Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77

Contents lists available at ScienceDirect

Journal of Atmospheric and Solar–Terrestrial Physics

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

Temporal variation of Black Carbon concentration using Aethalometer MARK

observations and its relationships with meteorological variables in Karachi,
Pakistan
⁎
Samina Bibia, Khan Alama, , Farrukh Chishtieb, Humera Bibia, Said Rahmanc
a
Department of Physics, University of Peshawar, Khyber Pakhtunkhwa, Pakistan
b
Theoretical Research Institute, Pakistan Academy of Sciences, Islamabad, Pakistan
c
Pakistan Space and Upper Atmosphere Research Commission (SUPARCO), P.O. Box 8402, Off University Road, Karachi 75270, Pakistan

A R T I C L E I N F O A BS T RAC T

Keywords: Black Carbon (BC) mass concentration was measured continuously for every five-minute interval with ground-
Black Carbon based Aethalometer at an urban site in Karachi for the period from 2006 to 2008. In this study, the temporal
Aethalometer (diurnal, monthly and seasonal) variations of BC and its relationship with meteorological variables were
Temporal variation analyzed. Monthly averaged concentrations of BC ranged from 2.2 to 12.5 µg/m3, with maximum in the month
Karachi, Meteorological variables
of January 2007 and minimum in the month of June 2006. BC showed higher concentrations during the months
of January, February and November while lower during the months of May, June, July and August throughout
the years. It also displayed comparatively high concentrations during winter and postmonsoon, while moderate
during premonsoon and low during summer. Diurnal analysis of BC concentration showed sharp peaks between
07:00 and 09:00 LST and again around 22:00 during all the months. Moreover, the relationship between BC
concentration and meteorological variables such as Temperature (Temp), Relative Humidity (RH), Wind Speed
(WS), Visibility (VIS) and RainFall (RF) was found and it was observed that BC concentration showed an inverse
relationship with all these meteorological variables. Finally, the analysis of the Hybrid Single Particle
Lagrangian Integrated Trajectory (HYSPLIT) cluster trajectories revealed that almost all the clusters were
originating from southwest of the study site.

1. Introduction climate system by affecting the radiative balance of the atmosphere

Atmospheric aerosol or Particulate Matter (PM) is a complex mixture
of tiny particles which consists of solid fragments and small liquid
droplets. Black Carbon (BC) is one of the important light absorbing
component (~ 90% in PM2.5) of airborne PM, which is directly emitted
from incomplete combustion of carbonaceous fuel (fossil fuel, biomass
and agricultural waste and forest fire) (Cao et al., 2009; Tiwari et al., 2013;
Guha et al., 2015; Li et al., 2016). BC emissions vary by changing use of
fossil fuels and development of technology. In developing countries, the
estimated emission of BC is the highest due to the fossil fuel (Novakov
et al., 2003; Ramachandran and Kedia, 2010). The variation in fuel
consumption has produced large temporal variations in the absorption of
solar radiation by aerosols suggesting that absorbing aerosols in the
atmosphere such as BC may have taken a significant part in global
temperature variation in the last century (Sreekanth et al., 2007).
Much attention has been given to BC because it is a strong absorber
of shortwave solar radiations and plays a significant role in the Earth's
(IPCC, 2013; Wang et al., 2016). According to Petzold et al. (2013),
almost all light absorbing carbonaceous elements present in the
atmosphere can be qualitatively defines as BC. In fact, BC is considered
to be the second strongest contributor to current global warming next
to carbon dioxide (Bergstrom et al., 2007; Dumka et al., 2013;
Surendran et al., 2013). It absorbs the incoming solar radiation, heats
up the atmosphere and produces a heating effect in contrast to other
aerosols (sulfates, etc.) which reflect solar radiations leading to a
cooling effect (Bond et al., 2013; Adesina et al., 2015). BC causes
harmful health effects, reduces crop yields, contaminates construction
materials and adversely affects terrestrial ecosystems, because it
absorbs toxic substances due to its porous and absorptive nature
(Cao et al., 2009; Tiwari et al., 2013). Apart from this, it can also alter
precipitation, cloud lifetime, reflectivity and melting of snow and ice
(Dumka et al., 2013). The long lifetime (≳1 week in the lower
troposphere (Babu and Moorthy, 2001) allows BC aerosols long-range
transportation (Tiwari et al., 2013). BC aerosol particles are chemically

⁎
Corresponding author.
E-mail address: khanalamso@gmail.com (K. Alam).

http://dx.doi.org/10.1016/j.jastp.2017.03.017
Received 19 December 2016; Received in revised form 23 February 2017; Accepted 7 March 2017
Available online 05 April 2017
1364-6826/ © 2017 Elsevier Ltd. All rights reserved.
S. Bibi et al.
inert and the main removal process is wet deposition due to its fine size
(Aruna et al., 2013).
Interest in BC has increased because of its fine size, inert chemical
characteristics and absorptive nature. These features combined with
the growing sources of the BC, make it necessary to give proper and
appropriate attention to the monitoring of these aerosols from different
regions of South Asia. In recent years, carbonaceous aerosols have
become one of the key research areas, as it has many climatic impacts.
Therefore, numerous studies have been conducted related to BC
aerosols over different regions of the world (Cao et al., 2009; Aruna
et al., 2013; Tiwari et al., 2013; Wu et al., 2016), but very few studies
could be traced regarding carbonaceous aerosols in Pakistan and these
are conducted over a very short period of time (Husain et al., 2007;
Dutkiewicz et al., 2009). Due to lack of ground observations, therefore,
the present study has the potential to improve our knowledge about BC
aerosols in different processes in Pakistan and also their impacts on
regional and global climate change.
In this study, for the first time, we present a comprehensive data-
sets on BC aerosols, measured for period of three years (2006–2008)
covering the four seasons (winter, summer, premonsoon and post-
monsoon) for an urban site (Karachi). The temporal variations of BC
concentration in diurnal, monthly, seasonal and scales were examined.
These variations were explained on the basis of changing surface
meteorological variables such as Temperature (Temp), Relative
Humidity (RH), Wind Speed (WS), Wind Direction (WD), Visibility
(VIS) and RainFall (RF) and the relationships between these variables
and BC concentration were also analyzed.
2. Research methodology
2.1. Site description and local meteorology
Karachi is a major city of Pakistan situated (Lat 24° 51′N; Long 67°
02′E) on the coast of Arabian Sea. It has a large industrial base, with
rapid urbanization and growing number of factories, vehicles as well as
power plants, which are major sources of BC emission. The sampling
site is northwest of Karachi Airport, which is located at the University
of Karachi campus. Though moderate vehicular traffic is present in the
campus, but the study area is adjacent to highways which have heavy
traffic load. The region experiences four major seasons in each year:
winter (December-February), pre-monsoon (March-May), summer
(June-August) and postmonsoon (September-November). During win-
ter, an increase in BC concentration occurs due to frequent burning
processes mainly done by the local population for warmth against cold.
The data of meteorological variables used in the present study were
obtained from the Pakistan Meteorological Department, Karachi.
Monthly averaged Temp, RH, RF and VIS are shown in Fig. 1. The
monthly averaged Temp was in the range of 17–33 °C with high during
the June-August and low during December-February. The monthly
averaged RH showed strong monthly variability i.e. high during the
summer months and low during winter months. The RH was lowest in
January 2008 (43.1 ± 11.3%) and highest in August 2006 (81.5 ±
4.4%). It was observed that there was no prominent seasonal variation
in VIS. The rainy season in Karachi is from July to September. During
the study period from 2006 to 2008, the total RF was about 827 mm
out of which 80% fall between July and September with maximum
(250 mm) RF recorded in the month of August 2007.
Fig. 2(a–d) shows the averaged surface WS and WD during
summer, winter, premonsoon and postmonsoon. The seasonal cycle
of wind speed was significant with higher values during summer
(4.5 m/s) and premonsoon (3.6 m/s) followed by postmonsoon
(2.4 m/s) and weaker during the winter season (1.8 m/s). Analysis of
these wind datasets revealed that the observational site was mostly
under the influence of southwesterly winds. Whereas during winter,
some of the wind patterns start shifting in the direction from southwest
to northeast.
68
Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77
2.2. Aethalometer
The BC concentration measurements were obtained with a Magee
Scientific Model AE21 Aethalometer, which was operated from a second
floor window of the HEJ Research Institute of Chemistry building at the
University of Karachi campus. Aethalometer measures BC using a light
beam of high intensity at wavelengths of 880 and 370 nm by measuring
the its attenuation caused by BC particles that accumulate on a rolled
quartz filter strip with a cellulose fiber backing (Dutkiewicz et al., 2009).
The mass concentrations were measured at 880 nm because BC is a major
absorber at this wavelength and this wavelength signifies an accurate
value of BC in the atmosphere. Aethalometer uses an optical transmission
and continuous filtration method for the real-time measurement of BC
concentration (Singh et al., 2010). A vacuum pump sucks air continuously
for the particles accumulated on filter strip of Aethalometer. The light
beam passes through the unloaded filter strip first and is then compared
with the loaded filter. In this study, the measurement sampling interval
was kept at five minutes from March 2006 to December 2008 at the
880 nm channel. BC concentration is calculated by measuring the change
in the transmittance through the quartz filter at which the BC particles are
deposited (Hansen et al., 1984). Attenuated light is linearly proportional
to the concentration of BC, which is placed on filter. The Aethalometer
was operated with a cyclone inlet (BGI corporation) equipped with an
insect and rain guard at a flow rate of 4 l/min making the cut-point ~
3.2 µm. It is noted that there are several systematic errors in filter based
absorption methods used for the measurement of BC concentration that
need to be corrected (Bond et al., 1999). The uncertainty in the
measurement of BC concentration is ~ 10% (Babu and Moorthy, 2002)
and particularly the absorption coefficient used for the 880 and 370 nm
are 16.6 and 39.5 m2/g, respectively (Dutkiewicz et al., 2009). The various
uncertainties in the measurement of BC mass concentration using
Aethalometer arises due to the changes in filter scattering occurs by
aerosol loading and by increasing the filter loads, the Aethalometer
measuring signals reduces i.e. BC concentrations (Ramachandran and
Kedia, 2010).
3. Results and discussion
3.1. Temporal variation of BC concentration
BC aerosol mass concentrations measured for every five-minute
interval within 24 h per day during each month were averaged for the
entire study period (2006–2008) over Karachi. Monthly averaged BC
mass concentration and associated standard deviation over Karachi for
period 2006–2008 are tabulated in Table 1. Fig. 3 shows the monthly
averaged BC concentration with maximum (12.5 µg/m3) in the month
of January 2007 and minimum (2.2 µg/m3) in the month of June 2006.
It is clear from the figure that overall concentrations were higher
during the months of January, February and November while lower
during the months of May, June, July and August throughout the years
but with varying magnitudes. One of the reasons for high BC
concentration is indoor heating due to coal burning activities (Cao
et al., 2009). Moreover, local sources such as industrial and vehicular
emissions also take part in high BC concentration (Safai et al., 2007).
On the other side, the enriched wet deposition during monsoon rainfall
is responsible for the low concentration of BC (Zhang et al., 2010; Lu
et al., 2011; Li et al., 2016). Our results are comparable with the
previous findings over Ahmedabad, in which monthly averaged BC
concentration was enhanced in December (13.8 µg/m3) and reduced in
June (1.6 µg/m3) (Ramachandran and Kedia, 2010). Dutkiewicz et al.
(2009) also found higher values during November to February (~
10 µg/m3) and lower during June to September (~ 2 µg/m3) over
Karachi during 2006–2007. The results show that BC concentration
increases year by year due to rapid increase in urbanization, vehicular
emissions and deforestation. It is to be mentioned here that no
measurements of BC concentration data were made for the months
S. Bibi et al. Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77

Fig. 1. Monthly variation of Temp, RH, RF and VIS over Karachi during March 2006 to December 2008.

of January, February and December in 2006 due to the instrumental
failure as seen from the figure.
Recently, Li et al. (2016) documented the monthly averaged with
maximum BC concentration above 9 µg/m3 in December, January and
February and minimum concentration below 5 µg/m3 in June, July and
August in China. Similarly, Surendran et al. (2013) found the highest
monthly averaged BC concentration (15.9 µg/m3) in December 2010
and lowest (2.44 µg/m3) in July 2011 over Delhi. It should be noted
that the extreme maximum of monthly averaged BC concentration
occurred during December 2006 and extreme minimum during August
having the values 16.7 and 3.26 µg/m3, respectively, which are higher
than our result (Singh et al., 2010). Dumka et al. (2013) reported the
lowest BC concentration (1.88 µg/m3) in July and the highest (7.08 µg/
m3) in January over Hyderabad. The monthly averaged BC concentra-
tions lie between 3.1–14.8 µg/m3 with maximum concentration in
December 2004 and a minimum in July 2007 (Wu et al., 2009).

Fig. 2. Ground based surface WS and WD during different seasons (a) summer, (b) winter, (c) premonsoon and (d) postmonsoon over Karachi during March 2006 to December 2008.
The colored scale represents WS in m/s. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

69
S. Bibi et al. Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77

Table 1
Monthly averaged BC mass concentration and associated standard deviation over
Karachi for period 2006–2008.

Month BC concentration µg/m3 ± Standard deviation

2006 2007 2008

January 12.5 ± 9.7 7.6 ± 2.7

February 7.4 ± 6.7 10.1 ± 7.3
March 5.2 ± 2.6 5.9 ± 4.7 7.6 ± 4.7
April 3.2 ± 1.9 5.9 ± 3.8 5.9 ± 4.7
May 2.3 ± 1.6 3.8 ± 1.6 2.9 ± 1.5
June 2.2 ± 1.5 2.9 ± 1.4 3.5 ± 2.2
July 2.5 ± 1.2 3.2 ± 1.8 3.1 ± 2.1
August 3.1 ± 1.7 3.1 ± 1.4 3.1 ± 1.6
September 3.8 ± 3.5 4.2 ± 5.4 6.1 ± 5.9
October 5.2 ± 4.5 8.4 ± 6.9 5.6 ± 5.0
November 8.8 ± 6.7 9.9 ± 8.0 10.5 ± 8.4
December 7.5 ± 6.9 4.89 ± 2.3

Fig. 4. Seasonal variation of BC concentration over Karachi during March 2006 to

December 2008.

Fig. 3. Monthly averaged BC concentration over Karachi during March 2006 to
December 2008.
In addition to annual variation, BC concentrations also shows
robust seasonal variations. The monthly averaged BC mass concentra-
tions were in the range of 4.8–12.5 µg/m3, 2.3–5.9 µg/m3, 2.2–3.5 µg/
m3 and 3.7–10.5 µg/m3 during winter, premonsoon, summer and
postmonsoon, respectively. Fig. 4 reveals that high BC concentrations
were observed during winter (8.3 ± 2.4 µg/m3) and postmonsoon (6.9
± 2.3), while moderate concentrations were found during premonsoon
(4.6 ± 1.4) and lowest concentrations were obtained during summer
(2.9 ± 0.4). It was also noted that the BC concentration was highest
with pronounced maximum value (12.5 µg/m3) during winter in 2007
and lowest (3.5 µg/m3) during summer in 2006. During winter, an
increase in BC concentration occurred to anthropogenic activities
mainly because of frequent burning processes against the cold weather
(Safai et al., 2007; Dumka et al., 2013). Biomass burning is the major
source of BC, which is responsible for the production of fine particles in
winter (Tare et al., 2006). In addition, the boundary layer is narrow
during winter due to low temperature and provides a small area which
keeps the pollutants close to the surface resulting in higher BC
concentration as compared with summer (Nair et al., 2007). In contrast
to winter the lowest BC concentration may be attributed to scavenging
processes during summer (Surendran et al., 2013). Low BC concentra-
tion during premonsoon could be attributed to the significantly strong
winds in this season, while during postmonsoon, weak winds and less
precipitation causes comparatively high BC concentration
(Ramachandran and Kedia, 2010). Similar seasonal variations were
observed in Ahmedabad by Ramachandran and Kedia (2010) and
recorded the averaged seasonal concentrations during winter, premon-
soon, summer and postmonsoon as 11.6, 3.9, 2.1 and 10.9 µg/m3,
respectively. Nair et al. (2007) documented a pronounced concentra-
tion in winter than other seasons over Kanpur. It should be noticed that
winter season is known for static meteorological conditions and man-
made activities which causes the emission of large amount of fine
particles and increased impacts of aerosols including BC. Safai et al.
(2007) studied the BC concentration over Pune and found these were
high during winter and postmonsoon (7.3 and 6.04 µg/m3) as com-
pared to summer and pre-monsoon (3.2 and 1.3 µg/m3) which could be
attributed as predominant meteorological conditions of the particular
seasons. Our results are comparable with that of Tiwari et al. (2013)
observed over New Delhi, where concentration was a maximum during
winter (10.8 µg/m3) and post-monsoon (9.4 µg/m3) while minimum
during premonsoon (3.5 µg/m3) and monsoon (2.8 µg/m3). Recently,
Dumka et al. (2013) analyzed a similar seasonal variation and
documented that the BC mass concentration was 1.0 µg/m3 during
summer and this concentration has been increased to 12.5 µg/m3
during winter. The BC concentrations in China show significant
seasonal variations, varied from the lower values of about 1–5 µg/m3
during summer to a higher value up to 9 µg/m3 during winter in the
North China plain and the Sichuan Basin (Li et al., 2016). The observed
trend whereby BC concentration maximizes during winter and drops
during summer can be explained by varying emission sources, meteor-
ological condition and height of mixed layer depths (Cao et al., 2009).
The seasonal variation plays an important role to regulate the BC
concentration, which is maximum in winter and minimum in summer
(Gabbi et al., 2015). Similarly, summer minima and winter maxima
were obtained over Netherlands by Schaap and van der Gon (2007)
showing similarity in the seasonal variation. Sreekanth et al. (2007)
measured the averaged BC concentration over eastern India and found
that the concentration was relatively higher (8.01 µg/m3) during winter
as compare to summer (3.33 µg/m3), monsoon (1.67 µg/m3) and post
monsoon (0.43 µg/m3).
The monthly averaged diurnal variations of BC mass concentrations
were obtained from hourly averaged data during the entire study
period, which are shown in Fig. 5(a-i). In general, distinct diurnal
patterns show the steady build up in the BC concentration to sharp
peaks between 07:00 and 09:00, LST just after sunrise preceded by a
smooth decrease to low concentration at 11:00 to 17:00, LST and then
again starts increasing to attain another peak values around 22:00, LST
after sunset during all the months. The highest morning peak (23.2 µg/
m3) was found during January 2007 and lowest peak (2.9 µg/m3)
during June 2007. This BC diurnal profile of morning peaks was very
comparable to that of documented at Pune by Safai et al. (2007),
although the evening peaks measured here are higher in magnitude.

70
S. Bibi et al. Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77

Fig. 5. Monthly averaged diurnal variation of BC mass concentration over Karachi during March 2006 to December 2008.

It was observed that morning peaks were more pronounced with
variable magnitudes as compared with evening peaks for all months.
The relatively higher BC concentration is during morning, and is
attributed mainly due to the increase in anthropogenic activities (i.e.
biomass burning, industrial activities, vehicular emission) along with
the breaking of the nocturnal boundary layer resulting in the lifting up
the fine particles such as BC (Stull, 2012). At noon hours, solar heat
becomes intense and boundary layer becomes denser which help uplift
the pollutants, resulting in the constant decrease in BC concentration.
Later, the, concentration starts building up smoothly attaining an
evening peak due to the dissipation of the local boundary layer. Similar
variations in concentration which are affected by the varying local

71
S. Bibi et al. Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77

Fig. 5. (continued)

boundary layer were observed by Dumka et al. (2013) over Hyderabad,
Tiwari et al. (2013) over New Delhi and Surendran et al. (2013) over
Delhi. Aruna et al. (2013) observed by obtaining the same monthly
diurnal patterns that the primary morning peaks were broader from
May to December as compared with other months and the evening
peaks were noticeable from the months of October to February over
Chennai. The monthly diurnal BC trends reported in Karachi were
analogous to that measured over other locations such as, Nainital
(Srivastava et al., 2006), Lahore (Husain et al., 2007), Kanpur (Tripathi
et al., 2005) and Hefie (Zhang et al., 2015).

72
S. Bibi et al. Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77

evening around 21:00 to 23:00 LST, during winter, premonsoon,

summer and postmonsoon. Maximum hourly BC concentrations at
morning were about 15.3, 8.5, 4.3 and 12.4 µg/m3 and at evening were
around 10.6, 5.6, 3.2 and 9.0 µg/m3 during winter, premonsoon,
summer and postmonsoon, respectively displaying the highest value
during winter and the lowest value during summer which is in
agreement with the monthly variations discussed in Fig. 4. Similar
seasonal diurnal characteristic was observed over New Delhi by Tiwari
et al. (2013). They observed that morning and evening peaks of BC
concentration were around 14.7, 4.8, 3.5 and 12.1 µg/m3 and 12.9, 3.6,
2.8 and 11.3 µg/m3 during winter, summer, monsoon, postmonsoon,
respectively. The shape and nature of the seasonal variation of BC
concentration are consistent with findings of Cao et al. (2009) over
Xi’an, China. They obtained that the highest BC concentrations at
morning were around 24, 15, 12, and 17 µg/m3 and at evening were
approximately around 26, 17, 11 and 21 µg/m3 during winter, spring,
summer and autumn, respectively. It is noticed from the figure that the
diurnal variations were more pronounced in winter and less pro-
Fig. 6. Diurnal variation of BC concentration for summer, winter, premonsoon and nounced during postmonsoon and premonsoon while diminished
postmonsoon over Karachi during March 2006 to December 2008.
during summer. The above diurnal cycle could be explained by both
emission sources and meteorological conditions as discussed above.
Table 2 During all seasons the high BC morning peaks were due to increase in
The computed model parameters coefficients, standard error, t stat and P-value of
the level of anthropogenic activities like dense traffic, industrial
meteorological variables over Karachi.
activities and the fumigation effect in the local boundary layer, when
Coefficients Standard error t Stat P-value the pollutants stay near the earth's surface (Tripathi et al., 2005). Apart
from these sources, additional sources of high BC concentration during
Intercept 24.40 3.26 7.47 4.94E−08
winter are enhanced coal burning, residential cooking with firewood
Temp 0.02 0.08 0.26 0.796
RH −0.11 0.05 −2.31 0.028 and open burning of woods in the nearby surroundings for heating
WS −0.25 0.07 −3.76 0.0008 purposes along with lower speed (Baxla et al., 2009). During summer,
Visibility −1.45 0.52 −2.76 0.010 the BC diurnal variation was pathetic as well as the magnitude of BC
Rainfall 0.0002 0.005 0.05 0.956 concentration was decreased because the atmospheric boundary layer
lies away from the earth's surface hence, pollutants get dispersed.
Sreekanth et al. (2007) reported similar diurnal pattern of BC
Table 3
Regression statics computed for the multiple regression analysis for all data over Karachi. concentration and found the morning peaks of 13.4 and 6.5 µg/m3
while evening peaks of 10.3 and 3.9 µg/m3 during winter and summer.
Regression statistics
3.2. Impact of meteorological variables on BC
Multiple R 0.930021
R Square 0.86494
Adjusted R Square 0.839928 To evaluate the impacts of surface meteorological variables on BC
Standard Error 1.094817 concentration, such as Temp, RH, WS, VIS and RF, the multiple
Observations 33 regression analysis between these variables and BC concentration was
conducted. Multiple regression analysis provides the information about
impact of multiple meteorological variables on the BC concentrations.
Table 4 Following the approach of Järvi et al. (2008), multiple regression
Correlation of meteorological variables such as Temp, RH, WS, Vis and RF with BC analysis was performed for meteorological variables such as, Temp,
during different seasons. RH, WS, VIS and RF with respect to BC concentration. The coefficients
for each variable are given, so we can write the best fit model as:
R
Summer Winter Premonsoon Postmonsoon BC=a 0 + a1Temp+a 2RH+a 3WS+a4Vis+a5RF

Temp −0.13 −0.17 −0.65 −0.93 where a 0 is the intercept and a1, a 2 , a3, a4anda5 are the regression
RH −0.26 −0.61 −0.79 −0.86 coefficients of Temp, RH, WS, VIS and RF, respectively. Table 2 shows
WS −0.43 −0.98 −0.93 −0.75
the computed model parameters coefficients, standard error, t stat and
Visibility −0.79 −0.24 −0.68 −0.88
Rainfall 0.17 −0.90 0.25 −0.44 P-value of meteorological variables. Using this regression analysis, an
overall good correlation (R=0.93) amongst all the meteorological
variables with BC concentration variable was observed, as given in
Besides the monthly diurnal variability, BC concentration also the Table 3. The negative sign for all variables shows an inverse
exhibits a pronounced season to season variations. Diurnal variability relationship between these variables and BC concentrations: hence, the
of BC concentration shows significant seasonal dependency which lower the variable values, the higher the concentration levels. Table 4
could be ascribed to anthropogenic activity, atmospheric boundary reveals the correlation of meteorological variables such as Temp, RH,
layer condition and vehicular traffic (Baxla et al., 2009). The averaged WS, VIS and RF with BC during different seasons.
seasonally diurnal variations were determined by averaging the three Additionally, in order to further facilitate analysis, scattered plots
years of seasonal hour to hour data. Fig. 6 reveals that the BC between BC and meteorological variables were plotted as shown in
concentration was ascending to sharp peaks at morning centred on Fig. 7. Monthly averaged temperature was observed in the range of 17–
8:00 to 9:00 LST, accompanied with the gradual descending to 32 °C with highest in June 2006 (with minimum BC levels) and lowest
minimum values reached at about 15:00 to 16:00 LST and then in January 2007 (maximum BC levels), which was responsible for
followed by the increasing trend with somewhat broader peaks at significant negative BC-temperature correlation (−0.71) for the whole

73
S. Bibi et al.
Fig. 7. Scattered plot of BC concentration vs. (a) temp, (b) RH, (c) WS, (d) RF and (e) VIS over Karachi during March 2006 to December 2008.
studied period. Surface temperature decreases during winter and
pollutants get accumulated resulting in high BC concentration while
low concentration in summer is due to degradation in temperature
when pollutants are scattered (Cao et al., 2009; Tiwari et al., 2013). The
similar inverse correlation between BC and temperature was previously
observed by numerous authors (Tripathi et al., 2005; Baxla et al., 2009;
Aruna et al., 2013). Similarly, BC was observed to be inversely
74
Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77
correlated (r =−0.83) with the relative humidity having low (43–
66%) from September to February covering two seasons (winter and
postmonsoon) as compared to premonsoon months and high (67–
81%) during summer months (June to August). The winter season
remains mainly dry than summer, which is one of the reason of high BC
concentration in winter. The daily averaged wind speeds were generally
low in winter and postmonsoon (7.0 and 8.7 km/h) and relatively
S. Bibi et al.
Fig. 8. Cluster trajectories of air masses and their percentage contribution over Karachi (yellow star) at 500, 1000, and 1500 m above ground level (AGL) in each season during 2006–
2008. The error bars represent the latitudinal extent of trajectories in each cluster.
higher during premonsoon and summer (13.1 and 16.3 km/h) reaching
20.7 km/h in May showing the inverse relation (r=−0.87) with BC. Our
results are comparable with that of Srivastava et al. (2012) over
Manora peak and Delhi. These winds play an important role in
advection of pollutants from nearby locations (Aruna et al., 2013).
The wind speeds are weaker during winter and postmonsoon due to
which reduction in the ventilation of aerosol resulting in the increase in
BC concentration near surface (Cao et al., 2009; Ramachandran and
Kedia, 2010). A similar attempt was made earlier to examine the
probable correlation between BC and meteorological variables and
observed the negative correlation of BC with temperature, relative
75
Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77
humidity, and wind speed (Cao et al., 2009; Tiwari et al., 2013).
Correspondingly, the VIS correlation trend was similar to that of BC
with temperature, wind speed and relative humidity. The seasonal BC-
VIS correlations were −0.89, −0.68, −0.79, and −0.87 with correspond-
ing averaged VIS were 6.2, 6.8, 6.7 and 6.6 km during winter,
premonsoon, summer and postmonsoon, respectively. Previously,
Tiwari et al. (2013) reported the inverse relation of BC with VIS having
relatively more pronounced correlation during winter and postmon-
soon as compared to summer and premonsoon, which are consistent
with the result obtained in this study. Visibility is considered to a
primary key of air quality in urban areas (Watson, 2002). Sloane and
S. Bibi et al.
White (1986) recommended that the reduction of visibility is a
certainly measured indicator of pollutant present in the air, rising
from the decrease of contrast between the atmosphere and the particle
and dilution of the light coming from the particle due to absorption and
scattering of light by fine aerosols.
The monthly average RF mostly recorded in summer, but monthly
averaged BC was not strongly correlated (−0.36) with RF as compared
to other variables over Karachi. This correlation (r=−0.35) is compar-
able to the correlation found by Ramachandran and Rajesh, (2007)
over Ahmedabad, in India and Cao et al. (2009) in Xi'an, China.
However, monthly BC concentration is strongly negative correlated
(r=−0.74) with RF over Trivandrum (Babu and Moorthy, 2002) and no
significant correlation (r=−0.01) was reported between these two
variables (Tiwari et al., 2013).
3.3. Cluster trajectory analysis
The backward trajectories arriving at the study site were investi-
gated over the period 2006–2008 at three levels, 500, 1000, 1500 m
Above Ground Level (AGL) using the Hybrid Single Particle Lagrangian
Integrated Trajectory (HYSPLIT) model (Draxler and Rolph, 2003). To
examine the source regions of air masses, the trajectories were
separated into three clusters based on their pathway. Fig. 8 shows
the cluster trajectories of air mass and their percentage contribution
during summer, winter, premonsoon and postmonsoon at 500, 1000,
1500 m. An important feature shown is that all the clusters have long
range overpass originating from southwest of the study site. This is
especially so in the case of clusters at 1000 and 1500 m during all the
seasons. Whereas, at 500 m, aerosols advected from nearby regions
and transported to the study site. The majority of the trajectory
pathways were originating from Iran, Arabian Peninsula and Arabian
Sea, whereas considerably lesser number of trajectories were observed
from Caspian Sea, India and Afghanistan. To investigate the source
regions and the possible transport of air masses, similar cluster
trajectories were also conducted by previous researchers (Dumka
et al., 2013; Aruna et al., 2013).
4. Conclusions
The measurement of BC mass concentrations at five-minute inter-
vals over an urban environment of Karachi for the long data-sets 2006–
2008) were examined. In the present study, the analysis for diurnally,
monthly and seasonal temporal scales and the impacts of meteorolo-
gical variables on BC concentration was carried out. The salient
conclusions of the present study can be summarized as follows.
• The monthly averaged BC concentration was maximum (12.5 µg/
m3) in the month of January 2007 and minimum (2.2 µg/m3) in the
month of June 2006
• The high BC concentration was observed during winter and post-
monsoon, while moderate during premonsoon and low during
summer.
• It was also noted that the BC concentration was highest with a
pronounced maximum value (12.5 µg/m3) during winter in 2007
and minimum (3.5 µg/m3) during summer in 2006.
• From the diurnal analysis, it is found that, a steady builds up occur
in the BC concentrations to sharp peaks between 07:00 and 09:00,
LST just after sunrise proceeded by a smooth decrease to low
concentration at 11:00 to 17:00, LST and then again starts increas-
ing to attain another peak values around 22:00, LST after sunset
during all the months. The highest morning peak (23.2 µg/m3) was
found during January 2007 and lowest peak (2.9 µg/m3) during
June 2007.
• Similarly, BC concentrations were ascending to sharp peaks at
morning centred at 8:00 to 9:00 LST, accompanied with the gradual
descending to minimum values reached at about 15:00 to 16:00 LST
76
Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77
and then followed by increasing trend with a somewhat broader
peaks at evening around 21:00 to 23:00 LST, during winter,
premonsoon, summer and postmonsoon
• As BC mass concentration is influenced by local meteorological
variables such as Temp, RH, WS, RF and Vis, so the relationships
between these variables and BC concentration were found. It was
observed that BC concentrations have an inverse correlation with all
these variables.
• It is observed from the HYSPLIT cluster trajectory analysis that the
majority of the trajectory pathways were initiating from Iran,
Arabian Peninsula and Arabian Sea, and considerably less number
of trajectories were observed from Caspian Sea, India and
Afghanistan.
Acknowledgment
We gratefully acknowledge the working team (http://ready.arl.
noaa.gov) for the HYSPLIT trajectories. The authors sincerely thank
Mr. Piyushkumar Patel from Indian Space Research Organization
(ISRO), Ahmedabad, India for his help in making HYSPLIT cluster
trajectories. The authors would like to thank the anonymous reviewers
for the critical comments and constructive suggestions toward
improvement of the paper.
References
Adesina, A.J., Kumar, K.R., Sivakumar, V., 2015. Variability in aerosol optical properties
and radiative forcing over Gorongosa (18.97 °S, 34.35 °E) in Mozambique. Meteorol.
Atmos. Phys. 127, 217–228.
Aruna, K., Kumar, T.L., Rao, D.N., Murthy, B.K., Babu, S.S., Moorthy, K.K., 2013. Black
carbon aerosols in a tropical semi-urban coastal environment: Effects of boundary
layer dynamics and long range transport. J. Atmos. Sol. -Terr. Phys. 104, 116–125.
Babu, S.S., Moorthy, K.K., 2001. Anthropogenic impact on aerosol black carbon mass
concentration at a tropical coastal station: a case study. Curr. Sci. 81, 1208–1214.
Babu, S.S., Moorthy, K.K., 2002. Aerosol black carbon over a tropical coastal station in
India. Geophys. Res. Lett., 29.
Baxla, S., Roy, A., Gupta, T., Tripathi, S., Bandyopadhyaya, R., 2009. Analysis of diurnal
and seasonal variation of submicron outdoor aerosol mass and size distribution in a
northern Indian city and its correlation to black carbon. Aerosol Air Qual. Res. 9,
458–469.
Bergstrom, R.W., Pilewskie, P., Russell, P., Redemann, J., Bond, T., Quinn, P., Sierau, B.,
2007. Spectral absorption properties of atmospheric aerosols. Atmos. Chem. Phys. 7,
5937–5943.
Bond, T.C., Anderson, T.L., Campbell, D., 1999. Calibration and intercomparison of
filter-based measurements of visible light absorption by aerosols. Aerosol Sci.
Technol. 30, 582–600.
Bond, T.C., Doherty, S.J., Fahey, D., Forster, P., Berntsen, T., DeAngelo, B., Flanner, M.,
Ghan, S., Kärcher, B., Koch, D., 2013. Bounding the role of black carbon in the
climate system: a scientific assessment. J. Geophys. Res. 118, 5380–5552.
Cao, J.-J., Zhu, C.-S., Chow, J.C., Watson, J.G., Han, Y.-M., Wang, G.-h., Shen, Z.-x., An,
Z.-S., 2009. Black carbon relationships with emissions and meteorology in Xi'an.
China Atmos. Res. 94, 194–202.
Draxler, R. R.; Rolph, G., 2003. HYSPLIT (HYbrid Single-Particle Lagrangian Integrated
Trajectory) model access via NOAA ARL READY website .〈http://www.arl.noaa.gov/
ready/hysplit4.html〉. NOAA Air Resources Laboratory, Silver Spring. In Md: 2003.
Dumka, U., Manchanda, R., Sinha, P., Sreenivasan, S., Moorthy, K.K., Babu, S.S., 2013.
Temporal variability and radiative impact of black carbon aerosol over tropical urban
station Hyderabad. J. Atmos. Sol. -Terr. Phys. 105, 81–90.
Dutkiewicz, V.A., Alvi, S., Ghauri, B.M., Choudhary, M.I., Husain, L., 2009. Black carbon
aerosols in urban air in South Asia. Atmos. Environ. 43, 1737–1744.
Gabbi, J., Huss, M., Bauder, A., Cao, F., Schwikowski, M., 2015. The impact of Saharan
dust and black carbon on albedo and long-term mass balance of an Alpine glacier.
Cryosphere 9, 1385–1400.
Guha, A., De, B.K., Dhar, P., Banik, T., Chakraborty, M., Roy, R., Choudhury, A., Gogoi,
M.M., Babu, S.S., Moorthy, K.K., 2015. Seasonal characteristics of aerosol black
carbon in relation to long range transport over Tripura in Northeast India. Aerosol
Air Qual. Res. 15, 786–798.
Hansen, A., Rosen, H., Novakov, T., 1984. The aethalometer—an instrument for the real-
time measurement of optical absorption by aerosol particles. Sci. Total. Environ. 36,
191–196.
Husain, L., Dutkiewicz, V.A., Khan, A., Ghauri, B.M., 2007. Characterization of
carbonaceous aerosols in urban air. Atmos. Environ. 41, 6872–6883.
IPCC, 2013. The Physical Science Basis. Working Group I Contribution In: Proceedings of
the IPCC 5th Assessment Report, Cambridge, United Kingdom and New York, NY, USA.
Järvi, L., Junninen, H., Karppinen, A., Hillamo, R., Virkkula, A., Mäkelä, T., Pakkanen,
T., Kulmala, M., 2008. Temporal variations in black carbon concentrations with
different time scales in Helsinki during 1996–2005. Atmos. Chem. Phys. 8,
1017–1027.
S. Bibi et al.
Li, K., Liao, H., Mao, Y., Ridley, D.A., 2016. Source sector and region contributions to
concentration and direct radiative forcing of black carbon in China. Atmos. Environ.
124, 351–366.
Lu, Z., Zhang, Q., Streets, D.G., 2011. Sulfur dioxide and primary carbonaceous aerosol
emissions in China and India, 1996–2010. Atmos. Chem. Phys. 11, 9839–9864.
Nair, V.S., Moorthy, K.K., Alappattu, D.P., Kunhikrishnan, P., George, S., Nair, P.R.,
Babu, S.S., Abish, B., Satheesh, S., Tripathi, S.N., 2007. Wintertime aerosol
characteristics over the Indo‐Gangetic Plain (IGP): impacts of local boundary layer
processes and long‐range transport. J. Geophys. Res., 112.
Novakov, T., Ramanathan, V., Hansen, J., Kirchstetter, T., Sato, M., Sinton, J., Sathaye,
J., 2003. Large historical changes of fossil‐fuel black carbon aerosols. Geophys. Res.
Lett., 30.
Petzold, A., Ogren, J.A., Fiebig, M., Laj, P., Li, S.-M., Baltensperger, U., Holzer-Popp, T.,
Kinne, S., Pappalardo, G., Sugimoto, N., 2013. Recommendations for reporting"
black carbon" measurements. Atmos. Chem. Phys. 13, 8365–8379.
Ramachandran, S., Kedia, S., 2010. Black carbon aerosols over an urban region: radiative
forcing and climate impact. J. Geophys. Res., 115.
Safai, P., Kewat, S., Praveen, P., Rao, P., Momin, G., Ali, K., Devara, P., 2007. Seasonal
variation of black carbon aerosols over a tropical urban city of Pune, India. Atmos.
Environ. 41, 2699–2709.
Schaap, M., van der Gon, H.D., 2007. On the variability of black smoke and carbonaceous
aerosols in the Netherlands. Atmos. Environ. 41, 5908–5920.
Singh, S., Soni, K., Bano, T., Tanwar, R., Nath, S., Arya, B., 2010. Clear-sky direct aerosol
radiative forcing variations over mega-city Delhi. Ann. Geophys. 28, 1157–1166.
Sloane, C.S., White, W.H., 1986. Visibility: an evolving issue. Environ. Sci. Technol. 20,
760–766.
Sreekanth, V., Niranjan, K., Madhavan, B., 2007. Radiative forcing of black carbon over
eastern India. Geophys. Res. Lett., 34.
Srivastava, A., Singh, S., Pant, P., Dumka, U., 2012. Characteristics of black carbon over
Delhi and Manora Peak—a comparative study. Atmos. Sci. Lett. 13, 223–230.
Srivastava, M.K., Singh, S., Saha, A., Dumka, U., Hegde, P., Singh, R., Pant, P., 2006.
Direct solar ultraviolet irradiance over Nainital, India, in the central Himalayas for
77
Journal of Atmospheric and Solar–Terrestrial Physics 157–158 (2017) 67–77
clear‐sky day conditions during December 2004. J. Geophys. Res., 111.
Stull, R.B., 2012. An Introduction to Boundary Layer Meteorology. Springer Science &
Business Media, (2012).
Surendran, D.E., Beig, G., Ghude, S.D., Panicker, A., Manoj, M., Chate, D.M., Ali, K.,
2013. Radiative forcing of black carbon over Delhi. Int. J. Photo., 2013.
Tare, V., Tripathi, S., Chinnam, N., Srivastava, A., Dey, S., Manar, M., Kanawade, V.P.,
Agarwal, A., Kishore, S., Lal, R., 2006. Measurements of atmospheric parameters
during Indian Space Research Organization Geosphere Biosphere Program Land
Campaign II at a typical location in the Ganga Basin: 2. chemical properties. J.
Geophys. Res., 111.
Tiwari, S., Srivastava, A., Bisht, D., Parmita, P., Srivastava, M.K., Attri, S., 2013. Diurnal
and seasonal variations of black carbon and PM 2.5 over new Delhi, India: influence
of meteorology. Atmos. Res. 125, 50–62.
Tripathi, S., Dey, S., Tare, V., Satheesh, S., 2005. Aerosol black carbon radiative forcing at
an industrial city in northern India. Geophys. Res. Lett., 32.
Wang, Q., Huang, R.-J., Cao, J., Tie, X., Shen, Z., Zhao, S., Han, Y., Li, G., Li, Z., Ni, H.,
2016. Contribution of regional transport to the black carbon aerosol during winter
haze period in Beijing. Atmos. Environ. 132, 11–18.
Watson, J.G., 2002. Visibility: science and regulation. J. Air Waste Manag. Assoc. 52,
628–713.
Wu, D., Mao, J., Deng, X., Tie, X., Zhang, Y., Zeng, L., Li, F., Tan, H., Bi, X., Huang, X.,
2009. Black carbon aerosols and their radiative properties in the Pearl River Delta
region. Sci. China Ser. D 52, 1152–1163.
Wu, Y., Zhang, R., Tian, P., Tao, J., Hsu, S.-C., Yan, P., Wang, Q., Cao, J., Zhang, X., Xia,
X., 2016. Effect of ambient humidity on the light absorption amplification of black
carbon in Beijing during January 2013. Atmos. Environ. 124, 217–223.
Zhang, L., Liao, H., Li, J., 2010. Impacts of Asian summer monsoon on seasonal and
interannual variations of aerosols over eastern China. J. Geophys. Res., 115.
Zhang, L., Henze, D. K., Grell, G. A., Bousserez, N., Zhang, Q., Torres, O., Ahn, C., Lu, Z.,
Cao, J., Mao, Y., 2015. Constraining black carbon aerosol over Asia using OMI
aerosol absorption optical depth and the adjoint of GEOS-Chem.