вход по аккаунту


Application of air dispersion modelling for exposure assessment from particulate matter pollution in mega city Delhi.

код для вставкиСкачать
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
Published online 9 July 2010 in Wiley Online Library
( DOI:10.1002/apj.468
Special Theme Research Article
Application of air dispersion modelling for exposure
assessment from particulate matter pollution in Mega City
Manju Mohan,1 * Shweta Bhati1 and Archana Rao2
Centre for Atmospheric Sciences, Indian Institute of Technology, Delhi, India
Tata Consultancy Services, Mumbai, India
Received 8 June 2009; Revised 3 May 2010; Accepted 4 May 2010
ABSTRACT: The severe health hazards of respirable suspended particulate matter (RSPM or PM10 ) are well known,
and control of RSPM is an integral part of any effective air quality management plan. The proposed study addresses the
implementation of an air quality model for exposure assessment subsequent to its validation for the particulate matter
in a megacity in India, namely Delhi, where RSPM concentrations most often exceed the ambient air quality standards.
The model validation has been accomplished here for the total suspended particulate matter (TSPM) for which an
extensive source inventory is available along with the requisite air quality data. A regulatory air dispersion model of
US Environmental Protection Agency, AERMOD (Aermic Dispersion Model version 07026), which is a steady-state
Gaussian Plume Model with improved atmospheric boundary layer physics, has been utilized for this purpose. Model
validation shows a satisfactory model performance based on various statistical indicators. Subsequently, RSPM values
are obtained from TSPM estimates and applied for exposure assessment for predefined emission control scenarios. The
results from the model have been used to analyse mortality change associated with two hypothetical scenarios, namely,
(1) reduction of vehicular traffic emissions by a fixed amount due to introduction of metro rail and compressed natural
gas based buses in the city and (2) shift to a cleaner fuel in all the thermal power plants. The results from the exposure
assessment study showed the importance of control of vehicular emission in improving the air quality and accrued
health benefits.  2010 Curtin University of Technology and John Wiley & Sons, Ltd.
KEYWORDS: air pollution modelling; model evaluation; particulate matter; regulatory models; exposure assessment;
health risk analysis
The rapid growth of Delhi in terms of population and
economic activity makes it vulnerable to environmental pollution problems. Delhi has been designated as
an air pollution control area by Ministry of Environment and Forests in recognition of the severity of
air pollution due to vehicular, industrial and domestic
sources.[1] Particulate matter is one of the key constituents of the pollutants in ambient air of Delhi. In year
2004, the annual average concentration of respirable
suspended particulate matter (PM10 ) was observed to
be 138 µg/m3[2] against the national ambient air quality
standard of 60 µg/m[3] . In the year 2009, the ambient
levels have increased to 165.2 µg/m3[3] and concentrations below standards have still not been achieved. The
*Correspondence to: Manju Mohan, Centre for Atmospheric Sciences, Indian Institute of Technology, Hauz Khas, New Delhi110016, India. E-mail:
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Curtin University is a trademark of Curtin University of Technology
particulate matter pollution mainly arises due to emissions from vehicles, coal-based thermal power plants,
biomass incineration, re-suspension of traffic dust, and
commercial and domestic use of fuels. Air quality management is a growing concern in Delhi. Increasing use
of vehicles increases the emission of air pollutants and
in turn degrades the air quality of the city. Studies
have indicated relationships between particulate matter
and respiratory and cardiovascular morbidity as well as
mortality,[4,5] especially PM10 [6,7] and PM2.5 [8,9] .
Air quality models and exposure assessment studies provide a tool to understand the implications of
pollutant emissions and aid in deciding control and
management strategies. Thus, there arises a need to
evaluate the available regulatory models as suited to the
local site conditions. Exposure assessment studies are
usually based on pollutant concentration data obtained
from ambient air monitors. However, as Bell[10] pointed
out, studies based on the monitoring networks face certain limitations. Measurements may be taken in an area
which may not be representative of either individual
or community-level exposure. Also, ambient monitors
may not provide full temporal and spatial coverage for
a range of pollutants. The present study, thus, evaluates
the performance of USEPA regulatory model AERMOD
(version 07 026) for modelling concentrations of total
suspended particulate matter (TSPM) in Delhi for the
year 2004. The concentration estimates from the model
have been applied to dose response equation used in an
earlier study by Ostro[11] for assessing health effects in
terms of change in mortality.
The capital city of Delhi is located at latitude
28◦ 38 17 and longitude 77◦ 15 51 with an altitude of
215 m above sea level. Delhi has three distinct seasons namely summer, monsoon and winter. The maximum temperature of Delhi ranges from 41 45 ◦ C in
peak summer season and the minimum temperature in
winter season is in the range of 3–6 ◦ C in the coldest period of December–January. The summer season
(March, April and May) is governed by high temperature and hot, high-speed winds. Another important feature of this season is prevalence of dust-storms which
suddenly increase ambient particulate matter concentrations. However, such scenarios can only be captured
appropriately by detailed numerical atmospheric chemical models where concentrations are estimated in a
time-dependent manner from seconds to minutes. However, most regulatory models are steady-state models
where detailed chemical modelling is absent or parameterized in a simple manner. The monsoon (July, August
and September) is dominated by rains and since rain
acts as a cleaning device for the environment, incidences of rain result in lower levels of pollutants in air.
The most important season in Delhi, from air quality
point of view, is the winter, which starts in November
and ends at the end of February. This period is dominated by cold, dry air and ground-based inversion with
low wind conditions, which occur very frequently and
increase the concentrations of pollutants. In the period
of years 2008–2009, the average ambient PM10 levels were in the range of 227–247 µg/m3 in the winter
months of November–February. The lowest monthly
average (192 µg/m3 ) was observed in the month of
July 2008 which shows consistent exceedence of standards of particulate levels which are 100 µg/m3 for 24-h
average and 60 µg/m3 for annual average.[12] The winter season is also characterized by sudden increase in
number of asthma, emphysema and chronic obstructive
pulmonary disease cases.[13,14] Especially in the case
of Indian children in Delhi, prevalence of respiratoryassociated symptoms is high in winter season (36%)
compared to average of other seasons (33%).[15]
The Central Pollution Control Board (CPCB), the
nodal agency responsible for monitoring and regulating
the pollution scenario, measures the particulate matter
concentration at seven monitoring stations in Delhi,
namely, Ashok Vihar, Siri Fort, Nizamuddin, Shahzada
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pacific Journal of Chemical Engineering
Figure 1. The study area of Delhi and location of seven
monitoring stations of CPCB. This figure is available in
colour online at
Baug, Janak Puri, Shahadara and ITO (Fig. 1). Shahdara
is an industrial area and ITO is one of the busiest traffic
intersections of Delhi. All other monitoring sites are
located in residential areas.
Applied model
In the present study, AERMOD (version 07 026)[16] has
been applied for simulating the TSPM concentrations.
AERMOD is a steady-state plume model developed
by the US Environmental Protection Agency (USEPA)
and incorporates planetary boundary layer concepts. It
includes a wide range of options for modelling air quality impacts of pollution sources. AERMOD is used for
the simulation of pollutant concentration from point
sources, line source and area sources. It is one of the
models recommended by USEPA to be used in policy
implementation programs. AERMOD is an improvement over the former recommended model ISCST3
(Industrial Source Complex Short Term Model, Version 3) by USEPA in which only surface meteorological
data are utilized. AERMOD can adapt multiple levels of
data to various stack and plume heights and can create
profiles of wind, temperature and turbulence, using all
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
available measurement levels. Use of turbulence-based
plume growth with height dependence rather than that
based upon stability class provides AERMOD with a
substantial advancement over the ISCST3 treatment.[17]
In India, both ISCST3 and AERMOD are amongst the
recommended models for regulatory purposes.[1] However, ISCST is most often used perhaps due to unavailability of requisite data for AERMOD or other improved
models at all sites. As the measurements improve in
the country, it is expected that AERMOD can find a
favourable place amongst the modelers and regulators.
Hence, the application of AERMOD to Indian conditions can serve a useful purpose for all concerned.
Kesarkar et al .[18] applied coupled WRF–AERMOD
model for estimation of PM10 in the city of Pune in
Maharashtra state of India, for a period of seven days.
However, there are still not enough studies regarding
the application of AERMOD for Indian cities for a long
time span as per the regulatory requirements.
AERMOD models a system with two separate
components: AERMOD and AERMET (AERMOD
Meteorological Preprocessor). The AERMET is the
meteorological processor for the AERMOD. Input data
for AERMET includes hourly cloud cover observations,
surface meteorological observations, and twice-a-day
upper air soundings. Output includes surface meteorological observations and parameters and vertical profiles
of several atmospheric parameters. All the variables and
input parameters required as well as the output variables
from an AERMOD run can be written using the following options: dispersion options, source options, receptor
options, meteorological options and output options.
Requisite hourly surface meteorological data for
Delhi for the year 2004 was obtained from Indian
Meteorological Department. The upper air data was
accessed from online global Radiosonde Database of
National Climatic Data Center (NCDC) of National
Oceanic and Atmospheric Administration (US-NOAA).
The model was run for two types of receptor options:
(1) over the entire grid network of Delhi and (2) for
discrete specified points, i.e. for the location of monitoring stations so that comparisons between estimated
and observed concentration could be made.
The output was generated in the form of 24-h average,
monthly average and annual average TSPM concentrations. The model was run to get the concentrations
from each individual source (i.e. transport, power plants
and others) as well as all sources combined for annual
Source emissions
Particulate matter emissions mainly arise from three
sources: transportation sector, power plants, and other
sources which include domestic and waste sector. Emissions from power plants have been considered under
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Table 1. Type of motor vehicles in Delhi as % Share
for Year 2004.
Type of vehicle
Number and % of total
Motor cycles and scooters
Cars and jeeps
Goods vehicles
Auto rickshaws
2 659 449
1 243 759
116 719
73 438
24 073
16 233
4 765
1 857
20 535
point source options, while emissions from transportation and other sectors have been modelled under area
In the present study, a grid network comprising of
2 km × 2 km cells was constructed for 26 × 30 km2
area of Delhi. The selected area covers that part of
Delhi where most of the urban activities take place
and includes all major sources of air pollution, sizeable
receptor population and the seven monitoring stations
of CPCB. Keeping these criteria in consideration, emissions from 173 cells of the grid network were input
into the model. Figure 1 displays the grid network over
the city of Delhi which was used for the study. The
shaded cells are the ones for which the emissions were
considered. The unshaded areas were poorly developed and with negligible emissions. Particulate matter
emissions for the year 2004 from the sectors such as
transportation,[19] power plants[20] and other, i.e. domestic and waste, etc.,[21] sources were calculated for each
2 km × 2 km cell of the grid. The methodology to
obtain total emission estimates was based on an earlier
work by Gurjar et al .[22]
The calculation of emission from vehicles is based
on the data on emission factor for the specific vehicle
type, the distance travelled by a particular vehicle type,
number of vehicles and their distribution in the type of
the fuel used. The emission from vehicles is calculated
using the following:
(Vehj × Dj ) × Ei ;j ;km
Ei =
where Ei is the emission of compound (i ), Vehj the
number of vehicles per type (j ), Dj the distance travelled
in a year per different vehicle type (j ) and Ei ;j ;km the
emission of compound (i ), vehicle type (j ) per driven
Table 1 shows the profile of vehicular population in
the city for the year 2004.[2] The emission factors of different pollutants for each vehicle type have been calculated in earlier studies conducted by organizations such
as CPCB[23] and Central Road Research Institute.[24]
The annual growth of petrol cars and diesel cars is 8.5%
and 16.6%, respectively.[25,26]
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
Three coal-based power plants (Badarpur,
Indraprastha and Rajghat) have been considered for
estimating emissions from power plants. The coal consumption in 000 tons for these power plants was
obtained from Central Electricity Authority performance review report.[27] The emission factors for different types of coal that have been used for calculations
are based on Gurjar et al .[22] The total emissions from
power plants are thus calculated as
Total Emission = coal consumption (in 000 ton)
× emission factor for that power plant
The domestic and waste sectors include emissions from
fuel consumption in households and emissions from
waste burning. While cooking gas is the major domestic
fuel, kerosene oil is also usually burnt in small stoves;
other energy sources for domestic sector that have been
considered are biomass such as fuel wood, crop waste
and dung. Emissions from domestic sector have been
calculated as
(Fuelj × EFij )
Ei =
Model evaluation
Data for the observed ambient air concentrations of
TSPM was collected for the year 2004 in terms of
average daily concentrations from the seven monitoring stations of CPCB in Delhi. The monitored daily
averaged air quality data of seven stations in the city
was used for comparison with the model results, and
the analysis of the monthly data was done. Statistical
performance measures were used to evaluate the performance of the model. Measures such as the fractional
bias, the normalized mean square error, the geometric
mean, the geometric variance, the correlation coefficient and the fraction of predictions within a factor of
two of observations are suggested in many validation
studies.[31 – 35]
In the present study, the evaluation of performance
of the model was based on statistical measures such as
scatter plots, quantile–quantile plots, mean square error,
fractional bias, index of agreement and geometric mean
and variance.[33,34] .
Exposure assessment
where Ei is the emission per compound (i ), Fuelj the
consumption of fuel per fuel type (j ) and EFi ,j the
emissions of compound (i ) per unit of energy (j ).
Fuel consumption data given in Delhi Statistical
Handbook[21] and emission factors used in Gurjar
et al .[22] have been used in calculations. The solid waste
generated in Delhi is 0.5 kg/capita/day.[28] Approximately 80% of the garbage generated in Delhi is collected with the remainder left on the streets or in small
dumps.[22] Particulate emissions from waste sector are
mainly produced from open burning of the remaining
waste and are considered in this study based on emission
factors from Gurjar et al .[22]
The emissions from small-scale industries have not
been taken into account for this study because, according to a Supreme Court decision in 1996, polluting
industries in Delhi were closed in 2000 and others
that were not hazardous were relocated, and there is
absence of factual information about emissions of relocated industries. Moreover, Gurjar et al .[22] estimated
the contribution of small-scale industries towards total
particulate matter emissions to be of negligible order
(< ∼ 1%) and thus, in this study, they have not been
accounted for.
Methodology for the preparation of the gridded
emission inventory is based on an earlier work by
Mohan et al .[29] The emission from transport is divided
into 173 grid networks according to the fraction of road
area in each grid cell. The total emission from domestic
and waste has been divided according to the ratio of cell
population towards total population of 15 275 000[30] in
the study area in Delhi.
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
As discussed initially, the health effects of PM10 pollution range from minor symptoms like irritation of the
airways, coughing, or difficulty in breathing to major
respiratory symptoms like decreased lung function,
aggravated asthma, development of chronic bronchitis,
irregular heartbeat, nonfatal heart attacks and premature deaths.[36] There is a distribution of susceptibility
to the effects of air pollution in any population. People who are in a weakened physical state or who have
a history of chronic obstructive pulmonary disease or
cardio-pulmonary problems are thought to be most vulnerable and most likely to die in case of rise in pollution.
Since death is the most clearly defined health end-point,
mortality cases are more extensively analysed in exposure assessment studies.[37] In this study, dose–response
equation used by Ostro[11] has been applied to estimate change in mortality cases in different scenarios.
The equation can be applied for outdoor ambient PM10
concentrations. The present study, however, considers only total suspended particulate emissions as the
inventory was prepared for TSPM rather than PM10 .
Shandilya et al .[38] collected baseline concentrations of
respirable suspended particulates (PM10 ), non-respirable
suspended particulates and PM2.5 for an urban-industrial
site in Delhi and reported the PM10 /TSPM ratio as
varying from 0.3 to 0.5. Measurements taken at three
different sites in Delhi in another study[39] found out
the average ratio of PM10 to TSPM as 0.53. According
to the monitoring data of seven air quality stations of
CPCB in Delhi,[3] the PM10 /TSPM ratio ranges from
0.42 to 0.61 for year 2001–2004. In the present study,
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
for deriving PM10 concentrations from TSPM concentrations estimated by the model, the ratio of 0.53 for
PM10 /TSPM has been used for exposure assessment in
the study. The ratio, which was observed in the earlier
study,[39] is also close to the average observed in the
studies that were surveyed for the present study.
With these considerations and using the available
data, the mortality calculations were based on the
following equation used by Ostro[11] to calculate change
in mortality due to incremental changes in ambient
PM10 concentrations:
Change in mortality = Cr × (PM10 )
× Population × CMortality
where Cr is concentration–response coefficient and
CMortality is the crude mortality rate. Ostro[11] computed
an average value of Cr based on the mean effect of
a 10 µg/m3 change in PM10 implied by various earlier studies in terms of mortality change, and a mean
of 0.096 was obtained. In the present study, the value
of Cr was calculated from Eqn (3) by substituting all
other parameters (i.e. change in respiratory disease
related mortality, PM10 , population and CMortality ) for
Delhi for a time period of 10 years (1991–2000) for
which these parameters were known. The concentration–response coefficient was calculated for each of
the 10 years using Eqn (4) and the average value was
obtained as 0.001231. Thus, the average Cr value of
0.001231 as derived here has been used in this study
for assessing mortality impact. The number of respiratory deaths every year for 10 years, i.e. 1991–2000
was taken from ‘The Annual Report on Registration of
Births and Deaths’, 2003,[40] Govt of National Capital
Territory of Delhi, and the population data was collected from Delhi Statistical Handbook, 2006.[21] The
crude mortality rate for Delhi has been taken as 0.0041
as projected in this Handbook. Equation (4) was applied
to each cell of the grid network of the study area. Hence,
the equation applicable to each cell is
Di = Cr × (PM10 )i × Pi × CMortality
Overall performance of AERMOD
Scatter plot of predicted vs observed concentrations is
shown in Fig. 2a,b along with the limits of factor of 2.
Most results from AERMOD agreed with the measured
concentration statistics to within a factor of two for daily
average concentrations (Fig. 2a). However, a scatter
plot for monthly averages reveals the tendency of the
model to underpredict as can be seen in Fig. 2b.
The model performance was evaluated by measuring
some statistical parameters like fractional bias (FB),
normalized root mean square error (NMSE), correlation
coefficient (r), index of agreement (d), geometric mean
(MG), geometric variance (VG), root mean square
where Di is the reduction (or increase) in number of
death cases associated with cell i of the grid network
where population Pi is exposed to a decrease (or
increase) of (PM10 )i in ambient concentration level.
Thus, total change in mortality over the entire study
area (TMortality ) has been calculated as
TMortality =
i =1
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Figure 2. (a) Scatter plot of observed and estimated daily
average total suspended particulate matter concentrations.
(b) Scatter plot of observed and estimated monthly average
TSPM concentrations.
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
Table 2. Performance of statistical indicators for concentration estimations by AERMOD for year 2004.
(ideal value)
Siri Fort
Correlation coefficient (1)
Index of agreement (1)
Fractional bias (0)
FAC2 (1)
NMSE (0)
Geometric mean bias (1)
Geometric variance (1)
RMSE (0)
error (RMSE) and fraction of data for which model
predictions lie within a factor of two of the observations
(FAC2). The results for the performance of statistical
indicators can be seen in Table 2. Summarily, FB varies
from 0.19 to 0.33, NMSE varies from 0.06 to 0.29, MG
varies from 1.19 to 1.34, r ranges from 0.54 to 0.91, d
varies from 0.64 to 0.95, VG from 1.0 to 1.2 and RMSE
varies from 70.4 to 245.6.
Values of NMSE and FB satisfy the model reliability criteria which have been used in some earlier
studies.[41,42] RMSE values, which account for the difference in each pair of values in modelled and observed
concentrations, are high. However, low values of NMSE
indicate that the overall deviations are less. Values for
correlation coefficient and index of agreement as shown
above indicate that the predicted values follow the trend
of the observed values.
Considering the overall performance of all statistical
measures, it can be seen that concentration estimations
of AERMOD were best for the site of residential area
Comparatively higher values of fractional bias,
NMSE and geometric variance are observed for the site
of ITO. ITO is one of the major traffic intersections
of Delhi and transport vehicles are a highly variable
source of emission. Models such as CALINE which
model vehicular emissions on a highway or motorway
as linear sources are better suited to estimate concentrations at such traffic intersections. However, this study
is aimed at analysing particulate matter pollution due to
all possible sources at a location, and in such scenarios
emission inventory is best presented in the form of area
sources in a grid network. Nonetheless, concentration
estimates of AERMOD for ITO can still be rated as
satisfactory considering performance of all considered
statistical measures taken together.
Greater prevalence of positive fractional bias values
for all the sites indicates that the model has a tendency
towards underprediction as compared to the observed
values. Overprediction or underprediction of the model
is explained further in the plots of VG vs MG.[34] If the
points are on the right side of the Y -axis it shows that
the model results are underpredicted, and if the values
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Figure 3. Plot of geometric variance (VG) vs geometric mean
bias (MG) for TSPM concentrations estimated by AERMOD.
are on the left side then it is overpredicted. From the
plot shown (Fig. 3), it can be inferred that the model
mostly underpredicts. Quantile–quantile (Q–Q) plots
also explain the model behaviour in terms of overprediction or underprediction. In Q–Q plots performance
of model is evaluated by comparing the estimated distribution of concentrations with the observed distribution
through scatter plots between observations ranked from
high to low and similarly ranked model estimates. Such
plots are referred to as Q–Q plots because each point
corresponds to a specific quantile of the data set.[43]
Figure 4 shows a Q–Q plot for 24-h average particulate matter observed and estimated concentrations. It
can be seen from the Q–Q plots that the extent of
underprediction by the model increases as the observed
concentration increases.
Figure 5 displays the spatial distribution of annual
average concentration of TSPM for the year 2004 in
the grid network. The location with the highest TSPM
value, ITO, has been marked on the map. As stated
earlier, ITO is a busy traffic intersection of Delhi. Also,
ITO is in close proximity to two coal-based thermal
power stations of Delhi, thereby leading to higher
ambient levels of TSPM.
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
Figure 4. Q–Q plot of modelled (by AERMOD) and observed
total suspended particulate matter concentrations.
Dispersion models can substantially underpredict
impacts because the theoretical and physical assumptions on which they are based do not match a particular
environmental or meteorological situation. Moreover,
the meteorological (weather) inputs provided may be
too few or too limited to allow the model to function
properly.[44] Often emission inventory that comprises
of large number of variety of sources finds it difficult to suitably account for all of these sources that
may affect the model bias. The TSPM concentration in
Delhi gets affected at times by long-range transport of
dust by dust-storms due to its close proximity to desert
land in the nearby state of Rajasthan. Such random and
irregular phenomenon is not being accounted for in the
emission inventory presently. Also, as stated in the section on Annual Average Concentration Due to Each
Emission Category, polluting industries in Delhi were
relocated in accordance with the Supreme Court ruling. However, certain small factories are still expected
to be operational within city boundary limits. Another
important reason is that the activities under sectors
such as transport and waste, which lead to particulate
Figure 5. Concentration isopleths of annual average TSPM concentration (year
2004) as estimated by AERMOD. This figure is available in colour online at
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
Table 3. Annual average (2004) concentration estimates by AERMOD for each emission category and all sources.
Source sector
Power plants
Domestic and waste
All sources
Percentage contribution
towards total emission
(µg/m3 )
Percentage contribution
towards concentration
matter emissions, are under the purview of regulatory
authorities and hence a close estimate of total emissions from these sectors can be obtained. However,
activities under domestic sector (such as domestic fuel
usage) cannot be surveyed in entirety as a large section of low-income-group people live in unauthorized
slums and colonies in Delhi which are not under legal
purview. The emissions from these sections are significant but their quantitative estimation is based on many
assumptions.[22,37] The monitored ambient data, however, would measure concentration due to all sources
and thus observed concentrations are usually higher than
those estimated by the models.
In certain cases, the model results exceeded the monitored values; this could be due to some disturbances in
the local activities. The emission data which serves as
an input to the models has been derived from suitable
averaging of the annual emission data. Hence, the emission data for each grid is taken to be constant throughout
the year. But this is not possible in the real scenario.
Annual average concentration due to each
emission category
Table 3 shows the annual average TSPM concentrations
calculated by averaging concentration output of each
grid cell of the entire study area of Delhi, obtained
from the model estimates when the model was run
for each source category separately and then for all
sources. It can be seen that the transportation sector has
the greatest contribution (∼66.4%) towards total particulate matter concentration followed by domestic and
waste (∼30.8%) and power plants (∼2.7%). Srivastava
et al .[45] carried out measurements at various sites for
studying source apportionment of suspended particulate
matters in Delhi. The range of contribution of vehicular
pollution was from 24% to 69% while crustal dust was
the other dominant factor. Industrial sources, including
power plants, accounted for 1–7%.
Ground level concentration due to power plants is
usually low as they are elevated sources. The small
contribution of power plants towards total ambient
particulate matter contribution can also be attributed
to implementation of better environmental management
policies. However, in vicinity of power plants the
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
maximum daily average concentration was observed
as 547 µg/m3 . The transport sector has the greatest
contribution owing enormous increase in the number
of diesel- and petrol-based vehicles.
Exposure assessment
In this study, exposure assessment has been conducted
from the view point of change in mortality associated
with change in particulate matter concentration. Considering the availability of the required data, particulate
matter–mortality relationship equation[11] has been used
with the modified Cr value as discussed in the section
on Exposure Assessment to estimate change in mortality
in two scenarios:
Case 1: Change of production of power from coalbased sources to natural gas.
Replacement of fuel from coal to natural gas leads
to significant reduction in TSPM which would subsequently reduce the PM10 as well. This case study
demonstrates the effect of this control measure on
overall reduction in the expected mortality. Total fuel
consumption by the three coal-based power plants
in Delhi ranges from 4777 to 5923 kt to generate
about 6253 GWh of electric power. The fourth power
plant (Pragati Gas Turbine Station) has a consumption of 201–327 million m3 of natural gas to produce
630–1053 GWh of energy. Natural gas is a clean fuel
and thus particulate matter emissions from gas-based
power plant are greatly reduced. Assuming that total
electric energy is generated using natural gas as a fuel,
the decrease in particulate matter emissions is approximately estimated out to be 94% based on an earlier
study by Gurjar et al .[22]
Considering 94% reduction in emissions from the
power sector, total particulate matter concentrations
were estimated for each cell of the study area. Change
in mortality was then estimated using Eqn (6).
Equation (6) yielded a change of 482 deaths per year
in this scenario. Thus, a decrease of 482 deaths per year
can be expected if there is a complete shift from coalbased power production to gas-based power production.
Case 2: Twenty percent decrease in emissions from
transportation sector.
Increased use of public transport like CNG (compressed natural gas) buses and Metro rail will result in
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
a decrease in emissions from sources like motor cycles
and petrol- and diesel-based cars. The environmental
impact assessment report[46] for Delhi Metro projects
a decrease of 188.54 t/year of particulates against the
present load of 1503.1 t/year which amounts to approximately 13% reduction in particulate pollution. Goyal
et al .[47] reported an improvement in air quality after
CNG fuel implementation with a 14.3% decrease in
TSPM levels. However, estimation of actual change in
emissions due to a shift towards combined use of various available means of public transport was outside the
scope of this study. Thus, a decrease of 20% in emissions from transport sector due to combined effect of
development of various public transport means has been
taken up as a case. As in case (1), total particulate matter concentrations were estimated for each cell of the
study area grid considering 20% reduction in emissions
from transport sector, and subsequently the PM10 due
to this reduction was calculated.
The decrease in the number of mortality cases was
estimated to be 2527 using Eqn (5). In other words, if
an efficient and less polluting public transport system
leads to a decrease of 20% in ambient particulate matter
levels, we can expect a reduction of about 2527 deaths
per year in the city.
These calculations are based on the averaged data
and many assumptions; hence, the results might not
represent the exact figures in reality. These calculations
were done on the basis of short-term immediate effects
of the pollutant and taking the whole population into
account. Nevertheless, it is clear from the results
of cases 1 and 2 that it is the emissions from the
transport sector in Delhi which need consideration for
reduction in terms of particulate matter pollution from
the viewpoint of public health. Reductions of about 94%
emissions from power plants due to switch over to a
cleaner fuel leads to a decrease of 482 deaths per year,
whereas 20% reduction in transport emissions leads to
a decrease of about 2527 deaths, i.e. about five times
greater decrease in mortality as compared to Case 1.
The results highlight the impact of control measures
of important emission sectors towards total particulate
matter concentration and consequent health benefits in
this city.
There is always an uncertainty associated with such
dose–response relationships. The ratio of PM10 /TSPM
keeps on varying and its estimation is also based on
many assumptions. We can rely on model results only
if we are confirmed about accuracy of our emission
input. Validation of model is possible with monitored
data of the past years but predictions for future should
be considered with caution. Mortality is a complex
phenomenon which cannot be attributed to a handful of
parameters. Change in mortality in different scenarios
taken above assumes that while one scenario changes,
all other aspects of urban activities leading to particulate
matter pollution remain constant. Possibility of such
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
an assumption to be true in actual conditions is again
very remote. However, in the present study, we have
made an attempt to integrate air quality modelling
with health risk analysis to assess their application for
formulation of air quality management strategies, and
this first attempt has given a reasonable estimate of
• AERMOD 07 026 has been used to model TSPM
concentrations at seven locations in Delhi for the
year 2004, and the estimated concentrations after
validation of the model have been used for assessing
mortality change in different scenarios.
• Though the model has a tendency towards underprediction of concentration, values estimated by the
model agreed quite well with the observed concentrations considering the performance of several statistical measures.
• Agreement of estimated particulate matter concentrations with observed concentrations is better within
lower ranges of observed concentrations. At higher
monitored concentrations, the degree of underprediction of concentration estimates by the model
• The reasons for overall underprediction can be
attributed to the lack of consideration of dust transported from the neighbouring states and also lack of
precise data for all the small-scale industries.
• Vehicular emissions have been found to be the greatest contributor towards ambient particulate matter
pollution followed by others (i.e. domestic and waste)
and the power plants.
• A hypothetical case-specific mortality assessment
revealed that a small decrease in vehicular emissions
leads to five times greater reduction in mortality
count as compared to a major shift from coal to
natural gas sources in power production sector.
Hence, a need of introduction of efficient public
transport system is emphasized upon.
• Overall, the performance of AERMOD for estimation
of particulate matter concentration was rated as
satisfactory for a sub-tropical environment such as
megacity Delhi.
This research was partially supported by Ministry
of Human Resource Development (MHRD), Govt. of
India, as a research grant. We thankfully acknowledge
India Meteorological Department (IMD), Delhi, for providing the meteorological data. We also thank Central
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
DOI: 10.1002/apj
Pollution Control Board, Delhi, for making available
the air quality data. We also wish to thank the anonymous reviewers for their critical review that has helped
in improving the quality of the manuscript.
[1] Ministry of Environment & Forests (MoEF). White Paper on
Pollution in Delhi with an Action Plan, Government of India,
[2] Economic Survey of Delhi 2005-2006 (ESD), Govt. of NCT of
Delhi, 2006;
[3] Central Pollution Control Board. Environmental Data Bank,
[4] A.G. Russel, B. Brunekreef. Environ. Sci. Technol., 2009; 43,
[5] K. Sawyer, S. Mundandhara, A.J. Ghio, M.C. Madden.
J. Toxicol. Environ. Health, 2010; 73, 41–57.
[6] E. Samoli, R. Peng, T. Ramsay, M. Pipikou, G. Touloumi,
F. Dominici. Environ. Health Perspect., 2008; 116,
[7] G. Jonathan, N. Kulkarni, N. Pierse, L. Rushton, C.
O’Callaghan, A. Rutman. The Health Effects Institute, 2008;
Research Report 134.
[8] L.C. Chen. HEI Annual Meeting 2005;, Accessed: 9 april
[9] J.M. Cavallari, E.A. Eisen, S.C. Fang, J. Schwartz, R. Hauser,
R.F. Herrick, D.C. Christiani. Environ. Health, 2008; 7, 36,
[10] M.L. Bell. Environ. Int., 2006; 32, 586–593.
[11] B. Ostro. World Bank Policy Research Working Paper No.
1301, 1994.
[12] Delhi Pollution Control Committee. Monthly Ambient Air
Monitoring, 2009;,
Accessed April 10, 2010.
[13] R. Agarwal, G. Jayaraman, S. Anand, P. Marimuthu. Environ.
Monit. Assess., 2006; 114, 489–504.
[14] S.K. Chhabra, C.K. Gupta, P. Chhabra, S. Rajpal. Ann.
Allergy Asthma Immunol., 1999; 83, 385–390.
[15] World Health Organisation-India. Delhi Children Final Report, Epidemio
logy air poll ruralchild 4.pdf. Accesed April 8, 2010.
[16] USEPA Support Center for Regulatory Atmospheric Modeling. Preferred/Recommended Models,
scram001/dispersion prefrec.htm#aermod. Accessed March
29, 2010.
[17] United States Environmental Protection Agency. AERMOD:
Latest Features and Evaluation Results, 2003; http://www. mep.pdf. Accessed
March 28, 2010.
[18] A. Kesarkar, M. Dalvi, A. Kaginalkar, A. Ojha. Atmos.
Environ., 2007; 41, 1976–1988.
[19] Central Pollution Control Board (CPCB). Vehicular air
pollution in Delhi-A preliminary study, Government of India,
[20] Central Electricity Authority (CEA), Government of India. All
India Electricity Statistics: General Review, 2006.
[21] Delhi Statistical Handbook, Directorate of Economics &
Statistics (DES), Govt. of NCT of Delhi, Delhi, 2006.
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pacific Journal of Chemical Engineering
[22] B.R. Gurjar, J.A. van Aardenne, J. Lelieveld, M. Mohan.
Atmos. Environ., 2004; 38, 5663–5681.
[23] Central Pollution Control Board. Transport Fuel Quality for
Year 2005; 2006.
[24] S.A. Jalihal, T.S. Reddy. J. Sci. Ind. Res., 2006; 65, 426–431.
[25] Delhi Statistical Handbook, Directorate of Economics &
Statistics (DES), Govt. of NCT of Delhi, Delhi, 2008.
[26] Economic Survey of Delhi 2006-2007 (ESD), Govt. of NCT
of Delhi, 2007.
[27] Performance review of thermal power plants; 2005; Performance Review/
index Thermal Performance Review.html.
[28] V. Tayal, R.P. Dhiya, T.R. Sreekrishnan. State of municipal
solid waste management in Delhi, the capital of India. Delhi
Urban Environment and infrastructure improvement project;
[29] M. Mohan, L. Dagar, B.R. Gurjar. Environ. Monit. Assess.,
2007; 130, 323–339.
[30] Directorate of Economics & Statistics, Govt. of NCT of
Delhi. Ward Wise Final Result of Census 2001; http://des. Accessed 20 March
[31] D.G. Fox. Bull. Am. Meteorol. Soc., 1984; 65, 27–36.
[32] S.R. Hanna. J. Air Pollution Control Assoc., 1988; 38,
[33] American Society for Testing and Materials. ASTM Standard
Guide for Statistical Evaluation of Atmospheric Dispersion
Model Performance, American Society for Testing and
Materials: Pennsylvania, USA, 2000.
[34] S.R. Hanna, J.C. Chang, D.G. Strimaitis. Atmos. Environ.,
1993; 27A, 2265–2285.
[35] M. Mohan, T.S. Panwar, M.P. Singh. Atmos. Environ., 1995;
29, 2075–2087.
[36] United States Environmental Protection Agency (USEPA)., Accessed
22 February 2008.
[37] M. Kandlikar, G. Ramachandran. Ann. Rev. Energy Environ.,
2000; 25, 629–684.
[38] K.K. Shandilya, M. Khare, A.B. Gupta. Environ. Monit.
Assess., 2007; 128, 431–445.
[39] M. Cropper, N.B. Simon, A. Alberini, P.K. Sharma. World
Bank Policy Research Working Paper No. 1860, 1997.
[40] Directorate of Economic and Social Statistics (DES), Govt of
National Capital Territory of Delhi. The Annual Report on
Registration of Births and Deaths, 2003.
[41] J.C. Chang, S.R. Hanna. Meteorol. Atmos. Phys., 2004; 87,
[42] A. Kumar, J. Luo, G. Bennett. Process Safety Progress, 1993;
12, 1–11.
[43] A. Venkatram, R. Brode, A. Cimorelli, R. Lee, R. Paine,
S. Perry, W. Peters, W.J. Jeffrey, R. Wilson. Atmos. Environ.,
2001; 35, 4211–4221.
[44] Alliance of Residents Concerning O’Hare (AReCO), Myths
and Facts of Pollution, 2002;
[45] A. Srivastava, S. Gupta, V.K. Jain. Atmos. Res., 2009; 92,
[46] Delhi Metro Environmental Impact Assessment Report. Positive Environmental Impacts.
corporates/ecofriendly/Chapter%205.pdf. Accessed April 15,
[47] P. Goyal, Sidhartha. Atmos. Environ., 2003; 37, 5423–5431.
Asia-Pac. J. Chem. Eng. 2011; 6: 85–94
DOI: 10.1002/apj
Без категории
Размер файла
414 Кб
air, particular, delhis, matter, application, exposure, mega, modellina, city, assessment, dispersion, pollution
Пожаловаться на содержимое документа