Statistical techniques represent a reliable tool for classifying, modelling and interpreting surface water quality monitoring data, particularly for lakes. The complexity associated with the analysis of a large number of measured variables, however, is a major problem in water quality assessments. Multivariate analysis, such as cluster analysis and factor analysis (FA), was utilized in this study for the analysis of water quality data (including water discharges and 28 water quality parameters) for Akkulam-Veli Lake, a tropical coastal lake system in Kerala, India. This lake is partially divided into two sub-systems, namely Veli Lake and Akkulam Lake. Akkulam Lake exhibits freshwater characteristics, in contrast to Veli Lake, which exhibits saline water characteristics because of its close proximity to the sea. Thus, studying this lake provides insights into water quality variations in both a freshwater and saline water lake in a tropical region. Water quality patterns and variations in Akkulam-Vela Lake over three seasons, including pre-monsoon (PRM), monsoon (MON) and post-monsoon (POM), also were studied, utilizing multivariate techniques. The organic pollution factor played a significant role on lake water quality during PRM. The influence of organic pollution tends to decrease during MON and POM, a particular situation faced by urban lakes in tropical regions. Polluted stretches in a lake system during different seasons can easily be ascertained by hierarchical cluster analysis. Further, the factors affecting a lake system as a whole, as well as for a particular sampling site, can easily be identified by FA. Improved water quality can be observed during POM. Akkulam and Vela lakes exhibit a wide variation in water quality during all seasons, a finding that corroborates a water flow obstruction from Akkulam Lake to Veli Lake because of the bund existing between the two lakes. The location of the bund is identified as the major reason for different hydrochemical processes in A-V Lake.
Ranked set sampling, Morgenstern type bivariate exponential distribution, Best linear unbiased estimator, Multistage ranked set sampling, Concomitants of order statistics,
Ranked set sampling is applicable whenever ranking of a set of sampling units can be done easily by a judgement method or based on the measurement of an auxiliary variable on the units selected. In this work, we derive different estimators of a parameter associated with the distribution of the study variate Y, based on a ranked-set sample obtained by using an auxiliary variable X correlated with Y for ranking the sample units, when (X, Y) follows a bivariate Pareto distribution. Efficiency comparisons among these estimators are also made. Real-life data have been used to illustrate the application of the results obtained.Ranked set sampling, bivariate Pareto distribution, best linear unbiased estimator,
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.