Position Statement 3: Passive Acoustic Monitoring

Position Statement 3: Passive Acoustic Monitoring

The Marine Mammal Observer Association (MMOA) recommends that Passive Acoustic Monitoring (PAM) equipment supplied for marine mammal mitigation purposes should comprise (at least) the following components:

  • An appropriate length of cable to deploy the hydrophone elements at a suitable distance from engine noise in order to be effective. At least one spare cable is recommended in case damage is incurred during deployments.

  • Hydrophone elements with the correct frequency range for the Species of Concern (SoC). This may mean multiple elements with differing frequency sensitivities.

  • Sound cards (and back up drivers) with the appropriate sampling rate for the SoC and full instructions on their configuration.

  • Pairs of hydrophones elements for each frequency range needed in order to provide bearing information for tracking animals.

  • Full details of separation distances of each pair of hydrophone elements for bearing determination.

  • GPS for tracking vocalisations (needed for range and bearing determination).

  • Calibrated depth sensor and full instructions on configuration.

  • Filtering system for unwanted noise and full instructions on configuration.

  • Professional noise cancelling headphones (with battery charger and spare batteries).

  • Appropriate software for real time monitoring which shows bearing and range to the animals and species classification, where possible.

  • Every component of the PAM equipment in duplicate (including the smallest connectors) in case of equipment failures.

  • Full detailed manual of assembly and calibration.

  • Full inventory of contents.

  • Properly labelled shipping containers and reels.

  • 24 hr. support from the PAM equipment provider for technical problems and configuration advice.

Using PAM alongside visual monitoring by Marine Mammal Observers (MMOs) greatly improves the possibility of certain marine mammal species being detected, such as harbour porpoises and sperm whales which vocalise frequently and distinctively. However, clients, regulators and PAM Operator provider companies should be aware of several inherent limitations with the use of PAM for mitigation purposes. These points are vital to understand when considering whether PAM is suitable for use as a sole monitoring method during darkness or other periods of limited visibility (i.e. whether it has a reasonable likelihood of reliably detecting the SoC present within a site).

 

Passive Acoustic Monitoring (PAM) has the following limitations:

  • Many potential SoC will not be detected using PAM equipment because they do not produce vocalisations, only vocalise above water or vocalise only infrequently, for example sea turtles, sirenians, baleen whales and basking sharks, and many species of pinniped. Even cetacean species may not vocalise all the time and may pass close to a PAM system but remain undetected. For example, minke whales in the north Atlantic are known to be very difficult to detect acoustically on their summer feeding grounds, while bowhead whales in Alaska alter their calling behaviour in response to air-gun sounds, ceasing calling altogether when cumulative sound levels reach ~160 dB re 1 µ Pa[1]. Many MMOs/PAM Operators will attest to having visually observed dolphins close to a platform without detecting them acoustically.

  • Limited ability to detect baleen whale vocalisations due to the fact that PAM equipment used for mitigation purposes is usually deployed astern of large commercial vessels and in proximity to continuous low frequency engine noise. Propeller noise (together with water noise and the low frequency sound emitted by the airguns (or other sound source) themselves) masks low frequency biological signals and renders the detection of baleen whale vocalisations virtually impossible using the currently-available PAM systems.

  • In cases where baleen whales are occasionally detected, for example singing humpback whales, the nature of these vocalisations, which travel over many kilometres, make it very problematic to determine range and bearing using current equipment and a PAM Operator may be unable to say with any level of reliability whether an animal is within a MZ or not.

  • Species that vocalise at high frequencies will not be detected beyond short distances. For example, the reliable detection range for harbour porpoise signals is approximately 200-300m. Therefore, a harbour porpoise may be present in the outer region of a MZ (e.g. 300-500m from the source) but not detected by a PAM system.

  • The vocal repertoires of many marine mammal species are poorly-described or unknown and efforts need to be made to address this.

  • Many marine mammal vocalisations, particularly the echolocation click trains produced by odontocetes, are directional in nature and not easily detected by PAM equipment if the animals are facing away from the hydrophone. For example, bow-riding dolphins are often undetected by PAM systems which are deployed astern, while harbour porpoises (which often show avoidance reactions to platforms) will not be detected when swimming away from a vessel’s track line.

  • Range determination is only possible for some species through the tracking of bearing information over time, using the timing of signals arriving at pairs of hydrophone elements deployed at known distances from the ship’s stern and the ship’s GPS position. Current PAM systems with simple pairs of hydrophones are often inadequate at providing suitable bearing information for range determination. In the majority of cases, the current PAM systems used during mitigation surveys do not provide adequate range Information to allow a PAM Operator to determine whether an animal is inside or outside of the MZ. Furthermore, accurate range estimation using current software relies on a hydrophone being deployed in a straight line and at set distances astern of the vessel. It is usually the case during pre-shoot mitigation watches that the vessel is engaged in a slow turn between seismic lines, and consequently the hydrophone will also be angled and curved to one side of the vessel track line causing inaccuracies in range estimation.

The MMOA believes that PAM equipment can be used with reasonable effectiveness during mitigation for some cetacean species, in particular:

  1. The harbour porpoise and other specific small odontocetes (e.g. porpoise species and Cephalorhynchus dolphins) known to emit regular high-frequency echolocation clicks. If these clicks are detected then animals can almost certainly be concluded to be within a few hundred metres of the PAM system, making them a good candidate for PAM-based mitigation.

  2. The sperm whale, which emits very regular clicks throughout its dives and embarks on dives of long duration which make it a less suitable species for visual tracking. The clicks of sperm whales are distinctive, regular and by their nature are amongst the most likely sounds to provide good bearing and range estimation using the currently-available software.

However, aside from these species, the MMOA believes that the use of PAM equipment for mitigation purposes for other SoC should not be considered to represent a reliable sole method but rather should be used to enhance other mitigation measures.


PAM Detections and Noise Samples

The MMOA recommends that the use of PAM during every mitigation project should be objectively assessed for detection rates, to determine whether acceptable signal-to-noise ratios are achieved for mitigation purposes and to test the sound quality of the PAM system. This may be achieved with PAM recordings made during made during mitigation surveys, or sound sample records for quality assessments (i.e. to gauge water noise and masking and determine how effective the PAM system was likely to be at detecting marine mammals in practice). The effectiveness of PAM for marine mammal mitigation purposes cannot be assessed without understanding the acoustic environment that the systems are being deployed in and the corresponding likelihood of detecting an animal.


Submission of PAM Data to Regulatory Bodies

It is not currently a requirement for recordings of PAM detections to be submitted to a regulatory body. However, the MMOA considers this desirable in order to:

  • Assess the likelihood of acoustic detection for different SoC.

  • Provide a means of checking and validating the species identification of acoustic detections recorded.

  • Provide a means of assessing the mitigation decisions made based on particular acoustic detections (i.e. double-checking of range and bearing information or amplitude).

  • Provide opportunity for acoustic analysis research.

At present, PAM data is neither checked nor verified and its effectiveness in mitigation surveys (in terms of equipment functioning properly, adequate ambient noise levels for making detections, the species and number of detections made, the provision of adequate range and bearing information and the consequent implementation of mitigation decisions) has not been examined.

 

Training for PAM Operators

A PAM Operator needs to be able to assemble (and sometimes repair) PAM equipment, deploy it (from a variety of platforms) safely, maximise the signal-to-noise ratio for the detection of marine mammals (and other SoC), configure PAM software with respect to monitoring screens and hardware settings (including for range and bearing information), identify acoustic signals and track bearing information to determine range (if possible). PAM is therefore a technical skill that needs thorough prior training and much experience at sea to accomplish. Please see the MMOA’s Position Statement 5 - Passive Acoustic Monitoring (PAM) Operator Qualifications - for more information on this subject.

The MMOA emphasises that the skills required to work effectively as a PAM Operator are difficult to acquire through short training courses, and require practical time at sea. Consequently, the MMOA recommends that organisations hiring PAM Operators for mitigation surveys seek to use personnel with proven previous field expertise with marine mammal acoustics. The MMOA does not promote the use of inexperienced PAM personnel as trainees offshore unless those persons are “extra” to the required number of experienced PAM Operators needed to cover 24 hr mitigation (i.e. trainees should be supplementary to, and learning from, the mitigation team, rather than working as part of it). The MMOA recommends that persons seeking acoustic monitoring experience achieve this by joining dedicated research projects (paid or otherwise) and through mentoring opportunities.

 

This is one of ten position statements produced by the MMOA. All of the MMOA Position Statements are available for download in a single document in addition to viewing on this website. To download this document please click here

 

 


[1] Blackwell, S.B., Nations, C.S., McDonald, T.L., Thode, A.M., Mathias, D., Kim, K.H., Greene Jr, C.R., Macrander, A.M. (2015). Effects of airgun sounds on bowhead whale calling rates: evidence for two behavioral thresholds. PLoS ONE 10(6): e0125720. Doi: 10.1371/journal.pone.0125720