Draft Industrial Noise Guideline
Technical Background Paper
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
© 2015 Copyright State of NSW and Environment Protection Authority
Disclaimer
EPA has compiled this guideline in good faith, exercising all due care and attention.
No representation is made about the accuracy, completeness or suitability of the
information in this publication for any particular purpose. EPA shall not be liable for
any damage which may occur to any person or organisation taking action or not on
the basis of this publication. Readers should seek appropriate advice when applying
the information to their specific needs.
Published by:
NSW Environment Protection Authority
59 Goulburn Street, Sydney NSW 2000
PO Box A290, Sydney South NSW 1232
Report pollution and environmental incidents
Environment Line: 131 555 (NSW only) or [email protected]
See also www.epa.nsw.gov.au
Phone: +61 2 9995 5000 (switchboard)
Phone: 131 555 (NSW only – environment information and publications requests)
Fax: +61 2 9995 5999
TTY users, phone: 133 677, then ask for 131 555
Email: [email protected]
Website: www.environment.nsw.gov.au
ISBN 978 1 74359 957 0
EPA 2015/0207
May 2015
ii
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
Contents
1
Introduction ................................................................................................... 1
2
Background to the Industrial Noise Policy ....................................................... 1
3
Scope of the review ........................................................................................ 1
4
Key issues....................................................................................................... 2
4.1 Retaining the dual criteria approach................................................................... 2
4.1.1 Intrusiveness criterion ..................................................................................... 2
4.1.2 Amenity criterion ............................................................................................. 2
4.1.3 A dual criteria approach .................................................................................. 2
4.2 Assumed minimum background noise level of 30 dB(A) (refer Industrial Noise
Policy, Section 3.1.2) ......................................................................................... 3
4.3 Amenity criteria ................................................................................................. 5
4.3.1 Adequacy of existing ‘acceptable noise levels’................................................ 5
4.3.2 Deriving the amenity criterion ......................................................................... 6
4.3.3 Amenity approach for clusters of industry ...................................................... 8
4.3.4 Precinct approach ............................................................................................ 8
4.4 Meteorological conditions ................................................................................. 9
4.4.1 How to define the presence and strength of a temperature inversion ............... 9
4.4.2 How to directly measure temperature gradients (inversion strength) .............. 10
4.4.3 Meteorological conditions used in environmental assessment documents
(i.e. documents accompanying applications) ................................................ 10
4.4.4 Meteorological conditions used at development planning stage versus
meteorological conditions included in consent and licence conditions ....... 12
4.5 Identifying non-compliances (sustained non-compliance) ................................. 13
Non-compliances with noise conditions .................................................................. 13
Managing non-compliances ..................................................................................... 13
When is a development in breach of a noise condition?......................................... 13
4.6 Modifying factor adjustments .......................................................................... 14
4.6.1 Tonality .......................................................................................................... 14
4.6.2 Low frequency noise ...................................................................................... 14
4.6.3 Impulsive noise .............................................................................................. 21
4.6.4 Duration ......................................................................................................... 21
4.6.5 Intermittent noise and maximum adjustments............................................. 22
4.7 Sleep disturbance ............................................................................................ 22
4.8 Land acquisition and home acoustic treatments ............................................... 24
4.9 Relationship between project-specific noise levels and statutory limits ............. 26
4.10 Acceptable noise levels do not equate to inaudibility........................................ 26
4.11 Incorporating where relevant the existing NSW application notes..................... 26
5
References ....................................................................................................28
Attachment 1 – Original justification for the minimum assumed background
level in INP, 2000 .................................................................................................30
EPA criteria............................................................................................................. 30
Minimum background............................................................................................. 30
Community response .............................................................................................. 31
References ............................................................................................................. 31
Attachment 2: Analysis of proposed amenity noise level approach .......................32
www.epa.nsw.gov.au
iii
Draft Industrial Noise Guideline Technical Background Paper
1 Introduction
A review of the Industrial Noise Policy (INP) has been undertaken and has resulted in
the preparation of a revised draft Industrial Noise Guideline (ING). This technical
background paper identifies issues with the INP that have been raised over time by
stakeholders, together with potential solutions which have been developed and
incorporated into the draft ING. The issues and solutions are not set in stone and are
open for discussion, exclusion, amendment, etcetera as part of the draft ING
consultation process.
2 Background to the Industrial Noise Policy
The NSW Industrial Noise Policy (INP) is a whole-of-government policy adopted in
2000 for the assessment of noise from industrial premises.
The scope of the policy is identified in Section 1.3 of the INP as follows:
‘The policy is specifically aimed at assessing noise from industrial premises scheduled
under the Protection of the Environment Operations Act 1997. It is used as a guide by
Environment Protection Authority (EPA) officers for setting statutory limits in licences
for these sources. Local government is an independent regulator for noise under the
legislation, and thus has discretion in dealing with noise within its area of
responsibility. The policy is designed for large and complex industrial sources and
specifies substantial monitoring and assessment procedures that may not always be
applicable to the types of sources councils need to address. However, local
government may find the policy helpful in the carrying out of its land-use planning
responsibilities (for example, the setting of targets in local and regional environmental
plans).’
Figures 1.1 to 1.6 in the INP effectively summarise the existing policy framework. It is
recommended that these flow charts be consulted before further considering the
contents of this technical background paper. It is essential that the existing policy
framework is well understood before considering policy shortcomings and possible
improvements. The INP can be viewed/downloaded at:
www.environment.nsw.gov.au/noise/industrial.htm.
Since the introduction of the INP in 2000, it has been augmented with ‘policy
application notes’. These application notes were developed to assist industry and
acoustical consultants produce noise impact assessments, apply the provisions of the
INP and reduce processing time. The application notes clarify and augment the
existing provisions in the INP. They have been published on the EPA’s website and
are available at: www.environment.nsw.gov.au/noise/applicnotesindustnoise.htm.
3 Scope of the review
The review of the Industrial Noise Policy (INP) has focussed on addressing key issues
identified by stakeholders during more than a decade of application of the policy. The
issues raised have not suggested that the policy is fundamentally flawed or requires
wholesale change, rather, they indicate that clarification and increased certainty of
policy requirements is needed.
However, the identification of key issues with the existing policy and potential
solutions for these issues should not be construed as indicating that a fundamental
change in the policy framework is outside the scope of the review. The draft Industrial
Noise Guideline was considered the best way to convey potential solutions to issues
with the INP that have been raised over time.
www.epa.nsw.gov.au
1
Draft Industrial Noise Guideline Technical Background Paper
4 Key issues
The following issues have been identified as potential ‘key issues’ that need to be
considered and addressed as part of the review process. They have not been
presented in order of priority as that would be largely a matter of perspective and
would vary between stakeholders.
4.1 Retaining the dual criteria approach
The current approach in the INP is to derive two criteria, the intrusiveness and
amenity criteria. The more stringent of the two criteria becomes the project specific
noise levels (PSNL). The PSNL provides the principal means to assess impacts and
evaluate the need for, and effectiveness of, noise mitigation measures.
4.1.1 Intrusiveness criterion
The intrusiveness criterion seeks to protect against significant changes in existing
noise levels and to ensure that a single noise source is not intrusive. The intrusiveness
criterion is based on the prevailing background noise level plus 5 decibel (A-weighted)
(dB(A)). The background noise is quantified as the rating background noise level
(RBL), which is a statistical processing of long-term LA90,15minute levels over day, evening
and night periods designed to ensure that intrusive criteria are relevant for 90 per cent
of the time. Typically one week of background data is required to establish the RBL.
The intrusiveness criterion is expressed as an LAeq,15minute level.
4.1.2 Amenity criterion
The amenity criterion seeks to maintain long-term total industrial noise at acceptable
levels. These levels are outlined in Table 2.1 of the INP and are referred to as
recommended acceptable noise levels (ANLs) and maximum noise levels (MNLs). A
common misconception is that the amenity criteria are the ANLs; however, this is not
always the case.
The ANLs are the cumulative goals for total industrial noise from all industrial sources,
whereas the amenity criterion is the amount of noise that an individual industrial use
can contribute to the total industrial noise at a receiver location. The higher the
existing industrial noise in an area is, the more stringent the amenity criterion will be
for a new industrial noise source.
Typically, the intrusiveness criterion will inform the PSNL in areas with little industrial
noise, whereas the amenity criterion will inform the PSNL as industrial noise in an
area grows. The amenity criterion seeks to cap the growth of industrial noise.
4.1.3 A dual criteria approach
The intrusiveness criterion seeks to protect against significant changes or intrusion
(unwanted noticeability) of an individual noise source (premises) above pre-existing
levels. The degree of change in noise levels in an area has been identified in the
scientific literature as a factor in how a community will respond to noise as opposed to
simply the absolute level of the noise2,7,11,13.
The amenity criterion seeks to achieve a long-term balance between industrial
noise and a satisfactory level of amenity for communities. The ANLs have been
established on the basis of annoyance dose response relationships from socioacoustic surveys of communities exposed to industrial noise. These are sometimes
described as steady state studies. The adequacy of the actual ANLs will be
evaluated later in this technical paper.
2
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
Van Kamp and Brown, 200323 conducted a review of the literature regarding how
communities will respond to changes in noise level. They concluded that:
‘All available studies demonstrate, with remarkable consistency, an excess response in
situations of both increments and decrements of noise exposure: respondents whose
noise exposure has increased report more annoyance than expected from steady-state
studies; respondents whose noise exposure has decreased report less annoyance than
expected from steady state studies. The effect is present even for quite small changes in
noise exposure.’
Based on the current literature it is clear that the degree of change in noise levels is
an important factor to consider along with total noise exposure when considering how
a community will respond to a new industrial noise source.
Proposal 1
The dual criteria approach remains relevant for noise assessment purposes and has
been maintained in the revised draft ING.
4.2 Assumed minimum background noise level of 30 dB(A)
(refer Industrial Noise Policy, Section 3.1.2)
The INP states that: ‘Where the rating background level is found to be less than 30
dB(A), then it is set to 30 dB(A)’. The consequence of this is that the minimum
intrusive noise criterion, which is set at background noise plus 5 dB(A), is LAeq,15minutes
35 dB(A). The original basis for the minimum assumed background is outlined in
Attachment 1. The rationale is that the vast majority of the community will be
protected from unacceptable impacts at a level of LAeq,15minutes 35 dB(A).
The approach of applying a minimum background noise level has drawn criticism
based on two markedly different viewpoints. The first is that the criterion is too high for
areas that have low background noise levels and the intrusive criterion should be set
at actual background plus 5 dB(A). This could result in an intrusive criterion as low as
LAeq,15minutes 25 dB(A). The alternative view is that the criterion is too stringent,
especially when applied for daytime hours, and is well below the level at which
unacceptable impacts are established based on annoyance dose response
relationships for industrial noise15.
Noise criteria in NSW are normally based on seeking to protect 90 per cent of an
exposed population from being highly annoyed17,19,20 and in the case of industrial noise
to also control the intrusiveness of noise from individual premises. The annoyance of a
noise source is largely dependent on the type of noise, for example industrial noise has
been shown to be more annoying than transportation noise15. Based on dose response
relationships for industrial noise postulated by Miedema & Voss (2004)15, an industrial
noise level of 35/35/35 dB(A) day/evening/night (i.e. the minimum intrusiveness
criterion) equates to 1.5 per cent of a community being highly annoyed.
This is of course a steady state relationship; that is, how a community would respond
to this level of noise in the long term. It does not reflect how a community would
respond to a newly exposed source of noise. Indeed, ISO 1996-1:2003 Acoustics –
Description, measurement and assessment of environmental noise – Part 1: Basic
quantities and assessment procedures states:
‘Equation D.1 [Schultz dose response curve] is only applicable in existing situation; in
newly created situations, especially when the community is not familiar with the sound
source in question, higher community annoyance can be expected. This difference may
be equivalent to up to 5 dB; research has shown that there is a greater expectation for
and value placed on ‘peace and quiet’ in quiet rural settings. In quiet rural areas, this
greater expectation for peace and quiet may be equivalent to up to 10 dB. The above
two factors are additive. A new, unfamiliar sound source sited in a quiet rural area can
www.epa.nsw.gov.au
3
Draft Industrial Noise Guideline Technical Background Paper
engender much greater annoyance levels than are normally estimated by relationships
like equation (D.1). This increase in annoyance may be equivalent to adding up to 15 dB
to the measured or predicted levels.’
The 10 per cent highly annoyed level from Miedema and Voss, 200415 for industrial
noise is approximately a day-evening-night level (DENL) of 56 dB(A). The DENL is a
composite whole day rating level with a 5 dB(A) and 10 dB(A) penalty (addition)
respectively for the evening and night-time period. This DENL level therefore could
approximate levels of LAeq,day,evening,night 56/51/46 dB(A). It is generally accepted that an
LAeq,15 minute is approximately equal to the LAeq, period + 2 dB. If we apply the above
principles from ISO 1996.1-2003 and the equivalence between a period and fifteen
minute noise descriptor, these levels would approximately equate to LAeq,15minute
43/38/33 dB(A) for the day evening and night period respectively for a newly exposed
community in a rural environment. These are above the 35 dB(A) minimum intrusive
criterion in the INP for the day and evening periods but marginally below it for the
night period.
In examining the question of whether a lower minimum night-time level is warranted for
the night-time period for an individual source (premises), it is relevant to consider the
World Health Organization Night Noise Guidelines for Europe (WHO – NNGE, 2009)24.
It should be noted that the WHO NNGE, 2009:
•
was not developed for assessment of specific projects
•
adopts long averaging times for the purpose of developing long-term noise
management strategies
•
adopt a conservative health-based end point of lowest observed adverse effect
level, which is essentially no impact.
WHO recommend a yearly average Lnight,outside of 40 dB(A) for the protection of the public
from sub-clinical health effects, with an interim target of 55 dB(A) Lnight,outside. The Lnight,outside
noise descriptor is essentially a year average of LAeq,night levels. As a very conservative
approximation the Lnight,outside 40 dB(A) would equate to a LAeq,15minute 42 dB(A).
The WHO NNGE refers to a study by Passchier–Vermeer W, et al. (2002)21 that
indicates that the average outside to inside noise difference considered in the
guideline is around 21 dB(A), due to windows being closed only some of the time. The
study notes windows were open for 75 per cent of the time and when windows were
slightly open in Europe the outside sound levels were usually reduced by 10–15 dB
indoors. Therefore, the guideline acknowledges that internal levels in the order of 25–
30 dB(A) for a significant proportion of the time across a year are consistent with the
conservative health-based WHO guideline value of Lnight,outside 40 dB(A).
On the basis of a conservative reduction of 10 dB(A) across a façade (outside to
inside noise difference) with a partially open window, the WHO NNGE infers that
an external criterion lower than 35 dB(A) is not warranted for the night-time
period. However, while the minimum background level of 30 dB(A) (and hence
minimum intrusive criterion of 35 dB(A)) appears to remain appropriate for nighttime, the question of whether it should apply to the day and evening period needs
to be examined.
A comparison with other national jurisdictions in terms of minimum or base-level
criteria is outlined in Table 1.
4
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
Table 1: Minimum/base criteria in other Australian jurisdictions
State
Day, dB(A)
Evening, dB(A)
Night, dB(A)
Victoria (metro)
LAeq,day 45
LAeq,evening 40
LAeq,night 35
Victoria (urban)
LAeq,day 45
LAeq,evening 37
LAeq,night 32
Queensland (petroleum
industry)
LAeq,1hr 40
LAeq,1hr 35
LAeq,1hr 28*
Queensland (general industry
– very rural)
LAeq,1hr 40
LAeq,1hr 35
LAeq,1hr 28*
Queensland (general industry
– rural residential)
LAeq,1hr 43
LAeq,1hr 38
LAeq,1hr 33
Western Australia
LA10,day (45-5**) = 40
LA10,evening (40–5) = 35
LA10,night (35–5) = 30
South Australia (rural living)
52–5**= 47
45–5** = 40
South Australia (residential)
47–5** = 42
40–5** = 35
* currently under review and proposed to rise to 33 dB(A); ** minus 5 to go from cumulative to individual premises.
The jurisdictions examined apply higher minimum criteria for the evening and daytime
periods typically in the order of plus 5 dB for the evening and plus 10 dB for the
daytime as compared to night. A minimum daytime intrusive criterion of LAeq,15minutes 40
dB(A) is generally consistent with other Australian jurisdictions and is proposed to be
adopted in the draft ING.
Proposal 2
No change to the minimum assumed background noise level of 30 dB(A) is proposed
for the night-time and evening period. However, the minimum assumed background
for the daytime is proposed to be raised to 35 dB(A), delivering a minimum daytime
intrusive criterion of LAeq,15minutes 40 dB(A).
4.3 Amenity criteria
4.3.1 Adequacy of existing ‘acceptable noise levels’
Fundamental to the existing amenity criterion in the INP are the cumulative
‘acceptable industrial noise levels’ (ANLs) outlined in Table 2.1 of the INP. To
determine whether the ANLs are consistent with contemporary measures of
annoyance for industrial noise they have been assessed relative to the dose–
response relationship of Miedema and Voss 200415 (Table 2).
Table 2: Percentage of people expected to be highly annoyed relative to existing
acceptable industrial noise levels
1
Receiver category
Equivalent DENL , dB(A)
Per cent highly annoyed
Rural acceptable
50
5
Suburban acceptable
53
7.5
Urban acceptable
58
12
Urban industrial interface
(acceptable)
63
18
2
Note: DENL = day-evening-night level; dB(A) = decibel (A-weighted); receiver category is the category sourced
from Industrial Noise Policy, Table 2.1. Day, evening and night acceptable industrial noise levels converted to DENL
level, referenced to ISO 1996–20039. The DENL is a composite energy whole-day rating level with a 5 dB(A) and 10
dB(A) penalty respectively for the evening and night-time period. 2. From Miedema and Voss (2004)15 dose–
response curve.
www.epa.nsw.gov.au
5
Draft Industrial Noise Guideline Technical Background Paper
The results of Table 2 demonstrate that ANLs for rural, suburban, urban and
industrial interface residential areas equate to between 82 and 95 per cent of
communities being protected from high annoyance from industrial noise. This should
not be taken to suggest that a higher level of protection is being afforded to rural
versus suburban and urban areas. It acknowledges the phenomenon stated in ISO
1996-1:20039, that the prevailing noise climate will have a bearing on the expected
community response, with rural dwellers expecting a quieter environment than
people in towns and cities, which will likely be manifested in the community’s
response and tolerance of noise.
The above results confirm that the ANLs currently in the INP align well with the stated
objective of seeking to protect 90 per cent of the community from being highly
annoyed by noise.
It should, however, be noted that the maximum ANLs currently in the INP are not
proposed to be retained in the revised ING as the practical use or application of
recommended ‘maximum’ levels is not readily apparent. The maximum levels are not
referenced in the existing INP criteria derivation process.
Proposal 3.1
The recommended acceptable noise levels in the INP are proposed to be retained;
however, the recommended maximum levels are proposed to be removed as they
serve no practical purpose. The industrial interface criterion is proposed to be applied
as a plus 5 dB(A) level for each residential receiver category (i.e. rural, suburban and
urban) as provided for in the existing INP application note – ‘When to apply the
urban/industrial interface category’.
4.3.2 Deriving the amenity criterion
The amenity criterion for an individual industrial site is essentially the level of noise
that it can emit that, in combination with other existing industrial noise, will result in
the relevant ANLs from INP, Table 2.1, being met. It is derived on the basis of a
logarithmic relationship approximated in INP, Table 2.2, that considers the existing
level of industrial noise and the relevant ANL.
The existing level of industrial noise at a receiver location is often difficult to
meaningfully define due to a range of factors including high ambient (non-industrial)
noise levels and temporal variations in noise due to changed noise output from
premises and meteorological conditions.
It is proposed to eliminate the need to quantify the existing industrial noise level to
derive the amenity criterion. The recommended acceptable noise levels in Table 2.1
in the INP are proposed to be retained; however, the ‘amenity criterion’ for an
individual activity will simply be derived by subtracting 5 dB from the acceptable
noise levels in Table 2.1 for the relevant receiver category. Similar approaches are
applied in other jurisdictions; for example, Western Australia and South Australia.
This will significantly reduce both the complexity of establishing the existing industrial
noise level and the associated variability of the derived amenity criterion. It will not,
however, result in significantly different outcomes. As an example, consider the
situations outlined in Table 3.
6
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
Table 3: Comparison in outcomes between existing and alternative amenity criterion
derivation processes.
Recommended
acceptable noise
level for rural
night, dB(A)
Existing industrial
noise, dB(A)
Amenity criterion,
dB(A)
Total noise (existing
industrial noise
plus amenity
criterion) and
difference, dB(A)
40 (existing approach)
50
30 or 40
50 or 50.5
40 (proposed
approach)
50
40–5 = 35
50.2
Difference
<0.5 dB
Example 1
Example 2
40 (existing approach)
43
30 or 33
43.2 or 43.4
40 (proposed
approach)
43
40–5 = 35
43.6
Difference
<0.5 dB
Example 3
40 (existing approach)
40
32
40.6
40 (proposed
approach)
40
35
41.2
Difference
+0.6 dB
Example 4
40 (existing approach)
35
38
39.8
40 (proposed
approach)
35
35
38
Difference
–1.8 dB
Example 5
40 (existing approach)
< 30
40
40.4
40 (proposed
approach)
< 30
35
36.2
Difference
–4.2
Notes: dB(A) = decibel (A-weighted).
In examples 1 and 2, where there are existing levels of industrial noise, if the
proposed industry commenced and emitted noise levels at the proposed amenity
criterion level (i.e. ANL – 5 dB(A)), the total noise would be within 0.5 dB of what the
current approach in the INP would deliver. This is an insignificant difference. In
example 3, where there is existing industrial noise at the ANL, the proposed
approach would deliver total industrial noise approximately 0.5 dB(A) above current
approaches. In examples 4 and 5, where there is little or no existing industrial noise,
the proposed approach would deliver less total industrial noise. This is appropriate as
it removes the potential for one industry to emit noise up to the ANL leaving no
additional headroom for other industrial development. In these cases the intrusive
noise criterion would generally apply anyway.
A more detailed analysis of outcomes based on a rural residential acceptable noise
level referenced to the current INP, Tables 2.1 and 2.2 is outlined at Attachment 2.
www.epa.nsw.gov.au
7
Draft Industrial Noise Guideline Technical Background Paper
Proposal 3.2
It is proposed to remove the need to quantify the existing level of industrial noise to
determine the amenity criterion. It is proposed to simply determine the amenity
criterion on the basis of the recommended acceptable noise level minus 5 dB(A). This
will significantly simplify the process while delivering comparable cumulative industrial
noise outcomes.
However, the need to establish existing levels of industrial noise will remain in
situations where the project-specific noise levels (proposed to be termed project noise
trigger levels, PNTL) cannot be met. This is because the assessment of residual noise
impacts (i.e. noise remaining above PNTL after all noise mitigation has been applied)
will require an assessment of cumulative industrial noise.
4.3.3 Amenity approach for clusters of industry
The recommended acceptable noise level from INP, Table 2.1, represents the ideal
total noise level from industry that should be met by a proposed development and any
future developments in the area. The approach of deriving the amenity criterion on the
basis of the recommended acceptable noise level minus 5 dB is based on a receiver
not being principally impacted by more than three to four industrial noise sources at a
given time.
Where an existing agglomeration of industry, for example an industrial estate or port
facility, is undergoing redevelopment and/or expansion, the default amenity criterion
approach can be applied where the proposed development constitutes new single
premises with no other redevelopment occurring in the foreseeable future.
However, in the case of a greenfield development or redevelopment of an existing
cluster of industry consisting of multiple new noise generating premises, the proposed
amenity approach (ANL – 5 dB) is not applicable. In these circumstances, the amenity
criterion for each industrial premise are proposed to be derived using equation 1.
Equation 1: Multiple premises redevelopment of existing agglomeration of industry
(ANL–5 dB/10)
Individual premise amenity criterion = 10Log (10
/N)
Where:
ANL = relevant acceptable noise level from Table 2.1
N = number of proposed additional premises
dB = decibel.
However, where a greenfield development is proposed and it can be demonstrated
that existing levels of industrial noise are less than the relevant recommended ANL,
equation 1 can be modified to reflect ‘ANL’ in lieu of ‘ANL-5 dB’.
Proposal 3.3
The proposed amenity approach for clusters of industry is incorporated into the
draft ING.
4.3.4 Precinct approach
There is a need to improve the tools available to manage the noise relationships
between major agglomerations of industry and existing residential uses more
broadly and efficiently than that provided by the INP at present. The existing
requirement allows for industrial interface requirements to extend to residences
within a zone that extends from the industrial boundary to a point where the
industrial noise has reduced by 5 dB. It does not, however, clearly provide the
flexibility to allow noise to be managed across an area for efficient land use within
8
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
acceptable noise outcomes for local communities. A provision to allow
consideration of a formal and structured approach to managing an agglomeration
of industry in a more flexible manner is proposed.
4.4 Meteorological conditions
Certain meteorological conditions may increase noise levels by focussing sound-wave
propagation paths at a single point. Such refraction of sound waves will occur during
temperature inversions (atmospheric conditions where temperatures increase with
height above ground level) and where there is a wind gradient (that is, wind velocities
increasing with height) and wind direction is from the source to the receiver. These
are commonly referred to as noise-enhancing meteorological conditions.
These meteorological effects typically increase noise levels by 5 to 10 dB(A), and
have been known to increase noise levels by as much as 20 dB(A) in extreme
conditions, thereby having the potential to significantly increase noise impacts.
Broadly, the INP requires that noise enhancing meteorological conditions be
considered in an Environmental Assessment when the condition is present for more
than 30 per cent of the time in any season and for any period (day, evening and
night) in the case of wind effects, and for more than 30 per cent of the time in winter
(night-time only) in the case of inversion conditions.
The INP provides for the adoption of either default noise-enhancing meteorological
conditions or for site-specific parameters to be developed from analysis of site
specific data.
Several issues have been identified in relation to meteorological effects; they are
summarised below.
4.4.1 How to define the presence and strength of a temperature inversion
The strength of a temperature inversion, for the purposes of noise assessment, is
generally defined or expressed in terms of the change in temperature in the first 100
metres of elevation from ground level. This is often expressed as a positive temperature
lapse rate. For example, the default parameter for non-arid areas in the INP is 3ºC/100
metres (moderate inversion) and 8ºC/100 metres for arid regions (strong inversion).
The INP also provides for determining inversion conditions based on the Pasquil–Gifford
(P–G) atmospheric stability class system, with F class indicating a moderate inversion
and G class indicating a strong inversion. However, the temperature lapse rates
expected to occur under these atmospheric stability conditions do not directly align with
the default parameters in the INP. For example, the range of lapse rates expected during
F class atmospheric stability conditions are between 1.5 and 4ºC/100 metres.
This has led to some regulatory uncertainty. For example, if a noise limit is stipulated
to apply under temperature lapse rates up to 3ºC/100 metres, but the licensee
determines the strength of an inversion via P–G atmospheric stability class, there is
uncertainty as the actual inversion strength may indeed exceed 3ºC/100 metres under
F class conditions.
P–G stability classes can be established from parameters measured by commonly
available weather stations and from ground-based observations; for example, cloud
cover. Therefore a move to P–G stability categories to define inversion strengths will
simplify and clarify the process.
Additionally, commonly used noise models, for example those based on ISO961326
and CONCAWE25 algorithms, rely on broad meteorological parameters as opposed to
specific temperature lapse rates. This will result in a better alignment with the draft
ING requirements and the capability of current noise models.
www.epa.nsw.gov.au
9
Draft Industrial Noise Guideline Technical Background Paper
Proposal 4.1
It is proposed to stipulate inversion conditions within the draft ING for both
environmental assessment purposes and for noise conditions in consents and
licences based on P–G atmospheric stability class. This will obviously not prohibit
direct measurement of temperature gradient as the P–G category can still be
determined via this method. It will, however, serve to increase the default non-arid
temperature lapse rate conditions in the INP to 4ºC/100 metres.
4.4.2 How to directly measure temperature gradients (inversion strength)
The INP presents a method to directly measure temperature gradients and assess the
significance of inversion conditions at Appendix E, E2. Very generally, this method
relies on temperature measurements at two heights, the first between 1.5 and 10
metres above ground level (AGL) and the second between 50 and 60 metres AGL
over a 50-metre height interval.
Various interpretations of these provisions have caused issues for stakeholders
and have been found to lead to significant differences in outcomes. An
Environment Trust Project was undertaken in 2010–2011 by the Noise Policy
Section (NPS) of the NSW, EPA18 to define a more standard and repeatable
approach for the direct measurement of temperature gradients. The study was
undertaken in the Hunter Valley using a helium balloon and tether sonde. The study
findings have assisted in better defining how temperature gradients should be
measured and the following changes are proposed.
Proposal 4.2
It is proposed to amend the current approach in the INP (Appendix E2) in the following
ways:
1. stipulate the minimum measurement height as 10 metres as opposed to 1.5 metres
2. stipulate a measurement height interval of 50 metres
3. stipulate that where land forms are used to measure at two different heights (i.e.
the measurement locations are not directly above/below one another) that this
approach be initially verified by a measurement campaign using, for example, a
balloon and tether sonde (i.e. measurements directly above/below one another).
4.4.3 Meteorological conditions used in environmental assessment
documents (i.e. documents accompanying applications)
The INP currently requires meteorological effects to be considered by either adopting
default noise-enhancing meteorological parameters or by site-specific analysis to
determine whether noise enhancing meteorology is significant.
The INP determines the significance of noise enhancing meteorological conditions for
environmental assessment (EA) purposes (i.e. development planning stage) on the
basis of a 30 per cent threshold. Where the condition is found to occur for more than
30 per cent of the time, noise predictions should be made under noise enhancing
conditions. The significance test is considered a reasonable threshold and is not
proposed to be changed.
Under the current INP, where meteorological conditions are deemed significant, the
assessment may proceed using default meteorological conditions or actual measured
meteorological conditions.
The default inversion conditions (relevant for night-time period only) currently in the
INP are reproduced in Table 4.
10
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
Table 4: Industrial Noise Policy default inversion parameters
Non-arid areas (annual rainfall greater
than 500mm)
º
3 C/100 m temperature inversion strength for all
receivers, plus a 2 m/s source-to-receiver
component drainage flow wind speed for those
receivers where applicable.
Arid and semi-arid areas (annual average
rainfall less than 500mm)
º
8 C/100 m temperature inversion strength for all
receivers, plus a 1m/s source-to-receiver
component drainage flow wind speed for those
receivers where applicable.
Note: m = metres; mm = millimetres; m/s = metres per second.
The default wind speed condition currently in the INP is reproduced in Table 5.
Table 5: Industrial Noise Policy default wind parameters
Three metres per second (at 10 metre height) is the default value for assessing noise
impacts caused by gradient winds.
Commonly used noise prediction models/calculation algorithms, for example
CONCAWE25 and ISO961326, consider meteorological effects on the basis of
predetermined meteorological scenarios that do not directly align with the current INP
default conditions.
For example, the CONCAWE25 model provides six meteorological categories as
shown in Table 6.
Table 6: CONCAWE meteorological categories
Meteorological
category
Atmospheric category
Unstable
Normal
Stable
1
v<–3
-
-
2
–3<v<–0.5
v<–3
-
3
–0.5<v<+0.5
–3<v<–0.5
-
4
+0.5<v<+3
–0.5<v<+0.5
–3<v<–0.5
5
v>3
+0.5<v<+3
–0.5<v<+0.5
6
-
v>3
+0.5<v<+3
Note: v = wind speed in m/s with a negative value meaning wind from receiver to source and a positive value
meaning wind from the source to the receiver.
The meteorological categories represent combinations of atmospheric stability and
wind vectors. Manning, 198125 indicates that ‘unstable’ represents P–G categories A
and B; ‘normal’ represents P–G categories C–E; and ‘stable’ represents P–G
categories F and G. It is relevant to note that the CONCAWE26 model is based on
empirical observations and hence the noise propagation algorithms are based on lines
of best fit. Therefore some professional judgement and experience will be needed in
selecting the meteorological category relevant for assessment purposes.
ISO9613-2:199626 states:
‘The method predicts the equivalent continuous A-weighted sound pressure level under
meteorological conditions favourable to propagation from sources of known sound
emission. These conditions are for downwind propagation, as specified 5.4.3.3 of ISO
1996-2:1987 or, equivalently propagation under a well-developed moderate ground
based temperature inversion, such as commonly occurring at night.’
It is evident that both CONCAWE and IS9613-2 based models could not model or
readily distinguish between the INP default inversion parameters for ‘non-arid’ and
‘arid and semi-arid’ areas. Additionally, both models are generally unable to take
account of temperature lapse rate directly. Therefore a move towards standardised
www.epa.nsw.gov.au
11
Draft Industrial Noise Guideline Technical Background Paper
meteorological conditions better aligned with current generation noise prediction
software has advantages addition to those already identified under s.4.2 of this
technical paper.
Therefore it is proposed to amend the INP to require assessment under one of two
meteorological conditions. They are termed ‘standard meteorological conditions’ and
‘noise enhancing meteorological conditions’ and are proposed to be defined as
described in Table 7.
Table 7: Proposed meteorological assessment conditions
Meteorological condition
Meteorological parameters
Standard meteorological conditions
Day/evening/night: stability category A–E wind speed
up to 0.5 m/s at 10 metres AGL.
Noise enhancing meteorological conditions
Daytime/evening: stability category A–E with light
winds (0.5–3 m/s), night-time stability category A–E
with light winds (0.5–3 m/s) and stability category F.
Note: AGL = above ground level; m/s = metres per second.
The procedure will be to determine whether noise-enhancing meteorological
conditions are significant on the basis of the 30 per cent threshold. Where found to be
significant, noise predictions will adopt these conditions. Otherwise standard
meteorological conditions will be adopted. This is designed to simplify consideration of
meteorological effects and to allow for better harmonisation between the
meteorological conditions used in noise assessments and those applied to noise limits
in consents and licences (see also proposal 4.4).
Proposal 4.3
To adopt a simplified meteorological condition assessment approach based on two
broad meteorological categories termed standard and noise-enhancing
meteorological conditions. This will better align with current noise prediction
capabilities; will not result in any loss of noise protection; and is proposed to be
augmented with further proposals (see section 4.4) as outlined below to ensure that
noise requirements also apply for meteorological conditions outside the range of the
two proposed meteorological conditions.
4.4.4 Meteorological conditions used at development planning stage
versus meteorological conditions included in consent and licence
conditions
It has been EPA operational practice to include noise-enhancing meteorological
conditions in licence limits regardless of whether the condition was found to be
significant at the environmental assessment (EA) stage.
The rationale for this approach is twofold:
1. that the proponent and not the community should assume the risk of a
meteorological assessment that is generally only based on a limited data set,
typically one year’s worth of data; and
2. that compliance testing needs to be able to be performed under a reasonable
range of meteorological conditions.
This has been considered by some proponents and consultants to be a ‘shifting of
the goal posts’ between the EA stage and operational stage of a development.
Additionally, it is potentially imposing a limit that the proponent has not demonstrated
can be met and/or has not considered appropriate noise mitigation for.
12
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
In addition to these problems, the practice of stipulating meteorological conditions
under which consent and licence noise conditions apply essentially results in limits
that do not apply under a range of meteorological conditions. For example, a noise
limit condition may indicate that the limits apply for wind speeds up to and including
three metres per second. The corollary of this is that limits do not apply for wind
speeds in excess of three metres per second, leaving a potential regulatory gap.
Proposal 4.4
As indicated in proposal 4.3 it is proposed that noise assessments are undertaken
using either ‘standard’ or ‘noise enhancing’ meteorological conditions. The
meteorological conditions used will be on the basis of a significance test to determine
whether noise-enhancing conditions are a significant feature of the area (i.e. >30%
occurrence). The meteorological conditions which the noise impact assessment was
based on, and hence the acceptability of noise impacts determined, will be reflected in
the consent or licence, thereby giving consistency. However, it is proposed to include
an upper bound of noise limit plus 5 dB(A) for meteorological conditions outside those
adopted for assessment. In this way noise limits will apply for all meteorological
conditions, giving increased certainty and accountability for all stakeholders.
4.5 Identifying non-compliances (sustained non-compliance)
Chapter 11 of the INP provides guidance on non-compliance with noise conditions.
Section 11.1.3 is reproduced below:
Non-compliances with noise conditions
When is a development in non-compliance with a noise condition?
A development will be deemed to be in non-compliance with a noise consent or licence
condition if the monitored noise level is more than 2 dB above the statutory noise limit
specified in the consent or licence condition. This may occur for the following two
reasons:
•
The noise from the development is excessive, in which case the development
is truly not complying with its consent or licence condition.
•
The noise was increased by extreme, non-standard weather effects – in which
case the development is not considered to be in non-compliance with its consent or
licence condition. Non-standard weather effects can be considered to be present during
monitoring if the cloud cover is less than 40 per cent and the wind speed (at 10m
height) is less than 1 metre per second (representing an extremely adverse weather
condition for noise) – during the period from 6 pm to 7 am in non-arid areas (see
Section 9.2).
In this later case, further monitoring at a later date is required to determine compliance
under the meteorological conditions specified in the consent/licence condition.
Managing non-compliances
Where the noise exceedance is found to be caused by excessive noise levels from a
development, the development is considered to be in non-compliance with its consent or
licence condition. Strategies for mitigation of noise need to be considered and
implemented, or a further negotiation process could be commenced.
When is a development in breach of a noise condition?
A development will be in breach of a noise consent or licence condition if sustained noncompliances are not addressed.
The general language and approach in this section does not align with the EPA’s
contemporary approach to managing compliance. There are two other issues with this
text in the INP. The first is that the 2 dB allowance is essentially seen as raising a
www.epa.nsw.gov.au
13
Draft Industrial Noise Guideline Technical Background Paper
noise limit by 2 dB, that is, a limit of 35 dB is essentially 37 dB. The 2 dB tolerance
was included in the INP to account for uncertainty including instrumentation
accuracy/error.
The second and more vexing aspect of this text is that ‘sustained’ is not defined and
therefore has been largely left to subjective and varying interpretations. This problem
has been compounded by the advent of continuous type noise monitoring systems
where compliance has been sought to be determined on the basis of running
averages and other statistical approaches.
Proposal 5
Subjective terms such as sustained non-compliance are proposed to be removed
from the guideline. Determining compliance or otherwise with a noise limit will be
consistent with the EPA’s approach to licence compliance of simply comparing the
measured/predicted level to the consent or licence noise limits. All noise levels should
be rounded to the nearest integer number before comparing to a noise limit.
The regulatory action that may be taken for a non-compliance is largely dependent on
the individual circumstances in consultation with guidance document(s) (if any)
prepared or adopted by the regulator. In the case of the Environment Protection
Authority, the primary documents would be the EPA Compliance Policy, May 2013
www.epa.nsw.gov.au/resources/legislation/130251epacompol.pdf and the EPA
Prosecution Guidelines, March 2013
www.epa.nsw.gov.au/resources/legislation/20130141EPAProsGuide.pdf.
As a guide only, a non-compliance that occurs for greater than 10 per cent of an
assessment period (i.e. day, evening or night-time period) would likely require a
regulatory response requiring the licensee to investigate and mitigate the cause.
4.6 Modifying factor adjustments
Chapter 4 in the INP outlines processes to determine whether a noise contains
particular characteristics that may increase or reduce the annoyance of a noise.
These characteristics include: tonality, low frequency noise (LFN), impulsive noise,
intermittent noise and duration.
4.6.1 Tonality
The tonality provisions in the INP appear to be working well and are not proposed to be
significantly changed at this stage. However, it is recommended that they be modified
slightly to align with the way tonality is expressed in ISO1996.2:2007 Acoustics –
Description, measurement and assessment of environmental noise – Determination of
environmental noise levels10. The current approach of adding a 5 dB(A) adjustment to
measured/predicted noise levels for tonality is proposed to be retained.
Proposal 6.1
That the tonality provisions in the revised ING be slightly modified to reflect
ISO1996.2:2007. The approach of adding 5 dB(A) to measured/predicted noise levels
that trigger the tonality thresholds is proposed to be retained.
4.6.2 Low frequency noise
Low frequency noise is currently defined in the INP as noise containing major
components within the low frequency range (20 Hz to 250 Hz) of the frequency
spectrum. The current provision in the INP applies a positive 5 dB(A) modification to
the measured/predicted A-weighted noise levels when the C-weighted levels minus
the A-weighted levels exceed 15 dB. This is quite a conservative criterion that would
be exceeded in many environments not associated with significant LFN problems; for
14
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
example, air conditioned environments. In addition, the ‘C minus A’ differential will
naturally increase as you move away from a noise source due to higher attenuation
rates of higher frequencies versus lower frequencies. This can lead to a perverse
outcome where a low frequency modification may not apply near to a noise source,
but will apply at more removed distances, even though the amplitude of the LFN
spectrum has reduced.
In 2010 the NSW Department of Planning and Infrastructure recommended an
alternative approach based on work by Dr Norm Broner (SKM)1, a world recognised
leader in LFN assessment. The criteria recommended by Broner are based on overall Cweighted levels: LCeq 65 dB daytime and LCeq 60 dB night-time. The recommended
criteria relied largely on the work by Hessler 20045, which relates to emissions from gas
turbines, which characteristically have significant emissions in the lower end of the low
frequency range; i.e. nominally dominant low frequency energy from 16 Hz to 63 Hz.
Both the C minus A and overall C-weighted approaches have potential limitations.
Some examples of these limitations are outlined below.
Example 1: Tonal LFN at 20 Hz at a level of say 70 dB could give a C minus A of
greater than 40 dB and an overall C weighted level of greater than 60 dB(C). Both of
these descriptors would indicate the potential for a LFN impact. However 70 dB @ 20
Hz is below the human hearing threshold in ISO226, 20038 and therefore would be
unlikely to represent an impact.
Example 2: Tonal LFN at 160 Hz at a level of say 50 dB could give a C minus A of
just less than 15 dB and an overall C weighted level as low as 50 dB(C). Both of
these descriptors would indicate that there is no low frequency impact. However,
the level of 50 dB @ 160 Hz would likely exceed internal annoyance indicators for
LFN; for example, that proposed by the UK Department for Environment, Food and
Regional Affairs (DEFRA)16 when assuming a modest transmission loss across a
façade of minus 10 dB.
Leventhal, 200430 indicates:
‘Attempts to assess low frequency noise by conventional wide-band noise methods
often fail, so illustrating the inadequacy of these methods for low frequencies.’
and
‘The difference between C- and A-weightings has also been considered as a predictor of
annoyance (Broner, 1979; Broner and Leventhall, 1983; Kjellberg et al., 1997), as this
difference is an indication of the amount of low frequency energy in the noise. If the
difference is greater than 20 dB, there is the potential for a low frequency noise problem.
Kjellberg et al used existing noise in work places (offices, laboratories, industry etc) with
508 subjects. Three sub-groups were obtained with a maximum difference in low and high
frequency exposure. The conclusions on correlations of (dB(C) – dB(A)) difference and
annoyance were that the difference is of limited value, but, when the difference exceeds 15
dB, an addition of 6 dB to the A-weighted level is a simple rating procedure. However, the
difference breaks down when the levels are low, since the low frequencies may then be
below threshold. The (dB(C) – dB(A)) difference cannot be used as an annoyance
predictor, but is a simple indicator of whether further investigations may be necessary.’
LFN criteria and approaches applied in Europe, are typically based on one-third
octave assessment curves generally from 10 Hz to 200 Hz, as opposed to overall Cweighted noise levels and/or ‘C minus A’ approaches. As eluded to above, the UK
Department for Environment, Food and Regional Affairs (DEFRA)17 also conducted a
review of European LFN criteria as part of the development of an assessment
approach for LFN designed to assist UK Environmental Health Officers when
undertaking investigations of alleged LFN nuisances.
www.epa.nsw.gov.au
15
Draft Industrial Noise Guideline Technical Background Paper
The DEFRA approach presents a criterion curve, reproduced below in Table 8, with
threshold levels from 10 Hz to 160 Hz. These levels can be relaxed by 5 dB if the
sound only occurs for the daytime period. It can also be relaxed by 5 dB if the sound
is steady and not fluctuating.
Table 8: UK Department for Environment, Food and Regional Affairs fluctuating
criterion values
Frequency Hz/dB
10
12.5
16
20
25
31.5
40
50
63
80
100
125
160
92
87
83
74
64
56
49
43
42
40
38
36
34
Note: dB = decibel; Hz = hertz. The criterion values may be raised by 5 dB for the daytime period or where the noise
is steady.
DEFRA, 200516 states:
‘The procedure is intended to assist in the evaluation of existing problems. It is not
intended as a means of predicting when disturbance might occur, for example in a
planning situation, and would not be reliable to use as such. This is because disturbance
by LFN depends on a number of factors, such as the character of the sound, whose
effects are neither well understood, nor readily predictable. Levels of sound above
criteria based on the average threshold of hearing are frequently found to be
acceptable and levels falling marginally below can occasionally cause
disturbance, so no generic approach appears to be possible.’
The underlined phrase suggests that the DEFRA LFN criterion is likely to be
conservative if used in a predictive/planning sense. For that reason the significance of
exceedance of the DEFRA fluctuating curve (base curve) would likely be less than the
significance of exceedance of the steady/daytime curve. Figure 1 compares the
DEFRA fluctuating curve (identified as UK) with other curves used in Europe to
assess LFN impacts.
Figure 1: UK Department for Environment, Food and Regional Affairs fluctuating curve
compared to other jurisdictions. dB = decibel; Hz = hertz.
Typically the curves are below the hearing threshold below 50 Hz and are reasonably
closely grouped. Above 50 Hz the curves move apart with a range of 20 dB at 125 Hz
and with some curves below and above the hearing threshold. Note that the Danish
16
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
curve is derived at each one-third octave from the Danish LFN criteria LpA,LF(10–160Hz)
20 dB(A) and is therefore an approximate interpretation.
Noise criteria in NSW typically apply external to a building and in the free field (i.e.
typically more than 3 metres from a reflective vertical surface). The use of internal
criteria would raise a number of complexities including the need to access buildings
for compliance assessment purposes.
Therefore, if the DEFRA criteria were proposed to be used as an external measure of
LFN, there would need to be reasonably conservative assumptions made about the
relationship between indoor and outdoor noise levels, that is, the attenuation (noise
reduction) effect of the façade. It is commonly accepted that typical residential
buildings provide lower reductions for noise in the low frequency range (10 Hz to 200
Hz) than in mid to high frequencies.
There is considerable scientific literature looking at overall façade noise reductions
using A-weighted decibels; however, this reveals very little in terms of how the
façade performs in the low frequency range. Literature specific to the low frequency
range is sparser.
Some recent work undertaken by Hansen et al, 20124 presents indoor to outdoor
differences in noise near a wind farm in Australia relative to two wind vectors at
the house (presumably microphone height) being 0 metres per second and 1.1
metres per second with windows open. Figures 2a and 2 b represent Figures 8
and 9 from Hansen et al, 2012. The operating conditions of the wind farm were
not available to the investigators.
Figure 2a: Sound pressure level (SPL) difference from outdoors to indoors near a wind
4
farm for wind speeds of 0 metres per second (Source: Hansen et al 2012 , Figure 8).
www.epa.nsw.gov.au
17
Draft Industrial Noise Guideline Technical Background Paper
Figure 2b: Sound pressure level (SPL) difference from outdoors to indoors near a wind
4
farm for wind speeds of 1.1 metres per second (Source: Hansen et al 2012 , Figure 9).
Denmark introduced LFN requirements for wind farms in 2012. It requires that 20
dB(A), band limited to 10 Hz to 160 Hz, not be exceeded inside a residence. The
limits apply for 6 metres per second and 8 metres per second and are based on
predicted noise levels following a prescribed approach. Part of this approach
necessarily includes façade noise reduction values from 10Hz to 160Hz.
The figure below is an extract from Hoffmeyer and Jakobsen, 20106 presenting a
summary of results from façade transmission loss studies in Denmark, the
Netherlands, Great Britain and United States. This study informed the façade
transmission loss for the façade noise reduction values in the Danish approach.
Details about whether windows were open or closed were not provided.
Figure 3: Summary of results from façade transmission loss studies in Denmark, the
6
Netherlands, Great Britain and United States (source: Hoffmeyer and Jakobsen, 2010 ,
Figure 6).
18
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
Additionally Jakobsen, 201212 presented data from measurements undertaken in 14
Danish residences using three internal measurement positions, one near the junction
of the wall and floor and the other two at other locations to represent normal use of the
room. The external noise source was provided by a loudspeaker facing the affected
façade. Figure 4 reproduces Figure 2 from Jakobsen, 2012, which presents the results
of all measurements of outdoor/indoor level difference and a selected ‘level difference
curve’ (solid dark line) that is exceeded by 67 per cent of the measurements, which is
used in the Danish Regulation of Low Frequency Noise from Wind Turbines. Details
about whether windows were open or closed were not provided.
Figure 4: Outdoor/indoor level difference and ‘level difference curve’ – Danish
Regulation of Low Frequency Noise (Source: From Jakobsen, 201212, Figure 2).
Shindo, et al, 200827 reported outdoor/indoor differences in the very low frequency range
by Ochiai, 2001 considering various configurations of construction and window
frame/sash arrangements. Figure 5 presents the result of these measurements. Details
about whether windows were open or closed were not provided.
www.epa.nsw.gov.au
19
Draft Industrial Noise Guideline Technical Background Paper
Figure 5: Outdoor/indoor sound pressure level differences for various window sash
configurations (Source: Shindo, et al 200827, Figure 3).
Kelley ND 198728 also reported transmission loss across five typical American homes
exposed to noise from a gas-fired turbine in the low frequency range for both
impulsive and non-impulsive noise. Details about whether windows were open or
closed were not provided. Kelly indicates that: ‘Typically, 5–7 dB of attenuation occurs
in the 10–160 Hz band range for a non-impulsive source excitation’.
Figure 6: Indoor/outdoor attenuation in five typical American homes near a gas fired
power station, non-impulsive source (Source: Kelley, 198729, Figure 4).
While in some circumstances it is clear that LFN can be amplified from outside to
inside due to structural resonance and room configuration, in most cases modest
reductions occur.
20
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
The correction factors outlined below in Table 9 have been derived based on the
lowest tenth percentile façade transmission loss from the above data. They have been
used to adjust the DEFRA criteria from inside to outside.
Table 9: Assumed façade transmission loss (10–160Hz)
Frequency Hz/dB
10
12.5
16
20
25
31.5
40
50
63
80
100
125
160
0
–2
–3
–3
–5
–5
–5
–7
–8
–8
–10
–10
–10
Note: These adjustments would be positive when correcting criterion values and negative when adjusting measured values.
Proposal 6.2
It is proposed to retain the C minus A of 15 dB or greater as an initial screening
approach. Where a C minus A of 15 dB or greater is measured/predicted, either a 5
dB(A) positive adjustment is applied to measured or predicted noise levels, or external
one-third octave band measurements or predictions are undertaken from 10 Hz to 160
Hz and compared to the frequency specific criteria in Table 10.
Table 10: Third octave low frequency noise adjustment thresholds, dB RE: 10
–6
Pa
Frequency (Hz), Leq,15minute values
10
12.5
16
20
25
31.5
40
50
63
80
100
125
160
92
89
86
77
69
61
54
50
50
48
48
46
44
•
Where any of the 1/3 octave noise levels in Table 10 are exceeded by up to 5
dB and cannot be mitigated, a 2 dB(A) positive adjustment to measured/predicted A
weighted levels applies for the evening/night period.
•
Where any of the 1/3 octave noise levels in Table 10 are exceeded by more
than 5 dB and cannot be mitigated, a 5 dB(A) positive adjustment to
measured/predicted A weighted levels applies for the evening/night period and a 2
dB positive adjustment applies for the daytime period.
4.6.3 Impulsive noise
The current provisions in the INP are based on the difference between the LAmax
measured in both fast and impulse time response. When the difference is greater
than 2 dB a correction is required which is equal to the difference up to a maximum
of 5 dB. It is proposed to seek input from acoustic practitioners as part of the review
process to assist with decisions regarding this aspect of the policy review.
Proposal 6.3
It is proposed to seek the views of acoustical practitioners as part of the
review/consultation process. It is not currently included in the draft ING.
4.6.4 Duration
The existing duration modification includes adjustments for noises that are present for
less than 15 minutes duration. The allowable increase in noise applies for the duration
of the noisy event. For noisy events with a duration of less than fifteen minutes, it is
proposed to clarify the allowable increase for both the duration of the event and more
importantly for limits that would normally be based on an LAeq,15minute basis.
www.epa.nsw.gov.au
21
Draft Industrial Noise Guideline Technical Background Paper
Proposal 6.4
The duration correction approach in the INP is proposed to be amended so that the
effect of duration allowances for noisy events of less than 15 minutes can be
determined on LAeq,15minute limit based conditions using equation 2.
criterion/10
Equation 2: Adjusted LAeq,15minute equivalent project noise trigger level = 10Log((10
(criterion + allowable exceedance/10)
x (Y/900s)))
(10
Where:
X = 900s – duration of event
Y = duration of event.
x (X/900s)) +
4.6.5 Intermittent noise and maximum adjustments
No changes are proposed to the existing requirements in the INP for intermittent noise
or maximum adjustment.
4.7 Sleep disturbance
The INP does not currently include provisions for sleep disturbance assessment.
However, an extract from an INP application note provides a screening level
assessment tool for sleep disturbance as follows:
Sleep disturbance
Peak noise level events, such as reversing beepers, noise from heavy items being
dropped or other high noise level events, have the potential to cause sleep disturbance.
The potential for high noise level events at night and effects on sleep should be
addressed in noise assessments for both the construction and operational phases of a
development. The INP does not specifically address sleep disturbance from high noise
level events.
Research on sleep disturbance is reviewed in the NSW Road Noise Policy. This review
concluded that the range of results is sufficiently diverse that it was not reasonable to
issue new noise criteria for sleep disturbance.
From the research, the EPA recognised that the current sleep disturbance criterion of an
LA1, (1 minute) not exceeding the LA90,(15 minute) by more than 15 dB(A) is not ideal.
Nevertheless, as there is insufficient evidence to determine what should replace it, the
EPA will continue to use it as a guide to identify the likelihood of sleep disturbance. This
means that where the criterion is met, sleep disturbance is not likely, but where it is not
met, a more detailed analysis is required.
The detailed analysis should cover the maximum noise level or LA1,(1 minute), that is, the
extent to which the maximum noise level exceeds the background level and the number
of times this happens during the night-time period. Some guidance on possible impact is
contained in the review of research results in the NSW Road Noise Policy. Other factors
that may be important in assessing the extent of impacts on sleep include:
•
how often high noise events will occur
•
time of day (normally between 10 pm and 7 am)
•
whether there are times of day when there is a clear change in the noise
environment (such as during early morning shoulder periods).
The LA1,(1 minute) descriptor is meant to represent a maximum noise level measured under
‘fast’ time response. The EPA will accept analysis based on either LA1,(1 minute) or LA,(Max).
More recent work has been done on the health effects of sleep disturbance by the
World Health Organization (WHO). The WHO Night Noise Guidelines for Europe
(WHO, NNG-2009)24 discusses the suitability of both the Lnight,outside and LAmax
descriptors for the assessment of sleep disturbance.
22
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
The Lnight,outside descriptor, which is similar to a yearly average Leq,night descriptor, is an
equal energy descriptor that inherently has regard for both the level and the number of
events. The WHO, NNG indicates that the Lnight,outside descriptor ‘ties the LAmax related
effects to a maximum and therefore allows for a protective/conservative approach’
(NNG, S1.3.2. page 8). This suggests that that an equal energy descriptor, for
example Leq, which has regard for both level and number of events, is useful for sleep
disturbance assessment.
However, the WHO, NNG also states the LAmax descriptor is also useful because it is
easy to explain to the public, intuitively understandable, and historically widely used
for assessing sleep disturbance.
There is merit in including criteria for both descriptors to ensure that the maximum
level and the number and frequency of events are taken account of when considering
sleep disturbance.
The WHO, NNG recommends a yearly average Lnight,outside of 40 dB(A). However, this
criterion has been specifically derived in relation to long-term exposure to noise and
the relationship with health effects. The WHO criteria are not intended for use as
criteria for assessment of the impacts of a specific project and must be used with
caution. The criteria represent a health-based threshold based on the lowest observed
adverse effect level (LOAEL), a very conservative health end point.
The WHO, NNG also indicates that LAmax 42 dB inside a bedroom aligns with the
LOAEL as this level is identified as the levels that may cause awakenings from sleep.
Based on the conservative assumption of a 10 dB(A) noise reduction across a façade
with a partially open window, this results in an external level of LAmax 52 dB.
The current practice of deriving screening level sleep disturbance assessment criteria
on the basis of background plus 15 dB can lead to screening criteria as low as LAmax 45
dB(A), which is well below the LOAEL recommended by WHO. Therefore, it is
proposed to raise the base screening level criteria for the maximum noise level
descriptor to LAmax 52 dB(A) to align with the WHO, NNG. Like all trigger levels in the
draft ING, this should not be construed as the level at which unacceptable impacts
occur, but rather the level at which feasible and reasonable mitigation measures need
to be considered as part of a detailed assessment.
It has therefore been proposed in the draft ING to adopt the following screening level
assessment criteria approach and trigger levels.
Where the subject development can satisfy the following two noise level event trigger
levels for the night-time period, no additional assessment or evaluation of sleep
disturbance is required:
1. a night-time project noise trigger level of LAeq,15minutes 40 dB(A)
2. a maximum noise level screening criteria of LAmax 52 dB(A) when assessed or
predicted at 1 metre from the façade of a residence containing a window.
Where the night-time noise levels are predicted to exceed one or both of the
maximum event noise trigger levels above, a detailed analysis should be undertaken.
The detailed analysis should cover the maximum noise levels, the extent to which the
maximum noise levels exceed the background level (RBL) and the number of times
this happens during the night-time period. Some guidance on possible impacts is
contained in the review of research results in the NSW Road Noise Policy.
www.epa.nsw.gov.au
23
Draft Industrial Noise Guideline Technical Background Paper
Other factors that may be important in assessing the extent of impacts on sleep
include:
•
how often high noise events will occur
•
the distribution of likely events across the night-time period and the existing
ambient maximum events in the absence of the subject development
•
whether there are times of day when there is a clear change in the noise
environment (such as during early morning shoulder periods)
•
current scientific literature available at the time of the assessment regarding the
impact of maximum noise level events at night.
Maximum noise level event assessments should be based on the LAmax descriptor on
an event basis under 'fast' time response.
The detailed assessment should consider all feasible and reasonable noise mitigation
measures with a goal of achieving the above trigger levels.
Proposal 7
A sleep disturbance assessment will not be required as part of an ING assessment
when a development is demonstrated to satisfy an LAeq,15minute 40 dB(A) and LAmax 52
dB(A) at night.
Where the night-time noise performance of a premises is predicted to be higher than
LAeq,15minute 40 dB(A) and or LAmax 52 dB(A), an assessment of all feasible and
reasonable noise mitigation measures and the potential for sleep disturbance should
be further evaluated.
4.8 Land acquisition and home acoustic treatments
The decision to require proponents to offer home acoustic treatments or land
acquisition rests with the consent authority approving the proponent’s development
under the Environmental Planning and Assessment Act 1979. The role of the EPA
and the ING is to identify the level of impact associated with residual noise impacts,
that is noise levels above the project noise trigger levels after all feasible and
reasonable noise mitigation strategies have been applied.
The residual noise impact assessment thresholds proposed in Table 11 are
generally based on an agreed practice developed by the then Department of
Planning & Infrastructure and Environment Protection Authority since the release of
the INP. It is also consistent with the NSW Government’s Voluntary Land
Acquisition and Mitigation Policy, For Sate significance mining, petroleum and
extractive industry developments31.
24
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
Table 11: Significance of residual noise impacts
If the predicted noise level
minus the project noise
trigger level is
And the total cumulative
industrial noise level is
Then the
significance of
residual noise level
is
≤ 2 dB (A)
≤ recommended amenity noise level
or
> recommended amenity noise level
but the increase in recommended
amenity noise levels resulting from the
development is less than 1 dB
Negligible
≤ 2 dB (A)
> recommended amenity noise level
and the increase in amenity noise
levels resulting from the development
is more than 1 dB
Marginal
≥ 3 but ≤ 5 dB (A)
< recommended amenity noise level
or
> recommended amenity noise level
but the increase in recommended
amenity noise level resulting from the
development is less than 1 dB
Marginal
≥ 3 but ≤ 5 dB (A)
> recommended amenity noise level
and the increase in recommended
amenity noise levels resulting from the
development is more than 1 dB
Moderate
>5 dB (A)
Significant
Note: dB = decibel; dB(A) = decibel (A-weighted). This approach is designed for new and substantially modified
developments and should be applied with caution to assessments of existing operations.
Examples of noise mitigation at a residence that may be required by Planning
Authorities to mitigate residual noise impacts are outlined in Table 12.
Table 12: Examples of receiver based treatments to mitigate residual noise impacts
Significance of residual noise level
Example of potential treatment
Negligible
The exceedances would not be discernable by the
average listener and therefore would not warrant
receiver-based treatments or controls
Marginal
Provide mechanical ventilation/comfort condition systems
to enable windows to be closed without compromising
internal air quality/amenity.
Moderate
As for marginal but also upgraded façade elements like
windows, doors, roof insulation etc. to further increase
the ability of the building façade to reduce noise levels.
Significant
Offers of acquisition
www.epa.nsw.gov.au
25
Draft Industrial Noise Guideline Technical Background Paper
4.9
Relationship between project-specific noise levels and
statutory limits
The general approach in the INP and as proposed in the draft ING is to derive project
noise trigger levels (PNTL) that are used for impact assessment purposes and
essentially mark the point at which noise mitigation needs to be considered. Where
an activity can satisfy the PNTL using feasible and reasonable noise mitigation
measures, an unacceptable level of noise impact is not expected. Where noise levels
above the PNTL are predicted a further assessment of the significance of that
impact, further feasible and reasonable on-site mitigation and the need for off-site
mitigation must be carried out and justified. This allows authorities to weigh up all
social, environmental and economic factors in making a decision; the PNTL approach
ensures all relevant factors and management options to be considered.
Therefore the PNTL is not the sole determiner of an activity’s acceptability. The draft
ING, like the current INP, acknowledges that a project should be determined on the
basis of overall social, economic and environmental factors. Therefore the PNTL is
not necessarily the number that will inform a projects statutory noise limit in a
consent or licence.
Limits in statutory documents are the levels that are determined after all factors are
considered and are achievable and accepted as not representing an unacceptable
environmental impact. It is proposed to clarify this point in the ING.
4.10 Acceptable noise levels do not equate to inaudibility
It needs to be made clear at the policy and project level that the draft ING criteria are
not, and should not, be about making industrial noise inaudible. Effectively
communicating this will assist in eliminating false expectations. This point has been
clarified in the draft ING.
4.11 Incorporating where relevant the existing NSW
application notes
Table 13 identifies the existing application notes and whether they are proposed to
be incorporated without change, with modification, or not at all.
26
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
Table 13: Status of the requirements in the NSW Industrial Noise Policy application
notes
Application note
Status
Identifying the existing level of noise from
industry
Incorporated into draft ING (see Fact Sheet
A2)
Assessing noise at industrial/commercial
receivers
Clarified that intrusiveness noise level applies
only to residential premises (see section 2.1
of the draft ING)
When to apply the urban/industrial interface
amenity category
Clarified/incorporated into draft ING (section
2.6)
Identifying the appropriate receiver amenity
category
Additional guidance provided in draft ING
(see Table 2.2)
Amenity criteria in high traffic-noise areas
Clarified/incorporated into draft ING (see
section 2.4.1)
Dealing with cumulative noise from multiple
developments
Clarified/incorporated into draft ING (see
s.2.4.2)
Identifying which of the amenity or intrusive
criteria apply
Clarified/incorporated into draft ING (see
section 2.1)
Assessing background noise levels
Clarified/incorporated into draft ING (see Fact
Sheet A)
When the RBL for evening or night is higher
than the RBL for daytime
Clarified/incorporated into draft ING (see
section 2.3)
Maximum noise levels during shoulder
periods
Redundant on basis of draft approach for
maximum noise level assessment (see draft
ING section 2.5)
Tonality – sliding scale test
Clarified/incorporated into draft ING (see Fact
Sheet C)
Duration correction
Clarified/incorporated into draft ING (see fact
Sheet C)
Determining what weather conditions
should be used when predicting noise
Clarified/incorporated into draft ING (see Fact
Sheet D)
How ‘calm’ is defined
Redundant on basis of revised meteorological
approach (see draft ING Fact Sheet D)
Presenting predicted noise impacts
Redundant on basis of revised meteorological
approach (see draft ING section 6.1.1.)
Noise impact assessment for the
modification of existing industrial premises
Clarified/incorporated into draft ING (see Fact
Sheet D)
Where noise from the existing premises
exceeds the project specific noise level
(PSNL)
Clarified/incorporated into draft ING (see
Section 6)
Prosecution guidelines
Clarified/incorporated into draft ING (see
glossary definition of breach and noncompliance)
Using Appendix D
Deleted from draft ING
How to account for operations that only
occur for part of the day, evening or night
Clarified/incorporated into draft ING through
revised amenity noise level approach
Sleep disturbance
Clarified/incorporated into draft ING (see
section 2.5)
Addressing privately owned haul roads
Clarified/incorporated into draft ING (see
section 1.4)
www.epa.nsw.gov.au
27
Draft Industrial Noise Guideline Technical Background Paper
5 References
1.
Broner N 2011, A simple outdoor criterion for assessment of low frequency
noise emissions. Acoustics Australia, vol 39, 1–7.
2.
Draft BS4142, 2014 Method for rating industrial noise affecting mixed
residential and industrial areas, BSI Standards Limited, London.
3.
European Environment Agency, 2010, Good practice guide on noise exposure
and potential health effects, EEA Technical Report No. 11/2010.
4.
Hansen K, Henrys N, Hansen C, Doolan C and Moreau D 2012, Wind Farm
Noise – What’s a reasonable limit in rural areas. Proceedings of Acoustics
2012, Fremantle.
5.
Hessler G Jr. 2004, Proposed criteria in residential communities for lowfrequency noise emissions from industrial sources, Noise Control Engineering
Journal 52 (4), Virginia, USA.
6.
Hoffmeyer D and Jakobsen J 2010, Sound insulation of dwellings at low
frequencies. Journal of Low Frequency Noise, Vibration and Active Noise
Control, vol. 29 (1) 15–23.
7.
Horonjeff RD 1997, Attitudinal responses to changes on noise environments in
residential communities. HMH Report 293350.
8.
ISO 2003, ISO 226, 2003 – Acoustics – Normal equal loudness level contours.
International Organization for Standardization, Geneva.
9.
ISO 2003, Acoustics – Description, measurement and assessment of
environmental noise – Part 1 Basic Quantities, ISO 1996.1:2003, International
Organization for Standardization, Geneva.
10.
ISO 2007, Acoustics – Description, measurement and assessment of
environmental noise – Determination of environmental noise levels, ISO
1996.2:2007, International Organization for Standardization, Geneva.
11.
Langdon J and Griffiths I 1982, Subjective effects of traffic noise exposure. II:
Comparison of noise indices, Response scales and the effects of changes in
noise levels. Journal of Sound and Vibration 2, vol. 82, 171–180.
12.
Jakobsen J 2012. Danish regulation of low frequency noise for wind turbines.
Journal Low Frequency Noise and Vibration, vol. 31(4).
13.
Job RFS 1996, The Influence of subjective reactions to noise on health effects
of the noise, Environment International, vol. 22(1), 93–104.
14.
Jones K, Rhodes DP 2013, Aircraft noise, sleep disturbance and health effects:
a review. Environmental Research and Consultancy Department (ECRD
Report 1208) Civil Aviation Authority, United Kingdom.
15.
Miedema HME and Vos H 2004, Noise annoyance from stationary sources:
Relationships with exposure metric DENL and t heir confidence intervals.
Journal Acoustical Society America, vol. 116, 334–343.
16.
Moorhouse A, Waddington D and Adams M 2011, Procedure for the
assessment of low frequency noise complaints. DEFRA (Contract No.
NANR45, Revision 1), DEFRA (Contract No. NANR45, Revision 1), London.
28
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
17.
NSW EPA, 2000, New South Wales Industrial Noise Policy, NSW Environment
Protection Authority, Sydney.
18.
NSW EPA, 2014, Discussion Paper: Validation of inversion strength estimation
method www.epa.nsw.gov.au/noise/140011invstrength.htm.
19.
NSW EPA, 2013, Rail Noise Policy, NSW Environment Protection Authority,
Sydney.
20.
NSW DECCW, 2012, Road Noise Policy, NSW Department of Environment,
Climate Change and Water, Sydney.
21.
Passchier–Vermeer W, Vos H, Steenbekkers J, van der Ploeg F, Groothuis–
Oudshoom K 2002, Sleep disturbance and aircraft noise exposure. TNO Inro 4
Report No. 2002.027, Leiden.
22.
Tachibana H, Yano H, Fukushima A and Sueoka S 2014, National field
measurements of wind turbine noise in Japan. Noise Control Engineering
Journal 62(2), March–April 2014.
23.
van Kamp I and Brown L 2003, Response to changed dose of environmental
noise: Diverse results and explanations in the literature. Proceedings of ICBEN
2003 CDROM, Rotterdam.
24.
World Health Organization, 2009, Night noise guidelines for Europe, WHO,
Geneva.
25.
Manning CJ 1981, The propagation of noise from petroleum and petrochemical
complexes to neighbouring communities, Report 4/81, CONCAWE, Den Haag.
26.
ISO 1995, ISO9613-2: 1996 – Attenuation of sound during propagation
outdoors. Part 2: General method of propagation, International Standard.
International Organization for Standardization, Geneva
27.
Shindo S, Ishikawa T and Kunimatsu S 2008, Activity report of the Working
Group on evaluation of habitability to low frequency noise and vibration,
Architectural Institute of Japan, 13th International meeting on low frequency
noise and vibration and its control, Tokyo Japan 21–23 October, 2008.
28.
Kelley ND 1 987, A proposed metric for assessing the potential of annoyance
for wind turbine low frequency noise emissions. SERI/TP-217-3261 UC
Category: 60 DE88001113, San Francisco, California.
29.
ISO 2005, ISO 389-7: 2005, Acoustics — reference zero for the calibration of
audiometric equipment — part 7: reference threshold of hearing under free-field
and diffuse-field listening conditions, International Organization for
Standardization, Geneva.
30.
Leventhall HG 2004, Low frequency noise and annoyance. Noise Health
2004;6:59–72.
31.
NSW Government, 2014: Voluntary Land Acquisition and Mitigation Policy, for
State Significant Mining, Petroleum and Extractive Industry Developments.
NSW Government, Sydney.
www.epa.nsw.gov.au
29
Draft Industrial Noise Guideline Technical Background Paper
Attachment 1 – Original justification for the minimum
assumed background level in INP, 2000
Original basis for an assumed background of 30 dB(A) where the background is
measured at less than 30 dB(A)
There are three parts that are worthy of mention in gaining an understanding of why a
minimum background noise level of 30 dB(A) is applied:
•
first, an explanation of the noise criteria, in particular, how the ‘background plus
5 dB’ applies to rural areas;
•
second, why a minimum background noise level of 30 dB(A) is used; and
•
third, the variation in community response to noise and how this relates to the
‘background plus 5 dB’ criterion.
EPA criteria
The noise criteria used by the EPA to assess noise amenity implications for any new
proposal consists of two components. These components are:
•
that the contributed LAeq noise levels emitted by the development not exceed
the LA90 background noise level by more than 5dB – protects against intrusive
impacts; and
•
that the noise contributed by the new proposal not raise the ambient noise level
of the area above the target amenity noise criteria for the appropriate land-use to
account for noise amenity effects – protects amenity by preventing continually
increasing ambient levels.
This process is described in the NSW Industrial Noise Policy. It should be noted that
both components of the criteria are applied and the level of noise control is normally
set by whichever component of the noise criteria is the most stringent.
In rural areas the first component is generally the most stringent and hence, typically
forms the criterion for new proposed industries in these areas. This criterion is based
on the results of social surveys that have indicated that noise from any particular
source will be noticeable (audible) to many people in the community when that noise
exceeds the background noise level by more than 5 dB.
Minimum background
The EPA currently specifies that where the existing background noise level is below
30 dB(A), the figure of 30 dB(A) should be applied as the background level. There are
two basic reasons for stipulating a minimum background noise level of 30dB(A). The
first is to ensure that amenity is protected and the second is to avoid the situation
where applying a very low background noise level would not improve the level of
protection but may impose very strict requirements on a new development.
Currently, there is no real evidence in the literature to suggest that noise impacts on
amenity occur at levels lower than 35 dB(A), and in using the minimum background
noise level of 30 dB LA90, the intrusive noise criterion becomes 35 dB LAeq. The EPA is
also conscious that the World Health Organization (WHO) (eds. Berglund & Lindvall,
1995) recommends the use of a steady internal noise level of 30–35 dB(A) for
continuous sources and a maximum internal level of 45 dB(A) for intermittent noise
sources to protect against sleep disturbance.
30
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
Accounting for a noise reduction of approximately 10 dB through an open window,
these levels equate to steady external noise levels of 40–45 dB(A) for continuous
noise and a maximum external level of 55 dB(A) for intermittent noise.
These ‘recommended’ external noise levels are at least 5 to 10 dB above the
‘background + 5 dB’ criterion level of 35 dB(A). Hence, the minimum background of 30
dB(A) is considered by the EPA to be a sufficiently conservative figure for the
protection of community amenity.
Community response
Because of the widely varying nature of community reaction to noise it is difficult to
assign a criterion level that would satisfy everyone, in fact research shows that there
is no such level.
Applying a minimum background noise level of 30 dB(A) it is a very conservative
criterion, hardly used elsewhere in the world. The minimum background of 30 dB(A)
rule aims to bring EPA policy more in line with world practice without compromising
community amenity.
This is not to say that a noise would be inaudible at a level of 35 dB(A). However, it
should be noted that just because a person can hear a sound it does not necessarily
follow that their amenity will be affected to an unacceptable level. There are many
noises we hear every day that do not cause annoyance.
Many other non-acoustic factors play a significant role in how people react to the
noise they hear. Kosten & Van Os (1962) presented a good summary of the factors
that determine whether or not an individual will complain about noise:
•
characteristics of the noise (sound pressure level, frequency content, pure
tones, continuous or intermittent, impulsive, low frequency content)
•
characteristics of the individual (personal likes and dislikes)
•
characteristics of the environment (very quiet suburban, suburban, residential
urban, near industry etc)
•
miscellaneous circumstances (noise avoidance possibilities, cognitive noise,
unpleasant associations etc)
•
human activity (sleep, communicating, reading, working, listening to radio and
TV, recreation) being interrupted.
Even with the variation in response to noise and the range of factors that can
influence the response on the basis of available scientific evidence, the setting of a
minimum background noise level of 30 dB(A) will ensure that the vast majority of the
community is protected from unacceptable impacts on their amenity.
References
a.
Berglund B and Lindvall T (eds) 1995, Community Noise, Stockholm
University and Karolinska Institute Archives of the Centre for Sensory Research on
behalf of the World Health Organization, vol. 2(1), ISBN 91-887-8402-9.
b.
Kosten CW and Van Os GJ 1962, Community reaction criteria for external
noises, National Physical Laboratory Symposium 12, pp 373–382.
c.
NSW EPA 1999, New South Wales Industrial Noise Policy, Environment
Protection Authority, Sydney.
d.
WHO 1999, Guidelines for Community Noise, World Health Organization,
Geneva.
www.epa.nsw.gov.au
31
Draft Industrial Noise Guideline Technical Background Paper
Attachment 2: Analysis of proposed amenity noise level approach
Detailed analysis of cumulative outcomes of moving to a revised amenity noise level approach based on ANL – 5dB. Based on rural residential
night-time acceptable noise level (ANL).
ANL
Existing
industrial
noise,
dB(A)
INP
Amenity
1
criteria ,
dB(A)
INP
amenity
2
criteria ,
dB(A)
Proposed
amenity
criteria (ANL5dB), dB(A)
Proposed
amenity
criteria minus
1
INP , dB(A)
Change in cumulative
industrial noise level
(Proposed amenity
1
criteria relative to INP ),
dB(A)
Proposed
amenity criteria
2
minus INP ,
dB(A)
Change in
cumulative industrial
noise level
(Proposed amenity
criteria relative to
2
INP ), dB(A)
40
31
40
40
35
–5
–4.1
–5
–4.1
40
32
40
40
35
–5
–3.9
–5
–3.9
40
33
40
40
35
–5
–3.7
–5
–3.7
40
34
39
39
35
–4
–2.7
–4
–2.7
40
35
38
38
35
–3
–1.8
–3
–1.8
40
36
38
38
35
–3
–1.6
–3
–1.6
40
37
37
37
35
–2
–0.9
–2
–0.9
40
38
36
36
35
–1
–0.4
–1
–0.4
40
39
34
34
35
1
0.3
1
0.3
40
40
32
32
35
3
0.6
3
0.6
40
41
32
32
35
3
0.5
3
0.5
40
42
30
32
35
5
0.5
3
0.1
40
43
30
33
35
5
0.4
2
0.2
40
44
30
34
35
5
0.3
1
0.2
40
45
30
35
35
5
0.3
0
0.3
40
46
30
36
35
5
0.2
–1
–0.3
32
www.epa.nsw.gov.au
Draft Industrial Noise Guideline Technical Background Paper
ANL
Existing
industrial
noise,
dB(A)
INP
Amenity
1
criteria ,
dB(A)
INP
amenity
2
criteria ,
dB(A)
Proposed
amenity
criteria (ANL5dB), dB(A)
Proposed
amenity
criteria minus
1
INP , dB(A)
Change in cumulative
industrial noise level
(Proposed amenity
1
criteria relative to INP ),
dB(A)
Proposed
amenity criteria
2
minus INP ,
dB(A)
Change in
cumulative industrial
noise level
(Proposed amenity
criteria relative to
2
INP ), dB(A)
40
47
30
37
35
5
0.2
–2
–0.3
40
48
30
38
35
5
0.1
–3
–0.3
40
49
30
39
35
5
0.1
–4
–0.4
40
50
30
40
35
5
0.1
–5
–0.4
40
51
30
41
35
5
0.1
–6
–0.4
40
52
30
42
35
5
0.1
–7
–0.4
40
53
30
43
35
5
0.0
–8
–0.4
40
54
30
44
35
5
0.0
–9
–0.4
40
55
30
45
35
5
0.0
–10
–0.4
Note: 1. Based on current Industrial Noise Policy (INP) amenity criteria approach using corrections to ANLs in Table 2.1. 2. Based on current INP amenity criteria approach using corrections to
existing industrial noise levels. 3. Negative numbers in columns 7 and 9 show that cumulative industrial noise will be lower whereas positive numbers denote that cumulative noise will be higher.
www.epa.nsw.gov.au
33