1 / 36

High-resolution modelling in mountainous areas: MAP results

High-resolution modelling in mountainous areas: MAP results. Evelyne Richard Laboratoire d’Aérologie CNRS / Univ. Paul Sabatier Toulouse, France. What are the skills of high-resolution models to forecast orographically influenced precipitation?

plato-moore
Download Presentation

High-resolution modelling in mountainous areas: MAP results

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. High-resolution modelling in mountainous areas:MAP results Evelyne Richard Laboratoire d’Aérologie CNRS / Univ. Paul Sabatier Toulouse, France

  2. What are the skills of high-resolution models to forecast orographically influenced precipitation? Does explicit (versus parameterized) convection lead to a gain in predictability? IOP 2a – 17 September 1999 A short, intense, isolated, convective event 70 mm within 12 hours Sensitivity experiments performed with Meso-NH

  3. MAP IOP 2a: IR Meteosat

  4. MAP target area

  5. Composite radar reflectivity @ z = 2km MAP – IOP 2A 18:00 UT 19:00 UT 20:00 UT 250 km 22:00 UT 23:00 UT 21:00 UT

  6. Observation Simulation (Δx =2km) Toce Ticino watershed 20:00 UT 23:00 UT Reflectivity @ 2000m

  7. ECMWF: OP. ANA 1999 RADAR SIMULATION 12 hour accumulated precipitation

  8. Z > 60 dBz Composite radar reflectivity @ z = 2km 17:00 UT 18:00 UT 250 km 19:00 UT 20:00 UT

  9. Radar Retrieval (S-Pol) Simulation (Meso-NH) 18:00 UT 12 km 19:00 UT 20:00 UT (x) hail + graupel graupel hail (o) hail rain rain 100 km

  10. Great ! My model is doing a good job

  11. RADAR OBSERVATIONS ECMWF: OP. ANA 1999 ARPEGE: OP. ANA. 1999 ECMWF: OP. ANA. 2002 ECMWF: REANALYSIS MAP - IOP2A: Intense Convection Strong sensitivity to initial state Low predictability

  12. ANA. OP. 1999 REANALYSIS REANA (NO MAP DATA) IOP 2a 850hPa water vapor mixing ratio : 17 September 1999 12UTC

  13. Lascaux et al., 2004 MAP IOP 2a Streamlines at 1000 m, 17/09/99 12 UTC ECMWF ANALYSIS MAP ECMWF REANALYSIS Low-level convergence between the Ligurian and Adriatic flows Increase in the model resolution -> higher mountains -> the Ligurian flow is blocked

  14. Prectitability ? Still a long way to go !

  15. How do the models compare with each other ? IOP2b 20/21 September 1999: Orographic enhancement of a frontal system 200 mm within 30h Model intercomparison : MC2, MM5, MOLOCH, Meso-NH

  16. 19 Sept. 1999 12:00 • 20 Sept. 1999 12:00 METEOSAT infrared

  17. Sensitivity to the analysis ECMWF Op. Analysis MAP Reanalysis Max: 512 mm Max: 482 mm Mean: 78 mm Mean: 87 mm

  18. The different models: • MESO-NH 10 KM + 2.5 KM • MOLOCH 10 KM + 2 KM • MM5-RE 27 KM + 9 KM + 3 KM • MM5-E1 18 KM + 6 KM + 2 KM • MC2 40 KM + 10 KM +2 KM Initial and boundary conditions from ECMWF operational analyses From 19 Sep. 12 UTC to 20 Sep. 18 UTC (30 hours)

  19. Toce-Ticino watershed MAP - IOP2B - 19-20 Sep. 1999 Intercomparison exercise 4 non-hydrostatic models with horizontal resolution of 2 to 3 km Initialization based upon ECMWF operational analysis Accumulated precipitation from the 19th 15 UTC to the 20th 18UTC

  20. Time evolution of the mean hourly precipitation rate Rain gauges Radar

  21. Time evolution of the correlation coeffecient (wrt rain gauges)

  22. Heidke skill scores as a function of precip. class 1h precip. 27h precip.

  23. Comparison with rain gauge measurements (121 points)

  24. Hydrological response Toce watershed 1532 km2

  25. Grossi et al., 2004

  26. Grossi et al., 2004

  27. How does the flow over complex terrain modify the growth mechanisms of precipitation particles? • Three Doppler radars • Monte Lema – Ronsard – S Pol • Dual Doppler analysis • 3D wind fields rerievals • Microphysical retrievals Monte Lema S Pol Ronsard

  28. Blocked and stable case dry snow U/Nh < 1 wet snow light rain Unblocked and unstable case graupel riming U/Nh > 1 coalescence heavy rain Medina and Houze, 2003

  29. To what extend the models able to reproduce this contrasted behaviour in the microphysics ?

  30. IOP2A IOP8 Ice Snow Graupel Snow Hail Cloud Rain • IOP2a ( Strong convection) • - Deep system • Large amount of hail and • graupel IOP8 ( Stratiform event) - Shallow system - Large amount of snow Mean vertical distribution of the hydrometeors

  31. Dominant microphysical processes: IOP2a IOP8 ACCRETION of cloud droplets by raindrops MELTING-CONVERSION of the snow (into graupel) Growth of graupel by RIMING DEPOSITION on ice (and sublimation) AUTOCONVERSION of pristine ice Depletion of graupel by WET GROWTH of hail

  32. Conclusion: • IOP 2A (Predictability) • The use of non-hydrostatic high-resolution models will improve the precipitation forecast but only to some extend. • Further improvement is tied to the improvement of the model initial state • Adding mesoscale data in a global assimilation system is insufficient • Mesoscale data assimilation system • Limited area ensemble forecast • (MAP D-PHASE)

  33. IOP 2b (Model Intercomparison) • Very good consistency of the accumulated precipitation pattern • Model results over/under estimate the total precipitation by a factor ranging from +30% to -30% • The accuracy of the model precipitation is rather weak for the hourly rainfall but fairly reasonable for the precipitation accumulated over the 30h time period of the event • However model results are not yet accurate enough to be used for hydrological forecast on small watersheds

  34. Explicit microphysical schemes do provide fairly realistic results … • The contrasted microphysical behaviour between different IOPs is reasonably reproduced • Convective – flow over ----> • Strong riming and coalescence • Stratiform – blocked flow ----> • Melting of snow

  35. http://www.aero.obs-mip.fr/map/MAP_wgnum N. Asencio (2), R. Benoit (3), A. Buzzi (4), R. Ferretti (5), F. Lascaux (1), P. Malguzzi (4), S. Serafin (6), G. Zängl (7), J-F. Georgis (1), R. Ranzi (8), G. Grossi (8), N. Kouwen (9) (1) LA CNRS/UPS, Toulouse, France (2) CNRM, Météo-France, Toulouse, France (3) RPN, Montréal, Canada (4) ISAC, CNR, Bologna, Italy (5) University of L ’Aquila, Italy (6) University of Milano, Italy (7) University of Munich, Germany (8) University of Brescia, Italy (9) University of Waterloo, Canada

More Related