Proposal for the continuation of the

Short Range Numerical Weather Prediction (SRNWP) Programme

as enlarged EUMETNET Optional Programme

by

Jeanette Onvlee, KNMI, Leader of the Redaction Committee;

Francois Bouttier, Météo-France;  Jean-Francois Geleyn, Météo-France; Mathias Rotach, MeteoSuisse; Hans-Joachim Koppert, DWD; Per Unden, SMHI; Fritz Neuwirth, ZAMG; Dijana Klaric, DHMZ; Terry Davies, Met Office; Alan Dickinson, Met Office.

Table of contents

1. Background

2. Past and future form of the SRNWP Programme

3. Propositions for the third phase of the SRNWP Programme

            3.1       Deliverables of the new programme

3.2       Accompanying Projects

            3.3       New governance of the Programme

                                    3.3.1    Programme Manager

3.3.2    Advisory Committee

                                    3.3.3    Scientific Working Groups

                                    3.3.4    EWGLAM meetings

            3.4       Collaboration with Academia

4. Start and length of the Programme

5. Proposed budget

1. Background

At the 24th EUMETNET Council meeting, the Chief Executive of the UK Met Office, Dr. David Rogers, presented a document titled "A Vision for Numerical Weather Prediction in Europe" in which he expresses his desire and ideas for more coordination and collaboration in Numerical Weather Prediction (NWP) between the Institutions of the European Meteorological Infrastructure. This has been the starting point of a discussion on the future of NWP in Europe.

In this context, a NWP Vision workshop was held at ECMWF on 15-17 March 2006. At this meeting, 6 areas or themes were identified where collaboration in NWP should be strengthened:

- Enhanced interoperability of NWP systems

- Improved framework for collaboration on NWP in Europe

- Socio-economic impacts

- European interactive multi-model LAM (LAM = Limited Area Modeling)

- Adapting to the enhanced NWP capability

- Quick wins.

For each of these themes, recommendations were formulated (cf. Document EMN/C27/Doc12), all of which have been considered and approved by the EUMETNET Council. Under the theme "Improved Framework for Collaboration", one of the recommendations reads: "Initiate the reorganisation of the SRNWP Programme" in order to enlarge its scope and strengthen its activity.

The discussions at the Vision Meeting made it clear that the present format of the SRNWP Programme as well as the financial means at its disposal (€ 35'000.- a year) do not correspond to the importance that the NWP has today in the works and duties of the NMS. In order to make the Programme stronger, more effective and more visible, it was deemed necessary to reassess the structure and tasks of the Programme.

This reassessment took place at the SRNWP Annual Meeting 2006 held in Zurich (Switzerland) on the 12th of October 2006. At that meeting, the aims, key deliverables and a governance structure formulated for a new SRNWP Programme were considered, and it was decided to formulate proposals for 3 specific projects that should accompany the Programme.

2. Past and future form of the SRNWP Programme

Following the decision taken by the EUMETNET Council at its 8th Meeting (9th of June 1999) to establish an Optional Programme supporting a Co-ordinator for short-range NWP, the Programme C-SRNWP (Coordination - Short Range Numerical Weather Prediction) started on the 1st of January 2000.

The Swiss Federal Office of Meteorology and Climatology (in short: MeteoSwiss) became Responsible Member, and Mr. Jean Quiby was designated as Programme Manager who had to devote 30% of his working time to the Programme. The duration of the Programme was a period of 4 years, with a budget of € 35'000 per year.

Before the creation of the Programme, an informal "European SRNWP Network" was already in existence, created in October 1993.  In that period, the Network had no co-ordinator and its secretariat was part of the EWGLAM secretariat. This latter was rotating on a yearly basis between the NMS in charge of the organisation of the annual EWGLAM Meetings.

At its 19th Meeting (12th of September 2003 in Paris), Council adopted the Programme Decision EMN/Dec/C19-2, which entailed the continuation of the C-SRNWP Programme for 4 more years (2004-2007) without changes, apart from the addition of the following tasks in the duties of the Programme Manager:

- to promote standardisation in the operating NWP environments

- to maintain the SRNWP web site http://srnwp.cscs.ch

The C-SRNWP programme that ends on 31 December 2007 is a very low-cost, low-profile programme, with a strong emphasis on promoting and facilitating general information exchange, rather than on attaining specific objectives, such as is the case in most other EUMETNET programmes. It functions primarily as a network of informal contacts, and as such it has really been quite successful. At the Vision workshop, it was clearly stated that this aspect has been a great strength of the programme, which should definitely be maintained. In addition, however, it was proposed to broaden the scope of the programme by the introduction of specific optional projects, focussed on achieving important targets on the themes identified in the Vision workshop. These conclusions have led to the proposal outlined below for a new, more ambitious and targeted SRNWP Programme.

3. Propositions for the third phase of the SRNWP Programme

3.1 Objectives and deliverables of the new programme

The main objectives of the programme are identified as:

• Enhanced scientific cooperation in the field of NWP in Europe, through the initiation and execution of joint projects, specifically targeted at achieving elements of the EUMETNET Vision on NWP. These projects are to be carried out under the umbrella of SRNWP. Three areas for which coordinated activities are highly desirable have already been identified during the 2006 Vision workshop and the 2006 SRNWP Meeting, and three project proposals have been, or are being, formulated accordingly along those lines (see section 3.2).

• Enhanced operational cooperation through harmonization of standards and increased interoperability between models.  This has been one of the original goals of the C-SRNWP programme, which has never materialized. It is aimed to achieve this through:
• The formulation and coordination of specific projects aimed at harmonization in e.g. data formats, data exchange, common model interfaces etc. One of the project proposals described in section 3.2 is directly committed to these issues.
• Tasking the working groups within the SRNWP program to identify and coordinate activities aimed at increased interoperability.

• Effective diffusion of NWP knowledge and enhanced practical cooperation in NWP through the information exchange. This is to be achieved through:
• The organization of workshops and the yearly EWGLAM/SRNWP meeting
• The organization of thematic working groups, responsible for the coordination of cross-consortium cooperative activities in their field
• The SRNWP web site

Note that these items essentially represent the continuation of the activities hitherto carried out within C-SRNWP, as recommended in the Vision workshop.

3.2 Accompanying Projects

As recommended by the Vision Meeting, the SRNWP Programme should be accompanied by Projects - officially EUMETNET Programmes – aimed at producing specific deliverables for the Participating Members.

At the Zurich SRNWP Meeting (12th of October 2006), it has been decided to initiate proposals for dedicated projects on the 3 following subjects:

• Increased interoperability between the numerical forecasting systems of ALADIN, HIRLAM, COSMO and Met Office.

This project is directly targeted towards one of the main recommendations of the 2006 Vision workshop. Its aim is to increase the harmonization of operational practices and the interoperability between the main elements of the European NWP forecasting systems.

One goal should be that all regional models can be driven by any of the European global models (ECMWF, Unified Model, Arpege and GME). Also, interfaces to standardized field formats should allow the use by all regional models of a common verification package (see below). Other aspects to be considered in this respect are e.g. the harmonization of data exchange formats, the arrangement of data exchange agreements, the definition and development of common interfaces between model components, and the adoption of common post-processing, verification and intercomparison tools.

• Realization of an operational European LAM-EPS - EurEPS

As for Interoperability, this specific project follows directly a strong recommendation of the Vision Meeting. It is believed that a “multi-centre” ensemble of ensembles, is well suited for utilizing the unevenly distributed computer resources amongst the NMSs, and that such a grand ensemble should provide access to more skilful and more frequent ensemble forecasts to participants than single countries could produce on their own. EurEPS aims at investigating these assumtions by inter alia: (1) gaining pre-operational experience with a European distributed grand ensemble system based on existing plans in the consortia; and (2) quantifying benefits for operational probabilistic forecasting of a grand pan-European ensemble system compared to benefits of sub-European ensemble systems that NMS can afford to run individually. EurEPS will simultaneously define a European contribution to the THORPEX Project TIGGE-LAM.

• Common verification package and model intercomparison

The main deliverable of this project will be a common verification methodology and software for the verification of deterministic model results. Every NMS that will participate to this project should be able to use it for its model without further adaptations. This will be possible thanks to the interface programmes that will be developed by the Project Interoperability. It will also be the duty of the Responsible Member of this Programme to run with the common verification package an operational model intercomparison between the models ALADIN, HIRLAM, COSMO and Unified Model.

These three project proposals will be presented to the EUMETNET Council as proposals for three independent EUMETNET Programmes, each with its own Responsible Member and budget. However, they should be considered as forming a unity with the SNRWP programme, as they will yield the main specific deliverables aimed at realizing the recommendations of the NWP Vision workshop for EUMETNET members in the coming years (an approach comparable to the one used in the EUMETNET ECSN-programme). The activities within these projects will require a strong coordination by the SNRWP programme, in order to achieve the required synergy between the projects and optimize efficiency.  

3.3 New governance of the Programme

A more extensive and project-oriented scope for the new programme also has consequences for the manner in which it is governed. A particularity of the SRNWP Programme until now is that it has never had a supervising body as a Management Board or something similar; the level of reporting and supervision of the Programme are weak, compared to other EUMETNET programmes; in view of the small size of the programme, a “heavier” structure did not appear to be justified. However, with the larger ambitions and extended scope of the new programme, a stronger management structure is considered necessary, in particular with regards to the initiation and supervision of projects to be carried out under the SRNWP umbrella.

Much of the activities in SRNWP research and operations in Europe take place within the LAM consortia. In order to strengthen cross-consortia cooperation and coordination, it is proposed to create an Advisory Committee consisting of the managers of the LAM consortia, or representatives on their behalf. In conjunction with the Programme Manager, this Advisory Committee regularly reviews the strategic developments within NWP in Europe, and the ensuing demands and opportunities for enhanced cooperation.  Their main responsibility is to monitor the progress within, and synergy between, SRNWP projects, and to identify new areas where cooperative projects are desirable.

Apart from the yearly EWGLAM/SRNWP meeting and SRNWP workshops, the main mechanism for exchange of information and promotion of collaboration within SRNWP will be the formation of scientific working groups (WGs) on specific projects or themes. The prime responsibility of these WGs will be to identify areas of practical cooperation between consortia in their field of work; they can make detailed proposals for, and subsequently coordinate, joint activities between participating Programme Members. The WGs consist of experts from the LAM consortia, who are responsible for communication on the WG activities and plans within their own consortium. 

Each of the projects formulated under the umbrella of the SRNWP programme will have its own working group. In addition, thematic working groups can be formed on areas of research, such as Dynamics, Physics, Data Assimilation, Surface Processes and Data. One of the main decisions for the SRNWP plenary meeting will be to decide on subject areas for WG. The initiation of a WG can be considered suitable once a critical mass of activity is identified.

Initiatives for formulating proposals for projects aimed at achieving specific deliverables for EUMETNET members, can originate from the Programme Manager, the Advisory Committee and the WGs. Each project proposal should include deliverables and milestones and identify a manager or coordinator for the activity.

3.3.1 Programme Manager

The main duties for the programme manager are as follows:

- to work towards the achievement of the 18 recommendations made in the 2006 NWP Vision meeting, which "identified a number of existing areas where there is potential for increased efficiencies through better cooperation, and new areas where we have the opportunity to bring skills together, and work in a more efficient and effective way" (Document EMN/C27/Doc12).

- To take initiatives to define concrete projects and activities aimed at achieving these recommendations, and to be actively engaged in monitoring and steering those projects and activities

-  to organize between the Programme Members the exchange of information related to scientific as well as operational aspects of NWP

More specifically, the programme manager is tasked with:

• the organization of Advisory Committee meetings, and providing the AC with information of the status of activities within the Programme;
• taking the initiative for initiating projects focussed on enhancing cooperation on NWP in Europe, and monitoring the progress of projects under the SRNWP umbrella and the level to which their goals are being achieved;
• liasing with the Working Groups, collate information and take initiatives, where necessary, to ensure that the WGs are functioning;
• reporting to the EUMETNET CO and Council, and liaison with other EUMETNET programmes;
• liaise with local organizers to prepare the programme and time table for EWGLAM meetings;;
• intensify collaboration with ECMWF and EUMETSAT, and promote contacts with non-European NMS’s and institutions;
• collecting and maintaining information on European NWP activities on the SRNWP web site.

Presently the function of Programme Manager is officially a 30% job, which is not enough even for the fulfilment of the tasks listed in the actual Programme Decision of the present, low-profile programme.

In view of the increased tasks of the Programme Manager in the present proposal with regards to the realization of the NWP vision, however, the new SRNWP programme is expected to require a full time Programme Manager.

3.3.2 SRNWP Advisory Committee

The envisaged tasks of the "SRNWP Advisory Committee" (in short "Committee") are:

- to regularly review the strategic developments within European NWP and identify the need for specific projects or coordinated activities,

- to define and initiate such projects or activities wherever necessary, and to monitor their progress, together with the Programme Manager 

- to act as an advisory instance to the Programme Manager for all matters concerning the Programme

The Committee will elect its own Chairperson among its Members.The Committee should meet at least once a year, during or adjacent to the EWGLAM/SRNWP meeting. The Committee Chairperson or the Programme Manager may call supplementary meetings if needed. Experts, particularly from ECMWF and EUMETSAT, can be invited to participate to the meetings. The EUMETNET CO will always be invited to participate.

3.3.3 Scientific Working Groups

The scientific working groups (WGs) are tasked with:

- Planning and organizing all SRNWP activities in their area of responsibility;

- In particular, planning and organization of workshops and meetings to foster the exchange of information on the state of the art in their field;

- Reporting back on the activities in their area to the Programme Manager and to the EWGLAM/SRNWP meeting;

- Identifying suitable areas for cooperation projects aimed at achieving specific deliverables for EUMETNET. 

The WGs consist of experts in the particular field of activity of the WG.  For any WG, each consortium will identify a contact/representative through which all communications regarding the WG will take place. Ideally, the consortia representative will be a programme manager from the equivalent consortia activity. It is up to the consortia to decide how to communicate and work with the information within the consortia.

Each WG is free to decide how often and when they get together. It is expected that much of the communication within the WG will take place by email.

Each WG will nominate a chairperson whose main responsibility will be to report to the SRNWP annual meeting, and to arrange for project proposals or reports on project activities to be presented to EUMETNET. When drawing up a project proposal the WG will be expected to consult widely within consortia.

The organization of workshops always has been an important activity of the SRNWP Programme, and this deserves to be continued. The primary role of the SRNWP Workshops is the exchange of scientific experience and information between the Consortia. It will be the duty of the Working Groups to plan and organise SRNWP Workshops in the fields of their respective speciality.

Each WG should be aware of the conferences, symposia and workshops in the field of its competence, organised in the recent past or announced for the future. Based on this knowledge, each WG will determine when it is appropriate to organize workshops in their area or, for an emerging topic, whether it would be useful to organise a first meeting. These SRNWP workshops will be opened to all the scientists of the SRNWP Members, to non-European NWP researchers and to Academia (See point 3.4 below).

3.3.4 EWGLAM meetings

The aim of the EWGLAM meetings has primarily been the exchange of information between the European NMS in all matters related to NWP: the scientific aspects, the computational aspects and the operational aspects. Initiated by SMHI, the first EWGLAM (European Working Group on Limited Area Modeling) Meeting took place in Norrköping in autumn 1979; this year, the 29th EWGLAM meeting (8-11 October 2007 will be held in Dubrovnik. Thus, the EWGLAM meetings have a solid tradition in the European NWP community, and they are highly valued for the key role they have played in facilitating cooperation and coordination of NWP research in Europe. They are presently the only Consortium-independent regular meetings assembling all the NWP groups of the European NMS. As such, it is felt that the practice of organizing the EWGLAM meetings should be continued in the new SRNWP programme.

It will be the responsibility of the Programme Manager to ensure that these meetings will continue to exist and it will remain his or her task to write the minutes of the "EWGLAM final discussion".

3.4 Collaboration with Academia

In Europe, the collaboration of the NMS with Academia should be intensified. This was the wish of the "NWP Vision Meeting" of March 2006 at the ECMWF which has passed a recommendation to "Develop better working relationship with Academia". Academia in Europe is very active in dynamical meteorology and modeling. There is a large number of "Institutes of Meteorology" or "Institutes for Atmospheric Physics" where high quality work is made in numerical simulation of atmospheric processes, including the ones that have to be parameterized.

It has been proposed at the Zurich meeting that the contact in NWP between the NMS and the Academia should be primarily realised through the Working Groups, as the WG members will know the colleagues working at Universities in the same field of research as their own. Thus, a good starting point for an enhanced collaboration with Academia would be to invite them to the SRNWP Workshops.

4.       Start and length of the Programme

The third phase of the SRNWP Programme will start the 1st of January 2008 and will last until the 31st of December 2011 (4 years).

5.       Proposed budget

The annual cost amounts to 100'000 € for a Programme Manager working full time.

From this amount, € 10'000 are foreseen on one side for the travelling expenses of the Programme Manager and on the other side for invitations of experts for the meetings of the Advisory Committee as well as for the SRNWP Workshops.