sim-air working paper series

Simple Interactive Models for Better Air Quality

Multi‐Pollutant Emissions Inventory for the National Capital Region of Delhi

Dr. Sarath GuttikundaDr. Giuseppe Calori

March, 2012

SIM‐air Working Paper Series: 38‐2012

Analysis & errors are sole responsibility of the authors.© www.urbanemissions.info

Please send your questions and comments tosim‐[email protected]

This report is distributed under the SIM‐air working paper series for free and requests an appropriate citation as

Guttikunda, S.K. and G. Calori, 2012. “Multi‐Pollutant Emissions Inventory for the National Capital Region of Delhi”, UrbanEmissions.Info (Ed.), SIM‐air Working Paper Series, 38‐2012, New Delhi, India.

MultiCapita Dr. SarathDr. Giuse

Delhi (Indincluding collectivepast 20 y(Census‐Inhousing –and Bell, GuttikundrespiratorKhaturia a Over the introduce

• Thfoth20pre20an

• Bofbala

• ThfrInci

• CDci

Whdtolage

‐Pollutaal Regio

h Guttikundappe Calori, A

dia), host city parts of thly referred asears ‐‐ in 199ndia, 2012). – that have co2006; Kandl

da and Gurjary ailments (Cand Khan, 20

past decade, ed to address he largest evor more than hree wheeler000’s this reollution, withetrofitting ap002; Kathuriand Devotta, 2efore the 20f the retrofituses and thelong with imppilot for futuhe city also rom 65 km nternational Aities of Gurgaonversion of

Delhi, 2010) anity outskirts, f

While these inealth for theaunting challo alternative ack of enoughenerator sets

ant Emison of De

, UrbanEmissAriaNet srl., M

for the 2010e neighborins the Nationa90, the total As India’s caontributed tolikar, 2007; Mar, 2011). ThChhabra, et a07).

a number ofcity’s pollutio

ver compresse100,000 pubrs, and taxisesulted in sh the largest pproximately a, 2005; Kuma2007). 010 Commonwted fleet wase fleet size iplementationure bus rapid tbenefitted frin Phase I tAirport. This on and Noidacoal based tnd relocationfollowing the

nitiatives hele citizens of enges posed fuels like CNGh public transs, and industr

ssions Inelhi

sions.info, NeMilano, Italy

0 Commonweng states of l Capital Regipopulation o

apital, Delhi o an increaseMohan et alis, in turn, hal., 2001; Pan

f green initiaton problem, ied natural gablic transport ) (DTC, 2010significant deimprovemen3,000 diese

ar and Foster

wealth Games replaced wincreased to of special tratransport approm the como 180 km, iresulted in aa. thermal pown of the coal a Supreme Co

ped improveDelhi, they by the growG is outdone sport buses, ial growth.

nventor

ew Delhi, Ind

ealth Games, Haryana, Ution of Delhi (Nof NCR stoodhas grown ae in air pollutl., 2007; Guthas increasednde et al., 200

tives have bencluding as (CNG) switvehicles (bus0). In the eaecrease in Pnt coming froel buses (DTr, 2007; Chela

es, a large pawith newer CNaround 5,00ansport corriplication.

mpletion of thncluding an a drop in on

wer plants witand fuel oil burt orders, (N

e the quality have neverthing sources oby the increagrowing dem

ry for th

dia

covers an arettar PradeshNCR). The regd at ~8.6 millcross all section (Bose, 19ttikunda, 200d health risk02;

en

tch ses, rly PM om TE, ani

art NG 00; dors during t

he Metro Phaexpress line

n‐road vehicle

thin Delhi to based industrNarain and B

of air in theheless fallen of air pollutioasing numbermand for elec

Fig.01

he Natio

ea of ~2,500 h, and Rajastgion has growion and in 20tors ‐ indust996; Gurjar e09; Firdaus aks, reflected

the Games, w

ase‐II, increae from the ce density tow

gas based pies, includingell, 2006).

e city and thshort in kee

on. The benefr of passengectricity leadin

1: Mail Today

onal

square kilomthan. The arwn rapidly ove011 at ~22 mtry, transportt al., 2004; Nand Ahmad, by an increa

which succeed

asing the covcity center twards the sa

power plants g brick kilns, t

us the respireping up witfits of leapfror vehicles on ng to use of i

y, March 3rd, 2

meters rea is er the million t, and Narain 2011; ase in

ded as

verage o the tellite

(SoE‐to the

ratory th the ogging road, n‐situ

2009

For the pe

g/m3 an

diameter

annual ampollution due to inc

In Decemfour statio

Fig.05:

Apte et al~1.5 timeenvironmare knowdevelopedcontributicontributirole of veburning, monitorinthese are necessary

Fig.02: Old B

eriod of 2008

nd PM10 ave

< 10m and

mbient air stlevels are wocreased emis

ber 2011, thons in Delhi.

PM2.5 Measu

l., (2011) highes the ambienents (Chowdn (Chhabra, ed a compreheion for the citions based onehicle exhauand industring sites, whenot for the c

y to assess the

Bus

8 and 2011, a

eraged 208 ±

PM2.5 refers

tandards for orse in the wisions from he

e daily avera

urements in D

hlighted the rnt levels for Phury et al., 2et al., 2001; Pensive emissity as a wholen establishedst emissionsal emissions ere the sourccity as a wholeir impacts on

Fig

at the seven m

± 137 g/m3.

s to particula

PM10 and nter months eating, and m

age PM2.5 wa

December 201

risks of exposPM2.5. Enrich2007; ShridhaPande et al., ions inventor. The Central PM pollution, re‐suspensiand discuss

ce apportionmle. Knowledgn the ambien

- 2 -

g.03: New Bu

monitoring st

. PM10 refer

ate matter wi

PM2.5 are 6with concentmeteorologica

as 267 ± 105

11 Fig.06

sure to air pohed trace eler et al., 20102002; HEI (2ry for the crit Pollution Con source appion of road sed an emissment sampline about the snt pollution le

SIM

us

tations in the

rs to particu

ith aerodynam

60 g/m3 antrations at leaal conditions

5 g/m3 and P

6: PM10 Meas

ollutants on thements are al0). While the 2004 and 201teria pollutanntrol Board (ortionment mdust, diesel sions inventong was condspatial spreadevels and hea

M‐air Working P

Fig.04: M

e city PM2.5 a

late matter

mic diameter

d 40 g/m3ast double the(Guttikunda

PM10 was 36

urements in

he roads of Dlso reported health impac0)), relativelynts and invesCPCB) of Indimethodologiegenerator seory for areasucted (CPCB,d of emissionlth (Johnson e

Paper Series 38

Metro System

averaged 123

with aerodyn

r < 2.5m, an

respectivelye annual averand Gurjar, 2

68 ± 116 g/

December 20

Delhi, which cin urban andcts of air poly few studiestigated the sia published ses, highlightinets, domestics surroundin, 2010). Howns across the et al., 2011).

8.2012

m

3 ± 87

namic

nd the

y. The rages, 2011).

/m3 at

011

can be d rural lution s have ource sector ng the c fuel g the wever, city is

Emissions Inventory Analysis for Delhi

- 3 -

Sources of Air Pollution in Delhi In this paper, we present results of a bottom‐up emissions inventory, an analysis of source‐receptor relationships for PM10 and PM2.5 for residential and industrial areas in the region, and implications of sector based interventions on the air pollution control policy, for the city of Delhi. The study domain, presented in Fig.07 covers Delhi and its satellite cities – Gurgaon, Noida, Greater Noida, Faridabad, and Ghaziabad, between 76.85°E to 77.65°E longitude and 28.2°N to 29.0°N latitude. Also in the figure are the two ring roads, main highways, locations of brick kilns (mostly located north and east of the city limits), power plants, and some points of interest. We compiled an emissions inventory for base year 2010 based on fuel consumption data and emission factors for transport, industrial, and domestic sectors. We used the activity based method to estimate the emissions inventory for PM, SO2, NOx, CO and VOCs is estimated using a well established activity based methodology. The same method has been used for building similar regional and urban inventories (Timilsina and Shrestha, 2009; Zhang et al., 2008; Kan et al., 2004; Tung et al., 2011; EEA, 2002 and 2010; Ramachandra and Shwetmala, 2009; Reddy and Venkataram, 2002; Garg et al., 2006; Singh et al., 2008; Baidya and Borken‐Kleefeld, 2009; Majumdar and Gajghate, 2011). Recently, similar methodology was utilized in India for the cities of Delhi, Mumbai, Pune, Bangalore, Kanpur, and Chennai for the base year 2006‐07 (CPCB, 2010). We acknowledge the uncertainties involved in the use of available emission factors instead of us calculating emission factors for every activity which is an expensive and time consuming process. Our method allows us to better understand the pollution sources in the city and the accordingly formulate policy to improve air quality (Schwela et al., 2006; Johnson et al., 2011).

Vehicle Exhaust We used the ASIF methodology by Schipper et al., (2000) to calculate vehicular emissions. In this method total travel activity (A) and modal shares (S) describe how much people travel by mode (in vehicle‐km traveled per day), modal energy intensity (I) represents energy use per kilometer, and the emission factor (F) is the emitted mass per vehicle‐km travelled. Table 1 lists a summary of registered vehicles in the NCR. The types of vehicles included are passenger cars (30%), taxis (<1%), 2‐wheelers (motorcycles and scooters, 61%), 3‐wheelers (<1%), buses (urban and inter‐state, ~2%), and multi‐utility and commercial vehicles (~4%). We obtained passenger travel statistics from MoUD (2008). The Ministry of Urban Development (MoUD, Government of India) studied these patterns in 30 big, medium, and small scale cities in India. The age mix of on‐road vehicles is calculated using data from the “pollution under check” program, under which all passenger and para‐transit vehicles are required to undergo emission tests and receive an inspection and maintenance certificate. We did not utilize the emission rate results from these tests, as they are based free‐acceleration tests conducted along the road‐side for compliance

Fig.07: Study domain over the NCR Delhi

76.9 77.0 77.1 77.2 77.3 77.4 77.5 77.628.2

28.3

28.4

28.5

28.6

28.7

28.8

28.9

29.0

Rohini

Faridabad

Greater Noida

NOIDA

South Delhi

Satellite cities

PP ‐ Dadri

PP = power plants

PP ‐ gas based

Brick kilns

and do n(2007) anwhich meintegrated

Ca

M

Tw

LM

Bu

HD

LM

To

So

M(a(c)pecainc

Road D Re‐suspenmost Asia30% to ththe trans2004; Srivfew monempirical States, Eu2003; Gil2011; Beron roads suggests (this is ho(Apte et agrams perand 10.5 and mix o

ot include a nd CPCB (201easured emisd with DIESEL

Table 1: Num

ars

MUV (Passenge

wo Wheelers

MV (Passenge

uses d

DV (Goods) e

MV (Goods) f

otal

ource: Ministry

UV = multiple ) includes jeep) includes threer trip, and locaarriers; (e) inclucludes three a

ust

nsion of roadan cities (Johhe ambient PMport corridorvastava et alnitoring and functions th

urope, and Chlies et al., 2rger and Denbusing the USits applicatioowever 2 timal., 2011)). Thr vehicle km tfor main roadof vehicles, sil

full driving c10). In this wsions factorsL (2008) and G

mber of regis

M

er) a b

er) c

y of Road Trans

utility vehiclesps, vans, and spe and four wheal and radio taudes trucks, lornd four wheele

d dust is a signson et al., 2M10 concentrs and residel., 2009; Shridsource app

at estimate rhina (Abu‐Alla005; USEPA, by, 2011). WeEPA AP‐42 mon for averagmes more thahe average rotraveled for fds, are calcult loading, and

cycle. We usework, we mos between CNGAINS (2010)

stered vehiclecapital re

March, 2007

1,536,900

77,850

3,377,100

193,300

127,850

87,800

91,700

5,493,000

sport and High

s; LMV = light mport utility veheeler passengexis; (d) includerries, ambulaner goods vehic

gnificant sour2011) and corations in Deential areas (dhar et al., 2portionment re‐suspensionaban et al., 202006; EEA,

e estimated rmethodology (ge road speedan the obserad dust re‐sufeeder roads, ated based od vehicle spee

- 4 -

ed emission fstly refer to tNG, petrol, an.

es in 2007, 20egion of Delh

March, 200

1,668,90

78,75

3,616,45

212,00

129,00

89,10

105,35

5,900,00

ways, Governm

motor vehicle; icles; (b) includer vehicles, whes public transpces, and policeles.

rce of air polntributes as lhi, particular(Balachandra2010; CPCB, 2studies est

n rates in the003; Etymezia2009; Amatore‐suspension(USEPA, 2006ds less than rved speeds uspension rate5.8 for arter

on the vehicleeds, on each

SIM

factors for ththe latest stund diesel bas

008, 2009, anhi, India

08 March,

00 1,80

50 7

50 3,84

00 23

00 13

00 8

50 12

00 6,30

ment of India (M

HDV = heavy ddes mopeds, schich can carry tport, contract, e and other sta

lution in much as rly along an et al., 2010). A tablished e United an et al., o et al., n of dust 6), which 55 mph in Delhi es of 4.8 ial roads, e density of these

M‐air Working P

he Indian vehudies of Reynsed 3‐wheele

nd 2010 in the

, 2009 M

02,300

78,900

46,750

30,050

30,550

89,650

24,200

03,000

(MoRTH, 2011)

duty vehicle; cooters, and mthree to seven school, and prate owned veh

Fig.08: D

Paper Series 38

hicle fleet bynolds et al. (ers on Delhi r

e national

March, 2010

1,946,450

81,300

4,126,500

244,750

134,850

92,250

141,600

6,768,000

)

motorcycles; passengers rivate stage hicles; (f)

Dust on roads

8.2012

y ARAI (2011) roads,

s

Emissions

road typeNew Delhprimarily passenge

Industri NCR has sand pharm(in the ealocated inNew Delhfuel oil (3previous s The constcase of Nblack dots~25,000 bnorth of tdominateemployedbull trenctechnologSimilar kilBangalorethis sectoused the SO2; 2.5 fobserved primarily construct

Inventory Ana

es. We used vhi, India). Thtravel on thr cars plying o

ial and Con

several commmaceuticals ost), Janakpurn these clustehi, India. The t35%), gas (7%studies (Redd

truction sectoCR, the bricks in Fig.07 shbricks per dathe outer ringed by small ind on a seasonch kilns that gies, as Hoffmlns are founde, Chennai, aor based on bfollowing emfor NOx; 64.0 at the brick due to differion activities

lysis for Delhi

vehicle densithe density fige highways on a mix of ro

nstruction

mercial areasoperate. Thesri (in the westers and the lototal energy %), and diesedy et al., 2002

or is rapidly k kilns are locow the locatiy, using a mig road containdividual openal basis (Gupare more p

mann, high dr in most of Nnd Hyderabarick producti

mission factorsfor CO; and kilns in Dhakarences in fueto estimate r

F

ty figures frogures accounand ring roaoads.

Activities

where smeltse clusters (Fit), and Gurgaocation informconsumptionel (33%) (GAIN2; Gurjar et al

growing in Icated just ouion of ~1,000ix of coal andins most of therators, eachpta, 2003). Molluting and raught or vertNorthern Indiaad (in South Ion rates for ts (in grams p420 for CO2 a, operating l mix (World re‐suspension

Fig.09 Brick k

- 5 -

om studies bynt for type ods and the l

ters, tanningig.07) are locon (in the somation is gatn is approximaNS, 2010; CPl., 2004; GAIN

ndia. This inctside the Del0 kilns in the sd biomass. The kilns (~60%h consisting oMost of the inenergy‐inefftical shaft bria, along the Inndia) (Isebellthe manufacter brick prod(Maithel et aunder similarBank, 2007;

n of dust that

kilns around N

y the Central of vehicle, folight duty co

g, textiles, macated in Fariduth). SoE (20thered from tately ~43 PJ, PCB, 2010). WNS, 2010; CPC

cludes brick lhi city limitsstudy domainThe area cove%). Brick manof 200 to 300nstallations aficient as comick kilns (CAI‐ndo‐Gangeticle et al., 200turing seasonduced) of 3.4 al., 2012). Thr conditions aGuttikunda

t has an impa

NCR Delhi

Road Researor instance hommercial ve

anufacturing, dabad (in the 010) lists ~6,0the EICHER mfueled by co

We used emiCB, 2010).

and cement , mostly alonn, with a prodered by the bnufacturing in0 daily wageare conventiompared to th‐Asia, 2008; Wc plain, and a07). We calcun (non‐monsofor PM2.5; 4.ese rates areand utilizing set al., 2012).ct on PM em

rch Institute (heavy duty tehicles, buses

chemicals, psouth), Ghaz

000 individualmapping divisal and coke (ssion factors

manufacturing the borderduction capacblack boxes dn northern Ine workers peronal fixed chihe newer, clWorld Bank, 2round the citlated emissiooonal months6 for PM10; 1e lower than similar techno. We also incissions.

(CRRI, trucks s, and

paper, ziabad l units ion in (25%), from

ng. In r. The city of drawn ndia is r kiln, imney eaner 2010). ties of ons of s) and 1.5 for those ology, cluded

Electrici Six majortheir locaplants locoutside (FMW, with(Kansal et

While moFaridabadcapacity cinemas, consumptdiesel pemonth. Th

Emissions

ity Genera

r power planttion, fuel concated in the Faridabad, Bah an average t al., 2009).

ost of the ed, and Ghaziagenerators iand farm hoution. For examr month andhe total diese

s factors for p

ation

ts are locatensumption, ancity (Indrapradapur, and fuel consum

Fig.10 Powe

lectricity neeabad, supplein hotels, houses are a somple, a five‐sd a campus lel consumptio

power plants a

d within the nd flue gas chastha, RajghaDadri) are coption rate of

er Plants in th

eds are met ement their pospitals, malource of emissstar hotel or ike the Indiaon in the in‐si

Fig.11 Diese

and diesel ge

- 6 -

modeling doharacteristics at, and Pragaoal based. Th800 kg of co

he modeling d

by the powpower needsls, markets, sions. These a big hospitaan Institute otu generator

el generators

nerators are

SIM

omain of thisis presented ati) are natuheir combineoal and 250 m

domain of NC

wer plants, as using in‐sitlarge instituare estimateal is estimateof Technologysets is estima

s in use

from GAINS (

M‐air Working P

s study (Fig.0in SoE‐Delhi ral gas baseded generationm3 of natural

CR Delhi

reas such asu diesel genutions, aparted using on‐sid to consumy consumes ated at ~60 P

(2010).

Paper Series 38

07). A summa(2010). The pd and while n capacity is gas per MW

s Gurgaon, Nerator sets. tment compite surveys foe ~30,000 lit~80,000 literPJ.

8.2012

ary of power those 2,700

W‐hour

Noida, Large lexes, or fuel ers of rs per

Emissions

Domest We estimhas informemissionswith highmix of fu(Novembecoal, andheating fr

Waste M There areprocessinat ~9,000residentiaemissionsdistrict lecollection

Inventory Ana

tic Sector

ate domesticmation on ms from cookin density areaels including er to Februa wood. We rom Zhang et

Managem

e three activeg capacity of 0 tons per daal areas, colles are estimateevel. The secn rates (World

lysis for Delhi

c emissions frmix of fuels ang activities inas utilizing mcoal, biomasry) the domeobtain activit al. (1999), Bh

Fig.

ment Sector

e landfills in ~5,000 tons

ay (SoE‐Delhi,ection sites, roed for varyinctors with thd Bank, 2006;

rom the distrand daily usento urban andostly liquefiess, cow dungestic sector aty based emhattacharya e

12 Winter he

r

Delhi at Okhper day. How 2010). A pooadside sitesg collection ehe highest po CPCB, 2010)

Fig.13 L

- 7 -

ribution of poe at the sub‐d rural areas ed petroleumg, wood, keroalso includes missions factoet al. (2000),

eating and wa

la Phase I, Jhwever, the totortion of the , and in someefficiencies bopulation de.

Landfills in De

opulation in t‐district level based on the gas (LPG) anosene, and Lemissions frors for variouZhang et al. (

aste burning

hangir puri, atal garbage gcollected gae cases, at thbased on the ensity often

elhi

the city. The (IFMR, 2011e population nd low densitLPG. During tom heating tus fuels used(2000), and G

and Ghazipur,generation forbage is reguhe landfill sitepopulation dexperience t

2001 Census1). We segredensity, per ty areas utilizthe winter mthat uses biod for cookingGAINS (2010).

, with a comr NCR is estimularly burnt ie. Garbage budensity at thethe highest w

s data gated grid ‐ zing a onths ofuels, g and

bined mated in the urning e sub‐waste

Air Traf The domefor a weealso includomestic passenge

Emissio A summaTable 2, 133,900 t(CO), and

Table 2:

Transport (T

Domestic (D

Diesel Gen S

Brick Kilns (B

Industries (IN

Construction

Waste Burni

Road Dust (R

Power Plant

Total

ffic

estic and intek was collectdes emissionterminal andr cars.

ons Invento

ry of the totaalong with stons of PM10332,700 tons

An activity b

PM

R) 17,

OM) 7,3

Sets (DG) 3,2

BK) 9,2

ND) 9,0

n (CON) 2,4

ng (WB) 3,8

RD) 6,3

(PP) 10,

69,

rnational airped from the fns from bus od a fraction

ory and Se

al emissions rsector contri0, 37,000 tonss of volatile o

ased emissio

M2.5 P

,750 (26%) 2

300 (11%) 8

200 (5%) 4

250 (13%) 1

000 (13%) 1

450 (4%) 8

850 (6%) 5

300 (9%) 4

,150 (15%) 1

,050 1

ports in Delhiflight status inoperations foof idling em

Fig.14 Fligh

ector Contr

resulting for yibutions. Tots of SO2, 492organic compo

ons inventoryDelhi,

PM10

23,800 (18%)

8,800 (7%)

4,300 (3%)

12,400 (9%)

12,650 (9%)

8,050 (6%)

5,450 (4%)

41,750 (31%)

16,850 (13%)

133,900

- 8 -

operate ~65nformation aor shuttling pmissions at th

ht statistics fo

ributions

year 2010 foral emissions ,250 tons of ounds (VOCs)

y (tons/year) , India, in 201

SO2

950 (3%)

2,050 (6%)

1,050 (3%)

4,000 (11%)

8,500 (23%)

100 (1%)

250 (1%)

20,250 (55%)

37,000

SIM

0 flights dailyvailable in thpassengers tohe arrival and

or Delhi

r the Nationa are estimatNOx, 1.52 m).

by sector for10

NOx

329,750 (67%

2,350 (1%)

81,300 (17%

6,750 (1%)

41,500 (8%)

2,150 (1%)

1,450 (1%)

27,200 (6%)

492,250

M‐air Working P

y. The landinghe public domo and from td departure

l Capital Regted as 69,05illion tons of

r the national

CO

%) 421,450 (2

161,200 (

%) 85,100 (6%

171,850 (

219,600 (

2,700 (1%

20,050 (1%

442,150 (2

1,524,050

Paper Series 38

g and take‐offmain. The invehe aircrafts asections from

ion is present50 tons of PCarbon Mon

l capital regio

VOC

28%) 208,900

11%) 18,300

%) 31,600

11%) 24,200

14%) 13,250

%) 50 (1%)

%) 1,600 (

29%) 34,900

0 332,700

8.2012

f data entory at the m the

ted in M2.5, noxide

on of

0 (63%)

(6%)

(9%)

(7%)

(4%)

)

%)

(10%)

0

Emissions

Overall, rothe vehicldust (31%industriesthe domin(15% and 11%) resppower pla In the traand light dexcept foDuring thduty vehithe benefto CNG basales of ddiesel basthe diesel The six pototal SO2 for in‐situthe hotelapartmenof PM2.5

emissionsemissionspopulated The brickDue to thproductiothe vicinitannual PFollowingallowed tevident frborder anDelhi. The emissfor dispediurnal cythe rush hours forhours for

Inventory Ana

oad transportle exhaust or%) and vehicls (9%), and bnant sources,29%), brick kpectively. Forants (55%) an

ansport sectoduty trucks isor some lighe CNG shift cles were retfits of conveased vehicles diesel based psed passengel subsidy prog

ower plants i(54%) and COu power gens, hospitals, nt complexes,. Although, s is small, whes are substd areas.

k kilns operathe recent boon rates and tty of the cityPM2.5, 11% og an ordinancto operate wrom Fig.07 thnd a major co

sion inventorersion modeycles for the tand non‐ru

r the industrthe domestic

lysis for Delhi

t sector remar indirectly frole exhaust (1rick kilns (9%, with substankilns (13% anr NOx emissid industries (

or, freight ms the largest cht duty vehiin the early trofitted to orting the busare now lostpassenger veer vehicle sagram (SIAM, 2

n the modeliO (29%) emisseration fromlarge institut, is estimatedthe overa

en spatially setantial, esp

te only betwoom in the cthe number y and estimatof CO and ce by the Supwithin the cihat these kilnontributor to

ry also includel‐ready inpuransport sectsh hours forial sector, ac sector.

ains one of thom the re‐sus18%) remain %). For PM2.5 antial contribud 11%), indusons, vehicle (23%), the lar

movement viacontributor. Micles, operat2000’s, someperate on CNses, taxis, and, due to an inehicles. An incles is primar2012).

ng domain, dsions. Diesel

m diesel genetions, marketd to contribull percentagegregated, thecially in t

een Novembconstruction of kilns haveted to accoun11% of SO

preme Court, ity limits. Hos are locatedthe air pollu

des sector‐speut preparatiotor to distingur all modes,and cooking

- 9 -

he major sourspension of dthe major soand CO emissutions from vestries (13% aexhaust (67%rgest coal use

a heavy dutyMost of thesee on diesel.e of the lightNG. However,d 3‐wheelersncrease in thecrease in therily driven by

dominate theconsumptionerator sets ints, malls, andute about 5%ge of thesehese low lyingthe densely

ber and May.industry, the increased innt for 13% ofO2 emissions.kilns are notowever, it isd close to thetion levels in

ecific profileson; such asuish between, operationaland heating

rces of all thedust on the rource, followsions, diesel ehicle exhausnd 14%), and%) remains ters in the regi

y e, . t , s e e y

e n n d % e g y

. e n f . t s e n

s s n l g

Fig.

e emissions, eoads. For PM

wed by the pand biomassst (26% and 2d the domestthe dominanton, are the d

.15: Sector C

either directlyM10 emissions,ower plant ( based secto28%), power pic sector (11%t source. Forominant sour

Contributions

y from , road (13%), rs are plants % and r SO2, rces.

s

Previou An earlier(2007) for(68%) andsector. Adand solid (2004) wathe powepassenge The Minissource ap2010). Thper day odifferencemain cityinventoriethe sourcareas suractivities.clusters lcontributiinventory A numberconductedKhillare eSrivastava2007b; SrChelani et2004; ShrPM pollutdust, vecombustioaerosols, among otkey contrCPCB (201in the pascontributivaried froprovide amonitorinthe expercomplime

us studies

r emissions inr the time ped total suspendditionally, agwaste disposas much smaler plants locar and comme

stry of Enviropportionmentis study for Dof SO2. Since, es in the met, covering anes were prime apportionmrounding the In case of Deocated ~20 ion to air poly for the regio

r of receptor d in Delhi (et al., 2004;a and Jain, 2rivastava et at al., 2010; CPridhar et al.,tion in Delhi ehicle exhaon, open rand constructher sources,ributor to PM10), which nost (Pant and Hions estimateom study to a scientific bng sites. Howeriment for maentary (Johnso

nventory for riod of 1990 tnded particlegriculture wasal the main sller than the ated in Delhiercial activitie

onment and t study in six Delhi estimatethese figuresthodologies. Cn area of 32 arily estimatement study ane main districelhi, the largekm away frolution concenon at 1.67km x

modeling stu(Balachandra; Chowdhury2007a; Srivaal., 2009; TiwPCB, 2010; M 2010). Key have been idaust, coal refuse burnction activitieLPG gas wasM2.5 in resido other studieHarrison, 2012ed from the rstudy, depenackground aever, the rigoany locations,on et al., 201

Delhi, preparto 2000, conces (80%), whias the major csource of meone used in t have been s has at least

Forests (MoEcities: Delhi, ed ~147 tons s are below tCPCB (2010)km x 32 kmed for an areand then extrapct where the est missed coom the city ntrations ovex 1.67km grid

udies have ben et al., 20y et al., 20stava and Ja

wari et al., 20önkönnen et contributorsentified as roand bioming, secondes. Interestings identified aential areas es have report2). Although treceptor modnding on thend a methoorous method, in which cas1).

- 10 -

red by Gurjarcluded that ple NOx, CO acontributor foethane CH4 (8the current sconverted frt doubled (SoE

EF) of India aKanpur, ChePM10 per dathe estimatesis for the bas; which is ~2a of 2 km x 2polated to thindustrial ac

ontribution sedistrict bouner the whole d resolution, a

een 000; 007; ain, 009; al., to oad ass ary gly, as a by ted the dels e season andd to validatedology is finase the source

F

SIM

r et al. (2004)power plants and VOC emisor ammonia 80%). The mostudy; anotherom coal to CE‐Delhi, 2010

and CPCB, caennai, Mumbaay, ~460 tonss from our stuse year of 2025% of the a km around te city districtctivity is geneeems to arise ndaries (Fig.0NCR. Sahu etalso fails to d

location of e the sourcencially and tee modeling st

Fig.16: Recep

M‐air Working P

) and updateare the primassions came f(70%) and niodeling domaer change is tCNG, and the0).

arried out paai, Pune, and per day of Nudy, it is imp006‐07 and rearea covered the monitorint area. This doerally larger t from brick k07), yet theyt al. (2011) pistinguish the

monitoring se‐receptor reechnically demtudies such as

ptor modelin

Paper Series 38

ed by Mohan ary sources fofrom the trantrous oxide (ain in Gurjar hat since 200e vehicle flee

articulate pold Bangalore (NOx, and ~268portant to notepresents onin our study

ng site selecteoes not includthan the in‐diln emissionsy make a deublished a gre brick kiln se

site, these stelationships amanding to rs this study c

g schematics

8.2012

et al. or SO2 nsport (50%), et al., 00‐05, et for

lution CPCB, 8 tons te the ly the y. The ed for de the istrict , with efinite ridded ector.

tudies at the epeat can be

s

Emissions

Spatial We develfurther usin each demissions

For the paverage p We used depots, trapartmenbased poemissions Emissionsindustriesscattered Four panekilns), doand const

Inventory Ana

Distributio

loped the emse in atmosphdirection at 0s for each sec

Fig.17: Schem

population depopulation de

GIS data fromraffic signals, nt complexes,opulation dens on feeder, a

s from powers, brick kilns,garbage burn

els for griddemestic (inclutruction activ

lysis for Delhi

on of Emis

missions inveheric modelin.01° resolutiotor on the gri

matics utilize

ensity maps, ensity is more

m EICHER (Neand landmar, large institunsity and vehrterial and m

r plants were, and generaning emission

ed PM10 emisding heating,vities) are pr

ssions

ntory on a Gng. Hence we on, correspoid ‐ summariz

ed for spatial

we used da than 5,000 p

ew Delhi, Indiarks), industrietions, and fahicle density

main roads.

directly assigator sets wens were distri

ssions (in ton, waste burnresented belo

- 11 -

GIS based plasubdivided tnding to 1 kmzed below.

segregation

ata from GRUper km2 in the

a) to map roaes, brick kilnsrm houses. Insurveys con

gned to theirere spatially buted using t

ns/year/grid) ing, and winow. In case

atform to spahe domain unm. We used

of total emis

UMP (2010)e main distric

ads (includings, hotels, hospn case of thenducted by C

r respective lallocated accthe populatio

for vehicles,ter heating),of vehicle ex

atially segregnder study (Fspatial proxi

ssions over N

at 30” spatits.

g informationpitals, marke transport seCRRI (New D

ocations, whcording to ton density and

industries (i and dust (inxhaust, the h

gate emissionFig.07) into 80ies to allocat

CR Delhi

ial resolution

n on bus stopts, malls, cineector, we useDelhi) to distr

ile emissionsheir locationd land‐use da

ncluding thencluding roadhighest dens

ns for 0 cells te the

n. The

s, bus emas, d grid ribute

s from n. The ata.

brick d dust sity of

SIM‐air Working Paper Series 38.2012

- 12 -

emissions is observed along the major road intersections, which is linked to the density of the feeder roads in the grid. A combination of vehicle, road, industrial, and population density is used to assign weights for congestion emissions. For convenience, only PM10 emissions are presented below. However, gridded fields were also developed for PM2.5, SO2, NOx, CO and VOC.

Fig.18: Schematics utilized for spatial segregation of total emissions over NCR Delhi

While it is important to know the footprint of total emissions for a city, it is also important to understand the hotspots and their pollution loads in residential and industrial regions. Table 3 presents the percentage contributions by sector to total annual PM2.5 emissions for six select regions evidenced in Figure 3. Note that the other pollutants are not presented in this graph. The regions are selected with equal areas, so that we can compare the mix of activities in the main district and in the satellite cities. Gurgaon (GURG) and West Delhi (WDEL) experience the most in terms of the total emissions, due to the large contribution of industries (>40%). Industries account for (35%) in nearby Faridabad (FARD). In the industrial regions, vehicular emissions are attributed to movement of heavy duty and light duty trucks.

76.9 77.0 77.1 77.2 77.3 77.4 77.5 77.628.2

28.3

28.4

28.5

28.6

28.7

28.8

28.9

29.0

(a)

0.00 to 0.010.01 to 0.100.10 to 1.001.00 to 5.005.00 to 50.0050.00 to 400.00

76.9 77.0 77.1 77.2 77.3 77.4 77.5 77.628.2

28.3

28.4

28.5

28.6

28.7

28.8

28.9

29.0

SDEL

FARD

NEBK

WDEL

GNOD

GURG

(b)

0.00 to 0.010.01 to 0.100.10 to 1.001.00 to 5.005.00 to 50.0050.00 to 400.00

76.9 77.0 77.1 77.2 77.3 77.4 77.5 77.628.2

28.3

28.4

28.5

28.6

28.7

28.8

28.9

29.0

(c)

0.00 to 0.010.01 to 0.100.10 to 1.001.00 to 5.005.00 to 50.0050.00 to 400.00

76.9 77.0 77.1 77.2 77.3 77.4 77.5 77.628.2

28.3

28.4

28.5

28.6

28.7

28.8

28.9

29.0

(d)

0.00 to 0.010.01 to 0.100.10 to 1.001.00 to 5.005.00 to 50.0050.00 to 400.00


- 13 -

Table 3: Contributions to total PM2.5 emissions in select residential and industrial regions for year 2010 in Delhi, India

SDEL FARD GNOD NEBK WDEL GURG

Total PM2.5 (tons/year) 2,080 3,700 1,680 2,700 5,360 4,700

Transport 48% 35% 40% 10% 31% 27%

Domestic 7% 5% 16% 12% 5% 5%

Diesel Gen Sets 13% 6% 2% 6% 6%

Brick Kilns 8% 68% 1%

Industries 10% 35% 45% 43%

Construction 2% 3% 10% 1% 6%

Waste Burning 5% 4% 6% 4% 4% 3%

Road Dust 14% 13% 17% 5% 9% 10%

In South Delhi (SDEL), an area with high population density, vehicle exhaust emissions are the largest contributors (48%), while in less dense parts of the city like Northeast Delhi (NEBK), brick kilns contribute about 68% of the emissions. In addition, most of the vehicle exhaust emissions are related to freight movement, originating or ending either at the brick kilns or at the industries. Among satellite cities, Gurgaon and Greater Noida (GNOD) experience the most significant contribution from construction activities (6% and 10% respectively), also linked to the lack of paved roads in the area. Road dust re‐suspension and waste burning are uniformly present in all the regions. It is important to note that the road dust particles are predominantly in the coarse size range (particle diameter between 2.5

and 10 m), so the percent contribution to total PM10 emissions is much larger than those estimated for total PM2.5 emissions. The use of diesel generator sets is the highest in the densely populated areas (>6% in SDEL, WDEL, FARD, and GURG), due to the high number of hotels, hospitals, malls, and institutions. The industrialized areas (WDEL, FARD, and GURG) cannot meet all the demand for electricity and there are instances when large institutions operate on diesel generators for more than 12 hours a day.


- 14 -

Particulate Matter Concentrations We used the Atmospheric Transport Modeling System (ATMoS) dispersion model ‐ a meso‐scale three‐layer forward trajectory Lagrangian Puff‐transport model (Calori and Carmichael, 1999) to model PM concentrations for the base year 2010. This model was previously utilized in pollution management studies in Asia at a regional scale (Arndt et al., 1998; Streets et al., 2000; Guttikunda et al., 2001; Holloway et al., 2002; Carmichael et al., 2008) and urban scale (Kan et al., 2004; Guttikunda et al., 2003; Guttikunda and Gurjar, 2011). The ATMoS model is a modified version of the USA National Oceanic Atmospheric Administration, Branch Atmospheric Trajectory (BAT) model (Heffter, 1983). The layers include a surface layer, boundary layer (designated as the mixing layer height) and a top reservoir layer. The multiple layers allow the model to differentiate the contributions of near‐ground diffused area sources, like transport and domestic combustion emissions and of elevated sources like industrial, brick kilns, and power plant stacks. The model also includes first order chemical reactions for SO2 and NOx emissions to estimate the secondary contributions in the form of sulfates and nitrates, added to the total PM2.5 concentrations. The model has flexible temporal and spatial resolution and can run for periods ranging from one month to a year and from regional to urban scales. Input meteorological data for Delhi analysis (3D wind, temperature, and pressure, as well as surface heat flux and precipitation fields) is derived from the National Center for Environmental Prediction (NCEP) global reanalysis (Kalnay et al. 1996), interpolated on the model grid. The summer time mixing heights are estimated to be as high as 2,500 meters during the daytime compared to the lows of less than 100 meters in the winter months (Guttikunda and Gurjar, 2011).

Fig.19 Mixing layer height in 2008 over NCR Delhi, estimated from NCEP Reanalysis data (thick line indicates a 15 day moving average)

Mixing Height (meters)

0

500

1000

1500

2000

2500

Jan Feb Mar Apr May Jun Jul Jul Aug Sep Oct Nov Dec


- 15 -

Fig. 20: Wind speed and wind direction for the Delhi domain @ 6 hour interval, estimated from NCEP reanalysis data for year 2008

We present below the modeled annual average concentrations for PM2.5 and PM10. We ran the dispersion model by sector and then aggregated the total PM2.5 and PM10 bins. These totals include primary PM and contributions from chemical transformation of SO2 and NOx emissions in the form of sulfates and nitrates (Guttikunda et al., 2001; Holloway et al., 2002) and do not include any effects of long‐range transport from the regions outside modeling domain.

0

45

90

135

180

225

270

315

0% 2% 4% 6% 8%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

January

0

45

90

135

180

225

270

315

0% 4% 8% 12% 16%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

February

0

45

90

135

180

225

270

315

0% 4% 8% 12% 16%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5>5

March

0

45

90

135

180

225

270

315

0% 4% 8% 12% 16%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

April

0

45

90

135

180

225

270

315

0% 4% 8% 12% 16%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

May

0

45

90

135

180

225

270

315

0% 5% 10% 15% 20% 25%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

June

0

45

90

135

180

225

270

315

0% 5% 10% 15% 20% 25%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

July

0

45

90

135

180

225

270

315

0% 4% 8% 12% 16% 20%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

August0

45

90

135

180

225

270

315

0% 4% 8% 12% 16%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

September

0

45

90

135

180

225

270

315

0% 4% 8% 12% 16%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

October

0

45

90

135

180

225

270

315

0% 2% 4% 6% 8% 10%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

November

0

45

90

135

180

225

270

315

0% 2% 4% 6% 8% 10%

<=1

>1 - 2

>2 - 3

>3 - 4

>4 - 5

>5

December


- 16 -

Fig.21 Modeled annual average (a) PM10 and (b) PM2.5 concentrations over the NCR of Delhi

For most of NCR the estimated concentrations exceeded the national ambient standards for yearly

averages, 60 g/m3 for PM10 and 40 g/m3 for PM2.5. The resulting modeled annual average

concentration for the main Delhi district (the area inside outer ring road) is ~120 g/m3 for PM2.5 and

~160 g/m3 for PM10. For the regional boxes designated in Figure 3, the estimated annual average PM2.5

concentrations are ‐ 131 g/m3 for SDEL, 88 g/m3 for GURG, 50 g/m3 for GNOD, 89 g/m3 for FARD,

102 g/m3 for WDEL, 46 g/m3 around the brick kiln clusters in the Northeast (NEBK).

Comparison with Monitoring Data We also compare the modeled concentrations against monitoring data from six continuous monitoring stations (presented below). The monitoring data analyzed in this paper is from the continuous air monitoring stations operated by CPCB and the Delhi Pollution Control Committee. This data is from (1) Delhi College of Engineering (DCE, university campus) (2) Netaji Subhas Institute of Technology in Dwarka (NSIT, university campus) (3) Income Tax Office (ITO, traffic junction) (4) Shadipur (mixed residential, industrial, and traffic) (5) Institute of Human Behaviour & Allied Sciences (IHBAS, mixed residential and traffic) (6) Mandir Marg (mixed traffic and industrial) (7) RK Puram (mixed traffic and residential) and is available in the public domain. These stations measure SO2, NOx, and CO, not all stations measure both PM2.5 and PM10. Ozone and PM2.5 were added to the list of criteria pollutants in November, 2009. The uncertainties in measured and in modeled data are not undermined and these results are presented to ensure that the emission estimates and the dispersion model schematics are representative of the geographical and meteorological conditions prevalent in the region. All the stations measure only one fraction of PM (either PM2.5 or PM10) and the data collection efficiencies are lower than 40% in a year. So, the measurements are an average of all the data available during the period of 2009 to 2011. The stations at Mandir Marg (MM) and RK Puram (RK) both measuring PM2.5 are new and operational since April, 2011; while the station at the income tax office (ITO) is operational since 2006. However, for PM10, stations at IHBAS, Shadipur (SHAD), and in Dwarka (NSIT) are operational since 2009. The annual

average measured concentrations ranged 130 ± 92 g/m3 for PM2.5 at ITO; 89 ± 50 g/m3 for PM2.5 at

Emissions

MM; 91 ±

and 205 ±

Fig.22 Comont

The graphof the daithe monimeasuremthe modeobserved behaviorsunfavorabbetween emissionsinto grids,

Inventory Ana

± 51 g/m3 fo

± 171 g/m3 f

omparison of thly average

h, also presenily averages oitoring statioments are moeled concentrat the diffe

s, with maximble dispersionthe measures inventory es, and (c) unce

lysis for Delhi

or PM2.5 at RK

for PM10 at SH

PM10 and PMconcentratio

nts the variatover each moon for the mostly represenrations are inrent sites arema in the coln conditions ed and the mstimation (b)ertainty in the

K; 162 ± 97 gHAD.

M2.5 measuremons over a 3km

tions in the month for the mmodeled resntative of theherently gride reasonablyld season as during these

modeled num uncertainty e dispersion m

- 17 -

g/m3 for PM1

ments from sm x 3km area

measured andmeasurementults. It is ine monitoring l averages. Fry captured, aa combined months (Gutbers can be in the spatiamodeling resu

10 NSIT; 240 ±

six stations ina covering th

d the modelets and the van fact to belocation and rom the compas well the mresult of thettikunda andexplained fual disaggregatults.

± 155 g/m3 f

n Delhi with te monitoring

d concentratriation over 9e considered the surroundparison it seemain featurese increased ed Gurjar, 2011rther by (a) tion of the em

for PM10 at IH

the modeled g station

tions: the var9 cells surrou that the stding sources; ems that the s in their moemissions an1). The differuncertainty imissions inve

HBAS;

riation nding tation while levels onthly nd the ences in the entory

Source‐ Due to prthe sourcemissionsshows theinformatio3x3 grid cimpact fa

Fig.23becabric

winter

Area andexhaust (uopen washeight anwith a poclusters pkilns. The proportio

‐Receptor R

revalent metee contributios at the samee sector conton is presentcells, centeredctor (HEI, 201

3: Contributioause of emissck kilns (BK), r burning), ro

diffused soup to 50%) anste burning. Wd intensity oopulation denpresented in monitoring sn of vehicle e

Relationsh

eorological coons to ambiene location. Astributions to ed for six mod on the locat10), percent s

ons to total Psions from podiesel generad dust (RD),

(

urces contribnd followed bWhile the briof emissions, nsity of ~10,0Figure 1 andstations are oexhaust.

hips by Reg

onditions detnt concentrats presented ithe modeledonitoring statition of the stshares for onl

PM2.5 and PMower plants (ator sets (DG, vehicle exha(b) around th

bute the moby road dust, ck kilns are lthe effects a000 people pd yet experienoften located

- 18 -

gion

termining theions are not in Table 3 w ambient conions used fortation. For coy PM2.5 conce

M10 concentraPP), construc

G), waste burnaust (VEH) fo

he monitoring

ost to the PMdiesel generalimited to cerare felt fartheper km2, is ances 10% of closer to the

SIM

e role of eminecessarily thith total emincentrations r comparisonsnvenience anentrations are

tions over section activitiening (WB), do

or (a) six selecg stations

M2.5 concentator sets, andrtain pocketser from theirapproximatelyfine PM2.5 potraffic junctio

M‐air Working P

ssions from she same as thssions by regover the sams, using the mnd because ofe presented h

elect regions es (CON), induomestic (DOMct residential

rations, domd domestic ems of the city, r locations. Fy 30km awayollution origions and thus

Paper Series 38

surrounding ahose estimategion, Figure bme regions. Smodeled dataf the higher hhere.

in Delhi, Indiustries (IND),M, including and industri

minated by vemissions, incldue to their or example, y from any onating from capturing a h

8.2012

areas, ed for below imilar a from health

a ,

als

ehicle uding stack SDEL, of the these higher

Emissions

Implica For the n133,900 tVOCs; andsectors. Tthe deter From thisof the dayfrom the emissionsmaintenatraffic ma Our studyquality antanneriesMoreoverthe 6 millwith the gtechnologenforcem The passconsumptexposed t(Apte et transportspecial buareas, suc2010 CWG(UNEP, 20transport and an ex

Inventory Ana

tions to Ai

national capitons of PM10, d further spaThe monitoriniorating PM p

study (and py – pollution city and natios and pollutionce. This appanagement, w

y highlights thnd better pu, brick kilns, sr, it is easier lion vehicles growing popugies that reduent of an insp

enger and ction, emissioto the vehicleal., 2011). A. After only aus corridors focceeded as aG in Delhi we009), but servsector got a

xtension of th

Fig.24 Sp

lysis for Delhi

ir Pollution

tal region, th37,000 tons otially segregang data frompollution leve

previously puis an externaonal authoriton, implemenplies to all secwaste manage

he low hangiublic health. smelters, andto inspect anplying on theulation and cuce the emisspection and m

commercial vns, and harme exhaust anA number ofa 6 km pilot for the CWG ca good pilot fere not as strved as an exaboost from de metro lines

pecial lanes d

n Control i

he total emisof SO2, 492,25ated into 80x the stationsls in the regio

blished studility (a public ies – such as tation of necctors ‐ industrement, road m

ng fruit in teThe urban c

d metalworkinnd maintain 1e roads. Whiity size, a mosion rates at tmaintenance

vehicles are mful exposurend the road df interventionor bus rapid covering 80 kfor future BRringent as thoample for pubdoubling of fles.

uring the 201

- 19 -

in Delhi

ssions for 2050 tons of NOx80 grids at 0 and the dispon, from a mi

es), we knowbad) that cansetting emisscessary intervry, public tranmaintenance,

erms of policyclusters of sng shops acco1,000 brick kle relocation ore promisingthe brick kilnprogram.

responsible e. In dense tdust pollutionns are in platransport (BRm between vRT applicationose observedblic awareneseet, along wit

10 Commonw

10 are estimOx, 1.52 millio0.01° resolutiopersion modeix of sources.

w the areas thnnot be addresion and ambventions and nsport, perso, and domest

y changes thsmall‐scale mount for a largkilns or 6,000of industries

g approach ws and the ind

for an increraffic zones on, sometimesace or beingRT) corridor ivenues, Gamen. The intervd in Beijing duss on air qualth the introd

wealth Games

mated as 69,0on tons CO, anon for each oeling results

hat need actiessed withoubient standarenforcementnal transportic.

at can be mamanufacturersge portion of other industs proved benwould be to industrial boiler

easing portioof Delhi, largs double the g promoted fin 2008, the ies village, andventions introuring the 200ity issues in tuction of air

s and BRT in

050 tons of nd 332,700 toof the contribfurther emph

on, but at thut concerted ads, monitorint of inspectiot, freight tran

ade to impros, such as lef pollution in tries, comparneficial in thentroduce emers, followed b

on of the ege populationambient polfor the passimplementatd major residoduced durin08 Olympic Gthe city. The pconditioned

2008

PM2.5, ons of buting hasize

e end action ng the n and sport,

ve air eather Delhi. red to past, erging by the

energy ns are lution enger ion of ential ng the Games public buses

The para‐from the heavy duemission raw and emissionsdebris and Another mconstructas it depconstructlike wet vegetatioroad worpavementdepartmehas finish A sourcemanagemknowledgcollectionhas been conditionmanagemof peoplereduce th The electrdiesel genup new penergy, an It is oftenthat can bof sourcean operat(CWG), nsystems walert noticontrol ai In an effopresentedinconsistecorrected

‐transit sectorexpansion oty trucks areloads and expfinished pros, and this pod sand, which

major source ion activities.pends on theion sites, andsweeping

n in dry areark that oftents that are ent (telephoned their work

which neement and gage of best prn and managein adapting s. In Delh

ment is highly e, which meane garbage bu

ricity demandnerator sets apower plantsnd environme

n observed thbe made by b and receptotional air quaow maintainwhile advantaces, also pror pollution.

rt to continud in this paencies in the or suppleme

r, 3‐wheelersof metro by le banned froposure levelsoducts, constollution tend h is often carr

of pollution i. This source,e vehicle mod meteorologof street

as, paving roan time resuleft as is aftne, sewer, elk.

ds further aarbage burnractices to imement exists. these practichi and othlabor intensins we need aurning emissio

d in the domeand related ems in the regioent ministries

at emission ibuilding monitor modeling slity forecastined, updated,ageous for mevided better

ously improvper will be procedures,

ented with ad

s and taxis halinking the feom entering s during the druction debrto linger everied without a

n most Indian a large part ovement on ical conditionroads, promads and complts in ditcheter the conclectricity, and

attention is ning. Considmprove the The basic pr

ces to specificher cities, ive, and proma consolidatedons.

estic, commemissions. Thison and requis.

nventories artoring capacitudies. The eng system for, and validatega events likunderstandi

e the quality made availa such as emdditional data

- 20 -

ve also benefeeding lanes the city limitday time. Howris, sand, anden after the tany covers, te

n cities is roaof the coarsethe roads,

ns. However, moting pleting es and cerned d gas)

waste erable waste oblem c local waste

mises basic emd effort betw

rcial, and inds is a sector wires a consen

re never comty, continuedemissions invr NCR, launched by CPCB ke CWG, whicng of the inf

of the data, table via the ission factors as they beco

SIM

fitted from thto the residts between wever, the ded bricks, hastrucks have send to add to

d dust (CPCB,e PM10 pollutiroad types, this source c

mployment oween the com

dustrial sectorwhich cannot nted dialogu

mplete and red identificatioventory presehed before th(Guttikunda

ch benefited ffluencing sou

the emission internet as s and spatialome available

Fig.25

M‐air Working P

he expansiondential points6AM and 9Pensity of thess increased, stopped opero the silt load

B, 2010), incluion, is often dsilt loading

can be manag

opportunities mmunities and

rs is fueling tbe addressede between p

quire refinemon of pollutioented in this he 2010 Commet al., 2011from 48‐hoururces for long

inventory anan Excel fill weights fore.

5: Sweeping o

Paper Series 38

of their fleets. The commPM, to reduce vehicles caresulting in rating at 6AMing on these

uding that frodifficult to quon roads a

ged with mea

for large numd manageme

he need for id by simply sepower, petro

ments and upn sources, anpaper is in‐umonwealth G). These modr short term hg term planni

nd activity datle. The remar gridding, m

on roads

8.2012

ts and mercial ce the rrying more

M. The roads.

m the antify nd at asures

mbers ent, to

n‐situ etting leum,

pdates nd use se for Games deling health ing to

tasets aining ay be

sim-air working paper series

Documents