Featured Stories

View of the SAFL building from the river


*Reprinted with permission from the University of Minnesota Aerospace Engineering and Mechanics Department*


How fast does rain fall? How fast does snow? After centuries looking at the sky and trying to make predictions on weather and precipitations, one might think we should know the answer. In fact, we don't.

How fast does a heavy sphere fall through a still fluid? That shouldn't be a hard question for anyone who took a basic physics course. If the sphere is microscopic, let's say a solid particle or a droplet, the answer should be even easier: small objects shed small wakes, which means the fluid is weakly perturbed by their fall, greatly simplifying the problem. It turns out we know the answer only if the flow is perfectly still, which is a condition hardly ever met in reality. If instead the fluid is characterized by chaotic and vortical motion (that is, if it is turbulent), then we are again walking in the dark. This is unfortunate, since our skies (and rivers, and oceans) are very turbulent, as everyone who has ever been on an airplane knows. The interplay of turbulence and heavy particles is a complex one, with the latter departing from the fluid trajectories just enough to make their behavior even more elusive  (if possible) than that of the underlying flow. Hence our ignorance about the fall speed of rain and snow  - and hail, aerosol, dust in sandstorms, etc. Without this knowledge, making prediction on weather, precipitation, pollution and contamination in the environment is just a lot harder than it could be.

Motivated by this unsolved problem, Professor Filippo Coletti and PhD students Douglas Carter and Alec Petersen have built an installation at St. Anthony Falls Lab (SAFL) able to generate a large volume (about one cubic meter) of homogeneous air turbulence, in which they inject solid particles and water droplets. The facility, which they called CloudIA (Cloud of Inertial Aerosol), consists of a chamber featuring two facing panels that accommodate 256 individually controlled air jets (see pictures at right). These are actuated in random sequence, producing high velocity fluctuations but negligible mean flow. The apparatus is unique in that an observer inside the chamber sees conditions similar to what he/she would experience flying in high altitude atmosphere and moving with the wind. Using techniques such as Particle Image Velocimetry (PIV) and Particle Tracking Velocimetry (PTV), Douglas and Alec are discovering that the fall speed of heavy particles (such as microscopic glass beads) in turbulent air can be about three times higher than in still air, greatly exceeding qualitative trends reported by previous theoretical and numerical studies.

In an effort to bridge the gap between laboratory studies and real natural phenomena, Professor Coletti and his postdoctoral associate Andras Nemes have joined SAFL members Professors Michele Guala and Jiarong Hong, and ventured to the field station of Rosemount, MN, to capture the motion of actual snowflakes. Using stage lights and high speed cameras at night time, they tracked snow particles (pictures at left) falling through the turbulent atmosphere at tens of meters from the ground, simultaneously monitoring the wind properties via a meteorological tower. Snowflakes appear to behave much like heavy particles do in the turbulence reproduced in laboratory or by numerical simulations, and in particular they fall about four times faster than they would in still air.



This axis of research in Coletti's group has just taken off. He and his students are looking forward to exploring important and poorly understood aspects of the problem: What is the effect of the particle shape? How much the particles (or snowflakes, or rain drops) alter the air flow itself? How much does the picture change when the small object waltzing through turbulence are not falling dead, but propel themselves, like swarming insects?