基于GAM模型的太湖叶绿素a与营养盐相关性研究

通过分析2013年1月—2015年7月的太湖水体叶绿素a(Chl-a)以及其他指标数据,发现太湖水质存在区域性差异,据此将太湖分为梅梁湾、贡湖湾、竺山湾和主湖区四大区域,引入广义加性模型(GAM模型)对营养盐、环境因子与Chl-a的关系进行分析。结果表明:梅梁湾只有TP与Chl-a浓度的相关性较强,且呈显著的非线性相关;贡湖湾TP浓度对Chl-a浓度的影响是线性的,TN浓度为非线性的,且TN浓度的影响可能更大;竺山湾COD_Mn和TP与Chl-a浓度均呈显著非线性相关,其中以COD_Mn的影响更为显著,可能原因是竺山湾历年来一直是有机污染排放重灾区;主湖区TN和TP对Chl-a浓度的影响均较大,呈显著非线性相关。太湖各区域富营养化爆发的条件不一致,不同的环境因素导致富营养化的条件也不相同。

Relationship of chlorophyll-a and nutrients in Taihu Lake based on GAM model

Data of chlorophyll-a and other environmental factors from January 2013 to July 2015 in Taihu Lake were sampled. It was found there existed significant difference of water quality in different parts of Taihu Lake. Therefore, the lake was divided into four areas, i.e. Meiliang Bay, Gonghu Bay, Zhushan Bay and Main Lake Area. The generalized additive model (GAM) was introduced to analyze the relationship between chlorophyll-a and nutrients and other environmental factors. The results showed that there was significant nonlinear relationship between chlorophyll-a was total phosphorus (TP) in Meiliang Bay. The relationship between the chlorophyll-a and TP was linear, while the relationship between chlorophyll-a and total nitrigon (TN) was non-linear in Gonghu Bay, and TN was possibly the main limiting factor in the bay. There was significant non-liner relationship between chlorophyll-a and TP and CODMn in Zhushan Bay, and CODMn was a more significant factor; the reason may be that Zhushan Bay was the key area of organic pollutants emission. Both TN and TP had a strong effect on chlorophyll-a in main area of Taihu Lake and there was significant non-liner relationship between TN, TP and chlorophyll-a. The outbreaks of eutrophication are inconsistent in different areas of Taihu Lake, and the different environmental factors have resulted in different conditions of entrophication.