Guest “it’s tough to make predictions, especially about the future” by David Middleton
Climate Predictions “Worse Than We Thought”
By Patrick J. Michaels
July 14, 2020
As the temperature of the eastern U.S. normally reaches its summer maximum around the last week of July, every year at this time we are bombarded with tired “climate change is worse than we thought” (WTWT) stories. These stories take time to produce, from imagination to final copy to editing to publication, so they have usually been submitted well in advance of the summer peak. Hence, orchestrated fear.
For once, I’m in agreement about the WTWT meme, but it’s about the climate models, not the climate itself.
Pat Michaels goes on to discuss John Christy’s analysis of the CMIP5 models which showed that only the Russian INM-CM4 model, and its low sensitivity (~2 ⁰C per doubling), was close to reality. This is from Andy May’s November 2018 post:
Then he notes that the new CMIP6 models are even worse than CNMIP5:
You’d think that, in the seven-year period between CMIP5 and CMIP6, the modeling community would address the critical errors that all the other models were making. But CMIP6 models are out and are indeed “worse than we thought.” While their error in the tropical atmosphere is very close to the same as it was in CMIP5 (which isn’t good), the range of global predictions is even larger than before. This is also not a good sign, especially because it has been documented that the models are “tuned” to give an answer that the modeling team wants.
The funny thing is that he cites a Carbon Brief article by Zeke Hausfather which is actually very good. This plot is from the article:
Mr. Hausfather notes that high sensitivity models often don’t even get the past right:
Many high sensitivity models have poor hindcasts
Climate models provide both projections of future warming and “hindcasts” of past temperatures. These hindcasts can be used as a tool to evaluate the performance of models, though historical temperatures are only one of many hundreds of different variables that climate models generate.
A number of the higher sensitivity models in CMIP6 have had trouble accurately “hindcasting” historical temperatures. Some show almost no warming over the 20th century — with cooling effects from aerosols almost completely counterbalancing rising atmospheric greenhouse gas concentrations — followed by a massive warming spike in recent decades. Others show too much warming over the past 150 years.
And he provided this excellent comparison of climate sensitivities derived from various different methods.
Climate sensitivities derived from actual observational data (instrumental) yield climate sensitivities ranging from innocuous to mildly concerning. It’s also important to note that equilibrium climate sensitivity (ECS) is really a fake parameter. The transient climate response (TCR) is what matters, and it’s generally only about 2/3 of the ECS. So a 2 ⁰C ECS would probably equate to a 1.3 ⁰C TCR.
In the 3.5 °C ECS case, about 2.0 °C of warming occurs by the time of the doubling of atmospheric CO2. The remaining 1.5 °C of warming supposedly will occur over the subsequent 500 years… But will probably be well-within the noise level of natural variability… And, honestly, no one will ever bother to check these predictions in 2520.
As the great Yogi Berra may have said:
But he also said this: