Can Drones Help Our Bird Populations?
– Good evening everyone, welcome to the lab of ornithology’s Monday night seminars. I’m Kevin McGowan, in the education section, and I am the host tonight. It’s a great pleasure to introduce my good friend David Bird. David is currently an emeritus professor of wildlife biology, and director of the Avian Science and Conservation center of McGill University in Montreal, Quebec. His research has covered endangered species, captive populations of birds of prey, human-wildlife conflict, and a bunch of other things. And tonight he’s gonna be talking about his new interest in using, I can’t use the unmanned… I know them as drones, so I don’t know what the… David will talk to you a little bit more about that. I’ve known David for a very long time. We’re both old. And I’m not sure exactly when we met, but we’ve known each other through constant meetings with the American Ornithologist Union, now the American Ornithologist Society. And I remember that at least, we met by 1993. And one of the things I want to mention about David is that he’s not just a run-of-the-mill scientist. He has a wicked sense of humor, and is thoroughly invested in talking about birds to everybody. Everybody he sees he just has to talk about birds, he can’t stop. And what happened was, I think we both noticed at the AOU meetings that each of us, we use humor in our talks. And I remember he was senior to me. He was at McGill at that point, at a faculty position when I was in grad school, and I remember him coming up and complementing me on my talk. And I was of course very flattered by that. And I remember specifically the talk that was the first time he ever came up, he just fully engaged me in a conversation. And that was when I was talking about, I forget what the exact title was, something about nest sanitation in corvids, do parents eat poop because they have to or because they like it? And it was a serious study. Since that time we’ve had some mutual friends, we’ve hung out together at these things. It’s always been a great pleasure to me to run into David wherever we go. David, as I say, is really interested in, uh oh, that was just the power cord, no problem. Very interested in getting out there and doing things with the public and with birds. He’s written a number of books, probably the one I’ve seen on most people’s shelves is “The Bird Almanac: A Guide to Essential Facts “and Figures on the World’s Birds.” He’s just recently authored a new book coming on birds of Canada. So a number of other things. He’s done a spectacular amount of research. When I heard about him first he was running a Kestrel colony, a captive colony of Kestrels at McGill, that he says had up to 300, 500 captive birds in there. And they did a number of innovative studies resulting in more than a 100 publications on just those captive birds alone. Things like toxicology, and behavior, just all kinds of stuff. And the other thing that I noted that was pretty cool, was they were doing some captive breeding of Loggerhead Shrikes. And you may know that we lost Loggerhead Shrike as a breeding bird in New York State 20 years ago, and Ontario and Quebec, basically the same thing. So David started trying to raise Shrikes in captivity and then releasing them into Canada. And that was, I thought, a very cool and ambitious project. David’s been a lot things, he has recently won the Doris H. Spiers Award for outstanding lifetime contributions to Canadian ornithology, and I have to say, knowing David, he totally, totally deserves that. He’s a past president of the Raptor Research Foundation. Also a past president of the Society of Canadian Ornithologists, and elected fellow in the American Ornithological Society. He’s also a birder, and was on the board of directors of the American Birding Association. He does a lot of good things. He’s still doing a lot of good things. One of the things about David is that he has what I would call restless energy. Is that he can’t stop thinking about stuff, and he retired several years ago and he’s been more active than ever, getting into this whole drones stuff, starting his own journal. And he travels relentlessly, giving talks and outreach to the general public. So much so, I mean he’s actually has his own column in Bird Watchers Digest, and in the magazine Canadian Wildlife that he’s done for years. He’s done of radio call-in shows about birds. His website is Ask Professor Bird. And that’s what he’s done for years and years and years, and I don’t expect that to change anytime soon. So as I say, the drone thing is kind of a new deal in ornithology and in wildlife science. And David’s really taken to it with gusto, and I look forward to hearing what he has to say about this evolving thing. I actually tried out drones. I was given a drone a couple years ago that I thought we would try to use. See if we could use it to check crow nests without having to climb the trees and disturb the adults. Didn’t work so well, and my drone ended up in the woods, somewhere on the floor of the woods for two years before somebody found it and gave it back to me. So I’m looking forward to hearing from somebody who actually has success using drones for things. And I think you’re going to enjoy this. Please join me in welcoming David Bird to talk tonight about, can drones help our bird populations. – I thought Kevin was gonna drone on forever about me. I had to say it, everybody’s always using that joke on me so I’m gonna use it on him. Yeah, I have to thank Kevin very much for inviting me here. It’s not the first time I’ve spoke to the Cornell group. I forget what the topic was the last time I was here, but it might have been urban wildlife or something like that. It was a lot of years ago. Kevin just gave me a tour of this facility today, and I have to tell you, I’ve never been here since it was, I’m ashamed to say, since it was first built. And it is like the Disneyland of bird centers. It is absolutely mind boggling, I’m still blown away and I feel so intimidated now about everything. This was the original title of my talk on drones about bird populations. But I’ve since changed it to this. Are drones essentially, are they a invaluable technology? Or are they just a new form of disturbance? Cause I have a lot of people asking me, aren’t drones just disturbing wildlife or that result. I’m gonna try and convince you that they are an invaluable technology with this talk tonight. And the reason that drones have got such a bad name, first of all, is because of the very name. We did try to get the public to accept unmanned vehicle systems, that’s what you were trying to think of. But I knew from the get-go, going to all those drone conferences and they kept on saying, oh don’t call them drones because the public is afraid of the word drones because they’ve been used to spy on people and kill people with smart bombs and so on and so forth, by the military. So they started off that way and then all of a sudden scientists started getting interested in them for other reasons. Of course, there’s also the invasion of privacy thing, the Big Brother thing. People are worried that the police are going to be spying into our bedrooms or taking photographs of those who were in protests in the street and so on, so they can chase after them after. And then of course, there’s safety issues. I mean, these things don’t always stay in the sky forever. Sometimes they do fly away. Sometimes engines just quit and they fall out. There was a boy that lost his eye. And a lot of us in the drone world, all awaiting the accident, where a third party is actually killed by a drone, maybe taking down a plane or something like that. But the regulations now have gotten so tight in Canada, the US and other countries around the world that this is becoming less and less of a worry. And of course there’s the final thing. And these are the ones my friends all say to me. How can you work with drones and promote them when they could be used by terrorists and so on. The fact is, that’s still a possibility and there have been some attempts to use them in that way. But that’s not what I’m looking for. I’m looking to use them in a good way. And I’ve categorized disturbance from drones in various different categories here, in different titles here, which I’m gonna go through. It’s the structure of my talk. And I’m gonna get rid of the first one right away. Unnecessary and unwanted form of disturbance. If you’re doing any of these things here, the last thing that you want is one of these things to come along and hover over you. And there have been all sort of court cases where people have done that very thing. And if you’re out having solace and enjoying the pristine wilderness or whatever in some place, the last thing you want is to have one of these come along. That’s happened to me, and I don’t like it either, and I think drones are an amazing tool. I think there’s a place for them, the bottom line. – [Kevin] There’s a cursor on the screen. – One tourist actually got banned from a national park in Australia for life because he used a drone. So they’re getting serious about this. When I was up at Hawk Mountain just three days ago and walked into the parking lot, there was a sign up there, a brand new sign saying, no dogs, no this, no that, no drones as well. So people are taking it serious. Let’s get into necessary disturbance but you’ve minimized, because you want to do it for a professional reason. Say for nest surveys or counts. Now, these are some advantages of drones for ornithologists and wildlife biologists in general. Initially they were much more expensive. When the military had them they were like hundreds of thousands to millions of dollars. And now you can buy a Mavic, for example, which I’ll talk about later, for about $1,000. They fly at lower altitudes than manned aircraft do, and you fly at night. Most pilots will not fly their planes and helicopters in the dark. You get better data, you get much better imagery, and you can even drape the images and stich them together. Less obtrusive to wildlife. I mean, you compare a small drone to a manned helicopter or a Cessna flying over it, obviously less obtrusive. And of course, more green overall. And you can put all kinds of sensors on these drones, as I’m going to explain. But the number one reason for a biologist might want to use a drone over going up in a manned helicopter or an airplane, is that it’s safer. The number one source of brutality for wildlife biologists is dying in a plane or helicopter crash. So the less time you put biologists up in the air, the better. One of the things that biologists like to do is count birds. And so the very first people to get involved in counting birds with drones were your US geological survey. And sandhill cranes were the first target. And they got some amazing pictures using drones that they inherited from the US military. They had these machines called Ravens, and they didn’t know what to do with them. They were sort of outdated, so they gave them to the biologists and the USGS, and they used them to count cranes and do other things, as I’ll mention. I got involved with them way back about 11 years ago, and we were looking at geese flocks on our campus, and specifically Canada geese and snow geese. And this was for my very first master’s student. And this was a picture of our campus there taken from a drone, a fixed wing drone, not a rotary drone. We were mostly into fixed wings in those days. And you get a picture like this, and you can count each of those individual Canada geese, except if they are down against the brown background near the edge of the water, very hard to see. Contrast that with these snow geese in this farmer’s field in a non-snowy environment. You take a picture then, you can count every single one of those snow geese, as well as even the blue morphs. But if those snow geese had been on a white background like this, then it would have been much, much more difficult. Which means that detectability is a big deal with drones. It’s related to a lot of factors. Some animals are more visible in photographic imagery, and others are very hard to see. It all depends on backgrounds, or they’re camouflaged and so on. And the other problem you can have with drones, which we’re still working on, but a lot of people are getting some success with it, is you get thousands and thousands of images on the cameras on these things. Which means, unless you’ve got some kind of a computer program to go through them and find out what you’re specifically looking for, then you’ve gotta sit there and go through thousands of them to try and find what you wanna see. I was bugging my grad students to say, look it, why don’t you do this kind of study for your master’s? Why don’t we put out some decoys, a specific number of different kinds of habitats, different kinds of backgrounds, different numbers, clumps and individually, and so on, and do a study. But I couldn’t get them interested. They all wanted to photograph live animals, whatever. So this guy right, Jarrod Hodgson of the University of Adelaide in Australia, actually stunned the world and had his PhD picked up by the news media everywhere. He put out thousands of shorebird decoys in three different populations, and he compared the counting ability of the drone taking pictures and then using machine learning to count the little birds, against human observers in blinds. And he found that the drone and the machine learning system beat the humans way, way… Were just much, much more accurate. And then Gail Schofield, also an Australian, out of Deakin University, she’s a turtle lady and she was studying loggerhead turtles for her PhD. And she used to use a boat and go out on the water to look for them and count them. And so she compared using a drone, and when she used the boat she got 40 pairs of turtles, but when she used the drone, she got 320. So she was obviously missing a lot of animals by using just the boat. So where I got involved with this with waterbirds with common terns, and I wasn’t the only person to be doing this sort of work. Just to give credit where credit’s due. The US Geological Survey again got involved in counting pelicans and cormorants and all sorts of things using Ravens, and they’ve got all sorts of drones now. And also, this fellow here, David Johnston is also a big drone guy, professor at Duke University. He’s got a whole center with drones and all sorts of fancy gear and students. And he focuses on Antarctica and penguins and seals. And this fellow here, Hans-Ulrich Peter, from Germany is also a big drone user. And he goes down to Antarctica and counts penguins and seals down there as well. My former PhD student who you’re gonna meet little bit later on, Dominique Chabot, before he even finished his PhD, he was drafted by this company called ING Robotics that made all their money using drones over in the Canadian military. But when our Canadian military pulled out of Afghanistan, there was no money anymore, so he started getting into environmental studies. And he hired Dominique taking pictures of Dunlins on the west coast in British Columbia. And this is a picture taken by a drone, and you’re thinking, well what is that, just a bunch of beach sand? Well if you blow that up you can count every single one of the Dunlins. And here’s a picture of Dominique right here. For his PhD, he stayed on after the geese work for a PhD, to look at common tern colonies. There’s a couple of islands off the coast of New Brunswick in Canada called the Tern Islands, aptly named. It’s the second largest common tern colony in North America. And the way they typically count them is they form a human chain like this, everybody has to wear hard hats, because the birds are constantly flying off their nests. And anybody who’s done a tern survey knows how dangerous it can. They’re trying to peck you head and draw blood but there’s also pooping on you, but there’s also stressed like crazy. And occasionally people step on their nests and so on. And even bother the red-breasted merganser, which coexist with them. So I said, well what about using a drone to count them instead of making those human chains and disturbing the birds? So we took our AI multi fixed wing machine here, and if you look at this picture here, you see the yellow dotted line. That’s the path of the drone flying about 300 feet over the tern colony. The first time it flew over, almost all of the terns rose off their nest. The second time it went around, about half. And the third time, hardly any. They basically habituated to it because it wasn’t doing any harm to them whatsoever. So it looked like we had a pretty good technique now where we’re gonna get a good count. Well, here are the pictures that it took 300 feet up. You’re saying, well where are the terns? Everyone of those little white dots is a tern on the ground. And so the count we got to compare with the human chain thing, the ground count was 2,372 and the drone count was 2,426. Why the higher number? Because not all of those turns were nesting, some of them we just loafing. So you have to build in like an error factor in there, probably about 5%. Other studies have done the same. Me being a raptor guy for 40 years though, the very first thing I thought of when I saw a drone in action, especially a rotary one, I thought, oh my god I can put those things over the top of eagle nests and so on and look into them and count the young. So that was one of the first things we got involved with. On the top left, we flew a drone, illegally I might add, over a hydroelectric tower with a bald eagle nest in it in British Columbia. But you can see, you wouldn’t be able to climb up that thing and look in the nest. It would be very dangerous. And then gyrfalcons, on the top right, sometimes nest in pretty hard to get to places. And of course, in Africa you get vultures and eagles nesting in sort of thorny trees with a lot of biting insects, they’re not easy to climb. So putting a drone over those things would be an ideal way to get a count of the young. Now the way that we normally would count these things, is we would either use maybe a cherry picker, if we were looking at osprey nests out in the open somewhere, where you could drive a truck in. Or you might get a climber in there, either climbing up a tree or belaying down a cliff. And all of those things are time consuming and cost a lot of money as well. But particularly, they’re stressful to the birds, cause while they’re happening it’s about an hour to make a climb down to a nest, to get up to the top and then climb down. So we decided to try it with drones. Now there were two questions with the drones. Would we get good enough images? And secondly, what would the birds do? How would they react to the drones? And so I brought in a grad student named James Junda and we worked with Erick Greene at the University of Montana in Missoula, because he had discovered the potential of drones exactly the same time as I did independently. So we joined forces, and we looked at four species. Ospreys, all in Montana, all nesting on poles. And then we had bald eagles, ferruginous hawks, and red-tailed hawks, bottom right, in Saskatchewan. And this was just one of our earlier studies. You don’t have to go to the bottom of the nest tree with the drone. You could launch your drone, maybe nowadays even a half mile away to fly over the nest. The technology’s become that sophisticated. And in this particular nest we found actually a live chick and a dead chick. It was something we did like 300 yards away. And we got quite good at doing this in a very timely fashion. We could put our drone down, about 200 yards away, power it up, have it fly over the nest, take a picture, be back at our feet, be picking it up and walking away with the bird back on the nest on an average of about six and a half minutes. That’s 10 times less than if we had to walk to that nest and climb up it and climb down. So that is a good thing. Now how successful were the flights in getting pictures? With the ospreys, we were successful 80 out of 82 times. Bald eagles, seven out of eight times. Red-tailed hawks, three out of five. And ferruginous hawks, 10 out of 11. We could of got 100%, if we’d gone back the next day. It’s just that that’s all we had time for. So we were constrained by that. Now you’ll notice that there’s a couple asterisks by the osprey, cause I know what some of you are thinking. Did any of them attack the drones? Well the only ones that did were the ospreys. They’re very aggressive birds, particularly in Montana anyway, and they would attack, in fact, Eric told me that they hate these drones so much, they will attack them even when they’re whirring on the ground. They hate them that much. So he developed a system that we adopted, where he put the drone up underneath the nest and then he spooked the osprey off the nest, gently, and then quickly before the osprey could even wheel around, put it above, take pictures, and be gone. And that’s what we did. But there were two occasions where there was miscommunication or pilot error where the drone actually, actually in one case the osprey took the drone down costing about $2,000 in damage. And in the other case, my student had a very difficult drone to fly. It was one of the earlier versions and so on, without GPS, and he actually crashed it into the nest. Well the osprey hit it and knocked it into the nest even further. But those were the only two times, and the ospreys were not hurt in any way. And bald eagles, what they do is they escort the drones over their nest territory. They don’t touch them, they just escort them around the area. But the red-tailed and ferruginous, ferruginous don’t even care about them at all. In fact, they went and sat on a perch nearby and just sort of quietly mewed there while we did the work. So here’s some sample pictures right here. And this is with a point and shoot camera back about seven or eight years ago. They have much more sophisticated cameras now. And you can see the osprey nest in the top left. And we have goose eggs in the top right, so we were able to tell the difference between those. And of course the osprey chicks with the racing stripes on them in the bottom. And then for the other species, here are some other ones. You can easily count them and age them and decide, hey these birds are ready for leg banding, let’s go up and do it. If you ask me today what would be the best drone to do this with, you don’t have to look any further than this DJI plane called the Mavic. It’s about $1,000, you can buy two of them, so if one breaks down you got another one in your packsack. And it has a great camera, a very impressive camera and a gimbal, which means you can direct the camera lens, make it move around. It’s up in the air for about 25 minutes, which is way longer than you really need, and it’s very easy to fly. These are some pictures that Dan took. He’s the guy that worked with us in Saskatchewan, with his Mavic, only two years ago. And you can see the quality of the pictures, they’re even better. And here’s some bald eagle chicks here up close and personal, and in the top left you can even see a dead heron in the nest. That’s how much new information you can get. But this is what really blew me away with Dan’s pictures. Normally Dan, he’s a raptor bander, normally Dan would have to climb, hike all the way up to the top of that cliff and then belay down to look into the nest to say, okay they’re ready to band or they’re not. If they’re not, then he’s gotta do it all again the next two days. With the Mavic, he can just stand a the bottom of the cliff, fly his Mavic up there and get a picture like this, and know that the chicks are ready to band. So it really is a big time saver, and less stress for the birds. But I wasn’t the only guy that had the idea of flying drones over raptor nests. Eugene Potapov, who’s a former Russian now living in Philadelphia, was using them to look into Steller’s sea eagle nests and I talked to him last year at a conference about whether his eagles… And Steller’s sea eagles are even bigger than your bald eagles, and they didn’t even care about the drones. They didn’t attack them in any way, or be aggressive much like your bald eagles. And then in Finland, Tapio Osala is using drones to look into the nest of ospreys and white-tailed sea eagles, again with no interest by the birds and the drone whatsoever. And what he told me about his ospreys is, they’re not aggressive to the drones at all. Which is either a contrast to the ones in Montana, I don’t understand why that is. So what about raptors that nest in thick canopy? I knew that was gonna be a bit of a tough chore. So I hired this guy in 2012 to go to an abandoned red-shouldered hawk nest in the local arboretum. And the first thing he found out that it was difficult to fly in among the branches and so on, to get good pictures of the red-shouldered hawk nest. But he also found, he had a GPS on his, that the canopy blocked out the GPS signals from these orbiting satellites, and then that made it even harder for him to control the machine. So I’m thinking that maybe these not gonna work so well for things like red-shouldered hawks, or broad-winged hawks. And what about hole nesting kestrels? You might think that’s impossible, but I’ve got a plan for this. I studied kestrels for 40 years and I’m thinking if I mount a camera on a long stem in the front, and for good enough pilots, he can stick it in the hole, we might be able to get a picture like that. Just stay tuned on that one. And then, here’s another problem, which I’ve already talked about with the ospreys. There are some birds of prey out there that are much more aggressive than even ospreys, that’s goshawks and great horned owls. If you go near their nests when they’ve got young in them, they will try to kill you. They will attack a Sherman tank. And so what are they gonna do with drones? Well I haven’t tried either one of them yet, but I do get worried about the birds getting hurt, of course. That’s the main thing. There’s another bird out there in Australia called the wedge-tailed eagle. And wedge-tailed eagles have taken down at least 46 drones used by various companies to maybe monitor forest losses or look for minerals or whatever. And some of these drones are worth $80,000. So they’re not very happy about these wedge-tailed eagles. And they’re not even going near their nests. They’re just flying by their territory, the wedge-tailed eagles just hate them, and they just take them down. And so another thing that’s going on with the canopy nesting things, Shane McPherson has been using drones to study the most fierce eagle in all of Africa, the crowned eagle. These things kill monkeys for a living, and they’ve got huge feet on them. But he’s never had a problem with these things attacking his drones whatsoever, they just sort of sit near by and let him take his pictures and so on. And they are nesting in the canopy, but fortunately for him, they nest in the top of the canopy, so it works for them. And the wedge-tailed eagle situation has gotten so bad in Australia, that they’re actually now attacking paragliders and hang gliders. Could you imagine how frightened you would be if you were up there, in a paraglider or a hang glider, and an eagle came by and ripped a big hole in your sail cloths? That’s why the finger, he’s just pointing out where the eagle is flying away up there, if you just look above it, just kidding. So what can you do to prevent birds from getting hurt by drones? Well you could put cowls and cages around the blades, that’s one thing we’re looking into. Also, you could put a Kevlar cage around it like this guy did in France. That wasn’t anything to do with birds. That had to do with it not getting wrecked when he was flying it near buildings and so on. So we published a paper, two papers actually to be specific, on how to fly drones safely around raptor nests, and which ones might be aggressive toward it. And I’m actually collaborating with a lot of these different people I’ve just told you about around the world to write a review paper on it and tell about their experiences, what works, what doesn’t work. And the ones I’m really interested in flying a drone around, of course, are peregrine falcons and gyrfalcons on the side of a cliff. No one’s actually done that yet. I think it would be a no-brainer to use a drone to fly over tree-nesting water birds like storks and cranes and herons, because they don’t attack things that come near their nest. They would just simply sit near by. So this would be an easy thing to do, I would think. I think somebody flirted with the idea in California, but I’ve never seen a study published on it. But what about seabirds? Seabirds are nesting in these humongous colonies on the side of a cliff… Sort of the premise of the idea. The USGS has already flown drones over sage-grouse and counted the birds. You can see those gray little dots there, very easily using a thermal infrared camera. And so FLIR, which is a company that makes the camera, it’s named Forward Looking Infrared Camera, basically have made these much, much smaller now and much more affordable. And so we actually took one of them to Africa. And if you’ll look at the picture in the top left, we got a wonderful picture of a barn owl sitting on a post, and on the bottom right of an ostrich. Look at the heat signature coming from that ostrich’s neck and his head, you’d be able to count those in the dark quite easily. So I got a phone call from a gentleman farmer named Tom Franklin, top right. And he’s a high school teacher, and he’s also a conservationist. And he’s got a hay farm, and so he’s got bobolinks in Ontario nesting in his hay farm. And of course, every time he starts to do his mowing, unfortunately some of the young get sliced and diced in there. And he wasn’t happy about it, so he wanted to know where the nests were, so he could avoid them. So he did a crowd funding thing, and got himself a Phantom and bought a cheap FLIR type camera. And he wasn’t able to get any kind of heat signatures from the birds. So my partner and Kyle Elliott went out there and did it with some bobolink nests in Quebec, and they had a really difficult time getting heat signatures from bobolink nests. Why? Because their nests are often covered in grass. And that could make the difference. And so we were able to find bobolink nests, but in another way though, because the drone would spook them up and we’d get a GPS number and then we’d know exactly where the nest was. But that was not the idea behind it. We wanted to know about it in another way. And so my partner Paul Pace, who you’ll meet at the end of this talk on the slides, has now found a way to affix one of these little tiny FLIR cameras to a Mavic, something that no one’s ever done before. And so using this Mavic would be an easy way to do this. It turns out that just recently, this summer, a professor named Dr. Darren Proppe of Calvin University in Grand Rapids, Michigan has actually took up using the drones to try and find ground-nesting songbirds as well. And he actually had some success. And you can see that arrow there pointing out, and there’s the nest of the birds here. And here’s another one here, you can see the nest up there even without the arrow. And that’s what the nest was. He was looking at different kinds of sparrows in the nest, and those nests didn’t have any grass over them. He did not work with bobolink. So we’re now joining forces with him to see if we can make it work in bobolinks. What about mapping bird habitats? You look at the kind of habitat that birds use. This might be applicable to endangered species, nuisance species. Maybe if you’ve got changes in habitat happening because of climate warming, invasive species or something, or inaccessible habitats like the Antarctica or deserts and so on. Well Andrea Laliberte deserves credit for being a pioneer in this area. She’s actually not working with drones anymore. But she published several papers mapping wildlife habitat. The top left there, that is a normal photograph taken with like a regular lens, and down below, that colorful one was taken by a drone with a hyper-spectral lens on it. Which gives you a colored picture, and each of those colors represents a different kind of vegetation or bare ground or whatever. And then you look in the right there, and she was able to map those out to see how much of the habitat was bare ground, how much was sage, shrubs, etc. etc. And so Mathew Kamm, an American student at Tufts University, got working with American Kestrels, which caught my interest. And he just finished his master’s recently. He’s published a paper on it. He was just simply wanting to know what kind of habitats were kestrels assessing when they were deciding whether they want to hunt in an area, or nest in an area, where there was bare ground, grass, and so on. And I’m not gonna go to all these details here and so on, but by and large, he found that it was very easy to use an off-the-shelf drone, like a Phantom and so on, and the regular lens that came with it. Which means you could probably have citizen science do this kind of thing. And he was able to basically get great pictures showing different kinds of habitats. Now he does say that differentiating the different vegetative cover substrates is a big obstacle, because you have to go up and there and then, okay say which habitat is green, which is purple and so on. But it’s doable, it’s certainly doable. And then Rhonda Milliken gave a post presentation at the International Ornithological Congress in Vancouver in 2018. And this is her paper right here. I’ve worked with Rhonda in other things, she’s a really smart lady, and she was able to use a drone to assess critical habitat of a sage thrasher, which is a species that’s in trouble in our country. What about mapping the bird use of habitat? How a bird uses habitat? Well this guy, just like Erick Greene at University of Montana, independently came up with the idea of using drones as well. And he was a former postdoc of mine, his name is Juan Jose Negro. He thinks outside the box. And so what they did is this, it was very clever, they captured a lesser kestrel, it’s a falcon in Spain that’s actually in trouble, and they put a data logger on its back, and they let it fly around for a whole day, over the countryside and even over the streets of Seville, where he lived. How he got permission to do that, I don’t know, because you wouldn’t be able to get it here in North America. And so they caught the bird up at the end of the day. They took the data logger, took all the GPS points, and put them into their drone and let the drone fly the same pattern using the GPS points and taking pictures of everything that the kestrel saw. So they were able to map out the day in the life of a lesser kestrel. It was absolutely incredible stuff. What about mapping bird distributions within habitats? Well, the only people that have done this with any success, just because people have not been so interested in it yet, are again the USGS, and they’ve been studying pelicans. And everyone of those little red dots in there you see is a pelican nest. And so they can take a picture like this and turn it into somewhat of a 3D picture and basically say, okay there’s where the pelicans like to nest in this place, the base of that cliff, or scattered over here but not in this area right here. But coming back to the common terns, I wasn’t gonna give Dominique Chabot a PhD, just because he had a technique to count terns. He had to do a lot more than that. So what he did, is he mapped all the habitat on those two Tern Islands, and characterized all the different habitats. He had control nests and real nests. And then when he flew the drone over, he got those pictures, and he knew where the nests were. And basically, he ended up getting these wonderful results like this, where he could actually categorize what parts of the Tern Islands did the common terns prefer to nest in the most. Was it near the shore, was it in the dead grass, the dry grass, or the live grass or wherever? And that’s the kind of results he would get. So it was a pretty neat thing. But I asked him to do one more study. We were asked by the Canadian Wildlife Service to study the threatened least bittern. The military in Canada have control over a compound, it’s actually a large marsh in Southern Quebec. And that’s it on the bottom right-hand side. And they wanted to know this. They had gone out in a canoe and used playback recordings, and they knew where all the nesting territories of the least bitterns were. That’s what all those colored dots are. They all represent a territory. What they wanted to do though, was they wanted to know why were the territories spaced out like that. What kind of habitats were they preferring? And so Dom launched his drone there and flew it over the area, and gave them pictures which they could overlay over the territories. And so they could then come up with a regression analysis, to say okay the birds preferred bulrushes, I’m having a senior moment. Basically they just corroborated what was already known about these bitterns anyway. No new revelation there. So what about using drones to track and detect birds? Besides counting birds, biologists love to track them I mean what we do now is we put transmitters on these birds. We’ve done everything from putting on satellite tags on large birds like condors and eagles, right on down to ducks and so on, to now putting on little things called nanotags. Which are so small you can even put them on a monarch butterfly. So normally, to track those animals carrying those things you gotta sort of walk around with a Yagi Antenna, or put one on your vehicle, or maybe on a plane, or maybe in a boat, and it’s a lot of work. And sometimes the habitats are quite inaccessible. And so I thought, well what about putting a drone up in the air with a receiver and antenna on it, and letting it do all the work of picking up the signals. And you could use rotary machine like the top left, or a fixed wing in the top right, or maybe even a dirigible, because those are also classified as drone, or even a kite. That one there has a camera on it, that’s a drone, it’s an unmanned aerial vehicle, essentially. And the Australians were the first to beat us to the punch. They designed a drone, which is in that picture right there, that actually picked up great signals from endangered mammals, I’m not sure about birds, but endangered animals in Australia. The trouble with it is, is that this thing costs about, like hundreds of thousands of dollars. And most wildlife biologists, except for the ones that work for Cornell, don’t have the money to be able to afford that. Certainly I couldn’t anyway. And so that was a bit… I was more interested in what could we do with off-the-shelf, like cheap drones and so on. Not stuff like that. So we basically draped a Yagi antenna from the bottom of our Sky-Hero. That’s me on the right basically catching it out of the sky, and for practicing. And the first thing we ran into right away, was we go interference from the motor of the drone. So all we got, instead of getting all we got was , like that. So we had to find a way to get rid of that interference. So Paul Pace, my partner who’s an aeronautical engineer and a military scientist with 40 years of experience working with drones, basically kept working on this problem. He was like a terrier with a bone. And he’s gotten rid of the problem with the static interference. So we’re now just looking for a project, and we’ve got one we’re gonna be doing with bats actually, coming out next summer. But meanwhile, I fell in with these guys here. These are biologists from Quebec that were doing a really neat study in this habitat here in Northern Quebec on Bicknell’s thrush and Swainson’s thrush. And they were using these things called Motus towers. I’m sure some of you’ve heard of these things now. It’s the hottest thing now for tracking birds. You essentially put these towers up, and you put little nanotags, there only about $200 a piece, on the backs of quite small birds, and if they fly within a certain distance of that Motus tower, they’re presence get picked up. And so essentially you can pick up when they leave, when they come back. It doesn’t triangulate where their nest is, it doesn’t do that, but it just let’s you know that they are in the area, if they’ve got one of those nanotags on. And so they even had a Land Rover driving around with a Yagi antenna as well. So we compared taking a drone and putting on a Sensorgnome and antenna, it’s kind of a receiver, to see how it would compare. And we picked up nine signals right away from Bicknell’s thrushes and Swainson’s Thrushes. In other words, it was a big success. And that was in the days before we had technology where we could actually have the drone sort of going over the trees, and not flying into the trees like we did with one, on one occasion. But this worked out real well. So the next thing we wanted to do is, could we use a drone to actually find the bird’s nest or their location near their nest. And Paul tried very hard to do this. He’s got it down to much better than this right now, but the error is still too high. But they’ve now put these nanotags, we’ve actually flown our drone, our Sky-Hero over a whole bunch of released monarch butterflies wearing little tags on them, and we picked up the signals for them. We’re about to do the same thing, we did do it with bats, but the thing is we’d like to try and find out where the bats are roosting. And that means that you have to put the drone up in the air out of view, and that is something that’s very hard to get permission to do, in your country and in my country, most countries. It’s called Beyond Visual Line of Sight. That’s one of the drawbacks to using drones right now is we just can’t get easy permission to put a drone up and fly over the forest and look for crow nests and that sort of thing. Because once it gets out of sight, there’s not guarantees that it might not hit something. So the reason that we want to know about how wildlife reacts to drones is because there’s a lot of professional users out there who are flying drones for various reasons. It could be as a real estate agents, taking pictures of a house or something, first responders doing crime scenes, and firefighters, film productions, foresters, they’re all using drones now. And they need to know where they can fly these drones safely without disturbing the wildlife. So we have to create that data for them. There’s also this problem. There are millions and millions of recreational drone users out there. Now, finally they’re on a leash, they weren’t before. Actually the toy industry got in there before the FAA and Transport Canada in my country could react. And everybody was buying these things for Christmas gifts and everybody was worried about what the heck was gonna happen with all those things. And so we need to do studies using drones deliberately on wildlife to find out how they’re gonna react. And all also, testing stress effects of the flights is another thing we need to do. This is deliberate disturbance of the animals, or disturbance of nuisance birds. Excuse me. Kevin and I were just talking about this before I started talking. This study came out in the news. It was a study done by a team of guys, Mark Ditmer. And what they did is they took four bears and they put a heart telemeter in each of them, and then they flew a drone 60 feet over them. The heart telemeter told them that the heart rate of the bears went up to as high as 400 beats a minute. So yes, but they didn’t compare it to what would happen if you used a manned helicopter over the bears, or an airplane or whatever, or even a human walking around the corner. Maybe the same thing would have happened. So I was a little bit dismayed because this is the headlines that came out of the newspapers, Drones Bothers Bears, Nearly Triggering Heart Attacks. Now I’m trying to give drones a good name, this gives drones a very bad name. So I was really upset about this when it came out. So I feel that if you’re gonna do studies then at least do good science. So the last section I want to talk to you about is nuisance birds. And this is another way you may want to deliberately frighten birds, because this is a billion dollar growth industry. Give you some examples, we’ve got birds at airports and town landfills that we’re trying to control. We had snow geese eating winter wheat in the US. Quelea and rice crops, quelea are the small little weaver birds. They’re actually devastating villages, causing starvation by eating all the rice crops in an afternoon. It’s the most numerous bird in the world. Parrots are eating crops in Australia. Blackbirds are eating expensive experimental crops. Cormorants and aquaculture facilities, and of course the ubiquitous Canada geese in parks and on golf courses. All of those things are just some of the examples of why it’s a billion dollar growth industry. And this is some of the typical ways that we control them. Trained falcons, the public loves, it’s the whole idea of using a trained falcon. It’s what they’re meant to do, and that’s fine, they don’t always kill something but they scare the bajeebies out of the birds. Secondly, a terrier dog, if you wanna scare Canada geese out of a park, then just set a dog on it, and they can’t stand them. But you gotta have somebody to look after both the falcon and the dog in that case. And of course there’s these loud Banger guns, and they work to a certain degree, but as you’re gonna find out, not that great. And of course, on the right, shotguns. You know, actually killing birds, that is the most effective tool. Jamaica Bay, for example, is a very good example of that. They kill hundreds and hundreds of gulls there, but the problem is the public doesn’t like it. So what about drones? Well the idea of flying a drone that looks like a hawk is not a new idea. Back in the 60s in Canada, Transport Canada, our FAA, did a lot of studies on flying radio controlled planes just like this, it looked like a goshawk. And it worked great, for the first few days. After a while the birds just said, ah here comes that noisy thing again, it’s not gonna hurt us, let’s just keep feeding. And that’s exactly what they did. They habituated to it. And that’s still a problem today, let me explain. We have these new drones out now. They’re called Robo-raptors, and the guy in the top left there from the Netherlands designed, he took two species and made drones out of them, peregrine falcon, and inexplicably he chose the bald eagle. I don’t know why he would choose that, and say not a golden eagle or something like that, even with the white head and everything else. He is now getting contracts here in North America in Edmonton International Airport in my country, and some of the ones in your country. And flying these things, they fly just like a raptor. And they are fairly effective for now. But as long as nobody’s getting hurt in any way, I’m not sure if they’ll work long term. And this guy here up in Ottawa, Ontario, he’s got a company there called GooseBusters. And he’s hired by cities to basically fly a drone. He doesn’t have it look like a hawk, he just simply hazes and harasses the geese and the gulls, and they don’t like these things of course, and they go away. But you still need that guy to be there to do it. So he could just as easily go there with a terrier and do the same thing. So I got involved with the drones in the very beginning because a friend of mind, a student of mine, wanted me to help her father design a drone to look like raptor, this was about 12 years ago, to scare away starlings from vineyards. And so I’ve come full circle now. Paul Pace, top left there, my bright partner, he managed to work out a way to have two bear bangers be shot by a drone. I don’t know if you know what a bear banger is, but they’re little orange pencil-like guns. They’re very cheap, you can buy them in various hunting and fishing stores, and hiking stores. And the idea is, of course, if you got a bear coming towards you, you don’t shoot the little orange thing in front of them, or behind them, you shoot it in front of them so the bear goes away, he doesn’t come running towards you. So he put two of these, a two-shot pistol on our drone. The first thing he notices when we went to the vineyards in one of our top grape-growing villages or towns in our country, was that we couldn’t find any starlings in any of the vineyards. Why, because they’re not there all the time. They’re only there on occasion. In other words they’ll decide to go to this guy’s one day and another guy’s another day. And so that means that we pretty much had to stay there and babysit our drone, waiting for the starlings to show up. So that was one problem. So in the end, to test out our bangers we had to go to a local feedlot, in amongst all of those things. And there were about thousands of pigeons, thousands of pigeons on the ground feeding, and thousands of blackbirds in the perches above sitting there. We put the drone up, fired off the bear banger and all the birds took off. In 60 seconds, all the pigeons were back on the ground in exactly the same spot, and in 10 minutes, all of the blackbirds were back. So they’re habituated. Why are they habituated to loud noises? Because they hear propane cannons all day long in every vineyard in the two towns that grow grapes. So they’re used to it, so that’s a problem. The other problem is about the babysitting. So wouldn’t it be nice if you could lease or sell to a farmer a drone like this that could charge itself on one of those plates. Those are two different systems there. It’s just kind of like one of those… I forget the name, the Star Wars things where the beam, you know beam down. So what happens is the drone goes up, flies low over the blueberry crops, or vineyards or whatever, comes back, lands exactly on that charging pad, charges itself up and then takes off again an hour later. So you have it all timed out by computer. The problem with it is, is this, is that the government will not let these things sit there un-babysat. In other words, they gotta have somebody there. So you might just as well hire a young kid with a shotgun to walk around in the fields, and fire a shotgun off every hour. It’s just unfortunately, it’s put people off on using them in this way. And I think what we have to do, is we have to make those drones a little bit more dangerous for the birds. Maybe shoot rock salt or rice grains, or something that stings, but it’s very hard to hit these birds using a drone. So that’s probably a nonstarter. The other possibility is having them emit high-pitched sounds, alarm calls, raptor calls, that might work. But I think probably the ideal way will be firing laser beams. Have a turret gun firing off a whole bunch of green laser beams cause birds don’t like them. And in fact, in California a lot of the farmers are now employing guns just like that one up there, on poles, and they’re firing green laser beams over top of their crops, and they’re very effective at scaring away birds. The problem is, is the Transport Canada in my country doesn’t like drones flying near airports, and they certainly don’t like laser beams either, because of what the kids do with the damn things. So putting a laser beam on a drone is almost a nonstarter, for now, we’ll see what happens. I’m certainly not gonna do this. I’m not gonna put a machine gun like these idiots did on a thing. And I’m gonna do what this guy did, where he took his pet cat and he turned it into a quad rover when it was dead, mind you. Now I know that this was scare birds away . But the public isn’t going to like it. I wouldn’t be surprised if Pete Mara doesn’t have one of those. And I do know about this story, speaking of birds of prey, I’m right near the end here, of training an eagle, the Dutch started it first, and then the French and so on. The problem with this is that, the idea was to take down drones used by bad guys, you know, delivering drugs into a jail or something like that. And the idea would be to take the drone out. Nowadays we have drones that actually take out drones. I’ve gone to drones conferences and seen them, two different kinds. And the problem with this is if I’m a bad guy and I know the police have got one of these things. I’m just simply gonna use a more powerful drone and sharpen the blades, and cut the legs off of the eagle. The other problem is, by the time you get that eagle out of its cage, in the truck, and over to the airport or wherever the heck the thing is being flown, the drone is gone because it take them quite a while. The drone can only stay up for 20 minutes. So here’s the thing that’s slowing down the drone industry and making things rough for me, is that we need to get a way to use Beyond Visual line of Sight. That way we would be able to fly drones way along power lines, looking for electrocuted birds. We’ll be able to fly over turbine fields. We’ll be able to fly over large expanse of territory and so on. And these are the two guys, the two regulatory bodies that we have to deal with, Transport Canada in my country and the FAA, and even Homeland Security in your country. And so I think what we need ae a lot more studies published in good journals, using empirical data, like these to try and get some of these answers. And to put my money where my mouth is, as Kevin said, I actually found, six years ago, a Journal of Unmanned Vehicle Systems. Which was the second of its kind, there was one on micro vehicles out of Florida, and it’s doing fairly well, and I’m about to step down as editor and my grad student, Dominique Chabot it taking over as editor. So I’m rather proud of him. And just the last slide here, this is Paul Pace. I always dedicate my talk to him because without him I wouldn’t be able to do all this stuff. I’m the dreamer, I’m the one that gives him the ideas, and he’s the one that makes them happen because he knows how to build these things, how to fix them, and how to fly them. Thanks very much. – [Kevin] Thank you David, I’m sure we have time for questions. People are probably curious about some things. – It’s a technical talk, I know, yes sir? – [Man] Do you have realtime video feeds from the drones? – Do I have realtime video feeds? Yeah you can get that, yeah. – [Man] But is that what you use, though, when you’re… – No, the guy that I showed you, Juan Jose Negro, that thinks outside the box, I was over in Spain watching him fly his drone, and he had a pair of glasses that you could put on so you could see what the drone sees. He was actually counting geese with them in a desert area in Donana. But he had that sort of a thing. I have not had experience with that. But the Mavics, the little Mavics which are amazing little machines, my partner Paul Pace, who knows a lot about technology, thinks that their wonderful machines and worth every penny. And that’s why I’ve been touting them a lot. I’m not getting any kind of kickback from DGI for that at all. But you can buy, for I think an extra $500, special glasses so that you see what it sees. That’s a good question. – [Kevin] So I’ve seen people using smartphones that they are sort of driving with the… – Yeah you can do that. A lot of my friends that have Mavics for fun are using their smartphones or a tablet to do that with. You just put programs on them. It’s amazing, you know, when I go to these drone conferences, I mentioned to you about how there’s now drones out there that catch drones. That was the latest thing I saw. I’ve been going to drone conferences now for about 10 years. And they all have exhibits, lots of exhibits. Everybody’s trying to make a buck out of this. So every time I go, I see new companies springing up. Companies that sell insurance for drone users, or landing mats for, sometimes it’s little things, or even for using a tablet, sun guards and all that sort of stuff. The latest thing I saw in Las Vegas last year was drone catchers, and there’s two kinds out there. If there’s a drone up there doing something bad, then you send this other drone up and it uses a big Spider Man net and catches it and takes it out of action, or it actually fires some kind of a beam out that deactivates the electronics. I don’t know where the drone goes when it falls down. I like to think it isn’t over a crowded beach. – [Kevin] So I was curious a little bit about the fixed wing, and what actually is powering them? It’s obvious with the quad drones, but how do the other ones work? – The thing is about using a drone, what is it about fixed wings over the rotary drones, Kevin’s asked. What powers those? You can have them powered mostly with an electric motor. The one that we had was an electric motor. It makes a whining noise, so you can barely, barely hear it. But there are others that are a little bit noisier. But you can also put gas engines on them. We went to Labrador to count Caribou, to track them. They had VHF transmitters on them. And we had this $650,000 drone, beautiful carbon fiber thing, it looked like a big giant airplane with wingspan longer than my arms. And they were supposed to put a gas engine on it. If they’d put the gas engine on it, we could have stayed up in the air for about 18 hours. So we could have flown for hours and hours and hours out in the countryside. And we were flying in a military compound, where the military control the airspace. So they would phone us up and say, okay it’s one o’clock, you’ve got two hours to fly your drone, there’s no commercial traffic, no military traffic in the area. And so unfortunately didn’t have the gas engine with us. And it does make a lot more noise mind you, but it doesn’t matter the thing was like, oh gosh, about 400 feet, 500 feet up in the air. Caribou wouldn’t even know it was there. I think the ideal drone for wildlife biologists, and people are working on this, is one that takes off up and down, like vertically up like that, and then can go straight like this, like a Harrier jet. And there are people working on that. Trouble is, they’re still a bit expensive for guys like me, but that’s the ideal one to use. And there are many ways of launching these things. Because the big one we used up in Labrador, we had to launch that with a pneumatic gun. Almost like a thing that would just fire it off. Others, as you saw, you can launch with your hand, there are other ones that launch with slingshots out of the hand. All sorts of different ways of doing it. And they can belly-land on the ground. There are some that can land on the water now. And there are some that they’re caught by a US Navy ship for example in a big net, just like the same way they catch the planes landing on the aircraft carriers. All that sort of stuff, it’s pretty neat. So you’re all gonna get out and buy a drone now so you can go count birds in your neighborhood? I don’t think so. – [Kevin] I had one more question. Do you fly the drones? – I’m ashamed to say, that I’m not a drone pilot. I have flown a Mavic, they’re so simple, I could probably teach everyone in this room how to fly one of those, it’s so easy to use. But I’m just not a great pilot. It takes a certain kind of skill. I think these kids nowadays that have this ability to use their phones and everything else, that’s why they can fly these things so well. But Paul is quite a good, he’s a former manned aircraft pilot as well, and so he’s flown all sorts of things, and done all sort of things with drones, which I’m allowed to repeat in this room. So no, the quick answer is I’m quite happy to let Paul fly or students fly the drones. – [Kevin] All right, well thank you very much David. – Yeah, thank you very much for coming, appreciate it, thank you. – [Kevin] My experience with drones was that, the first time it’s like this is really fun, and then it’s like, but I can’t make it do what I want to make it do, and I was thinking about it for a while.
End of transcriptSmall unmanned vehicle systems (UVS), sometimes referred to as “drones” and formerly exclusive to militaries, are rapidly advancing in sophistication and availability to civilians. Ranging from hand-launched autonomous airplanes to terrestrial robots to underwater machines, they are increasingly being employed in such areas as agriculture, emergency services, meteorology, oceanography and now, small UVS are being used in the field of bird research and management, for example conducting population surveys, tracking radio-tagged birds, sensing and observing birds in inaccessible or dangerous places, mapping and monitoring bird habitats, and deterring nuisance bird species. Join Dr. David M. Bird as he explores these applications of UAS for research, management, and conservation in the world of birds.