Validation and diagnostics tools used in countries

1
Validation and diagnostics tools used in
countries

• Maria Derkova
• Neva Pristov

HIRLAM-ALADIN workshop on physics and
diagnostics, Oslo, 12-13/12/2005
2
ALADIN validation tools

• ALADIN has specific output file format (FA),
  containing both spectral and gridpoint fields
• internal tools (available via/during model
  execution)
• fullpos
• movie
• echkevo
• external tools
• frodo, edf
• ecto
• pseudoTEMPs
• ASCS
• ALADIN verification project
• comment usually these tools have common software
  basis, but visualisation part adapted by each NMS

3
FULLPOS

• the post-processing package - our main diagnostic
  tool
• consistent with model equations (physics),
  interpolations, constants, climatology etc.
• allows transformation to various
• levels (p, z, PV, ?, model)
• domains grids (gauss, model, latlon, lelam)
• multiple fields
• available in-line or off-line

4
FULLPOS NORMS

• available in the output listing
• output frequency controled by namelists
• during integration
• SPECTRAL NORMS - SURFACE PRESSURE
  0.115013153860930E02 OROGRAPHY
  0.359236075962300E04
• LEV VORTICITY DIVERGENCE
  TEMPERATURE HUMIDITY
  KINETIC ENERGY
• AVE 0.590326020073889E-04 0.324345455324753E-04
  0.246393827139165E03 0.137562159965699E-02
  0.115190052015067E03
• specific to fullpos
• FULL-POS GPNORMS
• AVERAGE
  MINIMUM MAXIMUM
• SURFNEBUL.TOTALE/SHMU 0.478621864507254E00
  0.000000000000000E00 0.999759213216640E00
• CLSVENT.ZONAL /SHMU 0.341629148922581E00
  -.136312375001790E02 0.120025700924380E02
• CLSVENT.MERIDIEN/SHMU -.147638570158226E01
  -.181590558251937E02 0.126187709690653E02
• CLSTEMPERATURE /SHMU 0.278541842802290E03
  0.257494067681197E03 0.295056541570276E03
• CLSMINI.TEMPERAT/SHMU 0.278422883357361E03
  0.257322981260511E03 0.295056541570276E03

5
MOVIE

• pseudosatellite image simulated by the thermal
  radiation at the model top
• local file fort.91 is produced during model
  integration via namelist switches
• NAMPPC
• LMOVIE.TRUE.,
• LMOVPH.TRUE.,
• NAMCT0
• NFRISP2, (frequency of animation events in
  time-steps)

• the fort.91 file is read by HARPE program which
  creates ppm then mpeg movie is created from ppm
  files with mpeg_encode
• HARPE program (author P. LAMBOLEY) is controlled
  via namelist
• max size for pictures (pixels)
• number of pictures to generate, frequency
• picture files format RIG, RAW, PGM, PPM, GIF,
  ASC or BIN
• projection Cylindrique,Stereo
  polaire,Orthographique,Aladin
• selection of colors (palette)

6
MOVIE example in grey color (by N. Pristov)
7
(No Transcript)
8
ECHKEVO

• check/diagnostics of the gridpoint evolution
• enables to print out (into external FA file
  ICMSHEXPNCHKOUT2) the values of defined model
  variables at predefined points
• NAMCHK
• LECHKEVO.TRUE.,
• NFRQCHK, NGPCHK, NXCHK(i), NYCHK(i), NFLDCHK,
  NFLDCHK(j)
• needs external program to get ascii output
  (fort.70)
• visualisation (by e.g. xmgrace)
• does it work in CY28T3?

9
ECHKEVO example 1usage of digital filter

• time evolution of the surface pressure, point
  40,60

10
ECHKEVO example 2work of M. Tudor stability
testsPBL height diagnosed via Ri
11
ECHKEVO example 2Linear instability tests
acpluie(by M. Tudor)
reference
test
The pictures show time evolution of temperature
on the two lowest model levels and the difference
between them for one chosen point. The values for
each time-step are presented, x-axis represents
hours of the integration and y axis temperature
in Kelvin.
12
ECHKEVO example 2 (cont)Linear instability
tests - acpluie

• EVAP0, FONT0, evaporation and melting were
  switched off fibrillations completely
  disappeared
• FONT0, melting switched off, fibrillations
  occurred with the same intensity as in basic test
  
• EVAP0, evaporation switched off, fibrillations
  completely disappeared

13
ECHKEVO example 3 (cont)Linear instability
tests - acpluie
stability hidden in the evaporation of snow
computations of the acpluie gt decreasing the
ratio of the speed of evaporation of
precipitation ice/water REVGSL (80gt 4-15)
14
Comparison of single point values to
measurements(by M. Tudor)

• Comparison of the modelled 2m temperature
  evolution for 00 UTC run on 14th December 2004
  with measured data from synoptic station with
  operational radiation scheme (left) and including
  NER (right).

reference rand max rmnew RH XR cloud random old
RH
15
PALADIN

• Portable Auxillary Library And Development tools
  for aladIN
• package by J.-D. Gril
• portable version of fa/lfi and eggx tools
  modules, makefiles
• contains various utilities (frodo, edf, ecto,
  pseudo, domolalo, coneo)
• http//www.cnrm.meteo.fr/aladin/concept/tools.html
  

16
Tools to check the FA file

• FRODO checks the header of the file
• geometry, dates, vertical levels, fields
• usage FRODO FA_file
• EDF (Edit Field) prints out the values at the
  gridpoints (even of spectral fields!)
• control via namelist (fields, levels, zoom)
• usage EDF namelist FA_file(s)
• example of output
• LON LAT SURFIND.TERREMER
  SPECSURFGEOPOTEN S022TEMPERATURE
• -10.07 44.95 0.0000000
  -13.1869965 279.9895866
• -9.92 44.97 0.0000000
  21.8817933
  279.9214952
• -9.76 44.99 0.0000000
  -8.8118075
  279.8732009
• ...............
• http//www.cnrm.meteo.fr/aladin/concept/docEDF.v1.
  10

17
ECTOplasme

• computes spectral energy for the effective wave
  number over the slices of the ellipse
• control via namelist (variables, levels, wave
  numbers)
• usage ecto namelist FA_file(s)
• output is the ascii file x, y1 (y2, y3...) to
  be visualised (e.g. by xmgrace)
• http//www.cnrm.meteo.fr/aladin/concept/docECTO.v1
  .10

18
ECTOplasme examples
19
pseudo

• production of the parts B and D of the
  prognostic TEMP (pseudoTEMP, psTEMP) messages for
  predefined set of points (file GEOIN)
• usage pseudo list_of_files
• example
• IIBB SKBRA 56069 98000 99482 10172 17987 01771
  00981 02806 11979 02808 22974 03610 33965 03810
  44953 03608 55939 03205 66922 02604 77902 01805
  88880 01006 99856 00108 11830 01709 22802 03310
  33773 05112 44742 06514 55710 08117 66677 10122
  77643 12727 88608 15728 99572 18726 11536 21726
  22500 25128 33463 29131 44427 33736 55391 38739
  66355 44542 77320 51339 88286 57939 99253 58161
  11220 54575 22189 53195 33159 54199 44131 56791
  55105 59387 21212 00981 02506 11979 02507 22974
  03009 33965 04011 44953 05011 55939 05512 66922
  06012 77902 07012 88880 07511 99856 08010 11830
  08509 22802 09509 33773 10509 44742 12510 55710
  14012 66677 15014 77643 15514 88608 16015 99572
  17018 11536 18023 22500 18529 33463 18536 44427
  19042 55391 19047 66355 20054 77320 20062 88286
  21067 99253 22058 11220 23041 22189 23032 33159
  24030 44131 24529 55105 25531
• IIDD SKBRA 5606/ 98000 99482 10172 17987 01771
  11812 60786 22589 62386 33389 63786 44212 63791
  55050 367// 21212 11812 26532 22589 27537 33389
  28541 44212 28055 55050 27157
• further processing usually based on HRID
  program
• time/z cross-section
• vertical and time interpolations
• computation of additional derived energy,
  thermodynamics and stability parameters
• visualisation (originally NCAR-based)
• advantage if HRID processing/visualisation is
  used also for TEMP

20
psTEMP HRID visualisation cold front passage
(by J. Masek)
21
psTEMP TEMP HRID visualisation validation of
new model version (J. Masek M. Derkova)
psTEMP old_oper
psTEMP new_oper
TEMP
station Vienna 19/03/2005
22
psTEMP TEMP HRID visualisation validation of
new model version (J. Masek M. Derkova)
TEMP
psTEMP old_oper
psTEMP new_oper
station Vienna 19/03/2005
23
ASCS

• ALADIN VERTICAL (SPACE) CROSS SECTION
• needs input postprocessed to model levels,
  containing surface geopotential, the model levels
  geopotential, plus any other field on model
  levels. All these fields should be in
  grid-points, i.e. special fullpos has to be run
  prior to ASCS
• ASCS plots these fields/variables on the vertical
  plane between two selected geographical points
  (portion of the circle on the geographical
  sphere)
• driven by namelist with possible tcl/tk interface
• (originally) based on ncar graphics

24
ASCS example of bura wind event in Adriatic sea
(by N. Pristov)
25
SLHD case study of the false forecast of the
Adriatic cyclone (by F. Vana)
26
(No Transcript)
27
ASCS case study of the false forecast of the
Adriatic cyclone (by F. Vana A. Simon)
28
ASCS 19/11/2004 case study of the High Tatras
severe windstorm - downslope wind (by A. Simon
J. Vivoda)
29
ALADIN Verification Project (AVP)seen as a tool
for model validation

• common ALADIN project operated in Slovenia
  ww.arso.gov.si/verification/
• data from operational models and parallel suites
  collected
• client software installed at NMSs, data sent via
  e-mail stored in database
• web user interface to plot the scores (predefined
  set of bulletins once per month planned)
• currently in testing mode (performance problems)

30
AVP example of possible usage

• ALADIN EPS multimodel meteogram
• for selected station
• for selected variables

31
AVP example of various models forecasts