Scottish Marine and Freshwater Science Volume 4 Number 5: Modelling of Noise Effects of Operational Offshore Wind Turbines including noise transmission through various foundation types
This report presents modelling of the acoustic output of operational off-shore wind
turbines and its dependence on the type of foundation structure used.
5 Effect of acoustic output on marine life
5.1 Marine mammals
The potential effect that noise emitted from different foundation types may have on marine life can be examined by comparing the modelled near- and far-field sound pressure level to curves representing the hearing and behavioural response of marine species. Curves that characterise the hearing and behaviour of four marine mammals were examined: bottlenose dolphins, porpoise, minke whales and grey and harbour seals (together). These species represent marine mammals that are common in Scottish waters. Composite audiograms that represent the low frequency hearing thresholds of the marine mammal species are shown in Figure 5-1 (see section 2.5.1 for details of audiogram compilation).
Figure 5-1 - Audiograms of marine mammals; see section 2.5.1 for information on audiogram compilation.
Table 5-1 - Parameters representing the behavioural response and risk of injury to marine mammal species [ SMRU REFS]
| Species | Seals | Harbour porpoise | Bottlenose dolphin | Minke Whale | |
|---|---|---|---|---|---|
| Functional hearing group | Pinniped in water | High-frequency cetacean | Mid-frequency cetacean | Low-frequency cetacean | |
| M-weighting | Estimated auditory band minimum | 75 Hz | 200 Hz | 150 Hz | 7 Hz |
| Estimated auditory band maximum | 75 kHz | 180 kHz | 160 kHz | 22 kHz | |
| Sensation : Added to one-third octave bands | Lower | 45 | 49 | 49 | 49 |
| Upper | 59 | ||||
| Behavioural response SPLs ( RMS) | 10% response | 135 | 90 | 120 | 120 |
| 50% response | -- | 120 | 140 | 140 | |
| 90% response | 144 | 140 | 160 | 160 | |
The sound field produced by each wind farm can be examined to determine where the SPL is greater than the hearing threshold of each species for any given frequency. The audiograms shown in Figure 5-1 were applied in this way to the modelled far-field sound field and the maximum range at which marine mammals could hear the wind farm determined as a function of frequency ( Figure 5-2 to Figure 5-5). It is assumed that if the background noise exceeds the SPL produced by the wind farm that the noise from the farm is masked and cannot be detected by marine species. Thus, the maximum range at which species can hear the wind farm is less than or equal to the range shown in Figure 4-7.
Of the species of interest considered here, the minke whale has the most sensitive (i.e. best) hearing at low frequency (<2000 Hz) ( Figure 5-1). The modelling outputs predict that minke whales are able to detect the wind farm (with monopile or gravity foundations) at least 18 km away ( Figure 5-3) at most frequencies below 800 Hz under all three wind conditions modelled. A minke whale could also detect a wind farm founded on jackets at 630 Hz at all three wind speeds ( Figure 5-3) at large ranges. The seal species have less sensitive hearing than minke whales, particularly at very low frequencies (<100 Hz) ( Figure 5-1), resulting in seals not being able to detect the wind farm below 100 Hz independent of wind speed or foundation type ( Figure 5-2). However, seals can detect the wind farm up to at least 18 km in one third octave bands between 125 and 630 Hz at all wind speed conditions ( Figure 5-2). Bottlenose dolphins and harbour porpoises are less sensitive to low frequency sound than either minke whales or the seals species. However both species can still detect the operating wind farm under different foundation and wind speed scenarios. Harbour porpoises can only detect the jacket foundation wind turbines in the 630 Hz band at wind speeds of 10 ms -1 and 15 ms -1 out to 4 km and 11 km respectively but can detect the gravity and monopile foundations out to at least 18 km ( Figure 5-4). A bottlenose dolphin can detect a wind farm mounted on a gravity base 4 km away in wind speeds of 10 ms -1 and 15 km at 15 ms -1; though it can detect jackets and monopiles only at close ranges of ~1 km ( Figure 5-5).
Figure 5-2 - Maximum range at which a harbour seal could hear a wind farm at different wind speeds. Gravity base, jacket and monopile foundations are compared. It is assumed that if the SPL is below the background noise that a seal could not hear the wind farm. The range is measured to the centre of the wind farm.
Figure 5-3 - Maximum range at which a minke whale could hear a wind farm at different wind speeds. Gravity base, jacket and monopile foundations are compared. It is assumed that if the SPL is below the background noise that a minke whale could not hear the wind farm. The range is measured to the centre of the wind farm.
Figure 5-4 - Maximum range at which a porpoise could hear a wind farm at different wind speeds. Gravity base, jacket and monopile foundations are compared. It is assumed that if the SPL is below the background noise that a porpoise could not hear the wind farm. The range is measured to the centre of the wind farm.
Figure 5-5 - Maximum range at which a bottlenose dolphin could hear a wind farm at different wind speeds. Gravity base, jacket and monopile foundations are compared. It is assumed that if the SPL is below the background noise that a bottlenose dolphin could not hear the wind farm. The range is measured to the centre of the wind farm.
5.1.2 Behavioural response zones
The behavioural response to noise emitted by wind turbines is examined in two ways. Firstly a sensation parameter is added to the sound-field to determine behavioural response as a function of frequency. The upper and lower range of the sensation parameters used for each species are shown in Table 5-1. To determine the sensation level that may result in a behavioural response in seals the sensation parameters were added to the sound level integrated across the two neighbouring one-third octave bands either side of the band of interest. In the case of cetaceans the sensation level was found by adding the sensation parameters to the SPL integrated across the four neighbouring bands on the dominant side of the band of interest.
Potential behavioural response zones were calculated for each of the five marine mammal species of interest using three different metrics ( section 2.4.1). This behavioural response zone considers masking by background noise, i.e. if the background noise exceeds the noise produce by operational wind turbines then the marine species are assumed not to change their behaviour. The results suggest a marked variation in response between the species ( Table 5-2). Neither seal species nor bottlenose dolphins were predicted to exhibit a behavioural response to the sounds generated under any of the operational wind turbine scenarios. This means the predicted sound levels were lower than those required to elicit a behavioural response.
Harbour porpoises were only predicted to exhibit an aversive behavioural response using the M-weighting metric where 10% of animals encountering the noise field were expected to move away ( Table 5-2). The '10% avoidance ranges' varied depending on the foundation used and the wind speed. At low wind speeds (i.e. 5 ms -1) ranges were between 0 (jacket) and 1.7 km (monopile). At higher wind speeds (10 & 15 ms -1), avoidance ranges were predicted to be between 9.45 km (jacket foundation / 10 ms -1) and 18.84 km (monopile foundation / 10 & 15 ms -1). Avoidance ranges where 50% or 90% of porpoises were predicted to respond were not generated in any of the scenarios and therefore most harbour porpoises are not expected to respond to the operational noise.
Minke whales were determined to be more sensitive to the wind turbine noise than the other species of interest. For all metrics, the behavioural response ranges were largest for the monopile foundation, followed by the gravity base and lowest for the jacket foundation. As with the harbour porpoises, the minke whale behavioural response ranges increased as wind speed increased. The sensation level metric used indicated ranges of up to 18.84 km ( Table 5-2, Figure 5-6 and Figure 5-7). However, the reverse-audiogram weighting and M-weighting approaches suggest that 10% of animals encountering the noise field were expected to move away at ranges between 3.7 km ( RA-weighting) and 12.71 km (M-weighting from the source (both when wind speed is 15 ms -1).
Table 5-2 - Behavioural response zones for each species, foundation and wind speed scenario modelling here. The sensation levels and weighted (M-weighting and reverse audiogram weighting) response ranges are shown (in km). For the Reverse Audiogram and Southall m-weighting, the thresholds correspond to the SPLs at which 10%, 50% and 90% of animals encountering it, are predicted to respond. The absolute threshold is shown for each species for reference.
| Species | Foundation | Wind speed | Lower | Upper | 10% | 50% | 90% | 10% | 50% | 90% |
|---|---|---|---|---|---|---|---|---|---|---|
| Minke whale | 49 | -- | 120 | 140 | 160 | 120 | 140 | 160 | ||
| Gravity | 5ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 18.57 | -- | 0 | 0 | 0 | 1.58 | 0 | 0 | ||
| 15ms | 18.84 | -- | 1.7 | 0 | 0 | 4.67 | 0 | 0 | ||
| Jacket | 5ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 7.36 | -- | 0 | 0 | 0 | 0 | 0 | 0 | ||
| 15ms-1 | 18.18 | -- | 0 | 0 | 0 | 0 | 0 | 0 | ||
| Monopile | 5ms-1 | 2 | -- | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 18.84 | -- | 1.7 | 0 | 0 | 4.81 | 0 | 0 | ||
| 15ms-1 | 18.84 | -- | 3.7 | 0 | 0 | 12.71 | 0 | 0 | ||
| Seals | 45 | 59 | -- | 135 | 144 | -- | 135 | 144 | ||
| Gravity | 5ms-1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | ||
| 15ms-1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | ||
| Jacket | 5ms-1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | ||
| 15ms-1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | ||
| Monopile | 5ms-1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | ||
| 15ms-1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | ||
| Bottlenose dolphin | 49 | -- | 120 | 140 | 160 | 120 | 140 | 160 | ||
| Gravity | 5ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | ||
| 15ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | ||
| Jacket | 5ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | ||
| 15ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | ||
| Monopile | 5ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | ||
| 15ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | ||
| Harbour porpoise | 49 | -- | 90 | 120 | 140 | 90 | 120 | 140 | ||
| Gravity | 5ms-1 | 0 | -- | 0 | 0 | 0 | 1 | 0 | 0 | |
| 10ms-1 | 0 | -- | 0 | 0 | 0 | 19 | 0 | 0 | ||
| 15ms-1 | 0 | -- | 0 | 0 | 0 | 19 | 0 | 0 | ||
| Jacket | 5ms-1 | 0 | -- | 0 | 0 | 0 | 0 | 0 | 0 | |
| 10ms-1 | 0 | -- | 0 | 0 | 0 | 9.45 | 0 | 0 | ||
| 15ms-1 | 0 | -- | 0 | 0 | 0 | 18.58 | 0 | 0 | ||
| Monopile | 5ms-1 | 0 | -- | 0 | 0 | 0 | 1.7 | 0 | 0 | |
| 10ms-1 | 0 | -- | 0 | 0 | 0 | 18.84 | 0 | 0 | ||
| 15ms-1 | 0 | -- | 0 | 0 | 0 | 18.84 | 0 | 0 | ||
Figure 5-6 - Maximum range that sound emitted by a wind farm may have produced a behavioural response in the most sensitive minke (lower sensation range). Gravity base, jacket and monopile foundations are compared. The range is measured to the centre of the wind farm.
Figure 5-7 - Maximum range that sound emitted by a wind farm may have produced a behavioural response in the least sensitive minke (upper sensation range). Gravity base, jacket and monopile foundations are compared. The range is measured to the centre of the wind farm.
5.2 Fish
Audiograms of four species of fish were applied to the far-field sound field to determine the range at which they could detect the wind farm. The species examined were European eels, allis shad, sea trout and Atlantic salmon ( Figure 5-8). Of the four species examined European eel is most sensitive to low frequency sound (<300 Hz) ( Figure 5-8) resulting in eel being able to detect sound produced by the modelled wind farm at the greatest range at low frequency ( Figure 5-9 to Figure 5-12). Eels can detect wind turbines founded on monopiles up to at least 18 km operating at all wind speeds modelled ( Figure 5-9). Eels can detect turbines mounted on gravity bases operating in 10 ms -1 and 15 ms -1, but are unable to detect turbines mounted on jackets at all in the far-field ( Figure 5-9). Salmon cannot detect far-field noise from any of the foundation types at 5 ms -1, but can detect those founded on monopiles at 13 km in 10 ms -1 and gravity bases up to 14 km in 15 ms -1 ( Figure 5-10). Both shad and sea trout are relatively insensitive to the far-field sound produced by wind turbines ( Figure 5-11 and Figure 5-12) with only trout able to detect jackets within 3 km and gravity bases within 2 km at wind speeds of 15 ms -1 ( Figure 5-8).
Figure 5-8 - Audiograms of fish: eels based on Jenko, et al. (1989), shad based on Mann, et al. (2001), Atlantic salmon based on Hawkins and Myrberg (1983) and sea trout based on Horodysky, et al. (2008).
Figure 5-9 - Maximum range at which a European eel could hear a wind farm at different wind speeds. Gravity base, jacket and monopile foundations are compared. It is assumed that if the SPL is below the background noise that an eel could not hear the wind farm. The range is measured to the centre of the wind farm.
Figure 5-10 - Maximum range at which an Atlantic salmon could hear a wind farm at different wind speeds. Gravity base, jacket and monopile foundations are compared. It is assumed that if the SPL is below the background noise that a salmon could not hear the wind farm. The range is measured to the centre of the wind farm.
Figure 5-11 - Maximum range at which a shad could hear a wind farm at different wind speeds. Gravity base, jacket and monopile foundations are compared. It is assumed that if the SPL is below the background noise that a shad could not hear the wind farm. The range is measured to the centre of the wind farm.
Figure 5-12 - Maximum range at which a sea trout could hear a wind farm at different wind speeds. Gravity base, jacket and monopile foundations are compared. It is assumed that if the SPL is below the background noise that a sea trout could not hear the wind farm. The range is measured to the centre of the wind farm.
Contact
There is a problem
Thanks for your feedback