The Guardian’s Dana Nuccitelli uses pseudo-science to libel Dr. John Christy
By Paul Homewood
You may recall that Dana Nutticcelli ran a piece in the subsidised, left wing Guardian the other day, which attempted to debunk John Christy’s testimony to Congress. The deluded drones who read the Grauniad naturally sucked it all up.
I did not respond at the time because I knew that Lord Monckton would make a much better job of it than I ever would.
Here is his response in WUWT:
One Dana Nuccitelli, a co-author of the 2013 paper that found 0.5% consensus to the effect that recent global warming was mostly manmade and reported it as 97.1%, leading Queensland police to inform a Brisbane citizen who had complained to them that a “deception” had been perpetrated, has published an article in the British newspaper The Guardian making numerous inaccurate assertions calculated to libel Dr John Christy of the University of Alabama in connection with his now-famous chart showing the ever-growing discrepancy between models’ wild predictions and the slow, harmless, unexciting rise in global temperature since 1979.
The chart, described by Nuccitelli as “simply another example of cherry picked data … presented in a multiply misleading way”, shows his comments. Each comment is then given in more detail in bold face, followed by the truth in Roman face.
1. “The data are misleadingly misaligned” to start in 1979, so as “to visually exaggerate any difference between the models and data”. Instead, Mr Nuccitelli opines that they should have been aligned to a common baseline some decades in length.
Altering the baselines does not alter the trends. Nevertheless, to test Mr Nucccitelli’s allegation that Dr Christy had “misleadingly misaligned” the data, trends on the models’ predictions (red), satellites’ observations (green) and radiosondes’ measurements (blue) were expressed as centennial-equivalent warming rates of 2.22, 1.00 and 0.86 Celsius degrees respectively. The warming rate predicted by the models is thus some 2.2–2.5 times the warming rates observed by the satellites and radiosondes. The graph, therefore, correctly reflects a real and widening discrepancy between prediction and observation. Note also that the CMIP5 predictions were made in about 2010, so that nearly all the red curve represents hindcasts: yet still the models’ trend is excessive.
2. “No uncertainty ranges are shown whatsoever”. When they are taken into account, “the observations are consistent with the range of model projections”.
Data since 1979 for the CMIP5 models were not to hand. However, in 1990 IPCC (AR1, p. xxiv), on the basis of “substantial confidence” that the models on which it relied had captured all essential features of the climate, predicted near-linear warming of 1.0 [0.7, 1.5] Celsius degrees over the 36 years 1990-2025, equivalent to 2.78 [1.94, 4.17] Cº/century. The boundary between the two zones, marked with the red needle in the clock-graph below, is the IPCC’s then best prediction: warming equivalent to about 2.8 C°/century by now.
The very wide range of predictions made by the IPCC is shown as orange and red regions. The observed warming on the RSS and UAH satellite datasets, again expressed as centennial equivalents, is shown by the two green needles. The HadCRUT4 dataset, to Dr Jones’ credit, publishes its combined measurement, coverage and bias uncertainties, which are about 0.16 Celsius degrees either side of the central estimate. The satellite uncertainties are smaller. It is plain that there is no overlap whatsoever between the exaggerated predictions made by IPCC in 1990 and the rates of global warming since then shown by the satellites.
3. “Observational data disagreements are hidden,” because “Christy’s graph also averages together multiple different observational datasets, which aren’t in terribly close agreement.”
In the present context, disagreements between trends on the RSS and UAH satellite datasets, for instance, would only be material if either of the datasets showed a trend close to the trend on the models’ predictions: otherwise, such differences would be inconsequential when set against the far wider difference between the trend on each observational dataset and the trend on the models’ predictions.
To test whether the two satellite datasets “aren’t in terribly close agreement”, their spline-curves and trends from 1979-2015 were separately determined and plotted. Results showed that the two curves are visibly in reasonable agreement.
To verify this, copy each graph on to a PowerPoint slide, start the presentation and then use the up and down arrows in rapid succession to make a blink-comparator.
Their centennial-equivalent trends are within a tenth of a degree of one another, whereas the differences between each of the two observed trends and the model-predicted trend are each an order of magnitude greater than the difference between them.
4. “The chart isn’t peer-reviewed or easily reproducible”, in that “Christy doesn’t say which observational data sets he’s averaging together”.
Mr Nuccitelli did not email Dr Christy and simply ask for the information. On one occasion when I asked Dr Christy for some data to assist me in a paper I was writing, I received the requested data within 24 hours. My questions about the data were answered promptly, courteously, fully and helpfully. Furthermore, the chart is plainly labeled indicating that it was prepared using the online and publicly available Climate Explorer program and data maintained by the Royal Netherlands Meteorological Institute.
Had Mr Nuccitelli done a little homework, he would have been able to find the following widely-circulated graph that actually lists 73 of the models used by Dr Christy, and shows IPCC’s ever-increasing confidence in the “consensus” proposition that recent global warming was mostly manmade. In fact, as Mr Nuccitelli knows full well (for his own data file of 11,944 climate science papers shows it), the “consensus” is only 0.5%. But that is by the bye: the main point here is that it is the trends on the predictions compared with those on the observational data that matter, and, on all 73 models, the trends are higher than those on the real-world data.
Read the rest here.