Many studies related to air quality management in HCMC have been done recently, including “Air pollution forecast for Ho Chi Minh City, Vietnam in 2015 and 2020”[4]to simulate pollutants
Trang 1Original Research Article
impacts on human health
Bang Quoc Ho
Department of Air Pollution and Climate Change, Institute of Environment & Resources, Vietnam National University, Ho Chi Minh City 0084, Viet Nam
a r t i c l e i n f o
Article history:
Received 6 April 2016
Received in revised form
10 August 2016
Accepted 12 October 2016
Available online xxx
Keywords:
Air pollution
Health effects
Mortality
Ho Chi Minh City
a b s t r a c t
According to World Health Organization (WHO) and Global Burden of Disease, ambient air pollution is estimated to be responsible for 3.7 million premature deaths in 2012[1] Therefore, it is urgent to es-timate the impact of air pollution on public health and economic damage The objectives of this research are: study the distribution of PM10concentration over Ho Chi Minh city (HCMC) and relationship to public health and for proposing solutions of diseases prevention in HCM, Vietnam EMIssion SENSitivity model was applied to conduct air emission inventory for transportation sector Then, Finite Volume Model and Transport and Photochemistry Mesoscale Model were used to simulate the meteorology and the spatial distribution of PM10in HCMC Together with disease data obtained, the US Environmental Benefits Mapping and Analysis Model was applied for calculating the number of deaths and estimating economic losses due to PM10pollution Finally, solutions to reduce PM10pollution and protect public health are proposed The results showed that the highest 1-h average concentration of PM10 is
240mg m3in North Eastern of HCMC The concentration of PM10for annual average in District 5 ranged from 17 to 49mg m3 There are 12 wards of District 5 with PM10concentration exceeding the WHO guidelines (20mg m3for annual average of PM10and 50mg m3for 24-h average) The high concen-tration of PM10causes 5 deaths yr1in District 5 and 204 deaths yr1in HCMC, and it causes economic losses of 1.84 billion of USD
© 2017 Chinese Institute of Environmental Engineering, Taiwan Production and hosting by Elsevier B.V This is an open access article under the CC BY-NC-ND license (
http://creativecommons.org/licenses/by-nc-nd/4.0/)
1 Introduction
Ho Chi Minh City (HCMC) is the most dynamic area as a social,
cultural and economic center of Vietnam District 5 located at South
of HCMC, has an area of 4.27 km2with population of 176,890 people
and density of 41,426 person km2 It is one of the most density
areas with energetic economic activities Services, exchange of
goods and traffic in this area have been expanded for a long time
which drive economic growth Consequently, environment has
become polluted, and the citizens have to face many environmental
problems Especially in traffic jam conditions, air quality becomes
worse which directly affects people's health In a recent study of
relationship between air pollution and human health, over 90%
children less than 5 years old in HCMC were infected to respiratory
disease[2] According to World Economic Forum in 2012, Vietnam
is one of 10 countries which has the worst air pollution in world-wide[3]and in urban area as HCMC, traffic is the major contributor
to air pollution [4] Pollutants emitted from these sources is considered as hazardous pollutants by US Environmental Protec-tion Agency (USEPA), especially particulate matter PM2.5is believed
to cause respiratory disease, lung cancer and mortality In 2012, International Agency for Research on Cancer (IARC) has classified diesel engine emission to Group 1 Carcinogenic to humans[1] The IARC also reported that emissions from diesel engines from trucks, cars, train or boat were one of the major cause of lung and bladder cancer For those reasons, this study was conducted to clarify the relationship between PM10concentration and mortality in HCMC especially in District 5 and estimate the economic losses
Many studies related to air quality management in HCMC have been done recently, including “Air pollution forecast for Ho Chi Minh City, Vietnam in 2015 and 2020”[4]to simulate pollutants over HCMC (NOx, CO, SO2, O3, but not PM10);“Optimal Methodology
to Generate Road Traffic Emissions for Air Quality Modeling: Application to Ho Chi Minh City”[5]focusing on air emission for
E-mail address: bangquoc@yahoo.com
Peer review under responsibility of Chinese Institute of Environmental
Engineering.
Contents lists available atScienceDirect
Sustainable Environment Research
j o u r n a l h o me p a g e : w w w j o u r n a l s e l s e v i e r c o m / s u s t a i n a b l e
-e n v i r o n m -e n t - r -e s -e a r c h /
http://dx.doi.org/10.1016/j.serj.2017.01.001
2468-2039/© 2017 Chinese Institute of Environmental Engineering, Taiwan Production and hosting by Elsevier B.V This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ).
Sustainable Environment Research xxx (2017) 1e8
Trang 2road transportation for Optimal Methodology to Generate Road
Traffic Emissions for Air Quality Modeling: Application to Ho Chi
Minh City; and“Estimation of Road Traffic Emission Factors from a
Long Term Tracer Study” focusing on calculating air emission factor
for volatile organic compounds and NOx The previous researches
have not focused on PM10and there is no impact study of PM10on
human health conducted in Vietnam
2 Method and data
2.1 Method
In general, the method of this study is described inFig 1 First, all
information on traffic, industrial activities and household's fuel
consumption data were collected for calculation of PM10emission
inventory Then FVM (Finite Volume Model) e a meteorology
model and TAPOM (Transport and Photochemistry Mesoscale
Model) - a dispersion model were applied to make spatial
con-centration distribution of PM10 Based on WHO guideline for PM10
(20 mg m3 for annual mean), areas with PM10 concentration
exceeding WHO guideline were made Finally, applying theory of
the Environmental Benefits Mapping and Analysis Program
(Ben-MAP), a GIS computer program developed by USEPA to simulate
impact of air pollution change to human health and economic losses and to estimate the mortality cause by PM10in study area
2.2 Data for PM10emission inventory The PM10emission inventory was carried out for HCMC for the year of 2012
2.2.1 Traffic source EMISENS (EMIssion SENSitivity model) model was used to calculate emissions of PM10from traffic; the model combined two approaches of Bottom-up and Top-down This model is appropriate for developing countries as Vietnam with lack of information Traffic data were collected from survey for road type, vehicle shares, vehicle age, annual daily traffic, etc which was from pre-vious study for calculating traffic emission for HCMC conducted by the author[4]
2.2.2 Industrial source For industrial sources/point sources, PM10was estimated from
282 major sources (such as Iron, Steel, Cokes, Refinery, and Cement, among others) in boilers, chimneys, generators, etc in HCMC Ac-tivity data for each source were collected as fuel consumption, fuel
Fig 1 Research flow chart.
B.Q Ho / Sustainable Environment Research xxx (2017) 1e8 2
Trang 3type, treatment methods, etc The small industrial sources using
charcoal emit few emissions and were estimated in area source
Main information required is fuel consumption for each source and
also conducted in study from stationary emission sources of Ho Chi
Minh Environmental Protection Agency (HEPA)/HCMC DoNRE
(Department of Natural Resources and Environment)[5,6].Table 1
shows emission factor for PM10 by vehicle type from various
reference sources[7e9] PM10for industrial sources was estimated
(Eq.(1)) using the emission factors shown inTable 2
Gi¼X
Kj Nj
(1)
where Giis PM10emission (kg d1); Kiis emission factor of fuel
(kg kg1or kg m1); and Njfuel consumption of type j (kg d1or
m3d1)
2.2.3 Area sources
PM10generating from area sources are household activities and
small restaurants from fuel combustion (gases, LPG, charcoal, and
firewood) To estimate PM10emission, we used Eq.(2)and emission
factors fromTable 2
where E is emission of pollutant i (kgid1), ei emission rate of
pollutant i (kgid1person1), and P population (person)
In addition, according to WHO, fuel consumption rate in urban
area is 3.5 GJ person1yr1(with gas and charcoal) and in rural
11.7 GJ person1 yr1(with charcoal andfirewood) While PM10
emissions from gas is 3.7 g GJ1and from charcoal is 695 g GJ1and
population in District 5 is 176,890 persons The emission rate per
person for area source is described inTable 3
2.3 Using BenMAP to estimate mortality rate
Disease data were collected in District 5 since this is the area
which has the high population density and economic growth
Surveys were conducted over 15 wards of District 5 with total of
200 households (average number of persons in a house is 4.5),
collecting data for the averaged exposure time to PM10and number
of death related to respiratory system and strokes The number of
interviews corresponds with population distribution and
popula-tion density; Ward 1 has the highest populapopula-tion with 21,913 people,
density of 43,960 persons km2and the number of interviews is 25,
while ward 14 only has 1781 persons, being the lowest population
and accounting for only 3 interviews The equation used for
calculating the number of mortality related to air pollutant
con-centration is as follows:
Where:
Air pollution level of pollutant i (mg m3): is the value of PM10
concentration in study area higher than WHO guideline for
pollutant i; or the different concentration of different scenario In
this study, annual average concentration is for PM10and compare
with WHO guideline (20mg m3)
Baseline mortality for air pollutant i (mg m3)1is mortality
change when pollutant i increase or decrease in air quality The
mortality study was conducted by USEPA in North Carolina, USA, dealing with human from European, Africans and Asian in various ages For the present study, baseline mortality of 0.00075 was used when PM10increases 1mg m3comparing with WHO guideline, which means that annually there are 7.5 death in 10,000 people when annual average concentration of PM10increases by 1mg m3 Percentage of death in area is percentage of death related to res-piratory and cardio in total death in study area Number of exposed (person) is the population in study area
Only PM10 concentration was applied in Eq.(3) In the study domain, the concentration of PM10for annual average in District 5 ranged from 17 to 49mg m3 There are 12 wards of District 5 with
PM10concentration exceeding the WHO guidelines
3 Results and discussion 3.1 Result of emission inventory
Table 4demonstrates the sharing of emission sources, in which major source is traffic with proportion more than 83% Meanwhile,
EMISENS model also shows (i) generating mostly from hot emission stage with 98% total amount of traffic source, and (ii) HDV and motorcycle are two main contributors
3.2 Uncertain analysis for air emission inventory Uncertainty is one of the main issues in the development of an emission inventory In this study, there are many parameters which
Number of mortality¼ Air pollution level of pollutant img m3
Baseline mortality for air pollutant img m31
Table 1 Emission factor for PM 10 by vehicle type.
Heavy duty vehicle (HDV) 236* Light duty vehicle (LDV) 1.6
*Source: [7e9]
Table 2
PM 10 emission factors for in industrial sources.
PM 10 emission factor (g kg1) 0.00063* 0.00075* 0.00693* 0.0036* Specific weight (kg L 1 ) 0.87 0.96 e e
*Source: [10] , note: “DO is Diesel Oil, FO is Fuel Oil”
Table 3 Emission rate of PM 10 per person for area sources.
Fuel Consumption rate
(GJ person1yr1)
Emission factor
of PM 10 (g GJ1)
Emission rate of
PM 10 (kg person1yr1)
Source: [6]
Trang 4can be affected by the uncertainties of the results of the emission
inventory A discussion of those uncertainties follows:
Most of the emission factors for this inventory were derived
from EMEP/EEA Air Pollution Emission Inventory Guidebook
(2009), updated in 2013, while some emission factors were
ob-tained from other literature sources It should be noted the
avail-ability of local emission factors for Vietnam was very limited
The completeness of data from field surveys is not always
consistent as interviews may vary from source to source
Some-times the data obtained were from verbal communication with the
authorities or company officers There was a lack of official data
from local agencies, especially for industrial data Although the
emission inventory group put effort into including all potential
sources into the inventory, there might be some sources missing
and this could potentially underestimate the total emissions Air
emissions from burning pesticide's plastic bag were not included in
this inventory due to the lack of available information although
there are few plants Some investigations on these sources are
needed in the future
There was a lack of traffic information on the side streets, which
is one of the key features of road transport in the study domain Many potential particulate emission sources are omitted For example, uncontrolled biomass burning, unpaved roads and other natural sources
3.3 Spatial distribution of PM10
After calculating emission of PM10, MapInfo software was used
to make PM10 spatial distribution map with grid cell 1 km2 (1 1 km), domain with 34 30 cell for x and y direction PM10
emissions in each cell are the input data for TAPOM This domain is for PM10distribution in HCMC area and for air quality modeling, and then extract thefinal result of PM10concentration map For industrial sources, emissions of PM10are distributed depending on location of plants and factories InFig 2the green spots stand for plants and color in each cell show the amount emission of PM10 (g h1km2) Therefore, the greener spots the more emissions of
PM10 Result shows that PM10 emission range from 0 to 1.27 g h1km2 Distribution of PM10for area sources is showed in
Fig 3which is based on population density The result of PM10
emission ranged from 0 to 2.02 g h1km2 For traffic sources, emissions of PM10distribution are based on road length in each cell Therefore, the higher traffic density, the more emission of PM10.Fig 4demonstrates emission load of PM10
ranging from 0 to 1.84 g h1km2 Meteorology model FVM used data from National Center for Environment Prediction and US Geological Survey for boundary and initial condition (temperature, geopotential height, wind speed, etc.) as well as land data and
Table 4
Result of PM 10 emission inventory for HCMC and district 5.
Source PM 10 emission for
HCMC (ton yr1)
Percentage (%)
PM 10 emission for district 5 (ton yr1)
Percentage (%)
Household
activities
Fig 2 Location of industrial sources (green circle) and spatial distribution of PM 10 for industrial source (different color is emission load of PM 10 in g h1km2) in HCMC; the bold black line is the location map of HCMC.
B.Q Ho / Sustainable Environment Research xxx (2017) 1e8 4
Trang 5Fig 3 Spatial distribution of PM 10 for household activities source (different color is emission load of PM 10 in g h1km2) in HCMC; the bold black line is the location map of HCMC.
Fig 4 Road traffic network for spatial distribution of PM 10 for traffic source; the bold black line is the location map of HCMC.
Trang 6surface data For FVM, time period is from January to July 2012 And
the time for air pollution model TAPOM 01/07/2012 to 03/07/2012
since this is the most polluted period The calibration and validation
of FVM model were conducted Meteorological parameters of
temperature, wind components from FVM simulation models are
calibrated and validated compared with those from observation
stations in Nha Be period from 1/7/2012 to 3/7/2012 The data from
the simulation results of the model were compared with actual
data measured at stations with a high correlation coefficient for
daily value R2¼ 0.693 Simulation models very good day and night
temperatures in the research area
3.4 Meteorology and air pollution
Result of meteorology and air quality in domain area (HCMC) is
shown inFigs 5 and 6 The TAPOM model which is also calibrated
and validated was applied to simulated air quality over HCMC The
comparison between simulation and measurement at the Nha Be
station shows the model simulates quite well PM10concentrations
at research area with the correlation coefficient R2¼ 0.72 The Nha
Be station is the station where we measure the background of air
quality This station has less impact from human activities Low
monitoring value of the background makes higher error values
Nonetheless, the correlation coefficient of 0.72 is acceptable
This station is not located in the city center and should be less
affected by emissions from the operation of industrial sources,
transportation and residential area Then, air pollution map was
established for District 5 based on map of HCMC Because the District 5 has a small area Meteorology and air pollution modeling cannot be done directly The result shows that almost all wards have the PM10pollution exceeding WHO guidelines except for ward
13, 14 and 15
3.5 Mortality result With all prepared data, and based on BenMAP model theory, number of mortality is about 5 person yr1for District 5 for total pollution of 194,228 persons or mortality rate 0.0025% (Fig 7) Ward 10 has the highest mortality, followed by Ward 9 and 5 With 5 person die every year related to PM10, District 5 will lose
900 billion VND (45 million USD since life costs for 1 person is 9 million USD which is used by USEPA to estimate economic cost due
to air pollution)
The results of extrapolation show that the death rate related to
PM10 in HCMC (population of HCMC is 7,955,000 inhabitants) is about 204 persons yr1and economic losses of 1.84 billon of USD for HCMC
3.6 Measures to reduce PM10pollution and protect public health Together with provided decision on reducing air emission, the citizen needs to protect themselves from PM10by using specialized face mask which canfilter 80e90% of PM20and replace the polluted
Fig 5 Wind direction on surface at 4 a.m (upper left), 10 a.m (upper right), 12 a.m (lower left) and at 10 p.m (lower right) on 2/7/2012; the red line is the location map of HCMC.
B.Q Ho / Sustainable Environment Research xxx (2017) 1e8 6
Trang 7Fig 6 PM 10 concentration distribution for HCMC (mg m3); the red line is the location map of HCMC.
Fig 7 Mortality map for the study domain (in person); Phuong is “Ward in the map” (Phuong 13 is the ward 13 in the map).
Trang 8fuel use (charcoal and wood) as well as wood-burning stove by
cleaner fuel as gas or LPG
The main cause of air pollution in HCMC is traffic; especially
from motorcycles with more than 30% motorcycles can not meet
the emission requirement Meanwhile, the fuel use has the low
quality Therefore, the solutions for reducing PM10 include: (i)
regulating use duration for motorcycle to reduce the number and
circulation of old vehicle; (ii) providing a technical standard on gas
emission for motorcycle registry; (iii) implement registration and
inspection for motorcycle; (iv) improving fuel quality according to
Euro 4 standard in 2016 and Euro 5 in 2021; (v) improving the
automatic air quality monitoring system in HCMC to alert the
cit-izen when there is high air pollution level
4 Conclusions
Emissions of PM10for Ho Chi Minh City, Vietnam was calculated
in this research Result of PM10emissions in HCMC point out traffic
is the main source of PM10emissions and accounted for more than
83% For traffic source, the motorcycle is the main contributor and
occupied about 24% of traffic emission, heavy trucks are 23%, light
trucks accounted for 19% and remaining for cars and buses
Simu-lation results PM10dispersion in HCMC for maximum 1-h average is
250mg m3 The concentration in annual average in District 5 is
30 mg m3 District 5 has 12/15 wards which has the air quality
exceeding WHO guidelines (20mg m3), the highest annual average
concentration is 48.9 mg m3 in Ward 4 and the lowest is
16.8mg m3in Ward 13 The results from calculation shows that
number of deaths related to PM10in District 5 is 5 persons yr1,
occupied for 0.0025% in total population of 194,228 and cause
economic losses of more than 45 million of USD With the results of
extrapolation for HCMC, death rate related to PM10is 204 person's
yr1 in total population of 7,955,000, caused economic losses of 1.836 billion of USD
Acknowledgement Authors thank to Vietnam National University - Ho Chi Minh City (VNU-HCM) for providing the funding with grant number C2016_24_03
References [1] Carlos D Health imperative for urgent action on air pollution In: Integrated conference of BAQ 2014 and intergovernmental 8th regional EST Forum in Asia; 2014 Nov 19-21 Colombo, Sri Lanka
[2] HEI Effects of short-term exposure to air pollution on hospital admissions of young children for acute lower respiratory infections in Ho Chi Minh City, Vietnam Boston, MA: Health Effects Institute; 2012
[3] WEF Air pollution threatens national health Davos, Switzerland: World Economic Forum; 2012
[4] Ho BQ, Clappier A, Francois G Air pollution forecast for Ho Chi Minh City, Vietnam in 2015 and 2020 Air Qual Atmos Health 2011;4:145e58 [5] HEPA Report 2006 air quality in Ho Chi Minh City Ho Chi Minh city, Vietnam:
Ho Chi Minh Environmental Protection Agency; 2006 [6] HEPA Report 2010 air quality in Ho Chi Minh City Ho Chi Minh city, Vietnam:
Ho Chi Minh Environmental Protection Agency; 2010 [7] Ho BQ Optimal methodology to generate road traffic emissions for air quality modeling: application to Ho Chi Minh City [Ph.D Dissertation] Lausanne (Switzerland): Federal Institute of Technology; 2010
[8] Belalcazar LC, Fuhrer O, Ho MD, Zarate E, Clappier A Estimation of road traffic emission factors from a long term tracer study Atmos Environ 2011; 2009(43):5830e7
[9] Zarate E, Belalcazar LC, Clappier A, Manzi V, Van den Bergh H Air quality modelling over Bogota, Colombia: combined techniques to estimate and evaluate emission inventories Atmos Environ 2007;41:6302e18
[10] EMEP/EEA The EMEP/EEA air pollutant emission inventory guidebook Copenhagen Denmark: European Environment Agency; 2009
B.Q Ho / Sustainable Environment Research xxx (2017) 1e8 8