Passive Acoustic Monitoring (PAM) is a non-invasive tool with rapidly growing applications in bioacoustics (e.g. www.staticacousticmonitoringsystems.co.uk and www.dolphindetectors.co.uk) and noise measurements (www.underwatersoundmeasurement.com). Since World War I, the development of PAM technology has made it possible for researchers to listen, record, store and analyse marine mammal sounds; however, due to high costs and limitations in technological development, only in the early 1990s have these systems become more accessible for most field biologists.
There is an extensive range of PAM systems used for all different types of animals and industrial activities. For marine mammal research and mitigation, two main types of PAM systems are used widely: towed hydrophone systems and Static Acoustic Monitoring Systems (SAMS). Towed hydrophones are often deployed as an array of multiple hydrophone elements along a single cable, and towed by a vessel (www.towedhydrophonearrays.com). Implied in its name, the other category of PAM system, known as SAMS, refer static (i.e. fixed) systems. There are also PAM systems that attach directly onto marine mammals such as e.g. DTAGs (Johnson & Tyack 2003; Tyack et al. 2006).
While most often towed behind a vessel, this PAM system can also be deployed as a vertical array off the side of a stationary vessel, platform or rig. Towed hydrophone arrays are used often in conjunction with visual observations by Marine Mammal Observers (MMOs; www.marinemammalobserver.co.uk), either for marine mammal research or mitigation measures during industrial activities.
Using a six element towed hydrophone array deployed vertically, Kyhn et al. (2013) showed that two species of porpoise with overlapping geographic ranges in British Columbia (BC), Canada, and one in Denmark, have only a slight difference in echolocation click mean centroid frequencies. In BC, Dall’s porpoises (Phocoenoides dalli) had an echolocation click mean centroid frequency of 137±3 kHz, harbour porpoises (Phocoena phocoena) 141±2 kHz, and harbour porpoises in Denmark 136±3 kHz. Researchers also found porpoise clicks to be highly directional. It is possible that these porpoises have not only evolved to operate an effective sonar system, but also to minimise the risk of killer whales detecting their click (http://www.mysciencework.com). Killer whales cannot hear well above 100 kHz and are known porpoise predators. By producing clicks above the hearing threshold of killer whales, porpoises are able to communicate and forage without being overheard by killer whales.
Multiple hydrophone arrays are most often used when trying to localise vocalising animals. One study was even able to use the time delays of Hawaiian spinner dolphin (Stenella longirostris) echolocation clicks and bust pulse sounds, travelling by way of the surface and bottom reflection paths (Aubauer et al. 2000). After measuring water and hydrophone depth and time delays relative to direct signals, researchers were able to measure distance and depth of Hawaiian spinner dolphins accurately.
Due to their costs, cabled SAMS are unusual, and are used only by navies (e.g. the US Navy’s SOund SUrveillance System (SOSUS), https://en.wikipedia.org) or Governmental agencies (Comprehensive Test Ban Treaty Organization (CTBTO), www.ctbto.org). These permanent or semi-permanent cabled PAM systems provide data continuously in near real time, allowing quick response to unusual events. Such PAM systems can also benefit marine mammal scientific research, but due to military access restrictions, data are not always easily accessible, and recordable bandwidth is often limited to low frequencies.
The U.S. Navy’s Atlantic Undersea Test and Evaluation Center (AUTEC) testing range (http://en.wikipedia.org), located in the Tongue of the Ocean (TOTO), Bahamas, has been used to calculate density estimates of sperm whales (Physeter macrocephalus) (Ward et al. 2012). Using 82 deep water bottom mounted hydrophones, marine mammal monitoring (M3R) software, visual observers (MMOs), and DTAGs (http://www.whoi.edu), density estimation of sperm whales at the AUTEC range was found to be 0.158 animals/1,000 km2. Given that the TOTO is a relatively small and isolated area, it is not surprising that this number is lower than other regions around the world (e.g. Whitehead 2002).
Radio-linked hydrophones are also used occasionally for marine mammal acoustic surveys. Data are transmitted back to shore instead of using cables or storing the data on the PAM system.
Navy surplus sonobuoys have been deployed and monitored successfully from an aircraft to localise North Pacific right whales (Eubalaena japonica), allowing researchers to collect photographs and biopsy samples, deploy satellite transmitter tags, and conduct foraging ecology studies on this endangered species (Rone et al. 2012).
Rankin et al. (2005) deployed Directional Frequency Analysis and Recording (DIFAR) sonobuoys (www.sonobuoytechsystems.com) in close proximity to blue whales (Balaenoptera musculus) for sub-species identification of Antarctic blue whales (B. m. intermedia) and pygmy blue whales (B. m. brevicauda). Researchers found that, while the known 3-unit vocalisation type could be attributed to Antarctic blue whales (which has also been verified elsewhere in other studies) the 2nd and 3rd units of the vocalisations were too variable for sub-species identification. Peak frequency of the 28 Hz tones of the 3-unit vocalisations were more stable, even with different signal strengths, and may be used for in field discrimination of Antarctic and pygmy blue whales.
Hydrophones moored on the sea floor can also be autonomous (i.e. are not operated by a person), either recording continuously or according to a set sampling regime. These SAMS are deployed semi permanently, and can be left in the field for extended periods of time. With data being stored internally, there are no cables or radio transmitters; data are retrieved with the device. Sousa-Lima et al. (2013) gives an in depth review of over 40 autonomous SAMS.
Acoustic Recording Packages (ARPs) have been deployed at four circumpolar locations to investigate the seasonal and spatial variations of blue and fin whale (Balaenoptera physalus) calls (Širović et al. 2009). Researchers found that blue whales were detected year round, may not all migrate at the same time, and that long duration calls were more common during migrations. Fin whales, on the other hand, were detected only from February until July. Results revealed two different fin whale calls, which suggest there are two separate populations.
While more uncommon, PAM is also possible as a means to investigate seal vocalisations. Over a two year period in the Beaufort Sea, passive acoustic recorders (Aural-M2, www.multielectronique.com) were used to compare the number of hours with bearded seal (Erignathus barbatus) vocalisations with water temperature and daily sea ice concentrations (MacIntyre et al. 2013). Bearded seal vocalisations were detected all year, with peaks from January to early July, which coincided with the formation of pack ice and the breeding season, and late September or early October. A decrease in detections was found from late June to early July, which was in line with a decrease in ice concentrations. Between the two years, there was a difference in the timing of seasonal sea ice formation and retreat, which was mirrored in bearded seal call activity.
Please see www.dolphindetectors.co.uk for more details on the use of C-PODs and T-PODs for detecting dolphins.
Another variant of autonomous SAMS uses small sensors tagged onto the animal directly. Several species of mysticete and odontocete whales, and others large marine mammal species such as elephant seals (Mirounga angustirostris), have been tagged using these PAM systems.
By using digital acoustic recording tags (DTAGs) Oliveira et al. (2013) was able to show that the behavioural context of sperm whale slow clicks in high latitudes is most likely long range acoustic communication, and not orientation or foraging. Slow clicks were produced most often during the surface phases of the whales’ dive cycles. This study was conducted in Arctic feeding grounds, where only males are present, but slow clicks are also produced in warmer water breeding grounds by males encountering females.
DTAGs have also been used to look at the behavioural response of Cuvier’s beaked whales (Ziphius cavirostris) and blue whales, to Mid-Frequency Active (MFA) sonar signals (DeRuiter et al. 2013; Goldbogen et al. 2013) (www.bbc.co.uk).
Advantages of PAM
Monitoring the presence of marine mammals by visual observation only is limited to good light and weather conditions. It can also be expensive, due to vessel and personnel costs required over extended periods of time. While PAM systems can be expensive initially, they are sometimes more cost effective when they can be left in the field for long periods of time without the requirement of daily observations.
Binoculars offer only a limited field of view; in addition many animals spend a limited amount of time at the surface. Fortunately, many marine mammal species produce loud distinctive and frequent vocalisations, such as harbour porpoises and sperm whales. Passive Acoustic Monitoring systems can enhance detection capability up to approximately eight times more than an MMO alone.
The use of SAMS is especially useful for long term monitoring in remote locations, monitoring under-ice or during inclement weather, as well as being able to operate effectively at night. Another advantage for using fixed PAM systems is that, often marine mammal vocalisations are masked by survey vessel noise. Depending on study objectives, SAMS can be deployed away from noise sources.
Disadvantages of PAM
It is important to remember that marine mammals can only be detected by PAM systems if they are vocalising, and range estimation can be more difficult than when done visually. Static Acoustic Monitoring Systems can only monitor for marine mammals within the immediate vicinity (though range depends on the vocalisation range of target species and ambient propagation conditions). Towed hydrophone arrays, on the other hand, can be used over large areas, but may be limited to the vessel working restrictions (i.e. weather may play a factor in when surveys can take place).