Underwater robots listen for whales

Underwater robots listen out for whales even in the middle of a raging storm
11 May 2011

Interview with 

Mark Baumgartner, Woods Hole Oceanographic Institution


Ocean glider


Mark - Well one of the reasons we wanted to look into autonomous vehicles, or you can think of them as robots, to go to sea to do work for us is that it's very difficult to access marine mammals often.

It's expensive to go out on ships, and there's lots of times when we can't actually observes marine mammals. So for instance during difficult weather and when animals go far off shore it becomes difficult to get ships out to access them.

Ocean gliderA friend of mine, who works here at the Woods Hole Oceanographic Institution in the physical oceanography department, he has built a career on using these autonomous underwater vehicles for his physical oceanographic research.

And so we got together with the idea of equipping these vehicles with passive acoustic recorders. So not only can they be collecting a lot of oceanographic information but they can be listening for marine mammals as well.

And so in 2005 we developed a digital acoustic recorder and we put it on a glider and tried it for 5 days out off the coast of Massachusetts here.

We didn't realize we were going to have such a spectacular demonstration of this technology. When we put these things in the we put these things in the water a Noreaster, which is a powerful storm here on the east coast of the US, blew through the area when the gliders where in the water.

We were actually at sea at the time, trying to do research on North Atlantic right whales and we had to leave the area and come home, the storm was so intense. There were gale force winds, seas were up I think over 15 feet. These are conditions that are just not possible for us to work. But the gliders - no problem at all!

So they proved why we wanted to use them, right off the bat.

And they did a great job. They stayed at sea, they collected a lot of information, and the recordings were just wonderful that we got back.

Helen - What do these AUVs, these autonomous underwater vehicles, or ocean gliders, what do they look like? And how do you control them?

Mark - So, the ocean gliders the have long endurance because there's no motors associated with them. So the way they work, is they go up and down in the water column using what's called a buoyancy pump. They just suck in some seawater to become heavier than the surrounding water, and they sink. And when they get to the bottom of their dive they spit that water back out and they become buoyant and they float.

They have little stubby wings attached so that when they're going up and down, those wings provide a little lift, exactly the way and airplane glider works. 

The gliders themselves are sort of 4-feet long, maybe less than a foot in diameter. And so they're not very big. We paint them bright yellow so that we can see them. The way we operate them, is that they go to see, and we usually give them something to do. So, go to point A please, or stay in this one location.

And so they'll dive regularly to the bottom or to some specified depth then come back up to the surface. And every once in a while when they come back to the surface they sort of stick their but out of the water, and they have a GPS a Global Positioning System receiver on board, so they know where they are. And they also have a satellite phone, so they can phone home and tell us where they are and what they've been doing and what they've measured. We can see in real time what the water is like.

At that point we can communicate with the glider, it's a 2-way communication. You can say ok you've arrived at point A, please now go to point B.

And so you can sort of drive them but you're not sitting in the drivers seat necessarily, you're just telling the vehicle what to do, and it works out which direction it needs to go, how fast it needs to go.

Helen - Fantastic. And you said that a couple of years ago, you've tried one out and it seemed to work all the way through this great big storm and was wonderful for a few days.

Mark - Yup.

Helen - Have you now started using them  for a longer time? And what sort of things are you finding out?

Mark - So, we were involved in a pilot project to take gliders to an area that's fairly inaccessible to us and ships, in the Gulf of Maine here off the NE coast of the United States. There have been aerial surveys, they've been surveying the Gulf of Maine for many years and they found this pocket of right whales in the middle of the Gulf of Maine that seemed to go there every single year, kind of late fall/early winter. That's a very difficult time for us to operate on ships because that's getting into the beginning of the stormy season.

And so you can maybe stay out a day or two before seas get really rough and you have to come home again.

So we were thinking how can we go out there are figure out what it is about this habitat that's so important to the right whales. And to do that we typically take a ship and we go there and we take all kinds of measurements about what ocean conditions are like, what the food for the whales is like there, how much food is there, where's the food distributed. All of those things are impossible to observe from a plane. So we sent one glider, and two other autonomous underwater vehicles called profiling floats, which are very much like gliders only they don't fly they just up and down in the water column, and they stayed out there for a month.

So they were able to go to the area where the whales were and collect a lot of information about the environment that the whales had gone to exploit, we had acoustic recorders on all of those vehicles so we could tell where the animals were and where they weren't.

And so again it was  a good demonstration of the capabilities. They could stay there for a month and the weather, the sighting conditions didn't matter. It could be foggy, it could be night. They could still do their work.

Helen - What plans do you have next for these ocean gliders? Do you have things in the pipeline that you're going to be working on?

Mark - We're working on a system now where there can be not only are you recording the information on the glider, but you're also running software to detect marine mammals sounds and to classify them and say, Oo, that's a right whale sound or ah that's a humpback whale sound. Collect that information onboard the glider and then using satellite communication, basically phone that information home.

So we could potentially have these vehicles out in the ocean basically telling us what they're hearing, almost in real time. So you could say to the glider, come to the surface every 2 or 3 hours and tell me what you heard. And as scientists we'd like that information because we could send the gliders out to sort of do surveys for us. And find out where the animals are and then we can just go to those areas with planes or ships and study the animals, saving us an enormous amount of time.

For management purposes there's a lot of interest in this to mitigate chronic problem interactions between people and whales.

Right whales for instance are really prone to getting hit by ships and entangled in fishing gear. And so if you have of these vehicles out doing these surveys for you, you can cover a broad area for a long period of time to tell you where the animals are. And then you can build in safety features by saying well, we're not going to deploy fishing gear in this particular area because we know there are a lot of right whales there. 

Or over time, if you decided, well this particular area really seems to be important to right whales and this happens to be right where shipping lanes are, maybe we should move the shipping lanes.

And that actually has happened. A shipping lanes that went directly through a right whale habitat in the Bay of Fundy up in Canada, and those shipping lanes were eventually moved.

So it's just another surveillance technology that we can use to find out where the whales are so that we can better manage our activities around the whales.

Find out more

Mark Baumgartner's lab

Automomous Systems Laboratory


Add a comment