Wednesday, May 23, 2012


cc: Peter Thorne <>, Ben Santer <>, Susan Solomon <>, John Lanzante <>, Melissa Free <>, Dian Seidel <>, Tom Wigley <>, Karl Taylor <>, Thomas R Karl <>, Carl Mears <>, "David C. Bader" <>, "'Francis W. Zwiers'" <>, Frank Wentz <>, Leopold Haimberger <>, "Michael C. MacCracken" <>, Phil Jones <>, Tim Osborn <>, Gavin Schmidt <>, "Hack, James J." <>
date: Fri, 18 Jan 2008 11:01:43 -0500
from: Steve Sherwood <>
subject: Re: The minus 1 (.png or .ps.gz)
to: Stephen Klein <>

Stephen et al.,

Stephen Klein wrote:
> Peter et al.,
> Thanks for this figure updating Santer et al. 2005. Although I am a
> bit rusty on this subject, there are two questions I have:
> 1) Does anybody have an explanation why there is a relative minimum
> (and some negative trends) between 500 and 700 hPa? No models with
> significant surface warming do this, and I can't think of a plausible
> physical reason for this. Do people feel the relative cooling of this
> layer relative to the surface is robust?
This feature is very dataset-dependent. It does not appear at all in my
homogenized radiosonde dataset (which is not currently in Peter's
figure), but appears very strongly in the Haimberger et al. datasets
(and, I believe, in the raw reanalysis product used to do the
homogenization for all three versions). It also appears somewhat in a
temperature reconstruction based on geostrophic winds that my student
has done, but only for the 1979-05 time period. I agree with you that
it would be bizarre, physically, on very long time scales but if you
look at individual model realizations you do find variability from one
run to the next that, over short (25 year) time periods, does show
warmings at different heights that deviate from a moist adiabat. I
suspect that the profiles are strictly adiabatic in regions of deep
convection but not quite so in tropical means because of decadal
variations in the stationary wave pattern and the poleward heat flux.
So it could be real but that wouldn't necessarily make it representative
of the longer-term trend.

> 2) Have you thought of producing a comparison of models to radiosonde
> observations for a longer term trend (late 1950s to 1999)? My
> recollection (from when I worked with John L. and Dian S.) is that the
> tropical radiosonde upper troposphere temperature trends were a bit
> stronger over the longer-term. Even if you add this to the paper, it
> might be nice to examine a figure of this type for the longer period
> to assess models, and perhaps comment on in the paper.
I agree that it is worth noting in the paper that the improved agreement
between radiosonde and anticipated trends when one reaches farther back
in time (already well documented) is consistent with the hypothesis that
there were problems with the data in the 1980's.


> Steve
> At 02:56 AM 1/17/2008, Peter Thorne wrote:
>> First hack at a caption:
>> Figure 6. Tropical trends from all available radiosonde products for
>> 1979-1999. Trends have been calculated by zonally averaging the gridded
>> data; applying a cos(lat) weighting to this zonal profile over 20N to
>> 20S to create a tropical mean timeseries; and then calculating a trend
>> from this timeseries using a median of pairwise slope estimator that is
>> robust to outliers (Lanzante, 1996). For RATPAC-A, which consists of a
>> much sparser network, the tropical mean timeseries available from their
>> website has been used to calculate the trend. Also shown are theoretical
>> expectations based upon the assumption that the tropical troposphere
>> behaves as a moist adiabat and climate model mean and 2 sigma estimates
>> as to what the true trend should be. Both these estimates are derived
>> from Figure 3 B of Santer et al. (2005), scaled by the available surface
>> estimates (taken from Santer et al., 2006). The implicit assumption is
>> made that these estimates adequately portray the real surface changes.
>> I'm away Friday and Monday so will look afresh based upon accrued
>> comments on Tuesday and make any mods then.
>> My thanks to Leo for all of his hard work to provide the RAOBCORE
>> products to me to calculate the trend values for this figure. And my
>> apologies that my temperamental computer served to delay this by 24
>> hours ... (don't ask!)
>> Peter
>> --
>> Peter Thorne Climate Research Scientist
>> Met Office Hadley Centre, FitzRoy Road, Exeter, EX1 3PB
>> tel. +44 1392 886552 fax +44 1392 885681

Steven Sherwood
Yale University ph: 203 432-3167
P. O. Box 208109 fax: 203 432-3134
New Haven, CT 06520-8109


No comments:

Post a Comment