journal of maritime researches vol. 3, no. 1 march …system (ais) and geographic information system...

Kobe University Repository : Kernel

タイトルTit le

Gaussian plume and puff model to est imate ship emission dispersionby combining automat ic ident ificat ion system(AIS) and geographicinformat ion system(GIS)

著者Author(s)

Ariana Made / Pitana, Trika / Artana, Ketut Buda / Dinariyana, Bagus /Masroeri, Agoes A.

掲載誌・巻号・ページCitat ion Journal of marit ime researches,3(1):1-13

刊行日Issue date 2013-03

資源タイプResource Type Departmental Bullet in Paper / 紀要論文

版区分Resource Version publisher

権利Rights

DOI

JaLCDOI 10.24546/81006882

URL http://www.lib.kobe-u.ac.jp/handle_kernel/81006882

PDF issue: 2020-03-25

Journal of Maritime Researches Vol. 3, No. 1 March 2013: 1-13.

1

GAUSSIAN PLUME AND PUFF MODEL TO ESTIMATE

SHIP EMISSION DISPERSION BY COMBINING

AUTOMATIC IDENTIFICATION SYSTEM (AIS) AND

GEOGRAPHIC INFORMATION SYSTEM (GIS)

Made ARIANA*

Trika PITANA*

Ketut Buda ARTANA*

Bagus DINARIYANA*

Agoes A. MASROERI*

ABSTRACT

The issue on emission produced by ships is one of attractive and major attentions

in the shipping and shipbuilding industries. The use of more environmentally friendly

fuel, weather routing, power optimization and the possibility of using alternative

energy to propel ships are some examples of current efforts in minimizing emission.

Apart from the effort to reduce emission, this paper, on the other hand, provides a new

method in combining the Automatic Identification System (AIS) and the Geographic

Identification System (GIS) to estimate the distribution of emission produced by ships.

Data taken from AIS was used as the source of operating conditions of the ships and

the shipping database was used to calculate the amount of emitted gases. The Gaussian

Plum and Gaussian Puff models were then used to estimate the emission dispersion

along the coastal area of Madura Strait. The amount of emitted gas produced by main

engine and auxiliary engine of the ships operated along the Madura Strait was

calculated by means of equations provided by Trozzi (2002). This research found that

the AIS is not merely usable as ship‘s tracking and identification purposes, but further,

it could be used as an additional tool in estimating emission distribution.

Keywords: AIS, GIS, Gaussian Plume, Gaussian Puff, Emission, Madura Strait

* Department of Marine Engineering, Faculty of Marine Technology, ITS Surabaya Kampus ITS

Sukolilo, Surabaya 60111 Indonesia. Email : [email protected]

Gaussian Plume And Puff Model To Estimate Ship Emission Dispersion By Combining

Automatic Identification System (AIS) And Geographic Information System (GIS).

Made ARIANA, Trika PITANA, Ketut Buda ARTANA, Bagus DINARIYANA, and Agoes A.

MASROERI

2

1. INTRODUCTION

Emitted gases produced by a ship’s main and auxiliary engines have been

considered as a major contributor in environmental and health problems as well as

global warming issues. Nitrogen oxides (NOx), carbon monoxide (CO), carbon

dioxides (CO2), sulfur dioxides (SOx), and particulate matter (PM) are the components

of emitted gases that have been the subject of much research. Those components have

also been believed to be main contributors that reduce the quality of air (Wang et al.,

2009). The amount of emitted gas from a ship’s engines has also been examined by

some researchers who aim to find some solutions for effective ways of reducing it

(Bracken et al., 2007; Ishida, 2003; Jalkanen et.al., 2009; Pingjian et.al., 2006, Pitana

et.al., 2010; Trozzi et.al., 2002). The behavior of emission distribution in the air is

another foremost agenda in which many researchers have different points of view of

methods in estimating the probabilities and amounts of dispersed emission.

Port of Tanjung Perak Surabaya – Indonesia, is one of the biggest and busiest

ports in Indonesia (see Fig. 1). Currently, not less than 27,000 ship calls are served by

the port annually and it is expected to double by 2030 (see Table 1). The anchorage

area is located around 21 miles (38 KM) from the port. Along the strait, some other

major facilities and industries exist. Considering the number and ship distribution as

listed in table 1, it can be roughly concluded that the gas emission produced by ships is

quite prominent, though a majority of ship sizes that pass through the channel is

considered small.

This paper presents a new method in combining the Automatic Identification

System (AIS) and Geographic Information System (GIS) to estimate the distribution of

emission produced by ships. The dynamic data (longitude, latitude, time, course, rate

of turn, speed over ground (SOG), navigation information, etc) and static data (MMSI,

vessel name, vessel call sign, ship length, draft, beam, IMO number, ship type, etc.)

taken from AIS was used as the source of operating conditions of the ships, and then a

shipping database was used to calculate the amount of emitted components.

Fig. 1 Madura Strait Indonesia

Port of Tanjung Perak

Madura Strait




MASROERI

3

Table 1 Ships call in 2012.

The Gaussian Plum and Gaussian Puff models were then used to estimate the

emission distribution along the coastal area of the Madura Strait. The amount of

emitted gases produced by the main and auxiliary engines of the ships operated along

the Madura Strait were calculated by means of equations provided by Trozzi (2002).

This research found that AIS is not merely usable for ship tracking and identification

purposes, but further, it could be used as an additional tool in estimating emission

distribution.

2. AUTOMATIC IDENTIFICATION SYSTEM (AIS) AND GEOGRAPHIC

INFORMATION SYSTEM (GIS)

The Automatic Identification System (AIS), as shown in Fig. 2, is an automatic

tracking system used on ships and by Vessel Traffic Services (VTS) for identifying and

locating vessels by electronically exchanging data with other nearby ships and AIS

stations. AIS information supplements marine radar, which continues to be the primary

method of collision avoidance for water transport. The function of AIS can

significantly increase the safety of ship operations via a visualization system that

system installed on board a ship presents the bearing and distance of nearby vessels in

a radar-like display format. AIS provides a graphical display on board a ship where

information provided by AIS equipment, such as unique identification, position, course,

and speed, can be displayed on a screen or an ECDIS. The AIS is intended to assist a

vessel's watch-standing officers and allow maritime authorities to track and monitor

vessel movements.

The International Maritime Organization's (IMO) International Convention for the

Safety of Life at Sea (SOLAS) requires AIS to be fitted aboard international voyaging

ships with a gross tonnage (GT) of 300 or more, and all passenger ships regardless of

VESSEL

TYPE

VESSEL TYPE

DESCRIPTION

ANNUAL

SHIP

PASSING

TOTAL

ANNUAL

SHIP

PASSING

GROUP % OF

TOTAL

FVS Fishing Vessel Small 2,306

17,570 A 65.07%

FVM Fishing Vessel Medium 6,918

FVL Fishing Vessel Large 4,610

PFS Passenger Ferry Small 1,384

SVS Supply Vessel Small 2,306

TNKT Tanker Tug 46

PFM Passenger Ferry Medium 460 460 B 1.70%

TVM Oil Tanker Vessel Medium 692 692 C 2.56%

PFL Passenger Ferry Small 460

3,462 D 12,82% SVM Supply Vessel Medium 1,154

NVM Navy Vessel Medium 1,848

TVL Oil Tanker Vessel Large 230 690 E 2.56%

NVL Navy Vessel Large 460

SVL Supply Vessel Large 1,154 3,002 F 11.12%

CTL Container Large 1,848

CTX Container Extra Large 460 460 G 1.70%

PAN Panamax 256 256 H 0.95%

PPAN Post-Panamax 394 394 I 1.46%

TNKP Tanker Primary 14 14 J 0.05%

Total 27,000

http://en.wikipedia.org/wiki/Automatic

http://en.wikipedia.org/wiki/System

http://en.wikipedia.org/wiki/Watercraft

http://en.wikipedia.org/wiki/Ship

http://en.wikipedia.org/wiki/Coordinate_system

http://en.wikipedia.org/wiki/Course_%28navigation%29

http://en.wikipedia.org/wiki/Speed

http://en.wikipedia.org/wiki/Electronic_Chart_Display_and_Information_System

http://en.wikipedia.org/wiki/Watchstanding

http://en.wikipedia.org/wiki/Sea

http://en.wikipedia.org/wiki/International_Maritime_Organization

http://en.wikipedia.org/wiki/International_Convention_for_the_Safety_of_Life_at_Sea

http://en.wikipedia.org/wiki/International_Convention_for_the_Safety_of_Life_at_Sea




MASROERI

4

size. It is estimated that more than 40,000 ships carried AIS class A equipment. In

2007, the new Class B AIS standard was introduced which enabled a new generation of

low cost AIS transceivers.

Fig. 2 Automatic Identification System (AIS) and retrieved data

The use of AIS in the maritime transportation sector has been significantly and

widely developed. From its basic idea in assisting navigation, vessel traffic services,

and search and rescue and collision avoidance, the application of AIS has been moving

to a more advanced application such as accident investigation, binary messages,

computing and networking, AIS data on the internet, range limitations and space-based

tracking, and others. This, consequently, opens a broader chance for R&D

perspectives.

According to Indonesian Government Decree No. 5/2010 concerning Navigation,

Article 14, point (1) to point (3), it is very clear that the Government of Indonesia

requires that all vessels operating in the Indonesian territory to report their identities

and voyage data to shore based radio station including call sign, Maritime Mobile

Services Identities (MMSI), tonnage, destination, speed, course and estimation of

arrival time by using AIS or Long Range Identification and Tracking of Ships/LRIT.

A geographic information system (GIS) integrates hardware, software, and data

for capturing, managing, analyzing, and displaying all forms of geographically

referenced information. GIS allows us to view, understand, question, interpret, and

visualize data in many ways that reveal relationships, patterns, and trends in the form

of maps, globes, reports, and charts. With these features, GIS can be arranged to

display a ship’s position, tracking, and other important information obtained from AIS.

3. EMISSION CALCULATION

This research implements emission calculation methods provided by the MEET

project (methodologies for estimating air pollutant emissions from transport-Trozzy,

2002). Three different operation conditions of ships are considered: maneuvering

phases, hotelling phase, and cruising phase. The emission itself is calculated based on




MASROERI

5

power that is generated in order to supply the ship’s propulsion power, lighting,

heating, refrigeration, ventilation, etc. To implement the method, this research also

combines the AIS and GIS data with a shipping database in which the principal

dimensions of ships are obtained, as well as its main and auxiliary engine power. The

equations used to calculate emissions is shown in Table 2. The emission factor

produced by the ships is shown in Table 3.

Tabel 2 Ship classes and fuel consumption factor

Ship class Fuel consumption (ton/day) as function of

Gross Tonnage (GT)

Solid Bulk Cjk = 20.1860 + 0.00049 × GT

Liquid Bulk /Tanker Cjk = 14.6850 + 0.00079 × GT

General Cargo Cjk = 9.8197 + 0.00143 × GT

Container Cjk = 8.0552 + 0.00235 × GT

Ro-Ro Cargo Cjk = 12.8340 + 0.00156 × GT

Passenger Cjk = 16.9040 + 0.00198 × GT

High Speed Ferry Cjk = 39.4830 + 0.00972 × GT

Inland Cargo Cjk = 9.8197 + 0.00143 × GT

Sail Ship Cjk = 0.4268 + 0.00100 × GT

Tugs Cjk = 5.6511 + 0.01048 × GT

Fishing Cjk = 1.9387 + 0.00448 × GT

Other Ships Cjk = 9.7126 + 0.00091 × GT

Tabel 3 Emission factor (kg emission/ton fuel)

Mode Engine / Bahan Bakar NOx CO CO2 VOC PM SOx

Cruising

SSD/BFO 87 7.4 3200 2.4 1.2 60

MSD/BFO 57 7.4 3200 2.4 1.2 60

HSD/MDO 70 9 3200 3 1.5 20

Maneuvering

SSD/BFO 78 28 3200 3.6 1.2 60

MSD/BFO 51 28 3200 3.6 1.2 60

HSD/MDO 63 34 3200 4.5 1.5 20

Hotelling

SSD/BFO 35 99 3200 23.1 1.2 60

MSD/BFO 23 99 3200 23.1 1.2 60

HSD/MDO 28 120 3200 28.9 1.5 20

SSD = Slow Speed Diesel Engine BFO = Bunker Fuel Oil PM = Particulate Matter

MSD = Medium Speed Diesel Engine MDO = Marine Diesel Oil

HSD = High Speed Diesel Engine VOC = Volatile Organic Compound

Furthermore, the emitted gases of a ship’s main and auxiliary engines are then

calculated by incorporating some variables such as the type of engine, fuel type, and

operating conditions of the ships as shown in Table 3.

The emissions produced by ship’s main engine are obtained as:




MASROERI

6

𝐸𝑖 = ∑𝑗𝑘𝑙𝑚 𝐸𝑖𝑗𝑘𝑙𝑚 ……………………………………………………………….(1)

𝐸𝑖𝑗𝑘𝑙𝑚 = 𝑆𝑗𝑘𝑚(𝐺𝑇). 𝑡𝑗𝑘𝑙𝑚. 𝐹𝑖𝑗𝑙𝑚…………………………………………..………….(2)

Where i is pollutant, j is fuel type, k is ship class for use in consumption classification,

l is engines type class for use in emission factors characterization, m is operating mode,

Ei is total emissions of pollutant I, Eijklm is total emissions of pollutant i from use of fuel

j on ship class k with engines type l in operating mode m, and Sjkm(GT) is daily

consumption of fuel j in ship class k in mode m as a function of gross tonnage, tjklm is

days in navigation of ships of class k with engines type l using fuel j in operating mode

m and finally the Fijlm is average emission factors of pollutant i from fuel j in engines

type l in mode m (for SOx, taking into account average sulfur content of fuel).

On the other hand, emissions obtained from the auxiliary engine are obtained by

implementing the following formula.

𝑓 = 0.2 𝑥 𝑂 𝑥 𝐿 …………………………………………………………….……….(3)

Where f is the ship’s fuel consumption (kg/ship/hr), O is rated output (PS/engine) and

L is load factor (cruising: 30%, hotelling (tanker): 60%, hotelling (other ship): 40%

and maneuvering: 50%).

4. GAUSSIAN PLUME AND GAUSSIAN PUFF MODEL FOR EMISSION

DISPERSION

The Gaussian model is the most commonly used model for estimating air

pollution dispersion. It assumes that the air pollutant dispersion has a Gaussian

distribution, meaning that the pollutant distribution has a normal probability

distribution. Gaussian models are most often used for predicting the dispersion of

continuous, buoyant air pollution Plumes originating from ground-level or elevated

sources. When it is required to use the model for predicting the dispersion of

non-continuous air pollution Plumes, the Gaussian Puff models are preferable.

One of the key assumptions of this model is that over short periods of time (such

as a few hours), a steady state condition exists with regard to air pollutant emissions

and meteorological changes. In this research, emitted gases are represented by an

idealized Plume coming from the top of ship funnel (exhaust gas) of some height and

diameter. One of the primary calculations is the effective funnel height.

http://en.wikipedia.org/wiki/Normal_distribution

http://en.wikipedia.org/wiki/Normal_distribution




MASROERI

7

Fig. 3 Gaussian model for air pollution dispersion (Altwicker, 2000)

As the exhaust gases are produced in the main and auxiliary engines, the hot

Plume will be thrust upward some distance above the top of the funnel. We need to be

able to calculate this vertical displacement, which depends on the funnel gas exit

velocity and temperature, and the temperature of the surrounding air.

Once the Plume has reached its effective height, dispersion will begin in three

dimensions. Dispersion in the downwind direction is a function of the mean wind

speed blowing across the Plume. Dispersion in the cross-wind direction and in the

vertical direction will be governed by the Gaussian Plume equations of lateral

dispersion. Lateral dispersion depends on a value known as the atmospheric condition,

which is a measure of the relative stability of the surrounding air. The model assumes

that dispersion in these two dimensions will take the form of a normal Gaussian curve,

with the maximum concentration in the center of the Plume.

The standard algorithm used in Plume studies is the Gaussian Plume model,

developed in 1932 (see Fig. 3). It was then further developed by Bracken, et al., 2007 ;

Pingjian, et al, 2006; Altwicker, 2000. The algorithm is as follows:

𝐶 (𝑥, 𝑦, 𝑧, 𝐻𝑒) = 𝑄

2𝜋𝜎y 𝜎z 𝑢s

x 𝑒(−𝑦2

2𝜎𝑦2 )

x {𝑒(−(𝑧−𝐻𝑒)

2

2𝜎𝑧2 )

+ 𝑒(−(𝑧−𝐻𝑒)

2

2𝜎𝑧2 )

………………….(4)

Where C(x,y,z) is the concentration of the emission (in micrograms per cubic

meter) at any point x meters downwind of the source, y meters laterally from the

centerline of the Plume, and z meters above ground level. Q is the quantity or mass of

the emission (in grams) per unit of time (seconds). u is the wind speed (in meters per

second), He is the height of the source above ground level (in meters) and y and

x are the standard deviations of a statistically normal Plume in the lateral and vertical

dimensions, respectively.




MASROERI

8

Furthermore, the stability of the atmosphere depends on the temperature

difference between an air parcel and the air surrounding it. Therefore, different levels

of stability can occur based on how large or small the temperature difference is

between the air parcel and the surrounding air. The atmosphere can be stable,

conditionally stable, neutral, conditionally unstable, or unstable. However, for

dispersion estimation and modeling purposes, these levels of stability are classified

into six stability classes based on five surface wind speed categories, three types of

daytime insulation, and two types of nighttime cloudiness. These stability classes are

referred to as Pasquill-Gifford stability classes and are depicted below. Table 4 shows

that stabilities A, B, and C refer to daytime hours with unstable conditions. Stability D

is representative of overcast days or nights with neutral conditions. Stabilities E and F

refer to nighttime stable conditions and are based on the amount of cloud cover. Thus,

classification A represents conditions of greatest instability, and classification F

reflects conditions of greatest stability.

Table 4 Pasquill-Gifford stability classes

Surface wind speed

At 10m (m/sec)

Day Night

Incoming solar radiation Thinly overcast Clear

Strong Moderate Slight Or 4/8 or3/8

<2 A A-B B D F

2-3 A-B B C E E

3-5 B B-C C D E

5-6 C C-D D D D

>6 C D D D D

Differing from a Plume model, when it is required to use the model for predicting

the dispersion of non-continuous air pollution, the Gaussian Puff models is preferable.

The Puff model releases emissions independent of the source, allowing the Puff to

respond to the meteorology immediately surrounding it. This also allows Puffs to be

tracked across multiple sampling periods until they have either completely diluted or

have tracked across the entire modeling domain and out of the computational area. The

Puff model can be written in the following equation form.

𝐶 =𝑄

(2𝜋) . 𝜎 𝜎𝑦𝜎𝑧 (

( )2

2𝜎 2 ) 𝑒𝑥 (

2

2𝜎𝑦2) [𝑒𝑥 (

( 𝑒)2

2𝜎𝑧2 ) + (

( 𝑒)2

2𝜎 2 )] .(5)

Where Cr is concentration of emission at receptor (g/m3), 𝑥r , 𝑦r, 𝑧r is the distance from

the origin in 𝑥, 𝑦, z (m), He is height of the funnel on ship’s deck, Q is level of emitted

gases (g/s), y, z, x is horizontal and vertical Plume deviation standard (m), U is

wind speed at the highest position on board (m/s), Δt is time different of dispersion and

t is time of dispersion.




MASROERI

9

5. RESULTS

In compiling data obtained from AIS, the software MySql Ver 5.0 was used. AIS

data compiled over a year was stored in the MySql server to ease data retrieval with

specific characteristics. Data from July 2010 to March 2011 was used, and among them,

specific times with the heaviest ship density was taken as a model. It was found that in

October 2010 the average ship density was 104 ships per day; October 22nd 2010 was

found to have the heaviest density with 120 ships on that day. Fig. 4 shows that at

17.00-18.00 PM was the time with heaviest ship density.

By using the emission calculation method developed by Trozzi as explained

previously, it was then found that the emitted gases produced by ships in 22nd October

2010 is as shown in Tables 5 and 6.

Fig. 4 Traffic density on 22

nd October 2010

50

55

60

65

70

75

80

85

90

00.0

0-01

.00

01.0

0-02

.00

02.0

0-03

.00

03.0

0-04

.00

04.0

0-05

.00

05.0

0-06

.00

06.0

0-07

.00

07.0

0-08

.00

08.0

0-09

.00

09.0

0-10

.00

10.0

0-11

.00

11.0

0-12

.00

12.0

0-13

.00

13.0

0-14

.00

14.0

0-15

.00

15.0

0-16

.00

16.0

0-17

.00

17.0

0-18

.00

18.0

0-19

.00

19.0

0-20

.00

20.0

0-21

.00

21.0

0-22

.00

22.0

0-23

.00

23.0

0-24

.00

Num

ber o

f shi

p

Time

Marine Traffic Density at Selat Madura 22 Oktober 2010 (00.00-24.00)




MASROERI

10

Tabel 5 Estimasi NOx, CO, CO2, VOC, SOx dan PM (kg/hour) 22 Oktober, 2010

Tabel 6 Estimation NOx, CO, CO2, VOC, SOx and PM (kg/hour) on 22 Oktober, 2010

Ship type NOx CO CO2 VOC SOx PM

General Cargo 195.91 434.08 155.20 101.62 291.00 5.82

Container 374.74 315.80 194.28 75.39 364.28 7.29

Bulk Carrier 62.97 178.11 57.57 41.56 107.94 2.16

Passenger Ship 23.37 21.90 15.70 5.31 29.43 0.59

Other 50.59 84.12 33.31 19.09 33.68 1.46

Tanker 189.32 317.57 129.62 74.49 243.04 4.86

TOTAL 896.90 1351.59 585.68 317.46 1069.38 22.18

After obtaining the amount of the emitted gases by using the Gaussian Plume and

Gaussian Puff model, we found the emission dispersion as described below. Fig. 5 and

Fig. 6 consecutively are the examples of emission distribution obtained by using the

Plume and Puff methods. Surfer 8, a contouring and 3D surface mapping program was

used to visualize the result. This paper only considers the NOx distribution instead of

all emission components. Fig. 5 shows that according to the Plume model, the NOx

distribution was concentrated in the Greasik and Tanjung Perak areas where the ports

of Surabaya and Port of Gresik are. However, Puff model results show a rather

different picture. It shows that NOx distribution mainly concentrated in the Madura

Strait and tends to move away from the origin. This is understandable considering that

Puff is a continuous model that provides time-based results, so it is very much affected

by wind direction and time of emission.

The CalPuff View (leading interface for Puff dispersion) shows that the emission

distribution of the Gaussian Puff model was rather independent of the origin. When the

Plume model accounts for conditions with high variability then a straight line steady

state of a certain distance from the origin is invalid. Therefore, for an unsteady state

condition, the Puff Model is more precise than Plume model. Conclusively, the Puff

Flag of Ship Freq. % NOX CO CO2 VOC SOX PM

Kg/hour

Indonesia 55 71% 422.82 839.08 317.80 195.22 579.66 12.04

Panama 5 6.5% 94.71 145.98 62.31 34.41 116.83 2.34

Antigua and Barbuda 2 2.6% 10.39 29.40 9.50 6.86 17.82 0.36

Liberia 2 2.6% 241.45 42.25 93.14 11.67 174.82 3.49

Marshal Island 2 2.6% 30.51 99.12 29.91 23.41 43.52 1.22

Netherland 2 2.6% 8.34 23.59 7.62 5.50 14.30 0.29

Norway 2 2.6% 27.18 76.88 24.85 17.94 46.60 0.93

Vietnam 1 1.3% 2.70 7.65 2.47 1.78 4.64 0.09

Cambodia 1 1.3% 28.66 2.44 10.54 0.79 19.77 0.40

Greece 1 1.3% 2.55 7.21 2.33 1.68 4.37 0.09

Hong Kong 1 1.3% 13.76 38.93 12.58 9.08 23.59 0.47

Iran 1 1.3% 0.95 2.69 0.87 0.63 1.63 0.03

South Korea 1 1.3% 11.73 33.19 10.73 7.74 20.11 0.04

St. Kitts and Nevis 1 1.3% 1.12 3.17 1.03 0.74 1.92 0.04

TOTAL 77 100% 896.90 1351.59 585.68 317.46 1069.38 22.18




MASROERI

11

model provides more precise results for a case with several characteristics as shown

below:

Fig. 5 NOx emission distribution using Plume model

Fig. 6 NOx emission distribution using Puff model at Δt = 45 minutes

Fig. 7 NOx distribution in Shoreline of the Madura Strait




MASROERI

12

1. Unsteady state case

2. There is possibility of interference from wind and complex topography

3. Non uniform land use pattern

4. Coastal effect is evident

5. Various wind directions

Fig. 7 shows the emission concentration in several areas along the Madura Strait

obtained by using the Puff model. It is shown that area around Kamal and Banyu Ujuh

in Madura Island has the highest NOx concentration of 0.176 μg/m3, followed by Gili

Barat in the second position with 0.132 μg/m3 and then Kalianak and Benowo in third

position with 0.11 μg/m3. Kebomas, Gresik, Port of Petrokimia, Lumpur, Kroman,

Sukodono and Tlogo Pojok are placed in fourth position with NOx concentration of

0.066 μg/m3.

6. CONCLUDING REMARKS

This paper demonstrates that the use of AIS in combination with GIS and

shipping database can be utilized to estimate the emission dispersion produce by ships.

The Gaussian Plume model can be used to perfectly simulate the emission distribution

for steady state conditions, while the Puff model enhances the results in case of

unsteady state conditions.

This kind of research will be even more applicable when it is designed based on

real time and dynamical situations so as to enable results to be used as parameters in

examining the quality of air and in controlling emission concentrations due to marine

traffic. Though this paper only considers the emission produced by ships, incorporating

those produced by industries will provide clearer and more sensible results.

ACKNOWLEDGEMENT

The authors would like to extend appreciation to Kobe University for all supports

including AIS equipment grants and comments and constructive benchmarks. The

author would also like to extend the sincere thankfulness to the Hitachi Scholarship

Foundation for making the collaboration with Japanese counterparts becomes possible.

REFERENCES

Altwicker, E.R., (2000) Air Pollution, Lewis Publisher, Muncie, Indiana.




MASROERI

13

Bracken, C., Carnemolla, A., Ritter, C., Zielke, E., (2007) An Analysis of Exhaust

Emission from a Large Ship Docked in Humbolt Bay, ENGR 416- Transport

Phenomena.

Cimorelli (2004) AERMOD : Description Of Formulation. United States

Environmental Protection Agency (EPA),

<http://www.epa.com/AERMOD/EPA-454R- 03-004.pdf>

Flang, Richard,C and Seinfeld J.H., (1998) Fundamental Of Air Pollution

Engineering, Prentice Halls, New Jersey.

Godish, T. (1985), Air Quality, Ball State University, Lewis Publishers,Muncie,

Indiana.

Ishida,T., (2003) Emission of Estimate Methods of Air Pollution and Green House

Gases from Ships, Journal of Japanese Institute of Marine Engineering, 37(1).

Jalkanen, JP [et al.] (2009) Modelling System for the Exhaust Emissions Of Marine

Traffic and Its Aplication In the Baltic Sea Area. Journal of Atmos. Chem. Phys.,

15229 - 15373 : Vol. IX.

Kerrie J. Long, Sue Ellen Haupt, George S. Young, and Chris T. Allen, (1994)

Characterizing Contaminant Source And Meteorological Forcing Using Data

Assimilation With A Genetic Algorithm, The Pennsylvania State University,

Department of Meteorology, University Park, Pennsylvania

Pingjian, L., Kobayashi, E., Ohsawa, T., Sakata, M. (2006) Case Study on Health

Assessments Related to a Modal Shift in Transportation, Journal of Marine

Science and Technology-JASNACE.

Pitana, T., Kobayashi, E., Wakabayashi, N. (2010) Estimation Of Exhaust Emission Of

Marine Traffic Using Automatic Identification System Data (Case Study : Madura

Strait Area, Indonesia), proc. OCEANS 2010 LEEE Sydney

Security Maritime AIS Ship Data (2007), IMO Maritime Safety Committee.

Seinfeld, J. H. (1975), Air Pollution; Physical and Chemical Fundamental. McGraw

Hill, inc.

Trozzi,C., Vaccaro,R., (2002) Methodologies For Estimating Air Pollutant Emission

From Ships, Tech Report MEET RF98b.

UNECE/EMEP (2002) Group 8: Other Mobile Source and Machinery, in

EMEP/CORRINAIR, Emission Inventory Guidebook-third ed. Technnical Report

no.30.

Wang, C., Callahan, J., Corbett, J.J., (2007) Geospatial Modeling of Ship Traffic and

Air Emissions, Proceeding of ESRI International Conference.

journal of maritime researches vol. 3, no. 1 march …system (ais) and geographic information system...

Documents