Risk Assessment And Early Warning Model Of Groundwater Pollution

In recent years,with the acceleration of urbanization and industrialization,groundwater in China has been seriously polluted,which seriously threatens the drinking water safety for urban and rural residents and restricts the development of regional society and economy.Pinggu District,located in the northeast of Beijing,is the link between Beijing and Tianjin.In view of the serious problems of groundwater pollution caused by human activities such as industrial and agricultural production,urbanization and other activities in recent years,the plain area in Pinggu was studied in four aspects:1)investigation of relevant data of physical geography and hydrogeological conditions,and detection and analysis of groundwater samples;2)assessment of groundwater quality and pollution;3)assessment of inherent vulnerability,pollution source load,function value and pollution risk of groundwater;4)construction of early warning model for groundwater pollution.Based on the above study,the following conclusions were drawn.(1)Single index assessment and comprehensive assessment method were used to assess groundwater quality and pollution.The results show that chloride,chromium(VI),manganese,arsenic,mercury,volatile phenol,cyanide and oxygen consumption in inorganic evaluation indexes of groundwater quality in plain area of Pinggu are I,II or III groundwater at all sampling points.Iron,nitrite nitrogen,nitrate nitrogen,fluoride,ammonia nitrogen and total hardness are IV and V groundwater at most sampling sites.With regard to comprehensive quality of groundwater,the samples of II and III groundwater account for 76.84%of the total samples,and the ones of IV and V groundwater account for 16.19%.Only 6.97%of the samples are I groundwater.The main pollution indexes of groundwater in the study area are nitrate,iron,ammonia nitrogen and total hardness.The comprehensive pollution degree of groundwater is mainly moderate pollution,followed by heavy pollution and light pollution,and only two serious pollution points appear.(2)The DRASTIC model was used to evaluate the inherent vulnerability of groundwater.The results show that the medium vulnerability area is the most distributed in plain area of Pinggu.The next are secondary and highest vulnerability areas,and the lower one is distributed in lowest area.The sensitivity analysis of groundwater inherent vulnerability assessment index shows that the most important factor affecting the inherent vulnerability in the study area is the lithology of aeration zone,followed by groundwater depth.(3)Assessment of groundwater pollution source load was done based on investigation results of groundwater pollution sources.The results show that the load of the groundwater pollution source in the plain area of Pinggu is mainly distributed in the middle and south-west areas,while the one in the east and west areas is relatively weak.Based on the results of the groundwater quality and enrichment evaluation,the assessment of the groundwater function value is carried out by using the matrix superposition method.The results show that the groundwater function value of the study area is at a high level and the groundwater resources are relatively short.The groundwater quality is relatively good and the exploitation degree is high.(4)Based on the assessment of inherent vulnerability of groundwater,pollution source load and function value,the matrix superposition method was used to assess the risk of groundwater pollution.The results show that the second highest pollution risk area is the most distributed in plain area of Pinggu,followed by medium and high pollution risk areas.The distribution range of lower pollution risk area is the smallest.The risk assessment results of groundwater pollution with nitrate as a characteristic pollutant were verified.The results show that there is a strong correlation between pollution risk and nitrate concentration,and the assessment results are effective and reliable.(5)The prediction model of groundwater pollution probability was established by using the method of logical regression analysis.The results show that the indexes in the groundwater of four aquifer groups in plain area of Pinggu which have a high impact on the occurrence of groundwater pollution or not are similar.The pseudo R~2 of four logical regression models all reaches a high level,which demonstrates the models fit the monitoring data well.The correct rate and overall prediction accuracy of the models for groundwater pollution are above 90%,indicating the prediction results are good.The assessment model of groundwater pollution degree was established by stepwise regression analysis.The results show that the indexes of significant impact on groundwater pollution degree in the four aquifer groups in the study area are basically the same.The fitting degree of the four stepwise regression models is above 0.3,demonstrating the models fit well to the index of pollution degree.The Durbin-Watson statistics of the models all meet the test standard,which indicates the models have a good ability to explain the degree of pollution.(6)Based on the quantitative assessment results of groundwater pollution possibility and pollution degree obtained by regression analysis,the early warning level of groundwater pollution in the study area was graded.The results show that the groundwater pollution warning level in plain area of Pinggu is mainly three-level early warning,followed by secondary and fourth-level ones.The first level and zero level early warning areas are less in spatial distribution.The fourth-level early warning area is mainly distributed in urban area of Pinggu and the southwest part of the study area.The pollution degree and the deterioration trend of groundwater quality in these areas are serious.