How to think about a looming climate threshold

What we can say about the 1.5 degrees Celsius limit today.

Arctic sea ice patterns. Credit: NASA/Kathryn Hansen, via Flickr. CC BY 2.0.

We’ve been living through something extraordinary.

In the middle of 2023, amidst a strengthening El Niño event, global temperatures began to soar. The second half of 2023 was very hot, and 2024 as a whole was even hotter. This latter year, when all evidence is considered, appears likely to have been the first on record to exceed a key temperature threshold, according to the World Meteorological Organization.

That level is 1.5 degrees Celsius (2.7 degrees Fahrenheit) above pre-industrial temperatures, the latter usually defined as the average temperature between 1850 and 1900.

The El Niño event is long over now — but 2025 hasn’t cooled down much yet. That’s certainly true if you consult one of the most popular datasets, from the European Union’s Copernicus Climate Change Service. Late last week Copernicus released a monthly summary putting April of 2025 at just above 1.5C — 1.51C, to be precise, making it the 2nd hottest April in their record, which dates back to 1940.

“There have been a total of 26 months where the global temperature exceeded 1.5°C above pre-industrial levels, most of them since mid-2023,” said Copernicus Climate Change Service senior scientist Julien Nicolas in an email.

In the figure below, I’ve displayed this full dataset by year and month, in the style Copernicus often uses. Then I highlight all the months with values of 1.5C or above:

Update: This post was originally published in May 2025, when the data for April 2025 were the most recently available. However, I have continued to update the chart since then to show additional months of data.

Twenty-six months sounds like a lot. However, scientists have long cautioned that whether the world has truly crossed 1.5C should be assessed over periods of many years, rather than on a monthly or yearly basis. This is to ensure that natural wobbles, like El Niño, don’t skew our sense of where we are.

How the datasets diverge on 1.5C

The Copernicus dataset, more formally known as ERA5, runs warmer than some others. For Copernicus, the full year of 2024 easily exceeded 1.5 degrees C, for instance. But both U.S. global temperature datasets — from NASA and NOAA — found it fell slightly short. NASA put the year at 1.47C.

The same goes for the monthly values. NASA also recently released its temperature for April 2025, similarly showing the second warmest April on record. But based on this explanation, the month would not appear to exceed 1.5C in the agency’s accounting. Update 5/13/2025: Meanwhile, on Tuesday Berkeley Earth also released their April 2025 temperature assessment, and put the month at 1.49 degrees C.

These little differences shouldn’t distract from the broader picture — which is that the Earth in the last few years has been strikingly warm, inescapably raising questions about how close we are now to the 1.5 degrees Celsius threshold.

Still, it is worth understanding some reasons for the disparity here.

The traditional temperature datasets integrate vast numbers of thermometer measurements on land with measurements of sea surface temperatures, or SSTs. ERA5 is very different: It is a reanalysis, perhaps best thought of applying a weather model to past climate data to create a seamless global picture. Along the way, it uses air temperatures over the ocean, rather than SSTs — which makes a difference.

“We know that SSTs generally vary more slowly than air temperatures,” Nicolas said.

Climate science tradecraft aside, though, the datasets are ultimately fairly close to one another.

Knowing when we’ve crossed

For more perspective on where we are, I reached out to climate scientist Zeke Hausfather, who works with Stripe and Berkeley Earth and Carbon Brief and also writes here at the Climate Brink.

“We’re on the cusp of 1.5 degrees,” Hausfather said. “Whether we formally pass it in 2026, or 2028, or 2030, to be honest, doesn’t really matter, because we’re going to pass it.”

In mid-2024, Hausfather wrote a valuable rundown of how to assess when we’re really at 1.5C. The official approach is to take the temperature average over many years — say, 20 of them. But there are many other methods. Just extrapolating from where we are, Copernicus is now suggesting the world could exceed the threshold as soon as mid-2029, for instance.

What’s notable, Hausfather said, is that the various estimates have been getting closer to the present recently. That’s because of the strong recent warming.

Two recent studies in Nature Climate Change also tried to look at this question. One found that we’re probably within a 20-year window whose temperatures will average 1.5C. Another — strikingly — found that we could already be past the year for which we will say (retroactively, of course) that the threshold was breached.

The question of why the Earth seems to have leapt into a hotter regime is also a huge matter of scientific discussion right now. Many experts feel the 2023-2024 El Niño, alone, can’t explain how hot things got, and how long they’ve stayed that way. Proposed additional causes include changing clouds, reductions in marine shipping emissions, and much else.

Hausfather thinks what we’re seeing reflects declining air pollution. It’s not just shipping: He notes that China, the world’s largest emitter, and a number of other nations have also lowered levels of sulfate pollution (which reflects sunlight back into space and can mask the true extent of warming).

“When you add together shipping and China and the rest of the world, you’re probably talking about .14C additional warming from sulfur reductions,” he said.

“It’s a pretty big change.”

Global emissions: Still rising

I covered the era during which 1.5C, rather than 2C, became the climate target we focused on the most. It was driven by two major events.

The first was the Paris climate agreement, in which small island nations and vulnerable least developed countries successfully pushed to have the 1.5C target included, as at least an ambition.

The second was the release of the U.N. Intergovernmental Panel on Climate Change’s special report on 1.5C in 2018. It found that exceeding this threshold would bring on major impacts, including losses of coral reefs and possible destabilization of the ice sheets.

But while the report confirmed that holding the Earth below 1.5C would have many benefits, it also showed this would be extraordinarily difficult to achieve.

In the crucial table 2.4, the document laid out the scale of emissions reductions required by the year 2030 to achieve the goal with only a “low overshoot” (i.e., not warming above 1.5C very much or for very long, and ending the century below that level). They were staggering. It essentially boiled down to cutting the world’s greenhouse gas emissions by almost half in a decade.

It is now 2025 and there’s no sign of anything like that kind of change. Indeed, 2024’s emissions appear to have reached another new high.

It has also become clear these various modeled scenarios – low overshoots, high overshoots, and so on -- make big assumptions about future technological changes. The biggest is that we will deploy massive scale carbon removal technologies to help avoid further warming, or even cool down the Earth, later in the century. That’s a dicey thing to bank on.

And this gets at the real challenge for the 1.5C goal. The latest temperatures get the headlines, but it’s really all about emissions continuing to rise when they’re supposed to be falling in the 1.5C scenarios.

And yet … the dynamics of emissions and subtractions create endless hypotheticals. One can always imagine overshooting by a large amount — reaching a warming level of 1.75C, let’s say — and then deploying utterly massive carbon removal to bring about global cooling.

Still, I was struck by the language used recently by emissions expert Glen Peters from the Center for International Climate Research in Oslo, Norway:

In the end, perhaps history gives the best perspective. Fairly recent climate history.

I did some reading to jog my memory, and it turns out that in 2015, we were talking about a year breaching the 1 degree Celsius threshold for the first time. The UK Met Office’s Hadley Centre put out a big release about it. When I covered 2015’s then record temperature, I quoted NASA and NOAA officials confirming this.

The tone was different then. But if you swap out the numbers in those stories with “2024” and “1.5 degrees C,” I’m struck by the parallels.

Now think forward to 2035. We can’t know what our future selves will be able to see about today’s position along the trajectory of temperatures. But look at where we’ve already been, and how quickly the records fall.