Mrs. Susan M. Hudson, Clerk

Paul Brouha

Ridge Protectors

92 Queen Elizabeth Farm Ln.

Sutton, VT05867

May 11, 2010

Mrs. Susan M. Hudson, Clerk

Vermont Public Service Board

112 State Street – Drawer 20

Montpelier, VT05620-2701

Re: Docket No. 7156—Comments on UPC Vermont Wind, LLC’sMarch 31, 2010 and April 12, 2010 filings.

Dear Mrs. Hudson:

We thank the Board for the opportunity to submit comments concerning the filings referenced above. We enclosethree copies of Ridge Protectors, Inc.’s comments and are sending a copy toall parties on the Service List.

Sound Monitoring Plan

During the extension provided by the Board, Ridge Protectorsretained the expert services of Les Blomberg, Executive Director, Noise Pollution Clearinghouse, P.O. Box 1137Montpelier, VT05601 to assist preparation of comments on the Sound Monitoring Plan as follow:

There are a number of deficiencies with the Sound Monitoring Plan offered by Vermont Wind on March 31, 2010 to the Public Service Board concerning the Sheffield Wind Project. Deficiencies and proposed corrections are listed under the headings of the original plan below.

3.0 Monitoring Plan

Measuring the noise of the wind turbines indoors is a relatively easy and straight forward process as long as one has (1) access to the indoor location, (2) sufficiently low background noise levels, (3) control of the noise source, and (4) an observer. That one needs access to the property to make indoor measurements is obvious. The requirement for low background levels is required because sound level meters do not discriminate between noise from the source one is trying to measure and the existing background noise. Sound level meters provide a combined value. It is important, therefore, that the background levels be low enough so that they can be subtracted from the total value reported to derive the source noise levels. Control of the noise source (the ability to turn it on and off) is needed to accurately measure the background noise levels (the noise level when the source is turned off) and the source noise levels (when the source is on). Finally, an observer is needed to determine if the noise is from the turbines or from some other source.

The proposed Sound Monitoring Plan seeks to estimate the interior noise levels while avoiding using all four of these key conditions. The result is a rough estimate of interior noise levels in which the results are highly skewed toward “false negative” results (i.e., no violation of the Board’s Noise Condition). The proposed Sound Monitoring Plan and Vermont Wind’s comments to the August 8, 2007 Board Order dwell on the problems related to the first two of the necessary conditions (access and background noise), but ignore the need to control the noise source and actually hear the noise. This is somewhat understandable—Vermont Wind is probably hesitant to turn the turbine on and off for noise monitoring, and attended noise monitoring is very expensive. Nevertheless, their reluctance to do so should not decrease the reliability of their noise assessment.

Meeting condition (1) is fairly straight forward. The KingGeorgeSchool will agree to make an unoccupied girls dorm room available for each of the four testing periods. Other neighbors have also expressed willingness to make unoccupied rooms available for testing.

Meeting condition (2) is also fairly straightforward for the KingGeorgeSchool, as the background noise levels are sufficiently low for many hours of the day. The dorm is quiet during the night, and unoccupied during the day (when students are bused to the main campus. Vermont Wind’s noise measurement metric (a 10 minute Leq) will likely be contaminated by background noise during the early morning, late afternoon and early evening when the students are present. The best way to deal with this problem is to make a digital recording of the noise as well as sound pressure measurements, eliminate those periods when background noise is a problem, and then recalculate the Leq (1).

Meeting condition (2) at the other locations is more problematic because the proposed monitoring stations (with the exception of Dareios 1 and possibly P8) are relatively noisy locations. They are not representative of the soundscape of Sheffield, Barton, and Sutton. Wind turbine noise is more likely to impact quieter areas, rather than those immediately adjacent to Interstate 91 or US Route 5. Even nighttime road noise, which may be intermittent, will contaminate a 10 minute Leq, resulting in the data being discarded. Assuming that the noise levels exceed 30 dBA leq for the entire day, discarding only one or two ten minute segments each hour would make finding an leq (1) in excess of 30 dBA impossible. .

Using the audio recording method described above would solve the night time noise issues at the Interstate 91 and US Route 5 locations, but would probably be too time intensive and expensive to work during the daytime.

A better alternative is to seek out locations with quieter background noise levels. Locations along New Duck Pond Road that are further from I 91 and along Anis Brook Road(but away from the stream) would satisfy this condition. Many additional locations east and south of the facility would also meet this condition. King George Farm Road, Queen Elizabeth Farm Lane,Kivimae Road, Union House Road, Morey Rd, and Anis Brook Road are much more representative of the quiet soundscape that will be impacted by wind turbine noise.

One result of the lost data due to contamination is that more monitoring time may be required to accurately assess the noise impacts. Two weeks of usable data, not two weeks of any type of data, should be the monitoring requirement for each season.

Meeting condition (3), control of the turbines and the ability to turn them on and off so that noise readings can be correlated to those conditions is probably most problematic. As already mentioned, Vermont Wind probably won’t agree to turning the turbines on and off in noisy conditions to test the noise levels. The proposed Sound Monitoring Plan seeks to use shielding (making background measurements behind the façade of a building to partially block the turbine noise) and distance (estimating the noise levels based on noise levels at other locations) to substitute for actually turning the turbines on and off.

The screening method will not work at the I-91 and US Route 5 locations. The “screened” side of the buildings at these locations faces the roadway, introducing elevated background noise levels. Also, the screening will only be partial.

Using distance gives a very imprecise estimate of the background levels and source noise levels, and is similarly unhelpful in establishing compliance with the Board’s indoor condition of 30 dBA Leq (1).

Time may also be used to correlate noise levels with turbine operation. Turbine noise levels fluctuate over time and as long as turbine operation is well documented, a correlation may be made. The key to this is having a digital audio recording of the noise events (as described above) so that noise from wind and other background sources are not considered. All periods when the turbine is not audible, for example, could easily be eliminated from consideration.

The digital audio recordings also meet Condition (4), an observer. A recording is a compromise between expensive long-term observed monitoring and unobserved monitoring.

The proposed Sound Monitoring Plan is really a last-resort contingency plan. It should be used only if there is no workable alternative. Moreover as envisioned, the Sound Monitoring Plan is seriously flawed (by background noise contamination and the inability to discriminate between source and background noise). Fortunately, alternatives like those mentioned above do exist.

3.3 Monitoring Logistics

The Board’s Conditions 8 requires that “indoor sound levels …not exceed 30 dBA (Leq) (1).” Vermont Wind’s proposed Sound Monitoring Plan does not propose to record indoor noise measurements that could determine if this condition is met. The proposed Sound Monitoring Plan vastly overstates the difficulty of making indoor measurements. The issues cited are access to interior space (10/1/07 Amended PSB Order, page 5) and background noise levels in those spaces causing “false positive” results (Proposed Monitoring Plan, page 3 and Appendix A).

KingGeorgeSchool as well as other nearby residents will allow access to their property for interior noise monitoring, eliminating the first concern. Moreover, the rooms used for monitoring will be vacant during the monitoring periods, thereby reducing interior background noise levels. Background levels in these vacant rooms are sufficiently low to allow accurate testing. Finally, digital audio recordings can be made concurrently with the monitoring. The recording allows the noise expert to determine the source of the noise recorded, eliminating uncertainty about the noise source.

3.4 Outside to Inside Noise Test

The best and preferred method to measure interior noise levels is to measure interior noise levels. An exterior to interior estimation based on exterior levels and wall attenuation is appropriate only as a last resort, back-up option when direct inside measurements cannot be made. This is not the case for the homes most likely to suffer the greatest impact from the turbines. In particular, the outside to inside noise measurement protocol is unnecessary for the King George School and residences to the south and east of the facility.

An Interior Measurement Protocol, however, is needed. An ANSI Type 2 or better integrating sound level meter with one third octave band capabilities (to check for pure tones) should be used. Since one of the major reasons for the 30 dBA Leq (1) standard is meant to protect against sleep interference, the most significant measurement location may be slightly above pillow height at any likely location within the room where a bed might be located. A noise survey of the room should be made with observed measurements used to determine the location of the highest wind turbine sound pressure readings in the room. This site should then be used for the two weeks for unobserved monitoring. If turbine noise is not present during the survey, a location at pillow height (30 inches) and 18 inches from the window nearest the source should be chosen. The measurement location should be moved further into the room only if several days of data show that background levels at this location are consistently too high to assess the turbine noise.

It should also be noted that in the event that access to interior spaces and background noise levels make interior measurements impossible, that the interior measurement locations and procedures in the ASTM standard are inappropriate for assessing compliance with the Board’s condition. If the ASTM Standard E966-04ASTM is used, the measurement location shall be specified as the noisiest location due to wind turbines. Moreover, consideration should be given to any location where a person is likely to sleep.

In addition, the ASTM standard is a very poor way to measure interior wind turbine noise levels, in part because wind turbine noise is an overhead noise, while the standard is primarily focused on noise from the side. As an overhead noise source, there may be multiple paths the wind turbine noise takes to enter the room (through the ceiling, walls, or windows). The ASTM Standard’s concerns about eliminating flanking noise (turbine noise entering the room by paths other than the closest wall to the turbines) must be disregarded, as interior turbine noise is exactly what the monitoring should record, whether the path is through the wall, ceiling, or from an open window in the next room. Outside speaker height would need to simulate the maximum height of the turbine blade at a particular location.

Finally, since windows are typically opened during the summer months, summer measurements must be taken with windows open. Window settings should be at the option of the home owner and may also be opened in the spring or fall depending on weather conditions. Windows on facades facing away from the turbines may be closed if they help lower background noise levels (which would be the case at the KingGeorgeSchoolgirls dorm and along Interstate 91 and US Route 5).

3.5 Exterior Measurement Protocol

In a properly executed interior measurement plan, there is no need for exterior levels to be measured.

3.6 Reporting

Beyond the circumstances of the test, results, and rationale for conclusions, the report must include a log of what wind turbines were turning, wind speed and direction, electric generation levels, and noise levels for each test location during the four seasonal tests. In order to provide maximum time for review and evaluation, reports need to be provided to all parties and to owners of occupied residences concurrently with submission to the PSB. Furthermore, all raw data collected must be available to the parties to analyze.

The operation of the turbines, wind speed and direction, and electric generation are needed to correlate noise levels with turbine operation. The raw data are needed to assure the integrity of the process and to allow the parties to check Vermont Wind’s results.

4.0 Complaint Monitoring

Residents fear that noise problems and complaints will go unanswered. Vermont Wind fears a costly string of noise complaints that do not rise above the Board’s threshold level. A good complaint process, one that quickly responds to the neighbor’s concerns, assesses the noise levels, and convincingly shows that the noise is within or exceeds the Board’s condition is the best way to address the concerns of neighbors and Vermont Wind.

It is understandable that Vermont Wind would want a threshold test to limit the number of complaints it must respond to. Given the widely varying environmental, topological, and housing characteristics, however, the noise level of the four seasonal test sites provides too small a sample size to extrapolate from. Results of such an extrapolation would be arbitrary.

The abuse of the complaint system should be limited by limiting the number of assessments that will be made to two investigations (not one as proposed). Ridge Protectors is sympathetic to Vermont Wind’s abuse concerns and would agree to an alternative screening method. Since Vermont Wind’s modeling indicates it is unlikely that even an outdoor violation of the indoor standard will occur (outdoor levels at most structures are below a 30 dBA max level in the modeling), it seems reasonable to use simple unattended outdoor monitoring with audio recordings to screen cases where outdoor turbine noise levels don’t exceed indoor levels. If they do, then more extensive indoor testing shall occur.

T he following Complaint Procedure is proposed:

1. Vermont Wind provides a noise complaint phone number.
2. All complaints will be responded to within one business day.
3. Neighbors will at that time be supplied the wind speed, direction, operation of turbines, and generation for the period of concern. (Alternatively, this information may be continuously furnished on a website.)
4. If the neighbors request, a two week outdoor screening assessment as described above will be made.
5. Results and raw data will be made available to the effected neighbor and Board within two weeks.
6. If the turbine noise level exceeds 30 dBA Leq (1) for any period during that time, the neighbors may request an interior assessment.
7. The two week monitoring process used for the four seasonal monitoring scenarios will be used to determine the facility’s compliance with the Board’s Condition. If the neighbor does not wish to grant long term access to their home, the last resort contingency methodology may be used, as long as corrections for speaker height and microphone location are made as noted above.
8. A report and the raw data will be provided to the Board and affected neighbor within 2 weeks.
9. In the event the standard limiting noise from the turbines at receptor locations is exceeded, Vermont Wind shall cease operation of the turbines during similar wind conditions as those that caused the violations until the noise can be reduced to acceptable levels.
10. The four season testing methodology will be used to determine when compliance is met. Tests should occur in each of the four seasons at locations where an exceedance occurred.
11. Valid complaints will not be counted against the total of two monitoring opportunities allotted to each neighbor.
12. Neighbors shall be permitted to hire independent noise experts to make noise measurements using the Sound Monitoring Plan methodology if they choose, replacing steps 4 through 7, or if they exceed their two test allotment. A report and raw data will be provided to Vermont Wind and the Board within 2 weeks

Decommissioning Review Trigger Reduction

Ridge Protectors strongly opposes the requested reduction. To us, the request means Vermont Wind is not likely to be able to produce nearly as much power as they predicted – else why would they care about the trigger number. Second, for the second 10 years of operation – i.e. after the Federal subsidy has expired – Vermont Wind is only under contract for 10% of the power which in no way qualifies as a substantial amount. At a minimum, Vermont Wind should be required to produce for public inspection all the documents in their possession which form the basis for their belief that a 65% trigger will cause them any problem.