How Sonar Works (Submarine Shadow Zone) – Smarter Every Day 249

Here’s a sanitized version of the provided YouTube transcript:

—

Do you guys do things like that? That’s not acceptable. That last 15 seconds has to be removed. Okay, cool. Nice.

When discussing weapon systems on a nuclear submarine, there’s a very fine line between what’s unclassified and what’s classified. What you just saw was me getting a reminder for asking a question that crossed that line. This is the next video in the Smarter Every Day deep dive into how nuclear submarines work. We’re on the USS Toledo, and I’m about to talk to a very important person.

Do you remember in a previous episode, we talked to Chief Luth? You don’t look excited about it. – That’s because there are no poppers out here yet. – (Destin laughs) Well, that happens to be the guy overseeing sonar for the entire nuclear submarine that we’re on. Sonar operators do their work in the control room. This information is processed and then distributed on monitors throughout the boat. For this discussion, we’ll be located in the officer’s wardroom.

So here’s how this interview went down. I interviewed Chief Luth on the boat. I had to hand over all my data to the Navy, which then reviewed what was classified and what wasn’t. Long story short, this is an amazing video, and we’re going to dive into the hardcore physics of how sonar works. Let’s go get smarter every day.

When you think about submarines, you think about sonar, but this is a very sensitive topic because if this is a cat-and-mouse game, this is how you do the cat part. And so I guess the mouse part too. To ensure we don’t say anything we’re not supposed to, we have Executive Officer Andrews here. Thank you.

So if I say something wrong or he says something he shouldn’t, you’re going to… – Yup! Okay. So this is a sensitive topic. Okay. I’m Destin. – Hi, I’m Garrett. – Garrett? Yep. Okay. And so what is your role on the boat? – I am the Sonar Division Lead and Chief Petty Officer. – Okay. That means you’re in charge of sonar for the whole boat? – Correct.

So my understanding of sonar is that it’s kind of like echolocation. You ping, listen for a return off an object, and based on the shape of the sound that comes back, you can figure out what’s out there. I would say that you are partially right. – Okay. – But mostly wrong. (Both laugh)

So, is it because of the ping part? – Uh, yes. You’re smart. And like me, you probably have a good idea of how sonar works. SONAR, which stands for Sound Navigation and Ranging, works like this: You’re going along, and then you ping and listen for an echo. The amount of time it takes for that echo to return to you and how much sound bounces back gives you an idea of how far away and how big the object is. This is exactly how dolphins and bats use echolocation to navigate using sound.

This method is called ACTIVE SONAR because you’re actively sending out a sound and waiting for a return. We’re going to talk more about that in a few minutes, but for now, we’re about to learn about something called PASSIVE SONAR, which is something I don’t know a whole lot about.

So this underway is not ideal for me to tell you that you’re wrong because we’re using a lot of active sonar. Since we’re under ice, we’re trying to navigate around icebergs and find ice keels to avoid running into anything. We’re using a lot of active sonar this underway. Normally, we’re a hundred percent passive.

The reason you’d want to have passive sonar is that anytime you emit anything from the boat, people can detect that, right? [Silently nods] That’s right? He’s very tight-lipped; I kind of like this. So, active sonar, as I understand it, means you ping, make a sound, and look for the return. Passive sonar… This is something I don’t know about.

So passive sonar means we’re just listening. We’re listening to see what’s in the water. Once we find something, we can determine its bearing and maneuver the ship based on the direction the sound is coming from. We can even classify what is making the sound based on how the sound was made.

Chief Luth is talking about acoustic signatures—the ability to identify what’s in the water just from the sound. When we listen to sound, it looks like this in my head. It’s just a waveform, and I’m like, “Oh, that’s louder. That’s not as loud,” but sound is actually much more complicated than that. To analyze true sound, you need something called a spectrogram.

You can see all the different frequencies displayed here, and it’s a function of time, showing which frequencies are prominent at different times. The lower frequencies are down here, and the higher frequencies are up here. By analyzing a spectrogram, you can discover all kinds of amazing things.

For example, let’s look at a spectrogram of an Alabama summer night and see all the different sounds we might discover. [Mockingbird chirps line up with the graph] If we zoom in, we can see the whistle of the bird as it changes its pitch, creating a bird call. We also see faint horizontal bands on the screen. Let’s take a look at what’s going on here. [Crickets chirping]

If we isolate the frequency and increase the gain, we can clearly hear the hum of the summertime crickets. If we zoom in on this second lower frequency band and boost the gain again, we can hear tree frogs and toads from the surrounding area. And then you can see something down here in the even lower frequencies.

You want to guess what that might be? [Distant train horn] Yep! That’s a train in the distance, but all the bird calls keep going on above it. You just did signature recognition with your ears. You compared a sound you heard in a certain frequency band to known sounds in your database. That’s exactly what the Navy does, except they do it on a much deeper level.

Think about this: Your phone can identify music based on what it hears. Imagine how much money and effort has gone into developing algorithms to operate on a submarine to identify what kind of boat is out there in the ocean. Human-made machinery stands out, but the Navy also has to filter out what they call “biologics,” sounds of pistol shrimp clicking or whales off in the distance.

All this is rolled into tremendous efforts in signal processing and machine learning algorithms just to figure out how the Navy can process all the sounds in the ocean. So passive sonar. I’m assuming you’re using multiple frequencies? And those frequencies will affect what you’re hearing come back? Is that true? [Clicks teeth… Inhales… thinking deeply] Uh, yes, it is true.

Are you trying to figure out what you can say? – Yes. – Okay. So… You can just avoid it. You can just say, “I don’t want to talk about it.” Let’s just… It’s just the question you asked. I can’t answer it without getting into classified stuff. Well, don’t worry about it.

So passive sonar means you’re using multiple frequencies to see shapes of different things. Is that the deal? Uh, we’re looking at a large band of frequencies to determine where a contact is coming from. So if you’re listening to a boat out there, let’s say you have something running off a diesel engine and you hear a knocking cylinder or something.

It’s one thing to hear that just by listening, but that’s not sonar. You’re using the term passive sonar, so that implies you’re hearing something bounce off something else? Uh… no. So I mean that is part of it. There are different propagation paths or ways that sound could take coming from the source to us.

I’ll draw you a picture real quick about what we’re looking at, and it’ll make a little more sense. Chief Luth is about to draw what’s called a bearing rate graph. This seems intimidating at first, but it’s not. I promise you will understand it. Just stick with it. This is one of the most important tools submariners use to see the world.

Okay, so we have a waterfall display. The newest data comes from the top, and the older stuff scrolls down. You have a line of bearing at the top. In this case, we’d be going course North. If we saw a line coming down right here, this could be a contact. So it would be a contact bearing zero nine zero.

My broadband operator would take his cursor, put it on top of this trace, and listen to see what it sounds like. Based on how it sounds, we can classify the contact. Okay, so we’re about to get into the deep water. Just say no if I go too far.

I’ve observed enough to ask questions here. So if this is the boat here, can we draw a boat? What are these called? The bow planes. About planes out front, stern planes back AFT. Okay. And you got the rudder back here? Yes, sir.

So this right here would be the line of bearing and boat? Correct. So if we have a contact right here, straight off the nose, and we’re going straight at it, then I would expect this. This is time, right? – It is time versus bearing, correct. – So I would expect the line from that to look like that. Is that true? Exactly.

So if we have something else, I think this is what you’re about to tell me… If we have a contact here and the boat’s going forward, can you please draw? Are we okay, sir? We’re good. Okay. Can you please draw what that would look like on that graph?

So we’re moving forward at a certain velocity, and the contact’s off to the right. The contact is going to be moving down here to the right. Correct. And it’s actually going to look a lot like the first thing I drew, and it’s range dependent.

You could have something that looks like this, like I had there originally, if it’s at mid-range. But if it’s at a much closer range, it’s still going to start at the same bearing, but it’s going to go through those bearings a lot faster. I always think about it like you’re walking down the mall, and if you see somebody all the way down or walking towards them, you don’t have to turn your head very far because they are pretty much in the same position relative to you.

But when they get real close, if you want to keep looking at them, you have to turn your head, and that’s kind of where this bearing rate comes from. Because you’re going through bearings relative to you faster. So bearing rate means a measure of angle over time.

So imagine you’re in the submarine, moving along in a straight line. If another boat comes along and you hear its sound with the hydrophones installed on the outside of the boat, the Navy calls this a CONTACT. Just like your two ears tell you the direction that a sound is coming from.

If you have multiple hydrophones set up correctly on the outside of your boat, you can also tell which direction the contact sound is coming from. That angle from the nose of your boat to that contact is called the bearing, and that is very important for submariners. In this case, the contact is moving.

So the bearing is 20 degrees, 30, 60, 90, 120, 150… And if it’s running parallel, it doesn’t quite make it to 180. And that’s how you measure the bearing, which is basically just the angle from your boat. Now let’s do something a little strange. Let’s unwrap this polar coordinate graph and add a time component to it by scrolling the graph down and plotting the bearing we see as a function of time.

Watch this closely. Contact. There’s 20 degrees, 30 degrees, 60, 90, 120, 150. And we’ve lost contact. The shapes you see on this graph could literally mean life or death for submariners. For example, if you see a vertical line on this graph, that could mean danger.

Let’s say you’re stationary and trying to hide. If someone figures out where you are and starts moving right at you, their bearing rate will remain constant. Look at the graph. There’s a vertical line. Flip that scenario around. Let’s say you’ve found a contact that you want to interrogate. As you close the gap between you and the contact, your bearing rate also remains constant, meaning it also graphs as a vertical line.

Vertical lines on a bearing rate chart mean you’re on a collision course. It means that probably one party or the other knows the other one’s there, and they’re going to check it out. If a boat is crossing in front of you, you’ll see a very different kind of curve, and a boat going behind you will look totally different.

The ability to understand these curves is a skill that submariners develop over time until it becomes second nature. Imagine how complicated this gets if you have multiple contacts in relatively close proximity. This is how submariners see the world.

Let me give you a visual aid and see if you can now see what was being plotted on the bearing rate graph. How did you do? Now let’s take this one step deeper. What happens when we start to change the heading of the boat? Not only does the chart get incredibly complicated, but you have to do a rotational coordinate transform in your head while keeping track of time.

This bearing rate graph is an incredible tool to use math to literally see beyond the hull. Every major room I went into on the Toledo had one of these hanging on the wall. How am I doing? Like, in terms of understanding it, where am I at? Am I like…? – You’re pretty good. – Pretty good. – I’m like at kindergarten though… – That’s all we want you to know. – That’s where you want me? – That’s where we want to keep it.

I’m impressed with how fast you picked that up or understood the bearing rate concept. – Okay. So, this is like first glance. This is where you start. And then I’ve worked with radar stuff before. There’s a thing called a waterfall chart. You’re dealing with stuff like that as well. – Yeah.

So if you’ve worked with radar before, radar and active sonar, which is not really what we’re talking about right here, but radar and active sonar are almost identical except for the frequency at which they’re happening. I mean, in a lot of ways, it is the same thing except for radar happening in the GHz range and sonar happening in the Hertz range. – Gotcha.

So you’re also… if you can’t answer, don’t answer. – That’s not acceptable. – That’s not acceptable? – No. – Okay.

So we started to dig deeper into active sonar and had conversations about the signal-to-noise ratio and how they filter out different signals. At some point, the conversation switched to what was classified and what wasn’t. Each hydrophone is… Hold on. I’m reading his body language. Don’t go that deep.

Yeah, no, that’s good. I can show you in the manual; it’s good to go, a hundred percent good to go. It’s an unclassified manual, RP 33. I can go run and get it real quick if you want. In the end, Executive Officer Andrews was very patient with me and didn’t shut down the conversation completely, allowing us to keep going.

So if we have something out here that we’re contacting, I know this is an issue with radar. I don’t know if this is an issue with sonar, but like, as we create, if we’re pinging and doing active sonar, and then we get that return coming back, what can happen, if I understand correctly, is some of this goes up, hits the ice, and then comes down.

So instead of getting a distance from here to here, back to here, you get multipath. So basically, there are multiple ways that you can get returns to the vessel. Is that true? – That is true. And even when we’re not under ice, the signal can interfere with the bottom. – The what? – The bottom of the ocean. – Okay.

And we can also get a return off the bottom of the ocean. – Oh, okay. So the problem is more complex than I thought. So you have ways to take this into account? – Yes. Multipath gets incredibly complicated geometrically.

To prove a point, we’re going to play a game. You are now the sonar operator on the USS Toledo. You’ve been authorized by the captain to use active sonar to tail a mysterious craft in front of you, the Nautilus. First of all, we’re going to do this in open ocean. You are behind the Nautilus and are going to issue a ping.

And you’re going to watch the return to see where the Nautilus is. [Sweeping sonar ping] [Echoey Return] Okay. By that return, it’s clear that it’s in front of you and off to the right a little bit. Right? But that’s not how sonar works. You’re under the ice and over a seafloor.

So let’s take a different view. This is what it’s actually like. [Sweeping Ping] [Return from floor] [Return from Ice] [Return from Nautilus] [Bounce echoes overlap] I’m going to slow it down and watch closely. The first return to the submarine is actually not from the Nautilus. It’s from the bottom of the ocean.

So how do you figure out which one is the real return? That’s the difficulty of multipath. There are multiple paths that the sound can take to the contact and on the return path. So that means one sonar contact gives you multiple returns at your hydrophones. If that’s true, how do you know it’s real? Like, how do you know it’s here and not here?

Furthermore, what if there’s actually a school of Nautilus out there? How do you know that it’s just one contact and not five? That is a challenge that I don’t know the answer to. We’re good. We’re good. You got to get the manual.

He’s got to make sure that it’s all clear. RP 33, you can actually Google this; it breaks it down a lot more, and there’s a lot of good stuff in here. I asked a ton of questions, and there were a lot of things they just weren’t willing to talk about.

But when I got off the sub, I found this document that details a lot of the things submarines have to consider. For example, and this is crazy, but this is real: The speed of sound in water changes with temperature, pressure, and even salt content. Because of that, it can bend sound.

So because these three things change in different places around the globe and even at different depths in the water, you have to understand the conditions of the water around you at all times. The Navy likes to display this data by graphing the speed of sound in water versus the depth of that water.

This is commonly referred to as the Sound Velocity Profile (SVP). To measure the sound velocity profile, a device called an expendable bathythermograph is used. Just looking at these things shows you that they’re really expensive, which tells you