Study On Water Quality Simulation Method Of Xiangxi River Based On Remote Sensing

The eutrophication of Xiangxi River in Yangtze River is on a rising trend after the impounding of Three Gorges Reservoir,which gradually becomes a essential factor affecting the water quality and ecological security of the Three Gorges reservoir area.In order to study the mechanism of eutrophication,many scholars have carried out a lot of numerical simulation research on water quality of Xiangxi River in recent years.Water quality mathematical model requires a lot of nutrients and other monitoring data as a support,large-scale,however,it is much difficult to obtain high-density and long-term water quality monitoring data.With the development of remote sensing inversion technology in water quality field getting more and more mature,establishing a water quality model based on remote sensing to generate continuous,full-coverage and good-quality data is of great scientific significance to study Xiangxi River eutrophication.Regarding Xiangxi River as study area,this thesis analyses the temporal and spatial distribution characteristics of water quality as well as the water spectrum characteristics,then builds a remote-sensing inversion model based on satellite data.In the meantime,a mathematical model is established to simulate the hydrodynamic and water quality process of Xiangxi River.By comparing and analyzing the two model results,a method of combining remote sensing inversion and water quality model is proposed,which aims to improve the simulation accuracy of water quality model and provide data support for Xiangxi River eutrophication research.The main contens of the thesis are as followed:(1)The water quality level is analyzed by using single factor evaluation method,comprehensive pollution index evaluation method and comparison method.The results show that different levels of pollution appearance in different monitoring spots,in general class IV;water quality indicators concentration differ in temporal and spatial distribution,in the view of continuity on time and space,there is a breakout of algal blooms from July to August in 2010;total nitrogen concentration distributes in high level in the downstream and low level in the upstream while opposite trend happens in total phosphorus.(2)Based on the spectrum data and synchronous water quality data,the analysis of water spectral characteristics is carried out.And the correlation between chlorophyll-a,total nitrogen,total phosphorus,silicate concentration and spectrum was further studied.The results show that chlorophyll-a concentration and reflectance are highly correlated,and clear spectral characteristics exist in Xiangxi River water bodies.It is feasible for Xiangxi River to inverse chlorophyll-a concentration with remote sensing method.Indirectly water quality indicators can also be inversed through its relation with chlorophyll-a.(3)Based on the Landsat satellite image and the measured water quality data,the thesis analyses the correlation between water quality indicators and reflectance of band combination and establishes a remote-sensing inversion model by using linear regression technique.Reliability and fault tolerance of the model is analyzed and we applies the model in Xiangxi River in further.The results show that the established model is with high-precision simulation while the spatial trend is same as actual situation.Satisfyingly,the point-source pollution location of phosphate rock and other residential areas can be reflected in some ways.(4)A three-dimensional hydrodynamic model of Xiangxi River is established,and two periods of 2010 are selected to calibrate and validate the model.As the calibration result,roughness of the study area is valued as 0.026.Fow velocity and water temperature are preferably simulated in the calibration and validation,which reflects reasonably the Xiangxi River hydrodynamic characteristics in the vertical layer.In the wind state,the water flow direction in the Xiangxi estuary is consistent with the windless state,and the velocity of the water surface is more obvious driven by wind.The middle and bottom flows are opposite direction to the surface flow and less weakly driven by wind.(5)The thesis establishes a three-dimensional water quality model and selects two periods to calibrate and validate the model.Total nitrogen and total phosphorus varying process are preferably simulated in the calibration and validation,which indicates the established model is suitable to simulate the concentration of chlorophyll-a,total nitrogen and total phosphorus in 2010.Moreover,chlorophyll-a and total phosphorus is more accurate than total nitrogen in simulation,within 25%averaged relative error.(6)Based on the comparison of remote sensing inversion and mathematical model simulation results,this thesis presents a method combining inversion result into mathematical model.The combining model simulation shows that the simulation accuracy of coupling model is higher than traditional model.In space aspect,the coupling model results not only reflect the actual effect of the phosphate rock and other residential areas on water quality,but also show the temporal and spatial changing process of Xiangxi River water quality.Compared to the separate remote sensing model and mathematical model results,the simulation effect has improved.