Looks like the climate modeling community may have a civil war on its hands. Some serious players are rejecting the new hot models, but probably not their owners. If so we will see modeler against modeler. Be still my heart.

The first loud public shot has been fired by the prestigious journal Science (actually it is more of a magazine but never mind). Science is devoutly alarmist but they reject the hot models in the strongest possible terms (in a lengthy article that is not paywalled).

Their blunt article title is “U.N. climate panel confronts implausibly hot forecasts of future warming“. When it comes to science, “implausibly hot” is very strong language. Scientific language is normally extremely polite. (The U.N. climate panel is of course the IPCC.)

But it gets even stronger in the text. Here we find NASA’s Gavin Schmidt, arguably America’s top alarmist scientist and a major modeler, saying that using these models “You end up with numbers for even the near-term that are insanely scary—and wrong.”

The full Science article is well worth reading, see https://www.sciencemag.org/news/2021/07/un-climate-panel-confronts-implausibly-hot-forecasts-future-warming

For a leading modeler to describe a lot of the latest climate models as INSANELY SCARY is incredible. I suspect this is an act of war because I cannot imagine the owners of these newly hot models agreeing.

Moreover, as I first pointed out two years ago, there is a similar civil war going on in the political alarmist community. See my https://www.cfact.org/2019/09/28/is-climate-alarmism-tearing-itself-apart/. We have what I call the moderates versus the radicals.

For example, when Greta Thunberg gave her famous “How dare you” denunciation she was addressing the Paris Accord negotiators, who certainly are mainstream alarmists. But they are now the moderates in the climate movement civil war. Greta is a spiritual leader of the radicals.

Getting back to the war of the models, some history will help to see the scope of this fascinating fiasco. The IPCC itself does not do climate modeling. Instead, major modeling centers around the world do it, in a highly coordinated way. About 100 different models from around the world take part in performing this massive, multiyear exercise.

The coordination is provided by another UN agency, the World Climate Research Program. The exercise itself is called CMIP which stands for Coupled Model Intercomparison Project. The CMIPs are timed to feed into the big IPCC reports, so the latest is CMIP6. The IPCC is just now finalizing its AR6 reports.

The big blowup is because roughly half of the CMIP6 models ran much hotter this time than they did in CMIP5. Thus the modeling community faces a deep dilemma. If the newly hot models are right then all of the models have been wrong in CMIPs 1 through 5, and half still are. Either that or the newly hot models are all wrong. (Of course all are probably wrong, past and present, but that is a separate issue.)

So basically we now have moderate models and radical hot models. The Science article is the moderates saying the radicals are wrong. But I see no sign that the radicals agree. There has been no recall, telling researchers not to use the hot results. In fact Schmidt complains that the hot results are being used. This is where what he calls the insanely scary and wrong, near-term results come from. I find nothing on the CMIP6 website that even addresses this hugely disruptive issue. See https://pcmdi.llnl.gov/CMIP6/.

My conjecture therefore is that the hot model radicals are standing their ground. Given the strong language of the Science article this means civil war in the modeling community.

In fact I am pretty sure that the radicals juiced up their models deliberately. It appears that all these different modeling centers made the same change, which cannot have been a coincidence. And the change is such that they knew going in that they would make things much hotter.

So I do not see these lavishly funded modeling centers telling their funders that they goofed and the last five years work just didn’t work out, so sorry. Most of these funders are national governments which might react negatively to such an admission.

Interestingly the Science article says the insanely scary and wrong near-term warming is due to a deep error that has been in the models during previous CMIPs. The CMIP6 changes just brought it to the surface, as it were. This casts deep doubt on the prior CMIP results as well.

How the IPCC handles this model mess remains to be seen. The bigger question is how will the modeling community handle what looks like a civil war emerging. That community has always been a solid wall of (false) certainty, a position it can no longer maintain.

Climate alarmism has always been based on modeling. Now the modeling is a mess. Can alarmism survive the loss of modeling? Stay tuned.


  • David Wojick, Ph.D. is an independent analyst working at the intersection of science, technology and policy. For origins see http://www.stemed.info/engineer_tackles_confusion.html For over 100 prior articles for CFACT see http://www.cfact.org/author/david-wojick-ph-d/ Available for confidential research and consulting.