Finally climate scientists are starting to ask how the models need to change in order to fit the data. Hans von Storch, Eduardo Zorita and authors in Germany pointedly acknowledge that even at the 2% confidence level the model predictions don’t match reality. The fact is, the model simulations predicted it would get warmer than it has from 1998-2012. Now some climate scientists admit that there is less than a 2% chance that the models are compatible with the 15-year warming pause, according to the assumptions in the models.
In a brief paper they go on to suggest three ways the models could be failing, but draw no conclusions. For the first time I can recall, the possibility that the data might be wrong is not even mentioned. It has been the excuse du jour for years.
Note in the chart that while the 10 year “pause” passed the basic 5% test of statistical significance, by 13 years, the pause was so long that only 2% of CMIP5 or CMIP3 models simulations could be said to agree with reality. By 16 years that will be 1% of simulations. If the pause continues for 20 years, there would be “zero” segments that match.
Figure 2. Consistency between the recent trend of the global mean annual temperature and simulations with climate models: the figure shows the proportion of simulated trends that are smaller or equal to the observed global annual trend in the period 1998-2012 in the HadCRUT4 data set, Rhadcrut15.= 0.0041 oC/year. The ensemble of simulated trends has been calculated from non-overlapping periods of length n in the period 2001-2060. The climate models were driven by the emission scenarios RCP4.5 (CMIP5) and A1B (CMIP3). The inset shows an expanded view of the range 0% to 2% .
Why do the models fail?
Von Storch et al suggest three reasons why the models are flawed. Essentially, there might be more natural wobbles within Earth’s climate than they expected (“stochastic” or “natural” variability), there might be another forcing the models don’t account for, or perhaps climate sensitivity to man-made changes is too high.
They don’t make any conclusions, but it appears that they feel the first is more likely. Natural internal variability means factors like ocean circulation, sea-ice, land changes or ENSO could be more important and thus, man-made factors less so. It would mean part of the recent warming of the 1980s and 1990s was more of a natural swing than a man-made one.
They can’t rule out an external forcing like volcanoes or solar insolation, but make no mention of cosmic rays or solar magnetic effects. I would be interested to know why.
Lastly, they acknowledge that reducing climate sensitivity will help but not solve much. Other factors would have to change too.
As far as I can tell, the models are not matching the turning points in the trends (because CO2 emissions don’t match the turning points in the trends), which means that it doesn’t matter how much the slope of the line is increased or decreased, it still won’t fit the data. It means they don’t understand what factors drove those changes. They have no paddle.
The paper’s summary:
Hat tip to The HockeySchtick via GWPF
REFERENCES
Hans von Storch, Armineh Barkhordarian, Klaus Hasselmann and Eduardo Zorita (2013) Can climate models explain the recent stagnation in global warming? Academia