Shadow Flicker Assessment for the Proposed Hermanville Wind Farm

Revision 0

Prepared for: PEI Energy Corporation

Prepared by: Frontier Power Systems Inc. 392 Church Street Alberton, PE

January 2013

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Table of Contents TABLE OF CONTENTS ............................................................................................................................................ 2 LIST OF TABLES ....................................................................................................................................................... 2 LIST OF FIGURES ..................................................................................................................................................... 2 1.0

INTRODUCTION ......................................................................................................................................... 3

2.0

PROJECT DETAILS .................................................................................................................................... 3

2.1 2.2 2.3

SITE DESCRIPTION ....................................................................................................................................... 3 TURBINE DESCRIPTION ................................................................................................................................ 3 TURBINE LOCATIONS .................................................................................................................................. 4

3.0

BACKGROUND ON SHADOW FLICKER .............................................................................................. 4

4.0

SHADOW RECEPTORS ............................................................................................................................. 5

5.0

SHADOW FLICKER ANALYSIS .............................................................................................................. 7

5.1 5.2 5.3 5.4 5.5 5.6

METHODOLOGY........................................................................................................................................... 7 MODEL PARAMETERS .................................................................................................................................. 8 WORST-CASE SHADOW FLICKER ................................................................................................................ 8 RESULTS .................................................................................................................................................... 10 DISCUSSION OF RESULTS ........................................................................................................................... 11 POTENTIAL MITIGATION MEASURES ......................................................................................................... 12

6.0

CONCLUSION AND RECOMMENDATIONS ....................................................................................... 13

7.0

REFERENCES ............................................................................................................................................ 14

List of Tables Table 1: Turbine General Specifications ....................................................................................... 4 Table 2: Turbine Coordinates ........................................................................................................ 4 Table 3: Shadow Receptor Coordinates ......................................................................................... 7 Table 4: De-rating Data ................................................................................................................. 9 Table 5: Shadow Flicker Results ................................................................................................. 10 List of Figures Figure 1: Turbine Sites and Shadow Receptors ............................................................................. 6 Figure 2: Shadow Flicker Map – Theoretical Maximum............................................................. 11

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 2

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

1.0 Introduction Frontier Power Systems Inc has been engaged by the PEI Energy Corp to conduct a shadow flicker impact assessment of the proposed Hermanville Wind Farm, located in Kings County, Prince Edward Island. The project will consist of 10 wind turbines and have a total generating capacity of 30MW. The proposed turbines have a 116m rotor diameter, a hub height of 92m, and generating capacity of 3MW each. Residents living near the proposed project may experience shadow flicker from the wind turbines to various degrees. This report: • Describes the project site and turbine layout • Provides a brief background on shadow flicker • Identifies the potential shadow receptors near the proposed project • Summarizes the shadow flicker analysis methodology • Presents the results from the shadow flicker analysis • Discuss potential mitigation measures to reduce shadow flicker impact

2.0 Project Details 2.1

Site Description The proposed Hermanville Wind Farm is located between the communities of Hermanville and Clearspring, in north eastern PEI, approximately 11 km north from the town of Souris. The project area is located in an expansive wooded area south of Route 16 (the Northside Rd), between the New Zealand Rd and the Souris Line Rd. The site and surrounding area is characterized primarily by forest, with the exception of some blueberry and agricultural land to the north. The terrain is relatively flat with few low hills and valleys. The shore of the Gulf of St Lawrence is approximately 2 km north of the site. A small number of residential properties are located in the vicinity of the project area, along the Northside Rd and the Souris Line Rd.

2.2

Turbine Description The Acciona Windpower AW3000/116 is the proposed wind turbine for the Hermanville Wind Farm. The AW3000/116 is a 3 bladed, up-wind, pitch regulated turbine, mounted on a tubular steel tower. The proposed configuration has a 116m rotor diameter, a generating capacity of 3.0MW, and a hub height of 92m. The cut-in wind speed is 3m/s and cut-out wind speed is 25m/s. Rated power is reached at a wind speed of approximately 13m/s. The rotational speed of the rotor ranges from 10.1 RPM to 15.5 RPM. These general specifications are summarized in table 1.

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 3

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Acciona AW3000/116 General Specifications Rotor Diameter (m)

116

Hub Height (m)

92

Generating Capacity (MW)

3.0

Cut-in Wind Speed (m/s)

3

Cut-out Wind Speed (m/s)

25

Rated Wind Speed (m/s)

13

Rotor Speed (RPM) 10.1 – 15.5 Table 1: Turbine General Specifications

2.3

Turbine Locations The proposed wind farm layout is shown in figure 1 and coordinates for the 10 turbines are summarized in table 2.

Turbine Coordinates (UTM Zone 20 NAD83) Easting Northing Elevation Turbine ID (m) (m) (m asl.) 1

551822

5144883

34

2

552428

5144850

39

3

553133

5144945

36

4

553902

5144900

36

5

554837

5144944

31

6

555417

5145353

30

7

552347

5144141

51

8

552848

5144106

47

9

553510

5143806

50

10

554389 5143748 Table 2: Turbine Coordinates

42

3.0 Background on Shadow Flicker Shadow flicker from a wind turbine can be defined as alternating changes in light intensity caused by shadows cast from the moving blades. Shadow flicker only occurs under certain conditions and only affects a limited area surrounding a wind farm. However, shadow flicker can be a disturbance to residents living near a wind farm and therefore it is important to assess and ensure that exposure is limited. In this report, a location that may be sensitive to shadow flicker is referred to as a shadow receptor.

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 4

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

For shadow flicker to occur, the sky must be clear, and the turbine must be operating, otherwise no moving shadows are cast. For shadow flicker to occur at the location of a shadow receptor, the turbine rotor must be located in an unobstructed line of sight from the receptor to the sun. As the position of the sun changes throughout the day and throughout the year, the area that is affected by shadow flicker also changes. Furthermore, for shadow flicker to be visible, the change in light intensity must be above the level of perception of the human eye. The distance between a wind turbine and a receptor affects the intensity of the flickering. Shadow flicker intensity decreases with greater separation from the receptor to the turbine, up to a point where the change in light intensity is below what the human eye can distinguish. Shadows cast close to a turbine are more intense, distinct and ‘focused’ because a greater proportion of the sun is intermittently blocked by the passing blades. As separation between the receptor and the turbine increases, the proportion of the sun that is blocked decreases and the shadows become less intense and less discernible. At a distance of about 10 times the diameter of the rotor, the intensity of shadow flicker is significantly reduced and becomes less perceptible to the human eye1. Shadow flicker intensity is also reduced if the plane of the rotor is at an angle other than perpendicular to the line of sight from the receptor to the sun, again because a smaller proportion of the sun is blocked by the passing blades. Ambient lighting conditions also affect the visibility of shadow flicker. Shadow flicker is more noticeable in a darkened room with a window facing the turbine than outdoors where ambient light levels are higher. The frequency or speed of the flickering is related to the rotor speed and the number of blades on the turbine. Modern utility sized turbines are typically 3-bladed with rotor speeds below 20 RPM. This translates to blade passing frequencies less than 1 Hz or 1 cycle per second. At these low frequencies, shadow flicker is a concern for nuisance reasons, but does not pose a health threat. According to the British Epilepsy Association, frequencies below 3Hz do not cause seizures in people with photosensitive epilepsy, and the flicker rates of modern wind turbines are unlikely to trigger epileptic seizures2. In general shadow flicker occurs during clear sky conditions, when the sun is low on the horizon, either at sun rise or sun set. As the elevation of the sun in the sky changes throughout the year, the location of the shadow flicker also changes, so a specific shadow receptor is only affected at certain times of day and at certain times of year. By considering the spatial relationship between the turbines and the receptors (geographic locations and ground elevations) as well as the geometry of the turbines (hub height and rotor size), the occurrence of shadow flicker can be accurately modeled and predicted to within a few minutes at any location around the wind farm.

4.0 Shadow Receptors The project area was assessed to identify all potential shadow receptors within 1500m of the proposed turbine sites. In total 18 receptors were identified, consisting of permanent _____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 5

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

residences, seasonal residences, several buildings which appeared to be uninhabited, and 1 business. Figure 1 shows the receptor database in relation to the proposed turbine sites. The receptor coordinates are summarized in table 3.

Figure 1: Turbine Sites and Shadow Receptors

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 6

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Shadow Receptor Coordinates (UTM Zone 20 NAD83) Receptor ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Easting (m) 550459 550992 552160 552597 552669 553135 553139 553747 553766 553795 553858 554408 554669 554796 555865 555876 555900 556621

Northing (m) 5145443 5145619 5145411 5145449 5145439 5145567 5146057 5145507 5145511 5145511 5145760 5145488 5145571 5145572 5144763 5144724 5144055 5145780

Elevation (m asl.) 28 22 34 36.3 36.5 29 19.4 34 34 34 30 34.4 32.2 31.1 22 22 36.4 20.9

Distance to Nearest Turbine (m) 1474 1109.1 621.7 622.4 636.6 622.2 1112.1 626.5 626.2 620.4 860.7 692.6 649.1 629.6 741.4 778.7 1385.2 1277.5

Table 3: Shadow Receptor Coordinates

5.0 Shadow Flicker Analysis 5.1

Methodology Shadow flicker modeling was performed with the GH WindFarmer software package. The WindFarmer shadow flicker model determines the theoretical maximum amount of shadow flicker, in total hours of flicker per year, at any point up to the maximum specified calculation distance from the turbines. By defining specific shadow receptor locations, the model can also determine the time of day, day of year, and duration of every possible occurrence of shadow flicker at a receptor3. The shadow flicker model uses the following inputs: • • • • •

Geographic location of the wind plant Turbine locations Receptor locations Digital terrain map (ground elevation data) Turbine geometry (hub height and rotor diameter)

The amount of shadow flicker determined by the model is the theoretical maximum amount due to the following assumptions: • •

Every day is sunny and cloudless The turbines are always operating

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 7

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

• • •

The rotor plane is always perpendicular to the sun There are no obstacles such as trees or walls between the receptors and the turbines (it is assumed that all receptors have windows facing all directions with unobstructed view of the turbines) The limits of human perception of changing light intensity are not considered

The theoretical maximum amount of shadow flicker will never occur due to the impossibility of the above combination of assumptions. The theoretical maximum shadow flicker can be statistically de-rated to be more representative of actual, worst-case conditions using the following climatological data: • •

Wind speed frequency distribution at hub height Sunshine hours from long term monthly reference data

The de-rated hours of shadow flicker per year are still a conservative, worst-case scenario, as there is no consideration given to the directional wind distribution, turbine availability, blocking obstacles, location of windows, or the intensity of the flicker.

5.2

Model Parameters The following parameters were used in the shadow flicker model: • • • • • • • • • • • •

5.3

Latitude 46 deg 26 min North Longitude 62 deg 17 min West Calculation time interval 10 Minutes Maximum distance from turbine 1200 m Minimum sun elevation 3 deg Height above ground level 2 m Year of calculation 2013 Consider distance between rotor and tower Yes Turbine orientation Rotor plane facing sun azimuth plus 180º Model the sun as a disc No Terrain: consider turbine visibility Yes Terrain: consider sun visibility Yes

Worst-Case Shadow Flicker The data summarized in table 4 was used to statistically de-rate the theoretical maximum shadow flicker hours to obtain more conceivable, worst-case results. On site wind speed data was used to determine the percentage of time when wind speed at hub height is greater than 3 m/s, which is the cut-in wind speed for the proposed wind turbine. Long term climate records from the Environment Canada weather station in Charlottetown (Charlottetown CDA, Climate ID 8300400) provided data for the percentage of daylight hours with bright sunshine4. The monthly wind speed and sunshine statistics were

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 8

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

combined to obtain monthly de-rating factors. The monthly factors were then averaged to get an overall de-rating factor to be applied to the theoretical maximum amount of shadow flicker at each receptor. It should be noted that the de-rated shadow flicker hours are still a very conservative, worst-case scenario as directional wind statistics (the time when the rotor is actually facing the receptor), turbine availability, blocking obstacles, window locations, and flicker intensity are not considered.

V>3m/s at 92m (% time) Bright sunshine (% of daylight hours) Monthly Derating Factors

Shadow Flicker De-rating Data Apr May Jun Jul Aug

Jan

Feb

Mar

Sep

Oct

Nov

Dec

Year

93.7

92.2

90.9

92.7

81.4

90.4

91.9

93.1

92.8

92.3

93.8

96.2

91.0

37.5

39.7

38.7

37.1

42.5

47.8

52

50.9

46.6

36.3

27.1

27.7

40.3

0.35

0.37

0.35

0.34 0.35 0.43 0.48 0.47 Table 4: De-rating Data

0.43

0.34

0.25

0.27

0.37

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 9

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

5.4

Results The shadow flicker results for the 18 receptors are presented in table 5. 14 out of the 18 receptors may be affected by shadow flicker to varying degrees. These results are discussed in more detail in the next section of the report. A shadow flicker contour map was also produced to show the theoretical maximum hours of shadow flicker throughout the project area at 2 meters above ground level. The shadow flicker contour map is shown in figure 2. A report containing the detailed shadow flicker data has also been produced to provide the time of day, day of year, and duration for every possible occurrence of shadow flicker at each receptor. This report is available upon request.

Shadow Flicker Results Receptor ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Theoretical Maximum Shadow Flicker (Hrs/yr) 0 22 19 64 55 44 0 37 37 45 0 47 46 43 10 11 0 0

Worst Case Shadow Flicker (Hrs/yr) 0.0 8.1 7.0 23.7 20.4 16.3 0.0 13.7 13.7 16.7 0.0 17.4 17.0 15.9 3.7 4.1 0.0 0.0

Maximum Minutes on Worst Day (minutes) 0 30 30 60 50 60 0 40 40 40 10 70 40 40 30 30 0 0

Table 5: Shadow Flicker Results

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 10

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Figure 2: Shadow Flicker Map – Theoretical Maximum

5.5

Discussion of Results A commonly used assessment criterion or allowable limit for shadow flicker is 30 hrs/yr and 30 minutes/day. The theoretical maximum amount of shadow flicker exceeds 30 hrs/yr at 9 of the 18 receptors; however the more realistic, worst-case shadow flicker levels do not exceed 30 hrs/yr for any of the receptors. Given the conservative assumptions used in the shadow flicker model, it is likely that site specific conditions will further reduce the amount of shadow flicker that is actually observed throughout the year. The maximum daily flicker may exceed 30 minutes/day at 9 of the 18 receptors, and may be as high as 70 minutes per day for 1 receptor. These receptors have been considered on an individual basis to determine if site specific conditions will mitigate potential shadow flicker issues. Site specific conditions that may mitigate shadow flicker impact include trees or buildings that block the line of sight to the proposed turbine locations, seasonal or intermittent use, or the absence of windows facing the direction of the wind farm.

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 11

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Receptors 4 and 5 have the highest predicted levels of flicker; however both of these receptors (as well as receptor 3) are located in heavily forested areas with tall trees in close proximity on all sides. In these surroundings is unlikely that these receptors will have a clear view to any of the turbines or experience any flicker at all. Furthermore, receptors 4 and 5 are believed to be seasonal residences and both are only affected by flicker in the late fall and winter months when they are less likely to be occupied. Receptor 6, 10, 13, and 14 are believed to be the only permanent residences in the vicinity of the wind farm that may experience shadow flicker. Receptor 6 may be affected by flicker from turbines 2 and 4, both of which are more than 1000m from the house. At this distance it is likely that the forested area south of receptor 6 will partially or completely block the line of sight to one or both of these turbines. Assuming 3m receptor height, 50m separation to the trees, and 10m tall trees, the line of sight to both turbines would be completely blocked. Receptor 10 may be affected by flicker from turbine 3. Given the proximity to the forested area between receptor 10 and turbine 3, it is unlikely that this receptor will have a clear view of the turbine. Receptor 10 may also be affected by a small amount of flicker from turbine 5; however turbine 5 is located almost 1200m from the house and at this distance the intensity of flicker will be reduced. Receptors 13 and 14 may be affected by flicker from turbines 4 and 6. Turbine 4 is located more 1000m from these houses and given the proximity of the forested area between the turbine and the houses, it is unlikely that they will have a clear view of the turbine. Receptors 8 and 9 may be affected by flicker from turbine 3; however both of these buildings are believed to be vacant. It is also unlikely that either of the buildings will have a clear view of turbine 3 due to the proximity of the forested area separating them. Receptor 12 is a local business operating a distillery and retail sales outlet. This receptor may be affected by flicker from turbines 4, 5, and 6 primarily during the fall and winter months. Flicker from turbine 4 may occur in the afternoon hours from Nov 27 to Jan 13, however a tree line between turbine 4 and receptor 12 may block the view of this turbine. Flicker from turbine 5 may occur in the morning hours from Dec 3 to Jan 8. Flicker from turbine 6 may occur in the early morning from Sept 9 to Oct 5, and from Mar 7 to Mar 22; however turbine 6 is more than 1000m from the distillery and there is likely not a clear view of this turbine.

5.6

Potential Mitigation Measures In the event that shadow flicker from the Hermanville Wind Farm is a nuisance for nearby receptors, there are various mitigation measures which could be employed. Mitigation measures could include planting vegetation or tree lines, which will block the line of sight to the turbines causing flicker, or installation of window blinds or awnings.

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 12

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

A more extreme mitigation measure would be to shut down the turbines which are known to cause problematic flicker, during the times when it is known to occur.

6.0 Conclusion and Recommendations Shadow flicker from the proposed Hermanville Wind Farm has been assessed using a shadow flicker model to determine the theoretical maximum amount of flicker that is possible at each receptor location. Wind speed data and long term sunshine data were also used to determine a more representative, worst-case value for shadow flicker hours at each receptor. The results from this analysis indicate that 14 out of the 18 receptors may be affected by shadow flicker to varying degrees. Of these 14 receptors, 9 have a theoretical maximum flicker level above 30 hours per year, or may experience more than 30 minutes of flicker per day. The more representative, worst-case values doe not exceed 30 hours per year for any of the receptors. Several receptors were considered on an individual basis to determine if site specific conditions may mitigate potential shadow flicker issues. Most are expected to observe only minimal levels of shadow flicker (or none at all) due to distance to the turbines and the presence of trees in the line of sight to the turbines. The results of this assessment indicate that shadow flicker from the proposed Hermanville Wind Farm should not be a significant issue. In the event that shadow flicker is a nuisance for nearby receptors there are various mitigation measures, as previously described, which could be employed. There may be a requirement for some mitigation efforts once the wind plant becomes operational and the impact of flicker can be more closely assessed on a case by case basis.

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 13

Hermanville Wind Farm

Shadow Flicker Assessment

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

7.0 References 1

Onshore Wind: Shadow Flicker, Department for Business Enterprise and Regulatory Reform (BERR), website: http://www.dti.gov.uk/energy/sources/renewables/planning/onshore-wind/shadowflicker/page18736.html London, UK, November 2009.

2

Photosensitive Epilepsy, Epilepsy Action (British Epilepsy Association), website: http://www.epilepsy.org.uk/info/photo.html Leeds, UK, November 2009.

3

GH WindFarmer Theory Manual, Garrad Hassan and Partners Ltd., Bristol, UK, May 2009.

4

Canadian Climate Normals 1971-2000, Environment Canada, website: http://climate.weatheroffice.ec.gc.ca/climate_normals/index_e.html Canada, November 2009.

_____________________________________________________________________________________________________________________

Frontier Power Systems Inc.

Page 14