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\ ------- |