In accordance with the data which were experiment of mixing-ratio in water tank and collecting water samples in situ from natural seawater and urban sewage discharged into the sea along Dalian coast of the northern Yellow Sea in February and April 2012, with quinine sulfate and sodium humate as a reference, the calibration curve was established among CDOM (Chromophoric dissolved organic matter) concentration and fluorescence intensity and reference wave absorption coefficient. To calibration curve as the foundation, the CDOM samples concentration of various sources was determined after analyzing CDOM sample from Dalian coast of the northern Yellow Sea sewage into the sea and natural sea. Based on the comparative analysis on CDOM fluorescence fingerprint, the main component of water CDOM were determined. The results showed that in Dalian coastal waters of the northern Yellow Sea, the main component of CDOM in natural seawater is tryptophan and in urban sewage discharged into the sea are tryptophan, tyrosine, and humic acid. On the basis of comprehensive analysis of CDOM fluorescence and absorption spectral, the thinking of synergy inversion of CDOM absorption spectral slope S by connecting fluorescence and ocean color remote sensing is put forward.
Petroleum pollution is a key indicator to monitor and assess water environment in petroleum fields. Five sessions of field
work were made in Liaohe River in Panjin city, Liaoning province of China in 2006 and 2007. Field water spectra and
concurrent water samples for laboratory measurements of chlorophyll, petroleum pollution, and suspended material were
collected. An important feature of water spectra influenced by petroleum pollution was found to show that there are three
peaks and two troughs in spectral curves. The peaks are at 570-590, 680-710, and 810-830nm, while troughs are at
650-680 and 740-760nm. The field spectra were used as to correspond to Landsat TM bands to establish estimate models
of petroleum pollution concentration. The models were applied to the Landsat/ TM image on 11th Oct 2006 to obtain the
distribution image of petroleum pollution. The accuracy is up to 80% for petroleum pollution estimation with the
validation of reserved samples. The result shows that the estimate models from remotely sensing data provide an
effective means to obtain rapidly and low-cost the distribution of petroleum pollution concentration in the study area.