New study links rare earth deposits to ancient buried collision zones

A fresh study reports that the vast majority of rare earth deposits - which supply metals vital for electronics and clean energy technologies - are positioned above ancient, buried collision zones. These zones developed where tectonic plates previously crashed together and were driven deep beneath continental crust.

The authors argue that today’s mining hotspots are, in effect, the surface trace of profound tectonic reworking that occurred long before later melting concentrated these metals into mineable ore.

Buried tectonic footprints

Using reconstructions of past continental layouts, the strongest alignments showed up where long-duration plate collisions once pushed into continental margins and left chemically modified regions at depth.

By following these hidden belts through geological time, Carl Spandler, a professor at the University of Adelaide, and co-authors recorded the same relationship across 412 documented locations.

Their analysis indicates that roughly 75 million square kilometres (about 29 million square miles) of continental crust overlies these altered deep domains. The most intense clustering occurred in areas where several ancient collision belts intersect or stack.

With such a pronounced concentration, the match is difficult to write off as chance and leads directly to the bigger issue: how did those long-buried zones ultimately become the source of ore-bearing rocks?

Why carbonatites matter

A large share of the highest-grade rare earth deposits are hosted by carbonatites - uncommon igneous rocks rich in carbonate minerals - rather than by more typical volcanic lavas.

The magmas involved originate far beneath continents, where small degrees of partial melting preferentially gather elements that do not readily enter common mineral structures.

Work by the U.S. Geological Survey (USGS) has characterised these rocks, since the 1960s, as the principal source of light rare earth elements.

In the new compilation, about 67% of the relevant host rocks lay within the same ancient zones, tying ore-forming magmas to deep tectonic inheritance.

Deep mantle changes

Where one plate descends beneath another, subduction - the mechanism that returns crust to the mantle - drives water, carbon and trace elements downwards.

Part of that transported material can later migrate back upwards into the mantle lithosphere, the rigid layer beneath continents, where it alters the region’s chemistry.

This chemical imprint reduces the temperature required for subsequent melting, allowing unusual magma types to form without needing exceptionally high heat.

Rather than producing ore at the moment of collision, the collision phase seems to stockpile the necessary ingredients deep in the crust and mantle, where they can persist for very long periods.

Timing of formation

The timing complicates any straightforward cause-and-effect story, because the deep “preparation” and the later melt event are often separated by enormous spans of time.

“This time lag is one of the most surprising aspects of our findings,” said Spandler.

In certain examples, the interval runs from millions of years to almost 2 billion years.

That gap isolates the early chemical conditioning from the later trigger and leaves room for multiple routes by which melting could eventually begin.

Where the overlap grows

The tightest spatial matches were found on continents shaped by repeated collisions, notably across North America, southern Africa and China.

Ancient, stable continental cores known as cratons - among the most durable surviving parts of continents - appear particularly effective at preserving these enriched deep regions.

Roughly 85% of the mapped fertile areas overlap one another, implying that several distinct ancient episodes layered their effects in the same places.

Regions concealed beneath Antarctica’s ice sheet may also follow this pattern, although confirming deposits there remains challenging.

Why plumes lose ground

Earlier interpretations commonly put mantle plumes - buoyant upwellings of hot rock - at the centre of explanations for these deposits.

Yet many carbonatites, the unusual volcanic rocks that host most rare earth deposits, lack an obvious association with plume-related heat, and their geochemistry instead supports formation at comparatively lower temperatures.

Because the new mapping links deposits primarily to ancient collision-related zones, it undercuts the argument that rising hot plumes provided most of the initial set-up.

This does not exclude plumes as possible later contributors, but it shifts them away from being the main driver.

Triggers after long delays

Even with an enriched mantle source, a later disturbance is still required, because enrichment alone does not automatically produce melt or a deposit.

Processes such as rifting, deformation, nearby heating or pressure reduction can each nudge the prepared rock beyond its reduced melting threshold.

Once melting starts, rare elements become concentrated because they preferentially remain in the liquid rather than being locked into common crystalline minerals.

This chain of events helps explain how ore bodies can form far from active plate boundaries while still retaining a much older tectonic signature.

Exploration gets narrower

For exploration teams, the study does more than reinterpret ancient geology: it also reduces the practical search space worldwide.

Only about 35% of continental crust falls within the mapped fertile zones, but those areas contain most known deposits.

“This research shows that the ingredients for these critical mineral deposits were put in place many million to even billions of years ago,” said Spandler.

On that basis, exploration can become more focused, with ancient tectonic belts offering companies and governments clearer targets and less reliance on guesswork.

Limits of the map

Even so, not every deposit lies within the delineated zones, and the framework intentionally leaves several ore-forming pathways outside its scope.

Brief subduction episodes, later crustal rearrangement, erosion and mantle plumes could all generate apparent mismatches or mask older signals.

In addition, the most ancient concealed source regions extend beyond the map’s 2-billion-year limit, meaning part of the deep-time record is not captured.

Despite these constraints, randomised tests fell within fertile zones only around one-third of the time - well below the observed match rate for real deposits.

Deep Earth legacy

Overall, the evidence suggests ancient collisions charged continents with the right chemistry, while much younger disturbances determined when those hidden ingredients finally melted.

More precise tectonic reconstructions could refine the targets further, particularly in ice-covered areas and in terrain older than the current mapping can reliably follow.