Proposed Rule 5 - Vermontpsb.vermont.gov/sites/psbnew/files/doc_library/RSG.pdf · Montpelier, VT...
Transcript of Proposed Rule 5 - Vermontpsb.vermont.gov/sites/psbnew/files/doc_library/RSG.pdf · Montpelier, VT...
COMMENTS & TECHNICAL INFORMATION
PROPOSED RULE 5.700
5.11.2017
55 Railroad Row White River Junction, VT 05001
802.295.4999 www.rsginc.com
PREPARED FOR:
VERMONT PUBLIC SERVICE BOARD
SUBMITTED BY:
RSG
55 Railroad Row 802.295.4999 White River Junction, Vermont 05001 www.rsginc.com
May 11, 2017
Vermont Public Service Board
112 State Street, 4th Floor
Montpelier, VT 05620-2701
RE: Rule 5.700, Wind Generation Facility Sound Rulemaking
Dear Public Service Board:
RSG would like to thank the Board for the opportunity to present at the workshop on May 4. As
stated at the workshop, we appreciate the hard work the Board is putting into this process, and we
recognize the difficult nature of the technical material covered in the proposed rule.
This document provides information that supports some of what we presented at the workshop and
some additional information based on topics discussed by other parties at the workshop. We also
provide in the Appendix a copy of our presentation from May 4, and a copy of the proposed rule
with suggested changes. We hope you find this information useful as you continue to develop the
proposed rule.
Sincerely,
RSG
EDDIE DUNCAN, INCE BD. CERT.
Director
PROPOSED RULE 5.700
PREPARED FOR: VERMONT PUBLIC SERVICE BOARD
i
CONTENTS
1.0 INTRODUCTION ............................................................................................................................... 1
2.0 SOUND PROPAGATION MODELING ............................................................................................. 2
3.0 BUILDING ATTENUATION WITH WINDOWS OPEN AND CLOSED ............................................ 4
3.1 | Vermont Measurments .................................................................................................................. 4
3.2 | Literature ....................................................................................................................................... 4
3.3 | Building Attenuation Summary ...................................................................................................... 6
4.0 ANNOYANCE RESEARCH SUMMARY .......................................................................................... 7
5.0 NOISE REDUCED OPERATION (NRO) & EFFECTIVE LIMITS ..................................................... 9
6.0 INFRASOUND ................................................................................................................................ 10
7.0 SUMMARY ...................................................................................................................................... 13
8.0 WORKS CITED ............................................................................................................................... 14
APPENDIX A. RSG PRESENTATION TO THE PSB AT THE RULEMAKING WORKSHOP .............. 16
APPENDIX B. RSG SUGGESTED EDITS TO THE PROPOSED RULE .............................................. 60
List of Figures
FIGURE 1:COMPARISON OF RECENT WIND TURBINE DOSE-RESPONSE STUDY RESULTS .................................................. 8
FIGURE 2: COMPARISON OF TURBINE ON (TON) AND TURBINE OFF (TOFF) SOUND LEVELS WITH CONFIDENCE
INTERVALS AT SITE 8B (MOUNTAINOUS MULTI-TURBINE SITE) FOR THREE WIND SPEED RANGES AT 650 METERS ..... 11
FIGURE 3: WIND TURBINE SOUND PERCEPTION LEVEL RESULTS FROM YOKOYAMA ET AL, 2014 ................................... 12
1
1.0 INTRODUCTION
At the rulemaking workshop on May 4, 2017, RSG presented information on five topics that
we thought would be helpful to the Board including:
• Post-construction compliance measurements
• Aesthetics, noise annoyance, and acoustical metrics
• Outdoor-to-indoor attenuation
• Noise reduced operation of wind turbines
• PSB precedent & the proposed rule – acoustical context
A copy of the presentation is attached to this document as Appendix A.
In addition to the topics discussed during the presentation, we are providing in this
document further technical and relevant on:
• Sound propagation modeling
• Building attenuation with windows open and closed
• Annoyance research for wind turbines
• Noise reduced operation of wind turbines
• Infrasound from wind turbines.
Lastly, given the information presented at the workshop and in this document, we provide,
in Appendix B, an edited version of the proposed rule with our suggested changes.
Vermont Public Service Board Proposed Rule 5.700
2
2.0 SOUND PROPAGATION MODELING
Although, ISO 9613-2 is the most widely accepted wind turbine noise modeling algorithm,
other algorithms that have been used in wind power projects include:
• CONCAWE;
• Nord2000;
• Harmonoise; and
• NZS 6808-1998.
Both Nord2000 and NZS 6808-1998 are the approved method for specific countries (New
Zealand and Australia for NZS 6808-1998 and Nordic countries for Nord2000). NZS 6808-
1998 is a simplified method that assumes hemispherical sound propagation and uses the air
absorption method from ISO 9613-2. Nord2000 is more in-depth, complicated, and is of
similar scope to ISO 9613-2.
Harmonoise, was originally based on Nord 2000 with some refinements and was developed
over several years with the aim of becoming the standard algorithm for noise predictions in
Europe. The algorithm is available as an open source code and is implemented in several
noise prediction software packages. Harmonoise allows modeling of various meteorological
conditions, beyond the capabilities of ISO 9613-2, along with more sophisticated methods
of handling shielding and ground effects. The use of this model for wind turbine noise has
been limited, with few studies validating its accuracy.
CONCAWE was originally developed for the petroleum energy industry in Europe.
Characteristics of the model that are unique, are the ability to predict sound levels for
particular wind speeds and stability classes. The model has been used internationally for
wind turbine noise with some validation studies, though ISO 9613-2 is still more widely used
and validated.
None of these algorithms were originally developed for wind turbine noise prediction.
In the United States ISO 9613-2 is by far the most common algorithm used for sound
propagation modeling, particularly for wind turbine noise. To our knowledge, the only other
algorithm used is CONCAWE, but only in conjunction with ISO 9613-2 for special cases of
modeling annualized sound levels under varying meteorological conditions.
ISO 9613-2 assumes downwind sound propagation between every source and every
receptor, consequently, all wind directions, including the prevailing wind directions, are
taken into account. Turbines should be modeled as a single point source, located at the hub-
height of the turbine.
Selection of appropriate model input parameters is crucial to achieve conservative yet
accurate modeling results with ISO 9613-2. There have been several studies comparing
3
measured sound levels with sound propagation modeling results using ISO 9613-2.1,2,3,4
Results have shown that modeling is most accurate with the assumption of either mixed
(G=0.5) or hard (G=0) ground. Mixed ground (G=0.5), with a 2 dB 95 percent confidence
interval added to the sound power, is appropriate where the ground between the turbines
and receivers is flat or constant-gradient.
For situations where the gradient between the source and receiver is concave, hard ground is
the most appropriate, without the 95 percent confidence interval added. The modeled sound
power should be the maximum guaranteed sound power published by the turbine
manufacturer, that does not include the “K factor.” This is usually called the “apparent”
sound power.
Different receiver heights result in different interference patterns. The 4-meter (13-foot)
receiver height mimics the height of a second story bedroom and generally results in 1 to 2
dB higher predictions than a 1.5-meter (5-foot) receiver height. If a 1.5-meter microphone
height is used for post-construction monitoring, then model results should be produced at
both the 4-meter and 1.5-meter receiver height, so the model can be verified with the post-
construction data collection.
Results calculated with the parameters discuss here represent the highest 1-hour equivalent
average sound level (Leq1h) that will be emitted by a wind power project.
1 Duncan, E., and Kaliski, K., “Improving Sound Propagation Modeling for Wind Power Projects”, Acoustics ’08, 2008, Paris, France. 2 Bowdler, Dick et al,. “Prediction and Assessment of Wind Turbine Noise: Agreement about Relevant Factors for Noise Assessment from Wind Energy Projects.” Acoustics Bulletin. 34(2), pp. 35-37.
3 Evans, Tom and Cooper, Jonathan. “Comparison of Predicted and Measured Wind Farm Noise Levels and Implications for Assessments of New Wind Farms.” Acoustics Australia: April 2012. Vol. 40, No. 1. 4 RSG, et al., “Massachusetts Study on Wind Turbine Acoustics,” Massachusetts Clean Energy Center and Massachusetts Department of Environmental Protection, 2016 Chapter 6.
Vermont Public Service Board Proposed Rule 5.700
4
3.0 BUILDING ATTENUATION WITH WINDOWS OPEN AND CLOSED
Recent Vermont Public Service Board (PSB) dockets for wind power projects have included
sound level limits for both inside and outside of residences. This has followed the World
Health Organization (WHO) Guidelines for Community Noise (1999) that have based
exterior sound level guidelines on interior health-based sound level thresholds and assumed
building façade attenuations. The WHO assumes that residential building facades with the
windows partially open will provide 15 dB of attenuation. As a result, WHO derives their
recommended 45 dBA L(8hr) exterior sound level limit from an ideal threshold of 30 dBA
L(8hr) inside bedrooms.
Here we discuss measured and assumed façade attenuations with windows open and closed.
3.1 | VERMONT MEASURMENTS
In Vermont, there have been two cases where a residence façade was tested to determine
windows-open and close attenuation – Brouha in Sheffield and Fitzgerald in Georgia. In
Brouha, the test resulted in a sound level reduction of less than 5 dB with the windows open.
The tested residence was an anomalous example, with a large window located in a small
bedroom, and window panes that could be rotated until perpendicular with the façade, and
facing directly at the wind project. In effect, this allowed a large portion of the wall for this
room to be fully open to the outside, resulting in a low attenuation. Measurements with the
windows closed showed a sound level reduction of 25 dB.
At Fitzgerald, RSG conducted outdoor-indoor noise reduction measurements near Georgia
Mountain Community Wind (GMCW) during the winter of 2013. Results from this
measurement are detailed in the RSG memo to Martha Staskus.5 Results from this memo
indicated that that the bedroom façade with windows open resulted in an overall sound level
reduction of 15 dB, with the windows closed, the sound level reduction was 29 dB.
3.2 | LITERATURE
To determine what more “typical” windows-open and window-close façade attenuations are,
RSG conducted a literature search. Most literature lists façade attenuation as either a
weighted overall attenuation, as defined by different standards, or as the sound level
reduction of a certain sound source. The sound level reduction is achieved by subtracting the
sound attenuation value of a façade from the spectrum of a particular sound source (such as
aircraft, traffic, and railway).
Waters-Fuller and Lurcock (2007)
A thorough study on this subject was performed by Tim Waters-Fuller and Daniel Lurcock.6
The study looked at the laboratory-measured sound attenuation of several different window
types in various operational positions that included closed, closed but unlatched, and three
5 RSG. “Complaint Resolution Update – Fitzgerald Residence.” April 16, 2013 6 Waters-Fuller and Lurcock, Department for Environment, Food and Rural Affairs, UK, 2007
5
levels of openness (0.05 m2, 0.1 m2, and 0.2 m2). This was reported as either overall sound
attenuation values (Rw from ISO 717) or sound level reductions for various transportation
sound sources. Along with actual measurement results there is an extensive literature review
of other studies on the same subject. This literature review shows sound level differences for
aircraft noise between 9 and 11 dB with a 200 mm (8 inch) window opening and between 14
and 16 dB for a 25 mm (1 inch) window opening. A different study found a level difference
of 16 dB for a 100 mm opening. For road traffic, a sound level difference of 9 to 11 dB was
found when the window opening was 7 to 8 percent of the façade. Values for windows-
closed sound level reduction ranged from 21 to 30 dB for road traffic noise.
Results from the actual study showed weighted level differences between 7 and 26 dB with
windows open 0.2 m2. Most values were in the 10 to 17 dB range. Since this paper did not
look at the specific sound level difference for wind turbine noise, RSG used the sound level
attenuation for the worst-case window configuration (0.2 m2 opening) and applied it to a
wind turbine sound spectrum. The result was a sound level difference of 14 dB. For the best
case window construction the level difference was 18 dB. For the windows closed situation,
weighted level differences ranged from 33 to 46 dB. The paper does not show octave band
attenuation levels for a windows closed condition, so the values shown are generalized
reductions.
Hayes Mckenzie Partnership (2006)
In 2006 Hayes Mckenzie Partnership, Ltd published a report for the UK’s Department of
Trade and Industry (DTI).7 The report primarily concerned measurements of wind turbine
low frequency noise, but included the only in-situ measurement of windows-open wind
turbine sound level reduction measurement that we are aware of. There was only one
location, where a windows-open value was measured, which was 10 dB. Windows-closed
attenuation was measured at two locations, with sound level reduction values of 15 and 16
dB.
Environmental Protection Agency (1974)
The United States Environmental Protection Agency (EPA) developed typical windows-
open sound level reductions for transportation noise in both warm and cold climates.8 For
warm climates a 12 dB sound level reduction was specified and for cold climates a 17 dB
sound level reduction was specified for cold climates. This assumes an open area of 2 ft2
(0.19 m2). With the windows closed, a sound level reduction of 24 dB was specified for
warm climates and 27 dB for cold climates.
7 Hayes McKenzie Partnership, Department of Trade and Industry, UK, 2006. 8 U.S. Environmental Protection Agency Office of Noise Abatement and Control. Information on Levels of Environmental Noise Requisite to Protect Public Health and Welfare. Arlington, Virginia, 1974.
Vermont Public Service Board Proposed Rule 5.700
6
Federal Highway Administration (2011)
The Federal Highway Administration has a specified a 10 dB windows-open “noise
reduction,” used to estimate interior levels of traffic noise.9 There is no mention in the
document of what this value is based on. Note that this is paired with a noise abatement
criteria level of 67 dBA Leq1h during the maximum traffic noise hour. With the windows
closed, noise reductions between 20 and 35 dB are specified depending on the type of
window and type of building construction.
3.3 | BUILDING ATTENUATION SUMMARY
In the majority of tests, windows-open wind turbine sound level reduction, were 10 dB or
greater. Sound level reductions in excess of 15 dB are also possible and in some cases, even
20 dB may occur. In light of this, the Brouha attenuation of less than 5 dB is just one data
point at the extreme end of the low attenuation range and shouldn’t be used as a basis for
developing noise policy.
With the windows closed, sound level reductions range between 16 and 46 dB with most
values ranging between 20 and 30 dB. This value will also depend on a variety of factors.
The windows open and closed sound level reductions will be dependent on many factors
including:
• The size of the window;
• The window type;
• The amount the window is open, if it is open;
• The relative area of the façade occupied by the window;
• Sound insulation of the wall;
• Sound insulation of the window;
• Bedroom size;
• Bedroom furnishings (i.e. sound absorption); and,
• Bedroom orientation relative to the sound source.
9 Federal Highway Administration. "Highway Traffic Noise: Analysis and Abatement Guidance." 2011.
7
4.0 ANNOYANCE RESEARCH SUMMARY
Old and Kaliski, 2017 compares results from recent major dose-response studies that looked
at the relationship between modeled wind turbine sound pressure levels and the subjective
response of residents exposed to those levels.10 There have been four large, well performed
dose-response studies or series of dose-response studies performed since the year 2000. Of
these, three of the studies provide detailed information on how turbine-only sound levels
were modeled. All of these studies looked at sites that were generally rural and would have
background sound levels that are similar to Vermont.
To determine the level of agreement between the studies, differences between different
sound propagation modeling methods and parameters, and the relationship between
different sound level metrics were determined. Results were then normalized to a common
sound propagation modeling method, parameter set, and sound level metric. The sound
propagation modeling algorithm used was ISO 9613-2, with mixed ground attenuation
(G=0.5), a 4-meter receiver height, and no measurement uncertainty added to the wind
turbine sound power. These parameters are used to represent a median Leq1h for wind
turbine noise in flat and constant-gradient terrain with all turbines emitting maximum sound
power. These parameters are approximately 2 dB less conservative than what has typically
been used in Vermont for wind power project permitting. Modeling for Vermont projects
has tended to model a worst-case result instead of a median result. So, a 40 dBA result in this
paper would be equivalent to the 42 dBA modeled result for most Vermont projects.
Results are shown in Figure 1. For an exterior level of 40 dBA Leq1h (42 dBA using typically
modeled parameters in Vermont), between 7 and 10 percent are expected to be highly
annoyed outdoors and approximately 4 percent are expected to be highly annoyed indoors.
At an exterior level of 35 dBA Leq1h (37 dBA using VT modeling parameters), between 3 and
5 percent are expected to be highly annoyed outdoors and approximately 1 percent indoors.
For the sound level limit used at current utility wind power projects in Vermont (45 dBA
Leq1h) the percent highly annoyed is expected to be between 10 and 17 percent outdoors and
approximately 7 percent indoors. These levels of “percent highly annoyed” are typical for
other sources of noise. Some populations will exhibit higher or lower levels of response,
depending on non-acoustic factors.
10 Old, I., and Kaliski, K., (2017). Wind turbine noise dose response – comparison of recent studies. Proceedings of the 7th International Conference on Wind Turbine Noise, Rotterdam.
Vermont Public Service Board Proposed Rule 5.700
8
FIGURE 1:COMPARISON OF RECENT WIND TURBINE DOSE-RESPONSE STUDY RESULTS
9
5.0 NOISE REDUCED OPERATION (NRO) & EFFECTIVE LIMITS
It was not until after the rulemaking workshop on May 4 that we realized that Section 5.705
of the proposed rule does not allow for the use of NRO mode in the pre-construction
modeling. It states:
“[…] shall include a sound model developed for the proposed facility that reports the
expected maximum project sound levels, without using NRO mode, experienced out to a
distance where such levels are not greater than 30 dBA.” (emphasis added)
At the workshop, we contemplated how the limits to NRO technology, which can generally
achieve a 1 to 4 dB reduction, effectively make the limits in the proposed rule 35 dBA
nighttime and 39 dBA daytime. If NRO is not allowed per the draft rule, then the effective
limit would actually be 35 dBA nighttime and 35 dBA daytime. Having a different daytime
limit and nighttime limit and disallowing the use of NRO, means that the only tool a
developer has to meet the quieter nighttime limit is turbine shutdowns.
If the rule disallows the use of NRO mode, it takes away one of the noise control engineer’s
primary tools to mitigate potential impacts. It would be like making a rule for cars to be
quiet, but disallowing the use of mufflers.
Vermont Public Service Board Proposed Rule 5.700
10
6.0 INFRASOUND
Wind turbine-caused infrasound has become a frequently discussed topic during wind power
project permitting. The assertion has been that infrasound from wind turbines, even though
it is below the threshold of human hearing, is capable of being sensed and can cause a series
of health effects for some people. These concerns have led to research on the subject. Some
studies have focused on measuring wind turbine infrasound levels and others have studied
the human body’s ability to detect infrasound through hearing and other paths.
Recent measurement projects have shown that wind turbine-caused infrasound is several
orders of magnitude below measured hearing thresholds. The graphs in Figure 2 were taken
from the Massachusetts Study on Wind Turbine Acoustics.11 This shows that while wind
turbine-caused infrasound is measurable, it is well below established hearing thresholds at
receiver distances. The exact levels measured vary depending on the turbine model, but wind
turbine sound is almost always below measured audibility thresholds for 1/3 octave bands
with the center frequency below 25 Hz.
The Japanese Working Group on wind turbine acoustics performed a study where a
recording of wind turbine noise was filtered to eliminate different parts of the sound
spectrum and then reproduced to test subjects to test the audibility of wind turbine-caused
sound.12 Results from the study, shown in Figure 3, indicate that the audibility of wind
turbine sound emulates that of measured pure-tone thresholds. A paper by Tonin and Brett
looked at perception of infrasound and the influence expectations have regarding the effects
of infrasound on humans.13 Results showed, that any response to infrasound was more tied
to expected effects than actual infrasound exposure.
11 RSG. "Massachusetts Study on Wind Turbine Acoustics." 2016. 12 Yokoyama, Sakae, Shinichi Sakamoto and Tachibana Hideki. "Perception of Low Frequency
Components in Wind Turbine Noise." Noise Control Engineering Journal 65.5 (2014): 295-305.
13 Tonin, Renzo, James Brett and Ben Colaguiri. "The Effect of Infrasound and Negative Expectations to Adverse Pathological Symptons from Wind Farms." Journal of Low Frequency Noise, Vibration, and Active Control 0.0 (2016): 1-14.
11
FIGURE 2: COMPARISON OF TURBINE ON (TON) AND TURBINE OFF (TOFF) SOUND LEVELS WITH CONFIDENCE INTERVALS AT SITE 8B (MOUNTAINOUS MULTI-TURBINE SITE) FOR THREE WIND SPEED RANGES AT 650 METERS
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
0.5 0.63 0.8 1 1.25 1.6 2 2.5 3.15 4 5 6.3 8 10 12.5 16 20
So
un
d P
res
su
re L
eve
l (d
BZ
)
1/3 - Octave Band Frequencies (Hz)
3-6 m/s (T - On) 3-6 m/s (T - Off)
90 dBG contour
Watanabe and Moller - 1990
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
0.5 0.63 0.8 1 1.25 1.6 2 2.5 3.15 4 5 6.3 8 10 12.5 16 20
So
un
d P
res
su
re L
eve
l (d
BZ
)
6-9 m/s (T - On) 6-9 m/s (T - Off)
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
0.5 0.63 0.8 1 1.25 1.6 2 2.5 3.15 4 5 6.3 8 10 12.5 16 20
So
un
d P
res
su
re L
eve
l (d
BZ
)
>9 m/s (T - On) >9 m/s (T - Off)
Vermont Public Service Board Proposed Rule 5.700
12
FIGURE 3: WIND TURBINE SOUND PERCEPTION LEVEL RESULTS FROM YOKOYAMA ET AL, 2014
13
7.0 SUMMARY
Based on the information presented at the workshop and in this document, we offer the
following comments:
1. Compliance monitoring must account for background sound levels. (see Appendix
A and the Department’s presentation by Aercoustics)
2. The current PSB precedent of 45 dBA one-hour maximum protects against public
health impacts and undue adverse impact on aesthetics per the Act 250 framework.
3. Based on our experience, we think Kingdom Community Wind and other projects
built under the 45 dBA precedent would not have been built under the proposed
rule.
4. If a different limit is contemplated for aesthetics, an evening limit (5 p.m. to 9 p.m.)
that differs from the rest of the 24-hour period may make the most sense.
5. Use of NRO in pre-construction modeling should be allowed by the rule.
6. If the rule is to have different specified limits by time-of-day, decrease the difference
to not more than 4 dB. The proposed rule currently has a difference of 7 dB
between daytime and nighttime.
7. Most outdoor-to-indoor attenuation values from a variety of papers show a
reduction of 10 to 17 dB with windows open.
Vermont Public Service Board Proposed Rule 5.700
14
8.0 WORKS CITED
Evans, Tom and Jonathan Cooper. "Comparison of Predicted and Measured Wind Farm
Noise Levels and Implications for Assessments of New Wind Farms." Acoustics
Australia (2012): 28-36.
Federal Highway Administration. "Highway Traffic Noise: Analysis and Abatement
Guidance." 2011.
Fukushima, Akinori, et al. "Study on the Amplitude Modulation of Wind Turbine Noise:
Part 1 - Physical Investigation." Internoise. Innsbruck, Austria, 2013.
Hayes Mckenzie Partmership. "The Measurement of Low Frequency Noise at Three UK
Wind Farms." 2006.
Janssen, Sabine, et al. "A Comparison Between Exposure-response Relationships for Wind
Turbine Annoyance and Annoyance Due to other Noise Sources." Journal of the
Acoustical Society of America (2011): 3746-3753.
Kuwano, Sonoko, et al. "Social Survey on Wind Turbine Noise in Japan." Noise Control
Engineering Journal (2014): 503-520.
Michaud, David, et al. "Exposure to Wind Turbine Noise: Perceptual Responses and
Reported Health Effects." Journal of the Acoustical Society of America (2016): 1443-1453.
—. "Self-reported and Measured Stress Related Responses Associated with Exposure to
Wind Turbine Noise." Journal of the Acoustical Society of America (2016): 1467-1479.
Pedersen, Eja and Kerstin Persson-Waye. "Perception and Annoyance Due to Wind Turbine
Noise - a Dose-response Relationship." Journal of the Acoustical Society of America
(2004): 3460-3470.
—. "Wind Turbine Noise, Annoyance and Self-Reported Health and Well-being in Different
Living Environments." Occupational and Environmental Medicine (2007): 480-486.
Pedersen, Eja, et al. "Response to Noise From Modern Wind Farms in the Netherlands."
Journal of the Acoustical Society of America (2009): 634-642.
RSG. "Massachusetts Study on Wind Turbine Acoustics." 2016.
Tonin, Renzo, James Brett and Ben Colaguiri. "The Effect of Infrasound and Negative
Expectations to Adverse Pathological Symptons from Wind Farms." Journal of Low
Frequency Noise, Vibration, and Active Control 0.0 (2016): 1-14.
U.S. Environmental Protection Agency Office of Noise Abatement and Control. Information
on Levels of Environmental Noise Requisite to Protect Public Health and Welfare. Arlington,
Virginia, 1974.
Van Den Berg, Frits. "Criteria for Wind Farm Noise: Lmax and Lden." Acoustics. Paris:
European Acoustical Association, 2008. 4043-4048.
15
Waters-Fuller, Tim and Daniel Lurcock. NANR116: 'Open/Closed Window Research' Sound
Insulation Through Ventilated Domestic Windows. London, England: Department for
Environment, Food and Rural Affairs, 2007.
Yokoyama, Sakae, Shinichi Sakamoto and Tachibana Hideki. "Perception of Low Frequency
Components in Wind Turbine Noise." Noise Control Engineering Journal 65.5 (2014):
295-305.
Vermont Public Service Board Proposed Rule 5.700
16
APPENDIX A. RSG PRESENTATION TO THE PSB AT
THE RULEMAKING WORKSHOP
2
Introduction
Eddie Duncan, Director
• Board Certified, Institute of Noise Control Engineering
• Member of the Acoustical Society of America
- Technical Committee on Architectural Acoustics
• Education:
- M.S. Green Mountain College
Environmental Studies, Focus: Environmental Law &
Policy, Specifically Noise Policy
- B.S. Rensselaer Polytechnic Institute
Engineering Science, Focus: Acoustics
3
Introduction
RSG’s Experience
• Involved in noise assessments of wind power since 1993.
- Maine Land Use Regulatory Commission
• Studied over 80 proposed or installed wind power projects.
- Maine to Hawaii
- Including Deerfield Wind, Kingdom Community Wind, Georgia
Mountain Community Wind, and others in development
• Conduct research on wind turbine acoustics.
- Massachusetts Clean Energy Center
- Lawrence Berkeley National Laboratory (U.S. DOE)
• Staff regularly publish papers and technical presentations on wind
turbine acoustics.
• Staff co-chair of INCE Wind Turbine Technical Activity Committee.
4
Introduction
Presentation Topics
• Post-Construction Compliance Measurements
• Aesthetics, Noise Annoyance, and Acoustical Metrics
• Outdoor-to-Indoor Attenuation
• Noise Reduced Operation of Wind Turbines
• PSB Precedent & the Proposed Rule – Acoustical Context
6
Post-Construction Compliance Measurements
RSG’s experience is that the proposed methodology does not cost less than
other alternatives and will not necessarily yield accurate results.
Proposed Rule’s Economic Impact Statement
• The Board’s rule results in “…Compliance costs that are relatively
lower than other alternatives considered.”
• “…by requiring that monitoring occur under worst-case conditions
where turbine sound levels will be at their loudest output, and
background sound levels at their lowest.”
• Does away with accounting for background sound levels.
• Hypothesizes that the proposed methodology, “…allows for
monitoring campaigns to be of significantly shorter duration…”
7
Post-Construction Compliance Measurements
Proposed rule is similar to Maine’s compliance procedure.
• Arithmetic average of twelve, 10-minute intervals from the same
measurement period. (5.704)
• Measurements when wind turbine sound is dominant.
- Nighttime
- Downwind – within 45° of the acoustic center of the five nearest
turbines
- Maximum surface wind speeds (at 10 meters) of 6 mph or less
- Hub height wind speeds able to generate maximum turbine
sound power ±1 dB
This requires:
• Long-term monitoring similar to other methods because finding these
conditions can be very difficult.
• Installation of a temporary 10-meter mast in a cleared location.
8
Post-Construction Compliance Measurements
Example1: Maine Project
• 4 compliance monitor
locations = 4 wind directions
• Weather forecasts monitored
on a weekly basis for nine
months.
• Monitored over 7 periods for
53 total days.
• Valid Periods
– Monitor A: 7, not 12
– Monitor B: 0, not 12
– Monitor C: 8, not 12
– Monitor D: 0, not 12
9
Post-Construction Compliance Measurements
Example 2: Maine Project• 1 continuous sound monitor
• Valid Periods
– Year 1: 5 days of data analyzed to
find 12 periods
– Year 2: 5 days of data analyzed to
find 12 periods
– Year 3: 11 days of data analyzed
to find 12 periods
– Year 4: 5 days of data analyzed to
find 12 periods
– Year 5: 8 days of data analyzed to
find 12 periods
• Still had to filter out extraneous
events such as bird calls.
10
Post-Construction Compliance Measurements
Proposed rule is similar to Maine’s compliance procedure.
• Arithmetic average of twelve, 10-minute intervals from the same
measurement period. (5.704)
• Measurements when wind turbine sound is dominant.
- Nighttime
- Downwind – within 45° of the acoustic center of the five nearest
turbines
- Maximum surface wind speeds (at 10 meters) of 6 mph or less
- Hub height wind speeds able to generate maximum turbine
sound power ±1 dB
Quite problematic to capture.
Significant data analysis required.
Amounts to a Continuous Monitoring Exercise.
11
Post-Construction Compliance Measurements
Turbines are not the only
sound sources that are aloft
Wind
Gradient
12
Post-Construction Compliance Measurements
Turbines are not the only
sound sources that are aloft
Wind
Gradient
In a forested landscape with
hills or mountains, high
winds aloft and low winds
below results in sound
generated not only from
wind turbines but from the
forest as well.
High winds in a forest, particularly with no
leaves, can easily be confused with wind
turbine sound.
13
Post-Construction Compliance Measurements
Recommendations
1. Account for background sound levels.
a) Turbine shut-down method works well.
b) Shielding method also works if locations are selected properly.
c) Proxy monitor locations are problematic for hilly terrain and
heterogeneous landscapes. Don’t seem to work well in the
Northeast.
2. Keep the current instrumentation, personnel, and calibration
requirements in Section 5.707.
3. Use the post-construction measurements to verify and modify, if
necessary, the pre-construction sound modeling.
15
Aesthetics, Noise Annoyance, & Metrics
Generally, aesthetics is not something the professional acoustics community
studies or talks about.
• Sound quality – typically applied to product design
• Natural & cultural sounds as a natural resource – National Park Service
• Acoustical aesthetics in rural working landscapes – not addressed
Except in Act 250
• Criterion 1 – Air Pollution – Noise covered as a health impact.
• Criterion 8 – Aesthetics
- Noise not explicitly mentioned in the statute
- Case law covers it as an aesthetics issue
16
Aesthetics, Noise Annoyance, & Metrics
Quechee Test
• Developed by landscape architects for the Environmental Board in
Quechee Lakes Corporation, 1985
• Two Part Test
1. Is the project adverse? Does it fit the context of the area?
2. If found to be adverse, Is the project unduly adverse?
a. Does the project violate a clear, written community standard
intended to preserve the aesthetics or scenic natural beauty of
the area?
b. Does the project offend the sensibilities of the average
person?
c. Has the Applicant failed to take generally available mitigating
steps which a reasonable person would take to improve the
harmony of the proposed project with its surroundings?
17
Aesthetics, Noise Annoyance, & Metrics
Quechee Test
b. Does the project offend the sensibilities of the average
person?
Threshold: Would the sound be considered shocking and
offensive to the average person?
18
Aesthetics, Noise Annoyance, & Metrics
Quechee Test
b. Does the project offend the sensibilities of the average
person?
Threshold: Would the sound be considered shocking and
offensive to the average person?
If the Board is considering aesthetics in it’s decision-making
process, is a daytime limit of 42 dBA and a nighttime limit of 35
dBA necessary to keep the average person from being shocked
and offended?
19
Aesthetics, Noise Annoyance, & Metrics
Noise Annoyance
• More commonly studied in acoustics than aesthetics
• Fairly standardized methodologies (ISO/TS 15666:2003)1
- Social surveying methods
- 0 to 100 scale, 28/50/72 - lightly/moderately/highly annoyed2
- Dose-response relationships – at sound level of X dBA, causes % of
a population to be lightly, moderately, or highly annoyed.
• WHO’s Guidelines for Community Noise3
- Serious Annoyance, daytime and evening, 55 dBA Leq16hr
- Moderate Annoyance, daytime and evening, 50 dBA Leq16hr
1. ISO/TS 15666:2003. Acoustics – Assessment of noise annoyance by means of social and socio-acoustic surveys.
2. Miedema, H.M.E., & Vos, H. (2004). Noise annoyance from stationary sources: Relationships with exposure metric day-evening-night level (DENL) and their confidence
intervals. The Journal of the Acoustical Society of America, 116(1), 334-343.
3. Berglund, B., Lindvall, T., & Schwela, D.A. (1999). Guidelines for community noise. World Health Organization.
20
Aesthetics, Noise Annoyance, & Metrics
Noise Annoyance – Wind Turbine Specific Studies
• Several studies, but most use different acoustical metrics.1,2,3
• Results from Swedish & Dutch, Japanese, and Canadian studies have
been normalized to a common metric.4
- A-weighted hourly equivalent average sound level: LAeq1hr
- Modeled using ISO 9613-2, G=0.5, 4 meter high receivers
o These modeling parameters would yield results 2 dB lower than
what is currently used in Vermont.
o Doesn’t include turbine sound power uncertainty.
1. Janssen, S., Vos, H., Eisses, A., & Pedersen, E. (2011). A comparison between exposure-response relationships for wind turbine annoyance and annoyance due to other
noise sources. The Journal of the Acoustical Society of America, 130(6), 3746-3753.
2. Kuwano, S., Yano, T., Kageyama, T., Sueoka, S., and Tachibana, H., (2014). Social survey on wind turbine noise in Japan. Noise Control Engineering Journal, 62(6),
503-520.
3. Michaud, D., et.al. (2016). Exposure to wind turbine noise: Perceptual responses and reported health effects. The Journal of the Acoustical Society of America, 139(3),
1443-1454.
4. Old, I., and Kaliski, K., (2017). Wind turbine noise dose response – comparison of recent studies. Proceedings of the 7th International Conference on Wind Turbine Noise,
Rotterdam.
21
Aesthetics, Noise Annoyance, & Metrics
• At 43 dBA (equivalent to Vermont’s 45 dBA one-hour maximum):
- 15% highly annoyed - Swedish, Dutch, & Health Canada Studies
- 10% highly annoyed – Japanese Study
22
Aesthetics, Noise Annoyance, & Metrics
Noise Annoyance – Additional Observations
• Attitudinal variables strongly affect noise annoyance1,2,3,4
- Fear
- Belief that the noise could be prevented
- Perceived fairness in the decision making process
- Awareness of non-noise problems related to the noise source
- Perceived importance of the source of noise
- Personal benefit
• Annoyance occurs primarily when spending time outdoors with
activities such as relaxing, picnicking, or barbecuing.6
1. Fields, J.M. (1993). Effect of personal and situational variables on noise annoyance in residential areas. The Journal of the Acoustical Society of America, 93(5), 2753-
2763.
2. Miedema, H.M.E., & Vos, H. (1999). Demographic and attitudinal factors that modify annoyance from transportation noise. The Journal of the Acoustical Society of
America, 105(6), 3336-3344.
3. Miedema, H.M.E., & Vos, H. (2004). Noise annoyance from stationary sources: Relationships with exposure metric day-evening-night level (DENL) and their confidence
intervals. The Journal of the Acoustical Society of America, 116(1), 334-343.
4. Michaud, D.S., et.al. (2016). Personal and situational variables associated with wind turbine noise annoyance. The Journal of the Acoustical Society of America, 139(3),
1455-1466.
5. van Kamp, I., Job, R.F.S., Hatfield, J., Haines, M., Stellato, R.K., & Stansfeld, S.A. (2004). The role of noise sensitivity in the noise-response relation: A comparison of
three international airport studies. The Journal of the Acoustical Society of America, 116(6), 3471-3479.
6. Pedersen, E., & Waye, K.P. (2004). Perception and annoyance due to wind turbine noise—a does-response relationship. The Journal of the Acoustical Society of
America, 116(6), 3460-3470.
23
Aesthetics, Noise Annoyance, & Metrics
Given: Annoyance occurs primarily when spending time outdoors with
activities such as relaxing, picnicking, or barbecuing.
• Does it even make sense to have nighttime limits to address
aesthetics?
• Perhaps, if a different limit was needed for aesthetics, an evening limit
(5 p.m. to 9 p.m.) that differs from the rest of the 24 hour period would
make the most sense.
1. Pedersen, E., & Waye, K.P. (2004). Perception and annoyance due to wind turbine noise—a does-response relationship. The Journal of the Acoustical Society of
America, 116(6), 3460-3470.
24
Aesthetics, Noise Annoyance, & Metrics
Given:
• Per annoyance research, the current 45 dBA one-hour maximum level
limit used in Vermont results in 10 to 15% of population exposed to
those levels being highly annoyed.
• For the proposed rule:
- 35 dBA night: 2.5% highly annoyed (outdoors)
- 42 dBA day: 6 to 9% highly annoyed (outdoors)
Propose:
• The current PSB precedent protects against the average person being
shocked and offended.
• The current PSB precedent protects against undue adverse impact to
aesthetics.
1. Pedersen, E., & Waye, K.P. (2004). Perception and annoyance due to wind turbine noise—a does-response relationship. The Journal of the Acoustical Society of
America, 116(6), 3460-3470.
26
Outdoor-to-Indoor Attenuation
• Current PSB precedent has been based on the World Health
Organization (WHO) guideline of 45 dBA L8hr outside bedroom
windows which is derived from a threshold of 30 dBA L8hr inside
bedrooms to protect against sleep disturbance.
- 15 dB of attenuation for a partially open window
• Vermont tests (2 data points)
- Sheffield (Brouha): less than 5 dB of attenuation (windows open)
o Large windows located in a small bedroom
o Window panes that can rotate perpendicular with the façade
- Georgia (Fitzgerald): 15 dB of attenuation (windows open)
o Standard sized window
27
Outdoor-to-Indoor Attenuation
Literature Review (Additional Data Points)
• Waters-Fuller 7 Lurcock (2007)1
- 7 to 26 dB reduction with windows open 0.2 m2 (2.2 ft2)
- Most values between 10 to 17 dB reduction
- When attenuation values applied to a wind turbine sound
spectrum:
o 14 dB worst-case
o 18 dB best-case
• Hayes McKenzie Partnership (2006) 2
- Focused on wind turbine acoustics
- One window-open measurement: 10 dB reduction
1. Waters-Fuller and Lurcock, Department for Environment, Food and Rural Affairs, UK, 2007.
2. Hayes McKenzie Partnership, Department of Trade and Industry, UK, 2006.
28
Outdoor-to-Indoor Attenuation
Literature Review (Additional Data Points)
• Environmental Protection Agency (1974)
- 12 dB reduction for warm climates, windows open 0.19 m2 (2 ft2)
- 17 dB reduction for cold climates, windows open 0.19 m2 (2 ft2)
• Federal Highway Administration (2011)
- Uses a 10 dB reduction for windows open, all climates
29
Outdoor-to-Indoor Attenuation
Takeaway
• While 5 dB or less of attenuation is possible, it is only one data point.
• Reductions between 10 and 15 dB are more common.
• In some cases, 20 dB reductions may be present.
• Depends on a number of factors:
- Window size and type
- Amount open
- Window area relative to that of the façade
- Sound insulation of the wall and window
- Bedroom size, furnishings, and orientation to the sound source
31
Noise Reduced Operation of Wind Turbines
Section 5.703 of the proposed rule
• Daytime 42 dBA
• Nighttime 35 dBA
Sound Generation by Wind Turbines
• Aerodynamics – primary source
• Mechanical (nacelle) – secondary
• Noise Reduced Operation (NRO)
reduces aerodynamic noise.
32
Noise Reduced Operation of Wind Turbines
Designing a Wind Power Project
• Developers design the entire project to the most stringent sound
level limit.
- Array layout
- Turbine model
- NRO
- Shutdowns
• When daytime and nighttime standards vary, NRO is the tool that is
used to regulate sound emissions.
• Shutdowns effect the economics of a project too strongly making
them infeasible.
33
Noise Reduced Operation of Wind Turbines
How does NRO work?
• Blades are pitched
• Often a slight RPM reduction
• Often modest power losses
• Operational protocols, typically driven by software
- Time of day
- Wind direction
- Wind speed
- Different protocols for individual turbines
34
Noise Reduced Operation of Wind Turbines
Limits to it’s usefulness
• 1 to 3 dB reduction is typical
• Greater than 4 dB, only offered by one manufacturer
• 1 to 2 dB reduction - modest power losses
• 3 to 4 dB reduction – greater than modest power losses
Proposed Rule 5.700 Context
• 7 dB difference between daytime and nighttime limits (42 dBA to 35
dBA).
• Projects will need to be designed to 35 dBA, likely using NROs.
• If shutdowns are needed, project economics are affected too
strongly.
• With 4 dB NRO, the effective daytime limit is 39 dBA.
35
Noise Reduced Operation of Wind Turbines
Developers have tools to reduce sound emissions from wind turbines,
But
There are limits to the range of reductions that are achievable.
Recommendation
1. If Rule 5.700 is to have different specified limits by time-of-day,
decrease the difference to not more than 4 dB.
37
PSB Precedent & the Proposed Rule
From previous testimony, presentations, and submissions:
• Current precedent 45 dBA one-hour maximum, exterior, guards against
public health impacts.
• Same limit used in Kingdom Community Wind, Georgia Mountain
Community Wind, and Deerfield Wind.
Proposed Rule of 35 dBA nighttime and 42 dBA daytime goes beyond
protecting public health.
38
PSB Precedent & the Proposed Rule
Effective Limit is Lower than Proposed
• With limitations to NRO technology and how projects are designed,
the effective limit is 35 dBA nighttime and 40 dBA daytime.
• Since Section 5.705 requires potential model error to also be added
on to each source emission, the effective limits are even lower than
35 dBA nighttime and 40 dBA daytime.
39
PSB Precedent & the Proposed Rule
Closing Thoughts
• Under the proposed rule, Kingdom Community Wind, which has provided
over 700,000 MWhs of clean power to the Vermont grid and likely other
projects built under the 45 dBA precedent, would not have been built.
40
PSB Precedent & the Proposed Rule
Closing Thoughts
• Under the proposed rule, Kingdom Community Wind, which has provided
over 700,000 MWhs of clean power to the Vermont grid and likely other
projects built under the 45 dBA precedent, would not have been built.
• Compliance monitoring must account for background sound levels.
41
PSB Precedent & the Proposed Rule
Closing Thoughts
• Under the proposed rule, Kingdom Community Wind, which has provided
over 700,000 MWhs of clean power to the Vermont grid and likely other
projects built under the 45 dBA precedent, would not have been built.
• Compliance monitoring must account for background sound levels.
• The current PSB precedent of 45 dBA one-hour maximum protects against
public health impacts and undue adverse impact on aesthetics per Act 250
framework.
42
PSB Precedent & the Proposed Rule
Closing Thoughts
• Under the proposed rule, Kingdom Community Wind, which has provided
over 700,000 MWhs of clean power to the Vermont grid and likely other
projects built under the 45 dBA precedent, would not have been built.
• Compliance monitoring must account for background sound levels.
• The current PSB precedent of 45 dBA one-hour maximum protects against
public health impacts and undue adverse impact on aesthetics per Act 250
framework.
• If a different limit was needed for aesthetics, an evening limit (5 p.m. to 9
p.m.) that differs from the rest of the 24 hour period would make the most
sense.
www.rsginc.com
Contacts
www.rsginc.com
ContactsEddie Duncan, INCE Bd. Cert.Director
Vermont Public Service Board Proposed Rule 5.700
60
APPENDIX B. RSG SUGGESTED EDITS TO THE
PROPOSED RULE
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 1 of 11
5.700 RULE ON SOUND LEVELS FROM WIND GENERATION FACILITIES
5.701 Purpose and Applicability This rule establishes standards and procedures related to sound emissions from wind
generation facilities that apply for a certificate of public good (“CPG”) pursuant to 30 V.S.A.
§ 248 or § 8010 on or after July 1, 2017.
5.702 Definitions For the purposes of this Rule, the following definitions shall apply: (A) Board: the Vermont Public Service Board.
(B) CPG: certificate of public good.
(C) CPG Holder: a person or company who has received a CPG pursuant to 30 V.S.A.
§ 248 or § 8010 for a wind generation facility.
(D) dB: a unit used to measure the intensity of a sound wave using a logarithmic scale.
(E) dBA: A-weighted decibel
(F) Department: the Vermont Department of Public Service.
(G) Petitioner: a person or company who has filed a petition for a CPG pursuant to 30
V.S.A. § 248 or 8010 to construct and/or operate a wind generation facility.
(H) Plant capacity: pursuant to 30 V.S.A. § 8002, “plant capacity” means the rated
electrical nameplate for a wind generation facility.
(I) Residence: a permanent structure for human habitation that is occupied by one or
more people for a minimum of 90 days each year.
(J) SCADA: supervisory control and data acquisition or similar system capable of
measuring and recording turbine operation and meteorological data in one-minute
time intervals.
(K) Wind generation facility: a wind-driven electric generation facility for which a
petition for a CPG pursuant to 30 V.S.A. § 248 or § 8010 is submitted to the Board
on or after July 1, 2017.
(L) Leq: Continuous sound level in dB equivalent to the total sound energy over a given
period of time.
(M) LA90: Sound level exceeded during 90% of a measurement period.
(N) LA10: Sound level exceeded during 10% of a measurement period.
(O) LA50: Sound level exceeded during 50% of a measurement period. (P) Participating landowner: a landowner who has signed a written agreement with a
Petitioner stating that the sound emissions standards established by this rule do not
apply to the landowner’s property.
(Q) NRO mode: Noise Reduced Operation mode, in which the rotational speed of wind
turbines is limited in order to reduce their sound emissions.
(R) Facility-only sound: Sound attributable to a wind generation facility not including
background sound.
(S) Background sound: The sound attributed to all sources, other than the Facility-only
sound.
5.703 General Rule No wind generation facility shall emit sound levels in excess of the following during
operation:
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 2 of 11
(A) Facilities with a plant capacity of 150 kilowatts or less. Operation of facilities with a
plant capacity of 150 kilowatts (“kW”) or less shall not result in: (1) audible
prominent discrete-frequency tones pursuant to the latest revision of ANSI S1.13
Annex A at a distance of 100 feet from the residences of non-participating
landowners; and (2) Facility-only sound pressure levels in excess of 42 dBA
between the hours of
7 A.M. and 9 P.M. and 35 dBA between the hours of 9 P.M. and 7 A.M. at a
distance of 100 feet from the residences of non-participating landowners. In lieu of
demonstrating compliance with this limit, a petitioner may propose to locate a wind
generation facility such that every sound-producing element of the facility will be set
back horizontally no less than ten (10) times the turbine’s height, as measured from
base to the tip of a blade in the upright, vertical position, from the residences of non-
participating landowners.
(B) Facilities with a plant capacity of greater than 150 kW. Operation of facilities with a
plant capacity of greater than 150 kW shall not result in: (1) audible prominent
discrete-frequency tones pursuant to the latest revision of ANSI S1.13 Annex A at a
distance of 100 feet from the residences of non-participating landowners; and (2)
Facility-only sound pressure levels in excess of 42 *** dBA between the hours of 7
A.M. and 9 P.M. and 35 *** dBA between the hours of 9 P.M. and 7 A.M. at a
distance of 100 feet from the residences of non-participating landowners. Each
sound-producing element of such facilities shall be set back horizontally no less than
ten (10) times the turbines’ height, as measured from base to the tip of a blade in the
upright, vertical position, from the residences of non-participating landowners.
5.704 Compliance with the Sound Level Limits
Compliance with the sound level limits shall be determined in accordance with the following:
(A) Sound level data shall be aggregated in 10-minute measurement intervals within a
given compliance measurement period under the conditions set forth in Section 5.707
of this rule. Each hour of the compliance measurement period shall have six discrete
10-minute measurement intervals.
(B) Compliance will be demonstrated when the arithmetic average of the equivalent
sound level of, at a minimum, twelve 10-minute measurement intervals in a given
compliance measurement period is less than or equal to the sound level limit set forth
in Section 5.703. The loudest valid 10-minute measurement intervals shall be
included in the calculation of the arithmetic average.
(C) If a given compliance measurement period does not produce a minimum of twelve
10-minute measurement intervals under the atmospheric and site conditions set forth
in Section 5.708(E) of this rule, six or more 10-minute measurement intervals from
one compliance measurement period may be combined with six or more 10-minute
intervals from other compliance measurement periods (e.g., other days). Compliance
will be demonstrated when the arithmetic average of the combined 10-minute
measurement intervals is less than or equal to the applicable equivalent sound level
Commented [A1]: See Sections 3, 4, 5, and 7 and Appendix A of Comments & Technical Information provided by RSG in the May 11, 2017 submission to the Board. To the extent that the lower limits, are meant to public health impacts, the current limit of 45 dBA already does that. To the extent that the lower limits are meant to address aesthetic concerns, the current limit of 45 dBA already does that per Act 250’s Quechee test framework and scientific studies regarding noise annoyance due to wind turbine sound. To the extent that the lower limits are meant to indirectly address indoor sound levels based on assumptions that the indoor-outdoor attenuation is low, we refer the Board to the literature that shows 10-17 dB reduction with windows open is most typical. If the rule is to have different specified limits by time-of-day, the practical limit to the range is 4 dB based on NRO technology.
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 3 of 11
limit set forth in Section 5.703. The loudest valid 10-minute measurement intervals
shall be included in the calculation of the arithmetic average.
5.705 Pre-Construction Sound Modeling
All petitions to construct and operate a wind generation facility, except for those for a wind generation facility with a capacity of 50 kW or less, shall include a sound model developed
for the proposed facility that reports the expected maximum project sound levels, without using
NRO mode, experienced out to a distance where such levels are no greater than 30 dBA. A
petitioner must submit the following information with its petition:
(A) A map depicting the location of all proposed sound sources associated with the wind
generation facility, property boundaries for the proposed facility, and all residences
within the 30 dBA contour.
(B) A description of the major sound sources, including tonal sound sources, associated
with operation and maintenance of the facility. The sound model shall be based on
the technical specifications of the turbine model(s) with the highest manufacturer
apparent sound power level under consideration for use at the facility.
(C) The results of sound modeling pursuant to ISO 9613-2, including a description of the
equivalent continuous sound levels expected to be produced by the sound sources at a
distance of 100 feet from the residences of non-participating landowners. The
description shall include a full-page isopleths map depicting the predicted sound
pressure levels expected to be produced by the wind generation facility at a distance
of 100 feet from each residence of a non-participating landowner within the 30 dBA
isopleth. The predictive model used to generate the equivalent sound levels expected
to be produced by the sound sources shall be designed to represent the “predictable
worst case scenario.” All model inputs shall be the most realistic and conservative
available for each of the items listed below unless otherwise approved by the Board,
and shall include, at a minimum, the following:
(1) The maximum apparent sound power output (IEC 61400-11) of the sound
sources;
(2) Modeling in accordance with ISO 9613-2, with each turbine modeled as a
point source at hub height;
(3) All turbines operating at full rated sound output;
(4) Attenuation due to air absorption, with conditions set to 10ºC and 70%
relative humidity;
(5) Attenuation due to ground absorption/reflection, based on mixed ground
conditions (G=0.5) for propagation over land and hard conditions (G=0.0)
for propagation over water;
(6) Attenuation due to three-dimensional terrain;
(7) A receiver heights of four (4) meters and one and a half (1.5) meter;
(8) Attenuation due to meteorological factors such as relative wind speed and
direction (wind rose data), temperature/vertical profiles and relative
humidity, sky conditions, and atmospheric profiles;
Commented [A2]: See Workshop Presentation in Appendix A of Comments & Technical Information provided by RSG in the May 11, 2017 submission to the Board. We recommend rewriting this Section to reflect a methodology that accounts for background sound levels using either the turbine shutdown method, shielding method, or perhaps the method proposed by the Department’s expert.
Commented [A3]: See Section 5 of Comments & Technical Information provided by RSG in the May 11, 2017 submission to the Board.
Commented [A4]: See Section 2 of Comments & Technical Information provided by RSG in the May 11, 2017 submission to the Board. This is also in response to a question from the Board during the workshop on what height should be used. Four meters is recommended for preconstruction compliance demonstration, but one and a half meter should be provided as well, so that post-construction compliance monitoring can be conducted at 1.5 meters for model validation.
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 4 of 11
(9) An adjustment to the maximum rated output of the turbines to
account for turbine manufacturer uncertainty, determined in accordance
with the most recent version of the IEC 61400 Part 11 standard; and
(10) A disclosure of the model’s error, which is intended to account for
uncertainties in the modeling of sound propagation for wind energy
developments. This error shall be accounted for and incorporated as an
addition to the maximum rated output of the sound sources.
(D) A description of proposed major sound control measures, including their locations
and expected acoustical performance;
(E) A comparison of the expected sound pressure levels from the proposed wind
generation facility with the applicable sound pressure level limits of Section 5.703.
(F) A description and map identifying compliance testing locations on or near the
proposed wind generation facility site. The identified compliance testing locations
shall be selected to take advantage of prevailing downwind conditions and shall be
able to meet the site selection criteria outlined in Section 5.707(D). The identified
locations should include those locations that are expected to experience the highest
model-predicted equivalent sound levels. The locations should be free from sources
of material sound contamination.
(G) Prior to commencing site preparation or construction of the facility, a CPG Holder
shall update, supplement, and/or amend the sound modeling to reflect any changes to
the sound-producing elements of the facility. An opportunity to review and comment
on any change to the sound modeling, and to request a hearing, shall be given to all
parties to the 30 V.S.A. § 248 proceeding who have standing on the issue of sound.
The Board may, in its discretion, grant a hearing if a party who has standing on the
issue of sound demonstrates that the revised sound modeling represents a likelihood
of an exceedance of the applicable sound emissions standard specified in Section
5.703. If the Board holds a hearing, the CPG Holder may not commence site
preparation or construction of the facility until the Board resolves the issue.
5.706 Post-Construction Sound Monitoring
Sound monitoring shall take place during the times specified in section 5.708(D), in accordance with the requirements of this rule and any requirements of the CPG, which shall
specify the minimum number of compliance monitoring locations, the radius from the nearest
facility turbine in which monitoring locations may be selected, and the time period of
monitoring. The monitoring will be used to verify the accuracy of the pre-construction modeling
and facility compliance with CPG conditions and the requirements of this rule. In addition to the
requirements of this rule and the CPG, at its discretion, the Board may require additional
monitoring if the results of the initial post-construction sound monitoring or changes to the
facility or its operation indicate that exceedances of the sound-level limit are likely.
(A) Monitoring by the State. Post-construction sound monitoring shall be conducted
under the direct supervision and control of a State of Vermont agency or agencies
Commented [A5]: This is not needed per wind turbine model verification studies discussed in Section 2 of Comments & Technical Information provided by RSG in the May 11, 2017 submission to the Board.
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 5 of 11
designated by the Board. The post-construction sound monitoring shall be paid for by
the CPG Holder.
(B) Monitoring Locations. A petition for a CPG for a wind generation facility shall
include proposed monitoring locations for post-construction monitoring. The
proposed locations shall include residential locations that are expected to experience
the highest model-predicted equivalent sound levels and are consistent with the
requirements of Section 5.707(D). The proposed locations should be free from
sources of material sound contamination. Any change in monitoring locations must
be approved in advance by the Board.
(C) Modification of pre-construction sound modeling. A CPG Holder is required to
identify the appropriate inputs and/or assumptions, and modify the pre-construction
sound modeling if the post-construction sound monitoring indicates that there is a
reasonable likelihood that the expected highest sound levels at any of the monitoring
locations would be equal to or greater than 3 dBA above those modeled, or would
result in an exceedance of the sound level standard specified in Section 5.703. All
parties to the 30 V.S.A. § 248 or § 8010 proceeding who have standing on the issue of
sound shall be given an opportunity to review and comment on any change to the
sound modeling. The Board may, in its discretion, grant a hearing if a party who has
standing on the issue of sound demonstrates that the revised sound modeling indicates
a likelihood of an exceedance of the applicable sound emissions standard specified in
Section 5.703.
5.707 Sound Monitoring Methodology
(A) Measurement Personnel Measurements shall be supervised by personnel who are
well qualified by training and experience in measurement and evaluation of
environmental sound. Certification through the Institute of Noise Control Engineering
shall meet the qualification requirements of this section.
(B) Measurement Instrumentation The sound meter or alternative sound measurement
system used shall meet all appropriate industry standards and specifications. Each
monitoring site shall include installation of an anemometer and other equipment or
sensors capable of gathering and recording weather conditions at the microphone (10-
meter-level wind speed, wind direction, temperature, and precipitation) and be equipped
with enhanced-performance windscreens capable of significantly reducing or eliminating
wind-induced noise contamination over the microphone. The measurement
instrumentation shall meet the following specifications unless otherwise approved by the
Board:
1. A sound level meter or alternative sound level measurement system used
shall meet the Type 1 performance requirements of American National
Standard Specifications for Sound Level Meters, ANSI S1.4.
2. An integrating sound level meter (or measurement system) shall also meet the
Commented [A6]: 10-meter wind speed monitoring is not necessary and makes the post-construction monitoring process cumbersome. Monitoring wind speeds at microphone height would suffice.
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 6 of 11
Type 1 performance requirements for integrating/averaging in the
International Electrotechnical Commission Standard on Integrating-Averaging
Sound Level Meters, IEC Publication 61672-1.
3. A filter for determining the existence of tonal sounds shall meet all the
requirements of the American National Standard Specification for Octave-
Band and Fractional Octave-Band Analog and Digital Filters, ANSI S1.11 and
IEC 61260, Type 3-D performance.
4. The acoustical calibrator used shall be of a type recommended by the
manufacturer of the sound level meter and one that meets the requirements of
American National Standard Specification for Acoustical Calibrators, ANSI
S1.40.
5. Anemometer(s) used for surface (10 meter (m)) (32.8 feet) wind speeds shall
have a minimum manufacturer specified accuracy of ±1 mph providing data in
10-second integrations and 10 minute average/maximum values for the
evaluation of atmospheric stability.
6. Audio recording devices shall be time stamped (hh:mm:ss), recording the
sound signal output from the measurement microphone to be used for
identifying events. Audio recording and compliance data collection shall be
measured through the same microphone/sound meter and bear the same time
stamp.
(C) Equipment Calibration
1. The sound level meter shall have been calibrated to the manufacturer’s
specification no more than 24 months prior to completion of a measurement
campaign, and the microphone’s response shall be traceable to the National
Institute of Standards and Technology.
2. Field calibrations shall be recorded and documented in compliance monitoring
reports.
3. The 10-meter anemometer(s) and vane(s) shall have been calibrated to the
manufacturer’s specification no more than 24 months prior to completion of a
measurement campaign.
(D) Compliance Measurement Location, Configuration, and Environment
1. Compliance measurement locations shall be approved by the Board during its
review of a facility’s request for a CPG and shall be representative of the non-
participating residences expected to experience the highest equivalent sound
levels from routine operation of the wind generation facility, subject to
permission from the respective property owner(s).
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 7 of 11
a. To the greatest extent possible, compliance measurement locations shall
be at the center of unobstructed areas that are maintained free of
vegetation and other structures or material that is greater than 2 feet in
height for a 75-foot radius around the sound and audio monitoring
equipment.
b. To the greatest extent possible, meteorological measurement locations
shall be at the center of open flat terrain, inclusive of grass and minimum
number of obstacles that are greater than 6 feet in height for a 250-foot
radius around the anemometer location. Meteorological measurements
shall be taken at the monitoring location at or above the height of the
audio/acoustic microphone.
c. Meteorological measurements of wind speed and direction shall be
collected using anemometers at a 10-meter height (32.8 feet) above the
ground. Results shall be reported, based on 10-second integration
intervals, synchronously with turbine nacelle measurements and
measurements made at the sound-meter level at 10-minute measurement
intervals. The wind speed average and maximum for each 10-minute
interval shall be reported.
d. The sound microphone shall be positioned at a height of approximately 4
to 5 feet above the ground, and oriented in accordance with the
manufacturer’s recommendations.
e. When possible, measurement locations should be at least 50 feet from any
sound source. The proposed locations should be free from sources of
material sound contamination. Any non-facility sources of sound shall be
noted in the analysis.
4. The CPG Holder shall provide all relevant turbine operational data for the
monitoring period, including SCADA data for all turbines, the date, time, and
duration of any noise reduction operation or other operational changes that
occur during the compliance measurement period.
5.708 Compliance Data Collection, Measurement, and Retention Procedures
(A) Measurements of operational, sound, audio, and meteorological data shall occur as set forth in Section 5.707.
(B) All operational, sound, audio, and meteorological data collected shall be retained by
the State of Vermont agency or agencies designated by the Board for the life of the
project and subject to inspection upon request.
(C) Monitoring and data collection shall occur at a minimum:
1. Once during each of the first four years of facility operation, provided, however,
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 8 of 11
that if after three years the monitoring does not detect and the updated sound
model does not predict any exceedances of the applicable equivalent sound
pressure level in Section 5.703, the fourth year of monitoring and data collection
under this subsection shall not be required;
2. Once during each successive fifth year thereafter until the facility is
decommissioned; and
3. In response to a complaint if ordered by the Board.
a. The Board in its discretion may require sound monitoring for a wind
generation facility in response to a complaint if the Board determines that
a complaint raises a reasonable possibility that a wind generation facility
is operating in excess of the sound level limits required by this rule. In
making its determination, the Board shall consider:
i. The details of the complaint;
ii. Any response thereto filed by the operator of the wind generation
facility; and
iii. Any response and recommendation by the Department of Public
Service after its review of the complaint, the facility operator’s
response, and any attempts made to resolve the complaint under
the complaint response procedure(s) issued by the Vermont
Department of Public Service pursuant to Section 5c of Public Act
130 (2016 Vt., Adj. Sess.). As part of any recommendation, the
Department may propose a plan for additional sound monitoring of
the subject wind generation facility. Any such proposal should
incorporate the requirements and standards set forth in subsection
(b), below, or set forth an explanation why different requirements
and standards are being proposed.
b. Any monitoring ordered by the Board pursuant to this subsection:
i. Shall conform to the meteorological requirements set forth in
Section 5.708(E) of this rule, if possible.
ii. Shall be done under meteorological conditions as similar as
possible to the conditions existing at the time of the complaint.
iii. In the event that the monitoring cannot be performed pursuant to
the meteorological requirements set forth in Section 5.708(E) of
this rule due to prevailing meteorological or environmental
conditions at the time the complaint is filed and when the
monitoring will take place, then the Department of Public Service
may propose a plan of sound monitoring for review and approval
by the Board. Any such proposed monitoring plan should:
1. Require that sound monitoring be performed under
meteorological conditions similar to those that existed at
the time the complaint was made;
2. Provide for sound monitoring compliance testing consistent
with the requirements of Section 5.704 of this rule with
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 9 of 11
monitoring continuing until the requisite number of
measurement intervals are collected.
iv. The sound monitoring methodology for any such proposal shall be
consistent with the requirements of Section 5.707 of this rule.
v. Microphones shall be placed in locations that avoid material sound
contamination. All microphone locations must be approved by the
Board.
vi. Primary microphones shall not be placed such that any structure
blocks the line of sight between the microphone and the facility’s
turbines (if otherwise visible).
vii. Provide a process for determination of facility-only sound. In the
event the determination of facility-only sound will rely on
subtracting background sound levels from overall sound levels (i.e.
sound levels with the facility’s turbines in operation), such
background sound levels shall be determined by measurements
taken with the facility’s wind turbines shut down for a period of at
least 30 minutes both before and after sound monitoring is
performed to determine total sound levels with the facility in
operation.
viii. The monitoring shall be performed with at least 90% of the
facility’s turbines operating at within 1dB of maximum sound
power levels. Monitoring shall continue until the requisite
amount of data is collected under these operating conditions.
ix. Measurement intervals affected by increased biological activities,
leaf rustling, traffic, high water flow, aircraft flyovers, or other
extraneous ambient noise sources that affect the ability to
demonstrate compliance shall be excluded from all compliance
report data.
x. Reporting of the results of the monitoring shall be done consistent
with the requirements of section 5.709 of this rule.
(D) All operational (SCADA), sound level and meteorological data collected during a
compliance measurement period that meets or exceeds the specified wind speed
parameters shall be submitted by the State of Vermont agency or agencies designated
by the Board to the Board for review and approval. All data shall be submitted to the
Board within 60 days of completion of the monitoring period as part of the post-
monitoring report. Audio recordings will only be submitted upon request and may be
filtered to exclude private conversations and/or submitted under a confidentiality
order.
(E) Measurements shall be obtained during weather conditions when the wind turbine
sound is dominant and overall sound levels are not influenced by non-facility sounds.
Such conditions are generally expected at night, when the measurement location is
downwind of the wind generation facility and maximum surface wind speeds (10-
meter height) are equal to or less than 6 miles per hour (mph) with concurrent turbine
hub-elevation wind speeds sufficient to generate the highest continuous apparent
Commented [A7]: To increase the likelihood finding valid periods and reduce the amount of time needed to find valid monitoring periods, consider decreasing the percentage to less than 90% or allowing turbines to be operating to within 1 dB of maximum sound power levels.
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 10 of 11
sound power, +/- 1 dB, from the nearest wind turbines to the measurement location. A
downwind location is defined as within 45º of the direction between a specific
measurement location and the acoustic center of the five nearest wind turbines, or
fewer if the wind generation facility does not have five wind turbines. In some
circumstances, it may not be feasible to meet the wind speed and operations criteria
due to terrain features or limited elevation change between the wind turbines and
monitoring locations. In these cases, measurement periods are acceptable if the
following conditions are met:
1. The difference between the LA90 and LA10 during any 10-minute period is less than
5 dBA; and
2. The surface wind speed (10-meter height) (32.8 feet) is 6 mph or less for 80% of
the 10-minute measurement period and does not exceed 10 mph at any time, or
the turbines are shut down during the monitoring period and the difference in the
observed LA50 after shutdown is equal to or greater than 6 dBA; and
3. Observer logs or recorded sound files clearly indicate the dominance of wind
turbine(s).
(F) 4. Measurement intervals affected by increased biological activities, leaf rustling, traffic,
high water flow, aircraft flyovers, or other extraneous ambient noise sources that affect the
ability to demonstrate compliance shall be excluded from all compliance report data. The intent
is to obtain 10-minute measurement intervals that entirely meet the specific criteria and
represent facility-only sound pressure levels.
5.709 Reporting of Compliance Measurement Data
Compliance Reports shall be submitted to the Board within 60 days of the completion of the sound monitoring period. The Board will make the report publicly available. The report
shall include a certification that the required monitoring conditions were present and, at a
minimum, the following:
(A) A narrative description of the sound from the wind generation facility for the
compliance measurement period;
(B) The dates, days of the week, and hours of the day when measurements were made;
(C) The wind direction and speed, temperature, humidity, and sky condition;
(D) Identification of all measurement equipment by make, model, and serial number;
(E) All meteorological, sound, windscreen, and audio instrumentation specifications and
calibrations;
(F) All A-weighted equivalent sound levels for each 10-minute measurement interval;
Formatted: Indent: Left: 0", First line: 0"
Commented [A8]: See Workshop Presentation in Appendix A of Comments & Technical Information provided by RSG in the May 11, 2017 submission to the Board. We recommend rewriting this Section to reflect a methodology that accounts for background sound levels using either the turbine shutdown method, shielding method, or perhaps the method proposed by the Department’s expert.
Commented [A9]: If the Board decides to retain the methodology in the current proposed rule, this subsection should apply to all measurement intervals not just the circumstances where it is “not feasible to meet the wind speed and operations criteria …”, so we moved it from (E) 4. to (F)
Effective Date: Vermont
Public Service Board
Rule 5.700
Page 11 of 11
(G) Short-period sound level measurements (50 milliseconds or less);
(H) All LA10, LA50, and LA90 percentile levels for each 10-minute
measurement interval;
(I) All 10-minute 1/3 octave band unweighted and equivalent continuous sound levels
(dB);
(J) Should any sound data collection be observed by a trained attendant, the attendant’s
notes and observations shall be summarized and included with the Compliance
Report;
(K) All concurrent time-stamped, turbine-operational data including the date, time, and
duration of any noise-reduction operation or other interruptions in operations, if
present; and
(L) All other information determined necessary by the Board.
5.710 Complaint Response Procedures Complaints raised by residents located near the wind generation facility shall be
responded to in a manner consistent with the complaint response procedure(s) issued by the
Vermont Department of Public Service pursuant to Section 5c of Public Act 130 (2016 Vt., Adj.
Sess.)
Commented [A10]: This is not needed, and it is irrelevant to the proposed limits.
Commented [A11]: Required more specificity related to time interval.