Global Conservation through Sound

[Lisa Kopp] Welcome. So glad to have you joining today for webinar. We are going to get started in just one minute. We’re still seeing lots of people joining as we hit what is about to be noon in Ithaca, New York, where we are hosting this event from.

You are welcome to pop into the chat where you’re joining from. We always love to hear where people are Zooming in from. And we will we’ll get started. We have a really fascinating topic today, and I think you’re all going to really enjoy it.

All right, I am hitting noon, so I’m going to officially start. So welcome officially to this webinar hosted by the Lab of Ornithology. And today, we’re going to be discussing amazing research that’s coming out of the K. Lisa Yang Center for Conservation Bioacoustics. We’re going to have four panelists share updates on their research, which we will get to as soon as we can.

I’m just going to go through a couple of announcements as people are still joining and hearing where– and we get to hear from where you all are from. So I’m Lisa. I’m on the visitor center team at the Cornell Lab of Ornithology, and I’m just today’s facilitator and moderator. The stars are going to be the researchers that you’re hearing from.

And today, we’re joined by a team from, as I said, the K. Lisa Yang Center for Conservation Bioacoustics. And if you ladies want to pop on and give a wave, we’ll officially get everyone introduced in just a few minutes.

So as I mentioned, we’re hosting today’s webinar from Ithaca, New York, and although you’re going to hear about research being taking place all throughout the world, Ithaca is home to Cornell University, and I want to read a land acknowledgment, acknowledging the Indigenous people as the original inhabitants of this area.

Cornell University is located on the traditional homelands of the Gayogohó:no, the Cayuga Nation. The Gayogohó:no are members of the Haudenosaunee Confederacy, an alliance of six sovereign nations with historic and contemporary presence on this land. The Confederacy precedes the establishment of Cornell University, New York State, and the United States of America. And by reading this, we acknowledge the painful history of Gayogohó:no dispossession and honor the ongoing connection of Gayogohó:no people, past and present, to these lands and waters.

For those of you who aren’t familiar with who the Lab of Ornithology is, we are home to a community of researchers and supporters from all around the world, who appreciate birds and biodiversity and the integral role that they play in our ecosystem. Our mission is to advance leading-edge research, education, and citizen science that helps solve pressing conservation challenges.

So a few tech notes. We are going to be using– the closed captions are available on Zoom. If you need those, you can just scroll to the three buttons that say more on your Zoom screen and show transcripts.

For today’s program, each of our panelists are going to share an update on their research. And then at the end of all of the presentations, we’re going to take time for a Q&A. But throughout the time that the panelists are speaking, you are welcome to use the Q&A feature to add your questions. We’ve got some great people behind the scenes who can help answer questions. And then we will use the chat just for tech issues.

So you’re in for a treat today, and that you’re going to be able to get some really cool information on sound, which means that our panelists are going to be sharing sounds and videos. If, for any reason, you can’t hear it, let us know in the chat, and we can try to troubleshoot with you.

So let’s get started. So I just want to make a quick note, I’ve talked about how the Lab of Ornithology, we all love birds here. But birds are just a part of a much broader ecosystem. So birds can’t survive in isolation, and studying and understanding and protecting all biodiversity is really key.

So the Lab of Ornithology has a long history with acoustic monitoring and is home to the world’s premier institution for this kind of scientific study, the K. Lisa Yang Center for Conservation Bioacoustics. The Yang center explores the world of sound, from elephants, to birds, to whales, which we’ll get to hear some about today. But I will let our panelists get into all of that more in a minute.

So let’s say hello to our panelists. If you all want to turn on your videos and give a quick wave. We have Dawn, Sophie, Isha, and Loona joining us today, and we are going to start out with Dawn, who is going to be sharing a little bit of information on bioacoustics and the center for conservation bioacoustics broadly.

And then we are going to hear from Loona, Isha, and Sophie. So Dawn, do you want to kick us off and start out by sharing a little bit about who you are and how you landed in this world of research and sound?

[Dawn Parry] Yeah. Hi, everybody. I’m Dawn. I’m from Point Pleasant, New Jersey. I saw we have someone in the chat from the Jersey Shore, so hello. I’m from not too far away. I was lucky enough to get to go to a public high school for marine science in New Jersey, so that’s how I first got into marine science, in addition to just growing up at the Jersey Shore.

Then I pretty much knew I wanted to study whales. Didn’t know exactly how I wanted to do it. And then a scientist from NOAA came to talk to my undergrad college, and she gave this amazing presentation about bioacoustics and how you can know what animals are there and when to protect them just by looking at their sounds.

And that’s– I fell in love, and I was lucky enough to do an internship there and then that led me right through here to grad school. So now I’m a fifth year PhD candidate at the Yang Center, and I research fish and whale acoustics all for the purposes of conservation and management of the populations.

[Lisa Kopp] Amazing. I always love hearing the stories of how people ended up in the fields that they did. It sounds like you had a really early dose of this kind of work, and it stuck for you, which is wonderful. Loona, can we hear a little bit from you before we turn things over to Dawn?

[Loona Le Bourhis] Hi, everyone. I’m happy to be here. I hope you’re going to enjoy this. I’m Loona. I’m from France. I discovered the Yang Center a few months ago by doing my master’s thesis with them, and I work on the impact of climate change on bowhead whales.

I did this for my master’s thesis in ethology, and now I’m doing another master’s in bioacoustics in France. And I’m currently in Norway, so it’s all dark here, and yeah, I’m keep going, working on a new project on whales.

[Lisa Kopp] Wonderful. Thank you. Thank you for joining us from a very different time zone. Isha, can we hear about your background and what led you to this work?

[Isha Bopardikar] Hey, everyone. Thanks for having me over, and it’s really exciting to talk about our work. I’m Isha. I’m from India. And I got in touch with the Yang Center in 2015 when– I started studying marine mammals in 2012, and we started recording them just for curiosity and what they sound like.

And mostly I work with coastal animals, so we had some recordings of the dolphins along the Coast of India. And I got in touch with the Yang Center because I had no idea how to analyze all of these recordings. And then I came over for the sound analysis workshop all the way back in 2015, and then we’ve kept this association going.

And now I’m a PhD student at IISER in India, and I’m currently visiting the Yang Center as part of my analysis. So I’m happy to talk more about my research, studying coastal dolphins in India.

[Lisa Kopp] Amazing. Thank you. And Sophie, can we hear about your background?

[Sophia Trowbridge] Sure. Hi, everyone. I’m Sophie. I’m a research analyst with the elephant listening project here at the Yang Center. I’m also a veterinarian. My background is in conservation medicine, one health and planetary health, and I’ve– it’s been a dream of mine to work with the elephant listening project. I’ve loved this work for a very long time.

I originally got connected with them actually while in undergrad and did some senior thesis work with Elephant Listening Project then. And then went to vet school and got connected with many different conservationists and have ended up here after a circuitous route, but it’s really wonderful. The team is fantastic, and looking forward to sharing information about the forest elephant today.

[Lisa Kopp] Great. Well, thank you all for telling us a little bit about your story. We will say a brief goodbye to Sophie, Isha, and Loona while Dawn gives us a little bit of background information on the K. Lisa Yang Center for Conservation Bioacoustics and then talks to us about her research.

And I am also going to turn my video off and then I will pop back on to help us get to the next set of presentations. But in the meantime, I’m available in the chat and the Q&A to help answer things. But Dawn, I will go ahead and turn it over to you.

[Dawn Parry] Thank you, Lisa. Yeah, I’m super excited to talk to you guys today. So we’re doing a little eavesdropping today, eavesdropping on elephants, whales, and porpoises using animal sound to conserve species around the globe.

So let’s start with, what exactly is conservation bioacoustics? So bioacoustics is the study of animal sounds for understanding and protecting species and ecosystems. And conservation involves understanding and protecting biodiversity and habitats to inform effective management.

So acousticians, or scientists who study sound, like us, often use spectrograms to visualize sound. Spectrograms are graphs– so they’re graphs plotting the frequency or the pitch of sound over time. Most of our work is actually done by looking at spectrograms, not listening to the sound because different human-produced and animal sounds have characteristic appearances in spectrograms.

This is an example of a spectrogram of humpback whale song from my research in Bermuda. And humpback whales vocalize in a wide frequency range in a variety of patterns, ranging from complex repetitive song to other sounds like grunts, groans, and barks. Only male humpback whales sing. And let’s take a listen to what that sounds like. I’ll maybe click back.

[VIDEO PLAYBACK]

[GRUNTING]

[END PLAYBACK]

All righty. So if you go to the next slide– so what we get from our acoustic data is we get recordings of a soundscape. So a soundscape consists of all the sounds happening in the environment, produced by humans, animals, and other sounds like wind, rain, and seismic events. Here, we have a soundscape from Dzanga Bai in Central Africa, in which you can hear forest elephants as some of the low-frequency sound and birds and insects in the higher frequencies.

[VIDEO PLAYBACK]

[BIRDS CHIRPING]

[ELEPHANT TRUMPETING]

[GROWLING]

[CRICKETS CHIRPING]

[GRUNTING]

[SCREECHING]

[ROARING]

–

[END PLAYBACK]

[Dawn Parry] All righty. And we also have another soundscape from the Arctic Ocean, which contains bearded seal calls and some distant bowhead whale sound.

[VIDEO PLAYBACK]

[CALLING SOUNDS]

[END PLAYBACK]

[Dawn Parry] Yeah. So those are some great examples of some of the soundscapes that we collect here at the Yang Center. And the way we collect this soundscape data is through passive acoustic monitoring or PAM, a commonly used method in conservation bioacoustics. It’s a non-invasive way to monitor the environment by using recorders called hydrophones to record sounds from anthropogenic, which means human-produced sounds, like ship traffic, logging, and seismic exploration, to calls of fish marine mammals, birds, elephants, and more.

Hydrophones can be set into different types of acoustic recording devices. The main terrestrial recorder we use is called a SwiftOne, which can be strapped around a tree to record for several weeks or months. And we use several different types of underwater recorders, including the Glider and Rockhopper.

Gliders are mobile, so they can move through the water column at different depths and record sound over a wide area for weeks at a time. Rockhoppers are stationary recorders waited on the seafloor. They can be deployed for six months to a year, so they are a valuable tool for long-term data collection, especially in remote areas that would be too difficult and expensive to survey visually every week.

Visual monitoring of animals, like whale watching surveys or walking transect lines to observe birds, are also limited by light and weather conditions. So PAM provides a cost-effective way to collect non-invasive, long-term data, regardless of light or weather conditions.

So now we’re going to go into some applications of how we use PAM to conserve wildlife. We focus on the science on providing the best possible information to decision-makers and the public. So my research revolves around whale conservation in Bermuda, where the government has a plan to protect 20% of their waters and needs data on marine mammals to inform those protections.

We recorded data for nine months at the same site as previous studies were done in 1966 and 2013 to examine how sound levels changed over time. We’re studying which whale species are present to inform Bermuda when each species is in their waters and how human noise might be affecting these species over time.

So we have a few target species for this project, and I’d like to show you how some of them sound. So we’re looking at baleen whales and toothed whales. Baleen whales have plates made of keratin, like our hair, to filter food out of the water instead of teeth for catching prey. Most baleen whales vocalize in low frequency ranges, with a low number of call types. Sei whales produce down-sweeping calls.

[VIDEO PLAYBACK]

[WHALE CALLING]

[END PLAYBACK]

And Fin whales produce 20-hertz pulses.

[VIDEO PLAYBACK]

[WHALE CALLING]

[END PLAYBACK]

[Dawn Parry] So like I said, super low frequency, pretty quiet. And toothed whales usually have much more variety in their call types. One example are the beaked whales, which are not well studied, so we’re not always sure which call types can be attributed to which species, but they communicate with clicks.

[VIDEO PLAYBACK]

[CLICKING SOUND]

[END PLAYBACK]

[Dawn Parry] And dolphins, like this Atlantic-spotted dolphin, communicate with high-frequency buzzes, whistles, and clicks.

[VIDEO PLAYBACK]

[WHISTLING]

[END PLAYBACK]

[Dawn Parry] All righty. So from those nine months of soundscape recording, I have acoustic presence results for a few baleen whale species shown here. So acoustic presence means whales were present in the area vocalizing and picked up by the recorder. One limitation of acoustics is that we can’t detect animals that might be in the area but aren’t vocalizing.

So we saw humpback whales were present December through mid-May and Fin whales were present through mid-April. Sei whales had rare occurrences in December and January. This humpback data is very consistent with what has been found in the past but less research has been done on Sei and Fin whales. So it’s great to be able to provide that data.

I’m still working on my toothed whale analysis, looking for sperm whales, beaked whales, and dolphins. So there are many more results to come. And thank you guys so much for listening. I’m happy to turn things over to Loona, who’s going to take us to the Arctic.

[Lisa Kopp] Thank you, Dawn. We’ve got some really great questions in the Q&A for you to think about when we have time after everyone’s presented. And Loona, you want to hop on, and we can hear about your work? I’ll turn it over to you. Thank you.

[Loona Le Bourhis] OK. So let’s go– Dawn, thank you. To present the soundscape of Arctic, I’m going to dig into details and talk about listening under the ice the sound of bowheads during the spring migration. So OK.

To give you more information about– general information about bowhead whale, they are endemic Arctic species. They are in the Arctic Ocean all year round, and they do migration during spring and fall. They are also very dependent on sea ice.

Today, we’re going to focus on the Western Arctic population and on the spring migration. So this population leaves the wintering breeding grounds to reach out the summer feeding grounds in early spring. And previous colleagues’ works on assessing the state of this population after an era of heavy commercial whaling.

In the mid-19th century, as you can see on this population dynamics curve, the bowhead population drop a lot with this commercial whaling. Before, there were about 10,000 to 23,000 individuals, and after this intense activity of hunts, there were only less than 3,000 individuals.

So the project began at the beginning of the recovery for the population, and they were doing census, so they were counting how many whales remain after this heavy past. So it was in the ’80s. It’s an initiative by the native people of Alaska, and it was a great collaboration between Arctic communities and scientists of Point Barrow, that it’s a very good strategic point along the migration.

So native people of Alaska have been subsistence hunters for centuries, and they are very linked to bowhead. They are following their migration with attention. They also follow subsistence catches numbers developed by the International Whaling Commission. And those quotas to respect are based on the state of the population by using sensors, for example.

So this project, it’s using passive acoustic monitoring as Dawn presented to you. So it’s a real bioacoustic journey through generation. This project began in the ’80s until 2011, so you can be witness of almost the evolution of the science of bioacoustics across those three decades. And today, we are bringing back to the spotlight this incredible historical project to answer all the question about bowheads.

Now, we are working on the impact of climate change on the migratory phenology of this species. We are digging into the relationship between bowhead whales and sea ice cover because Arctic is an environment that’s changing very rapidly.

So how do we do that? We are using the sounds of bowhead to assess their presence along the migration. And you can split the vocalization of bowhead into two big categories, the song– like humpbacks, they do song. It’s a broadband sound. It’s a unit organized in structure into repeated phrases over time.

And also social calls that are generally under 500 Hertz, so pretty hard to catch. I have a cool example for the first category, the song.

[VIDEO PLAYBACK]

[HOWLING]

[Loona Le Bourhis] And obviously, you have bearded seal on the background, too.

[HOWLING]

[END PLAYBACK]

[Loona Le Bourhis] For social calls, I hope you have good headphones because they are quiet. They are difficult sometime to catch because they are very low-frequency. So I hope you are going to hear them.

[VIDEO PLAYBACK]

[LOW HOWLING]

[END PLAYBACK]

[Loona Le Bourhis] Very low frequency. So we annotated manually the sound of bowheads, but it’s a lot of work. It’s very time consuming because this species is very talkative. So we are in the process of building an automatic detector by using different call types, different categories of bowhead sounds.

And when we have access to the presence of bowheads with sounds, we can wonder how the whales vocalize during their migration. And when do they sing, for example? So in the spring, you can look at the presence of bowheads per hour per day.

And when we want to look at the question related to sea ice, we have just first results, preliminary results, because we analyzed just one year for the moment. But we found that when you have a high acoustic activity of bowheads, you often have a high concentration of sea ice.

That’s interesting. That can help us to do some hypothesis, like maybe bowheads have a preference for high concentration of sea ice during the migration, and this could be interesting for them because big packs of sea ice can be a good shelter from predators and also a good spot for food availability.

So the research goes on. We have an entire database to investigate. We have to do this multi-year analysis to be able to look at the impact of climate change on this three decades database.

But we already can have some insights on the impact of this climate change and this loss of sea ice. If this preference of bowheads for high concentration of sea ice exists, it’s pretty alarming because the sea ice is decreasing a lot and more earlier every year in the Arctic. This decrease of sea ice is also bringing new opportunity for other species that come in the Arctic, and they can come more earlier in the Arctic.

So it’s some competitors for bowhead whale for foods, for example humpbacks, or some risk of predation by killer whales, for example. And it’s also opened the access to human activity. For example, you have the increase of shipping traffic.

So all this is very important to study. We already know that the timing of migration has already changed for bowheads, becoming more and more earlier for spring migration. And this relationship with Arctic communities is very important.

So it’s important to well-known species to come back in time to understand this big phenomenon, these big changes, and to be able to have a correct management of the population to ensure the survival of the species but also the survival of the people who depends on this species. So thank you for your listening. I want to thank all the people who participate and help on this project. And if you have any question, I’m happy to answer at the end.

[Lisa Kopp] Thanks so much, Loona. It’s fascinating. And we’re going to stick with marine life for one more presentation with Isha, and then we’re going to switch gears and hear about elephants last. So Isha, take it away.

[Isha Bopardikar] Hey, everyone. Thanks for having me over. And that’s a picture from my field site. And as you can see, I work very, very close to the shore, and we’ll be talking more about that. So India, with a very long coastline, almost 7000 kilometers long, has about 26 species of marine mammals, and this is a wide range, from large baleen whales to bottlenose and spinner dolphins.

And they’re spread out across the marine ecosystem, so you have oceanic species that are found deeper off shore, and then there’s exclusively coastal ones, like the humpback dolphin and the Irrawaddy dolphin. And these coastal species are found in some of the busiest waters. Our shorelines are extremely populated. There’s a lot of anthropogenic activity that happens in these areas, right from industrial development to shipping, and a lot of intense fishing on different scales. You have small scale and large scale fishing.

And fisheries and fisheries interactions are immediate conservation concerns, especially as they become more and more unregulated and exploitative. It’s leaving the small scale fishers behind, and it’s bringing a lot of industrial large scale operations closer to shore. And my field site along the Southern West Coast of India is one such area, where we have a huge dominant fisheries industry, but it also has two resident species of marine mammals.

The humpback dolphin– and I’m talking about them first very briefly because I want to show you how easy they are to spot. They’re very big. They’re very surface active. And you can see them very clearly when it’s calm and when it’s not that calm.

The other one that I will talk about more is the finless porpoise. This is a very small animal. The adults are around 4 feet long. They are very shy, and that’s all you can see of them. The first time I saw one, I thought it was a piece of plastic floating in the water until it moved a few times, and then another one popped up and disappeared very quickly.

And throughout the range, finless porpoises are found right from waters of Iran and the Persian Gulf all the way to Japan. And everywhere they’re found, they are threatened by a lot of changes to their habitat. We have a lot of reports from finless porpoises with fishing gear entanglements, and all of this has pushed for a lot of research on this species.

So it’s becoming important to know where they are found, how they are distributed through the areas where they are found, and when they are present in these areas. And it’s crucial more and more to know how many animals there are present in an area.

But how do you do this for a species that you can barely see? It’s a bit tricky, but they are excellent candidates for acoustic monitoring. One [AUDIO OUT] use sound to find food, to communicate with each other, to sense their environment.

They make these high-frequency echolocation clicks that are very easy to pick out from anything else in their environment. They share their habitat with humpback dolphins, which produce all kinds of sounds. Like Dawn mentioned, dolphins whistle, click.

[VIDEO PLAYBACK]

[WHISTLING]

[Isha Bopardikar] So that’s a clip of the humpback dolphins. And if you can see, the finless porpoises only use echolocation clicks. So this makes them very easy to spot acoustically.

[END PLAYBACK]

[Isha Bopardikar] And we don’t really listen to them because it’s all high-frequency. It’s happening above our frequency hearing range, so we use spectrograms, like Dawn mentioned. And it’s very easy to pick out finless porpoises from the other species.

They also vocalize constantly because of the environment they’re in. It’s extremely murky. It’s turbid, so they use sound for navigation. Their eyesight is not really that great. And once you have acoustics, you’re not limited by weather conditions. It can get really windy on the boat. We have a very small vessel, but acoustics works just fine.

So what was the goal? We wanted to study finless porpoises, to count them, and know their distribution across an area using acoustic monitoring. How do we do this? We do visual and acoustic line transects. And for the visual part, you have your observer sitting out in front, looking for any signs of the animals.

But for the acoustic part, we use a long tube, in that we have multiple hydrophones, and they’re all connected to a recorder. And these hydrophones are laid out in a particular pattern. Here, they are all arranged in one line.

And when we do our visual surveys, we tow the hydrophone behind the boat, and it acts like an independent observer. It’s not getting tired. It’s constantly listening to its environment without any interruption, and it’s working irrespective of any external weather conditions. And that’s how we tow it behind the boat, so you can see how it works.

So with this acoustic setup, we can actually localize the animals we record. We do this using two methods. First, we get a series of bearing angles from the echolocation clicks of the animals from the different time of arrival at each of the hydrophones. And as the boat moves and the recorders move along the survey track, and because we are faster than the animals themselves, these bearings that we record along the track, as they pass the animal, they merge in space. And the point where they merge is where your animal is located.

And we did about 38 surveys for maybe more than 100 hours, and we covered thousands of kilometers in survey effort. And what we found was we always knew that the acoustic data would be more for the finless porpoises, but we didn’t think that we would have three times more detections from the acoustics rather than the visual surveys alone.

So we had about 39 visual detections, whereas our acoustic detections were about 149. And most of these acoustic detections were also recorded with the visual component, but it was not the opposite. And this is just a map of all our sighting effort.

And we use this data to create distribution maps through three years of survey, and we wanted to see if there are any changes in how many encounters we had through these years and how much change there is. And we do notice that there is some change and some reduction in encounters, and we need to look into this more, on why these changes are happening, whether there’s something seasonal or whether it’s something within the habitat that has changed with increasing anthropogenic pressure.

But we also found that some of the high-density hotspots for these animals stay the same. So it could be that these areas are important for these animals, especially for foraging, and this needs to be looked at with more detail, so that they can be placed into a conservation setup.

And while we map animal distribution, we also look at human-induced threats in the area. For finless porpoises, they overlap extensively with fishing grounds, and we wanted to look at how much of an overlap there is with fisheries and the animals themselves. So mostly, this is with large-scale operations, like purse seiners or trawl fishing, which is a bit ecologically destructive.

And now that we have this information of the animal distribution and the fisheries grounds, we aim to take this data back to the fishing community who we work with to come up with some conservation measures together. And that’s a picture of my team. Well, thank you, and I’m happy to take any questions.

[Lisa Kopp] Thank you, Isha. We’ve got some great questions in the Q&A that we’ll be able to get to after our final presenter. Sophie, do you want to take us to the elephants? Thank you.

[Sophia Trowbridge] All right. Awesome. So thank you to Dawn, and Luna, and Isha for– I could listen to your research all day. That’s awesome. So I’m going to be taking us out of the water and onto land. So we’re going to be talking about the African forest elephant today.

So this group, the Elephant Listening Project, is part of the K. Lisa Yang Center at the Conservation Bioacoustics lab at Cornell, and we’re specifically studying the African forest elephant. So when people think about African elephants, most often, they’re thinking about the African savanna elephant, Loxodonta africana.

But there are actually two distinct species of African elephants. There’s the African savanna elephant and there’s the African forest elephant. The African forest elephant lives in the remote rainforests of West and Central Africa, as opposed to the open savanna. And because of the dense rainforest that African forest elephants live in, there’s very little known about their biology, which is what makes acoustics such a perfect analysis technique for better understanding the conservation needs of this species.

So here is a spectrogram that Dawn had introduced in the beginning. So again, this is a visualization of sound, where we have frequency on the y-axis here and then time on the x-axis. And we’re going to play this in a second. This is a visualization of what we call rumbles.

So elephants make lots of different sounds. They make roars. They make trumpets, which are some of the sounds that people are most familiar with with elephants. But they’re actually bread and butter sounds. The sounds they make the most are these sounds called rumbles, and they sound like this.

[VIDEO PLAYBACK]

[ELEPHANT RUMBLES]

[Sophia Trowbridge] So very low-frequency. The low-frequency sounds travel farther, and they can communicate over greater distances.

[ELEPHANT RUMBLES]

You hear some sloshing around in the water there, too.

[ELEPHANT RUMBLES]

So elephants, and specifically forest elephants, are a keystone species for this area of the world. We refer to them as architects of the forest because they play a crucial role in seed dispersal of many tree species throughout West and Central Africa.

Their uniquely large bodies enable them to ingest very large seeds. They’re primarily frugivores, elephants, which they then would disperse across their large home ranges, covering vast areas of forest. In fact, there are some tree species that rely entirely on the forest elephant to germinate because unless they’ve been passed through an elephant digestive system, those tree species cannot grow and continue to be successful.

A lot of these tree species are globally important because they trap much of the carbon that would otherwise be released as CO2, which, in turn, affects our atmosphere and can accelerate rising global temperatures. So I always say if you live on planet Earth, and you breathe oxygen, you care about forest-elephant conservation.

[END PLAYBACK]

[Sophia Trowbridge] So forest elephants under threat. For the past two decades, poaching for ivory has caused the elephant populations to plummet. And because of this, the forest elephant has listed in the IUCN Red List as critically endangered. The forest elephant also faces significant habitat loss due to human development, logging, and fragmentation of landscapes.

And because, again, as I mentioned in the beginning, they live in these dense, difficult-to-get-to tropical rainforests, they’re very difficult to see and find. So their conservation needs are poorly understood because of this dense vegetation. And again, it’s also why passive acoustic monitoring is a perfect technique for being able to better understand the forest elephant.

So the project that I’m primarily working on as one of our study sites is in the North Republic of Congo, where we work with the Wildlife Conservation Society to maintain a landscape scale passive acoustic monitoring survey in the Nouabalé-Ndoki National Park, you see here on the map, that’s adjacent to logging concessions which are just South of the park, which you’ll see there on the map on the left.

Our survey area covers about 1,200 square kilometers, with 50 acoustic units that are in active use. And so far, this project has been going on successfully with continuous acoustic recordings at these sites for almost six years, which is a huge feat in terms of quantity of data.

So how this works. So you see here on the left side, there’s a green– sort of green box and then also black box. These are known as the rugged swift. So these boxes protect the recording units, the Swift recording units that are inside. You can imagine in a tropical rainforest, there’s lots to protect from, from humidity, rain, termites, also humans and elephants to a degree.

These boxes have to be installed pretty high up in trees because if humans or elephants can reach them, they will attempt to destroy them. I actually have– I don’t know if you still see my video on, but I have this box here. This is one of our previous boxes. This hole here, that’s a tusk hole. So if an elephant can reach a box and pull it down, they are infinitely curious [AUDIO OUT] with your research if you’re not careful.

So every four months, in this research site, our amazing Congolese WCS field team visits every single recorder, so every of those 50 recorders I spoke about, to replace batteries, swap out memory cards, and then load the sounds onto our hard drives, where they perform some of the initial sound analysis in the field. They also send that data to us at the Yang Center for continued sound evaluation.

And this wide-scale analysis directly informs anti-poaching protocols and management within the national park that can make this work more efficient and effective. So I’m going to play another sound here. The sound on the lower left is a very high-quality recording from Dzanga-Ndoki National Park of two related females greeting each other.

So you could sort of interpret this as two family members who haven’t seen each other in a while, and you’ll see there’s a lot of overlap in sound. And I would challenge you when the sound– you’ll see there’s a green line that’ll go across with the sound. And when the sound gets to this area here and this area here, see if you can actually hear the sound.

We can visualize it, but see if you can hear it because sometimes, most of the elephant sounds are made in the infrasound range, which are pitches below that humans can hear. So if you can hear that, I’ll be really, really impressed.

[VIDEO PLAYBACK]

[RUMBLING]

[Sophia Trowbridge] And then here.

[RUMBLING]

And again.

[END PLAYBACK]

[Sophia Trowbridge] Yeah, I can’t hear that, but if you can, kudos to you. So how we use this information. So these data allow us to visualize how elephants use entire landscapes and how this changes over time. So let me just pull this map up.

So this map here on the left shows the average number of rumbles recorded at each site per month. The more rumbles– so the pink into almost white color indicates more rumbles, while the light green into dark green is fewer rumbles. So this is really great information that allows us to see where elephants are over time and how they’re using the landscape.

And then on the right, we can see with acoustics, we’re able to provide unbiased information on when and where gun hunting has taken place. This map, we can see both elephant activity and gunshot activity that were recorded between April and May of 2018. The darker the blue, the more elephant presence, and the larger the red circle, the more gunshots that were detected in an area.

We can present maps and information like this to national parks, which can directly inform anti-poaching management. And we can also feed these data into our own statistical models to investigate how these elephants use their landscape and relationship to season, habitat type, and human disturbance.

So we can learn a lot from bioacoustic monitoring. And here, I’ve briefly talked about some of the applications, such as understanding forest elephant, landscape use and movement, understanding their vocal behavior and changes in behavior over time. We can also use it to understand age class and structure within a population. We can monitor for illegal gun hunting and evaluate the effects of our anti-poaching patrol efforts with our partnerships in the field teams on the ground.

So all of these pieces of information are deeply important for elephant conservation. And as we’ve chatted about today beyond forest elephants, there are so many ways passive acoustic monitoring can be beneficial for species monitoring and conservation.

And if any of this work has sparked your interest or you’d like to learn more and help support our continued work, feel free. And please check out our website through the K. Lisa Yang Center for Bioacoustics. That’s also my email address if you have questions and you want to follow up there, skt24@cornell.edu.

And I can’t leave today without sharing one final video. This is a forest elephant calf. This is a young female calf in Dzanga Bai, one of our other field sites. So she’s probably a week to two weeks old.

The calves are very infinitely fun to watch. They’re very social. They’re very playful. There’s a lot of– oh, she’s making some noises. There’s a rumble. She’s going to catch up with her mom in a moment. But thank you so much for your attention and for listening, and we’re going to take some questions in a moment.

[Lisa Kopp] Thank you, Sophie. Thank you, everyone, Dawn, Loona, Isha, if you guys want to hop back on for your amazing presentations. You have definitely sparked many, many questions with the audience, so we will spend the next 10 minutes or so answering as many of them as we possibly can.

The good news is that quite a few of them are sort of themed, and we will try to answer as many as we can. So Dawn, there were a couple of things that popped up when you were talking just about the basics of bioacoustics.

So there was one question about, what’s the difference between bioacoustics and a spectrogram? Just to get down– get us down with those language– those important differences.

[Dawn Parry] Yeah. So bioacoustics is the field of study. So it’s kind of the process and study of recording animal sound. And a spectrogram is one of the visualization techniques we use for sound. So it’s, bioacoustics is the field and a spectrogram is one of the tools, but also one of the products from the research.

[Lisa Kopp] Right. Thank you. And another thing that came up quite a bit is just this realization that the ocean is quite noisy. So Isha, Dawn, or Loona, I’d love to hear your interpretation of the fact that it is quite noisy. And then the follow-up question that a few people asked was, how do you filter out unwanted sound when you’re studying?

So Loona, in your case, ice, is it very noisy when you’re recording and there’s ice movements or breakage? And then, how can we help animals that are trying to communicate in a noisy ocean that has become far noisier because of human activity? So that’s a three-parter. So I don’t know how you all would want to split that up, but I’m open to whoever wants to take it.

[Dawn Parry] I can start with just the general noise and then we can pass it off to somebody else. Yeah, the ocean is very noisy. It’s pretty much getting louder all the time. There’s lots going on. A major source of noise is shipping traffic, so shipping vessels, cargo vessels. This is a big issue, especially around major ports in the US, but it’s also even an issue offshore.

Another big source of noise is seismic exploration for oil and gas that uses basically big– not harmful explosions themselves, but air bubble popping, which makes really, really loud noises. So that’s another source of noise. And yeah, there’s just a lot of human activity, and these activities, especially the oil and gas drilling, it’s being as technology improves, they’re able to move further and further and further offshore.

So they are heading into ecosystems they haven’t been in before. And I actually have a project in the Gulf of Mexico looking at the more off shore ecosystems and seeing how much quieter ecosystems are during hurricanes as a time to see how quiet the offshore ocean can be when there’s no human activity because that’s when the seismic surveys stop and that’s when the vessels aren’t in the areas, when there’s a hurricane in the area.

And we’re seeing that it can be– I think seismic noise can make the ocean over 16 times louder than when it’s not in the area.

[Lisa Kopp] Oh, my goodness. Thank you for that sad and scary but really valuable information and for the work that you’re doing to try to figure that out. Loona, do you want to speak a bit about sea ice and how you factor that into your research and analysis?

[Loona Le Bourhis] Yeah. Yeah, in the Arctic soundscape, it’s pretty buzzy, too. It’s buzzy with biophony, all the sound from animals. As we saw with Dawn, you have the bearded seals mixing with bowheads and sometimes, it’s a little crazy to find the difference.

For sea ice, I never find big sea ice noise. Normally, it’s a very long and very slow noise. But yeah, sometimes, it’s really difficult. It’s like a decision tree process. You find a sound, and you are trying to, OK, maybe it’s a predator, maybe it’s a whale. Is that a living animal or not? And so it’s very difficult.

I work a lot with Ashak, and he’s working on bowhead whales since so much here, and even him, sometimes, it’s really difficult to tell the difference and to be sure. So yeah, it’s a lot of work, a lot of practice, and yeah. But it’s very interesting at the same time because it’s always different.

[Lisa Kopp] Great. Thank you. Sophie, we had a comment or a question about the rumblings and whether they’re considered more like a warning to have either other elephants or people to stay away or if it’s sort of like a cat purring and it’s like a welcoming sound. And you played that recording of two elephants sort of speaking to each other and like a greeting, even though it was a deep rumble. So are you able to differentiate when it’s sort of a stay away or a warning call versus a greeting rumble?

[Sophia Trowbridge] Sure. Yeah. So actually, rumbles are– I wouldn’t necessarily associate them with warning at all. We associate, as humans, lower frequency sounds with scary things, but in the elephants’ world, that is their frame of reference for speaking. So lower frequency sounds would be like me talking to you today. That’s their everyday conversational frequency of speak.

The more alarm or warning-based sounds would actually be something like a trumpet or a roar, so sort of those higher-frequency sounds that we associate more with elephants. So the trumpet sound coming from the larynx and through the trunk comes out that way versus a roar that’s coming from the larynx, and they’re opening their mouth and they’re literally– it’s like yelling. It’s like elephants yelling versus a rumble, again, still coming from the larynx, but their mouth is closed, and you can think about it almost like in instruments, when you see a violin versus a cello versus a bass.

Because that instrument gets bigger, that frequency of sound can get lower. And the elephant can make that sound so low surely because they are so big. If they were small, they would not have the physiologic capacity to make that sound. So the rumbles– actually, the rumbles give me sort of a heartwarming feeling because it’s their every day talking.

[Lisa Kopp] Their conversational tone?

[Sophia Trowbridge] Conversational tone, yeah.

[Lisa Kopp] OK, great. Isha, we’ve gotten a couple questions about leveraging artificial intelligence and using it to analyze sound and then maybe potentially speed up efforts to protect marine life specifically. I know that you mentioned that you have been working on recording and analyzing sound recordings since 2015, and I wondered if you could speak a little bit to what goes into analyzing the sound recordings, and if you are either using or exploring or totally against the idea of artificial intelligence in furthering this work.

[Isha Bopardikar] Yeah. I’m not against using AI at all. It definitely has helped speed up a lot of things if you know how to use it well and if you have enough data to train those networks really well because they need a lot and a lot of data. Like how bird networks, we have tons and tons of ground truth data that goes into building these models, and they are made really well.

So you can use those to speed up a lot of the data processing. It helps you analyze months and months of data relatively faster rather than going through it all manually. And yeah, you can have a lot done very quickly, which can speed up your conservation efforts or help direct those efforts in a better way.

So yeah, definitely has helped quite a bit. I haven’t used much of it, but yeah, it is being used, and it’s the way to go.

[Lisa Kopp] Yeah. Yeah, it can make it so much faster for the important work you all are doing. We only have two minutes, which is hard to believe. So there’s quite a few questions about using bioacoustics to recognize and track specific animals, so identifying a single marine mammal or a single elephant. Would you all mind speaking to whether that’s being done or nearly impossible to do? Sort of just how that is possible or how that’s being used within your field. Can we– we can just kick it off with Dawn, if you don’t mind.

[Dawn Parry] Sorry, what was that one more time?

[Lisa Kopp] Oh, sure.

[Dawn Parry] I just wanted to review some of our last questions.

[Lisa Kopp] No. So I think part of what I think is so wonderful about your presentations is that marine mammals and elephants are, I think, animals that people feel really connected to. They feel like animals that you want to understand more about on a personal level.

And so there were a few questions about identifying individual animals from their sounds and whether you’re able to track– record and track and identify specific animals based on their calls.

[Sophia Trowbridge] Yeah. So we see this in some whale species, I believe mostly toothed whales. I think orcas and dolphins have– they kind of have equivalents of saying, this is where I’m from. This is the pod that I’m from. They have identifying calls for their pod.

I believe dolphins also have what they call signature whistles. So that’s kind of like their name. So if you hear that signature whistle, you know which animal is vocalizing. I believe sperm whales also have something similar. I think they have certain click patterns called codas that can convey individual identity, as well.

I’m not really sure of any progress on that from the baleen whale side. It makes it a little more challenging. They’re not as socially communicative in that exact way. Something that we see– part of what makes that complicated is with humpback whales, males in the same population all sing the same song each season, so it’s really hard to see who is who.

So that’s a little less done there. And in general, we, at the Yang Center and conservation bioacoustics, in general, is more focused on large ecosystems and populations of species. So individual tracking through sound is definitely possible in some of the ways that I just laid out. That’s not our specific main focus, but definitely does have its uses and gets looked into for sure.

[Lisa Kopp] That’s great. Thank you. That’s so helpful. I’m so sorry to say that we are a little over time, actually, and I want to get to a few announcements for the attendees. So first of all, huge thank you, Dawn, Isha, Sophie, and Loona for presenting this research. It’s inspiring and always a little harrowing to know what these animals are up against, but thank you for what you’re doing to try to solve these problems.

It’s really incredible, and I hope that it’s sparked something for those of you who are here in the audience. There is a ton of wonderful information on the K. Lisa Yang Center for Conservation Bioacoustics website, including information about how you can become a part of the work that people are doing and support what they do.

You, as Zoom registration attendees, will get a follow-up email with us, too, with a recording of this talk. And we’ll also include some helpful links for you to be able to explore more and keep in touch with the science that these people are doing.

And we hope that you’ll join us for our next round of webinars, and thank you, again, so much to Sophie and Dawn and Isha and Loona for your work. Hope everyone has a great rest of your day. Bye.

End of transcript