Temperature trends since 1998: Puzzling evidence?

In “Momentum on Climate Pact Is Elusive” (New York Times, 9/21/09) science reporter Andrew Revkin discusses the relative stability of temperatures over the last decade in the context of the UN climate summit this week. He posits that this short term trend may complicate efforts to achieve an international agreement on climate change this year.

While giving voice to scientists’ concerns about the difficulty of communicating to the public and policymakers the scientific basis of climate change, Revkin unfortunately also contributes to it. The article is premised on the notion that that the current, relative flat patch in temperatures may raise questions in the minds of the public and policymakers about the urgency of action to address climate change.

By implying that this stable period is meaningful in the context of climate change, Revkin is exacerbates confusion about the definition of climate. Climate is defined (see, e.g. http://climate.columbia.edu/?id=faq#1) as average weather in a specified geographic region over multiple decades- at least 30 years. By definition, then, short term trends such as the one Revkin points to do not constitute climate. Revkin obscures this delicate point by referring to “climate stability” in his discussion of global temperatures of the last 10 years.

The article is about the difficulty of conveying the difference between “climate” and “weather” to a non-expert audiences – Revkin is a science writer whose bailiwick is to interpret science for the public. He is obligated to be particularly careful in defining his terms.

The trajectory of temperatures over the last 130 years- according to instrumental records- looks like this:

Source: http://data.giss.nasa.gov/gistemp/graphs/

The graph depicts annual average departure from the average temperature over the period 1951-1980, for the years 1880-2008. Substantial inter-annual variability is readily apparent within the extremely strong long term warming trend.

The discussion of sources of data about past climate and projections of future climate only adds to the confusion. Revkin reports that proxy data from tree rings, ice cores and other long-lived substances and computer models “…[W]hile persuasive to most climate scientists, are not infallible”.

Of course these methods are not infallible. The scientific community is very careful in pointing out the many fallibilities and assumptions in any interpretation of data or model results. However, by painting model results and proxy data with the broad brush of “fallibility” Revkin fails to make a critical distinction: between an imperfect and categorically incorrect result (e.g. the earth will cool over the next century, not warm) and an imperfect result that is robust but contains inherent uncertainty (e.g. the earth will warm 4 +/- 2 degrees C over the next century). Such language can easily mislead the public by allowing readers to infer that paleo-proxies and models may be simply, categorically wrong.

The climate system is large and chaotic. Climate science, like all science, is filled with uncertainties. All uncertainties are not created equal, and in this article Revkin and the Times miss an opportunity to help people to understand and evaluate those risks and uncertainties.