Super Rapid Debunking Of Super Rapid Spanish Heatwave Analysis
World Weather Attribution are at it again! They’ve just published a super rapid analysis of the brief but extreme heatwave in Spain and Portugal which occurred near the end of April. So I’ve decided to do a super rapid exposé of what pseudoscientific nonsense it is.
Before I do, this is what the Guardian (of course!) says about it:
The record-shattering temperatures that hit the western Mediterranean last week would have been “almost impossible” without the climate crisis, according to scientists.
The heatwave across Spain, Portugal, Morocco and Algeria was made at least 100 times more likely by global heating, the researchers calculated. Before the climate crisis, such an extreme event would have been expected only once in a least 40,000 years, making it statistically impossible on human timescales.
‘Almost impossible without the climate crisis’. The study finds that it would also have been pretty rare even with the climate crisis - they calculated a return time of once in 400 years for the observed event and even this was a very uncertain estimate:
Given the extremity of the event compared to the length of the time series, the return period has a large uncertainty, with a best estimate of about 400 years, and a lower estimate of 60 years (and an infinite upper bound). A final value for the return period may change for reanalysis data rather than preliminary data, but given the uncertainty around it we do not consider this to be a problem for the current rapid analysis.
The upper bound was infinite, so the actual event might realistically have a return period greatly in excess of 400 years using better data, meaning that the event was statistically impossible even in the current warmer climate, meaning that something else must have caused it, other than ‘global heating’. But that’s not a problem as far as the authors are concerned.
The model runs are even worse. They all fail miserably to simulate the intensity of the observed event. Just look:
Apologies for the awful screenshot. It’s very difficult to get good screenshots from pdf files. Look at the chart on the left. The blue box is the observed event (with uncertainty defined by the width of the box). The light red boxes are the individual model simulations, the vivid red box is the model average and the purple box is the synthesis of the model runs and statistical analysis. The black vertical line is the central estimate. As you can see, none of the models which incorporate greenhouse gas forcing in their calculations got anywhere near simulating the observed event. So all of them comprehensively failed.
That’s the end of my super rapid response to WWA’s super rapid bullshit.
“ Almost impossible without climate crisis” really? Made 100 times worse a once in a 40,000 year event really.” Really you can trust these numbers because the scientist threw spaghetti on the wall and pulled numbers scientifically with pin point accuracy. The question is who funds this bull shit and I guess they just want to bamboozle us with their incredible accuracy.
I can’t see the graphs on my phone but the words are enough to tell me … when I was a kid we would exaggerate stories. It may not be lying but it’s bending the truth.
It is interesting that you should post on this study immediately following your report of the Koutsoyiannis paper on the Greek hydro meteorological records. Koutsoyiannis has published a lot of work on deterministic versus stochastic modelling of the climate. In 2010 he published a paper - A comparison of local and aggregated climate model outputs with observed data in the Hydrological Science Journal. His conclusions on the ability of deterministic models such as GCM’s to capture the variability of climate at local and regional scales over time periods ranging from times between monthly to climatic scales (30 years) is very poor. They lack any information on the stochastic nature of weather and climate events. He concludes “Do we have something better than GCMs when it comes to establishing policies for the future? Our answer is yes: we have stochastic approaches, and what is needed is a paradigm shift. We need to recognize the fact that the uncertainty is intrinsic, and shift our attention from reducing the uncertainty towards quantifying the uncertainty....” Of course, in the case of the Spanish heatwave (a 3 day event) we have no empirical data that can relate the magnitude of the event to the naturally occurring frequency of such events and thus lack information on the uncertainty that can be attributed to stochastic processes. The WWA study is just yet another example of models masquerading as reality.