How rain forms: an in situ experimentGuus Bertens¹, Jan Molacek¹, Gholamhossein Bagheri¹,

Eberhard Bodenschatz¹¹ Dept. LFPN, Max­Planck­Institute for Dynamics and Self­organisation, Göttingen, Germany

P2.02

IntroductionClouds play an important role in the weather and in climate systems. Not only dothey cause precipitation, they also block sunlight from reaching the Earth's surface,and thereby greatly affect the climate's energy balance. However, they are not wellunderstood. To illustrate this, consider that a process as ubiquitous as rainformation cannot be described based on first principles.

Rain formationIn order to form rain, small clouddroplets must grow to at least 50 um, atwhich point they are large enough tostart falling. Condensation, whichinitially forms small cloud droplets, isonly effective up to about 15 um, soturbulence must be responsible forbridging the gap between 15 and 50um. Exactly how turbulence does this,is not known, but several theories exist[1]. Two popular ones are prefential concentration [2], and the sling effect [3].

ExperimentHere we present an experiment that allows us todo in situ measurements of cloud dropletsdynamics. In particular we are able to trackindividual cloud droplets in 3D, which allows us tocompute several statistics such as dropletacceleration and clustering. The experiment iscurrently being developed, and first results areexpected this year.

The experimental setup is shown in the imagebelow. At the heart of the setup is a box withthree fast cameras, that all look at the same (3cm)³ volume from below. The cameras recordimages at 10 kHz. The volume is illuminatedusing a 150 W laser, that comes in from the topso that we observe forward­scattering. In order tokeep droplets in view for longer periods of timethe camerabox is mounted on a 6.5 m long railsystem, that allows it to move with the mean windspeed.

G.Be. J.M. E.B.G.Ba.

OutlookThe aim of the experiment is to see how turbulence affects droplet collision andcoalescence. We are soon able to obtain droplet tracks, and will then advance tocomputing several statistics related to e.g.droplet clustering.

The dynamic properties of droplets dependheavily on their size, but the experimental setupis currently not able to measure droplet sizes.Due to the short coherence length of our laser,we cannot use experimental techniques that relyon interference, such as IPI or GPD. Instead wewant to make use of the recorded dropletintensities and post­process our data using analgorithm that inverts the Mie­scattering.

AcknowledgementsThis project has received funding from the European Union Horizon2020 Research and Innovation Programme under the MarieSklodowska­Curie Actions, Grant Agreement No. 675675.

References[1] B.J. Devenish et al. Droplet growth in warmturbulent clouds. Q. J. Royal Meteorol. Soc.138(667):1401­1429, 2012.

[2] S. Balachandar et al. Turbulent dispersedmultiphase flow. Annu. Rev. Fluid Mech.,42(1):111­133, 2010.

[3] G.P. Bewley et al. Observation of the slingeffect. New J. Phys., 15(8):083051, 2013.

Recorded droplet images, 3 cameras side­by­side: