UWlxffC© STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON. D.C. 20460
January 27, 1981 OFFICE OF
AIR. NOISE. AND RADIATION
Dr. Richard Procunier
Noise Program Chief
Environmental Protection Agency
215 Fremont Street
San Francisco, CA 94105
Dear Dr. Procunier:
I have enclosed for your information the final draft of the forthcoming EPA
Guidelines for Noise Impact Analysis. Also enclosed is an accompanying Docket
Analysis. We are very grateful for your viewpoints and comments submitted in
response to our request for comments in June 1978. We apologize for the delay in
revising the Guidelines, but we have attempted to thoroughly address each issue
raised in the Docket.
As you may recall, the purpose of these Guidelines is to provide decision makers
in both the public and private sectors with analytic procedures which can be
uniformly used to express and quantify impacts from noise, so that such impacts
can be readily understood and fully considered within the comparative evaluations
which constitute noise environment decisions. The procedures contained in this
document are applicable to the preparation of environmental noise assessments,
and are entirely consistent with analytic procedures in current use by EPA. It
should be carefully noted that adherence to the procedures within the Guidelines is
strictly voluntary. The Guidelines are neither mandatory nor regulatory in intent.
We believe that the Guidelines offer a valuable aid to decision makers for
quantifying and evaluating noise environment changes.
The Guidelines as enclosed are based on the deliberations of the Committee on
Hearing, Bioacoustics and Biomechanics (CHABA) Working Group 69, from 1972 to
1976, in response to a request by the U.S. Environmental Protection Agency.
Recommended guidelines for preparing environmental impact statements on noise
were published by the National Academy of Sciences in early 1977. The procedures
in that document reflected a compromise among practicality, economy, desired
accuracy, and specificity. Subsequently, a request for comments on the CHABA
recommended procedures was solicited by EPA in the Federal Register in June
1 978, to which you raised questions or submitted remarks.
We have since carried out a detailed, step-by-step analysis of the issues raised
during the comment period in order to improve the overall accuracy, usage, and
general readability of the document. Conceptually, the enclosed final draft report
contains the same basic procedures delineated in the CHABA guidelines published
in 1977. The enclosed document does incorporate, however, some notable
refinements in both the general audible noise and high energy impulsive noise
assessment methodologies. In addition, the document now -contains improved
example cases demonstrating some applications of the recommended procedures.
-------
In light of the changes and additions noted above, we extend to you a further
opportunity to comment on the final draft version of the Guidelines. Please feel
free to raise any additional queries on the manner in which certain issues were
addressed, and to indicate to us any last minute refinements or considerations
which you feel would contribute to a more accurate and usable document.
Accordingly, we are requesting that if you wish to submit any further comments to
us on the Guidelines, that we receive these comments no later than March 20,
1981.
Please send your comments to:
Jeffrey Goldstein
Environmental Protection Agency
Office of Noise Abatement and Control
Washington, D.C. 20460 (ANR-471)
Telephone: (703)557-0592
V
It is our intention to finalize the Guidelines in the next four to six months, and
make copies available to the public through an appropriate notice in the Federal
Register at that time.
We very much appreciated your prompt and helpful assistance in reviewing and
commenting upon the CHABA recommended procedures, and we look forward to
receiving any further thought you may have in regard to the enclosed Guidelines.
Sincerel
Yarles LTElkins
Deputy Assistant Administrator
for Noise Control Programs
(ANR-471)
Enclosures
EJBD
ARCHIVE
EPA
402-
D-
81-
001
-------
-------
December 1980
DRAFT VERSION
GUIDELINES FOR NDISE IMPACT ANALYSIS
-------
CONTENTS
Section Title Page
1. Introduction 1
1.1 Purpose of the guidelines 1
1.2 Overview of the approach 3
1.3 Structure of the guidelines 7
1.3.1 Preliminaries 8
1.3.2 General audible noise 12
1.3.3 Special noises 13
1.3.4 Vibration 14
1.3.5 Potential changes in population 14
1.3.6 Examples 15
1.4 Other considerations 16
1.4.1 Projects which reduce noise 17
1.4.2 Temporary projects 17
1.4.3 Uncertainties in the analysis 18
2. General audible noise 19
2.1 Basic screening procedures 20
2.1.1 Measures for the decription of general
audible noise 23
2.1.2 Determining the yearly day-night sound level 25
2.1.3 Determining the population affected by the
noise of the proposed project 30
2.2 Health and welfare effects range 33
2.2.1 Human noise exposure criteria 33
2.2.2 Quantification of the noise impact 40
2.3 Severe health effects range 52
2.3.1 Human noise exposure criteria 54
2.3.2 Quantification of the impact 61
2.4 Environmental degradation 63
2.5 Treatment of temporary projects 64
2.6 Example application 66
3. Special noises 71
3.1 High energy impulse noise 71
3.1.1 Description of high energy impulse sounds 72
3.1.2 Human noise exposure effects of high energy
impulse noise 75
3.1.3 Structural damage criteria for impulse noise 80
3.2 Infrasound 85
3.2.1 Description of infrasound 85
3.2.2 ' Human noise exposure effects of infrasound 85
3.3 Ultrasound 87
3.3.1 Description of ultrasound 87
3.3.2 Human effects of ultrasound 87
3.4 Noises with information content 88
4. Vibration 89
4.1 Human effects of vibration 89
4.1.1 Description of building vibration 90
4.1.2 Human vibration exposure criteria 90
-------
Section
4.1.3
4.2
5.
5.1
5.2
5.3
5.4
CONTENDS (Continued)
Title Page
Quantification of the impact 94
Structural effects of vibration 97
Summary of noise impact analysis 99
Purpose and structure of the guidelines 99
Analysis of impacts of general audible noise 100
Analysis of impacts due to special noises 103
Analysis of impacts due to vibration 104
References
Appendix A
B
c.
D.
E.
R-1
Acoustical terms and symbols used in the guidelines,
and some mathematical formulations for them A-1
Procedures of other Federal agencies B-1
1. Environmental noise measures and their purposes
in Federal programs
2. Estimating I^n from other noise measures
B-2
B-3
Summary of human effects of general audible noise C-1
D-1
E-1
Measurement of and criteria for human vibration
exposure
Example application of guideline procedures for
general audible noise
vi
-------
List of Tables
Table Title Page
1 Sumnary of methods for noise impact analysis 6
2 Yearly day-night sound level as estimated by
population density 28
3 Values of the weighting function for general
adverse response 46
4 Average hearing loss as function of 8-hour Leg 58
5 Criterion function for severe health effects 59
6 Sample data presentation: future noise levels
without proposed project 67
7 Sample data presentation: future noise levels of
project alone 68
8 Sample data presentation: future levels from all
sources combined 69
9 Sample data presentation: special situations 70
10 Values of weighting function for high energy
impulse noise 79
11 Conversion of LQ^ to L^n via equal annoyance 81
12 Basic threshold acceleration values for acceptable
vibration environments 92
vii
-------
List of Figures
Figure Title Page
1 Preparation of a noise impact analysis 5
2 Flow chart and worksheet 9
3 Screening diagram 21
4 Summary of annoyance data from 12 surveys showing
close agreement 36
5 Comparison of new annoyance function with previous
functions 39
6 Suggested descriptors for various situations 50
7 Comparison of curvilinear function and fractional
impact linear function 53
8 Potential hearing damage risk for daily exposure to
8-hour equivalent sound levels . 60
9 Recommended relationship for predicting community
response to high energy inpulsive sounds 77
10 Infrasound criteria 86
11 Weighting characteristic for building vibration
in terms of human responses for the frequency range
1 to 80 Hz 91
12 Vibration criteria for residential areas 93
13 Percentage of population complaining as a function
of peak acceleration 95
14 Types of analyses suggested 102
viii
-------
Chapter 1
Introduction
It is the policy of the United States Government to consider the poten-
tial adverse impact en the environment of all proposed federal actions and
projects. Many states and local governments have similar policies. The pur-
pose of such policies is not merely to provide a catalog of the adverse
environmental impacts of a project (which has already received tacit approval),
Bather, the purpose is to provide a description of the environmental con-
sequences of a possible project, so that an understanding of those conse-
quences can be an integral part of the decision on the project. In order
for this to occur, it is necessary for the environmental effects to be
expressed in a manner which can be readily understood by the decision-maker,
and by the general public whose participation in such decisions is usually
encouraged by all levels of government.
1.1 Purpose of the Guidelines
One of the potential environmental consequences of many proposed actions
or projects is a change in the noise and vibration environment. The "action"
may be the building of a new refinery, development of a new mine, construc-
tion of a road, use of a new piece of machinery, etc. It may involve the
enlargement or the reduction in size of an existing facility, or an effort to
make a given facility more quiet. It may be the promulgation and enforcement
of a new noise abatement regulation. It may be the temporary noisy construc-
tion phase of an inherently quiet facility. Or, with no change in the noise
environment, the action may entail a change in land-use or population density
-------
in a neighborhood. Any proposed change that will significantly affect
either (a) the amount of noise generated or (b) the number of people exposed
to it, will result in noise-related environmental impacts. These guidelines
contain procedures which can be used to describe and quantify those noise
related impacts. These procedures are primarily intended for use during
initial planning stages of projects in order that the potential environ-
mental noise effects of proposed actions can be identified and considered
early in the decision process, and so that appropriate noise mitigation
measures can be conveniently implemented.
The users of this document are expected to be federal agencies, state
and local governmental agencies, industries, environmental groups, and
individuals. The procedures described here are applicable to the preparation
of environmental assessments and environmental impact statements, and to any
other situation in which a description of noise environment changes would be
useful. Although individual agencies have their own specific procedures, in
most instances the approach described here is consistent with those procedures.
These guidelines are not intended to replace existing approaches, but to
complement and extend them, by showing how to proceed from a description of
noise levels to a quantitative description of the impacts of noise on people.
It is hoped that this document will assist in achieving nationwide consistency
in dealing with noise problems, and provide an objective and uniform evalua-
tion of the noise impacts.
The approaches described in these guidelines are not mandatory, nor are
specific numbers which appear in the guidelines intended to be construed as
standards. The guidelines are offered as an aid to the treatment of noise
impacts in the preparation of environmental assessments, reviews, and impact
-------
statements. Paraphrasing a statement by the Council on Environmental Quality,
these guidelines are intended to help public officials make decisions which are
based on an understanding of environmental consequences, and to take actions that
protect, restore, and enhance the environment [1*, p. 25233]. The purpose of
these guidelines, then, is to present procedures which can be used to express
noise impacts in terms which are easily understood by decision-makers, so that
those impacts can be fully incorporated in the comparative evaluations which
constitute the decision.
1.2 Overview of the approach
The guidelines are based on the philosophy that the technical approach,
the descriptors of the noise environment, the measurement and prediction
methods, the evaluation criteria, and the techniques for impact assessment
should be as simple as possible consistent with reasonable accuracy. To the
extent that they are also uniform across different projects, public understand-
ing of noise impacts will be improved.
It appears feasible to follow these principles to arrive at an objec-
tive, and for most situations, quantitative definition of the noise impact.
In many situations, it will be possible to calculate a single number which
expresses the total noise impact of a proposed project on the population
exposed. When this single number index can be produced, the prospects are
enhanced for a more objective and rational comparison of noise with the other
impacts associated with specific projects. Quantitative tradeoff studies are
made possible—for example between noise impacts and societal benefits.
In some cases, this level of quantification might seem unwarranted, or
^Numbers in square brackets refer to the reference list at the end of the
main text of this report.
-------
overly mechanistic. For such cases, the guidelines suggest a tabulation, in
5 decibel (dB) increments, of the land area or number of people affected by
adverse noise levels. In addition, a traditional, non-quantitative descrip-
tion of the noise impact is encouraged, either as a supplement to these
numerical descriptions, or, in unusual cases, as the sole analysis of the
noise impacts.
The preparation of a noise impact analysis proceeds through several
distinct steps to arrive at these descriptions of the noise impact, which are
then used in the decision-making process (Figure 1). The methods proposed for
use in each of these steps (Table 1) are based, in part, on the work and
the progress achieved over the last few years by interagency committees, on
the recommendations of the National Academy of Sciences-National Research
Council, and on other scientific findings.
For measurement of the noise environment, use of the day-night sound
level (L^n), officially adopted by several Government agencies since publica-
tion of the Environmental Protection Agency's "Levels Document" [2], is recom-
mended as the primary measure of general audible noise. Circumstances calling
for the use of short-term measures of general audible noise are also discussed.
A modification of the day-night sound level for impulse noise is based on the
report of a National Academy of Sciences, Committee on Hearing, Bioacoustics
. o A and Biomechanics (CHABA) working group [3], and of an interagency task force
[4] on this subject. Measures to be used for infrasound, ultrasound, and
vibration are also described in these guidelines.
The quantification methods recommended for impact assessment in these
guidelines are further developments of the Fractional Impact Methodology
used by EPA for assessing the health and welfare effects of a noise
-------
Description of Project or Action
look for Noise-Related Effects of Project or Action
•.. —" * " ~ /
Does Noise rnVlrormienilCh'ahge?
Does Exposed Population Change?
Are Changes Significant Enough for Detailed Documentation?
i
Measurement and Documentation of Noise/Exposed Population
a. Definition of Existing Noise/Exposed Population
b. Projection of Future Noise/Exposed Population
c. Change in Noise/Impact of Project
I
Assessment of Impact
a. Health and Welfare Effects
b. Potential Loss of Hearing
c. Environmental Degradation
Discussion and Analysis of Results
Decision on Proposed Project
Figure 1. Preparation of a Noise Impact Analysis
-------
TABLE 1. SUMMARY OF METHODS FOR NOISE IMPACT ANALYSIS
TYPE OF
ENVIRONMENT
TYPE OF
CRITERIA
RECOMMENDED
NOISE MEASURE
SCREENING
LEVELS
ASSESSMENT
METHODOLOGY
GENERAL
AUDIBLE
NOISES
SPECIAL
NOISES !
HIGH
ENERGY ,
IMPULSE
NOISE
INFRASOUND
ULTRASOUND
VIBRATION
POTENTIAL FOR LOSS
OF HEARING
GENERAL ADVERSE
EFFECTS
ENVIRONMENTAL
DEGRADATION
STRUCTURAL
DAMAGE
ANNOYANCE DUE TO
AUDITORY STIMULATION
AND BUILDING
VIBRATION
ANNOYANCE AND
PHYSIOLOGICAL
STRUCTURAL
DAMAGE
ANNOYANCE AND
COMPLAINTS
8-HOUR AVERAGE SOUND LEVEL OR
24-HOUR AVERAGE SOUND LEVEL
DAY NIGHT SOUND LEVEL
PEAK PRESSURE
PEAK ACCELERATION
DAY-NIGHT SOUND
LEVEL USING C-WEIGHTED
SOUND EXPOSURE LEVEL. Lsc. FOR
IMPULSES
MAX-
6.1 Hz TO 20 Hi \ SOUND
20 kHl to 100 kHz f PRESSURE
LEVEL
PEAK ACCELERATION (WEIGHTED)
RMS ACCELERATION (WEIGHTED)
VERSUS TIME OF EXPOSURE
Ldn = 75dB
PROJECT LEVELS
HIGHER THAN
10 dB BELOW
THE EXISTING
LEVELS
EMPIRICAL
FORMULAS
1 m/sec2 INSIDE
l-ScOFBOilBFOR
DAYTIME. OR TtfdB .
FOR NIGHTTIME
0.1'HzTO5Hz:120dB '
6 Hz TO 20 Hz: 120-30 LOG-E
105 dB
1 m/ioc2 FOR MOST
STRUCTURES
0.6 m/tec* FOR SENSITIVE
STRUCTURES
0.05 m/Mc2 FOR CERTAIN
ANCIENT MONUMENTS
0.0036 m/KC2. OR
HIGHER DEPENDING ON
TIME OF DAY AND
TYPE OF PLACE
HEARING-LOSS-WEIGHTED'
POPULATION. HWP
SOUND-LEVEL-WEIGHTED
POPULATION. LWP
TABLES AND DESCRIPTION ONLY
TABLES AND DESCRIPTION ONLY
SOUND-LEVEL-WEIGHTED
POPULATION. LWP
DISCUSSION OF POSSIBLE EFFECTS.
NO TABULATION MADE
TABLES AND DESCRIPTIONS ONLY
VIBRATION-WEIGHTED POPULATION.
VWP
-------
environment. For the general adverse response to noise in the 55 to 75
dB (l£n) range, the function is based on data presented by Schultz in
a recent review paper [5]. Similar impact assessment methods are proposed
in these guidelines for quantifying the following: the potential for loss
of hearing at 24-hour equivalent sound levels in excess of 70 decibels;
the general adverse response to impulse noise; and the complaints caused
by vibration. For general audible noise in rural and wilderness areas,
and for infrasound and ultrasound, no quantification methods are sug-
gested, but more qualitafeative descriptions are recommended instead.
^>
The measures and methods listed in Table 1 and described in this
report are simplifications, and the recommendation for their use is not
intended to discourage more rigorous approaches. However, to provide a common
framework for comparison of different environmental noise assessments
f
(conducted by different persons in different parts of the country), it
is strongly recommended that the methods of these guidelines also be used
along with any other additional approach.
1.3 Structure of the guidelines
Three principal types of noise and vibration environments are considered:
general audible noise; special noises; and vibration. General audible noise
is noise as commonly encountered in our everyday living environment. It can
be adequately described by either the equivalent A-weighted sound level
/
(Leq) or its variation that includes a nighttime weighting, the day-night
sound level (I^n) • for most practical cases this type of noise measure will
adequately describe the noise environment, and much of the document concerns
the evaluation of general audible noise. Not all noises can be adequately
evaluated by average sound levels, however. Examples of such special noises
-------
are infrasound (frequency range of 0.1 to 20 Hz), ultrasound (frequency range
above 20 kHz), certain types of impulse noises (such as blasts and sonic
booms), and sounds that convey more information than random noise sources
with comparable average sound levels (such as voices, warning signals, or
%
barking dogs). Procedures are also included for evaluating the impact of
vibration on man. While the main reason for their inclusion here is to
account for vibration generated by airborne noise, the impact of certain
types of vibration can be assessed whether the transmission paths are air-
borne or structureborne.
There is a separate chapter for each of the three principal types of
environment. Each chapter covers four topics: the appropriate physical
measurement for that type of noise; methods for determining the existing
levels and for predicting the levels for the proposed project; human noise
exposure criteria; and procedures for quantifying the impact, usually in
terms of those criteria. All of the information necessary to deal with
one type of noise environment is thus in one place, to minimize the effort
required by a user to follow these guidelines.
1.3.1 Preliminaries
The logic of the structure of these guidelines has been set out in a
combined flow chart and worksheet (Figure 2), to provide guidance for using
this report and for carrying out the various parts of the noise impact
analysis. There are four principal branches in the flow chart (labeled A, B,
C and D) to be followed, depending on the nature of the proposed project and
its potential impact. There are exit points along each of the branches, at
-------
DESCRIBE
PROJECT OR
ALKRNAriVE 1
SOLUTION
DOES THE
PROJECT I
CHANGES IN
POPULATION?
MBEH5 IN.PAHENTHESH HEFIH TO THE. . Ka Honf
PROPaiAU SECTION OF THESE GUIDELINES. (r
-------
which the analysis for that branch may stop without the need for any further
analysis, since it is clear by then that there is no significant noise impact
with respect to the concern on that branch. At the right-hand edge of the
flow chart there are three columns that can be checked to indicate the
outcome of the analysis at each branch point. When this flow chart is used
as a worksheet, these columns summarize the noise impact analysis for the
project, showing the stages at which exit points occurred, and calling
attention to aspects of the noise impact receiving explicit evaluation
according to the methods of Chapters 2, 3, and 4.
Usually, there will be not just one version of the proposed project, but
a number of alternative proposals. Each of these alternatives must be analyzed
for noise impacts. There will thus be a flow chart worked through for each
of the alternative schemes. Each of the worksheets, by its summary columns
of checked boxes, will indicate what aspects of the noise impact of that
alternative received explicit consideration. A comparison of these columns
will facilitate choosing the project alternative with the least noise impact
on the environment.
1.3.1.1 Flow Chart
The following discussion of the use of the flow chart provides a brief
explanation of each of the branches. (The section of this report contain-
ing the more detailed discussion is indicated in parentheses on the flow
chart.)
The first step is to provide a general description of the proposed pro-
ject, including those aspects that are expected to contribute to noise
impact. The expected noise impact may be either adverse, if the noise environ-
ment would be worsened by the project, or beneficial, if the environment
10
-------
would be improved. Both the short term and long term effects expected from
the project should be described. For example, the construction of a new
airport or highway in a sparsely settled region would have as its initial
impact an increase in noise that would affect relatively few people. However,
the new facility will attract new people and business which will increase the
nearby population density, unless proper land-use planning and implementation
occur. Thus the ultimate noise impact may be significantly greater than that
projected on the basis of the initial effect alone. To evaluate an action
over time, it is suggested that a time interval of 20 years be used, unless
the project will be in existence less than 20 years, in which case the
project lifespan should be used. Thus, the initial impact and the expected
impact after 20 years should both be evaluated. To present a complete
picture, the impact after 5, 10, and 15 years might also be presented. When
comparing the impact between projects or alternatives or when assessing
cost-effectiveness, the average impact over a 20 year period may be used.
The first branch point in the flow chart occurs after the potential
noise impacts of the proposed project have been described. At branch
points such as this, each of the available branches (labeled A, B, C and D)
should be taken, and followed through to the appropriate indicators in the
righthand columns. At each of the question points following each branch, if
the project will entail no change at all, the 'NO* answer will be followed to
'EXIT*, and the analysis for that branch is complete. In that case, check
the box at the right-hand side of the page under 'No environmental change1.
If exit points have been found for each of the branches A, B, C, and D, the
noise analysis need not proceed further: four check marks will be in the
column labeled "No environmental change" at the right of the page and the
11
-------
noise analysis is finished. The environmental impact assessment on noise
will simply state this fact. Otherwise, the analysis continues in those
branches in which no exit point has been found.
1.3.2 General audible noise
If on the first branch the project was found to include a potential
change in general audible noise, chapter 2 is appropriate, describing how
to identify and quantify the noise impacts for such an environment. The
analysis begins with the screening step, to determine if the potential
change is large enough to pursue in a detailed analysis (section 2.1).
If it is not, the 'NO1 response is again followed to 'EXIT* and the analysis
for this branch is complete.*
If the potential change is large enough to warrant further analysis,
the next question is whether people will be exposed to the noise from
the project. If the answer is 'YES1, there are three branches to follow,
depending on the level of noise resulting from the project. The first
branch deals with the range bounded roughly by L$n values of 55 dB and
75 dB, in which general health and welfare criteria are the impacts of
interest (section 2.2). The second branch concerns projects which include
noise levels above 75 dB (I^n). Where this occurs, severe health effects
due to noise should be considered (section 2.3). The third branch is for
projects which result in levels less than 55 dB (I^n). Although these
*The flowchart and worksheet is designed primarily for those cases where
the noise (or vibration) impact is expected to be adverse, that is, the
noise environment is anticipated to be worsened by the project. If the
project entails a reduction in noise, thus improving the environment, the
flowchart and worksheet can still be used as a guide to carrying out any
noise assessment that may be desired to ascertain the degree of improve-
ment. See the discussion in section 1.4.1.
12
-------
are levels below which adverse health or welfare effects generally do not
occur, environmental degradation is of concern, and should be discussed
(section 2.4). For each of these three ranges of general audible noise, the
section identified provides a discussion of the human exposure criteria, and
methods for quantifying the impact in terms of these criteria. It should be
pointed out that at this branching point, at least one of the three sound
level ranges must lead to the requirement for a noise impact analysis, and
possibly more than one branch will do so. That is, these categories are
exhaustive (i.e., they cover all the possibilities), but they are not sin-
gularly exclusive (i.e., more than one can occur).
Even if people do not normally live or work in the area exposed to the
new noise levels, environmental degradation is still of concern. As such,
this is difficult to quantify, but it should be discussed, preferably in
terms of the principal uses made of the affected area (e.g., urban recreation,
wilderness recreation, wildlife).
1.3.3 Special noises
If the project is found to involve special noises, namely impulse
noise, infrasound, ultrasound, or sounds with negative information content,
it is necessary to follow branch B further (chapter 3). The screening step
verifies that the levels involved are high enough to warrant an analysis.*
These levels are discussed separately for each type of special noise:
impulse noise in section 3.1.1; infrasound in 3.2.1; ultrasound in 3.3.1;
and noises with information content in section 3.4. If people are exposed
to one or more of these special noises, the second part of the appropriate
*See footnote on page 12.
13
-------
section of chapter 3 is available, describing procedures to be used to
discuss and quantify the impacts. For impulse noise, there is also the
possibility of structural damage occurring (section 3.1.3).
This section of the flow chart is obviously somewhat simplified. If it
were drawn in full detail, there would be a branch such as this for each one
of the four special noises. Thus one should repeat this branch four times.
The reason for the simplification is that in most (but not all) cases encoun-
tered, no more than one type of special noise will generally be involved.
1.3.4 Vibration
Branch C is followed if the project involves vibration (chapter 4).
Again, there is a screening step, to allow an 'EXIT1 to 'NO ENVIRONMENTAL
CHANGE* if the levels are low enough (section 4.1.2).* If the levels are
*
higher than the cut-off, there are two branches to be pursued, the first
if people are exposed (section 4.1.3), and the second if buildings or monu-
ments are exposed (section 4.2). Thus for vibration there are both human
and structural criteria to be considered in assessing potential impact.
1.3.5 Potential changes in population
Some projects entail exposing new populations to existing noise levels,
for example the construction of a housing development in an area adjacent to
a major roadway. Branch D of Figure 2 describes the procedures to be fol-
lowed in such a situation. If the noise levels from existing sources are
presently below 55 dB (L^), and are expected to remain this low in the
future, then there is 'NO IMPACT', and no further analysis is required on
this branch (as long as there are also no special noises encountered). It
*See footnote on page 12.
14
-------
should be noted, however, that higher density development (whether residen-
tial, industrial, or commercial) usually brings with it increasing noise lev-
els, such that it is unlikely that sound levels after project completion will
be as low as they are at present. This 'EXIT1 is unlikely to be realistic for
any major development. If the noise levels are not below 55 dB (I^n), or
if special noises are present, the analysis follows the same steps as did
branch A, when new noises affected existing populations (sections 2.2, 2.3,
and chapter 3).
1.3.6 Examples
The following examples are presented to illustrate which of the major
branch(es) of the flow chart to use for various projects:
(1) A project that entails a change in land use may cause only a change
in the existing noise in an area, so only Branch A or B would be fol-
lowed; on the other hand, it may only involve relocation of some
of the population, in which case only Branch D would be followed.
If the project is expected to cause (or diminish) vibration, Branch
C would be followed. Most land use changes, however, will involve
a combination of A, B, C, and D.
(2) A project involving the installation of new equipment, or the replace-
ment of old equipment, is likely to require analysis of only branches
A, B, and/or C, since no population shift is likely to be involved.
(3) A project that consists of a new regulation, or a change in an exist-
ing regulation, might follow either A, B, or D (see discussion in
section 1.4.1). For example, a new regulation reducing the noise output
of heavy trucks would change the noise along a highway, and thus Branch
15
-------
A should be followed. On the other hand, a change in the noise policy
of the Department of Housing and Urban Development may alter the distri-
bution of future dwellings among neighborhoods with different levels of
existing noise; such a regulation would change the population exposed to
noise without affecting the noise anywhere, and hence would warrant
analysis along Branch D.
(4) A new airport, whose primary effect would be increased noise levels in
the neighborhood (Branch A), might impact only presently undeveloped land
that could be spoiled for later residential development by the airport
noise. The future magnitude of the impact would be quite different for
a prospective airport where land is purchased around the proposed site
for controlled leasing to non-noise-sensitive activities as compared to
one where such precaution was not taken.
(5) A project that causes a change in the interior noise of aircraft cabins
or a change in the noise insulation of automobile bodies would be ana-
lyzed on a path along Branch Af since it changes the noise environment
in existing spaces with a definable existing population.
1.4 Other considerations
The preceding summary of the structure of these guidelines, and of
the flow chart representing that structure, is written to deal with a proposed
project which will increase noise levels, or in which more people are exposed
to existing noise levels. These are not the only types of projects for
which noise impacts should be considered. This section describes an additional
situation in which noise impacts are a concern—projects aimed at reducing
noise levels. The section also discusses two topics which are relevant to
16
-------
all three of the chapters which follow: shortened analysis procedures for
temporary projects; and the treatment of uncertainties encountered in the
analysis.
1.4.1 Projects which reduce noise
Two of the examples of proposed actions with noise-related impacts
described en page 1 deal with the reduction of noise. If an action is pro-
posed in order to reduce noise-related impacts, it is immediately obvious
that an analysis of those impacts is called for. In most cases it is
equally obvious what kinds of impacts are of interest (e.g., general health
and welfare impacts). Hence the flowchart and worksheet are not really
needed as an aid to identifying the area of concern.
Instead, one may simply turn to the appropriate sections of these guide-
lines for a discussion of useful procedures for quantifying noise impacts.
In other words, the procedures (described subsequently) apply equally well to
projects which reduce noise as they do to projects which increase noise. The
summary of those procedures (described previously) is written only in terms
of projects which produce noise.
1.4.2 Temporary projects
At this stage of a noise impact analysis, the specific types of noise
impact requiring detailed documentation will have been identified. This
will have been accomplished either through the use of the worksheet (Figure
2), or by the fact that the action is intended to reduce noise impacts.
The next question to address is how far into the future the analysis should
go. Earlier, it was suggested that either the project duration or twenty
years, whichever is less, should be considered (section 1.3.1). This means
17
-------
that not only noise levels, but also affected populations, need to be pre-
dicted over the time period of interest. The impacts to consider are not
merely the immediate ones, but long-term ones as well.
Documentation of the impact for temporary projects is simplified by
the fact that population prediction is unnecessary; existing population
or land use information is sufficient. In this context, temporary is
taken to mean less than roughly two years duration. Beyond that length
of time, significant population changes may take place in an area, so that
population forecasts become important. This would be true even for longer
duration construction projects, which might be in place for 5-10 years.
1.4.3 Uncertainties in the analysis
There will almost always be areas of uncertainty in the noise impact
t
analysis, usually because of the unavailability of needed factual information.
For example, the projected future traffic volume for a proposed freeway may
be uncertain; the noise of a not-yet-built device may be only approximately
known; or the population estimated to be exposed to various sound levels from
the project may be subject to error. In all cases, a discussion of the
probable source and degree of these uncertainties should be included in the
analysis. Perhaps the most suitable approach for this purpose is to take the
upper and lower bound for each of the uncertain quantities that enter into
the analysis, and group the "most favorable" and "least favorable" bounds of
these quantities together to arrive at two estimates of the environmental
noise impact: the best and worst cases- that together bracket the range of
likely actual results of proceeding with the proposed project.
18
-------
Chapter 2
General Audible Noise
This chapter describes procedures to be followed for analyzing and docu-
menting the noise-related impacts of proposed projects which affect or are
affected by general audible noise. The first section outlines a screening
procedure for determining whether the expected noise levels of the proposed
project are high enough to warrant detailed analysis. As part of that discus-
sion, appropriate noise measures are identified, and methods for estimating
and predicting them are indicated. The second section discusses the general
health and welfare effects of noise on people, which serve as the criteria
for evaluation of noise in most urban and- suburban settings. It also pro-
vides a procedure which can be used to summarize these effects with a single
number impact descriptor. The third section discusses the severe health
effects which can be caused by higher levels of noise, and suggests a single
number impact descriptor for these. The fourth section discusses procedures
to follow for projects which will have reasonably low noise levels, but
which are located in very quiet areas—that is, projects for which environ-
mental degradation is the primary concern. Simplifications of these analysis
procedures which can be used for temporary projects are described in the fifth
section. The final section contains an example application of the procedures,
including samples of the types of tabulation which are recommended.
The criteria used in this chapter are not to be considered all-inclusive;
additional information should be used depending on the scope and magnitude of
the environmental change. The EPA Criteria and Levels documents [6,2] can be
consulted as additional reference sources as well as any other applicable
information.
19
-------
2.1 Basic screening procedures
Some proposed projects will obviously cause a severe noise impact on
their surroundings, others may obviously be so quiet as not to change the
noise environment at all. In the first case there is no doubt that a full
analysis of the noise impact is required; in the second case one would simply
state, with minimal documentation, that no impact is expected. About many
projects, however, there will be a question as to whether their noise impact
is significant enough to warrant a full noise impact analysis. This section
offers a screening test to determine how extensive a noise analysis is needed.
Figure 3 presents a screening diagram for use in determining whether a
full impact analysis is needed. The diagram is based on a comparison of the
existing noise environment and the noise environment due to the proposed
project alone.* This comparison should normally take place at the noise
sensitive location(s) in closest proximity to the proposed project.** So
long as the expected yearly L^ (see section 2.1.1 for explanation of
from the project is lower than 10 dB below the existing yearly L^n, the
* The meaning of "project alone" is clear when an entirely new facility is
to be built. But what about the expansion of an existing facility? In
such cases, "project alone" should be considered to be the total expanded
facility or project. For example, if the project is the widening of an
existing highway from two to six lanes, future noise levels from the
"project alone" would be the noise from the six-lane highway, not just the
noise from the additional four lanes. That is, the "project alone" is
the new six lane highway.
**There are some exceptions to this rule. If it is known that the
greatest noise impact will occur at a noise sensitive location farther
away, the comparison should take place at that point. An example would
be a close-in area protected from the noise by natural terrain, relative
to an unprotected point farther away. If the latter location receives
the greatest impact, the comparison should take place at that point.
20
-------
to
UJ
0
_J
<
u
UJ
3
cc
o.
Q_
o
J
Q
III
h-
u
Ul
Q.
X
Ul
j
70
60
50
40
'
h
S"
+
s~
~ ' DOCUMENTATION
FULL NOISE ENVIRONMENT
DOCUMENTATION FOR ALL
PROJECTS
/ DEPENDS ON
,' FUTURE OF 1
f EXISTING
- • S NOISE SOURCES
y
/
/
POSSIBLE NOISE /
_ DEGRADATION /
ANAL1. -IS /
' /
^ \s \ \
40 50 60
f
ALL PROJECTS
SCREENED OUT
1 . 1
70 80
\ ^
90
EXISTING Ldn(y)
FIGURE 3. SCREENING DIAGRAM
-------
project is screened out.* No further analysis is required, because the
change in the environment is not significant. The rationale for the location
of the screening line in Figure 3 is that for any project which is not
screened out, the total environmental noise level after project completion
will be increased from the existing levels. For example, if a project alone
is expected to produce the same level of noise as the existing yearly L, ,
then post-project total environmental noise will be increased by 3 dB. If the
difference obtained by subtracting the project alone noise levels from the
existing levels is greater than 10 dB, the post-project total noise level
will increase by less than 0.5 dB, which rounds off to a zero increase.
For any new project in which the expected I^n is greater than 55 dB,
the probability of significant noise reductions of some of the existing
sources should be considered. If the existing levels are high because of'one
major source which is likely to be quieted in the future, then the proposed
project should not be screened out; further analysis is needed since, in
the future, the proposed project could become relatively more dominant than
expected on the basis of existing noise levels. If the existing levels are
unlikely to be reduced, then the project can be screened out. However, even
if no reductions are likely an impact analysis can still be carried out, and
in many instances is strongly encouraged, based on idealized or hoped for
future noise levels for the area. In other words, noise impact analysis is
reconmended for noisy projects even if they are in already noisy areas.
*If the project is temporary with a duration of less than one year,
expected yearly 1^ should not be used. Rather, it is more appropriate
to use the day-night sound level averaged over the actual duration of the
project (see Section 2.5). In any case, existing yearly L^ is always
used.
22
-------
The diagonal line in Figure 3 continues as the indicator of screened-out
projects to very low noise levels. For proposed projects which are above
this line, but produce levels below 1^ = 45 dB, a modified form of noise
impact documentation is suggested. The modification is needed in part
because of the lack of data on the effects of very low level noise. Such low
levels are in fact extremely rare, if indeed they ever occur. Even on the
north run of the Grand Canyon 1^ was found to be close to 44 dB, due to
bird, animal, and insect noises [7]. Hence it is not expected that situations
in the lower left portion of the diagram will be encountered often, and the
diagram has been truncated accordingly.
2.1.1 Measures for the description of general audible noise
The primary measure for describing general audible noise is the day-night
sound level, symbolized as L, . The unit for L, is the decibel. The
day-night sound level is a 24-hour equivalent sound level in which nighttime
noise levels occurring between 10:00 p.m. and 7:00 a.m are increased by 10 dB
before calculation of the 24-hour average. Equivalent sound level is numerically
equal to the value of a steady sound level that would convey the same mean-square
A-weighted sound pressure level as does the actual time-varying sound in the
same time period. Equivalent sound level is also called average sound
level.
long term environmental impact is evaluated by the yearly day-night
sound level, symbolized as L^/yx* The yearly average is recommended on
the grounds that the noise metric used should be one which reflects any change
in the noise environment, and that this should be done consistently for differ-
ent sources. Yearly day-night sound level is analogous to the traffic
engineering concept of annual average daily traffic. In other words, it is
23
-------
meant to represent levels measured under average conditions, or, if condi-
tions vary during the year, weighted averages of levels at the different
times of year.
In some instances, a rough approximation to annual average conditions or
noise levels will be sufficient; in others, it will be necessary to be more
precise. For example, noise levels for some airports are reported for an
average busy day, rather than an annual average. If the project under
analysis is a land-use development, it is quite reasonable to use such
existing noise information, even though it is not exactly the annual average.
The error in the data is small enough that the cost of a more exact estimate
of the annual average is not warranted. On the other hand, if the project
being analyzed involves a change in airport use (for example Sunday flights
when there were previously none) the noise level typical of an average busy
day may lead to nonsensical results. (The busy day level would be reduced,
in the example, if the aircraft noise on Sunday was less than the average on
other busy days—even though over the year more noise was being produced
because of the added operations.) Approximations for the annual average L,
can be very useful shortcuts, but need to be applied with careful judgment.
Day-night sound level is the primary measure of general audible noise,
and is appropriate for noise environments that affect a community over an
entire 24-hour day. There are two kinds of situations where such a measure
is not appropriate, however. The first kind consists of those situations
in which it is desirable to assess the effect of a noise environment on an
activity of less than 24-hour duration. An example is the effect of noise
on speech communication in classrooms or in offices. In these situations it
is useful to consider the equivalent sound level, Ii ._ , over the time
24
-------
period of interest (T) , for example one hour or eight hours, (Leq(i) and
The second kind of situation covers those in which the noise is not
present for enough of the day to greatly affect the L, reading, but is
intrusive and disruptive when it is present. Examples of such noise sources
may include motorcycle passbys, trains, and specific aircraft flyovers. The
appropriate sound measure for such an event is the cumulated sound produced
by the single event, the A-weighted sound exposure level, L. It is a -
measure of accumulated, not average, sound energy.
Precise mathematical descriptions of all these measures are provided in
Appendix A. All are expressed in A-weighted decibels; the reference sound
pressure is 20 micropascals. •
2.1.2 Determining the yearly day-night sound level
For the screening procedure, two yearly L, values need to be deter-
mined: the existing levels; and the levels expected to be caused by the
proposed project. In addition, for the impact assessment it will be necessary
to estimate the future yearly L, values in the area if the project is not
constructed. (The total post-project noise level can then be calculated as the
(logarithmic) sum of the project levels and the future levels in the absence
of the project.)
Determining Isfri) by direct measurement. To establish the existing noise
exposure accurately, field measurements are oftentimes the preferred
approach. Unfortunately, such measurements can be expensive and time-consuming.
Nonetheless, measurements may be warranted. For example, if the present
average sound levels are already high, so that the noise impact of a new
project will not be much greater, or may be even less than the impact from the
25
-------
existing noise environment, it may behoove the applicant to conduct a
measurement program, so as to predict the noise impact more accurately.
When an existing noise environment is to be determined by direct measure-
ment, it will be necessary to make measurements at a number of locations suffi-
cient to establish a credible baseline for estimating total impact. The number
of measurement locations and their geographic disposition will depend on the
spatial extent of the impact expected to be produced by the project.
Measurement periods and the time intervals between them should be
determined by the characteristics of the existing noise, in order to obtain
a reliable estimate of yearly L, . If the existing noise is expected to be
substantially the same from day to day, measurements during a single typical
24-hour period may be adequate. Locations where the noise is caused primar-
ily by well-established motor vehicle traffic patterns are an example. In
other situations where strong daily, weekly, monthly, or seasonal effects
occur, it may be necessary to measure for a number of different daily periods
suitably chosen to account properly for these variations. In some particular
situations, the variations may be large enough to make measurement practi-
cally infeasible. A case in point might be in the vicinity of an airport
with more than one runway which has no on-going noise monitoring program.
The most reliable temporal data are obtained by techniques that approach
continuous measurement of the sound level over the time period in question.
In some instances it may be reasonable to obtain measurements over only
fractions of the total time—e.g., several minutes per hour. However, any
measurement method used to approximate continuous measurement of L, must be
justified by adequate technical reasons and data to show the accuracy of the
procedure when applied to the specific noise sources affecting the noise
environment being described.
26
-------
If field measurements are undertaken, they should be conducted in
accordance with accepted procedures [8] .
Determining I^n by the use of engineering prediction models.
Several kinds of noise have been extensively studied, particularly the
noise of transportation, and procedures have been developed for calculating
day-night sound levels based on the type of noise source and operational con-
siderations. Procedures for estimating the noise of specific sources such as
roadways [9,10,11] and aircraft near airports [12,13,14], are available and
may be easily adopted for those situations in which the existing noise envi-
ronment is dominated by a major noise source. A partial bibliography of some
of these engineering prediction models is included at the end of Appendix E.
Procedures for other sources, such as railroads, have been developed and will
be published shortly.
Determining ^jn from the population density. Where no dominant
source of this nature is present, the existing noise environment may be
considered to be caused primarily by local automotive traffic noise. For
these instances the day-night sound level may be estimated on the basis of
population density in accordance with the values listed in Table 2, which are
based on the equation:
Ldn - 10 log p+ 22 dB Eqn. 1 P
where p is the population density in persons per square mile. This relation-
ship was derived from measurements at 130 urban locations [15] . The equa-
tion has a standard error of 4 dB, which means that the 95% confidence
interval around the estimate is roughly +8 dB. The reliability of the
relationship is approximated by the correlation coefficient of 0.723 between
L, and the log of the population density over the 130 data points. This
27
-------
can be interpreted as indicating that the log of the population density
explains 52% of the variation in Ldn. This amount of uncertainty about
the true L, may or may not be acceptable for a given project. If it is
not, measurements or source-based predictions are recommended.
The levels shown in the table represent average values for residential
areas that are not in the vicinity of an especially noisy existing source
such as an airport, a freeway, a railroad, or a switching yard. If such a
noise source exists, its contribution to the existing L, should be
estimated separately, and then combined with the level given in Table 2. The
values in the table are representative of space average values over areas of
2
the order of 1 km (0.4 sq. mile), or larger, for typical urban conditions.
For purposes of estimating the existing noise in relation to permanent
changes in areas with population density greater than 20,000 persons/sq. mile,
the day-night sound level should be taken as 65 dB. Higher estimates of the
background noise by the use of equation 1 require specific justification such
as direct measurements or detailed calculations based on existing noise sources.
The reason for this suggestion is to avoid obtaining low numbers for the impact
TABLE 2
Yearly Day-Night Sound Level As Estimated by Population Density
(To be used only for residential neighborhoods where
there is no well-defined source of noise)
Population Density L. in dB Population Density
Description (People/Sq. Mi.) "" (People/Sq.km.>
Rural (undeveloped) 20 35 8
Rural (partially developed) 60 40 23
Quiet Suburban 200 45 77
Normal Suburban 600 50 232
Urban 2000 55 772
Noisy Urban 6000 60 2317
Very Noisy Urban 20000 65 7722
Note: Iyn estimates for population densities lower than 1000 persons/
sq. mi are extrapolations.
28
-------
of noisy projects in heavily populated areas. This is in line with the
discussion of accounting for existing noisy areas when using the screening
diagram. Particularly when an area is noisy because of high population
density it is important to consider very carefully, any project which will add
to the noise in the area, and therefore not to initially screen it out by
using high estimates of existing levels.
With respect to problems of estimation in rural areas, there simply is
not enough known about noise levels in such areas, since measurements such as
those used to calculate equation 1 are routinely conducted only in the absence
of wind, rain, and other natural sounds. Values obtained using equation 1 (or
Table 2) extrapolate beyond the data base, and should be used with caution.
Whenever possible, measurement of existing levels is recommended.
Estimation of future noise levels with and without the proposed project.
Most of these procedures which have been identified for estimating
existing noise levels can also be used, as appropriate, to estimate the
noise levels due to the proposed action or project. Prediction procedures,
approved by various federal agencies, are available for a number of typical
situations, including aircraft, motor vehicles, railroads, construction equip-
ment and other noise sources. In some instances those procedures do not pro-
vide predictions for L, . In those cases, the conversion equations provided
in Appendix B can be used to estimate the L, values, in order to use the
information in these guidelines to express the impact of the noise on people.
If prediction techniques are not available for a particular project, for
example a specific industrial installation, measurement of noise levels at a
similar existing installation is appropriate, although an engineering descrip-
tion should be included of the reasons for anticipating such similarities in
a new installation. Likewise, where the introduction of a new noise source
29
-------
is anticipated and neither an existing approved procedure nor a similar
installation is available, an engineering description of the procedure
employed to estimate noise emissions should be provided in adequate detail
for technical evaluation of its acceptability.
For predicting future noise levels in the absence of the proposed proj-
ect, the available methods are (1) extrapolation of existing levels to the
future, (2) the use of source-specific prediction techniques, or (3) the use
of equation 1.
2.1.3 Determining the population affected by the noise of the proposed
project
For each of the alternatives that involves the introduction of some form
of a new noise source, the affected population is defined as that population
experiencing sound levels produced by the new noise source above a specified
yearly L- . Ihis will be called the yearly L, , or base *v. . .. Hie
base yearly L, may be determined by references to existing yearly L, con-
tours in the area of interest (See section 2.1.2). Consistency with the
screening diagram requires the consideration of impacts whenever the* overall
post-project level is greater than the pre-project level, that is, when the
project alone L. is greater than pre-project L, less 10 dB. Thus, the
base L, will usually be 10 dB lower than the existing (pre-project) yearly
L. . For example, if the existing yearly L. is 60 dB, it is suggested
to start with a base L,. . of. 50 dB, if possible, in order to determine
the number of people affected by the project noise. In some instances, however,
it will not be feasible to predict project noise levels to such low values.
(An example of such an instance would be around commercial airports, where
existing prediction techniques are not particularly reliable below an L, of
65 dB (NEF 30), due to lack of information about aircraft flight track usage.)
30
-------
In such cases, it is still imperative to consider as large a range of
levels as is feasible. A difference of twenty decibels between the maximum
Ldn(v) for the P0^64* an^ the base Ldn(v) is a good ran9e to attemPt to
achieve, providing that it results in a base L,,. of 55 dB or less
since L^ = 55 dB has been identified as a point of significant adverse
noise effects. If the procedure results in a higher L, . . , a base
level of 55 dB should be chosen.
When several alternatives are compared, a common base area or base
population must be used for all alternatives. In such cases the base area
or population for all alternatives will be the largest area or population
affected by any alternative. In other words, the base population will
be determined by the project which has the highest yearly L, in a given
location in a given year. The reason for requiring a common base is that
several of the measures of relative impact, to be discussed subsequently,
will be meaningless if the total number of people over which they are cal-
culated changes from one alternative to the next. The base population,
therefore, should include all people who are affected by the noisiest alterna-
tive. As a consequence, for some of the less noisy alternatives the base
population will be considerably larger than the population actually affected
by those alternatives. If the base L^. . is consistent with the screen-
ing diagram, no person exposed to project noise levels less than the base
Ldn(v) wou^ be rsgarded as impacted.
There are cases when, over time, people will move into or out of a pro-
ject area at the same time the project is expanding and environmental noise
levels are increasing. Such changes in population may be entirely unrelated
to the project under analysis. In these cases it may be necessary to define
31
-------
several base populations or base areas, one for each year of interest. (See
Appendix E, section E.2 for an example of this type of analysis.)
There are actions that do not add new noise sources, but only change the
noise output of existing sources. In these cases, the changed source should
be treated as a new source for purposes of determining the base population.
There are actions that will neve people into noisy areas. For these cases,
the base population will be that population that is moved into an area in which
the existing yearly L^n is greater than 55 dB.
There are actions which affect large segments of the population that are
not easily related to specific areas. Laws and regulations that directly
affect mobile noise sources are examples of such actions. For actions affect-
ing regulation of noise sources in general, the base population might best be
described as the total population experiencing day-night sound levels
above 55 decibels from such sources. For actions affecting source control
for equipment operators, the base population might be only the users of the
specific noise source. In the final analysis, the preparer of a noise
impact analysis must use his or her judgment. In all cases, an explanation
should be included in the final report of how the base population was determined.
Population estimates for residential areas identified in the analysis
may be taken directly from census tract data, local master plans, or by
counting residential units identified on aerial photographs of the area.
Non-residential populations may be estimated from industrial, commercial,
or public facility employee statistics; student enrollments and employee
statistics can be used to estimate school populations. Population esti-
mates should strive to identify total populations within ±10 percent of
the true population. One way to deal with uncertainty in predicting future
32
-------
populations is to use the local community's land use plans or zoning designations,
whenever they exist, to estimate the most likely future population density.
2.2 Health and welfare effects range
This section deals with the most commonly encountered noise problems,
the general health and welfare effects of noise due to the noise environment
encountered in most urban and suburban areas. Those effects are the major
concern at yearly L, values which range approximately from 55 dB to 75 dB.
Above 75 dB, severe health effects need to be considered in addition to the
effects considered here (see section 2.3.). The first sub-section describes
the health and welfare criteria which apply in this range of general audible
noise, and the second covers the procedures to be used to quantify those
effects.
2.2.1 Human noise exposure criteria
As the primary criterion for evaluating the impact of noise on people, the
effect on "public health and welfare11 was selected in the Levels document [2].
Interference with speech communication, with general well-being, and with
sleep are related to the general annoyance produced by the noise environment,
and were accepted as indicators of effects on public health and welfare. The
same criteria are proposed here as the basis for environmental impact assessment.
A summary of the expected effects of noise on human activities for outdoor
yearly day-night sound levels of 55, 60, 65, 70, and 75 dB, in terms of
health effects, interference with speech communication, community reaction,
annoyance, and attitude towards the area is provided in Appendix C. Basic infor-
mation in these tables on speech intelligibility and general community reac-
tion was derived from the Levels document [2]. The relationships given in the
33
-------
Levels document between noise and annoyance have been modified in the light of
a substantially increased set of data subsequently available [5]. These tables
allow the preparer of a noise analysis to make an explicit statement as to the
expected impact of any day-night sound level.
In order to achieve the simplicity which these guidelines are intended
to promote, it is desirable bo be able to summarize these several health and
welfare effects with a single indicator. The response of interest is the
general adverse reaction of people to noise, which includes speech interfer-
ence, sleep interference, desire for a tranquil environment, and the ability
to use the telephone, radio, and television satisfactorily. A measure of this
response is the percentage of people in a population that feels high annoy-
ance about noise of a specified level. High annoyance is selected on both
theoretical and practical grounds. First, it arises as a consequence of
the activity interference and interruption caused by noise [16], and there-
fore summarizes all the effects better than any one of the direct effects
would. Second, there is available a large set of data which allows reponse,
expressed as percentage of a population highly annoyed, to be characterized
by a single functional relationship of the noise environment [5].
The percentage highly annoyed is used rather than the percentage at all
annoyed for a number of reasons [5, pp. 378-379]. Perhaps the most important
of these is that when people are highly annoyed by noise the effects of non-
acoustical variables are reduced, and the correlation between noise exposure
and the expressed subjective reaction is high. This is not to say that all
individuals have the same susceptibility to noise; they do not. Even groups
of people may vary in their response to noise, depending on previous exposure,
age, socio-economic status, political cohesiveness and other social variables.
34
-------
In the aggregate, however, for residential locations, the average response of
groups of people, as measured by the percentage highly annoyed, is quite stably
related to cumulative exposure to noise as expressed in a measure such as L,.
For schools, offices, and similar spaces where ease of speech communica-
tion is of primary concern, the same relationship can be used to estimate
the potential average response of people, as a group again, ignoring individ-
ual variations from person to person.
Data used to relate annoyance to noise environment in the Levels document
[2] was based on two social surveys around airports in the United States and
England. Data have now been analyzed from 19 social surveys (in 9 countries)
associated with aircraft, urban traffic, freeway traffic, and railroad noise
[5]. These data allow a much more definitive relationship to be developed
between percentage of the population highly annoyed and average noise level.
The data support the previous assumption that the statistical relationship
between population annoyance and noise level is essentially independent of
the type of noise source [17].
The results of this synthesis show quite clearly that the best fit of
response data to average sound level is provided by a curvilinear function;
usually a cubic equation was used in the regression analyses. Further, 12
of the surveys, covering aircraft, railroads, urban traffic, and expressway
traffic as noise sources, "clustered" closely around an average curve for the
set of data, as shown in Figure 4. The remaining 7 surveys showed similarly
shaped annoyance/sound level functions, but deviated in differing detail from
the 12 clustering surveys for various qualitative reasons [5]. It is worth
noting that the average of the non-clustering surveys was essentially the
same as the average for the clustering surveys.
35
-------
Q
ui
O
100
90
80
70
60
5 50
i
40
cc
ui
a.
30
20
10
-~*~» 1ST HEATHROW A/C (1961)
FRENCH A/C (1966)
MUN
PARI
SWEC
***»< SWIS.
LONC
SWIS!
FREf\
oo««» U.S. S
• LAX
w^
^
» •^^rf~J^- *
HEATHROW A/C (1967)-
ICH A/C (1969)
S STREET (1969)
)ISH A/C (1972)
5 ROAD (1972)
30N STREET (1972)
> A/C (1973)
CH RR (1973)
TREET(1974)
1974)
^
*^M
$&*
^\tf{
o'%$>
S&SS^
w
^
$r
A
1 /M . •
^'
^•••<
I
M
$/
Mf
f/:
/
// ./
//:/
%
/
1
/
/
40
50
60 70
Ldn (DECIBELS)
80
90
FIGURE 4. SUMMARY OF ANNOYANCE DATA FROM
12 SURVEYS SHOWING CLOSE AGREEMENT
SOURCE: SCHULTZ[5]
36
-------
Based on these data, Schultz proposes the following equation "as the
best currently available estimate of public annoyance due to transportation
noise of all kinds" [5, p. 382], relating percent highly annoyed, (%HA),
to day-night sound level:
%HA = 0.8553 L^ - 0.0401 L^2 + 0.00047 L^3 Bqn. 2a
This expression represents the least-squares fit of percent highly annoyed
to day-night sound level for the clustering survey data.
A second form of this equation, based on two power law functions, is
preferred on the grounds that it suggests an explanation for the behavior
represented in equation 2a. At lower levels, the first power function
represents increasing awareness, or arousal. At higher levels, annoyance
increases at the sane rate as the well-known loudness function, represented
by the second power function. The smoothed version of the function based
on the two power law functions is expressed as*:
(1.24 x 10~4)
%HA
(1.43 x 10"4) (10°'08I<3n) + (0.2) (IQ0'031^") Bqn. 2b
In the absence of any studies relating average response to noise level
for non-transportation sources, equation 2b has been adopted in these
guidelines for use as the criterion for all noise sources. If information
'Another very useful and simpler expression which approximates the annoyance
function is:
%HA = —
i j. J10-43 - 0.132 L. )
1 + e dn
This expression has the particular advantage of not allowing predicted values
to go below zero percent or above 100 percent.
37
W "A
-------
becomes available which identifies a different relationship for certain
sources, the guidelines will be revised accordingly.
This function for the percentage highly annoyed differs from previously
suggested equations, including the one in the Levels document [2, p. D-27],
as is illustrated in Figure 5. The relationship shown in the Levels document
was taken from a study by an EPA Task Group under the EPA Aircraft/Airport
Noise Study 'in 1973 [4]. In this study, social survey data from the first
study around Heathrow airport in England [18], and from the Tracer study of
U.S. airports [19] were combined to develop a relationship between "percent
highly annoyed" and day-night sound level. This function was expressed as:
% Highly Annoyed = 1.8 (L^ - 46) Eqn. 3
The Task Group also noted a similar relationship developed in an OECD study
[20] that used the relationship:
% Highly Annoyed = 2 (L^ - 50) Eqn. 4
This equation was also based on airport noise studies. The primary reason
for these differences is a redefinition of what is meant by highly annoyed.
In fact, the Heathrow study is included in the clustering surveys (Figure
4). As Schultz's paper makes clear [5, pp. 391-392], the annoyance scale
used in the first Heathrow study requires some interpretation: it is not
a direct question about degree of annoyance. The earlier analysis (Eqn. 3)
considered the top three scale points as highly annoyed; Schultz used only
the top two. His discussion is persuasive, and his function has been adopted
for these guidelines.
It is important to point out that this redefinition of annoyance does
not affect the conclusions reached in the Levels document, because that
document relied on speech and sleep interference indicators to identify the
38
-------
80 i-
60
O
UJ
>
O
I
O
X
40
20
40 •
LEVELS DOCUMENT
(HEATHROW & TRACOR)
/
/
X
OECD
NEW ANNOYANCE FUNCTION
50 60 70 80
DAY-NIGm SOUND LEVEL-DECIBELS
.90
FIGURE 5. COMPARISON OF NEW ANNOYANCE •
FUNCTION WITH PREVIOUS FUNCTIONS
39
-------
actual levels which were "requisite to protect the public health and welfare
with an adequate margin of safety." That approach led to the statements that
a day-night sound level of 55 decibels in residential areas will result in
negligible impact on public health and welfare and that the degree of impact
will increase as the day-night sound level increases. The EPA levels document
asserts that no significant effects on public health and welfare occur, for
the most sensitive portion of the population and with an adequate margin of
safety, if the prevailing day-night sound level is less than 55 decibels.
The difficulty with using annoyance for such a calculation is obvious in
Figure 5: there are still some people affected at sound levels as low as 45
dB (La). These guidelines, then, use as the criterion in populated areas
the function given by Eqn. 2, which shows some impact at levels as low as
45 dB, impact which is fairly low into levels in the low 60's (dB), and
impact which begins to increase fairly rapidly above 65 or 70 dB (L^J.
For those events in which the single event measure, sound exposure
level, is used to describe the noise environment, the previous discussion
will not apply. Information characterizing response as a function of sound
exposure level is not readily available. Some information can be approxi-
mated for sleep interference [21,22,23,24] and speech interference [22,24],
but it is not as easily dealt with as is the information on L, .
2.2.2 Quantification of the noise impact,.
The impact of a noise environment on people regularly experiencing the
environment is the degree to which the noise interferes with various activi-
ties such as speech, sleep, listening to radio and TV (i.e., the peaceful
pursuit of normal activities), and the degree to which it may impair health,
40
-------
through, for example, the inducement of hearing loss. Sound levels produced
by sources being considered in an environmental assessment will generally
vary with distance from the source, sometimes over a large geographic area.
As a consequence, people occupying different geographic areas will experience
different sound levels. The total impact of a particular noise environment
is a function of both sound level and the size of the population experiencing
a particular value of sound level.
The first step in describing the noise impact of an action is to tabulate
the number of people regularly experiencing various sound levels. In many
cases, particularly those in which noise impacts must be compared with a vari-
ety of other costs and benefits, such a tabulation is insufficient, because it
contains too much information. In those cases, it is desirable to derive a
single number which represents quantitatively the integrated impact of the
action on the total population experiencing the different sound levels. This
single number quantification is defined below as the sound level weighted popu-
lation, LWP. Sound level weighted population, together with the tabulations of
populations experiencing sound levels of a specified value, constitute the mini-
mum quantification of environmental impact of noise recommended in these guide-
lines. This subsection describes procedures for preparing the tabulations, and
for calculating the sound level weighted population. It also describes a useful
second descriptor of noise impact, the noise impact index, Nil, which is formed
by the ratio of sound level weighted population to the total base population.
The procedures proposed here do not rely on establishing specific criterion
noise levels for different land use categories. (For information on criterion
levels suggested by different organizations, see Appendix B.)
41
-------
a. Necessary tables - As a minimum the data characterizing the noise
impact should be tabulated in a set of summary tables. Typical tables are
included in the example in section 2.6 (Tables 6 through 9). For a given year
the areas and population are to be listed against the yearly day-night sound
level at increments not greater than five decibels, for the following esti-
mated noise environments:
(1) without the project's existence;
(2) due solely to the project action;
(3) due to all sources including the project action.
'All three tables may not always be necessary, especially if there are insig-
nificant differences between any two of the tables.
If the tables are properly constructed, the total population and/or land
area for each of the three conditions will be equal (i.e., will equal the base
population or area defined in section 2.1.3). The tables should include enough
increments of yearly L, that all residential populations, industrial, commer-
cial land and special situations experiencing L, values above the base
L^ are included.
The column headings might typically include: total land area, industrial/
commercial land area, residential land area, industrial/commercial employees,
residential population, and special situations. Depending upon local condi-
tions, different classifications of land use may be appropriate. Industrial/
commercial land area is meant to include all land not considered as residential
or associated with special functions. This land area would include farm land,
undeveloped land, industrial plants, and similar uses. Depending on local
plans, this category may be further broken down. Residential land includes
all land associated with a residential population. It may include land actually
42
-------
zoned oonntercial or industrial. For residences on farm lands, approximately
1 acre should be considered as residential land for each separate residence.
Special situations are those situations which must be highlighted or
treated separately in order to represent the impact properly. Situations
of this category can be religious facilities, outdoor auditoriums, schools,
precision laboratories, hospitals, etc. The detail to which each special
situation should be discussed will depend on the size of the project and the
size of the area being evaluated. Special situations should be combined as
necessary to keep the total number of special situations within reason (nor-
mally less than 20 or 30 items). One useful approach to the listing of spe-
cial situations is to number each one, and then to use this number in the
special situation column to indicate the corresponding L. for that
situation (see'examples, section 2.6, Tables 6 through 8).
If there are more than a few special situations, an additional table
summarizing them will also be useful (Table 9 in the example in section 2.6).
This should list the number of exposed people for each situation. At some
locations the population does not remain constant from day to day, week to
week, or month to month. Examples of such places are churches, parks, and
stadiums. In such situations the population entered in the special situation
table is the time-weighted average number of people present during the year.
This number should be calculated by summing the products of the number of
people using a facility, multiplied by the number of hours these people are
present in the facility during a year, and dividing by the total number of
hours in a year. If a noise measure other than yearly L, is being used,
the average number of people can be calculated similarly for that time period.
43
-------
such as the working day for office buildings. The concept of average number
should not be used for residential areas.
Formats other than that used in Tables 6 to 9 may be appropriate and
may be used; however, the information conveyed to the reader should be
effectively the same as or greater than is contained in these tables.
For each alternative of a permanent project or action, a separate set
of tables as outlined above should be prepared for (1) the first year of the
commencement of the project, (2) the last year before the end of the project
(or at the 20 year point, whichever is shorter), and (3) the worst case year
if such a year is not the first or last year. In many cases, only one of
these sets will be necessary because the conditions with respect to time
can be expected to remain reasonably constant. By "reasonably constant," it
is meant that the change in exposed population will be small enough so any
resulting errors are consistent with the error in the overall analysis.
In addition to the tables, it would be helpful to present a map or
drawing of the area including surrounding facilities such as airports, fac-
tories, highways, or electrical plants, with contours representing constant
values of yearly day-night sound level. In general, the decibel increments
between contours should be consistent with the tables as discussed above.
Other contours may be presented as needed. There should be a set of contours
for each of the alternatives studied.
b. Sound level weighted population - For those projects in which it
is necessary to compare or trade-off noise impacts with other costs and bene-
fits of a proposed project, a compilation of the data characterizing noise
impacts into the tables described above will usually not prove sufficient.
The tables contain too much information for easy comparison to be possible.
44
-------
A single number representation , combining the extensity (number of people
exposed) and intensity (severity of the exposure) of the noise impact is
desirable. Using the criterion function based on the percentage highly
annoyed (Eqn. 2), described in section 2.2.1., such a single-number index
can be constructed which summarizes the impact in terms of the total number
of people who respond adversely to the effects of noise.
Several assumptions are made in this method of analysis.
( 1 ) The intensity of human response is a measurable consequence of
equivalent sound level, and in the noise range of interest here (namely that
generally encountered in populated areas), is appropriately measured as the
percentage of the population which is highly annoyed.
(2) When measured this way, it is clear that the impact of high
noise levels on a small number of people is equivalent to the impact of
lower noise levels on a larger number of people in an overall evaluation,
when both yield the same number of people responding adversely. Thus the
properties of intensity (level of sound) and extensity (number of people
affected by the sound) can be combined mathematically.
(3) On the basis of these two assumptions one can ascribe differing
numerical degrees of impact to different segments of the population of con-
cern, depending on the equivalent sound level, and can sum over all of these
segments to obtain the total impact (total number responding adversely).
On the basis of these assumptions, the following equation is obtained
for the sound level weighted population, UWP:
.
r- LU/P
f Pt^Jn) • W^) ddfln) Eqn. 5 ^
>
where P(I^n) is the population distribution function, W(I^n) is the weighting
function described in Equation 2b, characterizing the severity of the impact
45
-------
TABLE 3 '
Values of the Weighting Function for General Adverse Response*
= (0.01)%HA]
Ldn W
-------
as a function of day-night sound level (Table 3), and dfl^fa) is the differen-
tial change in day-night sound level. Although Table 3 contains values for
I^ln as low as 35 dB, the values below 45 dB should be used only with great
caution, as they represent extrapolation beyond the range of the data [5] .
In any event, for most projects in populated areas, the future noise level even
without the project will probably be considerably higher than the 45 dB limit
of the data.
It is usually not necessary or possible to use the integral form to com-
pute LWP. Sufficient accuracy is obtained by taking average values of the
weighting function between equal decibel increments, up to 5 decibels in
size, and replacing the integrals by summations of successive increments in
average sound level:
U*P=I>
-------
d. Change in level-weighted population and relative change in impact -
A primary concern in an environmental noise assessment is a comparison of ( 1 )
the effect of the action on the noise environment with (2) the environment
before the action was to take place. Two additional descriptors of this
change due to the action are useful. The first descriptor is simply the
numerical change in sound level weighted populations before and after the
action, the change being an increase or decrease in sound level weighted
population (or the neutral effect case, no change). The second descriptor
is the relative change in impact (PCI), where the effect of the action is
expressed as the value of the change in the sound level weighted population
after the action, divided by the sound level weighted population before the
change:
a - LWP. __ ft
RCI = - *• - £ Bqn- 8
IWP.
D
where LWP is the impact after the action or project is in place, and LWP.
cl D
is the impact before the action is taken.
e. Level-weighted area - In those rare cases where it is known that an
area will be developed, but there is no information with which to predict the
future population, it may be necessary to calculate a level-weighted area, as
a proxy for the population impacts. Such a calculation would be equivalent
to that for level-weighted population (equation 6), but would use the tabula-
tion of area within decibel intervals, rather than population, assuming, in
effect, a constant and undefined population density.
f . General discussion - A number of different noise impact descrip-
tors are available, based on the four single-number indexes — level-weighted
population, noise impact index, change in level-weighted population, and
48
-------
relative change in impact—and the three noise characterizations—the project
alone, the environment without the project/ and the total future noise envi-
ronment obtained by combining the other two. Die result is almost a confu-
sion of supposedly simplifying descriptors. Two or three of these will be
most useful in each case, depending on the relationship between project levels
and expected levels without it (Figure 6). Where existing levels are already
high, the level-weighted population or noise impact index based on the project
noise alone is suggested as the best descriptor. The other descriptors will
minimize the impact by putting it on relative terms. Where project levels
are much higher than existing levels, the project levels will dominate the
combined levels, so either will give the same result. Where project levels
are similar to existing levels, it is necessary to use the combined levels to
identify the full impact.
For projects which will move people into areas with Lfa values above
55 dB, and for projects which reduce noise, Figure 6 is not applicable. Pro-
jects, such as housing developments, in areas with 1^ above 55 dB need to be
evaluated in terms of the level-weighted population or noise impact index
based on the non-residential noises to which they will be exposed (e.g. road
traffic or aircraft noise). The only basis for calculating the change in
level-weighted population or relative change in impact which might be useful
in such a situation is one based on the national average Nil, which has been
calculated to be 0.35. Projects which reduce noise, on the other hand,
should be evaluated on the basis of the change in level-weighted population,
or relative change in impact. Since the project is proposed to reduce noise,
it is obviously the reduction or change which is of interest.
49
-------
' 1
UJ
Z
0
^ tf\
•»- 70
u
UJ
0
£ 60
u.
O •
5 50
•o
Q
UJ
H- 40
U
UJ
I
LWP (COMBINED NOISE LEVELS)
^^ ^^ ^» m^ «^ ^^
/'
~~ OR Nil (COMBINED NOISE LEVELS) /
RCI (COMPARING COMBINED
LEVELS WITH Ll.VELS
EXPECTED WITHOUT
~ THE PROJECT) /
/
/
"*"* J
/
7
/
— SPECIAL /
ANALYSIS /
°- / X
x 4 X
"J LJ^. I/ 1 1
S LWP (PROJECT ALONE)
X OR
. ' Nil (PROJECT ALONE)
X
ALL PROJECTS
SCREENED OUT
111 w^
40 50
60
70
80
90
EXPECTED FUTURE Ldn(y) WITHOUT PROJECT
FIGURE 6. SUGGESTED DESCRIPTORS FOR VARIOUS SITUATIONS
-------
Relationships between annoyance and average sound level have been used
previously to define a weighting function for numerical evaluation of impact
analyses. It is useful to compare the present function (Eqn. 2 and Table 3)
to the one used earlier by EPA, which was first introduced in the fractional
inpact method developed initially for use in the analysis of highway noise
problems [25]. This method took into account the data and recommendations
both of the EPA Levels document [2], and of the earlier report on Impact
Characterization of Noise [4], which indicate that a community would not be
expected to exhibit significant reaction at noise exposures of L, = 55 dB or
below, but would be expected to show strong, organized reaction at L^ = 75 dB
and higher. Using these two anchor points, and the linear relationship of
Equations 3 and 4, a weighting function, "fractional impact,11 F.I., was defined £" T"
to be zero at L, = 55 dB, and unity at L, = 75 dB, varying linearly with
average sound level, such that:
P.I. = 0.05 (L^ - 55) Eqn. 9
The weighting function for F.I. has been used by EPA in impact analyses of
a number of potential regulatory actions.
Several features of equation 9 are unsatisfactory. It is not likely
that community response is adequately described with a linear function of
average noise level over a wide range of levels. Even though the data from
the individual social surveys are reasonably well fitted by linear regres-
sions over the limited range of levels represented in the separate surveys,
the individual survey results indicate that the rate of change of annoyance
with sound level is greater at higher sound levels than at lower sound levels.
Moreover, the choice of an arbitrary zero at L, = 55 dB is not easily justi-
fied. Finally, few data from noise sources other than aircraft were available
51
-------
at the time the original weighting functions were developed, and a weighting
function derived only from aircraft-related social surveys may not be satis-
factory for use in evaluating other sources of noise.
Despite these flaws, however, this linear function is quite similar in
its relative ratings to the curvilinear function used in this document (Egn.
2). If the two functions are placed on the same scale (Figure 7), it can be
seen that, in the day-night sound level range of 55 to 80 decibels, this
linear weighting function will generate relative values for level weighted
population that differ only by the order of one percent from the curvilinear
weighting function in many applications. The change in scales necessary to
make this comparison stems from the fact that fractional impact was defined
to be unity at L, = 75 dB, while the present function is based on the
number of people reporting a high degree of annoyance in a social survey
situation. Both are equally legitimate interpretations of available impact:
the first provides an indicator of absolute impact, while the second is more
easily understandable in comparisons with other costs and benefits of pro-
posed projects. Because the linear function (Eqn. 9) closely approximates
the curvilinear relationship (Eqn. 2) between the day-night sound level range
of 55 to 80 dB, the user may wish to employ the more simple linear relation-
ship in some cases.
2.3 Severe health effects range
In some high level noise environments people will be exposed regularly
to 24-hour equivalent sound levels in excess of 70 decibels. In these environ-
ments special consideration should be given to the potential for severe health
effects. This section discusses the criteria to be used for describing severe
52
-------
Q
ui
O
X
O,
5
10Q
90
80
70
."60
UJ _s>«.
O
DC 20'
UJ -••
Q-:
-' -Tdi
(0.000124HO°-103LdiLi
0.000143) 10°-08Ldn ^
F7j.°O.OS(Ldn-5!
80 .90
Ljn ('DECIBELS)'.
FIGTURE'TTJCOlyTP^RTSlO'lNrO'FlCURVIU_NEAR FUNlCTION
"Afro~FRA~CTIONAL
53
-------
health effects, and then describes a procedure for calculating a single number
index, analogous to the level-weighted population index, for statistically sum-
marizing expected severe health effects.
2.3.1 Human noise exposure criteria
The discussion of severe health effects in an environmental analysis
is meant bo supplement the discussion of general adverse effects, not to
replace it at high noise levels. The general adverse effects - speech inter-
ference, sleep interruption, annoyance - continue to be present at high noise
levels, and in fact increase more rapidly at the higher noise levels. Equa-
tion 2 (or Table 3) can be used to summarize these effects for 1^ values as
high as 85 dB. These effects, however, only include those of which people
are aware, and which have been articulated in attitudinal surveys. In many
instances, people are not aware of the potential severe health effects which
long-term noise exposure can cause. Hence a separate discussion of severe
health effects is necessary, which helps to emphasize the severity of the
problems caused by high noise levels.
Noise-induced hearing loss can begin bo occur at high noise levels. Other
noise-induced physiological effects and/or changes may occur. However, a firm
causal link between community noise and extra-auditory disease has not been
established at this time. Therefore, this document proceeds on the assumption
that protection against noise-induced hearing loss is sufficient to protect
against severe extra-auditory health effects.* However, one has to keep in
mind that as the noise level increases above the threshold for severe health
This is not to say that non-auditory physiological effects do not occur at
levels below those sufficient bo protect against hearing damage. In any event,
rigorous causal links between noise and extra-auditory health effects have not
yet been firmly established, but await further study.
54
-------
effects so does the probability that other health effects in addition to noise-
induced hearing loss might become important. The adverse effect of noise on
hearing rapidly accelerates as the noise exposure increases and it is reason-
able to use expected noise induced hearing loss as a basis for assessment of
severe health effects.
A problem arises in specifying the noise measure to be used when quanti-
fying severe health effects. If hearing loss is used as the indicator, the
noise measure needs to reflect at-ear measurement to be valid. Further, hear-
ing loss is properly expressed as a function of L , rather than of L, , but
it will not usually seem warranted to calculate and draw noise contours for
more than one noise measure. The data to be discussed below predict noise
induced permanent threshold shift (NIFTS) for 8-hour equivalent sound levels
(at-ear) starting at values of 75 dB. If the remaining 16 hours of the day
are spent in a noise environment of 70 dB L or lower, the at-ear 8-hour
equivalent sound level of 75 dB results in a 24-hour long L of 70 dB, at
the ear. For many proposed projects (particularly actions where assessment of
hearing damage risk is of primary concern) it will be appropriate to use one of
these two measures.
It is also important to be able to identify the 1^ values at which it
is appropriate to look for severe health effects. Those persons who have the
greatest outdoor activity, including young children, retired persons living
in warm climates, and people in certain outdoor occupations, are clearly the
people of major concern when outdoor 1^ is considered. For outdoor exposure,
daytime levels are the important ones for establishing at-ear values. The
values of 1^ corresponding to an A-weighted equivalent sound level of 75 dB
during daytime hours range between 73 and 81 dB. The lower value corresponds
55
-------
to a situation where the equivalent sound level during the night is 20 dB or more
lower than that occurring during the day, whereas the higher value corresponds
to the situation when the equivalent sound level during the night equals that
occurring during the day. The most probable difference between the daytime
and nighttime values of I^g is 4 dBf as shown for the noise levels of interest
in Fig. A-7 of the Levels document [2]. For this day-night difference, L^n
is three decibels above the daytime value of I^q, that is, L^ = 78 dB. This
value of 78 dB is considered to be the most probable value of L^n to be found
in real environments that have a daytime I^g of 75 dB. (This estimation is
based on that in reference 4, pp. B-8 - B-9.) However, due to the wide range of
possible values, it is recommended that an outdoor 1^ of 75 dB be used as the
threshold above which severe health effects are investigated. This has the
advantage of being an Lfa value for which contours will already be mapped, and
is therefore information readily available.
Consequently, for areas with L^n of 75 dB or above, it is important to
look for potential severe health effects. The way to do this is to estimate
the size of the population spending time outdoors, the length of time they
are outdoors, and the actual levels while they are outdoors. The last two of
these numbers can then be used to estimate the at-ear 24-hour Lgq for these
people (using the equation for I^g in Appendix A). As long as the outdoor
noise exposure exceeds 3 hours per day, the contribution of the indoor noise
environment may be neglected in computing the 24-hour I^g. This conclusion
does not depend greatly on the actual noise attenuation provided by the house
so long as the attenuation is greater than 10 dB [4, p. 8-9].
There have been numerous studies conducted for the purpose of determining
the long term effect of noise on the hearing ability of an exposed population.
56
-------
In particular, three studies [26,27,28] have provided reasonable predictive
models of the relationship between noise exposure and changes in the statistical
distribution of hearing levels of the exposed population. These changes are
called Noise Induced Permanent Threshold Shifts (NIPTS). The results of
these three studies were combined [29] and used in the EPA Levels document
[2, Table C-1], to provide a summary of the expected NIPTS that would occur
from a 40 year exposure beginning at an age of 20 years.
Inspection of Table C-1 in the EPA Levels Document [2] shows that as the
average sound level of the exposure increases, there is a widening of the fre-
quencies affected by the exposure. As would be expected, the average of 500
Hz, 1000 Hz, 2000 Hz and 4000 Hz does not show a uniform constant increase in
loss with a rising exposure level, but instead increases at an accelerated
pace with increasing average sound level. While use of the most sensitive
frequency is proper for the determination of an absolutely safe daily equiva-
lent sound level, assessment of the relative impact of exposure to higher equi-
valent sound levels requires that all audiometric frequencies be considered.
Therefore the average of 0.5 kHz, 1 kHz, 2 kHz, and 4 kHz is the recommended
measure. Since each of the four frequencies describe the center of the
preferred octave bands, there is no overlapping in octave bands as would be
the case if 3000 Hz was included.
Having selected a method to handle the question of frequency, the next
problem is time. One way to consider time is to select a point in time at
which the relative impact will be described. Selection of such a point is
somewhat arbitrary and not entirely meaningful. For instance one could argue
that it is more important to describe the effects of noise when a person is
middle-age, and not when a person is 60 years old. An alternative approach
57
-------
is to use the average NIPFS of the population during or over a normal working
lifetime. Averaging NIPTS with respect to time avoids arbitrarily selecting
any one point in time and provides a realistic assessment of the overall
effect of noise on hearing on a large population, recognizing that many indi-
viduals, because of differences in sensitivities and ages or lengths of expo-
surer may incur either more or less hearing loss than would be assessed using
this procedure.
A grand averaging of the NIPFS with respect to frequency (0.5 kHz, 1 kHz,
2 kHz, 4 kHz) and time (0 to 40 years of exposure) and percentiles (0.1 to 0.9
percent iles) from references 2 and 29 is listed in Table 4. These NIPTS data
can be very well described by the formula:
Ave NIPTS = (Leq(8) - 75)2/40 = (I^q(24) -70)2/40 Bqn. 10
The slight differences shown in Table 4 between equation 10 and the NIPFS
data should be considered insignificant, especially in view of the fact that
the original data were rounded to the nearest whole integer in any case.
Equation 10, then, is the criterion for estimating the potential severe
health effects due to a proposed project. For applications, it can be
TABLE 4
Average Hearing Loss as Function of 8-hour
8) Ave. Hearing Loss (LeafS) ~ 75)2/40
dB* dB
75 0 0.0
80 1 0.625
85 3 2.5
90 6 5.625
95 10 10.0
*Source: Table C-1 of Levels Document [2], and Johnson [29]
58
-------
calculated directly, read from Table 5, or read from Figure 8. The outdoor
day-night sound level, L^n, should be used only to identify potential problem
areas. Within those areas, an effort should be made to estimate the actual
exposure of groups of people. In any application, it should be remembered
that since this equation was developed from averaging the effects of noise
over frequency, time, and percentiles, it cannot estimate the effect on an
Table 5
Criterion Function for Severe Health Effects
or 1^-jn I%q(24) °B loss
(dB) (oB)
75 70 0
76 71 0.025
77 72 0.100
78 73 0.225
79 74 0.400
80 75 0.625
81 76 0.900
82 77 1.225
83 78 1.600
84 79 2.025
85 80 2.500
90 85 5.625
95 90 10.0
individual at one audiometric frequency at one point of time. This equation
should be used only to assess the average relative impact of exposure to
different equivalent sound levels.
It is also useful to look at individual susceptibility to noise induced
hearing loss. Therefore, a user may wish to consider the NIPTS for the most
sensitive ten percent of the population after 40 years of exposure. This infor-
mation can be read from the 'Max. NIPTS 90th Percentile1 curve of Figure 8.
59
-------
12
eg 10
u,
CO
Q
co
UJ.
cc.
H
Ul
cc
UJ
a..
Q
UJ.
u
a
z
CO
5'
6
0.
60:
RECOMMENDED
HEARING LOSS
WEIGHTING
FUNCTION
(Leq(8).75)2
40
90
100
110
DAILY 8 HOUR EQUIVALENT SOUND LEVEL (dB)
FIGURE 8. POTENTIAL HEARING DAMAGE RISK FOR DAILY
EXPOSURE TO 8 HOUR EQUIVALENT SOUND
LEVELS
60
-------
2.3.2 Quantification of the impact
The first step in quantifying the impact is to construct the tables indi
cating the number of people within decibel intervals. In many instances the
same tables setting out the extent of the general audible noise impact can
serve for this noise range also, but if there are very many people exposed to
high levels , smaller contour intervals are recommended for tabulating the
severe health effects.
As with the general adverse effects, it is desirable to quantify the
exposure of individuals to different levels by a single number. A term simi-
lar to the level-weighted population may be calculated using the hearing loss
function (eqn. 10) identified in the previous section. This would result in
a hearing loss-weighted population, HWP, measured in terms of hearing loss,
expressed as person-decibels:
x
HWP = f P(W8)) - H(Ioq(8)) . d(I«j(8)) Eqn. 11
where P(Ieq(8)) is the population distribution as a function of 8-hour Leq,
H(Leq(8)) is the weighting function given in equation 10 (and Figure 8 and
Table 5), and d(Leq(8)) is the differential change in 8-hour average sound
level. Replacing the integrals by summations of successive increments in
average sound level we have:
HWP = P(I) • H(I) Bn- 12
where i indexes the successive increments in average sound level. If the
Leq (24) measure is preferred for a particular application, summation would
start at 70 dB.
61
-------
The disadvantage of the hearing-weighted population is that it is not
easily understood: the product of persons and decibels of hearing loss is
not an intuitively obvious concept. A more understandable indicator of
severe health effects is the average potential hearing loss (PHL) which is
\r
analogous to the noise impact index for general audible noise:
ptotal
where the terms are as defined for equation 12, and Ptotal ^ equal to the
base population, which is normally the population exposed to levels above 75
dB. Care must be taken in defining the base population, however. If it is
to be used to compare alternatives, the same base population must be used for
all. Otherwise, the average hearing loss could be lowered by a project which
affected more people, and the indicator would not be a reliable measure of
impact. The simplest approach is to use as the base population the largest
total population subjected to severe health effects by any of the alternatives.
If this is done, PHL indicates the average hearing loss, in decibels, for
those people subjected to severe health effects due to noise.
Again, the above equations may be replaced by a summation over successive
increments of day-night sound level. It is recommended that increments of
day-night sound level less than five decibels (i.e., preferably one or two
decibels) be used in calculating values of PHL.
Further, analogous to the assessment of general audible noise, the change
in hearing loss-weighted population is a useful descriptor for many assessment
purposes, as is the relative change in impact defined in Equation 8 with HWP
substituted for IMP.
62
-------
2.4 Environmental degradation
Even in areas where no people are presently living, a significant increase
in noise over the existing conditions will constitute a noise impact. The envi-
ronment may be degraded either because the increased noise affects wildlife
or monuments, or because it destroys the tranquility of a wilderness area to
which urban dwellers wish to go for an escape from city noise, or because it
makes the area unsuitable for future residential or other noise sensitive
development. In each casef some of the quiet which is one of our national
natural resources is lost; the quality of the environment is lowered.
Unfortunately, there are no data available which express this reduction
in environmental quality as a function of noise level, or of the change in
noise level. Consequently, it is not possible to identify a special criterion
function for these areas, such as those identified in sections 2.2.1 and 2.3.1.
Instead, quantification of environmental degradation normally proceeds only
as far as the tabulation of the extent of the impact, The only modifications
necessary for the standard tabulation (such as the example Tables 6 through
9) is the likely deletion of the columns on residential and employee popula-
tions, and a revision in the use of the special situations column. Animal
species which are particularly vulnerable and recreational uses of the areas
will be the principal kinds of special situations to be listed. As a supple-
ment to this numeric quantification, a word description of the environmental
impact should be provided in terms of the expected change from the present
conditions, paying particular attention to the special situations. In some
circumstances, it may be useful to reduce this tabulation to a single number,
for example for comparisons or trade-offs with other planning criteria. In
those cases, a 'level-weighted area1 can be calculated by using the population
63
-------
weighting function of Table 3, which is the best available indicator of rela-
tive impact.
Rural areas can be treated by the methods of either this section or sec-
tion 2.2. That is, the analysis can stop with the tabulation of impacts, or
it can proceed to calculate the level-weighted population. The equation used
(equation 2b) shows some adverse response to general audible noise at levels
as low as 45 dB (I^n). However, because the percentage responding adversely
is so snail - less than 0.5% below 48 dB - and the number of people in most
rural areas is so low, the magnitude of the level-weighted population will
usually be so small as to be of little help in environmental assessment.
Although the single-number index can be used in such areas, it is not recom-
mended as strongly for them as it is for urban areas.
2.5 Treatment of temporary projects
The major simplification in the analysis for temporary projects has
already been mentioned (section 1.4.2): the fact that prediction of future
population in the affected area is unnecessary. For temporary projects
lasting more than one year, that is the only modification necessary.
For temporary projects, the same as for permanent noise environments,
the yearly day-night sound level should be used in computation of impact
indices. Impact assessment is done in the same manner as for permanent noise
environments by the use of tabulations and calculation of the sound level-
weighted population and noise impact index.
For temporary projects lasting less than a year, it is useful to compute
the level-weighted population for two situations:
(1) for the temporary noise environment as if it were permanent, but
also stating its actual duration; and
64
-------
(2) for the temporary noise environment in terms of its contribution
to the yearly day-night sound level.
For example, consider a population of 1000 experiencing a temporary day-night
sound level of 70 decibels for nine months due to a construction project,
after which the day-night sound level drops to 60 decibels on a long-term
basis. The following two situations would be described. .
1. During the nine-month construction period itself, the level-weighted
population is (0.245) (1000) = 245 persons responding adversely to the noise.
2. To calculate the effect of the construction activity on annual average
impact requires calculation of the yearly day-night average sound level:
1 70 60
lOn(y) = 10 logio 12" (9 x 10^) + (3 x 107"5") = 68.9 decibels Eqn. 14
On the basis of this I
-------
2.6 Example application
Sample tables to demonstrate the approach discussed in this chapter have
been drawn up for a simple example problem. The example is based on the pro-
posed expansion of a highway which runs through a suburban area, and is simpli-
fied to facilitate understanding of the suggested procedures. Details of the
example are contained in Appendix E, which also contains an additional example
application. This section is intended primarily to provide samples of the
appropriate tables. It does not cover all possible types of problems for
which these guidelines are appropriate.
As discussed in section 2.2.2, a number of tables are usually helpful.
The first table documents affected areas (i.e., the base population and base
area) for future noise levels without the proposed project (Table 6); the
second deals with project noise alone (Table 7); and the third tabulates
effects for the project noise together with all other sources (Table 8).
The final table provides details of various special situations, which may
be particularly affected by noise (Table 9). At the bottom, of Tables
6 to 8, several of the single number indexes are stated. By comparing the
single-number indexes presented in Tables 6 and 8, we see that the anticipated
change in impact is an increase of 111 more people responding strongly to the
adverse effects noise has on them. Likewise, the expected increase in
potentially severe health effects (Ldn 2. 75 dB) is in the range of 13 person-
decibels.
66
-------
Table 6
Sample Data Presentation:
Future Noise Levels Without Proposed Project
Yearly I85
80-85
75-80
70-75
65-70
60-65
55-60
50-55
45-50
Residential
Population
0
0
0
0
833
1389
2778
0
0
Industrial/
Commercial
Employees
0
10
40
130
470
2840
510
0
0
Total
Land Area
(sq km)
0
0.0156
0.0469
0.0625
0.3125
0.8542
0.7083
0
0
Residential
Land Area
(sq km)
0
0
0
0
0.1875
0.3125
0.6250
0
0
Industrial/
Commercial
Land Area
(sq km)
0
0.0156
0.0469
0.0625
0.1250
0.5417
0.0833
0
0
Special
Situations
(See Table)
-
-
-
-
-
8
1,2,3,4,5,6,7
-
-
5,000
4,000
2.0
1.125
Level Weighted Population (LWP) = 501 people
Noise Impact Index (Nil) =0.10
Hearing-loss Weighted Population (HWP) = 0
0.875
-------
CO
Table 7
Sample Data Presentation:
Future Noise Levels of Project Alone
Yearly L^n
>85
80-85
75-80
70-75
65-70
60-65
55-60
50-55
45-50
Residential
Population
0
0
83
150
350
717
1200
833
1667
Industrial/
Commercial
Employees
0
0
140
240
370
800
1500
380
570
Total
Land Area
(sq km)
0
0
0.050
0.090
0.160
0.340
0.610
0.250
0.500
Residential
Land Area
(sq km)
0
0
0.01875
0.03375
0.07875
0.. 161 25
0.27000
0.18750
0.37500
Industrial/
Commercial
Land Area
(sq km)
0
0
0.03125
0.05625
0.08125
0.17875
0.34000
0.06250
0.12500
Special
Situations
(See Table)
-
-
-
8
-
1,2
3,6
4,7
5
5,000
4,000
2.000
1.12500
0.8750
Level Weighted Population (LWP) = 362 people
Noise Impact Index (Nil) = 0.07
Hearing-Loss Weighted Population (HWP) = 13 person-decibels
Average Potential Hearing Loss (PHL) = 0.16 dB per person for 83 people
-------
Table 8
Sample Data Presentation:
Future Levels from all Sources Combined
Ok
vo
Yearly L^n
>85
80-85
75-80
70-75
65-70
60-65
55-60
50-55
45-50
Residential
Population
0
0
83
150
1278
1128
2361
0
0
5,000
Industrial/
Commercial
Employees
0
10
75
240
640
2535
500
0
0
4,000
Total
Land Area
(sq km)
0
0.0160
0.0969
0.1350
0.44875
0.6971
0.60625
0
0
2.000
Residential
Land Area
(sq km)
0
0
0.01875
0.03375
0.28750
0.25375
0.53125
0
0
1.12500
Industrial/
Commercial
Land Area
(sq km)
0
0.0160
0.07815
0.10125
0.16125
0.44335
0.0750
0
0
0.8750
Special
Situations
(See Table)
-
-
-
8
-
1,2
3,4,5,6,7
Level Weighted Population (LWP) =612 people
Noise Impact Index (Nil) = 0.12
Hearing-Loss Weighted Population (HWP) = 13 person-decibels
Average Potential Hearing Loss (PHL) = 0.16 dB per person for 83 people
-------
Table 9
Sample Data Presentation:
Special Situations
Average Population
Day Night
Area (sq km)
Comments
1. School 300
2. Playground 40
3. Park 30
4. Church 63
5. Nursing Home 200
6. School 1000
7. Library 25
8. School 500
0
0
0
200
150
5
Night Classes
70
-------
Chapter 3
Special Noises
Not all noises can be adequately evaluated by average A-weighted sound
levels. Examples of the special noises which require other measurement sys-
tems are the following: (1) infrasound, in the frequency range of 0.1 to 20
Hz; (2) ultrasound, frequency range above 20 kHz; (3) certain types of
impulse noises such as sonic booms and blasts; and (4) sounds that convey
more information than random noise sources with comparable average sound
levelsF such as voices, warning signals, or barking dogs. This chapter
contains a section discussing each of these four special noises. For the
first three, the section discusses measurement, screening levels, and human
effects. For the fourth, the section merely provides a brief description of
the nature of the problem and of how it might be treated in a noise impact
analysis.
3.1 High energy impulse noise
The assessment of impulse noise presents unusual problems. In many cases
the appropriate techniques and measures are applicable only to particular
situations. (For example, with respect to blast noise, damage to certain
types of buildings can be predicted in terms of non-acoustic parameters, such
as effective distance and the amount of explosive charge.) Moreover, the sig-
nificance of the noise impact cannot always be quantified for the same effects
suggested for general audible noises. Whereas low-level impulse noise is
accounted for as part of normal general audible noise, high energy impulses
require additional measurements for impact assessment. In many situations an
individual interpretation of the criteria is required.
71
-------
At present, high energy impulse noise comes primarily from sonic booms,
blasting operations, or artillery fire. Some limited community response data
for sonic booms and artillery fire are available [30, 31]. Noise measurement
instrumentation at the time of the sonic boom study (1965) was not as sophis-
ticated as it is now, so the physical measures from that study (peak overpres-
sure in pounds per square foot) need to be converted to more recently developed
measures. Tine conversion factors appear to be different for artillery and
blast noise as opposed to sonic booms. Consequently, the methods presented
in this section need to be verified with more data, some of which is being
collected at the time of this writing. The methods presented here are based
on the only available data, and should be applied with some caution.
3.1.1 Description of high energy impulse sounds
Day-night sound level is the primary descriptor for environmental noise.
High-energy impulse sounds, such as those produced by sonic booms, quarry
blasts, or artillery fire, in addition to the high-level audible sound,
can excite noticeable vibration of buildings and other structures. These
induced vibrations — caused by airborne sound or transmitted through the ground
or structures — may generate additional annoyance beyond that due to simple
audibility of the impulse, because of "house rattling11 and "startle," as well
as because of additional contributions to interference with speech or sleep.
The annoyance data which are used in the next section to summarize community
response to impulse noise are based on the annoyance caused by house rattle.
/
It has been general practice in the past to describe such high-energy
impulse sounds in terms of the peak sound pressure in a wide frequency band.
While the peak pressure may be satisfactory for assessment of impulses in a
restricted range of peak pressures and durations, it is not sufficient as
72
-------
a general descriptor for use in measurement or prediction of the combined
environmental effects of impulses having substantially different pressure-time
characteristics. Use of the peak pressure is also unwieldly or misleading
when a succession of impulses, sometimes overlapping, must be evaluated.
Two noise measures, similar to the measure used for general audible noise,
are recommended in these guidelines for assessing the environmental impact of
high-energy impulse noise. The choice between these two measures is based
primarily on the number of impulses occurring during a day. If only isolated
high-energy impulse noises occur, the C-weighted sound exposure level, symbol-
ized as LSO should be used. If there are more than perhaps eight occurrences
per day, it will be helpful to compute a C-weighted day-night sound level,
Icon, derived solely from the individual Lgc values.
There are two reasons for using a C-weighting. First, it does not dis-
count the low frequency components which are a major part of impulse noise and
of vibration, as the A-weighting network does. Second, subjective estimates
of blast noise magnitude conform with magnitude estimates of other noises
when the blast is measured by C-weighting and the other noises are measured by
A-weighting [32]. In general, C-weighting has been found to closely relate to
average human response to high-energy impulse noise [33].
The use of sound exposure level is recommended to facilitate combination
of data when more than one impulse noise event occurs per day, as is usually
the case. Further, it is consistent with subjective evaluations of sonic
booms where duration of the signal influences subjective response [34].
The assessment procedures suggested in this section should be used for
impulse sounds that have daytime C-weighted sound exposure levels greater
than about 80 dB. This corresponds to unweighted peak pressures for sonic
73
-------
booms greater than about 106 dB, which appear to be the threshold of adverse
community response on the basis of the data on sonic booms. This in turn
corresponds to unweighted peak pressures for quarry blasts and artillery fire
of about 100 dB. At night, the threshold of response should be reduced to a
Oweighted sound exposure level of 70 dB (corresponding to unweighted peak
pressure levels of 96 dB and 90 dB for sonic booms and blasts, respectively),
because of the decreased acceptability of nighttime impulsive exposures [30,
p. 150].* Impulse events with smaller peak pressure than described above are
assumed to elicit normal auditory responses and are assumed for most situa-
tions to be described adequately by 1^. For impulses with peak pressures
greater than 140 dB, assessment criteria based on actual physiological or
structural damage should also be applied. In addition, the effects of
groundborne vibration should be assessed (Chapter 4).
t
In most cases where impulse noise needs to be considered, the task for
the noise impact analysis is to predict or estimate in advance what the levels
of impulse noise will be. With rare exceptions (e.g., reference 35), there
are no reliable predictive techniques other than using a measurement of a
similar event occurring elsewhere. When the only data available are
expressed as peak pressure, useful approximations can be based on indications
that Lgc is roughly 26 dB lower than the peak pressure for sonic booms [36],
and 20 dB lower for blast noise [37, Fig. 29]. In those cases where it is
possible to conduct measurements of a similar event elsewhere, it is important
*For situations which are characterized by less than about eight impulse
occurrences per day, and where LJJC is computed rather than Icdm threshold
values 10 dB higher (90 dB daytime and 80 dB nighttime) may be used. Bow-
ever, the more stringent (80 and 70 dB) threshold values are recommended for
situations where there are many (greater than about eight) high-energy impulse
events per day, and where Icdn is calculated, in order to ensure that
relatively low level L^dn cases are not disregarded.
74
-------
to be able to distinguish impulse noise (such as sonic boom) from other high-
energy noise events (such as jet aircraft flyovers). A useful rule of thumb
to aid in making such a distinction is that for an impulse noise the maximum
C-weighted sound exposure level in any 2-second time period is 10 dB greater
than the C-weighted sound exposure level in any contiguous 2-second period of
the event.
3.1.2 Human noise exposure effects of high-energy impulse noise
The Oklahoma City sonic boom study [30] is the artillery fire study [31]
form the primary bases for the procedure proposed for assessment of the effects
of high-energy impulse sounds. In the sonic boom study [30], eight supersonic
overflights were performed on a daily basis for six months. Altitudes and
airspeeds were selected to obtain three different nominal overpressures, on an
increasing basis, during the tests. Personal interviews of respondents were
made during three time periods that corresponded to the three different nomi-
nal overpressures. Interviews were conducted at three different distances
from the ground projection of the flight path to obtain different exposures
for each of the three boom levels.
The questionnaire structure and response scaling used in the sonic boom
social survey are such that direct comparison with other surveys is difficult.
The responses to a question on the degree of annoyance due to "house rattles"
caused by the booms is used here as the primary measure to quantify community
response. The category "serious" annoyance is considered to be most compar-
able to the highly annoyed response used (in section 2.2.1) to summarize the
adverse effects of general audible noise. It should be noted that the percent
of respondents reportings serious annoyance at different boom levels was not
a percentage of the total population sample, but only of that fraction of the
75
-------
sample that believed it appropriate to complain about governmental actions.
This fraction is of the order of 60 percent. To compare these responses to
the total populations used in other surveys, an adjustment for the total pop-
ulation was made in the current analysis.
The noise measurements in the sonic boom study were collected in terms
of nominal peak overpressures. Conversion of nominal overpressures to C-
weighted sound exposure levels were made using the average difference of 26
decibels between peak overpressure, in decibels, and C-weighted sound expo-
sure level. The resulting values were then used to compute L^dn for tne eight
daytime sonic booms, using the approximation:
lojn = Isc + 10 log (Ifc + 10 Nn) - 49.4 Bqn. 16
where N^ and Nn represent the number of impulse events during the day and
night, respectively. Thus for eight sonic booms per day, equation 16 reduces
to:
ICdn = tGC - 40-5 Ban- 17
The resulting data for the percent serious annoyance at the computed C-weighted
day-night sound level values are plotted as filled-in squares in Figure 9.
In the artillery fire study [31], groups of residents were interviewed at
nine sites in the vicinity of an Army base where extensive artillery firing
training takes place. Six of the sites that were off-base were considered
here. Noise monitoring on a 24-hour basis took place at 17 locations for an
average of approximately 25 days per site. These measured average sound
levels in conjunction with computer based predictive models, were used to ob-
tain annual average C-weighted day-night average sound levels for blast noise
associated with the environments in which the survey respondents lived. The
social survey used scales similar to other recent surveys. The percent of
76
-------
80
70
60
50
40
30
20
%HA-
100
U1.17.0.151Lcd||>
Sonic Boom
45
50
60
65
70
" ^n 'For ' Eventil
i 90
95
10°
. 105
"°
115
IIS
120
125
130
135
140
I
130
.Peak Preouib
(Boom)
ikPranure
110 115 120 125
FIGURE 9. RECOMMENDED RELATIONSHIP FOR~PREDICTING COMMUNITY
RESPONSE TO HIGH ENERGY IMPULSIVE SOUNDS^
1 (Blan&
135 Artillery)
77
-------
respondents reporting high annoyance are plotted as filled-in circles on Fig-
ure 9.
Using annoyance data from both the Oklahoma City sonic boom study and
the artillery fire study, a function is plotted in Figure 9 which shows a
reasonably good fit of projected high annoyance against C-weighted day-night
sound levels. For use in situations where there are less than approximatley
eight booms or blasts per day, a separate abscissa for C-weighted sound expo-
sure level is also included in Figure 9. Further, because sonic boom and blast
noise data are oftentimes collected in terms of peak pressure, additional ab-
scissas are also included in Figure 9.
Over the range of data available, the function illustrated in Figure 9
provides a reasonably good prediction of the percentage of the population who
can be expected to be highly annoyed at given exposures to high-energy impulse
noise. Consequently, it is proposed to use the function shown in Figure 9
and presented in equation 18 below* for the assessment of high-energy impulse
noise, despite the fact that such applications may need to extrapolate beyond
existing data.
%HA 10°
- 0.153 I) E°n- 18
Quantification of adverse human response anticipated from high energy
impulse noise is performed in the same manner as for general audible noise
(Section 2.2.2). The appropriate weighting function describing the population
exposed to high energy impulses who are highly annoyed with the noise may be
computed from equation 18, or read from Figure 9 or Table 10. Level weighted
*Note that the format of equation 18 is similar to that footnoted on page 37
in section 2.2.1.
78
-------
Table 10
Values of Weighting Function for High Energy Impulse Noise
= (0.01)% HA]
45 0.014
46 0.016
47 0.018
48 0.021
49 0.025
50 0.029
51 0.033
52 0.039
53 0.045
54 0.052
55 0.060
56 0.069
57 0.080
58 0.091
59 0.105
60 0.120
61 0.137
62 0.157
63 0.178
64 0.201
65 0.227
66 0.255
67 0.285
68 0.317
69 0.351
70 0.387
71 0.424
72 0.462
73 0.500
74 0.538
75 0.576
79
-------
population may then be computed from equation 6. Likewise, Noise Impact
Index and Relative Change in Population may be calculated from equations 7
and 8, respectively.
In many situations, both iitpulse noise (measured in loan) and general
audible noise (measured in LA^) will be of concern, and it will be necessary
somehow to combine the results to obtain a total number of people affected
by all aspects of the noise. An assessment of the overall noise environment,
combining the effects of high energy impulse sounds and of general audible
noise, can be made by equating the degree of annoyance expected from the two
types of noise sources. Using Figures 4 and 9, then, it is possible to
identify a specific C-weighted L^ which causes as much annoyance as an
A-weighted day-night sound level. For example, an LQfo of 65 && is expected
to result in 23% of the exposed population being highly annoyed by the noise
r
(Figure 9). This same level of annoyance is reached at an A-weighted day-night
sound level of 69 dB (Figure 4). Thus the Lean may be converted to L^n
via equal annoyance (Table 11). This converted L^n is added, logarithmical-
ly, to the general audible noise already measured in terms of L^, and the
resulting composite noise level is used for assessment of the overall noise
environment, using Figure 4 and Table 3 as necessary. This procedure is
performed in order to avoid the double-counting of affected people which
could result if they were tallied separately for impulse noise and for
general audible noise.
3.1.3 Structural damage criteria for impulse noise
It is normally considered that the most sensitive parts of a structure
to airborne noise or overpressure are the structure's windows, although in
some cases it may be plastered walls or ceilings. Such noise or large pressure
80
-------
Table 11
Conversion of I<^n to L^n Via Equal Annoyance
* Highly Annoyed
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
1
2
2
2
3
3
3
4
4
5
6
7
8
9
10
12
14
16
18
20
23
25
28
32
39
42
46
50
54
58
45
49
49
49
. . 52_ . . .....
52
52
54
54
56
57
58
' 59
60
61
63
64
65
67
fifi
69
70
72
73
76
77
78
79
an _ .
81
81
-------
waves also introduce building vibration in addition to that due to ground
notion. Thus the effects of airborne sound on structures may need to be eval-
uated in terms of vibration criteria as well as in terms of criteria based on
peak overpressure. For most airborne sound, however, evaluation of the peak
overpressure is sufficient to determine the threshold of possible damage. On
the other hand, for some types of underground blasting and when the building
is close to the blast site, the vibration is transmitted essentially through
the ground. In this case the vibration inside the house must be predicted
and evaluated according to the vibration criteria (Chap. 4). This subsection
describes structural damage criteria for three kinds of impulse noise: blast
noise; sonic boom; and artillery fire. A brief paragraph is appended relating
to structural damage from continuous sounds.
For blast noises, the probability of broken windowpanes should be esti-
mated. Empirical formulas given below allow an estimate of "safe" distances
from the blast, beyond which window damage is negligible. These formulas
include sufficient safety factors to account for the negative influence of
such variables as wind direction, atmospheric temperature gradients, and win-
dowpane shape and size. These formulas are newly proposed and are somewhat
tentative [38]. They are suggested here essentially as screening tools: if
these equations suggest there will be no structural problems for a particular
project, the impact analysis needs to proceed no further. If these formulas
suggest a potential impact of blast noise on structures, then the analyst (or
blasting engineer) should undertake a more detailed analysis which involves
explicit consideration of the variables covered by a safety factor in these
formulas. In addition, a monitoring program might be recommended to identify
any damage, or lack of it, actually caused by an explosion.
82
-------
For surface explosions, window breakage in residential type structures is
expected to be negligible (less than 50% probability of even one broken pane)
if the charge per delay equivalent weight of high explosive (WHE) in kilograms
is less than that specified by the more appropriate of the following two
conditions:
(1) If /-the population is non-uniformly distributed, but is clustered,
then each population cluster, including the nearest residence,
should be checked. The amount of WHE for any cluster should be
less than 328 R /M where R is the distance in kilometers from the
explosion to the center of a cluster of residences and N is the
number of people residing in that cluster with the provison that N
must always be at least 4.
(2) If the population is reasonably uniformly distributed, then the
amount of WHE should be less than 40 R , where R is the distance
in kilometers to the nearest residence.
The use of these formulas requires some judgment as to what constitutes a pop-
ulation cluster and what constitutes a reasonably uniform distribution. In
some cases, both formulas might be checked and the one that predicts the least
allowable amount of WHE used.
For explosives buried deeper than 1.4 meter per (Kg) , the peak ampli-
tude will be attenuated by at least a factor of 5. For such underground ex-
plosions the preceding formulas need to be adjusted as follows:
(1) Population clusters - the amount of WHE should be less than 26430 R /N.
(2) Uniformly distributed population - the amount of WHE should be less
than 3200 R3.
"Weight per detonation where each detonation is delayed to go off in a
predetermined sequence (usually within a fraction of one second) for each
event. The duration of the total event is normally less than one second.
83
-------
For explosive charges greater than those determined by the above formulas,
the peak overpressure should be predicted and the number of broken windows
estimated. The statistical estimator (Q) for the number of "average typical"
panes broken is:
Q = 1.56 x 10~10 N(PK*)2*78 Eqn. 17
..i where N = number of people exposed (assuming 19 panes per person) and PK* is
peak-to-peak amplitude of the pressure variation (in pascals) at ground
k\
level. For convenience of measurement, the peak-to-peak pressure amplitude
reflected at ground level (PK*) may be used. The conversion between the peak
free air pressure (Ap) and PK* is given by the relation:
PK* = 2.7 AP Eqn. 18
However, the peak pressure may be amplified by a factor of 5 as the result of
refraction, ducting, and focusing; therefore, in the "worst case" condition
2 73
the number of broken panes, Q, may be multiplied by a factor as high as (5) *
or 88 to obtain Qmax- Atmospheric meteorological effects can increase this
factor further. In addition, for peak pressures (AP) above 140 dB (200 Pa),
structural damage other than window damage may occur. Measurement or predic-
tion of vibration should be accomplished.
For sonic boom and artillery fire, the amount of window damage can be esti-
mated by calculating Q and CL.. for the expected peak pressure, as discussed
for blasts. These formulas, however, should be used only for peak pressure
levels above 130 dB. Above 140 dB, structural damage should also be assessed
by prediction or measurement of vibration levels in the exposed structures.
For continuous sounds above sound pressure levels of 130 dB, there is the
possibility of structural damage due to excitation of structural resonances
for infrasound, as well as low and medium frequency sound. While certain
84
-------
frequencies (such as 30 Hz for window breakage) might be of more concern than
other frequencies, one may conservatively consider all sound lasting more than
1 sec above a sound pressure level of 130 dB (1 Hz to 1000 Hz) as potentially
damaging to structures.
3.2 Infrasound
3.2.1 Description of infrasound
Infrasound is defined as sound in the frequency range from 0.1 to 20 Hz.
The measurement of infrasound should be made with instrumentation having a flat
frequency response (+3 dB) from 0.1 Hz to 1000 Hz. The reason for the extended
measurement range is that in evaluating a noise that is composed of both infra-
sound and higher frequency sound, the higher frequency sound must also be mea-
sured for proper assessment of the infrasound, because sounds above 20 Hz can
mask the infrasonic sounds.
Although blasting operations cause infrasound as well as impulse noise
and vibration, it is not intended that all of these analyses be conducted.
Among other considerations, the necessary instrumentation is different for
each of these special noises. For blasting, an impulse noise evaluation is
adequate, covering both human and structural effects. Because infrasound can
be related to vibration, the vibration analysis (Chap. 4) also helps reduce
any need for a special infrasound analysis.
3.2.2 Human noise exposure effects of infrasound
On the basis of a summary of infrasound effects (Figure 10), compiled from
the Levels document [2] and more recent work [39, 40], it is suggested that for
exposures of less than 1 minute the maximum sound pressure level should be below
the following values:
85
-------
oo
SOUND PRESSURE LEVEL BELOW WHICH
INFRASOUND IS NOT EXPECTED TO
PRODUCE ANY ADVERSE PHYSIOLOGICAL
EFFECTS.
CURVE B1
CURVE B2
ESTIMATED
LOUDNESS OF
45 PHON
ANNOYANCE THRESHOLD
DUE TO BUILDING STRUCTURE
VIBRATION OR MIDDLE
EAR PRESSURE.
CURVEC
RANGE OF HEARING THRESHOLD
0.1 HZ 0.2
0.5 1HZ 2 5 10HZ 20HZ
FREQUENCY(HZ)
FIGURE 10. ll\ ASOUND CRITERIA
-------
0.1 Hz to 5 Hz ... 120 dB
5 Hz to 20 Hz . . . 120 dB - 30 log | Eqn. 19
where £ is the dominant frequency. For exposures longer than 1 minute and less >
than 100 minutes, the levels should be reduced by (10 log t) dB where t is £,
time of exposure in minutes. Exposure longer than 100 minutes should use the
100 minute limits. In other words, exposures 20 dB less than the one minute
criterion should be regarded as having no impact, regardless of exposure time.
The 100 minute criterion basically insures that the infrasound is inaudible.
These levels serve essentially as screening levels. As long as they are not
exceeded, infrasound does not need to be included in the noise analysis.
For evaluating the impact, if this screening criterion is exceeded, a
single-number index is not suitable. Instead, the impact should be qualita-
tively described; the effects that might occur at different sound levels are
given in Figure 10. Any assessment of the effects beyond those in Figure 10
is not contained in these guidelines and will require further research and
investigation.
3.3 Ultrasound
3.3.1 Description of ultrasound
Ultrasound is defined as sound at frequencies between 20 kHz and 100 kHz.
Seldom is ultrasound an environmental problem and, unless the level is expected
to exceed 105 dB, it can be ignored in an environmental noise analysis.
Measurement of ultrasound should be accomplished by instrumentation with
flat response (+_ 3 dB) from 10 kHz to 100 kHz.
3.3.2 Human effects of ultrasound
Ultrasound noise levels below 105 dB (frequencies above 20 kHz) are con-
sidered to have no significant impact on people. Noise levels above 105
87
-------
decibels should be reported in the analysis and individually evaluated based
on specific research studies. In particular, studies of effects on animals
may be important. No further quantification of the environmental impact of
ultrasound is recommended. Barely is ultrasound (except for some occupational
situations, e.g., ultrasonic cleaners) an environmental problem of practical
interest. Evaluation of ultrasound exposure above 105 dB requires additional
investigation and research.
3.4 Noises with information content
Some general audible noises are also more annoying than their level alone
would indicate, due in part to their information content or clear detectability.
Examples include barking dogs and back-up alarms, but the primary problem is
voice communication (live, amplified or recorded) that crosses residential
boundaries at high levels. There is no formal method for assessing the
impact of such sounds; each case must be assessed on its particular merits.
It is recommended, however, that the analyst mentions how, as a result of the
proposed action, the intrusion of understandable voices into some area might
cause loss of privacy and consequent undesirable effects. The actual content
of the typical messages or words might be stated along with the number of
people that are impacted.
88
-------
Chapter 4
Vibration
This chapter contains procedures for evaluating the impact of vibration
on nan. While the main reason for their inclusion here is to account for
vibration generated by airbone noise, the impact of certain types of vibra-
tion can be assessed whether the transmission paths are airborne or structure-
borne. The two sections of this chapter deal with the human effects of
vibration and the structural effects of vibration.
The material in the first section is based on an approved ISO standard
and its proposed amendments [41 ], and its United States Counterpart [42].
These are summarized in Appendix D, to provide the necessary background to
follow the recommendations in section 4.1. The recommendations in section 4.2
are based on consideration of that material and data contained in Bureau of
Mines Bulletin 656 [43].
4.1 Human effects of vibration
Vibration of structures may be due to airborne acoustical waves or solid-
borne vibration. Most problems caused by airborne impulse noise, when build-
ing vibrations are caused as a side effect of the primary auditory stimulus,
should be accounted for by the procedures of section 3.1. Nevertheless, at
certain times it may be necessary to assess separately the vibration caused by
such sources. Groundborne vibration which is quite likely to accompany some
mining, construction, and other industrial activities usually requires special
evaluation. A method to evaluate human response to vibration inside buildings
is presented which should be used to evaluate the impact of such activities.
The method applies to the frequency range between 1 Hz and 80 Hz.
89
-------
4.1.1 Description of building vibration
For continuous vibration environments, rms acceleration should be measured
along three orthogonal axes, one axis of which is normal to the surface being
measured. The acceleration will be weighted to account for the dependence
of human reaction on frequency by use of a low pass filter with a corner
frequency of 5.6 Hz (Figure 11). This accounts for the fact that human sensi-
tivity to acceleration decreases over the frequency range under considera-
tion; above 10 Hz this decrease is approximately proportional to frequency.
The assessment of the impact should be against greatest acceleration on any
of the three axes used.
For building measurements to be appropriate for the criteria of the next
subsection, the measurements should be taken on the floor at a point that has
the maximum amplitude of all the reasonable points of entry of the vibration
to the human occupants. Normally this point may be assumed to be at the
mid—span or center of a room.
For impulsive shock the measurement should be the same as for the
continuous vibration measurement, except that the peak acceleration, not
the rms value, should be used. The duration for impulsive shock excitation
will be determined by either the time the acceleration of an event exceeds
0.01 m/sec2 or by the time the acceleration is within one-tenth of the peak
value. ftoichever gives the snorter duration should be used.
4.1.2 Human vibration exposure criteria
Threshold levels are presented in Table 12 for most types of structures.
Not all types of buildings are classified, but common sense should suggest
the most appropriate classification.
The overall vibration that will not cause an adverse impact for any con-
dition and time period corresponds to rms acceleration values below
90
-------
24
ATTENUATION (dB) = 20 log >/1 + (f/5.6}2
20
16
00
•o
I12
Ul
8
15 20
30 40 50 60 80
FREQUENCY (Hz)
FIGURE 11. WEIGHTING CHARACTERISTIC FOR BUILDING
VIBRATION IN TERMS OF HUMAN RESPONSE
FOR THE FREQUENCY RANGE 1 TO 80 Hz.
Note: Electrical network for low frequency cutoff below
1 Hz and high frequency cutoff above 80 Hz not yet
standardized.
91
-------
3.6 x 10~3m/s2, evaluated by means of the weighting described in Figure
11. Ebr hospital operating areas and other such critical areas, no higher
levels should be permitted without analysis and justification of the
acceptability of such levels.
Ebr residential and other similar areas, continuous acceleration of
greater values are normally expected to cause virtually no complaints (less
than 1%). Even greater acceleration values could be permitted for shorter
times during the daytime (0700 to 2200 hours), as indicated by Table 12 and by
Figure 12. These also indicate that the maximum value of the impulsive shock
excitation that is expected to cause virtually no complaints can be raised,
dependent on the number of such impulses during the daytime. Ebr residential
areas or other areas where people sleep, the nighttime peak acceleration should
be less than 0.01 m/sec2 at any time and the continuous rms acceleration should
be below 0.005 m/sec2 if no complaints are to occur. No differentiation is
made as to the types of residential areas, i.e., city center, urban or rural.
Table 12
Basic Threshold Acceleration Values* for Acceptable Vibration Environments
Continuous or Impulsive Shock
Time Intermittent rms Excitation Peak
Type of Place of Day Acceleration (m/sec2) Acceleration (m/sec2)
Hospital Operating
Booms and Other Such
Critical Areas
Residential
Office
Factory and Workshop
Day
Night
Day
Night
Anytime
Anytime
0.0036
0.0036
0.072
t
0.005
0.14
t
0.28
—
0.005
0.005
0.1
N
0.01
0.2
N
0.4
"if
*Weighted as shown in Figure 11.
t = duration seconds of vibration, for durations greater than 100 sec, use
t as 100 sec.
N s is the number of discrete shock excitations that are one sec or less in
duration. Ebr more than 100 excitations, use N = 100.
Daytime is 7 am to 10 pm. Nighttime is 10 pm to 7 am
92
-------
vo
ACCELERATION (METERS/SEC2)
|_ 1 1 1 2. B § b g
~ 1 Threshold of risk of damage to normal dwelling-houses with plastered
.X \ ceiling and walls
*^x**x Threshold of risk
- ^~<>
ISO PROPOSAL 7*^ ,
^ 3 IMPULSES/DAY !
x^x
1
of damage to sensitive structures \
^ , ,
^ x Daytime peak impulses (complaints <20%)>^
\ Daytime rms (complaints <20%K^
** Daytime peak impulses ; (complaints <1%) *-*^
^"V^ Daytime rms (complaints o%) -^^
Nighttime rms (complaints <1%) • — ^
I I 1
1m/sec2
0.5m/sec2
0.01 m/sec2 „
0.0072m/sec^
0.005m/sec2
101
103
J SEC OR NR. OF
8
24
IMPULSES
(HOURS)
NUMBER OF IMPULSES PER DAY OR EXPOSURE TIME
FIGURE 12. VIBRATION CRITERIA FOR RESIDENTIAL AREAS
-------
For office type spaces, the threshold at which no adverse effects occur
is twice the daytime residential rms or peak value. No distinction is made
between daytime and nighttime exposure.
For factory and similar type spaces, the threshold at which no effects
occur is 4 times the daytime residential values. No distinction is made
between daytime and nighttime exposure.
Offices and workplaces may in many cases require vibration levels as
low as residential areas if any adverse reactions are to be avoided. In
certain critical areas, such as operating rooms and laboratories and possibly
research laboratories, standards rooms, tool rooms and the like, even lower
vibration exposures levels may be required than indicated by Table 12.
The acceleration values that are specified to cause less than 1% com-
plaints are near or at the perception threshold level of vibration during
normal activity and should serve as a realistic threshold of any adverse
reaction to the vibration. The percentage of complaints likely to occur for
higher levels of vibration are shown in Figure 13, which summarizes the
complaint history from the Salmon Nuclear Event [43]. For a single event
the number of complaints for residential areas varies roughly as 10 log K
(for peak acceleration range of 0.1 m/sec2 to 1.0 m/sec2), where K is
the ratio of the observed acceleration to 0.1 m/sec2.
4.1.3 Quantification of the impact
There is a lack of data related to the assessment of the severity of
the impact that results if the vibration guidelines proposed in this
section are exceeded. It is recoranended that the number of people exposed
to vibration levels above the "no complaint" value (Table 12) be estimated.
94
-------
0.1
5 10 15
PERCENTAGE COMPLAINTS
Figure-13. Percentage of Population Complaining as a Function of Peak Acceleration (Source: Reference 43)
95
-------
For a specific action, therefore, contours of the appropriate "no complaint"
acceleration values as determined by Table 12 should be predicted or measured.
For example, if an action causes a steady vibration that lasts a total of
25 seconds a day (during daytime hours), the contour of 0.014 m/sec^ should
be evaluated (0.072/ 25 = 0.014).
In addition to the mapping and tabulation of the impact, which cover
sensitive non-residential as well as residential buildings, single-number
indexes can be calculated which are similar to those suggested for general
audible noise (the level-weighted population and hearing-weighted population).
These indexes are based on the relationship for the percent complaining,
documented in Figure 13. It is suggested that this concept be tentatively
broadened to apply the vibration exposure to more than one impulse or to
intermittent/continuous exposures by using the ratio (k) of the actual
acceleration to the recommended "no complaint" acceleration value. A term
for the impact of vibration on residential areas can then be defined by using
a vibration weighting function. This function is described by:
V(k) = 20 log k Eqn. 20
where k is the ratio of the actual acceleration to the recommended "no
complaint" acceleration values listed in Table 12 for a specified time
period, and where k is limited to values from 1 to 20.
This function can be used to calculate a descriptor of the total
vibration impact of a project, by multiplying the number of people
exposed to each vibration condition by the vibration weighting function
for that condition, and then finding the sum of these products. This
Vibration-Weighted Population (VWP) is defined as:
k
VWP = /P(k) V(k) dk Eqn. 21
1
96
-------
where V(k) is the vibration weighting function described above, P(k) is the
population distribution function, and dk is the differential change in k. An
index, similar to the Noise Impact Index, but applied to vibration, is called
the Vibration Impact Index (VII) and is calculated as: \/ll
k
/* P(k) V(k) dk
VII = -! Eqn. 22
k
/ P(k) dk
1
where the denominator is based on the alternative affecting the largest
number of people. In other words, the base population for calculating the
vibration impact index needs to be constant across alternatives for the
number to be meaningful. Given that restriction, then changes in VWP and
VII can both be used to evaluate various alternatives and actions with
respect to vibration. The change can also be discussed by listing the
expected effects at the nearest residence.
4.2 Structural effects of vibration
A structural vibration velocity of 2 in/sec has commonly been used as
the safe limit, and certainly vibrations above this value will have a
very adverse environmental impact. Note that, except for frequencies below
3 Hz, if the acceleration measured with the weighting network of Figure 11
is less than 1 m/sec2, then the velocity will be 2 in/sec or less. For
frequencies from 10 Hz to 80 Hz a weighted acceleration of 1 m/sec2 is
essentially equivalent to a velocity of 1 in/sec. In most practical cases,
in which the acceleration is made up of several frequency components, an
acceleration of less than 1 m/sec2 will also mean that the resultant velocity
will be less than 2 in/sec, and possibly less than 1 in/sec, regardless of
97
-------
frequency. Therefore, it is recommended that 1 m/sec2 be used as the normally
safe acceleration with respect to structural damage. Vibrations above this
should be avoided, or special arrangements should be made with the owners
of the exposed structures. Since some minor damage has occasionally been
reported at vibration as low as 1 in/sec, (0.5 m/sec2 to 1 m/sec2), exposures
in the range between 0.5 m/sec2 and 1 m/sec2 should also be regarded as a
potentially adverse exposure with respect to structural damage. Finally, the
safe peak acceleration for ancient monuments or ruins should be considered
as 0.05 m/sec2. Higher exposure values for such ancient structures should
not be considered safe without a detailed structural analysis.
No single-number index is suggested for summarizing the structural
effects. Quantification of the impact will consist of a contour map and
tabulation, showing the number of structures above the potentially damaging
accelerations of 1 m/sec2 and 0.5 m/sec2. A description of the expected
damage and the likelihood of such damage occurring should be provided for
each type of structure. The information in Appendix D will be of some help
in making this assessment, but sufficient data will not often be available to
make this assessment fully. In such cases, a program for monitoring the
actual damage, or lack of it, may be necessary.
98
-------
Chapter 5
Sumnary Of Noise Impact Analysis
This chapter provides an overview of the analysis that might be expected
to characterize noise impact fully, by summarizing each of the four preceding
chapters. In addition, Figure 1 and Table 1 provide useful overviews of the
kinds of analyses suggested.
5.1 Purpose and structure of the guidelines
These guidelines contain procedures which can be used to describe and
quantify the noise related impacts of proposed projects. The resulting
description of noise impacts is intended to be easily understood by those
making decisions, so that consideration of these impacts can be an integral
part of the decision. The approach described here is applicable to any situa-
tion calling for the evaluation of noise-related impacts, such as EIS or envi-
ronmental assessment preparation for the NEPA process, and is consistent with
noise evaluation procedures used by EAA, EHWA, and HUD, among others. The
approach is not mandatory, but is meant to complement these other procedures
by showing how to proceed to a quantitative description of impacts on people
(which all procedures desire) from information on noise levels (which those
procedures require).
These guidelines provide procedures for arriving at verbal, tabular, and
single number descriptions of noise environments. The quantitative approaches
rely on tables detailing the affected area or population, and on a modifica-
tion of the earlier fractional impact method [44] to reduce the tabulated
information to a single number index. These descriptions should be applied
to future as well as to immediate impacts.
99
-------
Three principal types of noise environments are covered: general audible
noise; special noises; and vibration. There is a separate chapter for each,
which covers (1) the appropriate physical measurement, (2) methods for predict-
ing that measure for the proposed project and for determining the existing
levels, (3) human noise exposure criteria, and (4) procedures for quantifying
impact. Within the chapter on general audible noise, three subsections are
provided, entailing different approaches for human exposure criteria and
quantification procedures in different noise ranges: urban and suburban
settings (L, usually 55 to 75 dB); projects producing greater than 75 dB
(L, ); and rural and wilderness areas (L, usually less than 55 dB).
Additional types of proposed actions for which these guidelines will be
useful are projects which entail new populations to be introduced into noisy
areas, and actions which are intended to reduce noise. The impact of temporary
projects may be evaluated using a more simplified analysis. For all impact
analysis, the necessary estimation and prediction entail uncertainty. When
possible, the degree of uncertainty should be specified. In some circumstances,
optimistic and pessimistic forecasts can be used to bracket the estimate.
5.2 Analysis of impacts of general audible noise
General audible noise is noise as commonly encountered in the environ-
ment. Therefore, the material in this chapter should cover the great major-
ity of situations in which an evaluation of noise impacts is desired. The
primary measure of general audible noise is L, , and whenever possible,
an approximation to the annual average value should be used. In some
instances this measure is inappropriate, and shorter term measures such as
1-hour L or the sound exposure level should be used. The screening diagram
(Figure 3) shows that whenever the noise level after the project will be greater
100
-------
than the existing level a noise analysis is necessary (i.e., when the existing
level is less than 10 dB greater than the project noise level). The diagram
applies to both permanent and temporary projects.
Depending on the approximate range of L^ values, different types of
noise effects are of concern, and therefore different analyses are needed
(Figure 14). At levels generally encountered in populated areas (approx. L,
values of 45 dB to 80 dB), the general health and welfare effects of noise
are the primary concern. At levels above 75 dB (8-hour L at-ear) severe
64
health effects become important. The threshold level at which these should
be investigated is an L, of 75 dB. In rural or wilderness areas, with
very low residential populations, environmental degradation is as much a
concern as the effect of noise on residents. In such areas, judgment will
have to be used in deciding between a health and welfare analysis and an
environmental degradation analysis, depending more on characteristics of the
area than on the existing or project noise level.
Regardless of which noise effects are the focus, two elements are always
recommended for describing the impact. The first is a table (or set of tables)
setting out the number of people and total area affected as a function of
different noise levels. Five decibels is usually an appropriate interval to
use for those tables. The second element is a verbal, qualitative description
of the principal components of the impacts identified in the tables.
For general health and welfare effects and for severe health effects
the quantitative analysis can proceed further, to calculate a single-number
index which summarizes all the impacts. The human noise effects information
discussed in the Levels Document apply in the general health and welfare
effects range: speech interference; sleep interruption; annoyance; and
101
-------
SEVERE HEALTH EFFECTS (SEC. 2.3) VjS&X
HEALTH AND
WELFARE EFFECTS
HEALTH AND WELFARE
EFFECTS (SEC. 2.2)
: ENVIRONMENTAL
: DEGRADATION
(SEC. 2.4)
J5Q 70
EXISTING Ldn(y)
90
FIGURE 14. TYPES OF ANALYSES SUGGESTED
102
-------
possible health effects. Given the existence of Schultz's synthesis [5], it
is appropriate to use the percent of people who report being highly annoyed
as the indicator of general adverse response, and to use his equation to
summarize the total impact of noise on residential areas in terns of the
number of people responding adversely to the noise. In the severe health
effects range, the human noise exposure effects may include cardiovascular
effects and other stress related health problems. It is not known yet at
what levels these begin to occur, but it is known that at an 8-hour L of
^H
75 dB hearing damage (NIPTS) begins to occur. The curve for average NIPES
versus L .g. is used as the function to reduce tabulated data for
these extreme levels to a single-number indicator, because it is the only
direct health effect for which such a function has been established.
5.3 Analysis of impacts due to special noises
The special noises discussed in this document are impulse noise, infra-
sound, ultrasound, and noises with information content. Effects on humans,
structures, and animals all need to be considered.
For any special noise, the main task is to describe the noise environ-
ment for the population. As with general audible noise, tables such as those
in Chapter 2 may be needed. Except for large impulse sounds, only a word
description of the effects of the special noise is recommended. The criteria
of Chapter 3 should be referenced, but in many cases additional reference
material may be required. A discussion of previous experience with such
noises should be made, if available. For high energy impulse noise, the
analysis can be carried further and the expected percent highly annoyed, and
changes in this quantity, can be estimated.
For structures exposed to impulse noise, the noise environment should
be described for each building or set of buildings in terms of maximum
103
-------
sound pressure levels. Either a worst case or a statistical estimate of the
distribution of maximum levels should be provided. A discussion of the pos-
sible damaging effects of noise on structures or monuments is required. The
chance that such effects could occur should be estimated. Finally, the sig-
nificance of such damage, in monetary and/or non-monetary terms, should be
reviewed.
5.4 Analysis of impacts due to vibration
If people are exposed, the analysis should include documentation of the
vibration environment such that the expected vibration acceleration values
due to the action are provided for all residential and other sensitive areas
in which the weighted acceleration exceeds the "no complaint" level (Table 12).
The change in the vibration environment can be discussed both by using the
average Vibration Impact Index for the exposed population and by listing the
expected effects at the nearest residence. A discussion of the effects of
the vibration environment on sensitive non-residential buildings is also
needed.
When structures are exposed to potentially damaging vibration, a descrip-
tion of the expected damage and the likelihood of such damage occurring should
be provided for each type of structure. The information in Appendix C will
be of some help in making this assessment, but often enough data will not be
available to fully make this assessment. In such cases, a program for moni-
toring the actual damage, or lack of it, may be necessary.
104
-------
References
1. Council on Environmental Quality. National Environmental Policy Act-
Regulations, Federal Register, Vol. 43, No. 112, June 9, 1978,
pp. 25230 - 25247.
2. "Information on Levels of Environmental Noise .Requisite TO Protect Public
Health and Welfare with an Adequate Margin of-Safety," Report 550/9-74-004,
U.S. Environmental Protection Agency, March 1974.
3. "Human Response to Impulse Noise," Report.of Working Group 84, Committee
on Hearing, Bioacoustics and Biomechanics, The National Research Council,
Washington, D.C.
4. Von Gierke, H.G. (Task Group Chairman). "Impact Characterization of Noise
Including Implications of Identifying and Achieving Levels of Cumulative
Noise Exposure," Report NTID 73.4, U.S. Environmental Protection Agency,
Aircraft/Airport Noise Study Report, July 1973.
5. Schultz, T.J., "Synthesis of social surveys on noise annoyance," J. Acoust.
Soc. Am., 64 (2), 1978, pp. 377-405.
6. "Public Health and Welfare Criteria for.Noise," Report 550/9-73-002,
U.S. Environmental Protection Agency, July 1973.
7. "Community Noise," Report NTID 300.3, U.S. Environmental Protection Agency,
December 31, 1971.
8. "Code of current practices for enforcement of noise ordinances," forth-
coming report, U.S. Environmental Protection Agency.
9. Kurze, U.J., Levison, W.H., and Serben, S., "User's Manual for the
Prediction of Road Traffic Noise Computer Programs," Report DOT-TSC-315-1,
U.S. Department of Transportation, May 1972.
10. Gordon, C.G., Galloway, W.J., Kugler, B.A., and Nelson, D.L., "Highway Noise-
A Design Guide for Engineers," Report 117, National Cooperative Highway
Research Program, 1971.
11. Kugler, B.A., and Pierson, A.G., "Highway Noise - A Field Evaluation of
Traffic Noise Reduction Measures," Report 144, National Cooperative Highway
Research Program, 1973.
- . ' v.
12. "Calculation of Day-Night.Levels- (L,) Resulting from Civil Aircraft
Operations", Report 550/9-77;-450, .DVS. Environmental Protection Agency, 1977.
13. "FAA Integrated Noise Model Version 1:. User's 'Guide". Report FAA-EQ-78-01.
Federal Aviation Administration, 1978.
14. "Planning in the Noise Environment," draft joint services noise manual under
cover of the Environmental Planning Bulletin Series No. 12, dated Dec 1976.
ft-1
-------
15. Galloway, W.J., Eldred, K. McK., and Simpson, M.A., "Population Distri-
bution of the United States as a Function of Outdoor Noise Level,"
Report 550-9-74-009, U.S. Environmental Protection Agency, June 1974.
16. Borsky, P.M., "The use of social surveys for measuring community responses
to noise environments", in Transportation Noises, J.D. Chalupnik (ed).
University of Washington Press, Seattle, 1970, pp. 219-227.
17. Schultz, T.J., "Social surveys on noise annoyance - further considerations."
Paper presented at the International Congress on the Biological Effects of
Noise, Freiburg, Germany, October 1978.
18. "Noise-Final Report," H.M.S.O., Cmnd. 2056, London, July 1963.
19. Connor, W.K. and Patterson, H.P., "Comnunity Reaction to Aircraft Noise
Around Smaller City Airports", NASA CR-2104, August 1972.
20. "Social and Economic Impact of Aircraft Noise," Sector Group on the Urban
Environment, OECD, April 1973.
21. Lukas, J.S., "Measures of Noise Level: Their Relative Accuracy in Predicting
Objective and Subjective Responses to Noise during Sleep," Report
600/1-77-010. U.S. Environmental Protection Agency, February 1977.
22. Goldstein, J., "Assessing the impact of transportation noise: human response
measures," in Proceedings of the 1977 National Conference on Noise Control
Engineering, G.C. Maling (ed). Noise Control Foundation, Poughkeepsie,
New York, 1977, pp. 79-98.
23. Goldstein, J. and Lukas, J.S., "Noise and Sleep: Information Needs for
Noise Control," in Proceedings of the Third International Congress on
Noise as a Public Health Problem, J.V. Tobias (ed). American Speech
and Hearing Association, Washington, D.C., 1979.
24. "Noise Emission Standards for Surface Transportation Equipment: Regulatory
Analysis of the Noise Emission Regulations for Truck-Mounted Solid Waste
Compactors," Report 550/9-79-257, U.S. Environmental Protection Agency, August
1979.
25. Galloway, W.J., "Evaluating the Impact on the Public Health and Welfare of a
Change in Environmental Noise Exposure," Appendix D of "Design Guide for
Highway Noise Prediction and Control," Final Report Project 3-7/3, Submitted
to Transportation Research Board, NCHRP, National Academy of Sciences,
November 1974.
26. Baughn, W.L., "Relation Between Daily Noise Exposure and Hearing Loss Based
on the Evaluation of 6,835 Industrial Noise Exposure Cases," Joint EPA/USAF
Study, prepared for 6570th Aerospace Medical Research Laboratory, Wright-
Patterson AFB, Ohio. AMRL-TR-73-53, 1973.
R-2
-------
27. Passchier-Vermeer, W., "Hearing Loss Due to Exposure to Steady-State Broad-
band Noise," Instituut Voor Gezondheidstechniek, Sound & Light Division,
Report 35, April 1968 (with supplement, 1969).
28. Robinson, D.W., "Estimating the risk of hearing loss due to exposure to
continuous noise," Occupational Hearing Loss, British Acoustical Society
Special Volume No. 1, London and New York, Academic Press, 1971, pp. 43-62.
29. Johnson, D.L., "Prediction of NIPTS Due to Continuous Noise Exposure"
Report 550/9-73-001-B (or AMRL-TR-73-91), U.S. Environmental Protection
Agency, 1973.
30. Borsky, P.N., "Community Reactions to Sonic Booms in the Oklahoma City
Area," Report AMRL-TR-65-37, National Opinion Research Center, 1965.
31. Schemer, P.O., "Community Reaction to Impulse Noise: First Army Survey,"
Draft CERL Technical Report, U.S. Army Construction Engineering Research
Laboratory, 1980.
32. Schemer, P.O., "Evaluation of C-weighted L, for assessment of impulse
noise," J. Acoust. Soc. Am., 62 (2), 1977,^p. 396-399.
33. Schomer, P.D., "Human and Community Response to Impulse Noise: A
Literature Review," Illinois Institute for Environmental Quality
Document No. 78/07, March 1978.
34. Johnson, D.R. and Robinson, D.W., "The subjective evaluation of sonic
bangs," Acustica, 18, 1967, pp. 241-258.
35. Pawlowska, V. and Little, L., "The Blast Noise Prediction Program: User
Reference Manual," Interim Report N-75, Construction Engineering Research
Laboratory," August 1979.
36. Schomer, P.D., "Growth functions for human response to large-amplitude
impulse noise," J. Acoust. Soc. Am., 64 (6), 1978, pp. 1627-1632.
37. Schomer, P.O., Goff, R.J., and Little, L.M., "The Statistics of Amplitude
and Spectrum of Blasts Propagated in the Atmosphere," Vol. I., Report
CERL-TR-N-13, U.S. Army Construction Engineering Research Laboratory,
1976.
38. "Standard for Single Point Explosions in Air" draft report dated Aug 20,
1976 of Working Group S2-54, Committee on Mechanical Vibration and
Shock, American National Standards Institute.
39. Johnson, L., "Auditory and physiological effects of infrasound," Inter-
noise 75 Proceedings. Sendai, Japan, 1975, pp. 475-482.
40. Von Gierke, H.E. and Parker, D.E., "Infrasound", in Handbook of Sensory
Physiology, Vol. 5, Auditory System, Part 3, Springer-Verlag, Berlin-
Heidelberg-New York, 1976, pp. 585-624.
R-3
-------
41. "Guide for the Evaluation of Human Exposure to Whole-body Vibration,"
ISO 2631-1978, International Organization for Standardization, and
Addendum 2, "Vibration and Shock Limits for Occupants in Buildings."
42. "Guide for the Evaluation of Human Exposure to Whole-body Vibration,"
ANSI S3.18-1979, American National Standards Institute.
43. Nicholls, H.R., Johnson, C.F., and Ouvall, W.I., "Blasting Vibrations
and their Effects on Structures," Bulletin 656, U.S. Department of the
Interior, Bureau of Mines, 1971.
44. "Guidelines for Preparing Environmental Impact Statements on Noise,"
Report of Working Group 69, Committee on Hearing, Bioacoustics and
Biomechanics, The National Research Council, Washington, D.C., 1977.
R-4
-------
Appendix A
Acoustical Terms and Symbols Used in the Guidelines,
and Sane Mathematical Formulations for Them
A.I. Acoustical terms
Some acoustical terms are defined or described here, which have been
used in the main body of this report. They are arranged alphabetically, to
facilitate finding them as needed. Three key terms — sound level, equivalent
sound level, and sound exposure level — receive non-technical as well as
technical descriptions.
A. 1.1 C-weighted sound exposure level. In decibels, the level of the time
integral of C-weighted squared sound pressure, with reference to the square
of 20 micropascals and to one second.
A. 1.2 day-night sound level. The 24-hour equivalent sound level, in deci-
bels, obtained after addition of 10 decibels to sound levels in the night
from midnight up to 7 a.m. and from 10 p.m. to midnight (0000 up to 0700 and
2200 up to 2400 hours).
A. 1.3 day-night sound level contour. A curved line connecting places on a
map where the day-night sound level is the same. If only one kind of contour
is shown on the map the fact may be made known by a single legend, "Contours
of day-night sound level in decibels." In this case only the number of deci-
bels need be marked on a coutour.
A. 1.4 day sound level. Equivalent sound level over the 15-hour time period
from 7 a.m. up to 10 p.m. (0700 up to 2200 hours).
A. 1.5 decibel. A unit measure of sound level and other kinds of levels. It
is~a logarithmic measure. For sound level specifically it is equal to 10 log
(P VP ref) or 20 log (p/Pref).
A. 1.6 8-hour equivalent C-weighted sound level. Equivalent sound level, in
decibels, over a given 8-hour time period, measured with the C-frequency
weighting.
A. 1.7 8-hour equivalent sound level. Equivalent sound level, in decibels,
over an 8-hour period. The A-frequency weighting is understood.
A. 1.8 equivalent sound level. A sound level typical of the sound levels at
a certain place in stated time period. Technically, equivalent sound level
in decibels is the level of the mean-square A-weighted sound pressure during
the stated time period, with reference to the square of the standard refer-
ence sound pressure of 20 micropascals. Equivalent sound level differs from
sound level in that for equivalent sound level, equal emphasis is given to
all sounds within the stated averaging period, whereas for sound level an
exponential time weighting puts much more emphasis on sounds that have just
occurred than those which occurred earlier.
A-l
-------
A. 1.9 fast sound level. In decibels, the exponential-tine-average sound
level measured with the squared-pressure tine constant of 125 ms.
A. 1.10 hourly equivalent sound level. Equivalent sound level, in decibels,
over a one-hour time period, usually reckoned between integral hours. It may
be identified by the beginning and ending tines, or by the ending tine only.
A. 1.11 impulse sound level. In decibels, the exponential-tine-average sound
level obtained with a squared-pressure time constant of 35 milliseconds.
(But note that there is as yet no U.S. standard for this.)
A.1.12 instantaneous sound pressure, overpressure. Pressure at a place and
instant considered, minus the static pressure there.
A.1.13 maximum sound pressure level. Sane as peak sound pressure level,
provided that the tine interval considered is not less than a complete
period of a periodic wave.
A. 1.14 night sound level. Equivalent sound level, in decibels, over the nine-
hour period from midnight up to 7 a.m. and from 10 p.m. to midnight (0000 up
to 0700 and 2200 up to 2400 hours).
A.1.15 noise level. Sane as sound level, for sound in air. Some people use
"noise" only for sound that is undesirable. A sound level meter does not,
however, measure people's desires. Hence there is less likelihood of mis-
understanding, if what is measured by a sound level meter is called sound
level, rather than noise level.
A. 1.16 peak sound pressure. Greatest absolute instantaneous sound pressure
in a stated frequency band, during a given tine interval. (Also called
peak pressure.)
A. 1.17 peak sound pressure level. In decibels, twenty tines the cannon
logarithm of the ratio of a greatest absolute instantaneous sound pressure
to the reference sound pressure of twenty micropascals.
A. 1.18 slow C-weighted sound level. In decibels, the exponential time aver-
age sound level measured with the squared-pressure time constant of one second
and the C-frequency weighting of the sound level meter.
A.1.19 slow sound level. In decibels, the exponential-tine-average sound
level measured with the squared-pressure tine constant of one second.
A. 1.20 sound exposure. Tine integral of squared, A-frequency-weighted sound
pressure over a stated tine interval or event. The exponent of sound pres-
sure and the frequency weighting may be otherwise if clearly so specified.
A.1.21 sound exposure level. The level of sound accumulated over a given
tine period or event. It is particularly appropriate for a discrete event
such as the passage of an airplane, a railroad train, or a truck. Sound
exposure level is not an average, but a kind of sum. In contrast to equi-
valent sound level which may tend to stay relatively constant even though
A-2
-------
the sound fluctuates, sound exposure level increases continuously with the
passing of time. Technically, sound exposure level in decibels is the level
of the time integral of A-weighted squared sound pressure over a stated time
interval or event, with reference to the square of the standard reference
pressure of 20 micropascals and reference duration of one second.
A.1.22 sound level. The weighted sound pressure level, which reduces to a
single number the full information about sound pressure levels across the
frequency range 20 Hz to 20 kHz. It can be measured by a sound level meter
which meets the requirements of American National Standard Specification for
Sound Level Meters SI.4-1971. In these guidelines, fast tine-averaging and
A-frequency weighting are understood, unless others are specified. The
sound level meter with the A-weighting is progressively less sensitive to
sounds of frequency below 1000 hertz (cycles per second), somewhat as is the
ear. With fast time averaging the sound level meter responds particularly
to recent sounds almost as quickly as does the ear in judging the loudness
of a sound.
A. 1.23 sound pressure. Rcot-mean-square of instantaneous sound pressures
over a given time interval. The frequency bandwidth must be identified.
A. 1.24 sound pressure level. In decibels, twenty times the common logarithm
of the ratio of a sound pressure to the reference sound pressure of twenty
micropascals (0.0002 microbar). The frequency bandwidth must be identified.
A.I.25 (vibratory) acceleration. The rate of change of velocity of a vibra-
tion, in a specified direction. The frequency bandwidth must be identified.
A. 1.26 (vibratory) acceleration level. In decibels, tarenty times the common
logarithm of the ratio of a vibratory acceleration to the reference accel-
eration of ten micrometers per second squared (nearly one-millionth of the
standard acceleration of free fall). The frequency bandwidth must be
identified.
A.1.27 yearly day-night sound level. The day-night sound level, in decibels,
averaged over an entire calendar year.
A.2. Symbols used in the guidelines
A. 1.1 C-weighted sound exposure level
A.1.2 day-night sound level
A. 1.5 decibel
A.I.7 8-hour equivalent sound level
A.1.8 equivalent sound level Lea
A. 1.10 hourly equivalent sound level Leam
A.I.21 sound exposure level LT*
A. 1.23 sound pressure p
A.1.25 vibratory acceleration a
A.I.27 yearly day-night sound level Ldn(v)
A-3
-------
A.3. Mathematical Formulations for the Descriptors used in the guidelines
A.3.1 Equivalent sound level
T
Leq = 10
10
*/.
dt
Eqn A-1
where: T is the length of the tine interval during which the
average is taken, and L.(t) is the time varying
value of the A-weightea sound level during the time
interval T.
Note: Equivalent sound level may be calculated from the sound
exposure levels of individual events occurring within
the time interval T:
eq
10
10
Eqn A-2
where: Lg. is the sound exposure level of the i-th event, out
of a total of n events in time interval T. Lu is
defined in A.2.3.4. b
A.3.2 Day-night Sound Level
Ldn=10
10
1
86400
0700
10
dt
0000
• f
2200 2400
L.(t)/10
10 A dt +
0700
2200
10
Time t is in seconds, so the limits shown in hours and minutes
are actually interpreted in seconds. It is often convenient to
compute day-night sound level from hourly equivalent sound
levels obtained .during successive hours:
10910
Eqn A-4
where L^i is the hourly equivalent sound level for the i-th hour
of the day and LHJ is the hourly equivalent sound level for the
j-th hour of the night.
A-4
-------
A. 3. 3 Yearly Day-night Sound Level
t 365 L, ./10
Ldn(y) = 1 10 355
where: l^ni *-s tne day-night average sound level for the
i-th day out of one year.
A. 3. 4 Sound Exposure Level
Eqn A-6
where: LA(t) is the time-varying A-weighted sound level in
some time interval t. to t_.
The length of the time interval may be arbitrary, or it may
simply be large enough to encompass all the significant sound
of an event.
t
Note: The value of the above integral is usually approximated
* with sufficient accuracy by integrating LA(t) over the
time interval during which LA(t) is between 10 decibels
less than its maximum value and the maximum value, before
and after the maximum occurs.
A.3.5 C-weighted Sound Exposure Level
1^- 10 Iog10\ y 10 *- dt/ EqnA-7
\ti /
where: Lg(t) is the time-varying C-weighted sound level in
some time interval t-j to t2.
Note: In practice the integral is often approximated by inte-
gration within the time during which the sound level of
the event exceeds some threshold value such as 20 dB
less than the maximum sound pressure level.
A.3.6 C-weighted Day Night Sound Level
Analogous to the A-weighted L^n, with a nighttime penalty of 10 dBf the
C-weighted day-night average sound level is:
A-5
-------
^dn
- 10
J_
24
15 x 10
-
10
9 x 10
+ 10
10
Eqn A-8
is the average C-weighted sound level over the daytime period of
0*700 to 2200 hours, L is the C-weighted average level over the night-
time period of 2200 tff\)700 hours.
The C-weighted average level is most easily calculated from the
C-weighted sound exposure levels during the time of interest as follows:
10
1
15 x 3600
£10
10
for
>80
Eqn A-9
10 log
3600
n
10
for
>70
Eqn A-10
where L^ is the C-weighted sound exposure level of the i-th
discretfcTevent.
A-6
-------
Appendix B
Procedures of Other Federal Agencies
B-1
-------
bl. Knvironucntal Noise Measures and Their
Purposes in Federal Programs
ACtNCY
Ty|je of Program
or Policy
Key Documents
1. FIIUA
Highway Noise
Policy
FIlPM 7-773
(1976)
±. EPA
Health & Welfare
Guidance
EPA "Levels"
Document (1974)
3. HUD
HUD Noise
Policy
hUD Circular
1390.2 (1971)
4. DOD
Air Installation
Compatible Use
Zones (AICUZ)
Program
DOD Instruction
4165.57 (1977)
5. FAA
Airport 6 Land
Use compatibility
Planning
Advisory Circular
150/5050-b (1977)
b. VA
VA Noise
Policy
Section vm
Appraisal of
residential
properties
near Airports
(1969)
GO
PO
Title of
Levels
furpose of
Levels
Noise
Levels
source to
wh ich
applied
Noise
Descriptors
Used
Design Noise
Levels
These levels are
used in deter-
mining where
noise mitigation
on a particular
highway project
is warranted.
They reflect cost
and feasibility
considerations.
They are not
appropriate land
use criteria.
Location
Specific
Highway only
Levels which are
required to protect
the public health and
welfare with an ade-
quate margin of safety
These levels identify
in scientific terms
the threshold of
effect. While the
levels have relevance
for planning, they
do not in themselves
form the sole basis
for appropriate land
use actions because
they do not consider
cost, feasibility or
the development needs
of the community. The
user should make such
tradeoffs.
All sources
or L|0
for design hour
Level determin-
ing whether pro-
jected sites arc
eligible for HUD
assistance
See above. Levels
can be used as
general planning
levels. Considers
balance between
cost, feasibility
effect, community
developuent needs
and availability
of land for deve-
lopment . Com-
munity wide con-
sideration.
All sources
airport, highway
and rail
Various
(accepts Ldn)
Levels used as
"reasonable"
guidance to
communities in
planning
Guidance to com-
munities for
planning. Considers
balance between
cost, feasibility
effect, community
development needs,
and availability
of land for
development.
Community wide
consideration.
Levels used as
"starting points"
in determining
appropriate land/
use relationships
Guidance to com-
munities for plan-
ning. Considers
balance between
cost, feasibility
effect, community
development needs,
and availability
of land for
development.
Military
Airfields
Ldn
Civil Airports
Various
(including NEF
and Ldn)
Levels determi-
ning whether
projected sites
are eligible fur
VA assistance
See above. Consi
balance between
cost, feasibilit
effect, communit
development need
and availability
of land for deve
ment needs and
availability of
land for develop
ment.
Airports only
Various
(including Ldn)
-------
B2. Estimating L, from other Noise Measures
The equations listed here are approximations only, and are provided for
use in those situations in which measurement or prediction of the other noise
measure is already available. If no such information is available, it is
strongly recommended that L, be measured or predicted directly, instead
of using these equations.
NEP: L^ » NEF + 35 Eqn B-1
CNR: L, = CNR - 35 Eqn B-2
CNEL: L^ = CNEL Eqn B-3
24-hour L^:3 L^ . L +4 Eqn B-4
Peak (traffic) hour I^q: *dn =
Peak (traffic) hour L-josb I<3n a L10 "3 &$\ B-€
Notes;
aSource: [4], Parts II B, F, and Addendum A, approximated.
Source: Department of Housing and Urban Development. Notice of proposed
rulemaking. Environmental criteria and standards, Federal Register, Vol 43,
No. 249, December 29, 1978, p. 60399. "The day-night average sound level may
be estimated from the design hour L-JQ or Lgq values by [these] relationships,
provided heavy trucks do not exceed 10 percent of the total traffic flow in
vehicles per 24 hours and the traffic flow between 10 p.m. and 7 a.m. does not
exceed 15 percent of the average daily traffic flow in vehicles per 24 hours."
B-3
-------
APPENDIX C
Summary of Human Effects of General Audible Noise
TABLE Cl Summary of Human Effects for Outdoor Day-Night Sound Level of
75 Decibels
Type of Effect Magnitude of Effect
Hearing Loss May begin to occur in sensitive individuals, depending
on actual noise levels received at-ear.
Risk of non-auditory *
health effects
(stress)
7*
Speech - Indoors Some disturbance of normal conversation. Sentence
intelligibility (average) approximately 98%
- Outdoors Very significant disturbance of normal voice or relaxed
conversation with: 100% sentence intelligibility not
possible at any distance
*
or, 99% sentence intelligibility (average) at 0.15 meter
or, 95% sentence intelligibility (average) at 0.5 meter
High Annoyance Depending on attitude and other non-acoustical factors,
approximately 37% of the population will be highly annoyed.
Average Ccnnunity Very severe; 13 dB above level of significant "complaints
Reaction and threats of legal action" and at least 3 dB above
"vigorous action" (attitudes and other non-acoustical fac-
tors may modify this effect).
Attitudes Towards Noise is likely to be the most important of all adverse
Area aspects of the community environment.
^Research inplicates noise as one of several factors producing stress-related health
effects such as heart disease, high-blood pressure and stroke, ulcers and
other digestive disorders. The relationships between noise and these
effects have not yet been quantified, however.
7*
The speech effects data in these tables are drawn from the Levels
Document, as follows. Indoor effects are based on Table 3, and on Fig. D-l,
with 15 dB added to the indoor level to obtain the outdoor reading. Outdoor
effects come from Fig. D-2, using L, (as determined with Fig. A-7). Both
Figures D-l and D-2 are based on steady noise, not on L . Table D-3
shows that for a fluctuating noise, the average percent interference can be
higher or lower than for steady noise with the same L . The values
given in this report are the best estimates of the interference.
C-l
-------
TABLE C2 Summary of Human Effects for Outdoor Day-Night Sound Level of
70 Decibels
Type of Effect
Hearing Loss
Risk of non-auditory health
effects (stress)
Speech - Indoors
- Outdoors
High Annoyance
Average Community Reaction
Attitudes Towards Area
Magnitude of Effect
Will not likely occur
See Table C1
Slight disturbance of normal conversation
approximately 99% sentence intelligibility
(average)
Significant disturbance of normal voice or
relaxed conversation with 100% sentence
intelligibility (average) possible only at
distances less than 0.1 meter
or
99% sentence intelligibility (average) at
0.3 meter
or
95% sentence intelligibility (average) at
0.9 meter
Depending on attitude and other non-
acoustical factors, approximately 25
percent of the population will be highly
annoyed.
Severe; 8 dB above level of significant
"complaints and threats of legal action,"
but at least 2 dB below "vigorous action"
(attitudes and other non-acoustical
factors may modify this effect)
Noise is one of the most important adverse
aspects of the community environment
C-2
-------
TABLE C3 Summary of Hunan Effects for Outdoor Day-Night Sound Level of
65 Decibels
Type of Effect
Hearing Loss
Risk of non-auditory health
effects (stress)
Speech - Indoors
- Outdoors
High Annoyance
Average Community Reaction
Attitudes Towards Area
Magnitude of Effect
Will not occur
See Table C1
Slight disturbance of normal conversation
99% sentence intelligibility (average)
with a 4 dB margin of safety
Significant disturbance of normal voice or
relaxed conversation with 100% sentence
intelligibility (average) at 0.15 meter
or
99% sentence intelligibility (average) at
0.5 meter
or
95% sentence intelligibility (average) at
1.5 meters
Depending on attitude and other non-
acoustical factors, approximately 15
percent of the population will be highly
annoyed.
Significant; 3 dB above level of significant
"complaints and threats of legal action,"
but at least 7 dB below "vigorous action"
(attitudes and other non-acoustical
factors may modify this effect)
Noise is one of the important adverse
aspects of the community environment
C-3
-------
TABLE C4 Suitmary of Human Effects for Outdoor Day-Night Sound Level of
60 Decibels
Type of Effect
Hearing Loss
Risk of non-auditory health
effects (stress)
Speech - Indoors
- Outdoors
Magnitude of Effect
Will not occur
See Table C1
No disturbance of normal conversation
100% sentence intelligibility (average)
with no margin of safety
Moderate disturbance of normal voice or
relaxed conversation with 100% sentence
intelligibility (average) at 0.2 meter
or
99% sentence intelligibility (average) at
0.6 meter
or
High Annoyance
Average Community Reaction
Attitudes Towards Area
95% sentence intelligibility (average) at
2 meters
Depending on attitude and other non-
acoustical factors, approximately 9
percent of the population will be highly
annoyed.
Slight to moderate; 2 dB below level of
significant "complaints and threats of
legal act ion," but at least 11 dB below
"vigorous action" (attitudes and other
non-acoustical factors may modify this
effect)
Noise may be considered an adverse aspect
of the connunity environment
C-4
-------
TABLE C5 Summary of Human Effects for Outdoor Day-Night Sound Level of
55 Decibels
Type of Effect
Hearing Loss
Risk of non-auditory health
effects (stress)
Speech - Indoors
- Outdoors
High Annoyance
Average Community Reaction
Attitudes Towards Area
Magnitude of Effect
Will not occur
See Table C1
No disturbance of normal conversation
100% sentence intelligibility (average)
with a 5 dB margin of safety
Slight disturbance of normal voice or
relaxed conversation with: 100% sentence
intelligibility (average) at 0.35 meter
or
99% sentence intelligibility (average) at
1.0 meter
or
95% sentence intelligibility (average) at
3.5 meters
Depending on attitude and other non-
acoustical factors, approximately 4
percent of the population will be highly
annoyed.
None expected; 7 dB below level of signi-
ficant "complaints and threats of legal
action/1 but at least 16 dB below "vigorous
action" (attitudes and other non-acoustical
factors may modify this effect)
Noise considered no more important than
various other environmental factors
C-5
-------
APPENDIX D
Measurement of and Criteria for Human Vibration Exposure
D.I. Introduction
The criteria for vibration exposure in this appendix will address 3 types
of effects. These three types of effects are: (1) whole body vibration of
humans, (2) annoyance and interference caused by building vibration, and (3)
structural damage from building vibration.
The existing state of knowledge is not complete in any of the above
three areas; however, there are existing I.S.O. standards that have been
approved or proposed. Summaries of these standards, along with other data,
provide the content of this appendix. Some simplification of the proposed
standards on building vibration and structural damage have been made in order
to provide a simple, unified and reasonable method for assessing the effects
of vibration.
D.2. Whole Body Vibration Criteria (Summary of Approved ISO Standard 2631-1978)
D.2.1 The Three Criteria for Evaluation of Whole Body Vibration
Experimental data show that there are various rather complex factors
that determine the human response to vibration. Evaluation of all these
factors is difficult at this time because of the paucity of quantitative
data, concerning man's perception of vibration and his response to it. Never-
theless, there is an international standard which does provide provisional
guidance as to what is acceptable human exposure to vibration for some types
of vibration.
In general, there are four physical factors of primary importance in
determining the human response to vibration. These are intensity, frequency,
direction, and exposure time of the vibration. The current International
Standard for vibration addresses three main human criteria. These are:
1. Preservation of working efficiency
2. Preservation of health or safety
3. The preservation of comfort
For environmental problems, the preservation of comfort is considered
the best criteria for evaluation of whether or not vibration significantly
changes the environment.
D.2.2 Types of Vibration Transmissions.
The standard lists basically three kinds of human response to vibration,
namely:
D-1
-------
(a) Vibrations transmitted simultaneously to the whole body surface
or substantial parts of it. This occurs when the body is immersed in a
vibration medium. There are circumstances in which this is of practical
concern; for example, when high intensity sound in air or water excites
vibrations of the body.
(b) Vibration transmitted to the body as a whole through the supporting
surface, namely/ the feet of a standing man, the buttocks of a seated man or
the supporting area of a reclining man. This kind of vibration is usual in
vehicles, in vibrating buildings and in the vicinity of working machinery.
(c) Vibrations applied to particular parts of the body such as the
head or limbs; for example, by vibrating handles, pedals, or head-rests,
or by the wide variety of powered tools and appliances held in the hand.
It is also possible to recognize the condition in which an indirect
vibration nuisance is caused by the vibration of external objects in the
visual field (for example, an instrument panel).
The International Standard 2631, however, applies chiefly to the common
condition (b) above; and, in particular, where the vibration is applied
through the principal supporting surface to the body of a standing or seated
man. In the case of vibrations applied directly to a reclining or recumbent
man, insufficient data are available to make a firm recommendation; this is
particularly true of vibration transmitted directly to the head, when toler-
ability is generally reduced. Tolerance may also be reduced when conditions
(b) and (c) exist together. Provisionally, however, the limits for the
standing or seated man may also be used for the reclining or recumbent man.
It must be appreciated that some circumstances will arise in which the rigor-
ous application of these limits would be inappropriate.
D.2.3 Direction of Vibration
Rectilinear vibrations transmitted to man should be measured in the
appropriate directions of an orthogonal coordinate system centered at the
heart. The standard specifies separate criteria according to whether the
vibration is in the longitudinal (H- az) direction or transverse (± ax or ay)
plane. Accelerations in the foot (or buttocks) - to head (or longitudinal)
axis are designated + az: acceleration in the fore-and-aft (anteposterior
or chest-to-back) axis, + ax; and in the lateral (right-to-left side) axis,
+ ay. These axes are illustrated in Figure D-1.
D.2.4 Acceptable Whole Body Vibration
The ISO standard identifies the 24-hr comfort level for rms pure
(sinusoidal single) frequency or rms value in third octave band for random
vibration as given in Table 0-1. As long as the vibration levels are below
the 24-hr levels, vibration should be considered to have no direct impact
on an individual, regardless of the duration of the exposure. The standard
does allow for increased exposure levels for shorter exposure times. Such a
tradeoff is given by Table D-1 for 8-hr and 1 min exposures. For other
D-2.
-------
z
ax, 3y, az = acceleration in the directions of the x, y, z axes
x axis = back to chest
y axis = right to left side
z axis = foot for buttock(s)-to-head
FIGURE D—1. Directions of co-ordinate system for mechanical vibrations influencing humans
D-3
-------
TABLE D-1 - Numerical values of "comfort boundary" for vibration acceleration
in the longitudinal, a_, direction (foot (or buttocks)-to-head
direction) (see FigureT>-1 and in the transverse, a or a . direc-
tion (back-to-chest or side-to-side) y
Values define the boundary in terms of rms value of pure (sinusoidal) single
frequency vibration; or rms value in third-octave band for distributed vibration.
ACCELERATION m/sec
Frequency (Hz)
(Center Frequency
of 1/3 Octave Band)
1
1.25
1.6
2.0
2.5
3.15
4.0
5.0
6.3
8.0
10.0
12.5
16.0
20.0
25.0
31.5
40.0
50.0
63.0
80.0
a
z
J_
1
1
1
1
1
1
0
0
0
0
1
1
1
2
2
3
4
5
7
8
min
.78
.59
.43
.27
.13
.00
.89
.89
.89
.89
.13
.43
.78
.25
.86
.56
.44
.71
.11
.89
B
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
hr
.2
.18
.16
.14
.13
.11
.1
.1
.1
.1
.13
.16
.2
.25
.32
.40
.51
.63
.79
.0
24 hr
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.07
.06
.06
.05
.04
.04
.04
.04
.04
.04
.04
.06
.07
.09
.11
.14
.18
.23
.29
.36
axoray
1 min
0
0
0
0
0
1
1
1
2
2
3
3
5
6
7
10
12
15
20
25
.63
.63
.63
.63
.79
.0
.27
.59
.00
.54
.17
.97
.08
.35
.94
.00
.70
.87
.00
.40
B
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
2
2
hr
.07
.07
.07
.07
.09
.11
.14
.18
.24
.29
.36
.44
.57
.71
.89
.13
.43
.78
.25
.86
24 hr
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
.03
.03
.03
.03
.04
.05
.06
.08
.10
.13
.16
.20
.25
.32
.40
.51
.63
.79
.00
.27
-------
exposure times and for the concept of a vibration dose, the basic standard
should be consulted. For occupational and recreational situations, the values
of Table D-l can be raised by a factor of 3.15 (10 dB) to predict the boundary
at which working efficiency may start to decrease. Increasing the accelera-
tion listed in Table D-l by a factor of 6.3 (16 dB) will give the boundary
necessary for the preservation of health and safety. Thus the 1 min values of
Table D-l as multiplied by a factor of 6.3 provides the maximum recommended
continuous acceleration to which an individual should be subjected. However,
assessment of acceleration above the comfort levels listed in Table D-l should
be made only by direct reference to the ISO standard. In the ISO standard
there are many considerations and limitations with respect to human exposure
to acceleration that can cause reduced efficiency or health and safety
problems.
D.3. Vibration Criteria for Occupants in Buildings. (Summary of 1980 draft
addendum^ to ISO Standard 2631-1978)
D.3.1 Scope
The proposed standard takes into account the following factors:
1. Type of Excitation -. for example transient (shock) and/or
steady vibration;
2. Usage of the Occupied Space in Buildings - for example, hospital
operating theatres, residential, offices and factories;
3. Time of Day;
4. Limits of Acceptability - in a proposal of this type there is no
hard and fast line of acceptability, but guidance is given as to
the level of complaint to be achieved at different levels of
vibration. In cases where sensitive equipment or delicate opera-
tions impose more stringent limits than human comfort criteria,
then the more stringent criteria should be applied.
D.3.2 Characteristics of Building Vibration
D.3.2.1 Direction of vibration
Because a building may be used for many different activities, standing,
sitting and lying may all occur, hence vertical vibration of the building may
enter the body as either Z axis, X axis or Y axis vibration, as shown in Figure
D-l. The Standard is written for all three axes of vibration, however, in cases
where it is not clear which direction to apply, it is often more convenient to
consider the combined Standard detailed in Sections D.3.3.4 below.
D.3.2.2 Random or multi-frequency vibration
Random or multi-frequency vibration represents a particular problem which
fortunately does not often occur in buildings. There is evidence from research
concerning the building environment to suggest that there are interaction
D-5
-------
effects between different frequencies of vibration. Under these circumstances
and for random vibration, the proposed standard recommends an overall weighting
method such as that in section D.3.3.4.
D.3.2.3 The characterization of impulsive shock and intermittent vibration
Continuous vibration of a repetitive nature is easy to identify and
classify. The borderline between impulsive shock and intermittent vibration
is difficult to define. Impulsive shock is characterized by a rapid build-up
to a peak followed by decay, and is typically excited in buildings by blasting,
forging presses or pile driving using an impact device. Intermittent vibration
may only last a few seconds, but is characterized by a build-up to a level
which is maintained for a considerable number of cycles. Examples of this in
buildings would be traffic excited vibration and vibration generated inside a
building by machinery starting up or on intermittent service. Pile driving by
modern methods using vibrating columns would also be classified as continuous
or intermittent vibration and not as impulsive shock.
The proposed standard recommends that impulsive shock created by forging
presses or conventional pile drivers should be treated in a similar manner to
continuous and intermittent vibration. Research has shown that vibration
which only occurs at a specific instance, for example domestic building vibra-
tion by a passing bus, causes the same level of annoyance as continuous
vibration.
Blasting which occurs only up to three tines per day is a special case.
The proposed standard recommends that building operations of this nature
should never take place at night due to the disturbance and that during the
daytime they should be limited to a small number of occurrences. The levels
of vibration generated due to blasting are on an order of magnitude greater
than traffic and general building vibrations, and can only be accepted on the
basis of very limited exposure.
D.3.2.4 Classification of buildings and building areas
The criteria of classification are in the standard derived from the
human reaction to vibration. In the home the highest standards are required,
and this is characterized by an absence of detectable vibration. Under other
conditions, such as offices and factories, there is some tolerance to vibra-
tion disturbance.
In the proposed Standard no differentiation has been made between differ-
ent types of residential areas, i.e. city centre, urban or rural. It is con-
sidered that similar standards should be met for all occupants of residential
property. Some types of areas have not been classified, i.e. restaurants
or places of entertainment, but common sense suggests the most appropriate
classification—for example standards in a restaurant should be similar to
those in residential property. It should be noted that certain entertainment
areas in long span buildings present particular problems from self-generated
vibration, such as that from dancing.
D-6
-------
Hospitals have not been given more restrictive levels in general because
there is sane evidence that patients prefer to be in touch to some extent with
the outside world, but operating theatres and laboratories should be considered
as critical areas.
D.3.2.5 Measurement of vibration
The use of "root mean square11 acceleration is reconmended as the standard
unit of measurement. If possible building vibration should be measured in
acceleration terms, but in some cases it may be found necessary to measure in
velocity or displacement due to equipment limitations. For these situations
the vibration should be treated as sinusoidal and the appropriate correction
factors, which are a function of frequency, used to transform either the mea-
surement or the standard into compatible units.
In the case of impulsive vibration or shock the instantaneous peak value
of velocity or acceleration is the preferred unit of measurement. A trace of
the vibration should be obtained upon a suitable instrument and the peak level
estimated. The notion should then be considered sinusoidal and the correction
factors applied for the difference between peak and rms, and the frequency
dependent factors used to transform either measurement or standard into com-
patible units.
If frequency analysis of the vibration is required, third octave filters
are recommended. In certain circumstances it may be useful to analyze the
vibration in terms of narrow fixed band width filters.
Measurement of vibration should be taken on the floor at the point of
greatest amplitude, comnonly found at mid-span. This should be close to the
point of entry of vibration to the human subject. Measurement should be
taken along the three orthogonal axes, and reference made to the appropriate
human axis standard to determine whether limits have been exceeded. Alter-
natively the weighting network or combination curves (see Section D.3.3.4)
could be considered in relation to the worse case found.
In the case of impulsive shock caused by blasting, measurement may be
made at the foundations to check for structural damage. It is also necessary
to measure according to the technique given above in the areas of human
habitation.
D.3.3 Characterization of Building Vibration and Acceptable Limits
D.3.3.1 Acceptable Limits
All the following proposals are related to the recommendations for
general vibration on humans given in Section D.2. The presentation of infor-
mation is in the form of a basic rating which is given for the irost stringent
conditions. From this basic rating a multiplication factor is then applied
according to the tables for other more permissive situations.
The lowest basic rating has been defined in the area of the threshold of
human perception. It is based upon research work completed up to the end of
1975.
D-7
-------
Experience has shown in many countries that complaints of building
vibrations in residential situations are likely to arise from occupants if
the vibration levels are only slightly in excess of perception levels. In
general, the limits are related to the acceptance by the occupants and are
not determined by any other factors such as short-term health and work
efficiency. Indeed the levels are such that there is no possibility of
fatigue or other vibration induced syndromes.
D.3.3.2 Head to Foot ("Z" Axis) Vibration Limits
For Z axis the recommended vibration values proposed by the standard
is shown in Figure D-2. For frequencies between 4 Hz and 8 Hz the maximum
acceleration (rms) is 5 x 10~3 m/s2. At frequencies below 4 Hz the limit
changes at 3 dB/octave. For frequencies greater than 8 Hz the limit increases
by 6 dB/octave. For conditions other than the base curve a series of weight-
ing factors apply and these are given in Table D-2. For example, for residen-
tial property the weighting factor is two, hence at 4 to 8 Hz the maximum
recommended rms acceleration for residential property by day would be 10~2 m/s
0.3.3.3 Side to side or front to back (X or Y axis) vibration limits
For X and Y axis human vibration a different base curve applies which
is shown in Figure D-2. For frequencies from 1 - 2 Hz a maximum acceleration
level of 3.6 x 10~3 m/s2 will apply. At frequencies higher than 2 Hz the
acceptable acceleration level will increase at 6 dB/octave. This means that
for frequencies greater than 2 Hz a maximum rms velocity limit applies.
It will be noted that the standard for X or Y axis vibration is more
severe than the Z axis case at low frequencies. This is due to the sensi-
tivity of the human body towards sway at these low frequencies.
The table of weighting factors given in Table D-2 also applies to
X or Y axis vibration.
D.3.3.4 Combined standard - recommended limits for undefined axis of human
vibration exposure
D.3.3.4.1 Worst case combination curve
In many situations the same building area may be used in both the lying
and standing positions at different times of the day. If this is the case,
then a combined Standard using the worst case combination of both the Z axis
and X and Y axis conditions may be applied. This combination curve is shown
in Figure Dr2 and the same weighting factors given in Table D-2 still apply.
D.3.3.4.2 Proposed weighting network
The proposed standard also recommends a weighting network that closely
approximates the combination curve. For routine measurement and evaluation
of environmental vibration, this frequency weighting is recommended. The
weighting function proposed for combined or random vibrations is given by:
D-8
-------
CJ
v!"
E
VI
E
or
LJ
_i
LU
O
o
0.08
0.06
0.5
0.4
0.3
0.2
0.1
0.08
0.06
0.04
0.02
0.01
0.008
0.006
0.004
0.0036
0.003.
0.002
0.001
I I 1 i i
OMBINED WORST CASE X,Y OR Z AXIS .
DASHED LINE IS PROPOSED WEIGHTING FUNCTION
i i i I I i I
i i i I I i i
2 3456789 20
FREQUENCY (Hz }
40 60 80
I I * ^ *• >i«ta. I V %^ I \ t t fr i
FIGURE D—2. Building vibration criteria for occupants in buildings. All curves are for
hospital and critical working areas. See Table D—2 for proper scale factors.
D-9
-------
TABLE D-2 Weighting Factors for Acceptable Building Vibration
Place
Hospital operating
theatre & critical
working areas
Residential
(minimum
complaint
level)
Office
Workshop
Time
Day
Night
Day
Night
Day
Night
Day
Night
Continuous or
Intermittent
Vibration &
Repeated
Impulsive Shock
1
1
2
1.41
4
4
8
8
Impulsive Shock
Excitation with
not more than 3
Occurrences per day
1
1
16
1.41
128
128
128
128
Weighting Factors above basic level of Curve shown in Figure D-2
D-10
-------
G (JW) - 1+ J«-
11.2 ----- - Eqn D-1
where G (Jtu) is the transmissibility of the filter, J represents the square
root of -1, at represents the exciting frequency.
This mathematical expression defines the electronic weighting filter of the
low pass type. At low frequencies the transmissibility is zero, and at high
frequencies attenuation is at 6 dB/octave. The corner frequency is 5.6 Hz.
Accuracy - ^ 0.2 dB
Although the proposed standard recommends this function for preliminary
investigations, for practical evaluations of the overall environmental
impact of vibration on a community, the weighting function is a necessary
and useful simplification, especially with respect to residential areas,
that is not expected to introduce any significant errors.
D.4. Structural Damage from Building Vibration. (Summary of 1976 draft
standard ISO/TC 108/SC 2/WG3
D.4.1 General considerations
The proposed standard discusses the following general considerations:
Vibration in buildings (dwellings, offices, public buildings and factories)
is of increasing general importance, especially since the distances between
industrial areas with vibration exciting machines, blasts or other vibration
sources and residential areas are decreasing. Traffic on roads and railroads
also causes vibration troubles in nearby buildings.
Various methods of rating the severity of vibration in buildings and defining
limits based on laboratory or field data have been developed in the past.
However, none of these methods can be considered applicable in all situations
and consequently none have been universally accepted.
In view of the complex factors required to determine the response of a build-
ing due to vibrations and in view of the paucity of quantitative data, this
proposed Standard was prepared, first to facilitate the evaluation and com-
parison of data gained from continuing research in this field; and, second,
to give provisional guidance as to the acceptable values in order to avoid the
risk of damage. The limits proposed are a compromise of available data. They
satisfy the need for recommendations which are simple and suitable for
general application. These limits are defined explicitly in numerical
terms to avoid ambiguity and to encourage precise measurement in practice.
If the characteristics of the excitation vibration are known in relation
to the severity, position and direction of the building response—this may
be the case if the source of the vibration is within the building—and if
the parts of the buildings or the whole building influenced by the vibra-
tions can be idealized by a model, then it may be possible to estimate the
severity of the dynamic stresses by calculation.
D-11
-------
If vibrations are transmitted via the ground and the foundation into a building,
it may be possible to estimate dynamic stresses based on vibration measurements.
In addition to simple vibration there may be other factors which influence
vibration response (foundation conditions, dilatation due to temperature etc.)
and which result in damage to buildings. No general method exists at present
to take account into all such factors.
D.4.2 Categories of Damage
The proposed standard provides several phases of damage which can occur,
namely:
Category 1:
Threshold damage consists of visible cracks in non-structural members
such as partitions, facings, plasterwalls (e.g. loosening of mortar between
pantiles etc.). As a guideline visible cracks may be taken as those of a
width of 0.02 mm.
Category 2:
Minor damage consists of visible cracks in structural members such as
masonry walls, beams, columns, slabs and no serious reduction in load
carrying capacity.
Category 3:
Major damage consists of large permanent cracks in non-structural and
structural members; settlement and displacements of foundations which may
result in reduction of load carrying capacity.
The proposed standard applies chiefly to damage as described in categories
1 and 2. The limits of vibration specified in the standard were selected
to avoid the exceeding of the threshold of damage, but does include data
for estimating damage levels.
D.4.3 Measurement
D.4.3.1 Frequencies
The proposed standard recommends the following frequency ranges:
1. In the case of vibration caused by shock and quarry blasting and
the steady vibration of whole buildings: from about 1 Hz to about
100 Hz.
2. In the case of steady vibration of parts of a building, especially
floor and wall vibrations: from about 10 Hz to about 100 Hz.
D-12
-------
0.4.3.2 Measurement points
The standard recommends that for vibration caused by shock, especially
quarry blasting, should be measured on the foundation structure parallel to
its stiff-axes below ground level.
In only special cases are measurements of the floor vibration in vertical
direction and the horizontal vibration of the whole building recommended.
Such floor vibration measurements should be made in a manner similar to that
of section D.3.
In the case of steady vibration (e.g. floor vibration), the vibration
peak velocity, VMSX, at the place of highest amplitude shall be determined.
In floor vibration it is often the midspan, for whole building vibration it
is often the upper floor in horizontal direction.
D.4.3.3 Measurement quantity
Vibration can be measured by displacement, velocity or acceleration.
It is desirable to measure the quantity that is most simply and generally
related to damage as described below. While for steady vibration the
proposed standard provides curves related to velocity from 10 Hz to 80 Hz
(Figure D-3), it can be seen that for the frequency range of 10 to 80 Hz,
acceleration as weighted by the function in Chapter 3 is for all practical
purposes a measure of velocity. Plotting the weighted acceleration against
actual blast damage data, see Figure D-4, the weighted acceleration provides
a very reasonable fit to the data for frequencies below 10 Hz. For these
reasons, the use of the weighted acceleration is proposed in the main sections
of these guidelines for assessment of impact due to annoyance of building
occupants and building damage.
For shock the proposed standard recommends using the vector sum of
the maximum velocity along a set of orthogonal axis. The maximum velocity
along an axis is that measured at any time during an event. Such an approach
will be slightly more conservative than only using the maximum weighted
acceleration along the worst case axis. However, the differences between
the two approaches is not expected to be great (at the maximum they can
only differ by a factor of the square root of 3.
D.4.4 Vibration boundaries with respect to damage categories
D.4.4.1 Vibration caused by shock
In determining criteria for the onset of vibration damage to buildings,
the proposed standard indicates a number of factors which can affect the
results which are recorded.
These include:
- nature of the soil, clay, or rock, etc.
- stiffness of the building structure
- nature of the vibration, i.e. transient, intermittent, continuous,
vertical, horizontal, etc.
D-13
-------
WEIGHTED ACCELERATION BASED ON
ACCELERATION AT 80Hz OF 0.5m/s2
WEIGHTED ACCELERATION BASED ON
ACCELERATION AT 3Hi OF 0.5m/s2
I I I 1 I I I I I I L I L I I I I I I I
1.6 2.5 4 6.3 10 16 25 6O 63 100
.25 2 32 5 8 12.5 20 32 50 9O
FREQUENCY (Hz)
FIGURE D—3. Rough evaluation of vibrations of stationary floor vibrations by measurement
• of vibration displacements amplitude and frequency
Note: Amplitude is defined here as the maximum absolute value of the displacement
of the floor undergoing harmonic motion.
D-14
-------
0.2
0.08
0.06
I 0.04
1 0.02
a.
tn
0.01
0.008
0.006
0.004
0.002
0.001
1.0,—
0.6
0.4
I I
I T
0 Weighted Accelerations of 1 meter/sec2
oo
•Weighted Accelerations of 0.5 meter sec/
Major damage v = 7.6 in/sec.
Minor damage v = 5.4 in/sec.
Q
• ^XX • VI •V^'
••<. •••
Safe blasting criterion
v = 2.0 in/sec
— .O Bureau of Mines ^
c Langefors Major damage data\
A Edwards and Northwood \
* •
»
— • Bureau of Mines ) *\(
• Langefors Minor damage data
A Edwards and Northwood I
I
I I
I
J I
I
I
I
1
4 6
10
20
40 60 100 200 400 600
FREQUENCY, cps
FIGURE D—4. Displacement versus frequency, combined data with recommended safe blasting criterion
D-15
-------
With these uncertainties in mind, the proposed standard provides recommenda-
tions as to the maximum velocity to prevent damage for each of the three
categories. These velocities are listed in Table D-3.
TABLE D-3
Limiting values of the vector sum of the maximum velocities (in three
orthogonal axis) caused by quarry-blasting-vibration in dwellings and
offices in good physical conditions
Category of Damage range VR, onset of
(See Section D.4.2) damage, in mm/s
1 3 ... 5
2 5 ... 30
3 100
These values are based on measured foundation vibration in the
frequency range from about 3 Hz to about 100 Hz.
The standard cautions that:
(1) In the range between 30 mm/s and 100 mm/s the available data is
not sufficient to define the nature of the damage without regard
to the condition, type of structure and foundations.
(2) The limits apply only where differential settlement of the
structure has not been excessive.
(3) Special consideration shall be given where buildings are situated
on a slope or on soils which may be compacted or liquified by
vibration.
(4) When large dynamic displacements are found to exist in the whole
building or part of it then in addition to the reconmended measuring
points at the foundation additional measuring points located in the
structure shall be used for the evaluation of potential building
damage.
The standard recommends that the limits specified in Table D-3 be used for
the evaluation of vibration effects caused by pile drivers and foregoing
hammers when the time interval between two succeeding blows is so large that
the vibration of the building due to one blow dissipates before the effects
of the succeeding blow are observed. Dissipation is regarded as effective
when peak particle velocities have decayed 1/5 from their maximum.
D-16
-------
The standard proposed that the values specified in Table D-3 may also be
used to evaluate the effects of vibration in buildings caused by traffic;
however, when shakers and vibration pile drivers are the source of building
vibration, the values given in Table D-3 should not be applied.
Finally, the standard recommends that for the evaluation of transient
response of floors and walls, the vibration limits given for steady state
vibrations may be used in a modified form. Vfaen there is no danger of
fatigue the limits and values given in Figure D-3 may be increased by a
factor of 2.
D.4.4.2 Steady vibration of buildings
\
For steady building vibration, Figure D-3 summarizes the peak velocity
boundaries between the different categories of damage.
D.4.5 Comparison of the recommendation of the proposed standard to the
recommendations of these guidelines
The proposed standard recommends that 6 mm/s (5 to 30 mm for shock) be
considered as the upper limit of the threshold of damage. These velocities
are considerably lower than the 2 in/sec (50.8 mm/sec) that has commonly been
used in this country. Based on studies such as those shown in Figure D-4,
reducing the threshold from 50 mm/sec to 5 mm/sec does not appear warranted,
however, reduction of the threshold by a factor of 2 does seem reasonable.
All of the data points of Figure D-4 will be covered by use of a velocity of
1 in/sec and it is this velocity that is recommended in the main text of the
guidelines. Use of a weighted acceleration of 0.5 m/sec2 is consistent with
this velocity and is recommended.
D-17
-------
Appendix E
Example Application of the Guideline Procedures for
General Audible Noise
E.1 Proposed highway expansion
This example (presented briefly in section 2.6) concerns a section of
highway which runs for several kilometers through a suburban area (Figure
E-1). The present two lane roadway is operating at close to capacity, and
the proposal is to expand it to six lanes. Although many factors must be
considered before undertaking such an expansion, only the noise impacts of
the project will be discussed as an illustration of the use of these Guide-
lines. This example is divided into five sections as follows:
(1) Statement of the problem
(2) Using the screening diagram
(3) Determining the necessary number of figures and tables
(4) Completing the figures and tables
(5) Conclusions of the noise analysis
E.1.1 Statement of the problem
From Figure 2 in section 1.3.1, it has been assumed for the purposes of
this example that the only concern is general audible noise that may cause an
adverse impact. That is, special noises, vibration, and changes in population
location are not anticipated to be problems. Tables 6 to 9 in section 2.6
document the project impact over the total area (Figure E-1). However, to
illustrate in some detail the use of these Guidelines, this example focuses
on a small residential section only, as shown in Figure E-2. Each of the
residential buildings consists of two semi-detached townhouse units, with an
average population of five persons in each unit, or 50 persons in each row of
housing. Additionally, there are four special situations to be considered:
(1) a school with a student-teacher population of 2,500 in attendance 50 weeks
a year from 8:00 a.m. to 4:00 p.m.; (2) a playground where 400 children play
six hours each day; (3) a park where 160 people relax for one hour each day;
and (4) a church where 185 people meet for two hours each day, and 115 people
meet for one hour each evening. For a larger project area (such as in Figure
E-1), this amount of detail normally would not be obtained. Noise contours
would still be plotted, but populations could be estimated from average popu-
lation densities, census counts, or other such sources as discussed in section
2.1.3. The example is intended to provide an easy-to-follow description of
the Guideline procedures.
E-1
-------
Microarea
covered in
Highway
Example
n
i
10
RESIDENTIAL
FIELD
HIGHWAY
COMMERCIAL
INDUSTRIAL
RESIDENTIAL
School
Nursing
Home
| | School
COMMERCIAL
Library
FIGURE E— 1. Highway Expansion Example: Overview
7A
Area for the
specific example
-------
Ldn(y)
I
70dB
65clB
60dB
RESIDENTIAL
2 Lanes
SPECIAL
SITUATIONS
55dB
w
i
t.
0.5km l
dB
cm
•Q-
en
CD
CD
CD
cn
en
1=3
1 1
CD
CU
cn
[=3
CD
n
en
N
150
375
Total Population: 550
4.74km
SCHOOL
PLAYGROUND
CHURCH
Figure E-2. Sample Data Presentation for the Highway Example: Future Levels Without the Proposed Project
-------
E.1.2 Using the screening diagram
Is an environmental noise analysis necessary, and if so, what procedures
should be followed? Begin by examining the Screening Diagram (Figure E-3, and
discussed in section 2.1). This diagram is helpful for determining not only
whether a noise impact analysis is necessary, but also what type of analysis
should be conducted.*
E. 1.2.1 How to use the screening diagram. The values for the "existing I^jn(y)"
and the "expected L$n of [the] project alone" should be obtained at the loca-
tion of the noise sensitive land use nearest the project, or the point where
the impact of the project is likely to be the greatest. In this example, that
point would be the row of duplexes closest to the highway. The existing I^n(y)
at the closest noise sensitive point may be obtained either by direct measure-
ment or by use of a suitable traffic noise prediction model** as discussed in
section 2.1.2. It is assumed that future I^i was obtained through the use of
a prediction model. Once the values of "existing I^n(y)" and "future L^" are
obtained, they are plotted as coordinates on the screening diagram (Figure E-3).
Their point of intersection (for this example, 60 dB and 65 dB)*** determines
both the necessity for and type of noise analysis that should be conducted. In
this case, the coordinates fall into the cell that calls for a full noise envi-
ronment documentation. Therefore, a full noise analysis should be conducted.
E. 1.2.2 Other factors to consider before beginning the noise analysis. As
discussed in Chapter 2, there are a few other basic issues which should be
considered before beginning the noise analysis:
o How many projects or alternatives are we considering?
We are considering only one, an expansion from two to six lanes.
o Will the population of the residential area change in the future?
No, the population will remain the same (by assumption in this
example).
*For this example, it is not absolutely necessary to use the screening
diagram since it is more or less obvious that some increase in noise will
result, and at levels high enough for a full noise environment documenta-
tion. However, if in doubt, the screening diagram is a useful aid.
**There are in use several models for calculating day-night sound levels
based on the type of noise source and operational considerations. These
models are available from many sources, some of which are listed at the
end of this appendix.
***At the row of houses closest to the highway, the existing I^an(y) of 60 dB
is from Figure E-2, and the predicted I^n(y) of 65 dB from the project
alone (the 6-lane highway) is from Figure E-4.
E-4
-------
JL •'
w
m
LU
O
cc
Q.
C
TJ
Q
LU
OL
X
LU
70
60
50
FULL NOISE ENVIRONMENT
DOCUMENTATION FOR ALL
PROJECTS
X160. 65)
40
POSSIBLE NOISE
OEGHADATION
' ANALYSIS
O
40
DOCUMENTATION
DEPENDS ON
FUTURE OF
EXISTING
NOISE SOURCES
ALL PROJECTS
SCREENED OUT
50 60 70 80 90
EXISTING Ldn{y)
Figure E-3. Screening Diagram: Sample Application
-------
o Is this a temporary project, as defined in section 2.5?
No, this is a long-term project.
o Will the noise of the project change with time (after the
completion of the project)?
No, the noise of the project will remain the same. The
immediate demand for the added lanes will be sufficient
to fill them to capacity (by assumption in this example).
o Are existing noise levels in the area low enough that
"environmental degradation" is the only concern (as defined
in section 2.4)?
No, most of the area will be exposed to project Ldn(y) greater
than 55 dB. This is confirmed by the screening diagram.
E.1.3 Determining the necessary number of figures and tables. From the
discussion in section 2.2.2.a, it is clear that only three sets of contours
and corresponding tables are required.*
o Future levels in the area without the project, i.e., future
levels from the existing highway.
o Future levels due to the proposed project alone, i.e., the six-
lane highway alone.**
o Future levels resulting when the levels from the six-lane highway
are combined with the levels generated by other noise sources,
i.e., in this case by residential area activities.
*As noted in the text, the population density within the residential area
is not expected to change, nor will the noise from the highway change in
years subsequent to the proposed expansion. Because these conditions with
respect to time are expected to remain constant, additional sets of tables
and figures are not necessary. However, if these conditions were to change
over time, separate sets of tables should be prepared for (a) the first year
of the commencement of the project, (b) twenty years after the expansion
(or the latest year for which noise predictions can be reliably made), and
(c) the worst case year (if different from the preceding two).
**Note that when the proposed action is an expansion of an existing noise
source, the proposed project alone is interpreted to be the expanded
project (that is, the six-lane highway), not the amount of expansion (in
this case, the additional four lanes).
E-6
-------
E.1.4 Completing the figures and tables
The purpose of the analysis is to compare the future noise environment
with and without the proposed project. This comparison can be divided into
five steps: (1) drawing the noise contours, (2) determining the base area
and the base population, (3) transferring the data to the tables, (4) calcu-
lating the single-number comparison indices; and (5) noting special popula-
tions.
E. 1.4.1 Drawing the noise contours. As discussed in section E.1.3, three
sets of contours and tables are required. For purposes of this example, it is
assumed that contours describing the future noise environment in the area
without the highway expansion have been obtained by measurement, since in this
example future levels are the same as existing levels. It is also assumed
that the noise levels from the future six-lane highway alone have been
obtained from a suitable highway noise prediction model. These results are
illustrated in Figures E-2 and E-4, respectively.
To draw contours reflecting the combined future noise environment from
the project levels and levels generated from residential activity requires
additional information. That is, knowledge of residential area levels in
the absence of any highway noise. This information can be obtained in two
ways. An estimate can be made on the basis of population density using
equation 1 in section 2.1.1. Or, measurements can be taken at a large dis-
tance from the road (for example, where noise from the roadway is no longer
clearly noticeable), as long as the nature of the area is not expected to
change in the future. The background residential noise levels derived above
(assumed to be about 50 dB in this example) are then combined on a point-by-
point basis with the project alone levels presented in Figure E-4 to derive
contours depicting the total future noise environment (project plus back-
ground) as shown in Figure E-5.*
E.1.4.2 Determining the base area and base population. As noted in section
2.1.3, the base population is defined as the number of people living in areas
with outdoor noise levels produced by the project alone above a specified
I^n value. This is called the base 1^. The base Lfa is determined by
reference to the existing yearly 1^ contours in the residential area
(Figure E-2). The lowest 1^ in the residential area is about 50 dB near
*For a step-by-step explanation of the combination process, see the discussion
at the end of this appendix. Note also that the combination process may
result in contour lines in other than the desirable five decibel intervals.
Interpolation may be necessary to plot the information in the five decibel
bands.
E-7
-------
Ldn(y)
I
70clB
RESIDENTIAL
6 Lanes
SPECIAL
SITUATIONS
65dB
GOclB
CD
25
75
SCHOOL
CD
CD
z
PLAYGROUND
50(18
C=]
cm
CD
EZi
*fri*fc
150
PARK
0.5km I—L
CD
CD
ED
EH
CD
CD
CD
EH
CD
CD
EH
* EH
N
250
4.74km
Fiqure E-4. Sample Data Presentfition for the Hinhwav Examnle: Fnturn Nni«sR I PU«I« fmm tho Prn
CHURCH
-------
tl
I
NO
Ldn(y)
7
dB
65clB
60dB
55dB
0.5km
5(
MB
CZJ
en
^
1 1
a
cn
CD
RESIDENTIAL
6 Lanes
(ZZI
CD
•B
^3-
i 1
CD
CD
CZD
en
en
czi ^ a
N
50
75
fttttfc
150
275
4.74km
SPECIAL
SITUATIOS
SCH
OOL
i
PLAYGROUND
CHURCH
Figure E-5. Sample Data Presentation for the Highway Example: Future Levels from All Noise Sources Combined
-------
the bade row of houses. Therefore, from section 2.1.3, the optimum base
to use, if possible, in order to define the base population is 40 dB (that is,
10 dB below the existing I^n(y)). Next, we examine Figure E-4 which shows
the noise contours from the project alone. Applying the base I^n =
40 dB, we can derive the base area. In this example, none of the residents
are living in areas where the outdoor yearly day-night sound levels are below
45 dB (i.e, no people live within the 40-45 dB interval). Thus, the next best
thing is to effectively define the base area as the area exposed above
Itin(y) = 45 dB. In this case, there is only one proposed project, and the
base population is 550 people, in an area of 2.37 sq. km. (Fig. E-4).
E.1.4.3 Transferring the data to the tables. Tabulations of population and
area exposure information are provided in Tables E-1 , E-2, and E-3 for
Figures E-2, E-4, and E-5, respectively. The values in the tables are derived
by summing the number of people and land area within each five decibel band. If
a contour line bisected a row of duplexes, the residents were divided between
the noise bands. For example, from Figure E-5:
Noise band (I^n) Number of People
65 - 70 dB 50
60 - 65 dB 75
55 - 60 dB 150
50 - 55 dB 275
E.1.4.4 Calculating the single-number indices. For this comparison, three
measures of impact should be considered: (1) the sound level weighted
population (IMP); (2) the noise impact index (Nil); and (3) the relative
change in impact (FCI).* The indices LWP and Nil should be computed
for each of the tables (Tables E-1 thru E-3). For purposes of illustration,
detailed calculations will only be shown for Table E-1.
Calculation of the level weighted population was based on the values of
the weighting function of equation 2b, shown in Table 3 of the main text.
Thus, using equation 6:
LWP = [P(65-70) x W(67.5)] + [P(60-65) x W(62.5)] +
[P(55-60) x W(57.5)] + [P(50-55) x W(52.5)] +
[P( 45-50) x W(47.5)]
= [(0) x (0.194)] + [(25) x (0.116)] +
[(150) x (0.064)] + [(375) x (0.032)] +
[(0) x (0.015)]
= 24. 5 ~ 24 people
*In this example, since there are no outdoor exposures greater than L^n = 75 dB, it
is extremely unlikely that there will be any at-ear I^q(24) exposure greater than
70 dB. Therefore, we need not consider the single-number indices for severe health
effects (hearing-loss weighted population and average potential hearing loss) as
discussed in section 2.3.
E-10
-------
Table E-l
Sample Data Presentation for the Highway Example: Future Levels Without Proposed Project
Yearly Ldn (dB)
>70
65-70
60-65
55-60
50-55
45-50
Residential
Population
0
0
25
150
375
0
Residential
Land Area
(Sq km)
0
0.087
0.107
0.646
1.530
0
Industrial/
Commercial
Land Area
(Sq km)
0
0
0
0
0
0
Total Land
Area (Sq km)
0
0.087
0.107
0.646
1.530
0
Special
Situations
(See Table E-4)
-
-
-
1,2
3,4
—
550
2.370
2.370
Level Weighted Population (LWP) =24
Noise Impact Index (Nil) = 0.044
Hearing-Loss Weighted Population (HWP) = 0
Average Potential Hearing Loss (PHL) = 0
Corresponds to Fig. E-2
Includes: o Self-generated
neighborhood noise.
o Levels of noise from
the existing two-lane
highway.
-------
Table E-2
Sample Data Presentation for the Highway Example: Future Noise Levels of Project Alone
T
10
Yearly L70
65-70
60-65
55-60
50-55
45-50
Residential
Population
0
25
75
50
150
250
Residential
Land Area
(Sq km)
0
0.1232
0.2702
0.2465
0.3982
1.3320
Industrial/
Commercial
Land Area Total Land
(Sq km) Area (Sq km)
0
0
0
0
0
0
0
0.1232
0.2702
0.2465
0.3982
1 .3320
Special
Situations
(See Table E-4)
-
-
1
2
3
4
550
2.370
2.370
Level Weighted Population (LWP) - 25
Noise Impact Index (Nil) = 0.045
Hearing-loss Weighted Population (HWP) = 0
Average Potential Hearing Loss (PHL) = 0
Corresponds to Pig. E-4
Includes: o Levels of noise
from the proposed
six-lane highway.
-------
Table E-3
Sample Data Presentation for the Highway Example: Future Levels from All Noise Sources Combined
V
w
Yearly l#n (dB)
>70
65-70
60-65
55-60
50-55
45-50
Residential
Population
0
50
75
150
275
0
Residential
Land Area
(Sq km)
0
0.1232
0.2465
0.6716
1.3830
0
Industrial/
Commercial
Land Area
(Sq km)
0
0
0
0
0
0
Special
Total Land Situations
Area (Sq km) (See Table E-4)
0
0.1232 1
0.2465 1,2
0.6716 3
1.3830 4
0
550"
2.370
2.370
Level Weighted Population (LWP) = 37
Noise Impact Index (Nil) = 0.067
Hearing-Loss Weighted Population (HWP) = 0
Average Potential Hearing Loss (PHL) = 0
Corresponds to Fig. E-5
Includes: o Self-generated
neighborhood noise.
o Levels of noise
from the six-lane
highway.
-------
The Noise Impact Index, according to equation 7 in section 2.2.2.C, is
simply LWP divided by the total (base) population. Thus:
m
PTotal= 550 =0.044
From equation 8, the Relative Change in Impact between the case without
the proposed expansion (Table E-1 ) and the case with the expansion (Table E-3)
is computed as:
FCI
E.1.4.5 Noting special populations. Special populations do not affect the
calculation of the single-number indices. However, they are noted on the
figures and tables to give the reader additional information about the
affected area. As discussed in Section 2. 2. 2. a., the time weighted average
number of people present at these special locations during the year may be
computed as:
(number of people) x (time the people are present during the year)
(number of hours in a year)
For the school in our example where 2,500 students and faculty use the school eight
hours a day, five days a week, fifty weeks a year:
(2,500 student and teachers) x (8 hours) x (5 days) x (50 weeks)
8,760 hours in a year
For the playground where 400 children play six hours each day:
(400 children) x (6 hours) x (7 days) x (52 weeks) inn ____,
8,760 hours ^ a year = 1UU P60?16'
For the park where 160 people relax for one hour per person each day:
(160 people) x (1 hour) x (7 days) x (52 weeks) _ 7
8,760 hours in a year ~
For the church where 185 people meet for two hours each day, and 115 people meet for
one hour each night, consider the day and night population separately:
(185 people) x (2 hours) x (7 days) x (52 weeks) ,_ ___._,
8r760 hours in a year D people.
(115 people) x (1 hour) x (7 days) x (52 weeks)
8,760 =
The results for special populations are depicted in Table E-4
E-14
-------
s
m
Table E-4
Sample Data Presentation: Special situations
Average population Range of Ldn/v\
Day Night Current Future^M-th Project
1. Elementary school 571
2. School playground 100
3. Park 7
4. Church 15
0 55-60 dB
0 55-60 dB
0 50-55 dB
5 50-55 dB
60-70 dB
60-65 dB
55-60 dB
50-55 dB
Comments
Good acoustic
insulation
Evening meetings
-------
E.2 Proposed airport runway addition
This example concents the addition of a runway at an angle to an existing
runway. The additon will be completed in 1985. After 1985, airport operation
and noise will increase until maximum levels are attained in 2001. Bordering
the airport are a high and a low population density neighborhood. Both neighbor-
hoods will encroach on the airport between 1985 and 2001, as is shown in Fig. E-6.
The rest of the land near the airport is zoned commercial/industrial. Although
many factors must be considered before undertaking an airport expansion, only
the noise impacts will be discussed in order to illustrate the use of the Guide-
lines.
This example is divided into five sections:
(1) Statement of the problem
(2) Using the screening diagram
(3) Determining the necessary number of figures and tables
(4) Completing the figures and tables
(5) Conclusions of the noise analysis
E.2.1 Statement of the problem
'From Figure 2 in section 1.3.1, it has been assumed for the purposes of
this example that the only concern is general audible noise that may cause an
adverse or potentially severe impact. That is, special noises and vibration
are not anticipated to be problems. To illustrate in some detail the use of
these Guidelines, this example focuses on the potential noise impact upon each
of the residential areas in proximity to the airport as shown in Figure E-6.
It is assumed that the number of people living within each of these neighbor-
hoods is computed or estimated from census counts, average population
densities or other methods discussed in section 2.1.3.
E.2.2 Using the screening diagram
Is an environmental noise analysis necessary, and if so, what analytical
procedures should be followed? Begin by examining the Screening Diagram
(Figure E-7 and explained in section 2.1). This diagram is helpful for
determining not only whether a noise impact analysis is necessary, but also
what type of analysis should be conducted.
E.2.2.1 How to use the screening diagram. The values for the "existing
and the "expected L, of [the] Project Alone" should be
at the location ofnihe noise sensitive land used nearest the
*For this example,Tt is not absolutely necessary to use the Screening
Diagram since it is more or less obvious that some increase in noise will
result, and at levels high enough for a full noise environment documenta-
tion. However, if in doubt, the screening diagram is a useful aid.
E-16
-------
M
I
HIGHWAY
POPULATION: 31,689
HIGH DENSITY
NEIGHBORHOOD
LOW DENSITY
NEIGHBORHOOD
ORIGINAL RUNWAY
PRESENT
POPULATION: 31,689
PROJECT
RUNWAY
CQVPLETIONDATE: 1985
POPULATION: 68,291
GRaVTHOF
NEIGHBORHOODS
MAXIMUM USAGE AND NOISE LEVELS: 2001
Figure-E-6. Sample Data Presentation for the Airport Example: Schematic of the Existing Situation, 1985. and 2001.
(Note the Addition of the Project Runway and tho Encroachment of the Neighborhoods..)
-------
UI
1
80
70
60
u
uj
s
cc
a.
u.
O
JBO
a
UJ
U
UJ
X 40
u
X (58, 811
FULL NOISE ENVIRONMENT
DOCUMENTATION FOR ALL
PROJECTS
POSSIBLE NOISE
DEGRADATION
ANALYSIS
I
DOCUMENTATION
DEPENDS ON
FUTURE OF
EXISTING
NOISE SOURCES
ALL PROJECTS
SCREENED OUT
40
50
GO 70
EXISTING Ldn(y,
80
90
Figure E-7. Screening Diagram
E-18
-------
project, or the point where the impact of the project is likely to be the
greatest. Relevant noise level data is contained within Figures E-8 and E-9 ,
respectively. Because, as previously noted, airport noise will be greatest in
the year 2001, the point of greatest impact is taken from Figure E-9. In this
example, the point where the impact of the project is likely to be the
greatest is in the high population density neighborhood, and is designated on
Figure E-9 by the mark "X". The Itfn(y) at that location from the airport
alone will be about 81 dB.* Note that, in this example, the point of greatest
impact is not the closest noise sensitive location. The noise level at the
corresponding point on Figure E-8 is approximately 58 dB.**
Once the values of the "Existing ^3n(y)n and the "Expected 1^ of
the Project Alone" are obtained, they are plotted as coordinates on the
screening diagram (Fig. E-7). Their point of intersection (for this example,
58 dB and 81 dB) determines both the necessity for, and type of, noise analysis
that should be conducted. In this case, the coordinates fall into the cell
that calls for a full noise environment documentation. Therefore, a full
noise analysis should be conducted.
E.2.2.2 Other factors to consider before beginning the noise analysis. As
discussed in Chapter 2, there are a few other basic issues which should
be considered before beginning the noise analysis:
o How many projects are we considering?
We are considering only one option, the addition of a runway
with an increase in total airport operations.
o Will the population of the residential areas remain the same?
No, between the years 1985 and 2001 both residential areas
will increase in population and size. In the existing situation,
the self -generated (ambient) noise levels in the low density
neighborhood (780 people per square kilometer) is estimated
to be 55 dB. By assumption, the number of people in the low
density area will remain the same until 1985. By the year
2001 , even though the population will increase, the ambient
*The "Future Noise Levels of Project Alone; 2001" (Fig. E-9) was determined
by using a computer model, as discussed in section 2.1.2. There are
several models for calculating day-night sound levels based on the type
of noise source and operational considerations. These models are
available from many sources, some of which are listed at the end of this
appendix. These models were also used to predict all of the other future
ttin noise values in this example (except where otherwise noted).
**For purposes of this example, the "Existing I^nfy)" (Fi9* E~8)
was assumed to have been determined by direct measurement as
discussed in section 2.1.2.
E-19
-------
INCLUDES:
• NOISE FROM THE HIGHWAY
• SELF-GENERATED NOISE FROM
BOTH NEIGHBORHOODS
• NOISE FROM THE ORIGINAL RUNWAY
CORRESPONDS TO TABLE E-5
1km
55
60
65
Figure E-8. Sample Data Presentation for the Airport Example: Existing
-------
INCLUDES: 9 NOISE FROM BOTH ORIGINAL
AND PROJECT RUNWAY
CORRESPONDS TO TABLE E-6
1km
Fi9ure F,9. Sample Data Presentation for the Airport Example: Future Noise Levels of the Project Alone; 2001
E-21
-------
noise levels* will remain the sane because the population
density (people per square kilometer) will remain constant.
In the high density neighborhood, the existing and future ambient
is estimated to be 65 dB (7,700 people per square kilometer).
o Is this a temporary project, as defined in section 2.5?
No, this is a long-term project.
o Will the noise of the project change with time (after the
completion of the project in 1985)?
Yes, as previously stated, the noise levels will be increasing
between 1985 and 2001, reaching a maximum in the year 2001.
The increase in noise is caused by an increase in airport
operations.
o Are existing noise levels in the area low enough that
"environmental degradation" is the only concern (as defined
in section 2.4)?
No, surrounding the airport is a' commercial/industrial area with
a background ^jn(y) of 55 dB in many locations. Such a high
background noise level when plotted on the screening diagram is
outside of the cell concerning low noise areas. (This cell is
entitled "Possible Noise Degradation Analysis.") '
o Are there any special situations (as explained in section 2.2.2),
such as religious facilities, outdoor auditoriums, schools,
precision laboratories or hospitals?
By assumption in this example, there are no special situations.
E.2.3 Determining the necessary number of figures and tables
From the discussion in section 2.2.2.a, it is clear that a number of
sets of contours and corresponding tables are required.
o Existing noise levels
o Future levels without proposed project; 1985
o Future levels with the proposed project alone; 1985
o Future levels from all noise sources combined; 1985
o Future levels without proposed project; 2001
*Yearly day-night sound levels as estimated from population densities
are discussed in Section 2.1.2.
E-22
-------
o Future levels with the proposed project alone; 2001
o Future levels from all noise sources combined; 2001
E.2.3.1. Completing the figures and tables. The purpose of the analysis is to
compare the future noise environment with and without the proposed project.
This comparison can be divided into four steps: (1) drawing the noise
contours; (2) determining the base area and the base population; (3) transferring
the data to the tables; and (4) calculating the single number comparison indices.
E.2.3.1 Drawing the noise contours. As discussed in section E.2.3, a number
of sets of contours and tables are required. Ebr purposes of this example/ it
is assumed that the future noise levels of the project have been obtained from
a suitable airport noise prediction model taking into account the new generation
of quieter aircraft. To draw contours reflecting the combined future noise
environment from the project levels and levels generated from residential
activity requires additional information. A knowledge of residential area
levels in absence of any other noise (such as highway or airport noise) is
necessary. This information can be obtained in two ways. An estimation can be
made on the basis of population density using equation 1 from section 2.1.2, or
measurements can be taken at a large distance from the other noise sources. In
this example, the former method was used, as discussed in section E.2.2.2.
The required illustrations are discussed below:
o Figure E-8: Existing levels and future noise levels without
the proposed project; 1985
The noise levels in 1985 without the project are
the same as the noise levels in the existing situa-
tion. This lack of change is by assumption in this
example.
o Figure E-10: Future noise levels of the project alone; 1985
o Figure E-11: Future levels from all noise sources combined; 1985
The data contained in Figures E-8 and E-10 are
combined to create contours of total noise
exposure.**
*Note that when the proposed action is an expansion of an existing facility,
the proposed project alone is interpreted to be the expanded project (that
is, both the old and new runways), not the expansion alone (in this case,
the new runway alone).
**Iogarithmic combinations are discussed at the end of this appendix.
E-23
-------
o Figure E-12: Future levels without the proposed project; 2001
This represents the noise intrinsic to the neighbor-
hood which is expanded, the highway, and levels of
noise from the increased usage of the existing
single runway.
o Figure E-9: Future noise levels of the project alone; 2001
o Figure E-13: Future levels from all noise sources combined; 2001
The data contained in Figures E-9 and E-12 are
combined to create contours of total noise
exposure/
**Logarithmic combinations are discussed at the end of this appendix,
E-24
-------
INCLUDES:
• NOISE LEVELS FROM BOTH ORIGINAL AND PROJECT RUNWAYS
CORRESPONDS TO TABLE E-7
Figure E-10. Sample Data Presentation for Airport Example: Future Noise Levels of the Project Alone; 1985
E-25
-------
55 INCLUDES:
• NOISE FROM THE HIGHWAY
• SELF-GENERATED NOISE FROM
BOTH NEIGHBORHOODS
• NOISE FROM BOTH THE ORIGINAL
AND THE PROJECT RUNWAYS
CORRESPONDSTb TABLE E-8
\
\
Figure E-11. Sample Data Presentation for the Airport Example: Future Levels from All Noise Sources Combined: 1935
E-26
-------
INCLUDES:
• NOISE FROM THE HIGHWAY
• SELF-GENERATED NOISE FROM BOTH
EXPANDED NEIGHBORHOODS
• NOISE FKJM THE COMMERCIAL
INDUSTRIAL ZONE
• NOISE FROM THE ORIGINAL RUNWAY
CORRESPONDS TO TABLE E-9
1km
75
\
Figure E-12. Sample Data Presentation for the Airport Example: Future Noise Levels without the Proposed Project; 2001
E-27
-------
INCLUDES:
• NOISE FROM TOE HIGHWAY
• SELF-GENERATED NOISE FROM BOTH
EXPANDED NEIGHBORHOODS
• NOISE FROM BOTH THE ORIGINAL
AND THE PROJECT RUNWAYS
CORRESPONDS TO TABLE E-10
\
Figure E-13. Sample Data Presentation for the Airport Example: Future Levels of Noise
from All Noise Sources Combined; 2001
1km
E 28
-------
E.2.4.2 Determining the base area and base population. As noted in
section 2.1.3, the base population is defined as the number of people living
in areas with outdoor levels of noise produced by the project alone above a
specified L^ value. This is called the base 1^. The base L^n is
determined by reference to the existing yearly LJn contours in the residen-
tial area (Figure E-8). From Figure E-8, the lowest L^ in the residential
area is about 60 dB. Therefore, from section 2.1.3, the optimum base I^n to
use, if possible, in order to define the base population is 50 dB (that is,
10 dB below the existing I^n(y)). Next, we examine Figure E-9 which shows
the noise contours from the proposed project alone in 2001 (the year with the
greatest impact). Applying the base L^ = 50 dB, the base area should be
the area within the 50 dB contour. However, since the level of ambient noise
is assumed to be at least 55 dB, such a base area would extend far beyond the
boundaries of Figure E-9. In fact, the land within the 55 dB contour also
extends well beyond the boundaries of the figure. Since there are by assump-
tion no other residential areas in the vicinity, it is not necessary to choose
such a large base area. Instead, it would be more logical to consider as the
base population the people residing within the residential areas of interest.
E.2.4.3 Transferring the data to the tables. Tabulations of population and
the area exposed are provided in Tables E-5, -6, -7, -8, -9, and -10, correspond-
ing respectively to Figures E-8, -9, -10, -11, -12, and -13. The values in the
tables are derived by summing the number of people and the area of land within
each five decibel band. For example, in Figure E-13:
Residential
Noise Level (L^n) Population
80+ 1,233
75-80 30,799
70-75 30,346
65-70 3,942
60-65 1,971
55-60 0
50-55 0
E.2.4.4 Calculating the single number indices. For this comparison, five
single number indices should be considered: (1) the sound level-weighted
population (LWP); (2) the noise impact index (Nil); (3) the relative change
of impact (RCI); (4) hearing loss-weighted population (HWP); and (5) the
average potential hearing loss (PHL). The first three indices (explained
in section 2.2.2.b) concern the range of general adverse noise effects; the
latter two (explained in section 2.3.2) concerns noise levels which possibly
may cause severe health effects. The indices EWP, Nil, HWP and PHL should
be computed for each of the tables.
E-29
-------
Table E-5
Sample Data Presentation for Airport Example:
Future Levels Without Proposed Project; 1985
Yearly I#n
(dB)
80-85
75-80
70-75
65-70
n 60-65
u>
° 55-60
<55
Residential
Population
0
0
0
27,061*
4,628
0
0
31 ,689
Industrial/
Commercial Total Land
Employees Area (sq. km.)
0 0
0 0
1,550 8
7,264 42
10,822 19.2
6,814 135.8
41,678 95.0
68,128 300.0
Industrial/
Residential Land Commercial Land
Area (sq. km.) Area (sq. km.)
0 0
0 0
0 8.0
3.14 38.86
4.48 14.72
0 135.8
0 95.0
7.62 292.38
Level Weighted Population (LWP) = 5,787
Noise Impact Index (Nil) = 0.183
Hearing Loss-Weighted Population (HWP) = 0
Average Potential Hearing Loss (PHL) = 0
*Large number of people in the 65-70 dB band
is attributable to high ambient noise levels
(non-aircraft related) in the high density
residential area (See section E.2.2.2).
Corresponds to Figure E-8
Includes: o Highway noise
o Self-generated noise from both neighborhoods
o Noise from the original runway
Note: Single number indices are not computed on the
basis of industrial/commercial employees.
-------
Table E-6
Sample Data Presentation for Airport Example:
Future Noise Levels of Project Alone; 2001
Yearly L^n
(dB)
80-85
75-80
70-75
65-70
60-65
55-60
<55
Residential
Population
1,233
30,799
30,346
2,673
3,240
0
0
68,291
Industrial/
Commercial
Employees
303
19,473
22,921
25,431
0
0
0
68,128
Total Land
Area (sq. km.)
12.0
20.7
49.15
206.03
12.12
0
0
300.0
Residential Land
Area (sq. km.)
0.16
4
4.85
1.32
12.12
0
0
22.45
Industrial/
Commercial Land
Area (sq. km.)
11.84
16.7
44.3
204.71
22.97
0
0
277.55
Level Weighted Population (LWP) = 24,498
Noise Impact Index (Nil)- 0.359
Hearing Loss-Weighted Population (HWP) - 408
Average Potential Hearing Loss (PHL) =0.09
Corresponds to Figure E-9
Includes: o Noise from both the original and the
additional runways.
-------
Table E-7
Sample Data Presentation for Airport Example:
Future Noise Levels of Project Alone; 1985
Yearly l$n
(dB)
80-85
75-80
70-75
65-70
ra 60-65
10
55-60
<55
Residential
Population
0
0
0
5,358
26,331
0
0
31,689
Industrial/
Commercial
Employees
0
0
5,934
18,708
11,349
22,132
10,005
68,128
Total Land
Area (sq. km.)
0
0
25.6
82.76
74.0
92.64
25.0
300.0
Residential Land
Area (sq. km.)
0
0
0
2.12
5.50
0
0
7.62
Industrial/
Commercial Land
Area (sq. km.)
0
0
25.6
80.64
68.5
92.64
25.00
292.38
Level Weighted Population (LWP) = 4,094
Noise Impact Index (NII)= 0.129
Hearing Loss-Weighted Population (HWP) = 0
Average Potential Hearing Loss (PHL) = 0
Corresponds to Figure E-10
Includes: o Noise from both the original and the
additional runways.
-------
U)
UJ
Table E-8
Sample Data Presentation for Airport Example:
Future Levels From All Noise Sources Combined; 1985
Yearly L^n
(dB)
80-85
75-80
70-75
65-70
60-65
55-60
<55
Residential
Population
0
0
0
29,331
2,358
0
0
31,689
Industrial/
Commercial
Employees
0
0
5,934
18,708
17,354
26,132
0
68,128
Total Land
Area (sq. km.)
0
0
25.6
85.76
76.0
112.64
0
300.0
Residential Land
Area (sq. km.)
0
0
0
5.12
2.50
0
0
7.62
Industrial/
Commercial Land
Area (sq. km.)
0
0
25.6
80.64
73.5
112.64
0
292.38
Level Weighted Population (LWP) = 5,964
Noise Impact Index = 0.188
Hearing Loss-Weighted Population (HWP) = 0
Average Potential Hearing Loss (PHL) = 0
Corresponds to Figure E-11
Includes: o Highway noise
o Self-generated noise from both neighborhoods
o Noise from both the original and the project
runways
Note: Single number indices are not computed on the
basis of industrial/commercial employees.
-------
Table E-9
Sample Data Presentation for Airport Example:
Future Levels Without Proposed Project; 2001
Yearly L
-------
Table E-10
Sample Data Presentation for Airport Example:
Future Levels From All Noise Sources Combined; 2001
Yearly Ldn
(dB)
80-85
75-80
70-75
65-70
IJJ 60-65
K 55-60
55
Residential
Population
1,233
30,799
30,346
3,942
1,971
0
0
68,291
Industrial/
Commercial
Employees
303
19,473
22,921
25,431
0
0
0
68,128
Total Land
Area (sq. km.)
12.0
20.7
49.15
184.3
33.85
0
0
300.0
Industrial/
Residential Land Commercial Land
Area (sq. km.) Area (sq. km.)
0.16 11.84
4 16.7
4.85 44.3
2.56 181.74
10.88 22.97
0 0
0 0
22.45 277.55
Level Weighted Population (LWP) = 24,597
Noise Impact Index (NII)= 0.360
Hearing Loss-Weighted Population (HWP) = 408
Average Potential Hearing Loss (PHL) = 0.09
Corresponds to Figure E-13
Includes: o Highway noise
o Self-generated noise from both expanded
neighborhoods
o Noise from both the original and the project
runways
Note: Single number indices are not computed on the
basis of industrial/commercial employees.
-------
A summary of the assessment results is presented below. Calculation
of the level weighted population was based on the weighting function of
equation 2b, shown in Table 3 of the main text.
From Data In Example
Year LWP Table Number
Present (no project) 5,787 E-5
1985 with the project 5,787 E-5
1985 with the project 5,964 E-8
2001 without the project 12,647 E-9
2001 with the project 24,597 E-10
The Noise Impact Index, according to equation 7 in section 2.2.2.2, is
simply the LWP divided by the total population.
From Data In Example
Year %>tal LWP Nil Table Number
Present (no project) 31,689 5,787 0.183 E-5
1985 without the project 31,689 5,787 0.183 E-5
1985 with the project 31,689 5,964 0.188 E-8
2001 without the project 68,291 12,647 0.185 E-9
2001 with the project 68,291 24,597 0.360 E-10
The Relative Change in Impact is calculated from equation 8.
From Data on From Data in
Airport Example
Year LWPa Table Number LWPb Table Number RCI
1985 5,964 E-6 5,787 E-5 0.0306
2001 24,597 E-8 12,647 E-7 0.9449
As previously noted, the indices representing potential hearing damage
risk are not similar to the other three indices. In order to emphasize the
severity of the health problems caused by high noise levels, a separate
severe health effects single number index is used. As discussed in Section
2.3.1, for areas with an L^ of 75 dB* or above, the following information
should be estimated if possible: the population spending time out-of-doors;
the length of time they are out-of-doors; the actual noise levels while they
are out-of-doors. For this example, only the populations within the residen-
tial areas are being considered. Those people in other impacted areas of the
metropolitan area are assumed to remain indoors (because the metropolitan area
is entirely commercial/industrial, so they are not subjected to the noise.
The area for study of possible severe noise impact shown on Figure E-13
is expanded and shown as Figure E-14.
*As pointed out in Section 2.3.1, the L^ should be used only to identify
potential problem areas.
E-36
-------
fmnnmmummmnnntnni
INCLUDES:
O NOISE FROM THE PROPOSED PROJECT
• SELF-GENERATED NOISE FROM THE
HIGH POPULATION DENSITY
NEIGHBORHOOD
CORRESPONDS TO TABLE E-11
Figure E-14. Sampla Data Presentation for the Airport Example: Severe Health Analysis Area
K-37
-------
The additional information required is now assumed to have been collected
through additional estimates or survey work. It is listed on the next page as
Table E-ll.
The next step is to obtain the at-ear outdoor L values of the expo-
sure instead of the L,_ values. The best way is to tike additional noise
measurements. A much less preferable way, as explained in Section 3.2.1,
is to use the approximation:
Leq = Ldn(daytime) " 3
This approximation may be used if the difference between the daytime and
nighttime levels is the typical one where the nighttime level is approxi-
mately 4 dB lower. It has been assumed in this example that sufficient
measurements were taken to determine that the day-night difference in L
of that area is a typical one. Thus, the 3 dB correction term has been^
applied and the results entered in Table E-ll.
Next, since not all residents are exposed to exactly eight hours of out-
door noise, the data in Table E-ll are adjusted to the appropriate Le
values, using Table E-12. For example, the population of 7,521 e>
4 hours of L /4N = 73 dB has the equivalent of an L /8\ exposure of 70 dB.
These L 's nave been entered into the cells of Tablt^E-ll in brackets
([ ]). T$ow the number of people exposed to various levels of noise over
L ,0. = 75 may be read from Table E-ll:
eq(8)
Population Distribution as a
L.QJ Function of 8-hour L
76 (2542 + 691) = 3,233
77 (694 +287) = 981
78 176
79 (225 + 97) = 322
80 0
81 67
TOTAL 4,779
PHL may thus be computed from equation 13 as:
P(L )i . H(L
PHL = - -
Total
where P(L ) is the population distribution as a function of 8-hour L^^
(shown above), H(L ) is the corresponding weighting function shown onTTable
5 of the main textTor all ?(!•), where H( ). ^ O, and P-y^-i is the
population for the severe healfn effects area, i.e. , the sunrof all people exposed
to an L ,g. greater than 75 dB. Using the information in Table E-ll, the PHL is:
PHL =
4779
E-38
-------
Table E-11
Sample Data Presentation for Example the Airport: Information required to calculate the PHL
Residential Population
.—
[Exposures in Lg (8)]
Contour
Residential
Population
Median
Contour
Time Outdoors
4 hr
6 hr
8 hr
12 hr
Estimated
75-77
12,574
76
7,521
[75]
73
u* ,
1
77-79
9,474
78
6,378
[77]
75
79-81
8,953
80
5,899
[79]
77
81-83
1,031
82
691
[81]
79
TOTAL
32,032
20,489
5,560
3,997
1,986
Corresponds to Figure E-14
-------
Table E-12
Conversion of I^q(X) to Leq(8) and Leq(24)
- 9
- 6
I«q(8) - 3 = I«q(24) " 8
- 2 =
= Leq(24) - 5
3 = Lt>al?A\ - 2
E-40
-------
P. = [[(67)x(0.9)]+[(0)x(0.625)]+[(322)x(0.4)]+[(176)x(0.225)]+[(981)x(0.1)]+[(3233)x(0.025)]]
4779
PHL - ,!?5f-= 0.09
2.5 Conclusions of the Noise Analysis
The purpose of the noise analysis is to compare the number of people
affected by the noise levels with and without the project. To do this
comparison, the single number indices are used. In tabular form, they are
shown in Table E-13. These indices show: 177 more people in the residential
areas will be fully adversely affected by the noise created by the project
in 1985 (when compared with the existing situation or with 1985 without the
project); 18,810 more people in the residential areas will be fully adversely
affected by the noise from the project in 2001 (when compared with the existing
situation); 11,950 more people will be fully adversely affected by the noise
created by the project in 2001 (when compared with 2001 without the project).
In addition, the PHL for the year 2001 increases from 0 to 0.09.
FOOTNOTE TO THE HIGHWAY AND AIRPORT EXAMPLES:
STEP-BY-STEP EXPLANATION OF COMBINATION OF NOISE CONTOURS FROM
DIFFERENT SOURCES
How are the noise levels on Figures E-2 and E-4 combined?
Using Table E-14, combine the noise levels shown on Figures 2 and 4 by determining
the difference between levels at the same point, and adding the appropriate amount
from the table to the higher level.
E-41
-------
Table E-13
Summary of Airport Example
Single
Number
Index
IWP
(people)
Nil
FCI
PHL
Without the Project
Existing
and 1985 2001
5,787 12,647
0.183 0.185
- -
0 0
With the Project
1985 2001
5,964 24,597
0.188 0.360
- -
0 0.09
Difference
Between the
Existing &
1985 with
the Project
177
0.005
Difference
Between the
Existing &
2001 with
the Project
18,810
0.177
Difference
Between 1985
without the
Project and
1985 with
the Project
Difference
Between 200
without the
Project and
2001 with
the Project
177 11,950
0.005
0.175
For 1985: 0.0306 For 2001: 0.9449
0
0.09
0
0.09
to
-------
Table E-14
Number of decibels
Difference between to be added to
Levels in decibels Higher Level
0
1
2
3
4
5
6
7
8
9
10
12
14
16
3.0
2.6
2.1
1.8
1.5
1.2
1.0
0.8
0.6
0.5
0.4
0.3
0.2
0.1
For example, the noise level at the first row of houses is 60 dB in
Figure E-2 and 65 dB in Figure E-4, a difference of 5 decibels. Table
E-14 shows that for a difference of 5 dB, approximately 1 dB should be
added to the higher level in order to derive the total level. Therefore,
the noise level at the first row of duplexes in Figure E-5 is computed as
66 dB. Similarly, the noise level contour at the second row of houses is
64 dB. In summary:
E-43
-------
Add to
Higher
Duplex Row
1
2
3
4
5
6
7
8
9
10
11
Figure E-2
60
58
56
55
54
53
53
53
52
52
51
Figure E-4
65
62
58
54
53
51
49
48
48
47
47
Difference
5
4
2
1
1
2
4
5
4
5
4
Level
1.2 or 1
1.5 or 2
2.1 or 2
2.6 or 3
2.6 or 3
2.1 or 2
1.5 or 2
1.2 or 1
1.5 or 2
1.2 or 1
1.5 or 2
Figure E-5
66
64
60
58
57
55
55
54
54
53
53
E-44
-------
PARTIAL BIBLIOGRAPHY OF MODELS
FOR PREDICTING NOISE LEVELS FROM VARIOUS SOURCES
Highway Noise
U. S. Environmental Protection Agency, Office of Noise Abatement
and Control. Comparison of Highway Noise Prediction Models.
EPA 550/9-77-355. May, 1977.
U. S. Environmental Protection Agency, Office of Noise Abatement
and Control. Highway Noise Impact. EPA 550/9-77-356.
May, 1977.
Aircraft Noise
U. S. Environmental Protection Agency, Office of Noise Abatement
and Control. Calculation of Day-Night Levels (L.J
Resulting From Civil Aircraft Operations.EPA 550/9-77-450.
January, 1977.
U. S. Department of Transportation, Federal Aviation Administration,
Office of Environmental Quality. Federal Aviation
Administration Integrated Noise Model. April, 1978.
U. S. Department of Transportation, Federal Aviation Administration,
Office of Environmental Quality. Federal Aviation
Administration Integrated Noise Model Version I.
FAA-EQ-78-01. January, 1978.
U. S. Department oif Defense, Air Force Aerospace Medical Research
Laboratory. Community Noise Exposure Resulting From Air-
craft Operations; Computer Program Description, AMRL-TR-
73-109, November 1974. Computer Program Operator's Manual,
AMRL-TR-73-108, July 1974. NOISE MAP 3.4 Computer Program
Operator's Manual, AMRL-TR-78-109, December 1978.
Rapid Transit Noise
U. S. Department of Transportation, Urban Mass Transportation
Administration, Office of Technology Development and
Deployment, Office of Rail and Construction Technology,
Noise Rating Criteria for Elevated Rapid Transit Structures.
Report No. UMTA-MA-06-0099-79-3. May, 1979.
E-45
-------
Transmission Line Noise
Ccmber, M. G., and L. E. Zaffanella. "Audible Noise."
In: Transmission Line Reference Book 345 KV and above.
Published by Electric Power Research Institute,
3412 Hillview Avenue, Palo Alto, CA 94304, 1975.
Driscoll, D. A., and F. G. Haag. "Prevention and Control of
Environmental Noise Pollution in New York State."
In: Proceedings of a Workshop on Power Line Noise as
Related to Psychoacoustics. Published by the Institute
of Electrical and Electronics Engineers, Inc.,
345 East 47th St., New York, NY 10017, 1976.
Molino, J. A., G. A. Zerdy, N. D. Lerner, and 0. L. Harwood.
"Psychoacoustic Evaluation of the Audible Noise from
EHV Power Lines." To be published.
Kolcio, N., J. DiPlacido, P. M. Dietrich. "Apple Grove 750 KV
Project — Two-Year Statistical Analysis of Audible Noise
from Conductors at 775 KV and Ambient Noise Data."
In: IEEE Transactuibs ib Oiwer Appartus and Systems, 1977.
Taken together, these references provide enough information to predict
transmission line noise under different conditions. Of the four references,
the first is the most useful.
E-46
-------
-------
RESPONSE TO COMMENTS ON
Guidelines for Preparing Environmental Impact Statements on Noise
Background and Summary
In 1972, the Office of Noise Abatement and Control of the U.S. Environ-
mental Protection Agency requested the Committee on Hearing, Bioacoustics
and Biomechanics (CHABA) of the National Academy of Sciences to prepare a set
of procedural guidelines for evaluating the environmental impact of noise.
In response, CHABA formed Working Group 69 for this purpose, chaired by
Dr. Henning von Gierke. The Working Group completed its task in early 1977,
submitting the report entitled Guidelines for Preparing Environmental Impact
Statements on Noise (referred to in the rest of this report as the Guidelines),
During the summer of 1977, copies of the Guidelines were distributed to
federal agencies and other interested parties, with a request for comments.
On June 30, 1978, a request for comments was published in the Federal Register
(43 Fed. Reg. 28549).
As of 22 September 1978, a total of 45 letters, many with detailed
attachments, had been received in response to these requests for comments.
Thirty-three different organizations sent comments, of which 13 were indus-
trial concerns or trade organizations, 4 were state agencies, and 16 were
federal agencies (Table 1). In addition, comments were received either by
EPA or by the National Academy of Sciences from three individuals (Kryter (2),
Ward, and Young), from the chairman of Committee E-33 (on Environmental
Acoustics) of the American Society for Testing and Materials, and from
Transport Canada.
-------
Table 1
Commenting Organizations
Industrial
AMETEK
Columbia Gas Systems Service Corp. (Columbia Gas)a
Consolidated Edison Co. of N.Y. (Con Ed)
Dames and Moore
(3)b Edison Electric Institute (EEl)
Ford Motor Co. (Ford)
General Motors Corp. (GM)
(2) Motor Vehicle Manufacturers Association (MVMA)
Northern States Power Co. (NS Power)
R & F Coal Co.
United Engineers
Vibra-Tech Engineers
Wyle Laboratories
State
Kentucky Dept. of Mines and Minerals (Ky. Mines)
(2) New York State Dept. of Environmental Conservation (NY Envir)
New York State Dept. of Transportation (NY DoT)
(2) New York State Public Service Commission (NYPSC)
Federal
Dept. of Agriculture, Agricultural Research Service (Ag Res Serv)
(2) Dept. of Commerce, National Bureau of Standards, Institute for Basic
Standards (NBS)
Dept. of Defense, Deputy Assistant Secretary, Energy Environment and
Safety (DoD)
Air Force, Aerospace Medical Research Lab. (AMRL)
Army, Construction Engineering Research Lab (CERL)
Dept. of Housing and Urban Development, Office of Environmental
Quality (HUD)
Dept. of Interior, Deputy Assistant Secretary (Dol)
Dept. of Labor, Occupational Safety and Health Admin. (OSHA)
Dept. of Transportation, Assistant Secretary, Environment Safety and
Consumer Affairs (DoT)
Federal Aviation Administration (FAA)
Federal Highway Administration (FHWA)
Urban Mass Transportation Administration (UMTA)
National Aeronautics and Space Administration (NASA)
Naval Construction Battalion Center (NCBC)
Postal Service
Veterans Administration (VA)
.This abbreviation will be used to reference the comments.
Number of separate letters sent, if other than one (1).
-------
Many of those commenting raised excellent points regarding the clarity
or consistency of the publication. Others questioned the technical material
or the reasoning in the Guidelines. The major issues addressed by the com-
ments are:
(1) the title and purpose of the Guidelines,
(2) the inclusion of impulse noise and vibration,
(3) the treatment of noise increases in quiet areas, including effects
on wildlife and other animals,
(4) the use of yearly day-night sound level,
(5) the interpretation and empirical basis for calculations of level-
weighted population, and
(6) the organization and clarity of the document itself.
The responses to many of these are inter-related conceptually, although
they are written in separate and distinct sections in the body of this report.
We begin with the view that the comments about the title and purpose are
correct: noise is only one of many impacts to be considered in any environmental
assessment. The Guidelines are intended to help people treat noise impacts in
order that a decision-maker can understand them better, and can incorporate them
more fully in the comparative evaluation of all of the factors in the decision.
To facilitate such a treatment of noise impacts, the material in the
Guidelines has been reorganized around four topics: (1) the general adverse
response to audible noise; (2) hearing damage and other severe health effects
from higher level audible noise; (3) environmental degradation in rural and
natural areas; and (4) special considerations for ultrasound, infrasound,
impulse noise, and vibration. For the first topic, the approach in the
Guidelines is followed, and an effort is made to improve comprehension of
111
-------
"level-weighted population." For the second, the weighting function for loss
of hearing is based explicitly on at-ear exposure, which may be determined by
8-hour Lgq, 24-hour Leq, or outdoor L(jn depending on circumstances. For
the third topic, we are using the concept of environmental degradation but not
providing a procedure to calculate a single-number index for it. For the
fourth topic, we have tried to clarify when each of the special noises is to
be considered, and the reasons for the approach taken.
IV
-------
Response to Comments on Guidelines
Contents
Background and Summary
Introduction
Overall Assessments
Favorable Comments
Unfavorable Comments
General Comments
A. Simplification
B. Example Applications
C. Accountability
Title
1. Comments on Title
Front Matter
2. Membership of the CHABA Working Group
Chapter I: Introduction
3. Inclusion of wildlife
4. EIS
5. Purpose of Guidelines
6. Inclusion of Vibration
7. Mandatory Guidelines?
8. Conflict With Other Procedures
9. Federal Environmental Impact Review Process
10. Rationale
11. Hearing and Day-night Average Levels
12. Table 1-2
13. General Audible Noises
14. Temporary Changes
15. Feasibility of Forecasts
16. Chapter IV Omitted
Chapter II
17. Noise Due to Project
18. Fig. II-l
19. Noisy Sources
1
1
2
2
4
4
4
4
8
9
10
10
11
12
13
13
13
15
15
15
16
17
18
18
25
-------
Contents (Cont'd)
Chapter III
20. Fig. III-l 26
21. Project Life Span 26
22. Five-year Intervals 27
23. Noise Impacts Only 27
24. Simplified Procedures for Temporary Change 27
25. Uncertainties 28
Chapter IV
26. Any Change in Impact 29
27. Level of Detail 29
28. Noise Measures: YDNL 30
29. Other Noise Measures 36
30. Base YDNL 36
31. Estimating DNL 38
32. Summary Tables 40
33. Property Line Limits 41
34. Land Use Categories 42
35. Special Situations 42
36. Residential Average Number 43
37. Tables IV-2 and IV-3 43
Chapter V
38. CSEL for Impulse Noise 45
39. Measuring CSEL 46
40. CDNL 47
41. Infrasound 47
Chapter VI
42. Percent Highly Annoyed 49
43. Table VI-1: Criterion Levels for
Various Land Uses 50
44. Table VI-2, 3, and 4: Summaries of
Human Effects 51
45. Response Independent of Source 52
46. Degradation of Environment 54
47. Effects of Impulse Noise 55
48. Infrasound 56
49. Vibration Criteria 57
50. Blast Noise 58
Chapter VII
51. Single Number Impact Index 59
52. Level Weighted Population 60
53. PLH and Other Abbreviations 61
VI
-------
Contents (Cont'd)
54. Population Weighted Loss of Hearing 62
55. Normalizing Weighting Function 63
56. Sound Level Weighting Function 63
57. Weighting Function for Loss of Hearing 64
58. Change in LWP 64
59. Temporary Noise Environments 65
60. Vibration Impact Index 65
Chapter VIII
61. Noise Re Other Impacts 66
62. Structures Exposed 66
63. Sound Level Weighted Area 66
64. Descriptive Qualitative Evaluations 67
65. New Population Exposed 67
Appendix A
66. Terminology 69
Appendix B
67. Exponential Functions 69
68. Population Weighted Hearing Loss 70
Appendix C
69. ISO/TC 108/SC 2/WG 3 70
70. Figure C-4 71
vn
-------
Introduction
The comments are presented in the order one would encounter them in read-
ing through the Guidelines. Comments on the same topic from several sources
have been listed together, and appear the first time a topic is encountered.
In addition to the many comments on specific material, nearly half of those
commenting on the Guidelines included an overall assessment of the report.
These are summarized first. Also, several general comments were offered which
do not pertain to a specific part of the Guidelines. These are listed and
answered following the overall assessments, before turning to the more spe-
cific material.
Overall Assessments
Of the 33 different organizations which sent comments, 16 made explicit
statements about the overall usefulness or appropriateness of the Guidelines.
Eight of these were favorable; eight were negative, calling for delay or
withdrawal of publication. Extracts or precis of all 16 of these general
comments are listed here. Most of these contained particular questions or
comments as well as the overall appraisal. It should be kept in mind in
reading these that the praise was not without reservation, nor the criticism
without reasons.
Of the eight favorable comments, six were from federal government
agencies, and two from industry. No federal agencies sent unfavorable over-
all comments. Half of the eight suggestions for delay or withdrawal of publi-
cation came from public utilities—three from the electric power industry and
one from a gas company. Of the remaining four, two were from New York State
agencies, one from another state agency and one from a manufacturer.
1
-------
Favorable.Comments
"OSHA applauds this report and agrees that the suggested guidelines will
be useful in a wide variety of applications." (Occupational Safety and Health
Adminis trat ion)
"We agree with the approach...; the Guidelines should provide consider-
able assistance to the Navy..." (Naval Construction Battalion Center)
"I believe that it will be suitable for its intended function...[if]
the preparing group will update it as needed." (Ametek)
"We think it is an excellent report, representative of the best state-
of-the-art today." (National Bureau of Standards)
"As a technical guideline, this document provides excellent direction
for evaluating noise impacts." It should be published only as a guideline
and not as a regulation. (U.S. Postal Service)
The Guidelines "is a fine document which contains the necessary noise
impact criteria in one place." "...it is an important document which should
get wide dissemination." (Dames and Moore)
"Overall we believe that the guidelines can serve a useful purpose in
assisting agencies in their assessments of the environmental effects of
noise." (National Aeronautics and Space Administration)
"As you can see, most of our comments are minor in nature due to the
overall high quality of the Report." (Urban Mass Transportation Adminis-
tration)
Unfavorable Comments
"We have serious reservations regarding the validity of the Guidelines
and urge that U.S. EPA not adopt them as a 'uniform national method for
-------
noise impact assessment* (page iv)." (New York State Department of Environ-
mental Conservation)
"EEI recommends that EPA not publish or endorse the Guidelines for the
reasons cited in the attached comments." (Edison Electric Institute)
"NSP recommends that EPA not publish or endorse the Guidelines for the
reasons stated." (Northern States Power Company)
"...we must strenuously recommend that this document, which is in effect
a regulation, be required to have included an economic impact statement prior
to adoption." (Kentucky Department of Mines and Minerals)
"The docket should remain open for six months to allow for the completion
of these existing studies [at Battelle Columbus Laboratories] so that they may
be made available to EPA and the scientific community." (Ford Motor Company)
"Con Edison recommends that the Guidelines not be published or endorsed
by the U.S. Environmental Portection [sic] Agency (EPA) for the following
reasons." (Consolidated Edison Company of New York)
"...there are a number of serious flaws related to the reliability of
the impact assessment which I believe make it unwise to publish the Guidelines
in their present form." (State of New York Department of Public Service)
"...Columbia strongly recommends that the Committee withdraw its Proposed
Guidelines until after EPA prepares an L^n standard..., prepares an Economic
Impact Statement..., submits it for public review and comment, and publishes
a final recommended L,jn standard." (Columbia Gas System Service Corporation)
-------
General Comments
A. Comment (Simplification)
Dol, HUD, and UMTA all say that the Guidelines are complex and hard to
follow. UMTA complains about the foreword references, e.g., from Chapter IV
to VI and VII, which make the material hard to follow.
Response
The revision attempts to overcome these problems, primarily by putting
all material necessary for the treatment of a specific type of noise envir-
onment in one section.
B. Comment (Example Applications)
NASA, DoD, and NYSOoT all state that if EPA intends these Guidelines
to be widely used, they should demonstrate the merit of the procedures with
case histories, real-world applications, or field tests, which can show their
usefulness in the decision-making process. Dames and Moore echoes this in
their observation that the Guidelines are "not as applicable to the site
specific EIS as I would like," i.e., to those prepared "for local developers,
industrialists, or mine owners."
Response
More realistic examples are included in the revised version, to cover a
fairly wide range of possible applications.
C. Comment (Accountability)
New York Envir. points to "... a fundamental weakness in the environmental
impact assessment process, it ... fails to hold the project sponsor account-
able for the assumptions and predictions upon which the decision was based."
-------
Response
This is, of course, a weakness of the EIS process, not simply of these
Guidelines. Hopefully, the Guidelines (in both original and revised forms)
improve the probability of reasonably accurate assumptions by making those
assumptions explicit, at least in the noise assessment. In controversial
projects, weak assumptions are usually brought out by project opponents during
public hearings. Thus, while the Guidelines do not solve the accountability
problem, they do help to minimize it.
Title
1. Comment
Four federal agencies (HUD, DoD, DoT, FHWA) suggested that the existing
title of this publication is not appropriate, or is misleading. There
was considerable agreement on the reasons for this assertion among the four.
EIS's do not often address noise alone (FHWA, HUD). Noise is only one of
many environmental concerns in an EIS (FHWA, HUD, DoD, DoT). Noise assessments
are done in contexts other than simply an EIS (FHWA).
DoT proposes that "the title be revised to suggest that it provides guide-
lines for noise analysis and assessment as part of the comprehensive EIS
process." Two other agencies suggested new titles:
"Environmental Assessment of Noise" (FHWA); and
"Guidelines for Noise Impact Evaluation in the Environmental Impact
Analysis Process" (DoD).
Response
These points are all well-taken. Revision of the title is necessary.
The title of the revised document is "Guidelines for Noise Impact Analysis."
-------
Front Matter
2. Comments on the Membership of the CHABA Working Group
Some concern was expressed that the composition of the Working Group
which prepared the Guidelines was too limited. Five different writers made
this point, in varying degrees of strength. "The Guidelines were prepared
by a select group of 13 people who do not represent a consensus of the
acoustical profession. Almost half of Working Group 69 are employed by the
Federal government, and about one-quarter work for a government contractor.
Working Group 69 did not include a representative from a local or state
pollution control agency, nor any representation from the manufacturing,
construction, transportation, or utility industries. We recommend that
the Guidelines be reviewed by a consensus standards writing organization,
such as the ASTM E-33 committee..." (EEI; a very similar comment from Con
Ed). "The WG was clearly loaded with persons who have already been ordained
in the Total Energy religion." "I do hope...that in the future the composi-
tion of Working Groups is made broad enough to guarantee discussion of
alternative theories and approximations..." (Dixon Ward). "The EPA should
refer the comments submitted to the docket to a committee comprised of per-
sonnel with technical expertise (expanded from the original CHABA group)"
(Ford). "[N]one of the members of CHABA working group 69 is an established
authority in the area of blasting or impulsive vibrations ..." "Almost every
recommendation made in the proposed standard ... is contrary to the recommen-
dations of the U.S. Bureau of Mines and other authorities after 20 years
practical research ..." (Vibra-Tech).
-------
Response
The primary reason that EPA has requested comments on the report of
the CHABA Working Group was to provide an opportunity to consider additional
viewpoints and expertise in revising the proposed Guidelines. By carefully
reviewing comments submitted, and revising the Guidelines accordingly, any
problems of the kind raised are hopefully overcome. There would seem little
reason to rely on a consensus standards writing organization, since this docu-
ment is not a standard. It is a compilation of procedures for improving com-
munication about noise impacts, in hopes that decision-makers will be able to
incorporate them more fully into decisions. (Part of the cause of the mis-
understanding of this intent is perhaps addressed in comment 5 regarding the
purpose of the Guidelines.)
-------
Chapter I; Introduction
3. Comment (p. 1-1, para 1: inclusion of wildlife)
"We recommend deletion of Sections VI-E, VIII-A-1, and references to
'wildlife* and 'animals' on pages 1-1, III-2" (NS Power, EEI). "One cannot
assume that identical noise criteria can be utilized for people as well as
for wilderness animals. While there is little technical literature on this
subject, our terrestrial observations indicate that common species, i.e.,
squirrel, deer, seem unaffected by man's noises whereas certain others such
as bald eagle and timber wolf are highly sensitive to all disturbances by
man" (NS Power). "Protecting wildlife and domestic animals from noise vibra-
tion levels equivalent with human inhabitants is absurd" (Ky Mines). "There
is so little information on the effects of noise on animals, and effects
assumed in the Guidelines are gross speculation.... We recommend that EPA
sponsor additional studies on the effects of noise effects of noise [sic] on
animals..." (EEI). "A statement is even made to essentially ignore the impact
of ultrasound (>20 kHz) on animals, even though it is well known that dogs are
very sensitive to ultrasound" (United Engineers). "The proposal...'exposure
level identified to protect man will also protect animals' appears reasonable
but would be unduly restrictive if it were established as a maximum noise
restriction for animals" (Ag Res Serv).
Response
If these Guidelines are to serve as an aid to people who have found
noise to be a significant consideration in a proposed project, the document
should be comprehensive. It should also, however, be based on known facts,
and these are admittedly scarce for animals and wildlife. Qualitative
-------
statements about the effect of noise on animals are simply all that can be
made. (Examples are the statements NS Power makes about eagles and wolves,
and United Engineers makes about dogs.) If sensitive animal populations
are present in a project area, they should not be overlooked in the noise
assessment. On the other hand, if there is no information that suggests
noise problems for animals, it is not reasonable to assume they are more
sensitive to noise than people are. In general, then, the effects of noise on
animals will be listed as a "special situation" in the tabulation of impacts,
and will be described qualitatively in the text of the noise impact analysis.
(See the EPA publication Effects of Noise on Wildlife and Other Animals
(NTID 300.5, Dec. 1971).)
4. Comment (p. 1-1, para 1: EIS)
The reference to Environmental Impact Statements (EIS) is both incomplete
and misleading. "[T]he report does not give the uninitiated reader any clue
to what an EIS is, its objectives, and what it stems from" (HUD). The Guide-
lines "fail to take into consideration the mandates of NEPA in regard to
when a 'detailed statement1 or 'Environmental Impact Statement' is required"
(Columbia Gas).
Response
These comments are both correct. The Guidelines were written for people
involved in the NEPA process (or similar ones), so that a detailed descrip-
tion of that process was not required in the document. Therefore, the re-
vised version has not expanded the discussion of the NEPA process. Instead,
it has provided a broader statement of the rationale for these Guidelines.
(See comment 5 also).
-------
5. Comment (p. 1-1, Purpose of Guidelines)
Most of those who suggested altering the title also suggested revising
the purpose of the Guidelines, as did others (FHWA, DoT, OoD, FAA). The
purpose should read, "This report offers guidelines for the environmental
assessment of noise" (FHWA). "[T]he purpose of the document should be to
assist those evaluating the environmental impact of proposed actions and
to address noise as a contributing factor..." (DoD).
Response
This section has been rewritten to make exactly those points. The
revised section is similar to the following statement by CEQ: "The NEPA
process is intended to help public officials make decisions that are based on
understanding of environmental consequences, and take actions that protect,
restore, and enhance the environment" (Council on Environmental Quality,
Proposed amendments to regulations; Federal Register, June 9, 1978, p. 25232
Sec. 1500.l(c)). The purpose of the revised version, then, is to present
procedures which can be used to express noise impacts in terms which are easily
understood by decision-makers, so that those impacts can be fully incorporated
in the comparative evaluations which constitute the decision.
6. Comment (p. 1-1, para 2: inclusion of vibration)
Three writers, all from the electric power industry (Con Ed, EEI, NS Power),
suggested that the material on infrasound, ultrasound, and vibration should be
deleted from the Guidelines, because "the 1972 Noise Control Act gave EPA
authority to control noise only within the audible frequency range" (Con Ed).
Vibra-Tech also requested deletion of this material, on the grounds of the
Working Group composition. (See Comment 2.)
10
-------
Response
NCA 1972 defines environmental noise as "the intensity, duration, and the
character of sounds from all sources" (Sec. 3.11) but does not further define
'sound'. The question of whether or not there is sound if it cannot be heard
is an age-old semantic one, and no attempt is made to answer it here. Since
the revised Guidelines are offered simply as an aid to decision-making, and
their use is not mandatory, we feel it is entirely appropriate to offer
guidance for a variety of "non-audible situations."
7. Comment (p. 1-1, para 2: mandatory Guidelines?)
"One may assume that by adoption the guidelines take the character of a
'standard* and their use [becomes] mandatory" (NASA). EEI expands on this
notion: "EPA endorsement of the Guidelines would constitute de facto
regulation of environmental noise emissions from fixed facilities, which
appears to be beyond the authority granted EPA." Several of the comments
seemed to indicate a fear that these Guidelines would become mandatory,
thus entailing double work where other forms of noise assessment are also
required (FHWA, Ky Mines), or that they would be interpreted as noise level
standards by field personnel (Con Ed, EEI). Ky Mines also stated that since
the Guidelines are in effect regulations, there should be an economic impact
statement prior to their adoption.
Response
Given the purpose stated in comment 5, it should be obvious that these
guidelines are merely an aid for quantifying noise impacts, and should not
be considered as standards. A discussion in this regard is included in the
preface. It should be noted that the methods in the Guidelines are reasonably
11
-------
consistent with the procedures followed by most agencies at present. Because
the Guidelines are offered as only an aid to planning and decision-making, and
are not mandatory requirements, no economic impact statement is necessary.
8. Comment (p. 1-1, last para: conflict with other procedures)
Columbia Gas suggested that "If the Committee desires the Proposed
Guidelines to be used by federal agencies," delete the statement that
"Agency's guidelines would be expected to take precedence over those pro-
posed in this document" "where there is conflict." On the other hand,
NASA and FAA supported the statement in the Guidelines, and the FHWA goes
even further, saying that this is an important point which "should be stressed
in the summary." NYSDOT points out that because of this provision "the
Guidelines would have little if any application to our Department's activities"
since NYSDOT noise studies "presently follow procedures outlined in Federal
Highway Program Manual 7-7-3." Dol on the other hand observes that "... lack
of uniformity complicates efforts in seeking to comply with all applicable
noise control requirements," and refers specifically to quite different
requirements in Oregon and Washington.
Response
The response to comment 7 is appropriate here also. Because, the Guide-
lines are for the most part consistent with procedures of other agencies, the
question of making them mandatory should not arise. Furthermore, it should
be noted that many components of the procedures used or cited within the Guide-
lines have been standardized both nationally and internationally. The revised
Guidelines strongly recommend relating impacts to people, and suggest how to
do this, but they do not require a specific form of analysis. The point made
12
-------
by NASA, that "agency procedures need to retain sufficient flexibility to
be applicable to special conditions not addressed in these guidelines," is
a reasonable compromise position.
9. Comment (p. 1-2, para B.2; and Table 1-1: federal environmental impact
review process)
DoT suggested that Table 1-1 should be deleted or modified, "since the
EIS review process varies widely among Federal agencies." DoD, on the other
hand, suggests, "Additional drafting is needed to properly describe the NEPA
process."
Response
Expansion of the context of the guidelines should obviate the need
for a detailed discussion of the NEPA process. The procedures are useful
any time an analyst needs to investigate noise impacts. Specification
or description of all such occasions is unnecessary. Section I-B and Table
1-1 have been deleted.
10. Comment (Sec. I.C.: Rationale)
FHWA agrees completely with the rationale as stated. NBS would add the
phrase "consistent with reasonable accuracy" to "should be uniform and as
simple as possible."
Response
These comments are accepted. Other portions of the rationale have
been revised to be consistent with the new title and purpose.
11. Comment (p. 1-4: hearing and day-night average levels)
DoD notes that the 'Levels' document "does not endorse L^n as a suitable
technique for assessing the potential for hearing loss" but recommends Len
13
-------
instead. NBS raises similar concerns in conjunction with pages VI-8 to VI-11 of
the Guidelines, asking, "Ought not the 75 dB threshold be identified as an Leq(8)?
...Also, the need to use Leq(24) rather than L^n should be made explicit."
Response
Von Gierke's letter (6 Dec 77) responding to the DoD comment indicates
that Working Group 69 used L
-------
stated in the note to Figure VI-2 in the original version), the user of the
revised Guidelines may wish to use outdoor L
-------
longer than six months, but less than 10 years. Dol suggests that the
short term/long term division be at 2 years, and that the upper limit on long
term temporary changes be 25 years. Dames and Moore, on the other hand,
suggests 3 years as the upper limit of temporary change. Dol's reasoning is
that "it is common practice in the electric utility industry to install a
temporary transformer unit for a period of from six months to two years;" and
that "most electrical facilities have a useful service life of at least 25
years." NBS points out a further inconsistency in that the Guidelines state (on
p. III-9) : "Temporary changes in which the daily DNL is less than 90 dB may
modify, simplify, or eliminate full Noise Environment Documentation...."
Response
First, since the Guidelines do not seem to employ the distinction
between long and short term anywhere, the revision will speak only of
temporary changes. Second, 2 or 3 years seems a reasonable upper limit on
a usual temporary change, with the possible exception of some major construc-
tion efforts. For example, Dol speaks of "temporary transformers" in place
for up to 2 years. The 90 dB (Ldn) cut-off is a typographical error,
which should be 70 dB (to be consistent with Fig. II-l). However, a special
cut-off for temporary projects is not used in the revised document, consistent
with the new screening approach (see response to comment 18). Temporary
changes will call for a less involved noise impact analysis, as discussed in
comments 24 and 59.
15. Comment (p. 1-9, IV-8: feasibility of forecasts)
For permanent changes, the Guidelines require "a projection of popula-
tion and land use...over a twenty-year period." Dol points out that "the
16
-------
linear nature of [electrical] transmission facilities makes it impossible
to make the long range projections of adjacent populations and land uses,"
and suggests that such facilities "should be exempted from evaluations of
the effects of noise on projected adjacent populations and land uses." DoD
says the requirement for ^10% accuracy (p. IV-8) is unreasonable in view of
the 20-year horizon.
Response
In fact, transmission lines are not unique in finding it difficult
to make 20-year population projections. Even urban planners have great
difficulty (and no guarantee of success) in doing this, particularly at the
level of areal detail often required for noise impact analysis. Nonetheless,
it is important for a noise impact analysis to try to ascertain the long-
range implications of the project. Twenty years is suggested as an appropriate
time span. To avoid these prediction problems, the revised version suggests
using, where possible, the local community's land use plans or zoning designa-
tion for the areas in question, whenever such plans exist. Making this change
overcomes the criticisms (from HUD), that this document disregards community
goals, objectives, and plans. In locales without such plans, forecasts should be
attempted. When even this is not possible, the reasons need to be stated, since
it is almost always desirable to specify long-range as well as immediate
impacts.
16. Comment (p. 1-10: Chapter IV omitted)
An introduction to Chapter IV is absent from this summary (UMTA).
Response
The revised guidelines will contain appropriate introductory material.
17
-------
Chapter II
17. Comment (p. II-l: noise due to project)
The Guidelines states that the screening chart "is based on the relation
between the existing noise environment and the noise environment after the
project is completed." NBS suggests that "after the project" was intended to
say "due to the project."
Response
NBS is correct that the expected future noise levels for the screening
procedure (and for the base YDNL for impact assessment (p. IV-4)) are meant
to be calculated for the project alone, not for it added to the existing
levels. (Note that impact analysis is based on total resulting levels, how-
ever. See response to comment 18.) This seemingly minor editorial comment
points up the fact that it is extremely difficult to write these guidelines
concisely to apply equally well both to projects which produce noise and to
projects which reduce noise.. The revised version discusses the screening
diagram only in the context of projects which may increase noise. This, of
course, does not preclude use of the'methodology presented in the revised
Guidelines for those projects which may reduce noise.
18. Comment (Fig. II-l)
There were over a dozen separate comments on the rationale for or clarity
of Figure II-l, the Screening Diagram, and the text material explaining it.
The bulk of the comments are addressed to the levels identified for screening.
FHWA says the "levels shown are too low," because all highway projects would
exceed the 40 dBA lower limit. Dames and Moore say the screening "philosophy
encourages degradation of rural population density areas." Dol points out
18
-------
that the 40 dB threshold does not take into account areas where the existing
level is low, and the unique nature of wild lands, where even minor increases
in noise can "disturb the values for serenity possessed by these areas"
(Ool). DoD simply asks "What is the basis for the 40 decibel standard on
page II-l and the 50 and 60 dB standards on page II-3?" This is echoed in
Con Ed's contention that these thresholds will become de facto standards
(discussed in comment 7). Both Con Ed and EEI state that the figure
conflicts with NCA 1972 in that local governments are to set local noise
limits. Con Ed also asserts that "there should be no need for a noise
impact statement when the expected noise emissions are below...local
ambient sound levels." EEI concurs. Columbia Gas questions whether the
screening levels are consistent with NEPA Sec 102(C) which requires a
detailed statement on '"...major federal actions significantly affecting
the quality of the human environment."1 In line with this, Columbia notes
the reliance on the Levels document as the foundation for this screening
procedure, and asserts that the Levels document is being misinterpreted.
Dol also picks up on the apparent reliance on the Levels document, noting
that these "very stringent" levels were "derived based on the most sensitive
human response" and "also have a margin of safety associated with them."
A number of comments address the clarity of the diagram. DoD says it is
"difficult to understand," and that it "must be proven both scientifically and
practically." NBS points out that both axes have the same label—expected
YDNL—which is confusing. NASA points out that the double labelling of the
horizontal axis is confusing, which NBS would correct by putting the relation-
ship between population density and YDNL elsewhere. NASA also requests more
discussion of "the rationale for exempting potentially noisy projects from
detailed noise environment documentation."
19
-------
The shaded portion in the upper right of Fig. II-l, labelled "measurement
required" drew a number of comments. Columbia Gas notes that the title of the
figure says measurements "are advised," so at the very least, the shaded area
should be so labelled. HUD, NASA, and NBS all ask why measurement is needed
in this particular area. HUD goes on to suggest that "measurements should be
required based on the situation...(e.g. cases where there are aggregate
sources, where available models are not so reliable, etc.)."
Response
This diagram is obviously central to the content of the Guidelines, and it
is important, in responding to these comments, to keep in mind the totality of
the Working Group's approach. The starting point for clarification of Figure
II-l is the labelling of the axes. In his 6 Dec 77 reply to USAF comments, von
Gierke states, "This figure would be clearer, perhaps, if the ordinate was
labelled 'Expected Yearly Ljn of the Project Being Evaluated* and the abscissa
was labeled 'Existing or Expected Yearly L,
-------
Ill
-
111
€50
O
ill
40
MENTATION:
RILL NOISE ENVIRONMENT
DOCUMENTATION FOR ALL
PROJECTS '
.DEPENDS ON
-FUTURE OF
EXISTING
.NOISE SOURCES
POSSIBLE NOISE
- DEGRADATION'
ANALYSIS
»OUT
40
50
70
80
90
EXISTING 1^)
FIGURE 3. SCREENING DIAGRAM
21
-------
Given these labels, we must next address the levels themselves. Von
Gierke's letter also states, "[T]his figure is based on the observation that
if the existing background is 10 dB or more [greater] than the level produced
by a new noise source, the new noise level will increase less than one-half a
decibel." This explanation obviously applies to the diagonal line in the ori-
ginal Figure 11-1, which extends from the point representing project levels
of 40 and existing levels of 50 up to the point representing project levels of
55 and existing levels of 65. This explanation of the 10-dB-down diagonal is a
reasonable answer to Con Ed's assertion that if project levels are anywhere
below ambient, no noise impact analysis is needed. Con Ed and EE1 apparently
did not consider the fact that such situations would produce up to a 3 dB
increase in the ambient. The reasons for the 10 dB difference are stated in
the revised text. There may well be cases where departure from a rigid appli-
cation of such screening levels is warranted, which the user can explain given
a knowledge of the reason for those levels.
However, Columbia Gas rightly questions, in essence, whether a one-half
decibel increase constitutes "significantly affecting the human environment."
Von Gierke's letter goes on to say, "Thus the screening values were set con-
servatively so if a mistake was made, it would be requiring some additional
noise analysis for a project that turns out to cause little environmental
impact with respect to noise." Working Group 69, then, was operating, in
effect, on the basis of those sections of NEPA which mention "...the critical
importance of restoring and maintaining environmental quality..." (Sec.
101(a)), and which call for attaining "the widest range of beneficial uses of
the environment without degradation..." (Sec. 101(b)(3)). This seems an
appropriate perspective for a document which is intended to be a comprehensive
22
-------
guideline for assessing noise impacts. If there is any possibility of such
impacts, the method should be applicable.
In light of the comments on rural areas or wild lands, there seems a
good argument for extending the 10-dB-down diagonal below new project levels
of 40 dB. Von Gierke's letter (6 Dec 77) states, "To some extent the 40 dB
is arbitrary. It was based on...practical judgment and experience.... We
would be interested in specific cases where this was too high." The comments
appear to identify such cases, and to make the point that the screening dia-
gram should not ignore such situations, even though subsequent analysis may
show the impact to be quantitatively small. The revised screening diagram
allows for assessment of noise impact even for low project noise levels in
such cases, in recognition of the fact that quiet is a national resource, and
is one of the features of natural areas which makes them uniquely valuable.
It should be noted that such areas are valuable because they are rare: this
region of the screening diagram will not often be encountered. The revised
diagram uses a cut-off level of 45 dB for proposed projects in these quiet
areas. This level was set by combining the 10 dB down reasoning from the
previous discussion with the level of 55 dB (L. ) identified as protective
of public health and welfare in the Levels document. This 45 dB (L, ) level
is not, however, intended as a standard or regulatory level in any sense. The
expanded discussion of the rationale for the 10 dB difference should overcome
the tendency to read specific examples as standards, the way DoD apparently
did in their comment. The fact that the screening line is based on environ-
mental degradation in no way conflicts with any local noise limits. The
purpose of an environmental assessment is to allow a decision-maker to
evaluate comparatively effects on many factors, along with costs and, for
23
-------
example, services provided. The fact that noise is below local limits does not
mean a project is good: some other project may offer lower noise levels, along
with different levels of other effects, costs, and benefits. The same reason-
ing applies to the FHWA comment: noise effects should be a consideration in
all highway proposals if noise levels will increase as a result of the project.
The fact that the levels are not high enough to warrant federal cost sharing
of noise abatement efforts is not an indication that no noise impacts exist.
The termination of the diagonal line at project levels of 55 dB needs
some justification, which the existing document does not provide. In the re-
vised version, the diagonal line has not been terminated at 55 dB. Instead,
projects with levels above 55 dB, but below the diagonal, are indicated as
needing documentation depending on the likely future emissions of the existing
noise sources. In that region of the diagram, potential abatement of other
noise sources may reduce existing levels, making the proposed project the major
noise problem. This argument is made explicit in the revised version. It
should be recognized that screening in this region needs to be done with some
judgment. For example, the probability of significant noise reductions of the
other sources during the life of the project could be stated. If it is low
(as in central cities, where a multitude of noise sources are present), then a
project in this region of the screening diagram (new Fig. 3) may not require
further noise environment documentation. If existing levels are 70 dB and are
not expected to be abated, and one works through the analysis for a project
producing 55 dB, the result will be a finding of no impact. Since there seems
little point in doing a lot of work for a preordained outcome, the diagonal
line has been extended above project levels of 55 dB. It terminates at 75 dB
project levels because these are levels at which health effects are of concern.
24
-------
It is difficult to address the DoD comment that this diagram must
be scientifically and practically proven. Hopefully, a better explanation
of the rationale for the diagram will suffice.
On the matter of the advised or required measurements, there is no
available explanation for the location of the shaded portion. In the
revised version, that shaded portion of the diagram is omitted. Instead,
field measurement is identified as the preferred approach, and there is
a discussion of situations in which other estimation procedures might
reasonably be used.
In keeping with our aim to further clarify and simplify the Guidelines,
we have deleted from the screening diagram the distinction between permanent
and temporary projects, such that above the screening line (new Fig. 3), a
full noise environment documentation is recommended for all projects. This
decision is simply a recognition that nowhere do the Guidelines describe what
a "modified NED" is to consist of. It seems clear that documentation for a
temporary project would have at most only a two year, rather than twenty year,
horizon, and that this would be the time-saving modification apparently de-
sired.
19. Comment (p. II-3, noisy sources)
NASA suggests that some guidance would be useful on how to estimate the
effects of existing especially noisy sources.
Response
Agreed. In the revised version, references are made to existing (federal)
computer or manual techniques for estimating airport or expressway noise.
25
-------
Chapter III
20. Comment (Fig. III-l)
HUD says, "Since this [figure] is the key to the process, the structure of
the guidelines should follow [it] more closely and the outline should be keyed to
the Figure." HUD also states that this figure does not recognize community plans,
goals, or objectives. Animals appear as the top item in the detailed considera-
tions, and HUD suggests people should come first. (HUD's comment was made about
Chapter VIII, but since that chapter is based on the order of this figure, it is
addressed here.) NBS says, "A.4. shows exit of Ldn < 55 before examining possible
infrasound. Also B: no path if only infrasound or ultrasound is a problem."
Response
The organization of the revised report has been based on this diagram, as
suggested. Some re-ordering of the diagram was necessary, both to correct
the specific errors noted by the comments, and to improve the organization
of the Guidelines. Community plans have been brought into the Guidelines,
but it proved better to do so in the text rather than with one or two words
in this diagram. (See comment 15.)
21. Comment (p. III-3: project life span)
DoD points out that some actions may be obsolete or eliminated in 10 years
or less, so that a 20 year evaluation is unnecessary.
Response
Agreed. The impact assessment should be confined to the life-span of the
project, with the exception that when one is comparing alternatives with different
life-spans, the longest one should be applied to all of the alternatives.
26
-------
22. Comment (p. III-3: five-year intervals)
United Engineers writes, "The evaluation of impact after 5, 10, 15, and 20
years of operation can be speculative at best, and is definitely time-consuming.
...[I]t would appear to be less useful than potentially misleading."
Response
In many instances, the noise impacts may not occur at the same time for
all alternatives. In such instances, it is important to know the relative
timing of the impacts in making a decision among the alternatives. In
other cases, the comment is correct. The revised version documents the
reasons for this suggestion of evaluating impacts at five year intervals,
and suggests that judgment be used regarding its usefulness in each case.
23. Comment (pp. HI-4,5: noise impacts only)
FAA points out that this section appears to place noise as the sole
criterion in an EIS, which is contrary to statutory requirements.
Response
Agreed. This has been revised, as discussed in comments 4 and 5.
24. Comment (p. III-9: simplified procedures for temporary change)
NBS states, "Reference to Chapter VII is circular: Chapter VII refers
the reader back to this paragraph."
Response
The Guidelines are not clear on when to modify, simplify, or eliminate
the noise environment documentation, simply saying those things may be done
"as is reasonable" for "temporary changes in which the daily DNL is less than
27
-------
90 dB." The 9.0 is merely an uncorrected typographical error, and should be
70, to conform with Fig. II-l. The revision (comments 14 and 18) suggests
that all projects, temporary and permanent, conduct the same documentation,
which may be simplified as Chapter VII describes for temporary projects.
25. Comment (p. 111-10: uncertainties)
The Guidelines states that in all cases a discussion of probable uncer-
tainties must be provided, and suggests using two estimates: the best and
worst cases. NBS suggests "stating explicitly that use of 90% confidence
intervals (if it can be estimated)" is acceptable. R&F Coal quotes this part
of the Guidelines and says, "The report then proceeds to recommend specific
analytic methods which are predicated on 'worst* case data and assumptions."
Response
Given the nature of the uncertainties identified in section III.G. of
the Guidelines, there is almost no possibility of the kind of statistical
precision which would allow one to speak of 90% confidence intervals.
Inclusion of such a phrase in the Guidelines would therefore be misleading.
Users are of course welcome to use confidence intervals where feasible.
R&F Coal's comment is made primarily in the context of the impulse noise
and vibration material, where there is in fact some validity to it. The text
has been revised to make explicit the point which R&F Coal makes, and to
encourage the project proponent to conduct a best case analysis as well.
28
-------
Chapter IV
26. Comment (p. IV-1: any change in impact)
FAA asserts that the threshold levels which require a full noise analysis
are unreasonable: "... to describe any change in the impact of noise on
people...."
Response
Although there is some merit to this assertion, it is unfortunately
impossible to utilize a higher cut-off level without also requiring a full
noise analysis. That is, the alternative is to say (for example) a full noise
analysis is necessary only if more than 200 people are exposed to an increase
not less than 3 dB (or the LWP increases by 50). But in order to know whether
these limits are met, one must perform a noise analysis. FAA's comment seems
impossible to incorporate. (See also response 18.)
27. Comment (p. IV-1: level of detail)
FAA also asserts that "the level of detail required is unreasonable
(section IV)." NYSDOT says, "the methods and procedures ... are much too
complex and cumbersome for normal project development activities."
Response
Although this may be true for projects whose impact turns out to be
minor, it is felt that for projects with major impacts anything less than
is suggested in the Guidelines would not suffice.
29
-------
28. Comment (p. IV-2: noise measures: YDNL)
Fourteen different organizations commented on the assertions culminating
in the identification of YDNL, the yearly day-night average sound level, as
the primary measure to be used. None of the comments were laudatory—from 5
industrial, 2 state, and 6 federal agencies, plus 1 individual. The comments
dealt with A-weighting, I measurement problems, and yearly averaging.
NASA asserted that A-weighting is not "the best indicator of complaints
and annoyance due to low frequency noise (30-60 Hz)." Dol notes that Oregon
uses "octave band frequency requirements." Dol also suggests that the Guide-
lines are not useful for electric transmission facilities, because of the
"continuous 120 Hz low frequency hum of transformers and intermittent corona
noise from high-voltage lines." NYPSC and NS Power make similar comments.
Dixon Ward notes that "the use of A-weighted sound level and of total A-weighted
sound energy...are all quite arbitrary and challengeable." MVMA also makes the
point that the use of A-weighting is an assumption which "has not been widely
accepted," and which, according to one JASA article they cite, "is inadequate
for estimating speech interference.... MVMA does not propose a more suitable
scale..., but simply points out the inadequacy.... This example demonstrates
a need for further research."
MVMA objects to Leq/Ldn» on the grounds that "some short duration, but
highly annoying sounds" are not properly treated by them, and cites as a case
in point the EPA background document for proposed bus noise regulations.
Ford makes a similar point, saying that use of a single noise descriptor,
while commendable "... because of its simplicity, may in fact be misleading
and result in factually inaccurate noise assessments." MVMA also sent
copies of a report by Battelle Laboratories which they assert substantiates
30
-------
their position that "the energy average noise approach does not appropriately
account for the diverse effects of different noise sources." FHWA simply
states "our opposition to the Ljjn descriptor and the reasoning for our
position is well known" to ONAC. NYSDOT prefers "a critical hour approach."
With respect to measurement, both Con Ed and EEI note that there are
no consensus measurement standards for Leq and L^n, and Con Ed asserts
that publication of a document relying so heavily on such descriptors is
therefore premature. Dol also wants more information on measurement tech-
niques. Both EEI and DoD take exception to the stipulation that nighttime
levels need to be measured from 0000 to 0700 and 2200 to 2400 hours of one
calendar day.
Yearly averaging was objected to for both pragmatic and theoretical
reasons. Both United Engineers and NY Envir question how annual L
-------
than continually changing noise descriptors. NYPSC asserts that there is no
basis for assuming that people mentally average a year's noise "before deciding
whether to become annoyed—." DoD notes that the military has always used
the average "busy day" concept, and requests that this exception be written
into the Guidelines.
The final comment on L,jn was from UMTA, which expresses concern that
NEF, Leq, and CNEL received little or no treatment in the report.
Response
Data suggest that A-weighting is reasonable to use in almost all cases.
For that reason it has been widely accepted. MVMA1s second point, that
knowledge is not perfect, is a truism. One cannot, however, refrain from
action until complete knowledge is available. Certainly, some simplifica-
tions are necessary for a general document such as this, and in making them,
some sensitivity to subtleties is always lost. It is our opinion that the
advantages of simplicity outweigh the loss of potential information. Further,
if one has certain knowledge that there are special characteristics of the
proposed noise source, and that these constitute a problem above and beyond
the A-weigh ted level, one should certainly include a discussion of these
points in the assessment. The revised version makes this explicit, and the
original version recognizes it in the use of C-weighting for impulse noise.
However, there are as yet no definitive published studies documenting the
added impact in the community of pure tone (or low frequency) components.
Without such it is clearly premature to advocate the use of special proce-
dures for such noises.
32
-------
As to the problems with L
-------
Their traffic noise prediction method requires information on traffic
density. As this varies over the day, additional predictions are required.
Accurate convolution requires complete specification of exceedance levels
for individual sources. Until it can be shown that added information on
community response can be gained from this extra effort, there seems little
reason to undertake it, and to nullify all of the gains in simplification
(including uniform noise prediction approaches) which have been made in
recent years.
MVMA's comment on the use of other than L
-------
reasonable assumptions, and have been made explicit in the revision. In an
effort to reduce the problems of measurement, the revision explains yearly
day-night average sound level as being analogous to the traffic engineering
concept of annual average daily traffic. That is, they are meant to be
those levels measured under average conditions, or, if conditions vary
during the year, weighted averages of levels at the different times of
year. To reduce the proliferation of new sets of initials, the revision
uses Ldn(y) and yearly L{jn instead of Lrfny and YDNL. This is consis-
tent with the treatment of time in other metrics (e.g., Leq(8))» and with
the Acoustic Terminology Guide previously developed by this office.
With respect to problems of estimation in rural areas, there simply
is not enough known about noise levels in such areas, since measurement is
routinely conducted only in the absence of wind, rain, and other natural
sounds. Whenever possible, measurement of existing levels is recommended.
The comment by Dames and Moore is not fully understood. If the daily L
-------
DoD's use of average busy day, rather than annual average, means at most
an increase of 1-1/2 dB in the resulting value, and would appear to be
consistent with the underlying rationale. There seems no need to explicitly
acknowledge their variation, nor to prohibit it, particularly since the Guide-
lines are not to be construed as standards.
UMTA's request for a wider variety of descriptors runs counter to the overall
purpose of these Guidelines which is to present noise impacts in a way which
enhances their understandability and promotes their consideration in decisions.
The smaller the number of technical descriptors used, the more likely is this
to be accomplished.
29 . Comment (p. IV-3: other noise measures)
Both NYPSC and NASA call for further elaboration of the use of average
sound level over specific, short time periods. This need for elaboration is
also evident in HUD's question: "Why introduce SEL for individual noise
events...?" MVMA's comment is also appropriate here, on the use of other than
by EPA in the background document for the proposed bus noise regulation.
Response
This seems a reasonable request. The revision includes a clearer
statement of when the shorter term metrics are appropriate (instead of "In some
instances it is desirable..."). It also indicates where to get information so
that these metrics may be utilized in ways equivalent to the LWP approach.
30 . Comment (p. IV-4: base YDNL)
The discussion of 'base YDNL1 caused 8 comments, best summarized by that
from HUD: "We find discussion of base YDNL confusing." In particular, HUD
36
-------
questions the need for a common base among alternatives, and the requirement
for a 20 dB range of YDNL. The reliance on the Levels document to justify the
55 dB limit is objected to by both Columbia Gas and Dol (see comment 18 also).
The statement that base YDNL will never be greater than 55 dB draws 3 complaints.
DoD quotes EPA ONAC as having "stated that 'L
-------
rather Chan to introduce 55 dB, or a 20 dB range, to define the base Ldn(y)«
This would answer comments from HUD, Columbia Gas, Dol, EEI, DoD, and Dames
and Moore. It will still be emphasized in the revision, however, that a
range of noise levels needs to be considered. Twenty decibels is an example
of a good range to use. Practical feasibility, raised by DoD and CERL, may make
it advisable to relax this base level or range in many instances. Aircraft
noise prediction models, for example, are much more uncertain below an L
-------
Table B-3 from the 'Levels document1 except that the levels have been decreased
by five dB without comment." In its place they "urge use of the residual sound
level," as in ISO-R1996.
Response
The estimation procedure is derived in Population Distribution of the
United States as a Function of Outdoor Noise Level (550/9-74-009); this is
referred to in the revised Guidelines. That derivation identifies the standard
error of the estimate as roughly 4 dB. The ^10 dB mentioned in the Guidelines
is therefore somewhat misleading: the 95% confidence interval about an estimate
would be less than +6 dB. The reliability of the relationship is approximated
by the correlation coefficient of 0.723 between L
-------
levels is strongly advised for most projects. If measurement is not feasible,
then this table and its underlying equation are available. If these indicate
levels above 55 dB, and especially when levels are above 65 dB, the project
analyst needs to consider the probability of future reductions of those levels
due to noise abatement efforts. If there is such a probability, the impact
of the project noise should be calculated against these reduced future levels,
rather than against existing levels. (See the discussion of use of the screen-
ing diagram above project levels of 55 dB (L^), in the response to comment
18; and the discussion of appropriate comparisons for the single number index,
in the response to comment 51.)
32. Comment (p. IV-9: summary tables)
HUD suggests it is not really necessary to include all land areas in
the tables, just the noise sensitive uses, both existing and planned.
EEI makes a similar comment, proposing that evaluations be made only
beyond the projects' property line, since buffer zone acquisition is an
accepted way of abating noise in the community. EEI demonstrates with an
example that "a very small change in the estimated source sound power level
results in very large changes in the area between consecutive isopleths,"
—12
e.g. a 3 dB increase from 100 to 103 dB (re: 10 watts) leads to a 100%
increase in the area between the 55 and 60 dB contours. The fact that this
could lead to an over- or under-estimation of the true impact "serves to
cast doubt on the validity of the methodology."
Response
Certainly EEI's suggestion, that impacts be tabulated only beyond the
property line, is a reasonable one, and is made clear in the revised text.
40
-------
(An airport, for example, may attempt to contain the L
-------
which it will not. The second premise is unlikely to hold. Most noise limits
represent a trade-off between costs and benefits, for the general community.
They will therefore almost always be set at levels at which some adverse noise
effects may still result. The Guidelines, however, are meant to result in the
noise impacts being described in a manner which allows a decision-maker to
make those trade-offs anew for each potentially noisy project. There will be
instances, of course, where the appropriate decision-maker deems it not
worthwhile to make these trade-offs anew, and to use the existing local noise
limits. The Guidelines are merely pointing out that even where noise limits
do exist, major projects often warrant a reconsideration of the trade-offs.
34. Comment (land use categories)
HUD is concerned that the categories used do not consider local plans or
community goals, and also expresses interest at the determination that the
residential population is those "who sleep for four or more hours per day in a
residential area" (p. IV-10).
Response
These categories are offered as examples. The revised text emphasizes
use of local categories.
35. Comment (p. IV-10: special situations)
CERL notes that the separate treatment of special situations (such as
schools) can lead to excessive work, and gives the example of O'Hare Airport,
and a base DNL of 55, for which the "area is measured in one or more thousands
of square miles."
42
-------
Response
The concern seems justified (although it is probably easier to count
schools within a contour than, to estimate the total residential population).
It needs to be remembered that the treatment of special areas is offered as a
guideline for what is desirable, and that the project circumstances must
always be considered in using the Guidelines.
36. Comment (p. IV-11: residential average number)
The Guidelines, recognizing that the population varies over time, suggests
using a time-weighted average number of people for places such as churches,
parks, and stadiums, but specifies that this concept should not be used for
residential areas. United Engineers suggests that this will overestimate
impact "for residential areas during the working day."
Response
On the face of it, this seems a reasonable comment. However, when one
considers the data which lie behind the population weighting function, the
Guidelines are seen to be correct. The social survey data consist of responses
from a wide variety of people with respect to time spent at home. Thus the
data on average response have already incorporated, for residential areas,
the time-averaging which is recommended in the Guidelines for other areas.
37. Comment (Tables IV-2 and IV-3)
UMTA suggests including the category "industrial/commercial employees...
to maintain consistency" by dealing with people. Con Ed notes, and EEI
agrees, "The level of accuracy for calculating Nil as reported in [these
tables—i.e., 3 significant figures] is beyond the accuracy of the input
43
-------
data." NBS notes that Tables 2d and 3d are transposed. HUD asks why these ;
tables contain (in the definition of average population) a new day-night split
(at 0800 and 2000 hours).
Response
The inclusion of employees in the tables which summarize the impact is
a good suggestion. Con Ed and EEI are correct in limiting Nil to two digits,
as Table VI1-3 in the Guidelines indicates. NBS is correct about the trans-
posed tables. HUD's confusion stems from their inferred comparison with the
^dn split- The reason for the different time periods is explained more fully
in the revised text.
44
-------
Chapter V
38. Comment (p. V-2: CSEL for impulse noise)
Dames and Moore requests more explanation of CSEL. NBS notes that CSEL
is to be used if peak pressures "are greater than about 105 dB," and asks "is
this to be A-weighted pressure? C-weighted? unweighted?" DoD notes this
same cut-off, and says it is inconsistent with p. V-3, where 80 dB(C) is
indicated, "since the C-weighted sound exposure level for a blast is approxi-
mated by the maximum C-weighted slow level." Vibra-Tech says, "The C-weighted
network simply does not respond to much of the low frequency energy present in
air blast and which can produce a marked response in structures. ...US Bureau
of Mines has recommended the use of a linear frequency system...," which most
commercial recording systems use. NYPSC also points out that the Bureau of
Mines has recommended against C-weighting. On the other hand, CERL provides
three papers which "further support the use of C-weighting to assess sonic
booms, air blasts, and other similar impulsive sounds."
Response
More explanation of C-weighted sound exposure level (Lgg) is included in
the revision. The 105 dB is meant to be unweighted, and is consistent with the
screening level identified on page II-l. However, subsequent discussion has
led to the use of a 120 dB peak level. DoD's comment is incorrect: the
difference between the peak level (unweighted) of an impulse and the Lgc has
been empirically found to be about 20 and 30 dB for blast noise and sonic
booms, respectively. However, the confusion probably derives from DoD's
reading of the 105 dB as "maximum C-weighted slow level," since the text was
not explicit. The decision by Working Group 69 to use C-weighted L§ appears
45
-------
to have been based in large part on work by Kampermann Associates (for EPA)
and by CERL. This work is referenced in the revision, with better documenta
tion of the reasons for using
39 . Comment (p. V-3: measuring CSEL)
NBS says, "Implications of the 2 second contiguous time period require-
ment are unclear", and also questions the crest factor requirement of ANSI
SI .4, referring to IEC-79A. AMRL reports on an attempt to use the approach
identified on page V-3, which was less than successful, suggesting problems
with the method. R. W. Young questions the need for the thresholds included
in the 'Note*.
Response
The detailed discussion of measurement of impulse noise arises because
there is as yet no standard procedure for this. In general, for impact
assessment, measurement is not the problem; prediction is. For impulse noise,
however, the most effective prediction technique usually consists of measure-
ment of existing events similar to those in the proposed project. Hence,
measurement becomes an issue for the Guidelines. However, rather than being
a part of the main text, this material is included in Appendix A in the
revision.
The 2 second time period arises in an effort to specify a criterion
which could be used with unattended equipment to distinguish impulse noise
(such as blasting or sonic booms) from other high-energy noise events (such
as jet aircraft flyovers).
46
-------
40. Comment (p. V-3: CDNL)
HUD indicates confusion as to whether CSEL or CDNL should be used for
impulse noise. Young says, "My advice is to forget CDNL entirely." Both
DoD and United Engineers (UE) question the "physical basis" (UE) for adding
CDNL to DNL, "since they are two distinctly different measures" (DoD). UE
suggests that since this is an attempt to combine community reactions, the
combination should be done "during the impact-assessment stage, using non-
dimensional ratings."
Response
This portion of the text has been revised to clarify the reasoning
involved. In large part, the conversion of C-weighted sound exposure level
to C-weighted day-night sound level is based on the desire to be able to
simplify impact assessment in situations with more than one impulse noise
occurrence per day. Procedures for combining the community effects of impulse
noise and general audible noise are not spelled out in the revised version,
although it does recognize the need for these.
41. Comment (p. V-4: infrasound)
R & F Coal points out that "there is considerable confusion...on the method
of measuring mine blasts.... Because the frequency spectra of air blasts con-
tains [sic] infrasonic components it would appear that three different measure-
ments are required; 0.1 to 5 Hz, 5 Hz to 20 Hz, and C-weighted slow response."
They go on to conclude (paraphrasing a line from the preface), "With respect to
the use of explosives in surface mining the methods recommended are generally
not practical, certainly not economic, of doubtful accuracy...."
47
-------
Response
The confusion they identify is present in the Guidelines. It was the
intention of Working Group 69 that blasting be covered by CSEL measurements,
and that the infrasound material was not meant to apply in addition, but to
apply, alone, to other circumstances. This is made clear in the revised text,
48
-------
Chapter VI
42. Comment (p. VI-2: percent highly annoyed)
Ford notes, "This choice was based on the assumption that high annoyance
provides a reliable summary measure of other adverse effects of noise....
This is an assumption which has not been carefully tested in the research
literature." NBS says, "'percent highly annoyed* is really being used in a
very technical sense (e.g. the top two points of a 7 point annoyance scale)."
Consequently, they suggest "inventing a clearly technical term," or relating
this to "some estimate of just plain 'annoyed'."
Response
In the sense that there are few if any published papers testing this
assumption, Ford is correct. The statement in the Guidelines is more an asser-
tion than an assumption, however, indicating what response will be used to indi-
cate general adverse reaction. Further, the conceptual basis for this assertion
can be traced to Borsky's (1970) model of response to noise, which starts with
perception of the noise, moves to activity interruption or interference, which
results in annoyance, and may subsequently lead to complaints or other actions.
Thus annoyance is seen as the result of the direct impacts of noise, and conse-
quently a good summary indication of them. While this conceptual framework has
never been proven, it has also not been disproved or discarded. In fact, there
is some explicit support for this model in the report which Ford sent with
their comments (Hall, Birnie, and Taylor, Community Response to Road Traffic
Noise, p. 55; emphasis added).
Four main findings emerge from these results. Firstly, they
show that annoyance due to road traffic is significantly related to
the adverse impacts of the noise which an individual has experienced.
49
-------
To this extent the link between impacts and annoyance in the concep-
tual model is supported. Secondly, however, the results show that
this link is by no means a simple cause and effect relationship.
The proportion of the variation accounted for by the impacts whether
treated singly or in combination indicates that intervening factors
play an important role in determining annoyance. Thirdly, the results
indicate that reported health effects are in general more strongly
related to annoyance than are the various types of activity inter-
ference. Fourthly, the results reveal no major differences in the
strength or pattern of the relationships for reactions to road
traffic noise in general as compared with reactions to truck noise.
The NBS comment seems to arise because of inadequate explanation in the
Guidelines of the source of the generalized function (Schultz's synthesis).
First, 'highly annoyed1 is not a "technical" term. It is plain language
which is meant to be interpreted as such. In six of the 11 surveys in the
synthesis, the scales were fully labelled, such that "those people were
counted as 'highly annoyed' who said they were highly annoyed" (Schultz,
JASA, p. 380). The remaining 5 surveys did not use fully labelled scales,
hence judgment was used to equate those scales to the other 6. Second,
Schultz (1978) provides several good reasons for not using "just plain
annoyed," many of which are borne out by other publications (pp. 378-379).
43. Comment (Table VI-1: criterion levels for various land uses)
DoD says these "are practically unreasonable.... If published this way,
it will probably be used as a standard." Dol notes that the table does not
provide a category for wilderness areas, national parks, and other quiet places,
and feels they should be included. Or. Kryter suggests that it should be made
explicit that the level for school buildings is a one-hour Leq, not 24-hour,
and also feels that the 60 dB level is too high. NASA asserts, "It may be
questionable whether structural attenuation can be adequately described with
three discrete values," and raises several other questions about attenuation and
50
-------
masking of outdoor noise as received indoors, since most of these levels are
rationalized on the grounds that "the noise-sensitive activity in question is
usually indoors" (Guidelines, p. VI-3).
Response
It is doubtful that this table belongs in the Guidelines. Dr. von Gierke's
reply (6 Dec 77) to the DoD comment indicates, "In some respects this table
is a carry over from earlier versions.... In the final guidelines, this table
could easily be eliminated." In fact, this table would appear to contradict
the rationale for the screening diagram (comment 18). The table will be
deleted from the revised Guidelines. The revised Guidelines will refer the
reader to appropriate material in other documents.
44. Comment (Table VI-2, 3, and 4: summaries of human effects)
The fact that the basic information in these tables is drawn from the
Levels document bothers Dol, on the grounds that the Levels analysis included
a margin of safety. Dr. Kryter raises two objections (in correspondence with
Dr. von Gierke): "(1) the omission, without better justification, of the
benchmark function showing that 17% of the people are highly annoyed from
aircraft noise at an Ldn of 55; and (2) ...the failure to use as a criterion
measure, at a minimum, an hourly Leq related to specified noise-speech areas
in a classroom or lecture hall...." MVMA, on the other hand, wishes it to be
made very clear in the guidelines that "(1) they reflect essentially what was
concluded previously in the Criteria and Levels Documents of 1973 and 1974,
(2) little subsequent work has been done to substantiate those documents (see
'Guidelines' reference list), and (3) new information may demonstrate that
those Documents and the Guidelines may not be correct."
51
-------
Response
These tables do contain new data, and Dr. Kryter's suggestion is certainly
reasonable. A fuller explanation is provided in the revision. Further, the
material needs to be available in more detail if it is to be easily usable.
Kryter's concern about schools is a case in point. If we expect relationships
such as these to be used to describe impacts on schools, the relationship
needs to be available in more detail than 10 dB intervals. When this detail
is provided, it is clear that MVMA's comments are largely incorrect. Sub-
stantial new work on the effects of noise on residential communities has
been done in the past five years, and is incorporated in these Guidelines.
Dol's concern about the use of the Levels document is a legitimate one. It
is made clear in the revised version that what is being used is the data
summary of Appendix D, not the policy material of identified levels.
Kryter's comment about hourly Leq is further reason to expand the
discussion of the short-term metrics (comment 29), and to expand the material
in paragraph 4 of page VI-2, describing how to treat non-residential urban
land uses.
45. Comment (p. VI-8: response independent of source)
The Guidelines state that Schultz's data confirm the assumption that
the relationship between annoyance and noise is independent of noise source.
Ford notes, "This is a strong claim considering that the Schultz synthesis
involves only transportation noise surveys," a concern which is echoed by
United Engineers. Ford's position is summarized thus: "What has to be
questioned is whether our present state of knowledge about the relationships
between exposure to noise and human response supports the reliance upon a
single weighting function and specifically the one described in the Guidelines."
52
-------
Ford submitted a copy of a report from McMaster University, which indicates
lower percentages highly annoyed by road traffic than Schultz's curve would
suggest. From this, Ford draws the implication that, "because most of the
studies defining the Schultz curve are aircraft related, his average curve
does not accurately represent response to road traffic noise." In addition,
they cite several studies which show differences in response to road traffic
and aircraft noise, and point out that differences in the wording of the
questions reviewed by Schultz "serve to complicate further a synthesis of the
results." MVMA asserts that the logic in the Criteria document, "heavily
biased to aircraft exposure," is "carried into the 'Guidelines'." On the
other hand, Dr. Kryter says that "serious, substantial errors appear to have
been made . . . with respect to the effects of aircraft noise." Both Kryter
and Ford suggest that more than one function or set of criteria may be
required. Ford also suggests that the Guidelines should be "updated on a
semi-annual basis so that it is a truly living document."
Response
As noted earlier (comment 28), one cannot wait for complete information
before attempting to assess the environmental impacts of noise. At this
point in time there are no published studies which quantify any differences
in response to different noise sources, nor are there any studies relating
average response to noise level for non-transportation sources. (Dol states
that a study is in progress on electric power transmission line noise, but
its results are not yet available.) In the absence of any solid information
to the contrary, the Guidelines have adopted a single relationship for all
noise sources. If information becomes available which can be used to identify
a different relationship for certain sources, that should certainly be used.
53
-------
In that respect the Guidelines are open to revision as new information becomes
available, although that may not occur on a semi-annual basis as Ford suggests.
On the specific points raised, responses are as follows. In a more
recent paper, presented at the International Congress on the Biological
Effects of Noise, in Freiburg, Germany, in October 1978, Schultz constructed
separate curves for the 4 road traffic studies, the 6 aircraft surveys, and
the rail study in his synthesis. His conclusion is, "The road and air
traffic results are extremely close; while the railway data are perceptibly
lower than the other two, one cannot claim a significant difference." The
importance of the differences in wording of questions, noted by Ford, is
possibly responsible for differences between the McMaster results and Schultz's
synthesis. McMaster asked simply for a rating of the noise; many of the others
asked how annoyed the respondent was. Social survey experience suggests that,
relatively, a higher degree of response will be elicited by the latter ques-
tion. Also, Schultz has suggested that the difference arises because the
McMaster study uses a bipolar scale whereas most others have used simply an
annoyance scale (JASA, 1978, p. 403).
MVMA's assertion that the Guidelines are simply a repeat of data in
the Criteria document is obviously incorrect: eleven of the thirteen studies
in Schultz's synthesis were not included in the Criteria document. Kryter's
comment arises from the fact that the Guidelines do not explain Schultz's
synthesis well. The revised Guidelines are explicit about the reasons for
rejecting the earlier curve.
46. Comment (p. VI-11: Degradation of environment)
DoI would like some guidelines for assessing as a special case the
noise impacts where no people are living, such as wild lands. NBS remarks,
54
-------
"1C doesn't seem appropriate to use [data based on daily residential living]
to quantify impact on areas 'to which urban dwellers wish to go for an escape
from city noise....1" United Engineers makes a similar comment: "How can
you compute a percent population highly annoyed in an area where no people are
living?" NYPSC says this approach is inadequate, since "almost by definition,
'degradation1 must consider the extent to which the ambient sound level is
changed."
Response
These comments are all cogent. (See comments 3 and 18 also.) For special
areas valued for their quiet, a tabulation of the impacts and a qualitative
description are recommended. The tables would be similar to those used for
populated areas, but rather than focusing on number of people subjected to
various noise levels, the tables would focus on the size of the affected area.
Sensitive uses, such as for particular animal species, or for wilderness
recreation, could be treated the same way special situations are in the
populated areas. This proposed approach has the advantage of being able to
identify impacts even when there is no permanent population. In some cir-
cumstances, it may be useful to reduce this tabulation to a single number,
in which case the population weighting function is the best available indi-
cator of relative impact, and can be used to calculate a "level-weighted area."
If data become available which describe response as a function of noise in such
areas, the Guidelines will be revised accordingly.
47. Comment (p. VI-12: Effects of impulse noise)
R & F Coal questions the use of sonic boom data to estimate annoyance
for surface mine blasting. "Because people complain about military aircraft
55
-------
flying over their homes at supersonic speed it does not follow that they will
complain about a taxpaying, local employer, 'who is producing vital basic
materials just because the decibel level is the same."
Response
Data specific to mining blasts are not available. In the revised
version, recognition is given to the fact that estimates of C-weighted sound
exposure level (Lgc) from peak pressure are different for boom and blast
noises. As a result, booms are seen to cause adverse response at lower peak
sound pressure levels than blasts, for the same Lgc, as suggested by R & F
Coal.
48. Comment (p. VI-15: infrasound)
EEI recommends inclusion of "cut-off criteria for vibration, structure-
borne noise, ultrasound and infrasound." United Engineers notes that "the
fact that infrasound is sometimes related to structural vibration was not
even ment ioned."
Response
If EEI's comment is intended to point out the absence of a screening
diagram for these considerations, it is well taken. The numbers in the right-
hand column of Table 1-3, for impulse noise and vibration, are such cut-off
levels, which is stated explicitly in the revised version. United Engineers'
comment is correct, and has been noted in the revision.
56
-------
49. Comment (Fig. VI-6 and Table VI-5: Vibration criteria)
R & F Coal points out that "the annoyance curve suggests that people
will be more annoyed by numerous small blasts than by one blast properly
designed to do the task, and this preference is compatible with our exper-
ience." They then say that the Guidelines attempt to reduce both the charge
weights and the number of impulses per day, which can only be done by mining
less material, and that is not in the national interest. Ky Mines asserts
that acceleration measurements "do not correlate effectively with damage
levels," and that the level specified for vibration is "unrealistically
low."
Response
The comment by R & F Coal provides additional empirical confirmation of
the figure. The inference drawn from it, however, is a good example of a
misreading and abuse of the Guidelines. The Guidelines exist to assist one
in quantifying impacts. The question of trading off those impacts against
other concerns is not addressed in the Guidelines. For coal extraction,
energy concerns may outweigh vibration impacts. The important point is that
the impacts are identified, and are identified properly. The correlation of
acceleration with damage is irrelevant: The Guidelines bases the vibration
criteria on human response, as both the text and Figure VI-6 make clear.
For that reason, the levels undoubtedly are lower than those based only on
damage criteria.
57
-------
50. Comment; (p VI-21: blast noise)
R & F Coal notes that the Guidelines describe the formulas given for
blast size against safe distance as "newly proposed and somewhat tentative,"
and ask, therefore, why they should be used. R & F Coal also notes that
these are calculated for worst case conditions, and incorporate numerous
safety factors, such that no allowance is made for the careful surface mine
operator. They also claim the resulting charge weights are unduly restric-
tive, and are about one-fifth of what their experience indicates is safe and
reasonable. They question whether charge weight per delay or total explosives
in the blast is meant in the formula. DoD also says the formula is overly
conservative, and suggests 250^"w~as the distance beyond which there is
almost certain to be no glass breakage.
Response
These somewhat tentative guidelines for blasting are offered simply because
there are no other generally accepted, simplified guidelines. However, if
the blasting engineer is skilled and careful enough to be able to make more
detailed calculations and to (e.g.) wait for wind changes, then other formulas
can certainly be used. That is in no way contrary to the intent of the Guide-
lines, which attempt to provide simplified procedures for identifying
potential problems. (If this equation shows no potential impacts, for example,
there is no need to go into the more detailed analyses by a blasting engineer.)
The factor of five noted by R & F Coal is probably due in large part to the
safety factors built into the Guidelines equations. The equations should
specify charge weight per delay, an editorial omission which will be corrected.
The DoD formula appears quite consistent with the equation in the Guidelines,
once one corrects for different units (feet and pounds for DoD, kilometers
and kilograms for the Guidelines).
58
-------
Chapter VII .
51. Comment (p. VII-1: Single number impact index)
HUD disagrees that it is desirable to derive a single number which
represents the integrated impact. "[T]he approach is not necessary for an
adequate assessment of impact." FHWA made similar comments. NYSDOT suggests
that tabulating "the number of persons/residences at various levels" would
suffice. DoD also concurs (emphasis in original): "[N]o criteria are given
to evaluate the results of the calculations. How will they be used in the
decision-making process?" "Until it is tested..., we cannot concur in its
use."
Response
From the perspective of standards or limits (which HUD and FHWA both
employ), it is probably true that the notion of a single number impact index
offers little extra information. For such uses, the number of people within
each of the contour bands may be a reasonable end-point of the analysis.
However, for full and thorough environmental assessment, which these guide-
lines are written for, noise impacts must be compared with, and traded-off
against a host of other decision criteria. In those circumstances, a single
number index of noise impacts is obviously preferred over a full table of
numbers.
In reponse to DoD's comment, the revised version is more explicit about
how to use the calculated results. In particular, the revision suggests
which of the single-number indexes should be emphasized in any given situa-
tion, and suggests bases for comparison of the numbers. For clarity, the
different situations are identified on a reproduction of the screening
diagram (comment 18).
59
-------
52. Comment (p. VII-2: Level weighted population)
Dames and Moore note that references or discussions are needed to
justify the assumptions stated, because they are open to question and criticism.
NASA says, "the assumptions underlying Sound Level Weighted Population are
critical to the guidelines and should be demonstrated experimentally." FAA
says the paragraph in question "is completely unreasonable." Noise criteria
should protect people regardless of population density. HUD says the second
assumption "is absurd". EEI, in earlier correspondence on the Fractional
Impact Method (15 March 76), says the assumption is wrong, and while there
"is probably no reliable relationship..., non-linear values...make more sense
than linear values." More recent correspondence (29 Nov. 77) adds to this,
saying the assumption has no verification in the technical literature, and
that the assumption "implies a linear subjective response to noise and would
require a linear sound weighting function." Con Ed echoes EEl's concern for
"verification" of this "incorrect" assumption in the technical literature.
Ky Mines says this is "totally inconsistent with human rights, in that
according to the guidelines a group is entitled to more peace and quiet than
an individual." Columbia Gas says this assumption contradicts "the statement
on page B-2 that the rate of change of annoyances with sound level is greater
at high sound levels than at lower sound levels." FHWA points out that LWP
"is a number whose absolute value is meaningless. One of the serious defi-
ciencies of this report is how to explain to the layperson, in terms he will
understand, what the noise impact will be."
Response
In fact, FHWA has identified the fundamental problem with LWP, which is
behind all of these comments. In the Guidelines, it is explained in terms
60
-------
which make it poorly understood. This probably arises because it is explained
as a carryover from and minor modification to the earlier Fractional Impact
Method, which did in fact rest on some critical assumptions. The following
alternate explanation should overcome all of the comments, as will be shown.
The single number impact index under discussion here applies only to
general audible noise in populated areas, and is calculated in terms of the
previously identified 'general adverse impact1. In particular, it is based
on the use of a function (Schultz, JASA 1978) expressing percentage of people
highly annoyed as a function of outdoor residential Ldn. Clearly, if 4% of
people are highly annoyed at 55 dB, and 37% at 75 dB, one can identify the
total number of people exposed to each noise level, multiply by the appro-
priate percentage, and add over all noise levels to identify the total number
of people highly annoyed. This number is one which is easily explained
to laypeople because it has intrinsic and intuitive meaning.
If the single number is explained in this fashion, there is much less
chance of the kind of mis-interpretation both FAA and Ky Mines make.
Schultz's work provides the empirical demonstration requested by several. It
is obvious that linear or non-linear functions are equally possible, and that
this one is identically that discussed in Appendix B.
53. Comment (p. VII-4: PLH and other abbreviations)
NBS notes, "PLH and PHL are used interchangeably. Need to pick one and
stick with it." United Engineers concur. HUD says, "The non-technical
reader is introduced to such abbreviations as FI, PHL, LWP, Nil, DNL, YDNL,
CSEL, CDNL, TAVL, and SEL, not to mention all their derivations; these may
add to the mystique of the product, but leave much to be desired in a practical
guidance document."
61
-------
Response
Certainly when abbreviations are used, they should be used consistently.
The revision will be carefully edited. But HUD's point is well taken. Even
careful editing may not improve communication that much. Perhaps abbreviations
are needed for use in equations, but an excellent argument can be made for
writing the text in full, and for minimizing new jargon in Guidelines such as
these. The revised version uses the symbols set out in the EPA acoustic
terminology guide.
54. Comment (p. VII-4: Population weighted loss of hearing)
NBS asks why the denominator of PLH includes only those exposed to L
-------
55. Comment (p. VII-5: Normalizing weighting function)
Both Ford and MVMA question "the apparent 'scaling up1 of the Schultz
'percent highly annoyed1 curve in the Guidelines by a factor of approxi-
mately 3" (Ford).
Response
The revision uses Schultz*s curve directly. (See comment 52).
56. Comment (p. VII-5 and Fig. VII-7: Sound level weighting function)
Wyle points out that, with the proposed function, impacts could be
theoretically unbounded, since this function falls less rapidly than do noise
levels. NBS suggests that the mathematical expression of this function can
be simplified. Both Wyle and EEI suggest that the function should be zero
below roughly 55 L
-------
annoyed), and (c) in all eleven synthesized surveys, there is only 1 data
point below 45 dB (for which % HA is 0 (Fig. 36).
The effect of background noise remains a problem, although not as large
a one as NYPSC suggests. More data (Schultz's) are now available than were
when the Levels Document was compiled. The arbitrariness of the curvilinear
function is refuted by the fact that it is derived from data, and that others
besides Schultz have found non-linear relationships for many responses of
interest (e.g. Hall, Taylor & Birnie, in the report submitted by Ford).
However, as the Guidelines point out, this curve differs little from the
straight line used earlier by EPA to indicate relative impact. The re-
striction to outdoor environments noted by Wyle will be made explicit in
the revision.
57. Comment (p. VII-8: Weighting function for loss of hearing)
NBS says '"loss per ear* implies that one should multiply by the number
of ears... Suggest using 'average dB loss per person'".
Response
Agreed.
58. Comment (p. VII-9: Change in LWP)
NBS: "Percent change should be change in sound level weighted popula-
tion, divided by LWP before the change." United Engineers notes that as
defined this is a ratio, not a percentage.
Response
Agreed.
64
-------
59. Comment (p. VII-12: temporary noise environments)
Dames and Moore asserts, "YDNL should not be used for temporary noise
environments," because the yearly average may not change much for medium or
small projects. NYPSC says, "Actual impacts can be substantially greater than
would be predicted using an annual L^n," and implies that a better approach
to assess the impacts is to estimate them as if the project were permanent.
Response
The Guidelines specifically suggests using both the actual noise during
the project and the yearly average, in order that the decision-maker have
available the most appropriate information. It is instructive to note (using
the example on page VII-13) that impact calculated from the yearly day-night
sound level (LWP=601) will almost always be greater than impact for the year
calculated from a weighted average of project and non-project noise levels
(LWP=557, from 3/4 yr. of 664 and 1/4 yr. of 236).
60. Comment (p. VII-15: vibration impact index)
NBS: "Line 4 should read 'total number of people* (not residences).
Again, however, only the number of people in areas of significant vibration
are included in the denominator." (See comment 57.) VWP is Vibration-
weighted population.
Response
Agreed.
65
-------
Chapter VIII
61. Comment (p. VIII-1: noise re other impacts)
DMTA says the summary "could have been enhanced by discussing a mechanism
for evaluating noise impacts relative to other environmental impacts."
Response
In line with the revised context and title of the Guidelines, a discus-
sion of comparative evaluation of impacts would be desirable here. However, it
is not the task of these Guidelines to suggest mechanisms for making such
comparisons.
62. Comment (p. VIII-1: structures exposed)
EEI request specifying a lower limit below which there is no need to
discuss noise induced structural damage.
Response
It is made clear in the revised version that this consideration applies
only to impulse noise, infrasound, and ultrasound, not to general audible
noise. (For levels to use, see comment 48.)
63. Comment (p. VIII-2: sound level weighted area)
United Engineers notes that this concept was never defined.
Response
We agree that the summary is not the place to introduce new concepts.
This one is discussed earlier in the revised version. (See comment 18). As
defined, it may prove useful in two situations: to reduce tabulations of
unpopulated areas affected by noise; and for cases where it is known that an
area will be developed, but exact population projections cannot be forecast.
66
-------
64. Comment (p. VIII-3: descriptive qualitative evaluations)
HUD requests "more guidance" on doing descriptive evaluations, "e.g.
examples of questionnaires, where to obtain complaint listings, etc."
Response
While more guidance might be useful, it is not clear what can be
offered in a short section which would be generally applicable.
65. Comment (p. VIII-5: new population exposed)
NBS indicates this paragraph is not clear. "Is not the 'changes in the
existing environment...1 to be obtained from Table IV-1? Then what is the
meaning of comparing to the predictions of Table IV-1?" Further, NBS takes
issue with the assumption that the new population comes from an area of
Nil .35, suggesting some discussion of likely previous residential area be
required.
Response
The revised version clarifies this paragraph in line with the following
points. For projects which entail new populations in an area where noise is
otherwise unchanged, two possibilities arise. The first is that the existing
noise levels are greater than the levels which can be expected as a result
of the population cluster (Table IV-1); the second is that population noise
will be greater than the existing. In the first case, noise impacts are
obvious, and the site plan will probably include ways to reduce that impact.
In the second case, it needs to be pointed out that there are noise impacts,
and that, again, the site plan should look to ways to reduce them. Table
IV-1 can be used to show the expected sound level for the population density
67
-------
in question, and that can be compared with the levels predicted on the basis
of the thoughtful site plan.
On the basis of a somewhat cursory knowledge of the residential migra-
tion literature, we find it unlikely that one could say much useful about
the likely previous residences of a new housing development. The value
of Nil = 0.35 is based on the calculation in Table VII-3, and is intended to
represent the national average value. It is explained more clearly in the
revised Guidelines.
68
-------
Appendix A
66. Comment (Terminology)
NBS has offered suggested improvements to 13 of the terms in part 1.
United Engineers has also made a number of comments on the definitions, as has
DoD. Isolated points are raised by FAA and NASA. NBS asks what happened to
section 2 through 4 of this Appendix.
Response
Each suggested revision will be checked against EPA terminology and other
normal usages.
Appendix B
67. Comment (exponential functions)
NBS asks: "Why is equation B-6 better than B-5?"
Response
The Working Group apparently preferred the two exponential functions for
two reasons. First, at higher levels, this function "has the same rate of
growth as a loudness function" (page B-5). Second, the asymptotic behavior of
the function at very low noise levels provided a way to quantify environmental
degradation in such areas. The revised version retains equation B-6 since it
was indistinguishable from equation B-5 in the range of the (Schultz) data.
However, its use is discouraged (as noted in comment 56) below levels of 45
dB (Ldn).
69
-------
68. Comment (p. B-8: Population weighted hearing loss)
United Engineers makes several comments regarding calculation of pop-
ulation-weighted hearing loss. "[T]he time dependence should be kept in
parametric form, rather than averaged out. This is especially important if
impacts are to be assessed in 5-year increments." The assessment should be
"of NIPTS due to the new source alone.. .Also, the assessment was to be for a
20 year time span, not 40 years."
Response
Data are not available to assess NIPTS in 5-year increments, so there
is no point to leaving in the time dependence. Assessment of NIPTS for the
new source only would not be adequate, as the resulting level may be higher
than either the source or existing levels. Comparison of the resulting
level's NIPTS with that of the existing level might be reasonable. It is
true that the assessment covers at most a 20 year span, but the average used
here in fact approximates median response after 20 years.
Appendix C
69. Comment (ISO/TC 108/SC 2/WG 3)
Vibra-Tech Engineers main concern "is the inclusion of arbitrary levels
of vibration alleged to cause damage." The ISO draft standard, "the basis for
this criteria," "...was so out of step with more widely accepted practical
research done in the U.S.A. and other countries it was never adopted as an I.S.O,
standard by any member country."
70
-------
Response
The Guidelines do not adopt the ISO recommendation, either. Section
4.5 of Appendix C specifically states, "reducing the threshold from 50 mm/sec
to 5 mm/sec does not appear warranted." The data used to support a reduction
of 50% are given in Fig. C-4, from U.S. Bureau of Mines. (See next comment
also.)
70. Comment (Fig. C-4)
R & F Coal says, "This is a graph from U.S. Bureau of Mines Bulletin
656 which contains 4 points below the 2 ips safe level. The study group
is in error. The four minor damage points are not the result of blast damage.
They are from some experiments where eccentric motors were set up in doorways
and inside houses in order to induce damage. They should not be included at
all." Hence the 50% reduction of the commonly used 2 ips safe level is
uncalled for.
Response
R & F Coal is correct that the four points were from eccentric motor exper-
iments. R & F Coal do not mention that all 160 of the data points were from
that experiment. To throw out those data is to remove most of the informa-
tion which established the 2 ips level. Further, a close inspection of
Figures 3.1-3.4 of Bureau of Mines Bulletin 656 shows that careless drafting
left out of Fig. 3.4 a minor damage point at 40 cps and .0015 inches which
was in Fig. 3.1 and would not be covered by the proposed criterion.
71
-------
-------
7f
&*
*4 3_£
•• f •
^_p
Bi
Jgflt^Kcn'cj
if-
FL
If *( ..€«te«V»( ^ rff^'ffi
N JL A/jtee. Jjrywah_/«^« _%. 17
-------
r. k*« «g*3
It
^7
------- |