cc: naki@uea.ac.uk
date: Wed, 05 Nov 1997 10:09:42 +0100
from: Arnulf GRUBLER <gruebler@iiasa.ac.at>
subject: Re: SRES minutes
to: Mike Hulme <m.hulme@uea.ac.uk>

Dear Mike

Thanks for your precision concerning the minutes. I'll take a note and
report back at our next meeting.

I personally had never the illusion that new GCM runs AND
new scenarios could be used for the TAR. Personally I also do not think
this is a major issue, as long as CONSISTENCY
can be maintained.  Your suggestion of reinterpreting
the two GCM (1 and 0.5 % increase) runs for impact
assessment with data, storylines, etc. of the the new IS98 scenarios is
indeed excellent. This would be also of great use for impact assessment in
general as only the new IS98 scenarios provide the kind of contextual and
quantitative data (GDP structure, etc.) impact assessment requires.

The only major open consistency issue as you correctly state is that of
sulfur emissions, where numbers will be very different (lower) in the new
IS98 scenarios, at least
in the "success" scenarios compared to IS92. You will have to educate me if
it is possible to rescale some of the
GCM runs with lower sulfur emissions. If not, well then one needs to be
explicit about this shortcoming for the TAR
analysis and start a.s.a.p. a quick analysis of how big the
difference would be in case IS98 sulfur emissions had been included.
Somewhat naively I assume that you guys must have
a couple of runs with LOW sulfur emissions available (e.g. based on the
IIASA-WEC A3 scenario, or some other similar one).

Concerning first quantifications by March. I can only speak for our group
here at IIASA, and it is feasible to provide quantifications for most of
the new scenarios/variants by March (at least for A1, B1, and B2; A2 with
it's "fragmentation" story is a difficult nut to crack with formal models
that generally prefer stylized general relationships amenable to
quantitative modeling). I do not know yet which ones will come close to the
1% and 0.5%
GCM scenarios, but considering the many variants and subscenarios, it is a
reasonable expectation that matching pairs can be identified.

We have agreed on a minimum reporting standard of 4 regions for IS98 at
RIVM. Of course finer geographical detail
will be made available from different models (our models e.g. use 11
regions, for IS98 we aggregate them to 4, 
as simply no time is left to redesign a 4-region model, or to analyze 11
regions in detail and report them). 

I note your interest in sulfur data. Data can be made available for 11
world regions. I cannot provide emissions
by altitude; what is feasible however, is to seperate large point sources
(with high stack heights) from diffuse sources
(e.g. household fuel use). If this is of interest, please let me know
a.s.a.p. so we can change it in the models. Please indicate also further
data requirements beyond sulfur where higher spatial resolution is
essential rather than simply of interest, so I can try to plan for
providing the maximum possible.

Concerning the data distribution center I strongly suggest you stay in
touch with Naki on this. I personally feel that
extremely useful synergies could be exploited if you would also host
directly the SRES site.

In any case I stronlgy urge you to consider joining the SRES group again,
so we can continue exchange on all of these issues.

Best regards, Arnulf.


                     Arnulf GRUBLER
 International Institute for | Email: gruebler@iiasa.ac.at
  Applied  Systems Analysis  | Phone: +43  2236 807 470
  A-2361 Laxenburg, Austria  |   Fax: +43  2236 71313

