Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Contents lists available at ScienceDirect
Environment International
journal homepage: www.elsevier.com/locate/envint
Mapping urban air quality using mobile sampling with low-cost sensors and
machine learning in Seoul, South Korea
Chris C. Lim
a,⁎
, Ho Kim
b
, M.J. Ruzmyn Vilcassim
a
, George D. Thurston
a
, Terry Gordon
a
,
Lung-Chi Chen
a
, Kiyoung Lee
b
, Michael Heimbinder
c
, Sun-Young Kim
d
a
Department of Environmental Medicine, New York School of Medicine, New York, NY, United States of America
b
Graduate School of Public Health, Seoul National University, Seoul, South Korea
c
HabitatMap, Brooklyn, NY, United States of America
d
Graduate School of Cancer Science and Policy, National Cancer Center, Gyeonggi, South Korea
ARTICLE INFO
Handling Editor: Xavier Querol
ABSTRACT
Recent studies have demonstrated that mobile sampling can improve the spatial granularity of land use re-
gression (LUR) models. Mobile sampling campaigns deploying low-cost (< $300) air quality sensors could po-
tentially offer an inexpensive and practical approach to measure and model air pollution concentration levels. In
this study, we developed LUR models for street-level fine particulate matter (PM
2.5
) concentration levels in
Seoul, South Korea. 169 h of data were collected from an approximately three week long campaign across five
routes by ten volunteers sharing seven AirBeams, a low-cost ($250 per unit), smartphone-based particle counter,
while geospatial data were extracted from OpenStreetMap, an open-source and crowd-generated geographical
dataset. We applied and compared three statistical approaches in constructing the LUR models –linear re-
gression (LR), random forest (RF), and stacked ensemble (SE) combining multiple machine learning algorithms –
which resulted in cross-validation R
2
values of 0.63, 0.73, and 0.80, respectively, and identification of several
pollution ‘hotspots.’The high R
2
values suggest that study designs employing mobile sampling in conjunction
with multiple low-cost air quality monitors could be applied to characterize urban street-level air quality with
high spatial resolution, and that machine learning models could further improve model performance. Given this
study design's cost-effectiveness and ease of implementation, similar approaches may be especially suitable for
citizen science and community-based endeavors, or in regions bereft of air quality data and preexisting air
monitoring networks, such as developing countries.
1. Introduction
Ambient air pollution is a major global public health concern, with
the World Health Organization estimating that 4.2 million premature
deaths annually are attributable to fine particulate matter (PM
2.5
) ex-
posure (WHO, 2018). Government and regulatory agencies throughout
the world have traditionally relied on networks of fixed-site monitors in
order to measure air quality and establish standards. Owing to their
prohibitive equipment and operational costs, these monitors tend to be
sparsely located even in large metropolitan cities, or may be entirely
missing in many locales. However, as concentrations of air pollutants
can vary markedly over small distances and short time periods, the
urban environment cannot be fully characterized using information
from sparse, static networks of air pollution monitors (Kumar et al.,
2015). To empirically model and characterize the spatial or spatio-
temporal variability of PM
2.5
concentrations, land use regression (LUR)
models based on data from monitoring networks have been employed.
Recently, LUR models based on data collected from mobile sampling
designs –where predetermined locations or routes are repeatedly
sampled on modes of transport –have gained traction, offering im-
proved spatial resolution at a lower cost (e.g., Hankey and Marshall,
2015;Shi et al., 2016;Deville Cavellin et al., 2016).
Recent technological advancements and proliferation of air quality
sensors offer additional avenues to refine the spatiotemporal char-
acterization of air pollution levels (Morawska et al., 2018. Numerous
instruments from commercial entities, non-profits, and startups have
entered the market to date (Borghi et al., 2017;McKercher et al., 2017),
although the performance of these sensors can differ substantially
https://doi.org/10.1016/j.envint.2019.105022
Received 1 March 2019; Received in revised form 26 June 2019; Accepted 15 July 2019
⁎
Corresponding author at: Department of Environmental Medicine, New York University School of Medicine, 341 East 25th Street, New York, NY 10010, United
States of America.
E-mail address: ccl414@nyu.edu (C.C. Lim).
Environment International 131 (2019) 105022
Available online 27 July 2019
0160-4120/ © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
T
between the different models as well as between individual units, as
noted by evaluations in field and laboratory settings (Jiao et al., 2016;
Jerrett et al., 2017;Castell et al., 2017;Kelly et al., 2017;Feinberg
et al., 2018;Levy Zamora et al., 2019). Offering the capability to in-
expensively generate a large volume of data, distributed networks of
low-cost air quality sensors are beginning to be established to augment
existing monitoring networks or provide novel real-time data streams
(Gao et al., 2015;Schneider et al., 2017;Zikova et al., 2017). Note-
worthy examples of collaborative endeavors between government
agencies, research organizations, and communities include: ‘Open-
Sense’in Geneva, Switzerland (Hasenfratz et al., 2015), ‘Array of
Things’in Chicago, U.S (Catlett et al., 2017), and the Imperial County
Community Air Monitoring Network (English et al., 2017) in California,
U.S.
LUR models based on data collected from mobile sampling with
low-cost (< $300) consumer-based sensors are very limited thus far,
which could potentially offer a highly cost-effective approach to model
and map air pollution concentration levels. The main aim of this study
was to deploy multiple units of the smartphone-based particle counter
‘AirBeam’to measure and model street-level urban air quality in Seoul,
South Korea, a location with limited fixed regulatory monitoring sites
relative to the high population and diverse urban environments. The
individual AirBeam units were first collocated with a pDR-1500 within
a laboratory setting to adjust for intra-instrument variability and equate
particle counts to mass equivalents, and a mobile sampling campaign
was conducted by repeatedly walking across five routes during an ap-
proximately three-week period. The collected air pollution data, to-
gether with an openly available and crowd-sourced geographical data
source OpenStreetMap (OSM), were then used to construct LUR models
with both linear regression and machine learning methods. This work
explores the potential of mobile sampling with low-cost air quality
sensors, machine learning models, and ‘open data’sources to char-
acterize street-level air quality in urban locations with fine spatial re-
solution.
2. Materials and methods
2.1. Equipment description and intra-instrument variability adjustment
The internal optical particle sensor of the AirBeam (dimensions:
105 × 95 × 43.5 mm; weight: 198 g) is the PPD60PV-T2 (detectable
particle range: 0 to 400 μg/m
3
; detective particle size 0.5–2.5 μm) from
Shinyei Technology Co. LTD. (Kyoto, Japan), connected to an Android
OX smartphone running the AirCasting application (aircasting.org).
Supplemental Fig. 1 depicts the AirBeam, its specifications, and the
Android AirCasting app. This mobile system is capable of continuous
measurement (programmable intervals as little as per 1 s) and mapping
(by GPS and Google Maps). The platform code is open-source, and
collected data can be shared and mapped via an online platform,
‘Aircasting’(www.aircasting.org/map).
To adjust for potential intra-instrument variability and to convert
particle counts to PM
2.5
mass equivalents, the AirBeam units were
collocated with a DataRAM pDR-1500 (Thermo Scientific, Franklin,
MA) within a concentrated air particle (CAP) system in Sterling Forest,
New York. The system draws in and concentrates ambient air through a
cyclone inlet that first removes most of the particles larger than 2.5 μm
in aerodynamic diameter. The cyclone outflow is passed over the warm
bath of water and is then rapidly cooled in the condenser, resulting in
supersaturation and particle growth (Maciejczyk et al., 2005). The pDR-
1500 was initially calibrated with ambient particles via its internal
a) b)
c)
Fig. 1. (a) Locations of the five sampling routes in Seoul and government-run, fixed-site monitors (blue markers). Mean PM
2.5
concentration levels (μg/m
3
) during
the sampling period at each of the 100 m segments are also depicted. The red arrows point to an underground roadway, which was not included in analyses. We also
present close-up views of route C as an example to depict sampled data points, with (b) OpenStreetMap and (c) satellite backgrounds. (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of this article.)
C.C. Lim, et al. Environment International 131 (2019) 105022
2
gravimetric filter and pump system at a flow rate of 1.5 L/min in the
CAP chamber. The individual AirBeam units were then calibrated with
the pDR-1500; first, the individual AirBeam units were placed within
the CAPS chamber together with the pDR-1500 and tested for ap-
proximately 3 to 4 h periods per day and between 2 and 3 days per unit,
and separate linear regression models were then fit for each unit.
2.2. Sampling location and protocol
Seoul, the capital of South Korea and the 5th most populous me-
tropolitan area in the world, experiences one of the highest air pollution
concentration levels among cities in developed countries. The city is
characterized by extremely high urban density, abundance of high-rise
buildings and apartments, and a mountainous terrain. This study was
carried out in the southern part of Seoul, south of the Han River, in
three districts: Dongjak-gu (area = 16.35 km
2
; population den-
sity = 24,000/km
2
), Seocho-gu (area = 47.14 km
2
; density = 8300/
km
2
), and Gwanak-gu (area = 29.57 km
2
; density = 18,000/km
2
). The
sampling campaign was conducted during an approximately three-week
period (July 23rd to August 11th) in the summer of 2015, on weekdays
only (12 days total, on non-rainy days) during three different time
periods: morning (8–10 am), evening (6–8 pm), and night (9–11 pm).
Ten volunteers sharing 7 AirBeam units were instructed to repeatedly
sample the five routes without predetermined beginning/ending loca-
tions and times.
The five routes (Fig. 1), four of which were based near or around
government-run regulatory monitors, were designed to span various
neighborhoods and to obtain spatial coverage of a wide range of types
of geographical variables, such as major roads and highways, green
spaces, and both low and high density residential areas. Route A is
located in Sillim; the neighborhood is largely residential with low-rise
buildings and houses. Route B is in Sadang, which is also mainly re-
sidential with a large park and three major roads that surround the
neighborhood. Route C is in Seocho, where the central bus transport
terminal for Seoul is located, as well as the main city highway, a riv-
erside park, and high-rise apartment buildings. Route D is located at
Isu, where major highways and high-density residential areas are pre-
sent. Route E is located near Seoul National University, a large uni-
versity campus located at the base of a mountain; the area is hilly and
tree-covered, and has a relatively low volume of traffic, mainly con-
sisting of buses used for student transport. The lengths of the routes
ranged from 3.9 km to 4.9 km, and the total sum length of all the routes
was 21.5 km.
2.3. Data source for land use predictors
Geospatial data for the city of Seoul, South Korea were downloaded
from OpenStreetMap (OSM), a freely available, crowd-sourced and
user-generated online mapping system. The dataset included > 60
variables, grouped by the following categories: roads (cycleway,
footway, living, path, pedestrian, residential, primary, secondary, road,
secondary link, service, steps, subway, tertiary, trunk, trunk link, un-
classified); land use (cemetery, farm, footway, forest, garden, golf,
grass, hospital, island, park, parking, pitch, place of worship, play-
ground, residential, school, sports center, substation, university, wood);
buildings (apartments, cathedral, church, commercial, hospital, hotel,
house, public, residential, retail, school, university, identified/uni-
dentified), public amenities (fire station, fuel station, hospital, library,
police, school, town hall); transportation points (bus stop, motorway
junction, station, subway entrance); and water areas and waterways
(stream, river, riverbank, water). Several variables in different cate-
gories that repeatedly describe the same land use morphology –e.g.
“university”, which is counted as land use, buildings, and public ame-
nities –were all initially included in the analysis. After removing the
subway variable (as it describes underground paths), there were 67
predictor variables available for analysis (Supplemental Table 1).
2.4. Data reduction
As the frequency of data collection was in 1-second intervals, the
data points were first aggregated into 1-min averages to match the pDR-
1500 sampling frequency and to reduce data noise. Measurement points
with obvious GPS (e.g. located in middle of rivers) and sampling errors
(e.g. volunteer did not follow sampling route properly) were removed
by restricting data points to < 50 M away from the routes and also by
manually after visual inspection. We then employed a “snapping”
procedure to assign the collected data points to the nearest route seg-
ment on the basis of measured GPS coordinates to allow measurements
along the same segment to be analyzed as a group, as per previous
mobile LUR studies (Hankey and Marshall, 2015). Segments were first
defined by length from a starting point along a route, and buffers with
different radiuses were drawn around centroids of the route segments,
with geospatial data from OSM within the buffers then extracted. Each
road segment was thereby associated with land use, built, and natural
environment variables, calculated as different OSM variables within the
buffers of different sizes. We calculated road segments at 5 different
lengths (25 M, 50 M, 100 M, 150 M, 250 M) and 5 buffer radiuses (50 M,
100 M, 150 M, 350 M, 500 M) in order to build the LUR models as well
as to assess how these parameters influence the LUR model perfor-
mance.
2.5. Adjustment for background temporal trends
Previous mobile sampling investigations adjusted for potential
temporal bias through several approaches; for example, Tessum et al.
(2018) adjusted for between-day temporal trends by subtracting the
daily fifth percentile from all measured concentration values on a given
day. Deville Cavellin et al. (2016) used linear and quadratic terms for
temperature as independent variables in the model as adjustment for
potential temporal variability. We modified an approach applied by
multiple studies (Larson et al., 2009;Dons et al., 2012;Clougherty
et al., 2013;Van den Bossche et al., 2015;Apte et al., 2017) that used
background concentration levels from a nearby regulatory monitor to
adjust for temporal trends and normalize measured values. Leveraging
the available information on background PM
2.5
concentrations from
multiple fixed-site regulatory monitors nearby the sampling routes, we
adjusted each 1-min averaged measurements from AirBeams for each
day by applying a multiplicative hourly factor (defined as the ratio of
mean concentration level during the entire sampling period to corre-
sponding hour in which that measurement is taken) derived from the
nearby regulatory monitor. For route E, which was not designed around
a regulatory monitor, we used averaged values from the two nearby
monitors (approx. 2–4 km away) located by routes A and B. This re-
sulted in 6 factors per each sampling day for each of the 5 routes. Using
multiple nearby monitors, instead of a single monitor as done in past
studies, allowed for variable temporal adjustments across several lo-
cations. This approach minimizes the effect of day-to-day variations in
background air quality on the measurements, thereby decreasing the
amount of required sampling data (Van den Bossche et al., 2015).
Hourly measurements from regulatory monitors in Seoul revealed
considerable temporal variability during the study period, with hourly
PM
2.5
levels as low as 5 μg/m
3
and reaching 67 μg/m
3
during pollution
episodes (Fig. 2).
2.6. LUR model building
We first tested the potential effects of spatial aggregation by dif-
ferent route segment lengths and buffer sizes in the linear regression
model by including all available 67 variables into a linear regression
model, and we selected 100 m route segments to spatially aggregate the
collected data points based on the high adj-R
2
, resulting in 215 avail-
able segments for subsequent analyses. We then applied and compared
three statistical approaches for building the LUR model: linear
C.C. Lim, et al. Environment International 131 (2019) 105022
3
regression (LR), random forest (RF), and stacked ensemble (SE).
In the linear regression model, the GIS variables were retained for
multivariable models based on a distance-decay regression selection
strategy (ADDRESS) to screen and select informative candidate vari-
ables and corresponding buffer size from all of the available potential
variables (Su et al., 2015). We then applied a supervised forward search
approach, adding the variables one at a time in the LR model and
keeping the variable only if it increased the R
2
of the model by 1.0%
and if all predictor variables have statistically significant coefficients
(p< 0.05) (Van den Bossche et al., 2018). We also applied the random
forest (RF) model after first removing highly correlated variables (ab-
solute correlation > 0.8). Random forests, in brief, are an ensemble of
decision trees and each tree is constructed using the best split for each
node among a subset of predictors randomly chosen. Random search,
which randomly chooses combination of hyperparameters at every
iteration, was used to tune and optimize the model (Bergstra and
Bengio, 2012). Finally, we employed the stacked ensemble (SE) model,
a machine learning ensemble approach that involves training a learning
algorithm to combine the predictions of several other learning algo-
rithms; first, all of the other algorithms are trained using the available
data, then a ‘meta-classifier’algorithm (chosen from the list of algo-
rithms) is trained to make a final prediction combine all the predictions
of the other algorithms as additional inputs. We evaluated and selected
a diverse group of machine learning algorithms, including random
forest (‘rf’), Bayesian generalized linear model (‘bayesglm’), k-nearest
neighbors (‘knn’), recursive partitioning and regression trees (‘rpart’),
and partitioning using deletion, substitution, and addition moves
(‘partDSA’).
We applied 10-fold cross validation (with 500 repeats) to calculate
mean CV-R
2
(cross-validation R
2
; 1-(mean square error/variance)) and
root mean square errors (RMSE; a measure of the differences between
values predicted by a model and the values observed) for the three
methods to quantify their accuracy. We used packages ‘ggplot2’and
‘leaflet’for visualization and ‘caret’for statistical analyses in R (version
3.4.4).
3. Results
3.1. Adjustment for intra-instrument variability
We fit univariate linear regression models for each of the deployed
Airbeam unit in order to adjust for intra-unit variability and to convert
particle counts to PM
2.5
mass concentrations. During the collocated
sessions with the DataRAM pDR-1500 in the CAP chamber, the PM
2.5
concentration (as measured by pDR-1500) ranged from 0 to 81 μg/m
3
.
The AirBeams revealed strong agreements with the pDR-1500 (adj-
R
2
= 0.95–0.98) and noticeable differences in responses between the
individual units (Fig. 3). The regression models' intercepts, slopes, and
RMSE values varied across the units; detailed statistical summaries of
the models are presented in Table 1.
3.2. Mobile sampling summary statistics
The mobile sampling campaign yielded a total of 10,871 min of
data, of which after removing GPS and sampling errors, 10,177 min
(93.6%) of data remained, equaling > 169 h of total data across the 5
sampled routes (Table 2, Supplemental Tables 2 & 3). 1992 min (33.2 h)
of sampling data were collected at Route A; 2449 min (40.8h) at Route
B; 2313 min (38.6 h) at Route C; 1970 min (32.8 h) at Route D; and
1453 min (24.2 h) at Route E. Route D, which is located near major
roads and highways, had the highest concentration levels
(55.5 ± 27.7 μg/m
3
), while Route B (42.0 ± 24.2 μg/m
3
) and Route E
Fig. 2. Hourly (at 8 am, 9 am, 6 pm, 7 pm, 9 pm, 10 pm) PM
2.5
concentration levels during the sampling period (7/23/15 to 8/10/15) at the four regulatory
background monitors.
C.C. Lim, et al. Environment International 131 (2019) 105022
4
(48.4 ± 31.3 μg/m
3
) had the lowest concentration levels. Notable
differences between morning, evening, and night were also observed
across the five routes, especially for Route D, which had elevated levels
during morning (70.7 ± 25.5 μg/m
3
) compared to evening
(46.6 ± 28.3 μg/m
3
) and night (54.8 ± 24.1 μg/m
3
). The amount of
sampling data varied across the 215 segments, with a median of 44 min
per segment (minimum = 5; 25% percentile = 34; 75% percen-
tile = 55; maximum = 179). Summary statistics for minutes of sam-
pling per 100 m segment for each of the five routes are visualized as
boxplots in Fig. 4.
3.3. Model results
The LUR models were sensitive to different segment lengths and
buffer radiuses, with adj-R
2
generally increasing with larger buffer ra-
diuses (Fig. 5), while 100 m to 150 m segments for spatial aggregation
performed the best. Fitting individual equations to account for intra-
instrument variability for each AirBeam unit generally improved the
accuracy of the constructed LUR models, with an increase in CV-R
2
values by ~0.10–0.15.
In constructing the LR model, we screened and removed several
point variables (e.g. fire stations) that were not frequently present
across the sampling space but clustered near the pollution hotspots, as
these variables ended up having very strong influences on the models.
The final LR LUR model showed high goodness-of-fit with a CV-R
2
of
0.63 and RMSE of 7.01, and the following variables were included in
the model: wood, secondary link, residential road, cathedral, station,
pitch, and apartments (Table 3). The machine learning approaches
explained a greater proportion of the variance of PM
2.5
concentrations
than the LR model. The random forest model identified mostly different
variables as important (wood, residential road, living street, school,
park, apartments, residential, building, tertiary, and service) and also
revealed better performance metrics compared to the LR model, with
higher mean CV-R
2
(0.73) and lower RMSE (6.20). The stacked en-
semble model with random forest as the meta-predictor algorithm
performed the best, and the SE model outperformed both LR and RF
models, with higher CV-R
2
(0.80) and lower RMSE (5.22). Individual R
2
values for the algorithms in the ensemble were 0.74 for random forest,
0.45 for partDSA, 0.50 for rpart, 0.70 for bayesglm, and 0.69 for knn.
Adjusting for background temporal trends changed the overall
morning average concentration levels from 49.4 to 59.2 μg/m
3
; evening
from 46.4 to 45.7 μg/m
3
; and night from 51.5 to 47.3 μg/m
3
. The
changes in concentration levels after temporal adjustment during the
three sampling periods differed significantly across the routes
(Supplemental Table 4). This adjustment also improved the CV-R
2
for
the three approaches, as not doing so resulted in lower CV-R
2
values of
Fig. 3. DataRam pDR-1500 (mass; μg/m
3
) vs. 1-minute averaged AirBeam (hundreds of particles per cubic feet; hppcf) measurements in the concentrated air particle
chamber (CAP).
Table 1
Linear regression equations to convert particle counts to mass for each of the AirBeam unit.
Unit name Intercepts (standard error) Slope (standard error) RMSE Adj-R
2
99B −10.72 (0.29) 0.002616 (1.77 × 10–5) 23.48 0.95
B7E −11.69 (0.44) 0.001974 (1.41 × 10–5) 14.84 0.98
B99 −13.16 (0.42) 0.002102 (1.47 × 10–5) 15.50 0.98
C54 −6.68 (0.18) 0.001905 (1.35 × 10–5) 8.92 0.96
C58 −4.91 (0.16) 0.002000 (1.05 × 10–5) 9.16 0.98
C72 −9.94 (0.34) 0.002537 (3.17 × 10–5) 10.50 0.95
D46 −11.26 (0.37) 0.002049 (1.64 × 10–5) 11.39 0.98
C.C. Lim, et al. Environment International 131 (2019) 105022
5
0.54, 0.65, and 0.71 for the LR, RF, and SE models, respectively. The
constructed LUR models were used to create prediction maps of street-
level PM
2.5
concentration levels in Seoul nearby the sampled locations,
which revealed several ‘hotspots’with elevated PM
2.5
levels (Fig. 6).
The prediction maps revealed similar spatial patterns between the three
modeling approaches with emphasis on similar locations as hotspots,
especially at locations with major roads/highways and high population
density. Conversely, the lowest concentrations were predicted at
greenspace locations, such as parks and mountains. The three ap-
proaches resulted in relatively similar mean predicted values across the
exposure surface, at 47.31, 48.86, and 49.43 μg/m
3
, for LR, RF, and SE,
respectively. However, the LR prediction map predicted lower values
than machine learning approaches at the extremes (range:
26.36–68.96 μg/m
3
), while maps for RF (34.97–71.43 μg/m
3
) and
especially SE (33.50–83.19 μg/m
3
) models resulted in higher predicted
values.
4. Discussion
In this study, we conducted a mobile sampling campaign in Seoul,
South Korea deploying low-cost smartphone-based air quality sensors
and utilized the collected data to construct LUR models employing
three statistical approaches. The strengths of the resulting R
2
values
were comparable to recent, similar studies across multiple locations
around the world that utilized more advanced equipment. Our study is
unique for developing LUR models using multiple low-cost (< $300),
mobile sensors; priced at $250 per unit, AirBeams are order(s) of
magnitude less expensive than the commercially available portable (in
the thousands; the pDR-1500 used in this study cost ~$5700) and
federal standard (in the tens of thousands) instruments. AirBeam and its
operating platform, Aircasting, is also notable for being primarily de-
veloped for citizen science whereby users can upload their measure-
ments to share with the public, as well as for being open-sourced, al-
lowing developers and researchers to program and customize the
instruments and the smartphone app according to their needs and re-
quirements. Many similarly priced ($200–$300) sensors have entered
the market since the present study was conducted, underlining the
public's increasing interest in the capability to measure personalized
real-time exposure data (Caplin et al., 2019). Through deployment of
such low-cost sensors, we were able to characterize the spatial varia-
bility of street-level PM
2.5
in Seoul, the main source of which is likely to
be from traffic given the near-road sampling approach applied in this
study. Past source apportionment studies also identified the primary
source of PM
2.5
in Seoul as motor vehicle emissions and road dust (Heo
et al., 2008;Ryou et al., 2018).
Recent mobile sampling approaches for LUR model building have
employed a variety of study designs and instruments. For example,
Hankey and Marshall (2015) collected over 85 h of data on a bicycle-
based sampling platform in Minneapolis, MN and constructed LUR
models for particle size, black carbon, and PM
2.5
with modest goodness-
of-fit (adj-R
2
of ~0.5 for particle number and ~0.4 for PM
2.5
). Apte
et al. (2017) analyzed data collected from a Google Street View map-
ping vehicle equipped with air quality sensors that repeatedly sampled
every street in a 30-km
2
area of Oakland, CA, to model and reveal urban
air pollution patterns at 4–5 orders of magnitude greater spatial pre-
cision than possible with current central-site ambient monitoring. The
‘OpenSense’project in Zurich, Switzerland (Hasenfratz et al., 2015)
utilized mobile sensor nodes installed on top of public transport tram
vehicles in the city to create high-resolution pollution prediction maps
for ultrafine particles and particle counts. Vehicle-based mobile mea-
surements were also applied to create LUR models to estimate the
spatial variation of street-level PM
2.5
and PM
10
in the downtown area of
Hong Kong (Shi et al., 2016), and integration of urban/building mor-
phology as independent variables increased the adj-R
2
of the LUR
model, suggesting that incorporating detailed 3D characteristics of the
land use can improve the predictive power of such models.
Table 2
Summary statistics for measurements across the five routes.
Route ID Route name AirBeam units
deployed
Total Morning Evening Night
Minutes
sampled
Average (Std. Dev),
μg/m
3
IQR Minutes
sampled
Average (Std. Dev),
μg/m
3
IQR Minutes
sampled
Average (Std. Dev),
μg/m
3
IQR Minutes
sampled
Average (Std. Dev),
μg/m
3
IQR
A Sillim B7E, B99, C54,
D46
1992 51.3 (32.6) 56.3 901 43.3 (32.8) 40.4 462 58.6 (27.9) 49.6 629 57.4 (33.0) 47.4
B Sadang 99B, B7E, C58,
D46
2449 42.0 (24.2) 36.5 744 40.5 (24.8) 34.2 858 38.6 (21.2) 38.1 847 46.7 (25.8) 30.3
C Seocho 99B, C58, C72 2313 49.9 (31.4) 50.3 574 47.9 (35.2) 69.2 892 46.7 (29.4) 43.3 847 54.5 (30.1) 38.1
D Isu 99B, C72, D46 1970 55.5 (27.7) 33.9 477 70.7 (25.5) 39.6 755 46.6 (28.3) 43.3 738 54.8 (24.1) 26.1
E Seoul National
University
B7E, B99, C54,
D46
1453 48.4 (31.3) 48.4 396 56.8 (39.4) 74.3 528 47.4 (26.5) 23.2 529 43.0 (27.4) 54.0
C.C. Lim, et al. Environment International 131 (2019) 105022
6
Our study and sampling design highlight the potential advantages of
mobile sampling with low-cost and portable air quality sensors in
constructing the LUR models. The aforementioned studies were largely
based on sampling campaigns conducted on modes of transport (e.g.
cars) visiting a single location at a given time, which may potentially
result in a low number of visits per location. The results from this and
past studies found that mobile LUR models are highly sensitive to
parameters such as the number of route segments, radiuses of buffers,
and number of measurements per segment (Minet et al., 2017).
Hatzopoulou et al. (2017) evaluated the influence of the number of
sampling locations and durations of sampling on LUR model perfor-
mance, noting that mobile sampling campaigns can be inefficient due to
low sampling frequency at a large number of locations, and that spatial
variability may be more important than the numbers of locations when
designing the sampling routes. The authors also found that the LUR
models became relatively robust after 150–200 segments and 10–12
visits per segment. In the present study, walking at a slow speed, in-
stead of sampling on mechanical modes of transportation, resulted in
each route generally having a high number of data points
(median = 44) per segment. This approach also allows for assessing
personal-level exposure in urban areas where there are a larger number
of people on the streets than in cars. The disadvantage of shorter dis-
tances being covered when sampling on foot was offset by the low cost
and portability of AirBeams, which allowed for several units that could
be deployed simultaneously across multiple locations at a given time
and thereby maximize spatial coverage, as opposed to the majority of
past mobile sampling studies that were carried out on a single platform.
Simultaneous measurements within a structured sampling design could
Fig. 4. Boxplot demonstrating distribution of minutes of sampling per 100 m segment for each sampling route.
Fig. 5. Adjusted R
2
of LR LUR models (including all available 67 predictor variables) for mass, by segment radius and buffer sizes.
C.C. Lim, et al. Environment International 131 (2019) 105022
7
decrease the amount of collected data (and manpower) required to
construct robust models, whereas participatory sensing where sampling
is done ‘opportunistically’could lead to unstructured data that is more
difficult to interpret (Van den Bossche et al., 2016). Furthermore, Air-
Beam's ease of operation meant that minimal training (a few minutes at
most) was required prior to field deployment, resulting in a relatively
large volume of data being generated within the short sampling cam-
paign period during this study.
This study leveraged OpenStreetMap (OSM), an openly available
and crowd-sourced GIS dataset, which provided a rich and compre-
hensive source of geospatial data for a wide range of LUR variables.
OSM and other ‘open data’sources offer underexplored but valuable
information for data-driven methods to predict air pollution levels
(VoPham et al., 2018). Notably, the OSM GIS variables were highly
developed for Seoul and provided detailed and differentiated data for
the numerous types of roads and buildings, which are the land use
categories that usually provide the highest predictive power for air
pollution LUR models. Another advantageous aspect of crowd-sourced
data is that it is continually updated; for example, using an earlier
download of OSM from September 2015 (versus January 2018 in this
analysis) with less developed characterization of Seoul resulted inLUR
models with lower CV-R
2
, suggesting that in locations with lacking
geospatial data, crowd-sourced efforts to generate the relevant GIS
variables could be carried out in concert with the air pollution sampling
campaign to strengthen the predictive capability of LUR models. De-
spite recent endeavors to democratize data by agencies and organiza-
tions throughout the world as part of the ‘open data’movement, many
detailed GIS files remain proprietary and thereby cost-prohibitive, and
freely available data like OSM offer an alternative and important source
of detailed spatial data for researchers and communities.
Machine learning methods offered improved goodness-of-fit com-
pared to traditional stepwise linear regression in constructing the LUR
models. Prior work on machine learning applications in both national
(Hu et al., 2017;Di et al., 2016) and local-level (Adams and
Kanaroglou, 2016;Weichenthal et al., 2016;Brokamp et al., 2017)
predictions of air pollution concentration levels highlight the ad-
vantages associated with the approach, including higher accuracy and
identification of important variables. A recent example further under-
lines additional potential benefits; a study in Los Angeles, USA used a
multi-step and flexible spatial data mining approach using machine
learning to select for most important OSM geographic features and
predict PM
2.5
concentrations, removing the need for a priori selection of
predictors for exposure modeling (Lin et al., 2017). Similarly in our
analysis, applying the traditional step-wise linear regression LUR ap-
proach with the highly correlated OSM dataset, which also contained
several highly influential variables, required manual screening and re-
moval of predictor variables prior to input and during the model
building process. Notably, the stacked ensemble model combining
multiple machine learning algorithms outperformed both LR and RF in
this study. In recent years ensemble machine learning methods have
emerged as an important tool for modeling complex relationships and
have been applied successfully in various research areas (Yang et al.,
2010). Application of ensembles have been generally limited in air
pollution exposure assessment and modeling efforts to date, and the
results here suggest that ensemble-based approaches could further en-
hance the predictive performance of LUR models.
We note several potential weaknesses that are present in this study.
As we evaluated the AirBeam units in a carefully controlled experi-
mental chamber drawing in air from a forested and rural area (Tuxedo,
New York), the particle composition and the environmental conditions
(e.g., humidity and temperature) encountered during the experiment
are likely to be significantly different from the heavily urban location
where this study was carried out. Although the potential impacts of
these factors were not assessed in this study, previous performance
evaluations of AirBeams in various laboratory and field settings offer
insight. The initial manufacturer calibration was conducted in a simi-
larly urban setting (New York City), which revealed high correlations
with both gravimetric sampling and pDR-1500 (takingspace.org, 2014).
Comparison against federal equivalent method monitors showed high
agreements with GRIMM (R
2
~0.6–0.8) (Mukherjee et al., 2017;
SCAQMD, 2017;Feinberg et al., 2018), but mixed results were observed
with BAM (R
2
~0.2–0.7) (Jiao et al., 2016;SCAQMD, 2017). A study of
sensor responses to Arizona road dust, salt, and welding fumes (Sousan
et al., 2017) demonstrated that particle types had significant impacts on
AirBeam (and other low-cost sensors) measurements. Relative humidity
(RH) levels also influenced the measurements; a laboratory evaluation
found that bias was observed when both RH (> 65%) levels and con-
centration levels (> 100 μg/m
3
) were elevated (SCAQMD, 2017), while
another study (Feinberg et al., 2018) found that the particle counts
measurements were affected by higher humidity levels in a field setting.
Highly humid summers in Korea would likely influence the absolute
measurement values, but the potential impact on prediction model
performance is likely to be minimal as the spatial variability of hu-
midity levels is relatively uniform across a city. Nevertheless, these
findings emphasize the need to consider the potential influence of en-
vironmental factors in sensor deployments, and performance evalua-
tions at the study location is suggested for similar studies applying low-
cost sensors. In addition, the particle concentration levels encountered
during sampling in Seoul were higher than the range used for con-
structing calibration equations for the AirBeam units, which may ignore
Table 3
Selected LUR model predictor variables in the LR and RF models and associated statistics.
Variable name Variable type Buffer length Linear regression Random forest
a
ΒStd. error P-value Importance
Intercept 50.02 1.21 < 0.001
Wood Area 500 m −3.80 × 10
−5
4.25 × 10
−6
< 0.001 14.45
Residential Road Line 500 m 2.59 × 10
−5
5.88 × 10
−6
< 0.001 13.10
Secondary Link Line 500 m 6.88 × 10
−3
1.00 × 10
−3
< 0.001
Cathedral Point 500 m −2.47 × 10
−3
1.03 × 10
−3
0.02
Station Point 500 m −3.75 1.02 < 0.001
Pitch Area 350 m −1.88 × 10
−4
4.28 × 10
−5
< 0.001
Apartments Point 500 m 7.70 × 10
−5
4.02 × 10
−5
0.05 10.21
School Area 500 m 10.73
Living Street Line 500 m 10.85
Park Area 500 m 10.31
Residential Area 500 m 9.93
Building (Unclassified) Area 500 m 9.72
Tertiary Line 350 m 9.70
Service Line 350 m 8.97
a
Top ten variables by variable importance are shown in the table.
C.C. Lim, et al. Environment International 131 (2019) 105022
8
the potential nonlinearity of sensor responses. We also did not check for
potential sensor drift –a common issue for low-cost air quality sensors –
during and after the mobile sampling, although this is unlikely due to
the relatively short sampling period. These issues may have contributed
to predicted values that were significantly higher than observed values
from nearby fixed-site monitors, although it is also possible that such
differences are due to the fact that fixed-site monitors are often located
well above ground and tend to underestimate personal exposures when
walking near traffic(Deville Cavellin et al., 2016). Another potential
weakness is that the OSM data quality and density could be potentially
uneven across locations, as some areas could be characterized in more
detail than others. For example, in some of the sampled areas in this
study, several of the houses in residential areas were not captured in the
OSM file and thereby could have influenced model quality; however, as
OSM data coverage and quality continues to improve this should be-
come less of an issue over time.
5. Conclusions
Low-cost sensors represent an opportunity to bridge the data gap,
thereby promoting public discourse, influencing air pollution regula-
tions, and protecting public health (Amegah, 2018). This study high-
lights the advantages and potential of applying data collected from
mobile sampling with multiple low-cost sensors to model and map
street-level air pollution levels in urban locations, especially the cap-
ability to generate a large volume of sampling data with ease. The
predictive power of models developed here, despite deploying only a
limited number of significantly less expensive, consumer-based air
quality sensors, were comparable to the past mobile sampling LUR
studies, especially after adjusting for intra-instrument variability and
temporal trends. To minimize the potential influence of local particle
characteristics and environmental conditions, calibration with collo-
cated reference monitors at the sampling location is suggested for fu-
ture projects using similar low-cost sensors, as well as to convert par-
ticle counts to mass concentration, a unit of measurement that is more
readily transferable for policy-relevant metrics. Initial calibrations
should also carefully evaluate and adjust for the potential effects of
relative humidity levels, which can have significant influences on the
low-cost sensors. Overall, the findings here suggest that similar mobile
sampling designs using low-cost sensors and ‘open data’sources could
be applied to generate a large volume of data and construct LUR models
and maps with fine spatial granularity, and that machine learning
methods could further improve model performance. Our study design
and approach may be especially suitable for citizen science and com-
munity-based endeavors, or in locations without preexisting air mon-
itoring networks, such as developing countries.
Acknowledgement
This study was funded by the Basic Science Research Program
through the National Research Foundation of Korea (NRF) funded by
the South Korea Ministry of Education (2018R1A2B6004608), the NSF
East Asia and Pacific Summer Institute (EAPSI) Fellowship, the Air &
Waste Management Air Pollution Education and Research Grant
(APERG), the EPA STAR Graduate Fellowship, and by a grant from the
National Institutes of Environmental Health Sciences Center (ES00260).
This publication was developed under Assistance Agreement No.
FP917825 awarded by the U.S. Environmental Protection Agency to
Chris C. Lim. It has not been formally reviewed by EPA. The views
expressed in this document are solely those of the authors and do not
necessarily reflect those of the Agency. EPA does not endorse any
products or commercial services mentioned in this publication.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
Fig. 6. PM
2.5
prediction maps nearby sampled areas constructed applying (a)
linear regression, (b) random forest, and (c) stacked ensemble approaches.
C.C. Lim, et al. Environment International 131 (2019) 105022
9
doi.org/10.1016/j.envint.2019.105022.
References
Adams, M.D., Kanaroglou, P.S., 2016. Mapping real-time air pollution health risk for
environmental management: combining mobile and stationary air pollution mon-
itoring with neural network models. J. Environ. Manag. 168, 133–141.
Amegah, A.K., 2018. Proliferation of Low-Cost Sensors. What Prospects for Air Pollution
Epidemiologic Research in Sub-Saharan Africa? vol. 241. pp. 1132–1137.
Apte, J.S., Messier, K.P., Gani, S., Brauer, M., Kirchstetter, T.W., Lunden, M.M., ...
Hamburg, S.P., 2017. High-resolution air pollution mapping with Google street view
cars: exploiting big data. Environmental Science & Technology 6999–7008.
Bergstra, James, Bengio, Yoshua, 2012. Random search for hyper-parameter optimiza-
tion. J. Mach. Learn. Res. 13.Feb, 281–305.
Borghi, F., Spinazz, A., Rovelli, S., Campagnolo, D., Buono, L. Del, Cattaneo, A., &
Cavallo, D. M. (2017). Miniaturized monitors for assessment of exposure to air pol-
lutants: a review.
Brokamp, C., Jandarov, R., Rao, M.B., LeMasters, G., Ryan, P., 2017. Exposure assessment
models for elemental components of particulate matter in an urban environment: a
comparison of regression and random forest approaches. Atmos. Environ. 151, 1–11.
Caplin, A., Ghandehari, M., Lim, C., Glimcher, P., Thurston, G., 2019. Advancing en-
vironmental exposure assessment science to benefit society. Nat. Commun. 10 (1),
1236.
Castell, N., Dauge, F.R., Schneider, P., Vogt, M., Lerner, U., Fishbain, B., Bartonova, A.,
2017. Can commercial low-cost sensor platforms contribute to air quality monitoring
and exposure estimates? Environ. Int. 99, 293–302.
Catlett, C.E., Beckman, P.H., Sankaran, R., Galvin, K.K., 2017. Array of things: a scientific
research instrument in the public way: Platform design and early lessons learned. In:
Proceedings of the 2nd International Workshop on Science of Smart City Operations
and Platforms Engineering. ACM, pp. 26–33 April.
Clougherty, J.E., Kheirbek, I., Eisl, H.M., Ross, Z., Pezeshki, G., Gorczynski, J.E., Johnson,
S., Markowitz, S., Kass, D., Matte, T., 2013. Intra-urban spatial variability in win-
tertime street-level concentrations of multiple combustion- related air pollutants: the
New York City Community Air Survey (NYCCAS). J. Expo. Sci. Environ. Epidemiol.
23, 232e240.
Deville Cavellin, L., Weichenthal, S., Tack, R., Ragettli, M.S., Smargiassi, A., Hatzopoulou,
M., 2016. Investigating the use of portable air pollution sensors to capture the spatial
variability of traffic-related air pollution. Environmental Science & Technology 50
(1), 313–320.
Di, Q., Kloog, I., Koutrakis, P., Lyapustin, A., Wang, Y., Schwartz, J., 2016. Assessing PM
2.5 Exposures with High Spatiotemporal Resolution across the Continental United
States. Environ Sci Technol. 50 (9), 21–4712.
Dons, E., Int Panis, L., Van Poppel, M., Theunis, J., Wets, G., 2012. Personal exposure to
Black Carbon in transport microenvironments. Atmos. Environ. 55, 392–398.
English, P.B., Olmedo, L., Bejarano, E., Lugo, H., Murillo, E., Seto, E., Wong, M., King, G.,
Wilkie, A., Meltzer, D., Carvlin, G., 2017. The Imperial County Community Air
Monitoring Network: a model for community-based environmental monitoring for
public health action. Environ. Health Perspect. 125 (7).
Feinberg, S., et al., 2018. Long-term evaluation of air sensor technology under ambient
conditions in Denver, Colorado. Atmos. Meas. Tech. 11, 4605–4615.
Gao, M., Cao, J., Seto, E., 2015. A distributed network of low-cost continuous reading
sensors to measure spatiotemporal variations of PM
2.5
in Xi'an, China. Environ.
Pollut. 199, 56–65.
Hankey, S., Marshall, JD., 2015. Land Use Regression Models of On-Road Particulate Air
Pollution (Particle Number, Black Carbon, PM 2.5 , Particle Size) Using Mobile
Monitoring. Environ Sci Technol. 49 (15), 202–9194.
Hasenfratz, D., et al., 2015. Deriving high-resolution urban air pollution maps using
mobile sensor nodes. Pervasive Mob. Comput. 16, 268–285.
Hatzopoulou, M., Valois, M.F., Levy, I., Mihele, C., Lu, G., Bagg, S., ... Brook, J., 2017.
Robustness of land-use regression models developed from mobile air pollutant
measurements. Environ. Sci. Technol. 51 (7), 3938–3947.
Heo, J.B., Hopke, P.K., Yi, S.M., 2008. Source apportionment of PM
2.5
in Seoul.
Korea.Atmos. Chem. Phys. Discuss. 8, 20427e2046.
Hu, X., Belle, J. H., Meng, X., Wildani, A., Waller, L. A., & Strickland, M. J. (2017).
Estimating PM
2.5
concentrations in the conterminous United States using the random
forest approach.
Jerrett, M., et al., 2017. Validating novel air pollution sensors to improve exposure es-
timates for epidemiological analyses and citizen science. Environ. Res. 158, 286–294.
Jiao, W., et al., 2016. Community Air Sensor Network (CAIRSENSE) project: evaluation of
low-cost sensor performance in a suburban environment in the southeastern United
States. Atmos. Meas. Tech. Discuss. 1–24.
Kelly, K.E., et al., 2017. Ambient and laboratory evaluation of a low-cost particulate
matter sensor. Environ. Pollut. 221, 491–500. https://doi.org/10.1016/j.envpol.
2016.12.039. Feb.
Kumar, P., Morawska, L., Martani, C., Biskos, G., Neophytou, M., Di Sabatino, S., Bell, M.,
Norford, L., Britter, R., 2015. The rise of low-cost sensing for managing air pollution
in cities. Environ. Int. 75, 199–205.
Larson, T., Henderson, S.B., Brauer, M., 2009. Mobile Monitoring of Particle Light
Absorption Coefficient in an Urban Area as a Basis for Land Use Regression. Environ
Sci Technol. 43 (13), 8–4672.
Levy Zamora, Misti, et al., 2019. Field and laboratory evaluations of the low-cost plan-
tower particulate matter sensor. Environ. Sci. Technol. 53 (2), 838–849 American
Chemical Society.
Lin, Y., Chiang, Y.-Y., Pan, F., Stripelis, D., Ambite, J.L., Eckel, S.P., Habre, R., 2017.
Mining public datasets for modeling intra-city PM
2.5
concentrations at a fine spatial
resolution. In: Proceedings of the 25th ACM SIGSPATIAL International Conference on
Advances in Geographic Information Systems. ACM, Los Angeles area, CA, pp. 1–10.
Maciejczyk, P., Zhong, M., Li, Q., Xiong, J., Nadziejko, C., Chen, L.C., 2005. Effects of
subchronic exposures to concentrated ambient particles (CAPs) in mice: II. The design
of a CAPs exposure system for biometric telemetry monitoring. Inhal. Toxicol. 17
(4–5), 189–197.
McKercher, G.R., Salmond, J.A., Vanos, J.K., 2017. Characteristics and applications of
small, portable gaseous air pollution monitors. Environ. Pollut. https://doi.org/10.
1016/j.envpol.2016.12.045.
Minet, L., Gehr, R., Hatzopoulou, M., 2017. Capturing the sensitivity of land-use re-
gression models to short-term mobile monitoring campaigns using air pollution
micro-sensors. Environ. Pollut. 230, 280–290.
Morawska, L., et al., 2018. Applications of low-cost sensing technologies for air quality
monitoring and exposure assessment: how far have they gone? Environ. Int. 116,
286–299.
Mukherjee, Anondo, Stanton, Levi, Graham, Ashley, Roberts, Paul, August 5, 2017.
Assessing the utility of low-cost particulate matter sensors over a 12-week period in
the Cuyama Valley of California. Sensors 17 (8), 1805.
Ryou, H., Heo, J., Kim, S.Y., 2018. Source apportionment of PM10 and PM
2.5
air pollu-
tion, and possible impacts of study characteristics in South Korea. Environ. Pollut.
240, 963–972.
Schneider, P., Castell, N., Vogt, M., Dauge, F.R., Lahoz, W.A., Bartonova, A., 2017.
Mapping urban air quality in near real-time using observations from low-cost sensors
and model information. Environ. Int. 106 (December 2016), 234–247.
Shi, Y., Lau, K.K.L., Ng, E., 2016. Developing street-level PM
2.5
and PM
10
land use re-
gression models in high-density Hong Kong with urban morphological factors.
Environ. Sci. Technol. 50 (15), 8178–8187.
Sousan, S., Koehler, K., Hallett, L., Peters, T.M., 2017. Evaluation of consumer monitors to
measure particulate matter. J. Aerosol Sci. 107, 123–133.
South Coast AQMD, 2017. AirBeam summary report [online]. Available at. http://www.
aqmd.gov/aq-spec/sensordetail/airbeam, Accessed date: June 2019.
Su, J.G., Hopke, P.K., Tian, Y., Baldwin, N., Thurston, S.W., Evans, K., Rich, D.Q., 2015.
Modeling particulate matter concentrations measured through mobile monitoring in
a deletion/substitution/addition approach. Atmos. Environ. 122, 477–483.
Takingspace.org, 2014. AirBeam Technical Specifications, Operation & Performance:
Taking Space. [online] Available at. http://www.takingspace.org/airbeam-
technical-specifications-operation-performance/, Accessed date: June 2019.
Tessum, M.W., et al., 2018. Mobile and fixed-site measurements to identify spatial dis-
tributions of traffic-related pollution sources in Los Angeles. Environ. Sci. Technol.
https://doi.org/10.1021/acs.est.7b04889. acs.est.7b04889.
Van den Bossche, J., Peters, J., Verwaeren, J., Botteldooren, D., Theunis, J., De Baets, B.,
2015. Mobile monitoring for mapping spatial variation in urban air quality: devel-
opment and validation of a methodology based on an extensive dataset. Atmos.
Environ. 105, 148–161.
Van den Bossche, Joris, Theunis, Jan, Elen, Bart, Peters, Jan, Botteldooren, Dick, De
Baets, Bernard, 2016. Opportunistic mobile air pollution monitoring: a case study
with city wardens in Antwerp. Atmos. Environ. 141, 408–421.
Van den Bossche, J., De Baets, B., Verwaeren, J., Botteldooren, D., Theunis, J., 2018.
Development and evaluation of land use regression models for black carbon based on
bicycle and pedestrian measurements in the urban environment. Environ. Model.
Softw. 99, 58–69.
VoPham, T., Hart, J.E., Laden, F., Chiang, Y.-Y., 2018. Emerging trends in geospatial
artificial intelligence (geoAI): potential applications for environmental epidemiology.
Environmental Health: A Global Access Science Source 17 (1), 1–6.
Weichenthal, S., Ryswyk, K. Van, Goldstein, A., Bagg, S., Shekkarizfard, M., Hatzopoulou,
M., 2016. A land use regression model for ambient ultrafine particles in Montreal,
Canada: a comparison of linear regression and a machine learning approach. Environ.
Res. 146, 65–72.
World Health Organization, May 2, 2018. 9 out of 10 People Worldwide Breathe Polluted
Air, but More Countries Are Taking Action. Available at. https://www.who.int/
news-room/detail/02-05-2018-9-out-of-10-people-worldwide-breathe-polluted-air-
but-more-countries-are-taking-action, Accessed date: 16 September 2018.
Yang, P., Yang, Y.H., Zhou, B.B., Zomaya, A.Y., 2010. A review of ensemble methods in
bioinformatics. Curr. Bioinforma. 5, 296–308.
Zikova, N., et al., 2017. Estimating hourly concentrations of PM
2.5
across a metropolitan
area using low-cost particle monitors. Sensors 17, 1922.
C.C. Lim, et al. Environment International 131 (2019) 105022
10
