Big Data Is Saving Whales From Ship Collisions

Scientists are tapping information broadcast by vessels’ automated identification systems to prevent environmental disasters at sea.
A dead adult female humpback whale pinned to the bow of a Princess Cruises ship near Juneau, Alaska, in 2010. (Photo: Ed Lyman/NOAA/via Reuters)
Feb 19, 2016· 3 MIN READ
Richard Conniff is the author of House of Lost Worlds: Dinosaurs, Dynasties, and the Story of Life on Earth and other books.

The oceans cover 71 percent of Earth’s surface, and for much of human history they have appealed to sailors and poets alike as a wild refuge, beyond the reach of mundane human authority. But the scale of the oceans has also made them a haven for criminals and for some of the most destructive modern commercial practices imaginable, from dynamite fishing to industrial-scale bottom trawling. It’s one reason we live in an era of empty oceans and empty nets.

Even ships that operate legally can inadvertently do damage to wildlife—for instance, by fatal strikes, propeller noise, introduction of invasive species, and anchoring in sensitive habitats. But much of that destructive behavior could be avoided with help from big data, according to a new study in the journal Bulletin of Marine Science.

The potential for conservation and law enforcement to reach even into the most remote regions of the ocean comes from an automated identification system (AIS) for ships, originally introduced in 2000 as a tool for collision avoidance, coastal surveillance, and traffic management. With a network of small satellites and land-based antennae, that system monitors the name, position, course, and speed of large oceangoing vessels and all passenger ships. By tweaking the data, researchers argue, that system could also help reduce shipping damage to marine life, from whales to coral reefs.

The need for such a system became vividly evident for lead author Martin Robards of the Wildlife Conservation Society just from watching shipping traffic on the perilous Alaska coast. “You see that ship leap forward, you see it go over the horizon eight miles away, and once it’s out of sight you really don’t know what it’s doing, even though it’s still close to shore,” he said. “Now we have a method for monitoring what’s going on, not just for the safety of the ship but also for its impacts on wildlife.”

Even in its present form, said Robards, the AIS is producing conservation benefits. For instance, researchers used it in 2012 to track ships approaching the Pacific end of the Panama Canal in winter. Then they mapped that data onto the movements of 15 satellite-tagged humpback whales, a species “at the top of the death toll” for ship-whale collisions. More than half the whales turned out to have had close encounters with ships over just an 11-day period.

That study led the International Maritime Organization to establish a traffic-separation scheme at both ends of the canal and a recommended maximum speed, in season, of 10 knots per hour. In the waters around Boston, said Robards, similar studies led to a new traffic pattern that changed vessel routing by a trivial amount “and cut the risk to whales by 81 percent. It was incredible how little they had to impact the shipping traffic to have a real conservation impact.”

With minor adjustments, AIS could do much more, he said. “For us, the data needs to be of a higher quality. What is the vessel carrying? Is it a hazardous substance? Where is it going? Is this vessel trading between Asia and the United States? Or is it just going up the U.S. coast?”

While the current system is heavily automated, it also has too many elements that need to be provided by a ship’s master of watch, “and whenever you have a human element, it can be a disaster.” Robards said. The Alaska port of Dutch Harbor, for instance, has 10 or 20 names, in various languages. “But a drop-down menu listing different ports would be a way around that problem. There is a lot more we could do without causing great inconvenience to the shippers.”

Better automation could also help detect illegal activities on the open ocean. Ships engaged in fishing have recognizable activity patterns, and the Pew Charitable Trust and SkyTruth are working with Google to automate pattern recognition and ship identification using AIS data. SkyTruth has used AIS data to bust a ship in a 92-mile-long bilge-dumping incident off the coast of West Africa.

But it’s not about demonizing the shipping industry, said Robards. In many cases, legitimate shippers want conservation-minded changes. “A lot of this is helping them be environmentally aware,” he said. “Nobody wants to be running into a whale if they can help it. Nobody wants to be coming into harbor with a whale on the bow of the ship. It happens now, and then they just have a public relations nightmare.”

Safety and conservation are also tightly linked, he said, pointing to a 2004 incident in which the Malaysian freighter Selendang Ayu waited too long to report an engine failure and ended up aground in Alaska’s Aleutian Islands. Not only did six crew members die during a rescue attempt, but the ship also spilled 350,000 gallons of heating oil and fuel and enough soybeans to devastate shoreline bird habitat.

Today, said Robards, AIS would immediately notice that a ship had stopped at sea or even that it was headed for the wrong pass on the Great Circle route through the Aleutians. “The ship automatically gets pinged: ‘Are you aware that you are entering a non-recommended pass?’ And you watch on the screen and see them do a U-turn. It’s a little like Big Brother is watching. But you can see that people are getting the message and doing the right thing.”

As the system improves, the promise is that it will make the oceans and coastlines of the world safer for everyone.