Science of the Total Environment 715 (2020) 136258

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Identifying the wintertime sources of volatile organic compounds (VOCs)

from MAX-DOAS measured formaldehyde and glyoxal in Chongqing,
southwest China
Chengzhi Xing a,f,1, Cheng Liu a,b,c,d,e,⁎,1, Qihou Hu a, Qingyan Fu i, Hua Lin g, Shuntian Wang a, Wenjing Su f,
Weiwei Wang h, Zeeshan Javed j, Jianguo Liu a,c
a
Key Lab of Environmental Optics & Technology, Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China
b
Department of Precision Machinery and Precision Instrumentation, University of Science and Technology of China, Hefei 230026, China
c
Center for Excellence in Regional Atmospheric Environment, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China
d
Key Laboratory of Precision Scientific Instrumentation of Anhui Higher Education Institutes, University of Science and Technology of China, Hefei 230026, China
e
Anhui Province Key Laboratory of Polar Environment and Global Change, USTC, Hefei 230026, China
f
School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, China
g
School of Environmental Science and Optoelectronic Technology, University of Science and Technology of China, Hefei 230026, China
h
Chongqing Key Laboratory of Catalysis and New Environmental Materials, College of Environment and Resources, Chongqing Technology and Business University, Chongqing 400067, China
i
Shanghai Environmental Monitoring Center, Shanghai 200235, China
j
School of Environmental Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• The ratio of PM2.5/PM10 decreased when

PM2.5 pollution became worse
• HCHO and CHCOHO mainly distributed
in 0-100 m and 100-200 m, respectively.
• The sources of VOCs above 100 m are
mainly from biomass burning.

a r t i c l e i n f o a b s t r a c t

Article history: Ground-based multi-axis differential optical absorption spectroscopy (MAX-DOAS) observations were per-
Received 2 November 2019 formed from 27 December 2018 to 16 January 2019 in Changshou, one of subdistricts of Chongqing, China. Pri-
Received in revised form 19 December 2019 mary atmospheric pollutant in Changshou during wintertime was PM2.5, whose contribution averaged about
Accepted 19 December 2019
70.15% ± 9.5% of PM10. The ratio of PM2.5/PM10 decreased when PM2.5 pollution became worse, and it should at-
Available online 27 December 2019
tribute to biomass burning and the contribution of hygroscopic growth and enhanced heterogeneous chemistry
Editor: Pingqing Fu under high relative humidity condition. Moreover, nitrogen dioxide (NO2), formaldehyde (HCHO) and glyoxal
(CHOCHO) vertical profiles during the campaign period were retrieved separately. TROPOMI HCHO vertical col-
Keywords: umn densities (VCDs) and MAX-DOAS HCHO VCDs were correlated well (R = 0.93). In order to identify the
MAX-DOAS sources of volatile organic compound (VOC) in Changshou, the ratio of CHOCHO to HCHO (RGF) in five different

⁎ Corresponding author at: Department of Precision Machinery and Precision Instrumentation, University of Science and Technology of China, Hefei 230026, China.
E-mail address: [email protected] (C. Liu).
1
These authors contributed equally to this work.

https://doi.org/10.1016/j.scitotenv.2019.136258
0048-9697/© 2020 Elsevier B.V. All rights reserved.
2 C. Xing et al. / Science of the Total Environment 715 (2020) 136258

Formaldehyde layers were estimated. The estimated daily averaged RGF were 0.0205 ± 0.0077, 0.0727 ± 0.0286, 0.0864 ±
Glyoxal 0.0296, 0.0770 ± 0.0275 and 0.0746 ± 0.0263 in 0–100 m, 100–200 m, 300–400 m, 500–600 m and
Volatile organic compounds (VOCs) 700–800 m layers, respectively. The estimated RGF will increase when biomass burnings were dominated.
Biomass burning
Using NO2 as a tracer of anthropogenic emissions, we found the RGF values gradually decrease with the increase
of NO2 levels. RGF values in 0–100 m layer and all the other upper layers are 0.015–0.025 and 0.06–0.14, and that
means the dominant sources of VOCs in 0–100 m layer and all the other upper layers are biogenic emission and
anthropogenic emission (especially biomass burning), respectively. In addition, we found that RGF has site depen-
dence which is in compliance with several previous studies.
© 2020 Elsevier B.V. All rights reserved.

1. Introduction measuring atmospheric aerosol, NO2, HCHO and CHOCHO in the past
decades (Hendrick et al., 2014; Chan et al., 2015; Chan et al., 2017a;
In the atmosphere, volatile organic compounds (VOCs) mainly Chan et al., 2017b; Xing et al., 2017; Chan et al., 2019; Xing et al.,
source from biogenic, anthropogenic and biomass burning. 2019). MAX-DOAS measurements can also be used to validate satellite
They play significant roles in the formation of tropospheric ozone observations of atmospheric trace gases, i.e. HCHO (Irie et al., 2008;
and secondary organic aerosols (Houweling et al., 1998; Poisson et al., Tan et al., 2018; Zhang et al., 2019). In the future, portable, sensitive
2000; Kanakidou et al., 2005; Volkamer et al., 2006; Johnson et al., and machine-learning-based might be the new development directions
2006; Shao et al., 2009). High mixing ratios of ambient VOCs in Sichuan of MAX-DOAS (Wu et al., 2017; Su, 2018).
Basin were measured (Barletta et al., 2005), and the biomass burning In this paper, it is the first time to measure the vertical distributions
and transportation contribute N50% and ~25–30% of the total VOC emis- of HCHO and CHOCHO using MAX-DOAS in Chongqing, southwest
sions in this area (Bo et al., 2008; Zhang et al., 2009), respectively. Wood China. Several previous studies have successfully used the ratio of the
and crop burning were the primary contributors to the biomass burning ground surface concentration of CHOCHO to HCHO (RGF) to classify
in Sichuan Basin. To better understand the characteristics of VOCs, it is the VOC dominant-source types. In order to identify the sources of
helpful to improve our knowledge of their sources and removal path- VOCs at different heights, we will analyze the distributions of
ways. Formaldehyde (HCHO) and glyoxal (CHOCHO) are two valuable CHOCHO, HCHO and their ratios RGF in different layer heights in this
products from the oxidation of VOCs, and they could act tracers for the study.
VOCs sources.
In the atmosphere, HCHO is the most abundant aldehyde, and 2. Experimental
CHOCHO is one of the most abundant dihydroxy compounds. HCHO
has some primary sources, i.e. fossil fuel combustion, biomass burning, 2.1. Measurement site description
traffic and industrial activities (Lee et al., 1995; Holzinger et al., 1999;
Yokelson et al., 1999; Andreae and Merlet, 2001; Geiger et al., 2002), The field campaign was carried out from 27 December 2018 to 16
while CHOCHO is mainly produced from the oxidation of biogenic and January 2019 in Changshou (107.01°E, 29.83°N), Southwest China, and
anthropogenic VOCs (Fu et al., 2008; Myriokefalitakis et al., 2008). For it focused on exploring the source of wintertime pollutants in this
their secondary sources, isoprene is the major pathway producing area. Changshou is one of the subdistricts of Chongqing, which is located
HCHO and CHOCHO (around 50%), and HCHO is formed from the first in the northeast of Chongqing city center (approximately 50 km)
generation of isoprene oxidation whereas CHOCHO is produced in the (Fig. 1). The MAX-DOAS measurements were operated at the rural
third and fourth generation (Li et al., 2013). In addition, HCHO and part of Changshou, and the measurement site is surrounded by small
CHOCHO are also produced from the oxidation of alkenes and aromatics, forests while the local traffic was less during the campaign. In addition,
respectively. The removal processes of HCHO and CHOCHO are mainly there is a pipeline production factory in the north of the measurement
photolysis, oxidation by hydroxyl radical (OH) and wet deposition site (about 0.8 km). Outdoor biomass burning was occasionally ob-
(Arlander et al., 1995; Myriokefalitakis et al., 2008). Considering their served in the surrounding areas during the campaign, which is a tradi-
similar removal mechanisms and different sources, the ratio of tion that residents living in this area engaged in outdoor meat
CHOCHO to HCHO concentrations (RGF) is used to identify and classify smoking activities to preserve meat before the Chinese Spring Festival
biogenic and anthropogenic emissions of VOCs in different urban and (Chen et al., 2017a, 2017b). The campaign site was characterized by
rural sites. In previous studies, Vrekoussis et al. (2009, 2010) used satel- high relative humidity (mostly N80%) and low wind speed (mostly
lite measured VCDs of HCHO and CHOCHO to analyze RGF on a global b2 m/s) during the measurement period. Such meteorological condition
scale; Irie et al. (2011) and Ortega et al. (2015) analyzed RGF using is conducive to the formation and accumulation of air pollutants. More-
MAX-DOAS measurements; and Munger et al. (1995), Grosjean et al. over, the radiation is weak and the photochemical reaction might not be
(1996) and DiGangi et al. (2012) reported RGF from in situ measure- so active during the campaign, due to the large volume of cloud cover in
ments. They all found that the increased RGF values related to the emis- this region.
sions of VOCs from anthropogenic and biomass burning, while the low
RGF values usually occurred under the condition of biogenic emissions 2.2. MAX-DOAS
of VOCs. Moreover, some differences between RGF values in different
studies might attribute to altitude dependence of RGF (Hoque et al., A commercial MAX-DOAS instrument (Airyx, Heidelberg, Germany)
2018). was used during the campaign. The MAX-DOAS instrument was
Multi-Axis Differential Optical Absorption Spectroscopy (MAX- installed 4 m higher from the ground surface, with an azimuth angle
DOAS) as a passive remote sensing technique measures scattered sun- of 38° (North = 0°). It consists of three major parts: two spectrometers
light by changing the elevation angle of telescope and provides column (AvaSpec-ULS2048L-USB2, UV range: 296–408 nm, visible range:
densities of aerosols and trace gases (Platt and Stutz, 2008). In addition, 420–565 nm) at a fixed temperature of 20 °C with a deviation of
combined with forward radiative transfer model, vertical profiles of b0.05 °C, and the spectral resolutions are both 0.45 nm; a telescope
aerosol and trace gases could be retrieved (Hönninger and Platt, 2002; unit and a controlling computer. The telescope collected scattered sun-
Bobrowski et al., 2003; Hönninger et al., 2004; Wagner et al., 2004; light and then directs it to the spectrometer through a prism reflector
Wittrock et al., 2004). This method has been universally applied for and quartz fibres. Elevation angle (α) accuracy is b0.1° and the
 C. Xing et al. / Science of the Total Environment 715 (2020) 136258 3

Fig. 1. Map of Changshou in Sichuan Basin and the distribution of fire point from 27 December 2018 to 16 January 2019.

telescope field of view (opening angle) is b0.3°. Moreover, a charge- oxygen dimer (O4) and NO2 were analyzed in the interval between
coupled device detector camera (Sony ILX511 with 2048 individual 338 and 370 nm, and 336.5–359 nm wavelength interval was used for
pixels) is equipped to the optical spectrometer. In this campaign, a full HCHO absorption analysis, based on recommended settings from the
measurement sequence was set to 11 elevation angles, i.e., 1, 2, 3, 4, 5, CINDI intercomparison campaign (Roscoe et al., 2010; Kreher et al.,
6, 8, 10, 15, 30 and 90°, and one sequence took about 11 min. During 2019). Moreover, CHOCHO DSCDs were analyzed in the wavelength in-
daytime, the instrument collected scattered sunlight. During night, terval between 431 and 460 nm with lower root mean square of resid-
dark current and offset spectra were measured automatically through ual and error (Javed et al., 2019). Detailed DOAS fit settings for O4, NO2,
a script, and the dark current and offset will be removed from all of HCHO and CHOCHO have been listed in Table 1. The corrected I0 was
the measured spectrum. In addition, only spectra measured with solar used (Aliwell et al., 2002). Typical DOAS fit results of above four species
zenith angle (SZA) less than 75° were taken for analysis. are shown in Fig. 2. DOAS fit results were filtered out with a root mean
The QDOAS software developed by BIRA-IASB (http://uv-vis. square (rms) larger than 6.0 × 10−4 and 5.0 × 10−4 for (O4, NO2, HCHO)
aeronomie.be/software/QDOAS/, last access: 22 August 2019) was and CHOCHO, respectively. Afterwards, the corresponding qualified
used to analyze the filtered spectra. Differential slant column densities DSCD results remained 98.42%, 99.01%, 95.33% and 90.71%, respectively.
(DSCDs) will be firstly accepted, which is defined as the difference be- In order to remove the effects of cloud, we calculated the color index
tween the off-zenith and zenith slant column densities (SCDs) (CI), which is a ratio of solar spectral intensities at 330 nm to that in
(Wagner et al., 2011). We calculated the Ring spectrum for measured 390 nm (Wagner et al., 2016). Diurnal CI thresholds for each time in
spectrum, given the contribution of stratosphere to DSCDs. DSCDs of this campaign were obtained through fitting a fifth-order threshold

Table 1
DOAS retrieval settings used for O4, NO2, HCHO and CHOCHO DOAS spectral analysis.

Parameter Data source Fitting internal (nm)

O4 NO2 HCHO Glyoxal

Wavelength range 338–370 338–370 322.5–358 431–460

NO2 298 K, I0 correction (SCD of 1017 molecules cm−2); Vandaele et al. (1998) ✓ ✓ ✓ ✓
NO2 220 K, I0 correction (SCD of 1017 molecules cm−2), pre-orthogonalized; Vandaele et al. (1998) ✓ ✓ ✗ ✓
20 −2
O3 223 K, I0 correction (SCD of 10 molecules cm ); Serdyuchenko et al. (2014) ✓ ✓ ✓ ✓
20 −2
O3 243 K, I0 correction (SCD of 10 molecules cm ), pre-orthogonalized; Serdyuchenko et al. (2014) ✓ ✓ ✓ ✗
O4 293 K; Thalman and Volkamer (2013) ✓ ✓ ✓ ✓
BrO 223 K; Fleischmann et al. (2004) ✓ ✓ ✓ ✗
H2O 296 K, HITEMP; Rothman et al. (2009) ✗ ✗ ✗ ✓
HCHO 297 K; Meller and Moortgat (2000) ✓ ✓ ✓ ✗
Glyoxal 296 K, Volkamer et al. (2005) ✗ ✗ ✗ ✓
Ring Ring spectra calculated with QDOAS according to Chance and Spurr (1997) ✓ ✓ ✓ ✓
Polynomial degree Order 3 Order 3 Order 5 Order 5
Intensity offset Constant Constant Constant Constant
Wavelength calibration Based on a high-resolution solar reference spectrum (SAO2010 solar spectra); Chance and Kurucz (2010)
4 C. Xing et al. / Science of the Total Environment 715 (2020) 136258

Fig. 2. Example of DOAS fit results for (a) O4, (b) NO2, (c) HCHO and (d) CHOCHO. Red line and black line represent the fitted reference spectrum and the absorption of the target species,
respectively.

polynomial to CI data which is a function of time. When CI was b10% of monitoring station (29.516° N, 106.512° E), one of the stations of the na-
above threshold, the data was filtered out (Ryan et al., 2018). tional environmental monitoring network (http://beijingair.sinaapp.
Vertical profiles were retrieved from HEIPRO algorithm (Heidelberg com/, last access: 22 August 2019), located about ~5 km from our cam-
Profile, developed by IUP Heidelberg) (Frieß et al., 2006; Frieß et al., paign site. Fire point data was collected from satellite NPP-VIIRS.
2011). The inversion algorithm is based on the optimal estimation
method (Rodgers, 1990, 2000) and selected the radiative transfer 3. Results and discussion
model (RTM) SCIATRAN as the forward model. The cost function χ2 as
following is selecting to determine the maximum a posteriori state vec- 3.1. Overview of the observation
tor x.
The time series of aerosol extinction, PM2.5, PM10, NO2, CO, HCHO,
T T −1
χ2 ¼ ðy−F ðx; bÞÞ S −1
ε ðy− F ðx; bÞÞ þ ðx−xa Þ S a ðx−xa Þ ð1Þ CHOCHO and the corresponding meteorological parameters,
i.e., temperature, RH, wind speed and wind direction, from 27 December
where F(x,b) is the forward model. b represents meteorological param- 2018 to 16 January 2019 are presented in Fig. 3. Primary atmospheric
eters, i.e. atmospheric pressure and temperature vertical profiles. y rep- pollutant in Chongqing during the wintertime of 2018–2019 was
resents the measured DSCDs and xa is the a priori information. Sε and Sa PM2.5, which contributed averaged about 70.15% ± 9.5% of PM10. Aver-
are covariance matrices of y and xa, respectively. Two steps were used to aged PM2.5 concentration reached 58.12 ± 25.52 μg m−3 and peaked at
retrieve the vertical gradient of aerosol and trace gas. In the first step, 140.0 μg m−3 during the observation period. Similar conclusions were
aerosol vertical profiles were retrieved using measured O4 DSCDs of dif- also reported in previous studies (Chen et al., 2017a, 2017b; Tao et al.,
ferent elevation angles. In the second step, retrieved aerosol extinction 2017). Moreover, the evolution trend of aerosol extinction is similar to
profiles will play an input parameter to the RTM to retrieve NO2, PM2.5, except for 29 December 2018. It indicates PM2.5 is the main con-
HCHO and CHOCHO profiles. For this retrieval, the atmosphere was sep- tributor of aerosol extinction. On 29 December 2018, a fog day identified
arated into 30 layers from 0 to 4 km with a vertical resolution of 0.1 km using RH and visibility (Xing et al., 2019) (Fig. S3), the averaged RH
under 2 km and a vertical resolution of 0.2 km from 2 to 4 km. An expo- reached to 90.67% which means the high aerosol extinction might be re-
nential decreasing a priori with scale height of 1 km was selected for lated to the high concentration of water vapour. In addition, character-
both aerosol and trace gas profile retrieval. The a priori surface aerosol istic ranges for concentration were found to be 0.82 to 23.86 ppb and
extinction was set to 0.2 km−1, and the a priori surface concentrations averaged at 4.33 ppb for NO2, 0.49 to 17.64 ppb and averaged at
of NO2, HCHO and CHOCHO were set to 5 ppb, 2 ppb 0.2 ppb, respec- 4.95 ppb for HCHO, 0.03 to 0.17 ppb and averaged at 0.08 ppb for
tively. A priori uncertainties of aerosol and trace gases were all set to CHOCHO, respectively.
100%, and the correlation height was set to 0.5 km. Moreover, fixed As indicated in the gray areas of Fig. 3, two episodes of particulate
aerosol optical properties (asymmetry = 0.69, single scattering al- pollution during 30 December 2018 to 08 January 2019 and 13 to 16
bedo = 0.9, and surface albedo = 0.05) were use during the retrieval. January 2019 were identified. During these two episodes, aerosol ex-
Example retrieval results and errors from HEIPRO were plotted in tinction and PM2.5 concentrations typically increased and remained
Fig. S2. at a high level for several days. Aerosol extinction and PM2.5 concen-
trations increased to exceed the maximum values 1.88 km −1 and
2.3. Ancillary data 114 μg cm−3 during episode I. In episode II, aerosol extinction and
PM2.5 concentrations increased to exceed the maximum values
Meteorological data (temperature, relative humidity (RH), local sur- 2.01 km−1 and 140 μg cm −3 , respectively. The concentrations of
face wind speed and wind direction with a temporal resolution of NO2 were also increased in these two episodes, however, concentra-
1 min) were measured by a grid monitoring sensor installed in the cam- tions of HCHO and CHOCHO were decreased. Before episode I, the
paign site. CO concentrations were obtained from Yangjiaping dominant direction of wind was northeast with an average speed
 C. Xing et al. / Science of the Total Environment 715 (2020) 136258 5

Fig. 3. Time series of aerosol extinction, PM10, PM2.5, NO2, CO, HCHO and CHOCHO, and meteorological data (temperature, relative humidity, wind speed and wind direction) from 27
December 2018 to 16 January 2019.

1.73 m/s. The low concentrations of PM2.5 should be attributed to the matter (PM), especially between CO and PM10 as shown in Fig. S1.
cleaning effect of the cleaner air from northeast during this process. The elevation of CO levels should attribute to biomass burning in
The quickly decrease of PM2.5 concentrations on 09 December 2018 Changshou during our observation period. That means primary PM
was mainly due to the rain cleaning. In addition, we found the ratio emitted from burning is one of the major reasons to cause the parti-
of PM2.5/PM10 decreased when PM2.5 pollution became worse. On cle size increase. On the other hand, hygroscopic growth and en-
the one hand, we found suddenly increase of CO concentration al- hanced heterogeneous chemistry are also important pathways of
ways corresponded to the low values of PM2.5 /PM10 (Fig. 3), and the increasing of particle size under high RH (average value was
there are positive correlations between CO and coarse particulate 84.74%), low wind speed (average value was 0.17 m/s) and favorable

Fig. 4. (a) shows the time series of tropospheric HCHO VCDs from MAX-DOAS and TROPOMI. MAX-DOAS data are used just averaged around the TROPOMI overpass time. (b) shows
correlation between MAX-DOAS measured daily averaged tropospheric HCHO VCDs and USTC TROPOMI satellite data. The TROPOMI measurements are spatially averaged over the
grid cells within 3.5 km of ground location around the campaign site.
6 C. Xing et al. / Science of the Total Environment 715 (2020) 136258

solar radiation (average value was 114.74 W/m2) conditions (Meier CHOCHO all occurred in early morning hours and afternoon hours
et al., 2009; Zheng et al., 2015). after 15:00, which is different from the observation in Phimai,
In Fig. 4, we made a validation between MAX-DOAS HCHO VCDs Thailand (Hoque et al., 2018), but similar with that in Pearl River
and USTC TROPOMI HCHO VCDs. USTC (University of Science and Delta (PRD) region, China (Li et al., 2013). The maximum values of
Technology of China) TROPOMI (TROPOspheric Monitoring Instru- HCHO and CHOCHO in 0–100 m layers exceed 0.61 and 0.11 ppb, re-
ment) HCHO VCDs were converted from HCHO SCDs measured by spectively. The production ways of HCHO and CHOCHO are mainly the
TROPOMI and the calculated air mass factor (AMF), where the AMF oxidation of different VOCs, and their removal processes are mainly
was simulated from SCIATRAN using retrieved aerosol and trace from photolysis, OH oxidation and uptake by aerosols at night
gas vertical profiles as input. MAX-DOAS HCHO VCDs were averaged (Volkamer et al., 2007). The early appearance of HCHO and CHOCHO
over the TROPOMI overpass time from 12:30 to 14:30 LST, while the could be attributed to the oxidation of VOCs by OH radicals in the previ-
TROPOMI HCHO data are spatially averaged over pixels within ous night and the rerelease from the aerosols. The peaks of HCHO and
3.5 km of the campaign site. The temporal trends of HCHO VCDs CHOCHO after 15:00 are due to photo-oxidation of VOCs, especially iso-
show similar characteristics between MAX-DOAS and USTC prene, alkenes and aromatics (Hendrick et al., 2014; Gratsea et al.,
TROPOMI data and give a high correlation coefficient (R) of 0.93. 2016). In addition, we found the higher concentration of CHOCHO oc-
However, the HCHO VCDs measured by TROPOMI satellite were curred in 100–400 m, especially in 100–200 m layer, but the higher con-
higher than that measured by MAX-DOAS on some days, which is dif- centration of HCHO mainly distributed in lower heights. The vertical
ferent from some previous studies in Beijing and Shanghai (Liu et al., distribution of trace gases depends on their production and destruction
2016; Xing et al., 2017), but similar with some ship-based MAX- processes. Referring to Li et al. (2010) and Wagner et al. (2011), the
DOAS measurements (Tan et al., 2018; Hong et al., 2018). mixing heights of aerosol and trace gases were retrieved using multi-
parameter look-up tables, which are created by SCIATRAN. Parameters
3.2. Diurnal cycle of HCHO and CHOCHO used for creating look-up table in our research have been listed in
Table S1. In Fig. 6, we found the daily averaged mixing layer heights of
The analysis of the diurnal cycles of HCHO and CHOCHO is vital for HCHO are lower than those of CHOCHO. HCHO having possible primary
understanding their sources and the related atmospheric chemistry. surface source could be one of reasons of CHOCHO layer being higher
Fig. 5 shows the mean diurnal variations of CHOCHO and HCHO in dif- than HCHO layer. Moreover, as discussed in Li et al. (2013), the HCHO
ferent layers (0–100 m, 100–200 m, 300–400 m, 500–600 m and production occurring earlier, i.e. at lower heights, than CHOCHO pro-
700–800 m) during the entire measurement period. CHOCHO has the duction during the degradation of isoprene and the shorter lifetimes
similar diurnal variation trend in different layers, but HCHO acts differ- of isoprene and alkenes than aromatics caused CHOCHO to appear at
ently. In 0–100 m layer, the maximum mixing ratios of HCHO and higher layers.

Fig. 5. The first column shows the hourly averaged diurnal variations of HCHO volume mixing ratios in 0–100 m, 100–200 m, 300–400 m, 500–600 m and 700–800 m layers, respectively.
The same as the first column, the second column shows the hourly averaged diurnal variations of CHOCHO volume mixing ratios. The third column shows the hourly averaged diurnal
variations of RGF. The fourth column shows the hourly averaged diurnal variations of NO2 volume mixing ratios. The fifth column shows the correlations between NO2 and HCHO in
different layers. The last column shows the correlations between NO2 and CHOCHO in different layers.
 C. Xing et al. / Science of the Total Environment 715 (2020) 136258 7

Table 2
Estimated errors (%) of retrieved CHOCHO and HCHO in different layers.

Layers Component Random error Systematic error Total error

0–100 m CHOCHO 8 13 17
HCHO 12 19 24
100–200 m CHOCHO 8 13 17
HCHO 13 19 25
300–400 m CHOCHO 9 15 19
HCHO 13 21 27
500–600 m CHOCHO 11 15 20
HCHO 15 23 29
700–800 m CHOCHO 12 17 22
HCHO 15 24 30

minimum values usually appeared between 12:00 and 13:00. The diur-
nal RGF in 100–200 m, 300–400 m, 500–600 m and 700–800 m layers
Fig. 6. Time series of averaged mixing layer heights of aerosol extinction, HCHO, CHOCHO varied between 0.0557 and 0.1024, 0.0579 and 0.1306, 0.0582 and
and NO2 from individual elevation sequences.
0.1224 and 0.0459 and 0.1213, respectively. The obvious difference of
RGF in 0–100 m layer and other layers indicates there are different for-
Peak values of HCHO occurred between 12:00 and 13:00 in mation pathways for HCHO and CHOCHO on the surface and upper
100–200 m, 300–400 m, 500–600 m and 700–800 m layers (1.53, layers in Changshou. In addition, we described the time series of day-
1.64, 1.67 and 1.70 ppb, respectively). The direct emission of local com- time RGF values during the entire observation period in Fig. 7. We
bustion events is also one source of HCHO and the age of plume can also found that there is a consistent decrease for RGF in all layers during
impact the vertical distribution of HCHO. In Fig. 3, the averaged concen- 5–7 and 13–16 January 2019, which corresponds to above two episodes.
tration of CO after 10 January increased 45.45% than before. Moreover, The values of RGF on above 100 m on 08 January 2019 are obviously
according to the distribution of fire points shown in Fig. 1, combustion higher than that on other days. As shown in Fig. 3, we found the concen-
events are frequent during the observation time, especially after 10 Jan- tration of NO2 and CO on 08 January 2019 is significantly higher than
uary 2019. For example, two fire points with fire radiative power (FRP) that on other days. That means the biomass burning activities were
33.8 and 32.1 mW/m2 were detected during the noon time within 3 km more frequent on this day. CHOCHO maybe increase with the photolysis
of the observation site, and the detailed FRP information in and around of VOCs released from the frequent biomass burning plume, and that
the observation area during the measurement was listed in Table S2. It will cause RGF values increased on above 100 m.
possibly can be the outdoor meat smoking activities from nearby resi- Values of RGF could be different when the VOCs composition
dents according to previous study (Chen et al., 2017a, 2017b) and our changes. Certainly, RGF is determined by several factors instead of a sin-
observation record. With the occurrence of biomass burning, a large gle parameter, i.e. composition of VOCs, OH radicals and the radiative
amount of VOCs was released and the VOCs will be photolyzed with frequencies etc. Vrekoussis et al. (2010) reported that the dominant
the rise of the plume. The secondary HCHO maybe increase with the VOC sources are biogenic emissions when RGF is in the range of
photolysis of VOCs in above process and caused the peak values of 0.04–0.06, and RGF values b 0.03 indicate elevated anthropogenic VOC
HCHO in higher altitudes. Of course, this process depends on the types emissions based on GOME-2 observations. Schauer et al. (2001, 2002)
of released VOCs. Moreover, we can't exclude the possibility of short- found RGF N 0.5when biomass burning occurred, and RGF are 0.02–0.09
distance transmission of HCHO at above 100 m. in vehicle exhausts. DiGangi et al. (2012) suggest VOCs are under bio-
genic influence when RGF b 0.020, and VOCs are under anthropogenic
3.3. Estimation of the RGF influence when RGF N 0.025. Miller et al. (2014) told us VOCs are influ-
enced by anthropogenic if RGF is ~0.04, and VOCs are influenced by bio-
The precursor VOC structure and the concentration of NOx deter- genic emission if RGF is not equal to 0.04 (b0.04, isoprene controlling;
mine the yield of HCHO and CHOCHO from oxidation of VOCs. In previ- N0.04, monoterpenes controlling). We found that the conclusions
ous studies, the ratio between CHOCHO and HCHO (RGF) has been used based on RGF measured in different locations are different in previous
to identify the sources of VOCs due to CHOCHO and HCHO having differ- studies. Kaiser et al. (2015) also reported that there is an obvious differ-
ent formation pathways (Vrekoussis et al., 2010). The RGF depends on ence between satellite and in situ RGF observations. Compared with
the contribution of different sources to HCHO and CHOCHO. The esti- other rural observations, RGF in Changshou is not only strongly influ-
mated daily mean RGF in 0–100 m, 100–200 m, 300–400 m, enced by biogenic VOC emissions but also influenced by frequent bio-
500–600 m and 700–800 m layers in Changshou are 0.0205 ± 0.0077, mass burning due to the outdoor meat smoking activities near the
0.0727 ± 0.0286, 0.0864 ± 0.0296, 0.0770 ± 0.0275 and 0.0746 ± Spring Festival. In our study, values of RGF were N0.045 and N0.1 in
0.0263, respectively. The calculation of uncertainties of RGF was referred 0–100 m layer and other upper layers when biomass burning occurred
to Hoque et al. (2018). The uncertainties of RGF were estimated from the (referring to FRP data). That means anthropogenic emissions have obvi-
random and systematic errors of HCHO and CHOCHO, and the random ous influence on RGF, and that will be detailed discussed in Section 3.4.
and systematic errors of HCHO and CHOCHO in different layers have
been listed in Table 2. Previous studies were mostly concentrated on 3.4. Anthropogenic influence on the RGF
the investigation of RGF on the ground surface. The overview of winter-
time HCHO and CHOCHO concentrations and the corresponding RGF Fu et al. (2008) pointed out that CHOCHO produced from the oxida-
values in different sites all over the world were provided in Table 3. tion of isoprene or terpenes are much lower than that from anthropo-
The estimated RGF values in 0–100 layer in our study were within the genic VOCs, i.e. aromatics, oxidations. That means the value of RGF will
range of that reported in previous studies. increase under the condition of anthropogenic VOCs emissions, and
The averaged diurnal cycles of RGF in different layers was shown in the VOCs composition will change from isoprene dominant to aromatics
Fig. 5. The diurnal RGF varied between 0.0175 and 0.0242 and peaks at domination (Li et al., 2013). On the other hand, Lei et al. (2009) suggest
12:00 in 0–100 m layer. However, the maximum values of RGF in anthropogenic source could also cause RGF decrease, especially consid-
other layers always occurred in the early morning (~09:00), and the ering the increase of the primary emission of HCHO, i.e. industrial
8 C. Xing et al. / Science of the Total Environment 715 (2020) 136258

Table 3
Overview of field campaign wintertime average values of HCHO and CHOCHO and RGF in different locations all over the world.

Location Site HCHO (ppb) CHOCHO (ppb) RGF Reference

Las Vegas, USA Urban 0.79 0.14 0.177 Jing et al. (2001)
Elizabeth, New Jersey, USA Urban 5.67 0.45 0.079 Liu et al. (2006)
Phimai, Thailand Rural 4.96 0.13 0.026 Hoque et al. (2018)
Pantnagar, India Rural 3.90 0.11 0.029 Hoque et al. (2018)

activities. In order to further understand the impact of anthropogenic ~14.001 ppb, ~14.624 ppb and ~16.129 ppb in 100–200 m,
emissions on RGF, NO2 was chosen as an anthropogenic biomass burning 300–400 m, 500–600 m and 700–800 m layers, respectively. Moreover,
tracer due to its source being mainly from combustion of fossil fuels (Lee the averaged vertical column density of NO2 and RGF value are 2.18 × 10-
et al., 1998). Fig. 5 depicts the averaged diurnal variations of NO2 and 16
molec cm−2 and 0.022 in the surface layer during the observation, re-
the correlations of NO2 with HCHO and CHOCHO in different layers. spectively. Referred to the statistical results of NO2 and RGF in locations
We found NO2 concentrations usually peaks before 09:00 and after around the world investigated by Vrekoussis et al. (2010), we found
15:00, except for 500–600 m layer. The NO2 concentration distributed data measured in Changshou could fit well with the corresponding re-
in 100–200 m layer is much higher than that in other layers, and the sults in other locations (Table S3 and Fig. S4). That means the change
vertical distribution of NO2 is similar to CHCOHO. The hourly averaged of NO2 levels indeed have a potential impact on RGF values.
concentrations of HCHO and CHOCHO during the observation period As described above, the outdoor meat smoking activities are fre-
were relatively highly correlated with the hourly averaged NO2 concen- quent during the observation period in Changshou, and it could influ-
trations (R = 0.74 and 0.56, respectively), but not in other layers. ence the values of RGF in different layers. Due to the difference of the
Vrekoussis et al. (2010) clarified there will be a systematic impact on types of biomass burned and the age of plume, RGF values have site de-
RGF values when NO2 concentration changes. To further understand pendence. Hoque et al. (2018) reported lower values but DiGangi et al.
the relationship between NO2 concentration and RGF value, Fig. 8 illus- (2012) and Zarzana et al. (2017) reported higher values when fire
trates the RGF ratio as a function of various NO2 levels in different layers. events happened. Similar with our study, Baidar et al. (2013) also
The average and median RGF values were all depicted under four given found higher RGF values at a higher altitude. In general, RGF values are
NO2 ranges. We found the RGF values gradually decrease as the levels mainly 0.015–0.025 in 0–100 m layer, and these values are mainly
of NO2 increase, except for in 100–200 m layer. In this layer, the average 0.06–0.14 on above 100 m in this study. That means the dominant
RGF value under the condition of NO2 being 5–10 ppb increases about VOCs sources were biogenic emissions in 0–100 m, and the VOCs were
25.14% than that when NO2 level was in 0–5 ppb. That is because the in- mostly influenced by anthropogenic emissions, especially biomass
crease of the primary emission of HCHO from the pipeline production burning, on above 100 m.
factory will lead to the decrease of RGF, which will be more obvious
under lower NO2 conditions. In this rural site, the average RGF values 4. Summary
exceeded 0.023, 0.069, 0.108, 0.092 and 0.095 in 0–100 m,
100–200 m, 300–400 m, 500–600 m and 700–800 m layers under low We present measurements of NO2, HCHO and CHOCHO vertical pro-
NO2 concentration condition (b5 ppb), respectively. However, they files using MAX-DOAS in Changshou from 27 December 2018 to 16 Jan-
dropped below the limit of 0.003, 0.071, 0.046, 0.062 and 0.055 in uary 2019. PM2.5 contributed averaged about 70.15 ± 9.50% to PM10,
above five layers when NO2 concentrations increased to 10–20 ppb. and the ratio of PM2.5/PM10 decreased when PM2.5 pollution became
The decrease of RGF should be attributed to the small impact from an- worse during the observation period. That should be attributed to bio-
thropogenic sources of HCHO besides the secondary sources of HCHO mass burning, hygroscopic growth on PM and enhanced heterogeneous
and CHOCHO (Vrekoussis et al., 2010). In addition, we also referred to chemistry. High concentration of NO2 usually appeared with high PM2.5,
the approach described in Hoque et al. (2018). As shown in Fig. 9, con- while high concentration of HCHO and CHOCHO occurred with low
sidering extreme cases of NO2 levels, we could get the corresponding PM2.5. Moreover, we validated MAX-DOAS HCHO VCDs using USTC
RGF values in all five layers. When NO2 was at a low level of TROPOMI HCHO VCDs, and they correlated well (R = 0.93).
~0.815 ppb, the RGF was 0.023 in 0–100 m layer, and the RGF value From the diurnal cycles of HCHO and CHOCHO, we found that the
was 0.0129 when NO2 reached a relatively high level ~14.100 ppb in HCHO is mainly distributed in 0–100 m layer, while CHOCHO is mainly
this layer. Similarly, the RGF values are 0.072, 0.088, 0.153 and 0.083 distributed in 100–200 m layer. The estimated daily mean RGF are
when NO2 was at low levels of ~1.614 ppb, ~1.481 ppb, ~0.950 ppb 0.0205 ± 0.0077, 0.0727 ± 0.0286, 0.0864 ± 0.0296, 0.0770 ± 0.0275
and ~1.167 ppb in 100–200 m, 300–400 m, 500–600 m and and 0.0746 ± 0.0263 in 0–100 m, 100–200 m, 300–400 m,
700–800 m layers, respectively. The RGF values decrease to 0.046, 500–600 m and 700–800 m layers in Changshou, respectively. RGF
0.053, 0.071 and 0.058 when NO2 reached to ~28.098 ppb, were N0.045 and N0.1 in 0–100 m layer and the upper layers when

Fig. 7. Time series of RGF in five different layers from 27 December 2018 to 16 January 2019.
 C. Xing et al. / Science of the Total Environment 715 (2020) 136258 9

Fig. 8. Statistical information of average and median RGF ratios as a function of NO2 levels.

outdoor meat smoking activities were frequent. Using NO2 as an anthro- biomass burning happened, due to the difference of the plume age and
pogenic biomass burning tracer, we found RGF values mostly gradually types of biomass burned.
decrease as the levels of NO2 increase. The average RGF values exceeded
0.023, 0.069, 0.108, 0.092 and 0.095 under low NO2 condition (b5 ppb) Author contributions
in 0–100 m, 100–200 m, 300–400 m, 500–600 m, 700–800 m layers, re-
spectively. However, they dropped below the limit of 0.003, 0.071, CX and CL designed and implemented the research, as well as pre-
0.046, 0.062 and 0.055 when NO2 concentrations increased to pared the paper. QH contributed to HCHO and CHOCHO data analysis.
10–20 ppb in above five layers. In general, RGF values in 0–100 m layer CX, HL and QF carried out the MAX-DOAS observations at Changshou
and all the other upper layers are 0.015–0.025 and 0.06–0.14, and that site. WW contributed to the analysis of meteorological data. SW re-
means the dominant sources of VOCs in 0–100 m layer and all the trieved the FRP data and information of fire spots. WS contributed to re-
other upper layers are biogenic emission and anthropogenic emission trieve TROPOMI HCHO vertical density columns. ZJ helped to improve
(especially biomass burning), respectively. In compliance with the pre- the English language. JL provided constructive comments on this
vious studies, we found that the RGF values have site dependence when research.

Fig. 9. RGF values under extreme cases of NO2 in different layers. The extreme case of NO2 means the lowest and highest NO2 levels after which filtered by the relative error (N50%) and the
degree of freedom of signal (b1) during retrieval processes and the cloud information calculated using color index (CI) method. The vertical bars represent errors of RGF which were 2σ
standard errors.
10 C. Xing et al. / Science of the Total Environment 715 (2020) 136258

Declaration of competing interest Nanjing and a comparison to ozone monitoring instrument observations. Atmos.
Chem. Phys. 9, 10051–10071. https://doi.org/10.5194/acp-19-10051-2019.
Chance, K.V., Spurr, R.J, 1997. D.: Ring effect studies: Rayleigh scattering, including molec-
The authors declare that they have no known competing financial ular parameters for rotational Raman scattering, and the Fraunhofer spectrum. Appl.
interests or personal relationships that could have appeared to influ- Opt. 36, 5224–5230. https://doi.org/10.1364/AO.36.005224.
Chance, K., Kurucz, R., 2010. An improved high-resolution solar reference spectrum for
ence the work reported in this paper. earth’s atmosphere measurements in the ultraviolet, visible, and near infrared.
J. Quant. Spectrosc. Ra. 111, 1289–1295.
Acknowledgements Chen, Y., Wenger, J.C., Yang, F., Cao, J., Huang, R., Shi, G., et al., 2017a. Source characteriza-
tion of urban particles from meat smoking activities in Chongqing, China using single
particle aerosol mass spectrometry. Environ. Pollut. 228, 92–101.
This research was supported by grants from the National Key Re- Chen, Y., Xie, S., Luo, B., Zhai, C., 2017b. Particulate pollution in urban Chongqing of south-
search and Development Program of China (No. 2017YFC0210002, west China: historical trends of variation, chemical characteristics and source appor-
tionment. Sci. Total Environ. 584–585, 523–534. https://doi.org/10.1016/j.
2018YFC0213104 and 2016YFC0203302), Anhui Science and Technol-
scitotenv.2017.01.060.
ogy Department Major Project (18030801111), National Natural Sci- DiGangi, J., Henry, S., Kammrath, A., Boyle, E., Kaser, L., Schnitzhofer, R., et al., 2012. Obser-
ence Foundation of China (No. 41722501, 91544212, 51778596, vations of glyoxal and formaldehyde as metrics for the anthropogenic impact on rural
41575021 and 41977184), the Strategic Priority Research Program of photochemistry. Atmos. Chem. Phys. 12 (20), 9529–9543. https://doi.org/10.5194/
acp-12-9529-2012.
the Chinese Academy of Sciences (No. XDA23020301), the National Fleischmann, O.C., Hartmann, M., Burrows, J.P., Orphal, J., 2004. New ultraviolet absorp-
Key Project for Causes and Control of Heavy Air Pollution (No. tion cross-sections of BrO at atmospheric temperatures measured by time-
DQGG0102 and DQGG0205), the National High-Resolution Earth Obser- windowing Fourier transform spectroscopy. J. Photoch. Photobio. A 168, 117–132.
Frieß, U., Monks, P., Remedios, J., Rozanov, A., Sinreich, R., Wagner, T., Platt, U., 2006.
vation Project of China (No. 05-Y20A16-9001-15/17-2), Natural Science MAX-DOAS O4 measurements: a new technique to derive information on atmo-
Foundation of Guangzhou Province (No. 2016A030310115). We ac- spheric aerosols: 2. Modeling studies. J. Geophys. Res. 111, D14203. https://doi.org/
knowledge the Ministry of Environment Protection of the People's Re- 10.1029/2005JD006618.
Frieß, U., Sihler, H., Sander, R., Pöhler, D., Yilmaz, S., Platt, U., 2011. The vertical distribu-
public of China for making data measured by the atmospheric tion of BrO and aerosols in the Arctic: measurements by active and passive differen-
environment automatic monitoring stations available online. We thank tial optical absorption spectroscopy. J. Geophys. Res. 116, D00R04. https://doi.org/
the TROPOMI International Science Team for making TROPOMI data 10.1029/2011JD015938.
Fu, T.-M., Jacob, D.J., Wittrock, F., Burrows, J.P., Vrekoussis, M., Henze, D.K., 2008. Global
available for scientific researches. We would like to thank Chongqing budgets of atmospheric glyoxal and methylglyoxal, and implications for formation
Environment Monitoring Center for the meteorological data measured of secondary organic aerosols. J. Geophys. Res. 113, D15303. https://doi.org/
in Changshou. We thank IUP Heidelberg University to help us on 10.1029/2007JD009505.
Geiger, H., Kleffmann, J., Wiesen, P., 2002. Smog chamber studies on the influence of die-
HEIPRO. In addition, we would like to sincerely thank the great contri-
sel exhaust on photosmog formation. Atmos. Environ. 36 (11), 1737–1747. https://
bution of the SCIATRAN development team at the Institute of Remote doi.org/10.1016/S1352-2310(02)00175-9.
Sensing/Institute of Environmental Physics (IUP/IFE), University of Bre- Gratsea, M., Vrekoussis, M., Richter, A., Wittrock, F., Schönhardt, A., Burrows, J., Kazadzis,
men. We also thank the Belgian Institute for Space Aeronomy (BIRA- S., Mihalopoulos, N., Gerasopoulos, E., 2016. Slant column MAX-DOAS measurements
of nitrogen dioxide, formaldehyde, glyoxal and oxygen dimer in the urban environ-
IASB), Brussels, Belgium, for their freely accessible QDOAS software. ment of Athens. Atmos. Environ. 135, 118–131.
Grosjean, E., Grosjean, D., Fraser, M.P., Cass, G.R., 1996. Air quality model evaluation data
Appendix A. Supplementary data for organics. 2. C1–C14 carbonyls in Los Angeles air. Environmental Science & Tech-
nology 30 (9), 2687–2703. https://doi.org/10.1021/es950758w.
Hendrick, F., Müller, J.F., Clémer, K., Wang, P., De Mazière, M., Fayt, C., Gielen, C., Hermans,
Supplementary data to this article can be found online at https://doi. C., Ma, J.Z., Pinardi, G., et al., 2014. Four years of ground-based MAX-DOAS observa-
org/10.1016/j.scitotenv.2019.136258. tions of HONO and NO2 in the Beijing area. Atmos. Chem. Phys. 14, 765–781.
Holzinger, R., Warneke, C., Hansel, A., Jordan, A., Lindinger, W., Scharffe, D.H., et al., 1999.
Biomass burning as a source of formaldehyde, acetaldehyde, methanol, acetone, ace-
References tonitrile, and hydrogen cyanide. Geophys. Res. Lett. 26 (8), 1161–1164. https://doi.
org/10.1029/1999GL900156.
Aliwell, S.R., Van Roozendael, M., Johnston, P.V., Richter, A., Wagner, T., Arlander, D.W., Hong, Q., Liu, C., Chan, K.L., Hu, Q., Xie, Z., Liu, H., Si, F., Liu, J., 2018. Ship-based MAX-DOAS
Burrows, J.P., Fish, D.J., Jones, R.L., Tørnkvist, K.K., Lambert, J.-C., Pfeilsticker, K., measurements of tropospheric NO2, SO2, and HCHO distribution along the Yangtze
Pundt, I., 2002. Analysis for BrO in zenith-sky spectra: an intercomparison exercise River. Atmos. Chem. Phys. 18, 5931–5951. https://doi.org/10.5194/acp-18-5931-
for analysis improvement. J. Geophys. Res. 107. https://doi.org/10.1029/ 2018.
2001JD000329 ACH 10-1–ACH 10-20. Hönninger, G., Platt, U., 2002. Observations of BrO and its vertical distribution during sur-
Andreae, M.O., Merlet, P., 2001. Emission of trace gases and aerosols from biomass burn- face ozone depletion at Alert. Atmos. Environ. 36, 2481–2489.
ing. Glob. Biogeochem. Cycles 15 (4), 955–966. https://doi.org/10.1029/ Hönninger, G., von Friedeburg, C., Platt, U., 2004. Multi axis differential optical absorption
2000GB001382. spectroscopy (MAX-DOAS). Atmos. Chem. Phys. 4, 231–254. https://doi.org/10.5194/
Arlander, D., Brüning, D., Schmidt, U., Ehhalt, D., 1995. The tropospheric distribution of acp-4-231-2004.
formaldehyde during TROPOZ II. J. Atmos. Chem. 22 (3), 251–269. https://doi.org/ Hoque, H.M.S., Irie, H., Damiani, A., 2018. First MAX-DOAS observations of formaldehyde
10.1007/BF00696637. and glyoxal in Phimai, Thailand. J. Geophys. Res. 123 (17). https://doi.org/10.1029/
Baidar, S., Oetjen, H., Coburn, S., Dix, B., Ortega, I., Sinreich, R., Volkamer, R., 2013. The CU 2018JD028480.
airborne MAX-DOAS instrument: vertical profiling of aerosol extinction and trace Houweling, S., Dentener, F., Lelieveld, J., 1998. The impact of nonmethane hydrocarbon
gases. Atmospheric Measurement Techniques 6 (3), 719–739. https://doi.org/ compounds on tropospheric photochemistry. J. Geophys. Res.-Atmos. 103,
10.5194/amt-6-719-2013. 10673–10696.
Barletta, B., Meinardi, S., Sherwood Rowland, F., Chan, C.Y., Wang, X., Zou, S., Yin Chan, L., Irie, H., Kanaya, Y., Akimoto, H., Iwabuchi, H., Shimizu, A., Aoki, K., 2008. First retrieval of
Blake, D.R, 2005. Volatile organic compounds in 43 Chinese cities. Atmos. Environ 39, tropospheric aerosol profiles using MAX-DOAS and comparison with lidar and sky ra-
5979–5990. diometer measurements. Atmos. Chem. Phys. 8, 341–350. https://doi.org/10.5194/
Bo, Y., Cai, H., Xie, S.D., 2008. Spatial and temporal variation of historical anthropogenic acp-8-341-2008.
NMVOCs emission inventories in China. Atmos. Chem. Phys. 8, 7297e7316. https:// Irie, H., Takashima, H., Kanaya, Y., Boersma, K., Gast, L., Wittrock, F., et al., 2011. Eight-
doi.org/10.5194/acp-8-7297-2008. component retrievals from ground-based MAX-DOAS observations. Atmospheric
Bobrowski, N., Hönninger, G., Galle, B., Platt, U., 2003. Detection of bromine monoxide in a Measurement Techniques 4 (6), 1027–1044. https://doi.org/10.5194/amt-4-1027-
volcanic plume. Nature 423, 273–276. 2011.
Chan, K.L., Hartl, A., Lam, Y.F., Xie, P.H., Liu, W.Q., Cheung, H.M., Lampel, J., Pöhler, D., Li, A., Javed, Z., Liu, C., Khokhar, M.F., Xing, C., Tan, W., Subhani, M.A., Rehman, A., Tanvir, A.,
Xu, J., Zhou, H.J., Ning, Z., Wenig, M.O., 2015. Observations of tropospheric NO2 using 2019. Investigating the impact of glyoxal retrieval from MAX-DOAS observations dur-
ground based MAX-DOAS and OMI measurements during the Shanghai World Expo ing haze and non-haze conditions in Beijing. J. Environ. Sci. 80, 296–305.
2010. Atmos. Environ. 119, 45–58. https://doi.org/10.1016/j.atmosenv.2015.08.041. Jing, L.H., Steinberg, S.M., Johnson, B.J., 2001. Aldehyde and monocyclic aromatic hydro-
Chan, K.L., Wiegner, M., Wenig, M., Pöhler, D., 2017a. Observations of tropospheric aero- carbon mixing ratios at an urban site in Las Vegas, Nevada. J. Air Waste Manage.
sols and NO2 in Hong Kong over 5 years using ground based MAX-DOAS. Sci. Total Assoc. 51, 1359–1366.
Environ. 619, 1545–1556. https://doi.org/10.1016/j.scitotenv.2017.10.153. Johnson, D., Utembe, S.R., Jenkin, M.E., 2006. Simulating the detailed chemical composi-
Chan, K.L., Wang, S.S., Liu, C., Zhou, B., Wenig, M.O., Saiz-Lopez, A., 2017b. On the summer- tion of secondary organic aerosol formed on a regional scale during the TORCH
time air quality and related photochemical processes in megacity Shanghai, China. 2003 campaign in the southern UK. Atmos. Chem. Phys. 6, 419–431. https://doi.org/
Sci. Total Environ. 580, 974–983. https://doi.org/10.1016/j.scitotenv.2016.12.052. 10.5194/acp-6-419-2006.
Chan, K.L., Wang, Z., Ding, A., Heue, K.-P., Shen, Y., Wang, J., Zhang, F., Shi, Y., Hao, N., Kaiser, J., Wolfe, G.M., Min, K.E., Brown, S.S., Miller, C.C., Jacob, D.J., et al., 2015. Reassessing
Wenig, M., 2019. MAX-DOAS measurements of tropospheric NO2 and HCHO in the ratio of glyoxal to formaldehyde as an indicator of hydrocarbon precursor speci-
 C. Xing et al. / Science of the Total Environment 715 (2020) 136258 11

ation. Atmos. Chem. Phys. 15 (13), 7571–7583. https://doi.org/10.5194/acp-15-7571- Rodgers, C.D., 2000. Inverse methods for atmospheric sounding, theory and practice, Ser.
2015. Atmos., Oceanic, Planet. Phys. 2. World Sci, Singapore.
Kanakidou, M., Seinfeld, J.H., Pandis, S.N., Barnes, I., Dentener, F.J., Facchini, M.C., Van Roscoe, H.K., Van Roozendael, M., Fayt, C., du Piesanie, A., Abuhassan, N., Adams, C.,
Dingenen, R., Ervens, B., Nenes, A., Nielsen, C.J., Swietlicki, E., Putaud, J.P., Balkanski, Akrami, M., Cede, A., Chong, J., Clémer, K., Friess, U., Gil Ojeda, M., Goutail, F.,
Y., Fuzzi, S., Horth, J., Moortgat, G.K., Winterhalter, R., Myhre, C.E.L., Tsigaridis, K., Graves, R., Griesfeller, A., Grossmann, K., Hemerijckx, G., Hendrick, F., Herman, J.,
Vignati, E., Stephanou, E.G., Wilson, J., 2005. Organic aerosol and global climate Hermans, C., Irie, H., Johnston, P.V., Kanaya, Y., Kreher, K., Leigh, R., Merlaud, A.,
modelling: a review. Atmos. Chem. Phys. 5, 1053–1123. https://doi.org/10.5194/ Mount, G.H., Navarro, M., Oetjen, H., Pazmino, A., Perez-Camacho, M., Peters, E.,
acp-5-1053-2005. Pinardi, G., Puentedura, O., Richter, A., Schönhardt, A., Shaiganfar, R., Spinei, E.,
Kreher, Karin, Roozendael, Michel Van, Hendrick, Francois, Apituley, Arnoud, Strong, K., Takashima, H., Vlemmix, T., Vrekoussis, M., Wagner, T., Wittrock, F., Yela,
Dimitropoulou, Ermioni, Frieß, Udo, Richter, Andreas, Wagner, Thomas, Abuhassan, M., Yilmaz, S., Boersma, F., Hains, J., Kroon, M., Piters, A., Kim, Y.J., 2010. Intercompar-
Nader, Ang, Li, Anguas, Monica, Bais, Alkis, Benavent, Nuria, Bösch, Tim, Bognar, ison of slant column measurements of NO2 and O4 by MAX-DOAS and zenith-sky UV
Kristof, Borovski, Alexander, Bruchkouski, Ilya, Cede, Alexander, Ka, L. Chan, and visible spectrometers. Atmos. Meas. Tech. 3, 1629–1646. https://doi.org/10.5194/
Donner, Sebastian, Drosoglou, Theano, Fayt, Caroline, Finkenzeller, Henning, Garcia- amt-3-1629-2010.
Nieto, David, Gielen, Clio, Gómez-Martín, Laura, Hao, Nan, Herman, Jay R., Hermans, Rothman, L.S., Gordon, I.E., Barbe, A., Benner, D.C., Bernath, P.F., Birk, M., et al., 2009. The
Christian, Hoque, Syedul, Irie, Hitoshi, Jin, Junli, Johnston, Paul, Butt, Junaid HITRAN 2008 molecular spectroscopic database. J. Quant. Spectrosc. Radiat. Transf.
Khayyam, Khokhar, Fahim, Koenig, Theodore K., Kuhn, Jonas, Kumar, Vinod, 110 (9–10), 533–572. https://doi.org/10.1016/j.jsrt.2009.02.013.
Lampel, Johannes, Liu, Cheng, Ma, Jianzhong, Merlaud, Alexis, Mishra, Abhishek K., Ryan, Robert G., Rhodes, Steve, Tully, Matthew, Wilson, Stephen, Jones, Nicholas, Frieß,
Müller, Moritz, Navarro-Comas, Monica, Ostendorf, Mareike, Pazmino, Andrea, Udo, Schofie, Robyn, 2018. Daytime HONO, NO2 and aerosol distributions from
Peters, Enno, Pinardi, Gaia, Pinharanda, Manuel, Piters, Ankie, Platt, Ulrich, MAX-DOAS observations in Melbourne. Atmos. Chem. Phys. 18, 13969–13985.
Postylyakov, Oleg, Prados-Roman, Cristina, Puentedura, Olga, Querel, Richard, Saiz- https://doi.org/10.5194/acp-18-13969-2018.
Lopez, Alfonso, Schönhardt, Anja, Schreier, Stefan F., Seyler, Andre, Sinha, Vinayak, Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2001. Measurement of emissions
Spinei, Elena, Strong, Kimberly, Tack, Frederik, Tian, Xin, Tiefengraber, Martin, from air pollution sources. 3. C1–C29 organic compounds from fireplace combustion
Tirpitz, Jan-Lukas, van Gent, Jeron, Volkamer, Rainer, Vrekoussis, Mihalis, Wang, of wood. Environ. Sci. Technol. 35, 1716–1728. https://doi.org/10.1021/es001331e.
Shanshan, Wang, Zhuoru, Wenig, Mark, Wittrock, Folkard, Xie, Pinhua H., Jin, Xu, Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2002. Measurement of emissions
Yela, Margarita, Zhang, Chengxin, Zhao, Xiaoyi, 2019. Intercomparison of NO2, O4, from air pollution sources. 5. C1–C32 organic compounds from gasoline-powered
O3 and HCHO slant column measurements by MAX-DOAS and zenith-sky UV- motor vehicles. Environ. Sci. Technol. 36, 1169–1180. https://doi.org/10.1021/
Visible spectrometers during the CINDI-2 campaign. Atmos. Meas. Tech. Disc. es0108077.
https://doi.org/10.5194/amt-2019-157. Serdyuchenko, A., Gorshelev, V., Weber, M., Chehade, W., Burrows, J.P., 2014. High spec-
Lee, Y.N., Zhou, X., Hallock, K., 1995. Atmospheric carbonyl compounds at a rural south- tral resolution ozone absorption cross-sections – part 2: temperature dependence.
eastern United States site. J. Geophys. Res. 100 (D12), 25,933–25,944. https://doi. Atmos. Meas. Tech. 7, 625–636. https://doi.org/10.5194/amt-7-625-2014.
org/10.1029/95JD02605. Shao, M., Lu, S.H., Liu, Y., Xie, X., Chang, C.C., Huang, S., Chen, Z.M., 2009. Volatile organic
Lee, Y.N., Zhou, X., Kleinman, L.I., Nunnermacker, L.J., Springston, S.R., Daum, P.H., compounds measured in summer in Beijing and their role in ground-level ozone for-
Newman, L., Keigley, W.G., Holdren, M.W., Spicer, C.W., Young, V., Fu, B., Parrish, mation. J. Geophys. Res. Atmos. 114, D00g06. https://doi.org/10.1029/2008jd010863.
D.D., Holloway, J., Williams, J., Roberts, J.M., Ryerson, T.B., Fehsenfeld, F.C., 1998. At- Su, J., 2018. Portable and sensitive air pollution monitoring. Light Sci. Appl. 7 (3). https://
mospheric chemistry and distribution of formaldehyde and several multioxygenated doi.org/10.1038/s41377-018-0017-x.
carbonyl compounds during the 1995 Nashville/Middle Tennessee Ozone Study. Tan, W., Liu, C., Wang, S., Xing, C., Su, W., Zhang, C., Xia, C., Liu, H., Cai, Z., Liu, J., 2018. Tro-
J. Geophys. Res. 103, 22449–22462. https://doi.org/10.1029/98jd01251. pospheric NO2, SO2, and HCHO over the East China Sea, using ship-based MAX-DOAS
Lei, W., Zavala, M., de Foy, B., Volkamer, R., Molina, M.J., Molina, L.T., 2009. Impact of pri- observations and comparison with OMI and OMPS satellites data. Atmos. Chem. Phys.
mary formaldehyde on air pollution in the Mexico City Metropolitan Area. Atmos. 18 (20), 15387–15402. https://doi.org/10.5194/acp-2018-578.
Chem. Phys. 9, 2607–2618. https://doi.org/10.5194/acp-9-2607-2009. Tao, J., Zhang, L., Cao, J., Zhang, R., 2017. A review of current knowledge concerning
Li, X., Brauers, T., Shao, M., Garland, R.M., Wagner, T., Deutschmann, T., Wahner, A., 2010. PM2.5chemical composition, aerosol optical properties, and their relationships across
MAX-DOAS measurements in southern China: retrieval of aerosol extinctions and China. Atmos. Chem. Phys. Discuss. https://doi.org/10.5194/acp-2017-271.
validation using ground-based in-situ data. Atmos. Chem. Phys. 10, 2079–2089. Thalman, R.M., Volkamer, R., 2013. Temperature dependent absorption cross-sections of
https://doi.org/10.5194/acp-10-2079-2010. O2-O−2 collision pairs between 340 and 630nm and at atmospherically relevant
Li, X., Brauers, T., Hofzumahaus, A., Lu, K., Li, Y., Shao, M., et al., 2013. MAX-DOAS pressure. Phys. Chem. Chem. Phys. 15, 15371–15381. https://doi.org/10.1039/
measurements of NO 2 , HCHO and CHOCHO at a rural site in southern China. c3cp50968k.
Atmos. Chem. Phys. 13 (4), 2133–2151. https://doi.org/10.5194/acp-13-2133- Vandaele, A.C., Hermans, C., Simon, P.C., Carleer, M., Colin, R., Fally, S., Mérienne, M.-F.,
2013. Jenouvrier, A., Coquart, B., 1998. Measurements of the NO2 absorption cross section
Liu, W., Zhang, J., Kwon, J., Weisel, C., Turpin, B., Zhang, L., Korn, L., Morandi, M., Stock, T., from 42 000 cm−1 to 10000 cm−1 (238–1000 nm) at 220K and 294 K. J. Quant.
Colome, S., 2006. Concentrations and source characteristics of airborne carbonyl com- Spectrosc. Ra. 59, 171–184.
pounds measured outside urban residences. J. Air Waste Manage. Assoc. 56, Volkamer, R., Spietz, P., Burrows, J., Platt, U., 2005. High-resolution absorption cross-
1196–1204. section of glyoxal in the UV-vis and IR spectral ranges. J. Photoch. Photobio. A 172,
Liu, H., Liu, C., Xie, Z., Li, Y., Huang, X., Wang, S., Xu, J., Xie, P., 2016. A paradox for air pol- 35–46. https://doi.org/10.1016/j.jphotochem.2004.11.011.
lution controlling in China revealed by “APEC Blue” and “Parade Blue”. Sci. Rep. 6, Volkamer, R., Jimenez, J.L., San Martini, F., Dzepina, K., Zhang, Q., Salcedo, D., Molina, L.T.,
34408. https://doi.org/10.1038/srep34408. Worsnop, D.R., Molina, M.J., 2006. Secondary organic aerosol formation from anthro-
Meier, J., Wehner, B., Massling, A., Birmili, W., Nowak, A., Gnauk, T., Brueggemann, E., pogenic air pollution: rapid and higher than expected. Geophys. Res. Lett. 33, L17811.
Herrmann, H., Min, H., Wiedensohler, A., 2009. Hygroscopic growth of urban aerosol https://doi.org/10.1029/2006gl026899.
particles in Beijing (China) during wintertime: a comparison of three experimental Volkamer, R., San Martini, F., Molina, L.T., Salcedo, D., Jimenez, J.L., Molina, M.J., 2007. A
methods. Atmos. Chem. Phys. 9, 6865–6880. https://doi.org/10.5194/acp-9-6865- missing sink for gas-phase glyoxal in Mexico City: formation of secondary organic
2009. aerosol. Geophys. Res. Lett. 34, LI9807. https://doi.org/10.1029/2007GL030752.
Meller, R., Moortgat, G.K., 2000. Temperature dependence of the absorption cross sections Vrekoussis, M., Wittrock, F., Richter, A., Burrows, J.P., 2009. Temporal and spatial variabil-
of formaldehyde between 223 and 323K in the wavelength range 225–375 nm. ity of glyoxal as observed from space. Atmos. Chem. Phys. 9 (13), 4485–4504. https://
J. Geophys. Res. 105, 7089–7101. doi.org/10.5194/acp-9-4485-2009.
Miller, C., Gonzalez Abad, G., Wang, H., Liu, X., Kurosu, T., Jacob, D.J., Chance, K.V., 2014. Vrekoussis, M., Wittrock, F., Richter, A., Burrows, J.P., 2010. GOME-2 observations of oxy-
Glyoxal retrieval from the ozone monitoring instrument. Atmospheric Measurement genated VOCs: what can we learn from the ratio glyoxal to formaldehyde on a global
Techniques 7 (11), 3891–3907. https://doi.org/10.5194/amt-7-3891-2014. scale? Atmos. Chem. Phys. 10 (21), 10,145–10,160. https://doi.org/10.5194/acp-10-
Munger, J.W., Jacob, D.J., Daube, B.C., Horowitz, L., Keene, W., Heikes, B., 1995. Formalde- 10145-2010.
hyde, glyoxal, and methylglyoxal in air and cloud water at a rural mountain site in Wagner, T., Dix, B., Friedeburg, C.V., Frieß, U., Sanghavi, S., Sinreich, R., Platt, U., 2004.
central Virginia. J. Geophys. Res. 100 (D5), 9325–9333. https://doi.org/10.1029/ MAX-DOAS O4 measurements: a new technique to derive information on atmo-
95JD00508. spheric aerosols—principles and information content. J. Geophys. Res. 109, D22205.
Myriokefalitakis, S., Vrekoussis, M., Tsigaridis, K., Wittrock, F., Richter, A., Brühl, C., et al., https://doi.org/10.1029/2004JD004904.
2008. The influence of natural and anthropogenic secondary sources on the glyoxal Wagner, T., Beirle, S., Brauers, T., Deutschmann, T., Frieß, U., Hak, C., Halla, J.D., Heue, K.P.,
global distribution. Atmos. Chem. Phys. 8 (16), 4965–4981. https://doi.org/10.5194/ Junkermann, W., Li, X., Platt, U., Pundt-Gruber, I., 2011. Inversion of tropospheric pro-
acp-8-4965-2008. files of aerosol extinction and HCHO and NO2 mixing ratios from MAX-DOAS obser-
Ortega, I., Koenig, T., Sinreich, R., Thomson, D., Volkamer, R., 2015. The CU 2-D-MAX-DOAS vations in Milano during the summer of 2003 and comparison with independent
instrument—part 1: retrieval of 3-D distributions of NO2 and azimuth-dependent data sets. Atmos. Meas. Tech. 4, 2685–2715. https://doi.org/10.5194/amt-4-2685-
OVOC ratios. Atmospheric Measurement Techniques 8 (6), 2371–2395. https://doi. 2011.
org/10.5194/amt -8-2371-2015. Wagner, Thomas, Beirle, Steffen, Remmers, Julia, Shaiganfar, Reza, Wang, Yang, 2016. Ab-
Platt, U., Stutz, J., 2008. Differential Absorption Spectroscopy: Principles and Applications. solute calibration of the colour index and O4 absorption derived from Multi AXis
Springer, Berlin, pp. 497–498. (MAX-)DOAS measurements and their application to a standardised cloud classifica-
Poisson, N., Kanakidou, M., Crutzen, P.J., 2000. Impact of nonmethane hydrocarbons on tion algorithm. Atmos. Meas. Tech. 9, 4803–4823. https://doi.org/10.5194/amt-9-
tropospheric chemistry and the oxidizing power of the global troposphere: 3- 4803-2016.
dimensional modelling results. J. Atmos. Chem. 36, 157–230. Wittrock, F., Oetjen, H., Richter, A., Fietkau, S., Medeke, T., Rozanov, A., Burrows, J.P., 2004.
Rodgers, C.D., 1990. Characterization and error analysis of profiles retrieved from remote MAX-DOAS measurements of atmospheric trace gases in Ny-Ålesund – radiative
sounding measurements. J. Geophys. Res. 95, 5587–5595. https://doi.org/10.1029/ transfer studies and their application. Atmos. Chem. Phys. 4, 955–966. https://doi.
JD095iD05p05587. org/10.5194/acp-4-955-2004.
12 C. Xing et al. / Science of the Total Environment 715 (2020) 136258

Wu, Y., Shiledar, A., Li, Y., et al., 2017. Air quality monitoring using mobile microscopy and burning plumes sampled by aircraft. Environmental Science & Technology 51 (20),
machine learning. Light Sci. Appl. 6, e17046. https://doi.org/10.1038/lsa.2017.46. 11,761–11,770. https://doi.org/10.1021/acs.est.7b03517.
Xing, C., Liu, C., Wang, S., Chan, K.L., Gao, Y., Huang, X., Su, W., Zhang, C., Dong, Y., Fan, G., Zhang, Q., Streets, D.G., Carmichae, G.R., He, K.B., Huo, H., Kannari, A., Klimont, Z., Park, I.S.,
Zhang, T., Chen, Z., Hu, Q., Su, H., Xie, Z., Liu, J., 2017. Observations of the vertical dis- Reddy, S., Fu, J.S., Chen, D., Duan, L., Lei, Y., Wang, L.T., Yao, Z.L., 2009. Asian emissions
tributions of summertime atmospheric pollutants and the corresponding ozone pro- in 2006 for the NASA INTEX-B mission. Atmos. Chem. Phys. 9, 5131e5153. https://doi.
duction in Shanghai, China. Atmos. Chem. Phys. 17, 14275–14289. https://doi.org/ org/10.5194/acp-9-5131-2009.
10.5194/acp-17-14275-2017. Zhang, C., Liu, C., Hu, Q., et al., 2019. Satellite UV–Vis spectroscopy: implications for air
Xing, C., Liu, C., Wang, S., Hu, Q., Liu, H., Tan, W., Zhang, W., Li, B., Liu, J., 2019. A new quality trends and their driving forces in China during 2005–2017. Light Sci. Appl. 8
method to determine the aerosol optical properties from multiple-wavelength O4 ab- (100). https://doi.org/10.1038/s41377-019-0210-6.
sorptions by MAX-DOAS observation. Atmos. Meas. Tech. 12, 3289–3302. https://doi. Zheng, G.J., Duan, F.K., Su, H., Ma, Y.L., Cheng, Y., Zheng, B., Zhang, Q., Huang, T., Kimoto, T.,
org/10.5194/amt-12-3289-2019. Chang, D., Pöschl, U., Cheng, Y.F., He, K.B., 2015. Exploring the severe winter haze in
Yokelson, R.J., Goode, J.G., Ward, D.E., Susott, R.A., Babbitt, R.E., Wade, D.D., et al., 1999. Beijing: the impact of synoptic weather, regional transport and heterogeneous reac-
Emissions of formaldehyde, acetic acid, methanol, and other trace gases from biomass tions. Atmos. Chem. Phys. 15, 2969–2983. https://doi.org/10.5194/acp-15-2969-
fires in North Carolina measured by airborne Fourier transform infrared spectros- 2015.
copy. J. Geophys. Res. 104 (D23), 30,109–30,125. https://doi.org/10.1029/
1999JD900817.
Zarzana, K.J., Min, K., Washenfelder, R.A., Kaiser, J., Krawiec-Thayer, M., Peischl, J., et al.,
2017. Emissions of glyoxal and other carbonyl compounds from agricultural biomass