Two German scientists describe what many western governments have been basing their energy and environmental policies on. It’s not pretty. What follows is an excellent review of climate modeling so far.
Fun with Climate Models: Flops, Failures and Fumbles
By Dr. Sebastian Lüning and Prof. Fritz Vahrenholt
(Translated, edited by P Gosselin)
What’s great about science is that one can think up really neat models and see creativity come alive. And because there are many scientists, and not only just one, there are lots of alternative models. And things only get bad when the day of reckoning arrives, i.e. when the work gets graded. This is when the prognoses are compared to the real, observed measurements. So who was on the right path, and who needs go back to the drawing board?
When models turn out to be completely off, then they are said to have been falsified and thus are considered to have no value. The validation of models is one of the fundamental principles of science, Richard Feynman once said in a legendary lecture:
Failed hypotheses have been seen very often in science. A nice collection of the largest scientific flops is presented at WUWT. Unfortunately the climate sciences also belong to this category. Roy Spencer once compared an entire assortment of 73 climate models to the real observed temperature development, and they all ended up overshooting the target by far:
And already yet another model failure has appeared: In August 2009 Judith Lean and David Rind made a daring mid-term climate prognosis in the Geophysical Research Letters. They predicted a warming of 0.15° for the five-year period of 2009 to 2014. In truth it did not warm at all during the period. A bitter setback.
Over the last years it has started to dawn on scientists that perhaps something was missing in their models. The false prognoses stand out like a sore thumb. Not a single one of the once highly praised models saw the current 16-year stop in warming as possible. In September 2011 in an article in the Journal of Geophysical Research Crook & Forster admitted that the superficial reproduction of the real temperature development in a climate model hardly meant the mechanisms were completely understood. The freely adjustable parameters are just too multifaceted, and as a rule they are selected in a way to fabricate agreement. And just because there is an agreement, it does not mean predictive power can be automatically derived. What follows is an excerpt from the abstract by Crook & Foster (2011):
In this paper, we breakdown the temperature response of coupled ocean‐atmosphere climate models into components due to radiative forcing, climate feedback, and heat storage and transport to understand how well climate models reproduce the observed 20th century temperature record. Despite large differences between models’ feedback strength, they generally reproduce the temperature response well but for different reasons in each model.”
In a member journal of the American Geophysical Union (AGU), Eos, Colin Schultz took a look at the article and did not mince any words:
Climate model’s historical accuracy no guarantee of future success
To validate and rank the abilities of complex general circulation models (GCMs), emphasis has been placed on ensuring that they accurately reproduce the global climate of the past century. But because multiple paths can be taken to produce a given result, a model may get the right result but for the wrong reasons.”
Sobriety in the meantime has also spread over to IPCC-friendly blogs. On April 15, 2013, in a guest post at Real Climate Geert Jan van Oldenborgh, Francisco Doblas-Reyes, Sybren Drijfhout and Ed Hawkins made it clear that the models used in the 5th IPCC report were completely inadequate for regional climate prognoses:
To conclude, climate models can and have been verified against observations in a property that is most important for many users: the regional trends. This verification shows that many large-scale features of climate change are being simulated correctly, but smaller-scale observed trends are in the tails of the ensemble more often than predicted by chance fluctuations. The CMIP5 multi-model ensemble can therefore not be used as a probability forecast for future climate. We have to present the useful climate information in climate model ensembles in other ways until these problems have been resolved.”
– GCMs suffer from temperature-dependent biases
– This leads to an overestimation of projections of regional temperatures
– We estimate that 10-20% of projected warming is due to model deficiencies”
In January 2013 in the Journal of Climate Matthew Newman reported in an article “An Empirical Benchmark for Decadal Forecasts of Global Surface Temperature Anomalies” on the notable limitations of the models:
These results suggest that current coupled model decadal forecasts may not yet have much skill beyond that captured by multivariate red noise.”
In the prognosis time-frame of multiple decades, they do not perform better than noise. An embarrassment.
Also Frankignoul et al. 2013 expressed serious concerns in the Journal of Climate because of the unimpressive performance of the climate models. They graded the models plainly as “unrealistic” because they did not implement the role of ocean cycles correctly.
In July 2013 Ault et al. looked at a paper in the Geophysical Research Letters and at the models for the tropical Pacific region. They made an awful discovery: Not one of the current models is able to reproduce the climate history of the region during the past 850 years. Excerpts from the abstract:
[…] time series of the model and the reconstruction do not agree with each other. […] These findings imply that the response of the tropical Pacific to future forcings may be even more uncertain than portrayed by state-of-the-art models because there are potentially important sources of century-scale variability that these models do not simulate.”
Also Lienert et al. (2011) found problems with the North Pacific. And in July 2014 in an article in Environmetrics, McKitrick & Vogelsang documented a significant overestimation of the warming in the climate models for the tropical region over the past 60 years.
In March 2014 Steinhaeuser & Tsonis reported in Climate Dynamics on a comparison of 23 different climate models and the extent to which they were able to reproduce temperature, air pressure and precipitation over the 19th and 20th centuries. The surprise was great when the scientists found that the model results deviated widely from each other and were unable to give a correct account of reality. A more detailed discussion is available at The Hockey Schtick.
In a press release from September 17, 2012, scientists of the University of Arizona complained that as a rule climate models failed when looking at periods of three decades and less. Also attempts at prognoses for regional levels were unsuccessful:
UA Climate Scientists put predictions to the test
A new study has found that climate-prediction models are good at predicting long-term climate patterns on a global scale but lose their edge when applied to time frames shorter than three decades and on sub-continental scales.”
In October 2012 Klaus-Eckart Puls at EIKE warned that up to now the temperature prognoses of the climate models have been false for every atmospheric layer:
For some decades now climate models have been projecting trends (“scenarios”) for temperature for different layers of the atmosphere: near surface layer, troposphere, and stratosphere. From the near surface layer all the way to the upper troposphere it was supposed to get warmer according to the AGW hypothesis, and colder in the stratosphere. However meteorological measurements taken from all atmospheric layers show the exact opposite!”
So what is wrong with the models?
For one they still have not found a way to implement the empricially confirmed systematic impact of the ocean cycles into the models. Another problem of course is that the sun is missing in the models as its important impact on climate development continues to be denied. It’s still going to take some time before the sun finally gets a role in the models. But there are growing calls for the taking the sun into account and recognition that something is awry. In August 2014 in the Journal of Atmospheric Sciences a paper by Timothy Cronin appeared. It criticized the treatment of solar irradiance in the models. See more on this at The Hockey Schtick.
The poor prognosis-capability of climate models is giving more and more political leaders cause for concern. Maybe they should not have relied on the model results and developed far-reaching plans to change society. To some extent they have already began to implement these plans. Suddenly the very credibility of the climate protection measures finds itself at stake.
The best would be a moratorium on models. Something needs to be done. It is becoming increasingly clear that the present wild modeling simply cannot continue. It’s time to re-evaluate. The climate models so far are hardly distinguishable from computer games on climate change where one sits comfortably on the couch and shoots as many CO2 molecules out of the atmosphere as he can and then reaps the reward of a free private jet flight with climate activist Leonardo di Caprio.
Fritz Vahrenholt and Sebasian Lüning authored the climate science book The Forgotten Sun. In this book they examined the poor quality of climate models and why they will always fail.