Whales and Sonar

by Brian Dunning
December 14, 2010

Today we're going to sink into the deep ocean waters, the realm of great natural denizens such as beaked whales, and of the hulking artificial beasts called submarines. In this world of darkness and near-zero visibility, audio shimmers across the frequency spectrum, from the high-frequency chatter of dolphins to the great long distance low-frequency calls of whales. And, every once in a while, a tremendous electronic burst rips through the environment: a sonar ping. What happens next is debated worldwide. Some say the whales are driven mad, others say they are disoriented and beach themselves, some claim they are deafened or even killed outright. What's the truth? What is the real impact of Navy sonar on marine life?

Briefly, there are two basic kinds of sonar, active and passive. Active sonar sends out loud pings, much like radar, and the returning sound waves paint a picture of the surrounding environment including the location of enemy vessels. Passive sonar listens only without making any sound of its own. Active sonar is the type we're interested in.

There are different types of active sonar, primarily used in anti-submarine warfare (ASW). Low-frequency active sonar (LFAS), which operates around 300 Hz, was first suspected of causing harm to marine life, but upon further study, it was found there was no statistically significant correlation between the use of LFAS and whale strandings. Later we discovered that mid-frequency active sonar (MFAS), operating between 3000 and 4000 Hz, did in fact have such a correlation. So the specific culprit that we're looking at is mid-frequency active anti-submarine sonar.

The species that seem to be susceptible, according to observations, are pretty specific as well. They are the beaked whales, Ziphiidae. The danger to them is in the form of mass strandings. The definition of a mass stranding is pretty generous; two or more whales, within six days, within 74 kilometers, constitute a mass stranding. Whale strandings overall are rare enough that even a case of as few as two whales so far apart likely constitutes a related event. Whales need not die to be counted; often many of the whales are refloated and return to safety, but even these are considered to be part of the stranding event.

Yet, contrary to observations, some activists charge that much broader dangers exist. A 2010 article from the Environmental Protection Information Center, a California non-profit environmental watchdog group, states:

Sonar is extremely dangerous to marine mammals such as whales, dolphins and porpoises. These animals rely on their own sonar for food, navigation, mating and when high frequency sonar like the Navy is proposing to use reaches these mammals, they can be severly [sic] affected.

High frequency sonar has not been observed to have any effect on marine mammals at all, but it's also relatively new. High frequencies are absorbed very quickly by water, so its range is much shorter; thus, it hasn't been very useful for submarines. But with more recent software, it's now being used for high resolution imaging of ice and seafloors, as well as for close range detection of small objects like mines. Toothed whales, including dolphins and porpoises, use immediate-vicinity echolocation in the same way, but at frequencies much higher than the Navy, in fact much higher than the range of human hearing (40 kHz to 150 kHz). Social communication uses lower frequencies, because it carries much better in water.

The limited range of high frequency sonar, and the related fact that most marine life (including non-toothed whales) can't hear it at all, makes it much less likely to have any significant effect on mammals. It may, there's just no evidence for it yet; and acoustic science makes it unlikely.

The Environmental Protection Information Center's article continues:

Sonar has a huge impact on marine life anything from frying fish eggs, disorienting marine mammals causing them to be stranded, to permanently damaging their ears.

"Frying fish eggs" is just silly, and it's a little bizarre that they included such a statement. Other claims have said that whales are driven insane by the noise, which is an unfounded, unprecedented, and uninformed supposition. But the strandings and hearing damage are possibilities, so let's take a look at what we've learned from the latest research.

Scientists and the Navy agree that mid frequency sonar can cause whale strandings, but so far nobody's been able to determine why, and we're just beginning to understand in what circumstances. One hypothesis has been that the noise provokes them to suddenly surface, causing decompression sickness. This remains just a guess, since so far, none of the whale carcasses examined following sonar-associated strandings has exhibited any signs of decompression sickness. So if it is occurring, it's not happening anywhere we've ever observed. It could happen way out in the open ocean where we'd never find the carcasses; we just don't know.

All the other hypotheses we have are equally elusive. These include stress, hearing loss, and disruption of the whales' feeding. Dr. Darlene Ketten is a senior scientist at the Woods Hole Oceanographic Institution who studies sonar and whale strandings. She uses CT scanning on whales and biophysical models of hearing in marine animals, and has yet to find any evidence that whales' hearing has been damaged by sonar. In every case where a stranding was associated with sonar use, Dr. Ketten found that the only discernible cause of death was the injuries sustained from the beaching itself. The whales were well fed, their hearing was intact, and they had no signs of decompression sickness. There's been no evidence to support any hypothesis. The only problem anyone has been able to find has been the big one: The whales were dead.

The best clue yet lies in underwater geography. One of the most active places for the US Navy to use sonar in exercises is off the coast of southern California, where beaked whales abound. Yet, strangely, no cases of whales stranding or otherwise being affected associated with sonar activity off California has ever been discovered. In addition, such sonar is now used worldwide, not just by many nations' navies, but also by commercial vessels, and also other types of underwater noisemaking gear such as the towed airgun arrays used by oil companies searching for new oil fields. If sonar and other manmade sounds were always harmful to nearby marine mammals, we would expect to see a far larger number of such injuries all around the world. But we don't; when and where it happens seems to be highly specific.

The latest and most comprehensive publication on this is from 2009, in the journal Aquatic Mammals, entitled Correlating Military Sonar Use with Beaked Whale Mass Strandings: What Do the Historical Data Show? The paper includes graphical timeline representations from around the world of when naval sonar exercises took place, plotted alongside mass strandings in the local region. The findings to date are intriguing. There has been a correlation of sonar and strandings in the Mediterranean and the Caribbean, but so far, no correlation in California or Japan. The reason seems to be the underwater geography concerned, called the bathymetry. Bathymetric surveys of the areas reveal that steep dropoffs close to shore, especially when confined areas exist, are the places where beaked whales are at risk from sonar. Off the coasts of Japan and California, there are relatively broad shelves, with deeper trenches being farther offshore. Beaked whales do not appear to be at risk in areas with this type of bathymetry.

The research is quite clear that we don't yet have enough answers to fully understand the reasons for this. It may be as simple as how close to shore the whales' habitats are. Most mass strandings associated with sonar have happened where the shoreline is less than 80 kilometers away from water greater than 1000 meters deep. If they're spooked by the sonar, they may simply have less free space in which to avoid it. And of these stranding events, most of them happen on specific coasts where six or more such strandings have been recorded. The evidence strongly suggests that beaked whales are at the most risk from sonar associated strandings only on coastlines with specific bathymetric characteristics. No evidence yet supports any danger to these animals near other coastlines, such as California, or in the open ocean. That's not to claim the danger doesn't exist; merely that we don't have evidence that it does.

Navies continue to develop better techniques for managing this. Sonars are turned off when whales are detected nearby. Aerial surveys are also used to spot whales in advance, but this relies on the luck of happening to spot them near the surface. Scientists from the Woods Hole Oceanographic Institution funded by the Navy have attached non-invasive acoustic recording tags to beaked whales to learn the audio profile of their lifestyle and environment. This data helps us understand what kinds of sounds the whales make use of, and also provides better signatures for ships and submarines to listen for to help detect their presence. This research is still in its early stages, and it's work such as this that needs to be done before we'll have a complete understanding of the true risk factors.

Perhaps the most frustrating aspect of this question is that so much energy and money is spent on protests and lawsuits when neither side yet has sufficient science to support their position. Rather than spend this money learning the facts, both sides are spending money trying to enforce policy based on supposition. According to Dr. Peter Tyack at Woods Hole, "When the courts and the public do not get an accurate picture of the threats posed by different human activities to marine mammals and other wildlife, it distorts conservation priorities and does not serve the interests of the animals."

There's one big shocker that brings the question of whales and sonar into proper perspective. Dr. Barbara Taylor, a cetacean specialist at NOAA's Southwest Fisheries Science Center found that a total of about 200 whales have been killed in strandings associated with naval sonar in the past 40 years. That's about five per year. It's five too many; but it's still five. Many more than that are temporarily stranded or otherwise affected, but with no lasting effects. By stark contrast, 300,000 whales and dolphins are killed every year by fishing operations. That's sixty thousand times as many.

Meanwhile, the lawsuits, protests, and petitions continue raging, with good intentions but little informed science to support the charges. It seems unlikely that the world's navies will make do without sonar, so continued research will remain critical to minimizing or eliminating the risk to marine mammals. If you really want to help the whales, hug a scientist, or otherwise support the research.

Thanks to Barbara Taylor and Bob Brownell from NOAA who kindly provided the latest research on sonar related strandings on very short notice.

By Brian Dunning
Follow @BrianDunning