Status of ALADIN operational activities at SHMU (September 2006)

HARDWARE
--------

* Computer [no change]:

  - IBM Regatta
  - 32 CPUs of 1.7 GHz
  - 32 GB RAM
  - 1.5 TB disk array
  
* Archiving facility [no change]

  - IBM Total Storage 3584 Tape Library with IBM Tivoli Storage Manager
  - current capacity of tapes around 24 TB
  - used for automatic backup of ICMSH files, GRIBs and selected products

OPERATIONAL SUITE
-----------------

* Domain and geometry [no change]:

  - 309 x 277 points (C + I zone)
  - dx = 9.0 km
  - quadratic truncation
  - 37 vertical levels

* Operational model version [no change]:

  - al28t3 with "czech physics"

* Integrations [prolongated to 72/60 hours since 27-03-2006]:

  - 4 runs per day (00, 06, 12 UTC up to 72 hours, 18 UTC up to 60 hours)

OTHER OPERATIONAL ACTIVITIES
----------------------------

* Point verifications of 3-day ALADIN forecasts were carried on.
  There is slight deterioration visible with increasing forecast
  lead time and the forecast for third day was found to be valuable.
  There are several customers requesting 3-day forecasts with hourly
  frequency or high space resolution, who could not be satisfied
  without having ARPEGE LBC up to 72 hours.

* Non-envelope orography was tested in parsuite. It showed slight
  deterioration of 2 m temperature and 10 m wind. Precipitation 
  field was improved (decreased maxima over mountains, more 
  precipitation in valleys). Operational implementation is waiting
  for statistical adaptation of T2m, because lowering of mountain
  ranges significantly increases temperature bias for mountain 
  resorts.

* Epsgrams from multirun ensemble are produced operationaly since
  May, being available on intranet. They use 48-hour overlap of 5 
  subsequent ALADIN integrations. They were found to be useful
  tool for estimating (ARPEGE driven) uncertainity of ALADIN 
  forecast.

* Parts of INCA system provided by Austrian colleagues were ported.
  Currently T2m analysis is run operationaly over Slovak domain
  with 1 x 1 km horizontal resolution. Hourly outputs are visualised 
  by ACUVIS tool developed at ZAMG.

  Nowcasting precipitation module providing analysis and 6 hour
  forecast is technically working, being fed by 2-radar composite
  of precipitation rates, measurements from about 70 automatic 
  raingauges and ALADIN forecasts of precipitation and wind
  at 500 and 700 hPa levels. Tests with 1 hour step were not 
  successful (such step is too long for finding pattern correlations 
  and determining motion vectors). Tests with 15 minute step are 
  more promising, but they had to be done offline (data from 
  automatic stations are not collected in real time).

* Spectral blending is technically working. Basic tuning of blending
  truncation and DFI settings was done. Blending behaviour was tested
  on few cases. Inclusion of surface blending and operational
  implementation should follow soon.

* SYNOP observations are converted into OBSOUL format operationaly,
  TEMPs are in progress. Veral package works including visualisation, 
  but it is not ported to run_app environment yet.

PORTING STATUS
--------------

Highest validated cycle remains al29t2 (pre ALARO-0 tested). Cycles
al30t1 and al31t1 were ported but not validated.

ARPEGE LBC DOWNLOAD
-------------------

Both assimilation and production LBCs are downloaded 4 times per day.
Primary channel is internet/BDPE. Speed of internet connection
was increased to 10 Mbit/s.

After stopping RETIM LBC transfer, backup of production LBC files
was solved with help of Austrian colleagues. They regularly download 
backup files from ECMWF and make them available on machine at ZAMG.
>From ZAMG files are downloaded to SHMU via internet. Backup of
assimilation LBCs is still missing.

There were already two cases with unreliable connection to parme,
when backup LBCs downloaded via ZAMG were used by operational suite.