Orographic gravity waves: Lessons learnt from the DEEPWAVE field campaign

Transcription

1 Orographic gravity waves: Lessons learnt from the DEEPWAVE field campaign Andreas Dörnbrack DLR Oberpfaffenhofen Photo: Sonja Gisinger Photo: Petr Horalek

2 Hines (1974) 1. Motivation 2. DEEPWAVE Field Campaign 3. GW propagation from the Troposphere to the Stratosphere Radiosonde analyses 4. GW propagation from the Stratosphere to the Mesosphere Ground-based Lidar observations 5. Conclusions

3 1. Motivation - Internal gravity waves appear nearly everywhere in the atmosphere are often poorly represented in NWP and climate models

4

5 Prusa J. M., Smolarkiewicz P. K., Garcia R. R., 1996: On the propagation and breaking at high altitudes of gravity waves excited by tropospheric forcing. J. Atmos. Sci., 53,

6

7 z ~ 2 km z ~ 90 km Convective Waves above a single power plant, Northern Germany, 20 January km Baumgarten, G., and D. C. Fritts (2014), Quantifying Kelvin- Helmholtz instability dynamics observed in noctilucent clouds: 1. Methods and observations, J. Geophys. Res. Atmos., 119, , doi: /2014jd

8

9 Baseline: Θ (K) and w (ms -1 ) h m =100m, free slip (4h) Nearly identical results for all models (with exception of BLASIUS at upper levels due to anelastic assumptions)

10 Ex1000_fs: Θ (K) and w (ms -1 ) h m =1000 m, free slip (4h) Most models exhibit trapped waves (variations in number of crests) A few models show very weak vertical velocity (ASAM, RAMS )

11 Ex2500_fs: Θ (K) and w (ms -1 ) h m =2500 m, free slip (4h) A subset of models with a very strong windstorm, breaking response A few models show a much weaker response (ASAM, UM, RAMS)

12 The most uncertain aspect of climate modelling lies in the representation of unresolved (subgrid scale) processes such as clouds, convection, and boundary-layer and gravity-wave drag, and its sensitive interaction with large-scale dynamics. The divergence of model projections that arises from model errors means that it is essential to work towards reducing those errors, which are presumably associated with inadequate parameterizations of unresolved processes.

13 2. DEEPWAVE Field Campaign The Deep Propagating Gravity Wave Experiment (DEEPWAVE): An Airborne and Ground-Based Exploration of Gravity Wave Propagation and Effects from their Sources throughout the Lower and Middle Atmosphere David C. Fritts 1, Ronald B. Smith 2, Michael J. Taylor 3, James D. Doyle 4, Stephen D. Eckermann 5, Andreas Dörnbrack 6, Markus Rapp 6, Bifford P. Williams 1, P.-Dominique Pautet 3, Katrina Bossert 1, Neal R. Criddle 3, Carolyn A. Reynolds 4, P. Alex Reinecke 4, Michael Uddstrom 7, Michael J. Revell 7, Richard Turner 7, Bernd Kaifler 6, Johannes S. Wagner 6, Tyler Mixa 1, Christopher G. Kruse 2, Alison D. Nugent 2, Campbell D. Watson 2, Sonja Gisinger 6, Steven M. Smith 8, Ruth S. Lieberman 1, Brian Laughman 1, James J. Moore 9, William O. Brown 9, Julie A. Haggerty 9, Alison Rockwell 9, Gregory J. Stossmeister 9, Steven F. Williams 9, Gonzalo Hernandez 10, Damian J. Murphy 11, Andrew R. Klekociuk 11, Iain M. Reid 12, and Jun Ma 13 Bull. Am. Meteorol. Soc. (2015), in press

14 2. DEEPWAVE Field Campaign OBJECTIVES: - study dynamical coupling processes by gravity waves from the troposphere into the stratosphere and mesosphere by characterizing the complete life cycle of gravity waves: gravity wave excitation, propagation, and dissipation employing observational and modelling tools - improve GW parameterizations for use in general circulation models BMBF Research Initiative: ROMIC (Role of the Middle atmosphere In Climate) DFG Research Group: MSGwaves (Multiscale Dynamics of Gravity Waves)

15 Stratospheric GW hotspots Northern hemispheric winter Southern hemispheric winter Focus of GW-LCYCLE 2013, 2015/16 Jiang et al., 2003 US funded NFS project DEEPWAVE (moved from South America)

16 New Zealand: Gravity Wave Hot Spot in SH winter AIRS RMS Temperature at 2.5 hpa for June/July

17 Kim et al., 2003

18

19 150 hpa 5 hpa 12 July UTC 13 July 09 NZST 300 hpa 700 hpa

20 RF Jul 21 UTC 09 local

21 Horizontal average of horizontal wind over the South Island/NZ ECMWF T1279/L137 operational analyses (6 h) and 1 hourly high-resolution IFS predictions V HOR /ms -1

22 Horizontal average of horizontal wind over the South Island/NZ IOP1,2 IOP3,4 IOP5 IOP9 IOP6,7 IOP8 IOP10 IOP13 IOP14 IOP16 ECMWF T1279/L137 operational analyses (6 h) and 1 hourly high-resolution IFS predictions V HOR /ms -1

23 3. From the Troposphere to the Stratosphere Radiosonde analyses from Lauder/NZ (45 S, 169 E) o 98 soundings in total o mean height reached: 31.1 km o maximum height reached: 36.6 km Sonja Gisinger, DLR

24 3 hourly radiosonde profiles V HOR /ms -1 N 2 /s -2 Sonja Gisinger, DLR

25 Brunt-Väisälä Frequency N 2 and Tropopause Height above Lauder/NZ + radiosonde data ECMWF T1279/L137 operational analyses (6 h) and 1 hourly high-resolution IFS predictions Sonja Gisinger, DLR

26 Estimation of Gravity Wave Parameters solid lines: RS profiles dashed lines: background profiles Mean magnitude of horizontal wind perturbations: 6 8 m/s Mean magnitude of vertical wind perturbations: 1 2 m/s Perturbations mstrato: mid stratosphere lstrato: low stratosphere tropo: troposphere

27 Energies sensitive to different parts of the GW spectrum KE: sensitive to low frequency waves/inertial gravity waves VE: sensitive to high frequency gravity waves PE: mixed Sonja Gisinger, DLR

28 Gravity Wave Energies Troposphere (1 8 km) KE VE PE Sonja Gisinger, DLR

29 Gravity Wave Energies Lower Stratosphere (13 20 km) KE VE PE Sonja Gisinger, DLR

30 Gravity Wave Energies Mid Stratosphere (20 27 km) KE VE PE Sonja Gisinger, DLR

31 Vertical Energy (VE) Ratios tropo/lstrato lstrato/mstrato Ratio < 1: VE higher in upper layer input of GW energy in upper layer Ratio > 1: VE lower in upper layer GW dissipation, no conservative propagation Sonja Gisinger, DLR

32 Kinetic Energy (KE) Ratios tropo/lstrato lstrato/mstrato Ratio < 1: KE higher in upper layer input of GW energy in upper layer Ratio > 1: KE lower in upper layer GW dissipation, no conservative propagation Sonja Gisinger, DLR

33 Results: completely different behaviour of VE and KE : > tropopause regions seems to be a source/an amplifier of low frequency gravity waves (KE) and a filter for high-frequency waves Sonja Gisinger, DLR

34 Johannes Wagner, DLR

35 Johannes Wagner, DLR

36 Johannes Wagner, DLR

37 4. From the Stratosphere to the Mesosphere Ground-based Lidar observations DLR Rayleigh Lidar at Lauder Photo: Bernd Kaifler

38 Bernd & Natalie Kaifler, Benedikt Ehard, DLR DLR Rayleigh Lidar at Lauder Resolutions Raw profiles Δz= 10 m Δt = 10 ms

39 Rayleigh lidar and radiosonde daily mean temperature at Lauder, NZ Bernd & Natalie Kaifler, Benedikt Ehard, DLR

40 Questions How can we detect mountain waves in lidar data? What conditions are needed for deep gravity wave propagation?

41 Bernd & Natalie Kaifler, Benedikt Ehard, DLR Derivation of T' and E p from temperature profiles Temperature T Removal of background temperature T 0 by polynomial fitting method Temperature perturbation T' exp(z / 2.3H ) E p Gravity wave potential energy density:

42 Bernd & Natalie Kaifler, Benedikt Ehard, DLR Gravity wave potential energy density large variability modulation with a period of 1-3 weeks no deep propagation of large-amplitude mountain waves

43 Mountain Waves on 1 August 2014 (GB21) Bernd & Natalie Kaifler, Benedikt Ehard, DLR

44

45 1 hpa 300 hpa 1 August UTC 10 hpa 700 hpa

46 Mountain Waves on 1 August 2014 (GB21) High tropospheric wind speed Enhanced stratospheric E p Stationary waves with short vertical wavelength (~ 6 km) Bernd & Natalie Kaifler, Benedikt Ehard, DLR

47 Mountain Waves on 1 August 2014 (GB21) Benedikt Ehard, DLR

48 Mountain Waves on 1 August 2014 (GB21) Daily mean Temperature LIDAR ECMWF Background Temperature Benedikt Ehard, DLR

49 Mountain Waves on 1 August 2014 (GB21) Benedikt Ehard, DLR

50 Mountain Waves (?) on 4 July 2014 (IOP 10) Benedikt Ehard, DLR

51 Mountain Waves (?) on 4 July 2014 (IOP 10) Daily mean Temperature LIDAR ECMWF Background Temperature Benedikt Ehard, DLR

52 Mountain Waves (?) on 4 July 2014 (IOP 10) Benedikt Ehard, DLR

53 Distinction between GW types using 2d wavelets (I) Temperature perturbation (K) 2d wavelet analysis Directional 2d Morlet wavelets Wavelet spectrogram Wavelet scale log( variance ) (K 2 ) Wavelet Angle Kaifler, B., N. Kaifler, B. Ehard, A. Dörnbrack, M. Rapp, D. C. Fritts, 2015: Influences of source conditions on mountain wave penetration into the stratosphere and mesosphere, Geoph. Res. Letters, submitted

54 Distinction between GW types using 2d wavelets (I) Temperature perturbation (K) 2d wavelet analysis Directional 2d Morlet wavelets Variance normalized Wavelet spectrogram Back transformation Wavelet scale log( variance ) (K 2 ) Wavelet Angle Kaifler, B., N. Kaifler, B. Ehard, A. Dörnbrack, M. Rapp, D. C. Fritts, 2015: Influences of source conditions on mountain wave penetration into the stratosphere and mesosphere, Geoph. Res. Letters, submitted

55 Distinction between GW types using 2d wavelets (II) Wavelet scale small large Wavelet spectrogram Upwardpropagating GW Wavelet Angle Quasi-stationary GW = MW Downwardpropagating GW Partial back transformation log( variance ) (K 2 ) Kaifler, B., N. Kaifler, B. Ehard, A. Dörnbrack, M. Rapp, D. C. Fritts, 2015: Influences of source conditions on mountain wave penetration into the stratosphere and mesosphere, Geoph. Res. Letters, submitted

56 Distinction between GW types using 2d wavelets (III) mesosphere Upwardpropagating GW stratopause stratosphere Quasi-stationary GW = MW E p Downwardpropagating GW Estimate of EP for - MW, upward and downward propagating GW - every 3 h - 3 altitude ranges Kaifler, B., N. Kaifler, B. Ehard, A. Dörnbrack, M. Rapp, D. C. Fritts, 2015: Influences of source conditions on mountain wave penetration into the stratosphere and mesosphere, Geoph. Res. Letters, submitted

57 Lauder GW statistics Quasi-stationary GW = MW Upward-propagating GW Downward-propagating GW mesosphere stratopause stratosphere Kaifler, B., N. Kaifler, B. Ehard, A. Dörnbrack, M. Rapp, D. C. Fritts, 2015: Influences of source conditions on mountain wave penetration into the stratosphere and mesosphere, Geoph. Res. Letters, submitted

58 Correlation between stratospheric mountain wave E p and tropospheric forcing Cross-mountain flow at 1 km level Simple relationship: The stronger the forcing, the larger mountain waves energies in the stratosphere Kaifler, B., N. Kaifler, B. Ehard, A. Dörnbrack, M. Rapp, D. C. Fritts, 2015: Influences of source conditions on mountain wave penetration into the stratosphere and mesosphere, Geoph. Res. Letters, submitted

59 Correlation between mesospheric mountain wave E p and tropospheric forcing Cross-mountain flow at 1 km level Cross-mountain flow km Deep MW propagation occurs under condition of weak to moderate forcing and sufficiently stronger stratospheric winds Kaifler, B., N. Kaifler, B. Ehard, A. Dörnbrack, M. Rapp, D. C. Fritts, 2015: Influences of source conditions on mountain wave penetration into the stratosphere and mesosphere, Geoph. Res. Letters, submitted

60 5. Summary o surprisingly good agreement of observed gravity waves with ECMWF s IFS o deep vertical propagation of gravity waves depends critically on the magnitude of the stratospheric flow in an altitude range between 25 and 40 km o large-amplitude mountain waves in the stratosphere during strong tropospheric forcing o weak to moderate forcing and sufficiently stronger stratospheric winds needed for deep GW propagation o other sources not yet considered in NWP models: - GW-tide interaction may be an efficient secondary source for GWs - non-orographic GWs excited by polar night jet

61

62 Biff Williams & Dave Fritts, GATS

63

64

65

66 large deceleration of jet near exit/split region Intense Polar Jet

67 Thank you! Especially, ECMWF staff for excellent organization of the Annual Seminar 2015! Flower ducks, Insel Mainau, July 2015, Sonja Gisinger

68

69 Polar Stratospheric Clouds Above Scandinavia PSCs above Finland as seen from DLR Falcon, 26 Jan 2000

70 Kiruna, SWE Sodankylä, FIN MODIS 15 March UT

71 Lidar Backscatter Ratio at 1064 nm from MM5 hindcast Queney, 1948 Norwegian Sea Sweden Baltic Sea Finland 26 Jan UT -1-4 Ro=U/fL 15ms -1 /(10-4 s km)=0.6 Mesoscale T-anomalies generated by hydrostatic mountain waves hor MAX ~ km ~ 2000 m T ~ K, T MIN ~ 175K ( - 98 o C ) dt/dt < -50 K/h, t proc ~ 5.5 h

72 DLR Falcon 25 Jan SOLVE/THESEO 2001

73 DLR Falcon 25 Jan SOLVE/THESEO 2001 Ri = N 2/ S 2< 0.25