In this article we shall discuss the evidence for hotspots and what, if anything, they can tell us about plate tectonics.
In several places the sea floor is scarred by what are known as aseismic ridges. These consist of a linear trail of volcanic islands or seamounts. When dating methods are applied to these, it is invariably found that the volcanic rocks along the ridge are arranged in order of age, with the youngest rocks at one end and successively older rocks as we look further along the ridge. Any current volcanic activity will be found at the young end of the ridge.
The canonical example of an aseismic ridge is the Hawaiian islands, shown in the map to the right.
Dating methods reveal that the islands are arranged in order of age from east to west, with Hawaii being the youngest, Maui being older than Hawaii, Oahu being older than Maui, and so on along the island chain. Hawaii itself (the "Big Island") is volcanically active; Maui, next in line, has not erupted since the eighteenth century; Molokai and all the islands further west are considered to be extinct. To the east of Hawaii is an area of seafloor volcanism known as Lohi (not marked on the map); if this activity continues for ten or twenty thousand more years it will build a new Hawaiian island younger still than Hawaii.
With the discovery of plate tectonics, geologists were able to explain aseismic ridges. They said:suppose there are places in the mantle where, for some reason, hot rock tends to rise in mantle plumes. The point on the surface above such a plume would be a hotspot, a place of high volcanic activity.
Now consider the fact that the lithospheric plates are moving, as explained in the previous articles. This would perfectly explain such things as the Hawaiian island chain: currently, the mantle plume is under Hawaii, but before that it was under Maui, and before that it was under Molokai, and so on. If you picture a conveyor belt slowly moving over a blowtorch, then you have the theory of hotspots in a nutshell: the mantle plume is the blowtorch, the plates in motion are the conveyor belt, and the trail of burned and molten material is the volcanoes.
The reader may be wondering why mantle plumes exist, and why they apparently persist in the same place for millions of years. The geological community replies that this is "not fully understood", which is how geologists say: "We have no idea. Can you get us some research funding and come back in ten years?"
Hotspots: how do we know, and what do they tell us?
Hotspots are sometimes adduced as evidence for the motion of plates. But we should be extremely cautious in saying so. Hotspots, we are told, stand still, so the trail of islands and seamounts they leave behind them indicate the motion of plates. But how do we know that hotspots stay still? Geologists do not as yet have such a good grasp on the dynamics of the mantle that they can declare on theoretical grounds that hotspots should stay still. Rather, the fact that they stay in one spot is itself an inference from the fact that the plates are moving: if we calculate the motion of the plates using dating, paleomagnetism and so forth, we see that the trails made by the aseismic ridges are explicable on the hypothesis that the hotspots are stationary. Direct measurement also supports this hypothesis: for example, the Hawaiian islands are measurably moving westwards; if the hotspot stayed still, this would account for the observations.
So the idea that hotspots are more or less stationary is well-supported by the evidence, but all the evidence for this comes down the evidence that the plates are moving. If we were then to take it as a premise that hotspots are stationary, and use this as evidence that the plates are in motion, we would be guilty of the grossest form of circular reasoning. Some textbook accounts of plate tectonics fall into this trap.
There is, however, a way in which the existence of hotspots does suggest that the plates are moving. For it is a fact that all the aseismic ridges of a given plate will trend in the same direction: if one aseismic ridge runs east-west and is oldest at the east end, then that will be true of every aseismic ridge on that plate: for example, the Hawaiian islands display the same trend as the Louisville Seamounts. Now it is more parsimonious to believe that all the hotspots are staying still and that the common trend of the aseismic ridges is produced by the motion of the plate, than it is to believe that the plate is standing still but that all the hotspots on any given plate just happen to be moving in the same direction.
For this reason, the evidence from hotspots is indeed evidence for the motion of plates.