Earth is warming faster. Scientists are closing in on why

Gavin Schmidt, a leading climate modeller and the boss of NASA’s Goddard Institute for Space Science (GISS) in New York City, is not noted for his humility. Nevertheless, writing in Nature, a journal, in March 2024, he confessed to being humbled by his inability, and that of his colleagues, to understand the extraordinary year through which they had just lived—2023 had been around 0.2°C (0.4°F) hotter than had been expected.

Not just humbled: worried, too. If climate modellers’ accumulated knowledge and spiffy models could not explain what had just happened, it might mean that climate change had pushed the workings of Earth into “uncharted territory…fundamentally altering how the climate system operates”. Both the speed of climate change and the workings of the climate might be changing. The future might look even worse than it used to.

Nine months later, in Washington DC, Dr Schmidt and his colleagues returned to the subject at a recent meeting of the American Geophysical Union (AGU), the world’s largest annual gathering of Earth scientists. The sessions that took place on the topic felt at times like a murder inquiry, with the evidence for one suspect or another gone through meticulously. The probable verdict is now clearer than it was in March; some suspects have been ruled out and new clues have emerged which point to some others. The conclusion looks likely to be that the world can expect somewhat higher rates of warming. But the case is still not closed.

It was always going to be hot in 2023. Climate change forced by greenhouse gases means that all years can now be expected to be warm by past standards; in 2021 the Intergovernmental Panel on Climate Change put the rate of warming at 0.2°C a decade. What is more, the second half of 2023 saw an El Niño get under way.

El Niños are the warm phase of a seesaw of winds and ocean currents in the tropical Pacific called ENSO (El Niño-Southern Oscillation). The extra heat that such events can add to the overall warming trend means that El Niño years often set records for global temperatures. Because the El Niño that began in 2023 carried on into the following year, 2024 has therefore ended up being even hotter than the last year (see chart).

Whodunnit?

But if not the hottest year on record, 2023 still ranks as the strangest. For one thing, records were tumbling well before the El Niño kicked in in the second half of the year. For another, the scale of the warming compared with the year before was beyond what anyone would normally expect from an El Niño. For a third, the pattern of warming across various ocean basins was very peculiar.

At the time, several additional “forcings” were discussed. The underwater eruption of the Hunga Tonga-Hunga Ha’apai volcano in the South Pacific in January 2022 had put a huge amount of water vapour into the normally desiccated stratosphere; water vapour is a greenhouse gas, and in the stratosphere it sticks around for a long time. The Sun was reaching the peak of its 11-year sun-spot cycle; during such “solar maxima”, it provides around 0.05% more light than it does on average and its spectrum skews into the ultraviolet. And in 2020 new rules imposed by the International Maritime Organisation (IMO) had slashed the amount of sulphur allowed in the fuel of ships on the high seas.

Sulphur in ships’ fuel turns into sulphate particles rising from ships’ funnels. Some of those particles end up blowing ashore; producing fewer of them cleans the landlubbers’ air and saves lives. But the particles also encourage the formation of clouds, brighten clouds already there and reflect away sunlight even if the air is too dry for any clouds at all: all these effects cool the sea’s surface.

As soon as the IMO rules went into effect there were climate scientists keen to see what they did to temperatures. 2023’s spike added to the excitement. At the AGU Andrew Gettelman of the Pacific Northwest National Laboratory in Richland, Washington, presented an overview of studies of the phenomenon. Observations show that the distinctive lines of cloud—known as “ship tracks”—that can stretch out behind vessels burning sulphur-rich fuel are indeed much rarer now. Modelling suggests that, overall, this means something like 1.2 more watts per square metre of sunshine are warming the ocean.

That is enough to have a significant effect, but not enough to provide all the necessary warming for 2023. Nor can the concerted action of all the initial suspects suffice. The solar effect is smaller than the fuel effect. The volcano’s effect seems to point the other way. Volcanoes, too, throw sulphur up into the atmosphere. According to Mark Schoerberl, of the Science and Technology Corporation, the long-lived sulphate particles created in the stratosphere after the eruption of Hunga Tonga-Hunga Ha’apai more than counterbalanced the warming provided by the water vapour, providing a small net cooling.

Evidence against different culprits comes from work published recently in Science. Helge Goessling and his colleagues at the Alfred Wegener Institute in Bremerhaven used satellite data and weather records to show that over the course of this century Earth has gradually been reflecting less sunlight back into space than it used to. 2023 was the dimmest year to date. This was apparently due to paucity of cloud cover, particularly in the northern mid latitudes.

Part of this could be down to the new IMO rules, but the dimming is too strong to be explained by that alone. Bjorn Samset of CICERO, a Norwegian climate research institute, points to another possibility: the lack of sulphate emissions is not a result of cleaner ships, but of cleaner Chinese coal-fired power plants. Since 2014 China has been making progress in reducing sulphur emissions by closing particularly noxious power plants and scrubbing sulphur out of the flue gases at others. New data leads Dr Samset and colleagues to think the cleanup is having a marked effect across the North Pacific, where cleaner air and fewer clouds will mean more warming.

A lung-sparing dearth of sulphates may not be the only thing making Earth less reflective. As the climate warms, its workings change in all manner of ways. One is that the tropics expand, and the tracks of storms in temperate zones narrow. Narrower storm tracks mean less cloud. This sort of shift could be another reason why Earth is growing less reflective, and thus warming more.

Both the sulphur stories and the changing cloud patterns suggest that increased warming may be here for some time to come. Models expect warming to speed up as annual emissions get larger, which they continue to do, and as sulphate emissions fall, which they continue to do. Dr Schmidt’s predecessor as the boss of GISS, James Hansen, goes beyond what many of his colleagues are comfortable with when he argues that this effect is already apparent and large. That said, warming over the decade to 2023 was 0.26°C; not as high as the 0.32°C a decade rate that Dr Hansen thinks is the new normal, but well above what it used to be.

At the end of the AGU sessions Dr Schmidt felt that there had been real progress on the various possible culprits. In the next few weeks he expects modellers at GISS and elsewhere to start trying to pull them all together into a coherent narrative in new climate-model runs that use the most up-to-date data on both sulphur emissions and the reduction in reflected light. Picking over the results may allow scientists to say with some certainty what actually happened.