maps themselves are not just academic exercises. They feed directly into planning decisions: where to build new housing, where to install noise barriers, where to reroute truck traffic, and where to set speed limits. They also serve as baseline data for health research, since the mapped noise levels can be linked to population databases to estimate how many people are exposed to levels above recommended limits.
A key point researchers have emphasized is that the growing availability of strategic noise maps is expected to improve epidemiological studies on the health effects of noise. Standardized mapping software produces detailed noise levels for every building facade in a mapped area, giving health researchers much finer exposure estimates than they previously had access to.6Noise and Health. Cardiovascular effects of noise Before noise maps, studies often relied on crude proxies like distance from a road or self-reported noise annoyance.
Real-Time Monitoring With Sensor Networks
Traditional strategic noise maps are snapshots. They represent average conditions over a year, calculated from typical traffic patterns and standard meteorological assumptions. They do not capture the nighttime motorcycle that tears through a residential street, the construction project that starts at 7 a.m., or the weekend festival that changes an entire neighborhood’s sound environment. This is where real-time sensor networks come in.
Wireless acoustic sensor networks consist of microphone-equipped devices installed at fixed locations throughout a city, continuously measuring sound levels and transmitting the data to a central server. In the Spanish city of Linares, researchers deployed such a network and ran it continuously for ten months, demonstrating that the system could produce accurate, ongoing noise level maps for smart city applications.7PubMed Central. Wireless Acoustic Sensor Nodes for Noise Monitoring in the City of Linares (Jaén) The advantage of these systems is temporal resolution. Instead of a single annual average, city managers get hour-by-hour data that can reveal patterns invisible to traditional modeling, like noise spikes during school drop-off times or quiet periods during public holidays.
On the lower-cost end, researchers have built Internet of Things prototypes using inexpensive hardware like Arduino Nano boards paired with Wi-Fi modules and basic sound sensors. These devices measure ambient sound intensity and push the data to a web application that plots readings on a map in real time.8arXiv. IoT Based Real Time Noise Mapping System for Urban Sound Pollution Study The appeal of this approach is scalability. A network of cheap sensors can cover a much wider area than a handful of professional-grade sound level meters, though the trade-off is measurement precision. Professional instruments meeting international standards can cost thousands of dollars per unit, while a DIY sensor node might cost under fifty but lack the frequency response and calibration of a proper measurement device.
Crowdsourced Noise Maps From Your Phone
If fixed sensors are one way to make noise monitoring cheaper and more widespread, smartphones are another. The microphones built into modern phones are reasonably sensitive, and several research teams have explored whether data from thousands of phone users could substitute for expensive monitoring equipment. The central challenge is calibration. Phone microphones are designed for voice calls, not acoustic measurement, and their sensitivity varies between manufacturers, models, and even individual units.
One research group developed an in-house app for both Android and iOS that includes a calibration method to make smartphones accurate enough for environmental noise mapping, and their study demonstrated the viability of this approach.9Applied Acoustics. Crowdsourcing of environmental noise map using calibrated smartphones The calibration typically involves comparing the phone’s reading against a reference sound level meter under the same conditions. For best results, this reference reading should be within the phone’s optimal measurement range, which Android specifications define as between 72 and 102 decibels.10Building and Environment. An open-science crowdsourcing approach for producing community noise maps using smartphones
The most prominent crowdsourced noise mapping platform is the NoiseCapture app, developed as an open science project. After more than three years of operation, the application had collected a considerable amount of data from users worldwide, offering spatial and temporal noise analysis for the scientific community. But the researchers behind it are candid about its limitations: the data quality is constrained by the nature of voluntary contributions, inconsistent measurement protocols, variable technical performance across devices, and the fact that most users never calibrate their phones.11PubMed Central. A Smartphone-Based Crowd-Sourced Database for Environmental Noise Assessment An uncalibrated phone measurement can be off by several decibels, which sounds small but matters when you are trying to determine whether a location exceeds a regulatory limit of, say, 55 decibels at night.
Despite these limitations, crowdsourced data fills a genuine gap. Strategic noise maps are expensive to produce and are updated on five-year cycles in many jurisdictions. Smartphone data, even imperfect smartphone data, can flag emerging noise problems between official updates, identify locations where the model predictions seem wrong, and capture sources that traditional maps miss entirely, like construction noise or late-night entertainment districts.
Machine Learning and Frequency-Specific Mapping
Conventional noise maps express everything as a single number: the A-weighted decibel level, which mimics how the human ear perceives loudness. But two locations with the same overall decibel reading can sound very different. A low-frequency rumble from an industrial port feels different from the mid-frequency roar of highway traffic or the high-pitched whine of an electric vehicle’s inverter. The A-weighted average smooths over these differences.
Researchers in one study used mobile surveys and machine learning to go beyond A-weighted measurements, building prediction models for low-frequency (11 to 177 Hz), mid-frequency (177 Hz to 5.68 kHz), and high-frequency (5.68 to 22.72 kHz) bands at 30-meter resolution across an entire city. The models were driven by urban form variables like road density and building layout, and their accuracy ranged from moderate for low frequencies to reasonably good for mid frequencies, with mean absolute errors of about 4 to 5 decibels. The main predictors of noise were related to transport modes, but for low frequencies, port activities were also an important driver.12PubMed. Mobile surveys and machine learning can improve urban noise mapping: Beyond A-weighted measurements of exposure
This matters because frequency content affects both health outcomes and annoyance. Low-frequency noise penetrates building walls more easily than high-frequency noise, so a home near a port might have a moderate A-weighted level outside but still feel oppressively noisy inside. Frequency-specific maps give planners a more honest picture of where people are actually bothered, and they could eventually help health researchers tease apart which acoustic characteristics are most harmful.
Aircraft and Rail Noise
Road traffic dominates most urban noise maps, but airports and rail corridors demand specialized approaches. Aircraft noise is modeled differently because the source moves in three dimensions. Tools based on the ECAC Doc 29 method reconstruct each aircraft operation as a flight path composed of a ground track and a flight profile, using an aircraft noise database that covers about 150 reference aircraft models. The noise computation engine then estimates sound levels on the ground for each operation and accumulates them into contour maps around the airport.13Transportation Research Part D: Transport and Environment. Aircraft operation reconstruction and airport noise prediction from high-resolution flight tracking data
What makes aircraft noise mapping especially contentious is that the maps directly determine land-use restrictions around airports. A house inside the 65-decibel contour might be ineligible for certain types of development or federal funding in some countries. Property owners care intensely about where those lines fall, and small changes in the modeling assumptions or flight path data can shift the contours by hundreds of meters, affecting thousands of homes.
Railway noise involves its own complications. Rail noise includes both airborne sound (the rumble and squeal you hear above ground) and ground-borne vibration that transmits through the earth and re-radiates as low-frequency noise inside buildings. Predicting both requires numerical modeling that draws on data from conventional site investigations of the ground conditions along the rail corridor.14IntechOpen. Airborne and Ground-Borne Noise and Vibration from Urban Rail Transit Systems Residents near subway tunnels sometimes report feeling a rumble or hearing a hum that traditional airborne noise maps would not capture at all, because the energy enters the building through its foundations rather than through the air.
Underwater Noise Mapping
Noise mapping is not exclusively a land-based exercise. The ocean has become dramatically louder over the past century as global shipping traffic has grown, and marine biologists increasingly use noise maps to understand the impact on animals that depend on sound to communicate, navigate, and find food.
One landmark study mapped cumulative underwater acoustic energy from shipping throughout 2008 across the west Canadian Exclusive Economic Zone, using a sound transmission model combined with ship tracking data from the Automatic Identification System. The results showed high noise levels in critical habitats for endangered resident killer whales, with levels exceeding what the EU Marine Strategy Framework Directive considers compatible with “good conservation status.”15Journal of the Acoustical Society of America. Mapping cumulative noise from shipping to inform marine spatial planning
Building on that work, researchers created species-specific noise hotspot maps for ten marine mammal species in Canada’s Pacific region. They weighted ship noise levels by each species’ hearing sensitivity and overlaid the results with animal density maps, producing maps that show where noise is predicted to have the greatest impact. The Juan de Fuca and Haro Straits emerged as hotspots for every species in the area, simply because of the density of ship traffic, with secondary hotspots around Johnstone Strait and Prince Rupert.16PLOS ONE. Identifying Modeled Ship Noise Hotspots for Marine Mammals of Canada’s Pacific Region These maps give regulators a tool for deciding where to impose vessel speed reductions or reroute shipping lanes, measures that are already being tested in several busy waterways worldwide.
Green Space, Property Values, and How Maps Shape Cities
One of the more practical uses of noise maps is guiding where to put parks and green corridors. Vegetation and open space can buffer noise to some extent, but the relationship between green space and noise complaints is more nuanced than “more trees equals less noise.” A study examining the spatial pattern of noise complaints found that when a city’s green space is fragmented into many small patches, noise complaints actually go up. It is the total area of green cover, and particularly the presence of large continuous green spaces, that correlates with fewer noise complaints.17Sustainable Cities and Society. Does urban green space form influence the spatial pattern of noise complaints? Scattering tiny pocket parks throughout a noisy neighborhood does less good than consolidating green space into larger blocks that create genuine acoustic and perceptual relief.
Noise maps also feed into real estate markets, sometimes in ways homebuyers do not realize. Researchers in Poland used strategic noise map data as an input variable in property price analysis and found a relationship between mapped road traffic noise levels and residential property prices.18Real Estate Management and Valuation. The Impact of Road Traffic Noise on Housing Prices – Case Study in Poland The general pattern across similar studies in multiple countries is consistent: higher noise exposure on the map corresponds to lower property values, with estimates typically falling in the range of a fraction of a percent to about one percent of home value per additional decibel. This is not a trivial effect. A home on a busy road mapped at 70 decibels versus a side street at 55 decibels could be looking at a meaningful price difference, and noise maps make that exposure transparent in a way that a quick visit to the property on a Sunday afternoon would not.
What Noise Maps Get Wrong
For all their utility, noise maps have real limitations that users should understand. The models are only as good as their input data. Traffic counts might be outdated, building footprints might not reflect recent construction, and the maps often assume flat terrain or simplified ground conditions. Most strategic noise maps do not account for transient sources like construction, barking dogs, or neighborhood events. They represent long-term average exposure, not what you hear at any given moment.
There is also an inherent tension between resolution and accuracy. A map showing noise levels at a 10-meter grid looks impressively detailed, but the underlying model might not be validated at that resolution. The difference between the modeled value and a field measurement is commonly several decibels even in well-calibrated studies, and at the neighborhood scale, local features like an alley that funnels wind noise or a courtyard that traps sound reflections may not be captured at all. Researchers have acknowledged that discrepancies between modeled and measured levels need careful analysis before maps can be reliably applied to large and complex urban environments.
Perhaps the most important limitation is what the maps measure. The standard output is a decibel number weighted to approximate human hearing sensitivity. It tells you nothing about what kind of noise is present, whether people find it annoying or pleasant, or whether it is constant or intermittent. The hum of a distant highway and the laughter from a sidewalk café might register the same decibel level but produce very different subjective experiences. This is why some researchers are pushing for soundscape mapping, which tries to capture the quality of the sound environment rather than just its volume, but that work is still in relatively early stages compared to the well-established practice of decibel-based noise mapping.
Getting Access to Noise Maps for Your Area
If you live in the European Union, your city’s strategic noise map is public information. EU member states are required to publish them and make them accessible, and most are available through national environmental agency websites or interactive online viewers. Many cities outside the EU have followed suit, publishing noise maps voluntarily as part of urban planning transparency efforts.
When looking at a noise map, pay attention to which indicator is displayed. The most common are Lden (day-evening-night level, which adds a penalty for evening and nighttime noise to reflect the extra annoyance of noise during rest hours) and Lnight (the nighttime average, usually calculated for the hours between 11 p.m. and 7 a.m.). A location that looks acceptable on the Lden map might still have nighttime levels above recommended limits if a major freight route operates heavily after midnight. Also check the map’s reference year and data sources. A map based on 2017 traffic counts may not reflect a new highway interchange opened in 2022.
For locations not covered by official strategic maps, crowdsourced platforms and IoT monitoring networks are becoming viable alternatives for getting a rough sense of noise conditions, though their data should be treated as indicative rather than definitive. If you are making a major decision like buying a home, it is worth spending an hour at the property during peak traffic times with an inexpensive sound level meter app, calibrated against a known reference if possible, rather than relying solely on any map.