1 / 39

Ján Kaňák 1 , Richard Habrovsk ý 1 Jozef Csapl ár 1 , Alois Sokol 2

Implementation of EUMETSAT GII/RII software to SHM Ú ALADIN operational suite and first validation results. Ján Kaňák 1 , Richard Habrovsk ý 1 Jozef Csapl ár 1 , Alois Sokol 2 1 Slovak Hydrometeorological Institute, Jeséniova 17, 833 15 Bratislava, Slovakia, e-mail: Jan.Kanak @shmu.sk

albina
Download Presentation

Ján Kaňák 1 , Richard Habrovsk ý 1 Jozef Csapl ár 1 , Alois Sokol 2

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Implementation of EUMETSAT GII/RII software to SHMÚ ALADIN operational suite and first validation results Ján Kaňák1, Richard Habrovský1Jozef Csaplár1, Alois Sokol2 1Slovak Hydrometeorological Institute, Jeséniova 17, 833 15 Bratislava, Slovakia, e-mail: Jan.Kanak@shmu.sk 2Comenius University, FMFI, Mlynská dolina, 842 48 Bratislava, Slovakia, e-mail: lojzo.s@gmail.com EUMETSAT Convective Workshop, Landshut, Germany, 8-10 October 2009

  2. Parameters used for diagnostics of instability of the atmosphere The most precise information about temperature and humidity vertical profiles that are used for determination of stability of the atmosphere comes from aerological measurements. Unfortunately density of aerological stations provides only very rare information which can be used only locally or inside of certain airmass. In recent years new possibility in determination of atmospheric stability is to use multispectral data from geostationary satellite - measured brightness temperatures by IR channels to retrieve temperature and humidity profiles andto compute instability parameters in regular high resolution grid

  3. What is GII software GII is software offered by EUMETSAT and written by M. König. Aim of this programme is: -to use MSG IR brightness temperatures and NWP outputs as a first guess to determine realtime temperature and humidity profiles of the atmosphere -use determined profiles to compute indices of atmospheric instability GII version uses GLOBAL NWP temperature and humidity profiles RII version uses REGIONAL data based on outputs from limited area NWP models (Aladin, Slovakia domain) RII resolution is closer to MSG pixel resolution (3x3, 1x1) and local model provides us with more accurate forecast data

  4. Parameters suitable/considerable for instability diagnostics of the atmosphere There is several classical indices for which the satellite derived profiles can be used: Indices available by GII software, already validated in Central Europe conditions (IMGW): • K-index • Lifted index • Total precipitable water content (Running operationaly at SHMÚ) Microburst indices which have to be tested in Central Europe conditions: • WI (Wind Index) • TeD (Theta-e Difference) • WMSI (Wet Microburst Sevirity Index (Plannedto be implemented at SHMÚ, tested and validated) Definitions of microburst indices can be find here: http://www.star.nesdis.noaa.gov/smcd/opdb/aviation/mb.html Because all indices are empirical and depend on regionaland climate characteristicsit is necessary to test all potentialindices and find whether they are useful or not in given region.

  5. GII software overview • is based on RTTOV Library (NWP SAF) where Radiative Transfere Model is implemented • NWP outputs (Temperature, Humidity, Pressure) are used as a first guess • 6 IR SEVIRI channels brightness temperatures as input to radiative transfere model • The basis of the method is 1D-var retrieval iterative algorithm • Cloud mask is used to determine clear atmosphere where only the retrieval can be performed • When temperature and humidity profiles are retrieved indeces are computed

  6. IR channels available from MSG SEVIRI instrument and used as input to retrieve temperature and humidity profiles Ch6: 7,3 µm Ch7: 8,7 µm Ch5: 6,2 µm Ch9: 10,8 µm Ch10: 12,0 µm Ch11: 13,4 µm

  7. Normalised Weighting Functions for SEVIRI IR channels For details see: www.eumetsat.int MSG Interpretation Guide

  8. Details of local GII/RII Installation in Slovakia Model ALADIN inputs • Temperature field (3D)34 horizontal layers • Relative humidity (3D) 34 horizontal layers • Surface pressure (2D) • Surface temperature (2D) • Temperature in 2 m (2D) • Grid step0.049 deg (~5.5km) • Grid size400x250 points • Threshold of retrieval 1.5 K • Max number of retrieval iterations 4 • ALADIN domainlon 7.525E / 27.475E lat 41.275N / 53.725N

  9. Details of local GII/RII Installation in Slovakia Performance settings • Running in 1x1 MSG pixel resolution • every 15 minutes using MSG-2 and 1 hour Aladin inputsinterpolated into 15 minutes time slots • Processing on Intel Pentium-4 3GHz within 5 minutes

  10. Example of output of local SHMÚ GII/RII installation K-index and Total precipitable water contentcombined image for 25th June 2008 17:00 UTC High values of K-index and TPW are usualy overlapped.

  11. Preliminary validation results Methodology overview • Mesosynoptic analyses case by case • Comparison of K-index & TPW produced by GII against Aladin • Comparison of K-index & TPW against precipitation fields (radar, Inca analyses) • Comparison of K-index & TPW maxima against lightning intensity/density • Documentation of storm severity by in-situ evidence • 23 stormy days in 2008-2009 convection seasonover Slovakia/Central • Europe were indentified, reprocessed and are currently under validation. • List of selected cases: 1. 30.5.2008 2. 31.5.2008 3. 1.6.2008 4. 2.6.2008 5. 23.6.2008 6. 24.6.2008 7. 25.6.2008 8. 26.6.2008 9. 30.6.2008 10. 7.7.2008 11. 12.7.2008 12. 13.7.2008 13. 14.7.2008 14. 1.8.2008 15. 14.8.2008 16. 15.8.2008 17. 7.9.2008 18. 11.5.2009 19. 11.6.2009 20. 16.6.2009 21. 18.6.2009 22. 6.7.2009 23. 18.7.2009 K & TPW combined imagery is available as .avi files here: http://www.shmu.sk/sk/?page=1525

  12. Preliminary validation results Results of mesosynoptic analyses – part 1 • in high terrain areas • lower TPW values • low level water vapor missing • generally higher K values • higher T850 over terrain than in free atmosphere • high values of K-index and TPW are usualy overlapped • except mountains areas in some cases • very high K values (over 40) near cloud edges

  13. Preliminary validation results Results of mesosynoptic analyses – part 2 • continual real-time TPW and K-index fields development • almost all thunderstorms in high K and TPW areas • mesoscale boundaries detection based on vertical stability and water content • for example 1.6.2008 = case of „masked“ boundary in surface air temperature field • we can predict thunderstorm potential for next development (if t-storm already exists)

  14. Preliminary validation results Results of mezosynoptic analyses – part 3 • high K and TPW values does not guarantee thunderstorm occurence • there are many other factors which influence thunderstorm developing and especially trigerring • vertical wind shear • frontal boundaries • insolation • orography • we can expect thunderstorm occurence in high values areas but we can´t determine where exactly it will be • K index is very sensitive to relative humidity in middle troposphere • but lack of water vapor in these levels provides strong downbursts potential • comparison with TPW could be usefull

  15. Preliminary validation results Comparison of K-index produced by GII against Aladin K-index by ALADIN model (left) and byGII/RII software(right) for 25 June 2008 16:00 UTC: Critical differences are highlighted by cyan circles.

  16. Comparison of K-index produced by GII against Aladin 25 June 2008 09:00 1

  17. Comparison of K-index produced by GII against Aladin 25 June 2008 12:00 2 In general, pictures look similar… High values of K-index

  18. Comparison of K-index produced by GII against Aladin 25 June 2008 15:00 3 K-index values from Model are too low

  19. Comparison of K-index produced by GII against Aladin 25 June 2008 18:00 4 Cold front is moving to the east and convection intensifies

  20. Comparison of K-index produced by GII against Aladin 25 June 2008 21:00 5 While model is under- estimating K-index front is reaching Slovak territory

  21. Comparison of K-index produced by GII against Aladin 26 June 2008 00:00 6 Front is moving to the east…

  22. Comparison of K-index produced by GII against Aladin 26 June 2008 03:00 7 Convection is dissipating K-index values are lower

  23. Comparison of K-index produced by GII against Aladin 26 June 2008 06:00 8 Clouds are dissipating, satellite data can be used again…

  24. Comparison of K-index produced by GII against Aladin 26 June 2008 09:00 9 Values of K-index and TPW are quite high, airmass is still very moist and unstable

  25. Comparison of K-index produced by GII against Aladin 26 June 2008 12:00 10 Satellite shows very High values of K-index In Slovakia again…

  26. Comparison of K-index produced by GII against Aladin 26 June 2008 15:00 11 Satellite K-index values are growing up…

  27. Comparison of K-index produced by GII against Aladin 26 June 2008 18:00 12 Convection is intensifying…

  28. Comparison of K-index produced by GII against Aladin 26 June 2008 21:00 13 Convection is intensifying…

  29. Comparison of K-index produced by GII against Aladin 27 June 2008 00:00 14 Convection is intensifying…

  30. Preliminary validation results Case 25.-26.June 2008 – damages by strong winds in Bratislava Documentation of storm severity by in-situ evidence

  31. Preliminary validation results • Comparison of TPW against precipitation fields • Input data: • Raingauge measurements • Radar reflectivity, precipitation intensity & cumulated precipitation • Integrated precipitation data sources • Aim is to localise and isolate high precipitation amounts and to find • corresponding respons in GII products (TPW fields). Examples of localised heavy precipitation fields See summary table of results on the next slide.

  32. Preliminary validation results Comparison of K-index & TPW against precipitation fields Note: Values of K & TPW were readout some slots before development of convection started. In prevalent number of cases high values Of K-index and TPW indicated high preci- pitation amounts (highlighted by red). But there are also cases when no signi- ficant rain amount was detected however K-index and TPW values indicated severe event (highlighted by blue).

  33. Preliminary validation results Comparison of K-index & TPW maxima against lightning intensity/density Input data: Flash data from SHMÚ lightning detection system Aim is to localise and isolate intensive storm activity and to find corresponding respons in GII products (K and TPW fields) (Example and table of results on next slides)

  34. Comparison of K-index field against lightning intensity – case 1 June 2008 1 June 2008 10:30 UTC 1 June 2008 15:30 – after 5 hours Potentially severe convection area Lightning: Green– intracloud Red– “+” to ground Violet- “-” to ground • Total number of flashes: • Intracloud – 3700 • Pozitive to ground – 350 • Negative to ground - 200

  35. Preliminary validation results Comparison of K-index & TPW against lightning intensity Note: Values of K & TPW were readout some slots before development of convection started. Occurance of high lightning intensity is connected with high K-index values more then with high TPW values

  36. Conclusions • K - index based on GII/RII can control model forecast against real development • TPW content – is potentially important to control possible heavy rain intensities and useful because this parameter is not produced by our model • Reliability of satellite derived indices very close to cloudiness is limiting factor in practical use • Quite hard task to find convenient cases in given region for validation Future plans • Continue in collecting of severe convection cases and in validation activities • Implement and validate potentially useful new indices which can be derived on the base of satellite data • Convert satellite derived indices into grib format for more comfortable validation work • Use more precise colour scales and/or more exact method mapping maximum values of indices

  37. Acknowledgments Program code for computation of instability indices was developed by Marianne König from EUMETSAT and offered to SHMÚ. Many thanks to Marianne König also for guidance and number of priceless tips during adaptation of GII code to ALADIN Slovakia domain. Thanks to student of Comenius University, FMFI Bratislava Peter Hruška for data handling and processing of imagery materials.

  38. References Eyre, J.R., 1991: A fast radiative transfer model for satellite sounding systems. ECMWF. Technical Memorandum No. 176, 28pp. König M., 2002: Atmospheric Instability Parameters Derived from MSG SEVIRI Observations. EUMETSAT, Technical Memorandum No. 9. König M., Pajek M., Struzik P., 2007: MSG global instability indices for storm nowcasting – validation studies on product quality and analysis of sensibility to input model data. König M., 2007: GII: Global Instability Indices. EUMETSAT, Madrid Workshop on Physical Retrievals.

  39. References www.ecmwf.int/newsevents/training/ /meteorological_presentations/MET_DA.html - M.Fisher: Assimilation algoritms - T.McNally: Analysis of satelite data (1,2) - M.Matricardi: Infrared Radiative transfer K.N.Liou: An introduction to Atmospheric Radiation, Academic Press,2002 X.L.Ma,T.J Schmith, W.L.Schmith: A nonlinear physical retrieval algorithm – its application to the GOES-8/9 sounder. , Jour. of Appl.Meteor.38,501-513,1999

More Related