Spatial Cluster Analysis And Spatial Variability Assessment Of Groundwater Variables

How to use and protect the groundwater source has been causing social attention widely with the freshwater being in short day by day.Groundwater is not only precious natural resource to us all, but also one of the factors which have close relation to human lives. The scientific studies on groundwater, including its natural and disaster attribute, have been being one of the most important items that cause the geosciences experts' attention. Groundwater environmental planning is also an important part in water environmental planning. But the determination on the goal of the planning and decision and choice making of the programs are depended on the existing information acquisition. There are lots of problems in the implantation of monitoring points, such as the randomness and turbid hierarchical structure of the monitoring points , the redundancy of the monitoring data and so on, which have restricted the capability of the monitoring net to supply reliable and available information.The groundwater pollutants migration parameters have dual attributes of designability and randomicity. So it's very difficult or even impossible to study and describe such variables through deterministic mathematic methods and classic probabilistic methods. The appearance of the geostatistics has provided theoretical principles to the description and prediction of such parameters, which has provided the theoretical principles to the re- implantation of the monitoring net, and the possibility to improve the representativeness of the monitoring net.The conclusions are as follow: (1) Based on the theoretic analysis and consultation, by means of some principles such as global ,scientific, normative and so on, an indicators system of impact indicator system of spatial virariety is set up and three layers of total target layer ,rule target layer and specific indicators layer are established; (2) through the Analytic Hierarchy Process (AHP) method, we established a weight system to show the weight distribution of the indicators, which reflected the relative materiality of the indicators,it had been found that the attributes of solutes was the most important; (3) through the Interpretive Structural Modeling (ISM) method, we analyzed specific indicators ,and established ISM with four layers,the impact of certain factor had levels transitivity,which had provided basis for further study; (4) through the spatial clustering principles, we did cluster analysis of the monitoring points and the monitoring indicators of chose region,it had been found that the the spatial clustering was better than the traditional clustering ; (5) the assessment and contrast of the variability of the original data and the transacted data by cluster analysis had been done, which had proved the cluster analysis was reasonable and valid.