You probably have a lot of locations that are on top of each other and you disable different layers to show these locations? This is an archaic and inconvenient method, it's much easier to have a location nearby and just move the character there when descending to a new level....
In your case, you don't need multiple obstacle maps. You need to add new objects to the obstacle map when descending to a new floor, and recalculate the obstacle map.
This all sounds confusing, I think a small example with two floors would help explain it better.
This is all accurate, I guess I could move everything to the main layer, but I will need to bound the camera to the size of the main map, then let it unbound when the player's go to the underground (which will be off the visible main map).
"The obstacle maps are based off of square size and buffer size. If you have a unique square size (32x32) vs (30x30) or buffer size (-2 vs -1) for example they will record into a separate maps. Giving the map a different name doesn't make it different."
I saw this from another post, but couldn't get this to work. If I could change the buffer size on creation of the enemy, then maybe I could get this to work, but it doesn't look like there's an option for that. I think my best bet will be to move everything to one layer instead of overlapping assets. Very unfortunate that C3 doesn't have more flexibility with the obstacle maps/pathfinding.