28th EWGLAM and 13th SRNWP Meetings

9-12 October 2006, Zurich Switzerland, Local organizer: Swiss Federal Office of Meteorology and Climatology

See all meeting photographs


Jean-Marie Bettems, Gergely Bölöni, Massimo Bonavita, Mike Bush, Gerard Cats, Jean-Pierre Chalon, Guy de Morsier, Terry Davies, Maria Derkova, Pierre Eckert, Claude Fischer, Richard Forbes, Carl Fortelius, Jean-François Geleyn, Stefan Gollvik, Nils Gustafsson, James Hamilton, Met Éireann, Andras Horanyi, Mariano Hortal, Tamara Ivanova, Stjepan Ivatek-Sahdan, Trond Iversen, Dijana Klaric, Daniel Leuenberger, Bruce Macpherson, Aarne Männik, Chiara Marsigli, Jan Masek, Jean-Antoine, Maziejewski, Lars Meuller, Dmitrii Mironov, Deutscher Wetterdienst, Maria José Monteiro, Mark Naylor, Jeanette Onvlee, Bartolome Orfila, Tiziana Paccagnella, Ioannis Papageorgiou, Jean-Marcel Piriou, Patricia Pottier, Neva Pristov, Jean Quiby, Bent Sass, Francis Schubiger, Jan-Peter Schulz, Harald Seidl, Cornel Soci, Philippe Steiner, Jürgen Steppeler, Sander Tijm, Martina Tudor, Per Unden, Filip Vana, Josette Vanderborght, André Walser and Xiaohua Yang.

  1. Agenda and presentations
  2. Report of the SRNWP session
  3. Minutes of the EWGLAM final discussion
  4. List of the operational models in Europe in 2006
  5. the meeting NewsLetter

1. Agenda and presentations

Monday, 9th October 2006
9:00 Opening by the Director of MeteoSwiss, D. Keuerleber and organisational matters
09:15-11:30 Presentation of the Consortia and ECMWF, chair: Marco Arpagaus
09:15-09:30 Tiziana Paccagnella COSMO
09:30-09:45 Mike Bush UK Met Office
09:45-10:00 Jeanette Onvlee HIRLAM
10:00-10:15 Jean-François Geleyn ALADIN
10:15-10:30 Dijana Klaric LACE
10:30-11:00 Coffee break
11:00-11:30 Mariano Hortal ECMWF
11:30-14:45 Presentations of national posters, chair: Per Unden
Harald Seidl Austria
Josette Vanderborght Belgium
Stjepan Ivatek-Sahdan Croatia
Filip Vana Czech Republic
Bent Sass Denmark
Aarne Männik Estonia
Carl Fortelius Finland
Claude Fischer France
Jan-Peter Schulz Germany
Andras Horanyi Hungary
12:30-13:45Lunch break
13:34 James Hamilton Ireland
Massimo Bonavita Italy
Gerard Cats Netherlands
Maria José Monteiro Portugal
Cornel Soci Romania
Maria Derkova Slovakia
Neva Pristov Slovenia
Bartolome Orfila Spain
Lars Meuller Sweden
Philippe Steiner Switzerland
Mike Bush United Kingdom
14:45-15.15Poster session
15:15-15:45Coffee break
15:45-17:45 Scientific presentations on numerics and coupling numerics/physics, Chair: Claude Fischer
15:45-16:15 Terry Davies Lateral boundary conditions, variable resolution or both?
16:15-16:45 Filip Vana general Dynamics status in ALADIN and LACE
16:45-17:00 Jürgen Steppeler Overview and strategy on numerics. The LMZ_ including coupling Numerics & Physics
17:00-17:30 Jan Masek Idealised test of ALADIN NH dynamical core at very high resolution (and comparison of hydro­static and NH versions)
17:30-17:45 All Discussion
18:30City tour (start at hotel Basilea) with optional dinner
Tuesday, 10th October
08:30-11:50 Scientific presentations on data assimilation, Chair: Terry Davies
08:30-09:00 Nils Gustafsson Recent results and plans for HIRLAM 4D-VAR
09:00-09:15 Jürgen Steppeler Overview and strategy on data assimilation for LM
09:15-09:45 Bruce Macpherson Assimilation developments in North Atlantic and UK models
09:45-10:00 Mark Naylor Aspects of 4DVAR in North Atlantic model
10:00-10:30Coffee break
10:30-10:45 Daniel Leuenberger The LHN scheme of COSMO
10:45-11:15 Claude Fischer & Andras Horanyi Data assimilation - ALADIN
11:15-11:35 Gergely Bölöni Data assimilation-LACE specific
11:35-11:50 All Discussion
11:50-13:30 Lunch break with optional visit of the forecast service
13:30-17:30 Scientific presentations on Physics, Chair: Tiziana Paccagnella
13:30-14:00 Jean-Marcel Piriou Parametrisation in ALADIN
14:00-14:15 Marco Arpagaus Overview and strategy on LM Physics
14:15-14:30 Dmitrii Mironov The UTCS Project
14:30-15:00 Mike Bush Physics improvements in the North Atlantic Model
15:00-15:30Coffee break
15:30-16:00 Richard Forbes Recent results from parametrization develop­ments in the convective-scale (~1 km) Unified Model
16:00-16:30 Sander Tijm Physics developments in HIRLAM-A
16:30-16:55 Stefan Gollvik Snow, forest and lake aspects in the HIRLAM surface scheme
16:55-17:15 Neva Pristov ALARO part developed at LACE
17:15-17:30 All Discussion
18:00 Apero offered by MeteoSwiss on the occasion of its 125th anniversary
Wednesday, 11th October
08:30-11:00 Scientific presentations on predictability and EPS, Chair: Andras Horanyi
08:30-09:00 Harald Seidl LAMEPS Development and Plan of ALADIN­LACE
09:00-09:20 Trond Iversen A proposal for Grand Limited Area Model Ensem­ble Prediction System (GLAMEPS)
09:20-09:45 Bartolome Orfila Recent experiences with the INM multi-model EPS scheme
09:45-10:00 Chiara Marsigli The COSMO SREPS project
10:00-10:30Coffee break
10:30-10:45 Pierre Eckert COSMO ensemble systems
10:45-11:00 All Discussion
11:00-11:30 EWGLAM Final Discussion, Chair: Stjepan Ivatek-Sahdan
Preparation of the newsletters -Date and place of the next meeting (2007)
Place of the meeting 2008 (tentative)
11:30-12:00 SRNWP business meeting Jean Quiby
12:30Excursion and offered dinner
Thursday, 12th October
09:00-13:00 Discussion of the 3rd phase (from 1st January 2008) of the SRNWP Programme, Chair: Jean Quiby
Introduction to the NWP vision -draft Programme Proposal and Programme Decision
Procedure for the elaboration of a larger SRNWP scientific project
10:30-11:00Coffee break
11:00-13:00 Other proposals identified ad the Vision Workshop:
Common format for the product exchange
NWP requirements for next phase of EUMETNET obs pgm
Procedure for the prioritization of the proposals from the Vision Workshop
Work plan/timeline for programme proposal until spring Council
13:00End of the meetings

2. Minutes of the Final Discussion

For the minutes: Jean Quiby


It has been decided to produce a Newsletter under electronic form that everybody will be able to download and print if he or she wishes to do so.

The different contributions are limited to the following number of pages:

Contributions have to be sent either in MS Word or in Latex to 'ewglam2006@meteoswiss.ch'till the 30th of November 2006. If your contribution is larger than 4 MB, it must be transfered by ftp to 'ftp.meteoswiss.ch' (log in with user 'ewglam2006' and password 'ewglam4mch') and MeteoSwiss must be informed by e-mail at the address 'ewglam2006@meteoswiss.ch'with indication of the name of your file.

SRNWP Workshops

The Programme Manager showed the list of the SRNWP Workshops for the rest of 2006 and for 2007. It has been necessary to move the 'Third Workshop on Short-range EPS' from March 2007 to the autumn 2007.

Next SRNWP Meetings

At the invitation of the NMS of Croatia, the 29th EWGLAM Meeting together with the 14th SRNWP Meeting will take place in Dubrovnik the 9 - 12 October 2007 following the format defined at the 27th EWGLAM Meeting and applied with great succes at the Zurich Meeting.

The Representative of the INM has invited the Assembly to hold its 30th EWGLAM Meeting and its 15th SRNWP Meeting in Spain in 2007.

3. Lists of Decisions and Minutes

For the report: Jean Quiby, SRNWP Programme Manager

Decisions concerning the choice of the SRNWP Projects

1. Three topics have been selected as SRNWP Projects that should accompany the SRNWP Programme, with a dedicated financing for each Project. They are: Interoperability, Model Verification and European multi-model LAM EPS

2. For each Project, a 'Redaction Committee'will be formed for the preparation of the corresponding Project Proposal. Each Consortium communicates to the Programme Manager till the end October 2006 the name of its representative for each of the 3 Redaction Committees.

3. When the names of the Consortium representatives are known, the Programme Manager will contact the members of each Redaction Committee for the organisation of the work.

Decisions concerning the new governance

1. The Lead Centres are dissolved and replaced by Working Groups (WG) centred on NWP topics.

2. The SRNWP Workshops are maintained and their organisation goes under the responsibility of the WG. Each WG will have the responsibility to organise workshops on the field of its competence when it judges it appropriate. The rule 'a workshop every two years' is abandoned.

3. The EWGLAM Meetings remain annual. They follow the format defined at the 27th EWGLAM Meeting (2005 in Ljubljana). The idea to organise together with Academia 'European Conferences on dynamical meteorology and NWP' is abandoned.

4. A 'SRNWP Advisory Committee' will be created to better represent the Consortia in the Programme and to support the Programme Manager in his work.

5. A 'Programme Redaction Committee' will be formed for the preparation of the Programme Proposal. Each Consortium communicates to the Programme Manager till the end October 2006 the name of its 2 representatives.

6. When the names of the Consortium representatives are known, the Programme Manager will contact the members of the Programme Redaction Committee for the organisation of the work.

Minutes of the Meeting


56 participants from 25 European NMS. The EUMETNET CO, Jean-Pierre Chalon, participated to the meeting and explained how the financing of Programmes and Projects works by EUMETNET.

Aim of the Meeting

The aim of the Meeting was to take important decisions concerning the goals and the governance of the third phase of the EUMETNET SRNWP Programme starting January the 1st, 2008.

Goals for the 3rd phase of the Programme: Projects chosen

New for the third phase of the Programme will be the submission to Council of Projects on precise topics which should accompany the Programme with a dedicated financing for each Project.

The comments written and circulated as reactions to the draft Programme Proposal distributed by the Programme Manager contained a wealth of proposals where the Programme should be active.

All these proposals have been summarized by the Programme Manager in its presentation.

The proposals received could be classified in 5 chapters:

  1. Enhancement of the operational cooperation

  2. Enhancement of the scientific cooperation

  3. European multi-model LAM EPS

  4. Diffusion of the NWP knowledge

  5. Training and education activities in NWP

Enhancement of the operational cooperation

Numerous proposals have been made under this heading in the comments received. During the meeting it became immediately clear that the priority must be given to the interoperability.

Under the general term 'interoperability' we must understand primarily the definition or the harmonization of standards for the input and output of the:

Interoperability was also the most prominent theme identified for collaboration at the First Vision Workshop (15-17 March 2006). This underlines the absolute necessity to become active in this field if we want to significantly increase collaboration.

We should also add to our choice 'interoperability' the proposals made under the same theme at the Vision Workshop:

Enhancement of the scientific cooperation

Among the several propositions made, common model verification and model comparison stood in the fore front.

The wish to compare our main LAM models (HIRLAM, ALADIN, Local Model and Unified Model) has been already expressed by our Directors in the ICWED Meetings as well as in EUMETNET Council Meetings. But this was considered in the past as a ticklish issue, although it was accepted that model comparison could bring much for model improvement.

The Vision Workshop has also recognized the importance of model verification and has proposed to 'Initiate a project to enable us to work closer on verification of models and user benefits'.

European multi-model LAM EPS

As for the two propositions above, the Assembly has chosen as third proposition a theme which has also been selected at the Vision Workshop: European multi model LAM EPS.

It is hoped that the result of this Project will be the European contribution to TIGGE-LAM, a future component of the THORPEX initiative.

Diffusion of the NWP knowledge

All the comments received as well as the remarks made during the meeting stipulated that this activity - which is the main activity of the present programme - should continue.

The wish has been expressed that the liaison role of the Programme should be reinforced - particularly with the ECMWF - and extended to the climate community. But it has been decided not to make a dedicated Project of this activity as it will remain the basic activity of the NWP Programme.

Training and Education Activities in NWP

It has been considered important that the SRNWP Programme becomes active in this field: activity in training and education was proposed in all the comments and has been stressed again during the meeting. Several ideas have been suggested, for example the organisation of summer schools for the NMS and Academia.

Although considered as important, it has been decided not to make a dedicated EUMETNET Project of this activity. But inside SRNWP, a small WG should be formed to deal with this topic.

Submission of the Project Proposals

For each of the projects: interoperability, common model, verification, model comparison & European multi model LAM EPS, a Project Proposal stipulating among other things:

has to be written down and submitted to the EUMETNET Council.

The three Project Proposals must be sent to the EUMETNET Secretariat end of April 2007 at the latest if the Projects, after acceptance by Council, are going to start at the beginning of 2008.

It has been decided at the meeting to form a small 'Redaction Committee' for each Project.

Constitution of the Redaction Committees for the Projects:

Number of persons:

At the meeting, we agreed on 2 persons per Consortium. We have 5 Consortia. This would give 10 persons plus the help of the Programme Manager for each Redaction Committee.

As there is a steadily growing number of meetings and workshops where Consortia have to be represented and as the travelling budgets are in the NMS everywhere very tight, I think that one person per Consortium is sufficient. This would give for each of the 3 Projects a group of 5 persons for the formulation of the Proposal, with the Programme Manager helping to format the Proposals according to the EUMETNET practice.

Choice of the persons (opinions of the Programme Manager)

Although not discussed at the meeting, allow me the following remarks, knowing that the choice of the Consortium Representatives is of the sole responsibility of the Consortia.

Redaction Committee for Interoperability

In my view, the persons responsible of the maintenance and operation of the model reference versions would be very suitable for this work.

Redaction Committee for Common model verification

As we hope to use common verification schemes not only for biases and rmse, very experience scientists in verification with an interest for new and unconventional methods should be chosen.

Redaction Committee for the European multi-model LAM EPS

Most of the LAM EPS specialists will participate at the ALADIN-HIRLAM LAM EPS Workshop of Vienna (13-14 Nov. 2006). At this meeting, we should discuss the European participation to TIGGE-LAM. It would be efficient if the 'redactors' could be chosen among the participants to this meeting: they would have the latest information and we could organise the first meeting of this Redaction Committee in Vienna after the ALADIN-HIRLAM Workshop.

The names of the Consortium Representatives have to be communicated to the SRNWP Programme Manager till the 31st of October 2006.

After, reception of the names, the Programme Manager will liaise with the 3 groups in order to organise the work.

It has been said at the meeting that a first report will be provided at the end the year.

New Governance for the 3rd phase of the SRNWP Programme

At the meeting, we discussed the future role of the: Lead Centres, SRNWP Workshops, EWGLAM Meetings & Advisory Committee

Replacement of the Lead Centres (LC) by Working Groups (WG)

Three functions or duties have been defined for the Lead Centres when they have been created in 1999:

1. Organise every two years a workshop on their topic of responsibility. This function has been entirely fulfilled.

2. Inform regularly the other NMS on the state of the art in their topic of responsibility. This function has been only partially fulfilled.

It was foreseen that the LC would report at the annual EWGLAM/SRNWP meetings. But because of agenda constraints, only 15 minutes could generally be put at the disposal of a LC for its report. These reports were very divers, from very short to comprehensive. Many LC have simply reported on their activity of workshop organisers.

A thorough fulfilment of such a function necessitates resources: it is necessary to spend time studying the literature as well as to participate to conferences and workshops also outside Europe. This function is not at all supported financially by the Programme.

3. Act as consultant for the NMS in their topic of responsibility: This function has never been activated.

The reason for this is that in the last 6-8 years, the collaboration between the NMS inside the Consortia has steadily increased. Thus the interlocutor that a NMS would primarily consult for a scientific problem was, and still is, its Consortium.

In view of the experiences made in the past and considering that it will not be possible for EUMETNET to support financially the NMS hosting a Lead Centre, the Assembly decided to replace the Lead Centres by Working Groups centred on NWP topics.

Unfortunately, we had no time at the meeting to define the number of Working Groups we would like to have and the fields they should cover.

In order to save time and be ready for January 2008, we suggest mentioning in the Programme Proposal only a limited number of WG, as it will always be possible later for the Advisory Committee to modify this choice. The 'Programme Redaction Committee' will make a first choice mirroring the topics of the Consortia Working Groups or Project Teams.

Functions of the Working Groups

Existence of Working Groups made of specialists in a specific field will permit to much more easily find out what the possibilities of cooperation between Consortia in each particular field would be.

It will be the responsibility of the WG to organise SRNWP Workshops in their field. The rule 'a workshop every two years' is abandoned. Each WG will decide when it is appropriate to organise a workshop.

It will be the responsibility of the Chairperson of each WG to present each year at the EWGLAM Meeting the state-of-the-art in the field of competence of his/her WG.

As there will be at least one representative of each Consortium in each Working Group, the function of consultancy for the NMS is implicitly fulfilled.

SRNWP Workshops

The SRNWP Workshops are an important activity of the Programme. These workshops are of a high scientific level and well attended as this can be seen from the workshop reports in the Programme web site.

For each of the following topics

  1. variational methods in data assimilation
  2. non-hydrostatic modeling
  3. numerical techniques
  4. soil processes and assimilation of soil and surface variables
  5. statistical and physical adaptation of model results
  6. verification of high resolution model results
  7. short-range ensemble prediction systems

we have as a rule that a workshop should be organised every two years.

Two of these workshops are characterised by a regular participation of scientists outside Europe (mainly from Canada, USA and Japan): workshops on non-hydrostatic modeling & workshops on variational methods in data assimilation.

It has been decided to keep the SRNWP Workshops in existence. The responsibility to organise them goes to the WG. Each WG will have the duty to organise workshops in the field of its competence when it judges it appropriate. The rule 'a workshop every two years'is abandoned.

EWGLAM Meeting

It has been decided to keep the EWGLAM Meeting annually, following the format defined at the 27th EWGLAM Meeting (2005 in Ljubljana) and applied for the time this year with great success at the Zurich Meeting.

The idea to organise together with the Academia 'European Conferences on dynamical meteorology and NWP' is abandoned.

SRNWP Advisory Committee (AC)

This point gave rise to the longest discussion when the Assembly discussed the different aspects of the new governance. A vote took place and the set-up of a SRNWP Advisory Committee has been accepted by 20 votes again 2. The Advisory Committee will allow a better representation of the Consortia in the Programme and a better support of the activity of the Programme Manager.

Constitution of the SRNWP Advisory Committee

It appeared clearly that to constitute it with one Representative per SRNWP Member (27 NMS!) would give far too large a group. The Advisory Committee has to be centred on the Consortia even when the latter have no legal existence in EUMETNET (only the NMS are legal Members).

We had unfortunately no time to discuss the size of the AC: either a very small group in the sense of a 'daily management group' or a larger group that would concentrate its work on the important questions only. There has also been no time to discuss the rule or the formula to be used for the representation of the Consortia in the AC. The factors to be considered could or maybe should be the number of Members of the Consortia and its overall financial contribution. The 'Programme Redaction Committee' will study the question.

Contacts with the Academia

In his draft Programme Proposal the Programme Manager has raised the issue of the low degree of collaboration between the SRNWP Programme and the Academia. Quote: 'In Europe, the collaboration of the NMS with Academia should be intensified. This was also the wish of the NWP Vision Meeting of March 2006 at the ECMWF which has passed a recommendation in order to develop better working relationship with Academia'.

As the idea of common European Conferences on NWP will not be followed, it has been proposed at the meeting that the contact in NWP between the NMS and the Academia should be realised through the Working Groups. The WG members will know in their field of research their colleagues of the Universities. A good starting point for an enhanced collaboration would be to open the SRNWP Workshops to the relevant specialists working at Universities.

Preparation of the Programme Proposal and Programme Decision for the EUMETNET Council

The present phase of the SRNWP Programme ends the 31st of December 2007.

The Programme Proposal and the Programme Decision have to be sent to the EUMETNET Secretariat the 30th of April 2007 at the latest.

We can presently forget about the Programme Decision which can be considered as a summary of the Programme Proposal and can be rapidly prepared if the Programme Proposal treats accurately all the relevant issues.

It has been decided at the meeting to form a 'Programme Redaction Committee' for the preparation of the Programme Proposal. We agreed on 3 persons per Consortium. This would give of group of 15 persons, 16 with the Programme Manager. This is too large a group. Two representatives per Consortium would give together 11 persons. This would be more than enough.

The review of the present report by the Local Organizer of the Zurich Meeting is cordially acknowledged.

4. Operational NWP models in Europe November 2006

compiled by D. Majewski, DWD, Germany
Country Model Mesh size (km) # of gridpoints # of levels Computer
Denmark HIRLAM 17
Estonia NH-HIRLAM 11
Linux Cluster
12 nodes
Finland HIRLAM 22
SGI Altix
Ireland HIRLAM 16
Netherlands HIRLAM 11 816x650 60 SGI, Itanium
Norway HIRLAM 22
SGI Origin
Spain HIRLAM 17
Cray X1e
Sweden HIRLAM 22
Linux Cluster
Austria ALADIN 9.6 300x270 45 SGI Origin
Belgium ALADIN 7 240x240 46 SGI Altix 3700
Bulgaria ALADIN 12 90x72 41 SUN, Linux PC
Croatia ALADIN 8 229x205 37 SGI Altix
Czech Rep. ALADIN 9 309x277 43 NEC SX6/8A-64
46 Fujitsu VPP5000
Hungary ALADIN 8 349x309 49 SGI Altix 3700
Poland ALADIN 13.5 169x169 31 SGI Origin 2800
Portugal ALADIN 12.7 85x96 31 DEC Alpha ES40
Romania ALADIN 10 144x144 41 SUN E4500
Slovakia ALADIN 9 309x277 37 IBM p690
Slovenia ALADIN 9.6 258x244 37 Linux Cluster
14 nodes x 2 IntelXeon
Bulgaria HRM 14 97x73 20 SUN, Linux PC
Germany GME
IBM P5-575
Greece LM 14 95x113 35 Convex SPP16
Poland LM 14
SGI Origin
Romania HRM 14 41x37 20 Bull DPX, SUN Blade
Switzerland LM 7 385x325 45 NEC SX5
Yugoslavia ETA 16 245x305 32 Pentium III Cluster
United Kingdom UM
UM (UK4)
Poland UM 17 144x116 31 Cray SV1
ECMWF IFS 40 global 60 IBM SP