date: Wed, 28 Jun 2006 17:01:18 +0100
from: Tim Osborn <t.osbornatXYZxyz.ac.uk>
subject: Re: Simulation data, missing 1900 trend?
to: Alex Wright <alex.wrightatXYZxyzw.vu.nl>, Orson van de PLassche <Orson.van.de.Plassche@falw.vu.nl>
As far as I know the data are correct.
The sea level patterns came from Jason Lowe, and I guess Jonathan
Gregory was also involved in diagnosing them from the HadCM3 model.
But the extraction/manipulation of the data is all my doing, so blame
me not them if I've done something wrong. However I have followed
instructions from Jason in doing this, so hopefully I have done it ok.
I have included the global thermal expansion and global glacier melt
changes in the regional/local data that I sent you.
I have computed zonal means of sea level change from 'nat' and 'all',
and plotted these (and also plotted 'all' minus 'nat' to isolate the
anthropogenic component, 'anthro'). PDFs of these three are
attached. All changes are expressed as anomalies from the 1500-1549
mean of the 'nat' run (with 'all' adjusted to have the same local
means as 'nat' over the 1750-1799 period).
You will see that in the 'all' zonal means, there is a clear and
strong rise in sea level between 60S and 50N. Little increase is
simulated south of 60S (and I asked Jason why there is data at the
South Pole, but he didn't reply... of course we should just ignore
However, you will see that there is little overall trend between 55N
and 80N, which of course if where most of the data I sent you have
come from! It is higher than the 1500-1549 mean after 1980, but no
higher than the 1850-1900 period, which is why there appears to be
little sea level rise.
Whether the multi-decadal large variations in sea level in these
northern mid to high latitudes are caused by regional oceanic warming
and cooling due to either (i) regional sulphate aerosol cooling; or
(ii) internal variations of the ocean circulation; or (iii) errors in
model or data extraction, I cannot tell. I think probably not (iii)
[thought Jason I did raise some queries about the global-mean of the
pattern files showing larger variations than I expected, which you
were going to check when you had time]. The 'nat' zonal means show
multi-decadal variations in this region, whether due to internal
ocean circulation changes or volcanic/solar forcing I don't know (I'd
guess at the former).
So it is probably ok. And the time series I extracted for you are
consistent with the zonal mean of the data.
Such is the difficulty of interpreting *regional* sea level changes I
guess. I look forward to hearing how you interpret the comparison of
these changes with your palaeo and tide gauge series!
At 14:11 28/06/2006, you wrote:
>Have you looked at the simulation data at all?
>It is almost as if the sea-level trend from 1900 is missing from
>these regional simulations, when compared to the global mean sea
>I guess I need to know if the glacier melt model is included etc...
>Would you happen to know who submitted these runs to the database?
>I think I should check with them first before spending days (which I
>don't have) assimilating the data
Attachment Converted: "c:\eudora\attach\slzon_all.pdf"
Attachment Converted: "c:\eudora\attach\slzon_anthro.pdf"
Attachment Converted: "c:\eudora\attach\slzon_nat.pdf"
Dr Timothy J Osborn, Academic Fellow
Climatic Research Unit
School of Environmental Sciences, University of East Anglia
Norwich NR4 7TJ, UK
**Norwich -- City for Science:
**Hosting the BA Festival 2-9 September 2006