From	Ben Santer	Date	Fri, 22 Sep 2000 12:36:38 -0700
To	Tom Wigley, Erich Roeckner, ktaylor@zooks.llnl.gov, boyle@pcmdi.llnl.gov, sailes1@llnl.gov, Phil Jones, doutriau@pcmdi.llnl.gov, James Hansen, Jerry Meehl, bengtsson@dkrz.de
Subject	Status of our JGR paper
Dear All, I just wanted to keep you informed about the status of our draft JGR paper. First, thanks to all of you for your comments - they were very helpful. I am now in the process of revising the paper, and hope to have a new draft ready by Oct. 10th. After several discussions with Tom, I have decided to repeat the volcano/ENSO signal separation for the observed data and for the GSOP experiment. The reason for this is that there was a conceptual flaw in what I had done previously. The flaw related to the determination of the "pre-eruption" reference temperature, used as a baseline for estimating the maximum volcanically-induced cooling. Let's call this baseline temperature "TBASE". Previously, I was estimating TBASE for Pinatubo and El Chichon from either the raw or Gauss-filtered temperature data at time t=0 (the eruption month). If I was calculating TBASE from the filtered data, the estimate of TBASE was biased by "contamination" from post-eruption cooling. In other words, since I was using a 13-term Gaussian filter, temperature values from t=0 + 6 months were influencing TBASE, likely leading to an underestimate of the true TBASE value. I've now modified the program so that TBASE is not computed from the filtered data; instead, it is an average of the temperature anomalies in the MREF months prior to the eruption. There is some sensitivity to the choice of MREF (I've been experiment with values ranging from 6-18 months), which again underscores the uncertainties inherent in separating ENSO and volcanic signals. The maximum volcanically-induced cooling is still estimated using filtered data, but now I'm using a 5-term binomial filter rather than the 13-term Gaussian. These changes require repeating most of the analyses in the paper. Preliminary results indicate that the revised estimation of TBASE increases the ratio of the Chichon/Pinatubo maximum coolings, and brings this closer to the ratio of the Chichon/Pinatubo radiative forcings. Tom has also made a number of useful suggestions regarding reorganization and shortening of various sections of the manuscript. Hopefully the next iteration will be a little shorter than the current version of the paper! I will be out of my office next week, but should be back by October 2nd. With best regards, and thanks again for all your help, Ben -- ---------------------------------------------------------------------------- Benjamin D. Santer Program for Climate Model Diagnosis and Intercomparison Lawrence Livermore National Laboratory P.O. Box 808, Mail Stop L-264 Livermore, CA 94550, U.S.A. Tel: (925) 422-7638 FAX: (925) 422-7675 email: santer1@llnl.gov ----------------------------------------------------------------------------