Why Do Volcanoes Line Up in Rings and Chains?

Volcanoes line up because the things that make them are lined up. Most volcanoes form along the edges of tectonic plates, especially where one plate sinks beneath another, and the sinking edge is a long, curving line. A smaller number form in chains because a plate is slowly drifting over a fixed patch of unusually hot rock, which punches a new volcano through the surface every so often, like a sewing machine needle making stitches in moving cloth.

That's the short answer. The rest of this article is the long one, taken a step at a time. I'll start with a map you can picture in your head, then go underground and explain why rock melts where it does. We'll finish with a few things scientists are still working out. I'm still learning this subject myself, so where I'm not sure, I'll say so.

Start with the map: where volcanoes actually are

If you plotted every active volcano on a world map, you wouldn't get an even sprinkle of dots. You'd get lines and curves, with big empty areas in between. The most famous pattern is the "Ring of Fire" around the Pacific Ocean. It's really more of a horseshoe than a ring, and it runs through New Zealand, Japan, the Aleutian Islands, the Andes and more. Another important line runs down the middle of the Atlantic, mostly hidden under the sea.

Here in Aotearoa, you can see the pattern on a good day. Ruapehu, Ngāuruhoe and Tongariro stand in a row in the central North Island, and Taupō's great lake fills an old volcanic hollow nearby. The line carries on north-east out to sea. Further out, it continues as a chain of volcanic islands and undersea volcanoes towards Tonga. A line that long isn't a coincidence.

The key idea, and it's worth a moment to let it sink in, is that volcanoes mark places where the ground is doing something unusual. Mostly that something is the movement of tectonic plates.

A quick refresher on plates

The outer shell of the Earth, the lithosphere, is broken into large slabs called tectonic plates. They're rigid, and they ride on the hotter, softer rock below. "Softer" needs care here, and we'll come back to it. The plates move at a few centimetres a year, roughly the pace your fingernails grow, though different plates move at different speeds.

Where two plates meet, one of three things happens:

• They pull apart. Rock wells up into the gap, as along the mid-Atlantic ridge.
• They slide past each other. Think of New Zealand's Alpine Fault, which has lots of earthquakes but no volcanoes of its own.
• They push together. One plate usually ends up diving beneath the other. This is called subduction, and it's responsible for the best-known volcano lines on Earth.

Alfred Wegener first argued, early in the twentieth century, that continents drift. The fuller theory of plate tectonics came together in the mid-twentieth century, once the sea floor had been mapped and measured well enough to test the idea. It's now one of the best-established frameworks in all of Earth science.

Why one plate sliding under another makes volcanoes

This is the heart of the matter, so I'll go slowly.

Off the east coast of the North Island, the Pacific Plate is moving westwards and sliding down beneath the plate the North Island sits on. The ocean floor on the Pacific side is cold, dense and heavy, so it sinks, dragging itself down into the hotter mantle. That sinking slab is the engine of the whole system.

Now for the question that confused me for years: why does anything melt? As the slab descends, it gets hotter and the pressure on it climbs, so you might expect the rock to melt simply because it's deep and warm. But that's not the main story.

Here's why: water changes the melting point

The mantle is solid rock. Hold onto that, because it's the most common misconception I come across. People picture a sea of lava under the crust, with volcanoes as leaks. The mantle is almost entirely solid. It flows, but only over huge spans of time, like very stiff putty. Melting it takes special circumstances.

The ocean floor that's being dragged down is soaked with water. Some is trapped in the pores of the rock, and some is locked inside minerals. As the slab sinks, heat and pressure squeeze this water out. It rises into the hot mantle wedge sitting above the slab.

Water doesn't heat that rock up. What it does is lower the temperature at which the rock begins to melt. You already know a version of this from the kitchen and the road: scatter salt on an icy path and the ice melts, even though the air hasn't warmed. The salt lowers the freezing point. Water seeping into hot mantle rock works in a similar spirit, lowering the melting point so that a small fraction of the rock turns liquid. Geologists call this flux melting.

Two details are worth getting right:

1. Only a small portion of the rock melts. The result is more like a solid sponge with some liquid in its pores than a pool of liquid.
2. The melting mostly happens in the mantle above the slab, not in the slab itself. Many people (including me, when I first learned this) assume the sinking plate melts like a candle. Mostly it's the wedge above it that does.

From melt to mountain

Once a bit of melt forms, it's lighter than the solid rock around it. So it rises, slowly gathering and travelling upwards through cracks and channels. Some of it stalls and cools underground. Some reaches the surface, and then you have a volcano. The melt that arrives tends to be rich in dissolved gas, which helps explain why many arc volcanoes can erupt explosively, though details vary from one volcano to the next.

A quick check for understanding: if someone told you that volcanoes happen because the sinking plate melts from friction, what would you say? You might say that friction and heat do play a part, but the chief reason for melting is water released from the slab, which lowers the melting point of the mantle above it. If that came easily, you've got the central idea.

Why they line up: a worked example

Now we can answer the "line" part of the question. Melting needs the slab to be at the right depth. Under many volcanic arcs, that depth is roughly a hundred kilometres or so. It varies from place to place, and I'm giving a round figure rather than a precise one.

Imagine a slab diving at a constant angle. Volcanoes will appear above the place where the slab reaches that depth. Here's the arithmetic, using simple trigonometry:

• Suppose the slab dips at 45 degrees and the key depth is 100 km. Horizontal distance from the trench = depth ÷ tan(angle) = 100 ÷ tan 45° = 100 ÷ 1 = 100 km.
• Now suppose the slab dips more gently, at 30 degrees. Horizontal distance = 100 ÷ tan 30° = 100 ÷ 0.577 ≈ 173 km.

So a shallower dip pushes the volcanoes further from the trench, and a steeper dip pulls them closer. This is a simplification, because real slabs bend and the melting zone isn't a neat line. But it explains something useful: along a stretch where the slab dips at a roughly constant angle, the volcanoes sit at a roughly constant distance from the trench. A trench is a long, narrow valley in the sea floor, and it runs in a line. Parallel to it, at that steady distance, is another line, the volcanic arc.

That's the "chain" in a nutshell. The volcanoes line up because the depth contour on the slab lines up, and the trench it follows lines up too.

Why the arcs are curved

You might also have noticed that many arcs are curved rather than straight, which is why we call them arcs. The Aleutians and the islands of Japan are good examples. One standard explanation is pure geometry.

Take an orange, or a tennis ball, and push a flat card into it. Look at where the edge of the card meets the surface: it's a curve. The Earth is a sphere, and a flat sheet diving into a sphere meets the surface along a curve. Curvature of this kind is a natural result of a stiff plate sinking into a round planet.

I'd gently add that this isn't the whole story. Real arcs have kinks and segments and changes of direction, and local factors matter. But the orange test is a lovely, cheap experiment, and it's the sort of thing I like to try at the kitchen table.

Island chains in the middle of plates: hot spots

Not every volcano belongs to a plate boundary. Hawaii sits near the middle of the Pacific Plate, thousands of kilometres from the nearest edge. Yet its volcanoes also form a line.

The explanation, which Tuzo Wilson proposed in the 1960s, is the hot spot. The idea is that there's a long-lived source of unusually hot, rising rock in the mantle, sometimes pictured as a plume. It doesn't move much compared with the plate above, and the plate moves across it. Think of holding a sheet of paper over a candle flame and sliding the paper along. You'd get a line of scorch marks, one after another.

A volcano forms over the hot spot, then the plate carries it away, and a new one forms behind it. The result is a conveyor belt of volcanoes, with the youngest at one end and the oldest at the other. In Hawaii, the Big Island is the youngest and still active. The islands to the north-west are older, lower and more worn down, and beyond them are islands that have eroded below the waves. Water and weather wear volcanoes down over time, and that's one reason the age pattern is easy to read.

This is a pleasing link between geology and everyday observation. A farmer once taught me to look at a bank and read it from the bottom up, oldest layers first, youngest on top. An island chain is the same idea laid out sideways across the sea: the oldest stories at one end, the newest at the other.

Misconceptions and exceptions

A few things are worth gently correcting.

"All volcanoes are on plate edges." Not quite. Hot-spot volcanoes like Hawaii's are in plate interiors. And there are other volcanoes that don't fit neatly into either box. In New Zealand, for instance, there are old volcanoes around Banks Peninsula and Dunedin, and a field of small volcanoes under Auckland, none of them on a plate boundary. Scientists are still working out exactly why melt forms in such places, and I wouldn't want to claim more than that.

"The Ring of Fire is a literal ring." It's a horseshoe, and it's made of many separate segments. The pattern is real, but the name is a figure of speech.

"Underground there's a lake of lava." As above, the mantle is solid. Magma chambers are better pictured as crystal-rich, mushy zones than as open caverns, though that's a topic where the picture keeps being refined.

"Hot spots are fully understood." Also no. The basic idea, a plate moving over a persistent source of melt, is taught in textbooks. But how deep the source is, and how plumes behave, are still debated. A famous bend in the Hawaiian–Emperor chain, where the direction of the line changes, has prompted discussion about what changed and when. I haven't tried to go into that here, because the details are still being studied.

What I haven't covered, and what to wonder about

This article sticks to the broad picture, which has been tested for decades. I haven't covered how magma chambers are structured, why some eruptions are gentle and others violent, or how to forecast activity. Those are rich subjects, and parts of them are still active research.

If you'd like to try something, it needn't cost anything. Next time you look at a map of New Zealand, trace the line from Ruapehu north-east to the sea, and imagine the cold Pacific floor sinking beneath you, about a hundred kilometres down, squeezing out water as it goes. Then compare it with a map of the Hawaiian Islands, where the line is made by a different process, and try to work out which end is oldest.

And here's something to wonder about. The slab that's sinking beneath the North Island started its life at an ocean ridge far away, as fresh rock. It cooled, travelled, collected sediment and water, and is now returning to the mantle, where it will be slowly warmed and reworked over a span that makes our human timescales look small. Some of what leaves a volcano is, in a very roundabout way, the sea floor coming back up. How much, and by what routes, is something I'm still learning about, and I'd be glad if you were curious about it too.

Thank you for reading. If any step felt murky, that's normal. It took me a good few attempts, and a lot of oranges, to get it straight.