cc: Beck Christoph , Rudolf Bruno date: Tue May 3 08:25:41 2005 from: Phil Jones subject: Re: AW: Your views appreciated to: Grieser Jürgen Jurgen, Bruno and Christoph, Thanks for the comments. The'll be discussed at the WGI IPCC meeting next week (May 10-12). Cheers Phil At 16:42 29/04/2005, Grieser Jürgen wrote: Dear Phil, Bruno forwarded your todays email to me and I am very glad to be asked to reply. Pls find my comments in the attached word document. In case of further questions, comments, etc. pls don't hesitat to contact me again. Best regards from Christoph Beck, Bruno Rudolf and myself, Jürgen. ******************************************** Dr. Juergen Grieser Global Precipitation Climatology Centre GPCC Deutscher Wetterdienst P.O.Box 10 04 65 63004 Offenbach Germany Tel.: +49 -69 8062 2873 Fax: +49 -69 8062 3759 Web: [1]http://gpcc.dwd.de <[2]http://gpcc.dwd.de/> ******************************************** -----Ursprüngliche Nachricht----- Von: Rudolf Bruno Gesendet am: Freitag, 29. April 2005 13:39 An: 'Phil Jones' Cc: Grieser Jürgen Betreff: AW: Your views appreciated Hi Phil, thank you for contacting us. A very first quick reply: Yes, what you called the fixed stations versions should be used for the purpose discussed. We call it 50 year climatology, because it is optimized for homogeneity, and it is THE only of our products recommended to use it for climate variability (trend) studies. And relative anomaly is used for interpolation for this product. Juergen Grieser will send a more complete reply on the 50 year climatology. The Monitoring Product and Full Data Product are based on interpolation of precipitation totals because we had no normals for many of the stations with data for the recent years. For the recent years, most data are only available from SYNOP, and the accuracy of those is very low. Compared to normals, we partly obtain unrealistic anomalies, which are spatially exported by gridding. Maps resulting from the anomaly maps combined with normal maps showed some curious structures, whil the results from direct interpolation was still plausible. This assessment is based on earlier studies. We will do the methods comparison again for the currebt version of the full data basis. But whatever one does in analysis: if the gauge data base (number of stations) is very different in time, one will not get a homogeneous product (my opinion). Cheers, Bruno -----Ursprüngliche Nachricht----- Von: Phil Jones [[3]mailto:p.jones@uea.ac.uk] Gesendet am: Freitag, 29. April 2005 11:16 An: Rudolf Bruno Betreff: Your views appreciated Bruno, Shortly after AOPC, I was back with my IPCC hat on modifying text based on reviews of our first zero-order draft. I go to Beijing the week after next for the next Lead Author's meeting. By mid-August, we will have a new draft (the first-order one) which can be downloaded by anyone over an 8-week period then. I expect many hundreds will and there will be thousands of comments. That was some background ! I sent your email about VasClimO to some of the people we've involved as contributing authors (CAs) and also to our Lead Author for our precipitation section. Our LA for this is Dave Easterling. The details also went to Aiguo Dai (who is with Kevin Trenberth at NCAR). My intention was to alert this group to your latest work at GPCC. I said that your dataset from 1951 was likely the best available. Our chapter currently has trend maps for 1901-2004 and 1979-2004 from GHCN gridded fields (Dave Easterling produced these). Dave hasn't responded on this yet. The point of this email is to get your views on the email below from Aiguo Dai. We will almost certainly have some time series plots from different data sources including GPCC, GHCN, maybe CRU and Chen et al.. What I would like before I leave for China on May 7 (so by middle of next week) are your brief views on some of these other products (e.g. Chen et al. 2002, GPCPv2 from Adler et al., 2003)? Would you recommend using your fixed station version? I am very sympathetic to the view that you should grid using anomalies. Can you briefly say why you didn't? If the answers to all these questions are in the report, tell me and I'll read it. I've not had time to yet. I am just trying to get all the views together for the Beijing meeting. I expect Kevin Trenberth will want to accept what Aiguo says. For the moment, just reply to me. Cheers Phil Date: Thu, 28 Apr 2005 16:17:07 -0600 From: Aiguo Dai Organization: NCAR, Boulder, CO User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en To: Phil Jones CC: trenberth@ucar.edu, Peter Ambenje , Roxana Bojariu , David Easterling , David Parker , Fatemeh Rahimzadeh , Jim Renwick , Matilde Rusticucci , Brian Soden , Panmao Zhai , Albert Klein Tank , Aiguo Dai Subject: Re: Fwd: News about GPCC X-Spam-Score: 1.3 X-Spam-Level: + Dear Phil et al.: Based on my reading of their tech, report, GPCC is still gridding the total monthly precipitation amounts instead of anomalies. Many people (e.g., Jones and Hulme 1996; Chen et al. 2002) have shown that it is much better to grid the anomalies besides different gridding methods, even for temperature. GPCC argues that gridding monthly anomalies is better for some regions but worse for other regions, which I can not understand. Because of this griding method, one can not use their version 3 full data product for assessing climate changes (because addition or removal of wet/dry stations will have large impacts on regional estimates of precipitation), while their version with fixed stations (9343, but still 10% of the years may have missing data) does not make full use of the available data. For example Chen et al. (2002) and GHCHv2 use more than 10,000 gauges from 1948-around 1993 (over 15,000 during the 1960s and 1970s). I strongly recommend GPCC to grid anomalies using all available gauge data for each month. Maybe they can listen to people like Phil? There appears to be large differences among various estiamtes for global (land) precipitation for the last several years (1997-present), partly because of limited number of raingages available. The attached figure compares land precip during 1979-2002 from Chen et al. (2002), CRU (New et al., Mitchell et al.) and GPCP v2 (Adler et al. 2003, with climatological corrections for wind-induced undercatch). I think it would be useful to add both the Chen et al. (1948-present) and the new GPCC (1951-2000, the fixed station version) analyses into the precip plots of IPCC chapter 3. Regards, Aiguo Dai 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 ---------------------------------------------------------------------------- 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 ----------------------------------------------------------------------------