A remotely controlled sailboat that maps the ocean floor, collecting DNA samples and transmitting real-time data to scientists sounds like the stuff of science fiction. But it’s real and it’s actually pretty close to home.
Last month, Saildrone — an Alameda-based company that designs and builds unmanned surface vehicles — launched their first 72-foot vessel called the Saildrone Surveyor into the San Francisco Bay. Meant to replace the large ships that historically have been used for ocean mapping expeditions, the Surveyor is equipped with advanced sonar equipment that will allow it to map the ocean down to 7,000 meters (over 4 miles) beneath the surface and is expected to operate remotely on missions of six to 12 months.
“It’s kind of like a mission to Mars, right out in our Bay,” said Jim Birch, a researcher at the Monterey Bay Aquarium Research Institute who has been part of MBARI’s collaboration on the Surveyor’s development.
The Surveyor was developed in collaboration with researchers at MBARI and the University of New Hampshire through a grant from the National Oceanographic Partnership Program, which is sponsored by the National Ocean and Atmospheric Administration’s Office of Ocean Exploration and Research.
The vessel is going through some final engineering tests, but once it’s commissioned in a few weeks, it will set off on its maiden voyage to Hawaii and back, where it will map the ocean and collect DNA samples en route, in an unprecedentedly environmentally-friendly way.
“Ships are very, very dirty things,” said Richard Jenkins, founder and CEO of Saildrone. He estimates that a medium-sized ship burns between $2,000 and $5,000 a day in fuel.
By contrast, saildrones are completely wind and solar-powered. And because they’re silent, they don’t create noise pollution that can harm marine animals. Jenkins also said the saildrones are equipped with protocols to keep them from operating the sonar around marine mammals, many of which also use sonar to communicate.
Saildrones are also much cheaper than chartering a ship for research — which can cost tens of thousands of dollars per day — because Saildrone doesn’t sell their vessels, only the data the vessels collect. “It really dramatically reduces the cost of ocean exploration,” said Jenkins.
According to Birch, the reduced cost tremendously lowers the bar of getting into the field, which could help open it up to groups that have traditionally been excluded from such research efforts because of lack of funding.
“You don’t need to write a $3 million NSF grant. Maybe for $70,000, you can collect data that answers or gets at a question you might have,” said Birch. “It could really be a game-changer.”
‘Worse things happen at sea’
It hasn’t all been smooth sailing. Jenkins said one of the biggest challenges of designing the Surveyor was “cramming a lot of things that are normally distributed in a very large ship into a very small vehicle.”
Not only did all the sensors and machinery have to be incredibly condensed, but they had to be built to withstand the harsh environment of the open ocean.
“There’s a saying that says ‘worse things happen at sea,’” said Jenkins. And between the wind and waves and storms and hurricanes and corrosion caused by high salinity, he said it’s one of the toughest places in the world to operate, even with a crew supervising the equipment.
“On a ship, if something fails, you just go down and flick a switch or turn the computer on and off again,” said Jenkins. But on an uncrewed vehicle, there’s no one aboard to wiggle ethernet cables or plug wires back in. “So everything has to be incredibly robust and bulletproof,” he said.
The Surveyor’s hardware isn’t the only thing that needs careful engineering. Larry Mayer, director for the Center of Coastal and Ocean Mapping at the University of New Hampshire, and his group have been working to make sure the Surveyor’s software is also up to snuff.
The sonar collects millions of data points per hour, but it can’t transmit every point to researchers in real time because of limited satellite bandwidth. So Mayer’s group is studying the best ways for the Surveyor to condense or compress the information it sends — through tricks like combining similar measurements or leaving out glitchy data points.
They are also using machine learning to help the Surveyor sense and respond to its surroundings when it’s out at sea.
“We have to make sure the sonar recognizes if there’s a hazard coming up,” said Mayer. To do this, his team is working to integrate all the sensors and camera equipment with machine learning software. They want to see how well the Surveyor can distinguish between different sorts of objects in the water and if it can sense the speed and direction of other moving objects to avoid collisions.
‘Why can’t we bring the lab to the ocean?’
Acoustic sensing gives a lot of information, including giving rough estimates of how many animals are living in a given area (“You basically get reflections of the sound off fish or critters,” said Jenkins), but it can’t distinguish between different sorts of organisms.
That’s why Saildrone has been partnering with researchers at MBARI to equip the Surveyor with technology to collect “environmental DNA,” that is DNA that sloughs off from marine organisms and ends up floating in the water. Analyzing eDNA gives researchers a way to monitor diversity within an ocean habitat without needing to physically catch and count fish.
For years, researchers at MBARI have been doing different kinds of analysis on water samples, which used to involve collecting water in buckets and bringing the water back to shore to be filtered and then analyzed. Birch remembers Chris Scholin, now president and CEO of MBARI, being frustrated at how inefficient the process was. “Why can’t we bring the lab to the ocean?” Birch recalls him saying.
So Scholin and Birch started working on a device called the environmental sample processor, or ESP. Now in its third generation, the ESP is shaped like a cylinder about a foot in diameter and 2 feet long. The processor pumps water through a tiny filter, which catches tiny bits of debris floating in the water.
But the debris on the filter isn’t junk — it’s loaded with eDNA. The ESP stores the filter in a chemical preservative that locks all the genetic material in place, so it can be recovered and analyzed months later when the vessel returns.
Why map the ocean at all?
Despite oceans covering over 70% of the Earth’s surface, “we know more about the surface of Mars than we know about the deep ocean,” said Birch. Currently, about 19% of Earth’s ocean floors are mapped.
But the ocean needs to be mapped for more than just knowledge for knowledge’s sake.
According to Mayer, climate models are totally dependent on how heat is distributed through atmospheric and ocean currents — and how currents move depends on the shape of the ocean floor. Predicting hazards like tsunamis also require accurate maps of the sea floor.
Good ocean maps are also critical for maritime safety and navigation, said Mayer, recalling the 2005 incident when the USS San Francisco hit an uncharted underwater mountain, killing one crew member and seriously injuring nearly 100 others. More complete maps will help prevent disasters like this.
Ocean mapping is also essential for discovering and protecting natural resources like mineral deposits and delicate ecosystems, as well as accessing millennia of cultural legacies in the form of undiscovered shipwrecks.
“Who knows what the next great discovery is going to be?” asked Mayer. With 81% of the ocean still unmapped, “you’ve got good odds there’s going to be something there.”
