Marine seismic acquisition at Scott Reef: Minimizing environmental impacts in a sensitive location

MARK TAYLOR, JEREMY FITZPATRICK And RALPH WEISS, Woodside Energy Ltd.

Torosa gas field forms part of the proposed Browse LNG Development1, a project to commercialize three gas and condensate fields located on the outer continental shelf of northwestern Australia, approximately 400 km north of Broome. Two of these fields (Brecknock and Calliance fields) lie in deep water and have been fully imaged by modern 3D, towed streamer, seismic surveys. However, Torosa field partly underlies Scott Reef, which consists of two coral atolls separated by a deep channel.

The only permanently emergent land at Scott Reef is a small sand cay (Fig. 1), although the reef crests of both atolls are exposed at low tide. Outside the reefs, the seafloor drops away rapidly, with water depths of about 350 m to the east, increasing to more than 1,000 m to the west. South Scott Reef lagoon is open to the north, with water depths increasing to about 50 m before deepening abruptly into the channel between the two reefs. North Scott Reef lagoon is shallower—generally less than 25 m—and is connected to the ocean by two narrow channels. Semi-diurnal tides with a range of up to 4.6 m produce strong tidal currents in and near these channels. Small, steep-sided coral heads, or ‘bommies’, are common throughout the lagoons, especially in water less than 25 m deep.

Fig. 1. Illuminated Scott Reef bathymetry, showing Torosa field outline (dotted red line) plus Torosa 3D survey area (brown), Maxima 3D survey area (blue), Gigas 2D OBC survey lines (purple), Tridacna 3D OBC survey area (light blue) and Rosebud 3D area (red).

Prior to commencing the seismic program over the shallow-water parts of Torosa field, Woodside carried out an extensive airborne bathymetric survey of the entire Scott Reef system in early 2006. Additional bathymetric surveying was undertaken to fill some data gaps within the airborne bathymetry survey, resulting in a comprehensive, highly-detailed data set on a 4 sq m grid with vertical resolution of 0.1 m, covering both northern and southern Scott Reef to depths of about 50 m. Woodside also undertook extensive metocean work, including both tide and current modeling and model verification. These datasets were critical for the planning of the subsequent seismic surveys.

The Maxima 3D marine seismic survey, conducted in late 2007, was designed to extend 3D seismic coverage from the Torosa 3D area toward Scott Reef and into the southern Scott Reef lagoon using conventional towed streamer technology. A 4 3 4,500-m, solid streamer, dual-source geometry was selected as a balance between maneuverability and subsurface coverage. Minimum water depth for the towed streamer survey was 25 m.

The Maxima survey proposal was referred for both Western Australian and Commonwealth environmental assessments in 2006. Final approvals were received in September 2007 with a number of specific conditions that required, among other things, the completion of a field verification study into actual versus predicted impacts from airgun noise emissions on tropical marine fish and corals, to be conducted at Scott Reef prior to commencement of seismic acquisition. This study formed ‘Phase 1’ of the ‘Adaptive Management Program’ (AMP) for the Maxima survey (Grebe et al, 2009).

The AMP Phase 1 activities required engagement of a range of vessels within a short timeframe, and during a period of high industry demand. Vessels needed to operate safely in the remote location of Scott Reef. All had to pass inspection for possible invasive marine pests.

Phase 1 field studies commenced in late August 2007 and, at their peak, involved 123 people and eight vessels operating offshore. Coordinating the activities of such large numbers of vessels and people at Scott Reef was a major task, with the health and safety of personnel of paramount concern. Emergency response planning required very careful attention, especially for periods during which the seismic survey vessel (the only vessel equipped with a helipad) was not present. To provide the best possible medivac capability a fast vessel, capable of reaching shore in about six hours, was engaged for the duration of the Phase 1 work.

The most ambitious component of the research program was a field study to test for possible airgun effects on fish hearing, the first study of its kind–outside a laboratory–in the world (Hastings et al, 2008; Grebe et al, 2009). This demonstrated that airgun exposure did not result in observable effects on hearing for caged hearing-specialist fish at cumulative sound exposure levels up to 190 dB re 1µPa2sec.

Survey approvals also required “preliminary surveys to determine minimum air gun capacities required for seismic data acquisition,” with the minimum array to be used during the survey. These tests could not be done within the Maxima survey area, because the final survey approval was conditional upon completion of the tests. Fortunately, approval had been granted for another survey some 80 km to the south of Scott Reef, and the tests were done there. A single traverse was shot, using three different airgun array configurations. The smallest (2,055 in3) array gave adequate seismic data. This was verified on a single test line acquired in the south Scott Reef lagoon after completion of the fish exposure trials.

Safety performance and environmental compliance during the high-exposure Phase 1 period were excellent. The key study results included:

• No evidence of hearing damage to fish
• No evidence of damage to coral
• No evidence of fish mortality
• Impacts less than initially modeled
• Successfully-demonstrated suitability of minimum airgun array size.
• The results were submitted as part of the final AMP on Sept. 20. After completion of Phase 1, five of the support vessels were demobilized, leaving a ‘conventional’ survey unit, consisting of the survey vessel, a tender and a scout vessel.

In addition to the pre-survey field trials, the approvals for Maxima included many specific operational restrictions, including:

• Temporal and spatial exclusions related to turtle nesting on Sandy Islet, and to coral spawning (including field studies to determine when coral spawning would occur)
• Temporal restrictions (which differed with water depth) on acquiring lines close to one another
• Restrictions on the total number of shots and on separation between shots
• Controls on the distribution of shots within state waters
• Pre-survey cetacean and marine fauna observations
• Special conditions for refueling.
• To assist in meeting compliance and monitoring requirements, Woodside engaged three qualified ‘marine fauna observers’ (MFOs), who made baseline observations at and around Scott Reef for three weeks, prior to commencement of Maxima. As the survey progressed, they made observations for cetaceans and other marine fauna, and also assisted with documentation and monitoring.

Seismic acquisition (Phase 2 of the AMP) commenced on Sept. 22 and was completed Nov. 17. Survey downtime during the coral spawning exclusion period was minimized by redeploying the survey unit to another 3D survey area nearby. Overall HS&E performance for the survey was very good.

Gigas 2D OBC survey. Shallow water and restricted access precluded conventional, towed, streamer seismic acquisition in the northern Scott Reef lagoon. In May 2008, the Gigas 2D Pilot OBC survey, designed to test the suitability of ocean bottom cable (OBC) acquisition for conditions at Scott Reef, was acquired in the shallow waters of North Reef. In an OBC survey, cables containing hydrophones and geophones are laid temporarily on the seafloor from small boats. These cables are then connected to recording instruments aboard a vessel moored near the seismic line. The airgun source is deployed from another vessel, which moves along the line discharging its source array. OBC surveying is, thus, a small boat operation; experience gained during the Maxima survey was critical in planning this activity in a survey area more than 300 km offshore.

The approval process for the Gigas survey also benefited from Maxima experience. Key concerns were possible acoustic impacts on coral from seismic sources at close ranges (the survey area included areas with water depths less than 5 m) and the potential for coral damage from cable laying and entanglement. Detailed field studies into possible acoustic effects on corals, the potential for coral damage due to cable laying and entanglement, and observations of fish behavior were thus integral parts of the pilot survey.

The survey was carried out in May 2008. Divers were required to be on standby to retrieve any entangled sections of cable: this required a support vessel capable of carrying and supporting a dive chamber and the dive team. The survey unit consisted of a survey support and accommodation vessel, an additional accommodation vessel, source boat, receiver boat, a scout vessel and its tender (which served as a dive platform), and four rigid inflatable boats (“DIBs”) for cable deployment and retrieval. Of these, only the accommodation and scout vessels were ‘in class’, i.e. able to operate independently at, or en route to, Scott Reef. The survey unit travelled to and from the survey area in convoy, with the DIBs deck-stowed aboard the support vessel.

The seismic crew was accommodated aboard the support vessel, which remained outside the north lagoon and did not anchor for the duration of the survey. Personnel were transferred to and from the lagoon via DIBs. The source and receiver, and scout boats, remained in the north lagoon for extended periods. The dive crew were housed aboard the second accommodation vessel and transported via tenders. All personnel transfers, seismic activities, transits through the reef and refueling were daylight-only operations.

The survey was designed to minimize the total operational footprint while quantifying any impacts that did occur. In sensitive areas, the proposed survey lines were surveyed with underwater video and, where necessary, were relocated to paths of minimum impact. Any cable segments found to be snagged were marked with buoys for later retrieval by divers. In all, only 13 cable segments needed to be recovered in this manner, out of more than 780 segments laid.

The dive team also carried out a detailed program of coral and fish monitoring. To investigate potential acoustic impacts on fish, 18 monitoring sites were established. Of these, 12 were exposure sites, and six were control sites that were not exposed to the seismic source. Five remote underwater video cameras were set up at each site, with six sites active at once (a total of 30 cameras deployed simultaneously). Each site was sampled three times: before exposure to the seismic source; immediately after exposure (within 96 hr); and five months after exposure.

A total of 106,436 individual fish observations, representing 37 families and 329 species, were recorded. Fish were identified, counted and their behaviors recorded. No statistically significant effects on fish behavior were found, regardless of the distance between the seismic source and the fish. The specific results were:

• No mortalities were observed
• No erratic swimming was observed
• No changes were detected in fish population structure, species richness or abundance, or in startle rate.
• The conclusion was that there was no observed impact on fish as a result of exposure to the seismic source.

To investigate potential acoustic impacts on corals, divers selected and established five monitoring sites (two exposure sites and three control sites) that had a range of coral types present. One hundred corals were identified, examined, tagged and photographed at each site. Each site was sampled three times: before exposure to the seismic source; immediately after exposure (within 96 hr); and five months after exposure. There was no observed impact on hard corals as a result of exposure to seismic sound.

A separate study examined the effect of ocean bottom cable deployment and retrieval on corals. Three exposure sites and three control sites were established. Visual assessments were made before laying cable, after cable retrieval, and five months after the survey, to check for disturbance and injuries (such as breakages, abrasions and overturns, and for changes in coral color). These observations showed that coral damage, due to cable laying and retrieval, was minimal, with less than 0.02% of coral colonies injured. As expected, damage was restricted to fragile branched corals. The area of coral habitat impacted by the survey was estimated to be 0.004 ha, of the total 15,025-ha survey area. The return visit to the reef five months after the survey showed clear evidence of coral recovery, Fig. 2.

Fig. 2. The Torosa gas field lies beneath the abundant coral of Australia’s Scott Reef. Photo courtesy of Woodside Energy.

Refueling requirements were complex. The prohibition on vessel-to-vessel refueling within 12 nm of the reef posed a particular challenge, because the source and receiver vessels needed to operate within northern Scott Reef lagoon for extended periods. The sea state at 12 nm from the reef, was often marginal for refueling, although conditions were often much more favorable close in to the lee of the reef. The smaller inflatable workboats were lifted aboard the support vessel and refueled in bunded areas on deck: 98 of these lifting operations were completed, without incident. The larger source and receiver vessels were also required to transit the reef every few days to discharge treated wastewater. Because no discharges were permitted within 4 nm, and holding tank capacities were limited, waste management on the smallest vessels required innovative solutions for waste transfer.

Six survey lines totaling 82 receiver line km were recorded over North Reef. Receivers were successfully deployed (and retrieved) within North Reef lagoon and over the reef crest, to a maximum depth of about 30 m. A small airgun array was used successfully within the lagoon and in deepwater outside the reef, off-end from the receiver lines. However, despite considerable effort, the source array could not be used over the reef crest, and the lack of these shallow water shots was a major disappointment for the survey.

Tridacna 3D OBC survey. Results of the research into seismic impacts in the marine environment, combined with operational experience gained during the Maxima and Gigas surveys, were invaluable in planning, and gaining regulatory approval for, the Tridacna 3D OBC survey, a rich-azimuth two-component 3D survey covering approximately 80% of North Reef, conducted in 2011.

As with the previous surveys, the operational envelope for the Tridacna 3D OBC survey was constrained by technical requirements, and environmental and climatic factors. Environmental constraints included spatial and temporal exclusions related to turtle nesting, mass spawning of corals and whale migrations. Weather considerations included strong prevailing southeasterly trade winds over the winter months (May to September) and the potential for tropical cyclones during the summer monsoon season (December to April). The survey was scheduled from August to November, to avoid the most severe winter trade winds and to ensure survey completion prior to the tropical cyclone season.

The ocean bottom cable technique was selected for Tridacna following a re-evaluation of potentially-suitable acquisition technologies. The separation of sources and receivers is a significant advantage for OBC over streamer technologies and allows a wide variety of acquisition design options. Various parallel (inline swath) and orthogonal (cross spread) acquisition geometries were evaluated; an orthogonal design was selected, with the survey to be acquired in a series of parallel receiver patches. This enabled rich-azimuth (or, effectively, full-azimuth) acquisition, with the main disadvantage relative to the inline swath design being the substantial shot overlap required to zipper adjacent parallel patches together. Patch size was dictated by the quantity of equipment available and the ability of the contractor to keep the spread ‘live,’ once it had been deployed in the water. The more live receiver spread available, the fewer patches would be required, minimizing the repetition of shots between adjacent patches and increasing overall survey efficiency.

A nominal live receiver patch of six receiver lines, each 6,000 m in length, was selected as providing the best balance between the number of patches and the quantity of equipment deployed at any one time. Eight patches would be required to cover the nominal survey area. Each patch was expected to take approximately two weeks to acquire. This allowed coordinating of seismic operations with the fortnightly spring-neap tidal cycle, to allow data acquisition over the shallowest-water emergent-reef areas during the highest (spring) tides.

Six specialist, shallow draft, rigid inflatable vessels were used for receiver pickup and deployment. Three specially configured source vessels enabled shots to be acquired efficiently throughout the reef area, and with some good overlap between them, gave good source redundancy. A very shallow draft, rigid inflatable source vessel, capable of deploying a shallow, side-slung airgun, was used for very shallow areas of the reef crest and within the lagoon. A shallow draft catamaran source vessel was used inside the North Reef lagoon—this vessel deployed a more conventional, towed source array. A larger monohull source vessel was used to acquire shots outside the reef.

The survey proposal was referred to the Department of Sustainability, Environment, Water, Population and Communities for approval under the Environment Protection and Biodiversity Conservation Act 1999. An Environment Plan was submitted to, and accepted by, the Western Australian Department of Mines and Petroleum, in accordance with the Offshore Petroleum and Greenhouse Gas Storage (Environment) Regulations 2009. Specific commitments to minimize environmental impacts during the Tridacna survey included detailed procedures for anchoring, refueling, and cable deployment and retrieval, that were designed to minimize impacts to benthic habitat, plus a detailed procedure for the discharge of acoustic sources, designed to minimize impacts to marine flora and fauna.

Seismic data acquisition commenced in August 2011. A high degree of specialized acquisition planning was required to optimize survey activities, patch by patch, throughout the tidal cycle and on a day-to-day basis. Detailed shooting plans were generated each night for the following day. This daily planning ensured that an efficient mix of shallow-water acquisition (for which high tide was essential), deepwater acquisition (which could be shot at any time), and cable retrieval and deployment, was available, so that waiting on tides was kept to a minimum. The survey commenced slightly later than expected, and initial production rates were lower than expected. Also, the tidal cycle became increasing unfavorable. As with Gigas, the Tridacna survey was a daylight-only activity and, after the spring equinox in September, the highest high tides occurred during darkness. Onsite planning was crucial in supporting a changeover to a nominal eight-receiver line patch for the last two patches, effectively increasing productivity by a third without an associated increase in cost.

The Tridacna survey was completed successfully, safely and with minimal environmental impact in November 2011. With the exception of one small area, the prime objective of obtaining full rich-azimuth 3D coverage across the entire survey area, including over the reef crest, was met.

Rosebud 3D survey. Despite the successful completion of the Maxima streamer survey and the Tridacna OBC survey, there were several small areas with incomplete data coverage in the narrow gap between the full-offset coverage of the two surveys. These areas were deficient in near and near-mid offset range traces that were considered important for seismic imaging beneath the steeply-dipping outer reef flanks. The Rosebud 3D survey was proposed to infill these areas.

Rosebud utilized a small seismic survey vessel with dual, conventional airgun, source arrays and two relatively short (3,000 m), conventional solid streamers. For some of the closest passes to the reef, the receiver array was reduced to a single 1,500-m streamer.

Rosebud was the first seismic survey to be acquired at Scott Reef following the commencement of safety and environment oversight by a new national regulator, the National Offshore Petroleum Safety and Environmental Management Agency (NOPSEMA).

Woodside prepared and submitted multiple environmental approvals in accordance with Commonwealth and State legislative requirements and received the following environmental approvals prior to the commencement of the Rosebud 3D MSS:

Rosebud 3D MSS Environment Plan in accordance with the Offshore Petroleum Greenhouse Gas Storage (Environment) Regulations 2009, administered by National Offshore Petroleum Safety and Environmental Management Agency (NOPSEMA) (covering the portion of the survey within Commonwealth waters)

Rosebud 3D MSS Environment Plan in accordance with the Petroleum (Submerged Lands) (Environment) Regulations 2012, administered by Department of Mines and Petroleum (DMP) (covering the portion of the survey within Western Australian state waters)

Rosebud 3D MSS Referral Decision (EPBC 2012/6493), in accordance with the Environmental Protection and Biodiversity Conservation (EPBC) Act 1999, administered by Department of Sustainability Environment, Water, Populations and Communities (SEWPaC) (with respect to matters of national environmental significance).

During the assessment process, regulators expressed heightened concern regarding the possibility of oil spills that might result from vessel collision or grounding, plus possible entanglement of in-sea equipment with the reef.

To minimize the risk of vessel grounding or entanglement, the highly-detailed Scott Reef bathymetric dataset was used to derive an operational polygon with an inner boundary based on a smoothed envelope around the 20-m bathymetric contour. This ensured that the water depth on the inboard edge of the operational area would be at least 20 m, regardless of tides. An additional buffer zone, 100 m wide, was established beyond the smoothed 20-m bathymetry envelope, to give an additional safety margin. In addition, a support vessel capable of taking the survey vessel in tow was engaged for the survey, and the focus of the first day of operations was tow planning and drills.

The Rosebud 3D survey was completed in October 2012. Total coverage was approximately 37 sq km.

The Maxima 3D survey (in 2007) and Gigas 2D OBC survey (in 2008) demonstrated that it is possible to acquire seismic reflection data in a coral reef environment with minimal environmental impact. Both surveys demonstrated successful management of the health and safety exposure inherent in small-vessel operations, in a remote offshore location, and responsible operation in highly-sensitive environments. Between them, the surveys also produced best-in-class research on seismic impacts on fish and corals, conducted in situ at Scott Reef. The research results formed a solid basis underpinning subsequent survey results proposals and addressed the concerns of environmental and safety regulators in assessing these proposals. This led directly to regulatory approval for, and successful completion of, the subsequent Tridacna 3D OBC survey in 2011 and the Rosebud 3D MSS in 2012. 

REFERENCES

1. The participants in the Browse LNG Development are: BP Developments Australia Pty Ltd., BHP Billiton Petroleum (North West Shelf) Pty Ltd., Japan Australia LNG (MIMI Browse) Pty Ltd., Shell Development Australia Pty Ltd. and Woodside Browse Pty.