This Newsletter 3 presents you the principal events concerning ALADIN during the second quarter of 1996.
The news about work or events outside Toulouse are based on informations that I have received from you and essentially concern the first quarter of 1996.
Please do bring to my notice anything that you would like to be mentioned in the Newsletters.
For any comments, please contact :
It has become obvious that, in the framework of the ALADIN project, our common intellectual property needs protection. Météo-France intends to protect the ALADIN software through its registration at the "Agence pour la Protection des Programmes".
All the Institutes who had a valuable contribution in the development of ALADIN are considered as co-authors of this software. Their directors have got a letter asking their permission for Météo-France to register the ALADIN software in the name of all partners.
This was the first step to a legal regognition of our common effort ...
The last (and future) evolutions of ALADIN make it now necessary to get a minimum of formalization around the project in order to keep it as successful as it has been during the last five years.
A draft proposal for a Memorandum of Understanding (MoU) between the National (Hydro)Meteorological Services (NMS) of Austria, Bulgaria, Croatia, Czech Republic, France, Hungary, Morocco, Poland, Romania, Slovakia, Slovenia and Spain concerning the Limited Area Spectral Numerical Weather Prediction System named "ALADIN" has been drawn up and sent to every Directors of NMSs for comments and proposals of improvement or modification. A version acceptable to everyone will be produced then and submitted to the signing procedure.
This MoU first recalls the current principles guiding research, development and maintenance of ALADIN (more precisely ARPEGE/ALADIN, to recall the interdependency of the two systems).
Then, it intends to precise the conditions of use of the ALADIN software.
Conditions of membership to the ALADIN Project (for present partners), conditions of adhesion to the ALADIN Project (for new partners) and conditions of withdrawal from the ALADIN Project are then defined.
An impromptu ALADIN Project Review took place on Monday 22th of April 1996 on the occasion of the ACTIM meeting. The ALADIN visitors present at that time in Toulouse and the permament Toulouse staff contributed to the success of this review. Ten directors of NMSs from Central and Eastern Europe attended it.
The presentations began at 18:00 with :
A document that summarizes these presentations was distributed to the participants.
The review of the collaboration was then pleasantly discussed around a cold buffet offered by Météo-France/CNRM.
A few weeks later, another ALADIN project review was presented to Météo-France Directeur Général et some directors from other Météo-France services. Some of the previous actors who had come back home were replaced by new ones ... and all the presentations were made in French (it was the first French presentation for many of them and, according to the French speaking people attending, they were very clear presentations). The program of this "Session d'information sur le project ALADIN" was :
1. INCO-COPERNICUS concerted action, so-called "ALADIN-INCO"
A concerted action proposal was submitted to the INCO-COPERNICUS "Environmental" call in Brussels at the end of February. Our proposal has been rejected as "average" in terms of conformity with the aims of the financing. This was of course one of the weakest points (if not the weakest one) of our proposal. The lesson is that we should not any more try to twist a subject to fit in the pre-chosen work plan if the latter is too far away from our own aims. Sorry to all of you for the work for no result and nevertheless thanks for your help. Please pass the information to your individual partners with the same thanks.
2. INCO-COPERNICUS keep-in-touch, so-called "ALADIN-KIT"
A keep-in-touch for the previous EU PECO-Action "Validation and pre-operational evaluation of a fine mesh spectral limited area numerical weather prediction model used in dynamical adaptation" was submitted to the same INCO-COPERNICUS call, under the Acronym "ALADIN-KIT'. This proposal has been evaluated by the Commission.
For the following criteria "ALADIN-KIT" was judged :
The global evaluation for "ALADIN-KIT" is "very good" and it has been classified in the field 6 (Information Technology) of the program.
In this field, 5 proposals have been classified "excellent" and will be financed. Half of the 35 "very good" proposals (including ALADIN-KIT) will be financed : so we have a 50 percent chance of being financed.
The final choice within the "very good" proposals will be completed soon.
3. "Réseau-Formation'Recherche" : PhD Studies
Last February, the French Ministry of "Education Nationale, Enseignement Supérieur et Recherche" accepted our proposal for a new "Réseau Formation Recherche"; it consists in financing five new PhD studies.
The beginning of study of Doina Banciu was delayed to August due to some administrative formalities. Filip Vana will begin his study at the same period and Mark Zagar later this autumn. The last two studies will begin at the end of this year or next year (Marta Janiskova and Ilian Gospodinov).
4. Embassies support
As usual, the documents for embassies support have been submitted by DGS/IE in Météo-France to the relevant authorities. The political and financial decisions have now been taken in the French Ministries. More details should be asked directly to DGS/IE (arlette.rigaud++at++meteo.fr).
This is the present situation ...
1. Pre-Operational test on ALADIN-FRANCE in Météo-France
ALADIN-France is in pre-operational test since the 11.03.96 on the CRAY C98 computer in Météo-France. The forecasters in Toulouse are now evaluating the ALADIN forecast available on SYNERGIE system.
No major change in ALADIN-France during these 3 months. The coupling files are still prepared via the configuration 926 (E926) that will be replaced by the "E927" version of Full-Pos in July.
Several meetings have been organized in regional centres of Météo-France (Lille, Aix en Provence, Paris) for presenting the ALADIN-France application to the regional and local forecasters. The dissemination of the ALADIN-France products to the local centres (more than 100 in France) will be based on one of the new RETIM channels, scheduled end of summer, while the 7 regional centres are about to receive them via the dedicated link they have with Toulouse, that have been upgraded end June to 128 KBit/s minimum.
2. Pre-Operational ALADIN-CEE in Météo-France
The pre-operational ALADIN-CEE who has been faithfully running on Météo-France C98 since the 31st May of 1994 was stopped on the 1st of July of 1996.
3. Pre-Operational ALADIN-MAROC in Météo-France
This version has been running in Toulouse on the Cray C98 since february 1996. As a temporary solution, before full replacement by ALADIN-Maroc/Casablanca, Météo-France has developed and put into operation the feeding of the SYNERGIE system installed in Casablanca by the results of ALADIN-Maroc/Toulouse using the 64KBit/s link.
4. ALADIN model and ALBACHIR forecasting suite in Maroc-Météo
The Moroccan forecasting suite based on the ALADIN model was named ALBACHIR* by "Sa Majesté Le Roi du Maroc".
The ALBACHIR suite will be run on a Cray J916 computer (also used for research tasks). It is expected to be operational this summer.
For the present pre-operational ALBACHIR suite, the ALADIN model is run twice a day (with assimilation cycle) : it uses the coupling files produced by Météo-France at 31 Km resolution and provides forecasts until 48 hour range on a domain covering Morocco (still at 31 Km resolution). These products are used by the forecasters through a SYNERGIE interface.
The future suite is also under test now. With the replacing of the E926 configuration by the EE927 one, the coupling files can be produced at 16 Km resolution. The ALADIN model will be run in semi-Lagrangian two-time-level with a 675 second time-step. The initialisation will be made using digital filters. 20000s CPU are necessary to compute a 48 hour forecast. FULLPOS will be then used to provide the forecasted fields at a 0.15 degree resolution visualized on a SYNERGIE station.
* BACHIR is a religious term and ALBACHIR can be translated in French by "le bon augure".
5. Preparing ALADIN-LACE in Météo-France
The April-June period has been intensively devoted to the preparation and installation of ALADIN-LACE. People involved in this work are the RC-LACE Toulouse team and the Météo-France Computer and Telecom Division. As it was planned the operational start of ALADIN-LACE (and the operational death of the old ALADIN-ECO after more than 2 years of existence) took place July 1st. The main differences with restpect to former ALADIN-ECO are:
Since RC-LACE and ZAMG (Austria) are paying for this improved application they have some privileges as far as the products are concerned, but the domain has been chosen for covering still Bulgaria, Romania and Poland, who have access to a large number of products, as a tribute to their participation to the ALADIN project.
6. Boundary conditions supply for various ALADIN versions
Boundary conditions for ALADIN-Romania and ALADIN-Slovenia are now produced on a regular basis. They are made available at Toulouse and fetched by the users. Therefore the current situation is 5 coupling tasks running on the operational suite on the C98, with ALADIN-France and the 2 versions for Morocco (ALADIN-Maroc/Toulouse having a smaller domain than ALADIN-Maroc/Casablanca). This should be reduced in the future when ALADIN-Maroc Casablanca will become fully operational (hence ALADIN-Maroc/Toulouse becoming useless), and also if ALADIN-Slovenia takes its boundary conditions from ALADIN-LACE rather than from ARPEGE.
On their DEC ALPHA Station 250/4, it takes 3.5 hours for 48 hour integration starting with 12, once per day. The domain resolution is 12.36 Km and contains 100*100 points and 27 levels.The time-step is 432s for the moment, with a 6 hour post-processing frequency. The libraries currently used are the last ones available before distributed memory cyle.
The main point to run the operational suite is the availability of the coupling files in Bucarest.
For the moment, ALADIN runs in configuration 001 perfectly, Full-Pos runs also but there are still some small problems probably with memory. The first results for pressure, model and height levels for lots of variables have been obtained ; an extensive testing is just going on. The EE927 configuration is also under process and, with the help of Ryad EL KHATIB (France) and Liviu DRAGULANESCU (Romania), it is expected to be able to produce the coupling files at the beginning of August.
The runs are realized on a DEC ALPHA Station 600-5/333 with an old ALADIN code (cycle 4) but with modifications provided by Liviu Dragulanescu (Romania) concerning EE927. Some performance tests were realized on a domain 24*24 points with 28 Km resolution (equivalent to the ARPEGE coupling files). It takes around 100 seconds real time on empty machine for 6 hour integration with ALADIN in 2 times-level semi-Lagrangian configuration and time-step of 432s (the same integration but in 3 times-level and a time-step of of 216s takes 30 seconds CPU time on CRAY C90).
On the 1st of July, you received the last ALADIN Quasi-Operational Bulletin. Since the 1st of June 1994, 30 bulletins gave you informations on the quasi-operational ALADIN-CEE suite.
Due to the successful work of the LACE team in Toulouse, it was possible to stop this suite ... and the corresponding bulletins.
ALADIN-LACE took the place of ALADIN-CEE and the ALADIN-LACE Bulletins appeared.
On the 2nd of July, the ALADIN-LACE Bulletin number 1 gave you notice of the delay of the afternoon operational suite due to an unexpected problem on the CRAY C90.
On the 4th of July, the ALADIN-LACE Bulletin number 2 was pleased to inform you that the youth problems in the geometry of ALADIN-LACE were solved and where the GRIB files produced by ALADIN-LACE could be found.
The last summaries of the participations in the ALADIN project were prepared at the beginning of July 1996. As explained in the M.of U., the total participation updated on the 30 of June 1996 is calculated considering the Toulouse part of the work until this date and the deported part until the previous quarter (31 of March).
The estimation has been realized according to the following rules (more details will be avaiblable in the next version of the MoU in appendix) :
In this part, you will find informations sent by the representative of each country for the national participations. As a very few of them were received for the second quarter of 1996, the corresponding informations will be part of the next Newsletter.
For more details, please contact directly the authors.
On the other hand, do send me any information that you would like to be indicated here.
1. Work for ALADIN in Austria
2. Work for ALADIN in Bulgaria
3. Work for ALADIN in Croatia
MSH of Croatia would like to express gratitude to Hungarian colleague Mr. Istvan Ihasz for his great help.
4. Work for ALADIN Czech Republic
5. Work for ALADIN in Hungary
6. Work for ALADIN in Morocco
7. Work for ALADIN in Poland
8. Work for ALADIN in Romania
9. Work for ALADIN in Spain
10. Work for ALADIN in Slovakia
11. Work for ALADIN in Slovenia
12. Deported work for ALADIN realized by Météo-France people
Last quarter, fifteen visitors (and one French student) joined their efforts to the Toulouse permanent ALADIN team. Their main developments were related to :
1. About scripts and libraries :
Thanks to the great effort provided this quarter we are close to the cycle 6 of ALADIN library. Some parts of the code are ready (semi-Lagrangien, non-Hydrostatic, coupling, CANARI analysis, reorganisation of spectral tables,..) but we are waiting for others ones to make final tests.
Many people were involved in this task : Jaouad BOUTAHAR (Morroco), Liviu DRAGULANESCU (Romania), Samir ISSARA (Morroco), Radmila BUBNOVA and Gabor RADNOTI (LACE team), Patrick LE MOIGNE and Ryad EL KHATIB (Météo-France, GMAP), Jean-François ESTRADE and Vijay SARAVANE (Météo-France, VALPARAISO team).
Some tests were realised by Uros STRAJNAR and Andrey VELKAVRH at high resolution (2.5 Km) over a domain around the nuclear powerplant of Krsko in Slovenia. Both Eulerian and semi-Lagrangian runs were realised in Hydrostatic option, with an attempt to debug the non-Hydrostatic version.
ALADIN is now running on a SUN machine : Oldrich SPANIEL (Slovakia), Jean-Marc AUDOIN (Météo-France, GMAP), Jean-François ESTRADE and Vijay SARAVANE (Météo-France, VALPARAISO team).
The operational ALADIN-LACE suite was prepared by the LACE team in Toulouse (Radmila BUBNOVA, Gabor RADNOTI) and, as expected, the first operational runs were computed on the 1st of July.
2. About diagnostics:
3. About physics :
4. About analysis and assimilation :
5. About ALADIN coupled with ALADIN :
These informations (and others) are available on a public ftp (updated in July). A WWW server is ... planed to display information on ALADIN. Later, you will be informed by e-mail about the direct access to the ALADIN Web pages.
 |
Home |