Geen diatitel

1

Ice-Cloud effective particle size
parameterization based on combined lidar, radar
reflectivity, and mean Radar Doppler velocity
Data

• Rational
• Brief outline of lidar/radar procedure.
• Theory (relationship between Doppler-vel and
  effective sizes)
• Application to ARM data.
• Behavior of size-dist with IWC and Temperature
• Conclusions

2
Rational
Large-scale models can not capture cloud
microphysics so parameterization are necessary.

• For cirrus clouds
• ReffF(T) or Reff(IWC) or ReffF(T,IWC)
• Previous studies have mainly used in-situ
  aircraft measurements
• Various studies have show different behavior
• Limited coverage
• Limitations of instrumentation

• Active remote sensing
• More indirect
• Much greater coverage/sampling !

3
Active (lidar/radar) cloud remote sensing
Radar
Lidar
l 3-100mm
l 350-1100nm
Difference in returns is a function of particle
size !!
4
Effective size for ice crystals
Exact treatment of scattering difficult
(impossible?)However

• Ice particles are large compared to llid (Optical
  scattering regime)
• Ice particles are small compared to lrad
  (Rayleigh scattering regime)

Confirmed using DDA and RT calculations
5
Lidar/Radar Inversion Procedure
Can be problematic
6
Add Doppler Velocity....So we have.
and

• Vt is calculated using the work of Mitchell
  (1996) where
• it is a function of
• D
• Mass(D)
• Area(D)
• Temperature
• Pressure

We want to infer !
7
D-vs-R'eff-vs-Reff and Vd (For single particles )
Habit relationships (or similar form)
8
Ice crystal size-dist model
From aircraft measurements it is known that
Ice-crystal size-distributions are generally
multi-modal
Model of Ivanova/Mitchel two-mode gamma
with fixed' small-mode parameters
Increasing rl
Increasing Nl
Small mode parameters
1
9
ARM-SGP data
Want to examine average behavior
with Temperature and IWC
10
Direct point by-point inversions possible. But...

• Doppler is noisy (updrafts downdrafts etc..)
• gt
• Many points can not be fit (30 to 40 depending
  on habit)
• Results will be biased

• Thus...
• We have binned whole data set according to IWC(')
  and
• temperature across different times, clouds etc.
• gt
• Affects of air motions on average values should
  be eliminated.

11
Bin by T and IWC' (not yet IWC)
IWC'IWC(R'eff/Reff)
Increasing IWC' g/m3
12
Fits of bi-modal model to averaged and T/IWC'
binned data (Complex PC assumed)
Nl/NsCR'eff
ltDlgtRoRaT
(for each IWC(') interval)
(9 IWC' intervals used-only 2 shown here)
Note Can not get a defensible fit for some
habits !!!
13
Comparisons (size distribution)
14
Average large-mode particle size
15
Reff-vs-Temperature
16
SummaryConclusions

• Average behavior of ice-crystal size-dist can be
  well-described (in terms of observed lidar/radar
  effective radius and mean Doppler velocity) by a
  bi-modal size-dist model.
• Small-mode fixed. Large mode parameters function
  of both IWC and Temperature !
• Results consistent with previous in-situ
  measurements.
• Results depend on assumed habit. However, some
  standard habits do not fit the data.

17
SummaryConclusions II

• Differences in average Reff-vs-T behavior between
  previous studies may be in part due to different
  IWC sampled. (reanalyses of in-situ data ?)
• Small-mode contributions can not (in general) be
  ignored !
• Small-mode effects most important at relatively
  lower IWC(Temperature) values.
• Results have been used to construct tables of
  Reff , visible extinction and mass-weighted
  fall-velocity-vs-T and IWC.

18
For bi-modal size distributions......
Compact Polycrystals (Mitchell. 1996)
Mean particle radii in large mode
19
Results For complex polycrystals
ltDlgtRoRaT
Nl/NsCR'eff
RoRo(IWC) RaRa(IWC)
20
For bi-modal size distributions......
Compact Polycrystals (Mitchell. 1996)
Increasing Nl
Non-unique relationship !
Unique Relationship !
21
Average Behavior with T and IWC !
Increasing IWC' g/m3
22
Bin by IWC' (not yet IWC)
IWC'IWC(R'eff/Reff)
Increasing IWC' g/m3