Modeling air quality in Hochiminh city and scenarios for reduction air pollution levels

<span class='text_page_counter'>(1)</span><div class='page_container' data-page=1>

VNU Journal of Sciencc, Earth Sciences 25 (2009) 179-191

M odeling air quality in Hochiminh city an d scenarios for

reduction air pollution levels

Ho Minh Dung*, Dinh X uan T hang

<i>Institute f o r E nvironm ent an d Resources, Vietnam N a tio n a l U niversity H o C hi M inh City</i>
Received 15 September 2009

A bstract. Air pollution in general, espccially air pollution from road traffic in particular in
Hochiminh City (HCMC) are at the alarming. Using models to simulate air quality is needed to ■
managc and predict the air pollution levels. Research results have prepared em ission inventory o f
air pollutants from road traffic, industry and domestic sources in HCMC. Besides, TAPOM and
FVM models vvere uscd to simulatc ihe meteorological conditions and air quality in HCMC.

Emission inventory results from road traffic show that em ission from motorcycles account for a
signiiìcant amount o f total load o f pollutant emissions from thai source. Sim ulation results o f air
quality give better results when using emission inventory with traffic em ission íactors were
cstimatcd in HCMC.

In addition, some scenarios to reducc pollution levels in general, especially air pollution ííom road
traffic in particular show that, if rcduce 50% number o f motorcycles (private transportation) and
incrcase 10 times numbcr o f bus (public transportation), air quality will m ore improve and can
rcducc trafíìc jam.

<i>Keywords: Road ưĩic, emission invcntory, models, scenarios.</i>

1. I n tr o d u e tio n

A ir q u ality m odel is an im p o rta n t tool to

m an ag e air p o llu tio n in u rban area s. W e c an use
m o dcls to p rc d ic t im p acts in the p ro cess o f
u rb anization, su ch a s d e v e lo p m e n t o f traffic
netw o rk , th e lo catio n o r cx p a n sio n o f
rcsidcn tial a n d in d u strial a r e a s , ...

In ou r co u n try in g eneral and H C M C in
particular, the stu dies aim to sim ulate air quality
w as initially im plem ented and ach icvcd som e
initial rcsults. T ypically thcre w erc scveral

research prcỹects a t all lev els an d m aster thesis.
H ow ever, o n c m a jo r lim itatio n in alm ost studies
o f d a ta o n th e e m issio n íacto rs (EF) o f air
p o llu tan ts fro m ro a d tra ffic w as used from
n e ig h b o u r co u n trics. In ad d itio n , due to traffic is
o n e o f th e m ain so ư rc e o f a ir p o llu tio n gcnerated
in H C M C s o th e s im u la tio n resu lts o f air quality
in th a t stu d ies h ad m o re o r less lim ited accuracy.
T h e re ío re , in th is stu d y , th e a u th o rs used results
o f th e stu d y e s tim a te d air p o llu tan ts E F in real
co n d itio n s o f tran sp o rla tio n activ ities in H C M C ,
so th e sim u latio n re su lts o f a ir q u ality w ill get
m ore p recisio n .

Corrcsponding author. Tel.: 84-8-38651132
E-mail:

</div>
<span class='text_page_counter'>(2)</span><div class='page_container' data-page=2>

180 <i>H .M . Dung, D .x . Thang ỉ VNU Ịournaỉ o f Science, Earth Sciences 25 (2009) Ĩ79-Ĩ91</i>

2. E m issio n in v e n to r y f o r a i r p o llu ta n ts

<i>2 .1. Em ission in vento ry daía</i>

T h e general m e th o d o lo g y fo r p rep aratio n o f
the cm ission in v e n to ry in clu d es th rce m ain
steps, id e n tiíy thc so u rces, so u rces
classification an d calcu latio n o f em ission s. A
tem poral reso lu tio n o f l h is uscd, and
calculation s are do n c for given w o rk in g day
Jan. 19, 2006. A s for th e spatial reso lu tion , the
EI is calcu lated in sid c grid u sed for m odeling
35km by 35km vvith 1 k m squ are cells.
E m issio n calcu la tio n s are in d iv id u ally d on c by
sourccs and h ad v alu c ch an g e s acco rd in g to the
each cells.

Three m ain sources are considered for the
emission calculations include: road traíTic,
industry and d om estic activitíes. Emission

T ablel. Road traffic emission

estim ations are done for param ctcrs o f air
pollutants N O x, c o , V O C s an d S 0 2. In particular,
the em issions from road traffic activities using
tvvo kind E F s data, EFs from C hina (used before)
and EFs determ incd in IIC M C .

<i>2.2. M ethodology a n d input data</i>

<i>a. R o a d tra ffìc so u rc e</i>

T raffic em issio n s are calcu latcd u sing three
m ain g ro u p s o f acliv ity d ata: the g eorefercnced

S tr e e t n e tw o rk , th e fleet co m p o sitio n , and
tem po ral a n d sp atial v aria tio n s o f circu latio n o f
this fleet. T h e cm issio n s are com p utcd h o u r per
hour, u sin g ho urly cm issio n co efficicn ts in day.
In this stu d y , w e use E F s o f a ir p o llu tan ts vvas
d ev clo ped in real co n d itio n s in H C M C by
au th o rs (H o M .D ., 2 0 0 8 ) [1,2].

factors in Hochiminh City

No. Pollutants MC (g/km.veh.) LDVs (g/km.veh.) HDVs (g/km..vch.)

InH CM C Chi n a ( ) In HCMC C h in af * InH CM C Chí na

1 NOx 0.05 ± 0.02 0.23 1.9 ±0.9 3.3 19.7 ± 5 .2 6.1

2 VOCs 2.34 ± 1.17 11.8 15.02 ±7.36 0.5 89.92 ±33.01 6.69

3 CO 21.85 ±8.67 17 <i>34.8 ± 15.5</i> 16.1 11.1 ± 5 .3 14.96

Note: MC (Motorcycle); LDVs (Light duty vehiclcs); HDVs (Heavy duty vehicles).
n : EFs from China (DOSTH, 2001 )[3].

<i>b. Ịn d u stry so u rc e</i>

U p to n o w , H C M C has th rec exporl

P r o c e s s i n g z o n e s a n d 12 i n d u s t r i a l a r e a s w i t h a

total area o f 2 ,3 5 4 ha. A t p resen t, H C M C has
abo ut 1,000 p la n ts, factories an d m ore than
33,000 sm all-scale p ro d u ctio n íacilities
han dicraíts. In H C M C , th e em issio n calculation
o f air pollution load b ased on the em ission
factors and the p ro d u ctio n p ro cess o f industries
that can bc applied:

G in = I K in. N jn (g /y ea r) (1)
W hich, Gin is em issio n o f p o llu tan t i for
sector n (g /y ea r); K m is cm issio n íactors o f
pollutant 1 for secto r n (g/ton s o f ravv m atcrials
o r products); Njn is a m o u n t m aterial o r fucl o f
íactory j for s e c to r n (ton s/year).

<i>c. D o m e stic so u rc e</i>

E m ission in v en to ry from h um an activ ities
p la y an im p o rtan t ro lc in m o d e lin g a ir quality.
S om e m a in activ ities g en era te p o llu tan ts such
as b u m in g fuel (D O , F O , LPG , co al, etc),
b u ild in g h o m e s, o ffic e s (p a in t and o th er o rg an ic
s o lv e n ts ,...)

<i>2.3. E m ission inventory resuỉts</i>

</div>
<span class='text_page_counter'>(3)</span><div class='page_container' data-page=3>

<i>H.M . Dung, p . x Thang</i> / <i>VNU lonrnal o f Science, Enrth Sciences 25 (2009)</i> 179-197

Table 2. Total emission o f a ư pollutants by sources in HCM C

181

No. Pollutants Road traíĩic (tons/year) Industry (tons/year)'2’ Domestic (tons/year)'31

EI-1 EI-2

1 NO, 30.161 33.822 41.310 3.878

2 CO 2.903.064 2.197.008 38.400 358.950

3 VOCs 405.062 931.188 30.900 44.213

4 <b>s o 2</b> <b>6.422</b> 80.370 7.110

Sources: (I)EMISENS; (2'3>INTEX-B (Zhang Q. và nnk, 2009) [4].
<i>2.4. Evaỉuaíion o f the EIs over a specific case </i>

<i>study in H CM C</i>

In o rd cr to assess thc tw o resu lts o f the EIs,
an a ir quality m odel (A Q M ) a t m esoscalc is
applicd for tw o d ay s Jan. 19& 20, 2006. T he goal
is to com pare th c co n cen tratio ns o f c o , N O x,
V O C s and O zone generatcd by A Q M for both
EIs (EI-1 and E I-2 ) w ith m easu rem en t values.

<i>a. M odeỉ description</i>

T h e m o dels T A P O M (T ran sp o rl and A ir
P o llu tio n M odcl, M artin i A . ct al., 2002 &
20 0 3 ) (5,6] a n d F V M (P in ite V o lu m e M odcl,
C la p p ic r A . et al., 1998) [7] d ev elo p cd at
L P A S -E P F L , a re u scd for this stu d y . T h cy arc
th rce d im cn sio n al E u lerian m o d els using terrain
fo llow ing grid a n d íìn ite v o lu m e d iscretizatio n .
T h e tran sp o rt and p h o to c h em istry m odel
T A P O M in clu d es the R A C M lu m p ed sp ccies
m c ch an ism chem ical so lv e r for gaseo u s phase
(S to ck w ell ct al., 1997) [8]. M eteo rolog ical
input d ata for T A P O M is o b tain s from thc
m odel F V M , w h o se b o rd ers can be forccd using

w in d and tc m p eratu re íìeld s fro m large scale
m odcl rcsu lts. FV M in clu d es an urban
tu rb u lcn ce m o d el w h ic h sp e c iíic a lly sim ulates
the cffccts o f urban areas on th e m eteorology
[9-14].

<i>b. </i> <i>C om parison o f sim ulated a n d observed </i>
<i>concentrations</i>

A íìrst c o m p a riso n w ith p rim ary pollutants
is co n d u cts ([(C|£i-Cob.)/Cob.]%). S im ulation
resu lts by EI-1 h av e v alu e is clo see r to the
m casu rem en ts, esp ccially for <b>c o , </b>prcsen ting an
avcragc p c rc c n ta g e d iffere n ce o f 20 % , w hcreas

for thc sim u latio n resu lts b y E I-2, the
p ercen tag c dilTerence p rc se n ts 35% . F or N O x,
th c p crcc n tag c d iffere n ce w ith resp ect to
ob scrv atio n s are sm aller for E I-1, alth ou gh not
so d iffere n t fro m EI-2 resu lts. T h is ind icates a
cle a r u n d erestim atio n o f <b>c o </b>em issio n s in EI-2,
w h ereas for N O x b o th EI-1 a n d EI-2 generate
sim ilar results.

1 4 .0

12.0

^

10.0
1» 8.0

J 6.0

<i><b>ằ 40</b></i>

2.0
0.0

1 5 9 1 3 17 21 2 5 2 9 3 3 3 7 41 4 5
H o u r ( h )

0.0

1 5 9 13 17 21 2 5 2 9 3 3 3 7 41 4 5
H o u r ( h )

200.0
16 0 .0
- 1 120.0

O)
&
<i>6</i> 8 0 .0

“ 4 0 .0

</div>
<span class='text_page_counter'>(4)</span><div class='page_container' data-page=4>

182 H.M. <i>D ung</i>, <i>D .x . Thang</i> / <i>VNU Ịoum al o f Science, Earth Sciences</i> 25 <i>(2009) 179-191</i>

S in ce th ere is no m easu rin g V O C s d ata, so
w e use co m p ariso n O zo n e co n cen tratio n as a
w a y to assess the ind irect to im pact o f the
ch an ge o f V O C s em issio n in p u t data. The
averag e con cen tratio n o f O zo n e in the
sim u lation ep iso d e an d th e m e asu re m en t values

at H B statio n sh o w th at for E I-1 , the p ercentage
d iffere n ce is ab o u t 10%, vvhereas for EI-2, the
p erce n tag e d ifferen ce is ab o u t 30% . T herefore,
\ve can con clu d e th at th e sim u latio n for EI-1
vvill g et re a so n a b lc lev els o f O zo n e

5-1
B-2
A O b s e r v e d

9 1 3 1 7 21 2 5 2 9 3 3 3 7 41

H o u r (h)

4 5

Fig. 2. Comparison 0 3 conc. at HB station on January 19&20, 2006.

<i>2.5. D iscussion a n d im plicatioỉis o f E I results</i>

<i>a. </i> <i>D istribution o f em issio n s b y s o u rc e a n d </i>
<i>regioìi</i>

E m ission in ventory an d d istrib u tio n o f air
p o llutants em ission from p o llu tio n so u rces in
H C M C for spatial an d tem po ral by u sing G IS
m ethod. D om ain w ith d im en sio n o f each cell
Ik m 2 and have 35 cell for X an d y d irectio n is
used in this study. D istrib u tio n th e em ission for
tem po ral estim ated as equ ation :

Eh = E a * fa * fw * fd /8 7 6 0 (2)
W hich, Eh, Ea are em ission load p er hour and
year, respectively; fa, fw, fd are coefficient o f
em ission distribution for each m onth, hour in day,
respcctively; and 8760 is total hours o f a year.

T h e em issio n load in each cell is
gam /km 2.h. W h ich, th e co efficien t o f em ission
distribu tion for each m onth, vveek, and hour
w a s estim ated for each d iffere n ce p ollution
sources.

T h e m o s t im p o rtan t c o n trib u tio n o f c o , V O C s
and N O x in H C M C is attrib u ted from road
traffic. A sim ilar relativ e so u rce stren gth is
found fo r o th er d ev elo p in g co u n tries, esp ecially
for C O , V O C s an d NO*.

F or S 0 2, N O x> C O an d V O C s, the 63, 30,
94 an d 6 9 % , o f th e total ro ad traffic em ission,
resp ectiv ely , co rresp o n d to m o to rcy cles. This
resu lt is en tirely reaso n ab le th a t 95% o f thc
total vo lu m e o f m o to r v e h ic le s in H C M C is
m o to rcy cles. T h e m ost im p o rtan t co ntribu tio n
o f C O , V O C s an d NOx in H C M C is attributed
from ro a d traffic. A sim ila r relativ e source
stren g th is fo u n d for o th er d ev elo p in g countries,
e sp e c ia lly fo r c o , V O C s an d N O x.

</div>
<span class='text_page_counter'>(5)</span><div class='page_container' data-page=5>

<i>II.M . D uny, D .x . Thang</i> / <i>V N U Ịournnl o f Science, Earth Sciences</i> 25 <i>(2009) 179-19Ì</i> 183

1 0 0%

90%
80%
70%
6 0 %
50%
40%
30%
20%

10%

0 %

■ M ortocycle □ LDVs □ HVDs ■ C ar □ B us

$02 N O x NM/OC c o

Fig. 3. Distribution o f the on-road vehicle emissions in HCMC by typc o f vehicle and by pollutant.

<i>b. </i> <i>L im ita tio n s o f th e m e tììo d a n d discussion </i>
W e can State t h a t u n c c r t a i n t i e s in o u r EI
co m c from three m ain so u rces: F irst, the q u ality
o f th e in p u t d a ta vve h a v e c o llc c tc d . S e c o n d , th e
ex trap o latio n b ased on thc cx istin g in fo m iatio n
to fill in thc rcm ain in g d a ta gaps. T hird, som e
a sp c c ts o f the m eth o đ o lo g y its e lf E M IS E N S
p ro cedu re, as it has b een d e v e lo p e d and initially
a p p lied in S o u th A m erican c o u n trie s an d S outh
V ietnam so it has so m e rcstric tio n s to im prove.
D csp ite o f lim itatio n s, all rcsu lts ab o ve allovvs
to co n clu d e that EI w ith E F from H C M C h av e
m o re reaso n ab le sim u latio n v alu es than EI w ith
E F from C hina.

3 . S im u la tio n o f m c tc o ro lo g y a n d a i r q u a lity

<i>3.1. Select sim ulation episode</i>

E pisode selcctcd for sim u latio n b ased on

several criteria:

In the dry seaso n (Jan u ary to A p ril) b ccau se
d u rin g th at tim e clo u d y sky, a p p ro p ria te for
F V M m odel;

C o n cen tratio n o f p rim ary a ir p o llu tan ts are
h ig h and stab le in the m o n ito rin g stations;

O zon e co n ce n tratio n is h ig h in the
m o n ito rin g station s an d o ften ex cccd stand ards
(1 8 0 ppb);

B ased on th c critcria above, th e period
ch o sen ío r sim u lation is Jan. 19 & 20, 2006.

<i>3.2. Settings in the m odel</i>

T o sim u late a ir q u ality in m esoscale
req uires p recisio n and reso lu tio n o f
m cteo ro lo g ical in p u t d ata. T o get the
req u ircm cn ts, FV M m odel is run b y u sing one
w ay n e stin g m eth o d w ith 5 dom ains.
D im cn sio n and reso lu tio n o f d o m ain s are
selected to sim u late m eteo ro lo g y co n d itio n s in
th e stud y a re a as follow s:

- D o m a in l ( D l) : D im en sio n o f d o m a in 2 0 X
2 0 c e l l s , s p a t i a l r e s o l u t i o n 1 5 0 k m X 1 5 0 k m .

T h is d o m a in co v ers a n area o ĩS o u th e a s t A sia
and a parl o f S o u th C h in a Sca;

- D o m ain 2 (D 2 ): D im en sio n o f d om ain 20

X 2 0 c e l l s , s p a t i a l r e s o l u l i o n 7 5 k m X 7 5 k m . T h i s

do m ain c o v e rs an arca o f S outh o f V ietnam ,
C am b o d ia, T h ailan d an d a part o f the South
C h ina Sea;

- D o m ain 3 (D 3 ): D im en sio n o f d om ain 33

X 3 3 c e l l s , s p a t i a l r c s o l u t i o n 1 6 k m X 1 6 k m . T h i s

do m ain c o v c rs a n arca o f th e Southern
p ro v in ce s and p a rts o f C entral S outhern
p ro v in ce s a n d S o u th C h in a S ca;

- D o m ain 4 (D 4 ): D im en sio n o f do m ain 35

X 3 5 c e l l s , s p a t i a l r e s o l u t i o n 7 k m X 7 k m . T h i s

</div>
<span class='text_page_counter'>(6)</span><div class='page_container' data-page=6>

184 H.M. <i>Dung, D .x . Thang ỉ V N U Ịournal o f Science, Earth Sciences</i> 25 <i>(2009) 179-191</i>

- D om ain 5 ( 0 5 ) : D im en sio n o f dom ain 38

X 3 8 c e l l s , spatial r e s o l u t i o n l k m X l k m .

<i>3.3. Bounclary a n d in itia l co n d itio n s</i>

T h is sim u latio n u scs 6 -h o u rly d ata from the
N C E P /N C A R (2 0 0 6 ) rcan a ly sis dataset for its
initial and b o u n d ary co n d itio n s. D ata have 2.5 <b>X </b>

2.5 degree global rcso lu tio n w ith 17 p r c s s u r e

levels at tim es 0 Z , 6 Z , 12Z an d 18Z.
<i>3.4. T o p o g ra p h y a n d ỉ a n d use d a ta</i>

Input data fo r F V M m ođel also includes
topography, land use, c h ara cteristics o f the soil,
roughness, h u m id ity an d th crm al. A ll d atabases
is to ok from U SG S vvith lk m resolutio n.

4. R csu lỉs a n d d is c u s s io n

<i>4.1. M eíeoroỉogical sim uỉations</i>

<i>a. </i> <i>B o u n d a ry a n d in iíia l co n d itio n s </i>
T he m odel is first applied to a 3.000km <b>X </b>

3.000km grid, aim in g to g enerate adequate

C entral o f t h i s d o in a in coiiìcidcs vvitlì the center
o f H C M C .

boundary and initial conditions for o u r m esocale
dom ain. A very sim ilar behav ior o f the vvind
p attcrns bctvveen th e tw o days o f the episodc in

sim ulated by th e m otlcl (in range N E to SE
dircction). From 5h to 12h, the m ainly vvind
dircction often betw een SE and N E directions,
sim ilar behav io r o f th e sea vvind. From 12h to
14h, as influenced by sea-con tinen lal vvind so the
m ain ly vvind d irectio n is SE.

<i>b. </i> <i>M eso sca le sim u la íio tĩ </i>

</div>
<span class='text_page_counter'>(7)</span><div class='page_container' data-page=7>

/ / M <i>D ung,</i>D.x. 77ww^ / <i>V N U louninl o f Science, Eartỉt Sciences 25 (2009) 179-191</i> 185

co rrectio n is too k fro m U SG S and land u se in
H C M C . D om ain D5 is ch o scn for sim ulate
m eteo ro lo g ical as m e so scale in cp iso de o f
study. M easu rem en t d ata from T a n S o n N hat
(T S N ) station vvas u scd to co m p are vvith the
sim ulatio n values.

<i>c. C o m p a re sim ulcttion re su lt a n d ob served</i>
<i>+ W ind direcíion a n d m n d s p e e d :</i>

S im ulation resu lts from F V M shovv w ind
speed depen d on the ty p c o f su ríace . O n the sea,
th e vvind v ecto r is sta b lc o n d irectio n and value,
\vind spccd in lo \v er \vhcn com e to the
co n tin en t. W ind d ircc tio n an d w in d speed in
co n tin en t is ch an g e a lot and d ep en d on the
d istrib u tio n o f the su ría c e th erm al. In urban
area, vvind sp eed is lo w cr b u t n o t clear.
C o m pare w ith th e m e asu rc m en ts sh o w w ind

direction co n sisten t relativ ely . C om p ariso n
b etw cen th c sim u latio n w ith m easurem ent

values in ep iso d c study havc co ư elatio n
c o e íĩic ie n t R = 0 .6 8 and 0.81 fo r \vind direction
and w in d sp ecd , respcctively .

<i>+ T em p era ỉu re</i>

S im u latio n rcsu lts show th e effects o f sea to
t h e s u r í a c e tem p eratu re d i s t r i b u t i o n , Coastal
areas have lo w an av crag e tem peratu re and
tem p cratu re v aria tio n s. W h cn g o in g into the
co n tin en t, te m p eratu re v aria tio n s an d average
tem p eratu re tc n d s to in c re ase. T h e highest
tem p eratu re in d o m ain D5 a p p e a r in the cen ter
o f H C M C , th c rcg io n h av e trccs an d \vater
su rface area is low . T h e m odel p red icts w ell the
tim e o f the d ay w hcn tem p eratu re start
in creasing d u e to the su n rise (aro u n d 6h - 7h),
as vvell as th c tim c o f th c m ax im u m value
(betw cen 12h - 13h). C o m p a risio n betw een
sim ulatio n a n d m e asu re m en t v alu es in episode
stu dy h a v e co rrelatio n c o e ffic ic n t R =0.92.

■ọ
<i>ị</i>
<i>ị</i>

<i>ề</i>

360.0
300.0
g 240.0
o
2 180.0
€
<i>ị</i> 120.0
<i>ề</i> _60.0
0.0

1 5 9 13 17 21 25 29 33 37 41 45

<b>Hour (h) </b> <b>Hour (h)</b>

Fig. 5. C om parison vvind speed and w ind d ừ e c lio n b etw een sim ulation a n d obscrv cd on Jan. 19& 20, 2006.

<i>4.2. A ir quality m odeliỉìg</i>

E m ission in v e n to ry is c a lc u la te d for 24
h o u rs in Jan . 19, 2 0 0 6 w ith l h tem poral
r e s o l u t i o n a n d l k m X l k m sp atial rcso lu tio n in
do m ain 3 5km X 3 5 k m . B o u n d ary an d initial
co n d itio n s w ere p re p a rc d w ith the sam e
datab ases. A p re-ru n o f o n e d ay vvith the sam e
em issio n s and w ind ficlds is c o n d u c te d for all
th e sim ulatio ns, in o rd e r to p ro vide m ore
realistic initial co n ditions.

<i>a. P r im a r y a ir p o llu ta n ts</i>

</div>
<span class='text_page_counter'>(8)</span><div class='page_container' data-page=8>

186 <i>H M . Dung, D .x . Thang</i> / <i>V N U Ịournal o f Science, Earth Sciences 25 (2009) 179-Ĩ9Ĩ</i>

c o n c e n tra tio n s is d iff e re n c e , th e p c a k is in the
m o m in g (7 -9 h ) a n d a t n ig h t (2 0 -2 lh ).

B o th s im u la te d a n d m e a s u re d
c o n c e n tra tio n s o f c o a n d N O x s h o w im p o rta n t
m o m in g p c a k . T h is p e a k is re la te d to h ig h
e m is s io n s fro m r o a d tra ffìc in th e m o m in g ru sh
h o u r a n d lovv m ix in g h e ig h t. T h e in te n s itie s o f
th e p e a k s o f b o th c o a n d N O x a re g o o d
a g re e m c n t w ith o b s e rv a tio n s a t a ir q u a lity

m o n ito rin g s ta tio n H B a n d D O . Im p o rta n t
n ie h tly p e a k s o f c o a n d N O x (a ro u n d 2 1 h),
a p p e a r b o th in th e s im u la tio n and
m e a s u r e m e n ts . T h is p e a k is a ls o re la te d to the
ro a d tra ffíc a n d it is s o m e tim e s o v e re s tim a te d
b y th e m o d e l.

W ith S 0 2 v a lu e , th e c o m p a ris o n betvveen
s im u la tio n a n d m e a s u re m e n t s h o w th e sam e
p a tte m .

<i>ỈU.V</i>
60.0
^ 50.0
" I 40.0
o 30.0

<b>c</b>

o 20.0

10.0

0.0

1 5 9 13 17 21 25 29 33 37 41 45

<b>Hour(h)</b>

0.0

1 5 9 13 17 21 25 29 33 37 41 45
H o u r ( h )

Fig. 6. C om parison c o conc. at D O station (left) and S 0 2 conc. (right) at T N station o n Jan. 19& 20, 2006.

<i>b. S eco n d a ry p o llu ta n t - O zơĩie </i>
<i>+ Spa tial distrib u tỉo n</i>

T h e s p a tia l d is trib u tio n o f O z o n e is
g e n c ra te d d c p e n d in g o n th e p rim a ry p o llu ta n t
( N O x & V O C s ) c o n c e n tra tio n s a n d the
m e te o ro lo g ic a l c o n d itio n s . F o r th is e p is o d e ,
p o llu ta n ts a re p u s h e d b y w in d c o m in g fro m S E
d ire c tio n , w h ile O z o n e is b e in g fo rm e d w ith
m a x im u m v a lu e s a t m id d a y . A f te r m id d a y ,

w in d d ire c tio n m o v e a little to E d ire c tio n . A t

</div>
<span class='text_page_counter'>(9)</span><div class='page_container' data-page=9>

/7.M. <i>D ung, D .x . Thang / VNU Ịottrnal o f Science, Earth Sciences 25 (2009) 179-191</i>

₁₈₇

25

15

5

J a n . 1 9 / 2 0 0 6 1 4 0 0 L T , O z o n , p p b

5 1 5 2 5

Fig. 7. M apo fO zo ne concenlration (ppb) indomain D5 at lOh, 12h, 14h& 16h on Jan. 19, 2006.

<i>+ C o m p a riso n m t h m ea su re m en t values </i>
T h e O zo n e m e asu re m cn ts c o ư o b o ra te the
p rc sc n c e o f th e plum e o v c r th e c ity , \vith high
co n ce n tratio n in Ja n . 19, 2 0 0 6 a t D O & IIB
statio n s (15 2.3 ^ig/rn3, 8 4 .8 fig /m \ respectivcly )
an d lo v cr v alu es in Jan . 20, 2 00 6 . T h is m ight
im p ly th a t th e city p lu m c rem a in s m ain ly in the
city c c n te r on 19 Jan . b u t is slig h tly m oved
to\vards S E on 20 Jan .. T h e sim u latio n shovvs
high O zo n e lev els at th c sam e statio n s as the
m e asu rc m en t v alu es o n 19 Jan u ary , in d icatin g a
goo d rep ro d u ctio n o f th e p lu m e p osition.

<b>5. S cen a rio s to r e d u c e a ir p o llu tio n levcls</b>

<i>5.1. Em ission scenarìos</i>

D u e to th e ro a d traffic is p lay im portant
sou rcc ro le to air p o llu tio n in H C M C in general,
so the p ro p o sed sc e n a rio s to re d u c e pollution
lev cls from th is so u rcc is n ecessary . T w o types
o f tran sp o rla tio n m e an s sp ecial interested is
m o to rcy cle an d b u s. S cen ario s to rcduce air
po llu tio n lev els is b a sc d on tw o m ạjo r criteria:
(a) ch an ge the n u m b e r o f ty p es o f transportation
and (b) ch ang e the ty p e o f fuel used.

J a n . 1 9 / 2 0 0 6 1 0 0 0 L T , O z o n , p p b J a n . 1 9 / 2 0 0 6 1 2 0 0 L T , O z o n . p p b

</div>
<span class='text_page_counter'>(10)</span><div class='page_container' data-page=10>

188 <i>H .M . Dung, D .x . Thang</i> / <i>V N U lournaỉ o f Science. Enrth Sciences 25 (2009) 179-191</i>

H C M C h a s 3.584 strects vvith total road
le n g th ab o u t 3 .6 7 0 km , th e area o f p avcm ent is
3 6 m illio n m 2, s o the ratio o f ro ad area đensity
an d c ity area o n ly abo ut 1.8% (k m /k m 2), m uch
l o w c r t h a n t h e c o m m o n S ta n d a r d o f d e v e l o p e d
c o u n trie s 10-20% .

A rea for a m o to rc y c le travel is 10 - 12m 2,
w h ile arca o f a se a t on th e bus o n ly 2rn arca o f
ro ad . W ith a b o u t 4 m illion m o to rb ik cs will
a c c o u n t for ap p ro x im a te ly 4 0 m illio n m 2 travel.
B ut, H C M C o n ly build a n cw o r ad d in g about

1 % o f ro ad s u ría c e e v ery year.

In ad d itio n , fuel co n su m p tio n in d icato r for a
p a sse n g e r is 0.015 liters/k m w hcn using
m o to rc y c lc s a n d 0 .0 0 4 4 liters/km vvhcn using
th c b u s. T h u s , w h en uscd as a m e an s o f
tra n sp o rt, m o to rcy cle s h av e a fucl co n su m ptio n
h ig h e r th an 3.4 tim es o f the bus.

T h e re ío re , o n ly increase thc b u s and reduce
th e m o to rc y c le vo lu m es can so lv e traffic
c o n g estio n in th e statu s o f th e cu rren t tig h t line,
savc fuel an d re d u c e po llutio n.

<i>a. C h a n g e th e n u m b e r o f veh icle types</i>

S o m c sc e n a rio s to red u ce air p ollution
le v e ls fro m ro a d traffic w as p ro p o sed as
follow s:

- S c e n a rio 1 (S c e n l): redu cc 50% o f
m o to rc y c le v o lu m e trav el th ro u g h o u t th e city at
th e sam e tim e s o the load o f p o llu tan ts em issio n
fro m m o to rc y c le s w ill d ecrease 50% .

- S c e n a rio 2 (S cen 2): red u ce 50% o f th e
m o to rc y c le v o lu m e, at th at tim e to fill the
<i>d cm an d fo r tra v e l Vĩ re s t o f th e p eo p le to </i>
in c re ase th e v o lu m e o f b u s 10 tim es o f the
c u rre n t n u m b e r. B esides, c h an g e sm all size

b u ses (3 5 -4 0 seats: B 3 5 -B 4 0 ) for th e ty p e s are
c o m m o n ly u s e d (55 -80 seats: B 55 -B 80). The
nevv siz e m c an s in acco rd an cc vvith th e existing
ro ad traffic in H C M C w here th e w h o le city has
o n ly 14% o f the ro ad 12 m e te rs vvidth
(c o n v e n ic n t fo r th e b u s B 5 5-B 80 ), 51% o f the

road from 7 -1 2 m w id th an d 35% o f th e road
w idth less than 7m . In ad d itio n , efficiency
buses m u st be fro m 8 0 -1 0 0 % (cu rren tly 40-
4 5 % for th e larg e bu ses).

- S ccn ario 3 (S cen 3): red u ce 50% o f the
m otorcycle v o lu m e, a t th at tim c to íìll thc
<i>dem an d for trav el Vi rest o f th e peo plc to </i>
increasc the v o lu m e o f b u s 05 tim cs hig h cr than
the cu rrcnt n um b er. E fficien cy b u ses have at
least 80% (cu rren tly 4 0 -4 5 % ).

<i>b. C hatìge th e ty p e o f fu e l</i>

C h angc u sin g clean fuel (L P G , C N G , ..)
instcad o f luel b c in g u sed is o n e o f th e solutions
no t o nly gct c co n o m ica l b e n e íit but also
co n trib u te to rc d u c e a ir p o llu tio n from road
traffic sourcc.

U sing C N G (C o m p re sse d N atu ral G as)
c o s ts o n ly a b o u t 5 0 -6 0 % o f th e tra n s p o rt th a n
gasolinc, oi!, b u t rc d u c in g to 3 5 % o f C*Hy, 60%

C O an d 10% N O x, ... e m ittc d into the
environm cnt. W ith so m e o th e r b c n e íits such as
an ti-abrasion , in c re a se d en g in e lifc, rcduccd
m ain tcnance costs.

C o m p a riso n w ith g aso lin e, m o to rcy cle run
on 4 0 % gas sav in g fuel co sts and cnv iron m cntal
p o llu tio n levels re d u c e d o v er 70% .

<i>5.2. Sim uỉation resu lts a ir q u a lity in th e scenarios</i>
W ith 03 scen ario s a s m e n tio n ed ab o v e show
th e em issio n load c a lc u la te d is a different
am o u n t o f em issio n load a t th e íìrs t o f study
(b ase casc). D ep en d in g on th e a ir p o llu tan ts and
scenarios th at c m issio n s o f air p o llu tan ts are
increase o r dccrease.

<i>a. C o n cen tra tio n s o f p r im a r y p o llu ta n ts</i>
S im u lation rc su lts o f air qu ality in 3
sccnarios ab o v e m c n tio n cd w ith prim ary
po llu tan ts c o n c e n tra tio n s co m m cn ts as follow s:

</div>
<span class='text_page_counter'>(11)</span><div class='page_container' data-page=11>

<i>H .M . Dutĩq, D .x Thang</i> / <i>V N U Ịoum al o f Science, Enrth Sciences 25 (2009) 179-191</i> 189

case). H o w ever, in sccn ario s 2 an d 3,
co n centration s o f N O x ìn crcasc vvith thc base
case value sim u latio n (1 .2 tim cs and 1.1 tim es
hig h er for S ccn2 an d S ccn3 , respectively)
bccause o f N O x em issio n s load in scenario s 2&
3 increase vvith thc v alu e calcu la ted em issions

load initially.

- C O co n ccn tratio n s sig n ifican tly rcd u ccd in
3 scenarios w ith b asc casc sim u latio n value,

I l o u r (h)

this co m p lete ly rc a so n a b le w h en the load
em issio ns c a lc u la tc d o f <b>c o </b>in th e 3 scen ario s
are low cr (o n ly 5 3 -5 4 % ) em issio n s load in the
íìrst calcu latio n .

- S 0 2 c o n c e n tra tio n from sim u latio n resu lts
in 03 sccn ario s arc lig h t d ecre ase vvith results
sim u latio n by th c initial lo ad b e c a u se S 0 2
em issio n s load c a lc u la te d in 3 sc e n a rio s are
red u ccd by 69 -7 4% .

1 5 9 13 17 21 25 29 33 37 41 45
H o u r (h)

Fig. 8. Comparison c o conc. at HB station (leít) and Ozone conc. at DO station on Jan. 19&20, 2006.

<i>b. </i> <i>C o n cen tra tio n o f s e c o n d a ry p o llu ta n ts </i>
C o n cen tratio n s o f seco n d ary pollutant
O zone from 3 sccn ario s sim u latio n are low er
w ith th e sim u latio n resu lts o f the first cm issio ns
calculated. C o m p ariso n o f O zone
co ncentration s sim u latcd from 03 scen ario s and
base case sim u latio n w cre im p lcm en ted at air

qu ality m o n ito rin g statio n s D O an d HB.

<i>5.3. Support m ethod</i>

B esides o n the p ro p o sc o f sccn ario s to
reduce air po llu tio n lev els fro m ro ad traffic by
ch ang ing the n u m b e r and ty pe o f vehiclcs and
fuel used, the su p p o rt m e th o d to rcd u ce air
po llu tio n fro m ro ad traffic sh ou ld bc also o f
interest:

- P ro p o se to d o inspection an d control
em issio n s p erio d ically for o th cr m can s o f

tran sp o rlatio n 1 tim cs p e r y ear. F or
m o to rcy cles, d u c to for thc h ig h v o lu m es so in
in itially c an b e re q u ire d fo r c a te g o ry o f
m o to rcy cle w ith ca p a c ity cn g in e < 50 cm 3.
T h ese m o to rcy cle s h av e h ig h p o llu tio n
em issio n le v c ls w ill b e req u ire d to u pg rade,
m ain ten an ce o r rc p la c e m e n t o f sp are p arts, the
eng in e o r ío rb id d e n travcl.

- E n co u rag e, e x te n siv e p ro p a g a n d a and
stron gly in th e p c o p lc a b o u t th e co n scq u en c es
o f the m o to rcy clc c x p lo sio n , the lo n g -term risk
to the d ev elo p m cn t o f th e city an d nex t living
g enerations.

</div>
<span class='text_page_counter'>(12)</span><div class='page_container' data-page=12>

190

H.M. <i>D u n g</i>, <i>D .x . Thang</i> / V7VƯ <i>Ịournaỉ o f Science, Earth Sciences 25 (2009) 179-19Ĩ</i>

- R estrictio n o f cars trav el in to the Central
c ity d u rin g th e m o m in g rush h o u rs (7 - 9h) and
ev e n in g (2 0 - 21h). T h e s e are tim es w ith high
c o n c e n tra tio n o f air p o llu tan ts in sim u latio n and
m easu rem cn t.

<b>C o n clu sio n</b>

R e se a rc h results o b ta in e d in th is study is to
d ev elo p th e em issio n s d ata o f air pollutants
fro m m a jo r o f air po llu tio n so u rces in H C M C ,
esp ecial fo r tran sp o rta tio n so u rce. B esides,
T A P O M an d F V M m o d els w ere used to
sim u la te th e m eteo ro lo g ical co n d itio n s an d air
q u ality in H C M C . E p iso d e w a s chose íor
sim u la tio n o n Jan uary 19& 20 20 0 6 , in the đry
seaso n o f year.

E m issio n in v en to ry o f air p o llu tan ts from
ro ad traffic sh o w th a t em issio n load o f
m o to rc y c le s accou nt for a sig n ifican t am o u n t o f
total em issio n o f th a t source. A ir quality
sim u la tio n get b etter resu lts w hen using
em issio n in v en to ry c a lc u la te d from the
e m issio n facto rs d ev elo p ed in H C M C com pared
w ith c a se u sing the em issio n íacto rs from
C h in a.

F V M an d T A P O M m o d els are ch o se to

sim u late m e teo ro lo g y an d a ir q u ality in H C M C .
T h e sim u latio n ep iso d e for re se a rc h results
s h o w th at, th e re are n o t so d iffere n ce betvveen
sim u la tio n an d m e asu re m en t v alu es. P articular,
at a ir q u ality m o n ito rin g statio n H B , c o conc.
is 5 % d iffere n ce b etw een sim u latio n and
m easu re m en t. F o r o th e rs p aram eters N O x , S 0 2
a n d O z o n e c o m p a ris o n re s u lts b e tw e e n
sim u latio n an d m e asu re m en t has sim ilar results.

B esid e s, b ased on th e sim u latio n resu lts and
c u rren t sta tu e o f a ir p o llu tio n in H C M C w ith air
p o llu tio n from ro ad tra ffic p la y s a n im portant
ro le in a ir p ollution in g en eral, th e authors
p ro p o s e d so m e scenarios to re d u c e a ir pollution

lev els in general an d a ir p o llu tio n from road
traffic in p articu la r. T h e sim u latio n resu lts from
scenarios shovv th at, i f red u ce 50% n u m b c r o f
m o to rcy cles an d in c re ase 10 tim es n um b er o f
buses, c o n cu rren tly , chan ge all thc present
bu ses (B 5 5 -B 8 0 ) to sm all size b u ses (25-30
seats), air q u a lity w ill b e im p ro v e an d can be
reduce tra ííìc ja m in rush hour.

A ll sim u latio n resu lts abo ve c an h elp us to
und erstand ab o u t air p o llu tio n in H C M C in
gcneral an d air p o llu tio n íro m ro ad traffic in
particular. B ased on th is study, vve can d evclop
research fu rlh e r ab o u t m eteoro log ical

co n d itio n s and a ir q u ality in lo n g e r pcriod to
get m ore p re c isio n results.

R e fe rc n c c s

[1] Ho Minh D ung, Dinh Xuan Thang, . “The
cstim ation o f cm ission íactors o f air poilutants
from road tra ffic in Ho Chi Minh C ity”, <i>Journaỉ </i>
<i>o f Science</i> - <i>Earth Sciences</i>, 24 (2009) 184.
[2] Hồ M inh D ùng, Đ inh X uân T háng, Nghiên cứu

xây dưng hệ s ố phát thải chất ô nhiễm từ phương
tiện giao thông đư ờng bộ tại thành phố Hồ Chí
M inh. <i>Tạp c h ỉ BtìLDy</i> s ố 170 (2009) 25.

[3] DOSTE (D epartm ent o f Science, Technology
and Environm cnt o f H ochim inh City), Urban
transport en crg y dcm and and em ission analysis -
Casc s tu d y o fH C M city. N ° 1 (phase II) (2001).
[4] Q. Z hang, D .G . Streets ct al., Asian em issions in

2006 for N A SA INTEX-B m ission. <i>Atmospheric </i>
<i>Chemistry and Physics</i> 9 (2009) 5131.

[5] M artilli A., C lappier A. and Rotach M. s , An
Urban Surface Exchangc param cterisation for
m csoscale m odels. <i>Boundary-Layer</i>
<i>Meteoroỉogy></i> 104 (2002) 261.

[6] Martilli A., R oulet Y., Junier M ., K irchncr F.,

On the im pact o f urban surĩace cxchangc
param ctcrisations on air quaỉity sim ulations: the
Athens case. <i>Atmos. Env.</i> 37 (2003) 4217.

</div>
<span class='text_page_counter'>(13)</span><div class='page_container' data-page=13>

H.M. <i>D ung, D .x . Thang</i> / <i>V N U Ịoum aỉ o f Science, Earth Sciences 25 (2009) 179-Ĩ9Ĩ</i> 191

đim cnsional transport. <i>Monthìy Weather Review </i>
1 2 6 (1 9 9 8 )2 3 2 .

[8] Stockwell w. R., K ircchncr F., Kuhn M. (1997).
A new m echanism for rcgional atm ospheric
chcm istry m odclling. <i>ủournaì o f Gcophysical </i>
<i>Research</i> 102 (D 22) 25847.

[9] A. Hauric, J.J.E. Kúblcr, A. Clappicr, A
M etam odeling approach for intcgratcd
asscssm cnt o f air quality policics. <i>Environmental </i>
<i>Modelitxg & Assessm ent</i> 9 (2004) 1.

[10] Junicr M ., G as phasc chcm istry m cchanism s for
air qu ality m odclling: gcncration and application
to case studics, <i>PhD Thesis</i> No. 2936 (2004),
EPFL, Sw itzcrland.

[11] M . Lin, T. O k i, T. Hollovvay, D.G. Strccts,
Long-rangc transport o f acid iíy in g substanccs in
East Asia-Part I: M odcl cvaluation and
scnsitivity studics. <i>Atmos. Env.</i> 42 (2008) 5939.
[12] M ullcr c., Im provcm cnt o f an U rban turbulcncc

param ctri/.ation for m eteorological opcrational
forccast and air Quality m odeling, PhD T hesis
N o.3766 (2007), EPFL, Sw itzcrland.

[13] Zaratc E., B clalcaxar L. c., C ỉappicr A ., Air
quality m odclling ovcr B ogota, C olom bia:
C om bincd tcchniqucs to cstim atc and cvaỉuatc
cm ission invcntorics. <i>Atmos. Env.</i> 41 (2007)
6302.

</div>

<!--links-->