Call me crazy, if you want, but I think the accepted plate mapping currently used is... interesting & in need of a lot of tweeking. There are "sea bottoms" near the tops of peaks.
Near the tops of some peaks. And the flood only lasted for a part of a year, not long enough at all for the development of anything like a seabottom. Go digging down in the mire at the bottom of a lake reservoir a year, or five, after the valley has been flooded. You will find nothing that looks like a seabottom, with seashells, plants, etc. You'll find a drowned piece of muddy land.
The flood didn't last long enough for there to be sea bottoms formed over most of the earth.
Beyond that, uplifting has raised the former seabed in some areas, but by no means all. We do not find uplifted sea bottoms in all the great mountain chains of the world, only here and there.
However, you do touch upon an interesting point. All paleogeology, paleogeography, paleobiology, esp. the crucial radioactive decay dating techniques used to date very old things, rely upon the uniformitarian assumption: that things happen today as they happened in the past, and that in the past things happened at the same speed. Certainly it's the most defensible assumption imaginable, because it looks at what is and has been recorded for 400 or so years and says: no variation in this time period, so extrapolating back is reasonable. Reasonable, but not completely bulletproof. When we look at starlight and say that such light was radiated millions of light-years ago, and thereby calculate the distance to those stars by multiplying the speed of light, we make the uniformitarian assumption.
If that assumption is wrong, then virtually anything could happen in the past at rates of speed that seem wildly accelerated today.
To return to your point about decendants of Adam, the traditional view would require Cain's wife to be his sister, or perhaps niece, grand-niece, et al.