Department of Civil and Environmental Engineering, University of Queensland, Brisbane, QLD, 4072, Australia.
Environ Monit Assess. 2011 Mar;174(1-4):529-45. doi: 10.1007/s10661-010-1476-6. Epub 2010 May 12.
The chemometric techniques were applied for evaluation of the seasonal variation of water qualities at 17 stations along a stretch of the Bagmati river of Nepal for 23 water quality parameters measured during 1999-2003. The application of discriminant analysis confirmed the classification of the water quality measurements into three seasons: pre-monsoon, monsoon, and post-monsoon affording 93.8% correct classification. Factor analysis and box-whisker plots facilitated to investigate the seasonal variation of water quality and the pattern of pollution sources. Application of FA revealed that the influence of water quality parameters changes from season to season. A parameter that is most important in contributing to water quality variation for one season may not be important for another season. Comparison of the discriminant analysis and factor analysis helped to identify the most important water quality parameters, as water temperature, DO, EC, COD, CL, Ca, alkalinity, PO(4)P, and TP, that are most important for seasonal variation and play a significant role in establishment of water quality control strategy.
采用化学计量学技术评估了尼泊尔巴格马蒂河 17 个站点的水质季节性变化，对 1999 年至 2003 年间测量的 23 个水质参数进行了分析。判别分析的应用证实了水质测量可分为三个季节：前季风期、季风期和后季风期，正确分类率为 93.8%。因子分析和箱线图有助于研究水质的季节性变化和污染源模式。FA 的应用表明，水质参数的影响因季节而异。一个在一个季节对水质变化贡献最大的参数，在另一个季节可能并不重要。判别分析和因子分析的比较有助于确定最重要的水质参数，如水温、溶解氧、电导率、化学需氧量、氯化物、钙、碱度、PO(4)P 和总磷，这些参数对季节性变化最为重要，对制定水质控制策略具有重要意义。
