Evaluating and Assessing the Relative Effectiveness of Acoustic Deterrent Devices and other Non-Lethal Measures on Marine Mammals

6 Collisions between Marine Mammals and Vessels

Marine mammals which are in collision with vessels (ship strikes) are often seriously injured or killed. This is certainly an animal welfare issue and for some whale populations, such as North Atlantic right whales ( Eubalaena glacialis) ( IWC- ACCOBAMS, 2011; Knowlton and Kraus, 2001) and Mediterranean fin whales ( Balaenoptera physalus) (David et al., 2011) and it is a major conservation issue as well. In some cases whale collision can also lead to costly damage to vessels and even injury or mortality for passengers and crew (Carrillo and Ritter, 2010; Kato et al., 2012). Small cetaceans, whales and pinnipeds all suffer mortality from vessel collisions (Byard et al., 2012) and indications are that the number of vessel strikes and the rate of mortality is growing. This is likely to be the result of a greater number of ships travelling at higher speeds (Gerstein et al., 1996; Laist et al., 2001). The marine mammal ship strike issue is receiving increasing attention in international fora, for example it is on the agendas of both the International Whaling Commission ( IWC) and the International Maritime Organisation ( IMO). The IWC maintains a database of known ship collisions and this problem, how to measure it and potential solutions, have been the subjects of a number of workshops (> e.g. IWC- ACCOBAMS, 2011).

A number of different solutions have been proposed. Clearly, separating vulnerable animals and ships is a sensible action in locations where shipping lanes and predictable areas of high marine mammal abundance coincide and it is feasible to move the shipping to pass through an area with a lower density (Vanderlaan et al., 2008). This approach has been applied successfully off the East Coast of Canada in the Bay of Fundy, where a small change in the route of traffic separation zones has resulted in a much reduced risk of collision. Another strategy is to reduce vessel speed and to request seafarers to maintain higher levels of vigilance, especially if vulnerable animals have been sighted in an area. This has been applied adaptively in the shipping lanes for vessels using Boston Harbour which pass through right whale habitat. While these measures will be effective in some locations they do not provide a general solution.

It may initially seem unexpected that marine mammals appear not to detect oncoming vessels and either dive or move out of their way. It could be that animals detect the ships but either do not perceive them as a threat or may lack an innate or learned behavioural repertoire to respond appropriately to vessels moving at these speeds. It could also be the case that they do not detect the vessels in time to take avoiding action. Although large vessels travelling at speed radiate high levels of noise, especially at low frequencies, this is not transmitted equally and there is hypothesised to be a quiet sector directly ahead of large vessels where the noise from the vessel's propellers and machinery are shielded by the boat's hull (see Figure 12). In addition, downward diffraction of sound and the Lloyd mirror effect (a form of destructive interference) may reduce sound levels near the surface (Blue et al., 2001; Gerstein et al., 2011). If this is the case, then an additional acoustic signal which would cause animals ahead of the vessel to move out of the way might be helpful in reducing collisions. Indeed, one potential configuration for this has been patented (Gerstein et al., 1996). Ideally, these alerting signals should be directional, so that only those animals ahead of the boat and at greatest risk of being hit would be induced to move, and the sound should cause animals to move to the side, or to dive.

One major concern would be the potential for substantial disturbance that could result if many vessels used such devices continuously, it might therefore be preferable to activate them only in areas of high risk.

The marine mammal for which there has been the most concerted effort to develop an acoustic alerting device intended to minimise ship strikes is the West Indian manatee. These slow moving animals are often hit by speed boats in inshore waters in Florida and some other southern states of the USA (O'Shea et al., 1985) and elsewhere in their range. Gerstein et al. (2011) described the development of acoustic alerting devices to warn manatees of the approach of vessels and reported on some encouraging initial trials.

Workshops convened to discuss the vessel strike problem with cetaceans and how to address it have given little attention to the use of acoustic alerting signals. For example IWC/ ASCOBANS did not list this as a mitigation option worthy of discussion, while Silber et al. (2009) quickly dismissed it, citing the lack of evidence to show that it would work, the potential for habituation and concerns about causing disturbance. One attempt has been made to measure the behavioural response of a right whale to a potential alerting signal (Nowacek et al., 2004). In a trial in the Bay of Fundy, focal animals responded to playback of the alerting sound, but not in a way that was thought likely to reduce collision risk: they tended to surface. Unfortunately this single preliminary set of experiments has not been extended to test other signals types or scenarios.

The lack of enthusiasm for using alerting signals is understandable, but may be premature. It is certainly sensible to prioritise the low risk management solution of separating whales and shipping in situations wherever this is feasible but this can only address a small subset of what is a global problem and more proactive measures may therefore be required to provide a more general solution. The first task will be to find signal types that can cause animals to move in an appropriate manner and that are suitable for broadcast from a bow-mounted projector. The research required to achieve this will involve controlled exposure experiments to a range of species of wild animals in real world locations, with observations of behavioural responses being made using an appropriate mixture of telemetry, direct visual observation and passive acoustic tracking. Later work to develop practical and cost effective hardware which can be retro-fitted to existing vessels and to make measurement of responses in increasingly realistic scenarios will be contingent on the results of these trials.

Figure 12 Hypothesised patterns of radiated sound from a ship underway from Gerstein et al. (2011). See also Trevorrow et al. (2008).