MULTIVARIATE ANALYSIS OF WATER QUALITY AND PHYSICAL CHARACTERISTICS OF SELECTED WATERSHEDS IN PUERTO RICO¹

Abstract

Multivariate analyses were used to develop equations that could predict certain water quality (WQ) conditions for unmonitored watersheds in Puerto Rico based on their physical characteristics. Long term WQ data were used to represent the WQ of 15 watersheds in Puerto Rico. A factor analysis (FA) was performed to reduce the number of chemical constituents. Cluster analysis (CA) was used to group watersheds with similar WQ characteristics. Finally, a discriminant analysis (DA) was performed to relate the WQ clusters to different physical parameters and generate predicting equations. The FA identified six factors (77 percent of variation explained): nutrients, dissolved ions, sodium and chloride, silicacious geology, red ox conditions, and discharge. From the FA, specific conductance, sodium, phosphorous, silica, and dissolved oxygen were selected to represent the WQ characteristics in the CA. The CA determined five groups of watersheds (forested, urban polluted, mixed urban/rural, forested plutonic, and limestone) with similar WQ properties. From the five WQ clusters, two categories can be observed: forested and urban watersheds. The DA found that changes in forest cover, percent of limestone, mean annual rainfall, and watershed shape factor were the most important physical features affecting the WQ of watersheds in Puerto Rico.

Spatio-temporal variations in water quality of Nullah Aik-tributary of the river Chenab, Pakistan

“…The multivariate statistical techniques have been widely used in various studies for the explanation of spatio-temporal variations, and interpretation of chemical/physical characteristics of water quality parameters (Simeonov et al 2000;Wunderlin et al 2001;Singh et al 2005b) in comparison to uni-variant techniques that usually fail to give adequate information on multivariate dataset (Santos-Roman et al 2003). In the present study three multivariate techniques such as HACA, PCA/FA and DFA were used for the water quality assessment and interpretation of the results.…”

On the use of multivariate statistical methods for combining in-stream monitoring data and spatial analysis to characterize water quality conditions in the White River Basin, Indiana, USA

“…These newly created equations can be used for assessing water quality at unmonitored streams based on their cluster membership obtained from the physical watershed characteristics (predictor variables; Santos-Roman et al 2003). For example, an unmonitored station with highest C j for cluster 2 would be assessed to have water quality most similar to the water quality of monitored stations in cluster 2.…”

“…However, water quality and hydrologic data is also generally spatio-temporal in nature, hence, a suite of multivariate methods coupled together can provide useful means for incorporating the effect of space and time into the multivariate analyses. For example, linear multivariate approaches that combine principal component analysis (PCA)/factor analysis (FA), cluster analysis (CA), and linear discriminant analysis (LDA) have been used in multiple water quality studies (Santos-Roman et al 2003;Zhou et al 2007;Zhang et al 2009;Iscen et al 2009;Sojka et al 2008). Santos-Roman et al (2003) and Iscen et al (2009) used FA to reduce the number of physical and chemical parameters into fewer variables.…”

“…For example, linear multivariate approaches that combine principal component analysis (PCA)/factor analysis (FA), cluster analysis (CA), and linear discriminant analysis (LDA) have been used in multiple water quality studies (Santos-Roman et al 2003;Zhou et al 2007;Zhang et al 2009;Iscen et al 2009;Sojka et al 2008). Santos-Roman et al (2003) and Iscen et al (2009) used FA to reduce the number of physical and chemical parameters into fewer variables. In addition to variable reduction, SantosRoman et al (2003) and Iscen et al (2009) used FA to characterize the new factors based on the water quality variables that are associated with them.…”

“…In addition to variable reduction, SantosRoman et al (2003) and Iscen et al (2009) used FA to characterize the new factors based on the water quality variables that are associated with them. The Santos-Roman et al (2003) study then used variables derived from the FA to perform a CA that grouped the watersheds into five clusters. A LDA using physical attributes of each watershed was then performed to predict membership into one of the five clusters.…”