There's not underground glaciers. Glaciers are always on the surface, and flow due to gravity. You're thinking of permafrost (which can form wedges and lenses o

There's not underground glaciers. Glaciers are always on the surface, and flow due to gravity. You're thinking of permafrost (which can form wedges and lenses of pure ice). Deeper permafrost responds slowly, but on the flip side of that, there never would have been deep permafrost in S. Quebec during the last glacial maximum. The permafrost situation also wouldn't affect isostatic rebound much, because the amount of water in the ground wouldn't change appreciably. A rule of thumb is that only the first two meters of ground responds appreciably to annual variability, which also means that only the very near surface temperatures have the possibility of affecting snowmelt. Thus, any remaining deep permafrost would have almost no effect. It's too much of an insulating overcoat to allow for meaningful heat flux on those timescales.