Deutscher Wetterdienst

Slide 1 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Field Experiments for NWP:

The LITFASS Experience

Frank Beyrich

Deutscher Wetterdienst

Meteorologisches Observatorium Lindenberg – Richard-Aßmann-Observatorium

Deutscher Wetterdienst

Slide 2 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Observations for NWP ….

data assimilation

model verification parameterization development

test of sensors, systems, and

measurement strategies

© C. Heret (TU Dresden)

Deutscher Wetterdienst

Slide 3 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Routine

operation

Field

campaign

Data assimilation X (X)

Model verification X (X)

Parameterization

development

(X) X

Test of sensors, systems,

measurement strategies

(X) X

Observations for NWP ….

Deutscher Wetterdienst

Slide 4 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Classical field campaigns

Micrometeorological / ABL experiments • Great Plains / O‘Neill (1953) • Wangara (1967) • Kansas (1968) • Minnesota (1973)

Monin-Obukhov Similarity Theory (MOST) universal functions

Deutscher Wetterdienst

Slide 5 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Classical field campaigns

Micrometeorological / ABL experiments • Instrumented tower (T, q, V) • Sonic anemometer • Frequent radiosondes • Tethered balloon and later: • sodar, lidar … • aircraft • several sites raw data usually stored

Deutscher Wetterdienst

Slide 6 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Supersites and Observatories

Many of these measurements are operational nowadays … Supersites and observatories are nothing more than an operational field experiment …

Deutscher Wetterdienst

Slide 7 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Supersites and Observatories

… broad spectrum of in-situ and remote sensing instruments

… even raw data often stored

… cover the full spectrum of weather situations and atmospheric conditions at a given site over a longer time period

© DWD (R. Leinweber)

Deutscher Wetterdienst

Slide 8 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

MOL-RAO contributions to NWP

(Long-term) validation on NWP model output

© DWD (C. Becker)

Deutscher Wetterdienst

Slide 9 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Cluster Analysis / Conditional Validation: CBL Evolution

Goal: Analyse CBL evolution with / without cloud formation Case selection: conditional data base request Classification according to season, stability, radiation, wind Calculation of mean diurnal cycles for each class Comparison data vs. model

Clear sky (Weak to moderate wind, medium stability): 3 different radiation classes

MOL-RAO contributions to NWP

© DWD (C. Becker)

Deutscher Wetterdienst

Slide 10 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Deutscher Wetterdienst

Modification of COSMO Cloud Scheme Using Cloudnet Data

Large discontinuity in cloud fraction

distribution above 5 km

Large discontinuity in cloud fraction

distribution above 5 km Large discontinuity in cloud fraction

distribution above 5 km

Mean cloud fraction (red lines) and their

probability distribution function (in %) for

COSMODE

forecasts (left) and radar observations (middle).

The cloud fraction differences (in %) are shown

on

the right.

Parametrization scheme of stratiform clouds:

Correction of cloud fraction for subvisible ice clouds at levels p < 500 hPa

a, b: empirical parameters (a=0.2, b=5 x 10-5), qi:

ice water content (iwc)

a, b: empirical parameters (a=0.2, b=5 x 10-5),

qi: ice water content (iwc)

Verification based on radar derived

cloud fraction and ice water content.

Improved scheme (Opt) Parameters: a=0.1 and b=8x10-6

MOL-RAO contributions to NWP

© DWD (A. Seifert / U. Görsdorf)

Deutscher Wetterdienst

Slide 11 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Field Experiments at Observatories

• Observatories may provide climatology, core measurement program, infrastructure, and logistics for field experiments

© DWD (R. Leinweber)

Deutscher Wetterdienst

Slide 12 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Field Experiments at Observatories

The place of field experiments

• Process studies (special instrumentation)

• Spatial dimension (large number of sites)

• Test of measurement systems and strategies (large number of instruments of the same type)

• Non-automatic measurements (e.g., tethered balloons, aircraft)

© DWD (R. Leinweber)

Deutscher Wetterdienst

Slide 13 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Lindenberg Inhomogeneous Terrain: Fluxes between Atmosphere and Surface – a long-term Study

- Orographie

- Land-

nutzung

- Boden

- RR-

Netzwerk

- Radar

- QG -

Netzwerk

- Satelliten-

Daten

- Mast

- RaSo

- Sodar

- LAP

- TWP

- Mast

- RaSo

- RASS

- Mast

- RaSo

- MW-Rad.

- Photometer

- Lidar

Vertikalprofile V T q

Erdoberfläche Externe Nieder-

Parameter schlag Strahlung

MESSUNG/BEOBACHTUNG

DATEN

BANK

MODELL/SIMULATION

- 99 m Mast

- Sodar

- LAP

- EBS

- GM

- SLS-40

- Monitoring

- EBS

- Monitoring

- (Satelliten-

Daten)

- LAS

- Flugzeug

- Budget

Methode

Boden Grenz- Integrierte

schicht- schicht-

Flüsse

Vegetations-

parameter

Boden-

parameter

(TS, qS, T0, G)

DATEN

BANK

Antrieb

Verifikation

LLM

Energiebilanzstation

LAS

Pluvio mit Globalstrahlung

Pluvio

Hellmann

Pegel

LITFASS

Deutscher Wetterdienst

Slide 14 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

LITFASS-98 / LITFASS-2003

Deutscher Wetterdienst

Slide 15 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

LITFASS-2009

Deutscher Wetterdienst

Slide 16 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Uncertainties of turbulent fluxes from eddy-covariance measurements

with state-of-the-art sensors is 5-10 % for H and 10-15% for LE, resp.,

data processing algorithms may add additional uncertainty of the

same magnitude harmonization of algorithms

LITFASS: Selected Results

© DWD (J.-P. Leps)

© DWD (J.-P. Leps)

Deutscher Wetterdienst

Slide 17 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

0

5

10

15

20

25

30

35

40

45

50

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Flux Ratio

Nu

mb

er

of C

ase

s

grass / farmland

rape / farmland

maize / farmland

cereals / farmland

Measurements over grass at GM Falkenberg can be considered as

representative for farmland

LITFASS: Selected Results

Deutscher Wetterdienst

Slide 18 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

LITFASS: Selected Results

Scintillometry is a technique able to provide area-representative

turbulent flux estimates at NWP grid scale

-50

0

50

100

150

200

250

300

-50 0 50 100 150 200 250 300

H source-area [W m -2 ]

H LA

S [W

m

-2 ]

1:1

y = 1.16x R 2 = 0.8

© W.M.L. Meijninger (Wageningen Univ.)

Deutscher Wetterdienst

Slide 19 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

© F. Ament (Univ. Bonn)

LITFASS: Selected Results

The tile and mosaic approaches are suitable strategies to take into

account land-surface heterogeneity in an NWP model grid cell

Deutscher Wetterdienst

Slide 20 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

© DWD (B. Stiller)

LITFASS: Selected Results

An estimate of the regional variablity of incoming shortwave radiation

can be obtained from the temporal variability measured at a given site.

Deutscher Wetterdienst

Slide 21 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Outlook

The place of field experiments

• Process studies: Cold pools, gusts, and coherent structures

• Spatial dimension: sub-mesoscale

• Test of measurement systems and strategies: Doppler lidar scanning strategies

• Non-automatic measurements: aircraft

© R. Leinweber

Deutscher Wetterdienst

Slide 22 of 22 Workshop „Observational campaigns for better weather forecasts“, Reading , 11.6.2019

Thank you for your attention!