Committed warming

Why would seas keep rising for centuries even if emissions stopped tomorrow?

Imagine every engine, furnace and power station on Earth stopped tonight. Not reduced — stopped. It feels as though the problem should then stop too, the way a room stops getting warmer the moment you switch off the heater. Yet the standard expectation is that the sea would go on rising for centuries. Why should anything continue once the cause is removed?

The tempting answer is "because the heat is already in the system", and that is roughly right — but only roughly, and following it carefully turns up something that surprises most people: the air would in fact stop warming almost at once. The commitment is not where intuition puts it.

Start with a question of storage. Greenhouse gases do not heat the air directly so much as they slow the escape of energy to space, leaving the planet absorbing a little more than it emits. That surplus has to go somewhere. Where would you expect it to end up — in the thin skin of air we live in, or in the ocean?

The ocean, overwhelmingly. The IPCC's Sixth Assessment puts about 91 per cent of the accumulated excess energy in the ocean, with only around one per cent in the atmosphere. The air is a poor reservoir: light, thin, and quick to come to a new balance. Water is the opposite. So already we have two clocks running at very different speeds, and any question about "how long until it stops" has to ask what stops.

Now consider what actually happens to concentrations if emissions go to zero. Carbon dioxide already in the air does not stay put — the ocean and land biosphere keep drawing it down. So the forcing itself begins to fall. Meanwhile the ocean keeps taking up heat, which pulls surface temperature up. Two opposing effects, and here is the interesting result: in current assessments they very nearly cancel. The zero-emissions commitment — the further warming after CO2 emissions cease — has a best estimate close to zero, though models spread by a few tenths of a degree either side and the sign is not certain.

That is worth pausing on, because it corrects an older number that still circulates. Studies in 2005 by Wigley and by Meehl found roughly half a degree of additional warming "in the pipeline" — but they held greenhouse concentrations constant, which quietly assumes continued emissions to top them up. Constant concentration and zero emission are different experiments, and only the second is the one people mean when they say "if we stopped tomorrow".

So if air temperature roughly levels off, what is left to keep moving? Ask where the heat that entered the ocean has actually got to. Most of it is still in the upper few hundred metres. Mixing it downward is slow — overturning circulation timescales run to centuries and beyond. Warm water occupies more volume than cold water, so every additional metre of depth that eventually warms adds its own increment of expansion, long after the surface has settled.

And the ice is slower still. Ice sheets respond to a warmer ocean and atmosphere over centuries to millennia, not decades, because heat has to reach grounding lines and because their own geometry keeps adjusting. The Sixth Assessment's long-view figures are stark: over about 2,000 years, roughly 2 to 3 metres of sea level rise if warming is held near 1.5 degrees, 2 to 6 metres near 2 degrees, and 19 to 22 metres at 5 degrees. Those are commitments already implied by a temperature, not by continuing emissions — and their timing, particularly for Antarctica, is among the least settled numbers in the field.

Three refinements hold this together.

First, "committed warming" and "committed sea level rise" are not the same claim, and conflating them is the usual error. The best current estimate is that surface temperature stabilises within a couple of decades of net-zero CO2; sea level does not stabilise for millennia. One system reaches equilibrium quickly and one does not.

Second, stabilising is not reversing. A levelling temperature means the ocean is still warmer than it was, and the ice is still adjusting to that. Stopping the cause halts the growth of the imbalance; it does not undo the storage.

Third, other gases behave differently. Methane's atmospheric lifetime is roughly a decade, so its forcing decays quickly and cutting it produces relatively fast relief. Sulphate aerosols, which cool, wash out in days to weeks — so an abrupt halt to all combustion would remove that masking cooling almost immediately, producing a short warming bump before the longer-term picture settles. That transient is a real feature of the scenario, not a rhetorical flourish.

• Why the zero-emissions commitment is near zero in models but still carries a sign uncertainty.
• What "irreversible on human timescales" means for Greenland versus West Antarctica.
• How carbon dioxide removal would change the picture, and what it could not undo.

Every term the collection defines is gathered in the glossary.