PowerPoint bemutat

1
Use and evaluation of medium-range forecast at
the Hungarian Meteorological Service. Ferenc
Wantuch Hungarian Meteorological Service
Kitaibel P. u. 1. H-1024 Budapest E-mailwantuch.f
_at_met.hu
2
Talk outline

• Introduction
• Verification
• Two interesting case studies
• New products based on ECMWF.

3
Data flow and structure
Visualisation
NWP data source
Netcdf file format
File format GRIB BUFR
Code generalisation process
4
Code generalisation by the help of different
method (PP, MOS) visibility forecast, max, min
temperatures cloudiness, ...
Forecaster output
NWP output
Text and data Control
Verification
Forecast text generator program
End users
5
Verification
A Complex Score was defined.
Weight Temperature 2525 Precipitation
22,5 Wind 10 Cloud Cover
17,5
25
25
25
20
17,5
15
12,5

10
10
6
4
5
0
Tmin
Tmax
P.Exist
P.Amounty
Áv.Windl
Wind gust
Cloud Covert
6
Verification 2004 (Hungary) ALADIN ECMWF
Forecaster
Tmin
Tmax
7
Verification results and trends
Verification results 2004 (Hungary) ALADIN
ECMWF Forecaster
Felhozet
8
Verification results and trends
Complex Score for the first day 2000. júl 2004.
dec. (Country mean) ALADIN ECMWF Synoptic
9
Minimum Temperature sesonal ME and RMSE (2001
2004 Hungary mean)
EC Forecaster
ME seasonal RMSE slowly decrease
10
Complex Score for the first day in 2004-ben
(country mean) ALADIN ECMWF Forecaster
11
Case study I. Inversion and cold air
pad. (2004. December )
MSG comp321 sat. picture. 12.12. 12 2004 UTC and
observations
ECMWF model forecast. 12 UTC-re dec.12 2004
100
20
SKC is predicted
The last ECMWF prediction.
Tops of mountains are clear
Stratus (cloud base 200-400 m, surface
temperature 1, -3 C
12
Cloud cover verification. ECMWF - ALADIN
Forecaster (1st day) 2004. October November
December
V
4
AL
EC
F
3
2
1
0
-1
-2
-3
(okta)
-4
-5
-6
-7
okt. 1.
okt. 8.
nov. 5.
dec. 3.
okt. 15.
okt. 22.
okt. 29.
nov. 12.
nov. 19.
nov. 26.
dec. 10.
dec. 17.
dec. 24.
dec. 31.
13
Case study II. Flash flood in Matra
Mountains. (18.04.2005. )
The ECMWF model involved the possibility of
extreme precipitation
14
Case study II.
Succesful MM5 modell forecast based on ECMWF
initial condition.
15
New clusterisation method and early time torch
construction depend on ECMWF at HMS.
Varying number of clusters for the Carpathian
Basin
Torch construction before the dissemination time.
16
New torch diagrams and probability forecasts for
energetic sector.
925 hPa,850 hPa,700 hPa, wind direction ,relativ
hum.,temperture, torches
Probablilty forecasts
17
Post processing application.

• Program demonstration
• Internet and business application
• End

18
(No Transcript)
19
(No Transcript)
20
Thank you for attention !