Environmental impact assessment guidelines for offshore petroleum exploration seismic surveys

Abstract

The ocean environment is filled with natural sound, but the last century has introduced many anthropogenic activities that have increased the levels of noise. Research on the impact of anthropogenic noise on marine fauna is now extensive. Levels of threat are well defined. Mitigation and monitoring guidelines exist in many parts of the world; especially for offshore petroleum exploration. In many jurisdictions, these guidelines rely on environmental impact assessments (EIAs) consideration by decision-makers, yet few jurisdictions stipulate what such assessments should contain. Sound propagation in the marine environment is complex, yet robust and defensible modelling is rarely conducted. Many impact assessments are inadequately checked. This stands in contrast to the equivalent process for land-based assessments. We argue that defensible EIAs should include modelling of the proposed noise impact in the region and under the conditions of planned activity. We articulate why clear guidelines about the content of EIAs are needed and propose a template for offshore petroleum exploration assessment.

Keywords:

• Anthropogenic marine noise
• offshore petroleum exploration
• environmental impact assessments
• Australian sea lion

Introduction

The ocean environment is filled with natural sound from animals and physical processes. Species living in this environment are adapted to these sounds. Many species rely on sound as a primary sense, using it for hunting, reproduction and navigation (Southall et al. 2000, 2007; Simmonds et al. 2014). Over the past century, many anthropogenic marine activities have increased levels of noise (Hildebrand 2009; André et al. 2011). These modern anthropogenic noises have the potential for physical, physiological and behavioural impacts on marine fauna – mammals, reptiles, fish and invertebrates (Moriyasu et al. 2004; Southall et al. 2007; Payne et al. 2008; Clark et al. 2009; Miller et al. 2009; André et al. 2010; CBD CBD SBSTTA 2012). One noise-producing industry is offshore petroleum exploration.

There are national and regional operational guidelines available to the offshore petroleum exploration industry, each detailing the impacts to avoid and mitigation measures to take during operations. These began with the United Kingdom’s Joint Nature Conservation Committee guidelines to minimise acoustic disturbance of marine mammals by oil and gas industry seismic surveys in 1995. Similar guidelines have been iteratively developed in the United States of America, Brazil, Canada, Australia and New Zealand (Castellote 2007; Weir & Dolman 2007; Compton et al. 2008). At a regional level, the intergovernmental Agreement on the Conservation of Cetaceans in the Black Sea, Mediterranean Sea and Contiguous Atlantic Area (ACCOBAMS) has established comprehensive guidelines for the Mediterranean region. Other regional and international instruments are gradually developing similar guidance.

These guidelines focus on mitigation measures during operations and rely on an assessment of risk having being considered and approved by decisions-makers before the operation starts. This is an important step in the process, yet there are few guidelines about the content of these environmental impact assessments (EIAs). Generalised assumptions about impact are often all that is presented. If an EIA is to be a good decision-aiding tool, it must provide decision-makers with a thorough and detailed understanding of the consequences of their decisions (Tenney et al. 2006).

The propagation of sound in water is complex and requires many variables to be carefully considered before it can be known if the proposal is appropriate or not. Despite this, proposals from the offshore petroleum exploration industry are presented to regulators with generalised, unsubstantiated information and often without having conducted basic consultation with other stakeholders reliant on the same environment.

These hollow submissions perpetuate because the expectation from government has not been carefully prescribed. Regulators are forced to approve or reject projects without robust, defensible and impartial information on which to base their decisions. Regulator decisions are often made based on erroneous information. Such decisions are vulnerable to criticism of bias or tokenism (Court et al. 1996; Tenney et al. 2006; Jay et al. 2007; Devlin & Yap 2008; Prideaux & Prideaux 2012; 2013b, 2013c, 2013d, 2013e, 2013f; Wright et al. 2013).

This paper provides a basic explanation of the complexities of sound propagation in the marine environment and shows why generalised assumptions are inadequate to assess impact. A brief description of the common technology employed by the offshore petroleum exploration industry is provided. The next section will give a broad outline of the range of species susceptible to loud anthropogenic noise pollution and a general summary of the impacts they experience. The final section explores the trends in current EIAs for offshore petroleum exploration and introduces a template for EIA guidelines.

Sound propagation in water is complex

Often, offshore petroleum exploration industry statements are made in EIAs that a sound-producing activity is ‘X’ distance from ‘Y’ species or habitat. In these cases, distance is used as a basic proxy for impact, but is rarely backed with scientifically modelled information. To present a defensible EIA for offshore petroleum exploration proposal, proponents need to have professionally modelled the noise of the proposed activity in the region and under the conditions they plan to operate.

The behaviour of sound in the marine environment is different from sound in air. The extent and way that sound travels (propagation) is affected by many factors, including the frequency of the sound, water depth and density differences within the water column that vary with temperature, salinity and pressure (Wagstaff 1981; Clay & Medwin 1997; Lurton 2010; Etter 2013). Seawater is roughly 800–1500 times denser than air and sound travels around five times faster in this medium (Lurton 2010, p. 16). Consequently, a sound arriving at an animal is subject to propagation conditions that are complex (McCauley et al. 2000; Calambokidis et al. 2002; Hildebrand 2009; Lurton 2010).

While noise modelling is common for land-based anthropogenic noise-producing activities, it is less common for proposals in the marine environment. The lack of rigorous noise modelling in the marine setting needs to be urgently addressed. Modelling of each individual proposal should be professionally and impartially conducted to provide decision-makers with credible and defensible information. It should provide a clear indication of sound dispersal characteristics, informed by local propagation features (Urick 1983; Etter 2013). With this information, species exclusion zones can be identified with descriptions of how noise propagation into these zones will be minimised.

Elasticity

The speed of sound is not a fixed numerical value. Sound wave speed varies widely and depends on the medium, or material, it is transmitted through such as solids, gas or liquids. Each medium has its own elasticity (or resistance to molecular deformity). This elasticity factor affects the sound wave’s movement significantly.

Sound waves move through a medium by transferring kinetic energy from one molecule to the next (Lurton 2010, pp. 14–20). Solid mediums, such as metal, transmit sound waves extremely fast because the solid molecules are tightly packed together, providing only tiny spaces for vibration. Sound waves move rapidly through this high elasticity medium, because the solid molecules act like small springs, aiding the wave’s movement across the medium. The speed of sound through aluminium, for example, is around 6319 ms⁻¹ (Goel 2007; Gottlieb 2007, pp. 22–23; Giordano 2012, p. 414). Gas, like air, naturally has large spaces between each molecule. As a result, sound waves take longer to move through a gas. Each air molecule vibrates at a slower speed after a sound wave passes through it, because there is more space surrounding the molecule. The gas molecule effectively deforms in shape from the passing sound wave, making gas reflect a low elasticity. Sound waves moving through air at a temperature of 20 °C will only travel around 342 ms⁻¹ (Goel 2007; Gottlieb 2007). Liquid molecules, such as seawater, bond together in a tighter formation compared with gas molecules allowing only small vibration movements. Sound waves do not deform the liquid molecules as severely as gas molecules, creating a higher elasticity level. Sound waves moving through water at 22 °C travel at around 1484 ms⁻¹ (Goel 2007; Gottlieb 2007).

Warmer temperatures across a medium also excite molecules. Molecules move faster under higher temperatures, transmitting sound waves more rapidly across the medium. Conversely, decreasing temperatures cause the molecules to vibrate at a slower pace, hindering the sound wave’s movement (Goel 2007; Gottlieb 2007, p. 23; Giordano 2012). The temperature of the seawater at different depths is therefore of importance to modelling.

Spherical spreading, cylindrical spreading and transmission loss

The way sound propagates is also important. Spherical spreading is simply sound leaving a point source in an expanding spherical shape (Urick 1983, p. 100; Lurton 2010, p. 22). As sound waves reach the sea surface and sea floor, they can no longer maintain their spherical shape and they begin to resemble the shape of an expanding cheese wheel. This is called cylindrical spreading (Urick 1983, p. 102). The transmission loss, or the decrease in the sound intensity levels, happens uniformly in all directions during spherical transmission. However, when sound is in a state of cylindrical transmission it cannot propagate uniformly. The sound is effectively contained between the sea surface and the sea floor, while the radius is still expanding uniformly (the sides of the cheese wheel) but the height is now fixed and so the sound intensity level decreases more slowly (Urick 1983, p. 102).

Given the seabed is rarely, if ever, flat and parallel to the sea surface, modelling cylindrical spreading in the marine environment is complex. Seabed characteristics must be known to model this spreading. Modelling must accommodate the water depth below the seismic survey, as well as the rise and fall of the seabed surrounding it (Lurton 2010, p. 13).

Sound Fixing and Ranging channels (SOFAR)

As well as spherical and cylindrical spreading, another variable can impact how far sound will be transmitted. This is usually called a SOFAR or deep sound channel and is a horizontal layer of water in the ocean at which depth, the speed of sound is at its minimum.

The SOFAR channel is created through the interactive effect of temperature and water pressure (and, to a smaller extent, salinity). This occurs because pressure in the ocean increases with depth, but temperature is more variable, generally falling rapidly in the main thermocline from the surface to around a thousand meters deep and then remaining almost unchanged from there to the ocean floor. Near the surface, the rapidly falling temperature causes a decrease in sound speed (or a negative sound speed gradient). With increasing depth, the increasing pressure causes an increase in sound speed (or a positive sound speed gradient). The depth where the sound speed is at a minimum is called the sound channel axis. The speed gradient above and below the sound channel axis acts like a lens, bending sound towards the depth of minimum speeds. The portion of sound that remains within the sound channel encounters no acoustic loss from reflection of the sea surface and sea floor. Because of this low transmission loss, very long distances can be obtained from moderate acoustic power (Urick 1983, p. 159; Lurton 2010, p. 58).

Offshore petroleum exploration

The commonly used surveying method used for offshore petroleum exploration is ‘seismic reflection’. This is simply sound energy discharged from a sound source (air gun array) at the sea surface that penetrates subsurface layers of the seabed and is reflected to the surface where it is detected by acoustic receivers (hydrophones). These surveys are typically conducted using specially equipped vessels that tow one or more cables (streamers) with hydrophones at constant intervals. For the seismic reflection process to work, there needs to be enough energy discharged from the air gun array to travel, sometimes several kilometres, to the sea floor and then to be refracted as it passes from liquid into solid to a prescribed depth. Some of the energy is reflected and begins a return journey being refracted from solid to liquid then to travel to the hydrophone streamers. The analysis of these reflections provides a profile of the underlying rock strata and helps industry to identify hydrocarbon accumulations or anomalies that may correspond to hydrocarbon deposits. The typical discharge of each pulse of an air gun array is around 230 dB (re 1 μPa² @ 1m) every 10–15 s, and surveys typically run more or less continuously over many weeks (Urick 1983; Clay & Medwin 1997; Caldwell & Dragoset 2000; Dragoset 2000; Lurton 2010). These operations are usually called ‘seismic surveys’.

Marine fauna susceptible to anthropogenic noise

Marine animals rely on sound for their vital life functions, such as communication, prey and predator detection, orientation and for sensing their surroundings (Simmonds et al. 2014). Noise affects the behaviour and physiology of animals in various ways, including disruptions in the neuroendocrine, cardiovascular and immune systems (Kight & Swaddle 2011).

Southall et al. (2007) reviewed the expanding literature on marine mammal hearing and their physiological and behavioural responses to anthropogenic noise. They developed predictions of noise exposure levels above which adverse effects, as either injury or behavioural disturbance, on various groups of marine mammals could be expected. While these researchers acknowledged limits in their proposed criteria, because of scarcity of information about some species, the work is valuable for establishing policy guidelines or regulations about anthropogenic noise.

An important recent Convention on Biological Diversity (CBD) Decision (XII/23) has recommended that further research is conducted for the remaining significant knowledge gaps. This includes knowledge about fish, invertebrates, turtles and birds. They also recommended research into the implications of cumulative and synergistic impacts of multiple sources of noise on marine species (CBD 2014).

Southall et al. (2007) highlighted that exposure criteria for single individuals and short-term (not chronic) exposure events are inadequate to describe the cumulative and ecosystem-level effects likely to result from repeated and/or sustained human input of sound into the marine environment and from potential interactions with other stressors. It is therefore critical that modelling of noise propagation is conducted to determine the potential received levels of noise for different species and the duration of exposure.

An important volume of solid research should be considered directly for more detail about the unique characteristics of each of the species groups. The following section provides a summary of this knowledge base.

Fish, crustaceans and cephalopods

Fishermen worldwide complain that seismic surveys produce economic losses by reducing captures of a wide range of commercial species. The impact of anthropogenic noise on commercial fisheries is slowly being quantified. Behavioural responses of fish and cephalopods vary to received levels of seismic noise. These include leaving the area of the noise, through changes in depth distribution, schooling behaviour and startle responses to short-range start-up or high-level sounds. In some cases, behavioural responses from fish were observed up to 5 km distance from the seismic air gun array (McCauley et al. 2000, 2003; Hassel et al. 2004; McCauley & Fewtrell 2008). Short exposures to intense seismic signals are known to increase mortality of fish larvae at short ranges. Sublethal physiological impacts have been observed in crustaceans potentially impacting reproduction and recruitment. Significant developmental delays and abnormalities have been shown in mollusc larvae, including malformations in soft body tissues (Parry & Gason 2006; Payne et al. 2008; de Soto et al. 2013). Noise exposure during critical growth intervals may contribute to stock vulnerability (de Soto et al. 2013).

Pinnipeds

Pinnipeds (seals, sea lions and walrus) live part of their lives in both air and in water. Their hearing is adapted to both mediums and they are likely to be susceptible to the harmful effects of loud noise in each. Behavioural responses to anthropogenic sound have been recorded including pinnipeds removing themselves from feeding activities. Disturbances in marine and terrestrial environments can cause pinnipeds to abandon colonies, which could have serious implications, especially for species that are already endangered. In most respects, noise-induced threshold shifts in pinnipeds follow trends similar to those observed in other mammals (Southall et al. 2007). Pinnipeds, like many land-based mammals, have vibrissae (whiskers), which are well supplied with nerves, blood vessels and muscles and may function to detect the subtle movements of fish and other aquatic organisms. Vibrissae have been shown (for example, in harbour seals, Phoca vitulina) to be sensitive to low-frequency waterborne vibrations (Bohne et al. 1985; Mathews 1994; Southall et al. 2000; Harris et al. 2001; Kastak et al. 2005).

Sirenians

Similarly, sirenians (dugong and manatee) may be displaced from key feeding habitats by exposure to noise. While most research has focused on boating traffic, their behavioural response to the noise of passing vessels supports that these animals are sensitive to noise and should be considered carefully (Hodgson & Marsh 2007).

Cetaceans

Cetaceans (whales, dolphins and porpoises) are perhaps the most studied group of marine species when considering the impact of anthropogenic noise. Different taxonomic groups of cetaceans adopt different strategies for responding to acoustic disturbance from seismic noise. Baleen whales are susceptible to temporary threshold shift at a kilometre or more from seismic surveys (Gordon et al. 2003; Nowacek et al. 2007; Weilgart 2007; Di Iorio & Clark 2009; Gedamke et al. 2011; Gray & Van Waerebeek 2011). Toothed cetaceans have also shown significant avoidance behaviour at a range of distances (Madsen et al. 2002; Stone & Tasker 2006; Miller et al. 2009; Gray & Van Waerebeek 2011). Researchers are concerned that reducing an individual’s ability to detect socially relevant signals could affect biologically important processes and they caution that short-term proxies, such as avoidance behaviour, are not sufficiently robust to assess the extent and biological significance of long-term individual and population-level impacts.

Sea turtles

Studies of the hearing capabilities of sea turtles show that they hear low-frequency sounds within the range of 100–1000 Hz with greatest sensitivity at 200–400 Hz for adult sea turtles, and 600 and 700 Hz for juveniles. Although sea turtles are poorly studied compared with cetacean and fish species, studies have demonstrated behavioural responses to received levels of seismic noise (O’Hara & Wilcox 1990; Moein Bartol & Musick 2003; Southwood et al. 2008).

The importance of considering stress

There is also need to consider the impact prolonged noise exposure may have on marine fauna beyond the direct physiological and behavioural impacts (Rolland et al. 2012). Chronic levels of stress can result in various pathological dysfunctions with possible damage to long-term health. This is especially relevant for resident species dependent on certain habitats, such as beluga, seals or sea lions.

Failures of current EIAs

The following sections build on the information we have provided about the complexities of sound propagation in the marine environment and overview of the range of species and types of impact that might occur. We comment about the depth of information provided in current EIAs and finally propose guidelines for EIAs.

Many jurisdictions have developed national and regional operational guidelines about mitigating anthropogenic noise on marine fauna and in particular noise produced by offshore petroleum exploration. These began with the United Kingdom’s Joint Nature Conservation Committee guidelines with similar guidelines being iteratively developed in the United States of America, Brazil, Canada, Australia and New Zealand (Castellote 2007; Weir & Dolman 2007; Compton et al. 2008).

Several intergovernmental bodies have also elaborated principles of what EIAs should present. Collectively, these principles have been adopted by 196 governments who, through the process of their adoption, have individually committed to reflecting these decisions in their domestic law. The ‘weight’ of these decisions taken by governments at an international level is considerable.

The most notable of these is the ‘Agreement on the Conservation of Cetaceans in the Black Sea, Mediterranean Sea and Contiguous Atlantic Area’ (ACCOBAMS). ACCOBAMS ‘Resolution 4.17: Guidelines to address the impact of anthropogenic noise on cetaceans in the ACCOBAMS area’ articulate specifics for the Mediterranean region and

[encourage] Parties: – to address fully the issue of anthropogenic noise in the marine environment, including cumulative effects, in the light of the best scientific information available and taking into consideration the applicable legislation of the Parties, particularly as regards the need for thorough environmental impact assessments being undertaken before granting approval to proposed noise-producing activities. (ACCOBAMS 2010)

The ACCOBAMS Noise Guidelines further prescribe specific considerations about seismic surveys, including the need for accurate modelling.

ACCOBAMS Resolution 5.15 calls on the Parties to:

•	ensure that EIAs take full account of the effects of activities on cetaceans;
•	implement the recommended use of Best Available Techniques and Best Environmental Practice in their efforts to reduce or mitigate marine noise pollution;
•	integrate the issue of anthropogenic noise into the management plans of marine protected areas.

Resolution 5.15 also underlines that EIAs should include specific details that mirror those articulated in the ACCOBAMS Noise Guidelines (ACCOBAMS 2013).

The Convention on Migratory Species (CMS) ‘Resolution 10.24: Further Steps to Abate Underwater Noise Pollution for the Protection of Cetaceans and Other Migratory Species’ also strongly urges CMS Parties to prevent adverse effects on marine species by restricting the emission of underwater noise to the lowest necessary level and urges CMS Parties to ensure that EIAs take full account of the effects of activities on marine fauna (CMS 2011).

Most recently, the CBD ‘Decision XII/23: Marine and coastal biodiversity: Impacts on marine and coastal biodiversity of anthropogenic underwater noise’ has specifically encouraged CBD Parties to take suitable measures to avoid, lessen and mitigate adverse impacts of anthropogenic underwater noise on marine and coastal biodiversity, including:

•	combining acoustic mapping with habitat mapping of sound-sensitive species when developing spatial risk assessments to identify areas where those species may be exposed to noise impact;
•	using spatio-temporal management, including detailed knowledge of species or population distribution patterns, to mitigate and manage noise activities and avoiding producing noise in the area at critical times;
•	conducting EIAs for activities that may have significant adverse impacts on noise-sensitive species. (CBD 2014)

Assessment of likely impacts is also an emerging legal requirement in the European Union. The European Parliament and Council ‘Environmental Impact Assessment Directive 2014/52/EU’ requires that EIAs are carried out before development consent is given to activities (2014/52/EU Art 2.1) to identify impacts to biodiversity with particular attention to species and habitat protected under Directive 92/43/EEC and Directive 2009/147/EC (2014/52/EU Art 3.1). The Directive introduction states that:

[w]ith a view to ensuring a high level of protection of the marine environment, especially species and habitats, environmental impact assessment and screening procedures for projects in the marine environment should take into account the characteristics of those projects with particular regard to the technologies used (for example seismic surveys using active sonars). (2014/52/EU)

Conducting EIAs is now a well-established governance and environmental management principle, institutionalised in over 100 countries (Court et al. 1996; Glasson et al. 2013). These four intergovernmental bodies provide significant clarity about the expectations to conduct EIAs and effectively manage impacts associated with offshore petroleum exploration activities, among other underwater noise-producing activities.

It is broadly accepted the basic intent of EIAs is to anticipate the significant environmental impacts of development proposals before any commitment to a particular course of action. However, often, the detail required within EIAs is poorly defined. Many legislative provisions for EIAs have been introduced without consideration of the institutional requirements: organisational structure, staffing and capacity development (Cashmore et al. 2004; Jay et al. 2007; Devlin & Yap 2008). Often the scientific basis and methods need sophisticated understanding.

Given this, it is not surprising the efficacy of many EIAs is being criticised (Slootweg & Kolhoff 2003; Cashmore et al. 2004, 2010; Devlin & Yap 2008). Indeed, the criticism of the ‘low bar’ requirements for EIAs in many jurisdictions might be, in part, a result of decision-makers themselves having limited understanding of the EIA purposes and potential (Cashmore et al. 2004; Jay et al. 2007) as well as the general poor quality of EIA information (Morgan 2012; Morrison-Saunders & Retief 2012).

This was revealed to be the case for offshore petroleum exploration EIAs by Wright et al. (2013). They found that many assessments were insufficiently researched, drawing heavily from previous EIAs. In a significant number of cases, approvals were given without careful consideration of the detail presented in the EIAs. Instances of duplicated information or missing species were not uncommon. Topics were dealt with by dismissal, often ignoring recent scientific literature, perpetuating misconceptions and containing analytical flaws. Discussions about wildlife often focused on lethal impact, with little or no consideration of sublethal impacts.

Our documentary examination of five EIAs, that spanned less than one year and took place within one regulatory jurisdiction, revealed similar trends to those highlighted by Wright et al. (2013). All were proposals for petroleum exploration in Australia’s Exclusive Economic Zone under the same regulatory process and all were given approval by the National Offshore Petroleum Safety and Environmental Management Authority’s (NOPSEMA) (Prideaux & Prideaux 2013b, 2013c, 2013d, 2013e, 2013f).

These five are by no means isolated cases. Since inception, 291 EIAs (so-called Environmental Plans) have been received by NOPSEMA. Most of these have been accepted by the authority. The authors have engaged in a correspondence trail with the authority to highlight significant errors, inaccuracies, misconceptions and analytical flaws in a number of the 291 submissions. Written responses from the authority confirm that their focus is on ensuring the industry commits to self-identified benchmarks. They assert the authority does not assess the efficacy of claims or assurances contained in the EIAs (correspondence on file with the authors).

An example of assessment relating to Australian sea lions

An example of assessments relating to Australian sea lions provides a useful illustration. The Australian sea lion (Neophoca cinerea) is Australia’s only endemic and least numerous seal species. The species is listed as Vulnerable under the national environment legislation and has an IUCN Red List Criteria of Endangered (A2bd + 3d). The Australian Government’s own ‘South-west Marine Bioregional Plan and Species Group Report Card – Pinnipeds’ identifies noise as a threat of concern (Department of Sustainability Environment, Water, Population & Communities 2012a, 2012b).

Under the ‘South-west Marine Bioregional Plan’ any individual Australian sea lion breeding colony is regarded as an important population. The government’s Plan directs that all attempts should be made to avoid biologically important areas for the Australian sea lion, particularly water surrounding breeding colonies and foraging areas used by female sea lions, for any applications for offshore development. The Plan specifically states that ‘actions with a real chance or possibility of increasing the ambient noise levels within female Neophoca cinerea foraging areas to a level that might result in site avoidance or other physiological or behavioural responses’ have a high risk of a significant impact on this species (Department of Sustainability Environment, Water, Population & Communities 2012a, 2012b)

Clearly, the Australian Government has decided the status the sea lion demands a precautionary approach to ensure that human activities, including anthropogenic noise do not further jeopardise the species. Despite this, in a two-year period, NOPSEMA has accepted four EIAs, in the form of Environmental Plans. Each has failed to consider the impact of noise generated by offshore petroleum exploration on Australian sea lion populations and each has been given the proponent approval to proceed. These will or have already produced sound intensity levels around 230 dB (re water) that will transmit many hundreds of kilometres, including into and through areas of sea lion foraging habitat.

Given that offshore petroleum exploration activities typically span six to eight weeks, it is likely that sea lion foraging behaviour will be or has been significantly impacted or abandoned altogether. There could be reduced food availability, animals might show signs of reduced condition and may have difficulty feeding their pups. Colonies may or have been abandoned temporarily or permanently, which could have serious implications for this already endangered species. Review of the published EIAs (available on www.nopsema.gov.au) reveals that no modelling of noise propagation has been considered and no assessment of impact has been carried out. There is no description of the well-known Australian sea lion colonies. There is no discussion of the foraging habitats of the species, nor is their recognition of the precaution flagged in the ‘South-west Marine Bioregional Plan’ and ‘Species Group Report Card – Pinnipeds’. NOPSEMA has accepted and approved the EIAs. Even though the information was inconclusive or incomplete, NOPSEMA has not required any monitoring be established.

Anecdotal evidence for other regions shows similar trends in other jurisdictions including Europe, West Africa and East Africa (on file with the authors). There is a failure of current EIAs for offshore petroleum exploration.

It is important that government decision-makers can rely on sufficient technical, detailed and impartial information being presented to them to ensure credible and defensible decisions are made about offshore petroleum exploration. The following section proposes template guidelines on the detail of information that should be sought to support robust and defensible decisions.

Environmental impact assessment for offshore petroleum exploration seismic surveys

This section is built on the foundations of three important previous works. These are an important study on impact mitigation of offshore petroleum exploration in the Sakhalin region of the North Pacific Ocean (Nowacek et al. 2013); a framework for assessment of noise impact in the Arctic (Moore et al.2012); and a workshop on the requirements for marine noise EIAs during the 2014 European Cetacean Society meeting (Evans 2015). This collective work has elaborated that assessments should:

•	collect baseline biological and environmental information to describe the area being impacted;
•	fully characterise operations, including describing the sound source in some detail, the local sound propagation features and potential cumulative effects from other sound sources as well as other human activities that may not generate noise but can add to the pressures on the local animal populations; and
•	describe how impacts will be monitored before, during and after the operation.

To provide regulators with greater technical detail about how to seek this level information, we have developed the proposed template through two important cross-disciplinary peer discussion forums:

(1)	The Joint CMS/ASCOBANS/ACCOBAMS Noise Working Group where the template was formally developed as a contribution to the ‘CBD Expert Workshop on Underwater Noise and its Impacts on Marine and Coastal Biodiversity’.
(2)	The 18th CMS Scientific Council Meeting, where the template was presented and comments and input sought.

The template has also sought the input more broadly from regulators and industry. The proposal that follows is a reflection of this iterative discussion with experts through these processes (Prideaux & Prideaux 2013a).

Environmental impact assessment guidelines for offshore petroleum exploration proposals

In addition to jurisdictional specific requirements for impact mitigation during operations, such as observers or passive acoustic monitoring, EIAs for offshore petroleum exploration should be developed early in the proposal’s development process and should transparently include:

(1)	Description of area (a) Detailed description of the spatial extent and nature of the survey – including seabed bathymetry and composition, description of known stratification characteristics and broad ecosystem descriptions – as well as the spatial area that will experience anthropogenic noise, generated by the proposed survey, above natural ambient sound levels (b) Details of baseline data that have been gathered before developing the EIA, including consultation with regulating bodies and stakeholders (c) Identification of previous surveys, their seasons and duration in the same or adjoining areas, and a review of survey finding and implications (d) Identification of previous test wells in the same or adjoining areas including comment about any wells that may breach
(2)	Description of the equipment to be used (a) Explanation of all survey technologies available and why the proposed technology is chosen (b) Detailed description of the survey technology to be used (c) Name and description of the survey vessel to be used (d) If an air gun array is proposed: (i) Number of arrays (ii) Number of air guns within each array (iii) Air gun charge pressure to be used (PSI) (iv) Volume of each air gun in cubic inches (v) Official calibration figures supplied by the survey vessel to be charted (vi) Modelled sound intensity level one metre from source derived from the official calibration figures (vii) Depth the air guns to be set (viii) Number of streamers (ix) Length of streamers (x) Distant set apart (xi) Depth the hydrophones are set
(3)	Details of consultation and independent review (a) Identification of stakeholders who have been consulted (b) Identification of independent experts – especially species experts – that have been consulted including their affiliation and their qualifications (c) Explanation of information provided to stakeholders and experts, any opportunities given for appropriate engagement and the timeframe given for them to provide feedback (d) Description of the comments, queries, requests and concerns received from each of the stakeholders and experts (e) Explanation of what amendments and changes have been made to the proposed survey to the comments, queries, requests and concerns (f) Explanation of which comments, queries, requests and concerns have not been accommodated and why
(4)	Comprehensive description of activity (a) Comprehensive description of the total area to be explored and the entire exploration plan (2D, 3D and test wells) and for each activity: (i) Specifics of the activity including anticipated nautical miles to be covered, track-lines, speed of vessels, duration of track-lines, start up and shutdown procedures, distance and procedures for vessel turns including any planned air gun power setting changes (ii) Computer modelling of sound dispersal in the same season/weather conditions as the proposed survey, local propagation features (spherical and cylindrical spreading, depth and type of sea bottom, local propagation paths related to thermal stratification) and out to a radius where the generated noise levels are close to natural ambient sound levels (iii) Identification of any SOFAR or natural channels characteristics (iv) Sound intensity level and frequencies (Hz) from a point source, as well as the duration of each pulse (milliseconds), interval between pulses (seconds) and expected duration of pulses (12/24 h days) for the survey (a) Identification and mapping of proposed species exclusion zones and description of how noise propagation into these zones will be minimised, taking into consideration the local propagation features (spherical and cylindrical spreading, depth and type of sea bottom, local propagation paths related to thermal stratification) (b) Identification of other impacting activities in the region during the planned survey, accompanied by the analysis and review of potential cumulative impacts
(5)	Species likely to be encountered or impacted (a) Description of all listed/protected species likely to be present and that will experience sound transmission generated by the proposed survey above natural ambient sound levels, the total time they will experience these sound levels and proposed measures being taken for each to minimise impact (b) Description of all fisheries likely to be present or to rely on prey that might be present and that will experience sound transmission generated by the proposed survey above natural ambient sound levels and proposed measures being taken for each to minimise impact
(6)	Details of likely impact for each listed/protected species, including: (a) Identification of safe/harmful exposure levels for various species that is precautionary enough to handle large levels of uncertainty and avoids erroneous conclusions (b) Type of impact predicted (direct, behavioural and the duration) as well as direct and indirect impacts to prey species (c) Soft start and shutdown protocols (d) Plans for 24 h visual detection, especially under conditions of poor visibility (including high winds, night conditions, sea spray or fog) (e) Plans for establishing exclusion zones to protect specific species. These should be established on a scientific and precautionary basis rather than as arbitrary and/or static designations
(7)	Details of independent and transparent monitoring of all at-sea activities and observer coverage (a) Details of transparent processes for regular real-time public reporting of activity progress and all impacts encountered (b) Details of scientific monitoring programmes, conducted during and after the seismic survey, to assess impact
(8)	Reporting plans (a) Details of plans for post operation reporting including verification of the effectiveness of mitigation

The information requested in this template is well within the current technical competencies of the petroleum and scientific community. The detail within the EIA should be robust enough for independent review and not placed under a seal of commercial in-confidence. This process should prove sufficiently robust to ensure that regulators and decision-makers have access to an appropriate level of information before making approval decisions. It will allow them to seek expert technical critiques of the information if they do not have sufficient expertise within their department.

Conclusion

The ocean environment is filled with natural sound produced by animals and physical processes but modern anthropogenic activities have increased the levels of noise. Offshore petroleum exploration is a significant contributor to this noise. Sound propagation in the marine environment is complex and it is especially important that government decision-makers can rely on sufficient technical, detailed and impartial information being presented to them to ensure credible and defensible decisions are made about the impact of this industry and individual proposals.

While noise modelling is common for land-based anthropogenic noise-producing activities, we have shown that modelling and indeed robust EIAs for offshore petroleum exploration are failing this base need. EIAs should provide a clear indication of the sound propagation features across the full area the noise will impact. Proponents should be required to model the noise propagation of the proposed activity in the region and under the conditions they plan to operate. The documentation should demonstrate a clear understanding of the species present, necessary exclusion zones and descriptions of how noise propagation into these zones will be minimised.

This paper has proposed ‘Environmental Impact Assessment Guidelines for Offshore Petroleum Exploration Proposals’. These template guidelines have been developed with the benefit of peer input and review through two official processes; to provide guidance about the specifics that should form the basis of appropriate assessments. In time, global noise standards may supersede such a need, but that time is still in the distant future and will need complex and controversial international oversight to be in place. For now, given the strong commitment of governments around the world to reducing anthropogenic marine noise, this information, if transparently supplied, would provide regulators and decision-makers with robust, defensible and impartial information on which to base their decisions.