大宁河叶绿素a的因子分值-多元线性回归预测模型研究

PREDICTION OF CHLOROPHYLL-A IN THE DANING RIVER BY PRINCIPAL COMPONENT SCORES IN MULTIPLE LINEAR REGRESSION MODELS

自从2003年三峡水库蓄水以后,其支流大宁河水华频发且频率呈上升趋势。叶绿素a(Chl-a)是指示水体中浮游植物初级生产力的有效指标。采用因子分值-多元线性回归方法研究了大宁河水体11个相关水质因子与Chl-a之间的相关关系。因子分析用于简化水质指标之间的相关性,以因子得分(Score values)为自变量用于多元线性回归分析中。结果发现log(Chl-a)与6个因子分值明显相关,所获得多元线性回归模型为:log(Chl-a)=0.579-0.191×(Score 1)-0.013×(Score 2)-0.013×(Score 3)+0.042×(Score 4)+0.134×(Score 5)-0.059×(Score 6),相关系数R=0.731、相关系数的平方R2=0.535,说明自变量可以解释因变量53.5%的差异性。实测数据验证结果表明:该模型能够较好的预测2010年1～10月水体中Chl-a浓度的峰值和基本变化趋势。

Since 2003 after impoundment of Three Gorges Reservoir its tributary-Daning River began with frequent water blooms,and its frequency is increasing and expanding.Chlorophyll-a（Chl-a） is a well-accepted index for phytoplankton abundance and primary productivity in an aquatic environment.The relationships between Chl-a and 11 water quality indices in Daning River were studied by principal component scores in multiple linear regression analysis.Principal component analysis was used to simplify the complexity of relations between water quality indices.Score values obtained by principal component scores were used as independent variables in the multiple linear regression models.The results showed that log（Chl-a） was found to have significant linear relationship with 6 score values from the principal component analysis of variables,and predicted Chl-a values of Daning River were obtained from the following model：log（Chl-a） = 0.579-0.191×（Score 1）-0.013×（Score 2）-0.013×（Score 3）＋0.042×（Score 4）＋0.134×（Score 5）-0.059×（Score 6）,R=0.731 and R2（goodness of fit）=0.535,indicating that 53.5% of variation in Chl-a could be estimated by this modeling approach.The results of further validation using a new dataset suggested that the modeling obtained in this study successfully simulated Chl-a peak concentrations and developed tendency from January to October in 2010.