Estimation of emission strengths of primary air pollutants in the city of Izmir, Turkey

Atmospheric Environment 38 (2004) 1851–1857

ARTICLE IN PRESS

AE International – Europe

*Correspond

E-mail addr

1352-2310/$ - se

doi:10.1016/j.at

Estimation of emission strengths of primary air pollutantsin the city of Izmir, Turkey

Tolga Elbir*, Aysen Muezzinoglu

Department of Environmental Engineering, Kaynaklar Campus, Dokuz Eylul University, 35160 Buca/Izmir, Turkey

Received 22 October 2003; received in revised form 30 December 2003; accepted 19 January 2004

Abstract

This paper presents the air pollutant emission inventory of primary pollutants for Izmir, which is a highly

industrialized area situated in the western part of Turkey. A proper emission inventory is very important for planning

pollution control programs, particularly in coastal sites like Izmir, where environmental quality is of growing concern

owing to their typical meteorological conditions. The sources were broadly classified as point, line and area sources in a

systematic way. The data on activity levels of industries, fuel consumption in vehicles and domestic activities along with

the respective emission factors were used for estimating the emissions for the year 2000. The results showed that

industry is the most polluting sector for sulfur dioxide (SO2) in the study area contributing about 88% of total

emissions. On the other hand, domestic heating is the most polluting sector contributing about 56% of total PM

emissions while traffic has the highest portion for NOX emissions. Especially, emissions from industries located outside

the metropolitan city center are much higher in amount. Industries located around the Izmir metropolitan center

contribute to the industrial SO2 emissions by 93%, PM emissions by 59% and NOX emissions by 80% of the total.

r 2004 Elsevier Ltd. All rights reserved.

Keywords: Emission; Emission inventory; Pollutant source; Air quality; Izmir

1. Introduction

Knowledge of the types of pollutants and their

emission rates is fundamental to the study and control

of air pollution because these rates, together with the

prevailing meteorological conditions and topographical

factors, determine the air quality in a region. The listing

and description of air pollutant-emitting sources, plus

the estimated pollutant emission quantities, comprise

the emission inventory. Emission inventories are im-

portant for developing emission control strategies,

determining the applicability of permitting and control

programs, ascertaining the effects of sources and

appropriate mitigation strategies, and a number of other

related applications by an array of users, including

central and local agencies, technical consultants to

several projects and industrial managers aiming at

ing author. Tel./fax: +90-232-4530922.

ess: [email protected] (T. Elbir).

e front matter r 2004 Elsevier Ltd. All rights reserve

mosenv.2004.01.015

testing the compliance of their facilities. Data from

source-specific emission tests or continuous emission

monitors are usually preferred for estimating quantities

of emissions as source monitoring activities provide the

best representation of the tested source’s emissions.

However, test data from individual sources are not

always available and, even then, they may not reflect the

general pattern for actual emissions over time. Thus,

emission calculations are usually the preferred method

for estimating emissions, although there may be limita-

tions in selecting the emission factors.

An emission inventory has been prepared for Izmir,

which is an important coastal city situated in the western

part of Turkey. Izmir province is located by the Aegean

Sea with longitude between 26�150 and 28�150, and

latitude between 37�450 and 39�150, covering a total area

of 12012 km2 (DIE, 2000). Metropolitan center of Izmir

of the province is the third biggest urban agglomeration

of Turkey and the acknowledged industrial and com-

mercial capital of the Aegean Region of Turkey. The

d.

ARTICLE IN PRESS

Fig. 1. Study area.

T. Elbir, A. Muezzinoglu / Atmospheric Environment 38 (2004) 1851–18571852

city with a 3.4 million population and sizeable economic

activities hosts many industries that are emitting high

quantities of air pollutants (Elbir, 2002, 2003; Dincer

et al., 2003).

As the aim of this study was to determine the air

pollutant emissions from major pollutant sources, a

local emission inventory was prepared with 1-h temporal

and 1-km spatial resolution within an area of

80 km� 100 km centred at the metropolitan area of

Izmir which covers an area of 50 km� 50 km (Fig. 1).

0

10

20

30

40

50

60

70

Leat

her

Foo

d

Met

al

Tex

tile

Tra

nspo

rteq

uipm

ent

Pai

nt

Cer

amic

Tob

acco

Bev

erag

e

Pap

er

Nu

mb

er o

f in

du

stri

es

Fig. 2. The number of industries according to manufacturing

sectors.

2. Source characteristics

In a systematic way, the emission sources are broadly

categorized as point, line and area sources, covering

industrial, vehicular and domestic sources, respectively.

Amounts of three major pollutants, namely the particu-

late matter (PM), sulfur dioxide (SO2) and NOX emitted

from these sources were determined. The contributions

from natural sources, diffuse emissions from industry,

non-road machinery and road suspension/resuspension

were neglected in these calculations.

2.1. Point sources

Most of the industries emit several air pollutants with

flue gases through their stacks. When the emission flow

rates above a certain amount, these stacks are referred to

as point sources. In the study area 374 industrial

facilities were identified contributing to air pollution

and included in the emission inventory. Of these 153 are

located in the metropolitan center. Others are distrib-

uted around the city. There is no end-of-pipe flue gas

desulfurization or denitrification facility in the area.

The main manufacturing sectors are leather, food,

textile, paper, metals, printing, chemical, mineral and

ceramics, petrochemical, petroleum refinery, electric and

electronic products, cement, iron and non-iron metal-

lurgy. In fact a much wider pattern of sectoral

distribution is available.

The mineral products sector including cement, sand

and gravel, lime, stone and quarrying facilities makes the

biggest contribution to PM emissions. In this sector

several different dust control equipment varying

from electrostatic precipitators to fabric filters are used

depending on the regulatory requirements. Therefore,

calculating the emissions from larger facilities in

this sector, such as the two cement plants, actual

emission measurement results were taken as the basis

(DEU, 2001).

For SO2 emissions the most important sector is

petroleum refinery and petrochemicals complex. In this

sector emissions calculated from stack gas measure-

ments were also used, too (Muezzinoglu et al., 2003).

For SO2 emissions smaller scale industrial combustion

sources of high sulfur fuel oils are in second place. Use

of high sulfur fuel oil in stationary sources is allowable

only outside the urban center. In cement and lime

factories actual SO2 emission concentrations were very

low regardless of the fuel type (Muezzinoglu et al.,

2003). Therefore, SO2 emissions due to cement and lime

industry sources were neglected.

As combustion temperatures were not high conse-

quent low NOX concentrations were measured in the

stack gases. Therefore, domestic heating and industrial

combustion were not the major source categories for

NOX emissions (DEU, 2001). For petroleum refinery,

petrochemicals, cement and lime facilities emission

measurement results were used in calculations (DEU,

2001). Fig. 2 shows the number of industries according

to types of sectors.

As the industrial facilities are usually agglomerated in

the organized industrial zones, most of the related data

required have been recorded by the zone management.

ARTICLE IN PRESS

0

500

1000

1500

2000

2500

3000

3500

4000

Fuel oil Lignite LPG Other

t d-1 Metropol

Outside

Fig. 3. Daily fuel consumptions of industries in the metropo-

litan area and its surroundings.

T. Elbir, A. Muezzinoglu / Atmospheric Environment 38 (2004) 1851–1857 1853

However, for all larger facilities even though some were

located in such zones, stack coordinates were located in

a grid system with 1-km spatial resolution. Therefore,

studies outside the urban area were referred for locating

them one by one as necessary, and during such visits,

questionnaires were given out and later collected.

Industrial source specific-information on production

capacities, raw materials used, manufacturing processes,

fuel consumption and stack characteristics (i.e. stack

height, outlet diameter, flue gas temperature and exit gas

velocity) was collected through questionnaires with the

help of Izmir Provincial Office of Ministry of Environ-

ment within the scope of a project (DEU, 2001) for the

year 2000.

The main fuel types used in industries are fuel oil,

lignite, liquefied petroleum gas (LPG) and others (wood,

biomass, petroleum coke, biogas and naphtha). The

total fuel consumption of industries located in the city of

Izmir and its surroundings on a daily basis are

separately shown in Fig. 3.

Lignite is the primary fuel type in the metropolitan

area while fuel oil is the most used fuel type outside

the city. Fuel oil usage is high due to the presence of

the major petroleum refinery (TUPRAS) and the huge

petrochemical plant (PETKIM) located in Aliaga

district (Fig. 1). These two cover 69% of overall fuel

oil use. The very high lignite consumption in the

metropolitan area is because of two high-capacity

cement plants: CIMENTAS and BATICIM. These two

plants cover 87% of overall lignite use.

2.2. Area sources

Emissions from sources, too small and difficult to

be surveyed individually, were considered collectively

as area sources. Thus domestic sources are categorized

as area sources and were evaluated and allocated on

the grid system taking into account the aerial distribu-

tion of the density of the population. Actually, the

fuel use pattern in the city is generally controlled by

the population density and purchasing capacity of the

inhabitants. For the calculation of domestic heating

emissions, the information collected included the num-

ber of inhabitants, type of fuel used, fuel consumption

statistics and combustion characteristics per grid.

Population data belonging to settlements were obtained

from the statistics of the last population census held by

State Institute of Statistics of Turkey (DIE, 1999).

Two major fuel types, lignite and residual petroleum

fuels named fuel oil number 6, were used for calculation

of domestic heating emissions due to lack of information

on the other less used cleaner fuels such as LPG, diesel

oil and kerosene. Consumption data for lignite coals

were collected through questionnaires (DEU, 2001).

There were no data for fuel oil consumption in the

domestic heating sector. To solve this problem, a

coefficient of fuel oil consumption per person calculated

within the scope of a participated international research

project (Barth et al., 2001) was used in this study. This

coefficient was calculated by dividing the amount of

total fuel oil consumption in the Izmir city center for

1996 by the population in that year. In this study,

domestic fuel oil consumption was calculated by multi-

plying this coefficient by study year population of Izmir.

This approach assumed that the space heating require-

ments and its breakdown into different types of fuels did

not significantly change between 1996 and 2000. This

assumption was based on the lack of official population

statistics for the year 2000.

The total fuel oil consumption of domestic sources in

the metropolitan area and its surroundings was calcu-

lated as 45,040 tons per year for the year 2000. The total

lignite consumption was determined as 514,846 tons per

year from questionnaires.

2.3. Line sources

Pollution due to vehicles is covered under line sources.

For calculating total emissions due to these sources,

data on consumption of gasoline and diesel fuel were

collected from questionnaires (DEU, 2001). Sulfur

contents of these two types of fuels were taken as the

suggested values in the emission factor lists (CITEPA,

1992; USEPA, 1995).

In the base year for this study the percentage of cars

with catalytic converters was unknown. As the statistics

covering this information was not available for the base

year in this study it was assumed that catalytic

converters were not used in cars. It is known that the

usage of catalytic converters is increasing as the

consumption of non-leaded gasoline is increasing. By

the beginning of 2004, no leaded gasoline will be sold in

the country and this will also increase the catalytic

converter use in the city of Izmir, too.

Road traffic emissions were calculated by means of

two different types of data sets and methods of

ARTICLE IN PRESST. Elbir, A. Muezzinoglu / Atmospheric Environment 38 (2004) 1851–18571854

calculation: highway traffic around the city of Izmir and

low speed urban traffic inside the city center. For the

first group State Highways Department was a good

source of information with the updated data on traffic

densities at major segments and cross-sections on the

express ways and highways around Izmir. Vehicle

counts were made by Demircioglu (2000) at the main

ports of the city traffic to see the distribution of traffic

loads with respect to hours of the day and days of the

week. In this part of the study the number of cars, buses,

trucks or other vehicles were apportioned into known

segment lengths and emissions calculated by using the

emission factors of different types of vehicles per unit

distance traveled.

For the traffic inside the city, however, this approach

could not be used as the number of vehicles traveling each

kilometer segment of the major roads could not be known

accurately. Therefore, the total traffic fuel usages within

the city borders per year were taken as the basis and this

was distributed over the time scale so that fuel usage for

transportation per hour was calculated. For geographical

distribution, however, population densities within the city

had to be used as a distribution criterion. Errors due to

fuel sales to intercity transport were assumed to be levied

by full tanks of the incoming vehicles.

3. Selection of emission factors

Selection of emission factors had to take into account

the fuel types, traffic vehicle properties (motor types,

weights, availability of exhaust gas control techniques,

Table 1

Emission factors used to calculate emissions (CITEPA, 1992; USEPA

Sector Fuel type Unit

Industry Lignite gGJ�1

Fuel oil gGJ�1

LPG gGJ�1

Wood gGJ�1

Biomass gGJ�1

Biogas gGJ�1

Petroleum coke gGJ�1

Domestic heating Lignite kg t�1

Fuel oil kgm�3

Traffic Gasoline g km�1

Diesel oil g km�1

Diesel oil g km�1

Diesel oil g km�1

Diesel oil g km�1

aA: ash content (%).bS: sulfur content (%).cH: heating value (MJkg�1).d r: control efficiency.

etc.), types of pollutants in accordance with technology

in use in each sector, production range and availability

of pollution control devices. In this study the emission

factors were taken from CORINAIR (CITEPA, 1992)

and US Environmental Protection Agency (USEPA)

source emission factors catalogues (USEPA, 1995). As

national conditions for each emission sector were more

or less differing from working conditions in Europe and

the United States for which the emission factors were

derived, more Turkish emission factors should have

been used in this work. But, as these were not available,

European emission factors had to be adopted and used.

Whenever European emission factors were insufficient

to indicate the industrial subcategories, EPA emission

factors were used. For combustion sector USEPA

emission factors were preferably used in order to adjust

for different particulate control efficiencies. Also, very

high ash- and sulfur-containing fuel combustion could

be better reflected by EPA factors rather than Corinair

emission factors. The emission factors used for indus-

tries, domestic and vehicular activities are summarized

in Table 1.

4. Results and discussion

The data on emissions from industries, fuel consump-

tion for vehicles and domestic activities along with

respective emission factors provide the emission inven-

tory presented in Table 2. The table shows that the

major air pollutant is SO2 having 2000 annual emissions

of 84,271 in the study area. 88% and 9% of these

, 1995)

PM SO2 NOX

3.4�Aa 20000�(Sb/Hc)�(1�rd) 150

1.12�S+0.37 490�S 180

0.07 — 100

4.4 — 200

Not available — 280

Not available — 100

Not available 500�S 300

1.5�A 15�S 2.9

0.3 17.24�S 2.2

— — 2.7

0.25 Not available 1.2

0.25 Not available 1.2

0.82 Not available 7.4

0.82 Not available 7.4

ARTICLE IN PRESST. Elbir, A. Muezzinoglu / Atmospheric Environment 38 (2004) 1851–1857 1855

emissions were estimated to come from industries and

domestic heating, respectively. The reason for such high

SO2 emissions is the use of fossil fuels with high sulfur

content. Although at the city center only the combustion

of lignite coals of o1% sulfur content is allowed for

domestic heating and industrial use only, residual oils of

about 3.5% sulfur outside the city center is widely used

in industrial facilities.

Industry is the most polluting sector contributing to

about 88% of total SO2 emissions while domestic

heating is the most polluting sector contributing to

56% of total PM emissions. Traffic is also the most

polluting sector for NOX emissions in the study area.

Especially, emissions from industries located outside the

metropolitan area were found to be much higher than

the emissions from within the city center. Out of the

total industrial emissions 93% of SO2, 59% of PM and

80% of NOX were coming from outside the Izmir

metropolitan area. For high PM emissions in the

Table 2

Total and sectoral emissions in the study area, ton yr�1

PM SO2 NOX

Domestic Metropolitan area 11,159 5693 1124

Surroundings 3538 1616 340

Subtotal 14,697 7309 1464

Industrial Metropolitan area 3941 5539 2631

Surroundings 5674 68,904 10,313

Subtotal 9615 74,443 12,944

Traffic Metropolitan area 1351 1862 19,418

Surroundings 730 657 10,293

Total 2081 2519 29,711

Total Metropolitan area 16,451 13,094 23,173

Surroundings 9942 71,177 20,946

Total 26,393 84,271 44,119

Table 3

First ten industries for SO2 emissions in the study area

Name Input fuels Type of

TUPRAS Fuel oil, fuel gas Petroleu

PETKIM Fuel oil, fuel gas Petroche

TEAS-ALIAGA Fuel oil Power p

IDC Fuel oil Iron-ste

HABAS Fuel oil Iron-ste

CUKUROVA Fuel oil, LPG Iron-ste

ALTIN KIREMIT Fuel oil Clay

CEBITAS Fuel oil Iron-ste

ALKIM Fuel oil Paper

IZMIR BASMA Fuel oil, lignite Textile

Others — Various

Total — —

metropolitan area two cement plants (CIMENTAS

and BATICIM) which have the highest PM emissions

in the study area are the main reason. Stack gas

measurements were mainly used for the PM emissions

from cement plants in the study.

When the emissions of individual industries were

examined one by one, the petroleum refinery and the

petrochemical industry were found to be the largest

sources of air pollution in the study area. They

contributed by 72% to the total industrial SO2 emissions

and by 66% to the overall SO2 emissions. The first ten

industries with the highest SO2 emissions in the study

area are presented in Table 3.

Due to their dense population, several regions in the

metropolitan area such as Alsancak, Konak, Karsiyaka

and Bornova have high emissions. Fig. 4 shows the

geographical distribution of PM emissions from domes-

tic heating per square grids at the metropolitan area. It

should be noted that all of the emissions from domestic

activities were assumed to be emitted between morning

and night, i.e. 7:00–23:00 local time and during the

winter period.

Major deviations of real emissions from calculated

emissions in this study come from lack of dependable

statistics. This is a serious problem especially for traffic

sources. Another source of error is the lack of Turkish

emission factors to be used in calculations.

5. Conclusion

This paper presents a local emission inventory

prepared with 1-h temporal and 1-km spatial resolution

within an area of 80 km� 100 km with the metropolitan

city of Izmir at the center. The inventory showed that

quantities of emissions for primary pollutants in Izmir

and its surroundings are high. This is in spite of the use

of European and American emission factors which

might have been somewhat underestimating the actual

industry SO2 emission, ton yr�1 %

m refinery 27,900 37.5

mical 25,730 34.6

lant 5382 7.2

el 1615 2.2

el 1540 2.1

el 956 1.3

784 1.1

el 625 0.8

502 0.7

496 0.7

8913 11.8

74,443 100

ARTICLE IN PRESS

Fig. 4. Geographical distribution of PM emissions from domestic heating per square grids at the metropolitan area, ton yr�1.

T. Elbir, A. Muezzinoglu / Atmospheric Environment 38 (2004) 1851–18571856

quantities of emissions in Turkish conditions. This is

especially important when the state of the combustion

technologies was compared.

As a conclusion it can be stated that unless immediate

technical measures such as in a control policy plan are

promulgated in accordance with some realistic target

values for reducing the emissions, Turkish industrial and

urban sites will undergo very severe air pollution

problems. So emission reduction programs including

control technology implementation, energy conservation

planning and pollution prevention techniques must be

urgently prepared and implemented. In this planning

work priorities of contributing source categories must be

taken into account.

In the Izmir study area it is expected that two such

developments will largely affect the emission inventory:

recent gradual replacement of industrial fuels with

imported natural gas beginning from 2003 and the

national purchase tax incentives in effect aiming at the

renewal of the car fleet since summer 2003. Natural gas

is presently available for use in the industrial agglom-

erations at the peripheral zones of the city of Izmir. And

the tax-driven car renewal campaign backed up by the

prohibition of leaded-gasoline sales in the country will

possibly increase the use of catalytic converters in cars.

It is expected that a better maintenance capacity will be

developed in the near future.

However, a policy plan involving regular follow-ups

for air pollutant emission inventory and pollution

reduction strategies is seriously and urgently needed

aiming at protection of the natural resources, agricul-

tural production, community health and cultural/

archaeological heritage in Izmir. This plan should also

work out the pollution prevention incentives in the

industry. Existing national Air Quality Assurance

Legislation should be implemented without concession

and accordingly revised to create a stricter inspection

infrastructure. Programs for assessment and control to

include higher and less polluting technologies for dust

control and desulfurization systems are urgently needed.

Present projects aiming at the use of natural gas on a

wider scale should be studied with similar methods to see

if they are feasible with respect to emission reduction.

Acknowledgements

The authors acknowledge the use of air quality data

from the Greater Izmir Municipality and emission

inventory produced by Air Pollution Control Labora-

tory, Dokuz Eylul University.

References

Barth, H.G., Baumbach, G., Muezzinoglu, A., Erdem, U.,

2001. Air quality and city planning in Izmir. Final Project

Report. Hannover University, Hannover, Germany.

ARTICLE IN PRESST. Elbir, A. Muezzinoglu / Atmospheric Environment 38 (2004) 1851–1857 1857

CITEPA (Centre Interprofessional Technique de la Pollution

Atmospherique), 1992. Corinair Inventory-Default emission

factors handbook, 2nd Edition. CEC-DGXI, Paris.

Demircioglu, H., 2000. Determination of emission loads of air

pollutants due to vehicles. M.Sc. Thesis. Dokuz Eylul

University, Izmir, 120pp.

DIE (State Institute of Statistics Prime Ministry Republic of

Turkey), 1999. 1997 Population Count, Administrative

Division. Ankara, Turkey.

DIE (State Institute of Statistics Prime Ministry Republic of

Turkey), 2000. Statistical yearbook of Turkey. Ankara, Turkey.

DEU (Dokuz Eyl .ul University), 2001. Clean air plan in Izmir,

project report. Izmir, Turkey.

Dincer, F., Muezzinoglu, A., Elbir, T., 2003. SO2 levels at

forested mountains around Izmir, Turkey and their possible

sources. Water, Air, and Soil Pollution 147, 331–341.

Elbir, T., 2002. Application of an ISCST3 model for predicting

urban air pollution in the Izmir metropolitan area.

International Journal of Environment and Pollution 18,

498–507.

Elbir, T., 2003. Comparison of model predictions with the data

of an urban air quality monitoring network in Izmir,

Turkey. Atmospheric Environment 37, 2149–2157.

Muezzinoglu, A., Elbir, T., Bayram, A., 2003. Air quality

management in Izmir region of Turkey as required by

clean air plans. Water, Air, and Soil Pollution: Focus 3,

307–323.

USEPA (US Environmental Protection Agency), 1995. Compi-

lation of air pollutant emission factors, Vol. I. Stationary

point and area sources. Rep.AP-42. Fifth Edition, Research

Triangle Park, NC.