A bubble in gel scatters light. I think it scatters an amount of light proportional to its cross-sectional area, which is proportional to the square of pressure. If so, the average amount of light scattered from the bubble is a function of the RMS of the pressure, rather than the average pressure; oscillating pressure will increase the total light scattered.
Gel without bubbles can operate as a fiber optic, restricting light to within itself by total internal reflection.
By connecting a piezoelectric hydrophone to a bubbly gel column, you can generate a traveling wave in the column. But if the column is closed, you can generate a standing wave, in which bubbles in some parts of the column move back and forth, while bubbles in other parts merely grow and shrink. I think you can in fact generate an arbitrary superposition of standing waves simply by taking the Fourier transform of the desired display scattering function. Then you can illuminate the gel column with total internal reflection.
A somewhat better approach is probably to illuminate the column with a small duty cycle so you don't have to deal with the multiple nodes and antinodes.