In November 2021, one of the main programmers of the NASA climate model Gavin Schmidt told readers of the Spectator that the track record of models going back to the 1970s, “shows they have skilfully predicted the trends of the past decades”. Now that laughter has finally subsided, we have an expert analysis of NASA’s GISS Model E with its 441,668 lines of pre-historic (circa 1983) FORTRAN code. With water that doesn’t freeze and “negative” cloud cover, it is said that the claim the model is ‘physics-based’ is a term used in the same way that Hollywood producers say a movie is ‘based on a true story’.
The detailed examination has been written by the experienced computer programmer Willis Eschenbach and his paper Climate Models and Climate Muddles has been published by Net Zero Watch (NZW). Andrew Montford of NZW discussed the paper in a recent edition of the Daily Sceptic, noting that climate models are at the centre of the global warming scare and back all the weather alarms promoting the collectivist Net Zero project. But what if the climate models were all junk, he asked. Somewhat alarmingly, Eschenbach’s work shows “this is indeed the case”.
Eschenbach argues that the current crop of computer climate models are far from being fit to be used to decide public policy. To verify this, he says, you only need to look at the endless string of bad, failed, crashed-and-burned predictions they have produced. Pay them no attention, he cautions. “Their main use is to add false legitimacy to the unrealistic fears of the programmers.” If you write a model under the working assumption that carbon dioxide controls the temperature, then guess what you’ll get.
According to Eschenbach, climate models have a hard time replicating the amazing stability of the climate system. They are ‘iterative’ models, meaning the output of one timestep is used as the input for the next. As a result any errors are carried over, making it easy for models to spiral the Earth into fire and snow balls. NASA gets around polar water refusing to freeze and ‘negative’ amounts of cloud forming (what do minus-two clouds look like?) during model runs by replacing bad values with corresponding maximum or minimum values. “Science at its finest,” comments Eschenbach. He notes that he is not picking on just NASA. The same issues, to a greater or lesser extent, exist within all complex iterative models. “I’m simply pointing out that these are not ‘physics-based’ – they are propped up and fenced in to keep them from crashing,” he observes.
This is the graph produced by Professor Nicola Scafetta plotting 38 of the major climate models showing their temperature predictions set against the thick green line of the actual satellite record.
Read More: Climate Models Come With ‘Dangerous’ CO2 Warming Baked In, Code Review Finds
