以发展近岸海域富营养化快速评价技术为目的，基于有色溶解有机物（CDOM）吸收光谱及三维荧光光谱（EEMs），结合溶氧(DO)、叶绿素a（Chl a）、浊度（Tur）等可现场检测的参数，利用平行因子分析（PARAFAC）、偏最小二乘回归（PLSR）等化学计量学方法，建立了近岸海域富营养化快速评价技术。利用PARAFAC解析CDOM的EEMs，得到3个荧光组分，即类腐殖质C1、C2及类蛋白质C3；选择a220、a254、a270、a350、C1、C2、C3等CDOM光学参数以及DO、Chl a、Tur等水质参数，考察其与总磷（TP）、总氮（TN）的相关性，在此基础上，以富营养化指数(TRIX)方法为参照，利用PLSR建立了富营养化评价模型，其训练集和验证集的相关系数分别为0.9721和0.9827，验证集的分类准确率为89.0%。结果表明以CDOM光学参数代替TP、TN发展近海富营养化快速评价技术具有可行性。

The aim of this study is to develop a rapid offshore eutrophication assessment technique by parallel factor analysis (PARAFAC) and partial least squares regression (PLSR) based on fluorescence excitation-emission matrix spectroscopy (EEMs) of chromophoric dissolved organic matter (CDOM), UV-visible spectroscopy of CDOM, dissolved oxygen (DO), chlorophyll a (Chl a) and turbidity (Tur). EEM-PARAFAC model identifies three fluorescent components, two humic-like (C1 and C2) and one protein-like (C3). a220, a254, a270, a350, C1, C2, C3, DO, Chl a and Tur are selected to study their correlations with total phosphorus (TP) and total nitrogen (TN). Using trophic index (TRIX) as the reference, the model for rapidly evaluating eutrophication status is developed by PLSR. The correlation coefficients of the PLSR model are 0.9721 for the training data and 0.9827 for the validation data，and the classification accuracy is 89.0% for the validation data. The results prove that optical parameters of CDOM can be as substitutes for TP and TN in rapid assessment of offshore eutrophication.