550/9-74-009
riNUTIBI DISTRIBUTION
OF THE UNITED STATES
IS II
FUNCTION OF OUTDOOR NOISE LEVEL
JUNE 1974
o.s. Environmental Protection Agency
Washington, o.c. 20460
-------
Document is available to the public through the National Technical
Information Service, Springfield, Virginia 22151
Document is available in limited quantities through the Environmental
Protection Agency, Office of Noise Abatement and Control, Arlington,
Virginia 20460
-------
550/9-74-009
POPULATION DISTRIBUTION OF THE
UNITED STATES AS A FUNCTION OF
OUTDOOR NOISE LEVEL
June 1974
W. J. Galloway
K. McK. Eldred
M. A. Simpson
Prepared For:
U. S. Environmental Protection Agency
Office of Noise Abatement and Control
Washington, D. C. 20460
Under Contract No. 68-01-1886
This report has been approved for general availability. The contents of this
report reflect the views of the contractor, who is responsible for the facts and
the accuracy of the data presented herein, and do not necessarily reflect the
official views or policy of EPA. This report does not constitute a standard,
specification, or regulation.
-------
ACKNOWLEDGMENTS
This report has profited from the contributions of many individuals.
At Bolt Beranek and Newman, Andrew S. Harris was responsible for the
noise measurements in the eastern half of the United States; processing
of all field data was also performed under his direction. The organization,
correlation, and further analysis of processed data was undertaken by
Richard H. Muehlenbeck. In addition to these individuals, the authors
would like to thank the BBN personnel who participated in the measurement
and analysis phases of this project.
The authors are also grateful to Dr. Simone L. Yaniv, Project Officer
in EPA's Office of Noise Abatement and Control, for her guidance and
enthusiasm throughout the course of this study.
111
-------
TABLE OF CONTENTS
Section Page
ACKNOWLEDGMENTS j i i
1 INTRODUCTION 1
2 SURVEY RATIONALE 3
3 MEASUREMENT PROCEDURE 7
SITE SELECTION . 7
NOISE DATA COLLECTION AND PROCESSING . . 9
CENSUS DATA PROCESSING 9
4 SURVEY RESULTS 11
SUMMARY OF MEASURED DATA 11
RELATIONSHIPS BETWEEN VARIOUS NOISE
MEASURES 11
5 DEVELOPMENT OF MODEL FOR URBAN NOISE AND
ESTIMATE OF POPULATION EXPOSED TO NOISE AT
VARIOUS LEVELS 19
MODEL FOR PREDICTING OUTDOOR NOISE
LEVELS IN URBAN AREAS 19
ESTIMATING NATIONAL NOISE EXPOSURE ... 21
6 CONCLUSIONS 27
7 RECOMMENDATIONS FOR FUTURE STUDY 31
REFERENCES R-l
APPENDICES
A - DATA ACQUISITION AND ANALYSIS
DETAILS A-l
B - SITE DATA B-l
C - SUMMARY OF INFORMATION FROM OTHER
SOURCES C-l
D - INSIDE NOISE DATA D-l
-------
LIST OF FIGURES
Figure Page
4-1 Day /Night Sound Level as a Function of Population
Density .............................. 15
4-2 Daytime NPL as a Function of Population Density
(100 EPA Sites) ........................ 16
4-3 Nighttime NPL as a Function of Population Density
(100 EPA Sites) ....... ................. 17
4-4 Comparison of the Difference Between L^ and Ln
with the Day/Night Sound Level, Ldn ......... 18
6-1 Population of the United States Exposed to Day /Night
, Sound Levels in Excess of a Specified Value
(Excluding Freeway and Airport Noise) ........ 28
6-2 Population of the United States Exposed to Day /Night
Sound Levels in Excess of a Specified Value
(Excluding Freeway and Airport Noise) ........ 29
A-l Block Diagram of Noise Monitoring
Instrumentation A-3
A-2 Block Diagram of Data Analysis System A-4
B-l Description of Kansas City Site 1303 B-3
B-2 Hourly Noise Data for Kansas City Site 1303 .... B-4
B-3 Day/Night Data for Kansas City Site 1303 B-5
C-l Plot of LJ_ vs. p Referencing Data from Previous
Studies C-2
D-l Inside/Outside L-Equivalent Values at Boston
Site 0007 D-2
D-2 Inside/Outside L-Equivalent Values at Boston
Site 0008 D-3
VI 1
-------
LIST OF FIGURES (CONT)
Figure
D-3
D-4
D-5
D-6
D-7
D-8
D-9
D-10
D-ll
D-12
D-13
D-14
D-15
Inside/Outside L-Equivalent Values at Denver
Site 1104
Inside/Outside L-Equivalent Values at Denver
Site 1107
Inside/Outside L-Equivalent Values at Denver
Site 1110
Inside/Outside L-Equivalent Values at St. Louis
Site 1201
Inside/Outside L-Equivalent Values at St. Louis
Site 1211
Inside/Outside L-Equivalent Values at Dallas
Site 1401
Inside/Outside L-Equivalent Values at Dallas
Site 1404
Inside/Outside L-Equivalent Values at Seattle
Site 1503
Inside/Outside L-Equivalent Values at Seattle
Site 1506
Inside/Outside L-Equivalent Values at Los Angeles
Site 1603
Inside/Outside L-Equivalent Values at Los Angeles
Site 1605
Inside/Outside L-Equivalent Values at Los Angeles
Site 1606
Inside/Outside L-Equivalent Values at Los Angeles
Site 1608
Page
D-4
D-5
D-6
D-7
D-8
D-9
D-10
D-ll
D-12
D-13
D-14
D-15
D-16
vi n
-------
LIST OF TABLES
Table Page
3-1 Cities in Which Noise Measurements Were
Obtained 8
4-1 Pearson Product Moment Correlations for
Various Measures at 100 Sites ............. 13
5-1 Population Distribution in Urban Areas by Census
Tract Population Density - People Per (Mile)
Class Intervals Are "Up To" - E.G. 100 PSI
0 To 100 ............................. 22
5-2 Fractional Distribution of Peoplejfor Different
Values of L Around a Mean L ........... 23
5-3 Distribution of Urban Population at or Greater Than
a Specified L^ ........................ 25
6-1 Summary of the Number of People in the United
States Exposed to Various Levels of Outdoor Day/
Night Sound Level ...................... 30
A-l Difference in L-Equivalent Values of 10 Minute
Samples for Variable Mierophone-to-Building
Distances Relative to a Distance of 1. 8 Meters . . . A-2
C-l Data from Previous Studies ................ C-3
C-2 Comparison of Ljju and Hourly L-Equivalent Values
Between Digital Monitoring System and Analog
System Data Processing .................. C-4
IX
-------
SECTION 1
INTRODUCTION
The Noise Control Act of 1972, in part, directs the Environmental Protec-
tion Agency (EPA) to develop criteria with respect to noise, to publish informa-
tion on the levels of environmental noise requisite to protect the public health
and welfare, and to identify major sources of noise. It also directs the EPA
to propose noise emission standards, where feasible, for identified major
sources of noise. The purpose of these actions is to bring about a containment
and reduction of the noise to which the public is exposed in its overall environment.
A major consideration in judging the effectiveness of any emission standards
is the degree to which such standards will contribute to the reduction of the
general level of environmental noise in the country, and how many people will
benefit from such reduction. In order to make such determinations it is essen-
tial to have knowledge of the present values of noise exposure experienced by
the population so that the effect of noise control of a particular class of noise
producers can be assessed.
Many modes of noise exposure are incurred by an individual as he moves
through his normal activities. He receives various noise doses as he travels
from place to place, at work or recreation, and at home. Each of these modes
of exposure needs to be considered in terms of its effect on the individual's
total noise dose, as is discussed in the EPA Report on Environmental Noise
Levels. However, most people spend more total time at their place of residence
than at any other location. Thus, a basic information requirement is to describe
the present noise exposure at home for the national population.
Specific estimates of the noise exposure due to airport operations and free-
o
way traffic have been provided in the EPA Airport /Aircraft Noise Report
based on previously developed noise models for these two sources. While
-------
airports and freeways provide -two of the most noisy environments in which
people live, they impact only a small proportion of the total population. The
largest part of the population is primarily impacted by noise from other sources.
It is this noise exposure that needs to be defined.
While the noise exposures from airports and freeways have been studied
extensively, little systematic knowledge exists on the general noise environ-
ment away from these two sources. Sample data on typical noise levels at
different places ranging from wilderness to dense central city areas were
reported in the EPA Report to Congress. 3 Representative noise level measure-
ments at 20 different urban/suburban locations were reported under an Automo-
bile Manufacturers Association program.4 Some additional data of similar
nature were available in the files of Bolt Beranek and Newman Inc. (BBN).
These data serve to describe the range of noise levels to be expected in urban
areas, but do not provide the bases for extrapolation to a general population.
A detailed description of the national noise environment would require an
extensive noise survey over a wide variety of geographic locations, taking
place over an extended time period. Such a survey should be performed.
However, the urgency of having baseline information for use by EPA in meeting
its immediate objectives required a carefully planned sample noise survey that
could be used to derive a model for predicting a first order estimate of the
national noise exposure. .
This report summarizes the planning, conduct, and results of a noise
survey of 100 sites in urban areas across the continental United States and the
use of these results, coupled with other existing data, to provide an estimate
of the outdoor noise levels experienced in residential areas by the national
population. It concludes that more than 75 million people are exposed to Day/
Night* sound levels in excess of 60 decibels, while 600,000 people are exposed
to greater than 80 decibels.
* Day/Night sound level is the equivalent A-weighted sound level over a 24-
hour period, with actual sound levels between 2200 and 0700 hours increased
by 10 decibels. The symbol used for this measure is
-------
SECTION 2
SURVEY RATIONALE
The basic purpose of the survey was to assess the noise climate in residen-
tial areas primarily exposed to noise from other than airports or freeways. The
results of the survey, when used with estimates of populations exposed to noise
c s*
from airports and freeways derived from other models, ' as reported in the
o
EPA Aircraft/Airport Noise Study, were to be used to estimate the outdoor
noise exposure of the national population when at home.
Several factors seemed evident in planning the survey:
1. It was desirable to obtain data in areas where 1970 census data could
be used in conjunction with any model derivations.
2. Measurements should be taken in the various geographic regions of
the country.
3. Sites should be selected to obtain data over the widest possible range
of living conditions.
A primary consideration in the planning was the evidence from community
noise surveys that surface transportation noise is reported to be the most evi-
4 7
dent single contributor to environmental noise. ' The predominant source of
8 9
surface transportation noise is motor vehicles. ' Over a wide range of popula-
tion densities and of total populations, the number of automobiles per person is
almost constant, the ration of trucks in service to automobiles in service is
almost constant, and the usage of vehicles is directly proportional to population
density.10
The above facts lead to the hypothesis that ±he number of motor vehicles in
use per person, over a wide range of population densities, is essentially constant.
Further, if non-freeway traffic is considered, the average speed of motor
-------
vehicles in urban areas is essentially constant. ®»•*•" Based on existing traffic
noise models, ^ the noise produced by traffic is directly proportional to 10
times the logarithm of the number of vehicles flowing past a point (with constant
speed). If this is the case, then a first order estimate of the space average
noise level in a community should be proportional to 10 times the logarithm of
the population density of the community. Variations from this space average
would be expected to occur as a function of local street structure, homogeneity
of population densities, type of building structures (e. g. single family resi-
dences as compared to high rise apartments).
This hypothesis was tested against 30 random samples of existing data
where population density could be determined. Over a range of population
densities from 1000 to 70,000 per square mile, it was found that the day/night
sound level could be correlated with 10 times the logarithm of population
density per square mile with an intercept at zero population of 22 decibels.
For example, at a population density of 5000 per square mile, the approximate
average for urban areas of the United States, the expected average day/night
sound level would be 59 decibels.
Based on the above preliminary analysis, it was determined that the 100
sites to be selected for noise measurements would be chosen so that a maximum
range of population densities could be explored, that data would be acquired
only in urban areas so that census data could be employed in developing a
national model, and that data would be acquired in each EPA region to ensure
national coverage.
The measure selected for defining the noise environment was A-weighted
sound level. It was determined that, due to time and budget constraints,
measurements would be restricted to one 24-hour period per site, to be con-
ducted during weekdays only (to avoid weekend variations in traffic). In order
to provide a data base with sufficient detail for future analysis, it was decided
to obtain a continuous statistical record of A-weighted sound level over the
entire 24 hour period. The data were to be analyzed to provide a cumulative
distribution of sound level for each hourly period. Based on these data,
-------
equivalent sound level, noise pollution level, traffic noise index, and day/night
sound level were to be computed for each site.
Outside noise levels were chosen as being most characteristic of the noise
environment produced in a residential area from sources beyond the control of
the resident. Inside noise levels, while frequently affected by exterior noises,
are generally dominated by the activities inside a home. Thus, inside noise
levels are extremely dependent on the life styles of individuals and not on the
external environment. In order to provide an indication of the relationships
between inside and outside noise levels, 15 sites were chosen for which inside
noise levels would be measured simultaneously with those outside. Comparison
between the two sets of data at a site shows how little the interior noise levels
are affected by exterior noise during waking hours.
In the following sections of this report we provide a description of the actual
site selection, the results of the measurements, and the derivation of the model
for urban noise.
-------
SECTION 3
MEASUREMENT PROCEDURE
SITE SELECTION
One hundred measurement sites were chosen in fourteen cities located
throughout the ten EPA regions. The various cities were selected so as to
cover a wide range of population densities. Table 3-1 lists the cities in which
measurements were obtained, the number of sites in each city, and the average
central city population density. Within each city, specific sites were selected
in census tracts with population densities ranging from a fraction of the average
population density to several times this average density. Only census tracts
with a homogeneous population distribution were utilized.
Note that no attempt was made to provide a systematic statistical descrip-
tion of noise in any particular city. The primary purpose was to obtain samples
in as wide a range of population densities as possible, with the specific choice
of sites in any geographical area being arbitrarily picked to maximize the
difference in population densities.
Sites were also selected to provide a variation in proximity to streets of
varying traffic density. Thus, for each population density, an attempt was
made to select sites that varied between being adjacent to major traffic arteries
to being located on only lightly traveled streets. Since the primary purpose of
the survey was to obtain data in general urban residential areas away from
freeways or airports, no site was selected closer than 300 meters from a free-
way or an area where aircraft noise was significant, e.g. at an estimated
of 70 or higher.
-------
TABLE 3-1
CITIES IN WHICH NOISE MEASUREMENTS WERE OBTAINED
EPA
Region
1
2
3
4
5
6
7
8
9
10
City
Boston
New York
Washington, D. C.
Pittsburgh
Atlanta
Miami
Chicago
Dallas
Kansas City
St. Louis
Denver
San Francisco
Los Angeles
Seattle
Central City
Average Density
(per sq. mile)
13, 900
26,300
12, 300
9,400
3,800
9,800
15, 100
3,200
2,100
10, 200
5,400
15,800
6,100
6,400
Number of
Measurement
Sites
8
11
6
5
4
6
8
6
4
9
8
11
8
6
-------
NOISE DATA COLLECTION AND PROCESSING
At each measurement site, the microphone was located out-of-doors at an
appropriate listener height. Thus, for single story residences, the microphone
was typically positioned 1. 5 meters above the ground. For multi-story resi-
dences, the microphone was placed at an elevation comparable to the height
where occupants lived. In all instances, the microphone was located 1.8 meters
from the building facade. Specific heights and microphone locations for each
measurement are listed in Volume 2 of this report.
Noise levels were monitored continuously for a full 24-hour period at each
site. The A-weighted noise level was automatically distributed into bins 1.25
dB wide spanning the range from 30 to 110 dBA. This discrimination width is
an equipment limitation set by the selection of the wide dynamic range. Thus,
a narrower bin width could have been used if a narrower dynamic range was
employed. It was decided that the 1.25 dB bin width was perfectly narrow
enough for this survey and that the dynamic range was a necessity.
At the end of each hour, the accumulated contents of each bin were recorded
on magnetic tape in digital format. The data tapes were later processed by
computer to yield various statistics and noise exposure levels for each hour,
as well as for daytime and nighttime periods. Appendix A describes in greater
detail the noise measurement instrumentation and data processing procedures.
The computer output listings for each site are presented in Volume 2 (a sample
listing for one site is given in Appendix B).
CENSUS DATA PROCESSING
Population densities in people per square mile were determined for each
census tract in which noise measurements were made. Generally, this informa-
tion, or data from which this information could be derived, was supplied by
county or regional planning agencies.
As discussed above, measurements were to be obtained in homogenous
census tracts. That is, it was desirable for modeling purposes to make noise
measurements at sites having as uniform a population density as possible over
-------
an entire census tract. In a few instances, however, circumstances dictated
acquiring measurements either on the border between two census tracts or in
census tracts with a skewed population distribution, such as in an area in
which a large portion of the tract is devoted to a park. In the former instance
the two tract densities were averaged, while in the latter the population densi-
ties were adjusted to reflect the density of the actual inhabited area.
10
-------
SECTION 4
SURVEY RESULTS
SUMMARY OF MEASURED DATA
The detailed results of the noise measurements at each site, and information
on the site itself, are provided separately in Volume 2 of this report. A sample
of the data format is provided, for one site, in Appendix B of this report. These
data, all in A-weighted sound level, are presented in two formats. The first
format lists, for each of the 24 hours, the values for maximum, minimum, Leq,
NPL, TNI, SIG, L]_, L3, L5, L^Q, L2Q» LSO» L40, L50, L60, L70, L80, L90,
Lgg, Lg7, and Lgg (see Appendix B for explanation of symbols).
The second format combines the hourly data into a daytime (0700-2200 hrs)
period and a nighttime (2200-0700 hrs) period and lists the same measures as
above for these two periods. A separate listing is provided for the combined
24 hour period with weightings of 0, 8, 10, and 12 decibels added to the night-
time noise levels. Note that L-^ is this 24-hour value using the 10 dB nighttime
level weighting factor. Finally, cumulative distributions for the daytime and
nighttime periods are also indicated on a probability scale. Recall that a straight
line on this scale indicates a statistically normal distribution.
RELATIONSHIPS BETWEEN VARIOUS NOISE MEASURES
The data acquired in the survey were primarily intended for use in develop-
ing a model for estimating outdoor noise exposures for urban populations. Their
use in developing such a model is discussed in Section 5 of this report. In addi-
tion to this use, the data provide an opportunity to explore the relationships
between various noise measures. For example, how well does Lj or L^0 predict
a maximum value or the relationships between L and other measures. Fur-
eq
ther, it was hypothesized that noise environments should be proportional to the
logarithm of population density (logp), so it is of interest to examine how
various measures are correlated to log p.
11
-------
The data for the 100 sites, separated by daytime and nighttime values per
site, were used to compute a number of different Pearson product moment
correlations. These results are listed in Table 4-1. These correlations show
a number of interesting relationships:
1. LJ, LJQ, 1,50 are highly correlated (i. e. , of the order of 0. 9 or higher)
withLeq.
2. The maximum value that occurs at any 1/8 second interval in the 9
hours of nighttime or 15 hours of daytime is not nearly as well correlated
with the other measures, as might be expected.
3. LIO» L50» ^90 an<^ ^dn are a^ correlated at greater than 0. 7 with
log p, and the slope of the accompanying regression line is 10 plus or
minus less than 2. On the other hand NPL is considerably less predict-
able by log p, having a correlation coefficient of 0.458 during daytime
and 0. 642 at nighttime.
4. The addition of the 30 data points previously available to the data from
the 100 sites does not significantly affect the correlation of the 100 sites
alone with log p.
In addition to the correlation computations described above, it is of interest
to examine the scattergrams for several of the measures. In Reference 1 the
measure L^ is used to define goals for environmental noise levels. The quantity
NPL has also been advocated for this purpose. The scattergrams for L^ and
NPL, calculated separately for day and night periods, are plotted against popula-
tion density on Figures 4-1 to 4-3. In Figure 4-4 the difference between L(j and
Lfl is plotted against L^.
12
-------
TABLE 4-1
PEARSON PRODUCT MOMENT CORRELATIONS FOR
VARIOUS MEASURES AT 100 SITES
J \
y
max
max
LI
L10
L90
L10
max
Ll
L10
L50
L1
JL
L10
Leg
L90
L
NPL
+ a2
X
Ll
L10
J_ V/
L10
L50
L50
V V
L90
L -
Leq
L
Leq
logp
logp
logp
logp
logp
logp
r =
Time
Period
day
night
day
night
day
night
day
night
day
night
day
night
day
night
day
night
day
night
day
night
day
night
day
night
day
night
day
night
day
night
day
night
correlation coefficient Se = standard error
of estimate
No. of
al a2 r se Samples
.86
.97
.71
.75
.84
.80
.82
.92
.98
.96
.74
.77
.93
1.00
.92
.97
1.02
1.10
1.10
.98
7.53
9.38
9.12
10.44
11.16
9.12
11.22
8.13
7.90
9.23
5.68
10.82
34.4
23.7
51.8
44.6
19.5
20.4
17.5
10.6
-3.9
-2.5
25.7
21.4
39.3
31.4
15.3
11.7
-0.3
-5.0
-13.2
-5.1
40.7
26.6
25.0
12.8
8.8
11.5
3.6
11.0
29.1
17.4
51.7
23.1
.637
.776
.588
.682
.936
.904
.937
.907
.960
.937
.860
.774
.712
.790
.954
.951
.947
.957
. 896
.863
.683
.725
.737
.714
.792
.631
.779
.550
.687
.724
.458
.642
5.6
4.9
5.9
5.7
1.9
2.7
2.1
3.0
2.0
2.5
3.1
4.5
5.1
4.8
1.6
1.9
1.9
2.1
3.0
3.5
3.9
4.3
4.1
5.0
4.2
5.4
4.4
6.0
4.1
4.3
5.3
6.3
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
13
-------
TABLE 4-1 (Cont'd)
y = a1x + a2
y
Ldn
Ldn
L^-Ln
X
logp
logp
L/i
r =
Time
Period
24 hrs
24 hrs
24 hrs
correlation coefficient Se = standard error
of estimate
al
8.33
9.00
-0.12
a2
29.4
25.8
13.3
r
.722
.723
.188
Se
3.9
4.2
3.2
No. of
Samples
100
130 *
100
*Includes 30 sites from previous studies
14
-------
100
;
J!
1
-,
E 20
I
-8 10
B
I
£-
I
£
CL
£
• CURRENT STUDY
0 PREVIOUS STUDIES
•If
10 log p + 22 dB
9 log p + 26 dB
45 o 50 ° 55 60 65 70
Day/Night Sound Level, d3
'5
i
Figure 4-1. Day /Night Sound Level as a Function of Population Density
15
-------
55
70 75 80
NPL for Daytime Period, dB
:5
90
17
Figure 4-2. Daytime NPL as a Function of Population Density (100 EPA Sites)
16
-------
2.6 53.7
2.4 51.5
41.9 92.6
65 70 75 80
NPL for Nighttime Period, dB
9C
I
Figure 4-3. Nighttime NPL as a Function of Population Density (100 EPA Sites)
17
-------
20
18
!6
M
JJ
I I
| J 8
I I
-2
o
0
o
o
o
o
0°
o
o
0 °
0 O Oft
«
o
o
o
o
o
0
00
<
>
o
0
o
o
o
o
o o
o
OQo oo
°8 o
o
Q, O
° 0°
°o° , o
o
o
O
0
° 0
°0
o
o
o
o
o
($>
o
0
o
o
0
O rt
oo
0 0
o
0
o
o
50 55
60
70 75
80
Day/Night Sound Level, L , in dB
on
Figure 4-4. Comparison of the Difference Between L, and L with the Day/Night
Sound Level, L
dn
18
-------
SECTION 5
DEVELOPMENT OF MODEL FOR URBAN NOISE AND ESTIMATE OF
POPULATION EXPOSED TO NOISE AT VARIOUS LEVELS
MODEL FOR PREDICTING OUTDOOR NOISE LEVELS IN URBAN
AREAS
In the previous section the results of the noise measurements were corre-
lated with population densities. It was found that L and NPL were not nearly
TllflrX
as well correlated with population density as LJQ, 1,50, 1.90 or L^. It was also
found that these latter measures are between themselves highly correlated. Of
these measures, L^ was picked by EPA in Reference 1 as the measure most
useful among these measures for relating environmental noise levels to human
response. On these bases it seems reasonable to develop a model for predict-
ing L , from population density.
It was shown in the previous section that, using the 100 EPA survey site
data, plus the data from the 30 other sites available, a correlation of 0. 723
was found with log p. The regression line computed for this relationship has
the form:
= 9. 00 log1Q p + 25. 8 (decibels) for p >1.
The original hypothesis that urban noise levels would vary with population
density was based on the assumptions that vehicular traffic density varies
directly with population density, and further that average traffic speed for urban
areas is relatively speed independent. Models for traffic noise** predict that
average noise level varies as 10 log., Q of the mean vehicle flow rate. A first
order estimate of urban noise related to population density would thus expect
average noise levels to vary as 10 log-0 p . Using a prediction equation for
Ldn Siven by:
19
-------
L^ = 10 Iog10 P + 22 (decibel) p > 1
is statistically insignificant (at the 0.05 level) from the regression line found
from the site measurements. For example, the difference between these two
equations is less than one decibel for values of p between 1 and 100 thousand
per square mile. This is shown on Figure 4-1 where the solid line is 10 log-, QP
and the dashed line is 9 Iog10 p . For simplicity in analysis the 10 Iog1np
relationship is used in the following development.
It was hoped in the initial planning for the survey that some information
would be obtained to allow a systematic description of the deviations from the
average correlation between noise level and population density to be derived in
terms of city structure or roadway configuration. Evidence was obtained to
indicate that such considerations do explain levels that are higher or lower than
the average, but no systematic structure seemed developable from the data such
that its use could be incorporated in a model at this time. Much more detailed
exploration of individual community sites will be required before such refinements
can be justified in any urban noise model.
Examination of the scatter in the data in Figure 4-1 indicates that the only
reasonable estimate for variation of level at a fixed p is to assume that noise
levels are normally distributed at each value of p, with the distribution being
characterized by a standard deviation of the same order as the standard error
obtained from the correlation computation. The standard error computed for
the regression line is 4.19 decibels. The standard deviations of L^ for a
series of values at p vary from 3.56 to 5.12, with an average of 4.01. For
model purposes, one can assume a standard deviation in L^, at any value of
p, of 4 decibels.
With a means for predicting an average value of L^ for any population
density, and an assumption that the distribution of L^ around the average is
normal, with a standard deviation of 4 decibels, all the steps necessary to
estimate outdoor day/night sound level are in hand. These factors can be com-
bined by using the standard format for a normal probability distribution function
to express a distribution function for L as a function of p as:
20
-------
(L, -L, )2
* an dir
Ldn ( P ) = ]= e 32 (decibels)
4/2TT
Where L~j^ = 10 log.,Q p + 22, the standard deviation, cr , is 4 decibels, and
p is in number of people per square mile.
ESTIMATING NATIONAL NOISE EXPOSURE
An estimate of the number of people exposed to different values of L(jn
can be generated if the number of people at each value of p is known. The
1970 census provides the population by census tract for the more than 35, 000
census tracts that account for the 134 million people living in areas defined as
"urban" in the United States. The areas of these tracts, however, are not
available directly in Census Bureau publications. It was learned that the basic
data necessary to compute a distribution function of population as a function of
population density was available in computer readable format at National Planning
Data Corporation. Arrangements were made with this organization to generate
the required distribution function. These data were provided as a cumulative
population by census tract population density.
The distribution function of population was then stratified to provide the
cumulative distribution of population in successive population density increments
of one decibel on a 10 log p basis. That is, the total population in each suc-
cessive increment from 100 or less people per square mile up to the maximum
of greater than 200,000 people per square mile was determined with each
increment being 1.26 times the previous .increment. This stratified distribu-
tion is tabulated in Table 5-1.
The number of people exposed to different noise levels was determined
for each increment in population density by assuming a distribution of noise
levels according to the model previously described in this section as applied
to the distribution of total population as a function of population density. The
number of people at each level was assigned according to the approximation
to a normal distribution listed in Table 5-2.
21
-------
TABLE 5-1
POPULATION DISTRIBUTION IN URBAN AREAS BY
CENSUS TRACT POPULATION DENSITY - PEOPLE PER (MILE)2
CLASS INTERVALS ARE "UP TO" - E.G. 100 IS 0 TO 100
p
100
126
159
200
252
318
400
504
635
800
1,000
1,260
1,600
2,017
2,541
3,200
4,000
5, 071
6,384
7,943
10, 000
12, 589
15, 848
20, 000
25, 178
31, 697
40, 000
50, 356
63, 393
79, 806
100, 000
125, 890
158,483
200, 000
> 200, 000
10 log p
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
%Pop.
5.00
6.00
7.00
8.00
9.00
11.00
12.00
13.85
15.00
17.00
19.70
22.01
25.00
28.53
32.40
37.00
44.11
51.01
59.01
67.00
75.38
81.05
85.00
88.40
91.08
93.05
94.65
95.94
96.96
97.69
98.47
99. 15
99.68
99.92
100. 00
A%
5.00
1.00
1.00
1.00
1.00
2.00
1.00
1.85
1.15
2.00
2.70
2.31
2.99
3.53
3.87
4.60
7.11
6.90
8.00
7.99
8.38
5.67
3.95
3.40
2.68
1.97
1.60
1.29
1.02
0.73
0.78
0.68
0.53
0.24
0.08
Total Population Counted: 134,089,789
22
-------
TABLE 5-2
FRACTIONAL DISTRIBUTION OF PEOPLE FOR
DIFFERENT VALUES OF
AROUND A MEAN L
dn
dB re L^
-7
-6
-5
-4
-3
_0
-1
0
+1
+2
+3
+4
+5
4£
+7
+8
% of Population
2.2
2.3
3.5
5.5
7
8.5
10
11
11
10
8.5
7
5.5
3.5
2.3
2.2
23
-------
The population distribution of Table 5-1 was then used, at each increment
in p , to generate the number of people exposed to different values of L^ by
applying the distribution of Table 5-2. For example, the increment in total
population between population densities of 3200 to 4000 as listed in Table 5-1
is 7. 11% of 134, 089, 789 people, or 9, 533, 720 people. Of these, 11%, or
1, 048,000, were assigned the L^ value for the mean,
= 10 log 4000 +22 = 58),
11% were assigned to 59 dB, 10% to 60 dB, 8. 5% to 61 dB, etc.
The final distribution of population at each noise level was then obtained
by summing the fractional populations assigned to each noise level over all
the values of p . The resulting values provide the incremental portion of the
population exposed to various values of L^, and cumulative distributions of
the total population either exposed to levels greater than some specific value
of I^jn or exposed to less than a specified value of L^. Of primary interest
is the total number of people exposed to day /night sound levels in excess of a
specified value. This distribution is provided in Table 5-3.
24
-------
TABLE 5-3
DISTRIBUTION OF URBAN POPULATION AT OR
GREATER THAN A SPECIFIED
Ldn
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
"58
Cum N
(in millions
of people)
134. 090
133. 942
133.758
133.463
132.987
132.337
131. 463
130.373
129. 040
127.528
125.872
124. 085
122. 187
120. 147
117.983
115. 642
113.011
110. 116
106. 803
102.975
98. 544
93. 427
87.665
81.237
74. 222
Ldn
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
Cum N
(in millions
of people)
66. 738
58. 997
61. 234
43. 668
36. 542
30. 061
24. 320
19. 352
15. 200
11.791
9.046
6.853
5.155
3.826
2.776
1.963
1.347
0.889
0.559
.332
.187
.093
.039
.012
.002
.0
25
-------
SECTION 6
CONCLUSIONS
The measurement program and subsequent analysis reported here provide
a first order estimate of the urban population of the United States (not exposed
to airport or freeway noise) exposed to different values of outdoor day/night
sound level. In the previous report, EPA has provided an estimate of the num-
2
ber of people primarily exposed to various levels of airport and freeway noise.
The data from this report and that of Reference 2 can be combined to provide
an estimate of the total outdoor environmental noise exposure for the urban
residents of the country.
It is reasonable to assume that the rural population of the country is
generally exposed to lower values of noise outdoors at home than the urban
population. While no data are available to assess the noise environment of
the rural population, extension of the model developed here would lead to the
assumption that even at population densities of zero per square mile, the
average noise level will be of the order of 22 decibels or greater due to wind,
rain, and other natural sounds. Thus an extrapolation from the 134 million
urban population to the slightly over 200 million total population would say
that all people are exposed to average sound levels greater than 20 decibels.
These considerations lead to the distribution curves plotted in Figures
6-1 and 6-2. These figures show the estimate of the number of people in the
country exposed to various values of L(jn, exclusive of those whose outdoor
residential noise environment is dominated by freeway or airport noise. In
Figure 6-1 these data are plotted on a linear scale of population. In order to
more easily determine from the plotted data the number of people exposed to
higher noise levels, the same data of Figure 6-1 have been replotted on a
logarithmic scale of population on Figure 6-2.
27
-------
Rural Estimate
Added to Urban Model
20 30 40 50 60 70 80 90
ry
Outdoor Day/Night Sound Level, dB re 20 JJL N/m
Figure 6-1. Population of the United States Exposed to Day/Night Sound
Levels in Excess of a Specified Value (Excluding Freeway and Airport Noise)
28
-------
Rural Estimate
Added to Urban Model
0.01
30 40 50 60 70 2 8°
Outdoor Day/Night Sound Level, dB re 20 JJL IM/m
Figure 6-2. Population of the United States Exposed to Day/Night Sound
Levels in Excess of a Specified Value (Excluding Freeway and Airport Noise)
29
-------
A total estimate of residential population exposed to various values of
outdoor noise levels must also include those people heavily affected by freeway
and airport noise. Estimates of these populations were provided in Reference 2.
These results are combined with the estimates developed in this report in
Table 6-1. The reader should be aware that these estimates have been made
on different bases with different assumptions on population densities, noise
levels, and modeling methods. While the authors believe the estimates are as
reasonable as can be provided with the information available, any policy deci-
sions influenced by these data should take into account the adequacy of their
derivation.
TABLE 6-1
SUMMARY OF THE NUMBER OF PEOPLE IN THE
UNITED STATES EXPOSED TO VARIOUS LEVELS OF
OUTDOOR DAY/NIGHT SOUND LEVEL
Ldn Exceeds
60
65
70
75
80
Number of People in Millions
Urban
Traffic
59.0
24.3
6.9
1.3
0.1
Freeway
Traffic
3.1
2.5
1.9
0.9
0.3
Aircraft
Operations
16.0
7.5
3.4
1.5
0.2
Total
78.1
34.3
12.2
3.7
0.6
30
-------
SECTION 7
RECOMMENDATIONS FOR FUTURE STUDY
The measurement program described in this report has provided a pre-
liminary insight on the magnitude and variability of urban noise for a wide
range of population densities. These data also provide a new basis for the
design of a national noise survey. With this in mind, the following recommen-
dations are made for future investigation.
1. The strong correlation of noise level with population density suggests
that a random sampling of sites for urban noise measurements should
be stratified in proportion to the population densities of the area to be
studied.
2. A specific program should be conducted at several fixed levels of popu-
lation density to determine if the assumed normality of noise level dis-
tribution is correct. This program should examine in detail the spatial
variation in noise levels, for a fixed population density, in order to
improve the knowledge of the effect of local variation in street structure
and traffic flow on noise levels.
3. The variation in noise level over weekly periods should be examined to
determine the difference, if any, in noise exposure on weekends as com-
pared to week days.
4. The wide variation in noise levels throughout the urban population should
engender quite different attitudes in residents exposed to these noises.
A social survey should be performed to investigate these differences for
low, medium, and high population density environments.
31
-------
REFERENCES
1. "Information on Levels of Environmental Noise Requisite to Protect
Public Health and Welfare with an Adequate Margin of Safety," EPA
550/9-74-004, March 1974.
2. "Report on Aircraft/Airport Noise, Report of the Administrator of the
Environmental Protection Agency to the Committee of Public Works,
U.S. Senate," July 1973.
3. "Report to the President and Congress on Noise," EPA NRC500.1,
December 31, 1971.
4. Galloway, W. J., "Motor Vehicle Noise: Identification and Analysis of
Situations Contributing to Annoyance," BBN Report 2082, June 1971.
s
5. Galloway, W. J., "Predicting the Reduction in Noise Exposure Around
Airports, " Inter-Noise 72 Proceedings, Washington, B.C., October 4-6,
1972.
6. Gordon, C. G., Galloway, W. J., Kugler, B. A., and Nelson, D. L.,
"Highway Noise: A Design Guide for Highway Engineers, "NCHRP 117,
1971.
7. OECD, "Urban Traffic Noise: Strategy for an Improved Environment,"
October 1969.
8. Wyle Laboratories, "Transportation Noise and Noise from Equipment
Powered by Internal Combustion Engines," EPA NTID300.13, December
31, 1971.
9. Galloway, W. J., and Knapton, D. A., "Trends in Road Vehicle Noise,"
BBN Report 2081, June 1971.
10. Wilbur Smith and Associates, "Motor Trucks in the Metropolis,"
August 1969.
R-l
-------
APPENDIX A
DATA ACQUISITION AND ANALYSIS DETAILS
-------
APPENDIX A
DATA ACQUISITION AND ANALYSIS DETAILS
Measurement of the outdoor noise environment at each of the one hundred
sites was accomplished using portable, unmanned monitoring systems developed
by BBN. Each system consists, basically, of a microphone and preamplifier,
a special noise monitoring unit, and a digital tape recorder. Figure A-l shows
a block diagram of this instrumentation. As can be seen in the diagram, the
noise signal is A-weighted, converted to a digital format, and the output distrib-
uted to one of sixty-four counters, each 1.25 decibels wide. The noise environ-
ment is sampled at a rate of eight times per second. Once an hour, the contents
of the sixty-four counters, as well as the time of day, are recorded onto the
digital tape cassette. At the beginning and conclusion of each day's measure-
ments, a calibration signal was recorded on the tape cassette. This signal was
analyzed during data reduction as a check on system performance and as a
calibration standard for the noise recordings.
Each tape cassette was processed by a time-shared computer (see Figure
A-2), which produced an output listing tabulating hourly and daily statistics and
noise exposure levels as indicated in Appendix B.
Noise data were acquired for a complete 24-hour period at each site.
However, data for occasional hours were deleted due to the occurrence of
adverse weather conditions (excessive wind, rain or thunder), non-typical
noise intrusions (e. g. bulldozers, lawnmowers, garbage trucks very close to
the microphone), or equipment malfunction. The omission of these data will
not significantly influence the daily noise levels at a particular site.
As discussed in the report text, the microphone location was standardized
at 1.8 meters from the facade of the building. In order to provide rationale for
this distance, a series of measurements were performed in which simultaneous
A-l
-------
sets of ten minute noise samples were obtained with one microphone positioned
1.8 meters from a building facade and another microphone placed at various
distances from the facade. These measurements were performed for a quiet
residential situation as well as in an environment in which there were frequent
passages of automobiles and trucks. Table A-l shows the differences between
the two microphones for both sets of measurements. Note that at the "noisy"
site, as the microphone moves closer to the street, the noise levels are higher,
as would be expected. Further, as the microphone is placed closer to the facade
of the building, reflections result in higher levels as well. At the quiet site,
microphone location is not significant. Thus, the choice of 1.8 meters as a
measurement distance appears, from this limited set of measurements, to be
quite appropriate.
TABLE A-l
DIFFERENCE IN L-EQUTVALENT VALUES OF 10 MINUTE
SAMPLES FOR VARIABLE MICROPHONE-TO-BUILDING
DISTANCES RELATIVE TO A DISTANCE OF 1. 8 METERS
Distance d from
Building Facade,
in meters
0.9
3.6
5.4
7.2
Le (d) - Le (1. 8m.), in
Site A
1.6
1.2
2.2
2.6
dB
Site B
-0.4
-0.8
-0.4
-0.4
Note: The noise environment at Site A was dominated by car and truck passages,
while Site B was located in a "quiet" residential area.
A-2
-------
\ Windscreen
GR 1560-P5
One-Inch Microphone
GR 1560- P-40 or P42
Microphone Preamplifier
Calibrator
GR 1562 A
NOISE MONITORING UNIT BBN 704
Input
Amplifier
Digital
Clock
Annotation
Keyboard
D
A-Welghtlng
Network
Log
Converter
Analog to
Digital
Converter
64
Counters
J_
Switch
Digital Cassette
- Recorder
MSI 2001
Figure A-l. Block Diagram of Noise Monitoring Instrumentation
A-3
-------
CASSETTE
TAPE
RECORDER
MODEM
CONTROL
TELETYPE
TIME SHARED
DEC POP-10
COMPUTER
PRINTED
OUTPUT
Figure A-2. Block Diagram of Data Analysis System
-------
APPENDIX B
SITE DATA
-------
APPENDIX B
SITE DATA
(Note: Appendix B is presented in full as a separate volume
[ Volume 2 ]. The text and a sample of the site descriptive
data and computer output listings are reproduced here for
reference.)
This appendix provides a description of each of the 100 measurement
sites utilized during this project. The computer output listings for each of the
sites are also presented. Data are presented by city, in alphabetical order.
Figures B-l through B-14 show maps of the fourteen cities in which
noise measurements were obtained. On each map the specific measurement
locations are indicated.
Figures B-15 through B-114 provide data for each of the sites. The
first page of each figure, labeled Figure B-xx(a) provides a physical descrip-
tion of the site. A photo and vicinity map are shown, and the address, popula-
tion density, and measured L^ value are given. Also listed are various param-
eters of the traffic flow in the general vicinity of the site. The street on
which the site is located and the street in the vicinity of the site which most
contributes to the noise environment of the site are both classified into one
of four categories: freeway, arterial, collector, and local. Also indicated
are the types of vehicles that traverse these streets. Noise sources other than
traffic that affect the noise environment at the site are also listed.
The second page of the figure, labeled Figure B-xx(b), lists various
statistics and noise levels for each hour of the day. Tabulated are the maximum
and minimum values occurring during the hour, the noise pollution level (NPL),
the standard deviation (SIG) of the distribution of levels occurring during the
hour, the L-equivalent level (Lea)» and the traffic noise index (TNI). Various
percentile levels ranging from L^ to L are also listed.
B-l
-------
Similar noise measures are tabulated for the daytime (0700-2200 hrs.)
and the nighttime (2200-0700 hrs.) periods on the final page, labeled Figure
B-xx(c). Plotted at the top of this page is the distribution of levels for the
daytime and nighttime periods. Also, the weighted 24-hour L-equivalent value,
with weighting factors of 0, 8, 10, and 12 decibels for the nighttime periods,
are listed. Note that the weighted L-equivalent value for a weighting factor of
10 decibels is the day/night sound level
B-2
-------
Ldn '
P :
60.0
2,700
dB
people
sq.mi .
SITE:
1303
Address: 5030 North Park
Kansas City North, Mo.
Microphone Height: 1.5 ID.
Distance to Curb: 14.0 m.
ROADWAY TYPE
At Site: Local
Vicinity: Arterial
TRAFFIC
At Site: Cars
Vicinity: Cars, Trucks
OTHER NOISE SOURCES:
Figure B-l. Description of Kansas City Site 1303
(Figure B-51(a) in Volume 2)
B-3
-------
SITE * 1303
HB MAX HUT
H?l SIS
LEO
T8I L1
13
L5
X.10 1.20
1
2
3
4
5
6
7
8
9
10
11
12
13
1U
15
16
17
18
19
20
21
22
23
24
68.8
68.8
71.3
62.5
76.3
68.8
76.3
70.0
71.3
81.3
70.0
83.8
82.5
77.5
76.3
75.0
81.3
86.3
98.8
92.5
93.8
90.0
65.0
67.5
42.5
42.5
41.3
43.0
U3.0
U3.0
«2.5
42.5
42.5
45.0
43.8
43.8
43.8
45.0
43.8
43.8
43.8
47.5
45.0
46.3
46.3
46.3
46.3
43.8
55'. 0
50'. 5
5«",8
50.0
54'. 0
54'. 0
62'. 3
56.7
57.8
67'. 5
56.7
63.5
65.5
63". 3
57". 8
60'. 0
60.9
70-. 1
84'. 5
79'. n
86'.0
66'. 6
53'. 7
52.8
2.4
1.6
2.8
1.8
2.7
2.8
3.7
2.7
2.8
4.2
2.4
3.5
3.9
3.7
2.6
2.9
3.2
4.4
7.0
5.9
7.2
3.7
1.4
1.6
48.9
46,4
47.7
45.3
47.0
46.8
52.9
49,9
50.6
56.9
50.7
54.6
55.6
53.9
51.0
52,5
52,7
58.9
66.6
63.9
6 7', 6
57.2
50.1
48.7
29.4
26.3
28.5
26.9
25.9
34.3
37.0
37.1
36.1
40.8
33.0
46.3
40.1
43.2
36.1
34.2
39.6
51.6
84.9
72.3
92.2
43.5
29.1
28.9
58.0
51.2
60.6
52.8
58.7
56.9
65.6
59.7
60.8
70.0
60.9
63.8
65. U
66.1
58.6
64.0
63.2
70.8
77.6
75.9
79.5
68.3
55.9
55.6
54.8
48.4
51.4
48.2
49,6
51.1
58.6
55.2
56,2
64,9
53.6
58,8
59,4
62.0
52.4
56.7
57.0
65.5
72.4
69.4
75.4
62,2
53.8
51.5
52.2
47.5
48.4
47.0
46.6
49.6
53.5
53.3
54.0
59.7
52.0
56.9
55.9
58.7
51.4
55.3
54.3
62.3
69.6
66.3
73.0
59.2
52.6
50.4
48.9
•7.2
46.8
»«.1
115,3
<»7,7
50,6
S1.2
S1.3
53,7
50.9
54.9
31.7
54.1
50,8
51.9
52.2
57.4
65.3
62.5
69.0
55.2
51.1
49.7
UI.3
46.4
US. 9
45.6
44.6
46.0
48.7
49.7
49.8
51.3
49.9
53.4
49.9
50.7
49.9
50.4
50.9
53.6
60.3
SB. 6
64.6
52.4
50.1
48.9
HH 130 L4C Lbf> 1,60 1,70 L80 L90 L95 L97
L99
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
U7.9
46.1
45.4
45.0
44.1
U5.0
48.0
48.7
49. 1
50.7
49.5
52.3
49.1
49.8
49.2
49.8
50.0
52.4
57.1
56.1
61.1
51.8
49.8
48.5
47.5
45.8
45.0
44. 8
43.6
44.5
U7.4
48.3
48.6
50.1
49.1
51.0
48.5
49.5
48.5
49.6
49.4
51.8
54.6
54.3
58.6
51.2
49.6
48.3
47.2
45.5
44.5
44'. 5
43.3
43'. 9
47.0
47.8
U8.2
49'. 7
48.7
49.9
48. 0
49.1
4 7'. 6
49'. 3
48.8
5V.2
52'. 7
52'. 9
5 6'. 5
5(9'. 8
49'. 4
48'. 0
46.9
45.3
44.1
44.2
43. 0
43. &
46.6
47.4
47.8
49.4
48.3
49.2
47.6
48.8
47.1
49.0
48.1
50.8
51.3
51.7
54'. 9
50.4
49.2
47.8
46.6
45.0
43.7
44.0
42.7
43.1
46.2
47.0
47,4
49.0
48.0
48.6
46.9
48.5
46.7
48.8
47,4
50,4
50,3
50.8
53.5
50.0
49,1
47,5
46.3
4U.5
43.3
43.6
42.3
42.8
45.7
46.6
46.9
48.7
47.6
48. ,2
46.2
48.1
46.3
48.3
46.9
50.0
49.5
50.0
52.3
49.5
48.9
47.0
45.3
44.1
42.8
42. b
41.7
42.2
45.2
4i.9
46.3
48.0
46.9
47.7
45.6
47.8
45.7
47.8
46.4
49.4
48.8
49.2
51.3
49.1
48.4
46.6
44,6
43,9
42,6
41,9
41,4
41,7
44,6
45,4
45.6
47,7
46.5
47.5
45,2
47,6
45.3
47,5
45,8
49,0
48,0
48,8
50.3
48,8
47,9
46,4
44.3
43.8
42.5
41.6
41.3
41.5
44.2
45.1
45.3
47.6
46.3
46.9
45.1
47.6
45.2
46.8
45.5
48.9
47.6
48.4
49.8
48.7
47.8
46.3
43,9
43,0
41,8
41,4
40.7
41.3
43,9
44,4
44,6
46,8
45. 5
46.3
44,7
47.2
45.0
US, 6
43,1
08.8
46,8
47,8
48.9
47.9
47,6
US.7
Figure B-2. Hourly Noise Data for Kansas City Site 1303
(Figure B-51(b) in Volume 2)
B-4
-------
tin 0 130J
030 0SS 040 045 050 055 060 065 070 075 080 085 090 095 100 105 110
»»t»t»t»»tt*t»t»t
•t
90
70
50
30
10
r •
i *
« i
t i
• t
• •
'• x
t»t»tttt»tt»»»»tt
030 035 0U0 045 PS? 055 060 065 070 075 080 085 090 095 100 105 110
CUMULATIVE DISTRIBUTION
DAY '.''. NIGHT ",", WEIGHTING WBIGKJBD leq
RAX 98.8 76.3 0. 58,7
KIN 42.5 40.0 8. 59.5
RP1 74.0 56.7 10. 60.0
SIS 5.2 3.1 12. 60.7
LEQ 60.6 48,8
TRI 60.u 39.6
11 71.9 57.6
13 66.0 52.5
15 62.7 50,8
110 57.8 49.6
120 53.3 48.5
130 51.0 47.6
140 50.5 46.7
150 U9.7 46.0
160 49.2 45.2
170 48.6 44.5
110 07.9 43.9
110 06.9 42.9
IIS 06.2 42.0
197 05.7 41.9
!•• 05.2 01.0
Figure B-3. Day/Night Noise Data for Kansas City Site 1303
(Figure B-51(c) in Volume 2)
B-5
-------
APPENDIX C
SUMMARY OF INFORMATION FROM OTHER SOURCES
-------
APPENDIX C
SUMMARY OF INFORMATION FROM OTHER SOURCES
Data from noise measurements made at thirty sites during previous studies
have been included in our analysis (see Figure 4-1). Figure C-l provides a ref-
erence for each of these measurement points. The values of L(jn and population
density for each of these sites are tabulated in Table C-l.
For the most part, noise data obtained at these sites were acquired by
recording the noise level for ten minutes out of each hour on an analog tape
recorder, with subsequent processing of the recorded signal through a distribu-
tion analyzer. The statistics resulting from this analysis were assumed to
apply for the entire hour during which the fen minute sample occurred. Typi-
cally, the window size of the distribution analyzer was either 2.5 or 5 dB wide.
In order to evaluate whether this method of acquiring noise data for a
24-hour period would result in different L^n values from the method which we
utilized in this study, described in Appendix A, the recorded noise samples for
two of the positions were played back through the BBN monitoring equipment
and analyzed as though they were recorded in the field. Table C-2 compares
the L-equivalent value determined by the monitoring instrumentation with the
level determined originally using the analog system. While differences in
individual hours range up to 3 dB, the average difference in hourly levels are
0.4 and 0. 2 dB for the two sites, and in terms of L, comparisons, the differ-
ences are of the same magnitude.
C-l
-------
100
50
D
E
I 20
sr
a.
"a.
i
a.
*o
•5 10
c
J
"s
o
O
g 5
o
|
,£
2
C
:
SFF T;
APPEN
—
—
—
—
—
a<
VBLE C-l AN
DIX C REFER
do
dq.
»
°b
od
°d
0
. c
p
ENCES
°d
do
•
"
oa •
•
• •
ol*
0
"
*
/ d
1
d
oa
.
,
• • '.
••
•
•a
-4a
*
d , °d
*•
.
S
•
'..
od.
.
"
•
*
•
"°b.
Dd
*
•
•
•
•
45 °C 50 Co 55
60 65 70
Day/Night Sound Level, d3
B
Figure C-l. Plot of L, vs. p Referencing Data from Previous Studies
C-2
-------
TABLE C-l
DATA FROM PREVIOUS STUDIES
City
Los Angeles
Boston
Detroit
Portland
Hawaii
Los Angeles
Reference
a
a
a
b
c
d
Site
1
3
4
6
11
18
24
9
14
15
17
23
1
8
9
1, 2, 3*
4
5,6,7*
8,9,10*
11
2
3
4
9,10*
G
J
L
M
N
O
Ldn'dB
49.8
52.7
60.6
54.8
52.9
56.7
62.3
57.9
68.1
70.2
63.9
63.3
57.9
55.6
56.8
58.5
65.0
57.5
53.5
66.5
44.5
54.5
53.6
47.5
60.0
62.0
59.7
57.3
50.9
55.2
people
P' sq. mile
2,500
4,000
12,500
15,900
12,900
10,200
11,800
31,700
40, 600
57,000
45,400
7,600
6,100
8,800
5,800
6,300
6,300
4,200
6,700
6,700
1,300
1,300
950
950
7,600
11,000
21,400
3,400
2,500
2,500
*Note: Ldn values were averaged for these sites, which are located within a
few blocks of each other and have similar noise environments.
C-3
-------
TABLE C-2
COMPARISON OF L^ AND HOURLY L-EQUIVALENT VALUES
BETWEEN DIGITAL MONITORING SYSTEM
AND ANALOG SYSTEM DATA PROCESSING
ALe in dB
Hour
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Hourly Average
ALan, dB
Los Angeles
Site 4
(ref. a)
0.3
-0.4
-0.6
0.9
0.4
-0.9
0.9
0.3
0.1
0.6
-0.3
1.5
1.7
0.9
1.1
0.1
-0.1
-0.2
-0.1
0.6
3.0
0.5
-0.7
0.5
0.4
0.5
Los Angeles
Site 6
(ref. a)
-0.3
-1.2
0.2
1.8
-0.3
1.0
-1.2
-0.5
.
-1.1
0.8
1.0
-0.1
-0.1
-1.0
-1.0
-1.2
-0.3
1.4
-0.2
-1.1
-0.8
0.7
-0.2
0.2
Note: Differences are monitoring system levels minus analog system
levels.
C-4
-------
APPENDIX C REFERENCES
a. "Community Noise Measurements in Los Angeles, Detroit and Boston,"
Bolt Beranek and Newman Report 2078, June 1971.
b. "1-205 Noise Impact Analysis, " Bolt Beranek and Newman Report 2200,
April 1972.
c. "Noise Assessment of Interstate Route H-3 from the Halawa Interchange
to the Halekou Interchange, " Bolt Beranek and Newman Report 2099,
November 1971.
d. "Community Noise," EPA NTID300. 3, December 1971.
C-5
-------
APPENDIX D
INSIDE NOISE DATA
-------
APPENDIX D
INSIDE NOISE DATA
At fifteen of the one hundred sites at which outside measurements were
obtained, measurements were simultaneously taken inside the residence. The
inside measurement location was in the family or living room for fourteen of
the sites, and in the bedroom for one site. The microphone was positioned at
typical listener level, and for the family or living room situation was located
where a resident might normally spend several hours sitting, relaxing, watching
TV, etc.
The noise measurement instrumentation used for the inside measurements
s
was slightly different from that utilized for outside measurements, in that the
A-weighted noise level was continuously monitored and recorded on digital tape
in the field, with distribution into various counters being performed by computer
subsequently. The data were analyzed to yield the L-equivalent value for each
hour of the day.
Figures D-l through D-15 show patterns of L-equivalent values for each
hour of the day both inside and outside of the various residences. Also indicated
on each figure are the inside and outside values of the daytime (0700-2200 hrs)
and nighttime (2200-0700 hrs) L-equivalent levels (indicated as L(j and LQ,
respectively).
D-l
-------
Figure D-l. Inside/Outside L-Equivalent Values at Boston Site 0007
D-2
-------
BO
70
M
50
4C
3
20
Outside
~T~
Inside
T~
Outside
53.4
42.5
62.6
57.7
.
c
:
:'
\ /
\ x-
\ /
V
Inside Living Room
i
1 !
I
1 1
9 12
Hour of Day
15
:
21
Figure D-2. Inside/Outside L-Equivalent Values at Boston Site 0008
D-3
-------
-
-
r
Figure D-3. Inside/Outside L-Equlvalent Values at Denver Site 1104
D-4
-------
Figure D-4. Inside/Outside L-Equivalent Values at Denver Site 1107
D-5
-------
T
>
':
r
Figure D-5. Inside/Outside L-Equivalent Values at Denver Site 1110
D-6
-------
Figure D-6. Inside/Out side L-Equivalent Values at St. Louis Site 1201
D-7
-------
Inside
Outside
60.8
36.4
62.4
54.8
Figure D-7. Inside/Outside L-Equlvalent Values at St. Louis Site 1211
D-8
-------
Inside
Outside
-
--
-
j
E
71.4
55.9
65.8
56.1
20
Figure D-8. Inside/Outside L-Equivalent Values at Dallas Site 1401
D-9
-------
Figure D-9. Inside/Outside L-Equivalent Values at Dallas Site 1404
D-10
-------
Inside
Outside
60.2
48.7
64.2
41.7
Figure D-10. Inside/Outside L-Equlvalent Values at Seattle Site 15C
D-ll
-------
-
r
Figure D-ll. Inside/Outside L-Equivalent Values at Seattle Site 1506
D-12
-------
Inside
Outside
62.2
49.8
54.6
47.8
2C
Figure D-12. Inside/Outside L-Equivalent Values at Los Angeles Site 1603
D-13
-------
Inside
Outside
I
50.2
31.9
56.5
50.7
Figure D-13. Inside/Outside L-Equivalent Values at Los Angeles Site 1605
D-14
-------
12 15
Hour of Day
Figure D-14. Inside/Outside L-Equivalent Values at Los Angeles Site 1606
D-15
-------
Inside
Outside
-
c
60.0
63.9
55.5
46.)
Figure D-15. Inside/Outside L-Equivalent Values at Los Angeles Site 1608
D-16
-------
BIBLIOGRAPHIC DATA
SHEET
1. Report No.
550/9-74-00$
3. Recipient's Accession No.
4. Title and Subtitle
Population Distribution of the United States as a Function
of Outdoor Noise Level
5. Kcport Date
June 1974
6.
7. Author(s)
W.J. Gallowav. K. McK. Eldred, & M.A. Simpson
8. Performing Organisation Kept.
No.
9. Performing Organization Name and Addtess
Bolt Beranek and Newman Inc.
10. Project/Task/Woik Unit No.
11. Contract/Gram No.
68-01-1886
12. Sponsoring Organi/ation Name and Address
Environmental Protection Agency
Office of Noise Abatement and Control
Crystal Mall #2
ArUngton. Virginia 20460
13. Type of Report & Period
Covered
Final
14.
15, Supplementary Notes
16, Abstracts
This report summarizes the planning, conduct and results of a noise survey in 100
sites in urban areas across the United States and the use of these results coupled
with other existing data, to provide an estimate of the outdoor noise levels
experienced in residential areas by the United States population. It concludes that
there are more than 90 million people living in areas in excess of 55 Ldn and 1.3
million in areas in excess of 75 L^n.
17. Key Words .ind P/ocument Analysis. 17a. Descriptors
population density
day-night sound level Ldn
noise exposure
urban residential noise levels
noise measurement
7b. Idemificrs/Open-Ended Terms
7c. COSATI FieKl/Group
8. Availability .Scnremcnt
•imited number of copies from EPA.ONAC, Arlington,
Virginia. Available from NTIS
19..Security Class (This
Report) X
UNCLASSIFIED
20. Security Class (This
UNCLASSIFIED
21- No. of Pages
22. Price
FOIVM NTIS-3B (IO-/O)
USCOMM-DC 40329-f»7\
------- |