Heavy Metal Content of Total Suspended Air Particlesin the Heavily Industrialized Town of Gebze, Turkey

Pinar Ergenekon • Kadir Ulutas

Received: 19 April 2013 / Accepted: 25 October 2013 / Published online: 5 November 2013

� Springer Science+Business Media New York 2013

Abstract Air pollution is a serious environmental prob-

lem in industrialized towns, where a significant portion of

the residents live in close proximity to factories and major

highways with high traffic load. In this study, the ambient

air quality in Gebze, an industrial region with an area of

438 km2 and a population of 300,000, was characterized in

terms for total suspended particulate matter and its com-

position of trace elements, i.e. Cd, Cr, Cu, Fe, Mn, Ni and

Pb. Samples were collected using high volume samplers

from March to June 2009 at two sites during the day and

the night. A significantly higher Cu concentrations during

night suggested that Cu emissions were the result of a local

source. The known air toxics, Cd and Ni, had average

concentrations (34 and 43 ng/m3, respectively) higher than

proposed by the European Union’s ambient air quality

standards. These results highlight the potential health risks

for the local population.

Keywords Air quality � Source apportionment �Trace elements � Total suspended particles � Urban

aerosol

It has been well-documented that particulate air pollution

has adverse health effects, especially regarding respiratory

illnesses. Many studies have shown that long term exposure

to particulate air pollution has resulted in increased mor-

tality (Finkelstein et al. 2003; Nafstad et al. 2004). A study

by Jerrett et al. (2005) found that chronic exposure to

particulate air pollution was significantly associated with

premature, all-cause, cardio-respiratory and cancer mortal-

ity in the industrial city of Hamilton, Canada. A recent study

by Liu and Zhang (2009) showed a significant positive

correlation between the concentration of total suspended

particles (TSP) and lung dysfunction in children.

In addition to particulate matter (PM) concentrations,

the PM chemical composition has also been widely

investigated because it can determine potential human

health risks (Billet et al. 2007). Knowing the composition

of PM can also provide information regarding their effects

on visibility, climate forcing, and living organisms (Sald-

arriaga-Norena et al. 2011; Mauderly and Chow 2008) as

well as important insight into the sources of PM pollution

and its formation mechanisms (Ragosta et al. 2008). Trace

elements like arsenic, beryllium, cadmium, chromium,

cobalt, manganese, lead, mercury, nickel and selenium are

listed on the USEPA’s hazardous air pollutants list. Among

these compounds arsenic, beryllium, cadmium, chromium

and nickel are categorized as carcinogenic by the Interna-

tional Agency for Research on Cancer (IARC).

Gebze is one of the largest industrialized towns in

Turkey. Main emission sources in the town are chemical

and metal industries, highways with heavy vehicle traffic

and fossil fuel use for domestic heating. Unfortunately,

there are no air quality monitoring stations to identify

current air quality conditions in the region and no short

term monitoring efforts have been established. This situa-

tion prevents the establishment of air quality management

plans and health risk analyses in Gebze.

The present study was conducted to assess the current air

quality in Gebze by measuring TSP levels and concentra-

tions of Cd, Cr, Cu, Fe, Mn, Ni and Pb. Spatial and temporal

variations in TSP and trace element concentrations, com-

bined with their correlations to meteorological factors were

P. Ergenekon (&) � K. Ulutas

Environmental Engineering Department, Gebze Institute of

Technology, 41400 Cayirova, Gebze, Kocaeli, Turkey

e-mail: p.kus@gyte.edu.tr

123

Bull Environ Contam Toxicol (2014) 92:90–95

DOI 10.1007/s00128-013-1148-7

investigated. In addition, principal component analysis

(PCA) was performed with the purpose of identifying the

sources contributing most to PM pollution in Gebze.

Materials and Methods

Gebze is one of the largest industrial towns in Turkey and

is located between two major industrial cities, Istanbul and

Kocaeli (Fig. 1). The town has an area of 438 km2 and is

inhabited by 300,000 people. Two sampling locations were

selected by considering the ease of access, transportation

and security (Fig. 1). The first site was a park area just to

the north of highway D-100 and the major metal industries

in the Beylikbagi district, denoted as BP (40�4802200N,

29�2300300E, elevation of 33 m). The sampler was located

on the highest part of the park area at a sampling height of

1.7 m. The second site was the Gebze Municipality

Building, denoted as GM (40�4801900N, 29�2602100E, ele-

vation of 183 m). The GM site is situated away from

industrial sources (1 km north of the D-100 highway). The

sampler was placed on the roof of the building at a height

of 20 m.

Samples were collected on glass fiber (GF/A) filters

(90 mm diameter) using two GPS-1 PUF samplers

(Thermo-Andersen Instrument, Inc.) with a flow rate of

130 L/min. During March and April, samples were col-

lected over 24 h, starting at 5 p.m. each day. To evaluate

the differences between PM pollution during the day and

night, samples were collected during the day and night

from May to June. In this period, daytime samples were

collected from 7 a.m. to 5 p.m. and nighttime samples were

collected from 5 p.m. to 7 a.m. The total number (N) of

daily and daytime–nighttime samples collected during the

entire campaign for the BP and GM sites were 70 and 102,

respectively.

Prior to sampling, all filters were placed in an oven at

105�C for 24 h and then placed in a desiccator to cool

under low humidity conditions in a temperature-controlled

laboratory (20�C) for further 24 h. Their masses prior to

deployment were determined using an analytical micro-

balance. After being wrapped loosely in aluminum foil, the

filters were placed in a container and taken to the sampling

locations. Used filters were returned to the laboratory in a

container and stored in a desiccator for 24 h prior to

gravimetric mass determination. TSP concentrations were

Fig. 1 Location of the sampling sites (GM and BP) and the Gebze Meteorological Station

Bull Environ Contam Toxicol (2014) 92:90–95 91

123

calculated by dividing the collected mass on the filters by

the total volume of sampled air.

Trace element concentrations in the TSP were deter-

mined using atomic absorption spectroscopy (AAS) after

chemical and thermal extraction of the samples. Loaded

filters were transferred to Teflon tubes and treated with

8 mL nitric acid, 1 mL perchloric acid and 2 mL hydro-

fluoric acid. Upon heating in a microwave oven for 30 min

at 150�C, the extracts were filtered and the final volume of

the extracts was brought up to 100 mL by adding high-

purity deionized water (Millipore). Cd, Cr, Cu, Fe, Mn, Ni

and Pb standard solutions were prepared and used for a five

point calibration. All reagents used were of analytical

grade and obtained from Fluka.

Field blank samples were collected four times during the

measurement campaign. Measured elemental concentra-

tions on the blanks were generally much lower than the

sample concentrations (\10 %). The extracts of the filter

blanks were found to have Pb, Cd, Cu, Cr, Ni, Mn, and Fe

concentrations of 0.01, 0.001, 0.01, 0.05, 0.01, 0.05, and

0.67 mg/L, respectively. Sample quantities were blank-

corrected by subtracting the mean blank concentrations

from the sample concentrations. Averaged sample-to-blank

ratios varied between 2 for Cr to 37 for Cu. Limit of

detection (LOD) values were calculated as three times the

standard deviation of the blank samples for each measured

element. LOD values based on elemental concentrations in

the extract were 0.0001, 0.004, 0.005, 0.01, 0.02, 0.04, and

0.12 mg/L for Cd, Ni, Pb, Cu, Cr, Mn, and Fe respectively.

The high volume samplers were calibrated using a cali-

brated orifice prior to usage and flow checks were per-

formed at the middle and end of the sampling period.

Meteorological data were obtained from Gebze Meteo-

rological Station (GMS), located 2 km from the GM

sampling site. Averaged monthly temperature, wind speed,

and relative humidity data for the sampling period are

given in Table 1. The relationship between wind direction

and element concentrations was determined for each ele-

ment by considering 8 wind categories made up of 45�intervals (0�–45� corresponds to NE, 46�–90� corresponds

to E, etc.).

All the statistical analyses were done using SPSS Statistical

Software. Correlations among the parameters were quantified

as Pearson’s correlation coefficient (r). Comparison of means

of measured concentrations was done by Student’s t test. The

PCA for source identification was carried out by using Vari-

max rotation and factors with eigenvalues higher than 1.0

were extracted.

Results and Discussions

Average daily TSP concentrations over the entire mea-

surement period at the GM and BP sites were 191.7 and

199.7 lg/m3, respectively. The entire sample of daily TSP

concentrations are presented in Table 2. Although the

highest TSP concentrations (over 400 lg/m3) were

observed at the BP site, no significant difference (p [ 0.20)

in concentrations was observed between the two sites

(paired t test). In addition, the correlation between daily

TSP concentrations at BP site and GM site was significant

(r = 0.60). The difference between TSP concentrations

during the day and night was not significant for either of

the sampling sites (p [ 0.10). The same result occurred for

workday versus weekend in terms of TSP concentrations.

This suggests that TSP sources in Gebze do not show much

temporal and spatial variation.

Average daily trace element concentrations and corre-

sponding standard deviations for the two sampling sites are

presented in Table 3. The most abundant trace element at

both sites was Fe, making up 75 % of the identified portion

of TSP at GM on average and 56 % at BP. The measured

trace element compositions at both sites showed similar

profiles except for Cu content. Cu was the second most

abundant element (29 %) at the BP site, while its portion

was only 4 % at the GM site.

From Table 3, it is evident that the ambient air quality in

Gebze is alarming, especially with regards to ambient Cu

and Ni concentrations, which were much higher in Gebze

than any of the other industrial towns presented. Cd con-

centration in Gebze is also very high compared to the

values given for the other industrial areas. We could find

only one study that reports a higher Cd concentration than

the one measured in Gebze (Von Schneidemesser et al.

2010).

Table 1 Monthly temperature, relative humidity and wind speed

records at Gebze for the sampling period

Month Temperature (�C) Relative

humidity (%)

Wind speed

(km/h)

Range Average Range Average Range Average

March -17.8 to

22.8

7.3 45–91 74 3–27 7.4

April 2.8–3.3 10.8 42–95 74 0–11 5.2

May 8.3–30.6 17.4 35–82 61 0–18 6.5

June 12.8–36.1 22.7 27–90 59 0–4 1.2

Table 2 Basic statistics of daily TSP concentrations (lg/m3) at GM

and BP sites

Sampling site Mean SD Range Percentiles

25 50 75

GM 191.7 69.9 88–390 143.1 187.9 232.5

BP 199.7 85.9 91–482 145.2 175.1 234.6

92 Bull Environ Contam Toxicol (2014) 92:90–95

123

Average trace element concentrations during the day and

night at the GM site are presented in Fig. 2a. Application of

paired t tests revealed that there was a significant difference

(p \ 0.01) between daytime and nighttime concentrations of

Cr and Fe at the GM site. Average trace element concen-

trations during the day and night at the BP site are presented

in Fig. 2b. There was no significant difference between

nighttime and daytime concentrations except for Cu

(p \ 0.01). Nighttime Cu concentrations were four times

higher than daytime concentrations, suggesting that a local

source was responsible for emitting large quantities of Cu

during the evening. A comparison of workday versus

weekend trace element concentrations revealed that there

was no significant difference at either site (p [ 0.10).

The relationship between wind direction and each

measured trace element concentration was analyzed using

the average concentration of the particular element for each

wind direction. Maximum Cd, Mn, Ni and Pb concentra-

tions were observed under westerly winds at both GM and

BP sites. Cu at the BP site also peaks when the wind comes

from the west. The emissions from the industrial facilities

located to the west of the sampling points (Fig. 1) might be

Ta

ble

3A

ver

aged

trac

eel

emen

tco

nce

ntr

atio

ns,

inn

g/m

3,

atth

eG

Man

dB

Psi

tes

and

ind

iffe

ren

tin

du

stri

alar

eas

thro

ug

ho

ut

the

wo

rld

wit

hco

rres

po

nd

ing

Eu

rop

ean

Un

ion

’sai

rq

ual

ity

stan

dar

ds

for

Pb

,C

d,

and

Ni

Sit

eT

his

stu

dy

Lah

or,

Pak

ista

na

Llo

dio

,S

pai

nb

Izm

irT

urk

eyc

Rio

de

Jan

eiro

Bra

zild

Sih

wa,

Ko

reae

Eu

rop

ean

Un

ion

air

qu

alit

yst

and

ard

PM

frac

tio

nT

SP

PM

10

PM

10

TS

PT

SP

PM

10

Sam

pli

ng

per

iod

Mar

ch–

Jun

e2

00

92

00

7–

20

08

20

01

20

00

–2

00

12

00

1–

20

02

20

04

–2

00

6A

nn

ual

Nu

mb

ero

fsa

mp

les

N=

50

BP

N=

49

GM

N=

60

N=

87

N=

24

N=

47

N=

11

0

Cd

36

±1

83

2±

14

77

1.2

80

.97

.15

Cr

50

±4

14

4±

41

30

24

.61

14

21

69

.6

Cu

1,2

53

±1

,45

01

21

±1

01

73

32

.61

54

33

5–

Fe

2,4

48

±1

,30

62

,13

1±

76

88

,20

0–

3,4

38

38

,90

31

,20

3

Mn

26

7±

15

12

58

±1

23

30

08

6.5

12

51

,21

61

61

Ni

41

±5

34

6±

85

18

33

39

0.5

34

.72

5

Pb

14

3±

22

02

10

±2

33

4,4

00

10

1.7

11

11

01

32

15

00

aV

on

Sch

nei

dem

esse

ret

al.

(20

10

)b

Mo

ren

oet

al.

(20

06

)c

Od

abas

iet

al.

(20

02)

dQ

uit

erio

etal

.(2

00

4)

eL

imet

al.

(20

10

)

1

10

100

1000

10000

Pb Cd Cu Cr Ni Mn Fe

Co

nce

ntr

atio

n(n

g/m

3)

Co

nce

ntr

atio

n(n

g/m

3)

a

b

Day

Night

*

*

1

10

100

1000

10000

Pb Cd Cu Cr Ni Mn Fe

Day

Night

*

Fig. 2 Average trace element concentrations and SD (indicated by

error bars) during the day and night at the GM site (a) and the BP site

(b). The number of day and night sample pairs is 23 and 22 for GM

and BP site, respectively. Asterisk indicates that the difference is

significant (p \ 0.01)

Bull Environ Contam Toxicol (2014) 92:90–95 93

123

the underlying reason for these observed maximum con-

centrations. The maximum concentrations of Cu and Cr, on

the other hand, were observed under southerly and north-

westerly winds at the GM site. Cr might be released from

the organized industrial areas northwest of the GM site

(Fig. 1, denoted by 1 and 2). At the BP site, Fe and Cr

concentrations were highest under southerly and south-

westerly winds, suggesting that metal industries (shown in

Fig. 1) located south and southwest of the BP site might be

responsible for the high concentrations of these elements.

There was a significant positive correlation (p = 0.01)

between temperature and all elements except Ni at the GM

site. This suggests that domestic heating is not a dominant

pollution source since no increase in concentration levels

were observed during colder days. The negative correlation

between wind speed and concentration was significant for

only Pb and Cr (lrl [ 0.4) at GM. At the BP site on the

other hand, meteorological parameters did not affect ele-

ment concentrations in general. Only Fe and Cd concen-

trations showed a significant temperature dependence

(r \ 0.45). Although the site is surrounded by residential

areas, no direct effect of the use of fossil-fuel for domestic

heating was observed at the BP site, either. Wind speed, on

the other hand, exhibited a negative correlation with only

Cu concentrations (r = 0.47). Increased wind speed prob-

ably results in lower Cu concentrations since the wind

disperses the Cu emissions from a source located very

close to the BP site. The dependence of trace element

concentrations on temperature and wind speed at the two

sites is not exactly comparable. This might be related to the

difference in the sampling heights and sources of air pol-

lution at each site. It must also be noted that meteorological

parameters could not be measured at the sampling points

but were obtained from a meteorological station near the

GM point (denoted as GMS in Fig. 1).

Copper concentrations were considerably different

between the two sites, confirming that a Cu emitting source

is present near the BP site. A copper cable manufacturing

factory located to the west of the BP site might be the main

source of the observed ambient Cu at this point. Cd con-

centrations on the other hand, do not vary significantly

between the two sites. It is known that coal combustion and

production of non-ferrous metals, iron and steel are the

major sources of Cd (Pacyna 1987). High Cd concentra-

tions can be the result of these metal producers found in

Gebze and its neighbor town Dilovasi (1 and 10 km away

from the sampling points). Among those there are several

leading iron and steel production facilities with global scale

capacities. These iron and steel facilities produce large

quantities of coke by combusting coal as well as they use

scrap metals as raw material. Moreover, if these scrap

metals were treated with Cd as a surface coating, this might

be an additional Cd source (Stigliani and Anderberg 1994).

Principal component analysis was performed on the data

obtained from both sites to identify the contributions of

different sources to the measured trace element concen-

trations (Table 4). Elements with the largest influence

([0.8) are shown in bold. The analyses showed that 64 %

of the total variability was captured for the GM site and this

variability was attributable to two factors. The first factor

captured 47 % of the total variance and is characterized

primarily by Cr (0.92) and Fe (0.95), and moderately by Cd

(0.58) and Mn (0.42). This factor, like in the BP site, was

considered to represent a combination of emissions from

industrial production facilities in the region (metallurgical

industries) and re-suspended crustal matter. The second

factor was characterized by Cu (0.76), Pb (0.62), Cd (0.59),

and Mn (0.69). This factor was considered to be the result

of traffic emissions. The total variability captured at the BP

site was 82 %, with a total of three factors. The first factor

was characterized mainly by Fe (0.95), Cr (0.91), and Mn

(0.83), and moderately by Cd (0.69) and Cu (0.49). Fe and

Mn are elements generally associated with soil (Marcazzan

et al. 2001; Arditsoglou and Samara 2005; Lin et al. 2010).

Table 4 Total variance and factor contributions from PCA for the GM and BP sites

GM point BP point

Total variance (%) 64.3 82.0

Factor 1 2 1 2 3

Factor variance (%) 47.2 17.1 49.2 18.5 14.3

Pb -0.08 0.62 -0.14 0.83 0.13

Cd 0.58 0.59 0.69 0.40 0.31

Cu 0.39 0.76 0.49 0.67 -0.14

Cr 0.93 0.04 0.91 -0.17 0.05

Ni 0.02 0.50 0.09 0.04 0.97

Mn 0.42 0.69 0.83 0.41 0.01

Fe 0.95 0.11 0.95 -0.04 0.06

Likely source Re-suspended soil ? industry Mobile sources Re-suspended soil ? industry Mobile sources Ni Coating industry

94 Bull Environ Contam Toxicol (2014) 92:90–95

123

Cr, Cu and Cd can be related to industrial activities (Ra-

gosta et al. 2008; Caggiano et al. 2010; Kothai et al. 2008).

Therefore this factor, which explained the largest part of

the total variance (49 %), was considered to be represen-

tative of both industrial emissions and re-suspended crustal

matter. The second factor was characterized mainly by Pb

(0.83) and Cu (0.67), and, to a lesser extent, by Mn and Cd.

Pb, Cu, and Cd are associated with vehicle-related emis-

sions (Na and Cocker 2009; Ragosta et al. 2008; Pacyna

and Pacyna 2001). Compounds of Mn have also been

identified as tailpipe emissions (Pfeifer et al. 2004).

Therefore the second factor represents motor vehicles. The

third factor was characterized by only Ni (0.97) and likely

represents the metal coating industry to the west of the BP

sampling point in Gebze. In both sites, the factor identified

as re-suspended soil and industry explained the largest part

of the total variance ([45 %) which means that these

sources are more likely to be the major contributors to the

trace element content of the ambient TSP in Gebze.

This 4-month monitoring study in an industrialized town

of Gebze revealed that concentrations of trace elements of

Cd, Ni and Cu in TSP are exceptionally high. Compared to

the proposed limits of European Commission concerning

airborne dust especially Cd pollution in Gebze is alarming.

The difference observed in Cu concentrations between day

and night time periods suggests that the future monitoring

strategies for air quality on a local scale have to take this

interim nature of emissions into account. As in the case of

Gebze, continuous PM monitoring and determination of

trace element concentrations is essential for all industrial

towns to better identify the sources and to contemplate

effective air quality management strategies.

Acknowledgments We thank Dr. Salim Oncel for his generous help

in measuring trace element concentrations and Prof. Dr. Kadir Alp for

providing one of the high volume samplers.

References

Arditsoglou A, Samara C (2005) Levels of total suspended particulate

matter and major trace elements in Kosovo: a source identifi-

cation and apportionment study. Chemosphere 59:669–678

Billet S, Garc G, Dagher Z, Verdin A, Ledoux F, Cazier F, Courcot D,

Aboukais A, Shirali P (2007) Ambient particulate matter

(PM2.5): physicochemical characterization and metabolic acti-

vation of the organic fraction in human lung epithelial cells.

Environ Res 105:212–223

Caggiano R, Macchiato M, Trippetta S (2010) Levels, chemical

composition and sources of fine aerosol particles (PM1) in an

area of the Mediterranean basin. Sci Total Environ 408:884–895

Finkelstein MM, Jerrett M, DeLuca P, Finkelstein N, Verma D,

Chapman K, Sears M, Sears M (2003) Environmental justice: a

cohort study of income air pollution and mortality. Can Med

Assoc J 169:397–402

Jerrett M, Buzzelli M, Burnett RT, DeLuca PF (2005) Particulate air

pollution, social confounders, and mortality in small areas of an

industrial city. Soc Sci Med 60:2845–2863

Kothai P, Saradhi IV, Pandit GG, Markwitz A, Puranik VD (2008)

Source apportionment of coarse and fine particulate matter at

Navi Mumbai, India. Aerosol Air Qual Res 8(4):423–436

Lim JM, Lee JH, Moon JH, Chung YS, Kim KH (2010) Airborne

PM10 and metals from multifarious sources in an industrial

complex area. Atmos Res 96:53–64

Lin CW, Chen YR, Yen CH, Kao TC (2010) Chemical mass balance

source apportionment of ambient total suspended particulate

matters near Jhuoshuei river in Central Taiwan. Environ

Forensics 11:216–222

Liu L, Zhang J (2009) Ambient air pollution and children’s lung

function in China. Environ Int 35:178–186

Marcazzan GM, Vaccaro S, Valli G, Vecchi R (2001) Characteriza-

tion of PM10 and PM2.5 particulate matter in the ambient air of

Milan (Italy). Atmos Environ 35:4639–4650

Mauderly JL, Chow JC (2008) Health effects of organic aerosols.

Inhal Toxicol 20(3):257–288

Moreno T, Querol X, Alastuey A, Viana M, Salvador P, Sanchez de la

Campa A, Artinano B, de la Rosa J, Gibbons W (2006)

Variations in atmospheric PM trace metal content in Spanish

towns: illustrating the chemical complexity of the inorganic

urban aerosol cocktail. Atmos Environ 40:6791–6803

Na K, Cocker DR (2009) Characterization and source identification of

trace elements in PM2.5 from Mira Loma, Southern California.

Atmos Res 93:793–800

Nafstad P, Haheim L, Wisløff T, Gram F, Oftedal B, Holme I, Hjermann

I, Leren P (2004) Urban air pollution and mortality in a cohort of

Norwegian men. Environ Health Perspect 112:610–615

Odabasi M, Muezzinoglu A, Bozlaker A (2002) Ambient concentra-

tions and dry deposition fluxes of trace elements in Izmir,

Turkey. Atmos Environ 36:5841–5851

Pacyna JM (1987) Atmospheric emissions of arsenic, cadmium, lead

and mercury from high temperature processes in power gener-

ation and industry. In: Hutchinson TC, Meema KM (eds) Lead,

Mercury, Cadmium and Arsenic in the environment. John Wiley

and Sons Ltd, New York, p 69

Pacyna JM, Pacyna EG (2001) An assessment of global and regional

emissions of trace metals to the atmosphere from anthropogenic

sources worldwide. Environ Rev 9:269–298

Pfeifer GD, Roper JM, Dorman D, Lynam DR (2004) Health and

environmental testing of manganese exhaust products from use

of methylcyclopentadienyl manganese tricarbonyl in gasoline.

Sci Total Environ 334–335:397–408

Quiterio SL, Sousa da Silva CR, Arbilla G, Escaleira V (2004) Metals in

airborne particulate matter in the industrial district of Santa Cruz,

Rio de Janeiro, in an annual period. Atmos Environ 38:321–331

Ragosta M, Caggiano R, Macchiato M, Sabia S, Trippetta S (2008)

Trace elements in daily collected aerosol: level characterization

and source identification in a four-year study. Atmos Res

89:206–217

Saldarriaga-Norena H, Hernandez-Mena L, Murillo-Tovar M, Lopez-

Lopez A, Ramirez-Muniz M (2011) Elemental contribution to

the mass of PM2.5 in Guadalajara City. Mexico. Bull Environ

Contam Toxicol 86:490–494

Stigliani WM, Anderberg S (1994) Industrial metabolism at the

regional level: the Rhine basin. In: Ayresand RU, Simonis UE

(eds) Industrial metabolism: restructuring for sustainable devel-

opment. The United Nations University Press, Tokyo, p 390

Von Schneidemesser E, Stone EA, Quraishi TA, Schauer JJ, Shafer

MM (2010) Toxic metals in the atmosphere in Lahore, Pakistan.

Sci Total Environ 408:1640–1648

Bull Environ Contam Toxicol (2014) 92:90–95 95

123