Bernoulli's Theorem
Look at the gas section. I would think that it would be possible knowing the mass of the particles in the volume and the heat pressures involved. Since everything tends to move from high pressure (high heat) to low pressure (low heat), you can then deduce that particles of higher temp will move towards the lower temp particles. If you want a more accurate simulation, you would have to take into account for thermodynamics in which you account for heat transfer. Looks like an interesting project...I'm just not up to the math...sorry!

Hmm.. I understood none of that post Evert . Can you translate to C/C++ for us non-math experts?

Thanks Evert and Don. I'm going to fiddle with these formulas and see if I can get some currents flowing through my current model. If all goes well, I should be able to accumulate clouds and possibly storms. I'll let you know how it goes.

I'm saving credits in case anybody else wants to pitch in.

********************************* [EDIT] ************************************

Okay, let's put some theory to pratice (mainly because I'm at work and bored). I haven't evaluated all the expressions too closely, see feel free to correct my logic or misunderstandings.

For the purpose of this thread, I'm going to keep it relatively simply. Let's go back to the top-down view, graph-like diagram.

* * * * *
* X * * *
* * * * * 
* * * * *
* * * * *

The *'s above represent the ground at BASE height (for simplicity, let's consider BASE height to be 0). The X represents raised ground. For now, let's assume the temperature is uniform throughout the diagram. I would assume, from what little I know, that the wind, or air particles that make up the wind, fall off the hill and spread out across the flat area:

- ^ - - - 
< X > - -
- V - - - 
- - - - -
- - - - -

The above would be my guess at the where the wind would blow, although the -'s are not necessarily uniform and would have some wind velocity (higher as you got closer to the hill). However, this would mean there's no wind on top of the hill, but that's because I'm looking at this two dimensionally, and technically there'd be many more layers of air. Here's an example of a graph where air rolls up over the X:

* * * * *        - ^ - ^ -
* X * Y *        < < < Y >
* * * * *        - V - V -
* * * * *        - - - - -
* * * * *        - - - - -

Where Y is a spot of ground that is higher than spot X.

Of course, all the above is examining a graph with no temperature differences. Going back to the original graph (with just X), let's pretend the temperature displays like this:

C C C C H
C C C H H
C H H H H
H H H H H
H H H H H

Where C is cold and H is hot. Now, my guess is the following wind model would be used for the graph and temperature pairing:

- ^ ^ ^ <
< X > < < 
^ ^ ^ ^ ^ 
^ ^ ^ ^ ^ 
- - - - -

Kind of messy, but my idea here is looking at the parts of the hill where one side is the same temperature as the top of it would act like the previous wind models. The rest is just kind of a continual flow from hot to cold.

Any comments? I'm going to see what I can accomplish with this much for the time being.

Actually, the area is the size of a planet. Consider each node to be a pixel of a 640X480 screen (or 307,200 nodes). If I know the direction and general velocity of wind at any given spot, I can direct clouds created by water node evaporations. This may seem like overkill, but the conditions of the planet creation are considerably flexible.

For the size of the model, I most likely will have to.

I was also wondering if I need to take water into account. Doesn't water do some weird affects to wind or is it just the temperature of water?

BTW, here's a screen shot of what the model (models are generated randomly at runtime) looks like. No, it doesn't look like realistic terrain, and that's okay with me. I'm not doing any advanced math (or shading), I'm just changing the rgb intensities based on height (which might be how height maps work? I don't know, never worked with one before).

http://comp.uark.edu/~spsilve/terrain.JPG

[EDIT] - typo and forgot some info.