Aerosol-cloud-climate interactions modeling and observations at the cloud

Aerosol-cloud-climate interactions: modeling and observations at the cloud scale Graham Feingold NOAA Earth System Research Laboratory Boulder, Colorado ISSAOS 2008 AEROSOLS and CLIMATE CHANGE 22 - 26 September 2008, L'Aquila, Italy

Why do we care about aerosol-cloud interactions? • Planetary albedo is strongly affected by clouds • Large uncertainty in aerosol effects on albedo and radiative forcing • Larger uncertainty in aerosol effects on cloud albedo and radiative forcing Radiative Forcing

How does aerosol affect clouds ? Drop concentration, Nd • Aerosol particles, or that component that act as cloud condensation nuclei (CCN) are a necessary ingredient of clouds • An increase in aerosol concentrations Nacauses an increase in drop concentration Nd: If nothing else changes, drops are smaller and cloud is more reflective Ramanathan et al. 2001 Aerosol concentration, Na

dynamics aerosol: transport dynamics aerosol dynamics: radiation dynamics cloud: updraft, LWP, drop activation, growth cloud dynamics: drizzle, evaporation aerosol cloud: A Complex, Coupled System with Myriad Feedbacks Activation, growth, precipitation cloud mphysics aerosol • Aerosol-Cloud-Dynamics-Radiation-Chemistry-Land-surface cloud aerosol: aq. chemistry, collection, washout, nucleation radiation aerosol: Aerosol absorbs, scatters Modifies heating profiles and surface fluxes radiation cloud: Surface heating drives convection; Reduced shortwave at the surface reduces clouds

Which clouds matter most ? • Low clouds provide strong shortwave forcing – Strong contrast with underlying dark ocean – Radiate at ~ same To as ocean therefore no longwave effect • High clouds (cirrus) provide longwave forcing by trapping outgoing longwave radiation • Persistent and frequently occurring clouds

Microphysical Pathways

Ingredients for Cloud Formation: Dynamics vs. Microphysics • Clouds are formed by dynamics – Updrafts due to convection, orographic uplift – Expansion, cooling, generation of supersaturation • Aerosol particles do not form clouds! • Aerosol particles (particularly CCN) are an essential ingredient for droplet formation • Aerosol particles modify cloud microphysical and optical properties

Aerosol - cloud drop - rain drop r = radius in mm n = number/litre v = fall velocity in cm s-1 Typical number concentrations: aerosol: 103 - 104 cm-3 droplets: 102 cm-3 raindrops: 10 -5 cm-3 Mc. Donald (1958) CCN: drawn 25 x larger • 6 order of magnitude change in mass from typical CCN to cloud droplet! • 10 orders of magnitude mass change from cloud droplet to large raindrop! • factor of 650 increase in fall velocity (droplet to raindrop)

The Scope of the Aerosol Cloud Problem • Involves complexity in both aerosol and clouds • Range of spatial scales – Aerosol particles 10 s – 1000 s nanometres – Cloud drops/ice particles: mm – cm – Cloud scales: ~ 102 m – 103 km • Range of temporal scales – Activation process (aerosol droplet): seconds – Time to generate precipitation ~ 30 min – Cloud systems: days • Coupled System – Multiple feedbacks

What is a Drop? supersaturation • Drops are an aerosol (suspended particles in the air) • Drops can be distinguished from dry or humidified particles using some (somewhat) arbitrary criteria: – Size (e. g diameter > 2 mm) – Volume of water vs. particle – Optical detection (e. g. , can a light-scattering device see them? ) – Activation (have they passed the critical radius on the Kohler curve) Mc. Figgans et al. 2006 wet droplet radius • Continuum from dry particle humidified haze particle droplet

What is a Cloud? Just as the distinction between aerosol and cloud is somewhat arbitrary, so is the distinction between the cloudy atmosphere and the “clear-sky” atmosphere Clouds have fuzzy edges Linear intensity scale Log intensity scale y-direction Cloud? x-direction Model results: Koren and Feingold 2008

What is a Drop? The Kohler Curve Relationship between the supersaturation over a droplet and its wet radius at equilibrium critical supersaturation Supersaturation Kelvin term (surface tension) Solute (Raoult) term critical radius wet droplet radius Kohler curve describing the equilibrium growth of a particle at a given supersaturation for one particle size and composition. It does NOT predict the size of a cloud droplet size distribution

The Kohler Curve: critical supersaturation Supersaturation Kelvin term (surface tension) Solute (Raoult) term critical radius wet droplet radius

The Kohler Curves: Equilibrium particle size Easier to activate droplets at: • lower ss • lower Ms • higher msolute • higher n. F smaller particle larger particle Wallace and Hobbs 2006 Rasool 1973 5 vs 2: effect of composition 2 vs 3 vs 4 effect of mass droplet radius surface tension mass of solute Molecular weight of water Molecular weight of solute “Van’t Hoff factor” ~ number of dissociated ions

CCN Relationship between particle diameter and critical supersaturation Adipic acid (not very soluble) Insoluble but wettable particle Ammonium sulfate (very soluble) Adapted from Hings et al. (2008) ACPD

Aerosol Size distribution unactivated Critical radius (derived from the Kohler equation) Activated particles

Cloud condensation Nuclei (CCN) Typical Measured CCN “activation spectra” Azores Sometimes approximated by N=CSk Florida Arctic Wallace and Hobbs 2006 (after Hudson and Yum 2002) Note logarithmic scales

Drop-size distributions Polluted cloud Pressure, h. Pa d. N/dlogr, cm-3 Clean cloud Polluted cloud polluted cloud effective radius Droplet radius, mm Garrett and Hobbs (1996) drop conc Warm clouds: • Drop size increases with height • Drop conc ~ constant with height • Polluted clouds: more numerous, smaller drops

How do we measure cloud and rain drops? In-Situ (typically airborne) • Size distribution, Liquid Water Content, Extinction Remote Sensing (Radar, radiometer) • Drop sizes, liquid water path

Observations of Aerosol Effects on Cloud mphysics Drop size decreases with increasing aerosol (at constant LWP) Drop size decreases with increasing aerosol (all LWP) Drop size remote measurements Drop size Slope = 0. 10 – 0. 15 Slope = 0. 04 – 0. 085 LWP = liquid water path aerosol extinction Nd In-situ measurements d. N/dlogr, cm-3 aerosol index Clean cloud polluted cloud Na Droplet radius, mm

Cloud optical depth Drop size Drop conc CCN conc Cloud depth Remote sensing Year Boers et al. 2006

Growth Processes • Condensation growth does not produce precipitation in warm clouds • Collision-Coalescence produces precipitation Collector drop (larger fall velocity Vx) condensation coalescence Collected droplets (small fall velocity Vy) “Gravitational Kernel” E(x, y) = collection efficiency Wallace and Hobbs 2006

Droplet coalescence • Coalescence does not become important until collector droplets reach sizes of 20 mm diameter • drop mass raindrop mass 10 orders of magnitude! Collector drop radius Small droplets don’t collide easily with large droplets Collected drop radius, mm Wallace and Hobbs 2006 E(x, y) = collection efficiency

Effect of Aerosol on Precipitation Formation Aerosol significantly reduces the ability of a cloud to generate precipitation (all else equal) (Gunn and Phillips 1957; Warner 1967) Clean Polluted Na = 50 cm-3 t=0 r=20 mm Na = 300 cm-3 r=20 mm t=10 min Drop radius, mm

Effect of Giant Aerosol on Precipitation Formation Giant CCN ~ few mm in (dry) size produce collector droplets r ~ 20 mm Clean clouds: active coalescence process anyhow; Giant nuclei have no effect Polluted clouds: more significantly affected by giant nuclei Clean Polluted Na = 50 cm-3 Na = 300 cm-3 t=0 r=20 mm t=10 min Including 1/l GCCN Drop radius, mm

Influence of Giant CCN on precipitation in Stratocumulus Liquid water path (1/litre) Precipitation • Significant increase in precip due to ~1/litre Giant CCN • More particles does not always mean smaller drops Feingold et al. 1999

Influence of Giant Nuclei on mphysics and Cloud Albedo No giant CCN Drop size Drop Number No giant CCN With Giant CCN Cloud Albedo Significant reduction in cloud albedo due to ~1/litre Giant CCN Feingold et al. 1999

Precipitation: Macrophysics vs Microphysics Rainrate, mm d-1 • Measurements show that Rainrate ~ H 3/N or LWP 1. 5/N • Some models suggest Rainrate ~ LWP 1. 6/N 0. 7 • Rain production is 2. 5 x more sensitive to changes in LWP than changes in Nd • re is a much less effective determinant of rainrate H 3 N-1, m 6 re , mm Van Zanten et al. 2005

Drop breakup Two mm-size drops collide • May enhance cloud’s ability to precipitate by “seeding” the cloud with more precip embryos • Affects the raindrop size distribution and the ability to measure rainrate from radar (radar reflectivity-rainrate relationships) • Affects subcloud evaporation (fragments evaporate more efficiently) • Drives stronger downdrafts Parent drops fragments Coalescence vs Breakup: Depends on the energy of the collision