date: Tue Sep 29 15:35:53 2009 from: Phil Jones subject: Re: South American Temp. reconstruction paper draft to: Raphael Neukom Ralphi, Jones, P.D., Raper, S.C.B., Cherry, B.S.G., Goodess, C.M. and Wigley, T.M.L., 1986: A Grid Point Surface Air Temperature Data Set for the Southern Hemisphere, 1851-1984, U.S. Dept. of Energy, Carbon Dioxide Research Division, Technical Report TR027, 73 pp. I'm not saying you should use these, but they might do better than those in GHCN. Hopefully it won't take long to check. I suspect there will be little in it, so you might be able to use what you already have. I'm fully aware of how sensitive a PCR program can be to the addition/deletion of one site! Cheers Phil At 15:11 29/09/2009, you wrote: Dear Phil, Thanks a lot! - Thanks for sending the CRU station data. so you recommend using these instead of the GHCN stations for our verification analysis (table 4)? how would I cite the data? - The issue of the importance of the proxies is indeed complex. I made a lot of tests. For instance I am aware that CAN 11 first lets the skills drop but over the entire period the skill is better if I take it. Also, often the REs of the mean and the fraction of locations with positive REs react in opposite direction by adding/removing one proxy. The influence of each proxy changes with every new proxy combination. (that is also the reason why I started now working with ensembles instead of just using one "optimized" proxy set). - I'll make a compilation of my proxy records and send it to you. and I'll send you the contacts for the confidential material. Cheers Raphi Phil Jones schrieb: Dear Raphael, Thanks for the quick reply. Here are a few extra thoughts within your reply. Looking forward to reading the paper later in the week. Cheers Phil At 12:47 29/09/2009, Raphael Neukom wrote: Content-Type: text/plain; charset=ISO-8859-1; format=flowed X-MIME-Autoconverted: from 8bit to quoted-printable by ueamailgate02.uea.ac.uk id n8TBl8IL029484 Dear Phil, Thank you for your comments/questions. - CLARIS-LPB. No problem, we can send them the submitted version. OK - UK bid on SH proxies: Sounds interesting! The raw data will generally be available except some of the series that I reveived from the authors confidentially, so I can not distribute them. most of the data are, however, already on the NOAA paleo DB. - I do have the raw tree ring cores from all sites (I detrended all series again using the same method). Again, the raw and detrended data will be available except a couple of confidental records. OK - you finish your thesis! When you've more time then perhaps I can get the series then. Or you could send them when you have some time. Maybe package all up in a zip file. If you then let me know the people who asked you not to pass them on then I can contact them. -"1. Can you send the data (monthly averages) for the series in Table 4?" The monthly data are attached in the csv file (directly imported from the GHCN server). Any longer or better data would of course be very welcome. Here's a file attached of the CRU data. They seem similar on a few spot checks. we have made alterations to the following Rio before 1940 We don't use Sao Paolo Punta Arenas before 1963 Asuncion before 1964 Salta several periods of change Ignore station 2 for Cordoba - 87344 - use other one Buenos Aires before 1955 Bahia Blanca before 1930 I could only find these ones quickly. Alterations are small and probably won't affect the year-to-year changes that much. I recall some of the others having many gaps. I presume your using the GHCN data after the NCDC homogenization checks. -"2. Can you also send the daily series for the 4 long Argentinean daily series from Vargas and Naumann (2008) if that is possible?" I just asked the authors to be sure they are ok with that, but probably I can send them to you, soon. Thanks - hopefully! -3. "regression weights for each location...". To be honest I have never than that and I am am not sure if I understand how I can do that after transforming the pc's back. just by doing a multiple regression at each location? Maybe Jürg can help me. or do you have an example script? Worth talking to Juerg about this as I'll mention this again when I read more carefully later in the week. You are doing regressions of one set of PCs (the predictors) on the other set PCs (the predictands). It is possible to transform all back to equations the left hand side as a function of the right hand side Don't worry about this. Juerg knows that for his European pressure reconstructions depend a lot on having Paris and Uppsala pressure in the earliest years. You can sort of see what your reconstructions depend on by looking at the jumps in your Figure 8. If you compare the start years with the jumps. Cariaco basin improves things when it starts, but this falls back when CAN 11 starts. If you plot some vertical lines on Fig 8 at the start years of the two sets of predictors. -"If we were to change CRU TS 3.0 how long would it take you to rerun things?" Well for temperature it would take a week with all plots etc. I guess. I would suggest only do it if the changes are excepted to be significant. I now also started to reconstruct precip (using an ensemble approach, bootstrapping different proxy sets and parameter settings) and I think an update version could improve things a lot for precip. However I have to hand in the thesis at the end of the year, so I have to run the reconstructions within the next few weeks... I don't think things will change that much, so ignore this thought. Agree you have to finish. Thanks again and if you need more (proxy) data just ask. Cheers Raphi Phil Jones schrieb: Raphael, I will try and look at this over the weekend and get some comments back to you. It all looks quite interesting. I have a few questions now though. I'm involved in an EU project called CLARIS-LPB. Would it be possible to send a later draft to some of the people in that group? There is a paleo group looking at Pampean Lakes. I'm not involved in this group, but in an instrumental WP within the project. I'm thinking of a version once submitted or sent around more widely. Second I'm putting together a large UK bid on SH proxies - for Dec 1. Maybe I can mention this paper and the EU project and NCCR funding? We won't hear about that bid till next April and then it wouldn't start till 2011. The point I'd like to mention is that you've got an excellent dataset together. So the question is will the raw data be available once your paper is submitted/published? Related to this, do you have all the individual tree cores for each tree-ring site, or do you just have the chronologies? The text from a very quick reading seems to suggest that you have some of the cores? Again would these be available to other projects in the future? Now 3 points about the paper. 1. Can you send the data (monthly averages) for the series in Table 4? I'd like to get someone here to look through the CRU archives to see if we can extend some of them and/.or check that we've not made adjustments to any of them. I know we have adjusted Punta Arenas, and have longer data for Ushuaia. I'm thinking mainly of the data pre-1901 data here. 2. Can you also send the daily series for the 4 long Argentinean daily series from Vargas and Naumann (2008) if that is possible? 3. Finally, in your PCR technique you should know the regression weights for each location, so by adding these up across the SSA grid you will know which are the important proxies. This might be worth mentioning in some way, provided you have the information. In the CRU program which is doing similar things, we can get the regression weights once everything is transformed back to real-world variables (i.e. not in PC mode). We are still working on CRUTS3.0. Things are not going very quickly as the people have other things to do. If you're not using any data post-2000 then things are unlikely to change. I have a question. If we were to change CRU TS 3.0 how long would it take you to rerun things? I don't think we will, but we may have to for precipitation. Again - it looks like a good paper. I'm sure when I read it in detail I will have some more questions. Cheers Phil At 12:53 28/09/2009, Raphael Neukom wrote: Dear Coauthors, please find attached the draft of our new paper dealing with climate field reconstructions from southern South America. Any comments, suggestions and corrections from your side are most welcome. If possible, please send them to me as track changes within the next 10 days. Thank you very much and cheers Raphael -- Raphael Neukom Climatology & Meteorology (Klimet) Institute of Geography University of Bern Hallerstrasse 12 CH-3012 Bern Switzerland Tel.: +41 31 631 8868 Prof. Phil Jones Climatic Research Unit Telephone +44 (0) 1603 592090 School of Environmental Sciences Fax +44 (0) 1603 507784 University of East Anglia Norwich Email p.jones@uea.ac.uk NR4 7TJ UK ---------------------------------------------------------------------------- -- Raphael Neukom Climatology & Meteorology (Klimet) Institute of Geography University of Bern Hallerstrasse 12 CH-3012 Bern Switzerland Tel.: +41 31 631 8868 Prof. Phil Jones Climatic Research Unit Telephone +44 (0) 1603 592090 School of Environmental Sciences Fax +44 (0) 1603 507784 University of East Anglia Norwich Email p.jones@uea.ac.uk NR4 7TJ UK ---------------------------------------------------------------------------- -- Raphael Neukom Climatology & Meteorology (Klimet) Institute of Geography University of Bern Hallerstrasse 12 CH-3012 Bern Switzerland Tel.: +41 31 631 8868 Prof. Phil Jones Climatic Research Unit Telephone +44 (0) 1603 592090 School of Environmental Sciences Fax +44 (0) 1603 507784 University of East Anglia Norwich Email p.jones@uea.ac.uk NR4 7TJ UK ----------------------------------------------------------------------------