One of the graphs that will not be making an appearance in our #CheerfulCharts series is fertility rates around the world, which are crashing to levels that foretell a depopulated dystopian future. While climate alarmists might find the reduction or elimination of CO2-spewing, planet-wrecking human units a cheerful prospect, here at CDN we do not because we think people are swell, especially babies. As population projections keep bending down, some researchers like Roger Pielke Jr. (h/t) are asking what that implies for global warming projections. We have referred many times to the off-the-scale scam known as RCP8.5, but the IPCC now uses the so-called SSP projections, which also include a lot of impudently off-the-scale impossible projections, but also one rather low projection called SSP2-4.5. Scientists ignore it because they think it foretells far too few emissions. But in a recent Substack Pielke Jr. points out that even SSP2-4.5 includes population growth assumptions that are now likely too high. Which reminded us of a monumental study from March 2024 by Italian climate scientist Nicola Scaffeta that’s been on our to-do list ever since. Scaffeta asks what climate models would project if we (first) selected only those that have been accurate over the post-1980 interval then (second) plugged SSP2-4.5 into them. The answer is modest and not-alarming warming between now and 2100. But Pielke Jr.’s post makes us wonder what if we (third) correct for the fact that even SSP2-4.5 population projections are too high? You’ll have to squint to see the resulting warming.

Scaffeta’s paper, “Impacts and risks of ‘realistic’ global warming projections for the 21st century”, is one of those studies that appears in a top quality peer-reviewed scientific journal, in this case Geoscience Frontiers. But because it’s off-message it might as well have been written in invisible ink and buried under the Antarctic ice cap for all the attention the press gave it.

Scaffeta does a very deep dive into the complexities of climate modeling and reviews an enormous amount of data on climate sensitivity, showing that if we ask for evidence that models can reproduce the warming of the past 40 years we find they fall into three categories. But it’s not a Goldilocks scenario because some of the models are way too hot, some are too hot, and some are about right. The three categories correspond, respectively and unsurprisingly, to CDN readers at least, with the models that have high, medium and low climate sensitivity to CO2. And for good measure Scaffeta also discusses studies looking at tropospheric warming that likewise find all the models warm too much, some far more than others, and only the models with low sensitivity manage to reproduce the recent past realistically.

He then goes over more ground that will be as familiar to CDN readers as it is obscure to most climate journalists, namely the abuse of RCP8.5 and the need to use realistic energy and emission scenarios when projecting 21st century warming. So Scaffeta uses SSP2-4.5 and then looks at model projections selected based on how well they reproduce surface or tropospheric warming. The results are as follows:

The top chart uses models checked against surface warming and yields just under 2.5 C warming (compared to 1850) by 2100. The bottom chart selected models checked against tropospheric warming and results in only about 1.6 C warming by 2100.

As noted above, Roger Pielke Jr. points out that even SSP2-4.5 now has unrealistic global population growth assumptions in it. Which means an even lower projection called SSP1 is now looking more realistic. Also RPJ refers to some new economic projections that indicate slower GDP growth over the century than was assumed in the climate scenarios. But swapping in SSP1 population growth and reduced GDP growth rates knocks a further half a degree off projected warming this century. Which for the satellite-calibrated model group above would result in under 1.5 C total warming compared to the preindustrial baseline, which means we come in under the Paris target without having to put in any costly and destructive climate policies.

While we hope the current demographic slump ends and gives way to a renewed vote of confidence in the human race in the form of bumper crops of babies, it meanwhile has important implications for climate policy, namely that in addition to being costly, inefficient, intrusive and annoying, it is also unnecessary, once you confine your attention to models that actually work and emission scenarios that bear some resemblance to reality.