1 / 33

Joint APSDEU-11/NAEDEX-23 Data Exchange Meeting (Boulder 2011)

This report discusses the status of the Deutscher Wetterdienst (DWD) and their numerical weather prediction models. It covers various global and local developments in weather forecasting and data assimilation techniques.

mcnaughton
Download Presentation

Joint APSDEU-11/NAEDEX-23 Data Exchange Meeting (Boulder 2011)

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. Joint APSDEU-11/NAEDEX-23 Data Exchange Meeting (Boulder 2011) Deutscher Wetterdienst (DWD) status report Alexander Cress Deutscher Wetterdienst, Frankfurter Strasse 135, 6003 Offenbach am Main, Germany alexander.cress@dwd.de

  2. Numerical Weather Prediction at DWD Global model GME Grid spacing: 30 km Layers: 60 Forecast range: 174 h at 00 and 12 UTC 48 h at 06 and 18 UTC 1 grid element: 778 km2 COSMO-EU Grid spacing: 7 km Layers: 40 Forecast range: 78 h at 00 and 12 UTC 48 h at 06 and 18 UTC 1 grid element: 49 km2 COSMO-DE Grid spacing: 2.8 km Layers: 50 Forecast range: 21 h at 00, 03, 06, 09, 12, 15, 18, 21 UTC 1 grid element: 8 km2 COSMO-DE EPS Pre-operational 20 members Grid spacing: 2.8 km Variations in: lateral boundaries, initial conditions, physics

  3. Assimilation schemes • Global: 3DVAR PSAS • Minimzation in observation space • Wavelet representation of B-Matrix • seperable 1D+2D Approach • vertical: NMC derived covariances • horizontal: wavelet representation • Observation usage: Synop, Temp/Pilot, Dropsonde, AMV, Buoy, Scatterometer, AMUSU-A/B, Aircraft, Radio Occultation • Time window: 3 hours • Local: • Continous nudging scheme and latent heat nudging • Time windows: 0.5 – 1 hour • Observation usage: Synop, Temp/Pilot, Dropsonde, Buoy, • Aircraft, Scatterometer, Windprofiler, • Radar precipitation

  4. New developments since last meeting • Global: • Change from 40 km / 40 L to 30 km / 60 L • Revised background error correlations for new model • Use of Radio Occultation (bending angles) from Grace METOP/GRAS, COSMICS, TerraSar-X • AMVs from GOES 13 and MTSAT-2 • Use of stratospheric ballon observations over antartica • Global soil moisture analysis • Use of AMUS-B/MHS data (pre-operationally) • Local: • Use of a lake model (FLAKE) • New sea/lake ice model • Humidity bias correction for radiosondes • VAD and GPS water vapour measurements(pre-operationally) • COSMO ART (Aerosol Transport Model) • COSMO-Climate (climate version of COSMO)

  5. GME 30 km / L60 31 forecasts from 01.-31.01.2010 ANOC pmsl NH BIAS pmsl NH GME 30km / L60 GME 40km / L40 GME 40km / L40 GME 30km / L60

  6. Use of GPS - radio occultation (bending angles) in the 3DVar-Assimilation of GME (since 03. Aug. 2010) • Advantages of GPS radio occultations (bending angles) • high vertical resolution  even vertical thinning of data required! • globally accessible, approximately equally spaced • not influenced by clouds • measurement of the bending angle is almost bias free, temporally stable, independent from the instrument • number of profiles is proportional to the product of the sending GNSS-satellites (GPS, Galileo, GLONASS) and receiving LEOs: • CHAMP, GRACE-A (research satellites) • FORMOSAT-3 / COSMIC ( 6 research satellites) • GRAS (Metop-A) • TerraSar (H. Anlauf, DWD)

  7. Use of GPS - radio occultation in the 3DVar-Assimilation of GME (~2000 Obs/day) geopotential in 500 hPa: anomaly correlation of southern hemisphere for July 2010 Strong impact on the SH Smaller impact on the NH Impact strongest in the stratosphere Large impact on temperature Minor impact on humidity, strongest in the upper troposhere Visible impact also on sea level pressure

  8. Monitoring of new GPS radio occulation systems Statistics for bending angles (January 2011) COSMIC-C/NOFS COSMIC-SAC-C

  9. Aircraft temperature bias correction • Aircraft temperatures have biases • Bias problem largest at flight level • Biases different for different flight phases • Biases different for different aircrafts • Stdev is small and independent of • temperature bias • Bias is fairly constant for each aircraft • If we can remove the bias than more aircrafts will pass the QC • Better quality of temperature analysis and forecasts ?

  10. Aircraft temperature bias correction Experiment using a dynamic/automatic bias correction scheme different for individual aircraft and flight phases Aircraft profiles Obs – FG (full line) Obs – Ana (dashed) Northern Hemisphere Sept. 2010 • Positive temperature bias in upper troposphere (0.1 – 0.3 K) is removed • RMS in biascor.experiment is smaller • 10% - 15% more aircraft temperatures are used

  11. Aircraft temperature bias correction Radiosonde Verification of OBS – FG (full) and OBS – ANA (dashed) Control (blue) Experiment (red) Northern Hemisphere Sept. 2010 • Negative temperature bias in upper troposphere is reduced (~0.2 K) • RMS in biascor.experiment is slightly reduced in aircraft flight level • 0.2% - 0.6% less radiosonde temperatures are used • Overall neutral forecast impact

  12. Aircraft humidity sensor • Humidity sensor name: WVSS-II • Manufacturer: Spectra Sensors Inc. (USA) • Sensor principle is based on Beers Law: • Transmittance (T) is a function of absorption T=I/Io = e –σlN • where l and lo are the intensity (power) of the incident light and the transmitted light, respectively; σ is cross section • of light absorption by a single particle, N is the density of absorbing particles • Infrared Absorption Spectrometry “2f-Method by use of Tunable Diode Laser (TDL) • TDL scans a water vapor absorption band near 1.37 um • Path length: 23 cm • Generates every 2 seconds a measurement • Measures the water vapor mixing ration

  13. Relative Humidity [%] bias statistics Aircraft: SESWUUZA Mean: 6.56 RMSE: 20.90 Aircraft: IKWWUUBA Mean: -3.77 RMSE: 10.04

  14. Relative Humidity [%] bias statistics US Aircrafts (new sensor) Mean: -0.50 RMSE: 3.87 LH Aircrafts (old sensor) Mean: -12.79 RMSE: 17.06

  15. AMDAR relative humidity statistics Obs minus FG/ANA Date: 2010090100 - 2010093021 straight line: active dashed line: used Area: North 75.0 South 35.0 West -160 East: -50.0

  16. Radiosonde relative humidity statistics Comparison between GME and Exp: 7817 (used) Date: 2010090100 - 2010093021 straight line: OBS-FG dashed line: OBS-ANA Area: North 75.0 South 35.0 West -160 East: -50.0

  17. Forecast Scores N_America rel. humidity 700 hPa

  18. Concordia experiment Measurement of meteorological conditions in the lower stratosphere with the help of stratospheric super-pressure ballons Ballons are loaded with temperature, wind and ozone sensors plus a particle counter and GPS receiver Measuring in flight level * some ballons Carried a gondola with dropsondes Measurement campaign over Antartica in autumm/winter 2010 Meteorological data over GTS in BUFR as ACARS Monitoring and impact experiment Use in Routine since Oktober 2010

  19. Observation coverage

  20. Monitoring results Gondola observations 2010090803 - 2010093021 Temperature [K] U-wind component [m/s] Bias: -2.81 RMS: 3.01 COR: 0.41 Bias: 0.31 RMS: 2.59 COR: 0.49 • Strong temperature bias, probably model bias • Small wind bias • Use of Gondola Temperature and wind measurements in Experiment

  21. Radiosonde Verifikation • Bias (left); RMS (right) • Antartic region • Comparison of Routine (red) against • Experiment using stratospheric balloon • measurements • Results: • Temperature- and Windspeedbias is • reduced over Antartica in the lower • stratosphere • RMS of temperature is reduced • considerably for both, OBS minus FG • and OBS minus Ana

  22. GNSS = Global Navigation Satellite System GPS (USA) GLONASS (Russland) GALILEO (EU) ... Delay of a signal due to the atmosphere Total Delay (integrated value) can be measured by calculating the time dalay between sending and receiving signal Wet part of the delay is proportional to integrated water vapour Assimilation of GNSS based integrated water vapour content Troller, M.R. (2004): "'GPS based Determination of the Integrated and Spatially Distributed Water Vapor in the Troposphere.”

  23. http://oflxs04.dwd.de/~keichler/GPS/html/ • Stationsnetz Europa • Stationen Modellgebiet • http://egvap.dmi.dk/

  24. precipitation verification Assimilation Scores (02. – 31.07.2009) • Fraction Skill Score - moving average over 5 GP • RR > 0.1 mm • Rot = Control • Blau =Crtl + GNSS • Equitable Thread Score • Consistent improvement of the scores throughout the assimialtion in summer • Frequency Bias

  25. Start und Ende der 21 h Vorhersage im Tagesgang besser Scores zum großen Teil besser precipitation verification00 UTC forecast for 21 hScores und daily course (02. – 31.07.2009) • Equitable Thread Score • Frequency Bias • Daily course • RR > 0.1 mm • Rot = Control • Blau = Crtl + GNSS • Fraction Skill Score – moving average over 201 GP • Fraction Skill Score - moving average over 5 GP • Mainly positive results (scores are better with GNSS)

  26. Concept: COSMO-ARTis online coupled. Identical methods are applied for all scalars as temperature, humidity, and concentrations of gases and aerosols to calculate the transport processes. This includes the treatment of deep convection (Tiedtke Scheme) and dry and wet deposition It has a modular structure. Modelling the Volcanic Ash Episode:Experiences with COSMO-ART COSMO–ART(ART = Aerosols and Reactive Trace Gases) Ext. Parameters = operational weather forecast model (DWD) Vogel et al., 2009, ACP

  27. Requirements to describe theEmission of Volcanic AshSource strength of particles Vertical distribution of the effective emission heights Size distribution of particles at the source Density of the particles … Forecast of these variables None of those requirements were fulfilled when we started our work 27

  28. Concentration field of 1 µm particlesfor 16.04.2010 09 UTC (after 57 h) in model layer 14 (approx. 7500 m): Satellite observations at roughly the same time: J. Förstner – COSMO-User Seminar, 02.03.2011 28

  29. LMU Lidar data

  30. During the first days of the eruption volcanic ash was injected into the atmosphere up to 11 km. A comparison of the simulated ash-plume with the satellite picture shows:The model captures the horizontal distribution of the ash-plume quite well. The simulation results show the capability of an operational weather forecast model that is extended by aerosol processes to simulate the spatial and temporal distribution of volcanic ash. As the source strength will not be known in future eruption events only a combination of ground based and satellite borne remote sensing instruments together with in-situ observations and model results facilitates the work of decision makers during future events. Summary and Preliminary Conclusions

  31. Future Plans ICON: Develop a non-hydrostatic global model with static local zooming option (ICON: ICOsahedral Non-hydrostatic; http://www.icon.enes.org/). Replace global model GME and regional model COSMO-EU by ICON with a high-resolution window over Europe. Dynamicla core of a Earth Model system at MPI Diverse numerical tests are conducted. Currently physical parametris. are build in and tested. KENDA: Km-scale ENsemble-based Data Assimilation (KENDA) Local Ensemble Transform Kalman Filter (LETKF) (Hundt et. al 2007) (because of its relatively low computational costs) Initial conditions for COSMO-DE_EPS First prototype using only conventional observations is implemented Global Model:EnKF/3DVAR Hybrid System is planned

  32. Future Plans • Use of IASI data in global and regional model • Use of SSM/I-SSM/IS data • Preparation for AEOLUS wind lidar observations • Develop a 3D radar oberator for radar reflectivities / radial velocities • Uso of ground-based GNSS slant delay observations • Develop cloud analysis based on conventional and satellite • observations • Assimilation of Aerosol observations in COSMO ART • Use of radiances over Land and/or cloudy conditions

  33. Thank you for your attention!Questions?

More Related