Underwater Soundscapes: What Can We Learn by Listening? Meet GLUBS
We’re all familiar with the hauntingly beautiful songs that Humpback whales sing and with clicks and beeps that dolphins make as they echolocate schools of fish. But have you ever wondered about the larger sonic landscapes or soundscapes that exist underwater? Scientists may know more about deep space than we know about the deep sea and we’re only beginning to even understand the right questions to ask. An exciting new project is being developed bringing together scientists (and citizens) from around the world to understand what we know – and what we don’t know – about the world of underwater soundscapes. The Global Library of Underwater Sounds (also known by the charming shorthand “GLUBS”) was announced in the journal Frontiers in Evolution and Ecology in February 2022. According to Audrey Looby, one of the co-authors, the project has the potential to transform what we know about our underwater world and has profound implications for conservation, restoration and our understanding of our fellow water-dwelling animal kin.
Audrey Looby is a PhD candidate at the University of Florida and is currently one of the co-leads on a project called Fishsounds, a quantitative and comprehensive inventory of all known fish sound production.
Bioneers Senior Director of Programs & Research, Teo Grossman, spoke with Audrey about GLUBS, soundscape ecology and the fascinating world of underwater biological sounds.
TEO GROSSMAN, BIONEERS: Let’s start with the basics. What is GLUBS and how did you become involved?
AUDREY LOOBY, UNIVERSITY OF FLORIDA: Through my work with Fishsounds, I got in touch with Miles Parsons of the Australian Institute of Marine Science, who is leading the Global Library of Underwater Biological Sounds effort, or GLUBS for short. He put together a group of researchers from all around the world who have different specialties, including data management, bio-acoustics, eco-acoustics, soundscapes, a community with general expertise in sounds underwater, to start thinking about creating a comprehensive library and integrated network of bio-acoustics or sound repositories, as well as other tools, including creating machine learning applications. This is an idea that many researchers have been talking about for decades.
TEO: Humans are very visual creatures, that’s our best sense, and, as a result, most of us are familiar with ecological landscapes. But your research and the nascent Global Biological Library of Underwater Sounds is looking at soundscapes, basically.
AUDREY: That’s the term used by the field for the past couple of decades now. Basically soundscape ecology encompasses any sounds in the environment around you and, specifically what is contributing to the sounds and what is hearing the sounds.
TEO: What can we actually learn about the world by way of soundscape ecology? What are we hoping to understand?
AUDREY: My main interest is soundscapes underwater, but sometimes it’s easier to talk about terrestrial sounds because that’s just more in our human experience. If you walk out into the forest, you might not be able to see everything that’s living in that forest, so it becomes helpful to use your other senses to be able to find out more about what’s going on.
In the case of birds, because a lot of their songs are species and even individual-animal specific, you can learn a lot about the biodiversity in a forest habitat, as well as sometimes even get abundance estimates, by listening.
For the ecology of it, birds make songs. Why do they do that? How do they do that? What other organisms might be listening in on that song? How does that affect them? What are the other sounds that might be influencing where the birds are singing or that could be hurting their ability to use their songs most effectively? This is when human impacts could come into play as well.
There’s an evolutionary aspect of the research, asking how we, as humans, evolved the ability to use sound for communication. How do other taxa use it? From an applied standpoint, we can use ecological soundscapes to learn more about the status of an area. In some cases we may use that knowledge to reintroduce sounds into a particular area for the purposes of ecological restoration or habitat enhancement, or to regulate the sounds that we put into a particular environment, as there’s more growing attention on the impacts of noise pollution.
While I was just talking about a forest and birds, all of what I just said holds true for underwater environments as well. Along with that, we’re fairly confident all marine mammals use sound for communication underwater. There are roughly 34,000 species of fish that are thought to exist and we know of at least a thousand species that use sound for communication. There are likely thousands more that we just haven’t documented yet. And even invertebrates can produce sound for communication as well.
TEO: As I was preparing for our conversation, my children wanted me to ask about some of the most unique or beautiful sounds that you’ve actually encountered in your research?
AUDREY: One of my favorites that I love to tell to people to make them giggle is that there are certain species of fish in the clupeidae family, like herrings, that create sound by expelling air out of their backsides. They’re essentially farting, and they’re able to communicate information to each other through their farts. They even use it to confuse predators that might be trying to eat them. Fish farting communication is a thing we can confirm. And even beyond those specific fish species, a lot of other fish sounds do kind of end up sounding like farts a little bit.
Gulf toadfish and other toadfish species have some of the more complex calls that you’ll find in the fish world. They’ll have different components, like a phrase where they’ll start with some grunts and then they’ll create a multi-harmonic call that is more prolonged. Those can vary, depending on the situation, and they can form whole choruses where a bunch of males will be singing and basically competing together. Sometimes you can even have certain species of toad fish do what’s called tagging, where they’ll hear like a subordinate male trying to call for females, and the more dominant male will basically grunt in the middle of the other fish’s call, to try to be like, “Hey, I’m way cooler than this other guy; don’t listen to him.
So there are a lot of quirkiness in fish sounds, for sure, and there’s a lot of variety. Then, of course with marine mammals, and whales in particular, you get these really complex songs that can be broken up into individual components, and then phrases, and then full songs that can evolve and change over time.
TEO: Jesse Ausubel, one of the co-founders of the project, suggests in an article that AI applied to marine animal recordings could yield some level of understanding, “Human song varieties include love and work songs, lullabies, chants, and anthems. Marine animals must sing love songs. Maybe AI applied to the Global Library can help us understand the lyrics of these and many others.”
This is perhaps pushing the boundaries of what we know and what we can interpret in animal science, but that said, I’m aware of research describing humpback whales transmitting different songs to different pods, meeting the human definition of culture by way of song-learning. Are there other examples of this? How much do we know and what are the leading edges?
AUDREY: To summarize that research briefly, whales within individual pods will teach each other songs, and there will be “fads” that come and go over time. I can’t remember exactly what the directionality of this is, but humpback whales in one region of the world will develop melodies or phrases that get passed to other humpback whale pods in other regions. It can even be the case that one of these regional pods is more of like a cultural go-getter that creates the fad that is found a couple of years later somewhere else in the globe. This is because whales develop their songs through learning, through exchange with other whales, and this isn’t always the case with all underwater organisms that use sound.
With fish, we haven’t really found evidence of learning so much yet. There are examples of certain species that exhibit regional dialects, where species in different geographic regions have slightly varied call structure than in other regions. This distinctness is also sometimes used to help provide evidence for speciation or emerging speciation. For example, difference in sounds between two regional populations of the Black Drum fish is being used as an argument in a recent paper to suggest a possible species differentiation. One of the primary pieces of supporting evidence for this is that the difference in sounds between two regional populations would make it hard for them to communicate information to each other.
We can also distinguish individuals, depending on the species. Some of them have more complex calls than others, but even an individual fish can sometimes be identified based on its call compared to other individuals. In a lot of ways, marine mammals lead the way on a lot of the underwater acoustics research and we’re increasingly now applying that research to other species as well.
TEO: Can you talk about the conservation and restoration applications of soundscape ecology in general and the GLUBS project in particular?
AUDREY: In the ocean, lakes, streams and other underwater environments, it is really hard to find organisms to study. We have a bunch of different methods including catching them with nets or counting them visually with scuba divers, or putting down cameras, all sorts of active acoustic tracking, but we still have so much that we don’t know and that we can’t detect. So what’s called passive acoustic monitoring or listening to sounds with recorders and other equipment, offers an additional tool to find out what’s going on underwater.
For example, if you have a species of fish that’s invasive, like the Lionfish or Freshwater Drum up in New York that are vocal callers, you can use their sounds to detect the spread of an invasive species. Or on the flip side, you can find an endangered species that might be hard to see or find. There are a lot of specific applied questions that can be asked, like looking at distributions or detecting success in fish spawning,
Marine mammals can be very hard to find, they dive really deep and a lot of those species are increasingly rare because they are in decline, so one of the easiest ways to study marine mammals is through their sounds.
On the conservation/restoration side of things, many species use sound to find suitable habitat, to interact with other organisms. Many fish and invertebrate larvae, for example baby fish, baby corals, baby oysters, swim around in the water and then they have to pick a place to settle to spend like the rest of their life, especially the ones who attach and don’t move after that. One of the big factors in that decision can be sound. If we are trying to restore degraded habitat, say like an oyster reef, suitable habitat can be put down but it can be difficult to attract species to it right away, just because they need other species there. Playing sounds of a healthy reef to help attract species to that particular area can help.
There’s a lot of other work that’s been done for all sorts of reasons detecting habitat complexity and degradation, or looking at the effects of human noise, anthropogenic noise on underwater environments. And all of this research requires information. We need to know what species are making what sounds; what different areas of the world sound like; how they sound different; what species we know a lot about; what species we do not know a lot about. All of this different information is required to make these tools as effective as possible, and so that’s where Fishsounds on the fish side of things comes into play to some extent. GLUBS will help even more by offering more extensive tools and information to get us even better at performing acoustic monitoring or other soundscape-related applications to conservation and restoration.
TEO: I know one of the applications of GLUBS is the community science component. Can you describe how people can engage with both the library and the research as it starts to develop.
AUDREY: One of the biggest successes for birds, amphibians and insects has been using citizen or community science to collect information. There are way more members of the public out there than there will ever be researchers and scientists with time to devote to this stuff. So it’s really, really helpful for people who have an interest in either finding a particular bird ID or things like that to collect information that we can then use to inform the science and the research.
There are so many different smaller efforts that have existed or are up and running now where there are gaps in what we’re able to do, and that’s what GLUBS will serve as, helping to connect us and share information more effectively, as well as create tools that, because of the scale and the way funding can work for these things, smaller efforts like Fishsounds wouldn’t be able to do on their own.
One of the things that I’m most excited about is the community science side of things. If you’ve ever heard of Merlin Bird ID from Cornell or Birdnet or iNaturalist, all of those are projects have provided so much information that I have used myself in my research that we wouldn’t have been able to collect otherwise. It’s only really been recently that underwater recording equipment has become more and more accessible cost-wise. You can buy a couple hundred dollar hydrophone (an underwater microphone) that you can just throw off your boat and listen on your iPhone. Cheaper options are coming out every day. Now is really the time to start creating a platform where people who are interested would be able to contribute and share data and information for their own benefits as well as research.
TEO: That’s fascinating. Given the growth in AI applications, is it conceivable that somebody could drop a hydrophone into the water, flip on an app, and figure out which species are within listening distance? They have fish sonar apps already that can tell the density of fish in the area but not species. Is that like down the road?
AUDREY: That is practically here right now for birds, and it’s only going to be a little bit longer before we’ll have that for marine mammals, and then for fish and invertebrates. It’s definitely happening.
There was a really cool project I got to see in action at Cornell where they had a recorder out in their forest, and there was a machine-learning program identifying the bird species’ call in live time on a computer monitor in their lab. It’s coming.
TEO: That is indeed incredibly cool. Thank you so much for your time and we’re all interested in what comes next.