community consensus). However, details of the causes of these biases should be clearly described and a stronger explicit case made for discounting the stratospheric results. More critically, it is important to assess if the stratospheric bias problem extends into the middle and upper troposphere, especially in the tropics. The vertical range for which the radiosonde data have least the uncertainties for trend analyses should be more carefully discussed. Finally, the prospects for removing the biases or more effectively utilizing these data should be discussed.

7. The recommendations in Chapter 6 are too non-specific, unprioritized, and largely disconnected from the findings in Chapters 1-5. We suggest that Chapter 6 be reorganized into two parts:

a. The first part should take findings from Chapters 1-5 to recommend specific opportunities to improve understanding of vertical temperature trends. These recommendations should focus on understanding remaining uncertainties in existing satellite and radiosonde data sets.

b. The second part should focus on future measurement opportunities in the context of the specific goals of the report for reconciling observation and understanding of temperature trends.

8. Changes in the presentation and content of the Executive Summary are needed to make key results more accessible to a wide audience and ensure traceability to the results in Chapters 1-6. A possible strategy is to bring forward key bullet points from each chapter as answers to the six main chapter questions, followed by brief explanatory text, key figures, and implications for understanding within each chapter. The Executive Summary should reflect an appropriate balance of new results and outstanding uncertainties. The first page of the Executive Summary should concisely summarize the key results of the report in a short abstract.

One further summary comment is relevant for the CCSP synthesis and assessment reports in general. The committee feels that the current report suffered to some degree from the author group assessing their own work and excluding other independent work. This is evidenced by a lack of critical evaluations on some key data issues and numerous citations to their own work. To the extent possible, the authors should not be put in a position of assessing work where they have a vested interest in the outcome. While it is not reasonable to revise author teams for this report, those preparing future CCSP reports should carefully consider this issue.



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