Paul J. Curran. Multispectral Scanner Data to Estimate and Map the Concentration of Sewage Released from Sea Outfalls[J]. Journal of Remote Sensing, 1987, (2): 126-139. DOI: 10.11834/jrs.1987019.
Multispectral Scanner Data to Estimate and Map the Concentration of Sewage Released from Sea Outfalls
To monitor the dispersion and dilution performance of a sewage outfall a dye can be injected into the sewage and the resultant plume tracked by boat. In recent years remotely sensed imagery has proved to be a useful adjunct to boat survey as it provides a synoptic view of a plume at an instant which
as a result of tide
wind and current changes
a boat survey is unable to do.The aim of this survey was to use airborne multispectral scanner (MSS) imagery and a methodology of modeling
inversion and accuracy assessment to estimate and map the concentration of three dyed plumes of sewage. After the radiometric
atmospheric and irradiance correction of the airborne MSS data
a standardised radiance index was highly correlated with dye concentration (r = 0.85
significant at the 99% confidence level). Following inversion and extrapolation of this relationship to all of the airborne MSS data
twenty two plume plots were produced. For a seven-class classification of dye concentration these plume plots had an accuracy of 49-90% at the 95% confidence level increasing to 68-100% at the 95% confidence level for a two-class classification.