中文摘要：

通过对我国主要红树林区表层沉积物中总汞含量和环境因素的测定,研究了红树林湿地沉积物中汞的分布规律和影响因子之间的关系.结果表明,我国红树林湿地沉积物中总汞的含量范围是2.3-903.6 ng·g^-1,平均值为（197±137.6）ng·g^-1,福建洛阳桥[（467.5±68.8）ng·g^-1]、浮宫[（438.2±147.0）ng·g^-1]、鹅湾[（264.3±89.2）ng·g^-1]、姚家屿[（125.4±27.1）ng·g^-1],海南三亚[（164.8±143.9）ng·g^-1]、东寨港[（314.1±335.7）ng·g^-1],广东深圳[（179.9±7.7）ng·g^-1]等7个红树林区的汞含量均显著高于当地土壤背景值,汞污染严重;而福建云霄[（63.3±43.9）ng·g^-1]、广东高桥[（178.6±127.0）ng·g^-1]、广西大冠沙[（26.1±18.8）ng·g^-1]、山口[（73.8±21.1）ng·g^-1]、北仑河口[（117.8±51.4）ng·g^-1]、钦州湾[（147.5±107.6）ng·g^-1]等红树林区汞含量接近于背景值,基本无污染.红树林湿地中汞含量的差异是众多环境因素和人类活动相互交叉影响的结果.红树林湿地中,pH值、粒度、有机质、Fe/Hg、Mn/Hg与总汞含量显著相关.其中,黏粒和Fe/Hg是影响沉积物中汞含量最显著的因素.环境因素对总汞含量的综合影响作用,可以用线性回归方程表示.

英文摘要：

Surficial sediment samples were collected from main mangrove wetlands of China, mercury, pH, salinity, organic matters, grain size, Fe/Hg, Mn/Hg were analyzed. Mercury content ranges from 2.3 to 903.6 ng·g^-1 , with a average of （197 ± 137.6）ng·g^-1 . Compared with local background level, serious Hg pollution with high Hg content was found in 7 areas, including Luoyangqiao [（467.5 ± 68.8） ng·g^-1], Fugong [（438.2± 147.0） ng·g^-1], Ewan [（264.3 ± 89.2） ng·g^-1], Yaojiayu [（125.4±27.1）ng·g^-1], Fujian Province; Sanya [（164.8± 143.9）ng·g^-1 ], Dongzhaigang [（314.1 ± 335.7） ng·g^-1 ], Hainan Province, Shenzhen [（179.9 ± 7.7） ng·g^-1 ], Guangdong Province. Hg content was similar with background value in the other 6 areas, including Yunxiao [ （63.3 ± 43.9） ng·g^-1 ], Fujian Province; Gaoqiao [ （ 178.6 ± 127.0） ng·g^-1 ], Guangdong Province; Daguansha [（26.1 ± 18.8） ng·g^-1 ], Shankou [（73.8 ± 21.1 ） ng·g^-1 ], Beilun estuary [ （ 117.8 ± 51.4） ng·g^-1 ], Qinzhou Bay [ （ 147.5 ± 107.6） ng·g^-1 ], Guangxi Autonomous Region. Discrepancy of total Hg concentrations in mangrove wetlands is due to many environmental factors and human activities. Parameters such as grain size, pH, organic matter, Fe/Hg, Mn/Hg are significantly correlated with total Hg respectively. Clay and Fe/Hg are obvious factors influencing total Hg concentration. Impact of environmental factors on mercury can be showed by regression equation.