Measuring and Estimating Noise

Because decibels are a logarithmic scale, an increase of 10 decibels (dB) is perceived as a doubling of noise volume. The number of decibels from multiple sources cannot simply be added to get the total number of decibels. If two trucks that each produce 90 dB of noise are passing the same location at the same time, they will produce a total of 93 dB, not 180 dB. As an approximation, use the following table to add two decibel amounts; the table is accurate within 1 dB of the exact value.

Addition of Two Different Decibel Values

Source: Hendricks, Rudy. Technical Noise Supplement: A Technical Supplement to the Traffic Noise Analysis Protocol. California Department of Transportation (Caltrans), October 1998.

Transportation agencies have different standards for measuring noise. For example, the Federal Aviation Administration (FAA) uses a standard called the day-night sound level (DNL), which measures sound exposures (such as a plane flying overhead) and combines them into a single decibel measure, with nighttime exposures weighted more heavily. Many federal agencies use the DNL standard to measure noise.

In contrast, the Federal Highway Administration (FHWA) and many state highway agencies, including Caltrans, use one of the following standards:

• • L₁₀, the sound level that is exceeded 10 percent of the time during a given period

  • L_eq, the total amount of sound during a given period of time averaged over that period

Caltrans offers many publications about traffic noise that provide detailed instructions for measuring noise.

Estimating Changes in Noise

Many projects, such as a new street, will increase noise for nearby residents and businesses. Other projects may shift noise from one area to another; for example, a traffic diversion project would reduce noise on streets from which traffic was diverted and increase noise on other streets.

State and federal agencies have computerized models that can estimate how much a project will change noise levels in an area. Multi-modal transportation projects often require a different model for each mode.

The Federal Highway Administration (FHWA) model, called the Traffic Noise Model® (TNM), assesses the noise impact of different types of vehicles, different pavement types, graded roadways, and other variables. Some state transportation agencies use TNM for their own projects.

In California, the Caltrans Division of Environmental Analysis has developed its own modeling program, called Sound 2000, based on the FHWA model. Unlike TNM, Sound 2000 is available for free. Caltrans also provides computer-based training videos that show how to use Sound 2000 effectively.

Monetizing Noise Costs

A number of studies have attempted to monetize traffic noise costs, that is, to calculate the dollar value that people place on a change in their exposure to traffic noise, for economic evaluation. These mostly use hedonic pricing method, which measures how a change in traffic noise affects nearby residential property values.

Most older U.S. studies indicate that traffic noise has relatively low costs, typically averaging a fraction of a cent per vehicle-mile traveled (Delucchi and Hsu 1998; FHWA 1997). However, these studies tend to be biased in various ways that underestimate traffic noise costs, particularly under urban conditions. Most U.S. studies measured the marginal noise cost of each additional highway vehicle, and so are inappropriate for evaluating the noise costs of traffic on urban area surface streets. Their noise level thresholds they used were arbitrary, the data used are often incomplete, they assumed that home buyers have accurate knowledge of noise exposure at each location, and they do not account for non-residential noise impacts (such as on businesses and pedestrians). Verhoef (1994) concludes that conventional highway noise cost studies only account for approximately 1/8th of the total traffic costs.

More recent studies tend to indicate much higher noise costs for urban driving, typically one to three cents per vehicle-mile, and even higher for noisy vehicles (motorcycles, large trucks and buses) and for night conditions. The table below compares noise cost estimates of various studies, with all values converted to 2007 U.S. dollars.

Noise Studies Summary Table – Selected Urban Values (Litman 2009)