Plates and their speeds

The floor of the Atlantic grows about 2.5 cm a year, which adds up to 25 km in a million years.

A broken shell

The theory of plate tectonics says that the Earth's rigid outer layer is broken into a dozen or more large and small plates that move relative to one another. BGS counts about 15 major plates. Some, like the Pacific Plate, are almost all ocean floor. Others carry whole continents along with the sea floor around them: the North American Plate reaches from California to the middle of the Atlantic. A plate is not a continent. It is a slab of lithosphere, and the coastline is just where its surface happens to be above sea level.

Most of the action happens at the edges. USGS sorts plate boundaries into divergent ones, where plates pull apart and new crust is made; convergent ones, where crust is destroyed as one plate dives under another; transform ones, where plates slide past each other and crust is neither made nor destroyed; and broad boundary zones where the picture is messy, such as the Mediterranean and the Alps. The next three units take the main types one at a time.

Centimetres per year

Plates move slowly by human standards and quickly by geological ones. USGS gives an average of about 2.5 cm per year for the spreading of the Mid-Atlantic Ridge, less than 2.5 cm per year for the Arctic Ridge and more than 15 cm per year for the East Pacific Rise near Easter Island. Along the San Andreas fault, the Pacific Plate grinds past the North American Plate at about 5 cm per year. Multiply by a long time and the numbers become large. At 2.5 cm per year, the Atlantic widens by 25 km in a million years, and over the 100 to 200 million years that seafloor spreading has been going on there, it grew from a narrow inlet into an ocean thousands of kilometres wide.

How do we know the speeds? There are two clocks. The first is in the rocks. The ocean floor records reversals of the Earth's magnetic field as stripes, and since the age of each reversal is known, the distance from the ridge to a stripe gives an average rate over millions of years. Rock formations of known age offset across a fault give the same kind of answer. The second clock is in space. Since the late 1970s, satellite methods, and above all GPS, have measured the distances between stations hundreds to thousands of kilometres apart, again and again. USGS notes that the rates measured over a few years agree well with the rates averaged over millions of years.

What drives them

The hot mantle under the plates slowly convects, rising where it is hot and sinking where it is cool, and for a long time this was pictured as a conveyor belt carrying the plates. BGS lists three proposed forces: mantle convection, ridge push, where new warm plate stands high at a ridge and slides away from it, and slab pull, where an old, cold, dense plate sinks at a subduction zone and drags the rest behind it. Research cited by BGS found that plates with more of their edges sinking move faster, which points to slab pull as the main force, with ridge push helping. This was the question Alfred Wegener could not answer in 1912, and parts of it are still debated.