作者: Adamu Mustapha , Ahmad Zaharin Aris , Mohammad Firuz Ramli , Hafizan Juahir
DOI: 10.1100/2012/294540
关键词: Partial correlation 、 Chemical oxygen demand 、 Surface water 、 Total dissolved solids 、 Linear regression 、 Chemistry 、 Animal science 、 Biochemical oxygen demand 、 Environmental chemistry 、 Dry season 、 Water quality
摘要: Robust statistical tools were applied on the water quality datasets with aim of determining most significance parameters and their contribution towards temporal variation. Surface samples collected from four different sampling points during dry wet seasons analyzed for physicochemical constituents. Discriminant analysis (DA) provided better results great discriminatory ability by using five ( P 0.05 ) season affording more than 96% correct assignation used six forward backward stepwise in data -value 68.20% 82%, respectively. Partial correlation revealed that there are strong r p = 0.829 moderate 0.614 relationships between five-day biochemical oxygen demand (BOD5) chemical (COD), total solids (TS) dissolved (DS) controlling linear effect nitrogen form ammonia (NH3) conductivity seasons, Multiple regression identified each variable significant values 0.988, R 2 0.976 0.970, 0.942 Repeated measure t-test confirmed surface varies significantly value .