Cloud Microphysics

1
Cloud Microphysics (in Warm Clouds) Knut von
Salzen Canadian Centre for Climate Modelling and
Analysis, Science and Technology
Branch, Environment Canada, Victoria, BC, Canada
2
Overview

• Introduction
• Formation of Cloud Droplets
• Cloud Condensation Nuclei
• Droplet Growth by Condensation
• Spectral Broadening
• Formation of Rain in Warm Clouds

3
References

• Rogers, R. R. and M. K. Yau, 1996 A short course
  in cloud physics, 3rd Edition, Butterworth-Heinema
  nn, Oxford, pp. 290
• Pruppacher, H. R. and J. D. Klett, 1997
  Microphysics of clouds and precipitation, 2nd
  Edition, Springer Netherlands, pp. 976. Online
  version www.springerlink.com/content/gtu521
• Seinfeld, J. H. and S. N. Pandis, 1998
  Atmospheric chemistry and physics, Wiley Sons,
  New York, pp. 1326

4
A Planet of Clouds
MODIS/NASA
5
Impact of Microphysical Processes on Clouds
Example Ship tracks off the West Coast of North
America (near IR)
Source Ackerman and Toon (2000)
6
A Closer Look at a Ship Track
Durkee et al. (2000)
7
Global-scale Aerosol/Cloud Interactions
Source Breon et al. (2002)
8
Radiative Forcing Components
Source IPCC (2007)
9
Particles Involved in Cloud-Microphysical
Processes
Rogers and Yau (1996)
10
Snow and Ice Crystal Shapes
11
Cloud Microphysical Processes in CCCma AGCM4
Water vapour
?
Qsub(0.50)
Qevp(0.03)
?
Qcnd(1.34)
Qdep(0.76)
?
?
Qfrh
(lt 0.01)
Qfrk
Qfrs
Cloud liquid water
Cloud ice
?
?
?
?
Qmlti(0.05)
Qagg (0.27)
Qaut(0.54)
?
?
Qsacl(0.47)
?
Qsaci(0.45)
Qracl(0.38)
?
?
Snow
Rain
?
Qmlts(0.47)
Rates in kg/m2/day
12
Cloud Microphysical Processes in CCCma
AGCM4 Simulated Zonal Mean Conversion Rates
Condensation
Autoconversion
Melting of Snow
Deposition
Accretion of Cloud Liq. Water by Snow
Accretion of Cloud Water by Rain
Accretion of Cloud Ice by Snow
Sublimation of Snow
Aggregation
Rates in kg/m2/day
13
Formation of Cloud Droplets
14
Clausius-Clapeyron Equation
Vapour pressure Pressure on a liquid or solid
surface due to the partial pressure of the
molecules of that substance in the gas phase
which surrounds the surface.
15
Clausius-Clapeyron Equation (cont.)
Saturation vapour mixing ratio Relevant quantity
for atmospheric transport processes and
thermodynamics. Derived from ideal gas law and
Daltons law.
16
Clausius-Clapeyron Equation (cont.)
Saturation specific humidity
17
(Very) Simplistic Evolution of Cloud Droplets
e - es 0 Equilibrium e - es gt 0
Growth e - es lt 0 Decay
18
Surface Tension
19
Curved Surface Kelvin Effect
Increased vapour pressure over particle (or
droplet) compared to flat surface.
20
Pure Water Droplets
High supersaturation (S) is required for small
droplets to be stable (in absence of other
processes).
Rogers and Yau (1996)
Homogeneous droplet nucleation - from random
collisions of water vapour molecules Requires
high Supersaturation.
21
Solute Effects on Droplets Raoults Law
High solute concentrations in droplets ( small
size) are associated with a reduced equilibrium
water vapour pressure relative to pure water
droplets
22
Solute Effects on Droplets Mixing Rules for
Solutes
Mass fraction of soluble material in dry aerosol
particle
Number of soluble salts
Mass of soluble material per particle
Number of ions in solution from dissociation of
solutes
Molecular weight of soluble material
Densities of soluble, respectively insoluble
material
23
Köhler Theory
Competing effects of droplet size and solute
concentration on equilibrium saturation ratio
Rogers and Yau (1996)
24
Aerosol Activation/Nucleation of Cloud Droplets
Courtesy U. Lohmann
25
Cloud Condensation Nuclei
26
Cloud Condensation Nuclei (CCN)

• CCNs are aerosol particles that activate at a
  specified value of the saturation ratio (e.g.
  1.02).
• CCN concentrations can be experimentally
  determined under clear-sky conditions (CCN
  chamber).
• CCN concentrations predominantly depend on
  aerosol chemical composition and (dry) particle
  size.
• CCN often used as proxy for the cloud droplet
  number concentration (CDNC). However, the CDNC
  depends on the actual value of the saturation
  ratio in the (non-adiabatic) cloud and processes
  other than condensation/evaporation.

27
Observations Near Toronto Aerosol Composition
November 14, 2100 2200
November 16, 800 900
Aerosol Size Distributions at Egbert (Abbatt,
Leaitch)
Mass (µg/m3)
Dva (nm)
Dva (nm)
28
Observations Near Toronto - CCN Concentrations
0 WSOC
Predicted vs. Measured CCN for Different
Assumptions about Water-Solubility of Organic
Carbon (Abbatt, Leaitch)
60 WSOC
29
Hygroscopic Aerosol

• Hygroscopic substances highly water-soluble
  substances (low volatility) Efficient
  activation in clouds
• When solutes are added to water droplets, solute
  molecules displace water molecules in the
  droplets.
• According to Raoults law, a reduction in the
  fraction of water molecules causes a reduction in
  water equilibrium vapour pressure.
• Hygroscopic aerosols Sea salt, sulphate,
  (organic carbon), any aged/oxidised aerosol

30
Aerosol Burdens Natural Aerosol
Mineral Dust
Sea Salt
http//nansen.ipsl.jussieu.fr/AEROCOM/aerocomhome.
html
31
Aerosol Burdens Anthropogenic Aerosol
Sulphate (SO4)
Black Carbon (BC)
Organic Carbon (OC)
http//nansen.ipsl.jussieu.fr/AEROCOM/aerocomhome.
html
32
Sulphur Emissions by Sources
Volcanic (explosive)
Wildland Fires
Off-Road
Volcanic (cont.)
Domestic
Roads
Power Plants
Industry
Shipping
33
Atmospheric Sulphur Cycle
34
Simulation of Sulphate Size Distributions
JJA
DJF
Ma and von Salzen, JGR (2006)
35
Droplet Growth by Condensation
36
Droplet Growth Vapour and Heat Diffusion

• Cloud droplets are located in a vapour field.
• Diffusion equation can be solved to obtain flux
  of vapour molecules at droplet surface.
• Condensation is associated with release of latent
  heat, which causes temperature of droplet to rise
  above ambient value.
• Transfer of sensible heat between droplet and
  surrounding air.
• Diffusion coefficient and coefficient of thermal
  conductivity increase with increasing temperature.

37
Particle Growth by Condensation of Water Vapour
38
Mass and Energy Constraints
39
Summary of Mass and Energy Constraints

• Supersaturation is determined by the difference
  between sources and sinks of water vapour.
• Sources of supersaturation Mixing,
  transport of water vapour, cooling of air,
  decrease in pressure.
• Sinks Condensation, mixing, transport of
  water vapour, warming of air, increase in
  pressure.
• Temperature is determined accordingly from
  internal (e.g. condensation) and external (e.g.
  radiative cooling) sources and sinks.

40
Evolution of Droplet Sizes Near Cloud Base
Rogers and Yau (1996)
41
Evolution of Droplet Sizes Near Cloud Base (cont.)
Rogers and Yau (1996)
42
Adiabatic Parcel Model Simulations
Cloud droplet number concentration
CDNC (cm-3)
SO4 (µg m-3)
Dry aerosol mass 0.01 100 µg m-3 Mode radius
0.01 0.1 µm Variance 1.2 2.2 Updraft speed
1 m s-1 Cloud depth 1000 m
Full red line GCM4 parameterization
43
Observed Cloud Droplet Number Concentrations
Marine cloud
Continental cumulus
Continental stratus
All data combined
Boucher and Lohmann (1995)
44
Parameterization of Cloud Droplet Number in Models
45
Spectral Broadening
46
Rate of Growth of Droplets by Condensation

• Rain drops (R gt 100 µm) may form in less than 30
  minutes.
• Time it may take to form sufficiently large cloud
  droplet by condensation to initiate rain
  formation 41,000s 11.4h
• Solution of condensation equation yields
  considerably narrower droplet size spectra than
  observed at higher levels in clouds.
• Something else must be happening.

47
Potential Factors Leading to Spectral Broadening

• Droplet collisions Larger cloud droplets grow by
  collision/coalescence (only thought to be
  efficient for R gt 20 µm, approximately).
• Ventilation effects Rates of mass and heat
  transfer greater on upstream side of falling
  droplet.
• Unsteady updraft Cloudy regions with variable
  updraft velocities may cause different rates of
  droplet formation and growth.
• Turbulence/Entrainment Local variations in water
  vapour, temperature, and particle numbers lead to
  different particle growth histories and sizes.

48
Entrainment in Shallow Cumulus The Mixing Line
Linear mixing for conserved cloud properties
(total water, liquid water static energy)
Cloud
Environment
Cloud core
Paluch (JAS, 1979)
49
The Mixing Line More Observations
Cloud core
Burnet and Brenguier, JAS (2007)
Cloud environment
50
Homogeneous and Inhomogeneous Mixing in
Clouds from Entrainment

• Relevant time scales
• te(1/R)(d R/d t) Droplet evaporation time
  scale
• tt(L2/e)1/3 Turbulent mixing time
  scale
• Homogeneous mixing te gtgt tt , Efficient
  turbulent mixing means that droplets are exposed
  to the same humidity and temperature. Droplet
  sizes are reduced by evaporation. Spectral
  broadening from mixing line.
• Extremely inhomogeneous mixing tt gtgt te ,
  Filaments of cloudy and non-cloudy air. Droplets
  experience different environmental conditions and
  may evaporate completely or stay intact. Dilution
  of cloudy air by entrainment of clear air reduces
  mean number of cloud droplets per volume of air.
  Little spectral broadening.

51
The Microphysics Mixing Diagram
Burnet and Brenguier, JAS (2007)
Diamonds Neutrally buoyant air for different
homogeneous mixing conditions
52
Cloud Droplet Sizes Observations
te/tt0.05
te/tt1.9
Burnet and Brenguier, JAS (2007)
53
Formation of Rain in Warm Clouds
54
Droplet Growth by Collision and Coalescence

• Collision by gravitational settling
• Dominant collision process in clouds.
• Large drops fall faster than small drops.
• Large drops overtake and capture a fraction of
  these small drops.
• Collision by electrical force
• Enhances the collection of small droplets.
• Usually strong local effect.
• Collision from turbulence
• Particles tend to cluster in turbulent flow
  field.
• Coalescence is subsequently required to
  permanently unite droplets after collision.

55
Factors Controlling Droplet Collision/Coalescence

• Collision efficiency increases strongly with
  droplet size in clouds ( R4), leading to
  accelerating pace of rain formation.
  Predominantly determined by size of collector and
  collecting droplets.
• Coalescence efficiency is usually less than 1
  (for uncharged droplets). Droplets may generally
• Become permanently united
• Bounce apart
• Coalesce temporarily, separate, retaining their
  identities
• Coalesce temporarily, break up into a number of
  small droplets
• Collection efficiency Collision efficiency x
  coalescence efficiency

56
Droplet Collision Efficiency
Collision efficiency Probability that a
collision will occur with a droplet located at
random in the swept volume
Rogers and Yau (1996)
57
Droplet Collision Efficiency Results
Rogers and Yau (1996)
58
Droplet Terminal Fall Speed
Rogers and Yau (1996)
59
Droplet Growth by Collision/Coalescence
Average number of droplets with radii between
rdr collected in unit time
Collision/coalescence less efficient for small
cloud droplet sizes Cloud water content
increases with CCN concentration (Albrecht effect)
60
Effects of Turbulence on Collision/Coalescence
Clustering of droplets in turbulent flow field
Geom. collision kernel
Increased probability of collision/coalescence
Eddy dissipation rate
Franklin et al. (2005)
61
Modelled vs. Observed Rain Drop Size Distributions
Rogers and Yau (1996)
62
Understanding Interactions between
Microphysical and Cloud Dynamical Processes
Ackerman et al. (2004)
63
End