According to a 2007 report by the U.N. Intergovernmental Panel on Climate Change (IPCC), a group of hundreds of scientists that assesses and reports on climate change research, state-of-the-art climate models indicate the odds are about 90% that manmade influences are and will continue to be the principal cause of global warming. Model estimates of global average temperature rise over the 21st century range from 3.2°F to 7.2°F.
To what extent should we accept these projections at face value? How certain is the stated range of uncertainty? Can today's climate models provide credible predictions of the regional impacts of climate change (e.g., on the scale of U.S. states or most European countries)?
These questions are addressed in a series of recent articles and exchange of comments in New Scientist.
Keep reading for more on the certainty (or lack there of) of climate models...
Lenny Smith, a statistics professor at the London School of Economics and Political Science (and someone who I've worked with in the past and regard highly), believes human activities are changing the global climate, but that climate scientists are "overselling" their results.
"...we must stop pretending that we know the details of how it will all play out," comments Lenny, who points out that the estimates of uncertainty -- based on the distribution of results from as many as 300 runs of global climate models -- are themselves uncertain. This is especially true, he says, in the extremes (or "tails") of the distribution, which are often particularly important for decision-makers. Thus, it's quite possible that future warming could be significantly more or less than the range indicated by the models.
Moreover, as I indicated in an earlier post, the credibility of climate models is especially questionable with regard to forecasting regional aspects of climate change.
In questions and comments responding to Lenny's remarks, the obvious issue raised is whether we should believe the reports of the IPCC and, by inference, any statements by climate scientists based on climate models. Lenny and others say broadly yes -- as long as the qualifiers are acknowledged and carefully read. To that end, they point to one item in particular -- buried deep in the report's first section (chapter 10) -- that has likewise bothered me. Namely, the report explicitly acknowledges that the range of uncertainty in warming is too narrow. But, to quote one commenter, "It's good that the qualifier is in there, but it is a hell of a qualifier to find on page 797."
So while climate models should not be ignored, the majority of climate scientists who believe that humans are contributing to global warming should be especially forthright in not overselling their case. To which I must add that global warming skeptics should be equally concerned about not overstating their position by exaggerating justifiable questions about the credibility of climate models.
An egregious example of such exaggeration is one contributor to the New Scientist articles who argues it's about time that "climate modelers may have to recognize that we have learned most of what we can from their number crunching." In truth, as data increases, research builds and technology advances, climate models are undoubtedly providing ever increasing knowledge and understanding of the fundamental processes that govern Earth's climate system.
Overall, there is agreement that climate models can be valuable decision-making tools. They can yield information on plausible risks and minimize vulnerability, although not necessarily provide totally reliable estimates of the odds. As Lenny puts it: "When I cross the street, average statistics about cars and how they are driven are of less value to me than the sound of a bus heading my way. Models help us listen for that bus." In other words, models can tell us what to listen for, such as increasing levels of atmospheric carbon dioxide.
To me, the most important question is how societies respond to the reasonable, albeit less-than-certain chance that human-caused global warming and its consequences will continue -- and that it might possibly reach extremes not now encompassed by climate models. This is a classic example of risk analysis and decision making -- i.e., weighing the costs of action to reduce carbon dioxide emissions, which conceivably might not be the cause of global warming or may cause less warming than predicted, versus the costs of inaction, which could have profound negative societal impacts if, as is more likely, carbon dioxide is a principal driver of global warming.
Whatever the case, careless use of results obtained from climate models, as well as unwarranted skepticism of climate scientists and their models, can unnecessarily muddy the waters and lead to delays in critically urgent policy decisions.