Some liquids can be compressed, others resist compression. I think the hydraulic oil in my car jack resists compression, as well as my brake fluid.

I also understand water resists compression, too.
There are many factors affecting level, pressure, etc... but I'd do a visual check as Simon has suggested, while cold. If you have the kit and the time, then get yourself set up to do a cold pressure test and re-inspect - your LCA may go off when you inflate the system if there's a big air pocket in there to compress, or if hoses are bulging under pressure. If nothing obvious, then either go straight for a bleed, or if you're feeling brave, run it up to temperature/pressure and re-inspect the hoses. You can always run it up to temperature with the cap off while bleeding, then attach a pump and inflate to 10psi or whatever - in my opinion, a safe way of ensuring there's no air in there
and inspecting hot, pressurised hoses. If it was my van and I had the time, this is pretty much what I'd do.
I can't see why level would rise if you were relieving -ve pressure and open up to atmosphere - imagine if you put your mouth over the header tank neck and sucked as hard as you could - level would only rise - definitely not fall - so when you stopped sucking, level would drop, if at all. +ve air pressure in the header tank will bear down the surface of the coolant, which will cause the level to drop until air pressure in the header tank drops, and/or the coolant pushes back just as hard, due to elasticity in hoses or an air pocket somewhere being compressed and pushing back up. (It's not a perfect picture of the physics, I know.)
If the van was stone cold, say you put the cap on at ambient 0 degrees and took it off at 20 degrees, I calculate you'd have just under 2psi in a healthy, well sealed system, and level would be higher by about 100ml, depending on initial fill level (not sure what that is in cm rise, but my point is that with rising ambient temperature, you'd hear a hiss opening the cap). But it hasn't been frosty yet, and there may be another explanation - take the rest of this post with a pinch of salt though...
I mention this, only because it happened to me and you got me thinking... a possibility is that you're over pressuring in operation - in my case, a leaky head gasket with which I drove 3000 miles before replacing it. Because it was marginal and the van was still driving, and because I hadn't properly diagnosed it yet, I hooked up a pressure sensor to the coolant system and tapped the voltage off the TPS, and logged them both to a memory card - the graphs I got clearly showed pressure increasing with a heavy right foot.
Anyway - not saying this is your problem, but what I'm getting at is that hot combustion gases entering the coolant system would increase the total volume of air, which would result in higher pressures (greater volume, greater temperature => greater pressure), until the header tank cap relieves pressure to keep the system at or under the caps rating of ~16psi. All things being equal, if you cool a healthy system down from it's happy running pressure of something like 10-13 psi (depending on fill level, coolant strength, hose condition, etc...), you're likely to end up back down to atmospheric pressure, or perhaps even this -1psi, where the negative valve in the cap sucks air back in again. If you cool a max-pressure 16psi system down, the coolant and air will contract as they cool, lowering the pressure, but if you were over pressure to begin with, I believe you'd end up with a +ve pressure when at ambient temperature.
So my experience suggests, and I might be wrong, that if you have a +vely pressured system at ambient temperature, then if you're van was stone cold and there hasn't been a large rise in ambient temperature since putting the cap
on last time, you may have had an increase in gas volume in the system.
As for the LCA - if +ve pressure in the tank dropped the level significantly to sound an LCA that turned off when you opend the cap, then Simon was spot on to suggest inspecting hoses and possibly bleeding. I'd bet on the latter, myself - can't see a few psi swelling cold hoses significantly.