Volcanic hotspots are scattered across the globe, both beneath oceans and under continents, and most form when columns of unusually hot rock rise from deep within the Earth’s mantle toward the surface. Unlike the volcanoes that line tectonic plate boundaries, hotspots punch through the interior of plates, producing isolated volcanic islands, underwater seamount chains, and continental calderas in places conventional plate tectonics cannot easily explain. The best-studied examples include Hawaii, Iceland, Yellowstone, Réunion, and the Galápagos, but researchers have catalogued dozens more, and the question of exactly how many qualify as “true” deep-sourced hotspots remains surprisingly contentious.
Where Hotspots Appear on the Earth’s Surface
If you plotted every proposed hotspot on a map, you would see them dotted across the Pacific Ocean floor, threading through the Atlantic, and popping up in continental interiors from East Africa to the western United States. Different researchers count different numbers depending on how strictly they define the term. One global survey identified 49 candidate hotspots but concluded that only seven of them met the criteria for a very deep origin: three in the Pacific basin and four in the Indo-Atlantic hemisphere.1Earth and Planetary Science Letters. Three distinct types of hotspots in the Earth’s mantle A separate seismic imaging study examined 60 major hotspots and found plume-like low-velocity anomalies extending through both the upper and lower mantle beneath 12 of them, including Hawaii, Iceland, Tahiti, Louisville, Cape Verde, Réunion, Kerguelen, Afar, and Eifel.2Gondwana Research. Seismic images under 60 hotspots: Search for mantle plumes
The geographic spread is not random. Oceanic hotspots tend to cluster in the central and south Pacific and in the Indian Ocean, while continental hotspots like Yellowstone and the East African Rift mark places where rising heat has interacted with thick continental crust. Iceland straddles the Mid-Atlantic Ridge, which makes it a hybrid: both a plate-boundary volcano and a hotspot sitting on top of a deep mantle plume. The fact that hotspots appear in such varied tectonic settings is part of what makes them scientifically interesting and, at times, hard to classify.
Roots at the Core-Mantle Boundary
The leading explanation for where hotspot plumes originate points almost 2,900 kilometers down, to the boundary between the Earth’s liquid iron core and its solid rocky mantle. This boundary is one of the most extreme thermal contrasts anywhere inside the planet, and it acts as the launching pad for columns of buoyant rock that rise toward the surface. Seismic studies have revealed two continent-sized blobs of unusually slow seismic-wave velocity sitting on this boundary, one beneath Africa and one beneath the Pacific. Geophysicists call them Large Low Shear Velocity Provinces, or LLSVPs.
These structures matter because hotspot plumes do not seem to rise from just anywhere along the core-mantle boundary. A study mapping the eruption sites of Large Igneous Provinces over the past 300 million years found that they sit vertically above the narrow margins of these two LLSVPs, specifically above contours where seismic shear-wave velocity drops by about 1%. The same study showed that 24 active hotspot volcanoes also project downward to those same narrow borders, defining what the authors called “Plume Generation Zones.”3Earth and Planetary Science Letters. Plume Generation Zones at the margins of Large Low Shear Velocity Provinces on the core–mantle boundary Numerical modeling work has reinforced this picture, finding that the original eruption sites of deep-sourced Large Igneous Provinces over the last 200 million years fall mostly above LLSVP margins.4Earth and Planetary Science Letters. The source location of mantle plumes from 3D spherical models of mantle convection
In other words, the planet’s deepest hotspot plumes are not randomly distributed throughout the mantle. They preferentially sprout from the edges of two massive deep-mantle structures, which likely represent chemically distinct piles of ancient material that have accumulated on the core over billions of years. Think of them as the roots of the roots: the LLSVPs create the thermal and compositional conditions that allow plumes to form and rise.
How a Mantle Plume Rises
A plume begins when rock at the core-mantle boundary heats up enough to become buoyant relative to its surroundings. Because the mantle is solid rock that flows extremely slowly over geologic time, the plume does not shoot upward like a geyser. It creeps, mushrooming into a broad head trailed by a thinner conduit, or tail, that can persist for tens of millions of years. The classic model predicts that when a plume head first reaches the surface, it triggers massive volcanism over a wide area. The ongoing tail then feeds a smaller, steadier volcanic output, which is the hotspot you see on a map today.5PubMed. Flood basalts and hot-spot tracks: plume heads and tails
The journey upward is not straightforward. Plumes may carry recycled oceanic crust in the form of a dense rock called eclogite, which significantly reduces their buoyancy. Numerical models show that major plumes can contain up to 15 to 20 percent recycled oceanic crust, making them far less buoyant than a purely thermal plume would be. Despite this, large enough thermochemical plumes can still rise through the entire mantle, though their tails remain bulky (over 200 kilometers in radius) and can linger in the upper mantle for a hundred million years.6PubMed Central. Low-buoyancy thermochemical plumes resolve controversy of classical mantle plume concept Two depth zones act as particular barriers to an eclogite-rich plume: the deep lower mantle and a stretch of the upper mantle between about 300 and 400 kilometers depth. The upper-mantle barrier arises because mineral phase transformations temporarily make the eclogite denser than the surrounding rock. To punch through these barriers, a plume needs a high excess temperature.7Nature Communications. Low-buoyancy thermochemical plumes resolve controversy of classical mantle plume concept
This thermochemical picture also helps explain a long-standing puzzle. Classic thermal plumes should push the surface upward by hundreds of meters before any lava appears, yet many hotspots show little pre-volcanic uplift. Eclogite-laden plumes, being denser and less buoyant, cause only negligible surface uplift as they rise, fitting the geologic record much better.
Volcanic Chains and the Fingerprint of Plate Motion
One of the most compelling lines of evidence for hotspots comes from the linear chains of volcanoes they leave behind. As a tectonic plate drifts over a relatively stationary plume, the plume punches through the plate at successive points, building a trail of volcanoes. The oldest volcanoes in the chain sit farthest from the current hotspot; the youngest are directly above it. The Hawaiian-Emperor seamount chain is the textbook example, stretching more than 6,000 kilometers across the Pacific floor. The Louisville Seamount Trail in the South Pacific is another striking case: a roughly 4,300-kilometer-long chain built over the past 80 million years as the Pacific plate moved over a persistent mantle melting anomaly.8Geochemistry, Geophysics, Geosystems. Louisville Seamount Trail: implications for geodynamic mantle flow models and the geochemical evolution of primary hotspots
These chains record not only the existence of the plume but also the speed and direction of plate movement. Age-dating the volcanic rocks along a chain and measuring the distances between them gives you a direct readout of how fast the plate was moving and whether it changed course. The famous bend in the Hawaiian-Emperor chain, where the trail kinks sharply about 47 million years ago, was long attributed entirely to a change in Pacific plate motion. As we will see, newer research complicates that story.
Studies of individual seamounts along these chains also reveal how the volcanic chemistry shifts as a volcano moves away from the plume source. Work on Nintoku Seamount in the Emperor chain showed that contributions from the oceanic lithosphere became more important with time as the seamount drifted away from the Hawaiian plume axis.9Geochemistry, Geophysics, Geosystems. Petrogenesis of Hawaiian postshield lavas: Evidence from Nintoku Seamount, Emperor Seamount Chain That transition from plume-dominated to lithosphere-influenced chemistry provides a geochemical timestamp that independent of radiometric dating confirms the hotspot model.
Hotspots That Move
The original hotspot concept treated plumes as essentially fixed points in the mantle, like burners on a stove beneath a sliding pan. That simplification worked well enough for first-order predictions, but decades of more precise dating and paleolatitude measurements have revealed that plumes themselves can drift. The Hawaiian hotspot has drawn the most attention. In the Indo-Atlantic region, hotspots appear to have remained fairly stationary relative to the Earth’s spin axis for the past 95 million years, but the Hawaiian plume underwent large-scale southward drift during the early part of the Cenozoic era.10Earth and Planetary Science Letters. Frontiers Relative hotspot motions versus True Polar Wander
Detailed comparisons among three long-lived Pacific hotspot chains, those of Hawaii, Louisville, and Rurutu, have quantified the drift. Between about 60 and 48 million years ago, the Hawaiian plume diverged from the Louisville plume by roughly 640 kilometers and from the Rurutu plume by about 684 kilometers, corresponding to relative motion rates of around 53 to 57 kilometers per million years. After about 48 million years ago, inter-hotspot distances stabilized, suggesting the Hawaiian plume slowed down or stopped its rapid drift. Meanwhile, the Louisville and Rurutu plumes stayed relatively close to each other, implying they were either both nearly immobile or drifting together in roughly the same direction.11Nature Communications. On the relative motions of long-lived Pacific mantle plumes
Even the iconic bend in the Hawaiian-Emperor chain turns out to be partly a product of plume motion rather than solely a change in plate direction. Recent mantle convection modeling shows the Hawaiian hotspot drifted westward by about six degrees of longitude and then eastward by about two degrees, in addition to the southward motion. The westward component was driven by ancient subduction zones in the northeast Pacific, and it contributed a substantial fraction, roughly 20 degrees, of the total bend angle, challenging the traditional interpretation that the bend is purely a record of plate reorganization.12PubMed Central. Dynamics of longitudinal Hawaiian hotspot motion and the formation of the Hawaiian-Emperor Bend
Geochemical Fingerprints of a Plume
You cannot directly sample a mantle plume. What researchers can sample are the lavas it produces once magma reaches the surface. Those lavas carry chemical signatures that differ from the basalt erupted at mid-ocean ridges, where the mantle source is generally shallower and more chemically depleted. Hotspot lavas tend to be enriched in certain trace elements and isotopes, reflecting a deeper, less degassed source that may include recycled oceanic crust, sediments, and even material that has been sitting near the core-mantle boundary for billions of years.13Annual Review of Earth and Planetary Sciences. Oceanic Island Basalts and Mantle Plumes: The Geochemical Perspective
One of the most telling tracers is the ratio of helium-3 to helium-4. Helium-3 is a primordial isotope trapped inside the Earth since its formation, while helium-4 accumulates over time from radioactive decay. A high helium-3/helium-4 ratio signals rock that has been relatively isolated from the surface for a very long time. Submarine basalts from the Galápagos Archipelago have helium-3/helium-4 ratios as high as 23 times the atmospheric value, placing the Galápagos alongside Hawaii, Iceland, and Samoa as localities with the most extreme ratios measured and pointing to a deep, undegassed mantle plume beneath them.14PubMed. Mantle plume helium in submarine basalts from the galapagos platform
These geochemical signatures complement the seismic imaging evidence. Where seismology shows a column of low-velocity rock extending deep into the mantle, and lava chemistry shows elevated helium-3 and enriched isotopic compositions, the case for a genuine deep mantle plume becomes strong. Where only one line of evidence exists, the interpretation is less certain.
When a Hotspot Meets a Mid-Ocean Ridge
Some hotspots sit close to or directly on a mid-ocean ridge, where tectonic plates are pulling apart. Iceland is the most prominent example: the island exists because a deep plume delivers extra heat and material to the Mid-Atlantic Ridge, creating far more volcanism than the spreading process alone would produce. Hotspot-ridge interactions cause many of the largest structural and chemical anomalies found in ocean basins around the world.15Reviews of Geophysics. Observational and theoretical studies of the dynamics of mantle plume–mid‐ocean ridge interaction
The plume material does not necessarily spread symmetrically along the ridge. Three-dimensional numerical models of the Iceland plume interacting with the Mid-Atlantic Ridge show that after an initial period of symmetric influence during continental breakup, the plume’s reach rapidly contracts and then develops varying degrees of asymmetry along different ridge segments. One model predicts the plume expands along the ridge to the northeast only about 70 to 80 percent as far as it does to the southwest, depending on how the lithosphere stiffens through water extraction near the base of the melting zone.16Earth and Planetary Science Letters. The origin of the asymmetry in the Iceland hotspot along the Mid-Atlantic Ridge from continental breakup to present-day The practical result is that the volcanic crust and geochemical anomalies around Iceland are not evenly distributed, a detail that matters for understanding the evolution of the North Atlantic seafloor.
Yellowstone and Continental Hotspots
Not every hotspot sits under an ocean. Yellowstone, perhaps the most famous continental hotspot, lies beneath the thick crust of the North American plate. The Snake River Plain, a broad swath of volcanic terrain stretching across southern Idaho, records about 17 million years of volcanic activity as the continent migrated over a relatively fixed magma source.17Earth and Planetary Science Letters. Yellowstone hotspot–continental lithosphere interaction Isotopic analysis of basalts collected along the Snake River Plain shows that more than 97 percent of the total variability in lead isotopes can be explained by mixing between a deep hotspot mantle source and two continental lithosphere components of different ages and compositions. The hotspot mantle source resembles what produced the early Columbia River Basalt Group, a massive outpouring of lava roughly 16 million years ago that likely marked the arrival of the plume head.
The interaction between a deep plume and thick continental crust produces a different style of volcanism compared to oceanic settings. At Yellowstone, basaltic magma ascending from the plume heats the silica-rich continental crust, generating partial melts of rhyolitic composition. This produces the explosive, caldera-forming eruptions Yellowstone is famous for, a style that has no real counterpart at oceanic hotspots like Hawaii.18Journal of Volcanology and Geothermal Research. The Yellowstone hotspot
Are All Hotspots Alike
The short answer is no, and this is where the science becomes genuinely messy. The classic mantle plume model works well for the handful of prominent, long-lived hotspots with clear age-progressive volcanic chains, deep seismic anomalies, and distinctive geochemistry. But many of the 49 or more proposed hotspots lack one or more of those features. Some volcanic chains in the South Pacific, for instance, show no clear age progression at all, and their volcanism appears short-lived and discontinuous. Researchers have argued that this type of activity could involve shallow “plumelets” shooting off the top of a broad superplume in the South Pacific mantle, rather than individual conduits extending all the way to the core-mantle boundary. Regional plate extension driven by changes in the global plate circuit may also play a role in switching this shallow volcanism on and off.19Geochemistry, Geophysics, Geosystems. Short‐lived and discontinuous intraplate volcanism in the South Pacific: Hot spots or extensional volcanism?
The recognition that hotspots can originate from different depths has led to a classification scheme. One proposal separates them into “primary” hotspots rooted at the core-mantle boundary, “secondary” hotspots sourced from the upper-lower mantle transition zone, and shallow hotspots linked to lithospheric processes or local mantle anomalies.1Earth and Planetary Science Letters. Three distinct types of hotspots in the Earth’s mantle Under this framework, some hotspots that the public thinks of as equivalent to Hawaii may actually tap a fundamentally shallower source. The early formulation of the mantle plume hypothesis acknowledged this murkiness, noting that while the model is reasonably successful at explaining the direction and velocity of volcanic migration, “there are a number of difficulties.”20Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences. Intra-plate volcanism
Flood Basalts, Mass Extinctions, and Environmental Upheaval
When a new plume head first arrives at the base of the lithosphere, the result can be catastrophic on a planetary scale. The initial burst of volcanism produces a Large Igneous Province: millions of cubic kilometers of lava erupted in less than a million years, sometimes accompanied by marked thinning or breakup of a continent.21PubMed. Linking mantle plumes, large igneous provinces and environmental catastrophes The Siberian Traps, the largest known continental Large Igneous Province, are the most dramatic example. They erupted roughly 252 million years ago on top of thick cratonic lithosphere and were synchronous with the Permo-Triassic mass extinction, the worst die-off in the fossil record.
The lava itself is not the main killing agent. Modeling suggests that massive degassing of carbon dioxide and hydrogen chloride, mostly released from recycled oceanic crust carried within the plume head, could alone trigger a mass extinction. The models predict that this degassing begins before the main eruptive phase, which matches the stratigraphic and timing data for the Siberian Traps and other Large Igneous Provinces.21PubMed. Linking mantle plumes, large igneous provinces and environmental catastrophes Other mass extinctions have also been tentatively linked to Large Igneous Province eruptions, including the Deccan Traps of India coinciding with the end-Cretaceous event 66 million years ago. The connection between plume heads and environmental catastrophe is one reason hotspot geology extends well beyond volcanology into paleontology and climate science.
Mineral Wealth Tied to Mantle Plumes
Hotspots and the Large Igneous Provinces they spawn are not just scientifically fascinating. They are economically important. Continental rifting and breakups induced by mantle plumes generate a wide range of mineral deposits, including nickel-copper-platinum group element ores, iron-titanium-vanadium deposits, chromium deposits, iron oxide-copper-gold deposits, carbonatites, and even hydrocarbon accumulations along the passive continental margins that form when continents split apart.22Geological Society, London, Special Publications. Mineral systems and their putative link with mantle plumes South Africa’s Bushveld Igneous Complex, one of the world’s most valuable ore deposits, is a prime example of a plume-related mineral system.
The connection between plumes and nickel-copper-platinum deposits specifically comes from the unusually high temperatures involved. Producing magmas rich enough in these metals to form economic ore bodies requires melting large volumes of peridotite mantle rock, which takes the kind of excess heat that a plume provides. These deposits are therefore intrinsically linked to the emplacement of Large Igneous Provinces in extensional tectonic settings.23Economic Geology. Geodynamic Controls on Magmatic Ni-Cu-Platinum Group Element Sulfide Mineral Systems Through Time The metals that end up in catalytic converters, batteries, and electronics can sometimes trace their geologic origin back to a plume that rose from near the core-mantle boundary hundreds of millions of years ago. It is a reminder that the processes operating at the deepest levels of the planet have surprisingly tangible consequences at the surface.