Interview: How to build a lionfish-killing robot
With well over a million infesting the waters of the eastern and southern coasts of North America, lionfish are slow-moving environmental disasters affecting ecosystems from Massachusetts to Mexico. Now Robots in the Service of the Environment (RISE) is working with partners to develop a lionfish-killing robot to help contain – or even reverse – the problem. New Atlas caught up with the executive director of RISE, John Rizzi, to find out more about this piscatorial Terminator.
According to NOAA, lionfish are a major environmental problem. Native to the Indo-Pacific region, the invasive species was unknown in American waters until about 25 years ago. No one is certain how they were introduced, but DNA analysis of lionfish caught in the Caribbean, Bahamas, and other places indicates that all the lionfish in the Americas are descended from only a few individuals. This indicates that they were probably the result of private collectors dumping their aquaria into the sea.
Lionfish have now spread as far north as New England and as far south as Venezuela. They're predators with no natural enemies, a voracious appetite that makes them eat almost half their own body weight every day, and a rapid reproductive cycle that sees them sexually mature in a year, with females spawning 30,000 eggs every four to five days. Worse, native fish aren't adapted to them, so the lionfish can simply swim up and eat just about anything it wants without raising an alarm.
The result is the concerning spread of a fish that, once it gets established, can reduce a reef's population by 64 percent in one year. Lionfish aren't very large, but they can strip an area of its small fish, decimate the young of commercial fish such as groupers and snappers, deprive larger fish of food, and reduce herbivore populations to the point where they can't keep down the algae that can smother a living reef.
In recent years, authorities in the United States and elsewhere have taken active measures against the lionfish. One of the most successful has been to encourage recreational divers to go on lionfish hunts with prizes going to the one with the biggest catch. The fact that lionfish are delicious is another incentive, with restaurants joining the fight by adding lionfish to their menus.
Unfortunately, such drives haven't been enough. There are only so many divers and there is a lot of ocean out there – much of it out of reach. To help with the hunt, RISE has joined with deep ocean research charity Nekton to test an underwater robot in Bermuda that's designed to hunt down and kill lionfish using a robotic claw that delivers an electric shock.
However, when we caught up with the executive director of RISE, John Rizzi, we found out that killing may not be the first thing on the robot's agenda.
New Atlas: Why a robot?
It is true, divers go after them and they've been successful as far as keeping reefs clear of lionfish, but there are problems with that. Recreational divers only go as deep as 50 to 80 ft (15 to 25 m). After that, you use up air more quickly and you face the need for decompression on your way back up. You will hear of divers going much deeper, but that becomes very technical and expensive.
Meanwhile, there are large populations of lionfish much deeper down, so you need a robot to do the job. You don't want a million dollar robot to do the job. You want something that can go down inexpensively, find the fish, kill them, and bring them back up again.
One other point is that at popular diving sites where commercial operators travel, you might find one or two lionfish. If you go to the less popular sites farther offshore or are a little harder to get to, then you'll find 30. Then there are loads of places no one ever goes – probably 90 percent of all the reefs. There simply aren't enough divers.
New Atlas: Are places like Bermuda likely to have pockets of deep-living lionfish?
Exactly. Such as with the Bahamas and places in Florida. Part of the reason we started in Bermuda is that it gets steep quickly. You can go out half a mile and be in 200 or 300 ft (60 to 90 m) of water. That's really convenient for us to test the equipment, but anywhere [in the sea] it can drop quickly to a hundred feet and even there you'll find a population.
New Atlas: Does the robot have a name? We've just been calling it the "lionfish killer."
We need a sexy name. If you can up with one, that would be great.
New Atlas: How far along is the robot? Is it still in the concept stage?
The people we've been working with have developed ROVs [remotely operated underwater vehicles] in the past, so the basic structure is coming from other equipment. We're not inventing anything very new.
There are three major components. There's the ROV, there's the kill mechanism, and the control system that makes it all work. The ROV and control systems are pretty well standard. They're not off the shelf by any means, but they've been built before, so we're assembling from our experience.
The ROV itself will be in the river on the [US] West Coast and the kill mechanism is being tested here on the East Coast. The two of them will come together in a few months and we are highly confident that the prototype will be able to hunt, kill, and capture lionfish.
New Atlas: Is it autonomous or a remote-controlled?
The first device will be remote controlled. It will be on a tether and we'll drive it from above.
New Atlas: It sounds like a mobile fish trap
Traps are different. Traps are static and they trap anything. People have been sweating this problem for years, Any environmentalist going after lionfish says "don't do any other harm." The challenge is to avoid by-catches. That's the common term in the fishing industry when you catch the wrong fish. An odd thing is that when fishermen are trapping other fish or even lobster, they're finding lionfish in their traps.
The problem is that lionfish generally don't go into traps. There's no known bait that attracts them, so catching them is more random than strategic and traps just don't work the way we want them to. That being said, more and more advanced traps are being developed and that may be the solution, but as of now, nothing yet has proven to really do the job.
In our case, we have a vision system on the device for moving up and giving the fish a shock. The truth is, all the other fish run away. The lionfish stays still. We're lucky in that little bit of biology that lionfish are static, so you can sneak up on them and shoot them, whereas no other fish is going to stay around. They swim off where the electric shock won't reach them, so no by-catch.
New Atlas: How does it collect them once stunned?
There's a pump that creates a current flow of water that slowly sucks the carcass into the cage.
New Atlas: How does the stunner work?
That is the most interesting and the newest part of the science. That's never been done before in this context, so we're testing different types of shock mechanisms, such as DC, pulse, AC, or something more like a Taser. We have to test them before the mechanism goes on the ROV because that affects the design of the power supply and the controls, so there might be tradeoffs for size and weight as to what might go on the ROV.
New Atlas: What do you see as the ideal version of the robot?
There may be well over a million lionfish [in American waters]. No one really knows, but it's a terrible infestation across tens of thousands of square miles of ocean, so we need a lot of devices for hunting the fish. So, our real goal is to build these so that they're inexpensive for someone to buy, and then they're safe, easy to operate, and reliable.
But also, we want to capture the fish alive because they're actually delicious and as all humanity has proven, the best way to wipe out a species is to consume it. We capture the fish to create a financial incentive for fish markets to want to buy the device, put it in the water wherever they are, and either benefit economically or at least break even. There's an economic cycle there that allows us to then, as a non-profit, build a revenue stream, become self-sustaining and build more robots, as well as advancing our cause and attacking other environmental issues.
New Atlas: Have you tried lionfish?
Yes. It's a very nice white fish. Not very different from haddock. Not as meaty as swordfish. It has a little bit of oil to it, but not as oily as a salmon or a mackerel – it's a really a nice, tender white fish. There are recipes and cookbooks on it now, you'll find some high-end restaurants with fancy chefs doing interesting things with it. We'll be having some cook-offs with it in Bermuda when the launch for the project becomes more formal.
What people are scared of about with lionfish are the poisonous spines. There are 18 poisonous spines on the fish, but the venom is only in the tips of the spines. You can cut them off with a pair of scissors and the fish is totally safe. There's nothing poisonous or venomous in the meat of the fish itself. But there's an educational process where you say that once you get the spines off, then the meat itself is terrific.