Campaign data for parameterization tests: Examples from MAP‘99 VERTIKATOR’02, AWIATOR‘03 Hans Volkert, Thorsten Fehr, Christoph Kiemle, Oliver Reitebuch,

Campaign data for parameterization tests: Examples from MAP‘99VERTIKATOR’02, AWIATOR‘03

Hans Volkert, Thorsten Fehr, Christoph Kiemle,

Oliver Reitebuch, Arnold Tafferner and Martin Weißmann

DLR Oberpfaffenhofen, DInstitut für Physik der Atmosphäre

Evelyne RichardLaboratoire d‘Aérologie, CNRS & Université Paul Sabatier Toulouse, F

“Whenever possible, parameterizations should … be quantitatively validated against observations“ (Peixoto & Oort 1991)

Textbook knowledge:

High resolution …… simulation models need special data,

e.g from dedicated field campaigns

here: wind,boundary layer, precipitation microphysics

Which variables and processes?

The campaigns MAP-SOP 1999 Mesoscale Alpine Programme

Special Observing Period„Weather and Alps“ 7 Sept. – 15 Nov 1999

cf. Bougeault et al. 2001, BAMS 82, 433–462

wind QJ, Jan. 2003 B (No. 588), 129, 341-895 VERTIKATOR-02 Vertikaler Austausch, Trans- port und Orographie„Alpine Pumping“ July 2002, north of the Alps wind & precipitation AWIATOR-03 Wake Vortices at airports„Wake Vortex“ Aug. 2003, north of Pyrénées wind

MAP-SOP 1999

IOP-15 8 Nov.

north Föhn with waves

•

•

•

Rhone

Aosta

(model) orography (x=1 km)

1.67–2.00

km

aircraft triptych: B-A-C-D , 8 Nov. 1999

L

North Föhn

8 Nov. 1999

ARPEGE ana. 12 UT

500 hPa: flow (max. 30 m/s) geopot. (=40 gpm)

model: Meso-NH with four nests (32, 8, 2, 0.5 km)

Drop-sounding: uniform wind direction

ff

dd

T

rh

dry

moist

Three drop-sondes in 2mins.: reproducibility

130355

130455

130555

vertical velocity: trans-Alpine sections

Rhone

Aosta

Rhone

Valtellina

ob

s

obs+

sim(--)

F C E

+2

-2

18

legs

4 legs

10

legs

vertical velocity: 2 & 1/2 km simulation

2 km

½ km

Rhone

Valtellina

Rhone

Aosta

F C E

+2

-2

water vapour triptych: lidar-obs. vs. simulation

100

350

100

350

w(C130)

mean diff. < 10% --> backs retrieval (s.p.=0.7)

sim.

obs.

100 ppmv 200

VERTIKATOR 2002

8, 9, 19 July

Alpine pumping &generation of

thunderstorms

10 µm-System WINDDLR/CNRS/CNES/Meteo-Francevert. res. 250 mhor. res. 3 - 15 kmaccuracy 0.5 - 2 m/s

2 µm-SystemDLR/CTI-MAG1100 m3 - 10 km0.2 - 2 m/s

Airborne Doppler Lidar at DLR

Flight Track Falcon 8 July 2002

13:05 - 15:34 LT

Data processing with averaging 3 scanner revolutions: 10 - 11 km

12.0

0.0

1.5

3.0

4.5

6.0

7.5

9.0

10.5

6.0

0.0

1.0

2.0

3.0

4.0

5.0

altitude [km ASL] wind speed [m/s]

360

0

45

90

180

270

315

6.0

0.0

1.0

2.0

3.0

4.0

5.0

2000 20 40 60 80 100 120 140 160 180

altitude [km ASL] wind direction [deg]

distance [km]

east-westgradient in speed(> 4 km)

southwest

northerly winds along the Alps up to 2.3 km

WIND July 8, 2002, 14:10 - 14:27 LTTrack Bodensee Chiemsee

wind speed below 4 m/s(< 2.5 km)

MM5 by L. Gantner,Uni München

12

0

2

4567

910

6.0

0.0

1.0

2.0

3.0

4.0

5.0

2000 25 50 75 100 125 150 175

altitude [km ASL] wind speed [m/s]

360

0

45

90120

180

220

270300330

6.0

0.0

1.0

2.0

3.0

4.0

5.0

2000 25 50 75 100 125 150 175

altitude [km ASL] wind direction [deg]

distance [km]distance [km]

WIND and MM5 on July 8, 2002 Track Bodensee ChiemseeWIND: top, 14:10 - 14:27 LT; MM5: bottom, 14:00 LT

19 July: 2µm Lidar and vertical velocity

shallow

medium

steep

40 km

9 July:Bistatic

Doppler andpolarimetric

radar

steep

shallow

medium 12 min

later ...

shallow

medium

steep

12:30 UT

Meso-NH 2 km res.

surf. wind,all hydromet.,accum. precip.

storm gene-ration @ Alps

13:00 UT

Meso-NH 2 km res.



13:30 UT

Meso-NH 2 km res.



14:00 UT

Meso-NH 2 km res.



14:30 UT

Meso-NH 2 km res.



15:00 UT

Meso-NH 2 km res.



15:30 UT

Meso-NH 2 km res.



16:00 UT

Meso-NH 2 km res.



16:30 UT

Meso-NH 2 km res.



17:00 UT

Meso-NH 2 km res.



17:30 UT

Meso-NH 2 km res.



18:00 UT

Meso-NH 2 km res.



18:30 UT

Meso-NH 2 km res.



19:00 UT

Meso-NH 2 km res.



19:30 UT

Meso-NH 2 km res.



20:00 UT

Meso-NH 2 km res.



20:30 UT

Meso-NH 2 km res.



21:00 UT

Meso-NH 2 km res.



21:30 UT

Meso-NH 2 km res.



AWIATOR 2003

27 August

Diurnal cycle of surface profiles

RASS vs. LM & MM5

Tarbes Airport

Airport

Tarbes

Pau

The Pyrenees

LM forecast domain (DWD) MM5 forecast domain

0 2 4 6 8 10 12 14 16 18 m/s

2000

1800

1600

1400

1200

1000

800

600

400

LM

MM5WakePredictor

P2P

MM5 vertical grid

Modelling chain: LM–MM5–P2P

00 UT

virt. temperature

wind speed

01 UT

virt. temperature

wind speed

02 UT

virt. temperature

wind speed

03 UT

virt. temperature

wind speed

04 UT

virt. temperature

wind speed

05 UT

virt. temperature

wind speed

06 UT

virt. temperature

wind speed

07 UT

virt. temperature

wind speed

08 UT

virt. temperature

wind speed

09 UT

virt. temperature

wind speed

10 UT

virt. temperature

wind speed

11 UT

virt. temperature

wind speed

12 UT

virt. temperature

wind speed

13 UT

virt. temperature

wind speed

14 UT

virt. temperature

wind speed

15 UT

virt. temperature

wind speed

16 UT

virt. temperature

wind speed

17 UT

virt. temperature

wind speed

18 UT

virt. temperature

wind speed

19 UT

virt. temperature

wind speed

20 UT

virt. temperature

wind speed

21 UT

virt. temperature

wind speed

22 UT

virt. temperature

wind speed

23 UT

virt. temperature

wind speed

24 UT

virt. temperature

wind speed

Concluding messages

Potentially useful, non-standard data are available for parameterization test ofhigh-resolution models;here: remotely sensed wind and precipitationA suitable number of case-studies may bemore revealing than standard statistics

Modelling and measuring camps have tocome closer together

DANKE for listening to me !!!

Campaign data for parameterization tests: Examples from MAP‘99 VERTIKATOR’02, AWIATOR‘03 Hans Volkert, Thorsten Fehr, Christoph Kiemle, Oliver Reitebuch,

Documents