利用三维荧光光谱-平行因子分析(EEM-PARAFAC)技术, 研究了2009年春、 秋季厦门湾有色溶解有机物(CDOM)的荧光组分特征, 并利用主成分分析方法对影响该海域CDOM分布的主控因素及其相对贡献进行了解析。 厦门湾CDOM中含有3个类腐殖质荧光组分(C1, C2和C5)及2个类蛋白质组分(C3和C4)。 所有类腐殖质组分之间、 以及所有类蛋白质组分之间均有很好的相关性, 表明同一类型的荧光组分具有相似的来源属性及地球化学行为。 类腐殖质组分的高值区分布在九龙江河口区上游, 而类蛋白质组分的高值区则位于厦门西海域北部, 低值区都位于东部的厦金海域。 排污口附近的局部海域存在污染输入的贡献。 对荧光组分进行的主成分分析结果显示, 陆源径流输入是厦门湾水体中CDOM荧光组分的主要来源; 海区现场生物活动的贡献不大。 这表明对PARAFAC识别的荧光组分进行主成分分析, 有助于实现对水环境中CDOM的不同来源及其相对贡献率的定量解析。

The fluorescent components of chromophoric dissolved matter ( CDOM) in water samples collected from Xiamen Bay in spring and autumn, 2009 were examined using excitation-emission matrix fluorescence spectroscopy combined with parallel factor analysis (EEMs-PARAFAC). PARAFAC decomposed the fluorescence matrices of CDOM into three humic-like (C1: 250, 345/454 nm; C2: 230, 310/374 nm; C5: 265, 424/478 nm) and two protein-like (C3: 230/342 nm; C4: 230, 275/322 nm) components. Good linear correlation occurred among three humic-like components and between two protein-like components, respectively. This demonstrated that the same types of components (humic-like or protein-like) have similar origin and geochemical behaviors. High abundances of humic-like components were found at the upstream zone of the Jiulong Estuary, while the high abundance of protein-like components occurred at the northern part of semi-enclosed Western Xiamen Harbor. The significant negative correlations were found between the abundances of all fluorescence components and salinity in the estuary area. However, the high contents of chlorophyll a were in line with the high abundances of C3 and C4 in non-estuarine area, which implies that phytoplankton activity could be another important source of protein-like components besides the river runoff. A principal component analysis(PCA) of fluorescent components revealed that terrestrial runoff was the dominant sources of CDOM fluorescence components in Xiamen Bay, while the contribution of the in situ biological processes was relatively lower. This study demonstrates that the combination use of PARAFAC modeling and chemometrics (i.e. PCA) is very useful in identifying the origin of CDOM and quantifying the primary factors influencing their distributions.