PAM equipment used most often in industry consists of a single in-line hydrophone array (i.e. all hydrophones are in-line along a single cable); a necessary design for industrial operations to minimise the risk of entanglement.
When working offshore as a Passive Acoustic Monitoring (PAM) Operator (www.pamoperator.co.uk), questions often arise as to how an animal is located using PAM equipment (www.towedhydrophonearrays.com). Determining location can be import during industry operations, as marine mammals detected within pre-defined exclusion zones may cause a delay or shut-down in operations (www.marinemammalmitigationplan.com).
PAM equipment used most often in industry consists of a single in-line hydrophone array (i.e. all hydrophones are in-line along a single cable); a necessary design for industrial operations to minimise the risk of entanglement. In reality, localising animals in three dimensions is impossible using a single in-line hydrophone array; however, a single in-line array of at least two hydrophones can achieve two-dimensional localisation using the signal Time Differences Of Arrival (TDOA) method.
The basic principle behind TDOA-based methods is that, TDOA between signals received at each hydrophone element is related to propagation path and position of the marine mammal sound source. The below schematic illustrates the TDOA principle, that a hydrophone nearest to a vocalising animal will detect signals before the hydrophone furthest away.
There are two main steps to TDOA-based methods: firstly, a Time-Delay Estimation (TDE), and secondly a time-space inversion, which forms the position of the ‘radiating marine mammal sound source’ from the group of estimated TDOA, related to the geometry of the array. For example, if a signal arrives at a hydrophone closest to the vessel first, then 3 ms later arrives at another hydrophone 7.5 m further down the array, the sound source originated from an angle of 53° (assuming a speed of sound in water of 1,500 m s-1). PAMGuard (the most common monitoring software used in industry; www.pamguard.org) calculates TDOA and displays bearings to vocalisations.
If towing an in-line hydrophone array from a vessel sailing in a straight line, it is not possible to determine which side of the vessel a sound source originates from (with any of the above methods), because TDOA is the same from either side, referred to as ‘left-right ambiguity’. Left-right ambiguity can be resolved by either turning the vessel (not possible on a seismic line, or during industrial operations generally) or by using a twin-line array (i.e. two linear arrays parallel to each other). Aside from Research and Development (R&D) PAM systems not available commercially, twin arrays are not used widely in industrial operations because of increased costs and limited deployment possibilities. Localisation methods are explained in more detail by Au and Hastings (2008) and Zimmer (2011), and are being improved continually for marine mammals, for example, Baggenstoss (2013), Nosal (2013), and Rideout et al. (2013).
The following illustration shows how bearings calculated from a single in-line array lead to left-right ambiguity. Left-right ambiguity, resolved with a turning vessel, is also illustrated. The bearing lines gather on one side, indicating whether the vocalisation originates from the left or right. A twin array works on TDOA principles, in that, if a signal reaches the left array before the right array, the sound originated from the left. Bearing lines will also line up.
Periodically, clients raise the question of determining marine mammal positions in relation to a noise source, so marine mammal localisation and left-right ambiguity are important concepts to comprehend. Clients are often unaware of PAM technology limitations, and may make unrealistic demands if these principles are not explained.
|Au W.W.L. & Hastings M.C. (2008) Principles of marine bioacoustics. Springer|
|Baggenstoss P.M. (2013) Processing advances for localization of beaked whales using time difference of arrival.|
|Journal of the Acoustical Society of America 133, 4065-76.|
|Nosal E.-M. (2013) Methods for tracking multiple marine mammals with wide-baseline passive acoustic arrays.|
|Journal of the Acoustical Society of America 134, 2383-92.|
|Rideout B.P., Dosso S.E. & Hannay D.E. (2013) Underwater passive acoustic localization of Pacific walruses in the|
|northeastern Chukchi Sea. Journal of the Acoustical Society of America 134, 2534-45.|
|Zimmer W.M.X. (2011) Passive Acoustic Monitoring of Cetaceans. Cambridge University Press, New York,|
|United States of America.|