Impact of Delhi s CNG Program on Air Quality Urvashi Narain Presentation at Transport, Health, Environment, and Equity in Indian Cities Conference at Indian Institute of Technology, New Delhi December 2007
Background In 1985, M.C. Mehta filed a public interest litigation asking the Indian Supreme Court to direct the government to improve Delhi s air quality. At first, the Supreme Court pushed the government to develop comprehensive policies to tackle the problem of air pollution. When the government did little to implement these policies, the Supreme Court passed a series of orders to force implementation. 2
Background (contd.) As a direct result of the Supreme Court orders, Delhi implemented a range of policies to reduce air pollution: Closure of hazardous and noxious (category-h) industries Reduction in the sulfur-content of diesel and petrol Introduction of stricter vehicle emission standards and mandatory catalytic converters Retirement of old (15 20 year old) commercial vehicles Conversion of all commercial passenger vehicles to compressed natural gas Increase in bus fleet from 6,000 to 10,000 and introduction of a mass rapid transit system (METRO) 3
SO2 0 5 10 15 20 25 30 35 40 45 Jan-97 Apr-97 Jul-97 Oct-97 Jan-98 Apr-98 Jul-98 Oct-98 Jan-99 Apr-99 Jul-99 Oct-99 Jan-00 Apr-00 Jul-00 Oct-00 Jan-01 Apr-01 Jul-01 Oct-01 Jan-02 Apr-02 Jul-02 Oct-02 Jan-03 Apr-03 Jul-03 Oct-03 Jan-04 Apr-04 Jul-04 Oct-04 Jan-05 Apr-05 Jul-05 Oct-05 NO2 0 20 40 60 80 100 120 140 Jan-97 Apr-97 Jul-97 Oct-97 Jan-98 Apr-98 Jul-98 Oct-98 Jan-99 Apr-99 Jul-99 Oct-99 Jan-00 Apr-00 Jul-00 Oct-00 Jan-01 Apr-01 Jul-01 Oct-01 Jan-02 Apr-02 Jul-02 Oct-02 Jan-03 Apr-03 Jul-03 Oct-03 Jan-04 Apr-04 Jul-04 Oct-04 Jan-05 Apr-05 Jul-05 Oct-05 CO 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 Jan-97 Apr-97 Jul-97 Oct-97 Jan-98 Apr-98 Jul-98 Oct-98 Jan-99 Apr-99 Jul-99 Oct-99 Jan-00 Apr-00 Jul-00 Oct-00 Jan-01 Apr-01 Jul-01 Oct-01 Jan-02 Apr-02 Jul-02 Oct-02 Jan-03 Apr-03 Jul-03 Oct-03 Jan-04 Apr-04 Jul-04 Oct-04 Jan-05 Apr-05 Jul-05 Oct-05 RSPM 0 100 200 300 400 500 600 Mar-98 Jun-98 Sep-98 Dec-98 Mar-99 Jun-99 Sep-99 Dec-99 Mar-00 Jun-00 Sep-00 Dec-00 Mar-01 Jun-01 Sep-01 Dec-01 Mar-02 Jun-02 Sep-02 Dec-02 Mar-03 Jun-03 Sep-03 Dec-03 Mar-04 Jun-04 Sep-04 Dec-04 Mar-05 Jun-05 Sep-05 Dec-05 ITO Pollution Concentrations
5 SO2 0 5 10 15 20 25 Jan-90 Jul-90 Jan-91 Jul-91 Jan-92 Jul-92 Jan-93 Jul-93 Jan-94 Jul-94 Jan-95 Jul-95 Jan-96 Jul-96 Jan-97 Jul-97 Jan-98 Jul-98 Jan-99 Jul-99 Jan-00 Jul-00 Jan-01 Jul-01 Jan-02 Jul-02 Jan-03 Jul-03 Jan-04 Jul-04 Jan-05 Jul-05 NOx 0 10 20 30 40 50 60 Jan-90 Jul-90 Jan-91 Jul-91 Jan-92 Jul-92 Jan-93 Jul-93 Jan-94 Jul-94 Jan-95 Jul-95 Jan-96 Jul-96 Jan-97 Jul-97 Jan-98 Jul-98 Jan-99 Jul-99 Jan-00 Jul-00 Jan-01 Jul-01 Jan-02 Jul-02 Jan-03 Jul-03 Jan-04 Jul-04 Jan-05 Jul-05 RSPM 0 50 100 150 200 250 300 Jun-00 Aug-00 Oct-00 Dec-00 Feb-01 Apr-01 Jun-01 Aug-01 Oct-01 Dec-01 Feb-02 Apr-02 Jun-02 Aug-02 Oct-02 Dec-02 Feb-03 Apr-03 Jun-03 Aug-03 Oct-03 Dec-03 Feb-04 Apr-04 Jun-04 Aug-04 Oct-04 Dec-04 Feb-05 Apr-05 Jun-05 Aug-05 Oct-05 Dec-05 CPCB Pollution Concentrations
Question What explains the change in CO, NO 2, SO 2 and RSPM levels? In particular, what has been the impact of the conversion to CNG on air quality? 6
Methodology Two methods are typically used to analyze the impact of policies on air quality --- top-down and bottom-up. Top-Down --- Uses actual data on air quality and its determinants and builds a regression model to explain observed changes in air quality. Bottom-Up --- First estimates changes in emission loads caused by a particular policy and then converts these into likely changes in concentrations using relevant air dispersion models. Top-down models provides estimates of the actual impact on air quality rather than predicted impact. This approach, however, requires more data, and can often have low predictive power. 7
Literature Review Chelani and Devotta (2005), Ravindra et al (2006), and Kandlikar (2007) look at trends in air quality before and after the conversion to CNG and attribute any change to the conversion, without controlling for confounding factors. Lack of a trend in RSPM levels is taken to imply that CNG has had no impact on this pollutant. Decline in CO and SO 2, and increase in NO 2, are all attributed to the CNG conversion. Kathuria (2005) controls for a subset of confounding factors and finds no statistical link between a dummy that reflects the date when the CNG order was implemented and RSPM and NO 2 levels over the next week, and a negative link between the dummy and CO levels. These studies suffer from a lack of data on many sources of emissions and are therefore unable to control for confounding factors. Our study fills this gap. 8
Sources of Air Pollution According to the CPCB, motor vehicles are responsible for 72% of Delhi s air pollution (especially NO x and CO), industry for 20%, and the domestic sector for the remaining 8% (White Paper 1997). Balachandran et al (2000) report that particles less than 2.5 µm are associated mainly with vehicular traffic while particles greater than 2.5 µm with soil re-suspension. According to Chowdhury et al (2004) primary source contributors to PM 2.5 mass concentrations in Delhi are diesel exhaust, petrol exhaust, road dust, coal combustion, and biomass burning. 9
List of Variables Measures of vehicular emissions Kilometers driven by vehicle-type and fuel-type Consumption of petrol, diesel and CNG Measures of industrial and power-plant emissions Output by industry-type, including construction and use of diesel generators Consumption of coal, furnace oil, light diesel oil Measures of biomass use including refuge burning Controls for major policy interventions Other variables --- weather patterns 10
Vehicles on the Road in Delhi There are no monthly estimates for the number of vehicles on the road in Delhi Using monthly registration data, we have estimated that there are currently about 2.7 million vehicles in use in Delhi 11
Conversion from Registration to Actual Number Assumed 50% of registered vehicles up to 1990 on road; Accounted for the retirement of old commercial vehicles; Accounted for the conversion of buses, three-wheelers and taxi to CNG; Assumed 6% attrition (vehicles taken off road) on annual basis. 12
Jan-05 Jul-05 25,000 20,000 15,000 10,000 5,000 0 Number of Buses (Private & DTS) in Delhi, 1990-2005 13 Jan-04 Jul-04 Jul-92 Jan-93 Jul-93 Jan-94 Jul-94 Jan-95 Jul-95 Jan-96 Jul-96 Jan-97 Jul-97 Jan-98 Jul-98 Jan-99 Jul-99 Jan-00 Jul-00 Jan-01 Jul-01 Jan-02 Jul-02 Jan-03 Jul-03 Date CNG Petrol Diesel Jan-92 Jul-91 Jan-91 Jan-90 Jul-90
3,000,000 2,500,000 2,000,000 1,500,000 1,000,000 500,000 - Number of Vehicles on the Road in Delhi, 1990-2005 14 Jan-90 Jul-90 Jan-91 Jul-91 Jan-92 Jul-92 Jan-93 Jul-93 Jan-94 Jul-94 Jan-95 Jul-95 Jan-96 Jul-96 Jan-97 Jul-97 Jan-98 Jul-98 Jan-99 Jul-99 Jan-00 Jul-00 Jan-01 Jul-01 Jan-02 Jul-02 Jan-03 Jul-03 Jan-04 Jul-04 Jan-05 Jul-05 Date CNG Petrol Diesel
Not Just Numbers But Kilometers Driven Matter Very little information available on km-driven by different vehicles and how this has changed over time; Have converted numbers into kilometers using available information CRRI report; Estimates shows that even though share of CNG vehicles in total number of vehicles is low, share in terms of vehicle km traveled which is critical for pollution terms is higher; 15
120,000,000 100,000,000 80,000,000 60,000,000 40,000,000 20,000,000 0 Estimated Kilometers Driven in Delhi, 1990-2005 16 Jan-90 Jul-90 Jan-91 Jul-91 Jan-92 Jul-92 Jan-93 Jul-93 Jan-94 Jul-94 Jan-95 Jul-95 Jan-96 Jul-96 Jan-97 Jul-97 Jan-98 Jul-98 Jan-99 Jul-99 Jan-00 Jul-00 Jan-01 Jul-01 Jan-02 Jul-02 Jan-03 Jul-03 Jan-04 Jul-04 Jan-05 Jul-05 Date CNG Petrol Diesel
Other Vehicular-Emission Variables Our estimates of kilometers driven by different types of vehicles by fuel type reflects three policies, namely Retirement of old commercial vehicles Conversion of commercial passenger vehicles to CNG Increase in the number of public buses In addition we have constructed variables to capture the effect of Reductions in the sulfur-content of diesel and petrol Introduction of pre-mix fuel for 2 stroke engines Tightening of emission standards and introduction of catalytic converters Commissioning of the Delhi Metro 17
Other Variables Power-plant emissions are proxied by the total power generated by Delhi s three coal-based power plants. Industrial emissions are proxied by The amount of light diesel oil and furnace oil supplied to Delhi Variable that captures the closure of category-h units Impact of meterological factors is captured by Max and Min Temperature Wind Speed Average Precipitation 18
Our study Uses a statistical model to estimate how much of the variation in air quality indicators is the result of changes in vehicle use, industrial production, and weather patterns; Separate models are estimated for each pollutant RSPM, NO 2, CO, and SO 2 ; Separate models are estimated for ITO and all other stations. 19
RSPM An increase in the proportion of CNG buses and trucks has lead to a decrease in RSPM levels; On the other hand, an increase in the proportion of diesel cars has lead to an increase in RSPM levels. Similarly, RSPM levels have increased with an increase in the consumption of kerosene and light fuel oil. 20
NO 2 Increase in the proportion of diesel cars and 4-stroke 2- wheelers is leading to an increase in NO 2 levels; The model suggests that unlike the conversion of TSRs, the conversion of buses is not leading to an increase in NO 2 levels; Irrespective of fuel used, the increase in the number of km traveled by all fuel types for the major classes of vehicles is leading to increase in NO 2 levels; The power plants in Delhi are also contributing to NO 2 levels. 21
CO The increase in the proportion of CNG buses and trucks is leading to a decline in CO levels; Similarly, a tightening of emission standards for new vehicles is also leading to a decrease in CO; However, irrespective of the fuel-type, an increase in the number of bus-km, car-km, and 2-wheelerkm traveled is leading to an increase in CO; 22
SO 2 Reduction in the sulfur-content of diesel and petrol has lead to a decrease in SO 2 ; Conversion of buses and TSRs to CNG has also lead to a decrease in SO 2 ; Finally, an increase in the proportion of diesel cars has helped to reduce SO 2 levels. 23
Main Results Results suggest that conversion of buses to CNG has helped reduce RSPM, CO, SO 2 and not contributed to the increase in NO 2 levels; Out of the other interventions, the reduction of sulfur in diesel and petrol has also had a significant impact. While more fuel-efficient diesel has helped to reduce SO 2, diesel cars are also increasing RSPM and NO 2. The study also suggests that the gains of these interventions could be negated by the increase in km traveled by all vehicle type. 24
Policy Implications Air pollution policies in different cities must consider the gains that can be made by switching fuels moving from diesel or petrol to gas as this single intervention, if targeted at gross polluters, can have significant impacts. The gains of better quality fuel in the case of diesel are being negated because of the increase in the numbers of diesel operated cars, which in turn is leading to increased NO 2 and RSPM. This suggests the need for stricter emission standards for diesel cars. The model has indicated that the gains of interventions could possibly be lost if the km traveled by vehicles increases. This then suggests the need to devise pollution control strategies, which account for both the pollution load per km and per passenger traveled. 25
Research Implications The data on pollution key pollutants and their sources is sketchy and hard to access. This is particularly important as Delhi is perhaps the most well monitored city. The data gaps need to be filled by regulatory institutions with improved methodologies (e.g. the registration database). The data gaps can also be filled by researchers who can contribute by putting information in the public domain, which will help further research. There is a need for a research-network to expand the work using different methodologies and to different cities so that policy can be better informed. 26
Thank You! We would like to thank Mr. Bhurelal, Ms. Sunita Narain, Dr. Sengupta, Mr. Madan, Dr. Day, Ms. Naini Jaiseelan, Mr. Tamankar, Prof. Mathur, and Ms. Anumita RoyChowdhry for their help in acquiring the data for this study and for their comments on earlier drafts. We are also grateful to the United States Department of Energy s Clean Citites International Program for their financial support.