Ocean waves are oscillations of the water and air near the air-sea interface that affect the surface elevation, velocities and pressure in both water and air. As a result waves can be measured by recording any of these variables. Most common in situ measuring devices are based on the record of accelerations of a floating buoy (vertical and horizontal accelerations, or vertical acceleration and surface slopes estimated from the pitch and roll of a buoy), pressure in the water column. From space, several techniques are routinely used today, mainly altimetry and synthetic aperture radar imagery.
Computing wave properties with a numerical model
Since the late 1950s and the pioneering work of the Casablanca wave forecasting service, numerical wave models have used a spectral decomposition of the sea state: the energy of the waves is broken up into different components that each have a particular direction of propagation and wave period. Because waves of different period propagate at different speeds, the computer program computes separately the energy of all these components and how they involve in time, and moves them from place to place: the ocean surface is discretized in a number of points. Nowedays, for global scale models, these are separated by 60 nautical miles (100 km) or less. But these may be only a few meters apart in coastal applications.
At each of these points the program manages the evolution of the energy of all the spectral components. Typical models today use 24 directions or more and 30 wave periods or so. This means that at each point ot the ocean we are dealing with 24 times 30 variables that have to be evolved.