Models get the core assumptions wrong– – the hot spot is missing

This is part of a series that Tony Cox and I are doing that references the most important points and papers, as a definitive resource about the evidence. The missing hotspot is not just another flaw in the theory, it proves the models are wrong: not just “unverified”, not just “uncertain”, but failed. Apologies to those who feel I harp on about this! This is a condensed review, squishing years of a scientific battleground down to it’s bare bones… — Jo

It is not well known that even the IPCC agrees that the direct effects of CO₂ will only increase world temperatures by 1.2°C. All of the projections above that (3.3°C , 6°C  etc) come from model projections based on assumptions of what water vapor and clouds will do (these are the feedback effects of the original 1.2°C).^[i] Are the feedbacks correct?

If the IPCC models are right about the feedbacks, we would see a hot spot 10km above the tropics. The theory is that with more heat, more water will evaporate and rise, keeping relative humidity constant at all heights in the troposphere. The point has been conclusively tested with 28 million weather balloons since 1959.^[ii]

The CCSP (Climate Change Science Project) published the predictions and observations as graphs in separate parts of its large 2006 report

The CCSP (Climate Change Science Project) published the predictions and observations as graphs in separate parts of its large 2006 report.^{[iii] ,[iv]}

The observations don’t match the predictions

Figure 7. Based on Figures 2 and 3, page 13 of McKitrick et al.

Douglass et al ^[v] compared models with observations and officially pointed out the discrepancy in 2007.  The paper was rebutted by Santer et al [2008]^[vi], but the Santer paper did not contain new observations, instead it “found more uncertainties” in model projections and in radiosonde results, which widened the error bars until they overlapped. Santer et al strangely used a truncated dataset, ending in 1999, even though more recent data was available.

McKitrick, McIntyre, Herman^[vii] – (2010) used the same datasets, but included newer data, extending the analysis, and used a more sophisticated statistical technique to show that the models all predict warming in the low to mid troposphere, and that their predictions are about 4 times higher and outside the error bars of what the weather balloons and satellites measure.  They answered the critics, with corrections in 2011, and the results became even stronger.^[viii] All observational trends were now significantly below the average model trend.

Christie et al [2010]^[ix] also showed the observations didn’t fit the model predictions. They developed the Scaling Ratio – a ratio of the atmospheric trend to the surface trend. This neatly removes the effects of El Nino variations from year to year. They showed that global climate models predict a scaling ratio of 1.4 ±0.8.  (i.e. the atmosphere should warm 40% faster than the surface). Instead the scaling ratio for real world data was 0.8 ± 0.3  (i.e. the atmosphere was probably not even warming as fast as the surface.)  Fu et al^[x] replicated the approach, largely coming to the same conclusion.

Not only are the model predictions exaggerated and feedback assumptions the wrong sign (positive, rather than negative), the small amount of recent warming was likely due to a natural climate shift in 1977 (McKitrick^[xi]). This climate shift has been noted by many other researchers (Stockwell^[xii] ) and implies man-made global warming is playing an even smaller role then predicted by the models.

The core assumptions of the IPCC favoured models are not supported by empirical measurements.

Miskolczi developed a theoretical explanation for the absence of a tropical hot-spot  in 2004, postulating that the production of entropy would already be maximized, therefore, an increase in one greenhouse gas will be matched by a decrease in another (namely water vapor) so that the efficiency of radiation leaving the Earth will not be changed^{[xiv] ,[xv]}.  He estimates that the greenhouse effect as shown by the atmospheric optical depth has not changed in 61 years (he used estimations of greenhouse gas concentrations, and radiosonde recordings of temperature and humidity).^[xiii] Other studies by Paltridge et al 2009 ^[xvi] have shown that water vapor levels have dropped as CO₂ levels have risen. Miskolczi’s 2010 work has not been challenged to date at a peer-reviewed level.