Remote Sensing Monitoring And Spatial-temporal Characteristics Of Colored Dissolved Organic Matter In Zhanjiang Bay

Colored dissolved organic matter?CDOM?,one of the important water color parameters,is an important research object of ocean color remote sensing.Remote sensing monitoring of CDOM in Case-?waters of coastal and how to construct CDOM retrieval model in Case-?waters of coastal which is applied to particular locations has always been a research challenge at home and abroad.Zhanjiang Bay is a natural deep-water port.Affected by busy port transportation,offshore steel and iron production,petrochemical industries,urban life and sewage discharges,Zhanjiang Bay is characterized by serious eutrophication in recent years.The water quality is rating at four or inferior four category frequently.It is of great significance to study the CDOM optical properties of waters in this area for better understanding the inherent optical properties of the Case-?waters and the changes of the marine ecological environment.Based on the survey data of the four voyages in Zhanjiang Bay from 2016 to 2018,the spatio-temporal characteristics of the optical properties and the source of colored dissolved organic matter in the waters of Zhanjiang Bay are discussed.The results show that:In terms of seasonal changes,it manifest that the colored dissolved organic matter absorption coefficient?a_g?355??and spectral slope(S_275-295)in the surface waters of the harbor of autumn are greater than that of summer while summer are greater than that of spring and minimum in winter.In terms of spatial distribution,the a_g?355?of the thalassic surface and bottom layers of the three seasons?spring,summer and winter?basically shows a decreasing trend from the north to the south and from the west to the east,meanwhile the S_275-29575-295 presents a patchy distribution in the horizontal space generally.According to the comprehensive analysis of the colored dissolved organic absorption coefficient,spectral slope,chlorophyll a,dissolved oxygen and turbidity,we can draw a conclusion that the colored dissolved organic matter in the Zhanjiang Bay waters mainly comes from terrestrial input and phytoplankton on-site production in the three seasons?spring,summer and autumn?,the sediments resuspension has a certain contribution to the source of colored dissolved organic matter in the spring and autumn.Furthermore,the concentration of colored dissolved organic matter was negatively correlated with DO in the three seasons?spring,summer and autumn?accompanying a certain indication function for water pollution and eutrophication.At the same time,the convection and diffusion of sea water,topography and residual currents in the bay will affect the spatial distribution of CDOM.Rainfall and wind speed may be one of the reasons for seasonal variation of CDOM.The correlation analysis of each band of Landsat-8 OLI image and its band combination with CDOM concentration a_g?440?shows that the band combination?B1+B2?/2 is the most relevant to CDOM concentration in autumn voyage and the band combination?B1+B3?/2 is the most relevant to CDOM concentration in winter voyage.The retrieval models of autumn and winter are established by choosing the combination of these two bands.The quality of remote sensing images in spring and summer is not good,and remote sensing retrieval is not carried out in these two seasons.Finally,the thematic map of remote sensing retrieval can be obtained by image retrieval.We use the water sample extracted from 20 stations of the Zhanjiang Bay,and live spectral measurements of these stations during the May 2017 voyage to analysis the correlation between normalized remote sensing reflectance and CDOM concentration?a_g?400??.The CDOM neural network model is established.The results show that the average relative error and root mean square error of the BP and RBF neural network model were far less than other models and that the fitting effect between the predicted values of the neural network model and the measured values was better than the effect came from the multiple linear regression model.That is to say the neural network model is more suitable for the remote sensing estimation of colored dissolved organic matter in Zhanjiang Bay.