Deported developments from October 1999 to ... mid-2000

Due to the new rules and, for this transitional Newsletter, the deported works reported here deal with the last quarter of 1999 and the two first quarters of 2000. "Deported work" means work realised "at home", in each country, not for LACE or for ALATNET (see LACE or ALATNET work reports). As requested during the Cracow workshop, the report of Deported work will be modified ... and will contain the reports I am sure you will send to me following the new rules ...( a good example of these expected reports is given by the report sent by D. Banciu for Romania - see below -). Thanks to the effort of each correspondant in each country, this chapter about deported work should become less accounting but much more interesting.

Many newcomers entered recently the ALADIN project, both through the ALATNET training in Radostovice and thanks to Toulouse or "at home" training. Welcome to :

• Alex Deckmyn and Philippe Nomerange (Belgium),
• Tomas Kalibera and A. Trajakova (Czech),
• Gergely Boloni (Hungary),
• Siham Sbii, Rashyd Zaaboul and Mohamed Jidane (Morocco),
• Margarida Belo Pereira and Vanda Sousa da Costa (Portugal),
• Constantin Rada (Romania).

1. In Austria 

• Objective verification of ALADIN/Vienna (K. Stadlbacher),
• Operational ALADIN/Vienna (Y. Wang).
• CYCORA tuning (S. Kreilberger)
• Convection index (H. Seidl)

2. In Belgium

• Validation of the ALADIN model, qualitative statistical verification, study of exceptional meteorological situations(J. Neméghaire, J. Vanderborght),
• Operational implementation of CANARI (O. Latinne),
• Performance analysis (F. Chomé),
• Tools for verification and verification of the winter and the spring periods (P. Termonia),
• Attempts to implement an MPI version of the code, Development of a prognostic version of the convective scheme, Preparation of Cracow workshop, ALADIN Physics Documentation (L. Gérard).

3. In Bulgaria 

• Implementation of ALADIN11, supporting of operational suit, files with forcing for wave model included in operational suite (A. Bogatchev).

4. In Croatia

• Diagnostic study - severe bora wind diagnosis and ALADIN forecast (A. Bajic, D. Glasnovic),
• Work at direct ALADIN model output -farther development of the HRID diagnostic package; diagnostic study - MAP SOP diagnosis and forecast of vertical atmospheric structur (D. Glasnovic, B. Ivancan-Picek),
• Verification - maximum and minimum temperature and wind velocity from the lowest level of pseudoTEMPs (Z. Vakula),
• Tuning of blending initialization (S. Ivatek-sahdan, D. Klaric),
• Successive forecasts consistency as a measure of skill (D. Glasnovic, B. Ivancan-picek),
• Diagnostic package - implementation of the land-sea contrast into the structure function under the ALADIN developments in Prague, Visualization for MAP related application (M. Zitouni).

5. In Czech Republic 

• Operational suite software development and monitoring (D. Dvorak),
• Development of Kalman filtering of ALADIN forecasts, Model Output Statistics, CMAFOC files (Z. Huthova),
• Verification of ALADIN/LACE (R. Mladek), of wind forecasts (M. Zak), of precipitation forecasts (E. Synkova),
• Study of extreme convection events (D. Merkova),
• Development of a new CHAGAL interface (T. Kalibera),
• Diffusive properties of the semi-Lagrangian scheme (F. Vana).

6. In Hungary

• Preparations for the ALADIN workshop in Ljubljana (A. Horanyi, I. Bonta, S. Jenki, I. Bartha, A. Fovenyi),
• Computation of background error statistics for ALADIN/HU (A. Horanyi),
• Implementation of AL11, AL12 in Budapest (A. Horanyi, G. Radnoti),
• Subjective evaluation of ALADIN model (I. Bonta),
• Surface parametrization -lake problems- (S. Kertesz),
• Switch to AL11, maintenance and improvements of the operational scripts, automatic forecast generation based on ALADIN model, implementation of AL11t2 for the non-Hydrostatic version of ALADIN (T. Szabo).

7. In Moldavia

• Nothing reported this quarter.

8. In Morocco

• Aladin/Maroc Model controls (Y2K transition), analysis of specific humidity in CANARI (introduction of necessary modifications in AL11T2 Cycle), adjustment of forecasting errors statistics in 3d-Var Aladin (W. Sadiki),
• AL11T2 implementation on CRAY J916 and on SUN ULTRA 7, Aladin Distributed Memory tests on Cray J916, Test and validation (on one situation) of Aladin on a small domain, included in Aladin/Morocco, with higher resolution (R. Ajjaji),
• Using statistical adaptation of Aladin/Maroc outputs for forecasting extreme temperatures: data processing, using statistical methods to investigate correlations between extreme temperatures and numerical model outputs, elaboration of the statistical model, model controls, introduction of new data (S. Sbii),
• Local Tuning of cloudiness scheme (comparison of the forecasted satellite picture with MeteoSat and comparison of total cloudiness with synop data) (M. El Abed),
• Impact of the frequency of coupling on the Aladin forecast at fine scale, first results of Aladin forecast with a 7km horizontal resolution (J. Boutahar),
• Test and validation of the latest version of convection parametrization over Morocco (R. Zaaboul),
• Exploratory and bibliographic studies on the new high resolution data base of vegetation over the Aladin/Maroc domain (M. Jidane).

9. In Poland 

• Administration and development of NWP operational system (A. Dziedzic, M. Jerczynski, M. Szczech, W. Owcarz),
• Development of post-processing-on-demand system (W. Owcarz),
• Development of verification software (A. Dziedzic, M. Szczech, J. Woyciechowska),
• Organisation of the 8th ALADIN Workshop (J. Woyciechowska).

10. Portugal

• Installation of new versions and adaptation to ALADIN/Portugal environment, maintenance (C. Madeira),
• Verification, archiving and case studies (F. Prates).

11. In Romania 

• Convection parameterization (D. Banciu)

The recent "ensembling entrainment", developed at CNRM/GMAP has been tested for the TOGA-COARE and CLEOPATRA cases. For more details see the Ph.D. report of D. Banciu.

• Ozone profile (C. Rada, A. Sima)

A new function has been tested to express the climatic vertical profile of the ozone in the Aladin model, for integration over Romania domain. For this purpose 10 years of SBUV/2 NOAA11 data and daily measured data at Bucharest have been fitted by a function, similar to the Arpege one but with one additional degree of freedom, a variable exponent. Monthly triplet of constants gave the new monthly variable profile function, as well as daily triplets fitted the daily measured values. For a period of time, in parallel to the operational run, two more runs were executed: a run with the monthly climatic profile and another with the daily fitted profile. The assumption made was the constancy of the profile in the integration domain: studies based on measured data show, usually, a slow latitudinal variation over Romania territory. A variation with latitude and longitude has been finally tested but this time by fitting satellite data, due to the fact that one single DOBSON spectrometer UMKEHR measurement data, at Bucharest, is available. The impact of the ozone vertical profile on the forecast has been tested. For more details see the Aladin Newsletter article: "Ozone profile fitted to Bucharest measured data" of C. Rada, A.Sima and M.Caian.

• Optimal time step for Aladin-Romania (R. Radu, A. Sima)

Numerical simulations have been made with operational Aladin (cycle 11) in Romania for different time steps. We tested time steps of 6,7,8,10,12,15 mn. The forecast skills have been computed with the operational verification package and then compared against similar skills obtained for the operational run. Dynamical and physical behaviour at time step change have also been analysed through grid points diagnostics. The 8 minutes (12% reduction) simulations gave the best skills and behaviour (better than the operational case - 9 minutes time step; for example uo to 2 hPa precision increase for mean sea level pressure), but spectral features are to be further analysed to confirm the results.

• DDH - bug correction for LAM case; output interface (M. Caian)

Reading of DDH files routines have been implemented (rdhfa.F90, readddh.F90, wdhfor.F90), and a procedure to extract, optionally, diagnostic Aladin fields. The routine sumddh.F90 has been corrected to allow point-type domains diagnostics computation in LAM case. The routine ppfidh.F90 has been corrected (bug in writing some variables in LFI file, i.e. condensation fluxes: convective and stratiform, for rain and snow).

• Study for writing LBCs gradients in intermediate files and deriving lateral boundary perturbation (C. Soci)

The problem of writing intermediate gradients of the forecast error norm with respect to the lateral boundary conditions in grid point files for each coupling moment has been studied. The ALADIN code place where the small perturbations derived using these gradients might be injected in the original lateral boundary has been taken into account as well. The lateral boundary gradients can be used to create modifications to the original lateral boundary conditions in analogy with the corresponding modifications to the initial conditions. The procedure for deriving small lateral boundary perturbations is to multiply the gradients with a scaling factor chosen by experimentation.

• Statistical adaptation (O. Diaconu)

The Kalman filter calibration for the ALADIN outputs has been carried out in order to correct the 3 hours temperatures for 140 Romanian meteorological stations. It is a standard Kalman filter procedure (Pearson, 1991) using a 40 days moving archive for observed and predicted values. The control procedures were developed in order to have monthly comparative measures of direct output and corrected forecasts.

• Coupling the VAGROM wave model with the Aladin model (S. Stefanescu)

The 48 hours forecast for the 10m wind field obtained from the numerical weather prediction model ARPEGE is used operationally as input to drive the numerical wave model VAGROM (an adapted version of the French wave model VAGUE for the Black Sea). The VAGROM model has a regular longitude-latitude grid covering the whole Black Sea. A higher resolution domain (5 x 5 minutes latitude-longitude grid) has been nested on the coarser grid over the western basin and an interface to couple (the coupling frequency is 3 hours) the Aladin model with the wave model has been developed.

• Code maintenance (C. Soci)

The last version of the Aladin model, "al12", has been implemented. Due to our workstation memory limitations it is not possible to use the distributed memory switch. An upgrading of the ALPHA DEC platform is foreseen for this year. The tests carried out with the new version of the model did not show significant modifications in comparison with the previous one, "al11t2".

• e923 configuration (S. Alexandru)

The configuration e923 was implemented on our workstation following the documentation of N. Pristov. Technically it was necessary to do some modifications of the code due to the memory limitation. For orography a sub-domain (a little bigger than Aladin-Romania domain) from global data set GTOPO30 (resolution 2'30") was used while for the other surface parameters the standard files from Toulouse were used. The results were compared with those obtained in Toulouse and no significant differences were noticed. To improve the orography of the domain Aladin-Romania, we work with two cost functions: Jercynzski (LNEWORO2=.T.) and Bouteloup (LNEWORO=.T.), and now we are testing which of these functions represents better the orography for our domain.

• Aladin vertical profiles (L. Dragulanescu, B. Caraman)

Software was developed to visualise the Aladin vertical profiles in the standard emagramme. It uses the GRIB format of the postprocessed Aladin fields on pressure levels. The levels of the lifting condensation (LCL), convective condensation (LCC) and of free convection are automatically computed. Several instability indexes as cross total index, total totals index, lifted index, severe weather index and Showalter index are also available.

• Lake model coupling (M. Caian, R. Radu)

An interface Aladin-Hostetler lake model has been developed. Hourly computation are done having as input the prognostic Aladin variables (temperature, specific humidity, surface pressure, wind) and precipitation. The output consists on surface fluxes, lake surface temperature, ice/snow thickness and the surface cover. First tests have been carried out with observed lake temperature as initial data.

12. In Slovakia

• Verification of ALADIN outputs (J. Vivoda),
• Implementation of AL12 and 1D model on DEC (J. Vivoda),
• End user and customers presentation of ALADIN products (www), automatic text forecast from ALADIN model (M. Bellus),
• PseudoTEMPs and ACS development (J. Masek),
• Workstation version ALADIN/Slovakia, operational maintenance (O. Spaniel).

13. In Slovenia 

• Organisation of the 7th ALADIN workshop, CAPE, Diag-Pack (J. Jerman, N. Pristov)
• Operational configuration of cluster, double suite, CMAFOC (N. Pristov),
• Dynamic adaptation and diagnostics of precipitation (M. Zagar),
• Research work on cluster installation, CANARI (J. Jerman).

14. Deported work by Météo-France people

• In Budapest (D. Giard, C. Fischer) and in Prague (E. Gérard) have already been reported in Newsletter 17 as it was mixed Toulouse-Deported work.
• In Ljubljana : R. El Khatib helped to the porting of ALADIN/SI to the cluster of workstations.

Home