To: email@example.com, firstname.lastname@example.org
Subject: Meeting on SRES Scenarios, 1 October 1998
Date: Thu, 01 Oct 1998 21:57:23 +0200
A meeting was held today on SRES scenarios during the IPCC plenary session
in Vienna. The meeting was organized by David Griggs, Fortunaat Joos,
Richard Moss, and Rob Swart. Also present were a number of delegates
including two Co-Chairs of IPCC, John Houghton from WGI and Bert Metz from
WGIII. Attached is a document with issues discussed during this meeting.
The meeting was very productive in my view, even though it was quite brief.
Two key issues were discussed that are listed in the attachment: (1)
incomplete information concerning SRES emissions as reported on the
website, and (2) consistency and plausibility of SRES scenarios and their
(1) Incomplete information
There appeared to be a general consensus that the range of CO2 emissions
(especially energy-related ones) are in quite good agreement across the
SRES scenarios once one adds the missing emissions categories to all model
runs. They are also in a relatively good agreement with the ranges given
The SRES ranges of CH4 and N2O emissions did not appear to be a problem in
themselves, but they are considerably lower than the ranges given in SAR.
It was agreed to ask the SRES writing team to further harmonize the ranges
for the base year and the period 1990 to 2000 across the scenarios for CO2,
CH4 and N2O. At the same time, David Griggs will contact the colleagues
from WGI to inquire whether the emissions ranges for these gases as given
in SAR have changed in the mean time and will inform the SRES colleagues
soon about the result. In particular, he will check whether the non-energy
CO2, CH4 and N2O emissions ranges are still appropriate as best guess for
the 1990 situation and about any new numbers about the ranges for more
recent years. It was also suggested that the SRES writing team discuss the
reasons for relatively low CH4 emissions in 1990 compared with the SAR range.
Most of the SRES models do not generate CFC and HFC emissions but these
emissions are important for climate models. It was agreed that David
Griggs will inquire with climate modelers whether they really need all
species of these gases or whether it is sufficient to report their joint
emissions. SRES team is to report whether these emissions could be added
to most of the model runs and over which time-scale. Joergen Fenhann is
in touch with a number of colleagues on this issue already and he is
planning to make a specific proposal how to handle this question across
SRES sulfur emissions are considerably lower than the IS92 range. There
are a number of reasons for this difference that were discussed at the
meeting. It was decided that this exchange should continue in the future
so that there is a better understanding of all issues involved. This is a
new aspect of SRES scenarios that represents an important change since
IS92a, a change that was also suggested by the 1994 IPCC review of
The concern raised by Hugh Pitcher (in the WGI scenario discussion group)
about high productivity growth in A1 scenarios was briefly mentioned. This
issue is to be settled within the SRES writing team, possibly by including
the formulation of alternative scenario variants.
(2) Consistency and Plausibility
Most participants of the meeting expressed the need to have emissions
trajectories that are somehow normalized for all SRES scenarios for 1990
and that have the same trends through 2000 and diverge only thereafter
across different scenarios. This would meet the need of climate modelers
to work with the same starting points for all scenarios they model. One
suggestion was that SRES team simply takes midpoints of emissions ranges in
1990 and renormalizes all SRES emissions. Another proposal is that climate
modelers suggest their preferred values for 1990 to be used in
renormalization. In any case, the method that is used would need to be
well documented and cited in the relevant IPCC reports. This is necessary
so as not to introduce an artificial impression that there is a full
agreement on base-year emissions across SRES scenarios.
There were no specific suggestions how to harmonize short-term emissions
through 2000. This issues is to be discussed within the SRES writing team
and within the climate modeling community in order to collect emissions
data for the last years that could be used for such harmonization.
The issue was discussed of generally lower CO2 and SO2 emissions across the
range of SRES scenarios and in particular for B2 marker. This results in
lower GHG forcing and lower "negative" SO2 forcing. The total forcing
remains roughly the same as in IS92a but has fundamentally different
implications especially at regional level.
Most of the climate models will be in the position to use just a few
scenarios, in some case, may be just two. Possible ways of avoiding the
impression that there is a "preferred" scenario were discussed and there
was a consensus that somehow the message needs to be conveyed that the
whole set of SRES scenarios is plausible and that there is really no single
"central" case that can be compared with IS92a.
Climate models need gridded SO2 emissions while SRES models generate SO2
emissions for a number world regions. Mike Schlesinger and Steve Smith
will attend the next SRES meeting and it was suggested that Mike would use
his method to produce gridded SO2 emissions and that Steve would use the
method proposed by Tom Wigley to do the same. This way there would be two
alternative gridded emissions patterns for all SRES scenarios available to
In conclusion, it was agreed that it would be useful to organize an
informal meeting where SRES colleagues could meet with potential user
groups from TAR (especially from WGI and WGII). Next possibility to do so
would be on the occasion of the WGI meeting in Paris, 30 November to 3
December. I am not quite sure that I got the dates right. The next
communication will be more precise.
Attachment Converted: "c:\eudora\attach\sres_w~1.rtf"