The Study Of Satellite Remote Sensing On Concentration And Temporal-spatial Distribution Of Suspended Sediment In The Yangtze Estuarine And Near-shore Waters

Medium Resolution Imaging Spectrometer (MERIS), as a payload of ENVISAT, was launched by the European Space Agency (ESA) on 1 March 2003. It is one of the most advanced satellite-borne ocean color sensors so far. In this thesis, MERIS data processing and MERIS-derived suspended sediment concentration (SSC) were discussed. The algorithms of SSC retrieval based on single-band and multi-band models were proposed. The contents are detailed as follows:·Preprocess of satellite images, including atmospheric correction and geometric correction. In this work, a MODTRAN code, based on a radiative transfer model, was performed to the atmospheric correction. Thus, a nearly-true remote sensing reflectance from top of atmosphere (TOA) radiance could be inferred.·Analysis of remote sensing reflectance spectra. The remote sensing reflectance was estimated on the basis of in situ measurements and tank-based experiments.·Modeling of remote sensing reflectance and SSC at MERIS bands. The logarithmic algorithms were established between SSC and the spectral reflectance at fixed single-band, shifted single-band and multi-band ratio, respectively.·Mapping spatial distributions of SSC. The SSC was retrieved from MERIS image taken on different time phases. The error between the retrieval SSC and the in situ SSC is about 50% using the logarithm model based on single band, which is lower than that from MERIS level 2 products. As a result, the spatio-temporal distribution of SSC can be affected by variations of river discharges due to the shift of wet and dry seasons, tidal currents and wind fields in the Yangtze River estuary.·Comparisons between MODIS-derived SSC and MERIS-derived SSC and between MODIS surface reflectance and MERIS surface reflectance inferred from TOA radiance. By compared with in situ SSC data, it was found that the SSC yielded by MODIS OC3 algorithm was underestimated. Therefore, surface reflectance of MODIS level-2 products was adopted for the proposed algorithms of this thesis. Consequently, it was shown that SSC retrieved from the MODIS surface reflectance could be applicable in highly turbid coastal waters, as the supplement of absence of MERIS FR data.