WG3a activities in cloud microphysics and radiation

1
WG3a activities in cloud microphysics and
radiation

• Ulrich Blahak, for the large part on behalf of
  Axel SeifertGerman Weather Service, Offenbach

2
Outline

• Microphysics
• Activities in the last year
• Short term plans
• Long term plans
• The same for radiation
• and revised cloud radiation coupling

3
Microphysics recent work

• Improvement of supercooled LWC in mixed phase
  clouds (F. Rieper, motivated by air craft icing
  forecasts at DWD)
• Development of improved treatment of cirrus ice
  clouds for current one-moment schemes (PhD of C.
  Köhler, DWD)
• Development of improved snow melting
  parameterization (PhD of C. Frick, DWD)
• Interface to 2-moment scheme in official Version
  code itself and necessary lookup table files
  available on DWD ftp server. Still missing User
  Guide

4
Improvement of Supercooled LWC

• Much too less supercooled LWC in mixed phase
  clouds
• Changes to the microphysics
• Reduced diagnostic number of ice particles as
  function of T (before, it was unplausibly too
  high), therefore reduction of Bergeron-Findeisen-p
  rocess in mixed phase clouds
• Reduction of rain freezing rates (before, it was
  unplausibly too high)
• Parameterization of subgrid scale effects on
  liquid water layers on top of ice cloud layers.

5
Result reduced underprediction
January 18, 2013 Lindenberg
Aerosol
Observation
Ice SLW
Ice
COSMO (old)
COSMO
COSMO (new Ni)
Slides by F. Rieper, DWD
6
Result impact on supercooled LWP
January 18, 2013 Lindenberg
Observation
good retrieval
COSMO-EU (old)
COSMO-EU (new NiSLW)
COSMO-EU (new Ni)
Slides by F. Rieper, DWD
7
Ice nucleation current status
Homogeneous Freezing
Heterogeneous Freezing
Cloud droplet
Liquid Aerosol ( )
Immersion-
Deposition-
Condensation-
Contact-
Thomas Leisner, Institut für Meteorologie und
Klimaforschung FZ Karlsruhe und Institut für
Umweltphysik, Uni Heidelberg (modified)
Slides by C. Köhler, DWD
8
Changes in the Cloud Ice Microphysics

• Implementation of state-of-the-art
  parameterisations for homogeneous and
  heterogeneous nucleation
• New prognostic model variables
• 2-moment cloud ice scheme (still monodisperse
  size distribution)
• Changes in treatment of depositional growth for
  cloud ice and snow
• Limitation of heterogeneous nucleation with
  prognostic budget variable for activated ice
  nuclei
• Introduction of cloud ice sedimentation

Slides by C. Köhler, DWD
DLR 06/06/2011
9
COSMO-DE Model
MSG/SEVIRI
COSMO-Ice
COSMO Modell

• Implemented in COSMO
• Version 4.18

Slides by C. Köhler, DWD
10
GME Model

• Overestimation of ice water content (IWC) of the
  GME is reduced
• RMSE reduced with the new cloud ice nucleation
  scheme

Slides by C. Köhler, DWD
11
GME Model

• Ice supersaturation improved due to new
  depositional growth
• In-cloud ice supersaturation is allowed for. IFS
  assumes in-cloud saturation, which is not
  compliant with in situ measurements

Slides by C. Köhler, DWD
12
New snow melting scheme

• New prognostic snowflake water content
• Enables better parameterization of the melting
  process
• Leads to slower melting, more in accordance with
  observations

Old
New
(Frick et al. 2013, GMDD, under review)
13
New snow melting scheme
Meltwater on snow flakes, same section as slide
before
Different day, different region
One can have partially melted snow on the ground.
Affects prediction of precipitation phase on the
ground in winter time!
(Frick et al. 2013, GMDD, under review)
14
Microphysics plans

• Short term (2013 2017)
• Common COSMO / ICON physics library
• Investigate dry precipitation bias in summer time
• Further testing of the works of C. Köhler and C.
  Frick
• Continuing work of F. Rieper regarding
  supercooled LWC, demand of aviation forecasters
  (aircraft icing)
• 2-moment scheme further studies to evaluate
  benefit in operational NWP consistent data
  assimilation, otherwise we will most likely not
  see the full benefit
• Towards explicit hail forecasting with 2-moment
  scheme
• improvement of hail melting / shedding following
  methodology of C. Frick (Postdoc work of V. Sant
  in HD(CP)2 , Hamburg)
• Resolution requirements?
• Longer term (2018 2020)
• Keep an eye on scientific developments regarding
  subgrid scale processes in cloud microphysics
  parameterization (effects of turbulence and
  spatial inhomogeneity)

15
Microphysics recent publications

• Frick, C., A. Seifert, and H. Wernli, A bulk
  parameterization of melting snowflakes with
  explicit liquid water fraction for the COSMO
  model version 4.14, Geosci. Model Dev. Discuss.,
  6, 2927-2966, 2013
• Köhler, C Cloud ice particle nucleation and
  atmospheric ice supersaturation in numerical
  weather prediction models, PhD thesis, 2013
• Seifert, A., Köhler, C., and Beheng, K. D.
  Aerosol-cloud-precipitation effects over Germany
  as simulated by a convective-scale numerical
  weather prediction model, Atmos. Chem. Phys., 12,
  709-725, doi10.5194/acp-12-709-2012, 2012
• Naumann, A. K., Seifert, A., and Mellado, J. P.
  A refined statistical cloud closure using
  double-Gaussian probability density functions,
  Geosci. Model Dev. Discuss., 6, 1085-1125,
  doi10.5194/gmdd-6-1085-2013, 2013

16
Radiation recent work

• New modis derived total albedo (Frank Brenner,
  DWD)
• External parameters for orographic radiation
  corrections finally available in new EXTPAR

17
Radiation plans

• Short Term (2013 2017)
• Common COSMO / ICON physics library
• Test parameterization of 3D radiation effects
  (work from Uni Munich within Extramural Research
  program), TICA, McICA
• Test McSI Monte Carlo Spectral Integration
  methods (B. Ritter)
• Revision of cloud - radiation feedback (U.
  Blahak)
• Test alternative scheme RRTM for COSMO (available
  via common physics library)
• Consistent coupling with ICON
• RG92 has problems for domain heights gt 25 km
• Longer Term (2015 2017)
• Likely Revision of cloud - radiation feedback
  cont.

18
Revision of cloud radiation feedback

• In the COSMO radiation scheme (Ritter Geleyn
  1992), the optical properties (extinction coeff.,
  single scattering albedo, asymmetry factor) only
  depend on QC respectively QI.
• Nowadays more modern parameterizations based on
  an effective radius Re are available.
• From inherent assumptions about N(D) and particle
  shapes in state-of-the-art microphysical models
  Re can be deduced.
• Then optical properties fct (QX, Re, ?)
• For the gridscale clouds, only a fraction k0.5
  of QI and/or QC is considered in the radiation
  scheme (tentative effective factor to take into
  account subgrid scale variability). Test larger,
  more realistic value for todays models.
• Radiation scheme is only aware of cloud droplets
  and cloud ice! Extend to precipitating species
  (most importantly snow)

19
Revision of cloud-radiation feedback
Total QS factor 5 and more !
Total QI
20
Revision of cloud-radiation feedback
Case study COSMO-DE, 13.6.2012, 00UTC run, domain
average T_2M
Control Tegen aerosols hydrometeor
radiative properties depending on Reff qg,
qs radqfact 0.5 -gt 0.75
Potential big impact, but it looks not always so
nice! Careful re-tuning of the model necessary.
Current model tuned to compensate the cloud
radiative bias!
- - - - Synop Obs.