How are dikes and sills formed?

Answer 1

To begin the formation of dikes and sills, you have to have a subsurface magma body. This magma body, at nearly any depth, is growing in some way: probably through the addition of new magma from beneath. As the chamber becomes over-pressurized, magma needs a place to go. The wall rock of the chamber, feeling the tremendous stress of its expanding contents, begins to crack and fracture. This cracking starts a process that can continue for hundreds of kilometers.

The moment that wall rock fractures, the open space between the rock (even if it's only a millimeter or two) is filled with the pressurized gas of the chamber. This gas acts as a wedge to push the crack farther back and widen it slightly. As this gas wedge moves along, forcing the crack to grow, magma begins entering the newly widened edifice. This magma, though it is liquid, is still very dense and acts as a second wedge moving through. Moreover, this hot magma can also melt that wall rock, widening the new dike.

This rock-working duo, the gas and magma, push their way through the surrounding rock. Fueled by the addition of new magma behind it, the process continues with the gas wedge breaking open a path for the magma to pass through. The size of this gas wedge is theoretical, probably variable, but also probably very small. This gas is the stuff that was exsolved from the magma wedge, and so we're not talking about enormous volumes of gas. Still, its presence is extremely important because if the front of the advancing magma wedge actually had to push against rock, the face of the advancing magma would harden and make the process many times more difficult.

But this is also an important concept to remember: as the magma advances further from the magma chamber, the surrounding rock gets colder. It no longer becomes an issue of melting the surrounding rock, but the surrounding rock freezing the magma. This creates an interesting situation, however: if the interior of the magma flow continued moving in a straight-line direction, the outside edge would freeze inward…and continue to freeze until the dike is literally squeezed shut. Something else must happen.

What do you know, something else does happen: thermally-driven convection. As the material along the wall cools, it sinks. The brief vacuum left by the sinking cold magma is replaced by hot magma from the interior of the dike. The brief vacuum left by this magma leaving is filled by that cold magma that has just sunk out. That magma is reheated to some degree by the surrounding interior while the hot stuff that just moved to the outside is being cooled. The process is ongoing and, while it keeps the overall flow from pinching shut, it does mean the overall magma flow is cooling.

The idea is, though, that the magma cools much more uniformly. That is not to say that cooler magma isn't still sticking to the dike walls, but that it is just happening in much smaller quantities. Also, with the exterior moving more slowly due to cooling and friction, the flow experiences shear. This shear is basically friction for fluids, and converts kinetic energy (movement) to heat. This means that, though the flow slows, it can continue moving even longer; to some extent, it is self-heating.

Well that solves that…except…that large rhyolite dikes, which are much cooler and more viscous, have been found to extend dozens of kilometers. Now I don't care how you look at it, that just doesn't seem right. Even from the vent, rhyolite is usually extruded like toothpaste from a tube. Now we find rhyolite squeezing through dikes for dozens of kilometers without freezing, being clogged, or stopped? That just doesn't make any sense. Unless…

Unless we have bimodal flow. We know about it now, and engineers have tested it. Bi means two and modal refers to "states" or "types." That is, through the dike we have two types of magma flowing. One is high viscosity, the other low. You might think that the high and low viscosity magmas would mix, but remember that the high viscosity one is not very fluid at all compared to the low. That means it behaves more like modelling clay in a river of mud (what would you do without these off-the-wall analogies?).

Through this same dike we have the two magmas; for our example, basalt and rhyolite. Why do we have the two? Perhaps the crack in the magma chamber is pulling rhyolitic mush-zone material out and also some basaltic chamber interior. As the material enters the dike, it probably flows rather chaotically. Soon, however, the low viscosity (highly fluid) magma surrounds the rhyolite. This means only fluid magma is contacting the walls, which means the high viscosity magma remains insulated and is moving with almost no friction to slow it down; the basalt acts like a lubricant along the walls.

As bizarre or unusual as that may sound, it's really not at all. The petroleum industry has utilized this phenomenon for years by injecting water into oil pipelines, allowing the oil to move much more quickly from point A to point B. If any two magmas are injected, at the same time, into a dike, this effect will always occur. This event may also be present during sizeable eruptions, with a lower viscosity magma coating the walls of the conduit. There is some good evidence for this occurring, although it is not widely acknowledged at the moment.