Spatiotemporal Analysis Of Water Quality In Taihu Lake Basin And Remote Sensing Inversion Of The Main Pollutant

Taihu Lake Basin is one of the regions with the largest population density and the most developed economy in China.However,in the process of rapid economic growth in the past 40 years,the water body has been subjected to heavy pollution load for many years,and the basin has been faced with perennial water shortage induced by pollution.In order to fully understand the water quality status of Taihu Lake Basin and the variations,based on the monthly water quality data of 565 sites covering the whole Taihu Lake Basin released by the Taihu Basin Authority in 2018,this paper explored the spatiotemporal variation pattern of water quality in the basin through correlation analysis and cluster analysis,found out the main parameters affecting the water quality and analyzed their spatial distribution,and identified the influencing factors of water quality through multivariate statistical technology.In addition,in view of the serious nitrogen load in water body caused by excessive application of nitrogen fertilizer in agriculture,this paper used remote sensing data to establish an inversion model of NH₃-N through partial least square regression,so as to carry out spatial continuous estimation of NH₃-N concentration in agricultural areas of this basin.The main research results include:（1）In the water quality evaluation results,only 29.09%reached the drinkable level,42.2%was mildly polluted,16.14%was moderately polluted,and 12.49%was severely polluted.As can be seen,the drinking water resources in the basin were insufficient,and mild pollution was the common water quality status.The temporal variation of the proportion of“drinkable”was closely related to the temporal variation of the proportion of“moderate and severe”,and the temporal variation pattern of them can be described as"seesaw effect".Therefore,the key to increase the“drinkable”water is to control the“moderate”and“severe”water.The areas with strong human activities（Shanghai,Wuxi,Changzhou）have the worst water quality,while the areas with low population density（Tiaoxi）have the best water quality.（2）Ammonia nitrogen（NH₃-N）,total nitrogen（TN）,total phosphorus（TP）,five days biochemical oxygen demand（BOD₅）,chemical oxygen demand（COD）,permanganate index(COD_Mn)and dissolved oxygen（DO）were the main parameters affecting the water quality of Taihu Lake Basin.Among them,the proportions of moderately and severely polluted annual exceeding standard sites of TN（lake,reservoir）and NH₃-N were 96%and 66%,respectively,while the annual exceeding standard sites of the other parameters were almost mildly polluted.NH₃-N was the main pollutant leading to the deterioration of water quality in the most of densely populated urban areas.The lakes and reservoirs with the worst water quality were severely polluted by TN.Therefore,NH₃-N and TN should be the main emission reduction targets of water environment treatment.（3）Compared with the four seasons,the spatial clustering of rivers in dry and rainy seasons can more clearly show the influence of rainfall and temperature on the spatial pattern of water quality.The sites located in the densely populated urban areas had the worst water quality and the water quality characteristics did not change with the seasons.The main factors affecting the water quality in urban areas were the large amount of domestic sewage and industrial wastewater discharge.In winter and dry season 1,anoxic sites were dispersed,mainly caused by the enrichment of oxygen-consuming nutrients.However,in summer and rainy season,anoxic sites were distributed in the downstream of rivers leading to the main streams and lakes.The slow flow in the downstream makes the pollution stay longer,and at high temperature,the biodegradable organic matter in the sites was fully decomposed by active microorganisms through consuming much dissolved oxygen.In addition,domestic sewage and aquaculture wastewater were the main pollution sources in rural and suburban rivers.The spatial pattern of water quality of lakes and reservoirs did not change with the seasons.Lakes in the northwest of the basin,the northwest of Taihu Lake,Dianshan Lake,and Qingshan Reservoir had the worst water quality.The main factors were agricultural non-point source pollution,domestic sewage and phytoplankton.（4）Excessive application of nitrogen fertilizer will make crops grow but the unabsorbed nitrogen fertilizer will flow into the rivers with surface runoff,which will increase the concentration of NH₃-N in rivers.Based on the indirect relationship between crops and the NH₃-N concentration,the inversion model between remote sensing variables and the measured NH₃-N concentration data was established by partial least squares regression.The corresponding inversion model can account for79.64%of the variation of NH₃-N concentration in the area where agricultural non-point source pollution was the main pollution source.In the rural areas with bigger population density,the corresponding inversion model can only explain 58.45%of the variation of NH₃-N concentration,because the pollution sources of NH₃-N also included various point source pollution.The spatial continuous estimation results of the inversion model showed that the NH₃-N pollution in the northern and southern rural areas was more serious than that in the eastern parts of the basin.In this paper,the variations of water quality in Taihu Lake Basin were studied in detail,which can also provide useful information for the decision-making of pollution control and water resources management in the basin.