基于多元统计的宁夏沙湖主要污染物季节性变化原因探究

Statistics based study on the seasonal variation of main pollutants in Shahu Lake,Ningxia

以西北地区典型高原封闭湖泊宁夏沙湖为研究对象,分春、夏、秋、冬季(4月、6月、8月、10月)对沙湖水体网格点14个区域主要污染物高锰酸盐指数、氨氮、总磷、总氮、氟化物及叶绿素等常规指标进行监测和统计,同时对不同季节主要污染空间分布特征进行研究.研究结果表明,沙湖高锰酸盐指数和氟化物100%超标,石油类春、夏、冬季100%超标,总氮春、夏、秋、冬超标率分别为43%、64%、7.1%和100%,总磷春、夏、秋、冬超标率分别为50%、100%、85%和85%.相关性分析结果表明,春季氟化物浓度随水温的升高而降低(r=-0.663,P<0.01),总磷浓度随水温的升高而降低(r=-0.673,P<0.01);高锰酸盐指数和石油类会降低水体透明度.夏季叶绿素和高锰酸盐指数存在较强的正相关性(r=0.744,P<0.01),高锰酸盐指数和石油类也存在较强的正相关性(r=0.763,P<0.01);秋季高锰酸盐指数随p H的升高而降低(r=-0.672,P<0.01),石油类和透明度呈现一定的正相关(r=0.784,P<0.01),高锰酸盐指数和总磷呈现一定的相关性(r=0.594,P<0.05);冬季各项监测指标之间没有明显的相关性. 14个监测点位春、冬季节聚为两类,夏、秋季节聚为三类.春季共提出2个主成分,揭示75%的污染来源;夏季共提出1个主成分,揭示53%的污染来源;秋季共提出2个主成分,揭示62%的污染来源;冬季共提出3个主成分,揭示80%的污染来源.

This study mainly focused on the statistical analysis of the relationships between the important water quality indicators of fourteen monitoring locations in Shahu Lake,which is a typical closed lake located on the plateau of Ningxia province. Permanganate index,ammonia nitrogen,total phosphorus,total nitrogen and chlorophyll were monitored and analyzed in the four seasons,including spring（ April）,summer（ June）,autumn（ August）,and winter（ October）.The statistical analysis showed that in all samples,the measured permanganate index and fluoride both exceeded the environmental standards. Around 43%,64%,7.1% and 100% of the total nitrogen measured in spring,summer,autumn and winter,respectively,exceeded the environmental standard value,while for the total phosphorus in the samples,the exceeding ratio were 50%,100%,85% and85%,respectively. The correlation analysis showed that concentrations of fluoride and the total phosphorus of the monitoring locations measered in spring both decreased with an increase of the water temperature（ fluoride： r =-0. 663,P〈 0. 01,TP： r =-0. 673,P 〈0. 01）. Strong positive correlations were found between chlorophyll and permanganate index（ r = 0. 744,P〈 0. 01）,and between permanganate index and petroleum（ r = 0. 763,P 〈0. 01）. The permanganate index measured in autumn decreased with an increase of p H（ r =-0.672,P〈0.01）. The transparency was also positively correlated with the petroleum index（ r = 0. 784,P 〈0. 01）,showing that high permanganate index could reduce the water transparency. In addition,there was a certain correlation between permanganate index and total phosphorus index（ r = 0. 594,P〈 0. 05）. However,no significant correlation was observed between these monitoring indicators in winter. Furthermore,the indicators of the fourteen monitoring locations in spring and winter could be clustered into two categories by the HCA method,while that in summer and autumn seasons could be clustered into three categories. Two components were extracted from the experimental data in spring by the PCA meathod,accounting for 75% of the variations（ the potential pollution sources）. However,only one principal component was extracted from the experimental data measured in summer,revealing 53%of the potential pollution sources. For the experimental data in autumn,two main components were extracted,revealing 62% of the potential pollution sources. Three main components were extracted from the experimental data in winter,revealing 80% of the potential pollution sources.