Counting Fish TRANSCRIPT

This is a transcript of the Gastropod episode Counting Fish, first released on October 4, 2016. It is provided as a courtesy and may contain errors.

HOFFMAN: They call this playing the drums when this happens. We’re playing the drums right now.

TWILLEY: No, Cynthia and I have not started a rock band. Although, you know, if this whole podcast thing doesn’t work out…

GRABER: At least we have a fallback plan. The drums in this case are the sound of chains of a fishing trawl net hitting the side of the boat. This episode, we’re tackling something that sounds like a book you read to little kids.

TWILLEY: One fish, two fish, red fish, blue fish, you mean?

GRABER: Exactly. Or, in this case, counting fish. You’re listening to Gastropod, the podcast that looks at food through the lens of science and history, I’m Cynthia Graber.

TWILLEY: And I’m Nicola Twilley. And this week, we are taking on one of the universe’s great mysteries: how many fish are in sea? And are there really plenty more? Because it’s rather a big deal.

GRABER: And if you stop to think about it, it seems almost impossible to figure out how many fish there are, but scientists do give us numbers. Where do those numbers come from?

TWILLEY: And how accurate are they? The fate of fish *and* fishermen—they both rely on us getting those numbers right. And as it turns out, counting fish is undergoing something of a revolution right now, with drones and robotic submarines and all kinds of new high tech gear.

GRABER: So let’s go count some fish.



TWILLEY: We got up at the actual literal crack of dawn to go out for a day of fish counting with Bill Hoffman of the Massachusetts Department of Fisheries.

HOFFMAN: Right. So this is a fishing vessel Miss Emily based out of Scituate, Massachusetts. Kevin Norton is the captain. Talking about sorting the fish into the buckets

GRABER: We joined two scientists and three fishermen out there on Cape Cod Bay. They planned to be out on the water for maybe as long as twenty hours to pull up a handful of trawls—basically they’re dragging the net through the water for a half hour or so at different spots in the bay.

TWILLEY: And those sounds that you hear—that’s Bill and his colleague Nick sorting and counting and weighing and measuring the fish we caught. By the end of the day, Cynthia and I put on rubber gloves and were learning how to do it too.

HOFFMAN: Now all the fish are entered into the system. We have all the weights and now we’re gonna start the biological sampling.

GRABER: First the crew sort all the fish into buckets and weigh them by species. Then Nick slides each fish one by one down the table to Bill.

HOFFMAN: I took a length of this. You see what I’m doing, I’m measuring, right, we’ll take a fork length on this fish. So I hold it right here, the base of the tail.

TWILLEY: Bill’s got the fish on this cool magnetic ruler, and he lines up its nose on the left and then he uses a magnet to stamp the spot where the fish’s tail forks. The board then bleeps so that he know it has registered the length and sends it to the onboard computer..

GRABER: And then Nick slides another fish down to Bill—each fish takes only a couple of seconds. They’ve got this whole thing down to a machine-like clip. Slide, measure, slide, measure. When we tried, it was kind of tricky to hold onto the slippery fish with those big rubber gloves. But Bill just grabs hold of each one, no problem.

TWILLEY: And after he’s measured it, he uses one hand to send it sailing into an orange bucket nearby. He doesn’t even need to look. For some of the species, Bill has to do a little more sampling—maybe take out a cod’s ear bone to see how old it is. Or if it’s an species that isn’t usually found in Massachusetts Bay, he might bring the whole fish back to shore for scientists to check out. But most of them just go back over the edge and make a lovely breakfast for the seagulls.

GRABER: But not all. Because this is a partnership between scientists and commercial fishermen, some of the fish that are big enough to be sold on the market get packed up and brought back to shore. The money is used to fund more research.

TWILLEY: Our boat trip is just part of a much larger national effort—actually, an international effort—to count the fish in the ocean.

GRABER: In the U.S., the government trawls at certain spots each year, year after year. They count the fish they pull up. And they use those numbers to extrapolate out –

TWILLEY: They know how many fish they found in the spots they trawled, they know how much of the ocean they sampled, and they put that together with information about fish biology to arrive at an estimate of how many fish are out there total.

GRABER: The U.S. government, actually the National Oceanic and Atmospheric Administration, or NOAA, they’ve been doing this trawl survey basically the same way since the 1960s. But already by the 1960s, scientists knew things were not going so well for fish. In the mid 1900s, boats got more and more powerful. Fishermen pulled more and more fish out of the water. Fish stocks were crashing.

TWILLEY: The World Wildlife Fund estimates that some of the most commercially valuable species have been reduced by up to 75 percent since the 1970s. That kind of declinethat could lead to entire species being wiped out, and then who knows what kind of ecosystem repercussions. But we haven’t just stood by and watched fish numbers fall off a cliff.

GRABER: Starting in the 1970s and until today, the U.S. government has been repeatedly limiting the amount of fish that fishermen can catch. The cuts started off small, but they got sharper and sharper.

TWILLEY: So like Kevin—the captain of the Miss Emily—he’s been fishing for 20 years now. And he was still doing okay up till 2010.

NORTON: And then the cuts started. First year the cuts started and it hurt, I made less money. And the next year they cut again. And then I think it was the next year, they wiped it out. They started with their 80 percent reductions and I mean, you’ll see today, you guys are out today, nobody’s fishing. They’ve put everybody out of business. At what point are they gonna let a few of us go back to work? That’s the worst part of this is a lot of these guys that work their whole lives just get put out of business. There’s no retirement, there’s no, they weren’t able to sell their boats and permits and retire because by the time they’ve gone to sell their boats and permits, they’re worthless. I mean, these permits were worth a half million dollars a few years ago. Now they’re—I don’t know if you could sell one.

TWILLEY: That’s another reason counting fish matters. It’s not just about the future of fish. Fishermen are going out of business, they’re going bankrupt. And that affects their lives and families but also the entire coastal community.

GRABER: And frankly, Kevin thinks the most recent cuts are far too dramatic. Because he’s seeing plenty of fish out there.

NORTON: There was a year where we had a warm year, a warm winter, a warm year. There wasn’t much cod around for a year. But then, of course, the following spring we started seeing cod everywhere. And it’s not like the stuff it disappeared, it was just a slow few months. And, you know, and I felt like doing a spring and a fall survey wasn’t enough. They’re not getting a true picture of what goes on in the Gulf of Maine.

GRABER: Here is where the battle lines have been drawn. Fishermen and scientists over the past few decades have been literally yelling at each other in public meetings.

TWILLEY: Fishermen like Kevin are seeing lots of fish, and they think the government is counting in the wrong places and at the wrong times. The government says of course Kevin and his fellow fishermen are seeing plenty of fish—because they go to where the fish are, that’s what fishermen are good at. And everyone is angry.

GRABER: New England is one of the biggest battlegrounds of all. It’s a major fishery. If you live in the U.S., you’ve probably eaten fish from New England: lobster, clams, cod, haddock. New England’s towns were built on money from fish. There’s a sacred wooden cod that hangs in the state house. Cape Cod is, yes, named for the region’s iconic fish.

TWILLEY: And, at the same time, it’s a major center for fisheries science. Out on Cape Cod, you have NOAA but you also have Woods Hole Oceanographic Institute, and then there are researchers at all the New England universities. People really care about fish out here.

GRABER: They all want to know the same thing: how many fish are out there. So who’s right, the fishermen or the scientists? And why is it so hard to figure out? Andy Pershing is chief scientific officer at the Gulf of Maine Research Institute.

PERSHING: It’s really hard, I mean, the hardest part about counting fish is the ocean. I mean, if we didn’t have the ocean it would be easy to count fish, because they would just be right there and you could see them.

TWILLEY: In fact, every single scientist gave us the same analogy to help us understand why it’s so difficult.

LEGAULT: There’s a great quote that goes something along the lines of “counting fish is like counting trees except the trees are invisible and they move around.” So it’s very difficult because we can’t actually see most of the fish that we’re trying to count. And the ocean is a big place. And so we can’t go out and count individually each individual fish, and so we have to get at it indirectly in order to figure out how many fish are out there.

TWILLEY: We have at least a dozen different scientists on tape saying exactly that: that counting fish is like counting trees, except imagine the trees are invisible and moving. This one is Chris Legault. He’s a fisheries scientist at NOAA out on Cape Cod.

GRABER: And so counting fish is hard—but despite Kevin’s frustration, the U.S. is actually sort of a role model in this field.

LEGAULT: I think the U.S. has made a very strong commitment to having science based management. And I think it’s been a leader around the world.

GRABER: Most other countries don’t have an entire governmental scientific agency dedicated to this question. Instead they rely on data from the fishermen themselves—it’s called catch per unit effort. Basically, how hard is it for fishermen to catch a certain amount of fish. All of that data relies on what fishermen report. But Chris at NOAA says that doesn’t actually reflect the number of fish out there.

LEGAULT: As the fishing gear changes, technology, GPS, this fish finder technology, improves their ability to catch fish keeps increasing over time. And so they can maintain their catch rates even as the population declines.

TWILLEY: And so fisherman still catch the same amount of fish with the same amount of time and effort, but it’s because of better fishing gear, not because there are the same number of fish in the sea. And the other problem with relying on catch per unit effort is that it’s not scientific. The fishermen know where the fish are, so they go there. That’s a pretty biassed sample. NOAA’s trawls are random samples, so they get a better picture of the sea as a whole.

GRABER: As Chris says, the U.S. has better data than most of the rest of the world. But, here’s one problem with it: those surveys only go back to the 1960s. And, as we said, humans had been taking increasingly large amounts of fish out of the oceans for a century or more before that.

POUL HOLM: So the normality of the present day is actually a historical error. It’s an indication of how badly we have treated the oceans in the last couple of hundred years. And that’s the problem that I’m often faced with when I talk to the fisheries managers. They will say, well, actually management policies are a success. We have been really good at rebuilding stocks in a number of cases. And I would put it to them that, well, perhaps your goals are simply too limited.

TWILLEY: That’s Poul Holm, he’s a professor of environmental history at Trinity College Dublin.

HOLM: We know as a rule of thumb that in the last hundred two hundred years we have basically eradicated nine-tenths of the biomass of the large fish and marine mammals.

GRABER: Nine-tenths, that sounds like an insane amount. But how do we know? We weren’t doing scientific fish surveys over the hundreds—thousands—of years we’ve been fishing. But we do have some sources, like fishermen’s logbooks.

HOLM: There was very little incentive, for example, the fish captains to doctor their logbooks. They kept the logbooks in order to be able to measure their own success, revisit the same places year after year and basically keep a record for their own use. So for example in the Gulf of Maine we had a fantastic study by Andy Rosenberg’s team which showed that the level of accuracy in these logbooks was simply astounding.

GRABER: Poul mentioned Andrew Rosenberg. He’s director of the Center for Science and Democracy at the Union of Concerned Scientists and he spent years as the Northeast Regional Administrator of NOAA’s Fisheries division.

TWILLEY: So Andrew and his colleagues went through all these log books and entered all the historical cod catch data into a statistical model based on what we know today about cod biology.

ROSENBERG: And what it showed us was that if you calculate back in time, there had to be an incredibly large abundance of cod on the Scotian shelf and in the Gulf of Maine because of what actually the fishing boats brought home. The only way that catch was feasible was if there was a very, very high abundance of cod. And when I say very high abundance I mean, you know, towards a million metric tons whereas now we’re talking about, you know, tens of thousands of metric tons. So orders of magnitude smaller now compared to what that historical abundance was.

TWILLEY: In fact, Andrew and his team showed that nine-tenths of the cod in the Gulf of Maine is gone. It’s like Poul Holm said, that nine-tenths reduction in large fish holds true almost everywhere.

GRABER: And they’re not just mining log books for historical data. Poul has colleagues around the world who are looking into archaeological sites and doing genetic studies on fish bones. They’re collecting restaurant menus to see what fish were common and cheap. All of these help paint a clearer picture of how many fish there used to be.

HOLM: You cannot really assess the health of any patient without knowing what was his or her condition prior to the present state. That’s true for the oceans as well. We need to know how much have we taken out from the sea in order to know what might actually be regained by building up fish stocks. So we need historical baselines in order to help managers get a sense of what is the potential ocean productivity.

TWILLEY: One reason that figuring out these historical baselines matters is that it helps us separate out how much of this decline in fish numbers is because of humans, and how much is a result of natural cycles. And within that human footprint, how much is down to fishermen catching too many fish, and how much should we blame on other ecosystem changes—pollution, building dams, all the stuff we humans do.

GRABER: Studies like Andrew’s help us tease this out. But they also give a sense of how productive the oceans could be. And that can help everyone make better management decisions.

ROSENBERG: So when we have these arguments as I used to have as a manager, when people would say, oh no, it’s impossible, we’d never see that abundance, there’s no point in regulating fisheries to a point because you have some fictitious idea of how productive the ocean is. Well, actually, it’s not a fictitious idea. We’ve seen historical examples of very much higher productivity, when we haven’t knocked down so many of the populations and the habitat to such a low level. Now it doesn’t mean that we’re going to rebuild to millions of metric tons. But it does mean that the potential productivity is very much higher than we normally think of it in the day-to-day battles over what should the fishery fishing rules be.

GRABER: So that’s one problem with NOAA’s surveys. If you’re using them to look at trends, they’re kind of blind to anything that happened before the 1960s. But there are other problems, too.

TWILLEY: The trawl doesn’t work for all fish. Some species are known to avoid the nets. Some, like tuna, cover huge distances. And some parts of the ocean just can’t be trawled.

ELIZABETH CLARKE: Trawls don’t get into rocky habitat. They hang up. You’re going to lose the trawl. you can’t get up to the ledge. So there is some habitat that’s not well surveyed.

TWILLEY: That’s Elizabeth Clarke. Like Chris, she’s a fisheries scientist with NOAA. But she’s based on the West Coast where there’s a lot of these rocky bottoms. Still, the biggest problem with these NOAA trawls is also their biggest strength: NOAA has been doing them pretty much the exact same way since the 1960s. That consistency is great in terms of seeing patterns over time, but really, is 1960s era technology the best we can do?

CLARE: It’s old technology. But there are also new technologies that can help us in a lot of ways.

GRABER: Today, Liz heads NOAA’s research project on robotic submarines. They’re actually called autonomous underwater vehicles, or AUVs. This project got started about a decade ago almost by accident.

HANUMAT SINGH: So there’s an interesting story there. You know I’m not a fisheries biologist I don’t know anything about fish. And there was a conference once and at the very last moment they called me up and said “Oh, Hanu, there’s a conference here on sensors and fisheries. Can you come and give a talk?” And you know at the last moment. “Sure.”

GRABER: Hanumant Singh is an ocean engineer with the Woods Hole Oceanographic Institute. He’d already designed a robot that could glide, untethered, underwater, but he had no plan to use it to count fish.

SINGH: You know I walk up there I give a talk. And at the end of the talk one of the scientists there came up and talk to me. Her name was Elizabeth Clarke.

CLARKE: And I thought, “Wow, that is very interesting.” He was using it a lot for archaeological work and he was using it to do some mapping of shipwrecks and things like this. And I thought, “Hmmm.”

SINGH: And so we started talking and then we did a couple of pilot programs with her you know over the course of three years

CLARKE: And so we went in together to start using an AUV that we were tweaking for purposes of trying to survey fish in their habitat.

TWILLEY: So Cynthia and I went to visit Hanu’s robotic submarine in its shed at Wood’s Hole.

SINGH: So this is the robot.

TWILLEY: It’s basically a six-foot, neon yellow torpedo. With a bunch of high tech gear built in.

SINGH: There’s sensors to measure currents. There’s cameras, there’s sonars, all that is sitting on the bottom.

GRABER: When Hanu and Liz take the robot out for a spin, they chart a path for it ahead of time. Then they put it in the water, there’s a small splash, and then it’s off.

SINGH: And, you know, then we watch it go down until it disappears into the deep. We’re holding our breath continuously, we don’t breathe for six hours. So we watch it go down and and work, work deep.

TWILLEY: The whole point is that the sub is autonomous—it’s following a preset path but it’s responsible for avoiding all the boulders and cliffs on its own. All Hanu and Liz can do is listen to it beep.

GRABER: And what are those pings?

SINGH: Those are those pings for location. So we send out a little tone at a particular frequency, the vehicle hears it and responds back. By calculating how long it took to get to the vehicle and back, we can figure out what the range to the vehicle is. And by using multiple of these pings we can triangulate to figure out the position.

GRABER: We pictured it speeding along the ocean floor, James Bond style. But actually…

SINGH: It doesn’t go fast. A big question to, a good question to ask is how slow can go, okay? And the great thing about that is it can come to a complete stop, and it can also go at speeds of about 30 centimeters per second which is a little bit like walking.

GRABER: And as the sub strolls along the ocean floor it’s using sonar to map the seafloor and the fish and it’s repeatedly snapping overlapping photos. Thousands and thousands of them.

CLARKE: And so you get a lot of photos. And you need to analyze how many fish or what the habitat is or the size of the fish or the spatial relationship between the fish and let’s say a coral or a sponge. Once you collect the photos, that’s just the start.

SINGH: Yes, of course you know the first time you go out and, you know, when we started doing this, you know, the first reaction is God, was all fantastic, “These images are better than we’ve ever seen.” And they’re in color and they’re a fantastic resolution and the next reaction was, “Oh, what are we going to do with all this data,” right? And typically there would be a biology graduate student or some student standing there you know doing all these annotations and they would take, you know, literally half an hour per image. But when you have 10,000 of those that adds up really fast. And so we’ve gone very rapidly from an area where we had very little data to collecting lots and lots of data. And now the big push is to go from that data to information. Okay, so they don’t really want to collect the images per se. What they want to collect is how many fish they are. What is their distribution? What are the different species? And so we’re working on techniques to actually automatically segment them out figure out what the different fish are, and of course none of these techniques are perfect.

TWILLEY: Hanu and his colleagues are trying to automate some of this using artificial intelligence. But it’s a hard problem. Fish that live at the bottom of the ocean tend to look like the bottom of the ocean.

GRABER: The team is taking out the AUV repeatedly over a few months this fall off the coast of San Diego. The robot is great for untrawlable habitats like the ones off California, it could help count fish that live in areas where the trawls can’t go because they’ll get caught on the rocky bottom. So the robotic submarine solves one problem. But what about species that migrate all over the world like tuna?


TWILLEY: Tuna can’t be surveyed by the NOAA trawl, so fishery managers rely on the numbers that fisherman catch. Those numbers are flawed, like we said. And the other problem is that tuna are highly migratory, they can range all over the Atlantic. And if New England fishermen aren’t catching them because the tuna are off Canada, instead, then, according to the data, there are no tuna.

GRABER: Molly Lutcavage is a biologist at the University of Massachusetts. She hadn’t been focusing on bluefin tuna a couple of decades ago, when the commercial fishermen first reached out to her.

LUTCAVAGE: And how we got into it is that at the time, this was 1992-93, stock assessments were representing that there were only tens of thousands of adult Bluefin existing in the Western population. And the issue was that our New England fishery, commercial fisherman and spotter planes that worked with them, said that they could see that many fish in one or two schools that were at the surface in the Gulf of Maine. And so they greatly disputed what was being represented to the Western population. But no one would listen to them. And they said, “Come fly with us, come look, and we’ll show you that this just isn’t right.” And within a matter of weeks of having our equipment on spotter planes, we learned that the fishermen and the spotter pilots were absolutely right, that there were tens of thousands of bluefin just at the surface in a small area of the Gulf of Maine and the Gulf of Maine is only a portion of the bluefin’s range in the Western Atlantic.

TWILLEY: Basically, the numbers that the fishery managers were using were just completely wrong. And the spotter pilots knew that.

GRABER: If you’ve never heard of spotter pilots, and I hadn’t, they fly out to find the schools of tuna as the huge fish surface. And then they let boat captains know where to find them. Tuna cover a lot of ground. They swim around the world, and they swim incredibly quickly. Without spotter pilots, fishermen would have a hard time finding them.

TWILLEY: So Molly started to wonder whether aerial photography from spotter planes might be a better, new way to count tuna. Initially Molly worked with spotter pilots and just put a camera on the airplane.

MIKE JECH: Initially Molly worked with spotter pilots and just put a camera on the airplane. But we had a lot of issues with that. The data, they give great images but when you actually start to get measurements from them, you don’t have accuracy and precision that you need.

GRABER: Mike Jech is a fisheries biologist at NOAA on Woods Hole. He’s working with Molly to try to count tuna. Spotter pilots are great for finding the fish, but the measurements aren’t scientific enough for NOAA to use. So Mike had another idea.

JECH: So we had to try to come up with names for them, and Wasabi was the first one. Tuna sushi—ginger and wasabi. So things to eat with tuna. So that’s we that’s how they got their names.

TWILLEY: Wasabi and Ginger are drones. Ginger is a little more reliable, as it happens, so that’s the one we flew.

JECH: It’s a little half-bowl shaped that has six arms sticking out of it and at the end of each arm is a rotor, like a helicopter rotor. And then it has two little legs, kind of, that that are underneath it to help keep it up and on those legs are your swimming spaghettis that give your little kids in the swimming pool. Noodles, there we go. And then those are made for, if by chance it does go in the water, it’ll float. It has a little small camera, digital camera underneath it. Just like your little pocket Olympus cameras with a 25 millimeter lens. So it weighs less than five pounds so it’s maybe 18 inches in diameter or so—16, 18 inches in diameter, maybe a foot tall. So it’s quite small, quite portable.

GRABER: Mike and his colleague Jennifer took Ginger out on the dock behind the NOAA building on Cape Cod. They spread out a big mat that looks like black and white keys on a piano so that the drone could calibrate its camera. And then Mike put on a helmet. He lifted the drone above his head. Jennifer set the hexacopter blades whirring. Mike let go, and it took off.

JENNNIFER: Okay, so rotate the bird around the pitch access until the clicking stops.

GRABER: And then Mike lifted up the helmet, he put it on his head. Jennifer set the hexicopter blades whirring. Mike let go and it took off.

TWILLEY: Once the drone was up in the air, Jennifer was driving it using a joystick, and Mike got under an old school style camera hood to look at the images it was sending back. We squeezed in there with him, but we did not see any tuna.

JECH: I don’t see any buoys or anything quite yet.

JENNIFER: Going higher.

JECH: Your altitude is 114 feet. Okay, stop there for a second. Oh yeah, there it is.

GRABER: There it is.

JECH: So you can see the coyotes swimming.

TWILLEY: There it is, there it is!

GRABER: So we didn’t see a tuna, but we did a coyote. Or as some people call it, coyote.

TWILLEY: Just one. We counted.

GRABER: Now, the drone is great, but the battery only works for about 15 minutes. And it can’t go out of site. But tuna swim really quickly and go really far, so Mike works with the spotter pilots.

JECH: The spotter pilot will go out and fly around the area and then find schools and then we, so we are on another boat, a 35-foot fishing boat, and he’ll steam right up to near the school, we’ll fly the hex, the drone hexacopter over those schools and do that as many times as we can.

TWILLEY: The images that Ginger the drone captures are amazing, and Jennifer has become super accurate at figuring out the size of the tuna from the photos. But the drones only capture the tuna while they’re at the surface. The next step for Molly and Mike is to figure out how to combine the drone data with information from beneath the surface. And that requires sonar technology.

LUTCAVAGE: And our goal this this summer, which is the last part of our funding before it all runs out, is to bring the acoustics together with the hexacopter and try to finally get the hexacopter over bluefin school while we have our split-beam sonar pointed at it, and see if in fact we can get a beautiful picture of the entire bluefin school.

GRABER: Flying drones with Mike and Jennifer was incredibly exciting—watching those pool noodles zipping over the water and seeing the real-time images of the coyote—amazing. But the drone isn’t quite ready for prime time yet.

JECH: Battery life is a key. And I think that’s for anything. You know, if we get Tesla to work on smaller batteries for this, that’d be great because it is the battery life. When you’re in close, you know, fifteen, twenty minutes goes by pretty quickly. But when you’re going for something like tuna that goes really quickly, and they can be further away and they’re fast swimming. So it’s battery life is a big thing. And then image processing, we get the images but Jennifer has to do a lot of the processing by hand. And so still a lot of manual work. But I think, another big step is image processing, computerized or automated image processing is another big step for this.

TWILLEY: Last year was proof of concept for the drone, just to show that Mike, Jennifer, and Molly could actually get usable images. Next year, they’re going to try working with population biologists to see how their numbers fit against existing estimates. In other words it’s going to be a while yet before drone surveys are part of the official tuna count.

GRABER: But in some cases these new technologies are already helping survey fish and and protect them. New high-tech underwater cameras not only improved the accuracy of the scallop count in New England recently, but the camera also found a region of young scallops. And so the policy makers could close that region down and protect it.

TWILLEY: So that’s one success story. But there are real challenges to integrating these new methods into the official count. Chris Legault told us that having more data actually, somewhat ironically, makes decision making harder.

LEGAULT: And so it’s almost a curse at times when as we gain more and more information, it becomes more clear how much we don’t understand.

TWILLEY: And we’re not just gaining new information from drones and robot subs. There are teams working on counting fish using scraps of DNA in the water, there are teams using new broadband sonar and passive acoustics to count fish. Counting fish is a hot area for research.

GRABER: However great all of these technologies are, another problem is that each one comes with its own bias. Some fish avoid nets. Others shy away from cameras. Still others are disturbed by sound. As Ginger the drone circled overhead, Mike Jech explained that each of the different methods has pros and cons.

JECH: And that’s that’s one of the key things we really try to push with these is they all give you different information, and hopefully you can use that information and combine all the information. So every sampling gear has its own biases and, you know, and error. Those are the inherent limitations. As long as you know those you can account for those, then that’s what you need to do. So it’s not that any one sampling gear is wrong. If you rely on one piece of gear for everything then that’s not good.

TWILLEY: And none of these new sampling methods is going to put an end to the time-honored NOAA trawl.

JECH: Like we can’t tell ages, what they’re feeding on, you can’t tell what they’re feeding on their ages, the maturity stages, you know whether they’re going to spawn or not spawn. Only way you can do there right now is a sampling. You have to capture them. So we have to use nets.

TWILLEY: Sometimes you just need to catch a fish to count it properly. But also, you need to continue that trawl to have a sense of larger population patterns.

CLARKE: You have to be able to integrate this with the existing data. There aren’t long time series for these new technologies, but there are long time series for bottom trawl surveys. So you can’t, you don’t want to throw the baby out with the bathwater. You want to be able to integrate this with bottom trawl, or more traditional methods. You need to somehow knit these old surveys and new surveys together in a logical, cost-effective way that improves stock assessment and the number of fish that you can do stock assessments on.

GRABER: As Liz Clarke points out, you have to do the counting the same way, year after year, for years and years, for the data to be considered scientifically accurate. And on top of that, you still need physical samples to figure out the sex and the age of the fish.

TWILLEY: So that means you have to keep doing the trawls, but to get better numbers, you have to integrate these new technologies alongside them. And doing both costs money. Obviously. And money is in short supply these days, especially for scientific research and government environmental agencies.

GRABER: And there’s still another issue here. Scientists don’t get to make these decisions on their own. The management process is a political one. And politicians and commercial fishermen sit at the same table with the scientists. And they get a voice in how the fisheries are managed.

TWILLEY: And fisheries management is the point of counting fish in the first place.

JECH: And so you do, there’s a rigorous, you go through the rigorous scientific process of getting your you know the estimates from the instrument accepted scientifically. And then you need to get them accepted into the management process. And that’s a whole another. That’s a whole other story. And so it is it can be challenging.

TWILLEY: Part of the problem is people are just thinking on different timescales. For years, Andy Rosenberg was in charge of mediating the management process. In other words, he refereed the shouting matches between scientists and fishermen.

ROSENBERG: But I mean fisheries management is a, you know, it’s a science informed political process. And of course, people are very focused on a tug of war between short term needs and long term goals. And so when I was lead manager you always were in the middle of that struggle. Fishermen, for good reason, are saying that’s fine, you’re saying what we want to do 10 or 20 years from now. I need to make my house payment on my boat payment or my kid’s tuition or whatever it is, this year or next year. I need to make it to retirement in five years. Their time frame is not going to be what necessarily what’s in the national interest, which is what the legislation that regulators have to work under calls for. You know, do something that’s in the national interest for the benefit of the nation as a whole. Do it over the long term. Sustainability is a long-term concept.

GRABER: To keep listing problems here, scientists are trying to plan for decades, even centuries in the future. And climate change throws a real wrench in that.

PERSHING: Because we’re now seeing ecosystems all over the world that are being pushed into, you know, new situations, new temperature levels, new abundances of the predators and prey, that are beyond kind of the historical reference that we have.

TWILLEY: That’s our other Andrew—Andy Pershing, at the Gulf of Maine Research Institute.

PERSHING: And so our fisheries management tends to really be kind of backwards facing and based on history. But we’re now in a world where, you know, history is not as useful a guide.

TWILLEY: Given the uncertainty created by climate change, and the political shenanigans of the management process, it’s actually reasonable to ask, are all these new counting methods even necessary? Do we even need better numbers?

ROSENBERG: None of them are you know this magic pill that means now we can manage well and we couldn’t before. Management is about overall pattern. Over the long term you need to know, you know, when I was doing it, the exploitation rate was removing sixty percent of the stock every year. and we knew that we should be around twenty. Well, I didn’t need to know whether it was twenty or twenty-two percent that we should be removing every year because we were removing sixty. I needed to know that it had to be a lot less. And so we spent years saying no, is it twenty, twenty-two, or twenty-four that we should be getting to? Well, that wasn’t actually an important question but it enabled people to say, “Oh, we shouldn’t take any action until we know whether it’s twenty or twenty-two.”

GRABER: As Andrew Rosenberg points out, when you focus on getting the most exact numbers possible, you can get paralyzed. But we know full well we’re taking out too many fish.

ROSENBERG: One of the the sort of grand old men of Fisheries Science was a man named John Gulland, who passed away in about 1990 or 91. And John used to say that fisheries management is an endless argument about how many fish there are in the sea until all doubt is removed and so are all the fish. And that’s basically the case. And so there’s this endless push to be more and more precise about counting fish because that delays action. And fundamentally what we should be doing is actually trying to control the amount of fishing pressure and not destroying the stocks by waiting until we know exactly how many there are.

TWILLEY: It’s hard to disagree with that. But even Andrew agrees that better counts are useful. We’ve already seen how they helped improve the scallop fishery management. And they can help get fishermen on board with the process, like Molly and Mike’s commercial spotter pilots.

GRABER: And that might lead to fewer shouting matches. But what’s really exciting is that these technologies can help us get beyond numbers. As Chris Legault pointed out, they can help us understand aspects of fish biology that we can’t get a handle on with just the trawls.

LEGAULT: Exactly, even just basic biology questions. Some of our fish change sex as they get older. And what causes that to happen? Well, there are probably a lot of different factors. But if we could understand better what some of those factors were, that might help us in managing the stock to prevent there being ever being a problem with there being only males or only females, which would obviously be a big problem for the species.

TWILLEY: And it’s not just basic biology. Molly, Mike, and Jennifer are finding that with their new drone technology, they can not only count tuna, but also learn about its behavior

LUTCAVAGE: We’re going to know a lot more about how they use the water column and how far they travel. We didn’t have that knowledge before. And so we’re going to know a lot about what we’re not assessing or counting or including in assessments that we didn’t know before, because of these other new tools.

JECH: And what’s interesting with these types of instruments and sonar that are different than nets is that you can really look at the fine scale spatial distributions of the animals and you can even start to look within schools at that—the behavior within the schools. So if, you know, the juveniles are acting differently than the adults, do you manage the juveniles differently than the adults?

GRABER: And there’s an even more exciting aspect of all these cool new toys. We know so little about the whole world under the ocean’s surface. But these are giving us a glimpse of the entire ecosystem.

CLARKE: So it’s pretty important that we start gathering information that is about the fish and where they live simultaneously, because it’s so important not only to get species specific information for individual stock assessments, but rather, to get the information needed for these more integrated ecosystem assessments that put the fish in the whole ecosystem context. And these methods are just coming online that we can use. I think it’s really exciting.

TWILLEY: We started this episode with a couple of questions: how do we count fish? And can we do it better? But it turns out maybe we were asking the wrong questions.

GRABER: When you count fish, you’re just looking at individual species. That doesn’t tell us anything about how they live in the water, what fish are swimming nearby, how the predators and prey interact with each other, how they interact with their entire habitat. And knowing all that is the only way we’re actually going to be able to keep eating fish and still have more for the future. Especially with the uncertainties of climate change and the world’s growing population. These new technologies are starting to shine a light on a world that’s been largely in the dark.

TWILLEY: So yes, the numbers matter. But there’s so much more to the history and future of our oceans than just the numbers. That, for Poul Holm is the point.

HOLM: We are only discovering the enormity of what we don’t know as we go out there. So it’s still a voyage of discovery in many ways.



TWILLEY: A huge thanks to the Fund for Environmental Journalism for a grant that allowed us to report this story in depth.

GRABER: A print version of this story with lots of additional images, maps, data, and details is appearing in the online multimedia magazine biographic. We have links on our website—you should definitely check it out.

TWILLEY: We have way too many people to thank this episode to list everyone by name—so many fishermen, and spotter pilots and scientists and fisheries managers spoke with us to help understand the challenges and the technology. We have lots of images and links on our website for more, but for now, just a thanks to the wonderful Shelley Dawicki of NOAA for setting us up with so many great people, and Bill Hoffman of Massachusetts Department of Fisheries for taking us along for the day.

GRABER: We’re going to get a little closer to shore next episode, as we wade into the world of oysters. Rowan Jacobsen has a new book out on the subject, and he has some advice for oyster newbies—as I was until a few years ago.

ROWAN JACOBSEN: So I’ve had a lot of success getting small salty oysters into oyster virgins and then they convert pretty quick. It’s really fun to to bring people over to the other side of the fence.

TWILLEY: I was an oyster virgin until my 30s, and now I snarf them down at every opportunity! But it’s not just that they’re delicious—there’s a lot of fascinating history and science to oyster cultivation and flavor.

GRABER: All will be revealed in a couple of weeks! Till then.