| one publication added to basket [17385] | Validation of semi-empirical ocean colour algorithms using subsurface reflectance R(-) data
Santos, S.V. (1996). Validation of semi-empirical ocean colour algorithms using subsurface reflectance R(-) data. MSc Thesis. VUB: Brussel. 40 pp. |
| Available in | Author | | Document type: Dissertation
|
| Abstract | During the Remote Sensing North Sea (1988-1989) REMONO and Particular Matter North Sea (1993-1995) PMNS campaigns carried out at the Netherlands Institute of Sea Research many optical (e.g. radiance L, irradiance E, superficial reflectance R(+) and subsurface reflectance R(-) calculated values) and non-optical (e.g. Bio-geochemical and physical parameters) have been measured and stored in a comprehensive Information system (an Optical data Bank) for further processing. All year semi-empirical Ocean Colour algorithms applied to Coastal zone environments have been developed during the PMNS project using surface reflectance values R(+) acquired with the Spectrascan Spectracolorimeter (PR650). These algorithms were developed using the suitable channels for the new forthcoming ocean colour satellites (SeaWIFS, MERIS) and the visible channel of NOAA-AVHRR. The ratio R(+)/R(-) was also determined during this project. During the REMONO project subsurface reflectance values R(-) were calculated from the measurements below the water by the Advanced Spectral Irradiance Meter (ASIR). Non-optical parameters such as Chlorophyll, Total suspended matter and Dissolved organic matter have also been collected. In this research we tried to use the PMNS algorithms with the REMONO data. A transformation of the R(+) algorithms to algorithms that use subsurface irradiance R(-) values was necessary for this because the information used has been derived from subsurface radiance R(-) values. Therefore a preliminary transformation was made in order to obtain the PMNS algorithms as functions of R(-) values. Statistical analyses have been applied to this transformed algorithms by validating the predicted values against measured in-situ data collected during the REMONO campaign. A selection of spectra has been made in order to increase the accuracy of our analysis. Finally a new validation was performed to establish the applicability of these algorithms over specific in situ data. |
|
