Cyber-Angel:
I had to think for a while to write a concise and accurate answer, so sorry for the delay in responding. The atmosphere is a fluid, a compressible one and unless terminology varies CFD in my experience refers to the flow of both compressible and incompressible fluids. In the atmosphere there is no need consider "hypervelocity" and the complications that introduces.
For atmospheric models the atmosphere can be considered incompressible for horizontal flow. We know the atmosphere is not incompressible and motion is rarely only horizontal, but considering the scale of forces and the magnitude of the velocity - the atmosphere behaves as if it is incompressible.
In the atmosphere it is vertical displacement what ever the cause, that makes weather. Even at its most vigorous vertical motion is much slower than the time it takes for the pressure of a rising or sinking parcel to adjust to the changing ambient pressure, so a parcel rising, for example in a thunderstorm updraft, is at the same pressure as the environment at the same altitude.
In general because of the number of computations involved, condensation is assumed to take place if there is sufficient moisture. The measure of that is relative humidity which indicates if the water vapor pressure is near the equilibrium value. (When the RH is 100% meteorologists often use the poorly chosen term saturation).
The composition of the particular cloud condensation nucleus of course will determine the degree vapor condenses in cloud models, but because a great deal of evidence is lacking about the mix of compositions of a population of cloud condensation nuclei, models assume that if sufficient moisture is available condensation will take place. "Sufficient" of course is a function of temperature.
The
Bergeron-Findeisen process describes the water vapor environment in many clouds. In a mixed-phase cloud both water drops and ice crystals coexist. Because the equilibrium vapor pressure in the microscopic water environment is greater than the equilibrium vapor pressure in the ice environment, the random molecular motion of water vapor molecules dictate that the ice crystal will grow efficiently and the water drop evaporate. Eventually ice crystals "clump" and exceed the mass the cloud updraft can keep aloft. If clumps survive long enough and fall below the freezing level, they melt and the many resulting drops then go through the
collision- coalescence process resulting in raindrops. This is not the only possibility because precipitation does fall from clouds composed completely of water and entirely of ice and as we know snow and rain are not the only two possibilities.
In atmospheric models phase change is considered in bulk, that is individual cloud condensation nuclei are not considered. If the cloud environment is below an assumed temperature the phase change is assumed to be vapor >> ice and the latent head of the phase transition is calculated. If the temperature is above an assumed value then the phase change is assumed to be vapor >> liquid
and the latent heat is calculated for that particular transition. It is much more complicated than this brief explanation.
There are many good references for cloud microphysics which treat the formation of precipitation and the Bergeron-Findeisen process. It is named for the Swedish meteorologist Tor Bergeron who proposed it and German meteorologist Walter Findeisen who helped refine the theory. Interestingly a very basic form of the theory was proposed in 1911 by German meteorologist Alfred Wegener who is now remembered for continental drift.
Here are a couple links for more information:
On Tor Bergeron
http://www.cimms.ou.edu/~schultz/papers/TorBergeron.pdfand this one is a great summary of cloud microphysics:
http://www.met.utah.edu/class/jimsteen/ATM619/lectures/cloud_microphysics.pdf***JDex, where in Cincinnati are you?***
-Steve Horstmeyer
