date: Mon, 20 Sep 1999 16:36:32 +0200 from: Inger Hanssen-Bauer subject: contribution TAR to: Hans von Storch Dear Hans, Please find, at the end of this mail, a possible contribution to chapter 10.6 in TAR. Our main messages are: 1) When developing empirical downscaling models, it is essential to use long training periods. 2) The early twentieth century warming and the warming during the last 3 decades seem to be quite different with many respects. In Norway and Spitsbergen, it is possible to explain most of the warming after the 1960s by changes in the atmospheric circulation. The warming prior to 1940 cannot be explained in this way. We give 2 references. The second of these is presently published as an “institute report” only (we will send you a copy if you want one), however we are presently preparing a “journal version” of this work. We send a copy of our suggested contribution also to Giorgi. Best regards Inger and Eirik Suggested contribution to ch. 10.6 in TAR: " Empirical downscaling models using SLP as the only predictor have been developed both for Svalbard in the Atlantic sector of the Arctic (Hanssen-Bauer and Førland 1998), and for Norway (Hanssen-Bauer 1999). The models could mainly reproduce the 20th century observed long-term trends and decadal scale variability for precipitation. The models were also able to reproduce temperature trends and variability after 1960. However, the early twentieth century warming, which was especially pronounced in these areas, was not satisfactorily modelled. One may thus conclude that, while the warming in these areas during the last 3 decades mainly may be explained by changes in the atmospheric circulation (at average increased zonal circulation in winter and spring), the early twentieth century warming cannot be explained in this way. A conclusion of more general character, is that rather long data-series are needed to ensure sufficient training and testing of empirical models. By using only the last 30-50 yr of the data series, the performance of the temperature models (Hanssen-Bauer and Førland 1998, and Hanssen-Bauer 1999) would apparently be satisfactory. Applying the models on earlier decades reveal that this is not the case." References: Hanssen-Bauer, I. and E.J. Førland 1998: Long-term trends in precipitation and temperature in the Norwegian Arctic: can they be explained by changes in atmospheric circulation patterns? Climate Research, 10, 143-153. Hanssen-Bauer, I. 1999: Downscaling of temperature and precipitation in Norway based upon multiple regression of the principal components of the SLP field. 1896-1997. DNMI-KLIMA report 21/99, 40 p, Norwegian Met. Inst., Box 43 Blindern, N-0313 Oslo ************************************************************************* Inger Hanssen-Bauer Telephone: +47 22963172 Senior Scientist Switchboard: +47 22963000 Norwegian Meteorological Institute Telefax: +47 22963050 P.O.Box 43 - Blindern, N-0313 Oslo E-mail: i.hanssen-bauer@dnmi.no ************************************************************************* ----- End of forwarded message from Inger Hanssen-Bauer ----- Hans von Storch Institute of Hydrophysics GKSS Research Center, Max-Planck-Strasse 1, PO Box, 21502 Geesthacht, Germany Phone: +49 (4152) 87-1831, Fax: +49 (4152) 87-2832 WWW: http://w3g.gkss.de/G/Mitarbeiter/storch/ e-mail: storch@gkss.de and storch@dkrz.de Home fax: +49 (4153) 582 522 FINALLY PUBLISHED: VON STORCH & ZWIERS, STATISTICAL ANALYSIS IN CLIMATE RESEARCH, CAMBRIDGE UNIVERSITY PRESS