Urban Climate 4 (2013) 6173

Contents lists available at SciVerse ScienceDirect

Urban Climate
journal homepage: www.elsevier.com/locate/uclim

Short Communication

Trafc induced emission estimates and trends (20002005) in megacity Delhi

Ajay S. Nagpure a,, Ketki Sharma b, Bhola R. Gurjar c,1
a b c

Hubert H. Humphrey School of Public Affairs, University of Minnesota, Twin Cities, USA School of Civil and Environmental Engineering, Georgia Institute of Technology, 200 Bobby Dodd Way, Atlanta, GA 30332, USA Department of civil Engineering, Indian Institute of Technology Roorkee, Roorkee 247667, India

a r t i c l e

i n f o

a b s t r a c t
Transport sector emission inventory for megacity Delhi has been developed for the period 20002005 to quantify vehicular emissions and evaluate the effect of relevant policy reforms on total emissions of various air pollutants like CO2, CO, HC, NOx, TSP, SO2, Pb and VOCs over the years to assist in future policy formulations. Emission factor and vehicle utilization factor based approach as recommended by IPCC (2006) have been used for estimating emissions. CO level were found to increase continuously during the study period, other pollutants like CO2, TSP, NOx and SO2 declined in the initial years, which clearly seem to be the result of stricter emission norms and compressed natural gas conversion of public transport. The levels of NOx and TSP did not show appreciable rise during the study period, which is an indicator of CNG effectiveness as an alternative fuel. However, two-wheelers population were found to be a major contributor towards the air pollution load. Published by Elsevier B.V.

Article history: Received 3 October 2012 Revised 13 March 2013 Accepted 25 April 2013

Keywords: Emission inventory Transportation India Emission trend Air pollution

1. Introduction Air quality in urban areas is gradually leading to violation of ambient air safe limits. Megacity Delhi the National Capital Region of India is one of the most polluted cities in the world (WHO, 1992) having transport as major source of criteria area pollutants (Gurjar et al., 2004). It has been observed
Corresponding author. Tel.: +1 919 513 1804, mobile: +1 919 800 8194.
E-mail addresses: nagpureajay@gmail.com, anagpure@umn.edu (A.S. Nagpure), ksharma7@gatech.edu (K. Sharma), bholafce@iitr.ernet.in, brgurjar@gmail.com (B.R. Gurjar). 1 Tel.: +91 1332 285881; fax: +91 1332 275568/273560. 2212-0955/$ - see front matter Published by Elsevier B.V. http://dx.doi.org/10.1016/j.uclim.2013.04.005

62

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

that the automobile trafc has increased at a rapid rate in the urban areas during last decades in India. Annual growth rate of motor vehicles in India has been around 10% during the last decade. The most disturbing fact is their rising concentration in selected urban areas. About 32% of these vehicles are plying in metropolitan cities, which constitute 11% of the total Indian vehicle population (Singh, 2005) and responsible for degrading quality of urban atmosphere. In terms of emissions of various pollutants, Delhi was among the top ve SO2 emitting megacities of the world in early nineties and transport sector was the prime culprit for it (Garg et al., 2001). Mashelkar et al. (2002) state that the emission range of NOx from transport sector is 6674% in Delhi. CPCB (Central Pollution Control Board) (CPCB, 1995) data shows that almost 50% of the emission in Delhi is from vehicular activities, followed by domestic, industrial, and power plants. It was found that from 199091 to 199596, annual NOx emission from gasoline consumption increased from 3.5 to 4.5 Gg, respectively, whereas it was 8 Gg in 199091 and 12.8 Gg in 199596 from diesel (Sharma et al., 2002). According to Xie and Shah (2000), diesel driven vehicles were the major source of NOx emission in Delhi, whereas least contribution was from two- and three-wheelers. In terms of suspended particulate matter (SPM) concentration, Delhi ranks the fourth most polluted city in the world (Gadhok, 2000). Gurjar et al. (2004) and Reddy and Venkataraman (2002) estimated that 15% of Delhis RSPM (Respirable Suspended Particulate Matter) emission results from automotive trafc. Transport contributed about 19% to TSP (Total Suspended Particulate) emission in the year 2000 and almost doubled from 15 Gg in 1990 to 28 Gg in 2000 (Gurjar et al., 2004). According to ADB (Asian Development Bank) (ADB, 2005), diesel driven vehicles are the major contributor of PM emission among all vehicle categories in Delhi. As a consequence, incidence of respiratory diseases in Delhi is 12 times the national average, and 30% of Delhis population suffers from respiratory disorders (Kandlikar and Ramachandran, 2000). Its poor air quality is responsible for about 18600 premature deaths per year (TERI, 2001). Although there are other methods (e.g., inverse modeling), as suggested by IPCC (Intergovernmental Panel on Climate Change) (IPCC, 2006) using bottom-up approach based on vehicle numbers, their utilization factor and emission factors, present study carries out quantication of the vehicular emission loads in Delhi for the period of 20002005. The study aims to estimate air pollutant emissions from road transport and to determine the changing trends in levels of various pollutants (e.g., CO2, CO, NOx, HC, TSP, SO2, Pb, and NH3) in Delhi due to a combination of factors (e.g., rising population and increasing travel demand) combined with various policies and technological interventions introduced during the study period. This will help in establishing effects of reforms already introduced to assist the policy makers and modelers in developing future strategies to reduce and control urban air pollution in Delhi, India From time to time, Government of Delhi has implemented various norms and policies for reducing trafc emissions in megacity Delhi. In present study most of the norms and policies are considered indirectly due to the use of country specic emission factors and policies oriented vehicle population. Following are the norms and policies, which considered for emission estimations 1. 2000: India 2000 (Euro-I equivalent) norms for all 4 wheelers Bharat Stage-II (Euro-II equivalent) norms for Passenger Cars & MUVs. 2. 2001: Bharat Stage-II (Euro-II equivalent) norms for Commercial Vehicles in Delhi & Kolkata (from 24th October, 2001). 3. 20012002: Introduction of clean fuel CNG (Compressed Natural Gas) and phasing out of old age vehicle. 4. 2005: Bharat stage-III emission norms for all categories of 4 wheelers implemented in 11 megacities. However, since there are certain limitations related to baseline data and methodology adopted in this study, the present emissions estimates have further scope to be revised as and when more precise vehicle population data and emission factors are available for Delhi. The introduction is followed by methodology of the emission estimation based on vehicular population, vehicle utilization factor and emission factor. In results and discussion, trends of emission estimates are analyzed and comparison is made with air quality measurements and ndings of similar studies. There is a brief section on the limitations and scope of the present study followed by conclusions.

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

63

2. Methodology Emission factors and activity-based approach recommended by IPCC (2006) have been used for the calculation of vehicular emissions from the road transport sector in Delhi (Eq. (1)). A spreadsheet model (Microsoft Excel 1997/2003/2007) has been used for calculating emissions from transport sector using Eq. (1).

Ei

X V ehj Dj Ei;j

km

Ei = Emission of pollutant (i), Vehj = Number of vehicles per type (j), Dj = Distance travelled by per vehicle in per year or vehicle utilization factor (j), Ei,j km = Emission of pollutant (i) from vehicle type (j) per driven kilometer. 2.1 Vehicle population Population data of different vehicle types (e.g., Bus, LCV, HCV, Texi Car, etc.) for Delhi is taken from the Economic Survey of Delhi (20032004). CNG vehicle population data for the study period has been taken from Khaiwal et al. (2006). It is noticeable that the number of vehicles registered in Delhi is not the actual gure plying on Delhis roads. A large number of vehicles commute daily from nearby towns. In addition, large number of goods vehicle pass through the city at night; none of them running on CNG and hence contributing signicantly to the rising air pollution level in Delhi. The data for external trafc (including trafc originating from the city, terminating within the city and passing through the city) was adapted from the Central Road Research Institute (CRRI) study for 2002 (Mashelkar et al., 2002) and projected for other years. The two-wheeler population was categorized into 2stroke (2S) and 4-strokes (4S). Since 4S two-wheelers emit less pollution load in comparison to 2S two-wheelers, for studying the relative advantage of 4S two-wheelers over 2S two-wheelers we have taken 8:1 ration of 2S4S two wheelers to calculate their respective numbers as per Biswas (2006). The goods vehicles population was similarly divided into Light Commercial Vehicles (LCVs) and Heavy Commercial Vehicles (HCVs) by considering the dividing ratio given by Bose (1999). The age wise distribution of Delhis vehicle population was taken from CRRI-2002 study (Mashelkar et al., 2002). The distribution was assumed to be constant from 2001 to 2005 and was not applied for compressed natural gas (CNG) vehicles which were assumed to be in 05 years age group since the conversion of public transport eet took place after 2001. In 2002 Delhi government strictly implemented phasing out program for all commercial vehicles (e.g., Auto Rickshaw, Buses, LCVs and HCVs) and cancelled their registration due this a signicant decrease in these vehicle population have been observed between 2001 and 2002. Also because of this policy demand of new commercial vehicles was unexpectedly increased between 2002 and 2003.The population of vehicles for different years considered in this study are given in Table 1. The total population has been estimated for entire period taking into account the external trafc with an assumption that percentage of external trafc in 2002 is constant for all the years. 2.2 Emission factors An emission factor is a representative value that attempts to relate the quantity of a pollutant released to the atmosphere with an activity associated with the release of that pollutant. For transport sector, these factors are usually expressed as the weight of pollutant divided by a unit distance traveled by vehicle (g/km). It was hard to nd the emission factors taking into account Delhis driving conditions; control measures for different vehicle categories, and a range of pollutants (CO2, CO, HC, NOx, TSP, SO2, Pb and VOC) from a single source. Further, several of the earlier studies had simply taken the norms introduced as the emission factors, which result in gross under estimation keeping in mind the actual driving conditions on Delhis roads. Therefore, to make a more realistic study, emission factors were compared from range of sources and nally compiled as given in Table 2 and used those representing Indian conditions. For calculating per year emissions from transport sector, we have used average distance traveled by different vehicles. As expressed in Eq. (1), number of vehicles types (Vehj)

64 Table 1 Vehicular population in Delhi. Vehicle types (j) Cars/taxi-total Cars/taxi-CNG Cars/taxi-other 2 Wheelers 2 Wheelers-2 stroke 2 Wheelers-4 stroke Auto-total Auto-CNG Auto-petrol Buses-total Buses-CNG Buses-diesel Light commercial vehicle Heavy commercial vehicle Total

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

2000 1150489 0 1150489 2370160 2093018 249560 110488 0 110488 56156 0 56133 116928 515484 3456579

2001 1212281 11700 1197947 2407799 2126256 253523 110488 14000 92705 64407 650 63501 119259 525761 3551690

2002 1429708 15166 1411128 2675396 2362562 281699 56331 35678 11013 47102 6396 38428 56971 251167 3790515

2003 1572748 15505 1553753 2832620 2501403 298254 95119 44164 39022 48813 13809 30108 100092 441258 4183609

2004 1707973 15876 1688523 3013505 2661137 317299 161417 56846 89212 54419 15239 33776 107163 472436 4542646

2005 1822742 16255 1802828 3216012 2839965 338622 171801 56846 99596 55747 15239 35104 114364 504186 4847911

Table 2 Emission factors (Ei,j E.F. (g/km) Cars/taxis Cars/taxis-CNG 2 W-2S 2 W-4S Auto-CNG Auto-gasoline Bus-CNG Bus-diesel LCV HCV

g km

) compiled for the study. CO

e

CO2 223.6 178.8m 42a 38a 60a 60a 412.16m 515.2e 515.2e 515.2e

HC
g

NOxp
g

TSP
g

SO2
g

Pb
c

VOC
c

1.98 0.786g 2.2g 2.2g 3.4g 4.3g 3.6f 4.5g 5.1g 3.6g

0.25 1.55g 2.13g 0.7g 2.07g 2.05g 0.44i 1.21g 0.14g 0.8 g

0.2 0.92g 0.07g 0.3g 0.25g 0.11g 9.6g 12g 1.28g 6.3g

0.03 0.02g 0.05a 0.05a 0.02a 0.08b 0.013j 0.56g 0.2g 0.28g

0.053 0h 0.023c 0.023c 0h 0.029c 0h 2h 0.37c 0.037c

0.003 0h 0.003c,l 0.002c,l 0b 0b 0h 0h 0c 0c

2.5d,e 1.9k 2.13a 0.7a 0.25a 2.05a 1.33 1.6e 1.6e 1.6e

Where, a Iyer (2002). b Bose (1999). c UNEP (1999). d EIA USA (1994). e Gurjar et al. (2004). f Value not available from either of the sources, so value taken as 19.4% less than that for gasoline vehicles as carbon-dioxide emissions are 19.4% less for CNG (Source: De, 2004). g Mashelkar et al. (2002). h TERI (2006). i Value not available from either of the sources, so value taken as 63.2% less than that for gasoline vehicles as carbon-dioxide emissions are 63.2% less for CNG (Source: De, 2004). j Value not available from either of the sources, so value taken as 97.7% less than that for gasoline vehicles as carbon-dioxide emissions are 97.7% less for CNG (Source: De, 2004). k Value not available from either of the sources, so value taken as 24% less than that for gasoline vehicles as carbon-dioxide emissions are 24% less for CNG (Source: De, 2004). l Value not available from either of the sources, so value taken as 16.67% less than that for gasoline vehicles as carbon-dioxide emissions are 16.67% less for CNG (Source: De, 2004). m Value not available from either of the sources, so value taken as 20% less than that for gasoline vehicles as carbon-dioxide emissions are 20% less for CNG (Source: EIA USA, 1994).

(Table 1), emission factors (Ei,j km) (Table 2) and vehicle utilization factor (Table 3; as given in CRRI report by Mashelkar et al. (2002); which is based on actual eld experiments) were used to estimate vehicular emissions.

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173 Table 3 Vehicle utilization factors (km traveled per day per vehicle type). Vehicle type Cars/jeeps 2 Wheelers Auto Buses LCV HCV Source: Mashelkar et al. (2002). km/day 45.1 36.4 66.5 157.2 46.9 45.5

65

3. Results and discussion Emission loads from transport sector are discussed below for the following compounds; CO2, CO, HC, NOx, TSP, SO2, VOC and Pb. Figs. 18 show that emissions of particular compounds are dominated by specic vehicle categories. For example, two-wheelers are major sources for emissions of CO, HC, TSP, VOC and Pb (Figs. 2, 3, 5, 7 and 8), whereas cars and goods vehicles emit most of the CO2 (Fig. 1) and buses plus goods vehicles dominate in emissions of NOx and SO2 (Figs. 4 and 6). CO2 emissions (Fig. 1) increased by about 24% from 13.36 Tg in 2000 to 16.62 Tg in 2005. From 2000 to 2002, emissions decreased by about 11% from 13.36 Tg in 2000 to 11.92 Tg in 2002 followed by rising trend till 2005. Implementations of various emission norms and phasing out of old vehicles during this period might be responsible for decreased emissions during 2001 and 2002. Increasing vehicle population led to rising emissions between 2002 and 2005. Among all vehicle categories in 2002 onwards, cars contributed highest amount of CO2 emission followed by goods vehicle, two-wheelers, buses and autos (Fig. 1). Emissions of CO (Fig. 2) increased by 78% from 197 Gg in 2000 to 350 Gg in 2005. Gradual increasing trend was observed in emissions from 2000 to 2001. Increasing two wheeler population might be responsible for the gradual increment in CO emissions during 2000 to 2005. Fig. 2 clearly indicates the effect of rising population on increasing emissions. Fig. 2 shows that two wheelers (56%) predominated for CO emissions among all vehicle categories followed by cars (20%), goods vehicle (14%), autos (5%) and buses (5%). Hydrocarbons (HC) emission (Fig. 3) increased 1.3 times from 96 Gg in 2000 to 222 Gg in 2005. Highest increment in HC emission was observed from 107 Gg in 2001 to 177 Gg in 2002 (65%). After 2002, HC emissions showed a constant rising trend. Increasing emission trend during study period can be easily correlate with gradually rising vehicle population. Approximate 86% of HC (Fig. 3) was reported only from two wheeler populations (from 2000 to 2005) in Delhi. While contributions from other vehicle categories were very less, e.g., cars (4.3%), goods vehicle (4%), autos (4%) and buses (2%). NOx emission trend, in comparison to other pollutants, showed a different scenario (Fig. 4). About 0.5% of decline was observed in NOx emission (105 Gg in 2000 to 104 Gg in2005). During 2000 2002, steep decreases were observed followed by steadily rising emission trend until 2005. Highest increase (23%) was found during 2002 to 2003; followed by 6% annual average increase till 2005. Goods vehicles were the predominated source of NOx emission (51%) followed by buses (36%), cars (6%), two wheelers (6%), and autos (1%) during the study period (Fig. 4). Total suspended particles (TSP) emissions, as illustrated in Fig. 5, increased from 8 Gg in 2000 to 10 Gg in 2005 (about 31% increment). However highest decrease (9%) was observed from 8.4 Gg in 2001 to 7.69 Gg in 2002 followed by gradual increase till 2005. This is the time when government launched protective measures for control the rising emissions from transport sector in megacity Delhi. Two-wheelers were responsible for high TSP emissions (40%) during 20022005 followed by goods vehicle (29%), buses (19%), cars (10%) and autos (2%). However, it is observed that TSP emissions from goods vehicles were higher before 2002. It shows the efcacy of CNG implementation and clean fuel related initiatives taken after 2001.

66

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

Figure 1. Annual CO2 emissions from transport sector, Delhi (20002005).

Figure 2. Annual CO emission from transport sector, Delhi (20002005).

Figure 3. Annual HC emissions from transport sector, Delhi (20002005).

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

67

Figure 4. Annual NOx emission from transport sector, Delhi (20002005).

Figure 5. Annual TSP emission from transport sector, Delhi (20002005).

Figure 6. Annual SO2 emissions from transport sector, Delhi (20002005).

68

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

Figure 7. Annual VOC emission from transport sector, Delhi (20002005).

Figure 8. Annual Pb emissions from transport sector, Delhi (20002005).

SO2 emissions from transport sector in Delhi demonstrated effective results of policies and protective measures (e.g., ultra low sulphur diesel fuel) introduced by the government. Approximately 9% of the decrease was observed from 12 Gg in 2000 to 11 Gg in 2005 (Fig. 6), with highest decrease (32%) between 2000 and 2002. After 2002, emissions increased with annual average rate of 7.69% per year, but were less in comparison to those of 2000. Bused contributed highest amount (45%) of SO2 followed by goods vehicle (31%), cars (12%), two-wheelers (11%) and auto (1%) (Fig. 6). Declining trends were observed in most of the commercial vehicle categories (e.g., Auto, Bus, Goods Vehicle) in year between 2001 and 2002. Emissions from buses decreased continuously from year 2001 to 2003; with no significant changes later on. Emission trends from cars, two-wheelers and autos followed almost constant trend after 2002. VOC emission, however, observed about 77% increment in Delhi from transport sector with 139 Gg in 2000 to 247 Gg in 2005 (Fig. 7). About 38% increase was observed between 2001 and 2002 followed by (10%) from 2002 to 2003. After 2002, emissions increased at the average rate of 8% annually. Trend showed that implementation of policies and norms were not much effective in controlling the emissions of VOC in comparison to other pollutants. Two-wheelers accounted for the highest amount of VOC emission (56%) followed by cars (31%), good vehicle (8%), autos (2%) and buses (2%) (Fig. 7). No declining trend was found in the emissions from cars during the entire study period. Emission

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

69

trends of autos and goods vehicle were not going in denite way, while there was no signicant change from buses after 2002. Pb emission increased by about 91% from 0.15 Gg in 2000 to 0.29 Gg in 2005. Approximately 4% of decrease in Pb emission was seen in Fig. 8 followed by sharp increment (62%) between 2001 and 2002. After 2002, Pb emission increased at 7% annual rate. This trend showed that norms and policies had no signicant impact for controlling Pb emission in Delhi. Two-wheelers were, however, the predominated source of Pb followed by cars, while other sources were negligible (Fig. 8). As shown in Fig. 9, a comparative analysis of the emissions from 2-stroke and 4-stroke engine technologies clearly points out the need for encouraging 4-stroke two wheeler as the difference in their relative contribution towards all the pollutants is astoundingly great due to the large number of 2S two-wheelers as well as their higher emission factors. This implies that the pollution levels can be brought down to safer levels in spite of the rising two-wheeler population if the 4-stroke technology for the two-wheeler segment is promoted. The 4-stroke technology would be most effective in checking the VOC and HC emissions, which is quite evident from Fig. 9. However, note that 4S two-wheelers may increase relative emission of NOx due to efcient burning of fuel in 4-stroke internal combustion engines. Above discussion implies that implementation of various policy measures in Delhi has given fruitful results for reducing emissions from transportation sector. For example, implementation of Euro norms and introduction of CNG in public transport has consistently reduced emissions of almost all pollutants between 2000 and 2002. Nevertheless, our study suggests that government should introduce more efcient emission norms for two-wheelers and develop some bypass for reducing heavyduty vehicle trafc from Delhi to improve its ambient air quality.

3.1 Comparison with other studies Table 4 presents our estimates along with the emission estimates for Delhi taken from other emission inventories. Almost all of these studies have used similar sources and baseline data for vehicle population estimation with minor differences such as inclusion or exclusion of external vehicles. Thus, it is assumed that sample size of vehicle population in our study is comparable with other studies. Most of these studies have concentrated on a narrow range of pollutants for a specic year. Gurjar et al. (2004) concentrated on a set of pollutants bit different from our study. Compared with other studies our estimates are generally higher; the major part of which could be attributed to the vehicle population taken for estimates. Except the Auto Fuel Policy Report, all the studies have taken projections for the vehicle population of Delhi, which came out to be much lower than the actual gures used from the Economic Survey of Delhi. For example, Das and Parikh (2004) have taken a gure

Figure 9. Relative contribution of 2-stroke and 4-stroke two-wheelers towards total emission.

70

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

Table 4 Emission estimates taken from other studies and present study. Studies Auto fuel policya,b Jalihal and Reddy (2006)a Bose (1999) UNEP (1999) Year 2002 2002 2000 2005 2001 2002 2003 2004 2005 2005 2000 2000 2001 2002 2003 2004 2005 CO2 2775 3449 4665 8555 13363 13524 11919 14445 15621 16621 CO 154 171 194 192 339 341 341 340 341 203 442 197 214 265 300 328 350 HC 67 74 82 81 104 104 104 104 104 76 96 106 176 191 207 221 NOx 40 36 40 50 210 217 217 216 218 39 133 105 114 74 91 98 104 TSP 4.7 5.6 8 12 13 13 13 13 5.4 28 8 8.4 7.7 8.8 9.6 10 SO2 60 3.5 27 12 13 8.9 9.6 10 11 Pb 0.1 0.1 0.2 0.2 0.2 0.2 0.3 0.3 0.3

Das and Parikh (2004) Gurjar et al. (2004) Our estimates

a b

Converted in Gg annum for comparison. Mashelkar et al. (2002).

of 3508366 for 2005, which were even less than 3551690 we took for 2001. Similarly, the values estimated by Bose (1999) turned out to be on much lower side. Furthermore, our gures were higher due to the inclusion of external trafc in carrying out our estimates, which were not accounted by most of the studies including Gurjar et al. (2004). Gurjar et al. (2004) took the same vehicle population as adapted from the Economic Survey of Delhi for our study. However, the results vary in certain pollutants category like CO2 and SO2. Although our choice of emission factors (for 2000) for CO2 and SO2 was similar to those of Gurjar et al. (2004), our estimates turned out to be much higher due to the inclusion of deterioration factors and external trafc. In fact, CO emissions were identical and NOx emissions were still comparable. The other reason was the higher emission factors used in the present study, which were compiled from a variety of sources. The estimates of Jalihal and Reddy (2006) were much lower as they took BSII norms, which are not representative of the actual road conditions of Delhi. Our emission estimates for CO, HC and SO2 were comparable to UNEP (1999), the values from other studies were too low. In fact, NOx emissions of UNEP were higher reecting their choice of higher emission factors for nitrogen oxides. 3.2 Comparison with air quality measurements Fig. 10 indicates the yearly average ambient air quality data, taken from CPCB (2006) for a major road intersection (i.e., ITO intersection) in Delhi for the period 20002005. The initial decrease in SO2 concentrations during 20002002 is also clearly visible in Fig. 10. Nevertheless, SO2 concentration trends show a continuous decrease in later period, which is not consistent with our calculations. Concentration trends of NOx hardly matched with our emission trends because of theinclusion of all nitrogen oxides for our study and not only nitrogen dioxide role of NOx in formation of ozone and conversion of NOx into nitric acid could also have caused some variations from our trends). However, both the studies point out to the check in NOx pollution levels. The rising and falling trends for suspended particulate matter (SPM or TSP) although hardly counterpart and in fact were not comparable since SPM includes particles having a diameter less than 100 micrometer whereas in the present study we have taken into account all range of particles.

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

71

Figure 10. SO2, NO2, SPM ambient air quality data, Delhi (20002005).

4. Limitations and further scope of the study Due to the unavailability of year wise detailed data related to transport, our methodology accompanied number of limitations. Firstly, there is no actual ratio available for 2-stroke and 4-stroke twowheelers and light and heavy commercial vehicles. This data has been assumed on the basis of ratio given in available literature. In addition, the vehicle utilization factors were also assumed to be constant for the ve-year period, which did not take into account the rising travel demand and the effect of starting of the Delhi metro in the year 2002. Besides this, selection of emission factors from a range of different sources is also a cause of uncertainty in our calculations. Some of the values were adapted from EIA USA (1994) in the absence of any other source, which would be an underestimation of the condition actually existing on Delhis roads. The deterioration factors were also not available for all the pollutants and all vehicle and fuel categories and were therefore assumed unity in that case. These limitations set a future scope to improve the present emission estimates as and when more accurate input data (e.g., vehicle population, age wise distribution, and emission factors) are available for megacity Delhi. Moreover, it is also required to validate the emission trends with more ambient air quality data monitored along the roadside locations.

5. Conclusions A six-year (20002005) emission inventory of several air pollutants including criteria pollutants, ozone precursors, and greenhouse gases emitted from transport sector has been prepared for megacity Delhi. It is observed that the emission loads for almost all the pollutants showed increasing trend from year 2000 to 2005. The emission loads for pollutants NOx, TSP and SO2 decreased between 2001 and 2002. This trend is a clear indicator of the impact of steps taken by the government to control the emissions in the form of stricter emission norms, e.g., CNG implementation and phasing out of old age vehicles. However, after 2002 levels for all the pollutants continued to increase slowly which can also be attributed to the spurt in annual growth rate of vehicles registered in Delhi (6% and 10% in the years 2003 and 2004 respectively compared to the previous years 3% and 2.75% for 2001 and 2002). This indicates that increase in the emission loads could have been much greater in the absence of the pollution control measures. The constant annual emission trend of NOx, SO2 and TSP can be related to the emission norms and CNG implementation. Two-wheelers stood as the major contributor for almost all the pollutants (except for NOx, SO2 and TSP). In case of NOx emissions, goods vehicles and buses had a major share, which pointed out the need for bypassing of external trafc passing

72

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

through the city, as a major part of the external trafc is comprised of trucks and buses entering the city none of which run on CNG. Similarly, goods vehicles were found to be the major contributors in TSP emission that was quite clear from the high emission factors for these vehicles. In addition to encouraging 4-stroke two-wheelers, the problem of increasing air pollution emissions from road trafc may likely be checked by a combined effective functioning of the Delhi metro and the public transport, which is being planned and executed by the government.

Acknowledgments This study was supported by the Max Planck Partner Group for Megacities and Global Change, which is established at IIT Roorkee, India, by the Max Planck Society, Munich, and Max Planck Institute for Chemistry, Mainz, Germany. Authors would like to sincerely thank three anonymous reviewers whose valuable comments and suggestions have greatly improved the manuscript.

References
ADB, Asian Development Bank, Strengthening Environmental Management at the State level Cluster, Component EStrengthening Environmental Management at Waste Bengal Pollution Control Board. Final Report, vol. V Air Quality Management, November 2005. Biswas, D., 2006. CNG Conversion Programme for Auto Rickshaws in Delhi. Central Pollution Control Board, Ministry of Environment and Forests, New Delhi, India. URL: http://www.adb.org/Documents/Events/2001/RETA5937/Hanoi/ documents/18_Biswas.pdf (accessed 26.06.06). Bose, R.K., 1999. Towards Better Urban Transport Planning-Problems and Policies, Case of Delhi. Workshop on- Challenges and Opportunities for Concerted Action between Asia and Europe in Urban Transport. Centre for Renewable Energy Resources, the Greek National Centre for Renewable Energy Sources, Athens, Greece (1518 September, 1999). URL: http:// static.teriin.org/urban/trapaper.pdf. CPCB, 1995. Annual Reports 199394. Central Pollution Control Board, Ministry of Environment and Forest, Delhi-32 (http:// www.cpcb.nic.in). CPCB, 2006. Central Pollution Control Board Ambient Air Quality of Delhi. Available at URL: http://www.cpcb.nic.in/bulletin/ bul.htm (accessed 28.06.06). Das, A., Parikh, J., 2004. Transport scenarios in two metropolitan cities in India: Delhi and Mumbai. Energy Conversion and Management 45, 26032625. De, A.K., 2004. The Development of CNG Infrastructure in India with Special Reference to the NCT of Delhi. Indraprastha Gas Limited. Presented on 23 February 2004. URL: http://www.cci.com/media-room/speeches-presentations/2004/feb/feb3sample-a.pdf (accessed 21.06.06). Economic Survey of Delhi, 20032004. Chapter 12, Transport. URL: http://delhiplanning.nic.in/Economic%20Survey/ Ecosur2001-02/Ecosur2001-02.htm (accessed 03.06.06). EIA USA, 1994. Energy Information Administration, USA. Alternatives to Traditional Transportation Fuels 1994, vol. 2, and Greenhouse gas Emissions, Executive Summary. URL: http://www.eia.doe.gov/cneaf/alternate/page/environment/ exec2.html (accessed 28.06.06). Gadhok, T.K., 2000. Risks in Delhi: Environmental concerns, HSMI (HUSDCO), New Delhi. http://www.gisdevelopment.net/ application/natural_hazards/overview/nho0019.htm. Garg, A., Shukla, P.R., Bhattacharya, S., Dadhwal, V.K., 2001. Sub-region (district) and sector level SO2 and NOx emissions for India: assessment of inventories and mitigation exibility. Atmospheric Environment 35, 703713. Gurjar, B.R., van Aardenne, J.A., Lelieveld, J., Mohan, M., 2004. Emission estimates and trends (19902000) for megacity Delhi and implications. Atmospheric Environment 38, 56635681. IPCC, 2006. IPCC Guideline for National Greenhouse Gas Inventories, vol. 2, Energy, Coordinating Lead Authors, Amit Garg and Tinus Pulles, Intergovernmental Panel for Climate Change National Greenhouse Gas Inventories Programme. Iyer, N.V., 2002. Emission Factors for Indian Two and Three Wheelers. Clean Air Initiative: Asia. URL: http:// www.cleanairnet.org/caiasia/1412/channel.html (accessed 09.06.06). Jalihal, S.A., Reddy, T.S., 2006. Assessment of the impact of improvement measures on air quality: case study of Delhi. Journal of Transportation Engineering 132. ASCE. Kandlikar, M., Ramachandran, G., 2000. The causes and consequences of particulate air pollution in urban India a synthesis of the science. Annual Review of Energy and the Environment 25, 629684. Khaiwal, R., Wauters, E., Tyagi, S.K., Mor, S., van Grieken, Rene, 2006. Assessment of air quality after the implementation of compressed natural gas (CNG) as fuel in public transport in Delhi. India Environmental Monitoring and Assessment 115 (1 3), 405417. Mashelkar, R.A. et al, 2002. Urban Road Trafc and Air Pollution in Major Cities, Report of the Expert Committee on Auto Fuel Policy. CRRI, New Delhi. Reddy, M.S., Venkataraman, C., 2002. Inventory of aerosol and sulphur dioxide emissions from India Part II Biomass combustion. Atmospheric Environment 36 (4), 677697. Sharma, C., Dasgupta, A., Mitra, A.P., 2002. Future scenarios of inventories of GHGs and urban pollutants from Delhi and Calcutta, Proceedings of IGES/APN Mega-City Project, 2325 January 2002 (Rihga Royal Hotel Kokura, Kitakyushu, Japan). Singh, S.K., 2005. Review of urban transportation in India. Journal of Public Transportation 8 (1), 7997.

A.S. Nagpure et al. / Urban Climate 4 (2013) 6173

73

TERI, 2001. The Energy and Resources Institute, State of Environment Report for Delhi 2001. Report No. 2000EE65, Supported by the Department of Environment, Government of National Capital Territory, Delhi., TERI, New Delhi. TERI, 2006. The Energy and Resources Institute, A Model to Establish a Policy Framework for Emissions Reduction, and Costeffectiveness of Alternative Measures. http://static.teriin.org/urban/model.htm (accessed 25.06.06). UNEP, 1999. United Nations Environment Programme, Environmental impacts of trade liberalization and policies for sustainable management of natural resources: A case study on Indias automobile sector, New York, United Nations. WHO, 1992. Air quality in mega cities of world. World health organization publications. Xie, J., Shah, J., 2000. Reducing Transport Air Pollution: The Case of Two-Stroke Engine Vehicles in Asian Cities. The World Bank, International Conference on Sustainable Transport and Clean Air, Jakarta, 2931 May.