Ice is pretty buggy unpredictable, so I don't think anyone can say for sure what will happen when you mix it with lava.
I do know that you can't remove glacial ramps from an area that will melt, or you'll be left with completely impassable tiles with no material.
In that case, I guess I'll work on the !!Science!! here, as I'm in an ideal location. I'm just wondering if the "pop-up" effect (step 5) is supposed to happen here; it makes no sense to me.
ETA: I think I might have figured out at least one of the ways the mechanics might work here.
So there are several points to this mechanics:
1) When ice melts
2) When ice refreezes
3) How lava and water flows mix
So 1 in most situations is pretty obvious - whenever there is lava within 1 square of the ice. However, when actual mixing happens, I've been seeing some weird stuff - in general, when lava is on top of ice, first ice melts, then the lava falls and mixes, and 2 ticks later, the corner-connected ice squares melt. From this, I think that I can get the following:
1) As soon as lava hits a square, whether or not it's about to mix, every ice square around it gets a "melt" signal (or, in more temperature terms, warms up).
2) When lava and water mix, the squares that are side-connected to it lose the melt signal (or lower their temperatures enough not to eventually melt), but the corner squares retain that signal (or stay warm enough).
The next phenomenon to explain is the "popup" effect. The only explanation I can think of is the following:
When lava and water meet, they form obsidian, but the "total fluid" count must stay there (although it seems to decrease by 1; I plan on testing this more extensively). Then when melting happens at the corners, this "flow" goes to other squares, which in turn "overflow" and pop up.
That's the only thing I can think of, and I'm thinking it doesn't quite work as I have said, but it's the only thing that makes sense to me so far.
=Uncool-