The use of multivariate statistical techniques for hydrogeochemical study in Suzhou, China

摘要

In order to interpret large and complex data, the multivariate statistical techniques factor analysis and cluster analysis are applied to twenty two groundwater samples from the second confined layer of Suzhou. Factor analysis indicates that three main factors explain over 75.01% of the total groundwater quality variation. Factor 1, consist of concentrations of Ca~(2+), Mg~(2+), Cl~-, Fe~(2+), CO_2, Mn~(2+) and pH, represents the influence of water-rock interaction on groundwater quality. Factor 2, composed of concentrations of NH/, COD, reflects the impact of reductive organic rich environment on groundwater chemical compositions. Factor 3 including concentration of Na~+ shows the effect of cation exchange on groundwater quality. Cluster analysis indicates two different clusters of similarity between the sampling sites. The first cluster comprises a sample influenced by human activities while the second cluster includes the rest samples affected by hydrogeochemical processes. The second cluster is classified into four groups (A, B, C, D) based on their similarity. The hydrogeochemical processes for the four groups are an oxidizing environment, lixiviation process, cation exchange process and an organic rich reducing environment, respectively.