3. It appears that some subjectivity is necessary in making the optimistic-sounding statements in the first two paragraphs. Given the very small absolute changes that have been observed over the 20 years of the focus period of 1979-1999 (~ 0.2K) and the larger natural variability that has occurred over that time (~ 0.5K), it seems that the data can neither reject nor confirm the hypothesis that the models are in some sense reliable. The Executive Summary focuses almost exclusively on the 1979-1999 period, a rather short period for which the trends discussed would not be significant at the 5% level, if the significance testing were done in the usual way. The choice to present the results without any statistical significance testing or confidence intervals is highly questionable and ordinarily not allowed in the scientific literature.

4. Some additional support for the conclusions in the first two paragraphs might be gained if the issue of separating the stratospheric and tropospheric signals in the Microwave Sounding Unit (MSU) channels were addressed more directly in the body of the report and carried forward to the Executive Summary. In this regard, the Fu et al. results appear important but are basically ignored in the explicit arguments used to formulate the conclusions. This should be remedied. The Fu et al. work may be central to the issue of measuring and interpreting the vertical profile of temperature change and is no more difficult or controversial than addressing the spurious radiosonde trends or the differences between the Remote Sensing Systems (RSS), University of Alabama, Huntsville (UAH), and University of Alabama (UMD) interpretations of the MSU data.

5. The traceability of the conclusions in the Executive Summary to the detailed arguments presented in the chapters is not entirely clear. In some cases the Executive Summary does not appear consistent with the corresponding chapters where the conclusions should have been developed. A way to clarify the source of the conclusions might be to develop major conclusions in the chapters and then move these statements forward to the Executive Summary.


1. The Executive Summary should start with a statement about why the reader should care about the subject of the report.

2. The first two paragraphs appear to be the abstract of the document and contain the main takeaway messages. Both paragraphs conclude with statements about the comparison of models with observations. In the second paragraph, it is not clear what consistent means. The observations seem to give qualitatively different conclusions on several key points. Does consistency merely mean that the observations and the model results have overlapping uncertainties?

3. In lines 71-73, the conclusion that the report increases confidence in our understanding of recent climate change seems optimistic and inconsistent with the supporting evidence. The evidence concerning this conclusion needs to be spelled out more clearly in the document and succinctly summarized in the Executive Summary. It must be possible and convenient to trace these conclusions back to arguments that can be evaluated scientifically.

4. Figure 1 is not effective. For example, the color scheme should be changed because the “multi-colored line” is mostly just one shade of blue.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement