Study On Early-warning Of Water Quality And Water Levels In Urban Groundwater

Groundwater comprises an important part of the water resource as a whole, and is an essential source of water for agricultural, industrial and domestic purposes. However, overexploitation of the groundwater resource has led to continuous deterioration in water quality and decline in water level. To maintain a harmonious relationship between water resource and economic development, the management of groundwater resource should be strengthened by conducting early-warning analysis on groundwater with modern scientific early-warning techniques.Based on a great number of surveys and in consideration of the actual situation of the area studied, the definitions and features of the early-warning on the groundwater have been proposed. The process of the early-warning of the groundwater consists of five parts, i.e. specifying the target, searching for the source, distinguishing the sign, predicting the degree and eliminating the menace. The difficulty lies in the latter two.Early warning must be grounded on assessment and prediction. Therefore the paper, on the basis of the data available, has analyzed the changes in the main index of groundwater quality and level, as well as assessments of the water quality using international F value method, composite index method, Nemerow Index method and fuzzy evaluation method respectively. After comparison it is found that the fuzzy evaluation method reflects more objectively the grade of groundwater quality. In addition the BP neural network is also applied in assessment of groundwater quality and the result is desirable. The early-warning degree for the groundwater quality and level are: best status, better status, general status, baddish status and worst status. The early-warning degree for the groundwater quality is in correspondence with the five categories of groundwater. And early-warning degree for the groundwater level is mainly accomplished by comparison between the current and warning water level.The prediction of quality and level of groundwater has adopted the research methodology of random statistics. Due to the limitations of data available and actual demand, the paper has only carried out a short-term prediction. The support vector machine is first applied, and the result is compared against that of the BP nervenetwork. The accuracy of prediction varies in the two methods.Based on the actual data from the area studied, the paper then carried out an early-warning on the current situation and change tendency in the quality and level of ground water. The studies show that the quality of groundwater in the area studied is too low-grade for drinking due to the excessive amount of florin in the area studied, but applicable for industrial and agricultural use. The tendency for changes in quality is not obvious. The tendency of changes in groundwater level varies in different aquifers. It is suggested that the exploitation of different wells be further optimally configured.Finally the causes and mechanisms for groundwater pollution in the area studied are analyzed, and suggestions for restoration are proposed in reference to the current situation of researches both at home and abroad.