cc: Dith Stone <stoned@atm.ox.ac.uk>, Phil Jones <p.jones@uea.ac.uk>, Peter Stott <stott.peter@googlemail.com>, Toru Nozawa <nozawa@nies.go.jp>, Alexey Karpechko <A.Karpechko@uea.ac.uk>, Michael Wehner <MFWehner@lbl.gov>
date: Fri, 30 May 2008 16:40:37 +0100
from: Gabi Hegerl <gabi.hegerl@ed.ac.uk>
subject: Re: Revised polar attribution paper
to: Nathan Gillett <n.gillett@uea.ac.uk>

<x-flowed>
Hi all, nature geoscience seems like a good choice to me, and the paper 
reads very well.
Similar to Alexey I am a bit ambiguous about showing results with SAM 
subtracted only,
given it makes it a little bit harder to interpret. On the other hand, 
this is not an issue if
discussed that results are not sensitive to subtracting the SAM, but 
that this addresses
the concern if SAM related trends are wellr epresented in the models
Also, the SAM subtracted pattern is
very impressive and could be emphasized a bit more.
I am also wondering why Arctic results are so jiggly with detection 
completely lost
for a number of cases with lots of EOFs - do we have any idea why this 
is happening?

writing suggestions:
Abstract could mention that temperature trends over Antarctica have been 
reported as mixed, raising the question
if Antarctica is not warming as predicted which would have huge 
importance on longterm sea level rise prospects. or so
(is this too speculative?)

p. 2 middle, should this explicitly mention that West Antarctica is 
still not covered due to hostile conditions?

p. 3 first para: 5-yr average spatial means to avoid two means in short 
sequence?
bottom 2nd line same page, exclusion of these data??

p 4 discussion of simulations: the anthro could be easily (kind of) be 
extended with scenarios, right, its the natural that
give us trouble after 1999?

p 5 top, add? 'residual observed trends show warming everywhere 
except..pole. THis makes them more similar to the model residual trends
than the raw trends (see supplementary figure 1).'
I am wondering if there is any way to quantify this, for example, by 
plotting your sectorial mean data model and obs raw and subtracted as
a line plot, or by plotting zonal means with the peninsula separated as 
an extra point? This would also possibly strengthen the case...

on p 6, bottom paragraph, I would also add at the end of the ANtarctica 
discussion that
...available and that circulation changes, which were largely 
antrhopgenic (citeyourpaper) have reduced warming rates over parts of 
Antarctica in models and observations
in the past' does ch10 say something about futur, I think consistent 
warming is expected in future? (the time intensive approach would be to 
get the
data for the future anthro runs and plot a bit into the future as well 
which might make this paper hotter for nature but I admit it does sound 
like work....)

Gabi


Nathan Gillett wrote:
> Hi all,
> Please find attached a revised version of the polar attribution paper,
> and supplementary info. In response to your earlier comments and
> suggestions, I've made the following changes:
>
> - I now subsample the monthly model data at observed locations, rather
> than doing the sampling on 5-yr means.
> - I mask the model data with its land mask, before applying the
> analysis, so that I'm comparing observed land temperatures with
> simulated land temperatures (except in the case of cells in which the
> model has no land, but observations exist, in which case I take the
> mean over the whole cell).
> - I've applied the D&A analysis to the SAM-residual temperature
> changes (see supplementary info).
> - I haven't applied the analysis to data to the present, because the
> data doesn't seem to be available yet.
>
> Subsampling the monthly model data makes the attribution analysis
> slightly less robust to variations in truncation, but we can still
> detect ANT. Applying the analysis to the SAM residual makes it easier
> to detect ANT in the Antarctic.
>
> I'm inclined to submit this to Nature Geosciences, partly since it's
> new and makes a change from GRL, and partly since I won't have to
> reformat it to submit there... Let me know if you disagree. Comments
> welcome - it would be good to get this resubmitted soon.
>
> Cheers,
>
> Nathan
>
>
>   


-- 
Dr Gabriele Hegerl 
School of GeoSciences
The University of Edinburgh
Grant Institute, The King's Buildings
West Mains Road
EDINBURGH EH9 3JW 
Phone: +44 (0) 131 6519092, FAX: +44 (0) 131 668 3184
Email: Gabi.Hegerl@ed.ac.uk 


The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

</x-flowed>
