Environmental noise caused by traffic can reduce property values. Planners,
policymakers, and legislators must look at noise damage costs caused by motor
vehicles when considering transportation options. Daniel Haling and Harry Cohen
provided a method to estimate this type of noise impact in "Residential Noise Damage
Costs Caused by Motor Vehicles" (Transportation Research Record 1559).
BACKGROUND
The majority of sounds detected by human hearing are within the range of 0 to 140
decibels (dB). The noise created by traffic normally resides in the range of 50 to 95 dB.
The effects of transportation noise are routinely measured using an A-weighted decibel
scale (designated dBA), which is useful for measuring the noise impact of a single
occurrence but not the impact of continuous noise. A frequently used measurement for
continuous noise is the equivalent sound level (Leq), known also as the energy mean
sound level. Leq includes both the intensity and length of all sounds occurring during a
given period; it indicates "the average acoustic intensity over time and is the equivalent
noise energy level of a steady, unvarying tone."
The Environmental Protection Agency has developed a measurement for a community's
exposure to noise (the average energy sound level) for a 24-hour period from midnight
to midnight. The measure of this day-night sound level, designated DNL or Ldn, is
commonly used to evaluate noise impacts on communities and residential areas.
NOISE PREDICTION MODEL / NOISE DAMAGE COST STUDIES
The most common model for estimating vehicle traffic noise levels is the Federal
Highway Administration's (FHWA's) STAMINA 2.0/OPTIMA. Derived from
long-standing research by the FHWA and the National Cooperative Highway Research
Program (NCHRP), the FHWA model "is a two-level coordinate system-based program,
based on energy-equivalent sound levels."
Studies in the 1970s "estimated that background noise in a typical urban neighborhood
was roughly 55 Ldn and that housing prices decreased by 0.2 to 0.6 percent for every
one unit increase in Ldn." A major study of noise costs conducted for the 1982 Federal
Cost Allocation Study "assumed a 0.4 percent decrease in the value of a housing unit
for each dBA (Leq) increase over a threshold value of 55 dBA."
NOISE COST CALCULATIONS
Calculating the impact of transportation noise on residential property values requires
constructing a model for estimating the value of property that includes an estimate of
traffic noise cost. One method for calculating noise impact cost is based on an
estimating procedure developed in 1981 and used in the 1982 Federal Highway Cost
Allocation Study. This procedure takes into consideration reduced residential property
values caused by noise from vehicles. It operates on the theories that people will pay
to avoid high noise levels and that housing values reflect location relative to a noisy
roadway.
The procedure for estimating noise damage uses three main components: (1) the
number of housing units affected, (2) the noise level in decibels above an established
noise threshold, and (3) the average change in property values per decibel that can be
attributed to the roadway. The number of housing units affected varies by location.
The noise emission level of vehicles changes depending on the type of vehicle, its
speed, its operating weight, and the volume of traffic on the roadway. The third
component of the calculation is constant for all housing units, based on a survey of
studies on residential property values affected by noise. Using these values, the noise
damage caused by each vehicle-kilometer can be calculated--subject to the type of
vehicle, its speed, the volume of traffic on the roadway, and the type of housing
development surrounding the roadway.
TRANSPORTATION NOISE LEVEL
Calculating the noise damage cost of a single vehicle requires estimating the noise
emission of that vehicle, as well as the noise emission of all vehicles on that segment
of road. Noise emission level estimates of single vehicles are based on two emission
equations developed by the FHWA--the first for large trucks and the second for
passenger cars and light trucks. Truck noise levels, which are significantly different
from those generated by passenger cars, are converted into noise passenger-car
equivalents (NPCEs) using factors developed through a vehicle emission equation and
a total noise level equation.
By combining transportation noise levels across vehicle classes, a composite noise
emission level for the roadway is produced. (It should be noted that decibels add
logarithmically rather than algebraically.)
The number of housing units affected by transportation noise depends on the density of
the housing population and how close the housing unit is to the roadway. Noise
distance ranges are estimated for each of the land development types shown in Table 1
below. The distance ranges are an estimated number of feet within which houses are
subject to a given noise level range. Three noise levels are established at 55-65 dBA,
65-75 dBA, and greater than 75 dBA. The noise distance ranges are labeled A, B, and
C, where C is closest to the roadway and assumed to begin at 9.14 m (30 ft), with no
housing units located closer than that to the roadway.
After noise distance ranges are estimated, housing densities are needed to calculate
the total number of housing units affected. Based on the 1981 noise cost study, Table
1 illustrates the housing densities per acre by land development type and noise
distance range. As noted earlier, previous noise impact studies estimated that housing
units lose 0.4 percent of their value for every decibel above the threshold level. The
most recent survey of housing values (1993) showed a median house value of $86,529.
Using this value annualized at a 10 percent discount rate and multiplied by the 0.4
value loss, the noise damage cost is found to be $34.61 per decibel per housing unit.
