Study Of Spatio-temporal Distribution Of Nitrogen And Phosphorus Pollutants Of Shu River In Linyi Based On Remote Sensing

Nitrogen and phosphorus pollutants are important indicators for measuring the quality of water environment,and are closely related to the level of water eutrophication and balance of aquatic biological.Recently,the state has carried out a series of remediation actions against heavily polluted river waters,and national water environment has been significantly improved.However,the situation is still not optimistic,and the problem of nitrogen and phosphorus pollution is still the focus of river water pollution prevention and control.Water quality monitoring is the main basis for water quality assessment and water pollution prevention.Compared with conventional water quality monitoring methods,remote sensing technology has the advantages of macroscopic,rapid and low cost,and it can not only carry out long-term continuous monitoring,but also find some pollution sources and pollutant migration characteristics that are difficult to reveal by conventional methods.Therefore,it will have important practical value for studying the spatio-temporal distribution of pollutants in large-scale river by application of remote sensing technology to monitor the concentration of nitrogen and phosphorus in water bodies.Shu river in Linyi is the main water supply source for industrial,agricultural production and people's domestic water in Linyi.It's water quality is directly related to the sustainable development of Linyi's economy and the healthy life of the people.With the acceleration of the industrial and agricultural process in Linyi,the discharge of sewage in the urban area has increased,resulting in the occurrence of excessive nitrogen and phosphorus in the Shu river.The ecological and civilized construction in Linyi will be affected.Taking large-scale basin of Shu river in Linyi as the study area,the dissertation established regression models by using the routine monitoring data of the water from January 2014 to May 2017 and 25 Landsat-8 OLI satellite remote sensing images.The regression models between spectral reflectance of multi-spectral remote sensing data and the measured concentrations of ammonia nitrogen?NH₃-N?and total phosphorus?TP?.The concentrations of NH₃-N and TP in the study area were monitored and inverted by remote sensing,and their spatio-temporal distribution were analyzed,to reveal the trend and the main sources of pollution.It provides a theoretical basis for the control of water environment for the Shu river in Liyin.The major research contributions are as follows:?1?Based on the detailed analysis of physical geography and water environment of the study area,the routine monitoring datas of water quality from January 2014 to May2017 were processed.The results showed that there were obvious variability in the concentrations of NH₃-N and TP in the river,which provided background support for the remote sensing inversion of water quality.Pre-processing analysis was carried out on 25 Landsat-8 OLI satellite remote sensing images,including geometric correction,radiometric calibration and atmospheric correction.And the water body information of the Shu river in Linyi was extracted by using the Modified Normalized Difference Water Index?MNDWI?.?2?Based on the routine monitoring data and multi-spectral data of Shu river in Linyi,the remote sensing inversion models of NH₃-N and TP concentration were constructed,and the accuracy of the established inversion models were tested by measured data.The results showed that:?a?The correlation coefficients?r?for NH₃-N using R/G were the highest for Shu river in Linyi,and the value of r was 0.835.The correlation coefficients?r?for TP using NIR/G were the highest for Shu river in Linyi,and the value of r was 0.833.?b?In view of this,the best inversion models of NH₃-N and TP were established by linear regression method.The co-efficients of determination?R²?were 0.6589 for NH₃-N and 0.7189 for TP.?c?The measured data confirmed that the average relative error for NH₃-N was 17.04%,RSME=12.27%,and the average relative error for TP was 18.74%,RSME=2.93%.It indicated that the established remote sensing inversion models of NH₃-N and TP basically meet the requirements of water quality remote sensing monitoring.?3?The established models of NH₃-N and TP were inverted by using ENVI and ArcGIS software,and their spatio-temporal distribution were analyzed to reveal the trend and the main sources of pollution.The results showed that:?a?from 2014 to 2017,the NH₃-N and TP concentrations showed an overall downward trend,with the seasons showing different variation characteristics.?b?The spatial distribution characteristics of NH₃-N and TP concentrations were basically the same.The NH₃-N and TP concentrations in the river,nearing the upstream,downstream and monitoring sites,were significantly higher than those in other reachs.And during the planting season,the NH₃-N and TP concentrations nearing the monitoring site had the partial increase.From the variation trend of pollutants,the point source pollution caused by domestic sewage and industrial wastewater will be gradually replaced.The non-point source pollution caused by agriculture and the endogenous pollution of the river will be the main factors causing nitrogen and phosphorus pollution of Shu River in Linyi.The local government should pay enough attention.