3.9 Seismic exploration – part 1

My name is Dr. Rebecca Bell, and I'm a lecturer in geology and geophysics here at Imperial College. My key research interest is in the use of subsurface data, so data which lets us see what's going on under the Earth’s surface, to explore both for oil and gas, but also to look for faults which might be seismically active and could cause earthquakes. So there’s that. I've got a kind of industry element to my research, and also a kind of seismic hazard aspect as well.

So you can collect seismic data both on land and at sea. And it’s maybe a bit counter-intuitive, but it's actually a lot easier to collect seismic data at sea than it is on land, the reason being on land, you have to put out all of your equipment, dig holes to put the equipment in, dig it up, move it on, dig holes again, and it's really quite laborious, whereas at sea, we can just pull all of the equipment behind a ship behind us. So what the oil industry will typically use is something called an air gun. When we're ready to fire a shot, gates are released. The air comes out of the air gun into the water and produces a giant air bubble.

That air bubble then produces a pressure wave, basically a sound wave, otherwise known as a seismic wave, which is where this method gets its name. And that sound wave then travels through the water, and the very first thing it encounters then is the seabed. So the seabed has really different properties to water. It’s got different density, and the speed that sounds travel through it is much different as well. So we've got differences in density, differences in velocity, and that causes the sound to get reflected. It basically echoes off the seabed, like, if you were to shout in an empty room, you get an echo. Those reflections are then detected by a string of hydrophones, which are basically underwater microphones.

Because the sound is so energetic, though, it doesn't just stop at the seabed. It also travels down into the rock. Whenever it meets rock layers of different types, which have different density and velocity, we also get reflection and an echo, which we can detect. So depending on the time it takes for these reflections to be returned, we can start to build up a picture of what the Earth looks like in terms of that time, the idea being the longer it takes the reflection to reach the hydrophone, the deeper that particular rock layer. So it’s exactly the same way that bats figure out the back of their cave. They set off a high-frequency pulse and wait to see how long that pulse takes to return. If it's very quick, the back of the cave’s close. If it takes a long time, it’s much, much further away.

Seismic data first became commonly used by the oil industry really back in the sixties, and back then, they were limited to only collecting data in two dimensions. So they would literally get a cross-section through the Earth in one particular location. That has lots of issues, though, because if you get reflections from any geological feature slightly off of that one profile, those reflections end up coming into your 2-D line, so you end up with really quite poor imaging.

So in the eighties, nineties, we then started to collect 3-D seismic data. So that was a major, major leap forward, and that also allowed the oil industry to start really being able to target exactly where the best places to drill would be, rather than doing a potshot drill, if you didn’t have any seismic data, or being forced to drill on your 2-D profile, if that's all you had. So that was a really, really large step forward.