date: Tue Jun 15 10:13:48 1999 from: Tim Osborn subject: Overpeck et al. (1997) reconstruction to: jto@ngdc.noaa.gov Dear Dr. Overpeck I just had a quick question about your reconstruction of Arctic temperatures (Science, 1997). In the paper, the time series was plotted in sigma units (i.e., as standardised anomalies). On the NGDC website, your series appears as being in degrees C - see http://www.ngdc.noaa.gov/paleo/globalwarming/paleolast.html and http://www.ngdc.noaa.gov/paleo/globalwarming/overpeck.html - yet has the same or similar magnitude values. Has it really been calibrated? The reason I'm interested is that I've just seen a manuscript (Huang, Pollack & Shen) that overlays your series on some reconstructions from boreholes - they say both series are calibrated into degrees C. They match quite well, but I'm not sure if that's a fluke if your's is not calibrated. You may have seen the piece by Keith Briffa and myself in Science last month (vol 284, p926-927), where we calibrated your series. When calibrated, it reduced the magnitude of the values by a factor of about 0.5 - but we calibrated it against a record of summer temperatures from all land north of 20N (to enable comparison with other reconstructions), and a calibration against a more representative series (i.e., Arctic, annual) might yield values closer to the original. That's why I'm unsure whether the values on the website and used by Huang et al. are indeed calibrated, or whether it's an error. Best regards Tim