One of the main problems with interpreting satellite data over coastal zones is that brightness variations due to water depth confuse the interpretation of substrate types. The above diagram demonstrates that the radiance received at the sensor is a complex interaction of sea-floor reflectance, water depth, water surface and water column effects, illumination and atmospheric parameters. The Geoscience Australia Shallow-water-mapping algorithm uses a simplifying equation to extract the effects of changing water depth so that both water depth and seafloor colour can be directly obtained. The algorithm can be applied to multispectral images from satellites or airborne sensors, and to conventional colour aerial photography.

A detailed description of the algorithm can be found in:

Bierwirth, P. N., Lee, T. J., and Burne, R. V. (1993). Shallow sea-floor reflectance and water depth derived by unmixing multispectral imagery. Photogrammetric Engineering and Remote Sensing, Vol. 59, No. 3, pp. 331-338.