date: Fri, 18 Sep 2009 19:59:03 UT from: jgr-atmospheres@agu.org subject: 2009JD013094 Joost de Gouw: Review Received by Journal of to: p.jones@uea.ac.uk Content-Disposition: inline Content-Length: 3736 Content-Transfer-Encoding: binary Content-Type: text/plain Dear Dr. Jones: Thank you for your review of "On the reliability of the U.S. Surface Temperature Record" by Matthew J. Menne, Claude Williams, Jr., and Micheal Palecki [Paper #2009JD013094], which we have safely received. A copy of this review is attached for your reference. Sincerely, Joost de Gouw Editor, Journal of Geophysical Research - Atmospheres ------------------------------------------------------------------------ Assessment: Category 2 Ranking: Very Good Confidential Re-Review: Yes Annotated Manuscript: No Highlight: No Highlight: Comments: Review of Menne et al General This is an extremely useful paper that should be published as soon as possible. I've separated my comments into major and minor, even though the major ones can be accommodated without any new work. Major Comments 1. The study highlights that the issue of poor exposure of instruments cannot be solved with photographs and metadata. Instead the data recorded at the sites needs to be considered as well. The authors make this point at the end in the conclusions, but I'd like this point to be made in the abstract as well. In particular there is the point that unless the data are looked at, the effect of inhomogeneities cannot be assumed as some of the results are counter intuitive. Possibly the authors are being a little defensive by saying that the adjustments for MMTS are somewhat inadequate, but they have looked at the data. 2. It should be clear in section 2 that not all sites in the USHCN have been assessed into these ratings. Perhaps the number at the present could be given, even if the exact number is somewhat difficult to ascertain. 3. Perhaps the implication of Vose and Menne (2004) could be more clearly stated around lines 123-125. The fact that a limited number of sites is able to reproduce the CONUS average clearly shows that the number of independent spatial degrees of freedom across the CONUS is significantly smaller than the full number of sites in the networks. Minor Comments 1. Not that keen on CONUS as an acronym, but it serves the purpose. 2. MMTS is not defined in the abstract. It is in Section 1 and maybe that is all that is needed. 3. There are lots of references to non-climatic artifacts and homogenization. I'd like to see a few more referred to in addition to Mitchell (1953), which is quite old now. Perhaps a reference could be made to a paper where homogeneity has been defined. 4. Line 136, add the 'also' in this sentence. 5. Perhaps there could be a more clear statement that exposure issues tend to have a greater impact on Tx as opposed to Tn. The effects are almost always opposite, so will cancel in the mean T and be amplified in DTR. 6. With Figure 5 (lines 178-181) I'd like to see reference to the figure in Menne et al. that shows the histograms of the adjustments. I think this is additional information to that provided in Figure 5. I'm not saying add this figure, just refer clearly to it. 7. Almost all the refs are in italics, but a few aren't - e.g. Hubbard and Lin (2006). 8. A map of the locations of the USCRN sites would be useful, or a direct reference to where this can be viewed. The fact that there are 114 stations at 107 locations is a little confusing. Also is there a need to mention precipitation. Presumably the USCRN sites are measuring many variables. 9. With Figure 6 there are only 5 points involved in these excellent r-squared values. A better plot might be the anomalies from all 60 months within the 5 years? 10. The final sentence could add in the clause - provided the raw data are adjusted according the methods details in Menne and Williams (2009) and Menne et al. (2009).