This is Science Friday. I'm Flora Lichtman. Today on the podcast. learning about birdsong from chirps and sheeps captured by birders around the world. You could think of it, for example, if you had an orchestra with lots and lots of different types of instruments, a really rich orchestra. Birds are the virtuosos of the animal kingdom. I mean, they can make a remarkable range of sounds. But why is a barred owl more of a tenor? while a cedar waxwing is more of a soprano.
My next guest analyzed 100,000 bird sounds from all over the world to try to peck out some kernels of knowledge about what determines a bird's vocal range and the sounds that they make. Joining me now is one of the study authors, Dr. Zuzana Bujavallava, Assistant Professor of Forest and Wildlife Ecology at the University of Wisconsin-Madison, based in Madison, Wisconsin. Welcome to Science Friday.
Hi, thank you for having me. Okay, so you analyzed 100,000 bird sounds. Where did they come from? So in this research... which was actually led by a PhD student in my lab, Satya Chandra Zagar, we analyzed more than hundreds of thousands of recordings that came from this amazing repository of bird sounds called Zeno Canto.
And this repository is where anyone can upload sounds of birds that they encounter in the forest, in their backyard, or anywhere. Wow, so these are birders who made this data set. that you use. Yes, they're birders, they're citizen scientists, they're nature lovers. Anyone can upload recordings. And they're from around the world?
Yes, they're from around the world. They are from every single continent, biome. Obviously, there are some areas that are better represented than others, but because so many people participate, there's a pretty good coverage, actually. So tell me what you found. Give us the bird's eye view. The bird's eye view, yes. So Zagra found a lot of different things in terms of testing the major hypotheses about what.
rules the bird's sounds. One of the most important ones, and that one has been known for quite a while, is that the bigger the bird, the bigger its body size, the lower the sound it makes. We found that, yes, this relationship is true, and we're able to confirm it across all these different species, showing that really the body mass of the bird rules how deep it can go.
Which is not that surprising, right? I mean, that's often true in other animals that we might be even more familiar with, like ourselves. really really low pitch sounds as low as infrasound and then if you think of mice making sounds at high frequencies, it's not something that surprising. But the novel thing we're able to do is to really isolate this effect and separate it, for example, from even the role of the beak size.
and other things that turned out to be important, like the latitude and type of habitat. Well, let's talk about beak size. How did beak size seem to influence or correlate with the types of sounds a bird made? As you can imagine, the bigger the bird, the bigger the beak in general. But even when we take this kind of correlation into account, still some birds might have disproportionately large beak. exercise.
So the beak size itself was also related to the frequency at which they make sound. Again, with bigger beaks having lower frequencies and smaller beaks being able to make sounds at higher. pitch. Can you give me an example of a bird I might know? So for a small one with a small beak you might know for example the winter wren. And for a big one, for example, all the hornbill species that are common in tropical regions are pretty big animals with even larger beaks.
Coincidentally, a lot of these hornbill species are actually threatened because the beak is being targeted by poachers who use it for all sorts of different things. What about geography? How did that play a role in the sounds that birds are making? Specifically the latitude at which the species occur played an interesting role. What we found was that...
At the equator, the bird species making sound had a smaller range. So they were, in a way, occupying a narrower bandwidth than birds at northern latitudes. Huh. Why would... Birds at the equator have a smaller range of sounds. This is one of the puzzles of bird songs. One potential explanation is that at the equator... there are many, many more species. And so the bird species might have to carve for themselves out a narrower niche, narrower...
bandwidth that they will be using so that they could effectively communicate with each other. You could think of it, for example, if you had an orchestra with lots and lots of different types of instruments at the equator. really rich orchestra, each instrument might be given a smaller... smaller range of frequencies at which it can play so that the listeners can hear each instrument. Whereas at the temperate zones, we might have an orchestra with.
Fewer instruments, and each instrument can cover a broader bandwidth. So basically, if you're in an environment where there's a lot of sound... You don't want to compete across a lot of different frequencies. You want to carve out your own little sound niche. frequency niche. Yes. And this idea of species might be competing for the acoustic space was proposed already more than 30 years ago by Bernie Krause. And we...
We looked at this question whether this is the case in several different ways. We looked at the latitude and we know that at the equator there are many species. And we did find this relationship. But then Sagara also calculated for every single point for where we had recordings. He calculated... How many other species are there likely to be, both for birds and frogs?
so that we could see whether the species richness actually correlates with this bandwidth. And we did not find that that was the case. So it's still a little bit of a mystery why exactly. This is happening across the latitudes. You also found that on average, smaller birds had a larger vocal range, like they could hit low notes and high notes. Why would that be?
This is something really interesting and we didn't really expect this. One of the reasons that this might be the case is that smaller birds might want to, on occasion... pretend that they're actually much bigger than they are. And the way to do this might be by lowering the frequency, so lowering the voice so that other...
Other individuals might perceive, oh, I can hear this really low pitch sound. I am about to encounter a big, scary bird, whereas actually it's a tiny bird. Now, you don't only study birds. I understand you lead something called the Sound Forest Lab. What is that? In the lab, we're interested in actually all animal sounds, not just bird sounds. We are using the sounds that animals make across tropical forests, but also in temperate zones to...
better be able to monitor biodiversity to understand how healthy forests are, what is the impact of different things that humans do, and all that using sounds. Where have you recorded? Oh, we have recorded at lots of different places and currently the lab is working on this. soundscape baselines project where we're trying to record soundscapes at the world's most intact remaining forests so some examples are in gabon in central africa in ecuador in peru
also in Brunei, and also at One Forest in Wisconsin and in Germany. Do you go there yourself and place the microphones? Usually what it looks like is that I go, for example, I went to Ecuador and Gabon and Peru, but there I work very closely with local scientists or community conservation practitioners. people who know the forest really really well and together we place the microphones and decide where should we be recording we have to take a quick break but when we come back
Can a forest soundscape amplify conservation? I consider it almost as a time capsule. We can record it now because otherwise we might lose it forever. Support for Science Friday comes from the Alfred P. Sloan Foundation, working to enhance public understanding of science, technology, and economics in the modern world. We've got a clip of one of your soundscapes from Borneo, the Borneo rainforest. Before we play it, what should I be listening for?
Oh, you should be trying to spot as many species as you can, or at least especially different groups of animals. One example you might listen out for. is the hornbill that we mentioned earlier, a bird with a really big beak. And you might be even able to hear its wingspan. I think in this recording, you can also hear some gibbons.
Okay, so what did you hear in that? Well, I got completely transported back to the rainforest in East Kalimantan in Borneo, and I could hear... the the hornbill right in the middle this kind of almost dinosaur-like sound uh you can also hear the gibbons singing because this is this is early in the morning just when they are waking up there are lots of different bird species i think i heard the white room shama
And then, of course, kind of omnipresent background of insect noise. Can you use sort of computer analysis or AI to pick out the different animals in there? We're using a lot of artificial intelligence to just sift through these recordings and be able to pick up different species. So for example, for bird species and some mammals. it is quite easy to train these AI algorithms to identify them.
But unfortunately, there is no machine learning approach yet that identifies every single species. There is... Just not enough known about the different species of insects, how they sound, to be able to teach a computer to recognize them. We need a Merlin for bugs. Yes, we do. What has surprised you about some of these soundscapes that you've captured? I'm always surprised by the soundscapes we capture. I think...
One surprising thing is just quite how noisy it is in the morning. And this is something that's consistent everywhere where I've been. The dawn is the time that you just can no longer sleep, which is good. It motivates me to get up and do research. But I've also been really surprised by some of the animals that... just seem to be mimicking other sounds. So some of the famous ones that people might be familiar with.
with our birds that can mimic even chainsaws and one thing that surprised me for example in Gabon was the patinos monkey that actually sounded really like gunshots. And it even tricked our machine learning algorithms at one point because their sounds got picked up as gunshots at first. Really?
Are you an audio junkie? Like, have you always been interested in sound? I have been always interested in rainforests. And I have been, I mean, I grew up in a... temperate forest but I think I've always used all my senses hearing also I love the smells of forest so not just the sounds but I think sounds are
Something that is not yet fully explored and it can tell us a lot about the biodiversity that we just can't pick up by just looking. I was just going to ask, are there advantages of capturing sound over video? Oh, absolutely. We have to listen. We can't just look because... Just using images such as from camera trips, which is amazing, but it only picks up those animals that are big enough, that are...
going to pass through the field of vision of the camera. So for example, a chimpanzee moving on the ground rather than sitting up in the canopy. And also the animal has to be moving slowly enough to trigger the camera. So images might not be great for insects or really quickly flying small birds up in the canopy. Whereas with sound recordings... we can capture the animal sounds essentially regardless of the direction. What are the occupational hazards? Have you had an...
Have you had an animal walk off with your microphone before? The occupational hazards, definitely. Tropical forests are pretty tough environments for recording devices. We have had elephants attack or take away the recorders from trees. We don't know if it's because it was upsetting or if they were just curious. But surprisingly, the biggest challenge is termites and ants. They just end up eating the microphone foam straps and pretty quickly, actually.
Do you have a soundscape wish list, a place you're dying to hear? I have a long wish list. I just wish that we can record soundscapes in all of these. intact forests because there are so few remaining. For example, in Madagascar, those forests are disappearing pretty fast. It's just something I consider it almost as a time capsule. We can record it now.
because otherwise we might lose it forever. Well, I wish you luck, and I hope to get to hear Madagascar sometime. Zuzana, thanks so much for coming on the show. Thank you for having me. Dr. Zuzana Buzhualova. Assistant Professor of Forest and Wildlife Ecology at the University of Wisconsin-Madison, based in Madison, Wisconsin. And that is about all we have time for.
Lots of folks helped make the show happen, including Annie Nero, Emma Gomez, Charles Bergquist, Danielle Johnson. I'm Flora Lichtman. Thanks for listening.