Outlier Detection in Urban Air Quality Sensor Networks

V.M. van Zoest (Corresponding Author), A. Stein, G. Hoek

Research output: Contribution to journalArticleAcademicpeer-review

2 Citations (Scopus)
27 Downloads (Pure)

Abstract

Low-cost urban air quality sensor networks are increasingly used to study the spatio-temporal variability in air pollutant concentrations. Recently installed low-cost urban sensors, however, are more prone to result in erroneous data than conventional monitors, e.g., leading to outliers. Commonly applied outlier detection methods are unsuitable for air pollutant measurements that have large spatial and temporal variations as occur in urban areas. We present a novel outlier detection method based upon a spatio-temporal classification, focusing on hourly NO2 concentrations. We divide a full year’s observations into 16 spatio-temporal classes, reflecting urban background vs. urban traffic stations, weekdays vs. weekends, and four periods per day. For each spatio-temporal class, we detect outliers using the mean and standard deviation of the normal distribution underlying the truncated normal distribution of the NO2 observations. Applying this method to a low-cost air quality sensor network in the city of Eindhoven, the Netherlands, we found 0.1–0.5% of outliers. Outliers could reflect measurement errors or unusual high air pollution events. Additional evaluation using expert knowledge is needed to decide on treatment of the identified outliers. We conclude that our method is able to detect outliers while maintaining the spatio-temporal variability of air pollutant concentrations in urban areas.
Original languageEnglish
Article number111
Pages (from-to)1-13
Number of pages13
JournalWater, air and soil pollution
Volume229
Issue number4
Early online date8 Mar 2018
DOIs
Publication statusPublished - 1 Apr 2018

Fingerprint

Air Pollutants
outlier
Air quality
Sensor networks
air quality
Normal distribution
sensor
Air
Costs
Measurement errors
Air pollution
detection method
urban area
Sensors
cost
detection
temporal variation
atmospheric pollution
spatial variation

Keywords

  • Air quality
  • Air pollution
  • Outlier Detection
  • NO2
  • Sensor Network
  • ITC-ISI-JOURNAL-ARTICLE
  • ITC-HYBRID

Cite this

@article{7a3284131d7b43b3bb406f644552615a,
title = "Outlier Detection in Urban Air Quality Sensor Networks",
abstract = "Low-cost urban air quality sensor networks are increasingly used to study the spatio-temporal variability in air pollutant concentrations. Recently installed low-cost urban sensors, however, are more prone to result in erroneous data than conventional monitors, e.g., leading to outliers. Commonly applied outlier detection methods are unsuitable for air pollutant measurements that have large spatial and temporal variations as occur in urban areas. We present a novel outlier detection method based upon a spatio-temporal classification, focusing on hourly NO2 concentrations. We divide a full year’s observations into 16 spatio-temporal classes, reflecting urban background vs. urban traffic stations, weekdays vs. weekends, and four periods per day. For each spatio-temporal class, we detect outliers using the mean and standard deviation of the normal distribution underlying the truncated normal distribution of the NO2 observations. Applying this method to a low-cost air quality sensor network in the city of Eindhoven, the Netherlands, we found 0.1–0.5{\%} of outliers. Outliers could reflect measurement errors or unusual high air pollution events. Additional evaluation using expert knowledge is needed to decide on treatment of the identified outliers. We conclude that our method is able to detect outliers while maintaining the spatio-temporal variability of air pollutant concentrations in urban areas.",
keywords = "Air quality, Air pollution, Outlier Detection, NO2, Sensor Network, ITC-ISI-JOURNAL-ARTICLE, ITC-HYBRID",
author = "{van Zoest}, V.M. and A. Stein and G. Hoek",
note = "Springer Deal",
year = "2018",
month = "4",
day = "1",
doi = "10.1007/s11270-018-3756-7",
language = "English",
volume = "229",
pages = "1--13",
journal = "Water, air and soil pollution",
issn = "0049-6979",
publisher = "Springer",
number = "4",

}

Outlier Detection in Urban Air Quality Sensor Networks. / van Zoest, V.M. (Corresponding Author); Stein, A.; Hoek, G.

In: Water, air and soil pollution, Vol. 229, No. 4, 111, 01.04.2018, p. 1-13.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Outlier Detection in Urban Air Quality Sensor Networks

AU - van Zoest, V.M.

AU - Stein, A.

AU - Hoek, G.

N1 - Springer Deal

PY - 2018/4/1

Y1 - 2018/4/1

N2 - Low-cost urban air quality sensor networks are increasingly used to study the spatio-temporal variability in air pollutant concentrations. Recently installed low-cost urban sensors, however, are more prone to result in erroneous data than conventional monitors, e.g., leading to outliers. Commonly applied outlier detection methods are unsuitable for air pollutant measurements that have large spatial and temporal variations as occur in urban areas. We present a novel outlier detection method based upon a spatio-temporal classification, focusing on hourly NO2 concentrations. We divide a full year’s observations into 16 spatio-temporal classes, reflecting urban background vs. urban traffic stations, weekdays vs. weekends, and four periods per day. For each spatio-temporal class, we detect outliers using the mean and standard deviation of the normal distribution underlying the truncated normal distribution of the NO2 observations. Applying this method to a low-cost air quality sensor network in the city of Eindhoven, the Netherlands, we found 0.1–0.5% of outliers. Outliers could reflect measurement errors or unusual high air pollution events. Additional evaluation using expert knowledge is needed to decide on treatment of the identified outliers. We conclude that our method is able to detect outliers while maintaining the spatio-temporal variability of air pollutant concentrations in urban areas.

AB - Low-cost urban air quality sensor networks are increasingly used to study the spatio-temporal variability in air pollutant concentrations. Recently installed low-cost urban sensors, however, are more prone to result in erroneous data than conventional monitors, e.g., leading to outliers. Commonly applied outlier detection methods are unsuitable for air pollutant measurements that have large spatial and temporal variations as occur in urban areas. We present a novel outlier detection method based upon a spatio-temporal classification, focusing on hourly NO2 concentrations. We divide a full year’s observations into 16 spatio-temporal classes, reflecting urban background vs. urban traffic stations, weekdays vs. weekends, and four periods per day. For each spatio-temporal class, we detect outliers using the mean and standard deviation of the normal distribution underlying the truncated normal distribution of the NO2 observations. Applying this method to a low-cost air quality sensor network in the city of Eindhoven, the Netherlands, we found 0.1–0.5% of outliers. Outliers could reflect measurement errors or unusual high air pollution events. Additional evaluation using expert knowledge is needed to decide on treatment of the identified outliers. We conclude that our method is able to detect outliers while maintaining the spatio-temporal variability of air pollutant concentrations in urban areas.

KW - Air quality

KW - Air pollution

KW - Outlier Detection

KW - NO2

KW - Sensor Network

KW - ITC-ISI-JOURNAL-ARTICLE

KW - ITC-HYBRID

UR - https://ezproxy2.utwente.nl/login?url=https://webapps.itc.utwente.nl/library/2018/isi/zoest_out.pdf

U2 - 10.1007/s11270-018-3756-7

DO - 10.1007/s11270-018-3756-7

M3 - Article

VL - 229

SP - 1

EP - 13

JO - Water, air and soil pollution

JF - Water, air and soil pollution

SN - 0049-6979

IS - 4

M1 - 111

ER -