INCOHERENT SURFACE TEMPERATURES?

INCOHERENT OR COHERENT?

Data stands at the foundation of scientific investigation. An element of the scientific process is the examination or challenging of results by independent investigators. By training and personality, good scientists are skeptics; as long as there is uncertainty, they continue to return to the problem. One of the responses (ZRR) to the previous blog pointed out a report "The Surface Record." This is a report which examines the way that the surface temperature record has been calculated and questions the accuracy of the calculation.

One of the concerns raised in the report is the possible inconsistency between satellite temperature observations and thermometers at the Earth's surface. This is at the core of a well publicized scientific debate. In a series of papers in the September 2, 2005 issue of "Science" the source of this inconsistency was exposed. Most importantly, the paper by Carl Mears and Frank Wentz pointed out an error in the processing of the satellite observations. With the correction of this error, the temperature data sets are, in fact, consistent. Not only are they consistent, but there is a powerful coherency revealed by the amplification of the warming at the surface in the troposphere (The paper by Ben Santer and colleagues, same issue of "Science.")

Another concern expressed in "The Surface Record" is the treatment of various sources of errors in the surface temperature record, for example, urban heat island effects. The treatment of these errors offers challenges, and there are several published strategies for overcoming these challenges. In addition, one of the beauties of having a modeled Earth is that it can be sampled exactly like the observing system. In that way the adequacy (or inadequacy) of the sampling network can be addressed. Error bars can be assigned; these are reported in the IPCC reports (see Figure 1). In fact, determination of urban heat island effects is robust enough that they are explicitly reported in "Climate Change 2007."

Scientists challenge results by training and by human nature. If you challenge "The Surface Record," then even at its time of publication there are papers which would answer many of the concerns raised in the report. More importantly, scientific results are not static; new information is obtained; our knowledge evolves and becomes more robust.

Striving for coherence,
ricky


Figure 1. From IPCC 2001. Variations of the Earth's Surface Temperature. In the bottom panel the gray region represents the 95% confidence range in the observations. The increase at older times represents the lower quality and less robust sampling. Sources of errors are all investigated and reported by scientists; if they are ignored, then it is not science.