I've been wanting to jump into this discussion, but haven't had time. I'll have to keep it (sort of) short for now, and can provide more details later.

This whole concept of relative motion is classic Lagrangian frame of reference, where the "observer" is a moving particle of material (liquid or solid). This can lead to great simplifications of the equations of motion, where the relative fluid motion is all that matters, independent of the absolute motion of the frame of reference.

Unfortunately, those simplifications only apply when the frame of reference is non-accelerating, and inertial forces are weak relative to other forces such as viscous friction and gravity. Factoring the equations of motion into non-dimensional variables leads to dimensionless parameters such as the Reynolds number, Peclet number, Froude number. Of the many equations and dimensionless variables, the Navier-Stokes equation and Reynolds number are most relevant here, and with low Reynolds number the equations can usually be simplified to the more simple Stokes equations that can often be solved analytically without computers. Reynolds number is the ratio of inertial forces to viscous forces:

Unfortunately the geometry of the open seas has such a large characteristic length (typically the depth of the body of water) that any motion at all leads to a high Reynolds number, meaning that flow has a lot of inertia and very little viscous dissipation. This means that the energy of fluid motion has nowhere to go except to create eddies and waves, which are basically turbulence. Flow in narrow channels (or pipes) has a much smaller characteristic length, and if it's slow enough it will have a low Reynolds number leading to laminar flow, free of any eddies. In terms of energy transport, what happens is that the two major components of flow (momentum and vorticity) both diffuse to the rigid surface that encloses the liquid, taking energy away from the liquid and preventing turbulence and minimizing waves. But in the open seas, there is no such rigid surface nearby to absorb the energy, so the water churns away.

High Reynolds number flows require the full Navier-Stokes equations, which include nonlinear terms which are what lead to the eddies and turbulence. But Lagrangian frame of reference is almost impossible in this situation, requiring Eulerian frame of reference instead (where coordinate system is at rest). Wave action thus becomes dominated by non-linear effects that are more complicated than wind speed relative to water. In such a case, 10 kt wind against 3 kt current is

__very__ different than 16 kt wind with 3 kt current.

In low Reynolds number "creeping" flows, if the force that causes the flow is removed, the motion stops instantaneously (because there is no inertia). This is the situation where relative motion is all that matters.

In high Reynolds number flows, removal of forces (such as wind that's creating the waves) will eventually allow the seas to calm, but not instantaneously. Inertia causes the waves to propagate, sometimes for days and over thousands of miles, particularly in very deep seas where the is no solid surface to absorb the momentum or vorticity. That's why a storm in the North Atlantic can cause heavy swells in the Caribbean a week later.

As for the motion of the Earth, I think the surface actually moves about 1000 mph at the equator (not 24,000 mph). IIRC, the Earth's circumference is about 24,000 miles, and it spins once every 24 hours, so 24,000/24=1000.

The reason why Lagrangian frame of reference works on land (despite the high speed) is that the land mass of the Earth is a solid, and solid mechanics are different from fluid mechanics.