• 关键词：微型生物群落  典型鉴别分析  分类监别分析  相关分析  转折区

• 基金项目：国家自然科学基金资助重点项目(编号39730070)

• 陈建
• 湖北大学生命科学学院生态学研究所, 武汉430062

• 沈韫芬
• 中国科学院水生生物研究所

• 摘要：基于PFU微型生物群落参数调查数据,对湖北省汉江两条污染带11个站点和湖南常德市水系15个站点所属水质组别进行了综合分析研究.所得典型鉴别函数能够对水质组别作出有效的鉴别,分类函数对站点的水质组别分类正确率达到96%;群落参数对化学综合污染指数(P)的响应不呈简单的线性关系,群落参数的变化存在一定模式.提出了“转折区”(turninginterval)概念.水质由洁净到严重污染的变化过程中,在转折区内,微型生物群落的5个参数与P的相关性将发生根本性改变:D和Seq与P由正相关转变为负相关,T90%由负相关转变为正相关,HI和G相关性的显著程度发生改变.本研究显示,这个转折区存在于水质污染区段.

• Abstract：By means of discriminant analysis,integrating data of five biotic parameters (HI,D,Seq,G and T_90%) of microbial communities in water system of 26 stations from two regions,the canonical discriminant functions and classification functions were established and analysis of correlation of paramenters of microbial communities with chemical combined pollution index (P) and correlation among five biotic paramenters have carried out.Groups to which stations belong respectively can be differentiated efficiently by canonical functions obtained.Classification functions obtained can be used directly to classify stations.Group predicted for all stations,except of one station,was same as group marked according to P.Further analysis of correlation of biotic parameters with P indicated that,along water unpolluted to seriously polluted,there was an interval within which characteristics of correlation of five biotic parameters with P changed sharply.In that interval,correlations of D and Seq with P turned from positive to negative, T_90% from negative correlation to positive correlation, HI from non-significant positive correlation to significant positive correlation, G from significant positive correlation to nonsignificant correlation.The turning interval to refer this interval which,according to the present study,may exist in the interval of water polluted.

• Key words：microbial communities  canonical discriminant analysis  classificatory discriminant analysis  correlation analysis  turning interval

• 
  摘要点击次数： 1833 全文下载次数： 3401