Spatial And Temporal Variation Characteristics Of Water Environment And Water Quality Prediction In Jiangsu Taihu Lake Basin

Jiangsu Taihu Lake basin is located in the core area of Yangtze River Delta,with dense population and developed social economy.In recent years,the water environmental protection and restoration work in the basin has achieved certain results,but the current form of water environmental protection in the basin is still severe.This paper takes Jiangsu Taihu Lake basin as the study area,based on the water quality monitoring data from 2015 to 2020,and studies the water pollution status and spatial and temporal changes of water quality in the basin from the perspective of spatial and temporal differentiation,and evaluate the eutrophication of the main lakes in the basin.At the same time,according to the characteristics of water quality changes in the basin,a water quality prediction model is established.In order to provide reference for water environment management and risk prevention in the basin.The specific research contents and results of the paper are as follows.First,based on the monitoring data of water quality in Jiangsu Taihu Lake basin from 2015to 2020,spatial and temporal analysis of water quality in the basin was conducted using cluster analysis and principal component analysis.The results show that the TN pollution of lake and reservoir waters in the basin is the most serious,but the main pollution indicators such as DO,COD_Mn,COD,BOD₅,NH₃-N,TN,TP,etc.all show a trend of improvement year by year,and show a certain intra-year cyclical change pattern.Cluster analysis divided the water quality of the basin into T1 and T2 in time,corresponding to the flood season（May-September）and non-flood season（January-April and October-December）in the in the basin respectively;the spatial distribution of water quality in the basin shows certain clustering characteristics,which can be divided into two groups,S1 group is mainly distributed in the eastern part of the basin and along the Yangtze River,and S2 group is mainly distributed in the western and northern part of the basin.The results of principal component analysis show that the overall water quality pollution in the basin is mainly nitrogen and phosphorus pollution and organic pollution,while there is a potential risk of heavy metal pollution.The organic pollution is more serious in the flood season,and the nitrogen and phosphorus pollution is more serious in the non-flood season,and the overall water environment quality of the basin shows a trend of high in the east and low in the west and north.Second,a comprehensive trophic state index method based on Monte Carlo simulation was proposed to evaluate the eutrophication of Taihu Lake,Ge Lake,Changdang Lake,Yangcheng Lake and Kuncheng Lake in Jiangsu Taihu Basin.The results show that Monte Carlo simulation can reduce the uncertainty of eutrophication assessment results caused by errors.Taihu Lake,Ge Lake,Changdang Lake,Yangcheng Lake and Kuncheng Lake in the basin have different degrees of eutrophication,most of the lakes are in mild eutrophication,and the eutrophication problem is the most serious in Gehu Lake,which was in a moderately eutrophic state in both 2019 and 2020.Chl.a shows high influence on the integrated trophic state index.Third,a WT-GRU water quality prediction model integrating wavelet transform（WT）and gated recurrent unit（GRU）neural network was proposed to predict DO,COD_Mn,TN,and TP in Jiangsu Taihu Lake basin.The results show that the WT-GRU model can accurately predict the water quality changes in Jiangsu Taihu Lake basin,and the model prediction accuracy is significantly higher than that of GRU and BP models,which proves that the wavelet transform can effectively separate the low and high frequency information in the water quality time series data and reduce the influence of noise.In addition,the prediction accuracy of the model is higher for DO and COD_Mn,which have more stable data distribution and more significant periodic change patterns.