Cherrypicking A Standstill

During the recent Munk Debate in Canada, Lord Lawson seemed to astound the other contributors when he said that the world’s temperature hadn’t increased for a decade. They were incredulous and dismissed the assertion that the world hasn’t warmed for a decade as being misinformation.

A day later, one of them, George Monbiot, wrote in the Guardian about Lord Lawson during the debate, “...he said some pretty daft things such as: “There has been no further warming this century.” When I pressed him on this, he said his contention was supported by the HadCRUT3 temperature series. What it actually shows is that eight out of the 10 warmest years since records began have occurred since 2001. Lawson, of course, was deploying that tired old trick of cherry-picking his starting date. If you began the series at 1998, you might indeed conclude that temperatures have fallen, since 1998 was the hottest year ever recorded. But you begin with 1997 or 1999 or any other year in the 20th century, you discover that there has been plenty of global warming this century. That wasn’t the most sophisticated ruse, was it?”

The Grantham Research Institute on Climate Change and the Environment said, “What Lawson neglected to mention and which no journalist sought to quiz him about was a graph featuring prominently at the top of every page on the foundation’s website, using data published by the Climatic Research Unit and the Met Office’s Hadley Centre. Nor did he reveal that the foundation had “hidden” the temperature record prior to 2001, so that visitors could not tell that eight of the 10 warmest years since the instrumental records began in the 19th century have occurred since 2000.”

Contrary to those comments, it is now well established that the world’s average temperature has not changed this century. Recently, the Journal Science said the pause in global temperatures is real. In State of the Climate in 2008, a special supplement to the August Bulletin of the American Meteorological Society, Dr Jeff Knight et al of the UK Met Office Hadley Centre confirmed that in the past ten years the HadCRUT3 temperature data shows no increase whatsoever. Their analysis showed that the world warmed by 0.07 +/- 0.07 deg C from 1999 to 2008, not the 0.20 deg C expected by the IPCC. Corrected for the large 1998 El Nino event (that made 1998 the hottest year on record) and its sister La Nina, the last decade’s trend is perfectly flat. It’s a finding that has been confirmed in many of the world’s top peer-reviewed journals. There has not been “plenty of global warming this century.” Official.

We live in the warmest decade, no one doubts that, and this explains why the world’s warmest years are clustered during that period. Look at the order of the warmest years and you will see they are jumbled up and sit well within each others errors of measurement. There is no upward trend, just a plateau.

But is it cherrypicking? Why not, as The Grantham Research Institute on Climate Change and the Environment suggests, pick another start date.

The HadCRUT3 graph of the global temperature anomaly since 1975 is obviously something that surprisingly few environmental campaigners are familiar with, nor are they with good scientific procedure. The HadCRUT3 graph shown here is 35 years long and follows a period since 1940 when global temperatures were constant. The IPCC states that all temperature variations before the current warm period as shown in the graph are due to natural variations. So all of mankind’s influence on global temperatures is shown on this graph. Note that the errors of measurement are quite large encompassing all of the variability since 2001 so that, statistically speaking, the only permitted ‘curve’ since 2001 that adequately describes the data is a constant straight line.

It is normal scientific practice to look at such data and see what it shows; where is it increasing, where decreasing, and where is it flat. This is not cherry-picking, this is reading the graph as any impartial scientist would do. Sometimes cherry-picking instead of artificially highlighting a feature in the data has the effect of diluting an inconvenient feature and loosing scientific information. Having identified features in the data one should then ask how statistically significant are those changes, as the Met Office’s Hadley Centre did before reaching their conclusion.

Whilst it is true that no climate computer model predicted the past decade’s standstill, they have been used with hindsight to explain it. Natural cycles, oceanic cooling and solar influences, are to blame it is suggested. The Met Office Hadley Centre ran a series of computer climate predictions all of which had programmed into them the 0.20 deg C long-term IPCC trend. They found that in many of the computer runs there were decade-long standstills but none of 15 years. Consequently, the Hadley Centre expects the warming to resume swiftly.

Bearing in mind that the proposed explanation for the hiatus was arrived at post-hoc and relies on the same computer models that failed to predict it, one should be cautious about completely accepting this explanation. Clearly, the data has to decide and the presence or not of rapid warming in the next five years will be crucial. However, if the temperature remains constant or starts to decrease then a fundamental reappraisal will be warranted. In the meantime, it would benefit the debate if campaigners were more familiar with the basic physical parameters of global warming, and normal scientific practice.