Urban Air Pollution Analysis

Simple Interactive Models for Better Air Quality
Urban Air Pollution Analysis

in India
- Pune (Maharashtra)
- Chennai (Tamil Nadu)
- Indore (Madhya Pradesh)
- Ahmedabad (Gujarat)
- Surat (Gujarat)
- Rajkot (Gujarat)
18.65
18.6
18.55
d
Roa
ass
Byp
une
P
iba
Mum
18.7
Pimpri Chinchwad
160
140
Airport
120
100
80
60
Race Course
NH-9
18.5
40
20
Davachi Urali Waste Site
18.45
10
0
Khadakwasla Dam
NH-4
18.4
73.7
73.75
73.8
73.85
73.9
73.95
74
Sarath Guttikunda
Puja Jawahar
September, 2011
UrbanEmissions.Info, New Delhi, India (2011)
Analysis & errors are sole responsibility of the author(s).

www.urbanemissions.info
TableofContents
Chapter1
Chapter2
Chapter3
Chapter4
Chapter5
Chapter6
Annex1
Annex2
Annex3
ExecutiveSummaryofAirPollutionAnalysis
ScopeofthisStudy
1.1Integratedairqualitymanagement
ParticulatePollution&Cobenefits
2.1Cobenefitsofparticulatepollutioncontrol
CitiesataGlance
3.1Meteorologicalconditions
3.2Particulatepollution
3.3Industrialestates
3.4Transport
3.5Construction
3.6Powerplants
EmissionsInventory&DispersionModeling
4.1Previousstudies
4.2Emissionsinventoryfor2010
7
8
11
14
17
20
23
25
27
32
33
35
36
38
4.3Dispersionmodelingfor2010baselineemissions 46
4.4Percentagesectoralcontributions
50
HealthImpactsofAirPollution
55
5.1Studyobservations
58
WhatifScenariosfor2020&CoBenefits
61
6.12020baselinescenarioanalysis
62
6.2Whatifemissionscenarios
64
6.3Cobenefitsofwhatifscenarios
66
6.4Studyobservations
66
6.5Studyextensions
68
Methodologyfortransportemissions
71
ATMoSdispersionmodel
73
Reportsofhealthimpactsduetoairpollution 75
Sources of Air Pollution in Indian Cities
ListofFigures
Figure Title
Page
1.1
1.2
2.1
2.2
2.3
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
3.11
3.12
3.13
4.1
4.2
8
10
13
14
15
17
19
20
21
22
23
23
24
29
30
32
33
34
35
37
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
4.12
4.13
4.14
5.1
6.1
Locationofsixcities
SchematicframeworkoftheSIMairprogram
Atypicalchemicalcompositionofparticulatepollution
Cobenefitsofmanagementoptions
Radiativeforcingofclimatebetween1750and2005
Locationofsixcities
Geographicaldomainsofthesixcities
Annualwindspeedandwinddirections(2008)
Monthlyprecipitationpatterns(2008)
Dailymixinglayerheights(2008)
Boxmodelillustrationoftheimpactofthemixinglayerheight
NationalambientmonitoringnetworkinIndia
MonitoredmonthlyaveragePM10concentrations
Vehiclefleetcompositioninthesixcities
PassengertravelstatisticsinIndia
AnarielviewofabrickkilnclusteroutsideChennai(GoogleEarth)
AnaerialviewofquarriesalongthePuneNashikhighway
AnaerialviewofpowerplantsinChennaiandAhmedabad
Urbanairpollutionanalysissteps
Contribution of major sources to emissions inventory (200607) from
CPCB/MoEFstudy
ContributionofmajorsourcestoPM10emissionsinventory
ContributionofmajorsourcestoPM2.5emissionsinventory
ContributionofmajorsourcestoCO2emissionsinventory
Estimatedpercentpassengerkilometerstraveled
Estimatedgriddedvehicleexhaustemissions
DailynumberofflightsatAirports
Physicsofatmosphericdispersionmodeling
ModeledAnnualAveragePM10Concentrations
ModeledAnnualAveragePM2.5Concentrations
Schematicsforestimatingsectoralcontributions
ContributionofmajorsourcestoPM10pollutionbasedonMoEF/CPCB
sourceapportionmentstudies(200607)
Modeled2010PM10sourcecontributionsaveragedoverselectregions
acrossthecity
Schematicsofinformationtoassesshealthimpacts
Maximizingbenefitswithinformed&earlyaction
-2-
38
39
39
40
42
45
46
47
48
51
52
53
55
61
ListofTables
Table Title
Page
3.1
3.2
4.1
4.2
4.3
18
28
38
41
49
4.4
5.1
5.2
5.3
6.1
6.2
A3.1
Citiesataglancegeographicalandtransportcharacteristics
Baselinevehicleinventoryforthesixcities
Summaryof2010emissionsinventoryforthesixcities
SummaryofvehicularemissionfactorsfromCPCB/MoEFstudy
Summary of the 2010 PM10 emissions inventory; modeled and
measuredconcentrations
SummaryofpercentcontributionstomodeledPM10concentrations
Averagedoseresponsefunctionsformorbidityendpoints
EstimatedMortalityandMorbidityduetoairpollutionfor2010
EstimatedcostsforMortality&Morbidityduetoairpollutionfor2010
EstimatedMortalityandMorbidityduetoairpollutionfor2020
Estimatedcobenefitsfrom2020whatifscenarios
ReportsofHealthImpactsDuetoAirPollution
-3-
50
57
57
58
63
66
75
ExecutiveSummaryofUrbanAirPollutionAnalysisinIndianCities
Studydomainsize(kmxkm)
Longitude(degrees)
Latitude(degrees)
LandSeaBreeze
Elevation(meters)
domainPopulation(million)
Cityarea(squarekm)
Numberofmonitoringstations
AnnualaveragePM10(g/m3)
PM2.5measurements
VehicleFleet(millions)
(numbersrounded)CarsandJeeps
2Wheelers
3wheelers
Buses+StageCarriers
HDV+LDV+Others
Powerplants
BrickKilns(number)
2010PM10emissions(tons/yr)
2010PM2.5emissions(tons/yr)
2010CO2emissions(miltons/yr)
EstimatedPrematureDeaths
MortalitypertonofPM10
Mortalitycosts(millionUSD)
Morbiditycosts(millionUSD)
2020PM10emissions(tons/yr)
EstimatedPrematureDeaths
PM10emissionsreducedundersix
interventions(tons/yr)
%comparedto2020baseline
Prematuredeathssaved
%comparedto2020baseline
Mortalitysavings(millionUSD)
Morbiditysavings(millionUSD)
CO2emissionsreducedundersix
interventions(milliontons/yr)
Pune
32x32
7348E
1828N
NO
560
6.5
450
5
60160
Limited
2.3(2008)
323,400
1,708,100
66,500
15,100
151,730
NO
400
36,600
16,650
11.9
3,600
0.1
151
246
38,000
4,300
13,900
Chennai
44x44
8016E
1352N
YES
7
8.5
1,200
6
60120
Limited
3.8(2010)
565,350
2,986,600
55,400
15,600
123,920
YES(2)
600
56,400
26,000
30.6
3,950
0.07
165
269
55,100
6,000
17,400
Indore
32x32
7532E
2225N
NO
550
3.3
134
3
60170
NO
1.2(2010)
127,300
907,000
14,000
35,200
93,200
NO
120
17,900
9,900
8.6
1,800
0.1
75
122
21,000
2,500
6,200
Ahmedabad
44x44
7235E
2302N
NO
53
7.8
700
6
80100
NO
1.4(2010)
213,500
1,038,000
65,500
5,400
75,860
YES(2)
320
35,100
17,500
22.4
4,950
0.14
207
336
31,800
7,850
8,800
Surat
44x44
7250E
2110N
YES
13
5.0
105
3
75100
NO
1.3(2007)
132,750
1,063,000
65,400
1,900
69,840
YES
200
19,950
12,000
11.6
1,250
0.06
52
85
23,200
2,050
8,200
Rajkot
24x24
7047E
2218N
NO
134
1.4
310
2
80120
NO
1.1(2010)
126,700
878,000
8,860
79
46,900
NO
14,000
7,750
7.4
300
0.02
13
21
18,500
670
7,900
37%
1,700
39%
71
114
3.0
31%
1,270
21%
53
87
5.7
30%
630
25%
26
44
1.8
27%
1,390
18%
57
94
2.5
35%
590
29%
24
40
2.4
42%
290
42%
12
20
1.4
Chapter1.0
ScopeofthisStudy
AirpollutionisresponsibleforanincreasingnumberofmortalityandmorbiditycasesinIndian
cities. In a recent study, based on published information on air quality monitoring from the
CentralPollutionControlBoard(Delhi,India)andhospitaladmissionsrecordsfromcitiesacross
India,thecityofDelhiwasdeclaredthe"AsthmaCapital"ofIndia.1Airpollutionisacomplex
issue, fuelled by multiple sources ranging from vehicle exhaust, resuspended dust on the
roads due to vehicle movements, industrial flumes, construction debris, garbage burning,
domestic cooking and heating, and some seasonal sources such as agricultural field residue
burning,duststormsandseasalt (forcoastal areas).Whilestateandnationalauthoritiesare
taking necessary action and introducing interventions in varying capacities to curb these
emissionsandreduceambientpollutionlevels,alackofcoherentpolicyaswellasunplanned
growthacrosssectors(construction,transport,industry)ishinderingtheseefforts.
Accelerating growth in the transport sector, a booming construction industry, and a growing
industrialsectorareresponsibleforworseningairpollutioninIndiancities.Whileestimatesof
healthimpactsareeffectiveinraisingoverallconcernaboutairquality,theydonotspecifically
answer the question of where the pollution is coming from and how much each of these
sources contributes towards air pollution. Further uncontrolled growth will lead to more
pollution and require large recurring investments to control pollution. This study under the
SIMairprogramwasinitiatedwithsupportfromtheClimateWorksFoundation(USA)andthe
ShaktiSustainableEnergyFoundation(India)tobetterunderstandthesourcesofairpollutionin
sixcitiesinIndia,tosupportanintegrateddialoguebetweenlocalpollutionmanagementand
climate policy in a cobenefits framework. The six cities selected for this study are Pune
(Maharashtra);Chennai(TamilNadu);Indore(MadhyaPradesh);andAhmedabad,Rajkot,and
Surat(Gujarat).
1
Read more on the Delhi - Asthma Capital of India

@ http://mailtoday.intoday.in/index.php?id=18601&option=com_content&task=view&sectionid=23
The main objectives of this study, using local air pollution as the primary indicator in a co
benefitsframeworkare
ToestablishabaselineemissionsinventoryforthecriteriapollutantsPM,SO2,andNOx,
andthegreenhousegas(CO2)fromtheknownsourcesofairpollutioninthesesixcities
To analyze pollution due to these emissions and associated health impacts, based on
dispersionmodelingfortheselectedcitydomain
To analyze select interventions (from emissions and pollution perspective) for health
benefitsandGHGemissionreductions
To identify information gaps while building the emission inventories, which could
furtherunderstandingofairpollutionsourcesinthesecities
1.1IntegratedAirQualityManagement
Figure 1.1: Location of the six cities
Integrated air quality management (AQM) is a

challengeindevelopingcountriesbecauseofthe
lack of information on sources of air pollution
and insufficient ambient monitoring data in the
public domain. While we understand these
technicalconstraints,theobjectiveofthisstudy
Ahmedabad
is to establish an information base and a
Rajkot
Indore
methodology for the cities to follow in building
Surat
emissions inventories, conduct an analysis of
Pune
interventionsutilizingavailableinformation,and
identify information gaps for further studies,
especially those necessary to support and
Chennai
implementaneffectiveairqualitymanagement
plan for a city. The methodologies employed in
developing an emissions inventory and further
analysisispresentedinrelevantsectionsofthisreport.
Toconductintegratedairqualitymanagementinthesixcities,theSimpleInteractiveModels
forbetterairquality(SIMair)programwasutilized.TheSIMairprogram(Figure1.2)isafamily
ofopensourceanalyticaltoolstoestablishbaselineemissionsinventory,toconductdispersion
modeling,toassessthehealthimpactsofairpollution,andtoevaluatebenefitsofinterventions
tocontrolpollution.Thetoolsaredesignedinplugnplaymodulestoestimateandsupported
by publicly available software (such as MS Excel) and Geographic Information Systems (GIS)
data2.PreviousurbanapplicationsofthesetoolsincludeHyderabadandDelhi(India);Bangkok
(Thailand); Lagos (Nigeria); Antananarivo (Madagascar); Shanghai and Shijiazhuang (China);
Hanoi(Vietnam);Dhaka(Bangladesh);andUlaanbaatar(Mongolia).
The details of the SIM-air program and the tools are available @ http://www.urbanemissions.info
-8-
Duetoinherentchallengesincollectingnecessarydata,disclosureissueswithvariousagencies
or lack of institutional and technical capability to maintain such information, few cities have
developed or have made their emission inventories publically available in India. In this study,
weacknowledgethischallengeanduncertaintyinvolvedinsuchinventories.Thisisalearning
whilebuildingexerciseandtheemissionsinventoriesdevelopedthroughthisexercisecanbe
further improved as more capacity is built within the collaborating groups (with planned
trainingeventsinthefuture).Duringthestudyperiod,informationavailablefromvariouslocal,
national, and international academic groups were utilized to establish these baselines,
understand current trends in the ambient pollution levels, and analytical tools were used to
assesshealthimpacts3.
Box1.1RecentstudiesinIndiatounderstandairpollutionsourcesandstrengths
Air quality monitoring is the primary activity for most air quality
management institutions. In India, ~350 monitoring stations are
operated by the Central Pollution Control Board (CPCB) and the
State Pollution Control Boards (SPCBs). A summary of the annual
averagePM10concentrationsmeasuredin2008ispresentedhere4.
When compared to the national annual average standard of 60
g/m3 for PM10, more than 60 percent of these cities are
experiencingcriticallevels.
Whilethedatafromthesemonitoringstationsislocationspecific,
representativeofthepollutionsourcessurroundingthemonitors,
theydonotgiveanyspecificinformationaboutthesourcesnorthe
source contribution to the ambient pollution measured. There is
an acute need for apportioning these contributions in order to
betterunderstandairpollutioninthecountryandimproveairqualitymanagement5.
In 2011, the Ministry of Environment and Forests of India (MoEF) and CPCB, completed and
published a detailed study on source apportionment and emissions inventory (for their study
domains)forsixcitiesinIndiaPune,Chennai,Bangalore,Mumbai,Kanpur,andDelhi.Thiswork
was conducted for the base year 200607 providing necessary background information on the
sourcesofairpollution,anemissionsinventoryforthecriteriapollutants(forthestudydomains),
the results of a source apportionment study based on monitoring data, and an action plan to
controlairpollution6.AsummaryofresultsfromPuneandChennai,thetwocitiesoverlappingwith
thisstudyispresentedinthelatersections.
Inthisstudy,theemissionsinventoryandhealthimpactsanalysisisconductedforthebaseyear
of2010,followingtheschematicspresentedinFigure 1.2andtheemissionsareprojectedto
Examples of emissions inventories available in the public domain

Global Emissions Inventory Activity @ http://www.geiacenter.org/
Emissions Database for Global Atmospheric Research @ http://www.mnp.nl/edgar/
Greenhouse Gas and Air Pollution Interactions and Synergies @ http://gains.iiasa.ac.at/index.php
4
CAI-Asia, 2010, India Air Quality Profile 2010 @ http://cleanairinitiative.org/portal/node/6591
-9-
2020toevaluatetheimpactsandthepotentialforpollutioncontrolintermsofhealthand
carbonbenefitsofsixinterventions
1. anincreaseinthenonmotorizedtransport(cyclingandbiking)sharesby~20percent
2. anincreaseinthepublictransportsharesby~20percent(forthetrips)
3. analternativefuelforthepublictransportand3wheelers
4. apotential50percentreductionintheroaddust
5. a technology change in the brick kilns improving the efficiency and reducing the
emissionsby~50percent
6. areductionof~20percentofthetruckmovementinthecitylimits
Figure 1.2: Schematic framework of the SIM-air program

Air
Quality
Dispersion
Modeling
Analysis
Monitoring
Emissions
Inventory
SIM-air
Tools
Health
Impacts
Analysis
Knowledge
Control
Options
Theevaluationofcobenefitsintermsofmortalityandmorbidity(duetoexposuretopollution
levels)andconcurrentreductionsingreenhousegasemissionsfortheselectinterventions(to
supportclimatepolicyatglobalandnationallevel)wereevaluatedforthebaselineandcontrol
scenarios. The health impacts of air pollution are calculated using methodologies and dose
responses functions established through epidemiological studies. The methodology details
alongwithadatabaseofdoseresponsecoefficientsandreferencesarepresentedinthelater
sections.
A detailed overview of the source apportionment techniques top-down based on the monitoring techniques and
bottom-up based on the fuel consumption and activity levels is presented in Tools for Improving Air Quality
Management - A Review of Top-down Source Apportionment Techniques by Johnson TM, Guttikunda SK,
Watson J, Russell AG, West J, Bond T, and Artaxo P., including a summary of the applications from Asia, Africa,
and Latin America. Full report is available @ http://www.esmap.org/esmap/node/1159
6
Detailed report on the CPCBs source apportionment study from the cities in India is available
@ http://cpcb.nic.in/Source_Apportionment_Studies.php
A source apportionment and co-benefits of air pollution management for Hyderabad was conducted and published in
2007-08 @ http://urbanemissions.info/model-tools/sim-air/hyderabad-india.html
- 10 -
Chapter2.0
ParticulatePollution&CoBenefits
ThefivemainpollutantsresponsibleforthemajorityofthehealthimpactsinIndiancitiesare
particulates(PM),sulfurdioxide(SO2),nitrogenoxides(NOx),carbonmonoxide(CO),andozone
(O3). Each of the pollutants has an ambient standard (related to the pollution observed at
various locations) and an emissions standard (related to the weight of the pollution at the
source)7.Otherpollutantswithregulatorystandardsarethevolatileorganiccompounds(VOCs).
Ofthesepollutants,PM,SO2,CO,NOxandVOCsareconsideredtheprimarypollutants,asthey
are directly emitted at the source, while ozone is considered a secondary pollutant, as it is
formedduetochemicalreactionsamongtheprimarypollutants.GivenamixofNOxandVOC
emissions, in the presence of strong sunlight during the day, this leads to a buildup of ozone
throughphotolysis8.
Consequences of exposure to these pollutants range from premature mortality due to

aggravatedmorbidityeffectssuchasasthma,chronicbronchitis,andoxygendeficiencyinblood.
A detailed review of the impacts of outdoor air pollution in the developing countries was
conducted by the Health Effects Institute (USA)9. Other health impacts include eye irritation,
respiratory illness and impacts on reproductive health; ecological impacts include loss of
productivityandslowerphotosynthesisamongvegetation;structuralimpactsincludecorrosion
ofmetal,acceleratederosiononbuildingsandmonuments;aestheticimpactsincludebadsmell,
reduced visibility, and accumulation of soot and dust on buildings. While these impacts are
7
Read more on the national ambient air quality standards in India

@ http://cpcb.nic.in/National_Ambient_Air_Quality_Standards.php
8
Refer to the books Atmospheric Chemistry and Physics by Seinfied and Pandis, and Fundamentals of
Atmospheric Modeling by Mark Jacobson, which provide detailed discussion on the photochemistry of NOx, VOCs,
and ozone, and more on the air pollution modeling
9
See a list of publications summarizing the impacts of outdoor air pollution on human health based on
epidemiological studies for the particulates and other primary pollutants conducted across the world is distributed by
the Health Effects Institute, USA @ http://www.healtheffects.org/
mostly experienced in the urban centers, regional impacts include the loss in the agricultural
yieldduetoacidrainandozoneexposure,andglobalimpactsincludeincreaseinemissionsthat
haveanimpactonclimatechange.
Each of the pollutants is associated with a range of health impacts. For example, PM is the
leadingcauseofrespiratoryillnesses,ozonepollutioncauseseyeandlungirritations,SO2(with
precipitation) results in acid rain, irritation along the respiratory track and bronchitis, NOx
enhancesthesymptomsforchronicbronchitis,andCOreducestheoxygensupplytothebrain
(and is in some cases fatal). We experience health impacts due to air pollution from all the
pollutants.However,thehealthimpactsduetoPMareconsideredthemostharmful,duetoits
size,composition,directlinkagestoaggravatingchronicandacuterespiratorydiseases10.
StudiesinIndiasuggestthatacuterespiratoryinfection(ARI)inchildrenbelow5yearsofageis
thelargestsinglediseasecategory,accountingforabout13percentofthenationalburdenof
disease11, and children living in households using solid fuels have 23 times more risk of ARI
thanunexposedchildren.In1995,airpollutioninChinafromfuelcombustionwasestimatedto
havecaused218,000prematuredeaths(equivalentto2.9millionlifeyearslost),2millionnew
cases of chronic bronchitis, 1.9 billion additional restricted activity days, and nearly 6 billion
additional cases of respiratory symptoms12. The main pollutant in both China and India is
believed to be fine PM. Recent studies put the risk of exposure to urban air pollution at
~900,000prematuredeathsinAsia13.
In this study, emissions inventory, pollution modeling, and health impacts assessments are
calculatedusingPMastheprimaryindicator.
While PM pollution is measured and estimated as a single entity, the composition of PM is

complex and accounts for contributions from other pollutants. By restricting the impact
assessments to PM, we avoid any double counting of impacts between pollutants. A typical
chemicalcompositionofPMispresentedinFigure2.1,whichconsistsofdirectemissionsfrom
sourcessuchasvehicleexhaust,industry,anddomesticfossilfuelcombustion;intheformof
metals,crustalelements,soot(alsoknownasblackcarbon)andsecondarycontributionsdueto
thechemicaltransformationofgaseouspollutantsintoaerosolssuchasSO2tosulfates,NOxto
nitrates, and VOCs to secondary organic aerosols (SOAs). Similarly, the contribution to ozone
pollution is also accounted in the form of SOAs which form during the photolysis reactions
betweenNOxandVOC.
10
A detailed explanation on the science and composition of the PM pollution is presented in SIM-air Working Paper
No.10 What is Particulate Matter: Composition and Science and No.16 Urban Particulate Pollution Source
Apportionment - Part 1. Definition, Methodology, and Resources @ http://www.urbanemissions.info
11
Comparative Quantification of Health Risks @ http://www.who.int/publications/cra/en
12
Lvovsky, et al. 2000. Environmental Costs of Fossil Fuels: A Rapid Assessment Method with Application to Six
Cities. Environment Department Paper No. 78, The World Bank, Washington DC, USA
13
A summary of the mortality and morbidity health impacts, estimated and surveyed, reported from cities and
countries across the world, along with a methodology to estimate health impacts of outdoor air pollution are
presented in Estimating health impacts of air pollution, SIM-air working paper series, SIM-06-2008, available
@ http://urbanemissions.info/sim-series-06.html
- 12 -
Figure 2.1: A typical chemical composition of particulate pollution

Primary Particles
(directly emitted)
Secondary Particles
(from precursor gases)
VOCs
Carbon
Carbon
(Soot)
(Soot)
Organic
Organic
Carbon
Carbon
Metals
Metals
Ammonium
Ammonium
Sulfate
Sulfate
Crustal
Crustal
(soil,dust)
(soil,dust)
SO2
Other
Other
(sea salt)
(sea salt)
Ammonium
Ammonium
Nitrate
Nitrate
Ammonia
NOx
ThePMsizeisalsocriticalforestimatingandstudyingitsimpactonhumanhealth.Generally
referredsizefractionsare
Totalsuspendedparticulates(TSP,withaerodynamicdiameter<~30microns(m)
PM10(withanaerodynamicdiameteroflessthan10m,alsoreferredtoascoarse)
PM2.5(withanaerodynamicdiameteroflessthan2.5m,alsoreferredtoasfine)14
UltrafinePMarethosewithadiameteroflessthan0.1micron
ThePMsizefractionsaffecttheatmosphericlifetime,spatialdistribution,temporalvariability,
and health impacts of particles. In general, smaller the particle size, higher the damage. For
example, the PM2.5 size fraction is considered more harmful than the PM10 size fraction,
becausethePM2.5particlescantravelfartherintothelungsandcancausemoredamagethan
thePM10particles15.
Thelifetimeoftheparticlesalsovariesbysizeandshape.Coarseparticleshaveanaveragelife
time (resuspension before they settle on a canopy or wet deposited due to precipitation)
betweenminutestohours,whilethefineparticlesaveragebetweendaystoweeks.Also,the
travel distances for coarse particles range from 1 to 10s of kilometers and the fine particles
averagefrom100sto1000sofkilometers.
14
In India, PM2.5 was added to the list of criteria pollutants in November, 2009, and the SPCBs are required to
monitor PM2.5 along with PM10 at all their monitoring stations. However, the capacity to do so is still limited and
only a handful of the cities are monitoring and reporting PM2.5 pollution levels.
15
While the PM2.5 is considered more harmful and an important fraction of the PM pollution, in this study, the
emissions and health impacts analysis is limited to PM10 due to data constraints in the monitoring data. The
monitoring data from the cities is available only for the PM10 fractions. While we recognize the constraint in the
monitoring data and all the modeling and comparisons are carried out for the PM10 pollution, the final emissions
inventories and modeling data is available for both the fractions for further use.
- 13 -
2.1CoBenefitsofParticulatePollutionControl
A cobenefits approach16is increasingly becoming a starting point for discussing integrated

programsbenefitingbothclimatechangeandairquality.Figure 2.2depictsascenariowhere
the cobenefits can aide decision making over a variety of control measures. For example,
policies designed to reduce the impact of transport on air quality by tackling congestion and
encouraging a shift to public transport, walking, and cycling also reduces CO2 emissions.
Measurestoimproveenergyefficiencyandcutenergydemand,reducesairpollutantsandGHG
emissionsalike.
Figure 2.2: Co-benefits of management options17
In most developing countries, the
cobenefits approach is being
+ Carbon
recognized as a practical and
Increased
Natural Gas
effective tool in technical, policy,
Natural Gas
motorization
Power (leaks)
buses
economic,
and
institutional
Carbon Increase
Worse Air Quality
Carbon Increase,
perspectives. With the growing
Improved Air Quality
awareness for clean air, besides
Increased fuel
consumption
the
pressure
to
support
infrastructure needs, local impacts
+ PM
- PM
Renewable
Energy
energy
Efficiency
of air pollution will have a higher
Worse Air Quality
priority over global impacts of
Carbon Savings
Improved Air Quality
Carbon Savings
GHGs. However, this shouldnt
Biomass
suggest that the concept of low
NMT
BRT
Burning
carbon society takes a back seat.
- Carbon
The interventions that cities are
trying to implement, including
improvingpublictransport(roadandrail)orpromotingnonmotorizedtransportandincreasing
theenergyefficiencyintheindustriesreducepollutantsthatcontributetobothlocalaswellas
globalenvironmentalimpacts.
Inanurbanenvironment,thehealtheffectsduetoairpollution(mortalityandmorbidity)are
theprimaryconcern.Atthesametime,someknownfractionsofPMpollution(suchasBlack
CarbonFigure2.1)andthesecondarypollutantssuchassulfates(fromSO2)andozone,also
affects regional and global atmospheric chemistry and the radiation balance. Figure 2.2
summarizestheradiativeforcingofgreenhousegases(GHGs)andvarioustypesofaerosols18.
Thepresenceoflargequantitiesofaerosolsandthelinkedphysicalandchemicalinteractions
areknowntochangetheearthsradiationbalance,affectinglocalandglobaltemperaturesand
possiblyprecipitation.Thelevelofinteractionsbetweenvarioustypesofaerosolsandclimatic
16
The theme of the 5th Better Air Quality Conference for Asian Cities, the largest air quality conference of Asia is
Co-Benefits, www.baq2008.org
17
Source: Modified from presentation by Dr. Cornie Huizenga, Executive Director, CAI-Asia, Manila, Philippines
18
Intergovernmental Panel on Climate Change. Climate Change 2007: The Physical Science Basis, Summary for
Policymakers
- 14 -
conditions, along with the presence of the greenhouse gases like CO2, N2O, and CH4 is an
emergingscience19.
Figure 2.3: Radiative Forcing of Climate

At the global level, the Figure 2.3 shows
between 1750 and 2005
thatalthoughGHGsarequiteimportantin
the overall picture, pollutants that are
usually considered only in the air quality
domain, such as aerosols and ozone, also
have considerable affect on climate
change. Note that the sources of GHGs
and the local air pollution are the same
fossil fuel combustion. Thus, in an urban
environment, an intervention introduced
to control air pollution, the combined
benefits (also known as cobenefits)
include considerable reductions in the
GHG emissions (like CO2) and health
benefits(duetoreductioninPM)andaide
in the reduction of the climate radiative
forcing.Inthisstudy,weconsiderthePM
pollutioninacityastheprimaryindicator
and the CO2 emission reductions of select
interventionsasacobenefit.Duetothenatureoftheanalysisinthisstudy,whichfocuseson
thehealthimpactsofPMpollutioninacity,noindependentanalysiswasconductedforblack
carbon(orotheraerosolfractions)andtheirimpactofclimateradiativeforcing;whichcanbe
extendedinfuturestudies.
19
Climate Impacts of Black Carbon A Review and Policy Implications A report by the Princeton University.
Full report is available @ http://www.bioenergylists.org/sellersprincetonbc
- 15 -
Chapter3.0
CitiesataGlance
Thesixcitiesselectedforthisstudy(presentedinFigure3.1)arePune(Maharashtra);Chennai
(TamilNadu);Indore(MadhyaPradesh);andAhmedabad,Rajkot,andSurat(Gujarat).Thestudy
domain,geographicalconditions,population,generalsourcecharacteristicsfromindustrialand
powergeneration,andvehicleandroadstatisticsarepresentedinTable3.1andsomeofthe
geographical characteristics for each city are presented in Figure 3.2. Each of the city maps,
representthemainroadsrunningthroughthecity,includinghighways,pointsofinterest,brick
kilnclusters(reddots),industrialestates,powerplants(forthreecitiesChennai,Ahmedabad,
andSurat)andthemaincitydistrictboundaries.
Inelevation,exceptforPuneandIndore(bothat
~550 meters), remaining cities are close to the
sealevel(rangingbetween6to50meters).The
studydomainsforeachofthecitiesareselected
to be large enough to cover the main district
area, the nearest satellite cities, and cluster
locations with sources that could influence the
airqualityinthemaindistrictareas.Thedomain
are designated at 44 km x 44 km for Chennai,
Ahmedabad,andSurat;32kmx32kmforPune
and Indore; 24 km x 24 km for Rajkot the
smallestofthesixcities.Forexample,theblack
polygons in the city map of Pune represent the
quarries, a primary source of construction
material for the city covering an area of 11
square km, a significant emissions source of
fugitive dust emissions from the manufacturing
Figure 3.1: Location of the six cities
Ahmedabad
Rajkot
Indore
Surat
Pune
Chennai
processes,useofdieselintheinsitupowergeneratorsets,andthevehicleexhaustfromthe
trucks moving in and out of the quarries. If only the main district area is considered for the
emissions and pollution analysis, this important source will be missed. Similarly, the city of
ChennaiisexpandedfarthertotheWesttoincludethebrickkilnclusters,animportantsource
ofcriteriaandGHGpollutantsfrommanufacturingandfreightmovement.Theemissionsand
pollutionanalysiswasconductedat1kmgridresolutionforallthecities.Thebigbluepatches
inthecitymapsofChennaiandSuratindicatetheBayofBengalandtheArabianSea;andthe
smallerinlandbluepatchesinPune,Chennai,Indore,andRajkotindicatelakesanddams.
Table 3.1: Cities at a glance geographical and transport characteristics

Pune
Chennai Indore Ahmedabad Surat
Studydomainsize(kmxkm) 32x32
44x44
32x32 44x44
44x44
Longitude(degrees) 7348E
8016E
7532E 7235E
7250E
Latitude(degrees) 1828N
1352N
2225N 2302N
2110N
LandSeaBreeze NO
YES
NO
NO
YES
Elevation(meters) 560
7
550
53
13

domainPopulation(million) 6.5
8.5
3.3
7.8
5.0
Cityarea(squarekm) 450
1,200
134
700
105

Powerplants NO
YES(2)
NO
YES(2)
YES
Installedcapacity(MW)
1,080
1,100
500
Dominantfuel NA
Coal
NA
Coal
Gas

IndustrialEstate Medium Large

Small
Medium
Large
BrickKilnsBullTrench NO
YES
NO
NO
NO
BrickKilns(number) 400
600
120
320
200

60120
60170 80100
75100
AnnualaveragePM10(g/m3) 60160
Numberofmonitoringstations 5
6
3
6
3
PM2.5measurements Limited
Limited
NO
NO
NO

VehicleFleet(millions) 2.3(2008)
3.8(2010)
1.2(2010) 1.4(2010)
1.3(2007)
(numbersrounded)CarsandJeeps 323,400
565,350
127,300 213,500
132,750
2Wheelers 1,708,100 2,986,600 907,000 1,038,000
1,063,000
3wheelers 66,500
55,400
14,000 65,500
65,400
Buses+StageCarriers 15,100
15,600
35,200 5,400
1,900
HDV+LDV+Others 151,730
123,920
93,200 75,860
69,840

%HDVoftrucks 26%
50%
NA
14%
23%
%2Wheelersintotalfleet 75%
79%
77%
74%
80%
%Carsintotalfleet 14%
15%
11%
15%
10%

NA=notapplicable;HDV=heavydutyvehicles
- 18 -
Rajkot
24x24
7047E
2218N
NO
134
1.4
310
NO
NA
Medium
80120
2
NO
1.1(2010)
126,700
878,000
8,860
79
46,900
47%
82%
12%
Figure 3.2: Geographical domains of the six cities
Rajkot
Ahmedabad
Indore
Surat
Chennai
Pune
Ineachofthegeographicalstudydomainspresentedabove,exceptforthecityofRajkot,thethickdarksquare
representsthemaindistrictareawithmostofthecitypopulation;thedotsrepresentthebrickkilnsandindustrial
units;theredlinesrepresentthemainroadsandhighwayspassingthroughthecity
- 19 -
3.1MeteorologicalConditions
The meteorological conditions wind speeds, wind directions, precipitation, and the mixing
layerheights,prevalentineachofthecitiesarepresentedhere.
Figure 3.3: Annual wind speed and wind directions (2008)
Pune, Maharashtra
Chennai, Tamilnadu
Annual 2008
Annual 2008
315
<=1
>1 - 2
>2 - 3
>3 - 4
>4 - 5
>5
45
270
315
90
0%
4%
8%
225
270
90
12%
0%
135
12%
16%
180
Ahmedabad, Gujarat
Annual 2008
Annual 2008
315
<=1
>1 - 2
>2 - 3
>3 - 4
>4 - 5
>5
45
270
315
90
2%
4%
6%
225
<=1
>1 - 2
>2 - 3
>3 - 4
>4 - 5
>5
45
270
90
8%
0%
135
2%
4%
6%
225
180
8%
10%
135
180
Surat, Gujarat
Rajkot, Gujarat
Annual 2008
Annual 2008
315
<=1
>1 - 2
>2 - 3
>3 - 4
>4 - 5
>5
45
270
315
90
225
8%
135
180
0%
4%
225
Indore, Madhya Pradesh
0%
<=2
>2 - 4
>4 - 6
>6 - 8
>8 - 10
>10
45
2%
4%
6%
8%
270
90
10%
0%
135
225
180
4%
8%
12%
135
180
- 20 -
<=1
>1 - 2
>2 - 3
>3 - 4
>4 - 5
>5
45
ThemeteorologicaldatapresentedinFigure3.3forannualwindspeedsanddirections,Figure
3.4 for monthly precipitation and Figure 3.5 for mixing layer heights are obtained from the
NCEP Reanalysis fields, an opensource database maintained and distributed by NCAR20. The
meteorological data for precipitation, wind speeds and wind directions are available for the
period from 1948 to 2010. For this study, data from year 2008 is extracted for the grid cells
coveringthecityboundariesofPune,Chennai,Indore,Ahmedabad,Surat,andRajkot21.
Figure 3.4: Monthly precipitation patterns (2008)

Pune, Maharashtra
Chennai, Tamilnadu
Precipitation(mm/month)Pune
300
Precipitation(mm/month)Chennai
250
250
200
200
150
150
100
100
50
50
0
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Oct
Nov
Dec
Oct
Nov
Dec
Ahmedabad, Gujarat
Precipitation(mm/month)Indore
300
Mar
Precipitation(mm/month)Ahmedabad
300
250
250
200
200
150
150
100
100
50
50
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Surat, Gujarat
May
Jun
Jul
Aug
Sep
Rajkot, Gujarat
Precipitation(mm/month)Surat
400
Apr
Precipitation(mm/month)Rajkot
300
350
250
300
250
200
200
150
150
100
100
50
50
0
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Jan
Dec
20
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
NCEP/NCAR Reanalysis data for a number of meteorological conditions is available for a period starting from
01/01/1948 @ http://www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html The meteorological processing
is conducted before the dispersion modeling.
21
To calculate mixing layer height, the NCEP reanalysis data is re-gridded and multiple parameters are utilized,
such as 3D wind and temperature, surface wind, temperature, and pressure, and 2D surface heat flux, and
precipitation fields.
- 21 -
The port cities of Chennai and Surat are the wettest and as illustrated by the wind rose
functions (in Figure 3.3), they also experience the strongest landsea breeze affects. These
strongwindsblowingawayfromthecity,for themostpartoftheyear,resultis dispersinga
significantamountofthepollutionfromtheindustriesandthepowerplantslocatedalongthe
coast. The precipitation fields (in Figure 3.4) are vital in dispersion modeling, as precipitation
accounts for the wet deposition of pollution and entrainment of suspended particles; and
reducestheambientpollutionlevelsnaturally.Thewetsurfaceconditionsarealsohinderre
suspensionofthedustparticlesontheroads.
Figure 3.5: Daily mixing layer heights (2008)

Pune, Maharashtra
2000
DayAverageMixingLayerHeight(m)Pune
Chennai, Tamilnadu
1800
1600
1600
1400
1400
1200
1200
1000
1000
800
800
600
600
400
400
200
DayAverageMixingLayerHeight(m)Chennai
2000
1800
200
0
Jan08 Feb08 Mar08 Apr08 May08 Jun08
Jul08 Aug08 Sep08 Oct08 Nov08 Dec08

2000
DayAverageMixingLayerHeight(m)Indore
Ahmedabad, Gujarat
2000
1800
1800
1600
1600
1400
1400
1200
1200
1000
1000
800
800
600
600
400
400
200
DayAverageMixingLayerHeight(m)Ahmedabad
200
Surat, Gujarat
2000
DayAverageMixingLayerHeight(m)Surat
Aug08 Sep08 Oct08 Nov08 Dec08
Rajkot, Gujarat
2000
1800
1800
1600
1600
1400
1400
1200
1200
1000
1000
800
800
600
600
400
400
200
DayAverageMixingLayerHeight(m)Rajkot
200
0
Jul08
0
Jul08
Aug08 Sep08 Oct08 Nov08 Dec08
The mixing height plays a critical role in dispersion of the emissions over the seasons. The
wintermonthsaremorepollutedthanthesummermonthsfortworeasonsone,precipitation
is usually higher during the summer monsoon and two, the mixing layer is lower during the
winter months. Mathematically, this is illustrated in Figure 3.6 as a box model. By definition,
the ambient concentration is defined as mass over volume. Assuming that the emissions are
- 22 -
equallymixedinanurbanenvironmentunderthemixinglayer,forthesameemissions,alower
mixing height means higher ambient concentrations. Similar to the mixing layer height, the
windspeedisalsoveryrelevant.Thehigherwindspeedsareresponsiblefordrivingpartofthe
pollutionoutofthecitylimitsandthusreducingtheaveragecontributionofthelocalemissions.
Ontheotherhand,theemissionsfromregionsoutsideofthecityalsocontributetothelocal
pollution, which is coined as long range transport component, affecting the background
concentrationsinthecity.
Figure 3.6: Box model illustration of the impact of the mixing layer height
E
C
u *l * H
H
w
E
C
u *l * h
= wind speed
h
w
l
ThemixinglayerheightsinFigure 3.5,exceptforChennai,followatypicalpatternofhighsin
thesummerandlowsinthewintermonths.Themixinglayerheightisvitalforthedispersionof
emissions in a city. In the winter months, a lower height hinders dispersion and thus
exacerbates the ambient pollution levels, unlike the summer months. In the caseof Chennai,
landseabreezeandstrongerwinds(comparedtoothercities)resultinamoreuniformmixing
layer height over the seasons. Meteorological parameters are also available at the seasonal
level(uponrequest).
3.2ParticulatePollution
The Central Pollution Control Board operates a

networkofapproximately400+stationsin130cities
across India, measuring ambient levels of criteria
pollutantssuchasSO2,NOx,andPM10.Amajorityof
thestationsaremanuallyoperated,whichresultsin
a discontinuity of data from these stations. In
November,2009theMoEFandCPCBproposednew
protocols for the States to improve monitoring at
thesestationsandimprovetheircapacitytosupport
policy decisions. New pollutants like PM2.5, Ozone,
and some carcinogenic volatile organic compounds
suchasBenzene,wereaddedtothemonitoringlist,
which adds strength to data collection efforts. In
major cities, continuous monitors are now
operational,whichprovideadditionalinformationon
- 23 -
Figure 3.7: National ambient

monitoring network in India
thesepollutants.TherangeofaveragePM10concentrationsmeasuredinPune,Chennai,Indore,
Ahmedabad, Surat, and Rajkot is presented in Table 3.1 (all the stations are operated by the
statepollutioncontrolboards)andasummaryofthemonthlyaverageconcentrationsfromall
stationsineachcityispresentedinFigure3.8.Allsixcitiesexceedtheannualambientstandard
of60g/m3.
Of the six cities, Pune, Chennai, and Ahmedabad are the only cities conducting PM2.5
monitoring,howeverthedataavailabilityislimited.ForPM2.5,thedailyambientstandardis60
g/m3andanannualambientstandardis40g/m3.
Figure 3.8: Monitored monthly average PM10 concentrations
Pune, Maharashtra
Chennai, Tamilnadu
180
180
150
150
120
120
90
90
60
60
30
30
0
Jul09 Aug09 Sep09 Oct09 Nov09 Dec09 Jan10
Ahmedabad, Gujarat
180
180
150
150
120
120
90
90
60
60
30
30
0
Jan10 Feb10 Mar10 Apr10 May10 Jun10 Jul10 Aug10 Sep10 Oct10 Nov10 Dec10
Surat, Gujarat
Rajkot, Gujarat
180
180
150
150
120
120
90
90
60
60
30
30
0
- 24 -
Thefollowingsectionincludesadiscussionofthemajorsectorsforeachcityandprovidesan
overviewofthecityscharacteristicsinacomparativeframework.
3.3IndustrialEstates
Puneisknownforseveraluniversitiesandeducationalinstitutions.Italsohasawellestablished
manufacturing, glass, sugar, and forging industries since the 195060s. Recently, the
informationtechnology(IT)andtheautoindustry22havegrownsubstantially23.TheHinjawadi
ITPark(officiallycalledtheRajeevGandhiITPark)(locatedwestofthePimpriChinchwadcityis
aprojectstartedbyMaharashtraIndustrialDevelopmentCorporation(MIDC)24.SimilarITparks
constructed under the SEZ schemes include the Magarpatta city to the east of the city with
closeto5,000householdsresidinginthearea.
Chennai is the most commercial of the six cities. With its proximity to the Bay of Bengal and
accesstothemarketsinEastAsia,thecityhasemergedasanimportantport.Apartfromtrade
and shipping, the automobile industry, software services, medical care, and manufacturing
form the foundation of the economic base for Chennai. In particular, the auto industry for
domestic use and export purposes has established itself around Chennai thus giving it the
label of Detroit of India. Manufacturers like Ford, Hyundai, HMMitsubishi, Ashok Leyland,
TAFE,etc.,havetakenadvantageoftheproximitytotheport,engineeringandmanufacturing
industryaswellastheskilledlabortoestablishthemselvestotheextentthatthecityaccounts
for ~30 percent of Indias auto industry. For example, Nissan indicated that exports are
expectedtoreach110,000unitsin201125.Similarly,thechemical,petrochemicalandmineral
industries have also established themselves in the outskirts of Chennai. The Ennore Port, the
first corporate major port, which was originally started for handling only oil, had now grown
intoonehandlingvariouscommodities.Themajorcargoincludescoal(mostofthesupplyisfor
thetwopowerplantswithdedicatedfeederlinesrunningfromtheports),oil,andironore.The
currentcapacityof15milliontonsofcargoisexpectedtoreach30millionby2012andtripleto
90milliontonsby202026.
IndoreisthecommercialcenterofMadhyaPradesh,~200kmfromthecapitalofBhopal.The
city is well connected to other parts of the state through a network of national and state
22
Automotive companies like Tata Motors, Mercedes Benz, Force Motors (Firodia-Group), Kinetic Motors, General
Motors, Volkswagen, and Fiat have manufacturing facilities in and near Pune. Several automotive component
manufacturers like Saint-Gobain Sekurit,TATA Autocomp Systems Limited, Robert Bosch GmbH, ZF
Friedrichshafen AG, Visteon, and Continental Corporation also set up facilities in and around Pune.
23
Details of the research work at ARAI are available @ https://www.araiindia.com. ARAI played an integral part in
testing and establishing the emission factors for 62 vehicle categories in the Indian fleet, as part of the six city
particulate pollution source apportionment studies conducted by Ministry of Environment and Forests and Central
Pollution Control Board of India. Details of the source apportionment studies are available
24
The Economic Times, Hinjewadi: The land of opportunity, December 7th, 2007
@ http://economictimes.indiatimes.com/articleshow/2604416.cms
25
The Business Line, August 31st, 2010, Nissan to export Micra thru Ennore port soon
@ http://www.thehindubusinessline.in/2010/09/01/stories/2010090151270200.htm
26
The Ennore port in Chennai @ http://www.ennoreport.gov.in/
- 25 -
highways,becauseofwhich,itisfastemergingasanimportanttransportandlogisticshubin
thecountry.ThemajornationalhighwayspassingthroughthecityareNationalHighwayNo.3
(NH3AgraBombay);NationalHighwayNo.59(NH59IndoreAhmedabad);NationalHighway
No.59A(IndoreBetul);StateHighwayNo.27(IndoretoBurhanpur);andStateHighwayNo.34
(Indore to Jhansi). The main economic activities in Indore are manufacturing & service
industries (soybean processing, automobile, software, and pharmaceutical). Major industrial
areas surrounding the city include the Pithampur special economic zone and the Sanwer
industrialbelt.
Ahmedabad city is the 7th largest city in India and was the capital of Gujarat in 196070s
(thereafterthecapitalwasshiftedtoGandhiNagar,30kmaway)andconnectedtomajorcities
likeMumbai,Pune,Vadodara,andSuratwithsevenmajorroadways,oneexpressway,andfive
rail networks. The Sabarmati River divides the city into the eastern and western regions. The
eastern bank of the river houses the old city and industries. As the population grew, the city
expanded to the west, with newer construction, educational institutions, residential areas,
shoppingmalls,andbusinessdistrictsclusteredaroundarterialroads.Thecitywasonceknown
for its textile industry, with as many as 66 mills employing a workforce of over ~100,000. A
rapidgrowthofchemicalandpetrochemicalindustriesinSouthGujaratdistrictswasobserved
within its municipal limits of Naroda, Odhav, Vatwa, and Behrampura, accounting a total of
~5,000industriesandemploying~300,000workers.Thecityaccountsfor~19percentofmain
urbanworkersinthestateand60percentinAhmedabadDistrict27.
Surat, lies between Ahmedabad and Mumbai on the river Tapi and like the city of Chennai,
includes an active port to the west. The Hazira industrial estate, closer to the port, hosts a
number of petrochemical and steel refining units. The Gujarat Industrial Development
Corporation (GIDC) has set up industrial estates in Pandesara, Khatodara, Udhana, Katargam,
SachinandBhestanareas.MuchoftheindustrialactivityinSuratsupportsthemajorindustries
in the port of Hazira. Economically, Surat is known for its textile manufacturing (accounts for
~40%ofmanmadefabricproductionand~33%ofmanmadefiberproduction),trade,diamond
cutting and polishing industries (accounts for ~75 % of the countrys total rough diamond
cuttingandpolishingand~43%ofdiamondexports),intricatezariworks,chemicalindustries
and the gas based industries at Hazira established by leading industry houses such as ONGC,
Reliance,ESSAR,andShell.Therearealargenumberofsmallandmediumsizedindustrialunits
(42,509units)thataccountforabout15percentofthesmallscaleindustrialunitswithinthe
stateofGujarat.MuchoftheindustrialdevelopmentislocatedwithinthelimitsofSuratcity
andover50percentoftheworkforceinSuratisengagedinmanufacturingrelatedactivity28.
Rajkotisarelativelysmallcitywithapopulationof1million.Ithasadryandaridclimate,yet
severalagriculturalgoodssuchascotton,groundnutandotheroilseedsaregrownintheregion.
Ithasalargeprocessingsector,especiallyforoils.About5largeedibleoilmillsoperateinthe
city,inadditiontomorethan25smalloilmillsthatextractedibleoilslikegroundnut,sesame
27
28
JNNURM, City Development Plan for Ahmedabad, Gujarat

JNNURM, City Development Plan for Surat, Gujarat
- 26 -
and cottonseed oils. Other small and medium industries include foundries, machinery,
engineeringandautomobilecomponents,castoroilprocessing,goldandsilverjewellery,most
of which are clustered around the two main industrial estates of Aji and Bhaktinagar. Of the
estimated8,000industrialunitsinRajkot,3,150unitsproduce~300,000oilenginesandparts
thereof.Theseenginesaremainlyusedinagricultureandfortheexportmarket.Thereare~400
foundries and forging units in the city. For the automobile sector, the units in Rajkot
manufacturecolors,ballbearings,etc.
3.4Transport
ThevehiclefleetcharacteristicsarepresentedasasummaryinTable3.1anddetailedbreakup
ofthefleetinTable3.2.Thevehiclefleetinformationutilizedinthisstudycomesfrommultiple
sourcesandcompiledformultipleyears(withthebaseyearmentionednexttothetotalvehicle
fleetsizeinTable3.1).
IncaseofPune29andChennai,informationcomesfromtherespectivepolicedepartments;for
Ahmedabad from their ministry of road transport; for Indore from the local pollution control
department; and for the cities of Rajkot and Surat, data is collated from their respective
JNNURMreportsandextrapolatedtothebaseyearof2010.
Since this information is sourced from multiple locations, there is lack of consistency in the
categoriesbetweenthecities.Thisshouldntimplythatanyofthecategoriesaremissingform
thelist,butmorelikelytohavebeenclubbedtogetherinbroadertitles.Forexample,incase
Indore,thereisnolistingfortheheavydutyvehicles,whichisunlikelyforacityasindustrial
andcommercialasIndore,withabuddingfreighthandlinghubinthevicinity.Uponenquiry,it
wasnotedthatallthetrucksarelistedunderthelightdutyvehiclesprimarilyduetolackof
segregatedinformationavailablewiththelocalpollutioncontrolboard.
Thisisanimportantinformationgap,whichcouldinducesomeuncertaintyintotheemissions
calculations. For the emissions inventory calculations, the categories are clubbed into cars,
multiutilityvehicles(includingjeeps,ambulances,policevans),motorcycles,3wheelers,city
buses,shorthaulbuses(includingschoolbuses),lightdutycommercialvehiclesandheavyduty
commercialvehicles(includingtrailors,tractors).ThetotalfleetsizeisthehighestforChennai
at~3.8millionregisteredvehiclesinthecityandfleetpercentagesarepresentedinFigure3.9.
29
Further statistics on the vehicle counts prior to 2004 and traffic management are available
@ http://www.punepolice.gov.in/statistic.htm#gipovdtp5y
The vehicle fleet compositions utilized in this analysis matched to the numbers presented in Environmental Status
Report for Pune city 2009-10 (in Marathi) available @ http://punecorporation.org/pmcwebn/index.aspx
- 27 -
Table 3.2: Baseline vehicle inventory for the six cities

Vehicle
Motorcycle
Scooter
Moped
Car
Jeep
StationWagon
Taxi/Cab
3Wheeler
Pune
(2008)
1,074,891
402,523
230,698
268,957
53,401
1,015
19,221
66,522
Chennai
(2010)
1,805,213
417,899
763,481
553,286
11,197
871
45,807
55,366
Indore
(2010)
907,062
122,943
4,392
4,348
13,963
Ahmedabad
(2009)
887,210
150,829
199,732
13,773
2,845
65,518
Surat
(2007)
1,062,949
121,862
10,890
2,298
65,385
StageCarrier
ContractCarrier
MiniBus
SchoolBus
PrivateServiceVehicle
Ambulance
HeavyDutyTruck
Tanker
LightDutyTruck
Tractor
5,845
6,864
545
1,835
1,321
39,128
3,938
65,171
24,889
6,268
499
170
3712
4,955
1,545
61,661
5,417
30,106
15,009
34,691
502
65,328
16,147
1,759
716
2,238
304
367
485
10,589
1,114
27,776
18,995
1,113
443
313
407
16,189
691
27,472
14,410
10,186
3,792,648
9,239
2,490
1,181,105
10,222
6,679
1,401,151
9,427
1,247
1,335,096
Trailor
Others
13,005
4,277
Total 2,284,046
Rajkot
(2010)
878,133
88,706
38,017
3,825
8,860
57
22
22,032
24,851
1,064,503
ForcomparativepurposesinTable3.1,thetruckfleetnumbersareclubbedforlightdutyand
heavy dutyvehicles, including tractors, ambulances, police vehicles, tankers, and trailers; and
the bus fleet numbers include the regular public transport vehicles, stage carriers, private
service vehicles, school buses, and mini buses. For all the cities, the 2wheeler population
dominates,whilerangingfrom0.8 millionfor Rajkotto3.0millionin Chennai.The2wheeler
fleetincludesmopeds,scooters,andmotorcycles,withamixof2strokeand4strokeengines.
The 4stroke engines are considered more fuel efficient and the manufacturing groups are
promotingthesemodels;withlittleinthewayofphasingoutthe2strokevehicles.Onaverage,
~3040%oftheinuse2wheelerfleetisestimatedtohave2strokeengines.
- 28 -
Figure 3.9: Vehicle fleet composition in the six cities

Pune
Chennai
14%
15%
15%
4%
1%
7%
74%
5%
0.4%
3%
0.4%
3%
79%
Ahmedabad
11%
Indore
75%
Surat
2%
3%
5%
Rajkot
10%
12%
1%
83%
8%
76%
2Ws
Trucks
0.1%
5%
80%
4%
5%
Buses
3Ws+Taxi
4Ws
Chennaibeingaportcity,thefreightmovementtoandfromthetwomajorportsissupported
withthemixofroadandrailtransportandtheheavydutyvehiclefraction(~50%)ofthetotal
truck fleet is the highest here. For other cities, the fractions are under 25% and largely
supportedbylightdutyvehicles.FormostcitiesinIndia,theheavydutyvehiclesarerestricted
fromenteringcitylimitsduringthedaytime.Thispolicywasintroducedprimarilytocutdown
diesel emissions and related health impacts due to chronic exposure during the daytime. An
unintendedconsequenceofthispolicyisanincreaseintheuseofthelightdutyvehiclesduring
thedaytimetomovethecargoin,out,andaroundthecities.
The public transport systems in the Tier 1 and Tier 2 cities are expanding. This was also
highlightedinthestudiesconductedbytheMinistryofUrbanDevelopmentofIndia,thatasthe
cities are expanding, the collective need for public transportation systems is further realized
and implemented. A summary of the data is presented inFigure 3.10. The big cities have at
leastdoubledtheiradministrativeboundariesinthelastdecade.This,combinedwithincreasing
incomes, has been the impetus for transport demand to increase exponentially. Figure 3.10,
top left panel, presents the relationship between the city population (on log scale) and city
travelserviceindex(definedasthepercentageofworktripsaccessibleinlessthan15minutes
oftraveltime).Themegacities(reddots)fairpoorlycomparedtothemediumsizecities(blue
dots)andthenthetertiarycities(turquoisecoloreddots).
- 29 -
Figure 3.10: Passenger travel statistics in India30

60
Raipur
Pondicherry
Chandigarh
90
80
Kochi
70
60
50
40
Nagpur
30
Bhubaneswar
Ahmedabad
Delhi Mumbai
Bangalore
Hyderabad
Kolkata
Chennai
20
10
Mumbai
Delhi
40
10
100
Chennai
Hubli
Shimla
10
Guwahati
Bhubaneswar
1000
% Share of Para-Transport (e.g., 3 Wheelers)
% Share of Non-Motorized Transport
Agra
10
Bikaner
Raipur
60
Shimla
Patna
Pondicherry
55
Amritsar
Madurai
Nagpur
Bhubaneswar
Agra
Kanpur
Triavndrum
Surat
Guwahati Hubli Bhopal
Varanasi
Chandigarh
Jaipur Ahmedabad
50
45
40
35
Delhi
Pune
Mumbai
Bangalore
Hyderabad
Chennai
Kolkata
30
25
Kochi
20
1
10
100
1000
Varanasi
14
Guwahati
12
Triavndrum
10
Chandigarh
Panaji
70
Kochi
50
40
30
20
10
Mumbai
Bangalore
Hyderabad
Bikaner
Pondicherry Amritsar
Jaipur
Bhubaneswar Raipur
Agra
Bhopal
Chandigarh
Nagpur
Kolkata
0
10
0.50
100
1000
Delhi
Mumbai
0.45
Pondicherry
Shimla
Bikaner
Bangalore Varanasi
Hyderabad
0.40
Amritsar
Madurai
Guwahati
Agra
Surat
Jaipur
Varanasi
Patna
Bhopal
Kanpur
Hubli
Nagpur
Bhubaneswar
Ahmedabad
Delhi
Bangalore
Mumbai
Kolkata
Hyderabad
Chennai
60
Chennai
Surat
Pune Ahmedabad
Kochi Kanpur Delhi
Population (in 100,000s)
Congestion Index
80
Patna
Madurai
Gangtok
Raipur
90
1000
16
100
100
70
Service Index
(% work trips accessible in < 15 mins)
Jaipur
Ahmedabad
Chandigarh
Bhopal
Madurai
Pune
Pondicherry Nagpur
Kanpur
Triavndrum
20
65
Bangalore
Hyderabad
30
0
1
Kolkata
50
Bikaner
Amritsar
Madurai
Kochi
Guwahati
Surat Pune
Triavndrum Agra
Jaipur
Patna
Varanasi
Bhopal
Hubli
Kanpur
Shimla
% Share of Public Transport
Service Index
(% work trips accessible in < 15 mins)
100
Pune
Kolkata
Kanpur
Chennai Guwahati
Bhubaneswar
Ahmedabad
Surat
Raipur
Nagpur
Jaipur
0.35
0.30
0.25
Pune
0.20
Kochi
Triavndrum
Patna
Hubli
GangtokBikaner
Bhopal
Amritsar Pondicherry
0.15
Shimla
0.10
Panaji
Madurai
Agra
0.05
0.00
0
20
40
60
80
% Share of Non-motorized Transport
20
40
60
80
% Share of Non-motorized Transport
Color code: red = mega cities; blue = secondary cities; turquoise = tertiary cities
30
Data is sourced from the report Traffic and Transportation Policies and Strategies in Urban Areas in India by
Ministry of Urban Development, Government of India, May, 2008 @
http://urbanindia.nic.in/moud/theministry/ministryofurbandevelopment/main.htm
- 30 -
Asthecitiesareexpandinggeographically,theneedformotorized(selforpublic)transportis
becomingimminent.ImportantmessagesinFigure3.10:
Ascitiesgrow,accesstotheworkplaceinlessthan15minutestraveltimedecreases
Ascitiesgrow,theshareofpublictransportintheformofbuses(percentofpassenger
trips)increase
Ascitiesgrow,theshareofnonmotorizedtransport(NMT)intheformofwalkingand
cycling(percentofpassengertrips)decreases
Ascitiesgrow,theshareofparatransit(including3wheelers)increased,butnotatthe
samerateastheothermodes.
Lowertheshareofnonmotorizedtransportinthecity,lowertheserviceindex(%trips
accessibleinlessthan15minutestraveltime)
LowertheshareofNMTinthecity,higherthecongestionindex,primarilydueincrease
inthepersonaltransport
The access to public transport is growing, but not enough to support the growth in travel
demand. Figure 3.10, top right panel, presents the share of passenger trips covered by the
publictransportagainstthepopulationinthecities.Theaccesstothepublictransportishighin
themegacities,andexpectedtogrowundertheJNNURMfunds31.However,delaysinthebus
manufacturingsectortosupplythenecessarynumberofbuses(currentlystandingat~70,000
buses) is hindering further promotion of public transport32. In India, the growth rate for the
motorvehicles(passengercarsand2wheelers)isapproximately10to12percent33.Whilethe
shareofpersonaltransportisgrowing,itisimportanttofocusontheNMT(walkingandcycling)
together,whichformsamajorportionofthepassengertrips,especiallytheshorttrips<3km.
An intervention that some cities are trying to implement is the Bus Rapid Transport (BRT)
systems.Anumberofcitiesacrosstheworldhavesuccessfullyimplementedthisdesignintothe
public transport systems with worlds best practices coming from Bogot, Colombia and
Curitiba, Brazil. Similar efforts have been attempted in Pune (with limited success) and
Ahmedabad.
TheYouthCommonwealthgameswerehostedinPuneandfortheevent,thecityundertooka
forayintoBRTsystemswiththeaimtoencouragetheuseofpublictransportation,inparticular
buses. Though the program was implemented with much enthusiasm, faulty design and
problemsinimplementationmeantthattheprojectdidnotturnoutasplanned.Thepilotwas
restrictedtoalimitedarea,enforcementoflanesforonlybusseswasinadequate,thedesignof
31
32
Jawaharlal Nehru National Urban Renewal Mission (JNNURM) @ http://jnnurm.nic.in/

Down to Earth, October, 2008, City bus: In demand, out of supply @
http://www.downtoearth.org.in/cover.asp?foldername=20081031&filename=news&sid=45&page=1&sec_id=9&p=1
Times of India, February 8th, 2009, BRTS dreams may go bust @

http://timesofindia.indiatimes.com/articleshow/msid-4096144,prtpage-1.cms
The Hindu, May 13th, 2009, Delhi Govt. faces cancellation of bus funding under JNNURM @
http://www.hindu.com/thehindu/holnus/004200905131452.htm
33
Times of India, May 12th, 2009, On growth track: Auto sales zoom 11% in April, 2009, report from SIAM
Society of Indian Automobile Manufacturer @
http://timesofindia.indiatimes.com/Car-sales-up-420-bikes-jump-1211-in-April/articleshow/4508229.cms
- 31 -
bus stops in the center of the road caused problems for commuters and in some cases, was
responsible for accidents, and the number of lowfloorbuses was much below the original
plannedfleet.Asaresultof,theBRTsystemisallbutdismantledinPuneCity34.Ahmedabad
launcheditsBRTsystemin200935.TheprojectwasapprovedinNovember,2006forthefirst
phaseof12.5kmfrom RTOto Pirana.Thenetworkisnowbeingexpandedwithanexpected
totallengthofBRTScorridorof~88kmbyMarch,2012.
3.5Construction
Thenumberofbrickkilnssurrounding
Figure 3.11: An ariel view of a brick kiln cluster
outside Chennai (Google Earth)
the vicinity of the six cities is
substantial with numbers ranging
from120forSuratto600forChennai
(Table3.1).Theclustersarepresented
intheFigure3.2asreddots.Thetype
of kilns also varies significantly
between the cities from scove or
clamp kilns to bulltrench kilns. The
former is an inefficient method of
baking bricks, which is the common
sight for Pune, Indore, Ahmedabad,
and Surat and the later is
comparatively an efficient method to
bake bricks with higher production
capacity.AnexampleofakilnclusterfromChennaiispresentedinFigure3.11.Thebulltrench
kilns are a common sight in this area and also commonly found around the cities of Delhi,
Varanasi,Patna,Kanpur,andothercitiesalongtheIndoGangeticplans.OfthesixcitiesRajkot
istheonlycitywithnoknownbrickkilnmanufacturingunits.
Inthebrickmanufacturingindustry,alongwiththeemissionsassociatedwithcoalandbiomass
burning at the kilns, emissions associated with the transfer of soil from the quarries to the
manufacturing units, and transfer of bricks from the manufacturing sites to the construction
sites,formasignificantportion.Theuseofthelightdutyandheavydutyvehiclesinandaround
thebrickkilnclustersishighlightedwhilespatiallyallocatingthetransportsectoremissionsto
the 1 km grids over the study domains. Around Pune, the total estimated brick production is
~1,800millionperyearwithanestimatedcoalconsumptionof0.27milliontons36.Thebrickkiln
cluster to the northwest of the city has ~400 active units. Of the six cities, largest brick
manufacturing clusters are located outside the Chennai district, counting ~600. The total
installedcapacityinthesixcitiesisestimatedat~3,000millionperyear.
34
India Together, 2007, Punes BRT stumbles at start @ http://www.indiatogether.org/2007/jan/eco-brtpune.htm

A Note on the BRT System in Ahmedabad , Mr. Sudhir Gota, CAI-Asia, Manila, Philippines
36
UNDP/GEF, 2007, Energy Efficiency Improvements in the Indian Brick Industry, project document
35
- 32 -
The pollution from the fossil and biomass fuel usage at the brick kilns is a growing problem
aroundthecities.Duringtheagriculturalharvestseason,thekilnsuseamixofbiomassandcoal
forheatingandintheoffseason,pasttheharvestseason,theheatingissupportedprimarilyby
coal.BurningoffossilandbiomassfuelsreleasesignificantamountsofSO2,NOx,PM,andCO2.
Inthevicinitiesoftheurbancenters,thePM,SO2,andNOxemissionshaveanimmediateaffect
on the ambient air and human health. The black Carbon (BC) emissions, a fraction of PM
emissions,haveaprominentrolein
Figure 3.12: An aerial view of quarry operations
theclimatescience.BC,ashortlived
along the Pune-Nashik highway
species, can help counteract the
impacts of growing GHGs and
support the climate policy dialogue
attheglobalscale37.Theuseofcoal
and
Besides the brick kilns, the stone
quarries are a common in these
cities (for example, ~11 square km
of area (black polygons to the
northeastofthePunecityinFigure
3.2) at Moshi near PuneNashik
highway. An aerial view of the
quarryoperationsispresentedinFigure3.12,whereonadailybasis~500truckloadsofstone,
blackboulder,andmurumareextractedandtransportedtovariouspartsofthedistrictandthe
state38.Thehealthrisksassociatedwithconstantexposuretothedustparticlesintheseareasis
understudy39.Inthisstudy,thequarrylocationsaremappedtoaccountfortheresuspensionof
dustduetocrushingandhandlingofrockandemissionsfromthetruckmovementandinsitu
dieselgeneratorusage.
3.6PowerPlants
Ofthesixcities,Chennai,Ahmedabad,andSurataretheonlycitieswithpowerplantswithin
thestudydomain.InChennai,electricitydemandissupportedbytwothermalpowerplants
theNorthChennaipowerstation(630MW)andtheEnnorepowerstation(450MW),bothof
which are located to the north of the Chennai district. In Ahmedabad, it is supported by two
thermalpowerstationsoneinthevicinityofthecityalongtheSabarmatiRiver,alsocalled
37
Climate Impacts of Black Carbon A Review and Policy Implications A report by the Princeton University.
Full report is available @ http://www.bioenergylists.org/sellersprincetonbc
38
Pune Mirror, February 6th, 2011, Quarrying Mafia at it Again
@ http://www.punemirror.in/article/2/201102062011020606193103549f862d/Quarrying-mafia-at-it-again.html
39
Express India, January 4th, 2008, Bastu Rege, coordinator of Santulan, an NGO, that conducts pashan shalas
(schools) for children at stone quarries says that these kids face constant health risks due to the environmental
pollutants. Rege, who is instrumental along with others in opening 70 such schools in Pune district with 2,500
children living at stone quarries, says that the children often suffer from skin diseases, sore eyes and gastroentritis
due to unhygienic water.
@ http://www.expressindia.com/latest-news/Children-in-stone-quarries-hit-by-lung-disease/257713/
- 33 -
Sabarmatipowerstation(Figure3.13)andabiggerpowerstationof~800MWinGandhiNagar.
The 400 MW Sabarmati Thermal Power Station in Ahmedabad is one of the oldest operating
power stations in the country. Total power generation capacity installed in Chennai and
Ahmedabadis~1,200MW,allofwhichissupportedbycoal.IncaseofSurat,majorityofthe
power generation is supported by cogeneration in the steel and petroleum refineries in the
Hajiraarea,withaninstalledcapacityof~400500MW.
Figure 3.13: An aerial view of power plants

Ennore PP (Chennai)
North Chennai PP (Chennai)
Sabarmati PP (Ahmedabad)
Gandhi Nagar PP (Ahmedabad)
- 34 -
Chapter4.0
EmissionsInventoryandDispersionModeling
TheSimpleInteractiveModelsforbetterairquality(SIMair)toolswereutilizedtoestablishthe
baseline emissions inventory for each city, followed by dispersion modeling to assess health
impactsandbaselineandselectinterventionsforsomeemissionsources40.TheSIMairfamily
oftoolsisdevelopedonanopensourceplatform,inaplugnplaymodetosupportemissions
and pollution analysis in the cities. The tools are supported by publicly available information
from all known sources of emissions (for example those available in Geographic Information
Systems(GIS)suchasroadmaps,populationmapsandindustriallocations).
Figure 4.1: Urban air pollution analysis steps
monitoring
Information
40
data collection
hot spots
apportionment
institutional
policy
Analysis
Options
emissions
dispersion
impacts
technical
economic
impacts
stakeholders
options
Dialogue
cost & benefits
prioritization
The details of the SIM-air program and the tools are available @ http://www.urbanemissions.info. Previous urban
applications of SIM-air tools include Hyderabad and Delhi (India); Bangkok (Thailand); Lagos (Nigeria);
Antananarivo (Madagascar); Shanghai and Shijiazhuang (China); Hanoi (Vietnam); Dhaka (Bangladesh); and
Ulaanbaatar (Mongolia).
Inthisstudy,theanalysisstepsundertakenarepresentedinFigure4.1,startingwith
Information most of the information utilized to establish the baseline emissions
inventoryfor2010isobtainedfromliteraturereview,correspondencewithlocalgroups
working on similar issues or specialized topics (like transport and health), and using
publicly available resources from national and international agencies. No additional
surveyswereconductedaspartofthisstudy.
Analysis the emissions inventory was developed using data collected from various
departments, which is then spatially distributed over the study domain at 1km x 1km
resolution using GIS based information on roads, population, activities (such as
industries, hotels, hospitals, apartment complexes, institutions, and markets), and
followedbydispersionmodelingtoconverttheemissionsintoambientconcentrations,
utilizingthemeteorologicalconditionspresentedinthepreviouschapter.Thegridcells
withconcentrationsexceedingthethresholdvalues,alongwithpopulationdatapergrid
cell,werethenutilizedtoevaluatethehealthimpacts.Furtherdetailsoneachofthese
stepsarepresentedinthefollowingsections.
Options for the select interventions (presented in Chapter 1), the emissions,
concentrations,andhealthimpactreductionswereevaluatedfollowingsamemethods
utilizedforbaselinecalculations.
Dialoguethebaselineandinterventionresultswillbedisseminated(tobedetermined)
tothecitystakeholders
4.1PreviousStudies
ForthecityofPune,anemissionsinventorywasdevelopedbytheAirQualityManagementCell,
an independent organization associated with the zonal pollution control board of Pune. This
work was carried out as part of the PREIS program41with technical support from the United
StatesEnvironmentalProtectionAgency(USEPA)toimproveairqualitymanagementpractices
in India. As part of this effort, significant training and field study activities have been
undertakenintheCityofPune.
FollowingthePREISstudy,in200607,MinistryofEnvironmentandForests(MoEF)andCentral
PollutionControlBoard(CPCB)ofIndia,carriedoutparticulatepollutionsourceapportionment
studyinsixcitiesDelhi,Kanpur,Chennai,Mumbai,Pune,andBangalore42.Aspartofthestudy,
an emissions inventory was also developed (and published in January, 2011). The emissions
estimatedunderforPuneaccountfor32.3tonsPM10perday,41.4tonsperdayofNOx,and7.1
tonsperdayofSO2.TheemissionsestimatedundertheCPCBstudyforChennaidistrictaccount
for 11.02 tons PM10 per day, 12.1 tons per day of NOx, and 1.3 tons per day of SO2. The low
lying sources like the road dust and vehicle exhaust tend to contribute more to the ambient
pollutioncomparedtotheelevatedsources,whichtendtotravellongerdistances.
41
Pune Regional Emissions Inventory Study (PREIS) @

http://www.unipune.ac.in/dept/science/environmental_science/es_webfiles/preis.htm
42
A summary report of the particulate pollution source apportionment studies carried out in Pune, Chennai, Kanpur,
Delhi, Mumbai, and Bangalore is available @ http://moef.nic.in
- 36 -
Figure 4.2: Contribution of major sources to emissions inventory (2006-07)

from CPCB/MoEF study
PUNE
NOx
PM10
Transport
18%
Industry
73%
Construt
4%
Domestic
7%
Transport
95%
Industry
1%
Road
Dust
62%
Domestic
2%
Others
8%
Industry
2%
Others
8%
Transport
13%
Domestic
6%
Others
1%
CHENNAI
NOx
PM10
SO2
Domestic
18%
Domestic
19%
Transport
14%
Road
Dust
73%
SO2
Industry
5%
Others
28%
Construt
9%
Domestic
1%
Industry
2%
Others
1%
Industr
11%
Others
1%
Transport
69%
Transport
49%
In CPCBs study, it is important to note the methodology employed in developing these

inventories. The inventory is for the baseyear of 200607 and represents only the sources in
themaincitydistrict.Thisdoesnotincludetheareassurroundingthemaindistrictwherethe
industrialactivityisgenerallylargerthantheindistrictactivities.Theinventorieswereprimarily
surveyedandestimatedforanareaof2kmx2kmaroundthemonitoringsiteselectedforthe
sourceapportionmentstudyandthenextrapolatedtothecitydistrictarea.Thiscouldleadto
somebiasinthefinalinventory,asthemodeledconcentrationswillnotaccountforthelong
range transport of emissions within the vicinity of the city districts. For example, in case of
Chennai,theretwolargecoalbasedthermalpowerplantswithagenerationcapacityof~1100
MWandanumberofpetrochemicalrefineriesandothermanufacturingwhichfalloutsidethe
cityadministrativedistrict,arenotincludedintheinventory.IncaseofPuneandChennai,the
inventorydoesnotaccountforthebrickkilnemissions.Inboththecities,theclustersare~20
kmawayfromthecitydistrictboundaries,thoughtheyareknowntocontributetoambientair
pollutioninthecitylimits.
- 37 -
4.2EmissionsInventoryfor2010
Inthisstudy,anemissionsinventoryisestablishedusingbottomupconsumptionpatternsfor
various sectors and using appropriate emission factors from the past and relevant literature.
TheemissionsinventorywasdevelopedforallthecriteriapollutantsincludingPM10,PM2.5,NOx,
SO2,CO,VOCsandCO2.Thesectorsincludedintheanalysisarevehicleexhaust,roaddust,
domestic solid fuel combustion (in the low income and high income groups), food kiosks,
generator usage in multiple venues (such as hospitals, hotels, markets, and apartment
complexes),industrialemissionsincludingthosefrombrickkilnsandrockquarries,construction
activitiesinthecity,andwasteburningalongtheroadsideandatthelandfills.
Table 4.1: Summary of 2010 emissions inventory for the six cities (numbers rounded)
City
PM2.5
PM10
SO2
NOx
CO
CO2
(tons)
(tons)
(tons)
(tons)
(tons)
(milliontons)
Pune
16,650
36,600
3,600
127,350
438,200
11.9
Chennai
26,000
56,400
15,100
268,200
857,500
30.6
Indore
9,900
17,900
2,550
146,600
263,300
8.6
Ahmedabad
17,500
35,100
10,500
175,300
510,150
22.4
Surat
12,000
19,950
3,350
146,450
371,400
11.6
Rajkot
7,750
14,000
2,200
91,750
236,700
7.4
Figure 4.3: Contribution of major sources to PM10 emissions inventory

PUNE
CHENNAI
INDORE
IND
6%
CON
7%
WB
3%
CON
6%
IND
2%
BK
6%
BK
15%
GS
3%
TR
18%
AHMEDABAD
GS
3%
IND
2%
PP
39%
TR
16%
CON
7%
WB
7%
PP
21%
BK
7%
IND
12%
DOM
7%
BK
7%
GS
3%
DOM
5%
QR
5%
RD
31%
SURAT
RD
22%
BK
7%
TR
31%
TR
15%
CON WB
2% 4%
GS
DOM 1%
11%
RD
36%
GS
2%
DOM
4%
RD
44%
DOM
7%
WB
5%
RAJKOT
CON
2%
IND
25%
WB
5%
CON
1%
WB
3%
GS
1%
DOM
3%
RD
34%
TR
30%
TR
26%
QR
11%
RD
30%
TR = transport; BK = brick kilns; GS = generator sets; DOM = domestic fuel combustion; CMQ = construction material
processing at quarries; RD = road dust; WB = open waste burning
- 38 -
Figure 4.4: Contribution of major sources to PM2.5 emissions inventory

PUNE
CHENNAI
INDORE
BK
26%
GS
5%
GS
DOM 4%
8%
DOM
14%
BK
10%
DOM
16%
IND
3%CON
4%
WB
7%
GS
2%
BK
11%
IND
4%
PP
18%
RD
15%
AHMEDABAD
BK
11%
GS
5%
DOM
9%
RD
8%
SURAT
INDCON
4% 1% WB
5%
RD
7%
DOM
10%
GS
3%
WB
9%
CON
4%
RAJKOT
BK
10%
DOM
4%
GS
1%
IND
36%
IND
20%
TR
42%
TR
40%
PP
31%
TR
27%
IND
3%
TR
34%
WB
5%
TR
27%
TR
47%
RD
12%
CON
4%
CON
WB 1%
RD
6%
8%
QR
6%
CON
RD WB 1%
8% 4%
Figure 4.5: Contribution of major sources to CO2 emissions inventory

PUNE
CHENNAI
INDORE
GS
5%
DOM
5%
TR
65%
TR
85%
BK
7%
DOM
9%
TR
53%
GS
8%
IND
17%
BK
7%
IND
11%
AHMEDABAD
GS
DOM 5%
6%
BK
3%
DOM
9%
PP
13%
SURAT
IND
18%
RAJKOT
DOM
3% GS
1%
TR
66%
TR
69%
DOM
7%
TR
35%
GS
BK 5%
3%
PP
33%
IND
29%
IND
16%
QR
1%
TR = transport; BK = brick kilns; GS = generator sets; DOM = domestic fuel combustion; CMQ = construction material
processing at quarries; RD = road dust; WB = open waste burning
- 39 -
Someinterestingnotesfromtheemissionsinventory
Among the fuel burning sources of PM, the transport sector is the dominant source,
especiallythedieselbasedtrucks
Overall,theresuspensionofdustduetoconstantvehicularmovementontheroadsisa
growingprobleminthecities.TheroaddustdominatesthecoarsefractionofPM(with
particle diameter between 2.5 and 10 micron meter) and thus its dominance in the
PM10emissionsinventorywithpercentagesrangingfrom22to44
All the cities, except for Rajkot are surrounded by brick kilns and their emissions
contribute 6 to 15 percent of the emissions and similar fractions to the ambient
pollution(presentedinthelatersections)
Among the size fractions, PM2.5 is considered more harmful than PM10. With PM2.5 as
thecontrollingpollutant,thedirectvehicleexhaustisthelargestcontributorandwith
PM10asthecontrollingpollutant,theroaddustisthelargestcontributor.
AmongthesilentcontributorstoPMandCO2emissions,wehavethedomesticcooking
andheatingemissions,especiallythelowincomegroups,outsidethecitydistrictareas,
where use of coal, biomass, and biofuels is at large; followed by the use of generator
sets within the city limits in the sectors of hotels, hospitals, institutions, apartment
complexes,andmarkets.
Onesourcewiththelargestuncertaintyintheemissionsinventoryisthewasteburning.
Due to lack of enough waste management programs, parts of the domestic waste is
burntandaccountingforPMandothercarcinogenicemissions.
Figure 4.6: Estimated percent passenger kilometers traveled
PUNE
CHENNAI
Passenger Kilometers Traveled

Walk
14%
MCs
23%

MCs
42%
Buses
23%
RAJKOT
Walk
11%
Cycle
20%
MCs
43%
Cars
8%
Taxis
1%
Cars
4%
Taxis
1%
3Ws
2%

Walk
13%
Cycle
20%
Buses
18%
Taxis
3%
SURAT

Walk
14%
3Ws
11%
Cycle
29%
MCs
22%
Cars
11%
3Ws
5% Buses
17%
AHMEDABAD
MCs
28%
Walk
19%
Cycle
13%
Cars
8%
Taxis
1%
Buses
27%
Walk
9%
Cycle
21%
3Ws
6%
INDORE
Cycle
17%
MCs
56%
Cars
7%
Taxis
Buses
0%
3%
3Ws
14%
- 40 -
Cars
10%
3Ws
4%
Thetransportsectoristhemajorcontributor. Inthisstudy,inventoryanalysisiscarried
out for the motorized incity passenger vehicles and the freight transport, covering the
currenttrendsinfivemodesoftransport(passengercars,motorcycles,3wheelers,buses,and
trucks) and estimated the energy consumption and emissions using the information existing
studies, the factors developed by ARAI as part of the CPCB source apportionment study, and
academic publications43. Table 3.2 presents a summary of the vehicles registered in the six
citiesandasummaryofthesharesofvehiclesinFigure3.7.
Thechanginggeographicalsettinginthecityandneedforbettermobilityisleadingthetraffic
demand and related congestion, reflected in the double digit growth rate in the number of
registrationsofcarsandmultiutilityvehicles,andmotorcycles(2wheelers).Whilethenumber
of 2wheelers, 3wheelers, and 4wheelers dominate the vehicle numbers, majority of the
populationisstillservedbythepublictransportand/orusethenonmotorizedtransport44.
Table 4.2: Summary of vehicular emission factors from CPCB/MoEF study45

Vehicle
Fuel
CO
HC
NOx
CO2
2Wheelers Petrol2Stroke
3.622.50
2.871.78
0.0280.011 25.96.5
2Wheelers Petrol4Stroke
1.330.85
0.550.19
0.2970.113 29.69.7
0.3180.150 64.97.8
3Wheelers Petrol
1.980.79
2.362.17
3Wheelers Diesel
8.5512.6
0.630.83
0.8900.510 145.224.8
3Wheelers CNG
0.840.21
1.161.27
0.3450.219 67.714.1
2.311.81
0.060.028
61.39.6
3Wheelers LPG
3.041.90
Cars
Petrol
2.911.67
0.360.28
0.3690.348 127.028.8
Cars
Diesel
0.520.33
0.190.07
0.5340.207 151.013.7
0.530.22
0.4260.375 141.09.3
Cars
CNG
0.500.40
Cars
LPG
4.752.87
0.540.43
0.3500.212 135.06.5
MUVs
Diesel
1.290.96
0.790.62
1.2200.829 213.037.7
1.630.55
2.5430.458 353.041.0
LCVs
Diesel
3.240.36
Buses
Diesel
8.404.86
1.101.03
11.133.572 752.0144.
Buses
CNG
3.720.00
3.750.00
6.2100.000 806.00.00
1.501.59
11.573.210 799.53.1
HCVs
Diesel
12.69.40
MUVs=multiutilityvehicles;LCVs=lightcommercialvehicles;HCVs=heavycommercialvehicles
43
PM
0.0540.019
0.0160.009
0.0480.036
0.5450.580
0.0660.072
0.4250.417
0.0070.006
0.1180.077
0.0030.002
0.0010.000
0.3610.242
0.7090.265
1.0800.725
0.000.00
1.6020.512
An analysis of the Emissions from India's Intercity and Intracity Road Transport is presented by the Clean Air
Initiative for Asian Cities (CAI-Asia). A draft report from May, 2009, is available @
http://www.cleanairnet.org/caiasia/1412/article-73353.html
A study of basic transport and air quality indicators was carried out by WRI/EMBARQ in 2009 @
http://www.embarq.org/en/book/export/html/427
ADB, 2006, Energy Efficiency and Climate Change considerations for on-road transport in Asia @
http://www.cleanairnet.org/caiasia/1412/articles-70656_finalreport.pdf
44
Times of India, May 12th, 2009, On growth track: Auto sales zoom 11% in April, 2009, report from SIAM
Society of Indian Automobile Manufacturer @
http://timesofindia.indiatimes.com/Car-sales-up-420-bikes-jump-1211-in-April/articleshow/4508229.cms
45
A detailed report on the emission factors for the Indian vehicle fleet is available
- 41 -
Figure 4.7: Estimated gridded vehicle exhaust emissions

PUNE (32km x 32km)
CHENNAI (44km x 44km)
18.7
18.6
18.55
Brick Kilns
Pimpri Chinchwad
PM10 Emissions
per square km
0.0 to 0.1
0.1 to 1.0
1.0 to 2.5
2.5 to 10.0
10.0 to 75.0
75.0 to 125.0
Airport
Power Plant
13.2
Industries
PM10 Emissions
per square km
13.15
13.1
Chennai Port
Brick Kilns
Industries
Race Course
13.05
NH-9
18.5
Airport
IIT
13
18.45
0.0 to 0.1
0.1 to 1.0
1.0 to 2.5
2.5 to 10.0
10.0 to 75.0
75.0 to 125.0
Bay of Bengal
oad
ss R
ypa
ne B
u
P
baiMum
13.25
18.65
12.95
Khadakwasla Dam
NH-4
18.4
12.9
80
73.7
73.75
73.8
73.85
73.9
73.95
80.05
80.1
80.15
80.2
80.25
80.3
80.35
80.4
74
INDORE (32km x 32km)
AHMEDABAD (44km x 44km)

23.25
Gandhi Nagar
22.8
23.2
22.75
PM10 Emissions
per square km
23.15
Airport
PM10 Emissions
per square km
0.0 to 0.1
0.1 to 1.0
1.0 to 2.5
2.5 to 10.0
10.0 to 75.0
75.0 to 125.0
23.1
NH59
NH59A
Bilavali Lake
22.65
0.0 to 0.1
0.1 to 1.0
1.0 to 2.5
2.5 to 10.0
10.0 to 75.0
75.0 to 125.0
23.05
23
IIM
22.6
Ring Road
22.7
Industries
NH3 Bypass
22.95
Mhow
22.55
22.9
75.7
75.75
75.8
75.85
75.9
75.95
72.35
72.4
SURAT (44km x 44km)
72.45
72.5
72.55
72.6
72.65
72.7
72.75
RAJKOT (24km x 24km)
21.4
22.4
21.35
PM10 Emissions
per square km
21.3
PM10 Emissions
per square km
22.35
0.0 to 0.1
0.1 to 1.0
1.0 to 2.5
2.5 to 10.0
10.0 to 75.0
75.0 to 125.0
21.25
Industrial Estates
21.2
0.0 to 0.1
0.1 to 1.0
1.0 to 2.5
2.5 to 10.0
10.0 to 75.0
75.0 to 125.0
22.3
Hajira
21.15
22.25
Brick Kilns
21.1
21.05
Arabian Sea
22.2
21
72.6
72.65
72.7
72.75
72.8
72.85
72.9
72.95
73
70.68
- 42 -
70.7
70.72
70.74
70.76
70.78
70.8
70.82
70.84
70.86
70.88
70.9
70.92
ThemethodologyemployedincalculatingtheemissionsispresentedinAnnex1.Theemissions
estimationresultsincludeaninventoryforthecriteriapollutantsPM,SO2,NOx,CO,VOCsand
CO2 (presented Table 4.1), percentage shares of passenger kilometers traveled (presented in
Figures4.6),percentagesharesbyvehicletypeandfuel.Theonroadvehicleexhaustemissions
arethendistributedtothegrids@1kmresolutionusingGISbaseddatabasesforthehighways,
mainroads,andarterialroads(presentedinFigure3.2).Thegriddedemissionsinventoryfrom
the transport sector is presented in Figure 4.7. For each vehicle type, the emissions are
weighted by road type to allow for appropriate spatial distribution of the emissions. For
example, the emissions from the truck movement are mostly distributed to highways,
comparedtothemotorcycles,whosemovementisexpectedmorealongthearterialroads,hot
spotlocations,industrialestates,andpopulatedareas.Theidlingemissionsduetocongestion
ontheroadsaredistributedtotheroads(weightedtotheroaddensityoverthegrids)anda
fraction over grids overlapping with hot spot locations like hotels, hospitals, markets,
apartmentcomplexes,andcinemas.
The road dust emissions are calculated for different road types highways, main roads,
and arterial roads, based on the estimated vehicle kilometers traveled. The methodology
utilizedfortheanalysisisbasedontheUSEPAAP42andcustomizedforthesiltloadingand
speed conditions in each city46. Of the total road dust emissions, major contribution comes
fromtrucks,buses,andcars,althoughthevehiclefleetisdominatedbythemotorcycles.Thisis
because of the empirical relationship between resuspension of dust with vehicle weight and
speed.
Itisimportanttonotethatresuspensionoftheroaddustissuppressedduringtherainyseason
andthisisaccountedforinthedispersionmodelingoftheemissionswheremeteorologyisan
important factor for dispersing and depositing pollutants. The monthly precipitation pattern
over the city is presented in Figure 3.4. The silt loading on the roads in most of the cities is
substantial enough to induce resuspension of the roads and higher the fleet size, like in the
caseofPuneandChennai,higheraretheemissions.IncaseofChennai,theportactivitiesalong
withalargeindustrialarea,whichaccountof~10,000heavydutyvehiclesmovinginandoutof
thecityaccountforthebulkoftheresuspendeddust.
Among industries emissions, brick kilns and the quarries dominate.Thegrowing

number of brick kilns and technology in place for firing bricks make this sector one of the
inefficientfuelusers.Theboomintheconstructionindustryisleadingtheofficialandunofficial
brickmanufacturingoperationsandaccountsforatotalestimatedbrickproductionof~3,000
millionperyearwithanestimatedcoalconsumptionof0.27milliontons47,alongwithbiomass
andotherunconventionalfuels.Mostofthebiomassutilizedcomesfromtheagriculturalwaste,
especiallyintheharvestseasons,giventhevicinityofthekilnclusterstoagriculturallands.
46
47
US EPA - 42 details @ http://www.epa.gov/ttn/chief/ap42/

UNDP/GEF, 2007, Energy Efficiency Improvements in the Indian Brick Industry, project document
- 43 -
The bottomup source apportionment results (based on emissions inventory) presented in

Figure 4.3 highlight the contributions from the brick kiln operations at ~15 percent in the
residentialareas.Theproximityofthebrickkilnstotheagriculturalareaisanimportantfeature
to note. This supports the availability of biomass (especially during the harvest season) to
support some of the clay brick burning operations, which is otherwise supported by coal. In
Pune,mostofthebrickbakingisconductedasapilewithnochimneytosupporttomovement
of the burning emissions whereas in Chennai the same process is supported with a chimney,
whichallowsforreleaseofemissionsathigheraltitude(andthusfartherdispersion).
The domestic sector emissions include the low income and high income group
fuelusage.TheseemissionsarecalculatedusingthegriddedpopulationdensityfromSEDAC48
calledGRUMPdatabase(GlobalRuralUrbanMappingProject)availableat30sec(~500meters
resolution)fortheworld.ThepopulationdensitymapsareoverlaidwithdatafromtheCensus
ofIndiatoestimatethesolidandliquidfuelusageinthelowandhighincomegroupsforthe
domains presented49. In most cities, LPG is the dominant energy source for cooking. The
amountofusagevaries,wherea15kgLPGcylinder(standardsizeacrossthecountry)lasts~25
days for an average high income household with 4 members to ~50 days for a low income
groups(amixofsolidfuelsarealsoinuse).Incaseoffoodkiosks,whicharecommoninthe
marketsandroadside,thesecylinderslastforless~10days.
Theinventorytotalsforthedomesticsectorincludetheemissionsfromresidentialusage,food
kiosks, and possible fuel usage for heating (especially in the winter season). The domestic
sectorisestimatedtoaccountfor~820percentoftheannualtotalPM10emissionsamongthe
sixcities.Theseemissionsaredistributedoverthegridsinusingthepopulationdensitymaps.
The high income group emissions are distributed over the populated areas (with the highest
density).
The diesel generator sets are a common sight for most parts of the city and a
significant source of criteria pollutant emissions and GHGs. A database of hotspots was
estimated using population databases (gridded and district level), which is later used to
estimatetheemissions.Theaverageemissionratesusedfordieselgeneratorusageutilizedin
thisstudyare60tons/PJofNOx,63tons/PJofPM10,and1,256tons/PJofCOand~2.56kg/litof
dieselconsumed.Totalemissionsfromthedieselgeneratorsetsaccountfor~8milliontonsof
CO2forthesixcities.
48
Gridded population of the world by SEDAC @ http://sedac.ciesin.columbia.edu/gpw/

The Census of India data is available in the GIS format for all of India at State and Tahsil level and disaggregated
into the income groups, fractions of solid and liquid fuel usage
49
- 44 -
0.2
0.5
0
0
The flight statistics for each city are obtained from the online resources @ http://www.flightstats.com
This does not include private or chattered flights
50
- 45 10 AM to 11 AM
6 PM to 7 PM
7 PM to 8 PM
8 PM to 9 PM
8 PM to 9 PM
9 PM to 10 PM
10 PM to 11 PM
11 PM to 12 AM
8 PM to 9 PM
9 PM to 10 PM
10 PM to 11 PM
11 PM to 12 AM
11 PM to 12 AM
10 PM to 11 PM
9 PM to 10 PM
3 PM to 4 PM
2 PM to 3 PM
1 PM to 2 PM
12 PM to 1 PM
11 AM to 12 PM
10 AM to 11 AM
9 AM to 10 AM
8 AM to 9 AM
7 AM to 8 AM
6 AM to 7 AM
5 AM to 6 AM
4 AM to 5 AM
3 AM to 4 AM
2 AM to 3 AM
1 AM to 2 AM
12 AM to 1 AM
11 PM to 12 AM
10 PM to 11 PM
9 PM to 10 PM
8 PM to 9 PM
7 PM to 8 PM
6 PM to 7 PM
5 PM to 6 PM
4 PM to 5 PM
3 PM to 4 PM
2 PM to 3 PM
1 PM to 2 PM
12 PM to 1 PM
11 AM to 12 PM
10 AM to 11 AM
9 AM to 10 AM
8 AM to 9 AM
7 AM to 8 AM
6 AM to 7 AM
5 AM to 6 AM
4 AM to 5 AM
7 PM to 8 PM
7 PM to 8 PM
5 PM to 6 PM
1.5
6 PM to 7 PM
0.6
6 PM to 7 PM
RAJKOT
4 PM to 5 PM
2.5
5 PM to 6 PM
5 PM to 6 PM
4 PM to 5 PM
4 PM to 5 PM
3 PM to 4 PM
10
3 PM to 4 PM
AHMEDABAD
2 PM to 3 PM
1 PM to 2 PM
12 PM to 1 PM
11 AM to 12 PM
10 AM to 11 AM
9 AM to 10 AM
8 AM to 9 AM
12
2 PM to 3 PM
1 PM to 2 PM
12 PM to 1 PM
11 AM to 12 PM
SURAT
9 AM to 10 AM
8 AM to 9 AM
7 AM to 8 AM
6 AM to 7 AM
5 AM to 6 AM
4 AM to 5 AM
3 AM to 4 AM
INDORE
7 AM to 8 AM
6 AM to 7 AM
0.4
5 AM to 6 AM
4 AM to 5 AM
0.8
3 AM to 4 AM
1.2
0
2 AM to 3 AM
3 AM to 4 AM
20
2 AM to 3 AM
0
2 AM to 3 AM
1 AM to 2 AM
10
1 AM to 2 AM
12 AM to 1 AM
12 AM to 1 AM
25
10
1 AM to 2 AM
11 PM to 12 AM
10 PM to 11 PM
9 PM to 10 PM
8 PM to 9 PM
7 PM to 8 PM
6 PM to 7 PM
5 PM to 6 PM
4 PM to 5 PM
3 PM to 4 PM
2 PM to 3 PM
1 PM to 2 PM
12 PM to 1 PM
11 AM to 12 PM
10 AM to 11 AM
9 AM to 10 AM
8 AM to 9 AM
7 AM to 8 AM
6 AM to 7 AM
5 AM to 6 AM
4 AM to 5 AM
3 AM to 4 AM
2 AM to 3 AM
1 AM to 2 AM
12 AM to 1 AM
12
12 AM to 1 AM
11 PM to 12 AM
10 PM to 11 PM
9 PM to 10 PM
8 PM to 9 PM
7 PM to 8 PM
6 PM to 7 PM
5 PM to 6 PM
4 PM to 5 PM
3 PM to 4 PM
2 PM to 3 PM
1 PM to 2 PM
12 PM to 1 PM
11 AM to 12 PM
10 AM to 11 AM
9 AM to 10 AM
8 AM to 9 AM
7 AM to 8 AM
6 AM to 7 AM
5 AM to 6 AM
4 AM to 5 AM
3 AM to 4 AM
2 AM to 3 AM
1 AM to 2 AM
9
8
7
6
5
4
3
2
1
0
12 AM to 1 AM
Besides the regular passenger transport statistics to evaluate the emissions from cars, buses,
trucks, 2wheelers, and3wheelers, statistics for the aviation sector were collected for
themainairportineachcity50.AsummaryoftheflightstatisticsbyhourispresentedinFigure
4.8.Ofthesixcities,Chennaiisthelargestinternationalairportwithanaveragedepartureand
arrival of 320 flights, followed by Pune at 120 and the lowest is for the city of Surat with 2
flightsperdaytoDelhiandback.
Figure 4.8: Daily number of flights at Airports
PUNE
CHENNAI
15
The emissions inventory for this sector includes take offs and landing, idling due to the
passengervehiclesatthedepartureandarrivallounges,andthebuses(basedondiesel)usedto
shuttlepassengersfromtheterminaltotheflightsandviceversa.BreakupsforonlyPM2.5and
CO2arepresentedinthisdocument.Puneisalsoahubformilitarytrainingandoperationsand
aportionofthecityisdesignatedascantonmentarea,whichisnotincludedintheanalysisfor
lackofinformationtosupportsuchcalculations.
The power plantsarethemajorsourceofenergyconsumptionandemissionsforthreeof
thesecitiesChennai,Ahmedabad,andSurat.Foreachofthepowerplants,theexactlocations
ofthestackswasobtained,alongwithstatisticsofcoalconsumptionandashproductionsrates,
andthusallocatingthepointemissionsdirectlytothelocation.
4.3DispersionModelingof2010BaselineEmissionsInventory
Typically, the emissions are released from (a individual point or an area) source, which after
entering the atmosphere, depending on the meteorological conditions like wind speed, wind
direction, pressure, temperature, moisture content, etc., interacts with other pollutants, and
dependingonthelocalcanopy,eitherdepositsonasurface(asdryorwet)orlingersintheair
intheformofpollution,whichwebreathe51.
Figure 4.9: Physics of atmospheric dispersion

The importance of emissions and
dispersion modeling and the effect of
long range transport of various
Emissions
pollutants are studied extensively at
Dispersion
urban, regional, national, and inter
continental levels 52 . Similar to the
emissions baseline development, the
Wet
dispersion modeling is an intense
Deposition
exercise, which requires both
computational power and data
Dry
Deposition
assimilation. Several modeling systems
areavailable,withvaryingcapacityand
complexitytoaddressthephysicalandchemicalaspectsofatmospherictransportofpollutant.
Inthisstudy,theAtmosphericTransportModelingSystem(ATMoS)dispersionmodelisutilized
for modeling the air quality using local specific meteorological data 53 . This model was
previouslyutilizedfornumerousairpollutionmodelingstudiesinAsia,includingtheRAINSand
GAINS integrated modeling system, sulfur pollution control in Asia, PM pollution analysis for
cities.ThedetailsofthemodelformulationandtheinputdataarepresentedAnnex 2.0.The
modelallowsformultipollutantanalysisinwhicheachoftheprimaryemissionsaremodeled
separately due to differences in their physical and chemical characteristics and aggregatedat
the end. This includes SO2 to sulfates and NOx to nitrates54. The total PM10 concentrations
comprise of all the subfractions (PMcoarse+PMfine+SO4+NO3) while the total PM2.5 comprise of
the finer subfractions (PMfine+SO4+NO3), thus providing a multipollutant aspect to the PM
pollution.TheestimatedannualaverageconcentrationsofPM10arepresentedinFigure 4.10
and PM2.5 are presented in Figure 4.11. The smaller bold black box highlights the most
populatedareasofthedomain(themaincitydistrict).
51
Refer to the books Atmospheric Chemistry and Physics by Seinfied and Pandis, and Fundamentals of
Atmospheric Modeling by Mark Jacobson, which provide detailed discussion on the air pollution modeling and
involved steps
52
Examples of Urban, Regional, and Global scale atmospheris dispersion modeling exercises
MEGAPOLI Study @ http://megapoli.dmi.dk/
Task Force for Hemispheric Transport of Air Pollution @ http://www.htap.org/
Better Air Quality for North America @ http://www.narsto.org/
European Monitoring and Evaluation Programme @ http://www.emep.int/
53
For details on the Atmospheric Transport Modeling System (ATMoS) visit @ http://www.urbanemissions.info
54
These rates are specified in the model input file and user has an option to either use the default value or change it
accordingly. No chemical transformation is applied to direct PM emissions.
- 46 -
Figure 4.10: Modeled Annual Average PM10 Concentrations

Pune, Maharashtra
Chennai, Tamilnadu
18.7
g/m3
Pimpri Chinchwad
ass
Byp
une
ai-P
b
m
Mu
Brick Kilns
18.6
18.55
160
Power Plant
160
13.2
Industries
140
Airport
120
140
120
13.15
100
100
13.1
80
Chennai Port
Brick Kilns
NH-9
18.5
60
13.05
40
Airport
20
IIT
13
10
18.45
80
Industries
60
Race Course
Bay of Bengal
d
Roa
13.25
18.65
12.95
Khadakwasla Dam
40
20
10
0
NH-4
18.4
12.9
73.7
73.75
73.8
73.85
73.9
73.95
74
80
80.05
80.1
80.15
80.2
80.25
80.3
80.35
80.4
Ahmedabad, Gujarat
23.25
Gandhi Nagar
g/m3
g/m
22.8
23.2
160
160
22.75
23.15
140
140
22.7
NH59
NH59A
23.1
100
100
80
Bilavali Lake
23.05
60
22.65
40
IIM
22.6
120
120
Ring Road
Airport
23
80
60
Industries
40
20
NH3 Bypass
20
22.95
10
Mhow
10
22.9
22.55
75.7
75.75
75.8
75.85
75.9
72.35
75.95
72.4
72.45
Surat, Gujarat
72.5
72.55
72.6
72.65
72.7
72.75
Rajkot Gujarat
21.4
g/m3
22.4
g/m3
21.35
160
21.3
140
160
22.35
140
120
21.25
120
100
Industrial Estates
21.2
80
80
60
Hajira
21.15
100
22.3
60
40
20
Brick Kilns
21.1
40
20
22.25
10
10
21.05
Arabian Sea
22.2
21
72.6
72.65
72.7
72.75
72.8
72.85
72.9
72.95
73
70.68
- 47 -
70.7
70.72
70.74
70.76
70.78
70.8
70.82
70.84
70.86
70.88
70.9
70.92
Figure 4.11: Modeled Annual Average PM2.5 Concentrations in

Pune, Maharashtra
Chennai, Tamilnadu
18.7
g/m3
Pimpri Chinchwad
ass
Byp
une
ai-P
b
m
Mu
Brick Kilns
18.6
18.55
120
Power Plant
120
13.2
Industries
100
Airport
80
100
80
13.15
60
60
13.1
50
Chennai Port
Brick Kilns
NH-9
18.5
40
13.05
30
Airport
20
IIT
13
10
18.45
50
Industries
40
Race Course
Bay of Bengal
d
Roa
13.25
18.65
12.95
Khadakwasla Dam
30
20
10
0
NH-4
18.4
12.9
73.7
73.75
73.8
73.85
73.9
73.95
74
80
80.05
80.1
80.15
80.2
80.25
80.3
80.35
80.4
Ahmedabad, Gujarat
23.25
Gandhi Nagar
g/m3
g/m
22.8
23.2
120
120
22.75
23.15
100
100
NH59
NH59A
23.1
60
60
50
Bilavali Lake
23.05
40
22.65
30
IIM
22.6
80
80
22.7
Ring Road
Airport
23
50
40
Industries
30
20
NH3 Bypass
20
10
Mhow
22.95
10
22.9
22.55
75.7
75.75
75.8
75.85
75.9
72.35
75.95
72.4
72.45
Surat, Gujarat
72.5
72.55
72.6
72.65
72.7
72.75
Rajkot Gujarat
21.4
g/m3
22.4
g/m3
21.35
120
21.3
100
120
22.35
100
80
21.25
80
60
Industrial Estates
21.2
50
50
40
Hajira
21.15
60
22.3
40
30
20
Brick Kilns
21.1
30
20
22.25
10
10
21.05
Arabian Sea
22.2
21
72.6
72.65
72.7
72.75
72.8
72.85
72.9
72.95
73
70.68
- 48 -
70.7
70.72
70.74
70.76
70.78
70.8
70.82
70.84
70.86
70.88
70.9
70.92
AsummaryofthemodeledaveragePM10concentrationsfromFigure 4.10andthemeasured
concentrations presented in Figure 3.8 are presented in Table 4.3 and the ranges are
comparable.Itisimportanttonotethatmeasurementsarepointvaluesreflectingthepollution
from the sources surrounding the sampling location and are usually representative of the
sourceswithin~23kmradius.Whilethedatafromthestationsislimited,theseareveryuseful
comparisons.Themeasurementdataisusedtocalibratethemodelingresultsandthussupport
furtherwhatifemissionsscenarioanalysis.
Table 4.3: Summary of the 2010 PM10 emissions inventory, modeled, and measured
concentrations
City
Domain size
(km x km)
Total Annual PM10

Emissions (tons/yr)
Pune
Chennai
Indore
Ahmedabad
Surat
Rajkot
32 x 32
44 x 44
32 x 32
44 x 44
44 x 44
24 x 24
36,600
56,400
18,100
35,100
19,900
14,000
Range of PM10 concentrations

(g/m3)
Modeled
Measured55
80-140
60-160
60-100
60-120
80-150
60-170
60-140
80-100
50-120
75-100
50-140
80-120
DuetolimitedinformationonthePM2.5monitoringdata,therearenocomparisonsavailableat
thistime.ThePM2.5fractions(inemissions)arevettedwiththeemissionsfactorspertinentto
the sectors and the dispersion modeling is conducted in a manner similar to PM10, with
differentphysicalandchemicalcharacteristics.Asandwhen,moremonitoringdataisavailable
from these cities, the PM2.5 results will be revisited for calibration and further analysis. In
Figure 4.11, most of the main city district areas exceed the annual average standard of 40
g/m3forPM2.5concentrations.
The physical location and topography of the city plays a crucial role for emissions and
concentrationchangespresentedinTable4.3.Forexample,thecitiesofChennaiandSuratare
ideallylocatednexttotheBayofBengalSeaandArabianSearespectively,whichprovidesthem
withanextremebufferzonetodispersealargeamountofthepollutioncomingfromthecity
sources,especiallythelargescaleindustries.Forexample,incaseoftheChennaicity,though
the estimated annual emissions are the highest of the six cities, the concentrations modeled
and measured are in the same range of a city half its size in geography and emissions (like
Rajkot). This can be attributed to the landsea breezes, which are very strong in Chennai,
evident in the meteorological fields presented in Figure 3.3, with Westerly winds (with a
significantamountoftimeabove10m/sec).Thesewindsarestrongenoughtodisperseanyof
thepollutiontothesea,fromthetwolargepowerplantsandthosefromtheindustrialestate
includingpetroleumrefineriesandalargeportandthusreducingtheloadoftheseemissions
on local pollution. Similarly, in Surat, though the existence of the largest industrial area
(includingsteelplants,textiles,refineries,andsmallscalepowerplants),thepollutionisdriven
totheseabythestrongEasterlywinds(Figure3.3)andthisreducingthepollutionloadofthese
55
The measured concentrations are obtained from the CPCB and State Pollution Control Boards. The ranges
represent highs and lows of the daily average PM10 concentrations over each month (presented in Figure 3.8).
- 49 -
emissionsonlocalpollution.ThetopographyplaysadifferentroleincaseofPune,wherethe
valleys restrict the movement of the pollution at higher altitudes and then thus creating
atmosphereforhigherconcentrationsformostpartoftheyear.
4.4PercentageSectoralContributions
Thepercentagecontributionsofvarioussectorstothemodeledannualaverageconcentrations
provide an interesting background for urban air quality management. These estimates are
dependentonthespatialdistributionofemissionsfromeachofthesectors.Incaseofindustrial
emissions,specificlocationinformationwasgatheredforknownindustries,brickkilns,quarries,
andpowerplants,whoseemissionsarethenallocateddirectlytotherespective1kmx1kmgrid
(presentedinFigure3.2).Thespatialdistributionofthevehicularandroaddustemissionsalong
with road density (for highways and arterial roads)56is utilized for spatially gridding the
emissionsandthusobtainingthecontribution distributions.Thesedatabasesweredeveloped
on a GIS platform for easy viewing, understanding, and studying the spatial patterns (vehicle
exhaustdistributionpatternsarepresentedinFigure4.7).Inthedispersionmodel,provisionis
madetoevaluatethecontributionofeachsectortothetotalambientpollution(toPM)based
ontheemissionsinventory.Asummaryoftherangeofestimatedpercentcontributionstothe
annual average PM10 concentrations is presented in Table 4.4. Note that the ranges for each
sectorareforthemaincitydistrictareaandthespatialspreadofthesecontributionsinclude
pollutionfromsourcesinsideandoutsideofthecitydistrictlimits
Table 4.4: Summary of percent contributions to annual average PM10 concentrations

City
Vehicle exhaust
Road dust
Industries (+kilns)
Power plants
Domestic
Pune
8-16
30-50
10-30
8-20
Chennai
10-20
10-30
5-30
0-5
15-40
Indore
15-25
15-25
10-30
20-40
Ahmedabad
10-20
10-20
10-50
0-15
5-10
Surat
10-30
10-30
10-30
5-15
Rajkot
10-40
10-30
10-70
0-10
In 200607, MoEF and CPCB conducted particulate pollution source apportionment studies in
Delhi, Mumbai, Pune, Chennai, Bangalore, and Kanpur. Details of the source apportionment
results and the emissions inventories were published in January 201157. In this study, we
overlap with the cities of Pune and Chennai and though both the studies aim at establishing
sourcecontributionstoambientpollution,therearemajordifferencesintheapproaches.The
studyconductedbyMoEF/CPCBisatopdownsourceapportionment(theirresultsforthecities
of Pune and Chennai are summarized in Figure 4.13), which is based on monitoring and
receptormodeling,whilethisstudyisabottomupsourceapportionment(resultsforthecities
of Pune and Chennai are presented in Figure 4.14), which is based on activity levels, fuel
consumption,andsourcemodeling.
56
Previous studies include those surveyed and presented as part of the JNNURM city development reports, available
@ http://jnnurm.nic.in/nurmudweb/missioncities.htm for the cities Delhi, Pune, Chennai, Ahmedabad, Surat, and
Rajkot,
57
A summary report of the particulate pollution source apportionment studies carried out in Pune, Chennai, Kanpur,
Delhi, Mumbai, and Bangalore is available @ http://moef.nic.in
- 50 -
While both represent source contributions to ambient air pollution, it is important to

understand the differences and advantages of these results (also see Annex 3.0), which are
establishedusingtwodifferentconcepts.
In a source apportionment study (carried out primarily for PM pollution) where the
concentrations of PM are measured over a period of time, the samples of which are
chemically analyzed in a lab for biomarkers for various sources and statistically
estimate the contribution of various sources (results in Figure 4.13 from MoEF/CPCB
study).
Inthelater,anemissionsinventoryisdevelopedusingconsumptiondata,multipliedby
anappropriateemissionfactor.Theemissionsinventoriesarethenprocessedthrougha
dispersion model in order to estimate the contributions to ambient pollution, which
wouldbeanappropriatenumbertocompare(resultspresentedinFigure4.14);instead
of comparing the percentages from an emissions inventory to those from a source
apportionmentstudy.
Figure 4.12: Schematics for estimating sectoral contributions58

Source
Profiles
Receptor
Modeling
Monitoring
Concentrations
Pollution
Control
Monitoring
Contributions
TopDown
Approach
(MoEF/CPCB)
Economic
Technical
Policy
Evaluation
Dispersion
Modeling
Emissions
Impacts
Costs&
Benefits
Pollution
Control
PM,SO2,
NOx,CO,
Ozone,VOC
Concentrations
BottomUp
Approach
(Thisstudy)
Economic
Technical
Policy
Contributions
Evaluation
Impacts
Costs&
Benefits
Decisions
Decisions
Both the processes have their differences and we should note the limitations while
interpreting the results for policy decisions. For example, The percent contributions
established using topdown source apportionment study are specific to the vicinity of
thesamplinglocationusuallyrepresentthesourcecontributionsfroma23kmradius,
andshouldnotbegeneralizedfortheentirecity.
Giventheconfidencelevelsintheemissionsinventory,meteorology,andthedispersion
modeling,astudy(likethisone)canpresentthepossible contributionsoveranentire
city.
58
A detailed overview of the techniques to source apportionment and a summary of results from studies across the
world are presented in, Johnson and Guttikunda et al., 2011, Tools for Improving Air Quality Management - A
Review of Top-down Source Apportionment Techniques and Their Application in Developing Countries The
World Bank, Washington D.C., USA.
Full document is available for download @ http://www.esmap.org/esmap/node/1159
- 51 -
Whileaccountingforsourcesaroundasamplingsite,wehavetodifferentiatebetween
thegroundlevelsourcessuchasvehicleexhaust,roaddust,domesticfuelburning,etc.,
whichtendtoimpactthegroundlevelconcentrationsmorethantheelevatedsources
likepowerplantsandindustrieswherethestackheightallowsforanelevatedreleaseof
emissionsanddispersemore.
Thestudiescanalsocomplimenteachotherinthelongrun.(a)forthetopdownstudy,
whensourceprofilesarebeinggenerated,thesamecanbeusedtoverifytheemission
factorforvarioussources,whichcanbeusedforemissionsinventorydevelopment(b)
followingdispersionmodeling,wecanestablishthehotspotlocationsaroundthecity,
which can be used for sampling and source apportionment (c) the receptor modeling
canprovideinsightsintomissingsources,whichcanbeinvestigatedwhiledevelopingan
emissionsinventory
Figure 4.13: Contribution of major sources to PM10 pollution based on

MoEF/CPCB source apportionment studies (2006-07)
PUNE
Curbside
Residential
Domestic
11%
DGSets
4% Transport
10%
Brick
Kilns
15%
Waste
Burning
3%
Road
Dust
57%
Road
Dust
64%
Waste
Burning
9%
Road
Dust
49%
CHENNAI
Curbside
Industrial
Transport
46%
Transport
35%
Bakeries
5%
Bakeries
4%
Others
8%
Road
Dust
6%
Brick
Kilns
28%
Domestic
8%
Waste
Burning
4%
Transport
47%
Domestic
21%
DGSets Transport
2%
4%
Brick
Kilns
7%
Residential
DGSets
14%
Industrial
Domestic
13%
DGSets
3%
Transport
9%
Others
13%
DGSets
16%
Domestic
4%
Bakeries
4%
Road
Dust
27%
Others
11%
DGSets
14%
Domestic
8%
Road
Dust
17%
From the source apportionment for the Pune city (Figure 4.18), it is evident that the major
contributortothepollutionproblemsisresuspensionofroaddustduetoconstantvehicular
movementandforChennai,thevehicularexhaustemissions.Themajordifferencebetweenthe
two cities is the meteorology and geography. In case of Chennai, high moisture content and
local landsea breezes keep the resuspension of dust at bay compared to those observed in
- 52 -
Pune.ThisisalsoreflectedinthemodeledestimatedpresentedFigure4.10andFigure4.11for
PM10 and PM2.5 respectively. The modeled source contributions following the procedures in
Figure 4.12 are presented in Figure 4.14 for the cities of Pune and Chennai. In the figures,
several regions are selected across the city for which the concentrations (presented in the
figure)andsourcecontributionsareaveraged.
Figure 4.14: Modeled 2010 PM10 source contributions averaged over select regions
across the city
59 g/m3
(centraldistrict)
WST
4%
CON
5%
BK
37%
18.65
QR
GNIND
WST 0%
3%
DOM2%
18.6
6%
BK
61%
DST
20%
BK
13%
QR
2%
32
BK
3%
g/m3
18.55
140
120
100
VEH
15%
80
NH-9
99 g/m3
DST
40%
WST
4%
73.8
73.85
73.9
73.95
Brick Kilns
GNIND
5%
74
44 g/m3
QR
1%
VEH
12%
CON
5%
100
80
70
IND
12%
VEH
16%
60
Brick Kilns
Chennai Port
40
13.05
Airport
IIT
13
PP OTH CON
5% 2% 5%
50
12.95
30
10
BK
6%
DOM
10%
DST
27%
20
WST
7%
GEN
6%
12.9
80
80.05
80.1
80.15
80.2
80.25
80.3
80.35
80.4
40mg/m3
BK
21%
GEN
11%
Power Plant
Industries
13.15
73mg/m3
WST
8%
g/m
13.25
BK
54%
VEH
10% OTH
PP
CON
0%4%
0%
DOM
12%
10
NH-4
73.75
DOM
11%
Industries
73.7
(centraldistrict)
BK
13%
13.1
Khadakwasla Dam
18.4
DST
16%
PP
34%
DST
24%
40
20
18.45
CON
10%
60
Race Course
PP OTHCON
6% 1% 4%
13.2
WST
4%
GNIND
QR
3%
BK
2%
7%
OTH
1%
76mg/m3
VEH
14%
IND
GEN
WST
7%
1%
DOM3%
5%
DOM
15%
160
CON
8%
VEH
9%
IND
22%
VEH
PPCON
9% OTH
1%
0%4%
Pimpri Chinchwad
Airport
DST
19%
BK
60%
DST
15%
DST
44%
18.5
VEH
5%CON
3%
CON
8%
18.7
ad
ss Ro
Bypa
une
bai-P
Mum
VEH
10% CON
4%
VEH
8%
GNIND
9%
VEH
13%
BK
4%
DST
14%
WST
7% GEN
5%
DOM
9%
IND
19%
DOM
6%
89mg/m3
IND
GEN
WST3%
1%
3%
DOM
4%
QR
32%
WST
4%
QR
2%
DST
26%
GEN
WST
1%
4%
56mg/m3
DOM
13%
GNIND
7%
37mg/m3
CHENNAI
WST
3%GNIND
3%
DST
31%
DST
41%
68 g/m3
DOM
10%
DOM
12%
111g/m3
Bay of Bengal
PUNE
IND
8%
VEH
18%
PP
OTH CON
0%
4% 5%
BK
15%
When comparing the results with Figure 4.13, we should consider only the shares for the
centraldistricts(markedinFigure4.14).Majordifferencesbetweenthetwostudieswhichwe
shouldconsiderwhilecomparingthetworesultsare
ThebaseyearforanalysisinMoEF/CPCBstudy(Figure4.13)is200607whilethisstudy
isconductedforthebaseyear2010(Figure4.14).Duringthisperiod,thesecitieshave
experiencedasignificantgrowthinconsumptionpatterns,numberofindustries,public
and private ventures (especially in Pune, associated with the InformationTechnology
sector), vehicle ownership, and infrastructure (such as roads and residential
construction).Theinventoriesdevelopedunderstudyreflectthesechangesandthough
the shares are qualitatively comparable, there are significant quantitative differences
forthesamereasons.
Whiletherewasanimprovementinthefuelqualityused(nowsetatEuroIVstandards
for major cities) and more stringent emission standards for new vehicles (now set at
EuroIIIformajorcities),thereanincreaseintheemissionsfromthetransportsector,
from2006to2010,becauseofanincreaseinthenumberofvehicles.Notethatthese
higherstandardsapplyonlytonewervehicleswhichformafractionoftheoverallfleet.
Themodelingdomainsinthetwostudiesaresignificantlydifferent.Whileanemissions
inventory and source apportionment study was conducted covering only the main
districtareasforboththecities,thisstudywasconductedforalargerarea(forallcities),
thus accounting for longrange transport of elevated sources, such as brick kilns and
powerplants,whicharelocatedatleast20kmawayfromthecentraldistrict.
- 53 -
The advantage of this methodology is to be able to account for sources and their
contributionsforawiderareaofconsideration(asshowninFigure4.14),thusallowing
foraholisticpollutioncontrolplan,specifictovariouslocations.
ThevehiclecontributionsarehighlightedinFigure4.13,mainlybecauseofthesampling
locations and their proximity to the roads. The low lying sources like road dust and
vehicle exhaust tend to contribute more to the ambient pollution compared to the
elevatedsources,whichtravellongerdistancesbeforesettledontheground.
Overall,theconclusionofthe200607studycarriedbytheMoEFandCPCBisthatthetransport
sector (directly and indirectly) is responsible for most of the ambient PM pollution in Pune,
followedbycontributionsfromthebrickkilnindustry,openwasteburning,anddomesticsector.
In this study, the percentage contributions from various sectors have remained similar with
additionalsectorslikebrickkilnsandquarriescontributingtothelocalpollutionproblems.
- 54 -
Chapter5.0
HealthImpactsofAirPollution
An important objective of this study was to assess the baseline health impacts of current air
pollution trends in the cities (based on the estimated emissions inventory and concentration
profilesoverthecity)andcomparefuturescenariosforpotentialcobenefitsbetweenhealth
impactsandcarbonemissions(forsixinterventionspresentedinChapter1.0).
Figure 5.1: Schematics of information to assess health impacts
Emissions
Sources
Exposure
Concentrations
Impacts
DoseResponse
Among the criteria pollutants, PM is a well established endpoint for assessing the health
impacts in a city. Epidemiological studies in developed and developing countries have shown
that elevated ambient PM levels lead to an increased risk of mortality and morbidity (a
summary of reference studies is presented in Annex 3.0). Health effects range from minor
irritation in eyes and upper respiratory system to chronic respiratory disease, heart disease,
lungcancer,andleadinguptoprematuredeath.ThePMconcentrations(forPM10andPM2.5)
representamixofsourcesprimaryandsecondary,whichbecomesacommondenominator
for the criteria pollutants, without double counting the possible impacts of mix of pollutants
(chemical composition and contributions of pollutants to PM concentrations is presented in
Figure2.1).
In this study, the modeled PM10 concentrations59(presented in Figure 4.10) were utilized to
assessthehealthimpactsfollowingthemethodologypresentedinEquation5.1.
E * C * P
Equation5.1
Where,
E =numberofestimatedhealtheffects(variousendpointsformortalityandmorbidity).
= the dose response function (DRF) for particular health endpoint; this is defined the
changeinnumbercasesperunitchangeinconcentrations.Thisisestablishedbased
onepidemiologicalstudiesconductedoveraperiodoftime,analyzingthetrendsin
hospitalrecordsandairpollutionmonitoring.
C = thechangeinconcentrations;theconcentrationsmodeledaboveacertainthreshold
value. Although, WHO claims that there is no threshold over which the health
impactsaremeasured.Ingeneral,theimpactsarefeltattheminutefluctuationsof
thepollution.Inthisstudy,thethresholdissetat20g/m3
P = thepopulationexposedtotheincrementalconcentrationsabove;thiscouldbeona
gridbygridbasis(usedinthisstudy)orforthecityorregiononawhole,depending
onthelevelofinformationavailable.
The Health Effects Institute (USA) conducted a detailed literature survey on the impact of
outdoorair pollutiononhumanhealth60includesanextensivelistofreferencesfor followup
on the dose response functions for various end points and methodologies to conduct
epidemiological studies to develop these dose response functions. Main Conclusion of their
lateststudyunderthePAPAprogramisthatthedoseresponsefunctionsforPM10andPM2.5are
theverysimilarthroughouttheworldandstatesthatthestudysfindingofa0.6%increasein
mortality for every 20g/m3 of exposure to particulate air pollution is strikingly similar to
comparablewesternresults(whichrangefrom0.4%to0.6%)andprovideincreasedconfidence
in the new Asian results.. Also see reports from California Air Resources Board on PM and
healthimpacts.
59
The finer fraction of PM (less than 2.5 micron) is considered more harmful of the PM fractions and more studies
are now highlighting the importance of PM2.5 in health impact studies. In this study, while we calculated the PM2.5
concentrations, presented in Figure 4.11, due to limited measurements available from the cities, these numbers were
not calibrated for further use. Hence, all the health impacts are calculated using PM10 as a measurable endpoint and
the results will be updated in the future simulations, as and when the measurements of PM2.5 are more prevalent in
these cities. More details on the modeled PM2.5 annual and seasonal patterns are available up on request.
60
Health Effects Institute (USA), 2004, Health Effects of Outdoor Air Pollution in Developing Countries of Asia:
A Literature Review @ http://pubs.healtheffects.org/view.php?id=3
PAPA Public Health and Air Pollution in Asia @ http://www.cleanairnet.org/caiasia/1412/article-48844.html
California Air Resources Board (2008) Methodology for Estimating Premature Deaths Associated with Long-term
Exposure to Fine PM in California @ http://www.arb.ca.gov/research/health/pm-mort/pm-mort.htm
- 56 -
Table 5.1: Average dose response functions for morbidity end points
Willingness
Willingness
MortalityandMorbidity
Doseresponse
toPay(INR)
toPay
HealthEndpoint
function61()
(USD)62
PrematureDeaths
0.0000134
1,880,000
41,700
AdultChronicBronchitis
0.000040
50,000
1,100
ChildAcuteBronchitis
0.000544
20,000
450
RespiratoryHospitalAdmission
0.000012
2,000
45
CardiacHospitalAdmission
0.000005
500,000
11,100
EmergencyRoomVisit
0.000235
4,000
89
AsthmaAttacks
0.002900
1,000
22
RestrictedActivityDays
0.038280
500
11
RespiratorySymptomDays
0.183000
30
1
Themorbiditycalculationsforconcentrationsexceedingthenationalambientstandardarealso
includedinthisstudyandthedoseresponsefunctionsutilizedarepresentedinTable5.1.For
the final health impacts analysis, the population data is taken from global GRUMP database,
distributed by SEDAC and the mortality and morbidity cases are estimated for the modeled
annualaverageconcentrations(inFigure4.10)exceedingtheWHOhealthguidelinevalueof20
g/m3.Theestimatedhealthimpactsbasedonthemodeledannualaverageconcentrationsis
presented in Table 5.2. The health impacts are then valuated based on willingness to pay
methodology,presentedinTable5.3.
Table 5.2: Estimated Mortality and Morbidity due to air pollution for 2010
(numbers rounded to nearest zero)
Mortality&Morbidity Pune
Domainsize(kmxkm) 32x32 44x44
32x32 44x44
44x44
StudyDomainPopulation(million) 6.5
8.5
3.3
7.8
5.0
LandSeaBreeze NO
YES
NO
NO
YES
2010PM10emissions(tons/yr) 36,600 56,400
17,900 35,100
19,950

PrematureDeaths 3,600
3,950
1,800
4,950
1,250
MortalitypertonofPM10 0.1
0.07
0.1
0.14
0.06

AdultChronicBronchitis 10,800 11,800

5,400
14,800
3,750
ChildAcuteBronchitis 79,250 86,600
39,300 108,300
27,400
RespiratoryHospitalAdmission 5,000
5,460
2,500
6,800
1,700
CardiacHospitalAdmission 1,350
1,480
670
1,850
470
EmergencyRoomVisit 97,800 106,900 48,500 133,700
33,800
AsthmaAttacks(million) 1.2
1.3
0.6
1.7
0.4
RestrictedActivityDays(million) 10.4
11.3
5.1
14.2
3.6
RespiratorySymptomDays(million) 49.7
54.1
24.5
67.6
17.1
Rajkot
24x24
1.4
NO
14,000
300
0.02
950
6,800
450
120
8,400
0.1
0.9
4.2
61
62
The dose response coefficient is defined as the number of effects/1g/m3 of change in concentration/per capita
Conversion utilized 1 USD = 45 INR
- 57 -
In 2010, the six cities account for an estimated annual 15,200 premature deaths due to
exposure to air pollution above the WHO guidelines. These levels are very significant and
comparabletothelevelsbeingexperiencedinthecitiesacrossAsia.Asummaryofthehealth
impactsacrosstheworldispresentedinAnnex3.0.
Table 5.3: Estimated costs for Mortality and Morbidity due to air pollution for 2010
Rajkot
Mortality(INRcrores) 680
750
340
930
240
60
Morbidity(INRcrores) 1,100
1,210
550
1,510
380
95

Mortality(millionUSD)63 151
165
75
207
52
13
Morbidity(millionUSD) 246
269
122
336
85
21
5.1StudyObservations
It is important to note the differences in the population and geographical sizes of the study
domains when comparing the results between the cities and with the previous studies. The
domains selected in this study are typically larger than the main district areas, surrounding
satellite locations with significant industrial loads to account for the nontransport sector,
which is often neglected in the incity based observations and analysis. For example, for the
Pune domain, the satellite city of Pimpri Chinchwad is included, for the Ahmedabad domain,
the city of Gandhi Nagar is included, and for the Chennai domain, the all the neighboring
industrialestateswithbrickkilnclusters(atleast2030kmawayfrommaindistrictarea)are
included. With the increase in the study domain sizes, the population size and the exposure
estimatesalsoincreaseandthisisreflectedintheestimatespresentedinTable5.2.
Themaincorrelationbetweentheemissions,pollution,exposure,andpossiblehealthimpacts
issimpleandstraight.Mostoftheemissionsinthecitiesarelowlyingsources,whichtendto
affect the immediate vicinity more than the areas 2030 km away from the source locations.
This is particularly true for the sources like vehicle exhaust, road dust, residential burning,
garbage burning, and emissions from the diesel generator sets. Even from the construction
sector, the types of technology differed from city to city. Only in Chennai, the brick kilns are
designedwithstacks,whichallowforthepollutiontotravelfartherthanthelowlyingsources.
The four power plants (two in Chennai and two Ahmedabad) also contribute to the local
pollution, but the contributions are low when compared to the transport and industrial
activities,becausethehighstackswhichdispersemostofthepollutiontovicinitiesoutsidethe
cities.
The landsea breeze is a significant quotient in these studies. Comparatively, of the six cities,
ChennaiandSuratarethemostindustrialandalsothecoastalcities,withsignificantemissions
from this sector and yet experience the lowest impacts relative to their geographical and
63
For this analysis, conversion of 1 USD = 45 INR is utilized; 1 crore = 10,000,000; 1 million = 1,000,000
- 58 -
populationsizes.ThisisprimarilyduetothebreezesWesterlywindsforChennaiandEasterly
windsforSurat,dispersingmostofthepollutiontheindustries(seeFigures3.3to3.5).
The dose response functions themselves are also subjective and represent a conservative
estimate based on the studies around the globe. A summary of these studies and references
utilizedforthisanalysisaresummarizedinAnnex3.0.
The mortality per ton of emissions presents an interesting perspective to the health impacts
analysis.WhencomparingbetweenChennaiandAhmedabad,theratioforthelaterisatleast
doubleandthisisprimarilyduetothemeteorologyandgeographicalstructureofthecity.The
winds are at least double in Chennai (Figure 4.4) and mostly Westerly to drive emissions
(especiallyfromtheindustriesandpowerplants,whicharemostlycoastal)outofthecitylimits.
Also, the city spread for Ahmedabad is very densely packed for the emissions over the
populatedareas,includingindustries,andhencethehigherratio.IncaseofRajkotandSurat,
significantportionoftheemissionsarecomingfromtheindustriesallocatedtotheestimates
andthusreducingtheoverallimpactonmortality.
The costs associated with the mortality and morbidity is based on willingness to pay
methodologyandissubjectedtoscrutiny.ThenumberspresentedinTable5.3areprovidedas
abaselineestimatetocompareagainstpossiblecostsofinterventionstocontrolairpollutionin
thesecities.Thekindofquestionsthatcanbeansweredfollowingthisanalysisare,forexample,
(1) what level of interventions are possible to reduce the health costs associated with the
pollution exposure, based on the current knowhow of the emissions and their strengths to
control (2) do the estimated health costs warrant an immediate response to the pollution
control?
- 59 -
Chapter6.0
WhatifScenariosfor2020&Cobenefits
air pollution
Figure 6.1: Maximizing benefits with

For a growing number of cities, the main
informed & early action
environmental concern still remains local
pollutants such as particulate matter (PM10
with aerodynamic diameter less than 10 m
the goal is to reduce air
.. and maximize
andPM2.5withaerodynamicdiameterlessthan
pollution below
environmental and social
benefits
effectively & fast
2.5 m), SO2, NOx, CO and Ozone. Current
trends in environmental regulation and
industrial development are converging in a
manner that encourages a thoughtful and
consistent approach to appropriate control
.. an informed
analysis and dialogue
measures which can benefit both the local
will help
pollution control and reduction of GHGs to
20 yrs
now
5 yrs
10 yrs
time
support the climate dialogue. These can range
from little or none on small sources (household stoves, small industrial boilers) to advanced
controls on modern power plants (e.g., electrostatic precipitators, ESPs) and vehicles (e.g.,
dieselparticlefilters)64.
Theobjectiveofthisstudy,givenabaselineemissionsinventorywasdevelopedforthesixcities
for year 2010, was to evaluate six whatif emission scenarios for 2020 targeting the
reductionoflocalpollution(particularlyPM)forhealthbenefitsandtoreduceGHGstosupport
climatedialogue.Notethatthewhatifscenariosarebeingevaluatedforhealthandcarbon
64
McGranahan et al. (2003) Air pollution and health in rapidly developing countries, Published by Earthscan
Publishers @ http://www.earthscan.co.uk/?tabid=994
Schwela, et. al. (2006) Urban Air Pollution in Asian Cities: Status, Challenges and Management. Earthscan
Publications, UK @ http://www.earthscan.co.uk/?tabid=1455
benefitspossiblebasedontheemissionstatisticsestimatedforthestudydomain(presentedin
Figure3.2,4.3,4.4,and4.5andTable4.1)andnotnecessarilyreflecttheinvestmentplansfor
the city. The analysis is carried to evaluate the possible benefits to air quality (related health
benefits)andnotfocusedonhowtheinterventionmaybeimplementedorhowmuchitmay
cost.
ArecentGAPconferenceonCobenefitsofAirPollutionconcludedthat65
Currentscienceemphasizestheurgentneedtoaddressairpollutionandclimatechange
inanintegratedway.Weshouldnolongertreatthesetwoissuesseparatelyaswestrive
to achieve sustainable development and a low carbon society. In both developing and
industrialized countries, abatement of air pollution and mitigation of climate change
havegenerallybeentreatedseparately.Thereare,however,largebenefitsinconsidering
the control options together; such approaches would mostly lead to increased health
and/orclimatebenefitsanddecreasedcosts.
6.12020BaselineScenarioAnalysis
For 2020 scenario analysis, first the baseline emissions inventory for projected from 2010
baselinesunderbusinessasusualconditionsandthencorrectedforthesixinterventions.For
the domestic sector, the gridded population for 2020 was utilized. The industrial sector is
estimatedtogrowat~10percentperannum,attherateofthenationalGDP.Incaseofthe
transportsector,thepassengervehicleswereextrapolatedat~8percent,motorcyclesat~10
percent,andtheremainingprivatesectorvehicleslikeshortbuses,commercialvehicles,etc.,at
~1percent.Itisalsoassumedthattheemissionstandardswillimproveinthecomingdecade,
including an improvement of onroad conditions for the vehicle movement, which will
inherently reduce the deterioration rates and thus the emission loads. Hence, though the
vehicle population is expected to increase at the rate of ~10 percent, the emissions are not
expectedtogrowatthesamerate.Theemissionsarethenprocessedthroughthedispersion
model to develop the maps similar to those presented in Figures 4.10 and 4.11, but not
presentedinthisreport.
TheemissionsinventoriesforeachscenarioarethenestimatedforeachcityandlabeledS1to
S6, processed through the dispersion model, and evaluated for health benefits using the
followingequation.
E * C * P
Equation6.1
Where,
E =numberofestimatedhealtheffects(variousendpointsformortalityandmorbidity).
65
Global Atmospheric Pollution Forum, Climate and Air Pollution Co-benefits (2008) conference, organized by
SEI in Stockholm, Sweden @ http://sei-international.org/projects?prid=140
- 62 -
= the dose response function (DRF) for particular health endpoint; this is defined the
changeinnumbercasesperunitchangeinconcentrations(describedintheprevious
section).
C = the change in concentrations = the difference between the baseline concentrations

for 2020 against the concentrations modeled for each of the scenarios. Unlike the
2010 baseline estimates where the health impacts were evaluated against a
threshold value and thus an estimate of number of cases incurred in a particular
year, in this case, the difference between the baseline and the scenario
concentrationswillprovideanestimateofthenumberofcaseswhichcanbesaved
duetotheimplementationofthescenario.
P = thepopulationexposedtotheincrementalconcentrationsabove;thiscouldbeona
gridbygridbasis(usedinthisstudy)orforthecityorregiononawhole,depending
onthelevelofinformationavailable.
The estimated health impacts based on the modeled annual average concentrations for year
2020arepresentedinTable6.1.
Table 6.1: Estimated Mortality and Morbidity due to air pollution for 2020
(estimates rounded to nearest zero)
32x32 44x44
44x44
10.5
4.3
10.3
6.2
LandSeaBreeze NO
YES
NO
NO
YES
21,000 31,800
23,200

PrematureDeaths 4,300
6,000
2,500
7,850
2,050

AdultChronicBronchitis 12,900 17,800

7,500
23,400
6,100
ChildAcuteBronchitis 94,500 130,400 54,500 171,500
44,900
RespiratoryHospitalAdmission 6,000
8,200
3,400
10,800
2,800
CardiacHospitalAdmission 1,600
2,200
930
2,900
770
EmergencyRoomVisit 116,650 161,000 67,200 211,650
55,400
AsthmaAttacks(million) 1.4
2.0
83.0
2.6
683.3
RestrictedActivityDays(million) 12.3
17.0
7.1
22.4
5.9
RespiratorySymptomDays(million) 59.1
81.5
34.0
107.1
28.0
Rajkot
24x24
1.9
NO
18,500
670
2,010
14,700
930
250
18,200
0.2
1.9
9.2
In 2020, the six cities account for an estimated annual 21,400 premature deaths due to
exposuretoairpollutionabovetheWHOguidelines.
- 63 -
6.2WhatifEmissionScenarios
Thewhatifscenariosevaluatedforcobenefitsare
(1) An increase in NMT shares, ~20 percent. Assuming that there isenough infrastructure to
support the move from motorized to the nonmotorized transport, the shift in the passenger
travelkilometersisexpectedtocomeinfractionsfrompassengercars,motorcycles,andbuses.
Incitieswithlarge3wheelerpopulation,afractionofpassengertravelisshiftedtoNMT.For
the emissions analysis, the fractions are deducted from the vehicle kilometers traveled from
eachmodetoaccommodatefortheequivalentincreaseintheNMTsharesforeachcity.
(2)Anincreaseinpublictransportshare,~20percent.Thisscenarioismodeledassumingthat
the fleet will at least double from the current numbers to accommodate the increase in the
passengerkilometerstraveledbybus.Thisshiftismodeledassuming70percentcomingfrom
themotorcyclesand30percentfromthepassengercars.Ingeneral,thereisanurgentneedto
update the bus fleet with newer vehicles in most of the cities. However, the supply of new
buses has been slow. Two bus manufacturers in India Tata and Ashok Leyland together
producenearly90percentofstandardbusesinthecountry.Attheircurrentrateofproduction
theycandeliveronlyabout100busesamonthunlesstheyrampupproduction66.Thisrateof
supplyisinadequatetomeetthedemandforbusesinIndiancitiesthatareinvestinginpublic
transport. To underscore this point, the demand for buses under the JNNURM funds alone is
approximately70,000busesforpublictransportpurposes67.
(3) Introducing alternate fuel for public transport and 3 wheeler sectors. Assuming the fuel
supply and infrastructure to support the demand are in place, this scenario is evaluated
assuming that the public transport buses and the 3 wheelers will be converted from their
current petrol and diesel based fleets to CNG based fleets. This intervention follows the
exampleofthe1998SupremeCourtrulingwhichledthecityofDelhitotakeconcretestepsto
addressairpollutioninthetransportsector68andthelargesteverCNGswitchintheworldfor
public vehicles. More than 100,000 vehicles (including the three wheelers and taxis) were
converted to CNG in Delhi over five years69. This resulted in significant decrease in the air
66
Times of India, February 8th, 2009, BRTS dreams may go bust @

http://timesofindia.indiatimes.com/articleshow/msid-4096144,prtpage-1.cms
67
Times of India, May 13th, 2009, Low floor buses delayed in Delhi @
http://timesofindia.indiatimes.com/Delhi/Low-floor-buses-delayed-Bluelines-get-a-breather/articleshow/4521904.cms
The Hindu, May 13th, 2009, Delhi Govt. faces cancellation of bus funding under JNNURM @
http://www.hindu.com/thehindu/holnus/004200905131452.htm
68
More on the Supreme Court ruling @
http://www.cleanairnet.org/infopool/1411/propertyvalue-19513.html
The timeline of implementation (in the transport and industrial sector) and the experience for instituting change
which has become a model for other Indian cities, by Narain, et al., 2005, Who Changed Delhi's Air? The Roles of
the Court and the Executive in Environmental Decisionmaking, RFF @
http://www.rff.org/Publications/Pages/PublicationDetails.aspx?PublicationID=17425
69
More on Delhi Transport Corporation @
http://en.wikipedia.org/wiki/Delhi_Transport_Corporation
- 64 -
pollution.Thelargestimprovementcamefromtheconversionof~3,000dieselbusestoCNG70.
Delhihassince2000alsoenforcedEuroIIemissionstandards,fiveyearsaheadofschedule,and
EuroIIIin2005forallpassengervehicles.
(4) A reduction in resuspended road dust, ~50 percent. The resuspended dust due to the
movementofthevehiclesontheroadsisbyfarthelargestculprittotheairpollutionproblem
in the cities. In 2020, it is assumed that there is some inherent improvement in the road
infrastructurewhichwillreducetheresuspensionontheroads.However,theneardoublingof
thefleetsexpectedin2020allowsforthefurtherresuspensionofirrespectiveofthereduction
in silt loadings on the road. For the reduction estimates, the silt loadings on the main and
arterialroadsaresignificantlyreducedtoachievethepossible50percentreductioninthere
suspendeddust.
(5) A change in technology used for baking bricks and improving efficiency by ~50 percent.
The construction is the fastest growing sector in the cities and the demand is expected to
furtherincreaseinthecomingdecade.Whilethetechnologyusedforbakingbricksineachcity
variesfromchimneystyle(inChennai)tonochimneypiledstyle(inPune),itisassumedthat
the technology will be improved to the latest available, such as the Vertical Shaft Brick Kilns
(VSBK) which are gradually being introduced in the developing country cities of Dhaka,
Kathmandu, and Cairo71. In case of Pune, Ahmedabad, Surat, and Indore, where pile style
burningisprevalent,thereductioninemissionsareexpectedintherangeof60percentwhere
as in case of Chennai where the chimney style burning is already prevalent, the change in
technologywillresultinatleast30percentimprovementinefficiencyandemissions.
(6) A reduction in the truck movement in the city limits, ~20 percent. The truck movement
accounts for the bulk of the pollutants in each of the cities. This includes the incity freight
movement via light duty vehicles and intercity freight movement via heavy duty trucks. This
quotient is more significant for the cities of Chennai and Surat, which host large ports and
industrialestates.Forthepotentialemissionsreductionsandbenefitanalysis,itisassumedthat
freightmovementwillbeeithersupportedbytherailwayservicesand/orviaabypassandnot
allowing the trucks to pass through the city. As important assumption here is that lesser the
number of truck kilometers traveled through the incity roads, lesser the emissions, and thus
lesserthepollutionanditsimpacts.
70
Down to Earth, March 2002, The Supreme Court Not to Budge on CNG Issue @
http://www.downtoearth.org.in/full6.asp?foldername=20020331&filename=News&sec_id=4&sid=6
71
More details on the VSBK technology, the manuals prepared and published by the World Bank (2010) are
available @ http://urbanemissions.info/model-tools/sim-air/dhaka-bangladesh.html
- 65 -
6.3CoBenefitsofWhatifScenarios
Note that the scenarios evaluated in this study are neither subjected to planning procedures
norinvestment.Also,theextrapolationoftheemissionsfrom2010to2020isspeculativebased
on the current economic, geographical, and population trends and subject for alternatives in
the many studies. The six scenarios are evaluated, assuming that they are mutually
independent activities, which may not be the case at the time of implementation. This was
assumed for the ease of calculations. For example, the interventions designed to promote
public transportation, in some form or the other, is expected to invariably affect the non
motorizedtransport,especiallyintheformofshorttripsandthuseffectingtheemissionsand
pollutionpatterns.However,inthisstudythepublictransportandthenonmotorizedtransport
benefitsareevaluatedindependentlytoestimatepossiblemutualbenefits.
Table 6.2: Estimated co-benefits from 2020 what-if scenarios

32x32 44x44
44x44
10.5
4.3
10.3
6.2
LandSeaBreeze NO
YES
NO
NO
YES
21,000 31,800
23,200

EstimatedPM10emissionsreduced 13,900 17,400

6,200
8,800
8,200
underwhatifs(tons/yr)
%comparedto2020baseline 37%
31%
30%
27%
35%

Prematuredeathssaved 1,700
1,270
630
1,390
590
%comparedto2020baseline 39%
21%
25%
18%
29%
Mortalitycostsavings(millionUSD) 71
53
26
57
24
Morbiditycostsavings(millionUSD) 114
87
44
94
40

EstimatedCO2emissionsreduced 3.0
5.7
1.8
2.5
2.4
underwhatifs(milliontons/yr)
Rajkot
24x24
1.9
NO
18,500
7,900
42%
290
42%
12
20
1.4
6.4StudyObservations
AsummaryofcombinedbenefitsfromallthescenariosispresentedinTable 6.2.Overallthe
benefits are substantial for health and carbon. Of the estimated 21,400 premature deaths
under the 2020 baseline, implementation of the six scenarios by 2020 can potentially save
5,870lives(~27%)annuallyandinduceanestimatedannualreductionof~16.8milliontonsof
CO2 in the six cities. The details of possible benefits in terms of emissions, concentrations,
mortality,andmorbidityfromthesixscenariosandforthesixcities,andtheassociatedhealth
costsavings,areavailableuponrequest.Importantstudyobservationsinclude
Thebenefitsofthesix interventionswereevaluatedforhealth(fromthereductionof
PM, SO2, and NOx emissions) and CO2 reductions. However, one of the interventions
targets only PM pollution and has no direct reductions for CO2 emissions. This is
scenario(4)interventiontoreduceroaddust.Whilewewereawareofthefactthat
- 66 -
this scenario will benefit mostly the health sector by reducing PM pollution, it was
importanttohighlightalowhangingfruitforairqualitymanagementinthecities.While
thevehicleexhaustandindustrialplumesaremajorsourcesofairpollutionduetofossil
fuelburning,weshouldnotneglecttheirindirectimpactresuspensionofdustdueto
constant vehicular movement, most often from the heavy duty trucks carrying raw
materialandfinishedproductsfortheindustrialsector.
Itisalsothescenariowiththehighesthealthbenefitsofthesixinterventions.Thiscould
be achieved by paving roads, thus reducing the silt loading on the roads or by wet
sweepingwhichwouldreduceresuspensionofsiltandthuslessenexposureeffects.
ThecontributionofroaddustemissionsishighertothePM10totalsthanthePM2.5totals
(Figure 4.3 and 4.4). Since, the health benefit calculations are based on the PM10
concentrations;itisevidentthatthelargestbenefitsareexpectedfromthissource.If
the baseline for comparisons were to shift from PM10 to PM2.5, the analysis focus will
shiftfromresuspendedroaddusttodirectvehicleexhaustandotherfossilfuelbased
sources. Nevertheless, it is important that the road dust emissions and exposure
impactsarenotneglected.
Incaseofdirectvehicleexhaustemissions,thelargestmarginofbenefitsforhealthand
carbonemissions,isinthemaintenanceofthetruckfleet,reductionintheirmovement
inthecitylimits,andingeneralinthemanagementoffreightmovementinthecities.
ForthecitieslikeChennaiandSurat,wheretheportsandtheindustriesdominatethe
geographicalsettingsandtheemissionloads(fortransportandindustrialsectors),the
freight management can be an effective and faster intervention to control local
pollutionandreduceGHGs.Theportactivityisalsolinkedtorailemissions.Asignificant
portion of the freight from Chennai and Surat is moved by rail. An effective mass
transport option for freight management is improving the rail links and thus reducing
theincityemissionsduetotruckmovement.
For the transport sector, all the interventions are interlinked and could lead to
unintended consequences, though the scenarios are designed and evaluated
independentlyfortheeaseofmodeling.Forexample,incaseofscenario(2),improving
publictransportsharesinthecities.Forthreecities,thebenefitsfrommovingashareof
passengercarsandmotorcycletripstobuses(andthusreducingtheiremissions)didnt
result in any reduction in PM10 emissions. This is due to an increase in the road dust
emissionsfromtheadditionalbuses.Therateofresuspensionofroaddustisgenerally
higher for the buses than the cars and the motorcycles. This is an unintended
consequence,whichshouldbeaddressedintandemwithinfrastructureprogramssuch
asbetterroadsandpavementsandalongwithimprovementsinemissionstandardsfor
thesevehicles.
The brick kilns, which are a common site for all city outskirts, are expected to yield
significant benefits in PM and CO2 emissions. However, this intervention is dependent
on the city and the national regulators for promoting emerging technologies, but also
thesocialaspectsoftheworkersandowners,whoaredependentondaytodaywages.
Forthisstudy,itwasassumedthatkilnclusterswilleventuallymovetoVSBKorsimilar
systems for higher energy efficiency and result in significant health benefits for the
locals and carbon emission reductions. A parallel study conducted by Indian and US
- 67 -
researchgroupsconcludedthatashiftfromthecurrentbulltrenchstylekilnstozigzag
orVSBKtechnologiescanyield~4050percentimprovementsintheheatefficiencyand
emissions.
Ofthesixinterventions,onewithconsistentreductionsinPMandcarbonemissionsis
promotion of nonmotorized transport. By definition, nomotorization means no
combustion, and thus noemissions to see and nopollution to breath. Each of the
interventionshasitslimitations,whicharenotdiscussedinthischapter.However,any
intervention,whichpromotesnoncombustion,isboundtoproducebettercobenefits
thantherest.PromotingNMTisthebestofthelot.
Possible intervention which was not included in this study is waste management. The
garbage burning along the roads (when it is left unpicked) and at the landfills is a
growing contributor to ambient pollution, carcinogens, and GHGs. This waste can be
bettermanagedbyeithercompostingthewestwasteorrecyclingthepossibleproducts,
andusingthewaste(dependingonitscalorificvalue)forheatandelectricitygeneration
inselectlocations.
Another mentionable intervention is the power generation. As the cities are growing,
besides the demand for diesel in the transport sector, a good deal of diesel demand
growth comes from diesel generators, usually for own use (1 kW to several MW) but
alsobyutilities(afewtotensofMWeach)inplacessuchashospitals,hotels,markets,
malls,apartments,institutions,etc,withlimitedandnosupplyofelectricity.
6.5StudyExtensions
The primary objective of the study was to establish a baseline emissions inventory for the
criteria pollutants and CO2 for the six cities Pune, Chennai, Indore, Ahmedabad, Surat, and
Rajkot and help support air quality management planning in these cities in a cobenefits
framework,wheretheinterventionscanbestudiedandpossiblyimplementedtobenefitboth
local health (via air pollution reduction) and climate policy (via carbon reductions). Six such
interventions were studied for extrapolated emissions into 2020, demonstrating cobenefits.
Whilebaselinesandanalysiswasconductedusingtheliteratureavailableandinterviewswith
officialsfromvariousdepartmentsinthecities,thereisroomtofurtherthisanalysis.
Aninformationdisseminationmechanismtosharetheseresultswiththelocalofficials
and decision makers in the air quality management community and thus
institutionalizing the analytical framework employed in this study. This will help in
improving the databases and further use of such frameworks for preparation of an
effectivepollutioncontrolstrategy.
A sector based analytical framework. In the cities, interventions with the largest and
overlappingbenefitsare(1)publictransport(2)nonmotorizedtransport(3)brickkilns
and(4)generatorsets.Eachofthesesources,whilecontributesignificantlytothelocal
airpollutionandGHGs;theyareintegraltothecentralplanningprograms,interlinked
with infrastructure and urban development. While it is important to study their
contributionsatthecitylevel,programmaticallytheyhavemoretoofferacrossstates
andatthenationallevel.
- 68 -
ThisstudyfocusedontheTier1cities(exceptforRajkot).Asthecitiesaregrowing,the
smallercitiesarefollowingtheexamplesofbiggercities;bothintermsofgrowthfactors
and in terms of interventions introduced for better environment. All this is linked to
havingrelevantinformationathandforthepolicymakerstomaketherightdecisions.
Theseanalyticalprogramsshouldbereplicatedinothercitiestoinformdecisionmaking
and to promote cobenefit frameworks, where cities can benefit from possible local
pollutionreductionandclimatepolicyfundsforimplementingtheseinterventions.
- 69 -
Annex1.0MethodologyforTransportEmissions
Amethodologypresentedinequation A1.1wasutilizedtoestimatetheemissiontrendswith
air quality and health as the primary indicator. The fundamental equation for calculating the
emissions is based on the activity level, which for the transport sector is equivalent of
Emissions=NumberofVehicles*Vehiclekilometerstraveled(km)*EmissionFactor(gm/km).
Theagemixofvehiclesisdeducedfromtheinformationavailableonthevehiclecountsfrom
the previous years and anticipated retirements for each mode. The emissions inventory
estimatedusingthismethodologydoesnotincluderoaddustresuspension.
NV t 1 NV t * (1 growth )
NV total
age 20
age 0
NV age
EF age EF new * (1 drate ) ^ 2

EF effective
Emissions
age 20
age 0
EF age *
Equation3.1
NV age
NV total
NV total * VKT * EF effective
Foragivenfleet,thetotalemissionsdependonthemixofthevehiclesonroad,e.g.,themake
and the age of the vehicles. The age mix of the vehicles is considered to account for the
deterioration of the vehicles and the associated emissions, and to account for the average
retirement of the vehicles. In this study, an average retirement age of 10 years for the
passenger cars, 15 years for the buses, 12 years for the 3 wheelers, and 8 years for the
motorcycles is assumed. The average retirement age doesnt mean that all the vehicles are
retiredinstantaneously,butafractionofthefleetisinterchangedforanewerfleettomaintain
theassumedagemix. Theinventorypresentedin Table3alsoincludesanestimateforidling
emissions, assuming an average of~20 minutes in congestion for passenger vehicles and ~40
minutesforbusesperday72.
For the criteria pollutants, trucks (heavy duty and light duty commercial vehicles) and 2
wheelers(motorcycles,scooters,andmopeds)dominatetheemissions,followedbybusesand
passengercars.ThetrucksremainthedominantsourceofemissionsintheIndiancitiesbecause
of the diesel based fleets and operating under loaded conditions for most of the time. The
motorcycleemissionsareamongthehighestinspiteofaloweraverageageforthefleetmainly
because of the sheer number of vehicles on road. The passenger car fleet is the newest and
adherestotheneweremissionnormsforpetrolanddieselvehicles,whichtendtobecleaner
onaveragewhencomparedolderfleetsofbusesandtrucks.
72
The methodology utilized for calculating the contribution of congestion to air emissions is presented in SIM-air
working paper No.18, Vehicle idling emissions @ http://www.urbanemissions.info
ThefleetaverageemissionfactorsareestablishedbyARAIIndia(inPune)andutilizedforthe
six city source apportionment study conducted by MoEF and CPCB of India. The emission
factors were also studies and utilized from the following sources SEI emissions inventory
handbook73,IIASAGAINSmodelingsystemforcobenefitsanalysis74,USEPAEmissionsFactors
handbook75,andHEATemissionsmodelingsystem76.
Theemissionsinventoryforthetransportsector(andothers)wascalculatedinassegregated
fashionaspossiblebyvehicletypeandfueltype,toallowforanalysisofthefuelconsumption
patterns.Themainfuelsconsideredinthisanalysisarediesel,petrol,liquefiedpetroleumgas
(LPG)andcompressednaturalgas(CNG).Becauseofitsusageforbusandtruckfleets,therole
of diesel to local criteria and GHG emissions is the largest. Of the total estimated fuel usage
(and for CO2 emissions), the diesel, especially from the trucks is the dominant source of
pollutionfollowedbypetrolandthenthealternativegaseslikeCNGandLPG.
73
Details available @ http://sei-international.org/news-and-media/1147

Details on the GAINS @ http://gains.iiasa.ac.at/index.php/home-page
75
US EPA - 42 details @ http://www.epa.gov/ttn/chief/ap42/
76
HEAT emissions tool @ http://heat.iclei.org
74
- 72 -
Annex2.0ATMoSDispersionModel
The Atmospheric Transport Modeling System (ATMoS) model is a UNIX/Linux based air
pollutiondispersionmodel,primarilydevelopedforsulfurpollutiondispersionmodelingaspart
oftheRegionalAirPollutionInformationSystemforAsia(RAINSAsiasoftwaredevelopedand
distributed by the International Institute of Applied Systems Analysis (IIASA), Laxenburg,
Austria),whichwaslaterupdatedtosupportPMpollution(primaryandsecondary).Thismodel
waspreviouslyutilizedfornumerousairpollutionmodelingstudiesinAsia,includingtheRAINS
and GAINS integratedmodeling system, sulfur pollution control in Asia, PM pollution analysis
forcities.AlistofthestudiesandrelevantliteratureispresentedinBoxA1.1.
BoxA1.1:ApplicationsusingtheATMoSModel
Adhikary,B.,1998,AnalysisofAdvancedControlTechnologiesinImprovingAmbientAirQuality
andHumanHealthinChongqing.B.EThesis,TheUniversityofIowa,USA.
Arndt,R.,etal.,1998,SeasonalSourceReceptorRelationshipsinAsia,AtmosphericEnvironment,
32,13971406.
Calori,G.,etal.,1999,AnurbantrajectorymodelforsulfurinAsianmegacities.Modelconcepts
andpreliminaryapplicationAtmosphericEnvironment,33,31093117.
Calori,G.,etal.,2001,InterannualvariabilityinsulfurdepositioninAsia,JournalofGlobal
EnvironmentalEngineering,7,116.
Guttikunda,S.K.,etal.,2001,SulfurDepositioninAsia:SeasonalBehaviorandContributionsfrom
VariousEnergySectors.Water,AirandSoilPollution,131,383406.
Guttikunda,S.K.,etal.,2005,TheContributionofMegacitiestoRegionalSulfurPollutioninAsia
AtmosphericEnvironment.
Heffter,J.L.,1983,BranchAtmosphericTrajectory(BAT)model,NOAATechnicalMomorandum,
ERLARL121.
Holloway,T.,etal.,2002.TransferofreactivenitrogeninAsia:developmentandevaluationofa
sourcereceptormodelAtmosphericEnvironment
IES(2008).CobenefitsofairpollutionandGHGemissionreductionsinHyderabad,India.
IntegratedEnvironmentalStrategiesIndiaprogram,USEPA,WashingtonDC,USA.
Li,J.,etal.,2004.CurbingAirPollutioninMegacity:HumanHealthBenefitsofAirPollution
ControlinShanghaiJournalofEnvironmentManagement,70,4962.
Themodelhasflexibletemporalandspatialresolutionandcanberunfortheperiodsranging
from months to a year. The model produces monthly average concentrations as output and
then converted to seasonal and yearly averages for further analysis. The meteorological data
forthisstudyisobtainedfromtheNCEPReanalysisfields,anopenlyavailableresource77.The
meteorologicaldatafortheprecipitation,windspeedsandwinddirectionsareavailableforthe
periodfrom1948to2010fromtheNCEPreanalysisdatafields.Anexampledatafortheyear
2008,thegridscoveringthecityofPune,Chennai,Indore,Ahmedabad,Surat,andRajkotareas
foryear2008arepresentedinChapter3.0forwindspeedsanddirections,mixinglayerheights,
andmonthlyprecipitation.
RemovalProcessing
Whilethechemicaltransformationofthespeciesprovidesaformofremovaloftheemissions,
two physical processes are included dry and wet deposition. The wet deposition scheme is
activatedonlyduringtheprecipitationperiod.
A net removal rate at the surface due to dry deposition is calculated using the equation
V
Rd d where,Rdisthedryremovalrate(persec),Vdisthedrydepositionvelocity(m/sec),
z
andzisthesurfacelayerheight(m).Constantdrydepositionratesareprescribedinthemodel.
Similarly,netremovalrateatthesurfaceduetowetdepositioniscalculatedusingtheequation
Rw S p * P where Rw is the wet removal rate (per sec), Sp is the precipitation scavenging
coefficient and p is the precipitation rate (mm/hr). The rates are average numbers and the
methodologyofcalculationisflexible,dependingontheavailablelocalinformation78.
Utilizing the gridded emissions inventory (presented in the previous section) and the
meteorologicalfields,thedispersionmodelingiscarriedoutforannualaverageconcentrations
overthestudydomainsforeachofthecities.
77
NCEP/NCAR Reanalysis data for a number of meteorological conditions is available for a period starting from
01/01/1948 @ http://www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html The meteorological processing
is conducted before the dispersion modeling.
78
An average set of dry and wet deposition rates are presented in the input file of ATMoS-4.0 and user has the
option to either keep the defaults or change
- 74 -
Annex3.0ReportsofHealthImpactsDuetoAirPollution
Country
City
Mortalityreported
Reference
Worldwide
Pakistan
Worldwide
Reuters,May29th,2009
TheNews,May15th,2009
Guardian,May12th,2009
India
Pakistan
UnitedStates
UAE
Worldwide
Bangladesh
Dhaka
Worldwide
India
Indonesia
byWMO
Delhi
Jakarta
Russia
Uganda
Climatechangecauses315,000deathsayear
Airpollutionkills~23,000annually
100millionprematuredeathscouldbe
preventedbycuttingglobalemissionsby50%
by2050
Over20millionpeoplehaveasthmaand
susceptibletohigherrisks
Pollutionkills55,000everyyear
Airpollutionshortenslife
Airpollutionisamajorpublichealthissue
Upto60,000prematuredeathsayear
worldwideareduetoPMemissionsfrom
oceangoingshipengines
Airpollutioniscausing~15,000premature
deathsayear
Lethalairpollutionboomsinemergingnations
Country'sasthmacapital
50%oftheprofessionalsmaybeliterallysickof
workduetoairquality
Airpollutionisresponsiblefor17%ofdiseases
inchildrenand10%inadults,andaffects44%
ofthepopulation
Dowehavetolivewithpollution?
Afghanistan
Kabul
Bahrain
HongKong
HongKong
Philippines
Manila
China
India
Beijing
Hyderabad
China
Worldwide
Mongolia
Worldwide
Airpollutionishasteningthedeathsof3,000
everyyear
Morethan10%ofthepopulationsuffersfrom
asthmaandthenumberissettoriseasair
pollutionincreases
Airpollutioncausedmorethen6,600
prematuredealths
Airpollutionkills5,000annually
Pollutionsparkshealthworries
Airpollutioncauses~2,500prematuredeaths
annually
Thecombinedhealthandnonhealthcostof
outdoorairandwaterpollutionforChina's
economycomestoaround$US100billiona
year
byWHO
Airpollutionintheworldcitiesiscausingsome
twomillionprematuredeathseveryyear
Ulaanbaatar Airpollutioncauses~7,000prematuredeaths
Anthropogenicclimatechangeofthepast
30yearsalreadyclaimsover150,000lives
annually
- 75 -
HindustanTimes,April25th,
2009
TheNation,April24th,2009
BBC,April12th,2009
GulfNews,April12th,2009
Guardian,April09th,2009
IRIN,April3rd,2009
AFP,March22nd,2009
MailToday,March1st,2009
JakartaGlobe,February19th,
2009
RussianNews&Information
Agency,February13th,2009
TheNewVision,February10th,
2009
IRIN,January29th,2009
GulfNews,February08th,2009
AFP,January22nd,2009
ManilaTimes,November07th,
2009
Telegraph,August07th,2007
IES,March15th,2008
WorldBank,July07th,2007
ENS,October06th,2006
UE,June15th,2007
Nature,November,2005
BoxA3.1:HealthImpactAnalysisStudies
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
HealthEffectsofOutdoorAirPollutioninDevelopingCountriesofAsia:ALiterature
Review(2004,2010)
@http://pubs.healtheffects.org/view.php?id=3
CaliforniaAirResourcesBoard(2008)MethodologyforEstimatingPrematureDeaths
AssociatedwithLongtermExposuretoFineAirborneParticulateMatterinCalifornia
@http://www.arb.ca.gov/research/health/pmmort/pmmort.htm
PAPAPublicHealthandAirPollutioninAsia
@http://www.cleanairnet.org/caiasia/1412/article48844.html
Lvovsky,etal.2000.EnvironmentalCostsofFossilFuels:ARapidAssessmentMethod
withApplicationtoSixCities.EnvironmentDepartmentPaperNo.78,TheWorldBank,
WashingtonDC,USA
Bell,etal.,2006.TheavoidablehealtheffectsofairpollutioninthreeLatinAmerican
cities:Santiago,SoPaulo,andMexicoCity.EnvironmentalResearch,100,March2006,
431440.
Pope,C.A.,IIIandDockery,D.W.2006.Healtheffectsoffineparticulateairpollution:
Linesthatconnect.JournaloftheAirWasteManagementAssoc.56(6):709742.
Ostro,etal.,1998.EstimatingtheHealthImpactofAirPollution:Methodologyandan
ApplicationtoJakarta.Workingpaperseries,TheWorldBank,WashingtonDC,USA
Li,J.,andS.K.Guttikunda,et.al.,2004.QuantifyingtheHumanHealthBenefitsof
CurbingAirPollutioninShanghai.JournalofEnvironmentalManagement.70,pp.4962
URBAIRAirQualityManagementSeries,TheWorldBank,WashingtonDC,USA
HEI,2004.HealthEffectsofOutdoorAirPollutioninDevelopingCountriesofAsia:A
LiteratureReview.HealthEffectsInstitute,Boston,USA
Ostro,etal.,1994.EstimatingtheHealthEffectsfromAirPollutants:AMethodWith
anApplicationtoJakarta.WorldBankPolicyResearchWorkingPaper#1301
Xu,etal.,1994,AirPollutionandDailyMortalityinResidentialAreasofBeijing,China.
ArchivesofEnvironmentalHealth,49,pp.216222
SAES,2000,ShanghaiEnergyOptionandHealthImpact.ReportpreparedbyShanghai
AcademyofEnvironmentalSciencesandShanghaiMedicalUniversity
CostofPollutioninChina,EastandPacificRegion,TheWorldBank,WashingtonDC
@http://go.worldbank.org/FFCJVBTP40
- 76 -

Urban Air Pollution Analysis - India 6 Cities 2011 09 13

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Urban Air Pollution Analysis - India 6 Cities 2011 09 13

Uploaded by

Copyright:

Available Formats

Simple Interactive Models for Better Air Quality

Davachi Urali Waste Site

UrbanEmissions.Info, New Delhi, India (2011)

Analysis & errors are sole responsibility of the author(s).

Sources of Air Pollution in Indian Cities

Read more on the Delhi - Asthma Capital of India

Sources of Air Pollution in Indian Cities

Figure 1.1: Location of the six cities

Integrated air quality management (AQM) is a

Examples of emissions inventories available in the public domain

Sources of Air Pollution in Indian Cities

Figure 1.2: Schematic framework of the SIM-air program

Consequences of exposure to these pollutants range from premature mortality due to

Read more on the national ambient air quality standards in India

Sources of Air Pollution in Indian Cities

While PM pollution is measured and estimated as a single entity, the composition of PM is

Figure 2.1: A typical chemical composition of particulate pollution

Sources of Air Pollution in Indian Cities

A cobenefits approach16is increasingly becoming a starting point for discussing integrated

Figure 2.3: Radiative Forcing of Climate

Figure 3.1: Location of the six cities

Sources of Air Pollution in Indian Cities

Table 3.1: Cities at a glance geographical and transport characteristics

IndustrialEstate Medium Large

Figure 3.2: Geographical domains of the six cities

Sources of Air Pollution in Indian Cities

Indore, Madhya Pradesh

Figure 3.4: Monthly precipitation patterns (2008)

Indore, Madhya Pradesh

Sources of Air Pollution in Indian Cities

Figure 3.5: Daily mixing layer heights (2008)

Jul08 Aug08 Sep08 Oct08 Nov08 Dec08

Jan08 Feb08 Mar08 Apr08 May08 Jun08

Indore, Madhya Pradesh

Jan08 Feb08 Mar08 Apr08 May08 Jun08

Jul08 Aug08 Sep08 Oct08 Nov08 Dec08

Jan08 Feb08 Mar08 Apr08 May08 Jun08

Jul08 Aug08 Sep08 Oct08 Nov08 Dec08

Aug08 Sep08 Oct08 Nov08 Dec08

Jul08 Aug08 Sep08 Oct08 Nov08 Dec08

Aug08 Sep08 Oct08 Nov08 Dec08

The Central Pollution Control Board operates a

Figure 3.7: National ambient

Sources of Air Pollution in Indian Cities

Jul09 Aug09 Sep09 Oct09 Nov09 Dec09 Jan10

Jan09 Feb09 Mar09 Apr09 May09 Jun09

Indore, Madhya Pradesh

Jul09 Aug09 Sep09 Oct09 Nov09 Dec09 Jan10

Jan09 Feb09 Mar09 Apr09 May09 Jun09

Jul09 Aug09 Sep09 Oct09 Nov09 Dec09 Jan10

Jul09 Aug09 Sep09 Oct09 Nov09 Dec09 Jan10

Jan09 Feb09 Mar09 Apr09 May09 Jun09

Jul09 Aug09 Sep09 Oct09 Nov09 Dec09 Jan10

Sources of Air Pollution in Indian Cities

JNNURM, City Development Plan for Ahmedabad, Gujarat

Sources of Air Pollution in Indian Cities

Table 3.2: Baseline vehicle inventory for the six cities

Figure 3.9: Vehicle fleet composition in the six cities

Sources of Air Pollution in Indian Cities

Figure 3.10: Passenger travel statistics in India30

% Share of Para-Transport (e.g., 3 Wheelers)

% Share of Non-Motorized Transport

Population (in 100,000s)