By Christopher Monckton of Brenchley
Pseudo-science by press release has become the norm among the Forces of Darkness. With as much fanfare as McGill University could muster, the recent paper by Professor Lovejoy was promoted via a typically head-banging instance of the genre.
The gushingly flatulent halation of the university PR-wallahs is typical of the verbal diarrhea habitual among practitioners of the Dark Arts. The ipsissima verba of Lovejoy himself in the press release are of particular interest.
Here is what he is quoted as saying:
“We’ve had a fluctuation in average temperature that’s just huge since 1880 – on the order of about 0.9 degrees Celsius. This study shows that the odds of that being caused by natural fluctuations are less than one in a hundred and are likely to be less than one in a thousand.
“While the statistical rejection of a hypothesis can’t generally be used to conclude the truth of any specific alternative, in many cases – including this one – the rejection of one greatly enhances the credibility of the other.”
Is the post-1880 variability in global temperatures “huge”? The most direct method to test this proposition is to examine the available temperature record since 1500, the starting date of Lovejoy’s analysis.
However, there is no direct global thermometer record going back that far. Accordingly, Lovejoy uses a ragbag of politically-correct reconstructions over various periods since then, inevitably including the long-discredited “hockey-stick” graph. Not only that, but an unpublished version of Richard Muller’s Berkeley Earth Surface Temperature record is dragged in as well.
There is one curious and significant omission on Lovejoy’s list of references. There is no mention of the Central England Temperature Record. Now, CET is the longest continuous regional temperature record we have. Despite its name, it covers most of England roughly from Lancashire in the north to Devon in the south. And it has been maintained since 1659.
No one doing a genuinely study of temperature change since 1500 would make ignore CET altogether. It has been maintained in a nation on which Fortune has smiled, so that there have been no interruptions owing to riot or revolution. It is on the right latitude to be – at least potentially – a respectable proxy for global temperature change over the period of study.
The most straightforward way to determine whether any proxy measurement – here, the part for the whole – provides a reasonable indication of how the Earth may have warmed and cooled over the period of record is to make a comparison between the proxy and the global record during the period of overlap.
First, let us apply this test to the “hockey stick” graph that Professor Lovejoy used. To establish that such a test is valuable, we go to no less authoritative a source than the World Meteorological Organization, one of the profiteering cheer-leaders of the panic pandemic.
Here, taken directly from the front cover of the WMO’s “Status of the Global Climate” for 1999, is a graph showing that three distinct series of tree-ring data faithfully replicate the sharp increase in global temperature over the 20th century.
Here is the “smoking gun” that proves how exceptional the growth in 20th-century was. There had been nothing like it in a millennium.
Just one problem with that. The original tree-ring data for the period 1960-1999 did not show temperature soaring. But the global temperature record for 1960-1999 did. So the tree-ring graphs, all three of them, were tampered with to truncate all three records in 1960 and to bolt on the real-world temperatures.
For the inconvenient truth was that the Jones and Mann data showed no increase in temperature after 1960 and the Briffa data showed a precipitate decline.
To “hide the decline”, the graph published on the WMO’s front cover in 1999 had had the real tree-ring data airbrushed out. All three tree-ring series were then falsely made to track the sharp increase in global temperature from 1960-1999.
Look closely at the WMO graph and it is clear that all three colors, one for each record, are shown as reaching all the way to the top right-hand corner of the graph.
Here, screen-capped from YouTube, is a slide from a presentation by Richard Muller showing what the data looked like before they were tampered with. The true graph show the tree-ring data as diverging markedly from global temperatures since 1960, indicating that dendrochronology is unable reliably to detect sharp increases in global temperature and should not have been used in Professor Lovejoy’s analysis.
There is no statement in the WMO report that the graph prominently displayed on its front cover had thus been altered. The WMO’s publication of the graph was, therefore, a criminal offense. It was fraud by misrepresentation and by breach of a position of trust, calculated profitably to deceive readers into thinking that the tree-ring data were reliable proxies for global temperature change and, therefore, that the inconvenient Medieval Warm Period had not been warmer than the present.
However, the WMO is one of the plethora of supranational organizations that are not subject to any jurisdiction. No one will go to jail for fraud by misrepresentation and breach of a position of trust, even though climatology now looks less like science than organized crime.
A similar overlap comparison test will now be applied to the Central England Temperature Record. We shall take two full 60-year cycles of the Pacific Decadal Oscillation, so as to cancel any distorting influence of the 30-year positive or warming phases and the 30-year negative or cooling phases (we are in the middle of a negative phase at the moment).
The analysis will cover the 120 calendar years January 1894 to December 2013. Calendar years are used so as to avoid any distortion of the trend on the Central England record caused by the fact that it is regional rather than global and is accordingly subject to seasonal variation.
The mean of the monthly anomalies for the combined GISS, HadCRUT4, and NCDC global-temperature datasets over the period shows warming 0f 0.89 Cº. The CET monthly anomalies over the same period show warming of 0.90 Cº. Therefore we shall not need to tamper with the data to fit a preconceived result: the Central England series, over a sufficiently long period, is a not unreasonable proxy for global temperature change.
Having calibrated the CET dataset against the global instrumental datasets, we look back through the record to find the greatest centennial rate of warming before the Industrial Revolution. Here is the graph from January 1663-December 1762:
The rate of warming was 0.9 Cº, and that rate occurred over 100 years rather than the 124 years 1880-2013 covered by Lovejoy’s statement that 0.9 Cº was a “huge” temperature increase. And it was entirely natural warming. As Professor Lovejoy might put it, it is 99.9% certain that we were not to blame.
In passing, it is worth noting how small a “huge” 0.9 Cº temperature trend is when compared with the seasonal variability of regional surface temperatures. This variability is concealed by the usual suspects’ habit of showing global temperature changes only, so that any trend – however small – looks more drastic than it is.
There is a further simple test that can be done, this time using the mean of the three global terrestrial datasets. In this test, we shall compare the 60-year period before we could have had any measurable influence on global temperature with the most recent 60-year period after it.
From January 1894-December 1953, the warming trend was 0.44 Cº, very nearly all of which must have been natural. From January 1954-December 2013, it was 0.77 Cº, some of which may have been anthropogenic. But the fact that there was almost half a Celsius degree of global warming in the early 20th century, before we could have had anything much to do with it, does cast further doubt on Professor Lovejoy’s conclusion.
A final test is to take the mean of all five global-temperature datasets and see how far back one can go before encountering even the most insignificant warming trend. The dataset of datasets shows a zero global warming trend for 13 years 2 months, or well over a sixth of a century.
Now, long periods without global warming do happen. On the HadCRUT4 data, there was a zero trend from 1850-1900, a remarkable period of 51 years.
However, the failure of global temperature to change at all in this millennium since January 2001 is also remarkable, for two reasons. First, this is the longest sustained period without global warming since we might first have influenced temperatures in 1950.
Secondly, CO2 concentration is rising at record rates, yet temperature is not responding at all, suggesting, at the very least, that CO2’s influence on temperature is so small that it is easily canceled by natural factors that might otherwise have produced cooling.
Thirdly, the wretched computer models did not predict the Pause. Oh boy, did they not predict the Pause. Here are their very latest predictions, each model run shown separately in a spaghetti graph from the IPCC’s 2013 Assessment Report. I have updated the real-world trend to bring the record up to the present.
Now, the computers are programmed to assume, just as Professor Lovejoy assumed, that it is our influence on global temperatures that has caused the global warming of recent decades. That, and that alone, is why the spaghetti graph of model-predicted near-term global warming soars at a rate equivalent to 2.33 [1.33, 3.33] Cº/century.
And the real-world trend since 1950? Well, it’s equivalent to less than 1.2 Cº/century, below the lower bound of the IPCC’s current predictions. Another five years without global warming and the Sixth Assessment Report (if there is one) will look even more like a costly and sick practical joke than its predecessors.
Based on these simple tests using real-world data rather than mere models, the notion that the odds of 0.9 Cº global warming having been caused by natural variability are “less than 1 in 100 and are likely to be less than 1 in 1000” is, to say the least, questionable.
It was not we who caused the 0.9 Cº warming in Central England and, inferentially, worldwide from 1663-1762. It may not have been we who caused the 0.9 Cº global warming since 1880 either.
For this reason among many others, Professor Lovejoy errs in contending that he has “rejected” the hypothesis that much of the 0.9 Cº global warming since 1880 is natural. At the very least, we know that the half of it that occurred before 1950 had very little to do with us.
In the McGill press release, Professor Lovejoy concludes:
“This study will be a blow to any remaining climate-change deniers. Their two most convincing arguments – that the warming is natural in origin, and that the computer models are wrong – are either directly contradicted by this analysis, or simply do not apply to it.”
Nice try, but No. As will be evident from the temperature graphs above, the climate does change. We do not deny it. Some element in the global warming since 1950 may well have been caused by us. We do not deny that either. Long-established theory and experiment have confirmed that there is such a thing as a greenhouse effect and that by enriching the atmosphere with CO2 we may – to some extent yet to be determined – be enhancing it.
What we do deny is that aprioristic modeling elaborately contrived at taxpayers’ great expense to achieve a preconceived result manifestly at odds with observation and measurement but congenial and profitable to the classe politique has anything whatsoever to do with true science.
But then, Lovejoy was not intending to do science. Note the last two bullet points from a November 2013 presentation by him (hat-tip to Willie Soon):