1/194
Looks like no tags are added yet.
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
General circulation
Horizontal temperature differences
Pressure differences
PGF
Coriolis force modifies motion on larger spatial scales – perpendicular to the direction of motion
Geostrophic balance = between PGF and CF
Frictional force slows down motion
Example of temperature driven horizontal movement
Land sea breeze
Occur due to the unequal heating rates of land and water
During the day the land surface heats up faster than ocean water surface
This warms the air just above each surface differently
This causes the air above the land to rise
Rising air causes horizontal movement drawing cooler ocean air to replace the rising warm air
This is a sea breeze and it occurs during the day
At night
The different relative heat capacities of the ocean and the land cause the reverse
Whereby the ocean surface is warmer than the land surface
And so air above the ocean becomes buoyant and rises
This causes the denser cool air over land to flow horizontal to replace the rising warmer air
This is a land breeze
Vertical differences in heating
cause unstable stratification in the atmosphere
Causes buoyancy forces – convergence and divergence
Radiative convective equilibrium
Radiation introduces instability as temperature falls with height so doesn’t the energy emitted at each layer of the atmosphere
No net horizontal transport of heat
Explains why warmer regions have a higher tropopause and why the tropopause expands under global warming
Processes which heat the atmosphere
Latent heat release from convective clouds in tropics dominates heating in free troposphere and drives large scale mean circulation and teleconnections
Clouds affect climate
Through latent heat or interaction with radiation
SVP
Saturation Vapour Pressure
If exceeded vapour condenses until SVP reached again
Depends on temperature
Cloud formation mechanisms
Any process that causes a cooling in the atmosphere until supersaturation is reached
Radiative cooling or lifting/mixing of air masses
Typical droplet sizes
CCN – 0.0002 mm
Cloud droplet – 0.02 mm
Drizzle – 0.1 mm < d < 0.5 mm
Rain – 0.5 < d < 5mm
Summary - general
Clouds = crucial for the climate system
Influence weather and climate through latent heat effects and interaction with radiation
Precipitation determines the net heating of clouds through latent heat release
Precipitation formation = complex
Important information in the atmosphere and ocean propagated through waves
Aerosol
liquid or solid particle suspended in the air
0.01-100 micrometres
But size depends on type of aerosol and changes with time – why not all CNN and cloud droplets the same time
Aerosols are ubiquitous features of the atmosphere – sulphur dioxide, sea salt, desert dust
They have many health impacts but we care about them for their cloud and climate impacts
Aerosol observations
In situ – collect and analyse – sparse but accurate
Remote sensing – radiation measurements, ground based, satellite based
Infer radiative properties of aerosol particles
Assumptions needed for retrieval algorithm
Less accurate but global coverage possible
Complex shapes and composition in contrast to assumptions of sphericality, and perfect mixture
This influences how the particles interact and CCN formation
Aerosol characteristics and types
Population highly variable in space, time, physical and chamical properties
Characterised by size distribution, composition, shape and optical properties
Primary – directly emitted – sea salt
Secondary – formed by gas-to-particle conversion
Sulphate aerosols produced through oxidation of SO2, H2S, DMS, COS and VOCs
Aerosol sources
Natural and anthropogenic
The estimates of global aerosol emissions vary hugely
Primary are less of a concern as they are bigger = lower residence time
Secondary smaller and so longer
Aerosol sources Boucher, 2015
Marine Aerosols (Sea spray aerosols)
Wind friction = salt water into the atmosphere – some water evaporates = more salty
Therefore, sea salt particles are differentially hydrated based on local humidity
Size = 100 nanometers – tens of micrometers
Desert dust
Wind friction = soil suspended in atmosphere
Especially the case in desert or arid regions where there is little slowing down the wind, and where there is little moisture and so little cohesion between particles
Size = 100 nanometers – tens of micrometers
Emission dependent on local environmental and meteorological conditions
Volcanic aerosols
Emission of ash during eruption
Size = micrometer – millimeter
Can be transported huge distances (1000s km)
But fall out of the atmosphere rapidly
Limited effect on climate
Sulphur rich gasses (SO2 and H2S)
Oxidised in the atmosphere to form submicronic sulphate aerosols
If in the troposphere = max residence time = weeks
If in the stratosphere = few months – more than a year (dep. Alt of injection and region)
Spatial and temporal aerosol distributions
Boucher, 2015
Once aerosols are in the atmosphere, they are transported but also removed
Removal pathways = atmospheric sinks
Globally, and over a long enough period sources and sinks should be in equilibrium
Sinks = dry deposition at the surface and wet deposition from precipitation
Aerosol properties also evolve during atmospheric transport
Aerosol concentrations and properties also vary in the vertical
Concentrations larger in the atmospheric boundary than the free troposphere
But can be lifted into the troposphere
And also, into the stratosphere (where sinks are less effective, and residence times are therefore longer)
Aerosol-Cloud-Radiation Interactions
Aerosol size distribution
NOAA, n.d.
Radius or diameter characterize size of one particles, but the particles may have complex shapes + radii vary by orders of magnitude => NOT one size but size distribution covering full spectrum of radius.
Aerosol distribution characterized by 3 modes:
fine mode (d < 2.5 m) and coarse mode (d > 2.5 m);
fine mode is divided on the nuclei mode (about 0.005 m < d < 0.1 m) accumulation mode (0.1m < d < 2.5 m).
Modes
Nucleation (Aitken)
Accumulation
Coarse
(Alfarra, 2004)
Coarse
Alfarra, 2004
mainly produced by mechanical processes and are introduced directly into the atmosphere from both natural and anthropogenic sources.
most significant source is the bursting of bubbles in the ocean, which creates coarse particles of sea salt
Accumulation
Alfarra, 2004
representing a region of particle growth mainly due to the coagulation of particles with diameters smaller than 0.1 µm and from condensation of vapours onto existing particles, causing them to grow into this size range
particle removal mechanisms are least efficient in this regime, causing particles to accumulate there until they are ultimately lost through rain or other forms of precipitation (wet deposition)
Aitken mode
Alfarra, 2004
Act as nuclei for condensation
Short lifetime
Account for the most particles by number
But due to their small size they account for a few percent of the total mass
Processes which change the size/number
Nucleation, condensation, coagulation, deposition
Determine the life/residence time
Nucleation
Formation of new particles from the gas phase without influence of existing particles
Homogenous nucleation of sulfate aerosols
Second organic carbon (SOC) aerosols believed to require the presence of sulfuric acid
Have to overcome nucleation barrier to push gad to particle phase
Rate is a function of temperature, WV and sulfuric acid – the colder, more EV and more sulfuric acid the better (Lohmann et al., 2016)
Condensation
Growth by uptake of gaseous species that condense on pre-existing particles
Preferential condensation > nucleation (due to nucleation barrier)
Nucleation vs condensation
Friedlander, 2000
Types of nucleation
Homogenous – no foreign nuclei or surfaces
Heterogenous – nucleation on a foreign substance – like an aerosol particle
Condensation
Growth of aerosol particles by uptake of vapours from the gas phase --> the most important mechanism of particle growth in the atmosphere
Nucleation vs condensation
Homogeneous nucleation = a new liquid / solid cluster is formed from vapor molecules in the absence of a surface
Heterogeneous nucleation = new liquid/solid cluster is formed on a preexisting surface
Condensation = vapor molecules go from the gas phase to existing liquid phase
In condensation there already exists a liquid phase of the condensing compound.
In heterogeneous nucleation there exists a surface, but it can be a cluster composed of some other compound or some planar surface (e.g. wall). In addition, heterogeneous nucleation also can occur between liquid and solid phases (e.g. ice nucleation).
Before condensation can occur, homogeneous/heterogeneous nucleation need to have happened.
Coagulation, collision and coalescence
Collisions in the atmosphere cause by Brownian motion, turbulent shear, differential fall velocity, electrostatic charges
Collision efficiency depends on size – more efficient bigger until air streams of other particles get involved (Seinfeld and Pandis, 1998)
Friedlander, 2000
Collisions + coalescence
Agglomerations = collision and sticking (no coalescence)
In the atmosphere, coagulation of sub-micron particles is governed by the Brownian motion of particles (Brownian coagulation).
For super-micron particles and cloud droplets gravitation, turbulence and wind shear also influence the coagulation rate.
Removal processes
Dry deposition
Removal in the absence of precipitation – function of size and density
Turbulent motion contributes to vertical transport – hit another particle and stick to it
0.1 micrometre < r < 1 micrometer least efficient
Wet deposition
In cloud scavenging – incorporation of aerosols in cloud droplets
Below cloud – by precipitation
Snow is super-efficient
Role of aerosols in cloud formation
Clouds – small water droplets/ice crystals suspended in air
Clouds are NOT WV
To understand microphysics, you need to understand the movement of the cloud particles
Microphysics
Processes on micrometre scales – phase transfers between vapour, liquid and ice
Nucleation, condensation, coagulation, auto conversion, accretion, riming and freezing
These determine the size distribution, density and shape – which determine cloud properties
Herzog et al., 1998
CDN - cloud droplet number
(Pringle et al., 2009)
For a given aerosol number concentration produces a wide range of cloud droplet concentrations due to variation in the shape of the aerosol size distribution
Prediction of CDN
Number of cloud droplets in a rising air parcel is dependent on the number, size and chemical composition of the aerosol particles and the meteorological conditions (e.g. updraft velocity)
Calculation simplified in climate models
Empirical relations derived from regional measurements and extrapolated to the globe and for past and future
By bypass detailed microphysical processes which control CDN
Cloud formation
Air saturated at SVP – no cloud formation – stable balance between condensation and evaporation
Super-saturated – existing water droplets grow by condensation and new water droplets may form
Super-saturation necessary but not sufficient – need CCN
SVP different for water and ice
Need something to produce supersaturation – radiative processes, lifting air masses, mixing two subsaturated
Cloud physics - Houze, 2014
Cloud physics = cloud microphysics and cloud dynamics
Warm clouds
Liquid cloud droplets begin as CCN
And grow to become precipitation particles
Nucleation - H, 2014
If the CCN happens to be composed of a material that is soluble in water, the efficacy of the nucleation process is further enhanced.
Since the saturation vapor pressure over the liquid solution is generally lower than that over a surface of pure water
Condensation, H 2014
Rate of condensation determined by diffusional ratio between WV and latent heat
Fallspeeds, H14
Cloud droplets are subject to downward gravitational force which leads to their fallout
As a particle accelerated down its motion is increasingly resisted by frictional force
Final speed = terminal fall velocity
Both cloud droplets and precipitation undergo sedimentation in this way
Larger rain drops become flattened into the shape of a disc as they fall – velocity increases nonlinearly
Continous collection, H14
Cloud drop growth by coalescence
Continuous collection – the mass of a falling particle increases continually as it falls thorough a cloud and coalesces with all particles in its cross-sectional area swept out – which is continually increasing
Early on stochastic collection leads to a broadening of the size distribution
Larger drops grow faster due to more frequnt collisions
Runaway growth process
Self reinforcing system
This is modified as larger particles break up due to aerodynamical instability and become a number of smaller drops
Cold clouds H14
Homogenous nucleation of ice particles
In theory ice particle can be homogenously nucleated from the vapour phase but requires very low temperatures –65 and high supersaturations ~1000%
This doesn’t occur
Homogenous from the liquid phase could occur at temps –35 to –40 add to from lecture
Heterogenous nucleation
Observations of ice crystals between 0 and –38
Homogenous doesn’t happen here so there must be a heterogenous process
The principal difficulty with the heterogeneous nucleation of the ice is that the molecules of the solid phase are arranged in a highly ordered crystal lattice
To allow the formation of an interfacial surface between the ice embryo and the foreign substance, the latter should have a lattice structure similar to that of ice
Ice nucleation complexity H14
An ice nucleus contained within a supercooled drop may initiate heterogeneous freezing when the temperature of the drop is lowered to the value at which the nucleus can be activated
If the CCN on which the drop forms is the ice nucleus, the process is called condensation nucleation
If the nucleation is caused by any other nucleus suspended in supercooled water, the process is referred to as immersion freezing
Drops may also be frozen if an ice nucleus in the air comes into contact with the drop; this process is called contact nucleation
the ice may be formed on a nucleus directly from the vapor phase, in which case the process is called deposition nucleation
The probability of ice-particle nucleation increases with decreasing temperature, and substances possessing a crystal lattice structure similar to ice provide best nucleating surface
Aggregation and riming, H14
If ice particles collect other ice particles, the process is called aggregation.
Temperature dependent 0 probability of adhesion is likely > -5 – ice crystal surfaces become sticky
If ice particles collect liquid drops, which freeze on contact, the process is called riming
Fog
(Gultepe et al., 2007)
Fog refers to a collection of suspended water droplets of ice crystals near the Earths surface that lead to a reduction in visibility below 1 km (NOAA, 1966)
There are many classifications – by formation
Radiation fog – forms over land due to nocturnal cooling
Advection fog – warm moist air over cold surfaces
And others (frontal, upslope)
Microphysics and nucelation processes
Formation occurs in aerosol laden surface air under high relative humidity conditions
Composed of haze (unactivated) particles
Fog droplets are generally smaller than cloud droplets due to the lack of updraft which means supersaturation remains low
Most important factors for fog formation (Duynkerke, 1991)
Cooling of moist air by radiative flux divergence
Mixing of heat and moisture
The presence of clouds increases the incoming longwave radiation at ground level and reduces longwave radiative cooling at the surface
Radiation fog occurs in sprint in coastal plains under the influence of advection inland of moist marine air during the previous afternoon
Advection fog tends to occur in the spring and summer months – when the occurrence of warm air flowing over the cold ocean is maximised
Cloud formation
Buoyance, convergence, topography, frontal lifting
Mixing
Due to convex shape of SVP line two subsaturated can mix to be supersaturated – e.g. at edge of frontal systems
Processes aren’t distinct
Mixing enhanced vertical motion and emission of latent heat adds energy and makes unstable – generated convection heating
The development of a particular cloud type depends on vertical temperature and humidity profiles
Orographic uplift and foehn effect
Moist adiabatic rise and precipitation on the windward side
dry adiabatic decent on the leeward side
leeward size = warmer than windward side = foehn effect
Other formation mechanisms
Gravity waves
Can form or change shape
Turbulent mixing
Boundary layer clouds
Stevens, 2005
Marine stratocumulus
Development
Dynamically evolving
Change forms
Cloud droplet nucleation
Energy barrier must be overcome for phase change
Homogenous – droplet formed solely from WV
Heterogenous – CCN facilitates condensation
Kelvin effect
If a droplet is in equilibrium with its environment at any time condensation and evaporation at the surface is equal
The ease of this dynamic equilibrium is dependent on the surface tension of the droplet which depends on size or curvature
SVP larger over curved surface
Curvature reduced the number of nearest neighbours (coordination number) which makes it easier for molecules to evaporate
Higher vapour pressure for smaller droplets – important for nucleation of new droplets and lifetime of small droplets
SVP over curved > over flat
Expresses as a ratio
If they grow past 0.12 micrometres then the ratio is essentially 1 and the Kelvin effect is less important (link to Koehler)
Homogenous nucleation (liquid)
Homogenous nucleation requires RH>400% when observed supersaturation is typically smaller than 1%
Raoult
Raoult’s law opposes the kelvin effect
The solution effect
Reduces SVP
Solution effect
If solute added, then liquid molecules at the surface are replaced by solute molecules and the SVP is reduced – easier for WV molecules to transfer from gas to liquid
For droplet
Smaller droplets have lower SVP for constant solute content
Rogers and Yau, 1989
Heterogenous nucleation (liquid)
Droplet larger than critical radius is said to be activated and continued to grow by condensation at successively lower supersaturation
Condensation on particles/aerosol surface = acting as CCN now
Activation depends on critical radius
Koehler curve
Bringing together Kelvin and Raoult’s law
Supersaturation of droplet with given solute content as a function of radius
Increase in SVP due to dominance of Kelvin at small scales and then decrease as Raoult comes into play
The hump of the curve is the critical supersaturation which the droplet much reach to grow into a cloud droplet
If this critical environmental humidity was never reached, then the particle would never become activated into a cloud droplet
This critical diameter and supersaturation depend on the amount of solute
The more into supersaturation the environment, the greater range of particle sizes activated
CCN
Aitken or nucleation mode
D<0.1 micrometer
Accumulation mode
01<d<1
Most important for cloud formation
Coarse mode
D<1 micrometer
Fog and cloud droplets
D> 10 micrometres
Typical CN
Radii > 0.1 micrometres
Supersaturation necessary but not sufficient – need CCN (aerosol particles – most CCN are soluble accumulation mode particles)
Atmospheric moist convection
Stevens, 2005
Unlike dry convection it involves phase changes of water leading to complex interactions with radiation, gravity waves and microphysical processes like precipitation
Two fluid problem – one fluid (unsaturated air) and transform itself into another (saturated air) simply through vertical displacement
Stratocumulus convection
Low lying stratiform clouds driven by radiative cooling at the cloud top
Turbulent mixing in the boundary layer driven by surface heating, wind shear and radiative cooling
Lift moist air parcels upward, cooling them adiabatically until they reach saturation
Cumulus cloud
Ahrens, 2006
Warm air rises
And condenses forming a cloud
There will be downward movement down the sides of the cloud
Due to cool air descending to replace the warm air
And evaporation around the outer edge which cools the air and makes it heavier
Subsiding air inhibits the formation of clouds and so cumulus clouds have space between them
Shading of the surface also cuts of surface warming and therefore continued development of the cloud
Causing it to dissipate (as the water droplets evaporate) and the cycle of warming to start again
Key idea of aerosl effect
Net cooling
Radiative forcing
Change in radiative flux (W/m2) at tropopause or TOA due to change in forcing agent
Climate systems not allowed to adjust to forcing
Net radiative flu = incoming – outgoing
Radiative effect
The net effect of a specified change in the system (cooling or warming without quantification)
Aerosol radiative effects
Aerosol-radiation interaction
Aerosol-cloud interaction
Effective radiative forcing
If radiative forcing calculated at the tropopause, stratosphere allowed to adjust to changes --> thus radiative forcing in a theoretical concept not directly observable
Climate effects of aerosols
Direct – scatter and absorb --> cool
Indirect
Aerosols increase number concentration of cloud droplets and ice particles --> cool
Aerosols decrease precipitation efficiency --> cool
There are other indirect and semi-indirect effects which both warm and cool
Effective radiative forcing
Effective radiative forcing quantifies the energy gained or lost by the earth system following a perturbation
Fundamental driver of changes in earths TOA energy budget
Bellouin et al., 2020
Very likely that ERF is negative
Aerosol radiation interactions = –0.22 [–0.47 to 0.04]
Aerosol cloud interactions = –0.84 [–1.45 to –0.25]
Anthropogenic aerosol particles primarily affect water clouds by serving as additional cloud condensation nuclei (CCN) and thus increasing cloud drop number concentration (Twomey, 1959).
Adjustment processes
For aerosol-radiation interactions = semi-direct effects (Johnson et al., 2004)
Semi-direct effect of absorbing aerosols on marine stratocumulus clouds – how their vertical distribution influences cloud properties and radiative forcing
used a large-eddy model, isolating the semi-direct effect by excluding microphysical (indirect) aerosol impacts
Found
Within the boundary layer
Heat the cloud layer, reducing low-cloud cover and liquid-water path (LWP) by ~10 g m⁻².
Cause a warming effect
Even mildly absorbing produce a semi-direct forcing 3x stronger and in the opposite sign than the direct aerosol forcing
Above the boundary layer
Increase LWP by 5–10 g m⁻², leading to a negative radiative forcing (cooling effect).
Strengthen the temperature inversion, reducing cloud-top entrainment and thickening the cloud layer.
W/in and above the BL
Result in a positive but weaker semi-direct forcing (half that of BL-only aerosols).
Since marine stratocumulus covers ~20% of the globe, the semi-direct effect of absorbing aerosols could significantly impact global radiative forcing.
Boundary layer = lowest part of the atmosphere – 1-2 km altitude
In the context of marine stratocumulus clouds (low sheet like couds over the ocean)
They form and persist within the BL
Aerosols within the BL are within the cloud layer
Aerosols above it cool the air above the cloud – strengthening the temperature inversion
Lifetime effect
Rosenfeld, 2006
The study challenges the traditional focus on the Twomey effect (aerosols increasing cloud albedo by making droplets smaller) and emphasizes the lifetime effect—where aerosols suppress precipitation, prolonging cloud lifetime and increasing cloud cover. This effect dominates aerosol-induced radiative forcing, particularly for marine stratocumulus and deep convective clouds.
Lifetime effect mechanisms
In shallow clouds – marine stratocumulus
Aerosols supress precipitation - More aerosols → smaller droplets → slower coalescence into raindrops → less cloud water loss
Cloud cover increases as a result
Deep convective clouds
Polluted clouds
Delayed precipitation – allows more water/ice to reach upper levels releasing latent heat higher up
Invigorating updrafts and generating taller clouds
Finds
Lifetime effect dwarfs albedo effect
The Twomey (albedo) effect contributes <20% of the total forcing; the rest is due to suppressed precipitation and increased cloud cover.
Twomey effect (microphysics driven – droplet optics)
Aerosols act as cloud condensation nuclei (CCN), increasing the number of cloud droplets for the same amount of liquid water.
This makes droplets smaller and more numerous, enhancing cloud reflectivity (albedo).
Shortwave cooling: More sunlight is reflected back to space, reducing surface warming.
Confined to shallow clouds: Deep clouds already reflect most sunlight, so the effect is minimal.
But lifetime effect (dynamics driven – precipitation and convection)
Smaller droplets (from aerosols) suppress precipitation because coalescence into raindrops slows.
Cloud water is not depleted by rain, prolonging cloud lifetime and increasing coverage/thickness.
Longwave and shortwave effects: Persisting clouds trap more thermal radiation (warming) while reflecting more sunlight (cooling). Net forcing depends on cloud type.
Too cool to be true?
When thinking about atmospheric forcing in general (applies to ghgs) assumption – forcing is related to GMST independently of what is causing the forcing – is that true for aerosols? No lol
Equivalence of surface and top of atmosphere
Purely scattering = if you know TOA you know BOA
Absorbing aerosols
You have less SW out
Through warming = more LW emissions
GH effect
Overall, with a net positive forcing – more radiation in than out
But not indicative of warming at the surface
The atmosphere warms but not the surface
Less reaches the surface as the warming as absorbed in the aerosol layer
So radiative forcing does not indicate surface temperature change
Indicative of wider atmospheric change
Complexity in aerosol forcing
Aerosol concentrations vary in space and time and so forcing comes with associated patterns which alter pressure and temperature gradients and change circulation
Winter warming
Kirchner et al., 1999
Post 1991 pintubo eruption and aerosol emission
Stratospheric heating
Aerosols absorb solar near-IR and terrestrial LW radiation, warming the lower stratosphere by ~4 K (simulated; observed ~2 K after accounting for QBO and ozone depletion)
Heating peaks in the tropics, enhancing the meridional temperature gradient.
Tropospheric cooling
Aerosols scatter solar radiation, reducing surface shortwave (SW) flux by ~3 W/m² (global average).
Causes summer cooling (direct effect).
Winter warming
Enhanced polar vortex
Stronger tropical heating → increased equator-pole temperature gradient → strengthens the stratospheric polar vortex (westerly winds).
Observed: Zonal winds at 60°N intensify by ~4 m/s in winter.
S-T coupling
A stronger polar vortex reflects planetary waves back into the troposphere, altering circulation.
Favors a positive phase of the North Atlantic Oscillation (NAO):
High-pressure anomaly over the North Atlantic → warmer advection over NH continents (Eurasia, North America).
Low-pressure anomaly over Greenland.
Surface expression
Winter warming: Simulated warming over NH land (e.g., Europe, North America) matches observations
Delayed response: Warming peaks in late winter (February–April) due to vortex persistence.
Semi-direct effect
Absorbing aerosol warms the aerosol layer and cools the surface
Stabilises the atmosphere and a reduction of low-level cloud cover (need more moisture to reach SVP)
Semi-direct effect leads to warming
Can be several times larger than the direct effect
Semi-direct effect case study
Koren et al., 2004
Urban air pollution and smoke from fires modelled to reduce cloud formation by absorbing sunlight and cooling the surface (and heating the atmosphere)
Inhibiting convection and cloud formation
Satellite data over the Amazon region during the biomass burning season showed that scattered cumulus cloud cover was reduced from 38%in clean conditions to 0%for heavy smoke
Smoke does some scattering but not as much as clouds
and so cannot make up for the cloud loss of scattering
and it also causes some warming
Relationship between cloud droplet and aerosol number
Gultepe and Isaac, 1999
Compare aerosol number and cloud droplet
Expect increase with CD as AN increases
But this does not align with measurements
There is a lot of noise as cloud rework – dynamic system
These is evolution and entrainment of air and feeding from moisture
Most increase in CD in clean atmosphere
Most vulnerable to aerosol perturbation
Polluted air less susceptible to the effects of aerosols on CDN
Also depends on the type of aerosol
Saturation effect
Large uncertainties even when the process (cloud brightening) is clear
ACE-2
Raes et al., 2000
Experiment which looked at clouds and aerosols
In a clean marine environment – low pollution
Polluted air from African continent coming through which can be compared
Used model to investigate aerosol indirect effects
High resolution in both vertical and horizontal
Cyclic boundary to get long term development
Twomey effect
More aerosols (e.g., pollution) increase cloud droplet number but reduce droplet size, making clouds more reflective (brightening effect).
Albrecht effect
Smaller droplets suppress rain formation, potentially prolonging cloud lifetime (more cloud cover).
ACE-2 Observed Both: Pollution plumes showed higher droplet concentrations and reduced drizzle, supporting these indirect effects.
Marine Stratocumulus Clouds:
Polluted air masses had higher droplet concentrations but no evidence of reduced cloud cover
In some cases, suppressed precipitation (Albrecht effect) could maintain cloud layers longer.
Key idea precipitation in the liquid
need a mechanism which accounts for the production of precipitation on timescales observe in the atmosphere – 30 minutes
Precip liquid growth
Condensation
Need supersaturation around the droplet
Difference (gradient) between near-field and far-field saturation
Growth dependent on the speed of diffusion from the far- to the near-field
Latent heat produced by condensation needs to diffuse away to maintain condensation
Speed
The larger the radius the slower the growth (double size, half growth)
Too slow for precipitation formation (even discounting for competition between droplets for supersaturation) (Rogers and Yau, 1989)
To reach 50 micrometer (drizzle – smallest precip possible) you need ~12 hours
PL - collision and coalescence
Collisions
Brownian motion, turbulent shear, differential fall velocity, electrostatic charges
For cloud droplets DFV is the most important – due to non-monodisperse size distribution
Collisions only efficient in clouds for r>20 micrometres
1mm raindrop needs collection of 100k droplets, need >20 micrometer to be efficient, only need a few of these large ones – only need one in 100k
18 micrometres is a magic number – collisional growth becomes stable
Coagulation
Collision doesn’t guarantee coalescence
Bounce apart, coalesce temporarily
Liquid = usually permanent coalescence
Terminal fall velocity
Gravitational force balanced by drag force
Drag depends on flow regime which is dependent on droplet size
Large = turbulent flow
Droplet spectrum
Droplet growth extremely sensitive to initial size distribution
Berry, Reinhardt, 1974
Marshall Palmer, 1948 distribution
Empirical study of the distribution of raindrop sizes
Exponential drop size distribution
Small drops are more common
The number of raindrops decreases rapidly as size decreases
And large drops are unstable
The shape of the distribution depends on rainfall rate
In light rain most drops are small
In heavy rain there are more medium and large, but still fewer large than small
Warm rain
Never homogenous formation
Initial cloud droplets (r~1micrometer) differ in size due to different properties of CCN, fluctuations in temperature and humidity
Condensational growth alone is too slow
Collision and coalescence drives warm rain process
Initial collisions determined by stochastic process
Then turbulence and condensational growth enhance initial collisions
Key idea - precipitation in the ice phase
Cool rain forms in similar but not the same ways, warm and cool rain processes interact
Freezing
Nucleation barrier needs to be overcome
Homogenous freezing
Need T<-38 degrees c for spontaneous freezing of cloud droplets
Need to get liquid water molecules to organise in an ice lattice
Through rearranging
Probability of formation of ice nucleus dependent on temperature
Once nucleus forms then it all goes and rearranges rapidly
The only thing we know is that when it gets cold enough then cloud droplets spontaneously freeze
Temps reached in the free troposphere
Homogenous deposition
None!
Solid ice and gas phase water forming on it
Never happens under atmospheric conditions
You would always get liquid water
There can be no formation of ice phase particles from gas phase
Heterogenous freezing
Aided by the presence of foreign surfaces or suspended particles
Becomes significant at –15 degrees
Dependent on surface properties, shape and chemistry
Statistical process – freezing depends on exposure time
No closed theoretical descroption
Ice Nuclei
Needed for heterogenous freezing
IN only small fration of aerosol population (1 in a billion
IN conc dependent on temp – only active if temps are low enough
Some INs are CCNs which reduces the INs available for freezing
Mineral dust = good IN bad CCN
Ice in clouds
Unlikely for temps above –5
More common in decaying than newly forming cumulus
More common in stratiform than cumulus with same cloud top temp
Because slow formation allows stoachastic processes to have more of a chance to be active
Observed ice crystal number larger than IN concentration – ice multiplication
Fracture of ice crystals/shattering or splintering of freezing drops
Models
Low level polar clouds often in liquid phase but models bias towards ice which is a problem for climate modelling
Hallet-Mossop/rime splintering
Secondary ice crystal formation process
Supercooled droplet of the right temp and size are captured by graupel (soft hail) then small droplets can be produced during freezing
-8<T<-3 and d<25 micrometre
Warm enough that freezing takes time to allow for layered growth, cold enough to freeze
(Atlas et al., 2022)
In clouds containing both liquid and ice with temperatures between −3°C and −8°C, liquid droplets collide with large ice crystals, freeze, and shatter, producing a plethora of small ice splinters.
These splinters act as new ice nuclei
Accelerating cloud glaciation
Enhancing processes like aggregation (sticking together) and riming (ice particles growing by collecting liquid droplets)
Earlier dissipation – early precipitation which speeds to transition from mixed-phase to fully glaciated cloud (which dissipates more quickly)
Causing clouds to reflect less sunlight and have shorter lifetimes
Diffusional growth of ice
Same diffusional balance as water (latent heat from WV to ice which is 10% larger release)
Ice crystals have non-spherical shape
Makes modelling v hard as cannot predict shape
Ventilation effect
Don’t have radial shape of temp and WV around the ice crystal
Means there is asymmetric flow of air around the particle
Enahned diffusion processes – poorly understood
Preferential growth on ledges and edges as more WV deposition than flat surface
Explains snowflake branches
Environmental conditions – shape depends on historical environmental conditions
Saturation over water and ice
Ice crystals grow at the expense of water droplets
SVP for ice and water is different
SVP over liquid surface is always larger than an ice surface
The absolute difference is meaningless a v low temps, max difference at –12
Bang on ice saturation but subsaturated for water
Water droplet evaporates and the ice gets more deposition
If ice crystals and water droplets both present
Then ice grows at the expense of water droplets
But we have lower IN conc than CCN conc
Form more cloud droplets than ice crystals
So have many cool ice droplets and few ice crystals
So diffusional growth of ice crystals can form very large ice particles in liquid clouds
Diffusional growth is too slow in water to produce precipitation
For ice it is so much faster
Ice vs water
Super cooled droplets w/o ice crystals
WV conc close to saturation
When ice starts to form
ice crystals form into supersaturated environments
rapid growth by diffusions
reduced WV conc in atmosphere
water droplets in subsaturated conditions
evaporate and become available to ice crystals as WV
Bergerson-Findeisen process
A theoretical explanation of the process by which precipitation particles may form within a mixed cloud (composed of both ice crystals and liquid water drops).
“cold-rain process”
Subsaturated environment for liquid water but a supersaturated environment for ice, resulting in rapid evaporation of liquid water and rapid ice crystal growth through vapor deposition.
If the number density of ice is small compared to liquid water, the ice crystals can grow large enough to fall out of the cloud, melting into rain drops if lower level temperatures are warm enough.
Accretion
Precipitation particle (in the ice-phase) captures a supercooled droplet
Contact freezing and riming
Aggregation = clumping of ice crystals to form snowflakes
Collision and capture
Similar to coagulation of droplets
Need to know fall velocities – shape and density plays an important role
Size distribution
Collection efficiency of ice crystal aggregation is strongly temperature dependent – only significant for T>-10
Observed size distributions for hail, graupel and snow follow power law -> MP distribution
But no breakup for hail and graupel so size is only limited by fall velocity vs updraft velocity
No absolute limit as no collisional breakup or aerodynamical instability
Only limit is fall velocity vs updraft velocity – thunderstorms = huge crystals
Comparison of warm vs cool rain
Warm rain process (collision-coalescence)
Warm rain formation is generally faster because droplets grow through frequent collisions and can produce rain within tens of minutes in intense convection.
Can sustain clouds for longer if updrafts keep droplets suspended.
Cool rain process (Bergeron-Findeisen process)
Cool rain formation is slower because it relies on ice crystal growth and secondary ice processes, often taking an hour or more to produce precipitation.
Can shorten cloud lifetime by enhancing ice crystal growth and precipitation.
Circulation and rotation
KI: circulation and vorticity are different and important
KI: Rotation is key for predicting motion in the atmosphere (this links nicely to a discussion on waves and then teleconnections)
Rotation general
Atmospheric motion is never along straight lines and is never circular
Constant change in direction means there is always an element of rotation involved
Tropical cyclones and midlatitude cyclones represent rotational features of the atmosphere
The gulf stream consists of many isolated vortices and eddies
General principles of atmospheric motion
Differential heating of the earths surface causes temperature gradients
This generates pressure gradients
Which cause horizontal motion along the PGF
The Coriolis force (the effect of the earths rotation) deflects moving objects
Resulting in flow perpendicular to the PGF
This describes flow generation but not flow development
Flow development
Change in flow in time = advection of flow by flow itself + CF + PGF + Gravitational force + frictional force
Development = non-linear transport + generation + G in the vertical + friction (including turbulence)
This is a key framework for discussion and essay structure
Asymmetry created by non-linear term (advection)
PGF
Only force which leads to an acceleration and increase in horizontal windspeed
In the vertical PG is balanced by gravitational/buoyancy forces
Friction
Near surface friction (turbulence) reduces flow
New equilibrium between CF, PGF and frictional force
Causes slowdown of flow
Also ,when there is a reduction in flow speed the PGF becomes out of balance w CF and so there is motion to balance this again
Cyclostrophic balance
Non-linear advection term includes centrifugal force
Perpendicular to radial component
Cyclostrophic balance between PGF, CF and centrifugal
Low pressure system
Centrifugal opposes PGF in the same direction as the CF
Lower flow velocity compared to geostrophic
High pressure system
Centrifugal in the same direction as the PGF and so higher flow velocity than geostrophic
Strength of high pressure system is limited – otherwise the sum of centrifugal plus PGF grows faster with increasing velocity than CF
Centrifugal force only changes direction and so has an indirect effect on flow speed (through generating a new balance)
Scale asymmetry
Compact intense low pressure systems
Large weak high pressure systems
Due to this asymmetry wind speeds are higher in high pressure systems than low pressure systems at the same pressure gradient
High pressure systems are more uncommon than low pressure systems
Concept of circulation
Describes rotation in a fluid without reference to a rotational axis
Positive for counterclockwise (cyclonic) rotation
Development of circulation
Changes with time if the integral flow along a closed contour changes
Due to transport, CF, pressure gradient and friction
In the absence of friction circulation can only change due to the PGF if density is not just a function of pressure (baroclinicity)
Baroclinic conditions = temperature not constant along surfaces of constant pressure
Temperature changes independently to pressure
Baroclinicity generates circulation through converting potential energy from temperature gradients into kinetic energy
Important for mid-latitudes
Circulation will change under baroclinic conditions even if no rotational component existed in the flow before
Thus, baroclinic instabilities are at the heart of fronts and cyclones in midlatitudes
Frontal systems indicate baroclinic flow
Baroclinicity
When the gradient of pressure is misaligned from the gradient of density in a fluid
This means that density is dependent on both temperature and pressure
Baroclinicity is what drives the land sea breeze
Circulation lynch and cassano, 2006
Circulation does not require an axis of rotation because it is a measure of the total rotational motion of a fluid along a closed path, rather than around a specific axis
Convenient when an axis of rotation is hard to identify
Circulation is a path-dependent property of fluid motion and does not require a specific axis, whereas rigid-body rotation does.
C is positive for counterclockwise flow
Kelvins circulation theorem
L&C 06
Circulation around a closed curve will be a function of both time and space
Circulation is constant in a barotropic, inviscid fluid
The theorem implies that if a fluid initially has no circulation, it will remain circulation-free unless external influences act.
The theorem implies that if a fluid initially has no circulation, it will remain circulation-free unless external influences act.
In the absence of viscosity, fluid elements cannot spontaneously develop rotation (vorticity), meaning rotational motion must be inherited from initial conditions or external forcing.
Another way of describing the curved motion of fluid parcels without reference to a centre of rotation
Simplifies analysis
No need to define a specific closed circuit
Relative vorticity = positive for counterclockwise rotation
Flow around a low-pressure centre in the NH – relative vorticity = positive
Temporal changes in vorticity tell us about cyclone development
Vorticity increases as cyclones spin up and decreases as they die
Spatial changes in vorticity can indicate topographic features or temperature gradients
Which can lead to the generation of circulation according to Bjerknes’ theorem
One way to address the spatial variation or vorticity in isolation is to develop a quantity which is conserved
Potential vorticity
Conservation of potential vorticity
L&C 06
Kelvin’s theorem
Inviscid barotropic flow
Circulation is conserved
In real cases circulation (or vortex strength) changes due to the presence of baroclinicity or friction
We can simplify
Remove friction by assuming we are far from the surface
And that motion is adiabatic (potential temperature is constant)
Potentiotropic – the density is a function of pressure alone
Vorticity gen
Circulation is tricky because it depends on a closed contour – it is a property of an area that can change even if circulation doesn’t
So, more intuitive to study circulation per unit are = vorticity
Circulation per unit area
Property of a flow at a given location
Circulation around a closed contour divided by the area enclosed by this contour
Vorticity can change even if circulation remains constant
To account for this = concepts of relative and absolute vorticity
Relative = vorticity for an observer on earth – change in wind speed perpendicular to the direction of motion
Absolute = relative plus Coriolis parameter
Vorticity due to horizontal wind shear
Any rotation in the same direction as the earth rotation in that hemisphere is termed positive vorticity
So counterclockwise in NH
Wind shear
For example, where westerly (flow from W – E) flow is increasing in meridional direction (N-S direction) causing cyclonic rotation
Vorticity in flow with curvature
A flow with curvature has change in meridional and latitudinal direction at the same time
For example, a low pressure system moving within a mean weaterly flow which meanders as a planetary RW
Ridge = clockwise anti-cyclonic/negative vorticity, minimum vorticity at the peak
Trough = counterclockwise positive vorticity, maximum vorticity at the peak
If a cyclone is moving through these troughs and ridges
It has positive vorticity
It will weaken through the ridge and strengthen through the trough
Development of vorticity
In divergent flow circulation will not change but vorticity will – why?
Divergence (convergence) reduces (increases) absolute vorticity
Producing anti-cyclonic (cyclonic) rotation even if relative vorticity is initially zero
Barotropic vorticity equation
Equivalent to Kelvins circulation theorem
In a barotropic, frictionless, divergence free fluid absolute vorticity is conserved
Fluid moving N-S direction
Coriolis parameter changes
So relative vorticity changes accordingly
Barotropic vorticity equation
Absolute vorticity is conserved in a barotropic, inviscid, divergence free fluid
Explains why low-pressure systems weaken (strengthen) when moving poleward (equatorward)
Vorticities interact with mean flow
A low-pressure system travelling with the meandering winds of the midlatitude westerlies will intensify when it travels south easterly towards a trough and will weaken when it travels north easterly towards a ridge
The opposite happens to a high pressure system which is characterised by negative vorticity – anticyclonic