date: Mon, 10 Jul 2006 03:52:43 -0400 from: "W.R. Peltier" subject: Re: [Wg1-ar4-ch06] IPCC Figure 6.6 - Last Interglacial to: ottobli@ucar.edu, wg1-ar4-ch06@joss.ucar.edu Hi Bette, Of course the left panel should be kept as a multi-model average. It would appear to be the best way to emphasize commonality and thereby to minimize divergence. Cheers Dick At 02:44 PM 07/07/2006, ottobli@ucar.edu wrote: >Dear All, > >There are several issues to discuss and resolve concerning the last >interglacial figure (6.6) in our chapter. > >Right panel: This panel shows the Greenland Ice Sheet (GIS) minimal extent >and ice thickness for the last interglacial. It is an average of the >minimal configurations of the GIS from three published results - Tarasov >and Peltier (2003), Lhomme et al. (2005), and Otto-Bliesner et al. (2006). >The colored dots represent an assessment of ice core observations on >whether ice disappeared at these ice core sites during some point in the >LIG, which I would like your views on: > >* White - Ice remained through LIG: N(NGRIP), S(Summit-GRIP and GISP2) >R(Renland) >* Black - Ice disappeared during some time in the LIG: A(Agassiz), De(Devon) >* Gray - Status of LIG ice at these sites is unresolved: C(Camp Century)?, >D(Dye3)? > >Any additional references that I should include in the figure legend would >also be useful. > >Left panel: This panel shows the summer (JJA) surface temperature change >from 2 proxy compilations and an average of 2 model simulations. The data >represents proxy estimates of peak summer warmth. Susan wanted the model >panel to be an average of results from more than one model. The two >simulations available for this average are CCSM, 130ka minus >present(1990), published in Otto-Bliesner et al. (2006); and ECHO-G, 125ka >minus preindustrial, published in Kaspar et al. (2005) [except using the >preindustrial simulation from the IPCC database rather than the >preindustrial simulation used in Kaspar et al. because of a problem with >snow buildup and very cold temperatures over Greenland in the Kaspar et >al. >preindustrial simulation]. We had rationalized at the time of the SOD that >these two modeling group results are roughly comparable for computing >Arctic summer surface temperature anomalies based on the following forcing >effects: > > CCSM ECHO-G > --------------------- --------------------- > 130ka 1990 R.F. 125ka PI R.F. >CO2 280 355 -1.27 270 280 -0.19 >CH4 600 1714 -0.53 630 700 -0.05 >N2O Pres Pres 0 260 265 ~0 > >del Solar (incoming divided by 4 times 0.7) >69N,MJJ +8.12 +6.95 >69N,JJA +1.25 +4.88 > >I do not really like averaging these two modeling results although we can >argue that this is somewhat justified based on the comparable GHG+Solar >radiative forcings for Arctic May-Jun-Jul (but not so for Jun-Jul-Aug). >Notice also the teardrop pattern of temperature anomalies in northern >Greenland, which are a feature of the ECHO-G differences. The results from >CCSM alone can be seen in the left panel of Figure TS-24 which is not yet >the multi-model figure. Averaging the two models also makes answering >comment 6-1060 problematic. Should we keep the left panel as a multi-model >average? > >Bette > > > > > > > > > >_______________________________________________ >Wg1-ar4-ch06 mailing list >Wg1-ar4-ch06@joss.ucar.edu >http://lists.joss.ucar.edu/mailman/listinfo/wg1-ar4-ch06 _______________________________________________ Wg1-ar4-ch06 mailing list Wg1-ar4-ch06@joss.ucar.edu http://lists.joss.ucar.edu/mailman/listinfo/wg1-ar4-ch06