For example, a wave that loses amplitude is said to dissipate. The precise nature of the effects depends on the nature of the wave: an atmospheric wave, for instance, may dissipate close to the surface due to friction with the land mass, and at higher levels due to radiative cooling.
Dissipating forces are those which can not be described by Hamiltonian formalism. Loosely speaking, friction and all similar forces which result in decoherency of energy, that is, conversion of coherent or directed energy flow into an indirected or more isotropic distribution of energy.
In computational physics, a numerical dissipation is also known as "artificial dissipation" or "artificial diffusion" or "numerical diffusion". They all mean this: when the pure advection equation--which, by definition, is free of dissipation--is solved by a numerical approximation method that reduces the amplitude and changes the shape of the initial wave in a way analogous to a diffusional process, the method is said to contain 'dissipation'.
A formal, mathematical definition of dissipation, as commonly used in the mathematical study of measure-preserving dynamical systems, is given in the article wandering set.