At the ARPA-E Energy Innovation Summit back In 2017, we met a company called Marine BioEnergy that was exploring a concept involving robotic submarines farming the open ocean for kelp to create carbon-neutral biofuel. The concept had a lot going for it: Kelp sucks up carbon as it grows, so any carbon that it later releases into the atmosphere is balanced out as new plants take root. What’s more, kelp can be turned into energy-dense liquid fuel, for which there is already a massive distribution infrastructure. And most importantly, kelp grows in the ocean, meaning that we wouldn’t have to fertilize it, give it fresh water, or let it compete for land space like wind and solar farms do.
The tricky bit with kelp farming is that kelp needs three things to grow: sunlight, nutrients, and something to hold onto. This combination can only be found naturally along coastlines, placing severe limitations on how much kelp you’d be able to farm. But Marine BioEnergy’s idea is to farm kelp out in the open ocean instead, using robot submarines to cycle the kelp from daytime sunlight to nighttime nutrient-rich water hundreds of meters beneath the surface. Whether this depth cycling would actually work with kelp was the big open question, but some recent experiments have put that question to rest.
Kelp doesn’t naturally depth-cycle itself. On its own, kelp will pick some nice rock in a shallow bit of coast, stick itself there, and grow straight upwards towards the sunlight. In order to keep itself vertical, the kelp produces floaty gas-filled bladders called pneumatocysts at the base of each leaf. Unfortunately, things that are filled with gas tend to implode when they descend deeper into the water. Nobody knew what would happen if kelp were to be grown while depth-cycling it; would those pneumatocysts even be able to form, and if not, what would that do to the rest of the plant?
To figure this out, Marine BioEnergy partnered with the USC Wrigley Institute for Environmental Studies on Santa Catalina Island, off the coast of California, to depth-cycle some baby kelp. Rather than using robot submarines, they instead put together a kelp elevator, consisting of an automated winch tethered to the seafloor. Attached to the winch was a scaffold that supported lots of little baby kelp plants. Every evening, the elevator lowered them 80 meters down into nutrient-rich waters to feed. In the morning, the whole contraption was winched back up into the sunlight.
After 100 days and nights of winching up and down, the testing showed the kelp had adapted to its depth cycling and was growing rapidly, as President of Marine BioEnergy Cindy Wilcox described to us in an email.
“As it turns out, the depth-cycled bladders were long and narrow and filled with a liquid, not gas. For the first time, this showed that at least one species of kelp (macrocystis, otherwise known as Giant Kelp) thrives when depth-cycled between sunlight at the surface in the daytime and submerged to the nutrients below the thermocline at night.”
The depth-cycled kelp produced about four times the biomass of a control group of kelp that was not depth-cycled, and although the experiment ended at 100 days, the kelp wasn’t even full grown at that point. Seeing exactly how big the mature kelp gets, and how quickly, will be the next phase of the experiment.
Ultimately, the idea is to disconnect production of kelp from the shore, using solar-powered robot submarines to depth-cycle giant rafts of kelp out in the open ocean. Every 90 days, the kelp (which grows continuously) would get trimmed, bagged, and delivered to a pickup point to get converted into biofuel, while the robot subs drag the freshly shorn kelp back out to start the cycle over again.
Diagram showing the life cycle of an ocean kelp farm.Image: Marine BioEnergy
The actual conversion of kelp into fuel happens through existing commercial processes, either hydrothermal liquefaction or anerobic digestion. About half the carbon in the kelp can be processed into gasoline or heating oil equivalents, while the other half is processed into methane that can be used to power the conversion process itself, or converted into hydrogen, or just sold off as a separate product. Since the carbon being released in this process is coming from the kelp itself, it’s not actually adding any carbon to the atmosphere, as Wilcox explains:
Our projections are that the kelp grown per drone submarine, over its 30-year life, is about 12,000 dry metric tons of biomass, which is over 200 times the mass of the drones and farm system. The energy contained in this biomass is over 160 times as great as that required to make and operate the drone and all associated farm equipment, including deployment and harvesting. When fuel from the kelp is burned, it releases CO2 that was absorbed from the environment only a few months before, and the carbon footprint of the farm itself is relatively minor since its mass is so small compared to the product. The vision is that, eventually, kelp-derived energy and organic feedstocks would provide all inputs for the relatively small mass of farm equipment and so no fossil fuels would be needed to sustain and grow the system beyond that point.
Replacing all liquid transportation fossil fuels used in the United States, Wilcox says, would require farming about 2.2 million square kilometers of kelp, representing less than 1.5% of the area of the Pacific. It may be a small percentage, but that’s still a lot of kelp, and some concerns have been raised about what effect that could have on other ocean life. According to Wilcox, the thermohaline circulation generates about 3.5 meters of nutrient upwelling across the entire ocean every year, and kelp farming would only suck up the nutrients in about 6 cm of that upwelling. Interestingly, by producing fertilizer as a biofuel byproduct, kelp could also be used to help bring deep-ocean nutrients back to land, a process that (as far as we know) currently only happens through volcanoes and salmon. “We expect that the main effect of the ocean farms will be to help reduce the damage from the human-caused flood of artificial nutrients that are making their way into the ocean,” Wilcox says, “but this needs more study.”
Over the next few years, Marine BioEnergy hopes to use funding from ARPA-E to prototype farm implements and perform large-scale ocean testing, after which the goal is to build the first farm and start producing kelp at scale.
Evan Ackerman is a senior editor at IEEE Spectrum. Since 2007, he has written over 6,000 articles on robotics and technology. He has a degree in Martian geology and is excellent at playing bagpipes.