The German media are reporting on findings of a study appearing in Nature coming from the Jochen Marotzke of the Max Planck Institute for Meteorology (MPI-M) in Hamburg and Piers Forster of University of Leeds, who both insist the climate models continue to be correct – despite having been wrong for 18 years.
Warming not showing up
The DPA, via Die Welt here, has an article titled: “Why the crass global warming is not showing up” and writes:
The earth’s temperature over the past 15 years has not warmed as much as the climate models projected. In the opinion of scientists this not due to poor climate modeling, but rather because of ‘random climate fluctuations’. “
Modelers blame “spontaneous climate variability”
The MPI-M press release claims there is “no evidence for systematic model error” and blames “spontaneous climate variability” arising from “chaotic processes in the climate system” for the growing divergence between observations and simulations since the conception of the climate models.
The MPI-M press release obviously concedes the climate models do not understand the short and mid-term natural factors at play in climate, saying that the 1998 to 2012 period “witnessed some extreme regional climate trends, such as a strong cooling in central Siberia during winter and an unparalleled strengthening of the trade winds in the Pacific causing cold ocean water to reach the surface. Such random extreme events are very rare and are hard to reproduce with models.”
Modelers have struck out on short and mid-term reliability
Clearly the modelers are rapidly abandoning any earlier claims made of short-term and mid-term model reliability, where they have plainly struck out, and are now left asking us to believe in their long-term reliability.
For years skeptics have warned alarmist scientists that they were making grave mistakes in ignoring the climate’s natural cyclic behavior and that the generally chaotic climate system was not predictable. Predictably, they scoffed and insisted that their models were “tweaked”, “fine-tuned” and as good as gold. But it turns out they have not been okay since the very start, and now the climate scientists are insisting the models are okay so long as you ignore all the short and mid-term factors that make them not okay.
The DPA German news agency writes:
In a first step the scientists compared the 114 simulations with the actual observations. The question was whether the simulations are too sensitive and certain factors too heavily weighted and as a result forecast too much warming. If that were the case, then the most sensitive models would have to forecast the most warming, explains Marotzke. That is not at all the case. Foremost they did not in principle react too sensitively to an increase in atmospheric CO2. Moreso it was chaos and random events in weather that make the work of climate scientists difficult.”
And that difficulty has been a huge thorn in the models’ side for about 17 years now. It’s panic time.
Variablilty? Or an inconvenient climate cycle?
The MPI-M press release insists that 15 years are too short because here “the differences between simulated and observed trends are dominated by spontaneous climate variability and that the sensitivity of climate in the models plays no role.”
The real question that needs to be asked is: Have the models been wrong for 15 years really because of weird random weather events, which is what the MPI-M seems to be hoping, or are these “random events” part of a powerful climate-driving, natural cycle that the warmist scientists are refusing to acknowledge?
How much longer before they throw in the towel?
Unfortunately the MPI-M failed to tell us how much longer the current divergence can continue on before their latest explanation falls apart. Kevin Trenberth told us 17 years was already enough to make the call. The MPI-M charts suggest it’s around 20 years. The MPI-M paper may buy the warmists a little bit more time, but convincing it is not.
Yet, MPM-M director Marotzke stubbornly continues his desperate cling to the models, and is quoted by the DPA:
Over the long term we can rely on the models,’ says Marotzke. ‘And they tell us that the warming is coming at us.'”
We’ll be watching, and I suspect waiting…and waiting…and waiting.
18 responses to “Last Ditch Clinging Effort…Scientists Plainly Have Struck Out On Short and Mid-Term Climate Model Reliability”
Actually, the models and their predictions have only been anywhere near reality at their lowest levels since James Hansen in 1988 and the IPCC FAR in 1990.
a) 1988 James Hansen based his predictions on three scenarios :-
Scenario A predicted an increase of 0.34-0.44 C/decade
Sceanrio B predicted an increase of 0.18-0.30 C/decade
Scenario C predicted an increase of 0.19-0.29 C/decade
(figures from http://www.realclimate.org/index.php/archives/2007/05/hansens-1988-projections/ )
Actual increase from 1988 to 2014 (mean of HadCRUT4, NCDC and GISS) = 0.13 C/decade
b) 1990 IPCC First Assessment Report Working Group 1 predicted 0.2-0.5 C/decade
Actual increase from 1990 to 2014 (mean of HadCRUT4, NCDC and GISS) = 0.12 C/decade
c) 1995 IPCC Second Assessment Report Working Group 1 predicted 0.1-0.35 C/decade
Actual increase from 1995 to 2014 (mean of HadCRUT4, NCDC and GISS) = 0.14 C/decade
d) 2001 IPCC Third Assessment Report Working Group 1 predicted 0.13-0.43 C/decade
Actual increase from 2001 to 2014 (mean of HadCRUT4, NCDC and GISS) = 0.11 C/decade
e) 2007 IPCC Assessment Report 4 Working Group 1 predicted 0.11-0.64 C/decade
Actual increase from 2007 to 2014 (mean of HadCRUT4, NCDC and NOAA) = 0.12 C/decade
…And scenario A assumed a CO2 emission reduction, not a growth faster than his business as usual Scenario B.
“Black Knight” Marotzke: “It’s nary a scratch”. (paraphrased)
Funnily enough the scientists do not consider the petty and short-lived rise in temperarure is due to ‘random climate fluctuations’
If it warms, it’s just confirming their prediction. If it cools, that’s random variation. Begging the question refined to an art form.
No talk any more of error rates. If their models were true, the probability is less than 5 percent that the observations are due to random variation (Type II error rate). If the null-model were true, the probability is less than 5 percent that the former observations were due to random variation (Type I error rate). So they decided that warming was ‘significant’. Principle of fair play: the non-warming is ‘significant’.
“who both insist the climate models continue to be correct – despite having been wrong for 18 years.”
I’m curious — why do you think climate models can make predictions over a decade or two?
The modelers I talk to say it can’t be done, because models aren’t spun up into the real climate state, but an arbitrary state. That state, over many decades, relaxes to the equilibrium state…. But ENSOs, PDO, AMOs, etc mean no climate model can ever predict the nearest decade or three, because they can’t predict ENSOs….
David, why can’t they predict the “ENSOs, PDO, AMOs etc.”?
No reply from David, as expected. No wonder since we do have more and more research that tells us that the ocean follows the beat of the sun:
hard times for (C)AGW-believers…
“We’re wrong over the short term, but the farther in the future you go the more exact it gets” (paraphrased), yeah David, hey, how about you learn about numerics and error propagation and the mathematical definition of chaos for a start…
I would also like to know why ENSOs, AMOs, PDOs, etc. (and solar activity for that matter) cannot be modelled over the period from 1880, and so be used to make predictions at the decadal level.
ENSOs occur on year-long scales and they are practically impossible to predict. However AMOs and PDOs have a pretty good 60-year cycle and absolutely should be built into the climate models. But there’s a problem here: Doing so would necessitate attributing much of the 1980 – 1998 global warming to these cycles, and not CO2. The global warming alarmist scientists want to avoid this at all costs.
Thanks for that. So it should be possible for one or more of the sceptical scientists to build a model that incorporates the AMO and PDO – or has that already been done ?
Prof. Fritz Vahrenholt and Sebastian Lüning recently wrote a book on the topic and introduced their own climate model that incorporated the AMO, PDO and solar cycles.
As a result they are forecasting modest cooling up to about 2040 and then warming by about 1°C by 2100. So far of course this is precisely what is shaping up.
The PDO at http://jisao.washington.edu/pdo/PDO.latest went up a lot last year. At first glance, the AMO at http://www.esrl.noaa.gov/psd/data/timeseries/AMO/ looks to be better correlated with temperature.
Actually the correlation of AMO with HadCRUT4 (which is very close to HadCRUT3) is less than 70% so it’s only a general indication that they tend to rise and fall at about the same times.
Probably the best “first order” guess for the future is a repetition of the past pattern (all other things being equal).
HADCRUT3 is a widely-respected summary of global surface temperatures (fewer adjustments than GISTEMP or HADCRUT4).
Satellite records only start in 1979.
If you squint a bit, you might see the following (roughly):
a temperature drop of 0.3 deg C (1880-1910)
a temperature rise of 0.6 deg C (1910-1940)
a temperature drop of 0.3 deg C (1940-1970)
a temperature rise of 0.6 deg C (1970-2000)
Now, you may not be a betting man, but wouldn’t:
a temperature drop of 0.3 deg C (2000-2030) be a reasonable guess?
Long-term trend rate since 1850 < 1 deg. C / century.
Remember, atmospheric CO2 concentration has apparently been rising steadily from 280 ppm since the late 18th Century; at least we know the smoothed rise rate in 1960 was about 1ppm / y and today it's about 2 ppm / y (from the Keeling Curve).
Leading Climate Scientists / IPCC have been saying for three decades that the global mean surface temperature will accelerate to dangerous and irreversible levels, due primarily to the rising concentration of human greenhouse gas emissions / concentration. But the temperature rise rate from the 1970-2000 was neither steeper nor longer in duration than that from 1910-1940 (despite far higher CO2 emissions and considerably higher CO2 concentration).
And I did that without a grant.
Yes, I saw from the NCDC / GISS / HadCRUT4 graph :-
Cooling 1880-1911 -0.29 C
Warming 1911-1944 +0.65 C
Cooling 1944-1976 -0.32 C
Warming 1976-2011 +0.65 C
It has gone up +0.14 C since then, so we shall see.
Perhaps not. They are somewhat correlated with solar activity. Strong solar activity seems to be correlated with stronger and more frequent El Niño … ocean heat transfer introduces some lag but the Pacific absorbs a great deal of solar radiation; absent clouds. When cloud cover changes, heat input changes substantially. Type of cloud cover and area seems to be correlated with solar activity (inversely).