Winter Weather Research

1
Improvements to the Microphysical Schemes in WRF
aimed at Improving Short Term Forecasts of
Storms
Key Investigators Ed Brandes Andy
Heymsfield Greg Thompson Paul Field Roy
Rasmussen Kyoko Ikeda Bill Hall

Leveraged funding FAA, STEP, Water Cycle
2

• Motivation
• QPF
• Explicit storm forecasting in WRF relies on the
  microphysical parameterization to produce
  precipitation (no convective parameterization).
• Cold pool formation
• - Important role of microphysics in producing
  cold pools due to melting and evaporation and
  precipitation loading. Dynamics of storms
  strongly dependent on the cold pool formation,
  depth, intensity, etc.

3
Cold pool formation critical to the formation and
propagation of MCS
Noon
Early next morning
Directional shear
MCS cumulonimbus family
Cumulo- nimbus
Mesoscale downdraft
Moister further east
To first order, elevated solar heating
determines start position start time of
traveling convection
1000 km
4

• Microphysical Schemes We Are Addressing
• One moment scheme of Thompson et al. (2004,2006)
• Two moment schemes of Axel Seifert, Bill Hall
• Bin microphysical scheme of Istvan Geresdi
  (currently implemented into older version of WRF)

5

• Approach
• Improving representation of processes in
  microphysical schemes (Greg Thompson, Bill Hall,
  Roy Rasmussen, Istvan Geresdi)
• Collecting and analyzing observation data that
  can directly address key uncertainties in the
  schemes (Ed Brandes, Kyoko Ikeda, Andy
  Heymsfield, Paul Field)
• - Size Distribution of hydrometeors
• - Density of the hydrometeors
• - Terminal Velocity of hydrometeors
• Comparison to Case Studies (Roy Rasmussen, Kyoko
  Ikeda, Bill Hall, Greg Thompson)
• Comparison of bulk one moment and two moment
  schemes to the bin microphysical scheme (Roy,
  Istvan, Greg).
• Inter-comparison studies (WMO Cloud Modeling
  Workshop)

6
Presentations today 1. Ground based
observations talk Ed Brandes 2. Airborne
observations talk Andy Heymsfield
(Collecting and analyzing observation data that
can directly address key uncertainties in the
schemes) 3. Improving representation of
processes in microphysical schemes Greg
Thompson 4. Comparison to case studies,
comparison of bulk one moment and two moment
schemes to bin microphysics, WMO Cloud Modeling
inter-comparison studies Roy Rasmussen


7

• Future Studies
• Continue improving representation of processes in
  microphysical schemes
• Participate in the ICE-L field program to help
  address the large uncertainty in ice initiation
  in the schemes.
• Continue collecting and analyzing observation
  data that can directly address key uncertainties
  in the schemes
• Continue to compare to case studies
• Participate in the 2008 WMO Cloud Modeling
  Workshop
• Implement aerosols explicitly into the schemes
• Transfer upgraded microphysical schemes into WRF

8
Development of an Improved Microphysical
Parameterization

• Roy Rasmussen
• Greg Thompson
• Bill Hall
• Kyoko Ikeda
• NCAR
• Istvan Geresdi
• University of Pecs, Hungary

9
Detailed microphysical model

• The detailed microphysical model of Geresdi
  (1998) was implemented into the MM5 mesoscale
  model to conduct two-dimensional simulations of
  precipitation formation in a stably stratified
  cloud (Rasmussen et al. 2002, JAS)
• Simple bell-shaped mountain used to generate 6
  10 cm/s uplift over 100 km horizontal scale.
• Detailed microphysical model enhanced to include
  ice phase.

10
Detailed Microphysical Model

• Five different hydrometeor species simulated
  with 36 size bins for each species
• Hydrometeor species Water drops, pristine ice
  crystals, rimed ice crystals, snowflake
  aggregates, and graupel
• Moment conserving technique of Tsvion et al.
  (1987, 1999) implemented to prevent artificial
  broadening of the hydrometeor distributions by
  numerical diffusion.
• Interactions allowed between the various
  hydrometeor types

11
Detailed Microphysical Model (cont.)

• Cloud droplets initiated from a CCN spectra as a
  function of supersaturation typical of
  continental and maritime clouds
• Ice initiated via
• - Deposition or condensation freezing using
  Cooper scheme
• Freezing of drops via Biggs freezing mechanism
• Contact nucleation

12
Detailed Microphysical Model simulations
13
Maritime case, Cooper ice initiation
Cloud water mixing ratio (g/kg)
0.35 g/kg
Height (km)
14
Rain water mixing ratio (g/kg)
0.08 g/kg
Height (km)
15
Rimed ice mixing ratio (g/kg)
Height (km)
0.025 g/kg
16
Aggregated ice crystals mixing ratio (g/kg)
Height (km)
0.00007 g/kg
17
Graupel mixing ratio (g/kg)
Height (km)
0.0005 g/kg
18
Bulk Microphysical Model simulations
19
Nc 300 cm-3
Nc 50 cm-3
20
Mass dependent snow Y intercept
Temperature dependent snow Y intercept
21
Cloud Water Mixing Ratio
Rain Water Mixing Ratio
Ice Mixing Ratio
Ice Conc.
Snow Mixing Ratio
Graupel Mixing Ratio