The Marine Mammal Observer Association (MMOA) has developed some generic advice and guiding principles to follow when drafting a Mitigation Plan (MP) for Species of Concern (SoC) during surveys using anthropogenic sound sources in the marine environment. This advice is based on the MMOA recognising that nominating single generic values for mitigation zones, delay durations and soft starts is inappropriate. Instead, the MMOA supports the establishment of survey and site-specific mitigation measures based on the water depths, oceanographic conditions, sound source and SoC likely to be encountered during a particular survey. Please note that in some geographic areas the regulatory bodies have developed a set of formalised mitigation procedures and where that is the case those procedures must be strictly adhered to. Before developing the appropriate MP, a full Environmental Impact Assessment (EIA) should be carried out.
The Principles of Establishing Mitigation Measures for SoC
This planning principle will be achieved by identifying seasons/areas of particular importance for SoC (through the EIA and, where necessary, baseline survey work), and scheduling survey activities around those times/areas. The MMOA believes that avoidance of critical habitat is the single most important mitigation measure and should be considered early in the survey planning process.
The MMOA believes that the primary principle adopted during real-time mitigation activities during a survey (hereafter the ‘Mitigation Principle’) should be:
“to minimise the likelihood that all Species of Concern (SoC) are exposed to source levels that have the potential to cause injury or death”
This mitigation principle should then be consistently applied throughout the procedures adopted. The remainder of this document is concerned with the implementation of this mitigation principle. More detailed advice on mitigation measures are outlined below:
Mitigation Zones (MZs)
An acceptable (‘safe’) level of source activity should be determined i.e., a sound level which scientific studies indicate is unlikely to be injurious to the identified SoC. The MZ will then represent the radius around the sound source within which sound levels are higher than this predetermined ‘safe’ level. More than one MZ may be appropriate when different species groups are being mitigated for, e.g., sea turtles, mysticetes, odontocetes, pinnipeds and sirenians. Alternatively, simplification for monitoring purposes can be achieved by selecting a MZ for all SoC in an area that is based on the most precautionary zone from impact modelling. Cumulative exposure metrics (i.e., SEL) should be considered when SoC may be exposed over long durations or to multiple pulses of impulsive sources.
The MZ should be established by acoustic modelling to determine the zone of influence for a particular source and activity for each functional hearing group. The MZ should be determined for each source by applying modelled transmission loss to the corresponding frequency band source levels. Transmission loss modelling should consider in situ measurements of local environmental conditions (e.g. water depth, seabed type, water temperature profiles) which affect the resulting sound propagation.
Where surveys are of long duration (> 1 month), MZ models should be updated regularly according to changes in localised oceanographic conditions within the site. For example, changes in water temperature and salinity are often measured regularly using T-S dips during seismic surveys and can be incorporated into updated models to calculate site-specific MZs over time.
If sound source modelling at the site of the actual project is not possible, the radius of the MZ should be determined by assuming that local conditions are optimal for sound propagation i.e. the worst case scenario which considers the hardest seabed sediment or substrate that is likely to be found in the area, the sound speed profile that is likely to give rise to the longest range propagation and the loudest noise likely to be generated by the sound source during the survey. In the case of seismic surveys, the firing of all guns simultaneously on the array but not including the spare guns should be considered i.e. the firing of spare guns should not be carried out unless other guns are dropped out first so that production volume is not exceeded).
Monitoring of SoC Within the Mitigation Zone Prior to and During Operations
Monitoring period prior to source activity – A minimum monitoring period should be adopted prior to the initiation of a sound source to detect any SoC in the area. The duration of the monitoring period can be determined by what is known about the surfacing rates and behaviour of SoC in the area. Various mitigation procedures worldwide recommend no less than a 30 minute watch period prior to source activation and the MMOA recognises this as an acceptable minimum period.
Monitoring during source activity – The MMOA recommends that monitoring for SoC continues throughout periods of source activity, in order to collect information on potential impacts. When shut-down procedures are part of the adopted mitigation protocol (as recommended by the MMOA for sound levels considered likely to cause injury/Permanent Threshold Shifts in hearing (PTS)) then monitoring will be required throughout periods of source activity in order to protect SoC.
Monitoring methods – Monitoring of the MZ may comprise visual monitoring during daylight hours, passive acoustic monitoring (PAM) during all hours and night vision/ infrared techniques during hours of darkness. The MMOA fully supports the development and use of novel detection technologies (remote PAM and thermal imaging to improve detection and mitigation). All monitoring methods should be assessed for their efficiency and likelihood of detecting the SoC within a particular area. Marine Mammal Observers (MMOs) and Passive Acoustic Monitoring (PAM) Operators should be positioned as near to the source as possible for effective monitoring of the MZ. The MMOA considers that the MMOs/PAM Operators should always be located on the source platform , unless an alternative platform has the potential to have a better view of the sound source. This would be essential if the distance of the sound source from the platform compromises the ability of the MMO/PAM Operator to effectively monitor the MZ for the relevant SoC. The relative detectability of each species group should be carefully considered, since some (e.g. sea turtles) may only be reliably detected within a few hundred metres of a visual observer. Consequently, a MMO located on a source platform from which the sound source is deployed several hundred metres astern (i.e., most 3D/4D seismic vessels) is unlikely to be able to reliably visually detect such fauna within the latter portion of a MZ. Where multiple sources are deployed from multiple survey platforms, a separate MZ and monitoring team are required for each source.
Use of Suitably Qualified Marine Mammal Observers (MMOs) and Passive Acoustic Monitoring (PAM) Operators
Monitoring should only be conducted by independent, experienced and dedicated MMOs and PAM Operators. The MMOA considers that, in this context, experience should reflect both field expertise with marine mammals (in order to develop detection and identification skills) and suitable accredited or regulator approved training courses (in order to develop proficiency with implementing mitigation procedures and completing data forms). Please see the MMOA Position Statements 4 & 5 – Marine Mammal Observer (MMO) Qualifications and Passive Acoustic Monitoring (PAM) Operator Qualifications - for further guidance.
There needs to be some quality assessment protocol to ensure that suitably qualified MMOs and PAM Operators are being employed before the project starts and the qualification of each should be outlined in the final MMO/PAM report. The regulatory body in place should ensure this and in the absence of a regulator, the Client should adopt this assessment protocol as “best practice”.
An assessment should be made of how many MMOs and PAM Operators are required to provide the necessary mitigation monitoring, allowing sufficient rest periods to avoid fatigue. As a rule, a MMO/PAM Operator should not be expected to work more than the standard 12 hr work per day of other seafarers – where observer duties include observations, data recording and reporting. Additionally, since visual observations require a high level of concentration, the MMOA recommends that a MMO team should be organised such that no MMO works for more than 2 hr continuously followed by a break of at least 1 hour.
Appropriate Real-time Mitigation Measures Should a SoC be Detected Inside the MZ Prior to and During Source Operations
Species of concern inside the mitigation Zone (MZ) - If a SoC is detected (visually or acoustically) within the MZ prior to source activation, commencement of the source should be delayed until such time that the animals depart the MZ. If it is unclear whether an animal has departed the MZ, an additional “buffer” time should be added to ensure a high likelihood that the animal is no longer in proximity to the sound source. Due to the limitations of passive acoustic monitoring (see Position Statement 3 – Passive Acoustic Monitoring) it is very unlikely that a PAM operator, during the hours of darkness, will be able to determine when SoC have left the MZ, unless experienced, (with the only possible exception being the ability, at times, to track vocalising sperm whales) and adding buffer times will be a standard procedure throughout night-time operations. Vessel speed, and the behaviour and swimming speeds of specific SoCs, should be considered to determine the likely length of time for a safe distance to be established. Different buffer times may be required according to the species groups being mitigated for, e.g. sea turtles, mysticetes, odontocetes, pinnipeds, or sirenians.
If a SoC is detected (visually or acoustically) within the MZ while a sound source is active, the source should be shut down until such time that the animals depart the MZ (see above for circumstances where it is unclear whether an animal has departed the MZ). The source should resume using a soft start procedure (see below).
Some regulators worldwide do not adopt shut down procedures, based on an unproven ‘common sense’ assumption that animals opting to approach a sound source voluntarily are indicative of a lack of discomfort and consequently do not require mitigation. This principle has also been adopted as ‘best practice’ in the mitigation plans for surveys in some geographic areas where no formal regulator guidance is in place. However, there is no scientific basis for the assumption that animals entering a mitigation zone (which has previously been defined as an injurious zone) ‘voluntarily’ while a source is active, are somehow less susceptible to impacts. This may particularly be the case when young, sensitive animals are accompanying adults and are involuntarily exposed to high sound levels. This would also apply to feeding animals where the drive for food means that the animals show greater levels of tolerance e.g. bowhead whales. Finally, some SoC, for example sea turtles basking at the surface for metabolic purposes, may not have sufficient mobility to move away appropriately from an approaching sound source. The MMOA believes that mitigation plans should aim to “minimise the likelihood that all species of concern are exposed to source levels that have the potential to cause injury” and therefore supports shutting down a sound source whenever a SoC enters the MZ for whatever reason.
The Incorporation of Additional Mitigation Measures to Minimise the Likelihood that Undetected SoCs are Exposed to Injurious Levels of Source Activity (e.g. due to night, poor weather or the limitations of monitoring techniques).
Adopt a ramp-up or soft start procedure  – The aim of the soft start is to initiate the source at the lowest possible sound level and gradually increase power over a pre-defined time period to reach the operational sound level. This procedure is intended to allow animals to depart the immediate vicinity of a sound source prior to being exposed to potentially injurious levels of sound. The soft start must commence at a sound level which does not have the potential to cause injury. If the source cannot be started at a sufficiently low (i.e. ‘safe’) power level, then an alternative method, such as deploying specially designed acoustic alarms, should be adopted. Acoustic alarms or pingers should be designed to cover the frequency ranges of all SoC and should also be activated gradually.
The duration of a soft start should be sufficient to allow an animal to move to a safe distance before full power is achieved. The behaviour and swimming speeds of the SoC should be considered to determine the length of time that it takes for an animal to move to a safe distance. Consideration should be given to situations when full soft starts might not be necessary – for example when testing low-power single airguns at source levels below those that are considered injurious.
In the case of seismic airgun arrays, the airgun soft start procedure should be automated wherever possible to avoid operator inconsistencies in start-up techniques. Seismic contractors should have, and follow, detailed protocols for the order in which each gun of specified volume/sound level is added-in over time and a graph showing cumulative volume/sound level of the array over time. This should expressly include procedures for the inclusion of ‘spare guns’ in the soft start (without exceeding normal operational volume during production) and procedures for soft starts during gun tests.
For other operations utilising sound sources, such as pile driving, military sonars and explosives, it is important to have pre-determined and detailed soft start procedures in place for all equipment that produces high-amplitude sound with the potential to cause injury to SoC.
Breaks in Source activity – Technical issues with equipment sometimes result in a “break in source activity” during surveys, where the sound source is temporarily shut down for maintenance. Various regional guidelines stipulate permitted short lengths of time within which a source can resume at full power following such a break in activity provided no SoC have been detected in the MZ during the break. If the source does not resume within that period of time, a soft start (with monitoring beforehand) must be carried out prior to resuming the survey. Determining an appropriate length of time for a permitted break in source activity (i.e. where a soft start is not required on source resumption) should take account of:
- Whether the MMOs and PAM Operators can be certain that the MZ was free of SoC (i.e. whether they were on watch, whether it was night or day, the detectability and mobility of the SoC).
- The speed of the platform and how far it has moved since shutting down.
The MMOA believes that, in most cases, a precautionary approach to mitigation should not permit breaks in source of activity of more than 5 minutes without requiring a soft start. This is particularly the case for SoC of low detectability such as sea turtles. The MMOA does not support resuming at full volume following a break in source activity if no visual watch for SoC was being maintained throughout the break by a MMO (i.e. at night or in poor visibility). This procedure should only be utilised during occasional, unexpected incidences of technical issues with source activity; it should not be a routine method of avoiding a subsequent soft start.
Noise records – The MMOA is aware that noise records of (usually) 1 or 2 minutes duration are routinely carried out during seismic surveys, prior to the start of each line. This involves shutting down the source and then resuming at full power. The routine occurrence of noise records should be acknowledged and specifically included within Mitigation Plans. The mitigation team should maintain a watch for SoC throughout the duration of noise records in case animals are observed within the MZ during the break in firing and mitigating actions are required. The duration of noise records should be kept to the minimum possible, and these should not be carried out and included within the duration of the soft start.
Use of a mitigation source - The practice of keeping sound sources active at reduced power in-between operations (for example during seismic survey line changes) is sometimes adopted by regulators or included in mitigation plans where no formal regulations are in place. This is called the “minimum source” or “mitigation source”, and is intended as a ‘common sense’ approach to warn animals of the location of a sound source. There is no currently available scientific evidence that “mitigation sources” are effective at deterring animals from the vicinity of an active source (indeed, there is equally the potential for attraction of animals towards low power sound sources). Consequently, the MMOA recommends that careful consideration should be given before adopting this unproven technique, particularly given the significant amount of additional sound that will be emitted into the marine environment as a result.
Minimising Source Use and Levels
Best practice should include both minimising overall use of the sound source throughout the survey and using the lowest practicable sound levels. The sound source should be switched off whenever the survey is not in production mode. Testing of sound sources should be kept to an absolute minimum. Start-up of the sound source (including soft start) should be run as close to the start of operations as is possible. The use of a mitigation source in between operations is discouraged until scientific evidence is available to support the procedure. Barkaszi and Kelly (2019) tested the effectiveness of mitigation methods and found mixed results between odontocetes and mysticetes, there is still research to be done.
It is important to clarify the communication links (and chain of command) between the mitigation team and the operators of the sound source, in order to ensure quick, clear and well-implemented mitigation actions when animals are observed.
The mitigation team should have direct contact with personnel operating source equipment so that delays and shutdowns are executed immediately. The MMOA recommends that the use of handheld radios (as opposed to intercom systems) is vital (especially for MMOs, who arelikely to be outside on deck during pre-shoot watches) to ensure that mitigation guidance is swiftly conveyed to the operations crew whilst allowing the observer to continue to monitor the animals. A sufficient number of handheld radios should therefore be made available to the MMOs and PAM Operators at all times.
Recording and Reporting Compliance to Appropriate Authorities
Prior to any project it should be determined how compliance with mitigation measures is to be reported to the appropriate authorities. This should include the provision to maintain the independence of the MMO/PAM reports from Client, Operators and MMO/PAM Provider companies and how this review and assessment of compliance and data quality is to be conducted. There should be a clear system for how non-compliance is dealt with. This could be achieved by allowing independent dialogue between MMOs and PAM operators and any regulatory body involved. Reports should also be available for public viewing.
What data are to be collected and how this is to be recorded should be determined from the outset. The data recording protocol will then need to be passed onto any appropriate authority for assessment.
The MMOA recommends that data collection during mitigation surveys should be standardised according to established protocols. For example, a set of standardised electronic MMO and PAM data forms have been developed by the E&P Sound and Marine Life Joint Industry Programme (JIP) (www.soundandmarinelife.org), a body affiliated with the International Association of Oil and Gas Producers (IOGP). The data forms have been frequently revised following extensive feedback from MMOs and PAM Operators working in the field, and are recommended for a global standard. These forms have been adopted by the Joint Nature Conservation Committee in the United Kingdom (and by other bodies in some other regions) and can be found online at: https://hub.jncc.gov.uk/assets/e2a46de5-43d4-43f0-b296-c62134397ce4 .
 Platform is defined throughout this document as any mobile or static base on which MMOs are located in order to carry out marine mammal mitigation duties. This potentially includes vessels, rigs, shore and airplanes, but most commonly refers to vessels.
 The terms ‘ramp-up’ and ‘soft start’ refer to the same operational procedure where a sound source is initiated at its lowest sound level and gradually increases in sound level over time to reach the desired operational level. The terms are used by different regulatory bodies worldwide and are inter-changeable. We use ‘soft start’ in the remainder of this document.