Strangerbarry wrote:Re: "Normal cloud bottoms are flat. This is because moist warm air that rises and cools will condense into water droplets at a specific temperature, which usually corresponds to a very specific height. "
Wouldn't this mean the TOPS should be flat ?
As different parcels of clear air rise, they all cool at the same rate and reach the condensation temperature at the same level, and the cloud's base appears flat. Then, as different parcels of cloudy air rise (towards the top), their vertical velocity will change based on how much they mix with the environment. If a cloud is still warm enough to be rising, it will not have a flat top, but a buoyant bubbly looking top.
The parcels of cloudy air will eventually reach a level where they are no longer buoyant; you'd think they'd all stop there and be flat. But some may still have positive vertical velocity by that level and overshoot the top, bouncing around a bit before spreading out flat. You are probably familiar with flat top clouds, anvil clouds. That's cloudy air that has reached it's non-buoyancy level at a very specific temperature, with corresponds to a very specific height because the atmosphere is so spread out horizontally. Explaining the distribution of cloud top heights is the fundamental question a convection parameterization answers for a climate model.
To explain mammatus clouds (which almost always form on the underside of anvils) you must explain their bubbly buoyant appearance. The text under the APOD uses the old explanation, that bigger cloud particles settle in the lobes and are sinking. Another explanation is that as the anvil spreads over warm dry land it is heated by radiation from the ground, and develops a convective bubbly appearance. The mammatus lobes are just what's left behind as other sections of the anvil are gently heated and begin to rise up. This would also explain why mammatus rarely form when it is very humid at levels below the anvil (water vapor absorbs the ground's radiation, reducing what the anvil is exposed to). Since this theory explains more but is slightly simpler, it is more compelling.
Signed,
A grad student reading too many papers about this
http://journals.ametsoc.org/doi/abs/10. ... lCode=atsc