The funny paper that dumps the Atlantic Meridional Oscillation (AMO) is so unconvincing it had to be published twice.
It’s been called a tectonic shift in climate science resulting in the junking of thousands of peer-reviewed climate research papers. It’s also been held up as a prime example of why peer-review of some aspects of climate science is broken. Strange that you might not have noticed. It was published in a major journal last week. There was also a press release.
It’s a paper whose prime author is Michael Mann of Penn State University. Its main conclusion is that there is no internal variability in the Earth’s climate system, at least not in the past thousand years. All variability is down to just two things, greenhouse gas forcing and aerosols from volcanoes. Nothing else is needed confirm computer models.
The main target of the paper is the Atlantic Meridional Oscillation or AMO (Mann coined the term but didn’t discover the effect), a periodic surface temperature cycle of the North Atlantic. It was shown to be an artefact, not real, the authors claimed.
The AMO met its nemesis via a computer model, a useful tool but which is often regarded as superior to reality. This is the case even when the models are so complicated they need the latest supercomputers and provide a range of outputs for a given set of input parameters.
Mann and company find variations in AMO data, specifically a 50–70-year spectral peak seen in an analysis of 27 proxy records dating back to 1400, although other studies have yielded differing results. Comparing them to a volcanic activity dataset the authors conclude that the AMO isn’t natural but is an artefact due to a combination of greenhouse gas forcing and cooling due to sulphate particles given off by volcanoes. Considering the last millennium, they say, the multi-decadal variations are due to pulses of volcanic activity. They also add that there is additional evidence for this conclusion because in the climate models there are no multi-decadal variations. That’s it then. QED.
Writing in her blog Judith Curry is unimpressed. “Wow. In one fell swoop, the pesky problems of the ‘grand hiatus’ in the mid 20th century, debates over the attribution of 20th century warming and the role of multidecadal internal variability, and the difficulty of attributing the recent increase in Atlantic hurricane activity to AGW, all go away. Brilliant! Almost as ‘brilliant’ as the Hockey Stick.”
She goes on to say, “Relying on global climate models, which don’t adequately simulate the multi-decadal internal variability, to ‘prove’ that such multi-decadal internal variability doesn’t exist, is circular reasoning (at best). How does this stuff get published in a journal like Science? Peer review is sooooo broken.”
Substantial discussion and disagreement
In fact, the research that dumps the AMO is so unconvincing it had to be published twice. As well as coming out in last week’s edition of Science, essentially the same findings were published in Nature Communications in January 2020. Michael Mann’s university even issued a press release about it then.
In the Nature Communications paper, the researchers say that if the AMO and the PDO (Pacific Decadal Oscillation) existed we would see them in our current simulations “given the current sophistication of climate models.” QED, again.
The Nature Communications paper was originally submitted in July 2019 (it had previously been submitted to another journal) and one Nature reviewer was lukewarm about its merits wondering if Nature Comms was the right place for it (“certainly interesting enough for a more specialised than Nature Communications.”) It was remarked that the computer models referred to were not good enough to be used to make definitive statements about the reality or otherwise of narrow-band natural multi-decadal or interdecadal variability.
This was pointed out by Kravtsov et al. (2018), who concluded that: “While climate models exhibit various levels of decadal climate variability and some regional similarities to observations, none of the model simulations considered match the observed signal in terms of its magnitude, spatial patterns and their sequential time development. These results highlight a substantial degree of uncertainty in our interpretation of the observed climate change using current generation of climate models.”
Judith Curry emphasises this point, “There is substantial discussion and disagreement in the climate dynamics community on this topic, which isn’t surprising given the apparent complex interactions between ocean circulations and the AMO, weather and interannual climate variability, and external forcing from the sun and volcanoes.” The paper had its supporters, and it did get through.
“In closing,” says Curry, “Mann’s quest to cancel the Medieval Warm Period and now the AMO, in the interests of showing that recent warming is 100% anthropogenic, is not at all convincing to scientists who understand anything about climate dynamics and global climate models.”
There is another way of looking a Mann’s findings. Roger Pielke Jnr remarked, “Maybe it’s just me, but it would seem that it should be much bigger news that 15,000+ peer-reviewed climate research papers published since 2000 are based on a non-existent phenomenon and are thus now discredited. “
Among them would be a recent paper by two Harvard University scientists.
They want to explain why sea surface temperatures were higher during World War 2, a longstanding topic of debate. They conclude that better corrections are needed for the way the temperature was measured using the intake of water from ships. They find a solution that accounts for the increase. Interesting then that both ways that account for the World War 2 ocean anomaly. Two examples of observational bias.