Fissure eruptions

As well as in the form of lava domes and seamounts (see earlier post), there is a third way magma can erupt onto the seafloor: as a fissure eruption. Fissure eruptions generally conjure up images of lava fountains erupting from long rifts in Iceland, and an underwater version of this eruption occurs 3000m deep at the Costa Rica Rift.

A fissure eruption at Krafla, Iceland

From our swath bathymetry data, we can identify a few large craters up to 2km wide dotting the seafloor, and their formation can be linked to these fissure eruptions. These craters (known as calderas) were probably once lava domes, formed by small eruptions from a shallow magma chamber when enough pressure had built up. 

Schematic diagram of a fissure eruption occurring along a fault, emptying a magma chamber that was previously feeding a lava dome, and forming a crater or caldera in its place

Occasionally the stress of the stretching crust pulling apart at the ridge will cause a fracture to open up perpendicular to the spreading direction. This fault will propagate through the crust laterally until the stress (known as tensile stress) has been dissipated. These fractures can sometimes propagate into the subsurface magma chambers, diverting the pressurised magma away from its usual eruption site at the lava dome. The magma will travel along this fracture, forming a dyke, and erupt in a long fissure eruption the length of the fault. These eruptions can often empty the magma chamber which reduces the pressure in the surrounding crust. A similar process to this occurs on land when water is extracted from an underground aquifer, resulting in subsidence. Here, the rapid exodus of magma causes a zone of decompressed crust to occur beneath the lava dome. The above crust will then collapse into this space compacting the magma chamber and resulting in a caldera forming at the surface, where once stood a 100m high lava dome.

Seismic reflection: imaging the upper crust

Over the last few days, we have been acquiring data in long transects from the young crust at the Costa Rica Rift, to older crust (up to 6 million years old) further from the spreading ridge. Since our OBSs for this stage still remain on the seafloor, we have predominantly been looking at our multichannel-streamer (MCS) data coming in. We mostly look at the reflections from this data, to study the upper structure of the oceanic crust.

A diagram of our MCS set-up. The source and receivers are towed behind the ship. Energy travels as waves down to the seabed and reflects back to be recorded by the hydrophones in the streamer.

Seismic reflection works by creating an acoustic (sound) pulse, which travels down to, and through the seabed, reflecting off boundaries as it goes. This reflected energy comes back to the surface and is recorded by our receivers in our streamer (currently 4.5 km long!). 

The data outputs in the form of time (from when the pulse was fired) against the amplitude of the sound wave recorded by our receivers. From this we can identify different waves of energy reaching the streamer: from the direct wave that travels straight through the water, to the seabed reflections and sub-seabed reflections.

Data recorded from a single acoustic pulse from the source. This shows the amplitude of waves recorded by the receivers, with receiver number along the streamer across the top (0-360), representing distance (0-4.5 km). Two-way travel time (TWTT) is down the side, called such as it represents the time for the energy to travel down through the water, reflect off a boundary, and travel back to the sea surface again. This can roughly be interpreted as depth if we have an idea of the speed which the waves are travelling at (~1500 m/s in sea water).

After some rough processing, we combine a large section of data to form a picture of the structure of the seafloor and crust, with horizontal distance against time, which can be roughly converted to depth.

So far we have found lots of interesting structures from our reflection data, including rough topography characteristic of mid-ocean ridges, faulting (fracturing) of the crust, and stratified sediment layers on the older parts of the crust. 

A roughly-processing seismic reflection image of a fault zone, with TWTT representing depth down the side, and CMP number (each CMP is 25m apart) representing horizontal distance across the top. You can see the rough topography with large ridges and valleys characteristic of these fracture zones, plus the thick layered sediment, showing that this is fairly old crust.

In other news, the first Science Café of JC114 was held a few days ago, with four short presentations to scientists and crew alike about the general purpose of OSCAR, and some of the exciting recent findings. Hopefully another will follow before the end of the cruise! We've also managed to lose the flock of red-footed boobies that were covering the A-frame just a few days ago. The chance of receiving some 'good luck' from above will not be sorely missed.

A few red-footed boobies perching on the met platform in the early morning