Mitigation measures (www.marinemammalmitigation.co.uk) are utilised to minimise potential effects of industrial activities on marine mammals. Marine Mammal Observers (MMO; www.marinemammalobservers.co.uk) and Passive Acoustic Monitoring (PAM) Operators (www.pamoperator.co.uk) are used commonly to minimise impact of industrial activities such as dredging, seismic surveys, pile driving, or navy SONAR. Monitoring the area prior to work commencing allows MMOs and PAM Operators to ensure no marine mammals are in close proximity (usually 500 m) to the sound source.
The mitigation method used commonly by fisheries and aquaculture sites, and prior to high intensity noise (e.g. explosions) are acoustic deterrents. Known as Pingers, Acoustic Mitigation Devices (AMD), Acoustic Deterrent Devices (ADD), Acoustic Harassment Devices (AHD), and ‘seal scarers/scrammers’, they emit loud, aversive sounds with the aim of provoking avoidance reactions in marine mammals, thus excluding them from the area. Aggregations of prey in fisheries or aquaculture sites make them attractive feeding grounds for marine mammals, so conflict is high and can result in injury or entanglement in nets. Other potential effects/losses at aquaculture sites include fish injury (which decreases fish value/increases disease susceptibility), reduced fish growth rates (due to increased stress levels), fish pen damage, increased maintenance and repair costs, loss of fish stock through tears in damaged pens, and two-way genetic contamination/disease-transmission between wild and farmed fish stocks.
Acoustic deterrent devices in use today vary substantially. To be effective, they must produce noise within the hearing range of the target species, thus those designed to deter seals should have different spectral properties to those focused on dolphins or porpoises. Reactions are likely to be situation specific, and depend upon motivation. For example, fish farms are a plentiful food source, so animals may be more likely to withstand the noise, than in other areas where benefits are minimal.
Despite widespread use of acoustic deterrents, they are not without downsides. One of the primary concerns is that they increase the level of anthropogenic noise in the ocean. Although some are designed with target species in mind, noise is detectable, and therefore a potential disturbance to other species. Johnston (2002) monitored the impact of AHDs, designed to deter seals, on harbour porpoises (Phocoena phocoena) in the Bay of Fundy, Canada. Fewer harbour porpoise sightings were recorded when the AHDs were active, and the closest approach distance increased from 364 m when the AHD was inactive to 991 m when it was active. Similarly, Morton and Symonds (2002) (http://bit.ly/1fjXooP) reported that killer whales (Orcinus orca) avoided AHDs deployed on salmon farms in the Broughton Archipelago (http://bit.ly/1jSLypl). Again, the primary purpose was to prevent seals from predating on fish. Once AHDs were removed, killer whale presence increased back to levels observed prior to AHD use. If multiple AHDs are in use within a small area, they could exclude marine mammals from a large proportion of their habitat. The level of impact this has will depend upon the importance of the area to the individual.
The long-term effectiveness of acoustic deterrents is uncertain because of the potential of marine mammals to get used to, or habituate to, the noise. Producing acoustic deterrents with highly randomised signals could reduce the chance of this happening, and will likely induce a stronger startle response.
Cox et al. (2001) (www.cetaceanbycatch.org) reported that harbour porpoises in the Bay of Fundy (http://en.wikipedia.org/wiki/Bay_of_Fundy) were initially displaced by a pinger, but that the level of displacement decreased over time. Minimum sound pressure level of the pinger was 132 dB re 1 µPa @ 1 m, and fundamental frequency was 10 kHz. In contrast, Crosby et al. (2013) (http://bit.ly/Neftu1) deployed one banana pinger, set on a 21-hour on/off cycle alongside a C-POD (www.cpodclickdetector.com) for eight months. A second C-POD was moored 150 m away. Both porpoise and dolphin detections reduced substantially when the pinger was active, and habituation was not evident. Banana pingers have source levels of 145 dB re 1 µPa @ 1 m, and frequencies of 50–120 kHz. When tested on fishing boats a lower number of detections were recorded by C-PODs on nets with pingers than without. These results indicate that banana pingers have the potential to reduce by-catch of porpoises and dolphins in fishing nets.
Carretta and Barlow (2011) assessed the long-term effectiveness of pingers operating at a drift gillnet fishery for swordfish and thresher shark in California. Pingers had a source level of 135 dB re 1 µPa @ 1 m rms (Root Mean Square), frequencies of 10–12 kHz, and harmonics up to 80 kHz. They were active at the fishery from 1996–2009. Fisheries Observers monitored levels of by-catch in nets without pingers and nets with ≥30 pingers. Short-beaked common dolphin (Delpinus delphis) by-catch rates were significantly lower in nets with pingers than without. Results were not significant for pinnipeds, but number of entanglements increased for California sea lions (Zalophus californianus), and decreased for Northern elephant seals (Mirounga angustirostris). The increase in California sea lion by-catch corresponded to a general increase in population size, so could not be attributed to pinger presence alone. Early (1996–2001) and late (2001–2009) periods of the study were compared to assess long-term effectiveness. By-catch rates did not differ significantly; suggesting habituation was not an issue.
PORPOISE ALARM (PAL)
As an alternative to alerting stimuli, a new mitigation device, Porpoise ALarm (PAL) uses synthesised harbour porpoise click trains to deter harbour porpoises from areas of harm. The click trains are designed to resemble aggressive clicks emitted by a female towards a male (Conrad et al. 2013) (http://bit.ly/1h0sKB1).
Click train characteristics are based upon research by Clausen et al. (2010) (http://bit.ly/MeC9t8), who analysed click trains and behaviour of captive harbour porpoises simultaneously. Aggressive interactions between a female and male were characterised by high repetition rate clicks of 150–1,100 clicks s-1. Received levels reached 180 dB re 1 µPa p-p (peak-to-peak).
The PAL is still in its infancy, but results of field trials suggest that harbour porpoises do not change their course when confronted with an active PAL, thus are neither attracted to nor repelled by the noise. Echolocation activity did, however, increase in response to the alarm. The theory is that by increasing echolocation, it increases the chance of harbour porpoises being detected acoustically during real-time mitigation prior to operation start-up, and for example, the chance that harbour porpoises will detect fishing nets (Conrad et al. 2013). To date, no evidence exists to support either claim.
Using biological sounds as a form of mitigation is questionable. The function of marine mammal vocalisations is mostly unknown, and interpretation of communicative functions of acoustic signals in any species requires information on the social and environmental contexts in which the sounds are used. For many species, including the harbour porpoise, research on the communicative function of sounds is very limited. The response of marine mammals to sounds of conspecifics in playback experiments varies substantially depending on a number of factors (Deecke 2006) (http://bit.ly/1eXyB5F). For example, vocalisations have specific functions, which influence how different sexes, age groups, or pods react. Sounds of female humpback whales (Megaptera novaeangliae) in social groups elicited strong attraction reactions in male humpback whales, but mother and calf pairs, and other active groups avoided the same sounds (Tyack 1983) (http://bit.ly/MeCvQu). A playback experiment on bottlenose dolphins (Tursiops truncatus) demonstrated that individuals react differently to calls of their own calves than to other young individuals (Sayigh et al. 1999), and killer whales have been found to respond differently to calls of their own pod, than to unfamiliar pods (Filatova et al. 2011).
Use of biological sounds has been trialled in the past, with minimal success. In an attempt to deter sperm whales (Physeter macrocephalus) from shipping lanes, Andŕe (1997) played back a series of artificial sounds, including codas (http://en.wikipedia.org/wiki/Sperm_whale#Types_of_vocalization). Sperm whales reacted to the synthesised codas by ceasing vocalising, but other changes in behaviour were not evident. Consequently, it was concluded that coda sounds would not be effective at reducing sperm whale interaction with vessels. Similarly, Nowacek et al. (2004) (http://bit.ly/1ffxC6O) tested the response of North Atlantic right whales (Eubalaena glacialis) to playbacks of vessel sound, conspecific social sounds and an acoustic alarm sound, with a view to using acoustics to reduce the number of ship strikes. The whales did not respond to vessel noise, showed minimal responses to sounds of conspecifics and reacted most strongly to the acoustic alarm.