How thick is the ground beneath you?

Dig down beneath any spot on Earth and you pass first through the crust, then into the mantle. For a while the mantle stays cold and stiff, welded to the crust above it as one rigid layer. Then, somewhere between a few tens and a few hundred kilometres down, it becomes hot enough to creep. That change marks the base of the lithosphere, the bottom of the plate you live on. Under a mid-ocean ridge it comes very close to the surface; under the old heart of a continent it can sit deeper than the distance from London to Birmingham.

This page started with a beautiful map by Fabio Crameri on s-Ink, drawn from a model by Clinton Conrad and Carolina Lithgow-Bertelloni (2006). The s-Ink download holds only images, not the numbers behind them, and I could not find an open copy of that grid. So I built my own map from two newer open models that do publish their numbers: LithoRef18 (Afonso and colleagues, 2019) and WINTERC-G (Fullea and colleagues, 2021). Where they agree, the picture is solid; where they disagree, the map lets you see it.

01 · The mapThe base of the plate, everywhere

Colours show how far down the base of the lithosphere lies: pale for thin, deep plum for thick. By default the map shows the average of the two models. Switch to either model alone, to where they disagree, to the age of the sea floor, or to the thickness of the crust. Search a place or click anywhere for a cross-section of the ground beneath it.

02 · Old heartsThe deep roots of the continents

The thickest lithosphere sits under the cratons, the oldest and most stable cores of the continents, many of them more than 2.5 billion years old. Geologists think these regions owe their survival to deep, cold, chemically depleted roots, sometimes called keels, which are buoyant enough to resist being dragged back into the mantle. In both models the cratons stand out as islands of deep plum: across 14 representative points, the median depth to the base is 178 km in LithoRef18 and 199 km in WINTERC-G, and 11 of the 14 are deeper than 150 km in both.

The exception proves the rule. The eastern North China craton is ancient at the surface but sits on lithosphere only about 66 km thick here. Geologists widely describe it as a craton that lost much of its root, probably during the last 200 million years. Its big cities, including Tianjin, are among the thinnest-floored of any city of a million people. Click a card to fly there.

03 · Born thinHow the ocean floor thickens with age

Oceanic plates are made at mid-ocean ridges, where hot mantle rises, melts and freezes into new sea floor. As the new plate moves away it cools from above, and the cold, rigid layer grows downwards. Simple cooling theory says its thickness should grow with the square root of its age: four times as old, twice as thick. On sea floor younger than 2 million years LithoRef18 puts the base at a median of 42 km, rising to 118 km on sea floor older than 150 million years. Across all the ocean, its thickness tracks the square root of age with a correlation of 0.88.

WINTERC-G tells the same story more quietly, from 70 to 99 km, partly because its nodes are about 225 km apart, wider than the thin strip along a ridge. Both curves flatten for the oldest sea floor, as plate-cooling models predict. The oldest crust in the age grid, about 339 million years old, lies in the eastern Mediterranean.

04 · Fire and iceWhere volcanoes and diamonds sit

Thin lithosphere and volcanoes go together. Most volcanoes grow where plates pull apart or where one dives beneath another, and in both settings the rigid lid is thin. Of the 1,828 volcanoes with a Smithsonian Global Volcanism Program number in Wikidata, 71% stand on lithosphere less than 100 km thick (averaging the two models), and only 1% on ground deeper than 200 km. Their median is 84 km, well below the continental median of 117 km.

Diamonds are the opposite. They form in the mantle at depths of roughly 150 km or more, where pressure is high enough, and survive only where a thick, cold root keeps them stable. They reach the surface in rare, fast eruptions of kimberlite. Of the 63 diamond mines in Wikidata with coordinates, 87% sit on lithosphere more than 150 km thick. Their median root, 207 km, is deeper than 92% of all continental ground. That is the pattern prospectors call Clifford's rule: look for diamonds on the old cratons.

Wikidata coverage of mines is thin and uneven (63 sites, mostly large or famous ones), and some volcano entries are submarine. The pattern is clear, but the exact shares would shift with a fuller list.

05 · Close to homeThe UK and Lancashire

Britain is not a craton. Its rocks were stitched together from several older fragments, stretched when the Atlantic opened and warmed by the volcanoes of the Hebrides some 60 million years ago. Its lithosphere is middling: the UK averages 115 km across the two models, 73rd thickest of the 154 countries with at least 20,000 km² of land, close to the continental median of 117 km. Thickness falls to the north and west.

Under Lancashire the base lies about 90 km down in LithoRef18 and 99 km in WINTERC-G, of which roughly 33 km is crust. The nearest ground where both models put the base below 200 km is in Sweden, about 1,480 km away: the edge of the Baltic Shield, one of Europe's ancient hearts.

06 · League tableThickest and thinnest countries

Averaged across their land, the countries on the East European craton come out thickest, led by Lithuania (248 km). The thinnest are along young rifts and plate boundaries, led by Djibouti (53 km), at the mouth of the Red Sea where Africa and Arabia are pulling apart. Click a row to see it on the map.

Area-weighted means over 0.25 degree cells inside simplified Natural Earth outlines, countries with at least 20,000 km² in the grid. WINTERC-G stops at 300 km, so the Baltic states may be thicker than shown.

07 · Two modelsWhere the experts' maps disagree

Nobody can drill to the base of the lithosphere. Every map of it is a model, built from how seismic waves travel, how heat flows out of the ground, how high the land stands and how gravity varies. LithoRef18 and WINTERC-G use overlapping kinds of evidence but different methods, and they agree on the broad pattern: a correlation of 0.74 across the globe, with a median difference of 19 km. They differ by more than 50 km over 8% of the Earth's surface.

The biggest arguments are under the East European plain, where WINTERC-G reaches its 300 km ceiling, and under Tibet, where WINTERC-G finds a thick root of about 297 km and LithoRef18 only 127 km. Geologists have long debated Tibet's deep structure, so this is honest uncertainty, not an error in either map. Choose "Disagreement" above the map to see where the two part company.

The map that started this

Fabio Crameri's map on s-Ink shows Conrad and Lithgow-Bertelloni's 2006 model, which set oceanic thickness from the square root of sea-floor age and continental thickness from seismic tomography, with 100 km as the boundary between the two. It is reproduced here under its CC BY-SA 4.0 licence. Its broad pattern matches the two newer models: the same cratons, the same thin ridges.