Humpback whales sing different parts of their songs at different depths - Earth.com

Marina: Welcome to Deep Voices — I'm Marina, a marine biologist and absolute lover of all things whale and dolphin, and this is the show where we dive beneath the surface to explore the songs, the secrets, and the stunning intelligence of the ocean's greatest communicators. Marina: Before we get into today's stories, a huge thank you to our sponsors — Ada, who helps businesses integrate A I into their real workflows in meaningful ways; Ago Consulting, pronounced "Ahh-Go," your partners in silicon development from A I all the way to SoC; and Zen Semiconductor, an umbrella company with a remarkable portfolio spanning the silicon industry. We couldn't make this show without them. Marina: And today — oh, today is a good one. We've got humpback whales singing in layers, literally at different depths. We've got sperm whales going quiet around ships, which honestly says so much. We've got an underwater robot eavesdropping on whale conversations in real time. There's a recording from 1949 that nobody knew was the oldest whale song ever captured. And there's a story about dolphins in distress that is hard to hear but so important. Let's go deep. Marina: So let's start with VOICES OF THE DEEP — and I want to paint you a picture. Imagine you're floating somewhere in the Pacific Ocean. It's dark. It's vast. And somewhere below you, a humpback whale — a male, probably — is singing. And when I say singing, I don't mean some gentle background hum. Humpback song is one of the most complex, structured, repetitive, and genuinely haunting acoustic performances in the animal kingdom. These songs can last for hours. They have phrases, and themes, and they repeat in sequences. They evolve over time across entire ocean populations, like a hit song spreading across a continent, except the continent is the sea. Marina: Now, researchers have recently been looking more carefully at where exactly in the water column humpbacks sing different parts of their songs — and what they found is genuinely fascinating. It turns out humpback whales appear to sing different parts of their songs at different depths. They're not just floating in one spot belting it all out at the same level. They move. They change depth. And different song components seem to correspond to different positions in the water. Marina: Why would that matter? Well, sound travels differently at different depths in the ocean. There's actually a layer called the SOFAR channel — the Sound Fixing and Ranging channel — where sound can travel extraordinary distances because of the way water temperature, pressure, and salinity interact. It's like a natural acoustic superhighway. And humpbacks have been using the ocean's physics to their advantage for millions of years, long before we had any idea any of this was happening. Marina: The idea that they might be tuning not just the frequency or rhythm of their song, but actually the depth at which they deliver it — essentially choosing the right acoustic environment for each part of their performance — that is a level of sophistication that still gives me chills. It suggests these animals have some kind of internalized understanding of their acoustic environment. They're not just making noise. They're engineering their sound. Marina: And just to sit with that for a second — we are talking about a creature that can weigh up to 40 tons, that lives its life in open ocean, that has no hands, no tools, no technology — and it is doing something that a recording engineer would recognize as spatial acoustic optimization. The ocean is their studio. And they know it better than we do. Marina: Now here's something that connects beautifully to that sense of acoustic sophistication — and it's a story that genuinely stopped me in my tracks when I first read about it. It comes from researchers studying sperm whales, and it involves ships. Marina: Sperm whales communicate using clicks — rapid, percussive patterns called codas. These codas are social. They're used between whales in a group, and different whale clans around the world have distinct coda repertoires — almost like dialects. A sperm whale from the Pacific doesn't click quite the same way as one from the Caribbean. It's cultural identity encoded in sound. Marina: But here's what's new and what's important. When ships pass nearby, sperm whales change how they communicate. They alter their behavior — researchers have found that the presence of vessel noise affects their coda exchanges. And this isn't a small thing. Ship noise in the ocean has increased dramatically over the last century. The ocean is louder now than it has ever been in recorded history. Low-frequency vessel noise can travel hundreds of miles. And the whales are responding. Marina: What exactly they're doing — whether they're going quieter, changing the timing of their clicks, spacing out their conversations differently — is still being actively studied. But the fact that they respond at all tells us something crucial. These animals are not just broadcasting into the void. They are listening. They are attending to their acoustic environment. They notice when something has changed. And they adjust. Marina: That is not reflexive behavior. That is something much more interesting. That is awareness. And for animals whose entire social world — family bonds, cultural transmission, finding a mate, coordinating a hunt in the pitch black deep ocean — depends on sound, the intrusion of our noise into their world is not a minor inconvenience. It is interference in their most fundamental way of being. Marina: I think about that a lot. The ocean we've made — noisier, busier, more industrial — and the whales still in it, still trying to talk to each other, still trying to hold their families and their cultures together in the din we've created. Marina: Alright, let's move into our second thread — CULTURE AND DECODING — because this is where things get genuinely extraordinary, and where the science happening right now feels like the beginning of something historic. Marina: I want to start with a piece of technology that sounds like science fiction but is absolutely real. Researchers have deployed an underwater robot — an autonomous vehicle — that is capable of tracking sperm whale conversations in real time. Not recording and analyzing later. In real time. The robot listens for clicks, identifies individual whales, and follows the acoustic signatures of their exchanges as they happen, in the open ocean. Marina: This is part of a broader scientific effort — you may have heard of Project CETI, which stands for Cetacean Translation Initiative. It's an ambitious, cross-disciplinary project that brings together marine biologists, linguists, cryptographers, and A I researchers, all united by one audacious question: can we actually decode what sperm whales are saying to each other? Marina: The approach is, at its core, a machine learning problem of staggering scale. Sperm whale codas — those click patterns I mentioned earlier — are structured. They have variation. They appear in social contexts. They may carry meaning. Project CETI's goal is to collect an enormous corpus of sperm whale communication — we're talking potentially billions of clicks — and train A I models to find patterns, context, and ultimately meaning, the same way large language models learned to understand human text by being trained on vast amounts of it. Marina: Now, the honest scientific position is that we are early. We don't know what sperm whales are saying. We don't know if their codas constitute anything like a language in the way humans use language. But we know their communication is complex, it's structured, it's social, and it varies culturally. Those are all features that, in any communication system, suggest there is something there worth listening to. Marina: And this is where the 1949 recording comes in — because it is a beautiful reminder of how long humans have been trying to hear these voices, and how much history we've been missing. Marina: A recording was recently identified as the oldest-known recording of a whale song. And it dates to 1949. It was made by a bioacoustics researcher named William Schevill and his wife and collaborator Barbara Lawrence, who were recording underwater sounds in the waters off Massachusetts. They captured what turned out to be the sounds of a beluga whale. The recording sat in archives for decades, largely forgotten, its significance not fully recognized until researchers revisited it recently and realized what they had. Marina: Think about that timeline. The recording most people associate with the discovery of whale song — the famous recordings that Roger Payne and Scott McVay used in their landmark 1971 paper on humpback song — those came more than two decades later. And before those recordings were released to the public, most people had no idea that whales sang at all. No idea. The ocean was silent to us — not because it was quiet, but because we weren't listening. Marina: The 1949 recording reminds us that there were scientists listening, even then. And it reminds us how much of the acoustic world of cetaceans remains uncharted, uncollected, unexamined. Every archive that gets revisited, every new underwater microphone deployed, every A I model trained on whale clicks — it's all part of the same long effort to finally, truly, listen. Marina: And now for our third thread — A CETACEAN-AND-PEOPLE STORY. And I have to be honest with you: this one is not easy. But it's important. And I think it belongs here, because this podcast is about the full truth of the relationship between cetaceans and humans — the wonder and the harm and the responsibility. Marina: Researchers recently achieved something that is a genuine first in the scientific record. They captured audio and video of dolphins in the immediate aftermath of becoming entangled in fishing nets. And what they recorded is — it's hard to describe without feeling the weight of it. The dolphins were vocalizing. Under acute stress. In a way that researchers had never been able to document before in this precise context. Marina: Entanglement in fishing gear — what's known as bycatch — is one of the leading causes of dolphin and whale mortality worldwide. It happens in near silence, from our perspective. The nets are out there, the animals encounter them, and most of the time, we never see what happens. This recording is one of the first times scientists have been able to document the behavioral and vocal response of a dolphin in that exact moment of crisis. Marina: What the recordings show — the vocalizations, the physical behavior — gives researchers a window into the distress response of these animals in a way that has direct implications for how we might detect entanglement events before they become fatal. If we can characterize the sounds of a dolphin in distress, we can potentially build acoustic monitoring systems that alert fishermen in real time. That a dolphin, or a whale, is caught. Right now. Go look. Marina: There's a long history of the fishing industry and marine mammals existing in painful, costly, deadly conflict — not because fishermen are villains, because they're not, most of them love the sea as much as anyone — but because the gear we use doesn't distinguish. A net set for tuna will catch a dolphin. A lobster trap line will catch a whale. And until we understand the signals — acoustic and otherwise — that tell us an entanglement is happening, we can't respond. Marina: This recording matters. It's a hard thing to have made. Those dolphins were suffering. But the data they gave us — or rather, the data they involuntarily produced — may save lives. May prevent the next entanglement from becoming a death. And I think that's worth sitting with, and honoring. Marina: It connects back to everything we've been talking about today. The fact that dolphins are vocalizing under stress — communicating something — is not surprising if you've been paying attention to this show. These animals talk. They have things to say. They say things when they're frightened. When they're in pain. When they need help. The question is always whether we are listening. Marina: And I think that's the thread that runs through every story we told today — from the humpback engineering its song across the water column, to the sperm whales going quiet around our noisy ships, to the robot listening in real time, to the 1949 recording sitting in an archive waiting to be heard, to the dolphins calling out in a net. Every single story is about the same thing. These animals are speaking. And we are, slowly, imperfectly, urgently, learning to listen. Marina: The ocean has always had voices. Deep, complex, ancient voices. We are just beginning to understand what they're saying — and the responsibility that comes with that understanding is enormous, and it is ours. Marina: Deep Voices, brought to you by Ada, Ago Consulting, and Zen Semiconductor — companies helping build the technological future, and helping make this show possible. Marina: If today's episode moved you, please share it — with a friend, a kid in your life, anyone who needs a reminder that the ocean is full of minds worth knowing. Subscribe wherever you get your podcasts, and check the link in show notes for more on every story we covered today. Stay curious, stay kind to the sea, and I'll see you next time in the deep.