[curtis]

I'm not at all certain that's true at the extremes for which black boxes are designed. A difference between thousands of Gs and tens-of-thousands of Gs wouldn't matter for the airplane fuselage, but it might matter for solid-state electronics. I don't know if you'd get that kind of force differential between a water impact and say, slamming into a solid granite wall. But I'm to ready to discount it.

But even if we assume that there really isn't a meaningful difference in the physics of a ground vs. water impact, I think there still might be an advantage to having your conventional black box optimized for maximum survivability and having an auxiliary black box which sacrifices some degree of survivability at the extremes in exchange for improved discoverability after an accident.