Skip to Content

Excavating the Deep

MIT engineers are developing technology to study shipwrecks located kilometers beneath the surface of the oceans.

David Mindell, PhD ‘96, was lying face down on a mattress in the belly of NR-1, the U.S. Navy’s only noncombatant nuclear submarine, 900 meters below the surface of the Mediterranean Sea. It was the summer of 1997, and he was on a deep-sea exploration with a group of researchers from Woods Hole Oceanographic Institution, MIT, and a handful of other research establishments. They were trying to determine whether ancient shipwrecks that had been discovered two years earlier were archaeologically valuable, and they were searching for other wrecks in the area. Mindell was staring at the seafloor through a seven-and-one-half-centimeter window on the bottom of the sub, waiting to see whether the sonar soundings the crew had wanted to pursue were natural outcroppings of the seafloor or another shipwreck.

The sub’s thallium iodide lights painted the seafloor a murky green. Suddenly, the scene below the window changed. The vessel was passing over two piles of ceramic jars, or amphorae, stacked in the now disintegrated hull of an ancient trading ship. It was, says Mindell, the most thrilling moment of his young career. On the basis of other ruins found in the area, he knew the wreck had to be at least 2,000 years old. As he watched the amphorae slip from view, he realized that those ancient jars had probably been resting in that exact spot on the sea bottom since before the birth of Jesus, the Middle Ages, the Renaissance, and the birth of modern science, and that he was the first person to have seen those jars since the ship had foundered so long ago. In fact, that wreck, which would be dubbed Skerki D-the fourth shipwreck to be discovered on the Skerki Bank between Sicily and Tunisia-was the largest ancient shipwreck found in deep water. The discovery of this Roman trading ship of the first-century b.c.e. and the work on the site during that expedition would mark the beginning of a new scientific field: deep-sea archaeology.

Since the discovery of the wreck of the Titanic in 1985, deep-sea exploration has caught the public’s interest. For engineers, oceanographers, archaeologists, and other scientists, finding such phantom ships is just the beginning of the quest to understand ancient cultures through the remains of the vessels. But the quest is no small feat. Doing painstaking archaeological work at depths of up to 6,000 meters presents the most challenging problems in ocean robotics today. Since the discovery of Skerki D, Mindell, an MIT professor of the history of technology and of engineering systems, has been working to solve some of those engineering problems, and in the process he has been laying the intellectual foundations of a new academic discipline. He has put together numerous seminars on deep-sea archaeology and organized two national conferences, all at MIT. He established the DeepArch Research Group at the Institute in 1998 and has attracted graduate students to work with him here. And since last year, the U.S. government has made federal grant money available for deep-sea archaeology. Mindell believes the Institute is the place for that research to flourish. “MIT is a place where there are a lot of engineers who like hard problems,” he says, “and there are plenty of hard problems in deep-sea archaeology.”

Innovative Engineering Solutions

Exploration of shipwrecks in the deep ocean is not new, but exploration is really just seeing a site, “planting the flag,” then going home. Field archaeology, on the other hand, is a precise science. For Mindell and the research team that discovered Skerki D, the fundamental questions are: How do you execute very precise and accurate scientific work thousands of meters below the ocean’s surface in a place you can never touch? And, How do you excavate those sites to exacting archaeological standards?

“One of the things that concerned me early on was there was too much technology driving the archaeology,” says Mindell. “This technology is big; it’s expensive; it’s impressive; it’s dazzling; and it’s very easy for people to get lost in the very momentum of it.” To make sure archaeological needs drive the engineering research, Mindell and his colleagues talk frequently with the archaeologists. “We spend a lot of time getting people in a room together and asking what kinds of questions archaeologists have and how they translate into an engineering requirement,” he says.

Mindell and members of his research lab are working on solutions for two of those requirements. They are developing a navigation system that allows an underwater research robot to take detailed pictures of a site, as well as a subbottom profiler that can penetrate the sediment around a ship to create a 3-D image of the entire wreck, including the portions that are buried.

During that expedition in 1997, Mindell’s navigation system, called “Exact,” was used in deep water to do a precision mapping of Skerki D. The system’s two high-frequency wireless transponders were positioned at either end of the shipwreck by a remotely operated vehicle. The transponders, which operate on the same principal as the Global Positioning System, guided the remotely operated vehicle in precise, parallel lines one meter apart over the ship while also mapping the site with digital and video cameras. About 200 of the still photographs these cameras snapped were later put together in a mosaic that showed the entire exposed surface of the wreck. To compensate for distortions in the photomosaic, the vehicle also produced an acoustic map. Together, the two mapping methods gave scientists a complete record of the site and a database that documented the location and size of each artifact.

This summer the next generation of the navigation system will be tested on expeditions to the Black Sea and to the wreck of the USS Monitor off the North Carolina coast. The new system uses digital signal processing akin to cell phone technology to produce stronger, more accurate signals and to improve the resolution of the maps made of the site.

In 1999 the DeepArch Lab tested its second engineering contribution, the subbottom profiler, on a shipwreck off the coast of Israel. The profiler uses ultrasound to “see” into the mud around the wreck. Subbottom profilers were not new to ocean exploration, but until Mindell came along, their beams were too broad for the precision work of archaeology. Mindell built a profiler that uses higher acoustic frequencies to create narrower beams and used it first to map the Tanit, an eighth-century b.c.e. Phoenician trading ship-the oldest deep-water shipwreck discovered to date-that rests 300 meters below the surface. Guided by Exact, the profiler produced a series of cross-sectional images of the Tanit, revealing artifacts buried in the mud. Eventually, as the navigation system becomes more precise and the distance shrinks between the computer-controlled track lines used for mapping, the profiler will be able to create 3-D images that allow “virtual excavation” of a site.

A photomosaic (top) of Skerki D shows the artifacts in place, and the colors in an acoustic map (bottom) indicate changes in depth across the site. (Images courtesy of woods hole oceanographic institution)

Working at Sea

Deep-sea archaeology is an expensive venture. Typical expeditions last two to three weeks and cost $40,000 to $50,000 a day. Researchers are, therefore, careful to choose areas that are most likely to yield valuable historical information. Areas that are historically interesting, logistically accessible, and promise a high probability of well-preserved sites rise to the top of the list. So far, most of the expeditions have been in the Black Sea and the Mediterranean.

Finding sites can be a tedious process. First, a side-scanning sonar trolls behind the research ship, searching a broad swath of the seafloor and measuring the height of objects that rise from the seabed. Next, a remotely operated vehicle, which is tethered to the ship by a fiber-optic cable, is lowered to examine areas of interest with a video camera. But discoveries are rare. “It’s like watching hours and hours of bad TV locked up in a loud steel box,” says Brendan Foley, a doctoral candidate in the DeepArch Research Group. He was bitten by the deep-sea archaeology bug on the 1997 Skerki expedition. “It’s boredom, boredom, boredom, and then something really exciting happens.”

And when it does, the remotely operated vehicle provides real-time feedback to the ship’s monitors, giving the team the opportunity to observe the ruins and discuss what they are seeing. On some expeditions, three-person submersibles motor off to verify a site while the rest of the team waits aboard ship for the scientists to return and share their discoveries.

In 2001, Mindell’s research team tested an autonomous underwater vehicle on an expedition in the Aegean Sea. Although such vehicles are not yet designed for the precise work of archaeological sites, Mindell believes they have great potential for the future. It is possible to program them to negotiate variable terrain-a challenge tethered vehicles meet with limited success-and they could provide more stability for high-resolution side-scan sonars than is available from a towed platform. Finally, because they don’t require fiber-optic cables or ships to tow them, autonomous vehicles could cut expedition costs by about 75 percent, says Mindell.

The research group is focusing on technology that can search the deep, but Brian Bingham, another doctoral candidate in the group, points out that the technology can be used at any depth especially if visibility is limited. “It’s really about the technology not the depth,” he says. “It’s about using remote sensing to deliver that precision for interpretation.”

No matter how this new field develops, MIT scientists and engineers will be at the head of the pack. Robert Ballard, who discovered the wreck of the Titanic and is one of Mindell’s research colleagues, commends Mindell on his leadership in helping to create a new academic discipline. “He fully understands all aspects of the challenge we face before deep-sea archaeology is mainstreamed within the academic community, and he is doing all within his power to ensure this challenge will be met.”

The sea and the intersection of his two loves-history and engineering-fascinate Mindell. “Most of the shipwrecks out there aren’t findable with today’s technology,” he says. “There’s a lot to be done in the future that we just can’t do today.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.