Friday, March 20, 2009

Another Look at Global Warming

Thanks to Watts Up With That, slides and commentary from Steve McIntyre of climateaudit.org. McIntyre has become famous for questioning the "hockey stick" graph that is supposed to present the twentieth century as hotter than earlier centuries, and in general he questions the quality of data, and the reliability of the conclusions that are drawn based on the data, in the climate change world.

Al Gore represents what might be called the official view of climate change: the IPCC, James Hansen, and a few studies that are constantly cited. What few people realize is the extent to which all claims to the effect that the 20th century was unusually warm are based, ultimately, on a remarkably small number of data sets. There is very little direct evidence of temperature before 1900, so everyone is forced to rely on proxies such as tree growth. To an amazing extent, the Gore studies depend on a very limited study of one kind of tree in California--bristlecone pines, which apparently showed some dramatic growth in the 20th century in comparison to past periods.

The famous "hockey stick," showing a dramatic increase in temperature in the 20th century after temperature remained more or less constant for a millenium, depended on these pines, and a few other data sets that have been cited many times. When the hockey stick was criticized for being based on limited and inconclusive data, the response was a "spaghetti graph," showing six or more trend lines very similar to the hockey stick. When criticism continued, the claim was made that "twelve independent studies" supported the main conclusion of the hockey stick argument. Nine of the twelve continued to depend on bristlecone pines, and the data continued to show that without the bristlecone pines, the data were inconclusive. The other three of the twelve didn't have even one set of data as conclusive as the bristlecone pines.

To say the least, there are a variety of possible explanations for the findings involved with the bristlecone pines, so even with the pines, the studies don't pass established statistical tests of significance. Certainly without the pines, and perhaps one or two other highly localized results that are less conclusive, the twentieth century has not been proven to be warmer than any previous century--and it may have been less warm than the medieval period. Now that's a consensus. McIntyre is not committed to the view that the medieval period was warmer than the 20th century, but he suggests there is plenty of data to support that view. On the other hand, toward the end of his presentation he says there are several areas where glacier retreat indicates the twentieth century was warmer than the medieval period; plant life is being exposed that was growing thousands of years ago, and presumably was not exposed during the medieval period. There is one study indicating that in ancient Roman times the Swiss Alps were green in vast areas where they have been white--with snow and ice--more recently.

Perhaps there was a really warm age from some thousands of years BCE to just before the CE, then an age almost as warm in the medieval period, and then another warm period in the twentieth century. If so, then of course the earlier warm periods had nothing to do with man-made CO2 emissions, and probably nothing to do with man-made anything. Also the obvious point needs to be made: there are always tremendous local variations on earth when it comes to temperature. Some of those taking the Gore view say: if there have been various warming periods, with rapid, dramatic and potentially dangerous temperature changes caused by God knows what, then this supports a precautionary approach: man-made CO2 just might cause dramatic etc. changes. But this is not far from saying the sky might be made of cheese; let's take precautions (whatever they might be) to be on the safe side.

It's surprising to me how few actual data sets are referred to here. Supposedly some famous people in the Al Gore camp have claimed it is too difficult to go out and get such things as tree samples. Some local results point toward warming, others toward cooling. There is a tendency to dismiss what doesn't fit one's theory, and say it is "merely local." But then the data that does fit your theory is probably local as well. Gore's defenders have admitted that he could hardly have made a worse choice of a glacier to emphasize than the one he chose: Kilimanjaro. There is a group of reputable scientists who are convinced the retreat of the glacier there results from local factors, not temperature change.

I guess to understand climate, and climate change, we would have to understand such things as water vapour and cosmic rays.

No comments:

Post a Comment