Analysis of the Information Obtained From PM_2.5 Concentration Measurements in an Urban Park

Abstract

In this article, parametric and nonparametric statistical inference analysis of a set the measurements of air pollution because of PM 2.5 concentrations was performed. The research work was carried out in an urban park in Quito, Ecuador. Specifically, the park that was chosen to perform the analysis was La Carolina Park. The analysis carried out here was aimed at obtaining the statistical models for parts of this urban park under study and some of its border streets. Furthermore, the park and its border streets were modeled as random variables that were finally classified according to the amount of PM 2.5 concentration levels they carry. This classification was performed by using a method based on both Friedman's test and the categories of the index of air quality of Quito. The results of this article showed that air pollution levels because of PM 2.5 concentrations in La Carolina Park are not in alert level. The worst case, considering the analysis tools used in this article, is that one of the streets that border the park is in caution level. The other streets and parts of the park that were analyzed are either in a desirable level or in an acceptable level. Furthermore, in this article, it has been shown that as pedestrian and temporary residents move further away from the trees and vegetation of the park, the level of exposure to PM 2.5 concentrations that they experience is higher.

Twelve-Year Analysis of NO2 Concentration Measurements at Belisario Station (Quito, Ecuador) Using Statistical Inference Techniques

“…Nonparametric statistical inference tools have also been used in the analysis of air pollution variables [ 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 ]. This has been important, because if the data do not follow parametric distributions, then classical statistical inference is unfounded, which is the case of study in this research.…”

“…Here, the Wilcoxon rank sum test [ 11 , 12 ] was used to test whether the observations come from distributions with equal medians. This test was also satisfactorily used in [ 4 , 20 , 21 , 25 , 26 ]. The null hypothesis was , and the alternative hypothesis was .…”

“…Finally, it is important to mention that nonparametric statistical methods have also been used to estimate the behavior of air pollution variables [ 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 ]. However, the results obtained using nonparametric methods are inferior to those obtained using robust methods, in the sense that robust methods are practically immune to the influence of extreme values.…”

Robust Estimation of Carbon Monoxide Measurements

“…In this paper, in order to study whether the distributions of the variables were the same or not, the Wilcoxon rank sum test [ 20 , 21 ] was used to test whether the data collected in the variables under study comes from distributions with equal medians, as was also done in [ 22 , 23 , 31 ].…”

“…With the purpose of conducting the above-mentioned study, nonparametric tests such as the Kruskal-Wallis test and the Mann-Whitney U test [ 20 , 21 ] were used [ 19 ]. Other examples of recent publications in which nonparametric tests have been used to analyze measurements of air pollution variables can be found in [ 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 ].…”