cc: jean-claude.duplessy@cfr.cnrs-gif.fr, ll@dmi.min.dk
date: Thu, 21 Nov 1996 17:46:17 +0001
from: Gerbrand Komen <Gerbrand.Komen@knmi.nl>
subject: past climate data, Abisko, 1 - 3 September 1997
to: bengtsson@dkrz.de, p.jones@uea.ac.uk

Dear Lennart, Phil

Below you find the next iteration in our workshop description.

I suppose all have a chance to react by e-mail.
Anyway,  Lennart and I agreed that

- Lennart would seek formal commitment from the organizing 
  comittee members in the next few days

- we would send out the invitation to participants in early
  December

Best regards,

Gerbrand Komen



                       Proposal for a Euroclivar workshop on


                                        Past Climate Data



Rationale
Climate modellers concerned with understanding the physics
behind low frequency variability and predictability of climate
and for the detection of anthropogenic climate change are in a
desperate need of better observational data. The situation is
serious. Global 3-dimensional data for the atmosphere exist
essentially only available for the last 18 years, for the
oceans they are mainly restricted to the upper ocean only. Data
for other aspects of the climate system; deep ocean,
atmospheric chemistry, biochemical cycles, land ices are still
only patchy. Surface data for temperature and pressure can at
best be broadly reconstructed (essentially populated areas
only) until the middle of the last century. Before that we have
mainly only indirect observations, which by different
techniques can be interpreted into geophysical quantities. We
have now wonderful new potential possibilities to use models to
bring all these sparse data together in a systematic way and
the modellers and the Paleo "detectives" must now work closely
together.

Two actions are required:

       1.    To establish a comprehensive global climate observing
             system. (At least we must provide data for future
             generations)

       2.    To provide the best possible available data set of
             past data.

The first of these two actions will be handled by GCOS. This
work is well in hand and a special international committee
(WMO, ICSU and UNEP) is presently developing such a concept.
Euroclivar wants to stimulate work on - and with - past data.
To this end a 3-day workshop will be organized as laid out
below.

1. Long term objective
To provide the best possible climate data of the past and to
use these for the validation of climate models and an
assessment of (natural) climate change.

2. Meeting products

*      A proposal for making (by whom?) data (what data?) better
       available (how?)

*      A plan of action for the validation of climate models and
       an assessment of (natural) climate change (to the extend
       possible).

3. Types of past data
Climate data can conveniently be divided in four different
groups; (i) 3-dim atmospheric global data 1979-1995, level 2
and 3, (ii) 3-dim Northern Hemisphere 1950-1995, level 2 and 3,
(iii) Surface data 1750-1995 ("200+"), level 2, (iv) Paleo-data
for the last 2000 years, level 1 and 2.

[level 1 = indirect information; level 2 = geophysical
parameters; level 3 = gridded data]

Data set (i) is expected to be provided by the ECMWF
re-analysis (ERA) project. The Euroclivar workshop 
will concentrate on (ii), (iii) and (iv). 

Assimilation of (mainly) Northern Hemispheric upper air data
1950(47?)-1995(98) is possible to achieve within the next
two-three years. NMC has made a first start, but the result
will be model dependent so at least another data set is needed.
ECMWF is a perfect candidate, they have the technical expertise
and they can also work reasonably well with external groups.
The assimilation should of course be global and it will be
interesting to see how well we can reconstruct/constrain the
circulation in the tropics and for the Southern Hemisphere.

For instrumental data (the last 200 years or so  - principally
surface) a distinction is made between 

       *     monthly data. These are needed for detection and
             validation. Much is available already. There are
             temperature and precipitation data sets available on
             a 5 by 5 grid-box basis. 

       *     data on a daily basis. At the moment it is not clear
              what is/might be available from (eg) ECSN.

For Proxy (Paleoclimatic) data distinction is made between

       *     Paleoclimatic time series  (Year by year
             reconstructions of for example summer temperature or
             summer precipitation totals or possibly annual/wate-
             ryear values). The key aspect here is time series
             available for a number of locations from around the
             world from trees, corals, ice cores and varves. 

       *     Spatial patterns of change for periods in the paleo
             past (ie  6K , 9K  and 18K years ago). This sort of
             data is needed for the PMIP  intercomparisons, 
             which can be thought of as
             validation - but only of spatial patterns. For
             detection/assessment the time series aspects are the
             ones that are most relevant.

4. Format of the meeting
Three day meeting

First Day:          

       *     Presentations (key) of instrumental and different
             kinds of proxy data.
             
       *     Availability of the data  - formats etc for inclusion
             in a data base.
             
       *     What sort of analyses can be done between these data
             and model output with respect to validation/detection

Second Day (Break into small groups)

       *     Reanalysis
       *     Last 200 years
       *     Modern data where daily timescale available
       *     Proxy - the last 2000 years
       *     Spatial proxy - PMIP uses

Third Day:          Report Writing / Final Presentations

5. Organizing committee
Bengtsson, Duplessy, Laursen, Jones, ECMWF (via Burridge), ECSN
(via Orfila) Wibjoern Karlen (Stockholm)
Organizing committee members should be willing to prepare
strawman texts, e.g.
       re-reanalysis (ECMWF)
       3d, 1950 - now (Bengtsson)
       200+  (Jones)
       paleo (Duplessy)
       validation of climate models and an assessment of
       (natural) climate change (statistician + modeller [Simon
       Tett (HC, UKMO) and/or Gabi Hegerl (MPI, Hamburg]).

6. Place and venue: Abisko, Sweden, 1 - 3 September 1997

7. Participants
Invited, 10 - 15 people: 

Bengtsson, Duplessy, Jones, Karlen, Laursen, ECMWF, ECSN, ICPO

Further

1) Input from currently funded EC projects
      
       von Storch (Millennia)
       Jouzel (Ice core project - Antarctica  + Claus Hammer re
       GRIP)
       European Coral worker (Patzold, Bremen)
       Other Instrumental people  (eg Folland/Parker HC UKMO,
       Brazil, CZ Rep, Povl Frich from DMI re the NACD efforts in
       N.Europe)

2) The following non-Europeans: Francis Zwiers (Canada),
Raymond Bradley (Dept Geosciences, Un. Mass.) and Tom Karl
(NCDC Asheville,USA).



    



	Gerbrand Komen			
	KNMI, P.O. Box 201		phone	:	+ 31 30 2206 676
	3730 AE De Bilt			fax	:	+ 31 30 2210 407
	The Netherlands			email	:	komen@knmi.nl
        http://www.knmi.nl/KLIMAAT/OCEAAN 
        (see also http://www.knmi.nl/euroclivar )
