4 ms·
Because "magnitude of the gravitational force" =/= "gravitational potential" and potential is what matters- the shape is stable if the potential is equal across
by gliese1337 13y ago
Because "magnitude of the gravitational force" =/= "gravitational potential" and potential is what matters- the shape is stable if the potential is equal across the entire surface. This corresponds to the gravitational force vector always being perpendicular to the surface (i.e., gravity always points 'down'). As long as the gravitational force vector is perpendicular to the surface, nothing will get pulled sideways, therefore nothing will move, therefore the shape is statically stable, and the magnitude of the perpendicular force can vary as much as you like.
Now, you may be thinking "but shouldn't the bits of planet-fluid under stronger gravity sink and push on the fluid around them and thus indirectly cause stuff to move around to the sides?" This is where equal potential really becomes important. For one bit of the planet to sink (losing potential energy), another bit of the planet would have to rise (gaining potential energy). If the shape of the planet currently conforms to an equipotential surface, any marginal redistribution of which bits are up and which bits are down will end up requiring at least as much total energy as the original configuration. Thus, the shape is a local optimum in the energy configuration space, and nothing will move.
- GhotiFish 13y ago> Now, you may be thinking "but shouldn't the bits of planet-fluid under stronger gravity sink and push on the fluid around them and thus indirectly cause stuff to move around to the sides?" that was what I was thinking. So the reason we see gradients is because areas that are experiencing low force have lots of mass, and areas experiencing high force have little mass. Looking at the diagrams, I guess I can see that is the case. Thanks for clearing that up!