Simulations with the DWD Lokal-Modell at different horizontal - - PowerPoint PPT Presentation

▶
simulations with the dwd lokal modell at different
SMART_READER_LITE
LIVE PREVIEW

Simulations with the DWD Lokal-Modell at different horizontal - - PowerPoint PPT Presentation

Sep 08, 2022 122 likes •255 views

WP4200 Simulations with the DWD Lokal-Modell at different horizontal resolutions Case studies during CNN2 and BCC Felix Ament, Susanne Crewell (University of Bonn) Model domains and selected cases of WP4200 a) CNN1: b) CNN2, BBC: 02 Aug.

slide-1
SLIDE 1

Simulations with the DWD Lokal-Modell at different horizontal resolutions

Case studies during CNN2 and BCC

Felix Ament, Susanne Crewell (University of Bonn)

WP4200

slide-2
SLIDE 2

Model domains and selected cases of WP4200

a) CNN1:

  • 02 Aug. 2000
  • 25 Aug. 2000
  • 04 Sep. 2000

b) CNN2, BBC:

  • 13 April 2001
  • 02 Aug. 2001
  • 23 Sep. 2001
  • size of model domain: 400x400km
  • horizontal grid spacing: 7.0, (5.5), 2.8, (2.2), 1.1 km
  • no additional information at small scales
slide-3
SLIDE 3

Available model data

+36h +12h +0h spin-up analysis Station output time series ∆t=40s..10s Standard output 3d-fields every hour

area ave.

Special analysis

  • mean quantities 5min
  • mean profiles 10min
  • cell detection 10min

time

slide-4
SLIDE 4

Example: 13. April 2001

average over model domain and 24h

50 100 150 200

Water vapour 6.5kg/m2 Ice 3.4g/m2 Liquid water 7.6g/m2 Cloud cover 59% Radiation 80W/m2 Evapo. 99W/m2 Rain 0.9mm/d

50 100 150 200 250 300

50 100 150 200

50 100 150 200 50 100 150 200 50 100 150 200 50 100 150 200

50 100 150 200

Budgets and fluxes I

Heatflux 36W/m2

Reference values:

7km run without convection scheme

relative deviations

runs with 7, 2.8 and 1.1 km grid spacings

Results:

  • water vapour,

cloud cover and surface fluxes remain unchanged

  • LWP and rain rate

increase due to refinement

Results:

  • water vapour,

cloud cover and surface fluxes remain unchanged

  • LWP and rain rate

increase due to refinement

slide-5
SLIDE 5

Water vapour 6.5kg/m2 Ice 3.4g/m2 Liquid water

7.6g/m2

Cloud cover 59% Radiation 80W/m2 Evapo. 99W/m2 Rain 0.9mm/d Heatflux 36W/m2

  • 13. April 2001

Water vapour 27.9kg/m2 Ice 7.5g/m2 Liquid water 21.2g/m2 Cloud cover 83% Radiation 138W/m2 Evapo. 99W/m2 Heatflux 20W/m2

02 Aug.2001

Water vapour 19,5kg/m2 Ice 1.7g/m2 Liquid water 57g/m2 Cloud cover 86% Radiation 40W/m2 Evapo. 60W/m2 Rain 3.4mm/d Heatflux 18W/m2

23 Sep.2001

Water vapour 28kg/m2 Liquid water 47g/m2 Cloud cover 74% Radiation 108W/m2 Evapo. 91W/m2 Rain 4.7mm/d Heatflux 0.6W/m2

2 Aug.2000

Water vapour 17kg/m2 Liquid water 43g/m2 Cloud cover 73% Radiation 78W/m2 Evapo. 54W/m2 Rain 0.3mm/d Heatflux 32W/m2

25 Aug.2000

Water vapour 17kg/m2 Liquid water 90g/m2 Cloud cover 76% Radiation 15W/m2 Evapo. 53W/m2 Rain 4.4mm/d Heatflux 10W/m2

4 Sep.2000

Budgets and fluxes II

Rain 0.5mm/d

7, 2.8, 1.1km

slide-6
SLIDE 6

Convection scheme on/off - influence on averaged quantities

Differences: Small deviations: No effect:

  • water vapour
  • cloud cover
  • radiative fluxes
  • sensible heat flux
  • latent heat flux

10 20 30 40 50 60 70 13 Apr 01 02 Aug 01 23 Sep 01 1 2 3 4 5 6 7 8 9 13 Apr 01 02 Aug 01 23 Sep 01 1 2 3 4 5 6 7 8 9 10 13 Apr 01 02 Aug 01 23 Sep 01

7, 2.8, 1.1km

no convection convection

LWP [g/m2] IWP [g/m2] RR [mm/d]

slide-7
SLIDE 7

IWV - measurement and simulation

Example: 02 August 2001, Station Cabauw Extreme wet air is advected during the afternoon from the south to Cabauw.

slide-8
SLIDE 8

Direct LWP-rain effect

grid refinement more LWP variations more RAIN ! nonlinear LWP-RR relation

Result of LM cloud scheme using idealized cloud profiles

slide-9
SLIDE 9

Direct LWP-rain effect

Example 14 April 01

25 50 75 100 125 150 175 200

averaged rainfall

Domain average Cabauw

Probably poor statistic!

slide-10
SLIDE 10

Comparison of LWP time series

microwave radiometer - model output

⇒ no better match, but statistic is improved! CNN1

slide-11
SLIDE 11

No rule without exception :-)

Example: 23 Sep 01

25 50 75 100 125 150 175 200

averaged rainfall

Domain average

but

==> additional mechanism:

  • vertical cloud structure, humidity below cloud
  • ???
slide-12
SLIDE 12

Results of WP4200 - so far ...

  • boundary effects can be larger than refinement effect

==> use the same boundary conditions at all resolutions!

  • Average cloud water content and rain rates are the only

mean quantities affected by refinement

  • LM simulations without convection scheme produce

quite realistic LWP values

  • The size of resolved convective cells depends on the

grid spacing at scales larger than 1km.

slide-13
SLIDE 13

Interesting issues

Increase of LWP and rain

  • understanding of mechanisms
  • (validation with measurements)

Convection scheme

  • Partitioning of resolved and parameterized convection at different

grid spacings

  • vertical transport

Cloud cover

  • Is there really no refinement effect?

Cell shrinking

  • Which parameters control the cell size? Horizontal diffusion?