Is there really a plume of hot rock under Hawaii?

Probably yes, but not quite the plume that textbooks used to draw. Hawaii sits more than 3,200 km from the nearest plate boundary, so plate edges cannot explain its volcanoes. In 1963 the geophysicist J. Tuzo Wilson proposed the hotspot theory for volcanism so far from regular conditions, and it came into wide acceptance. The classic version says a single, fixed plume builds volcanoes, and the Pacific plate carries each one away from its source.

The evidence is written in the islands themselves. The oldest rock in the main Hawaiian islands, on Kauaʻi, is about 5.5 million years old and deeply eroded, while Hawaiʻi Island is 0.7 million years old or less, with new lava constantly erupting at Kīlauea. Seafaring Hawaiians noticed differences in erosion, soil and vegetation, and deduced that the islands to the northwest were older than those to the southeast. The idea was handed down through the legend of Pele.

Where the classic story cracks is the long bend in the chain. It was blamed on a change in the Pacific plate's direction, a change that no plate record shows. Rocks drilled from the seamounts instead suggest that the hotspot itself drifted southward, and that its motion slowed about 47 million years ago.

And does a deep plume feed it at all? Some geologists doubt plumes in general: a debate called the great plume debate has pitted plumes against explanations from shallow plate processes. Reading seismic images of the deep mantle is often ambiguous, but for Hawaii one imaging method provides evidence for a plume, as it does for Yellowstone. For many other hotspots, the question is still open.