基于主成分分析法的梁子湖水质评价分析

摘要

This research analyzed the major pollutants distribution in Liangzi Lake,Hubei Province in July,October and December of 2012.Ammonia nitrogen (NH4+-N),total nitrogen (TN),total phosphorus (TP),potassium permanganate index (CODMn) and dissolved oxygen (DO) were selected as water quality parameters for comprehensive water quality evaluation during high water level,low water level,and normal level.Statistically,we used SPSS and principal component analysis (PCA),combined with ArcGIS and ordinary Kriging interpolation,and semi-variation function calculation to demonstrate the water quality variations spatially during high water level,low water level,and normal level.The results showed that Liangzi Lake is rich in nutrients and is in the process of becoming eutrophicated.The selected water quality parameters and pollutants were very high near the lake shores and inlets with less eutrophication and better water quality in the center of the lake.Water quality in Liangzi Lake is poor and worse during the low water level and better in the high lake water stage.SPSS and principal component analysis are good statistical methods in studying the combined water quality parameters to determine the stages of eutrophication,and ArcGIS and Kriging interpolations are very useful methods in displaying the spatial distribution and variation of water quality.%依据梁子湖区2012年7月、10月和12月的水质监测数据,分析不同时期水质的主要污染因子和水质空间分布特征.选取氨氮(Nh4+-N)、总氮(TN)、总磷(TP)、高锰酸盐指数(cODMn)和溶解氧(DO)等5个水质指标,利用SPSS软件和主成分分析法,分别得到丰水期、平水期、枯水期梁子湖区水质主要影响因子及综合水质评价函数;基于ArcGIS软件,采用普通克里金插值和半变异函数模型,分别得到丰水期、枯水期和平水期的水质等级分布图,并对不同时期梁子湖水质进行空间变异特征分析.分析得到,不同时期梁子湖水质主要影响因子和污染特点有所差异,主要以富营养污染为主,梁子湖的水质污染存在明显的区域性,湖区中心的水质较好,入湖口附近水质较差;不同时期梁子湖水质状况为,枯水期优于平水期,平水期优于丰水期.实验结果表明,基于SPSS软件,利用主成分分析方法对水质监测数据进行综合水质评价是一种较好的统计分析方法;基于ArcGIS软件,使用普通克里金插值方法对于显示水质的空间分布和水质变化具有较好的效果.