Seismic Surveys: What We Dont Know Can Hurt
This page contains the Executive Summary
All citations are contained in the full report
Download Full Report (30p):
Download Bibliography: [DOWNLOAD(doc)]
Download Executive Summary (5p): [DOWNLOAD (doc)]
A report by Jim Cummings, Acoustic Ecology Institute
Prepared for Greenpeace USA
Seismic surveys utilize airguns to produce explosive impulses of sound directed toward the ocean bottom. Echoes produced by these impulses are used to gather information on sub-surface geological features; this information is used by academic geologists and the oil and gas industry. Both researchers and the public have become concerned about ways that sounds created by the airguns may impact ocean creatures; while it is generally believed that the risk of physiological damage is low, there are many uncertainties in our understanding of both sound propagation and biological effects. In addition, the complexities of acoustics science and inconsistent measuring systems used within the research community have made it difficult for non-scientists to communicate their concerns in an informed way. Here we provide a primer for the non-scientist on current knowledge about airguns, propagation of their sounds over distance, and acoustic sensitivity of ocean dwelling creatures; a discussion of the need for a precautionary approach to regulation of anthropogenic sound in the seas; and a series of suggested mitigations and research programs to be undertaken in consort with future seismic surveys.
Seismic surveys utilizing airguns have been taking place in mineral-rich areas of the worlds oceans since 1968. Among the areas that have experienced the most intense survey activity are the North Sea, the Beaufort Sea (off Alaskas North Slope), and the Gulf of Mexico; areas around Australia and South America are also current hot-spots of activity. (for details of US OCS trends, see 2000 MMS report Geological and Geophysical Data Acquisition, Outer Continental Shelf Through 1999, [READ REPORT(pdf)])
The impulses created by the release of air from arrays of up to 24 airguns create low frequency sound waves powerful enough to penetrate up to 40km below the seafloor. The source level of these sound waves is generally over 200dB (and often 230dB or more), roughly comparable to a sound of at least 140-170dB in air. These sound waves become less intense as they travel away from the airgun array, due to both scattering off of the seafloor and acoustic spreading, by which the total sound energy is spread out into a larger and larger area as it spreads in a spherical or cylindrical fashion. The sound is generally falls to below 180dB at about 1km from the source; the reduction continues, but at a slower pace, as distance increases, leveling off at around 100dB at a distances of 40-50 km, and remaining audible for hundreds of kilometers (depending on background noise levels); many researchers have tales of acoustic studies being interrupted or even abandoned due to airgun sounds from distant surveys.
The central concerns about seismic airguns include physiological damage they may cause (ranging from catastrophic tissue damage to long-term hearing loss), behavioral responses such as causing fish or whales to move away from the survey area, and the added noise airguns contribute to the general din of the sea. Current mitigation measures, based largely on avoiding severe tissue damage, may deserve revision in the face of increasing evidence of widespread behavioral avoidance.
A note on dB and measurement
It is important to note that there are several acoustics measurement issues to take into account when considering the dB level of any waterborne sound. These issues are covered in some detail in the full report. Briefly, there are three main points to be kept in mind. First, due to the higher density of water, sounds of similar perceived volume have a higher pressure in water; the standard correction is about 60dB, as noted above. Second, there are several measurement systems used by acoustics researchers; each is useful for considering different sorts of biological effects. Peak-to-peak measurements are several dB higher than Mean-squared measurements, with in turn are a few dB higher than Equivalent Energy measurements. The total variation in the standard measurement systems can account to up to a 26dB difference, so when we are considering the effects of a given sound level, it is important to recognize which scale is being used. Finally, each species preferentially hears a certain range of frequencies more easily (for example, large whales hear and vocalize best in lower frequencies, while dolphins hear and vocalize at significantly higher frequencies). It is generally assumed that human-made noise in frequency ranges outside the primary range for a given animal will be either imperceptible, or will be heard relatively faintly, and therefore have relatively milder effects on behavior. However, high-power sound waves may also have physical effects beyond the effects on auditory perception. And, the actual hearing ranges of many sea creatures, especially large whales, is not known (it is often inferred from the range in which a given species vocalizes, but as we know from our own hearing, the range of perception is much wider than the range of vocalization). These two factors, acoustic measurement variation and a large degree of uncertainty about the physiological and perceptual hearing systems of sea creatures, both suggest that reliance on any specific dB measurement is problematic. All parties are on very thin ice in making any definitive claims about how sea creatures may respond to a given dB level. At best, an underwater dB rating is a decent estimate of the sounds physical power, and a somewhat useful benchmark by which to make informed guesses about its possible effects. This is why scientists tend to rely on behavioral responses and observed physiological damage caused by ocean-based sound, and to avoid making direct comparisons with human hearing, and especially human safety thresholds, in the air.
That being said, it is especially important to base our decisions about operational safety, mitigation, and regulation on careful observation of the actual responses of many species at a range of distances from seismic survey vessels, rather than to over-rely on ideas about the danger of specific dB levels or on models of how we think that the dB level will change as the sound moves out into the ocean environment. Nevertheless, researchers and regulatory agencies have tended to rely as best they can on the concrete numbers that have emerged from various studies of the impact of anthropogenic (human-made) sound on sea creatures. This report follows suit, albeit with a high degree of caution.
History of Seismic Surveys, Natural Ocean Noise
When considering the impacts of seismic surveys on sea creatures, it is also important to remember that through several decades of seismic airgun activity, there is little evidence of severe physiological damage occurring, and that behavioral effects seem to have not caused significant population declines. In the Gulf of Mexico, for example, close a million miles of seismic survey lines have been shot, while the area remains home to acoustically sensitive species such as sperm whales and dolphins. However, there are few if any studies that offer a control that can assure us of what population trends would be in the absence of airgun noise in these areas; likewise, we have no way of knowing whether local populations are experiencing more hearing loss than their kin in other parts of the world. Further, there is some concern that severely injured animals may sink rather than float to shore, so that our estimates of injuries may be understated. (One of the few mass death events that has been solidly tied to acoustic trauma, the 2001 beaked whale strandings in the Bahamas after Navy sonar testing, was compounded by the disappearance of the entire rest of the population, which had been under observation for several years.)
It is also true that the sea is not always a quiet place. The sounds of wind stirring the surface of the ocean, lightning strikes, animal vocalizations and noises (including whale calls, echolocation clicks, and snapping shrimp), and sea-quakes on the ocean floor can all create noise in the range of 60-180db (seaquakes and some whale calls can approach or top 200dB). Anthropogenic noise from shipping, explosives, construction, fish-finding sonar, and other sources is a constant feature of many or most ocean soundscapes. Shipping noise, in particular, often at source levels of 150-190dB, is a dominant noise in many areas.
Increased Scrutiny in the face of Uncertainty
The fact that seismic surveys have been taking place for decades, and they operate for only a few days to weeks at a time, often in areas that also experience heavy shipping traffic, can suggest that they do not deserve special scrutiny. However, we must bear in mind that the current attention on seismic activity is emerging within a social context that reflects a new and increased concern about ocean health in general, and ocean noise in particular, including questions about the cumulative impacts of many sound sources. In this context, it is understandable that our attention would first turn to some of the most extreme sources of human noise. This does not necessarily mean that all airgun activity should cease, but rather that closer attention to the effects of airgun noise offers a striking opportunity to learn more about the effects of noise in general.
Disturbances that may seem negligible taken individually are being seen within a larger context, one that includes a deeper consideration for the rights of other species to be free of acoustic harassment, as well as acknowledgement that there is much that we do not know about the lives (and the auditory systems) of specific species, or the ways that communities of sea creatures may use sound for navigation or perception of their world. Indeed, very little is known about the physiological or behavioral responses of cetaceans, fish, and other sea life to sound. We have no direct knowledge even of the hearing range of large marine mammals, and there is ample reason to suspect that our land-based conception of hearing (centered on the ears and on perception of discrete frequencies of sound) offers only a very partial picture of the acoustic perception of water-based creatures. It seems quite possible that many or most sea creatures perceive sonic vibrations throughout their bodies, and are capable of sensing minute changes in acoustical energy, beyond our perceptual imagination or scientific measurements. These changes in both scientific and social perspective are important to appreciating the current increase in concern about the effects of human noise in the sea.
Application of Harassment Regulations
Some observers have complained that the public, press, and even courts are misinterpreting the Marine Mammal Protection Act (MMPA) and National Marine Fisheries Service (NMFS) regulations governing small takes by incorrectly asserting that take means to kill marine mammals, while in fact takes include any sort incidental exposure of animals to human activity that may cause behavioral changes, including simply creating sounds that they may hear (such exposure is termed level 2 harassment under the statutes). In recent years, many projects have been challenged after the public learned that small take permits had been issued that allowed takes of hundreds or thousands of marine mammals. While there may indeed at times be some confusion on this, the fact is that the public, press, and courts are beginning to realize that harassment in the form of exposure to excessive sound is something worth being concerned about, and are calling on regulators to address such harassment as defined. This reflects a desire on the part of many citizens (as well as a growing number of researchers) that our regulations should not simply protect sea creatures from physical harm, but should reflect a consideration and respect for their right to hunt, rest, and travel through the oceans without being subjected to excessive human noise. Our desire to search for oil should not automatically trump their ability to live in peace.
Seas of Data, Seas of Sound
In surveying the existing research on seismic surveys, its easy to get lost in a sea of numbers, charts, and conclusions. The abstractions generated by models of acoustic propagation, threshold sound levels, and predicted consequences can lead to both a false sense of security (its not killing the whales) and false sense of uncertainty (theres no way to know what the animals are experiencing). Strangely, both responses lead to similar action, or more commonly, inaction.
As an antidote to the torpor that can be imposed by the deluge of data, it is especially revealing to step back and listen to what researchers say about their own experiences in the field. These anecdotal reports from people who have grounding in the objective data may be especially eye- (and ear-) opening. One longtime whale acoustician, Peter Delory, made the following observation at a conference:
The authors have been in the North and Norwegian Seas on many acoustic experiments, and listening to the raw output from hydrophones deployed from the research vessel, the entire soundscape is often dominated by the repetitive 'boom...boom...' of distant geophysical surveying. This incessant cacophony, an acoustic equivalent to the fabled Chinese water torture deeply disturbs some individuals who are exposed to the sound over long periods. Perhaps it does the same to whales. . . Recently, attempts were made to monitor baleen whales off the West Coast of the British Isles using an array of (deep-water) hydrophones. Levels of background noise were so high in the summer months due to oil-related seismic surveying that monitoring had to be abandoned for long periods. . . One can only assume that baleen whales' ability to monitor (their) acoustic environment might be similarly compromised by such noises.
There is much concern among environmental advocates about the possibility of loud human sound, especially low frequency noise, causing severe tissue damage such as that seen in conjunction with some low-frequency and mid-frequency military sonars that operate at intensities similar to airguns (source levels over 200dB). However, in contrast to several strandings (often with clear physiological effects such as bleeding from the ears) that have coincided with sonar tests, there has not been any particular increase in strandings in areas exposed to extended airgun operations. One such event, involving two beaked whales (the family that seems most susceptible to sonar) occurred in 2002 in the Gulf of California, although the bodies were too decomposed when found to allow testing that could determine whether the auditory system seemed damaged. (A series of mass strandings coincidental to a seismic survey in southeastern Australia during late 2003 has further fueled concern.) Still, given the many years of airgun operations worldwide, the current evidence is far less troubling than that which is accumulating regarding military sonars, especially the mid-frequency active sonar.
That being said, recent research on fish has identified the physiological mechanism by which loud sound (in the 180dB range) can cause long-term hearing damage. Tiny hairs in the ears of fish can be damaged, and holes left in the sensitive tissues within the auditory system.
In addition, evidence is beginning to accumulate that suggests that some populations of sea creatures may be experiencing long-term hearing loss. It is not possible to pinpoint exposure to loud noise as the source of this damage to the auditory systems, but it bears further scrutiny. Most striking is the observation by Darlene Ketten, a leading authority on cetacean hearing, that close to 50% of specimens examined after beaching seem to have some degree of compromise in their auditory systems.
The mildest (and likely the most common) physiological effect that is associated with exposure to excessive sound is acoustic masking. The more ambient sound there is in an environment, the harder it will be for sea creatures to communicate, perceive prey or predators, and navigate. While by far the dominant source of anthropogenic noise in the sea is shipping, accounting by some reports for over 90% of noise, the noise of ships ranges from 150-190dB at the source. Both naval sonar and airgun noise are significantly more intense than even the loudest ships, so that they can be heard above the shipping noise at close range (up to 10-30km), and become a significant factor in the distant ambient noise at greater ranges. Thus these louder sounds, though more transient (and pulsed, rather than continuous) are likely to cause additional masking of auditory information of biological importance, especially in the near to mid distance.
Much more prevalent than clear physiological damage are changes in behavior in response to airgun noise. It is not certain to what degree these changes lead to biologically significant effects; in many or most cases, forcing sea creatures to move up to several kilometers to get away from excessive noise is considered by researchers and regulatory agencies to be a negligible impact.
Nevertheless, it is clear that whales, fish, and sea turtles all exhibit avoidance behavior when encountering sounds above around 155dB (equivalent energy, 170dB RMS, 183dB peak). There is some variation in this threshold, but it is surprisingly consistent across species. While it seems clear that movement on the scale of 1-30km, as has been seen in various species, does not cause any immediate survival impacts on individual animals, the long-term effects on reproduction or species-level survival, as may be caused by stress, use of limited energy reserves, or temporary exclusion from preferred feeding grounds, is very difficult to know. In addition, as we become more aware of the extent of these modest but relatively consistent behavioral responses, many citizens and some scientists are raising the sorts of questions mentioned earlier, about whether it is ethical to impose our sounds on other creatures in this way.
When dealing with species that are either declining (eg, many fish species) or in the process of a gradual return from the brink of extinction (eg, many cetacean species), there is some concern that small and cumulative impacts from many sources, including noise, may be significant in affecting long-term population trends. For this reason, many environmental advocates, and some regulatory agencies, are beginning to apply precautionary approaches to regulation and management of such stocks. The precautionary approach is more likely consider the uncertainty in our knowledge as an important factor in making management or mitigation decisions, rather than waiting for conclusive causative evidence before limiting human activity such as noise-making.
Current mitigation measures include shutting down airguns when cetaceans are seen at very close range (100m to 1km, depending on the size of the air gun array and species of concern in the area), gradually ramping up the power of the airguns over a 15 minute period (to allow fish and whales to move away), and employing on-board observers to watch for whales at the surface. The generally accepted safe exposure level for sea creatures is 180dB or below (peak levels), which compares fairly well (if a bit loosely) to accepted terrestrial limits for impulse noise and observed behavioral responses of marine mammals and other ocean species, as well as being, conveniently enough, within the range that allows ship-board observers to spot whales (at least whales at the surface). This rough correlation, however, should not blind us to the many indications that the 180dB threshold, reflecting a zone 1-2km around the survey ship, may be set a bit too high.
Many innovative approaches to mitigation and monitoring are emerging from researchers and regulatory agencies worldwide. A best practices approach to mitigation would include:
- Case by case analysis and modeling. Because of the great variability in transmission loss across different sorts of ocean floor profiles, and the small but significant variation in the output levels of individual air gun arrays, we follow Robert McCauley, who (in a 2000 report commissioned by the Australian oil and gas industry) called for case-by-case modeling of likely propagation patterns: at present, predicting the horizontal sound propagation from any specified air-gun array source needs to be done on a case by case basis. . . Accurately predicting levels at specified ranges and water depths requires modeling of the source and local environment. We also encourage the use of cumulative effects modeling such as developed by McCauley, to better adapt each surveys operations to fit local biological and geophysical conditions (cumulative effects modeling involves charting the total number of airgun shots that will be heard at various dB levels in the survey area). Such case-by-case procedures may help to ameliorate the effects of current uncertainties, which can lead to seemingly excessive regulatory hurdles for very small single-gun site surveys.
- Consider establishment of a larger exclusion zone to reduce behavioral effects, especially on species with tenuous populations. It may be that observed behavioral disruptions will be better addressed by a 2-5km radius, rather than the current 1km radius. In the same way, the current 180 dB re 1mPa2 standard for acceptable received levels of sound (based on avoiding physiological damage) may need to be adjusted downward to avoid behavioral disruption.
- Require passive acoustic monitoring (Passive Acoustic Monitoring utilizes listening devices to eavesdrop on vocalizations beneath the surface; systems have been developed that can identify a large number of species based on such acoustic data): such monitoring can complementing visual observations by identifying animals vocalizing beneath the surface, down to the sea bottom. Passive acoustic monitoring can also extend the zone of effective observation, which would be important should we choose to apply a more precautionary approach to acceptable received sound levels.
- Incorporate Environmental Effects Monitoring into all active seismic surveys. EEM should include measurement of received sound levels at ranges of 1-25km, as well as both visual and passive acoustics monitoring of the responses of marine mammals, fish, and other species present in the area. One area that has needs more research is the effects of exposure to airgun noise experienced by bottom-dwelling creatures directly beneath the guns (one anecdotal report by a geologist noted that a seafloor profile that at first seemed soft, then hard, turned out to be caused by a sea-bottom worm population retreating into the safety of their hard casings upon being exposed to the gun sound; this was seen by the geologist as an interesting oddity, and there was no associated biology study that could follow up to see whether they re-emerged and thrived after the survey).
- Consider cumulative impacts over time in permitting and effects modeling; include consideration of seasonal and historical impacts from other activities (shipping, military, industrial, other seismic) in the specific survey area and nearby region. Develop databases that track the history of seismic and other industrial activities, using GIS mapping. The UK has initiated such a project, with maps available on a publicly-accessible website.
- Use of the lowest possible power array to meet local conditions and obtain information being sought.
- Extend ramp-up times where turtles are present; 30 minute ramp-ups are minimal and 60 minutes preferable, to accommodate turtles relatively slow swimming speeds. Evidence from fishery studies also suggests that smaller fish would benefit from slower ramp-ups.
- Adapt the sequencing of seismic lines to account for any predictable movements of fish across the survey area.
These best practices would require a greater investment in time and money from companies conducting seismic surveys; this is best seen as a realistic cost of pursuing new sources of fossil fuel, with an acceptance that the resulting cost of the products delivered will be higher. It represents a social choice to accept short-term increase in the cost of fuels (or, in the case of academic geological research, an increase in the cost of knowledge about, say, plate tectonics or earthquake prediction) in order to be more careful about avoiding environmental costs that cannot be replaced should we overstep our bounds. Over time, as our knowledge increases, it is likely that we will be able to design both technologies and implementation strategies that can reduce these added costs associated with seismic surveys.
In addition, we encourage efforts to establish more coordinated oversight of seismic surveys and of noise in the ocean in general. Recent reports from the National Research Council, the Whale and Dolphin Conservation Society, and the Office of Naval Research have all called for stronger coordination and commitment to carrying out much-needed research into the impacts of noise in the sea. Such basic information as a sound map charting worldwide noise levels and a sound budget which attempts to set caps on total noise so that vulnerable areas will be protected, are still only in the conceptual stage. There are several federal efforts underway to establish better coordination of research. We second this desire for better oversight, and encourage adequate funding for these important research and coordination programs.
An independent, government funded commission could coordinate efforts specifically directed at better understanding the impacts of seismic testing and development of improved mitigation and monitoring measures.
Although this report provides a summary of the current state of our knowledge, it is important to realize that all the research that has been done to date still offers only an extremely limited picture of the extent and impact of human noise in the oceans, and of the functioning of ocean creatures acoustic perceptual systems. Research scientists certainly do their best with what they have to work with, but there is a widely recognized lack of solid data in nearly all aspects of ocean acoustics. A 1998 conference presentation by biologists John Potter and Peter Delory raised this point clearly:
Marine mammals are perhaps the hardest mammalian group to study. Virtually all relevant aspects of their biology (including sensory capabilities, undisturbed behaviour and its adaptive significance, distribution and abundance) are only poorly understood. Conducting marine mammal research at sea is always difficult and costly. . . Given such a background of ignorance, it is extremely difficult to even establish a meaningful framework for estimating the impact of noise on these animals.
For this reason, its crucial to remember that the observations and studies that have taken place provide just some starting points for understanding, and that there is much more to be learned about the acoustic experience of ocean creatures. Just as it is imprudent to make sweeping conclusions based on current knowledge, likewise it is important not to lose sight of the idea that human activities in the ocean are likely having effects that are not yet recognized. Since many marine species are experiencing severe population declines (e.g., most fish species), or are in the tenuous process of recovery (e.g., many cetacean species), it is essential to act with caution, knowing that any errors we make now in assessing the ecological impacts of our actions could easily lead to biologically critical population stresses, including limited genetic diversity or extinction.
It is the authors hope that increased focus on the effects of human noise on ocean creatures will extend beyond the most extreme noise sources (airguns and navy sonar), to also include consideration of the effects of the most widespread noise sources, such as shipping and fish-finding sonar. The underlying goal should be to better understand the impacts of our actions, so as to minimize their effects on ocean creatures.