Turbulence and Clouds: A Measurement Perspective
(Raymond Shaw, Michigan Tech) 

Clouds are an integral part of the climate system because they couple the
dynamics, the transfer of radiation, and the distribution of water in the
atmosphere. Clouds are inherently turbulent and as a result of their
enormous Reynolds numbers fluctuations in energy dissipation and velocity
and scalar gradients are highly intermittent. Recent measurements of energy
dissipation rate at meter-scales, for example, are qualitatively consistent
with a lognormal (K62) distribution (Siebert et al. 2006b). How turbulent
fluctuations influence the evolution of cloud particle size distributions
is highly relevant to our ability to describe relevant cloud processes in
numerical and theoretical models of the atmosphere (both regional and
global). Taking rain formation via droplet coalescence as an example,
turbulence is thought to influence the droplet collision rate by modulating
relative droplet velocity, droplet collision and coalescence efficiencies,
and droplet spatial distributions. A minimalist model of rain formation
demonstrates that rain formation is highly sensitive to rare collisions
early in the lifetime of a cloud, and therefore fluctuations in these
quantities necessarily will dominate the time to initiate rain (Kostinski
and Shaw 2005).

These turbulence-cloud interactions have been studied mostly with
theoretical and computational tools, but some recent experimental results
are summarized here. Droplet spatial distributions are relevant to
collision rates, condensation growth rates, and to radiative transfer.
Briefly, it is expected that particles with Stokes numbers [St=(particle
inertial time scale)/(fluid time scale)] of order 1 will cluster in regions
of low vorticity in a turbulent flow (Shaw 2003). The hypothesis has been
validated in numerical models and in a recent experiment in a
high-Reynolds-number wind tunnel flow seeded with particles. Particle
clustering is observed to increase with decreasing scale in a power-law
form, as expected from theory, and at a given scale the clustering strength
increases with droplet Stokes number. Similar experiments have been carried
out in clouds and initial data analysis shows analogous clustering
signatures in regions of intense turbulence. Finally, recent cloud
measurements in collaboration with the ACTOS group from the Institute for
Tropospheric Research (Leipzig, Germany) show rich interactions between
thermodynamic and cloud microphysical processes at fine scales (Siebert et
al. 2006a). For example, we have observed sudden appearance of large
droplets in mixed regions of clouds when the conditions for mixing are
inhomogeneous (e.g., Andrejczuk et al. 2004). Laboratory and cloud
measurements reveal the complexity of cloud-turbulence interactions and are
poised to constrain and stimulate theoretical and computational work on
these problems.

M. Andrejczuk, W. W. Grabowski, S. P. Malinowski, and P. K. Smolarkiewicz
(2004): Numerical simulation of cloud-clear air interfacial mixing. J.
Atmos. Sci., 61, 1726-1739.

A. B. Kostinski, and R. A. Shaw (2005): Fluctuations and luck in droplet
growth by coalescence. Bull. Amer. Meteor. Soc., 86, 235-244.

R. A. Shaw (2003): Particle-turbulence interactions in atmospheric clouds.
Ann. Rev. Fluid Mech., 35, 183-227.

H. Siebert, H. Franke, K. Lehmann, R. Maser, E. W. Saw, R. A. Shaw, D.
Schell, and M. Wendisch (2006a): Probing fine-scale dynamics and
microphysics of clouds with helicopter-borne measurements. Bull. Amer.
Meteor. Soc., in review.

H. Siebert, K. Lehmann, and M. Wendisch (2006b): Observations of
small-scale turbulence and energy dissipation rates in the cloudy boundary
layer. J. Atmos. Sci., 63, 14511466.