What America's First Citywide Noise Study Found in Portland

In 2025, researchers from Oregon State University and the Multnomah County Health Department published the results of something no U.S. city had done before: a systematic, citywide, health-oriented noise measurement campaign, funded by a CDC grant and published in the Journal of Exposure Science & Environmental Epidemiology. Rather than estimating noise from traffic models — the usual shortcut — they went out and measured it, continuously, across Portland, Oregon.

The team monitored 217 locations, each chosen by weighted probability sampling rather than convenience, each recording continuously for at least five days with research-grade (Class 1) sound level meters. Nine additional sites ran nonstop for a full year to confirm the short-term measurements reflected long-term reality. The instruments were networked meters logging 24 hours a day to a cloud service — the study’s own description of its method is, in essence, continuous remote monitoring.

• 78% of sites exceeded the U.S. EPA outdoor residential guideline (55 dB day-night average level, from EPA’s 1974 “Levels Document”).
• 90% of sites exceeded WHO nighttime noise guidance.
• Day-night average levels ranged from 49.6 to 86.7 dB — from library-quiet residential blocks to levels associated in the research literature with elevated long-term health risk.
• Noise followed the equity gradient: census tracts with higher incomes and a higher proportion of white residents were measurably quieter. The loudest parts of the city were its most socially vulnerable — consistent with national-scale U.S. findings.
• Machine-learning clustering revealed four distinct noise profiles — steadily quiet, steadily loud (near highways), loud-on-average near major roads, and the interesting one: moderate average with high intermittency, the loud-burst pattern found a block or two off major roads. Crucially, the profiles did not align with zoning and could flip from one block to the next.

That last finding deserves a beat: a city’s official land-use map is a poor predictor of its actual acoustic reality. The only way to know what a specific location experiences is to measure that location.

The reason this had never been done is historical: the U.S. federal government largely stopped working on noise when the EPA’s Office of Noise Abatement and Control was defunded in the early 1980s. The operative federal reference values still come from the EPA’s 1974 Levels Document. Europe went the other way — the WHO’s Environmental Noise Guidelines and decades of European cohort studies link long-term noise exposure to sleep disturbance, cardiovascular disease, and other health outcomes. As the study’s authors put it, European research ranks noise as the second-biggest environmental threat to health after air pollution — while most American communities, Portland included, simply don’t know how noisy they are.

The study was explicitly designed as a replicable framework for other cities. A follow-up is in the works: a model of noise across all of Portland at 10-meter resolution, with Multnomah County preparing a public interactive noise map.

• Continuous beats spot checks. The study’s legitimacy comes from measuring around the clock for days — the same reason a venue’s compliance record or a construction site’s property-line case is only as strong as its coverage. A one-hour spot measurement can’t see the loud-burst profile that defined whole swaths of Portland.
• Your block is not your zone. If noise profiles flip block to block, then the measurement that matters is the one at your property line, your nearest neighbor, your front-of-house — not a citywide average or a zoning assumption.
• Expect more measurement, not less. A CDC-funded, Nature-family-journal framework plus a coming public noise map normalizes the idea that communities should know their sound environment. Operators who already hold continuous, calibrated records of their own noise will be ahead of that conversation instead of reacting to it.

The Portland study’s operating model — calibrated instruments recording continuously, streaming to the cloud, analyzed over long horizons — is precisely the model Beacons packages for venues, events, construction sites, and community studies: live monitoring against schedule-aware limits, a complete timestamped record, and regulator-ready reporting. If the study’s framework is the science, Beacons is the instrument layer that makes the same kind of measurement practical outside a research grant.

Curious what continuous monitoring would look like at your location? Book a demo.

• Mowrer C. et al., “Systematic measurement and machine learning-based profile characterization of community noise in a medium-large city in the United States,” Journal of Exposure Science & Environmental Epidemiology 36, 199–210 (2026)
• Oregon State University, “Study documents unhealthy noise in Portland, provides research framework for other cities” (press release, July 2025)
• WHO, Environmental Noise Guidelines for the European Region (2018)

Health findings describe population-level associations reported in the cited research; individual risk varies. Guideline values are health-based references, not local law — limits for a specific venue or site come from its license and local ordinance.