Abstract
In this article, parametric and nonparametric statistical inference analysis of a set the measurements of air pollution because of PM2.5 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 PM2.5 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 PM2.5 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 PM2.5 concentrations that they experience is higher.
Highlights
P ARTICULATE matter is a composition that is formed by fine particles of different kinds or nature, whose dimensions range from units of nanometers to units of micrometers.Manuscript received July 23, 2019; revised October 15, 2019; accepted January 3, 2020
The classification was performed in blocks to improve the accuracy and precision
As a result of the above-mentioned analysis, it was discovered that all the obtained statistical models were compatible with heavy-tailed distributions [27], [28], which tells us that there are many outliers regarding PM2.5 concentration in the region
Summary
P ARTICULATE matter is a composition that is formed by fine particles of different kinds or nature, whose dimensions range from units of nanometers to units of micrometers. These instruments aim to retrieve in real time the surface-level atmospheric composition, optical coefficients, PM2.5, aerosol optical depth (AOD), and particle size distribution Since these instruments are expensive and can only be deployed over very limited areas, the importance of using advanced modeling methods to provide better air quality predictions was made clear in [2]. The statistical methods used in [17] relied on a five-stage machine-learning approach, based on RF, which according to [17] was intended for: 1) “predicting PM2.5 and PM2.5−10 from PM10”; 2) “imputing missing multiangle implementation of atmospheric correction (MAIAC) AOD data from Copernicus Atmosphere Monitoring Service ()”; 3) “calibrating the spatiotemporal particulate matter concentrations with AOD, meteorology and land-use data”; 4) “predicting particulate matter over all 1 km grid cells of Italy”; 5) “improving particulate matter predictions by using small-scale predictors.”.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IEEE Transactions on Instrumentation and Measurement
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
