Probability analysis of water quality by turbidity

Abstract

The distribution of turbidity values in given sample is analyzed. The results demonstrated that the nature of the distribution of turbidity values during the year largely depends on the seasonal factor, hence the analysis of the distribution of turbidity is performed separately for each month. Order statistic (variation series) is computed and an empirical distribution function of turbidity values is derived. It is concluded, that the distribution of turbidity in given water sample differs from normal, log-normal and gamma distributions. However, it can be described with sufficient accuracy by a cubic polynomial function. The turbidity distribution hypothesis is tested by the Kolmogorov-Smirnov test. The water turbidity distribution function predicts the probability of exceeding the specified values of turbidity and enables numerical assessment of its likelihood.