Vacuum dumps to zero
While I'm no expert, this is typical of all internal combustion engines.
The "Source" of the vacuum reading in this case is taken at the intake manifold, where the intake stroke (piston moving down) during a constant RPM (like at idle) creates a negative pressure, or vacuum, within the manifold.
So, atmospheric pressure, or "0" vacuum, is attempting to force its way past the throttle and into the intake at this stage. This differential (or difference in pressure) is measured by the vacuum gauge. For lack of a better term, the downward stroke of the piston is attempting to "Suck Air" and fill the volume of the cyl, with the throttle "Preventing" the air from filling that volume. Thus, RPMs are controlled (Along with a number of other factors, such as fuel metering.)
So, when the throttle is opened rapidly, there is a slight delay in piston reaction, meaning, the rpms do not increase to compensate for the incoming rush of air right away. At this point, most or all vacuum diminishes (bleeds off) and there is no longer any differential to measure... Vacuum has been momentarily lost.
BUT.. The piston speeds or RPMs are increasing, and if the throttle is held partially open (no boost yet), they "Catch up" and a vacuum is once again created. When you snap the throttle closed, the air is almost shut off, but the piston speeds are still quite high... You will see a spike on the vac gauge... almost 30 in/hg... a perfect vacuum, before it levels off.
We haven't even talked about supercharging yet! Basically, supercharging is a method of further increasing the pressure differential. Rather than only relying on the pistons ability to "Suck" air in, we want to go beyond atmospheric pressure and force even more air into the cyl. volume... "Fake Cubic Inches" if you will!
If you really think about it, the relationship between "Bigger" camshafts and resulting lower vacuum readings begin the make sense... Not due to lift, but because of valve overlap (When the intake and exhaust valves are open at the same time) In this case, the more overlap you have, the more likely atmospheric pressure "sneaks its way in" from the exhaust side and lowers the pressure differential. The throttle is no longer the only "Point of entry" for holding back the atmospheric pressure.
I'm sure that others will correct my oversimplifications, but I hope this helps you out!