This file describes corrections to the paper:

Abbot, D.S., C.C. Walker, and E. Tziperman (2009), Can a convective
cloud feedback help to eliminate winter and spring sea ice at high CO2
concentrations? Journal of Climate, 22(21), 5719-5731, DOI:
10.1175/2009JCLI2854.1.


(1) For the CMIP project IPCC AR4 4XCO2 simulations some modeling
groups quadrupled CO2 starting at 280 ppm (including GFDL) while
others started at 355 ppm (including NCAR). In a certain sense results
from these simulations are comparable across different models because
a quadrupling of CO2 represents roughly the same increase in radiative
regardless of the initial CO2 value. The final CO2 at the end of the
experiment, however, is clearly different in the two cases. We did not
appreciate this when we wrote the paper, and the paper reflects this
misunderstanding.

(2) I made some transcription errors when copying climate
sensitivities for the AR4 models from this table:

http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch8s8-6-2-3.html#table-8-2

When I correct these errors, r^2=0.03 for Fig. 8a and r^2=0.1 for
Fig. 8b, so the correlation between March sea ice at the end of the
runs and climate sensitivities is still very low. Notice that Fig. 8c
was not affected by this error.


I thank Ian Eisenman and Kyle Amour for pointing out these errors to me.

(dsa, 6/2/11)