cc: jto@u.arizona.edu, Valerie.Masson@cea.fr, olgasolomina@yandex.ru, ricardo@lab.cricyt.edu.ar, Eystein Jansen date: Thu, 23 Feb 2006 11:02:00 +0100 from: Valérie Masson-Delmotte subject: Re: latest draft of 2000-year section text to: Keith Briffa Dear Keith, A few rapid comments on the section 6.6 revised text. I have enjoyed reading it, more concise, less defensive and key conclusions appear more solid. Sometimes the text is written in the past tense, sometimes in the present tense : it could be homogenised. Please remove the sentence page 6-15 "The paleohydrologic record of North America is the most complete and diverse of any of the world in part due to the proximity to many well equipped labs but also due to the concern of the frequent change in drought, flood...". This has nothing to do in a scientific assesment (equipement versus motivation). The same motivation should hold true for all tropical areas! It would be worth to discuss in one paragraph somewhere (possibly together with the text page 6-6 about the proxies) the methods of tree ring standardisation which seem to have changed over time and lead to larger low frequency signals in the tree ring width based reconstructions. Comments on the structure : 6.6.1 I think that the italic question for the section does not work. I suggest to add sub questions such as : What do early instrumental records tell us? (p6-2, lines 7 to 39) What new reconstruction efforts have been conducted since TAR for NH temperatures (6-2 lines 41 to 6-6 25) What are the main sources of uncertainties in large scale climate reconstructions (6-6 lines 27 to 49) - should refer to the section introduction / description of proxies What do NH temperature reconstructions tell us (6-6 lines 51 to 6-8 line 5) Regarding climate forcings and simulations (6.6.3 and 6.6.4) there must be a cross verification with chapter 9, have you looked at their revised text? The title 6.6.3 includes too much refereence to modelling. They have been also statistical efforts to relate forcings and respondes (not only physical models) which have to be mentioned. Then modelling should be in 6.6.4 only. Another way could be to combine both in one section : 6.6.3 would be model-data comparisons with 1) forcings and 2) simulations versus reconstructions. Section 6.6.5 is too long compared to the # of studies conducted here. Minor comments : 6-3 2 line 20 add "North European records" line 27 and onwards I think that Boehm reconstruction should be cited around the Alps back to 1780 (it really deserves to be cited). line 33 Chuine et al puts the French heat wave in a 700 perspective with grape harvest dates, which could be mentioned. line 36 shorten to "detailed changes in various climate forcings" line 44 : what are the documentary sources incorporated by Mann? I understand essentially early instrumental records. 6-3 line 49 : this paragraph is a bit vague. Maybe mention more clearly areas where no data are available. Goosse et al GRL 2004 used a synthesis of Antarctica data + simulations to discuss the pb of phase with Antarctica and could be mentioned. I suggest to replace "assimilated" which has a special meaning for meteorologists by "combined" 6-4 line 9 change"are" to "is" line 16 : how many such long records are available (= what are "very few"?) 6-3 line 39 : is it the rapidity of the 20th c warming or the level of late 20th c temperatures that have to be discusssed? 6-5 line 8 use reconstruction, not "series". I understand that one series is one proxy record and a mixture of records with various statistical methods is a reconstruction. Line 31 : add "many of the individual annually resolved proxy series". 6-6 line 30 change "over a fixed calendar based time window such as J-A or J-D" to "over a specific season" 6-8 line 29 : I propose to change the text about tropical ice cores. There are few strongly temperature-sensitive proxies from tropical latitudes. Water stable isotope records from high latitude tropical glaciers where first used as temperature proxies but recent calibration and modelling studies have confirmed that tropical precipitation isotopic composition is mostly sensitive to precipitation changes ("amount effect") at seasonal to decadal time scales both in south America and south Tibet. References : *Hoffmann G*, *Ramirez E*, Taupin JD, et al. Coherent isotope history of Andean ice cores over the last century GEOPHYSICAL RESEARCH LETTERS 30 (4): Art. No. 1179 FEB 25 2003 *Vuille M*, Werner M, Bradley RS, et al. Stable isotopes in precipitation in the Asian monsoon region JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 110 (D23): Art. No. D23108 DEC 8 2005 By the way, in the same paragraph, you cite tropical glacier retreat as caused by temperature changes. I suggest to refer to chapter 3 on this topic because many studies have also shown that precipitation / relative humidity / albedo effects can be very important for tropical glacier mass balance (see for instance Vincent et al, Comptes rendus Geosciences 2005). Page 6-8, ground surface temperatures : are there tropical records available that could be explicitely discussed? The problem of calibration mentioned line 29 (lack of the last decades of the 20th century) also holds true for many of the long tree ring records... should it be explicitely highlighted here? 6-9 : line 9-10, what is a "much longer warm period", I do not understand. I think that this could be shortened. I still suffer that Antarctica is not mentioned at all. In Goosse et al 2004 I made a stack of 6 records from East Antarctica. There is also one good borehole record from Law Dome (Dahl Jensen Annals of Glacio 1998) showing the same features. 6-10 line 28 : I do not think that it is appropriate to discuss the Solanki paper here. 6-10 and 11 : why mix volcanic and anthropogenic sulface aerosols rather than 2 sections? Why not discuss changes in surface occupation (land use) in the forcings for the last millenium at least in one sentence? 6-12, lines 38 and onwards : it seems that this is attribution and detection and should be a summary of chapter 9 or just a cross reference to chapter 9. Section 6.6.5 (6-12 and 13) is too long compared to the studies cited. Maybe Fortunat could help to make this section more punchy. Should the PhD thesis of MacFarling Meure be cited in this assessment? Remove "the best known aspects of the records" Refer to chapter XX for biogeochemical cycles The last paragraph is probably redondant with respect to the carbon cycle climate feedback discussed in that chapter. Page 6-14 line 43 : redundancy in this paragraph. Does the coldest European winter have to be discussed in such detail? I would skip this (remove line mid 42 to beg of 45 and keep the last sentence of the paragraph which basically says the same thing. The section on Asian monsoon variability is not focused on the last 2000 years but on millenial variability => mix with 6.4? Why not cite the Tibet ice core records here (ex Dasuopu 18O which should be a local precip record). There are also high res speleothem records with high resolution. Ramesh should help on this paragraph. I hope that you find this useful, congratulations for the large improvements of this section and taking into account a record number of comments... Valérie.