TY  - JOUR
T1  - Application of Multivariate Statistical Methods to Assessment of Water Quality in Selected Locations of the Lagos Lagoon, Nigeria
AU - Ladipo, M.K. AU - Ajibola, V.O. AU - Oniye, S.J. 
JO  - Environmental Research Journal
VL  - 6
IS  - 3
SP  - 141
EP  - 150
PY  - 2012
DA  - 2001/08/19
SN  - 1994-5396
DO  - erj.2012.141.150
UR  - https://makhillpublications.co/view-article.php?doi=erj.2012.141.150
KW  - multivariate analysis
KW  -spatial and temporal differences
KW  -Water quality
KW  -Lagos Lagoon
KW  -Nigeria
AB  - Multivariate statistical methods, i.e., Cluster Analysis (CA), Principal Component Analysis (PCA) and Discriminant Analysis (DA) were used to assess temporal and spatial variations in the water quality of the Lagos Lagoon during the wet period (July 2007 and 2008) anddry seasons (February 2008 and 2009). The study was focused on nine locations of the lagoon, specifically to describe the distribution of water physicochemical parameters and identify the parameter (s) that most influence the distributions observed. Physicochemical parameters (pH, EC, salinity, turbidity, DO, BOD<SUB>5</SUB>, COD, TSS, TDS, alkalinity, NO<SUB>3</SUB>, PO<SUB>4</SUB> and SO<SUB>4</SUB>) were used to study spatial and temporal variations in water quality of these locations. The descriptive statistics of the average values obtained for each location during the period of study were discussed. The results obtained from the detailed chemical analysis of water from the different sections of the lagoon confirmed the dynamic nature and diverse chemistry of the water. Multivariate analysis of obtained data during the periods of study further reflects this diversity during each of the periods samples were collected. The loading pattern of principal components showed some variations during each of the period of sample collection. The processes or sources associated with the principal components obtained during the different sampling periods are highly localized and contributed mainly by anthropogenic sources. Hierarchical CA grouped the nine locations into three based on the water characteristics during each period of sample collection. Hierarchical CA and PCA did not give a clear trend in temporal distribution of the parameters. As a result it was difficult to determine a constant similarity between locations during these periods however, DA showed EC and TDS were the only good predictors or discriminant variables in all the locations during the period of investigation.
ER  -