Science of the Total Environment 622–623 (2018) 1621–1629

Contents lists available at ScienceDirect

Science of the Total Environment

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

Spatial and temporal variations in the atmospheric concentrations of

“Stockholm Convention” organochlorine pesticides in Kuwait
Bondi Gevao ⁎, Massimiliano Porcelli, Smitha Rajagopalan, Divya Krishnan, Karell Martinez-Guijarro,
Hassan Alshemmari, Majed Bahloul, Jamal Zafar
Environmental Pollution and Climate Program, Environment and Life Sciences Research Center, Kuwait Institute for Scientific Research, P. O. Box 24885, Safat 13109, Kuwait

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

• First ever report of organochlorine pes-

ticide concentrations in air in Arabian
Gulf
• Levels higher at urban locations relative
to their concentrations at the remote lo-
cation
• HCB and PeCB concentrations negative-
ly correlated with temperature.
• Concentrations in Summer higher for
HCHs, chlordanes, dieldrin, DDTs, ex-
cept for pp-DDT

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

Article history: The study reports fortnightly atmospheric concentrations of organochlorine pesticides concomitantly measured
Received 12 July 2017 at an urban, “industrial” and a remote location over a twelve month period in Kuwait to examine seasonal vari-
Received in revised form 12 September 2017 ability and urban-rural concentration gradients. The average ± SD (and range) of the ΣOC concentrations mea-
Accepted 5 October 2017
sured throughout the study period in decreasing order were urban, 505 ± 305 (range, 33–1352) pg m−3, remote,
Available online 18 October 2017
204 ± 124 (4.5–556) pg m−3, and “industrial” 155 ± 103 (8.8–533) pg m−3. The concentrations of most OCs
Editor: Adrian Covaci measured in this study, except for DDT and its metabolites, were higher at urban locations relative to their con-
centrations at remote location, in line with the literature on POPs regarding urban conurbations being sources of
Keywords: industrial chemicals. The most abundant pesticides measured throughout this study were dieldrin, pentachloro-
Persistent organic pollutants benzene, hexachlorobenzene, ΣDDTs, ΣHCHs, and oxychlordane. Hexachlorobenzene concentrations were gen-
Organochlorine pesticides erally higher than those of pentachlorobenzene and are both strongly negatively correlated with temperature
Stockholm Convention (p b 0.05) at all sampling locations. Mean summertime concentrations were higher for hexachlorocyclohexanes,
Air, spatial and temporal trends chlordanes, dieldrin, and DDT isomers, except for pp-DDT. This may suggest that concentrations of these com-
pounds may be driven by temperature.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction

Organochlorine pesticides (OCPs) are a diverse group of persistent

compounds that were manufactured and used in agriculture and in pub-
⁎ Corresponding author. lic health (Barakat, 2004; Barra et al., 2006; Li et al., 1998; Prudente
E-mail address: bgevao@kisr.edu.kw (B. Gevao). et al., 2007). These chemicals have received intense international

https://doi.org/10.1016/j.scitotenv.2017.10.036
0048-9697/© 2017 Elsevier B.V. All rights reserved.
1622 B. Gevao et al. / Science of the Total Environment 622–623 (2018) 1621–1629
attention because of their ubiquity, persistence, high bioaccumulation
potential and harmful biological effects. The combination of their resis-
tance to metabolism and lipophilicity means that they will
bioaccumulate and be transported through food chains (Daley et al.,
2011; Ikemoto et al., 2008; Kelly et al., 2007). The OCPs included in
this study are either banned or restricted under the “Stockholm Conven-
tion” on persistent organic pollutants (POPs), yet they continue to be
measured in air around the world. Current sources of most of these pes-
ticides, particularly in Kuwait which has a very small agricultural sector,
include atmospheric transport from agricultural countries where some
are likely to still be in use or from volatilization from stockpiles
(Alamdar et al., 2014) and other contaminated sites (Fang et al.,
2017). As these chemicals are trans-boundary pollutants and undergo
long range transport from source to remote regions (Meng et al.,
2011; Scheringer et al., 2004), environmental data are needed from all
regions of the globe to better understand the key processes that influ-
ence their global distribution. Until recently, there was a paucity of reli-
able environmental data on the levels of most POP chemicals in the
Middle East, most of Africa, and Asia, from which to assess the effective-
ness of international efforts to minimize the release of these chemicals
to the environment. The main objective of this study therefore, was to
generate the first reliable concentrations of OC pesticides in air in
Kuwait which can form the baseline for the evaluation of the effective-
ness of international controls on their levels in the future. To achieve
this objective, high volume air samplers were used to determine the
seasonality in the atmospheric concentrations of OCPs listed in the
Stockholm Convention at three fixed monitoring sites over a
12 month period.
2. Materials and methods
2.1. Sampling
Three sites were selected for collecting high volume air samples
using the sampler from Tisch Environmental, Inc. The “urban” sampling
site at Shuwaikh was on the premises of the Kuwait Institute for Scien-
tific Research (29° 20′ 10.90″ N; 47° 54′ 15.80″ E) close to the meteoro-
logical station. The “remote” site (29° 58′ 7.30″ N; 47° 42′ 3.80″ E) was
on a farm in Abdali, close to the border with Iraq, was selected to reflect
background conditions. The third site located on the roof of a house in
Egaila (29° 9′51″ N; 48° 6′26″ E) was selected to represent “industrial”
atmosphere to assess the impact of industrial activities on the levels of
POPs in Kuwait. In the discussions that follow, the sampling locations
will be described by the location type (urban, remote, and industrial)
rather than the name of the area samples were collected. Twenty-four
hour air samples (n = 27, at each site) were collected every fortnight
from March 2013 to March 2014 to assess seasonal fluctuations in am-
bient concentrations of POPs. Approximately 800 ± 50 m3 of air was
pumped through a Whatman GFF (8″ × 10″) to trap particulate matter
and the vapor phase compounds trapped on two PUF plugs (85 mm in
diameter and 70 mm in length) located downstream of the GFF. The
exact air volumes for every sample were determined by using a calibrat-
ed Magnehelic gauge (Tisch Environmental, Inc.) to measure pressure at
the start and end of each sampling period. The average of the two read-
ings was used to calculate the air sampled during that measurement pe-
riod. Upon retrieval, the PUF and filter samples were stored in separate
cleaned amber glass jars and kept at −15 °C until extraction, to mini-
mize losses by photolysis and/or volatilization. The GFF was weighed
using a micro balance before and after deployment to determine the
total suspended particulate (TSP) concentration over that sampling pe-
riod, which is nominally defined as the difference between the two
measurements. Field blanks, collected fortnightly, consist of a PUF and
filter assembled in the sampler, immediately removed at the time of de-
ployment of the sampling, and processed in an identical manner like ac-
tual samples. Laboratory blanks comprise of pre-extracted PUF plugs
and baked filters were treated as samples.
2.2. Extraction and clean-up
Extraction and cleanup procedures followed previously published
protocols (Gevao et al., 2010). Briefly, samples were extracted in a
Soxhlet apparatus using 1:1 v/v mixture of DCM:hexane. Prior to extrac-
tion, the samples were spiked with a 13C-POPs surrogate cocktail pur-
chased from CIL (Tewksbury, MA, USA) containing all the compounds
in the expanded POPs list, (hexachlorobenzene, pentachlorobenzene,
aldrin, dieldrin, endrin, 4,4′-DDT, 4,4′-DDE, 4,4′-DDD, 2,4′-DDT, 2,4′-
DDE, 2,4′-DDD, α-chlordane, γ-chlordane, trans-nanochlor, cis-
nanochlor, oxychlordane, heptachlor, trans-heptachlor epoxide (A iso-
mer), cis-heptachlor epoxide (B isomer), mirex, kepone (chlordecone),
α-HCH, β-HCH, γ-HCH, δ-HCH, endosulfan I, endosulfan II); to monitor
analytical recoveries. The sample extracts were concentrated on a
Turbovap® II workstation (Hopkinton, MA, USA), solvent exchanged
to hexane and interfering compounds removed by column chromatog-
raphy (i.d. 9 mm) using 2 g of silica gel (Merck, 60–230 mesh) and 1 g
alumina (BDH, neutral alumina) and capped with baked anhydrous so-
dium sulphate to prevent the column from exposure to air. The samples
were then transferred to 100-μl glass inserts, and spiked with 20 ng
isodrin and analyzed for organochlorine pesticides (OCPs) using an
Agilent 7890B gas chromatograph coupled to a triple quadrupole mass
spectrometer operated in electron ionization mode.
2.3. Instrumental analysis
The analytes were separated on a 30-m DB5-ms column (0.25 mm
i.d., 0.25-μm film thickness, J&W Scientific) with helium as carrier gas.
The oven program was 70 °C for 2 min, ramped at 25 °C min−1 to 150
°C, 3 °C min−1 to 200 °C, and further ramped at 8 °C min−1 to 280 °C
and held for 10 min. Operating conditions were as follows: injector tem-
perature was set at 280 °C; ion source 300 °C; quadrupole 150 °C; trans-
fer line 310 °C. The carrier gas was helium at a flow rate of 2.6 ml/min
and the collision gas was nitrogen at a flow rate of 1.5 ml/min. The gas
chromatograph was fitted with a back flush capability. At the end of
each run a 5 min back flush was carried out during which the oven tem-
perature was maintained at 310 °C. The triple quadrupole was operated
in MS/MS mode, and analytes detected and confirmed using multiple
reaction monitoring (MRM) using two transitions per analyte. The
MRM transitions monitored for each compound is given in Table SI-1
(Supporting information). MassHunter analysis software was used for
data processing.
2.4. QA/QC
Strict quality-assurance and quality-control measures were put in
place to monitor the analytical process. Identification and quantification
was carried out against 5 calibration standards of known concentration.
A peak was positively identified if it was within ±0.05 min of the reten-
tion time in the calibration standard and quantified only if the S/N ≥ 3,
and the ratio of the ion to its qualifier ion was within ±20% of the stan-
dard value. There was no discernible difference between laboratory and
field blanks, indicating minimal contamination during transport, stor-
age and analysis. When analytes were present in a blank processed
along a batch of samples, the values were subtracted from those in the
sample extracts. The method detection limits (MDLs) were calculated
as the mean of the field blank + 3 × SD. In cases when the target
analytes were not detected in the field blanks, the assigned MDL value
was calculated as three times the signal-to-noise ratio of the lowest cal-
ibration. Surrogate recoveries were found to be 85 ± 12% for most re-
ported pesticides but the data reported in this study were not
recovery corrected (see Table SI-2). Since there are no reference mate-
rials for air, the accuracy and precision of the analytical method was
assessed by replicate analyses (n = 6) of a certified indoor dust refer-
ence material (SRM 2585). The results compared very well with the
 B. Gevao et al. / Science of the Total Environment 622–623 (2018) 1621–1629 1623

certified values of the certified pesticides reported in this study (see

Fig. SI-1).

3. Results and discussion

3.1. Introductory remarks

The concentrations of endosulphan, cis-nonachlor, and chlordecone

were not reported since they could not be determined with confidence.
The ΣOC concentrations reported here is the sum of all the other com-
pounds except for these three compounds, and aldrin, which was not
detected in any of the samples. Virtually all the pesticides measured in
this study were associated with the vapor phase and as such data on
particle phase is not reported. The concentrations of individual pesti-
cides measured at all three sampling locations between March 2013
and March 2014 are summarized in Table 1, whereas the ΣOC concen-
trations are presented as box and whisker plots in Fig. 1 for easy visual-
ization. The box contains the middle 50% of the data, whereas the top
and bottom ends of the box represents the 75th and 25th percentiles
of the data set, respectively. The extensions (“whiskers”) at either end
of the box indicate the 99 and 1 percentile and the outliers represent in-
dividual data that exceed 1.5 times the interquartile range. The median
concentrations are indicated by the horizontal line. In general, the size
of the box and the length of the whiskers is an indicator of the variability
in concentrations at a given site for a given compound. A small box
shows that the distribution is uniform over the entire sampling period
and vice versa. Fig. 1. Average concentration (pg m−3) of ΣOCPs determined by high volume air samples
at three locations in Kuwait.

3.2. Atmospheric concentrations of OC pesticides

The average ± SD (and range) of the ΣOC concentrations measured
throughout the study period in decreasing order were urban, 505 ± 305
(range, 33–1352) pg m−3, remote, 204 ± 124 (4.5–556) pg m−3, and
industrial 155 ± 103 (8.8–533) pg m−3 (see Table 1). The concentra-
tions of ΣOCs and those of individual pesticides, at the urban site,
were three fold higher than the concentrations measured at both the in-
dustrial site and the remote site. The elevated concentrations measured
in these samples may possibly be due to the revolatilization from repos-
itories like sediments or localized fugitive, but currently unknown,
sources. The urban site is located on the downwind a power plant,
and near a busy shipping port. The sampler was located at the edge of
the Arabian Gulf with mudflats exposed for long periods during the
day. The biweekly ΣOC concentrations are given in Figs. 2 to 4 for the
urban, industrial, and remote locations, respectively, whereas Figs. 5 to
7 summarize the compound specific concentrations throughout the
year, as box-and whisker plots at these sites. At the urban site, the con-
centrations generally varied between 200 and 600 pg m−3 with dieldrin
being the most abundant pesticide with an average concentrations of

Table 1
Average concentrations (pg m−3) of organochlorine pesticides.

Egaila (industrial sampling site) Shuwaikh (urban sampling site) Abdali (remote sampling site)

Average ± SD Minimum Maximum % DF Average ± SD Minimum Maximum % DF Average ± SD Minimum Maximum % DF

Pentachlorobenzene 7.4 ± 7.2 bd.l. 26.3 93 24.6 ± 24.1 2.2 108 100 8.7 ± 11.3 bd.l. 51.1 92.6
α-HCH 4.7 ± 5.0 0.23 26.5 100 15.4 ± 10.3 2.5 49 100 4.6 ± 3.1 bd.l. 12.7 96.3
Hexachlorobenzene 37.1 ± 42.5 bd.l. 173.4 96 108.4 ± 101 8.7 539 100 39.8 ± 53.8 bd.l. 259.4 96.3
β-HCH 2.9 ± 4.4 bd.l. 56.8 41 16.2 ± 36.7 bd.l. 158 67 0.8 ± 1.1 bd.l. 4.2 74.1
γ-HCH 5.5 ± 6.4 bd.l. 38.6 89 31.5 ± 27 2.5 112 100 3.0 ± 2.1 bd.l. 7.7 88.9
δ-HCH 0.3 ± 0.2 bd.l. 3.6 11 4.1 ± 13.3 bd.l. 65 22 bd.l. bd.l. bd.l. bd.l.
Heptachlor 1.1 ± 0.8 bd.l. 6.8 89 3.5 ± 3.0 bd.l. 13 96 1.1 ± 1.2 bd.l. 5.2 85.2
Oxychlordane 7.5 ± 8.1 bd.l. 62.5 63 20.1 ± 31.5 bd.l. 130 63 5.7 ± 6.7 bd.l. 27.3 63.0
Heptachlor exo-epoxide 0.8 ± 0.7 bd.l. 2.9 85 2.7 ± 2.2 bd.l. 8 89 0.8 ± 0.44 bd.l. 1.8 92.6
trans-Chlordane 1.5 ± 1.4 bd.l. 6.6 93 8.1 ± 5.9 bd.l. 20 96 1.4 ± 1.0 bd.l. 3.4 92.6
o,p′-DDE 1.4 ± 1.1 0.14 4.6 100 4.6 ± 2.1 0.5 10 100 1.5 ± 0.82 0.14 3.5 100.0
trans-Nonachlor 1.7 ± 0.9 0.00 4.4 96 6.0 ± 4.3 bd.l. 16 96 1.6 ± 0.6 bd.l. 2.6 3.7
o,p′-DDD 2.6 ± 3.3 bd.l. 15.6 70 11.1 ± 15.2 bd.l. 75 85 2.9 ± 6.5 bd.l. 33.6 70.4
p,p′-DDE 5.1 ± 3.9 0.04 16.7 100 30 ± 20 2.3 107 100 12.1 ± 8.1 0.15 33.6 100.0
Dieldrin 56.6 ± 59 4.75 212.5 100 161 ± 195 7.0 821 100 53.4 ± 73.4 bd.l. 295.3 92.6
Endrin 1.0 ± 5.1 bd.l. 26.6 96 3.3 ± 12.1 bd.l. 52 7 0.4 ± 2.3 bd.l. 11.9 3.7
p,p′-DDD 2.3 ± 1.7 bd.l. 6.1 78 7.6 ± 7.3 bd.l. 25 26 7.6 ± 7.9 bd.l. 28.5 66.7
o,p′-DDT 2.2 ± 1.5 bd.l. 5.7 85 7.5 ± 8.8 bd.l. 41 85 7.9 ± 7.6 bd.l. 28.1 85.2
cis-Nonachlor 6.5 ± 12.2 bd.l. 51.0 67 20 ± 39 bd.l. 142 56 5.9 ± 8.9 bd.l. 27.0 77.8
p,p′-DDT 6.9 ± 10 bd.l. 44.7 96 19 ± 27 bd.l. 129 7 44.5 ± 61.3 0.53 272.8 77.8
ΣOCs 155 ± 103 8.80 533.5 505 ± 305 33.0 1352 203.8 ± 124 4.49 555.6

SD: standard deviation; %DF: percent detection frequency.

1624 B. Gevao et al. / Science of the Total Environment 622–623 (2018) 1621–1629

Fig. 2. Annual variations in ΣOC concentrations (pg m−3) over the study period at the urban site.

160 (Range, 7–821) pg m−3, followed by hexachlorobenzene with an
average concentration of 108 (Range, 8–540) pg m−3. The seasonal var-
iations in concentrations at the industrial site were not that pro-
nounced, with ΣOC concentrations generally varying between 50 and
200 pg m−3. The dominant pesticides measured at this site were also
dieldrin with a mean concentration of 57 (Range, 5–212) pg m−3 and
hexachlorobenzene with a mean concentration of 37 (Range, bd.l.–
173) pg m−3. At the remote sampling location, the ΣOC concentrations
generally ranged between 100 and 250 pg m− 3. The most abundant
pesticides measured at the remote site were dieldrin with a mean con-
centration of 53 (Range, bd.l.–295) pg m−3 followed by pp′-DDT with a
mean concentration of 45 (Range, 0.5–273) pg m−3 and hexachloro-
benzene with a mean concentration of 40 (b d.l.–259) pg m−3.
The concentrations of ΣOCs was significantly correlated with tem-
perature at the urban monitoring site (R = 0.43, p = 0.025). At the in-
dustrial and remote sites, however, the correlations with temperature
were not significant. At the urban site, the concentrations of all the indi-
vidual pesticides were significantly positively correlated with tempera-
ture (p b 0.05) except for heptachlor (p = 0.509), oxychlordane (p =
0.433), op′-DDE (p = 0.276), pp′-DDT (p = 0.180) and cis-nonachlor
(p = 0.218). Positive correlations with temperature (p b 0.05) observed
for α-, β, γ-hexachlorocyclohexanes, heptachlor expoxide, trans-
chlordane, trans-nonachlor, op′-DDD, pp′-DDE, dieldrin, pp′-DDD, and
op′-DDT suggest that their concentrations may be regulated by air-
surface exchange at this sampling location. The concentrations of
pentachlorobenzene (R = − 0.4, p = 0.038) and hexachlorobenzene
(R = −0.46, p = 0.014) were significantly negatively correlated with
temperature. The concentrations of these two pesticides at the urban
site is clearly not temperature driven. Both of these compounds have
significant combustion sources and are unintentionally released from
thermal processes involving organic matter and chlorine and other
industrial processes (Bailey et al., 2009).
At the industrial sampling site, the concentration of ΣOC was not
correlated with ambient temperature (R = 0.02, p = 0.909) suggesting
that the most likely source of these compounds at this site may be from
long-range transport. The concentrations of pentachlorobenzene, hexa-
chlorobenzene, op′-DDE, and pp′-DDT were negatively correlated with
temperature suggesting that these compounds are likely used in winter
months either in Kuwait or in the region. The concentrations of dieldrin
and pp′-DDD are positively correlated with ambient temperature at

Fig. 3. Annual variations in ΣOC concentrations (pg m−3) over the study period at the industrial site.
 B. Gevao et al. / Science of the Total Environment 622–623 (2018) 1621–1629 1625

Fig. 4. Annual variations in ΣOC concentrations (pg m−3) over the study period at the remote site.

this site suggesting that for these compounds their concentrations may
be driven by air-surface exchange. At the remote sampling location, the
ΣOC concentration was also not correlated with temperature (R =
0.142, p = 0.478). However, the concentrations of op′-DDE, pentachlo-
robenzene and hexachlorobenzene were negatively correlated with
temperature whereas those of pp′-DDD, op′-DDT, and oxychlordane
were positively correlated with temperature (p b 0.05).
To investigate the driving forces modulating the atmospheric con-
centrations of the target compounds at these monitoring sites, we plot-
ted the mean summer/winter ratios for all pesticides analyzed at the
urban, industrial, and remote locations, respectively, and is shown in
Figs. 8, 9, and 10. Mean summertime concentrations were higher for
hexachlorocyclohexanes, chlordanes, dieldrin, and the DDT isomers ex-
cept for pp′-DDT. This provides further evidence that the concentrations

Fig. 5. Average concentrations of pesticides measured at the urban site between March Fig. 6. Average concentrations of pesticides measured at the industrial site between March
2013 and March 2014. 2013 and March 2014.
1626 B. Gevao et al. / Science of the Total Environment 622–623 (2018) 1621–1629
Fig. 7. Average concentrations of pesticides measured at the remote site between March
2013 and March 2014.
of these compounds may be driven by temperature, whereas the atmo-
spheric concentrations of compounds with a summer/winter ratio less
than unity (pentachlorobenzene, hexachlorobenzene, pp′-DDT,
Fig. 8. Summer/winter ratios for various pesticides at the urban site.
heptachlor) may be driven by sources. It is very possible that these com-
pounds are still being used in agriculture within the region.
3.3. Seasonal variability in concentrations of selected pesticides
3.3.1. Hexachlocyclohexane (HCHs)
The mean (and range) ƩHCH concentrations measured at the urban,
industrial, and remote locations were 67 (5 to 383), 13 (0.2 to 125) and
8 (bd.l. to 25) pg m−3, respectively (see Table 1). The average contribu-
tion to the measured ƩHCH from the various isomers in decreasing
order of importance at the urban location was γ-HCH (46%) N β-HCH
(24%) N α-HCH (22%) N δ-HCH (6%). A similar composition was also ob-
served at the industrial site with the highest contribution from γ-HCH
(41%) followed in decreasing order of importance by α-HCH (34%), β-
HCH (21%), and δ-HCH (2%). At the remote location, however, the com-
position was dominated by α-HCH (54%) followed in decreasing order
by γ-HCH (35%), and β-HCH (10%). No δ-HCH was detected at this loca-
tion throughout the year. The variations in ΣHCH concentrations at all
sites are given in Fig. SI-2. The composition of HCHs in this study may
be due to the composition of the formulation that may have been
used in the region. Technical HCH was manufactured and used as a pes-
ticide but banned in many countries in the mid 1970s (Chakraborty
et al., 2010; Cheng et al., 2007; Shunthirasingham et al., 2010). The com-
position of technical mixtures vary between manufacturers but general-
ly comprises of 60 to 70% of α-HCH, 5 to 12% β-HCH, 10 to 12% γ-HCH, 6
to 10% δ-HCH, and 3 to 4% ε-HCHs (Wurl and Obbard, 2005). Technical
HCH was replaced by lindane comprising 99% of the gamma isomer,
which is the only isomer with insecticidal property and traces of the
other isomers (Chakraborty et al., 2010; Cheng et al., 2007). Lindane
and technical HCH were used in the control of insects in fruit, vegeta-
bles, seed treatment, vector control (Chakraborty et al., 2010;
Shunthirasingham et al., 2010) and in health care programs (for the
treatment of head lice and scabies). The mean (and range) in the α/γ-
ratio at the urban site was 0.6 (0 to 1.3), the industrial site 0.8 (0 to
2), and at the remote site the ratio was 1.8 (0.8 to 7). The ratio was
slightly higher in winter than in the summer months. The α-HCH/γ-
HCH ratio in the technical HCH range was between 5 and 7. It is gener-
ally considered that a high α/γ-ratio (N4) is indicative of recent use of
 B. Gevao et al. / Science of the Total Environment 622–623 (2018) 1621–1629 1627

Fig. 9. Summer/winter ratios for various pesticides at the industrial site.

technical HCH. In this study, however, the ratios are very low suggesting
that fresh inputs of technical HCH may not be the source of HCHs the
urban site and the industrial site. At the remote site, on the other
hand, the high α/γ-ratio may suggest revolatilization of technical HCH
that may have been used or the photolytic conversion of γ-HCH to α-
HCH or biological degradation of lindane (Syed et al., 2013).
Hexachlorocyclohexanes (HCHs) were banned in most of Europe
and North America during the 1970s and 1980s but their use continued
in many developing countries in agriculture and in public health
(Barakat, 2004; Barra et al., 2006; Li et al., 1998; Prudente et al., 2007).
Well into the 1990s, for example, the use of technical HCH continued
in India, Vietnam, and other isolated parts of the world (Li, 1999).
Most countries made the switch to γ-HCH (lindane), although it is be-
lieved there are areas still using technical HCH including India, Central
America, and parts of Asia (Voldner and Li, 1995). A wide range of con-
centrations have been reported for these compounds around the globe.
In a global passive sampling survey, concentrations ranging between 17
and 150 pg m−3 were reported for α-HCH and 5 and 79 pg m−3 for γ-
HCH (Harner et al., 2006). Very high concentrations have been reported
from agricultural regions of the world. For example, α-HCH concentra-
tions of 1220 pg m− 3 in Ontario, Canada, 490 to 33,400 pg m− 3 in
Indian cities (Chakraborty et al., 2010), and 41 to 313 pg m− 3 in

Fig. 10. Summer/winter ratios for various pesticides at the remote site.
1628 B. Gevao et al. / Science of the Total Environment 622–623 (2018) 1621–1629
Urban Guangzhou, China have been reported. A wide range of concen-
trations have also been reported in Asia for γ-HCHs. For example,
Chakraborty et al., 2010 reported concentrations between 960 and
97,400 pg m−3 from major cities of India, whereas in urban areas of
Guangzhou, China the concentrations ranged between 5 and
2640 pg m−3. The concentrations of HCHs in this study are far lower
than those reported in other parts of the world indicating that HCH is
not currently used in Kuwait.
3.3.2. Dichlorodiphenyltrichloroethane (DDT)
The annual variations in the concentration of ΣDDT [DDT and
metabolites dichlorodiphenyldichloroethylene (DDE) and
dichlorodiphenyldichloroethane (DDD)] are given in Figs. SI 3–5 for
the industrial, urban, and remote sites respectively. The ΣDDT concen-
trations measured at the remote site were the highest with a mean
(and range) of 52 (1 to 272) pg m− 3, followed by the levels at the
urban site with a mean (and range) of 27 (1.8 to 129) pg m− 3, and
the lowest concentrations were measured at the industrial site with a
mean (and range) of 9 (0.29 to 45) pg m−3 (see Table 1). The DDT con-
centrations at all sampling sites were elevated in the winter months
with concentrations decreasing in the summer. The levels during this
study are lower than those reported previously (Chakraborty et al.,
2010), similar to those reported at a remote observatory in China
(Cheng et al., 2007) and in Singapore (Wurl and Obbard, 2005), but
two orders of magnitude lower than those reported recently in India
(Shunthirasingham et al., 2010). The period of peak concentrations co-
incides with periods of intense agricultural activities in Iraq and Egypt.
As air mass back trajectories suggest that air sampled in Kuwait pre-
dominantly originates from Egypt and Iraq (Gevao et al., 2010), the
most likely source of these compounds in Kuwait is their long-range
transport from these regions. DDT metabolizes in soils to DDE and
DDD. In general the DDT/DDE ratio of b1 indicates aged DDT, whereas
fresh applications of DDT will give a DDT/DDE ratio N1 (Chakraborty
et al., 2010; Shunthirasingham et al., 2010). In this study, the DDT/
DDE ratio at the urban site of 0.95 (range, 0.2 to 3.1) indicates that
there are no fresh sources of DDT at this location. At the industrial and
remote sites, however, the DDT/DDE ratios of 2.1 (range, 0.39 to 7.6)
and 4.1 (0.41 to 19) respectively, (Fig. SI-5) suggest fresh inputs of
DDT, mainly at the remote site and particularly during the winter
months, which coincide with the period of agricultural activities in
Egypt. DDT is used in agriculture and to control mosquitoes, the primary
vector for malaria and dengue fever, and the levels measured during
this study may be suggestive of its recent use. It seems more likely
that the presence is due to its use in agriculture and, or for disease vector
control, because the levels would have been higher in summer if the
source was volatilization from stockpiles.
3.3.3. Drins
Endrin, dieldrin, aldrin are all listed in Annex A of the Stockholm
Convention. They are a group of banned pesticides. Aldrin was not de-
tected in any sample. The concentrations of Endrin were very low at
all sampling locations. As shown in Table 1, the mean (and range) in
concentrations measured throughout the study period at the industrial
was 1.0 ± 5.1 (bd.l. to 26) pg m−3, whereas at the remote site, the con-
centrations ranged from b d.l. to 12 pg m−3. The concentrations were
slightly higher at the urban site with a mean (and range) of 3 ± 12
(b d.l. to 52) pg m−3. The concentration of dieldrin on the other hand
was high at all sites especially at the urban site with a mean
(and range) in concentration of 160 ± 195 (7 to 820) pg m−3. The con-
centrations of dieldrin at both the remote and industrial sites were sim-
ilar with concentrations of 53 ± 73 (b d.l. to 295) pg m−3 and 57 ± 59
(5 to 212) pg m− 3, respectively. These concentrations are low com-
pared with reported concentrations in countries where they were
used intensively as a soil insecticide.
3.3.4. Chlordanes
The concentrations of Ʃchlordanes (cis-chlordane, trans-chlordane,
cis-nonachlor, trans-nonachlor) in the samples were very low at the re-
mote and industrial sites with mean and range in concentrations being
4 (0.41 to 19) pg m−3 and 2 (0.39 to 7.6) pg m−3 respectively. These
concentrations at these two sampling locations are similar to those re-
ported for European background air (Halse et al., 2011). At the urban
site however, the concentrations of Σchlordanes were slightly higher
at mean and range in concentrations being 35 (b d.l. to 156) pg m−3.
The chlordane concentrations were elevated between March and July
at all sampling locations (Fig. SI-7). The chlordane concentrations mea-
sured at the urban site are identical to those reported for agricultural
sites near Delhi (Pozo et al., 2011) but much lower than those reported
for several Chinese cities (Liu et al., 2009).
3.3.5. Pentachlorobenzene and hexachlorobenzene
The annual variations in the concentrations of pentachlorobenzene
and hexachlorobenzene are plotted in Figs. SI8–10 for the remote, in-
dustrial and urban sites, respectively. Hexachlorobenzene concentra-
tions are generally higher than those of pentachlorobenzene and are
both strongly negatively correlated with temperature (p b 0.05) at all
sampling locations. The concentrations of both pentachlorobenzene
and hexachlorobenzene increase during the cold winter months. Con-
centrations for pentachlorobenzene ranged from 2 to 108 pg m− 3 at
the urban sampling site with a mean of 25 pg m− 3; at the industrial
site, the concentrations ranged between bd.l. and 26 pg m− 3 with a
mean of 7.5 pg m−3 whereas at the remote site the mean (and range)
in concentrations was 8.7 (bd.l. to 51) pg m−3. The levels of pentachlo-
robenzene in this study are similar to that reported in a global passive
sampling survey (Koblizkova et al., 2012) and another similar survey
in North America (Shen et al., 2005). Pentachlorobenze (PeCBz) is the
most volatile among the POPs that have so far been added to the
Stockholm Convention. It is not known to be manufactured for commer-
cial use, but it is produced as a byproduct of a range of industrial pro-
cesses and is also produced during solid waste combustion and
biomass burning (Bailey et al., 2009). Bailey et al. (2009) recently pub-
lished a detailed review on the “sources and prevalence of PeCBz in the
environment”. The concentration of PeCBz in the global atmosphere
was recently reported to range between 2.5 and 75 pg m−3 using SIP
disk passive samplers (Koblizkova et al., 2012). PeCBz is thought to be
uniformly distributed in the global environment as a result of its volatil-
ity, emission history, and long atmospheric residence time (Bailey et al.,
2009; Koblizkova et al., 2012).
The concentrations of hexachlorobenzene on the other hand, were
much higher. Mean (and range) in the concentrations in decreasing
order at the three sampling sites were as follows: urban, 108 ± 101
(8.7 to 539) pg m−3; remote, 39.8 ± 53.8 (bd.l. to 259) pg m−3), and
industrial, 37 ± 42 (b d.l. to 173) pg m−3. The concentrations of hexa-
chlorobenzene measured at the industrial and remote sites are similar
to those reported for European background air (Halse et al., 2011).
The concentrations of hexachlorobenzene, particularly in winter, are
similar to those reported in Asia. Recently, Koblizkova et al. (2012) re-
ported hexachlorobenzene concentrations ranging between 37 and
240 pg m−3 on a continental scale in line with the concentrations mea-
sured in this study. Similar concentrations to those reported here were
also reported in another global atmospheric pesticide survey
(Shunthirasingham et al., 2010). The presence of hexachlorobenzene
in the environment is linked to its use as fungicide on seeds, as a
wood preserving agent and several industrial processes, like its use in
dye manufacturing (Barber et al., 2005). Hexachlorobenzene is also
present as an impurity in the manufacture of chlorinated solvents and
pesticides and can also be produced during combustion of solid waste
and biomass burning (Liu et al., 2009). Hexachlorobenzene is
also known to be used as a fungicide. The concentrations of
pentachloronitrobenzene and hexachlorobenzene are also very strongly
 B. Gevao et al. / Science of the Total Environment 622–623 (2018) 1621–1629
positively correlated with each other at all sampling locations (R N 0.85;
p b 0.001) suggesting that both compounds have an identical source.
4. Conclusions
High-volume air samplers were used to concomitantly measure the
concentrations of OC pesticides at three sites in Kuwait with the view of
establishing baseline concentrations against which the effectiveness of
global control measures to reduce the levels of these compounds can
be assessed. The concentrations of most OCs measured in this study, ex-
cept for DDT and its metabolites, were higher at urban locations relative
to their concentrations at remote location, in line with the literature on
POPs regarding urban conurbations being sources of industrial
chemicals. The most abundant pesticides measured throughout this
study were dieldrin, pentachlorobenzene, hexachlorobenzene, ΣDDTs,
and ΣHCHs, and oxychlordane. Hexachlorobenzene concentrations are
generally higher than those of pentachlorobenzene and are both
strongly negatively correlated with temperature (p b 0.05) at all sam-
pling locations. Mean summertime concentrations were higher for
hexachlorocyclohexanes, chlordanes, dieldrin, and DDT isomers, except
for pp.-DDT. This may suggest that concentrations of these compounds
may be driven by temperature. The DDT concentrations at all sampling
sites were elevated in the winter months with concentrations decreas-
ing in the summer. The DDT/DDE ratio at the urban site of 0.95 (range,
0.2 to 3.1) indicates that there are no fresh sources of DDT at this loca-
tion. At the industrial and remote sites, however, the DDT/DDE ratios
of 2.1 (range, 0.39 to 7.6) and 4.1 (0.41 to 19), respectively, suggest
fresh inputs of DDT particularly at the remote site during the winter
months, which coincides with the period of agricultural activities in
the region.
Acknowledgements
We are grateful to the managements of Kuwait Institute for Scientific
Research, the Kuwait Foundation for the Advancement of Science, and
the Kuwait Environmental Public Authority for funding various aspects
of the data discussed in this paper.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.scitotenv.2017.10.036.
References
Alamdar, A., Syed, J.H., Malik, R.N., Katsoyiannis, A., Liu, J., Li, J., Zhang, G., Jones, K.C., 2014.
Organochlorine pesticides in surface soils from obsolete pesticide dumping ground in
Hyderabad City, Pakistan: contamination levels and their potential for air-soil ex-
change. Sci. Total Environ. 470-471, 733–741.
Bailey, R.E., van Wijk, D., Thomas, P.C., 2009. Sources and prevalence of pentachloroben-
zene in the environment. Chemosphere 75 (5), 555–564.
Barakat, A.O., 2004. Assessment of persistent toxic substances in the environment of
Egypt. Environ. Int. 30, 309–322.
1629
Barber, J.L., Sweetman, A.J., Van Wijk, D., Jones, K.C., 2005. Hexachlorobenzene in the glob-
al environment: emissions, levels, distribution, trends and processes. Sci. Total Envi-
ron. 349, 1–44.
Barra, R., Colombo, J.C., Eguren, G., Gamboa, N., Jardim, W.F., Mendoza, G., 2006. Persistent
organic pollutants (POPs) n Eastern and Western South American countries. Rev. En-
viron. Contam. Toxicol. 185, 1–33.
Chakraborty, P., et al., 2010. Selected organochlorine pesticides in the atmosphere of
major Indian cities: levels, regional versus local variations, and sources. Environ. Sci.
Technol. 44 (12), 8038–8043.
Cheng, H., et al., 2007. Organochlorine pesticides, polybrominated biphenyl ethers and
lead isotopes during spring time at the Waliguan Baseline Observatory, northwest
China: implication for long-range atmospheric transport. Atmos. Environ. 41,
4734–4747.
Daley, J.M., Corkum, L.D., Drouillard, K.G., 2011. Aquatic to terrestrial transfer of sediment
associated persistent organic pollutants is enhanced by bioamplification processes.
Environ. Toxicol. Chem. 30 (9):2167–2174. https://doi.org/10.1002/etc.608.
Fang, Y., Nie, Z.Q., Die, Q., Tian, Y., Liu, F., He, J., Huang, Q., 2017. Organochlorine pesticides
in soil, air, and vegetation at and around a contaminated site in southwestern China:
concentration, transmission, and risk evaluation. Chemosphere 178, 340–349.
Gevao, B., Jaward, F.M., MacLeod, M., Jones, K.C., 2010. Diurnal fluctuations in
polybrominated diphenyl ether concentrations during and after a severe dust storm
episode in Kuwait City, Kuwait. Environ. Sci. Technol. 44 (21), 8114–8120.
Halse, A.K., Schlabach, M., Eckhardt, S., Sweetman, A., Jones, K.C., Breivik, K., 2011. Spatial
variability of POPs in European background air. Atmos. Chem. Phys. 11, 1549–1564.
Harner, T., et al., 2006. Global pilot study for persistent organic pollutants (POPs) using
PUF disk passive air samplers. Environ. Pollut. 144 (2), 445–452.
Ikemoto, T., et al., 2008. Analysis of biomagnification of persistent organic pollutants in
the aquatic food web of the Mekong Delta, South Vietnam using stable carbon and ni-
trogen isotopes. Chemosphere 72 (1), 104–114.
Kelly, B.C., Ikonomou, M.G., Blair, J.D., Morin, A.E., Gobas, F.A., 2007. Food web-specific
biomagnification of persistent organic pollutants. Science 317 (5835), 236–239.
Koblizkova, M., Genualdi, S., Lee, S.C., Harner, T., 2012. Application of sorbent impregnated
polyurethane foam (SIP) disk passive air samplers for investigating organochlorine
pesticides and polybrominated diphenyl ethers at the global scale. Environ. Sci.
Technol. 46 (1), 391–396.
Li, Y.-F., 1999. Global technical hexachlorocyclohexane usage and its contamination con-
sequences in the environment: from 1948 to1997. Sci. Total Environ. 232, 121–158.
Li, Y.F., Cai, D.J., Singh, A., 1998. Technical hexachlorohexane use trends in China and their
impact on the environment. Arch. Environ. Contam. Toxicol. 35, 688–697.
Liu, X., et al., 2009. Seasonal patterns and current sources of DDTs, chlordanes, hexachlo-
robenzene, and endosulfan in the atmosphere of 37 Chinese cities. Environ. Sci.
Technol. 43, 1316–1321.
Meng, X.Z., Pan, Z.Y., JJ, Wu, Qiu, Y.L., Chen, L., Li, G.M., 2011. Occurrence of
polybrominated diphenyl ethers in soil from the central Loess Plateau, China: role
of regional range atmospheric transport. Chemosphere 83 (10), 1391–1397.
Pozo, K., et al., 2011. Assessing seasonal and spatial trends of persistent organic pollutants
(POPs) in Indian agricultural regions using PUF disk passive air samplers. Environ.
Pollut. 159 (2), 646–653.
Prudente, M., Malarvanna, G., Tanabe, S., 2007. Persistent toxic substances in the
Philippine environment. In: Li, A., Tanabe, S., Jiang, G., Giesy, J.P., PKS, Lam (Eds.), Per-
sistent Organic Pollutants in Asia Sources Distribution, Transport and Fate. Advances
in Environmental Sciences. vol. 7. Elsevier Ltd., pp. 559–585.
Scheringer, M., Salzmann, M., Stroebe, M., Wegmann, F., Fenner, K., Hungerbuhler, K.,
2004. Long-range transport and global fractionation of POPs: insights from multime-
dia modeling studies. Environ. Pollut. 128 (1–2), 177–188.
Shen, L., Wania, F., Lei, Y.D., Teixeira, C., Muir, D.C.G., Bidleman, T., 2005. Atmospheric dis-
tribution and long range transport behaviour of organochlorine pesticides in North
America. Environ. Sci. Technol. 39, 409–420.
Shunthirasingham, C., et al., 2010. Spatial and temporal pattern of pesticides in the global
atmosphere. J. Environ. Monit. 12, 1650–1657.
Syed, J.H., et al., 2013. Organochlorine pesticides in air and soil and estimated air-soil ex-
change in Punjab, Pakistan. Sci. Total Environ. 444, 491–497.
Voldner, E.C., Li, Y.-F., 1995. Global usage of selected persistent organochlorines. Sci. Total
Environ. 160 (/161), 201–210.
Wurl, O., Obbard, J.P., 2005. Organochlorine compounds in the marine atmosphere of
Singapore. Atmos. Environ. 39, 7202–7216.