Javier Vinós, our guest in a December 2024 podcast, recently posted a detailed essay on Judith Curry’s blog discussing the remarkable global warming event of 2023-24 which broke all kinds of records and left climate modelers scratching their digital heads in confusion. He’s been watching the process closely since 2024 when he argued that the warming was likely due to the 2023 underwater Hunga Tonga volcano injecting huge amounts of the key greenhouse gas dihydrogen oxide into the atmosphere (a topic we’ve discussed many times). But climate science officialdom said no, the volcano caused net cooling as above-water ones do, and the warming had to be due to other factors. You know, like say man-made CO2. In mid-2025 Vinós replied with a testable proposition: if he was right we should see the warming reverse over the coming two to three years rather than going higher and higher as implied by the official explanations. And now his essay points out that global sea surface temperatures are almost back to where they were in 2022. As he predicted, but the climate models did not. Their failure, he argues, shouldn’t surprise anyone since they never get volcanic effects right.

At least not big volcanoes. Vinós reviews climate model reproductions of the temperature effects of the 1815 Tambora volcano and shows that even when they get to peek at the answer in the back of the book they still can’t get the timing or magnitude of the climate effects right. So we shouldn’t suppose that climate models supplied with the details of the Hunga Tonga volcano, a large amount of sulphur aerosols and an even larger amount of water vapour, will predict its effects correctly.

In his paper Vinós reviews many weird weather events of 2023, including a large reduction in global cloud cover which was the immediate trigger for surface warming. And while he doesn’t claim to know what the physical mechanisms behind them all were, he does point out that the usual explanations just don’t work. Sometimes the timing is wrong, like trying to blame it all on El Niño. Blaming it on CO2 also doesn’t work because there was no unusual surge in CO2 levels. Some have argued it was due to a reduction in atmospheric sulphate levels caused by new emission regulations on marine tankers, another attempt to salvage the CO2 paradigm at almost any cost. But Vinós shows the data and points out that the actual reduction wasn’t large enough to affect the global climate.

All the way down the list the problem is the same: nothing happened in 2023 that would explain a temporary surge in ocean and atmosphere temperatures except the Hunga Tonga volcano. But climate science says that despite the volcano pumping massive amounts of water vapour into the upper atmosphere, it can’t have caused warming because the models say CO2 is the driver. Alternative explanations to Hunga Tonga for the 2023-24 warming imply that it will be either permanent or last a long time, which is again testable. But Vinós points to the inconvenient sea surface temperature data to say they failed the test:

As of the start of 2026 the oceans have cooled about 0.2 degrees C compared to the start of 2024, leaving them back where they were in mid-2022.

Vinós also goes over what the climate news summaries have had to say since 2022 and it’s not pretty. Up to early 2025 there were lots of stories about the warming surge. But since then none of them have discussed the rapid cooling. Because of course.

Vinós’s conclusion is that the past few years signal a big failure of climate science. A huge warming event occurred which models can’t explain. A likely cause presents itself but the models can’t handle it so it gets ruled out because orthodox climate scientists prefer the models to reality. In which the predictions don’t line up with the data.

The science ain’t settled. And sometimes it takes a well-informed outsider like Vinós to point out the problems.