A spectral model for three-dimensional ocean acoustic propagation, published in 2024, demonstrated that Chebyshev polynomial decomposition of the vertical sound field achieves exponential convergence where finite-difference methods stall at algebraic rates. The gain is not marginal. In a benchmark waveguide with a thermocline at 200 meters, the spectral scheme matched reference solutions to six significant figures using roughly one-tenth the grid points. The ocean, in other words, cooperates with modal descriptions of its own interior.
This is not an accident of mathematics. Sound in the sea propagates through layers whose temperature, salinity, and pressure vary smoothly in the vertical but shift slowly along the horizontal. Each layer acts as a waveguide channel, trapping energy into discrete modes that travel at their own group velocities. A pulse emitted near the surface does not simply expand outward — it is parsed by the medium into a finite set of traveling shapes, each carrying information about the depth structure it traversed. The spectral scheme works because it mirrors this physical reality. It decomposes the field the way the ocean already decomposes the sound.
The deeper point is that not all media are equally legible. Air scatters sound into a near-continuum; shallow soil absorbs it chaotically. But stratified water — kilometers deep, thermally layered, pressure-graded — functions as a natural spectrometer. The medium itself sorts the signal. This is why underwater acoustics has always been quietly ahead of atmospheric acoustics in mathematical elegance: the ocean imposes structure on the energy it carries, and the right numerical method merely consents to that structure rather than fighting it.
Every discipline eventually discovers that the most efficient description of a system is the one the system already uses to describe itself.