摘要
本文利用化学模式识别技术的主成分分析（PCA）法对厦门西港及香港维多利亚港的18个沉积物样品进行分类研究。所涉及的变量（环境指标）包括重金属Cu、Pb、Zn、Cd，有机污染物DDT、PAH、PCB、HCH，及碱性磷酸酶活性（APA）。正确的分类基于变量的适当选择，即选择代表来源不同的样品之典型特性的变量组合方式。结果表明，运用PCA技术可获知大量样品的统计特征，对分辨来源不同的样品颇具潜力。
Pattern recognition technique was was applied to classify the sediment samples fromXiamen Western Harbour and Victoria Harbour of HOng Kong. In this study the technique ofprincipal component analysis (PCA) of pattern recognition to distinguish the sediment sampleswas adopted. The variables included the heavey metals Cu.Pb. Zn.Cd, the organic pollutantsDDT.PAH .PCB .HCH, and alkaline phosphatase activity. The optimized classification is based onwell selecting the variables. The variables selected, on the other hand, most Often represert thetypical characteristics of the sediment samples collected from the different harbours. The resultsshow that PCA technique may offer the potential strategy in the distinguising classes of the environmental samples rome the different origins.
出处
《海洋环境科学》
CAS
CSCD
北大核心
1998年第3期61-66,共6页
Marine Environmental Science
基金
国家自然科学基金!No29477273
