Access to this full-text is provided by Copernicus Publications on behalf of European Geosciences Union.
Content available from Atmospheric Measurement Techniques
This content is subject to copyright.
Atmos. Meas. Tech., 13, 6427–6443, 2020
https://doi.org/10.5194/amt-13-6427-2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.
Evaluation of optical particulate matter sensors under realistic
conditions of strong and mild urban pollution
Adnan Masic1, Dzevad Bibic1, Boran Pikula1, Almir Blazevic1, Jasna Huremovic2, and Sabina Zero2
1Faculty of Mechanical Engineering, University of Sarajevo, 71000 Sarajevo, Bosnia-Herzegovina
2Faculty of Science, University of Sarajevo, 71000 Sarajevo, Bosnia-Herzegovina
Correspondence: Adnan Masic (masic@mef.unsa.ba)
Received: 16 June 2020 – Discussion started: 8 July 2020
Revised: 20 October 2020 – Accepted: 21 October 2020 – Published: 30 November 2020
Abstract. In this paper we evaluate characteristics of three
optical particulate matter sensors/sizers (OPS): high-end
spectrometer 11-D (Grimm, Germany), low-cost sensor
OPC-N2 (Alphasense, United Kingdom) and in-house devel-
oped MAQS (Mobile Air Quality System), which is based
on another low-cost sensor – PMS5003 (Plantower, China),
under realistic conditions of strong and mild urban pollu-
tion. Results were compared against a reference gravimet-
ric system, based on a Gemini (Dadolab, Italy), 2.3 m3h−1
air sampler, with two channels (simultaneously measuring
PM2.5and PM10 concentrations). The measurements were
performed in Sarajevo, the capital of Bosnia-Herzegovina,
from December 2019 until May 2020. This interval is di-
vided into period 1 – strong pollution – and period 2 – mild
pollution. The city of Sarajevo is one of the most polluted
cities in Europe in terms of particulate matter: the average
concentration of PM2.5during the period 1 was 83 µg m−3,
with daily average values exceeding 500 µg m−3. During pe-
riod 2, the average concentration of PM2.5was 20 µg m−3.
These conditions represent a good opportunity to test opti-
cal devices against the reference instrument in a wide range
of ambient particulate matter (PM) concentrations. The ef-
fect of an in-house developed diffusion dryer for 11-D is dis-
cussed as well. In order to analyse the mass distribution of
particles, a scanning mobility particle sizer (SMPS), which
together with the 11-D spectrometer gives the full spectrum
from nanoparticles of diameter 10 nm to coarse particles of
diameter 35 µm, was used. All tested devices showed excel-
lent correlation with the reference instrument in period 1,
with R2values between 0.90 and 0.99 for daily average PM
concentrations. However, in period 2, where the range of con-
centrations was much narrower, R2values decreased signif-
icantly, to values from 0.28 to 0.92. We have also included
results of a 13.5-month long-term comparison of our MAQS
sensor with a nearby beta attenuation monitor (BAM) 1020
(Met One Instruments, USA) operated by the United States
Environmental Protection Agency (US EPA), which showed
similar correlation and no observable change in performance
over time.
1 Introduction
Analysis of particulate matter represents a key element for
studies of air pollution. Various studies shed light on their
effect on health (Downward et al., 2018) and climate (Zhao
et al., 2019). In many cases particulate matter is a dominant
pollutant among other components of pollution. Therefore,
developing a strategy for reliable quantification of particulate
matter in ambient air is necessary. The traditional and most
accurate approach to measuring the particulate matter con-
centration in the air is the reference method, based on gravi-
metric measurements, after the collection of particulate mat-
ter by air samplers. The typical time resolution of such mea-
surements is 24 h. Although there are portable air samplers,
these measurements are usually performed at fixed locations,
such as research supersites. Reference systems are expensive
and require a lot of laboratory work. Results are not immedi-
ately available because of the time-consuming process of fil-
ter treatment. Taking that into account, various governmen-
tal institutions usually opt for more affordable and easier-
to-use and -maintain equivalent methods. These are usually
fixed, semi-automatic stations equipped with beta attenuation
monitors (BAMs). The typical time resolution of such sta-
Published by Copernicus Publications on behalf of the European Geosciences Union.
6428 A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions
tions is 1 h. If maintained and calibrated properly, the equiv-
alent methods should achieve an acceptable level of agree-
ment with the reference. For example, one long-term com-
prehensive study (Hafkenscheid and Vonk, 2014) performed
at 14 different locations across the Netherlands showed that
a linear correction y=0.91x−1.6, applied to raw readings
from BAM, was necessary to achieve the requirements of the
Guide to the Demonstration of Equivalence (ECWG, 2010).
Newer methods, based on optical particle sensors (OPS),
are nowadays increasingly more popular, particularly low-
cost variants (Zheng et al., 2019; Mukherjee et al., 2019;
Tanzer et al., 2019; Morawska et al., 2018). Their typical
time resolution is between 1 s and 1min, and because of their
price and size, they can be used in networks to provide better
spatial coverage (Martin et al., 2019; Li et al., 2019). Further-
more, they provide information about multiple mass fractions
of particulate matter simultaneously, unlike the concentration
of a single fraction in a gravimetric system or BAM. How-
ever, there are concerns about their suitability for measuring
mass concentrations of ambient PM, since there is a signifi-
cant measurement uncertainty arising from the principles of
their operation.
Most commercially available OPS use Mie scattering the-
ory (Mie, 1908) to determine the size and number of parti-
cles within the unit volume of air. Mie theory provides the
solution of the Maxwell equations for the scattering of plane
waves on spherical particles. The Mie solution is rather com-
plex, but in order to illustrate the non-linearity of the theory,
it will suffice to consider the case where particles are much
smaller than the wavelength (of light, since a red laser is com-
monly used in practice). In that case the intensity of scattered
radiation is given by
I=I01+cos2θ
2R22π
λ4d
26
m2−1
m2+2
2
,(1)
where I0is the intensity of the incident radiation, θis the
scattering angle, Ris the distance between the particle and
the observing point, λis the wavelength, dis the particle
diameter and mis the refractive index of the particle. Thus, in
order to calculate the diameter of the particle by measuring
the intensity of the scattered radiation, one must assume a
value for the refractive index of the particle. If the particle
absorbs nothing from incoming radiation, its refractive index
will be real; otherwise, it is written in the form
m=n+iκ, (2)
where κis called the extinction coefficient and is related to
the absorption coefficient α:
α=4πκ
λ.(3)
Once the size distribution is calculated across Kchannels
(bins), the total mass concentration of particles will be
cm=
K
X
i=1
wiρiViNi,(4)
where Viis the (average) volume, ρiis the density of the par-
ticles, Niis the number of particles per unit volume and wi
is the weighting factor for channel i. Here we have another
cause of OPS uncertainty: the density of particles must be
assumed. Regarding the weighting factors, sensor manufac-
turers calculate values to correct for certain effects, such as
the fact that OPS cannot detect particles which are too small.
Laboratory tests and calibrations of OPS are performed
under controlled conditions with known particles, such as
polystyrene latex spheres (Walser et al., 2017; Bezantakos
et al., 2018), continuously changing monodisperse particles
(Kuula et al., 2017; Kuula et al., 2020) or multi-modal par-
ticles (Cai et al., 2019). A burning chamber is used in some
investigations as well (Wang et al., 2015). However, Eq. (1)
is strongly non-linear in terms of refractive index, and in
most practical cases corrections for different particles’ op-
tical properties are impossible to implement. Furthermore,
densities appearing in Eq. (4) are not known a priori. That ex-
plains why it is difficult to calibrate OPS for realistic ambient
PM concentration measurements: any laboratory calibration
may or may not be applicable to the changing outdoor con-
ditions (Tryner et al., 2020; Crilley et al., 2020).
For outdoor applications, there is an additional prob-
lem: hygroscopic growth of particles (Jayaratne et al., 2018;
Granados-Muñoz et al., 2015; Di Antonio et al., 2018), which
leads to overshoots of OPS if the ambient air humidity is
(too) high. An obvious solution is to dry the air. However,
any proper drying system would cost more than many mod-
els of OPS, and it is rarely seen in combination with low-cost
sensors. Analytical corrections are often used: humidity sen-
sors are used to measure the relative humidity of ambient air,
and some analytical models, like Kohler’s theory (Castarède
and Thomson, 2018) or the Hänel equation (Hänel, 1976),
are applied. Later in this paper we will make some observa-
tions on this issue.
Due to all the above-mentioned factors, it is always inter-
esting to check how OPS perform in different realistic sce-
narios. Numerous papers deal with laboratory calibrations
and outdoor evaluations of OPS (Karagulian et al., 2019;
Borghi et al., 2018; Chatzidiakou et al., 2019; Magi et al.,
2020; Sousan et al., 2016b; Malings et al., 2020; Kelly et al.,
2017; Sayahi et al., 2019; Crilley et al., 2018; Zheng et al.,
2018; Tasic et al., 2012; Cavaliere et al., 2018; Mukher-
jee et al., 2017; Sousan et al., 2016a; Zhang et al., 2018;
Holstius et al., 2014; Badura et al., 2018). Reported re-
sults vary depending on the composition of particulate matter
pollution, range of concentrations and meteorological fac-
tors. In Mukherjee et al. (2017) OPC-N2, PMS7003 and
11-R were compared against BAM-1020 during 12 weeks
in the Cuyama Valley, California, USA. Grimm 11-R per-
Atmos. Meas. Tech., 13, 6427–6443, 2020 https://doi.org/10.5194/amt-13-6427-2020
A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions 6429
formed well, while both OPC-N2 and PMS7003 (which is a
miniaturized version of PMS5003) produced mediocre per-
formance with heavy low bias. PurpleAir (PMS5003) was
tested in Tryner et al. (2020) using laboratory and field tests.
High bias of PMS5003 was observed. In Magi et al. (2020)
PurpleAir (PMS5003) was analysed for 16 months in Char-
lotte, North Carolina, USA, against BAM-1022, and high
bias of PMS5003 that increases with humidity was reported.
High mean bias of PurpleAir (PMS5003) was reported in
Kosmopoulos et al. (2020) as well.
The novelty of this research is a unique combination of
instruments and conditions of extremely high urban pollu-
tion. The city of Sarajevo is situated in a valley and is af-
fected by strong temperature inversions that appear typically
150–300 m above ground level with a very strong temper-
ature gradient in the inversion layer, exceeding 30K km−1
(Masic et al., 2019). The inversion episodes were present
during most of January 2020. As a consequence, the av-
erage monthly concentration of PM2.5was very high:
167.3 µg m−3. In contrast to that, monthly average values for
March and April 2020 were 21.6 and 19.6 µg m−3, respec-
tively. This presented an excellent opportunity to test the per-
formance of OPS in very different pollution levels. Simulta-
neously with OPS and reference gravimetric measurements,
a scanning mobility particle sizer (SMPS) was employed to
detect nanoparticles. It can detect particles with diameters
from 10 nm up to 1µm. While an SMPS can count very small
particles, 11-D can count larger particles, from 0.25 to 35 µm
in diameter. When they work simultaneously, they can detect
(almost) the full range of particles’ diameters, with a span
of more than 3 orders of magnitude. This will give detailed
insights into the mass distribution of particles.
2 Methodology and experimental setup
The experimental facility was located at the Faculty of Me-
chanical Engineering in the central part of the Sarajevo val-
ley (43.85424◦N, 18.39607◦E; 540 m above sea level) and
represents well the overall conditions in the city. The ref-
erence instrument for measurements of PM concentrations
was a Dadolab Gemini air sampler (Fig. 1). It is a single de-
vice with two completely independent channels (PM2.5and
PM10 in this campaign). The filter preparation and gravimet-
ric analysis are performed in a separate laboratory of the Fac-
ulty of Science, Department of Chemistry. The air sampler,
gravimetric laboratory and all filter procedures satisfied re-
quirements of the standard EN 12341:2014. According to
requirements of the standard, all filters were conditioned at
relative humidity between 45% and 50 % and temperature
between 19 and 21 ◦C.
The Grimm 11-D is a high-end optical particle sizer, with
sophisticated construction and the ability to count individual
particles from 250 nm to 35 µm in 31 equidistant (on log-
arithmic scale) channels. It uses a proprietary algorithm and
the manufacturer does not share information about the refrac-
tive index, density or weighting factors. It was factory cali-
brated and equipped with firmware version 12.50. Data were
recorded in 1 min intervals (6s is also possible). Since we use
the common term OPS occasionally, it should be noted that
11-D belongs to a different category of devices (in compari-
son to low-cost sensors).
Alphasense OPC-N2 belongs to the category of low-cost
sensors. The manufacturer transparently shared most speci-
fications. It has a much simpler construction than 11-D: in-
stead of a regulated pump, air flow is provided by a 25mm
fan. The device has 16 channels, from 380nm to 17 µm.
Firmware version 18.2 was used. Refractive index was n=
1.50+i0 and density was 1.65 gcm−3. All other parameters,
including weighting factors, were used as firmware default
values.
The Plantower PMS5003 could be termed a very low-cost
sensor, since its price is lower by an order of magnitude
than that of the OPC-N2. Limited specifications do not re-
veal all operating parameters. From the specification sheet
we can conclude that the device uses Mie scattering the-
ory, with a detection limit of 300nm, and has six channels.
It uses a red semiconductor laser, a photodetector at a 90◦
scattering angle (Kuula et al., 2020) and a 32-bit processor
(Cypress CY8C4245, 48 MHz). According to (Tanzer et al.,
2019) PMS5003 is a nephelometer, not the particle counter.
Air flow is provided by a 20mm fan. The PMS5003 has two
data outputs, one called SM (standard material, CF =1) and
another AE (atmospheric environment). The latter mode is
used in our work, since the manufacturer recommends the
AE mode for ambient air measurements, without further ex-
planation. Figure 2 shows results from our laboratory test us-
ing the incense scents as the source of PM. Based on these
results, the relationships between the SM and AE modes are
SMPM2.5=
AEPM2.5for AEPM2.5≤30
non-linear for 30 <AEPM2.5≤50,
1.5×AEPM2.5for AEPM2.5>50
SMPM10 =
AEPM10 for AEPM10 ≤43
non-linear for 43 <AEPM10 ≤77.
1.5×AEPM10 for AEPM10 >77
(5)
Based on PMS5003, we have designed a MAQS (Mobile
Air Quality System) smart sensor. Essentially, it is a mod-
ular platform for PMS5003, with options for additional sen-
sors (pressure, temperature, humidity, carbon dioxide, wind
speed), GNSS receiver, flash memory, WiFi module and
3D-printed enclosure. Eight MAQS sensors were made and
tested prior to the main campaign in order to evaluate con-
sistency between units. Figure 3 shows the results of prelim-
inary outdoor measurements for a batch of eight MAQS sen-
sors. They showed very good consistency: the coefficient of
determination, R2, between any two sensors from the batch
was greater than 0.99, and average readings from all sensors
are within ±10% from the average value of the batch of sen-
https://doi.org/10.5194/amt-13-6427-2020 Atmos. Meas. Tech., 13, 6427–6443, 2020
6430 A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions
Figure 1. Experimental setup: (a) colocated air sampler and Stevenson screen; (b) devices under test inside of the Stevenson screen: 11-D
with dryer, OPC-N2 with SPI adapter and (white-orange) enclosure, MAQS (white enclosure with grey front panel), and (c) indoor SMPS
with dryer.
Figure 2. AE and SM modes of the PMS5003 sensor.
sors. Data were recorded every minute on a local SD card
and remote cloud server simultaneously. The recording inter-
val can be as short as 1 s, but there was no need for that.
The Grimm 11-D and Alphasense OPC-N2 could not be
used outdoors without shelter, while MAQS has a special
case which provides basic protection for outdoor use. Fur-
thermore, a netbook PC was used to record data from the
OPC-N2. Outdoor shelter had to be constructed to accom-
modate 11-D with power supply, OPC-N2 with PC and SPI
adapter, and MAQS (for better protection). A Stevenson
screen-like wooden structure was designed for that purpose.
Another MAQS sensor was used at a remote location, for
reasons that will be explained later.
For low-cost sensors (OPC-N2 and MAQS) there was no
air dryer or heater, since they are typically used in such con-
ditions. We have designed and constructed a diffusion dryer
Atmos. Meas. Tech., 13, 6427–6443, 2020 https://doi.org/10.5194/amt-13-6427-2020
A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions 6431
Figure 3. Preliminary test of eight MAQS sensors, outdoor measurements.
for application on 11-D, which consists of a porous stain-
less steel tube surrounded with 1 kg of silica gel. The dryer
is compact, 25 cm in length with an 8 cm external diameter,
and does not reduce the mobility of the instrument. It was
installed only during the period of mild pollution.
Meteorological parameters were measured using the Van-
tage Pro2 (Davis Instruments, USA) weather station with
recording intervals of 15 min.
An SMPS is a complex system which consists of a conden-
sation particle counter (CPC), a differential mobility anal-
yser (DMA) and a charge conditioner (often inadequately
called a “neutralizer”). Depending on the characteristics of
the DMA, the SMPS can be configured for a certain span of
particle diameters. We have used a Grimm 5.416 high-end
SMPS with a long DMA which is able to separate particles
from 10 to 1000 nm into 129 channels, equidistant on a log-
arithmic scale. Despite the fact that particles with diameters
below 10 nm play an important role in nucleation and growth
studies (Tiszenkel et al., 2019), their contribution to the mass
budget is negligible. Another (larger) in-house developed dif-
fusion dryer was installed at the inlet of the SMPS. A soft
X-ray device was used as the charge conditioner. Scanning
mode (alternating upscan and downscan) was used for all
measurements. One scan takes about 4 min (8 min for both
upscan and downscan). When working in parallel, the SMPS
and 11-D form a powerful wide-range spectrometer, which
covers a range of particle diameters from 10nm to 35 µm
in 160 channels. Additionally, there is an overlapping area
between 250 and 1000 nm where we can see how well these
two instruments match. The complex SMPS system was kept
indoors (an unavoidable necessity, since both X-ray charger
and DMA use very high operating voltages). The air was
sampled from outside using a conductive tube of the short-
est possible length to avoid particle losses. It was running
continuously, except for the periods of maintenance.
A rigorous data validation procedure was used. All in-
struments were inspected periodically and data logs were
analysed thoroughly. When calculating daily average values,
complete and consistent data series were required.
3 Results and discussion
During this campaign 296 filters were used in the reference
air sampler. After the removal of several blank filters used
for periodic verification and those with incomplete sampling
(the pneumatic system of the air sampler failed to load new
https://doi.org/10.5194/amt-13-6427-2020 Atmos. Meas. Tech., 13, 6427–6443, 2020
6432 A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions
filters automatically a couple of times), 288 filters remained:
143 PM2.5and 145 PM10 samples. Figure 2 shows PM2.5and
PM10 daily average concentrations, together with hourly and
daily values of ambient air temperature and relative humidity.
Some modifications of the shelter for 11-D and OPC-N2
were necessary, making those instruments unavailable peri-
odically during December and January. Additionally, more
frequent maintenance, such as cleaning of 11-D, was needed
when working in extreme conditions. The same goes for the
SMPS, which was maintained according to the recommen-
dations of the manufacturer. Taking into account difficult op-
erating conditions, the amount of data collected is satisfac-
tory during the period of strong pollution and excellent dur-
ing the period of mild pollution. The lower limit of detection
(LLoD) of PM2.5concentration for evaluated optical aerosol
devices is estimated based on their actual field performance.
Standard deviation (σ) was calculated for periods with near-
zero ambient PM concentration, and an average value of 3σ
is the estimated LLoD. For PMS5003 our final estimation
is 5 µg m−3. The same value is an estimation of Magi et al.
(2020), calculated by averaging segmented regressions, and
Bulot et al. (2019), by combining results from several previ-
ous studies. This method applied to OPC-N2 yields an LLoD
of 2 and 1 µg m−3for 11-D. For the reference gravimetric
system LLoD was calculated using the blank filters, which
were treated in exactly the same way as real samples (except
the sampling of particulate matter), and the calculated value
of LLoD is 0.7 µgm−3. All measurements below LLoD were
discarded during the quality assurance phase.
3.1 Strong urban pollution
During the period of strong urban pollution (2 December
2019–12 March 2020), the average value of PM2.5con-
centration was 82.9 µg m−3, with minimum daily average
value 1.3 µg m−3and maximum value 504.9 µg m−3(Fig. 4).
In the same period, the average PM10 concentration was
95.5 µg m−3, with minimum value 3.6 µg m−3and maximum
value 549.0 µg m−3. The ratio of average values of concen-
trations PM2.5/PM10 was 0.87. Very good correlations were
observed for all three OPS against the reference instrument
(Fig. 5). Such a range of ambient PM concentrations was
favourable for achievement of high R2values, but non-linear
effects of low-cost sensors were observed too.
The Grimm 11-D produced results with R2values of 0.988
and 0.985 for PM2.5and PM10 concentrations, respectively.
Absolute values were larger than the reference, on average
17.6 % for PM2.5and 25.5 % for PM10. The average ratio
PM2.5/PM10 measured by 11-D was 0.93. Mean absolute
error (MAE) was 13.4 µgm−3for PM2.5and 10.8 µg m−3for
PM10. Alphasense OPC-N2 undershoots with respect to the
reference values, on average 31.0% for PM2.5and 36.8 %
for PM10, but R2coefficients are relatively high: 0.903 and
0.920 for PM2.5and PM10, respectively. The OPC-N2 mea-
sured the ratio PM2.5/PM10 to be 0.97. MAE for this sen-
sor was 29.4 µg m−3for PM2.5and 34.8 µg m−3for PM10.
The MAQS sensor produced surprisingly good R2values of
0.975 for PM2.5and 0.950 for PM10. In terms of absolute val-
ues, it overshoots by 31.9% for PM2.5and 49.3 % for PM10
(on average). The calculated ratio PM2.5/PM10 was 0.76.
MAE was 35.9 µg m−3for PM2.5and 55.2 µg m−3for PM10 .
It seems that the Plantower PMS5003 cannot accurately de-
termine the PM10 fraction. One possible explanation is pro-
vided by a laboratory test of PMS5003, where it was found
that its size bin [2.5–10 µm] is noisy and inaccurate (Kuula
et al., 2020). Further investigation of this behaviour would be
useful.
None of the tested OPS were equipped with an air dryer,
and this certainly contributes to overprediction. However, Al-
phasense OPC-N2 with default firmware settings underpre-
dicts values, despite the particle hygroscopic growth effect.
3.2 Mild urban pollution
The correlation coefficients changed dramatically in the pe-
riod of mild pollution (13 March 2020–4 May 2020), as
Fig. 6 shows. The much narrower range of particulate mat-
ter concentrations plays an important role, and even the ref-
erence method is less accurate, since the mass difference
of loaded and blank filters becomes very small (smaller
than 1 mg for the 24 h sampling period if PM concentration
is below 18 µg m−3). The average concentration of PM2.5
was 19.7 µg m−3with a minimum daily average value of
7.1 µg m−3and a maximum value of 39.3 µg m−3. During
this period, the average value of the PM10 concentration was
24.2 µg m−3, with minimum and maximum values of 7.6 and
48.8 µg m−3, respectively. The ratio PM2.5/PM10 was 0.81
on average.
This time the Grimm 11-D was equipped with a dryer,
whose effects will be discussed in the next subsection. The
device produced relatively high R2values of 0.868 for
PM2.5and 0.917 for PM10. The absolute readings under-
estimated concentrations of PM2.5by 16.3 % on average,
while PM10 were underestimated by 10.9 % on average. The
PM2.5/PM10 ratio was 0.87. This test clearly shows that 11-
D is a completely different class of instrument (in compari-
son to low-cost sensors). When equipped with a dryer, 11-D
shows a level of performance comparable to BAM, at least
those reported by Hafkenscheid and Vonk (2014). MAE was
3.0 µg m−3for PM2.5and 4.1 µg m−3for PM10.
Alphasense OPC-N2 did not perform well during the pe-
riod of mild pollution. Coefficients of determination, R2,
were only 0.284 for PM2.5and 0.525 for PM10. Absolute
readings are worrying: the OPC-N2 underpredicted PM2.5
by 67.6 % and PM10 by 71.6 % on average. The ratio
PM2.5/PM10 was 0.73. MAE was 13.8 µg m−3for PM2.5
and 15.8 µg m−3for PM10.
The MAQS sensor demonstrated mediocre performance,
with R2values of 0.730 for PM2.5and 0.718 for PM10.
On average, this sensor overpredicted PM2.5by 30.5% and
Atmos. Meas. Tech., 13, 6427–6443, 2020 https://doi.org/10.5194/amt-13-6427-2020
A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions 6433
Figure 4. Reference PM concentrations with ambient air temperature and humidity.
PM10 by 32.6 %. The PM2.5/PM10 ratio was 0.83, very
close to the reference value (in contrast to the performance
of the sensor in the period of strong pollution). MAE was
7.1 µg m−3for PM2.5and 8.2 µg m−3for PM10.
It would be interesting to test low-cost sensors with a
proper dryer as well, but that combination is rarely seen in
practice.
3.3 Humidity influence
One of the important factors in ambient measurements of PM
concentrations is humidity, since the particles reflect more
light (i.e. appear larger) during measurements due to hygro-
scopic growth. This can be described using the Hänel equa-
tion:
fζ(RH)=1−RH
1−RHref −γ
,(6)
https://doi.org/10.5194/amt-13-6427-2020 Atmos. Meas. Tech., 13, 6427–6443, 2020
6434 A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions
Figure 5. OPS performance during the period of strong pollution (2 December 2019–12 March 2020).
where fζis the enhancement factor for particle property ζ.
Here RH represents the relative humidity and RHref is a ref-
erence relative humidity:
fζ(RH)=ζ(RH)
ζ(RHref).(7)
It is important to note that the coefficient γ, which is an indi-
cator of the hygroscopicity of particles, depends on the type
of particles (and changes whenever the composition of ambi-
ent particles is changed).
If we compare results produced by 11-D relative to the ref-
erence, during period 1 (without a dryer) and period 2 (with
a dryer), we can see that readings of 11-D were reduced by
more than 30 %. However, we cannot conclude whether it
was the effect of the dryer or the consequence of signifi-
cantly different ambient conditions. Unfortunately, we have
only one 11-D, so we could not measure simultaneously with
and without a dryer (that is the reason why we used the in-
strument with the dryer only in one of the two periods). If
we take into account two intervals with similar ambient con-
ditions, with and without a dryer we get the following val-
ues: from 27 February 2020 to 12 March 2020 the average
ambient concentration of PM2.5was 21.1 µg m−3, while 11-
D (without a dryer) measured 21.5 % more. In the second
interval, from 13 March 2020 to 1 April 2020, the ambient
concentration was similar, 21.0 µg m−3, while 11-D (with a
dryer) measured a 1.4 % smaller value. This comparison in-
dicates that the effect of the dryer could be around 23 %. A
similar analysis for PM10 concentrations gives an estimate of
about 20 % for the dryer effect.
Atmos. Meas. Tech., 13, 6427–6443, 2020 https://doi.org/10.5194/amt-13-6427-2020
A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions 6435
Figure 6. OPS performance during the period of mild pollution (13 March 2020–4 May 2020).
The Grimm 11-D has a very useful feature: an internal
temperature and humidity sensor. Figure 7 shows the self-
heating and diffusion dryer effect on 11-D by comparing in-
ternal and external measurements of temperature and humid-
ity. The average ambient air temperature from 27 February
2020 to 1 April 2020 was 7.02 ◦C, while the average 11-D
internal temperature was 14.27 ◦C, which shows a signifi-
cant difference of 7.25 ◦C. This self-heating effect reduces
internal humidity significantly, and we can see that it rarely
goes beyond 50 %. Once the dryer is installed, internal rela-
tive humidity is further reduced: the average value of inter-
nal humidity without a dryer (27 February 2020–12 March
2020) was 36.2 %, and with a dryer (13 March 2020–1 April
2020) it was 21.8 % (the ambient air humidity also dropped
in the later period, but nevertheless the effect of the dryer is
evident).
After roughly a month, the dryer’s performance degraded
and the silica gel needed a regeneration (but it was not per-
formed since we did not want to interrupt measurements
when the end of the campaign was near).
Figure 8 shows the long-term (13.5-month) comparison of
MAQS and BAM-1020 with a time resolution of 1 h, together
with measured values of ambient air humidity. By averaging
all these data we can estimate the influence of humidity on
the MAQS sensor: if we sort the measurements by humid-
ity, a subset of points where humidity is below 50 % has an
average bias of 14.3%, for humidity range 50 %–70%, bias
is 16.5 %, for humidity range 70 %–85 %, bias is 31.6 % and
for humidity range 85 %–100 %, bias is 37.3 %. If we sub-
https://doi.org/10.5194/amt-13-6427-2020 Atmos. Meas. Tech., 13, 6427–6443, 2020
6436 A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions
Figure 7. Self-heating and diffusion dryer effect of the Grimm 11-D. The dryer was installed on 12 March 2020 15:30LT (UTC+1).
tract the bias of the least humidity subset from the bias of the
highest humidity subset, we can estimate that humidity in-
fluence adds up to 23 % on PM2.5readings from the MAQS
sensor, which is a similar result to the analysis of humidity
influence on our 11-D with dryer installed. While this influ-
ence cannot be neglected, it is still relatively modest. A pos-
sible reason is the chemical composition of PM without too
many hygroscopic components, but that requires a different
type of analysis. A relatively modest humidity influence on
PMS5003 was also reported in Jayaratne et al. (2018), and a
surprisingly low influence is reported by Kosmopoulos et al.
(2020).
3.4 SMPS data and wide-range spectrometer
The wide-range spectrometer (SMPS+11D) produced very
valuable results. Figure 9 shows the continuous concentra-
tion and mass distributions. It is created from hourly aver-
age measurements from the SMPS and 11-D. A relative den-
sity of 1.65 was applied in SMPS software (based on Lab-
VIEW) for the mass calculation. No other corrections were
performed and all settings were factory defaults. Selected
histograms (hourly average values) are shown in Fig. 10.
What we can see from Figs. 9 and 10 is that in the period
of strong pollution the dominant mass contribution comes
from particles with diameters around 300 nm. In terms of
concentrations, particles around 100 nm appear in the great-
est numbers, with occasional secondary peaks coming from
even smaller particles.
In the period of mild pollution, however, we can see that
particles larger than 2.5 µm often appear on histograms (usu-
ally about 3 µm in diameter). Number concentrations still
have peaks of about 100nm, but sometimes the distribution
Atmos. Meas. Tech., 13, 6427–6443, 2020 https://doi.org/10.5194/amt-13-6427-2020
A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions 6437
Figure 8. Long-term comparisons of the MAQS sensor with BAM operated by the US EPA at the nearby location: hourly, daily, and monthly
average values and comparison of hourly and daily average values of two MAQS sensors: the first one (MAQS) at our main facility and the
second one (MAQS-FEE) at the Faculty of Electrical Engineering in the immediate vicinity of BAM-1020.
is different in favour of even smaller particles, as Fig. 10
shows. Again, the largest mass contribution comes from par-
ticles around 300 nm.
In the overlapping area, the SMPS and 11-D matched very
well, almost perfectly for concentrations. Their match was
not as good for mass calculations, but that is understandable,
taking into account all the factors explained in Sect. 1. Over-
all, the combination of the SMPS and 11-D worked very well
and gave the full spectrum of particles, both for number con-
centrations and mass distribution.
The obtained mass distribution of particles, especially dur-
ing the period of strong pollution, raises the question of the
suitability of OPS for measuring mass concentrations and
resolving different fractions, since they cannot detect small
particles that significantly contribute to the total mass. For
example, the Alphasense OPC-N2 has a detection limit of
380 nm and cannot detect the particles around 300 nm which
form the dominant contribution to the mass budget. The
Grimm 11-D, with a detection limit of 250 nm, has a far bet-
ter potential to resolve mass fractions.
3.5 OPS histograms and Aralkum Desert dust
All tested OPS have data bins, with different numbers of
channels, as described in Sect. 2. Figure 11 shows histograms
that compare data bins from 11-D, OPC-N2 and MAQS on
https://doi.org/10.5194/amt-13-6427-2020 Atmos. Meas. Tech., 13, 6427–6443, 2020
6438 A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions
Figure 9. Wide-range spectrometer (SMPS+11D) and hourly average values. Relative density 1.65 was applied to the SMPS to calculate
mass of particles.
18 January 2020 (strong pollution) and 16 April 2020 (mild
pollution). It should be noted that we compare here data bins
from devices with different specifications and categories.
As expected, 11-D has the ability to count particles below
300 nm, which appear in the greatest numbers. Counting ef-
ficiency of OPC-N2 is investigated in laboratory conditions
using PSL particles in Sousan et al. (2016a), and the results
were good for particles larger than 0.8 µm, while for parti-
cles with a diameter of 0.5 µm OPC-N2 the device showed a
lower detection efficiency (the detection limit of OPC-N2 is
0.38 µm). In our realistic scenario, the dominant contribution
to the mass comes from particles much smaller than 0.8 µm
(Figs. 9 and 10), which is not favourable to OPC-N2.
In contrast to OPC-N2, PMS5003 has problems with
coarse particles, as indicated in laboratory tests (Kuula et al.,
2020). If the fraction of coarse particles is small and steady,
PMS5003 performs much better. Ambient conditions in
Bosnia-Herzegovina are such most of the time, since the pri-
mary source of PM is combustion of coal and biomass. That
could explain why PMS5003 performs better than OPC-N2
most of the time. However, different conditions were ob-
served on 27 March 2020, when the dust from the Aralkum
Desert covered part of Europe, including our test location.
During this episode, OPC-N2 performed much better than
PMS5003, which was not able to determine a large fraction
of coarse particles correctly (Fig. 11). A similar observation
Atmos. Meas. Tech., 13, 6427–6443, 2020 https://doi.org/10.5194/amt-13-6427-2020
A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions 6439
Figure 10. Wide-range histograms and hourly average values. Relative density 1.65 was applied to the SMPS to calculate mass of particles.
about PMS5003 was reported by Kosmopoulos et al. (2020),
when Sahara dust covered Greece.
3.6 Long-term performance
Another question about OPS, especially low-cost types, is
the drift of performance over time. The PMS5003 sensor
uses a semiconductor laser (diode laser) which has a lim-
ited lifetime. We have some long-term comparisons of the
MAQS sensor with MetOne BAM-1020 operated at a nearby
location by the US EPA. Strictly speaking, their station is
not collocated with our equipment, but for a distance of
only 300 m it is reasonable to assume that the air compo-
sition is very similar at these two points, since they are lo-
cated in the same neighbourhood. In order to verify that as-
sumption, we have installed another MAQS sensor at the lo-
cation of the Faculty of Electrical Engineering, University
of Sarajevo, which is in the immediate vicinity of the US
https://doi.org/10.5194/amt-13-6427-2020 Atmos. Meas. Tech., 13, 6427–6443, 2020
6440 A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions
Figure 11. OPS histograms and Aralkum Desert dust episode, hourly average values.
EPA site and at the same distance from us (about 300 m).
Figure 8 shows long-term comparisons of the MAQS sen-
sor and BAM-1020 and additional verification of correla-
tion between readings of the two MAQS sensors, which was
very high (R2=0.970, MAE =4.7 µg m−3for hourly aver-
age values and R2=0.994, MAE =2.9 µg m−3for daily av-
erage values), confirming our assumption that these two lo-
cations share the same air in terms of PM concentrations and
properties.
Based on 13.5 months of continuous comparison of
MAQS and BAM-1020, hourly average values give R2co-
efficient 0.919 and MAE 16.7 µg m−3. Daily average val-
ues produce R2coefficient 0.980 and MAE 12.2 µg m−3,
while the monthly average values give R2=0.998 and
MAE =11.4 µg m−3(Fig. 8).
This leads us to the conclusion that time averaging reduces
by a lot the influence of variation of PM composition and
meteorological variations. If we use a longer time-average
period, we lose one of the major advantages of low-cost sen-
sors (time resolution), but it is a more natural approach to
correcting readings compared to using artificial algorithms
like neural networks (Badura et al., 2019) or machine learn-
ing (Si et al., 2020). An excellent viewpoint of this issue is
given by Hagler et al. (2018). The calibration of a larger num-
Atmos. Meas. Tech., 13, 6427–6443, 2020 https://doi.org/10.5194/amt-13-6427-2020
A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions 6441
ber of low-cost sensors can be simplified if they show simi-
lar relative performance (to each other) in the laboratory and
field (Sousan et al., 2018). Floating corrections, even physi-
cally justifiable interventions, such as the instantaneous cor-
rection for humidity growth of particles, insert a lot of noise,
and the benefit is questionable. Depending on ambient con-
ditions, self heating of the sensor and some other factors, rel-
ative humidity may not be accurately determined. Even if we
have a very accurate humidity measurement, the hygroscopic
growth coefficient will change whenever the composition of
PM changes, inevitably injecting noise into the results.
We can also see strong non-linear effects at very high con-
centrations, above 500µg m−3. In that case a quadratic re-
gression fit will be more suitable.
During this period of 13.5 months of continuous outdoor
operation, the MAQS sensor worked flawlessly without per-
formance drifts. Designed enclosure sufficiently protected
the sensor outdoors while not obstructing air sampling.
4 Conclusions
A comprehensive experimental study was carried out with
the aim of evaluating the performance of three very differ-
ent OPS: high-end Grimm 11-D, low-cost Alphasense OPC-
N2 and in-house developed MAQS sensor, which is based
on another low-cost sensor, the Plantower PMS5003 sensor.
The study was performed in realistic conditions of strong and
mild urban pollution. The reference instrument was a dual-
channel air sampler with gravimetric analysis in a separate
laboratory. In total 288 filters were collected from 2 Decem-
ber 2019 to 4 May 2020.
During the period of strong urban pollution all three in-
struments produced very high R2values. However, during
the period of mild urban pollution, these correlation factors
dropped significantly, especially for the Alphasense OPC-N2
sensor measuring the PM2.5parameter. The OPC-N2 under-
estimated the mass concentrations badly, especially during
the period of mild pollution. The MAQS sensor overshoots
PM2.5concentrations by approximately 30 % on average,
which is partially caused by hygroscopic growth.
The wide-range spectrometer, which consists of the SMPS
and 11-D, produced valuable information about the distribu-
tion of particles, both in number and mass concentrations.
Particles with diameters around 100 nm (and sometimes be-
low) represent the dominant fraction in pure numbers, while
particles with diameters of around 300 nm give the highest
contribution to mass. In the period of mild pollution, parti-
cles larger then 2.5 µm gave a larger contribution than in the
period of strong pollution.
The Grimm 11-D performed well in all conditions, and
when equipped with a dryer, it performed at a comparable
level to the beta attenuation monitor. For the calibration of
low-cost sensors, especially those based on PMS5003, we
propose a linear or quadratic correction (in the case of high
pollution levels) with steady coefficients, since the instanta-
neous corrections insert noise into results.
Future measurements should further investigate character-
istics of OPS in different ambient conditions, influence of
humidity and effect of micro-dryers specifically designed for
low-cost sensors and mass distributions by means of wide-
range spectrometers.
Data availability. The underlying datasets for this publication are
available at https://doi.org/10.5281/zenodo.3897379 (Masic et al.,
2020). Furthermore, data from the BAM measurements by the
US EPA are available at https://www.airnow.gov/?city=Sarajevo&
country=BIH (U.S. EPA, 2020).
Author contributions. AM participated in all phases of this research
and wrote the manuscript with contributions from all the co-authors.
DB and BP performed field work together with AM. AB designed,
manufactured and analysed the performance of two diffusion dry-
ers and evaluated the influence of humidity on the readings of 11-D.
SZ and JH performed gravimetric measurements, including precon-
ditioning and treatment of the filters, ensuring strict fulfilment of
the standard EN 12341:2014.
Competing interests. The authors declare that they have no conflict
of interest.
Acknowledgements. We would like to thank the Embassy of Swe-
den in Bosnia-Herzegovina for supporting this research and the
United States Environmental Protection Agency for sharing data
publicly.
Review statement. This paper was edited by Pierre Herckes and re-
viewed by three anonymous referees.
References
Badura, M., Batog, P., Drzeniecka-Osiadacz, A., and Modzel, P.:
Evaluation of low-cost sensors for ambient PM2.5monitoring,
J. Sensors, 2018, 1–16, https://doi.org/10.1155/2018/5096540,
2018.
Badura, M., Batog, P., Drzeniecka-Osiadacz, A., and Modzel, P.:
Regression methods in the calibration of low-cost sensors for am-
bient particulate matter measurements, SN Applied Sciences, 1,
622, https://doi.org/10.1007/s42452-019-0630-1, 2019.
Bezantakos, S., Schmidt-Ott, F., and Biskos, G.: Perfor-
mance evaluation of the cost-effective and lightweight
Alphasense optical particle counter for use onboard un-
manned aerial vehicles, Aerosol Sci. Tech., 52, 385–392,
https://doi.org/10.1080/02786826.2017.1412394, 2018.
Borghi, F., Spinazze, A., Campagnolo, D., Rovelli, S., Cattaneo, A.,
and Cavallo, D. M.: Precision and accuracy of a direct-reading
https://doi.org/10.5194/amt-13-6427-2020 Atmos. Meas. Tech., 13, 6427–6443, 2020
6442 A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions
miniaturized monitor in PM2.5exposure assessment, Sensors,
18, 3089, https://doi.org/10.3390/s18093089, 2018.
Bulot, F. M. J., Johnston, S. J., Basford, P. J., Easton, N. H. C.,
Apetroaie-Cristea, M., Foster, G. L., Morris, A. K. R., Cox,
S. J., and Loxham, M.: Long-term field comparison of multi-
ple low-cost particulate matter sensors in an outdoor urban envi-
ronment, Sci. Rep.-UK, 9, 7497, https://doi.org/10.1038/s41598-
019-43716-3, 2019.
Cai, C., Stebounova, L. V., Peate, D. W., and Peters, T. M.:
Evaluation of a Portable Aerosol Collector and Spec-
trometer to measure particle concentration by com-
position and size, Aerosol Sci. Tech., 53, 675–687,
https://doi.org/10.1080/02786826.2019.1600654, 2019.
Castarède, D. and Thomson, E. S.: A thermodynamic de-
scription for the hygroscopic growth of atmospheric
aerosol particles, Atmos. Chem. Phys., 18, 14939–14948,
https://doi.org/10.5194/acp-18-14939-2018, 2018.
Cavaliere, A., Carotenuto, F., Di Gennaro, F., Gioli, B., Gualtieri,
G., Martelli, F., Matese, A., Toscano, P., Vagnoli, C., and
Zaldei, A.: Development of Low-Cost Air Quality Stations
for Next Generation Monitoring Networks: Calibration and
Validation of PM2.5and PM10 Sensors, Sensors, 18, 2843,
https://doi.org/10.3390/s18092843, 2018.
Chatzidiakou, L., Krause, A., Popoola, O. A. M., Di Antonio, A.,
Kellaway, M., Han, Y., Squires, F. A., Wang, T., Zhang, H.,
Wang, Q., Fan, Y., Chen, S., Hu, M., Quint, J. K., Barratt, B.,
Kelly, F. J., Zhu, T., and Jones, R. L.: Characterising low-cost
sensors in highly portable platforms to quantify personal expo-
sure in diverse environments, Atmos. Meas. Tech., 12, 4643–
4657, https://doi.org/10.5194/amt-12-4643-2019, 2019.
Crilley, L. R., Shaw, M., Pound, R., Kramer, L. J., Price, R.,
Young, S., Lewis, A. C., and Pope, F. D.: Evaluation of a
low-cost optical particle counter (Alphasense OPC-N2) for
ambient air monitoring, Atmos. Meas. Tech., 11, 709–720,
https://doi.org/10.5194/amt-11-709-2018, 2018.
Crilley, L. R., Singh, A., Kramer, L. J., Shaw, M. D., Alam, M. S.,
Apte, J. S., Bloss, W. J., Hildebrandt Ruiz, L., Fu, P., Fu, W.,
Gani, S., Gatari, M., Ilyinskaya, E., Lewis, A. C., Ng’ang’a, D.,
Sun, Y., Whitty, R. C. W., Yue, S., Young, S., and Pope, F. D.: Ef-
fect of aerosol composition on the performance of low-cost op-
tical particle counter correction factors, Atmos. Meas. Tech., 13,
1181–1193, https://doi.org/10.5194/amt-13-1181-2020, 2020.
Di Antonio, A., Popoola, O. A. M., Ouyang, B., Saffell, J., and
Jones, R. L.: Developing a relative humidity correction for low-
cost sensors measuring ambient particulate matter, Sensors, 18,
2790, https://doi.org/10.3390/s18092790, 2018.
Downward, G. S., van Nunen, E. J. H. M., Kerckhoffs, J., Vineis,
P., Brunekreef, B., Boer, J. M. A., Messier, K. P., Roy, A., Ver-
schuren, W. M. M., van der Schouw, Y. T., Sluijs, I., Gulliver,
J., Hoek, G., and Vermeulen, R.: Long-term exposure to ultra-
fine particles and incidence of cardiovascular and cerebrovascu-
lar disease in a prospective study of a Dutch cohort, Environ.
Health Persp., 126, 1–8, https://doi.org/10.1289/EHP3047, 2018.
ECWG (European Commission Working Group): Guide to the
demonstration of equivalence of ambient air monitoring meth-
ods, Report, Brussels, Belgium, 92 pp., 2010.
Granados-Muñoz, M. J., Navas-Guzmán, F., Bravo-Aranda, J. A.,
Guerrero-Rascado, J. L., Lyamani, H., Valenzuela, A., Titos,
G., Fernández-Gálvez, J., and Alados-Arboledas, L.: Hygro-
scopic growth of atmospheric aerosol particles based on ac-
tive remote sensing and radiosounding measurements: selected
cases in southeastern Spain, Atmos. Meas. Tech., 8, 705–718,
https://doi.org/10.5194/amt-8-705-2015, 2015.
Hafkenscheid, T. L. and Vonk, J.: Evaluation of equivalence of the
MetOne BAM-1020 for the measurement of PM2.5in ambient
air, RIVM Letter report 2014-0078), National Institute for Public
Health and the Environment, Bilthoven, the Netherlands, 37 pp.,
2014.
Hagler, G. S. W., Williams, R., Papapostolou, V., and Polidori, A.:
Air Quality Sensors and Data Adjustment Algorithms: When Is It
No Longer a Measurement?, Environ. Sci. Tech., 52, 5530–5531,
https://doi.org/10.1021/acs.est.8b01826, 2018.
Hänel, G.: The properties of atmospheric aerosol particles as func-
tions of the relative humidity at thermodynamic equilibrium
with the surrounding moist air, Adv. Geophys., 19, 73–188,
https://doi.org/10.1016/S0065-2687(08)60142-9, 1976.
Holstius, D. M., Pillarisetti, A., Smith, K. R., and Seto, E.: Field
calibrations of a low-cost aerosol sensor at a regulatory mon-
itoring site in California, Atmos. Meas. Tech., 7, 1121–1131,
https://doi.org/10.5194/amt-7-1121-2014, 2014.
Jayaratne, R., Liu, X., Thai, P., Dunbabin, M., and Morawska, L.:
The influence of humidity on the performance of a low-cost
air particle mass sensor and the effect of atmospheric fog, At-
mos. Meas. Tech., 11, 4883–4890, https://doi.org/10.5194/amt-
11-4883-2018, 2018.
Karagulian, F., Barbiere, M., Kotsev, A., Spinelle, L., Gerboles,
M., Lagler, F., Redon, N., Crunaire, S., and Borowiak,
A.: Review of the Performance of Low-Cost Sensors
for Air Quality Monitoring, Atmosphere-Basel, 10, 506,
https://doi.org/10.3390/atmos10090506, 2019.
Kelly, K. E., Whitaker, J., Petty, A., Widmer, C., Dyb-
wad, A., Sleeth, D., Martin, R., and Butterfield, A.:
Ambient and laboratory evaluation of a low-cost par-
ticulate matter sensor, Environ. Pollut., 221, 491–500,
https://doi.org/10.1016/j.envpol.2016.12.039, 2017.
Kosmopoulos, G., Salamalikis, V., Pandis, S. N., Yannopoulos, P.,
Bloutsos, A. A., and Kazantzidis, A.: Low-cost sensors for mea-
suring airborne particulate matter: Field evaluation and calibra-
tion at a South-Eastern European site, Sci. Total Environ., 748,
141396, https://doi.org/10.1016/j.scitotenv.2020.141396, 2020.
Kuula, J., Makela, T., Hillamo, R., and Timonen, H.: Response char-
acterization of an inexpensive aerosol sensor, Sensors, 17, 2915,
https://doi.org/10.3390/s17122915, 2017.
Kuula, J., Mäkelä, T., Aurela, M., Teinilä, K., Varjonen,
S., González, Ó., and Timonen, H.: Laboratory evalua-
tion of particle-size selectivity of optical low-cost particu-
late matter sensors, Atmos. Meas. Tech., 13, 2413–2423,
https://doi.org/10.5194/amt-13-2413-2020, 2020.
Li, H. Z., Gu, P., Ye, Q., Zimmerman, N., Robinsona, E. S.,
Subramanian, R., Apte, J. S., Robinsona, A. L., and Presto,
A. A.: Spatially dense air pollutant sampling: Implica-
tions of spatial variability on the representativeness of sta-
tionary air pollutant monitors, Atmos. Environ., 2, 1–13,
https://doi.org/10.1016/j.aeaoa.2019.100012, 2019.
Magi, B. I., Cupini, C., Francis, J., Green, M., and Hauser, C.:
Evaluation of PM2.5measured in an urban setting using a low-
cost optical particle counter and a Federal Equivalent Method
Atmos. Meas. Tech., 13, 6427–6443, 2020 https://doi.org/10.5194/amt-13-6427-2020
A. Masic et al.: Evaluation of optical particulate matter sensors under realistic conditions 6443
Beta Attenuation Monitor, Aerosol Sci. Tech., 54, 147–159,
https://doi.org/10.1080/02786826.2019.1619915, 2020.
Malings, C., Tanzer, R., Hauryliuk, A., Saha, P. K., Robin-
son, A. L., Presto, A. A., and Subramanian, R.: Fine particle
mass monitoring with low-cost sensors: Corrections and long-
term performance evaluation, Aerosol Sci. Tech., 54, 160–174,
https://doi.org/10.1080/02786826.2019.1623863, 2020.
Martin, R. V., Brauer, M., van Donkelaar, A., Shaddick, G., Narain,
U., and Dey, S.: No one knows which city has the highest con-
centration of fine particulate matter, Atmos. Environ., 3, 1–5,
https://doi.org/10.1016/j.aeaoa.2019.100040, 2019.
Masic, A., Bibic, D., Pikula, B., Dzaferovic-Masic, E., and
Musemic, R.: Experimental study of temperature inversions
above urban area using unmanned aerial vehicle, Therm. Sci., 23,
3327–3338, https://doi.org/10.2298/TSCI180227250M, 2019.
Masic, A., Bibic, D., Pikula, B., Blazevic, A., Huremovic, J., and
Zero S.: Evaluation of optical particulate matter sensors under
realistic conditions of strong and mild urban pollution, Zenodo,
https://doi.org/10.5281/zenodo.3897379, 2020.
Mie, G.: Beiträge zur Optik trüber Medien, speziell kolloidaler
Metallösungen (contributions to the optics of diffuse media,
especially colloid metal solutions, Ann. Phys., 25, 377–445,
https://doi.org/10.1002/andp.19083300302, 1908 (in German).
Morawska, L., Thai, P. K., Liu, X., Asumadu-Sakyi, A., Ayoko, G.,
Bartonova, A., Bedini, A., Chai, F., Christensen, B., Dunbabin,
M., Gao, J., Hagler, G. S. W., Jayaratne, R., Kumar, P., Lau,
A. K. H., Louie, P. K. K., Mazaheri, M., Ning, Z., Motta, N.,
Mullins, B., Rahman, M. M., Ristovski, Z., Shafiei, M., Tjon-
dronegoro, D., Westerdahl, D., and Williams, R.: Applications of
low-cost sensing technologies for air quality monitoring and ex-
posure assessment: How far have they gone?, Environ. Int., 116,
286–299, https://doi.org/10.1016/j.envint.2018.04.018, 2018.
Mukherjee, A., Stanton, L. G., Graham, A. R., and Roberts, P. T.:
Assessing the utility of low-cost particulate matter sensors over a
12-week period in the Cuyama Valley of California, Sensors, 17,
1805, https://doi.org/10.3390/s17081805, 2017.
Mukherjee, A., Brown, S. G., McCarthy, M. C., Pavlovic, N. R.,
Stanton, L. G., Snyder, J. L., D’Andrea, S., and Hafner, H. R.:
Measuring Spatial and Temporal PM2.5Variations in Sacra-
mento, California, Communities Using a Network of Low-Cost
Sensors, Sensors, 19, 4701, https://doi.org/10.3390/s19214701,
2019.
Sayahi, T., Butterfield, A., and Kelly, K. E.: Long-term
field evaluation of the Plantower PMS low-cost par-
ticulate matter sensors, Environ. Poll., 245, 932–940,
https://doi.org/10.1016/j.envpol.2018.11.065, 2019.
Si, M., Xiong, Y., Du, S., and Du, K.: Evaluation and calibra-
tion of a low-cost particle sensor in ambient conditions using
machine-learning methods, Atmos. Meas. Tech., 13, 1693–1707,
https://doi.org/10.5194/amt-13-1693-2020, 2020.
Sousan, S., Koehler, K., Hallett, L., and Peters, T. M.:
Evaluation of the Alphasense optical particle counter
(OPC-N2) and the Grimm portable aerosol spectrom-
eter (PAS-1.108), Aerosol Sci. Tech., 50, 1352–1365,
https://doi.org/10.1080/02786826.2016.1232859, 2016a.
Sousan, S., Koehler, K., Thomas, G., Park, J. H., Hillman,
M., Halterman, A., and Peters, T. M.: Inter-comparison
of low-cost sensors for measuring the mass concentration
of occupational aerosols, Aerosol Sci. Tech., 50, 462–473,
https://doi.org/10.1080/02786826.2016.1162901, 2016b.
Sousan, S., Gray, A., Zuidema, C., Stebounova, L., Thomas, G.,
Koehler, K., and Peters, T.: Sensor selection to improve estimates
of particulate matter concentration from a low-cost network, Sen-
sors, 18, 3008, https://doi.org/10.3390/s18093008, 2018.
Tanzer, R., Malings, C., Hauryliuk, A., Subramanian, R.,
and Presto, A. A.: Demonstration of a low-cost multi-
pollutant network to quantify intra-urban spatial variations
in air pollutant source impacts and to evaluate environ-
mental justice, Int. J. Environ. Res. Public He., 16, 2523,
https://doi.org/10.3390/ijerph16142523, 2019.
Tasic, V., Jovasevic-Stojanovic, M., Vardoulakis, S., Milosevic,
N. Kovacevic, R., and Petrovic, J.: Comparative assessment of
a real-time particle monitor against the reference gravimetric
method for PM10 and PM2.5in indoor air, Atmos. Environ., 54,
358–364, https://doi.org/10.1016/j.atmosenv.2012.02.030, 2012.
Tiszenkel, L., Stangl, C., Krasnomowitz, J., Ouyang, Q., Yu, H.,
Apsokardu, M. J., Johnston, M. V., and Lee, S.-H.: Temperature
effects on sulfuric acid aerosol nucleation and growth: initial re-
sults from the TANGENT study, Atmos. Chem. Phys., 19, 8915–
8929, https://doi.org/10.5194/acp-19-8915-2019, 2019.
Tryner, J., L’Orange, C., Mehaffy, J., Miller-Lionberg, D., Hofstet-
ter, J. C., Wilson, A., and Volckens, J.: Laboratory evaluation of
low-cost PurpleAir PM monitors and in-field correction using co-
located portable filter samplers, Atmos. Environ., 220, 117067,
https://doi.org/10.1016/j.atmosenv.2019.117067, 2020.
U.S. EPA: AirNow, available at: https://www.airnow.gov/?city=
Sarajevo&country=BIH, last access: 26 November 2020.
Walser, A., Sauer, D., Spanu, A., Gasteiger, J., and Weinzierl,
B.: On the parametrization of optical particle counter response
including instrument-induced broadening of size spectra and
a self-consistent evaluation of calibration measurements, At-
mos. Meas. Tech., 10, 4341–4361, https://doi.org/10.5194/amt-
10-4341-2017, 2017.
Wang, Y., Li, J., Jing, H., Zhang, Q., Jiang, J., and Biswas, P.: Lab-
oratory evaluation and calibration of three low-cost particle sen-
sors for particulate matter measurement, Aerosol Sci. Tech., 49,
1063–1077, https://doi.org/10.1080/02786826.2015.1100710,
2015.
Zhang, J., Marto, J. P., and Schwab, J. J.: Exploring the applica-
bility and limitations of selected optical scattering instruments
for PM mass measurement, Atmos. Meas. Tech., 11, 2995–3005,
https://doi.org/10.5194/amt-11-2995-2018, 2018.
Zhao, A., Bollasina, M. A., Crippa, M., and Stevenson, D. S.: Sig-
nificant climate impacts of aerosol changes driven by growth in
energy use and advances in emission control technology, Atmos.
Chem. Phys., 19, 14517–14533, https://doi.org/10.5194/acp-19-
14517-2019, 2019.
Zheng, T., Bergin, M. H., Johnson, K. K., Tripathi, S. N., Shirod-
kar, S., Landis, M. S., Sutaria, R., and Carlson, D. E.: Field
evaluation of low-cost particulate matter sensors in high- and
low-concentration environments, Atmos. Meas. Tech., 11, 4823–
4846, https://doi.org/10.5194/amt-11-4823-2018, 2018.
Zheng, T., Bergin, M. H., Sutaria, R., Tripathi, S. N., Caldow, R.,
and Carlson, D. E.: Gaussian process regression model for dy-
namically calibrating and surveilling a wireless low-cost partic-
ulate matter sensor network in Delhi, Atmos. Meas. Tech., 12,
5161–5181, https://doi.org/10.5194/amt-12-5161-2019, 2019.
https://doi.org/10.5194/amt-13-6427-2020 Atmos. Meas. Tech., 13, 6427–6443, 2020
Content uploaded by Adnan Masic
Author content
All content in this area was uploaded by Adnan Masic on Nov 30, 2020
Content may be subject to copyright.
