Drive Clean Program Emissions Benefit Analysis and Reporting Light Duty Vehicles and Non-Diesel Heavy Duty Vehicles 1999 to 2003

Transcription

1 Drive Clean Program Emissions Benefit Analysis and Reporting Light Duty Vehicles and Non-Diesel Heavy Duty Vehicles 1999 to 2003 February 2005 Cette publication technique n'est disponible qu'en anglais Copyright: Queen's Printer for Ontario, 2005 This publication may be reproduced for non-commercial purposes with appropriate attribution. PIBS 5015e

2 Table of Contents 1.0 Summary Introduction Background Analysis and Reporting Results Discussion of Possible Biases Conclusions...44 Key Data Tables Tables 1A to 1F: Rates of test, failure, retest, and waiver by vehicle type and model year for individual vehicles for 2002 and 2003 Tables 2A to 2F: Modal failure rates for 2002 and 2003 Tables 3A to 3F: Calculation of emissions reductions by vehicle type and by age group for 2002 and 2003 Tables 4A to 4C: Overall emissions reductions for1999 to 2003 Table 5: Summary of Inspection Statistics for 2002 and 2003 Tables 6A to 6D: Annual Emissions Reductions for 1999 to Data Tables: (PDF Format 1375 KB) APPENDICES APPENDIX A... Baseline Emissions Inventory Estimates (MOBILE Models) APPENDIX B... Numbers of In-Use Light Duty Vehicles, 1999 to 2003 APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX APPENDIX C... Annual Kilometres Traveled D...Mass Emissions Factors E...Estimating Actual Emissions Inventories F...Comparing Baseline Emissions with Actual Emissions G... Detailed Example of Emissions Reduction Calculations H... Emissions and Repair Reductions by Vehicle Type and Age I...Effects of Adding the Idle Test J... Heavy duty Non-Diesel Vehicles APPENDIX K...Summary Statistics for 2002 and 2003 Appendices: (PDF Format 88 KB) Note: Links may not work if you have an older version of Adobe Acrobat. You can also paste file location into your browser, or you can update Adobe Acrobat at:

3 1.0 SUMMARY Drive Clean analysis shows major reductions in harmful vehicle emissions Vehicles are the largest domestic source of smog-causing pollutants in Ontario. Smog-causing pollutants include nitrogen oxides (NOx) and volatile organic compounds such as hydrocarbons (HC). Drive Clean reduces emissions of smog-causing pollutants as well as other pollutants by requiring designated vehicles to have their emissions tested before their licence plates can be renewed or ownership is transferred. The test identifies vehicles with emissions problems and Drive Clean requires them to be repaired. A new report describes results from an independent analysis by Stewart Brown Associates of emissions data collected from 1999 through 2003 from emissions tests of light-duty vehicles (LDVs) and non-diesel heavy-duty vehicles for the Drive Clean program. An improved computer model (MOBILE 6.2C) was used to analyze program results. During this period Drive Clean has reduced smog-causing emissions by more than 81,200 tonnes. Overall, smog-causing emissions from LDVs are estimated to have been reduced by more than 140,000 tonnes from 1999 through 2003 due to replacement of older cars, new emissions technology and cleaner fuel as well as repairs to vehicles that fail a Drive Clean test. This means that Drive Clean is responsible for nearly 60 per cent of the overall reduction figure. In 2003 alone, smog-causing emissions were reduced by the equivalent of removing more than 600,000 typical LDVs from Ontario s roads. In addition, from 1999 to 2003 emissions of carbon monoxide (CO), a poisonous gas, were reduced by over 690,000 tonnes due to Drive Clean Emissions of the climate-change gas carbon dioxide (CO2) were reduced by more than 100,000 tonnes, as a result of improved vehicle fuel efficiency of LDVs due to emissions system repairs required by Drive Clean. 3

4 During the same five year period, smog-causing emissions from less than 14,000 NDHDVs were reduced by more than 160 tonnes and CO by more than 760 tonnes. 4

5 2.0 INTRODUCTION Ontario s Drive Clean is a mandatory inspection and maintenance program for both light and heavy duty vehicles that are required to meet emissions standards as a condition of vehicle registration or transfer of ownership (resale). The following report describes results from an independent analysis by Stewart Brown Associates (SBA) of emissions data from dynamometer (treadmill) testing and curb-idle testing of non-diesel light duty vehicles (referred to in this report as LDVs), collected in 2002 and The report also updates results from a previous analysis for the years 1999 to An additional section is provided to describe results from twospeed idle testing of non-diesel heavy duty vehicles (NDHDVs). Results include estimates of reductions of smog-causing emissions of nitrogen oxides (NOx), and hydrocarbons (HC), and emissions of the poisonous gas carbon monoxide (CO) from LDVs in: Phase 1 (Greater Toronto Area and Hamilton, where mandatory emissions testing for Drive Clean began in April 1999); Phase 2 (urban and commuter areas from Windsor to Peterborough, where mandatory emissions testing began January 1, 2001); Phase 3 (the remainder of the southern Ontario smog zone from Windsor to Ottawa, where mandatory vehicle emissions testing began on July 1, 2002). Results also include estimates of reductions of HC and CO from NDHDVs in the same areas as the LDV program. In the previous report covering the period , reductions of smogcausing emissions were reported as percentages and tonnages achieved during that period in Phases 1 and 2. Tonnage reductions were estimated by subtracting vehicle emissions in tonnes, calculated separately for Phases 1 and 2 for the year 2001, from the estimated vehicle emissions during the start-up year for Drive Clean in each Phase. These start-up baseline emissions were the emissions that would have been expected without Drive Clean. They were estimated using the MOBILE 5C vehicle emissions computer model. The

6 2001 report described in detail how the vehicle emissions in 2001 were calculated. This report uses the current MOBILE 6.2C vehicle emissions model to re-estimate startup baseline emissions as well as ongoing baseline emissions. These baseline emissions include the effect of factors such as vehicle retirement, improvements in vehicle technologies and cleaner fuels that would have been expected if the Drive Clean program had not been implemented. In this report, the emissions reduction estimates for the period were also recalculated using the MOBILE 6.2C model. Migration to the newer vehicle emissions computer model is part of an ongoing process of continuous improvement in the estimation of emissions and emissions reductions. The current model provides a more accurate estimate of vehicle emissions, accounting for emissions reductions from vehicle retirement, improved emissions control technologies and cleaner fuels. The MOBILE 6.2C computer model is a Canadian version developed by Environment Canada with the co-operation of the Ontario Ministry of the Environment of the emissions reduction computer model currently accepted as the standard throughout North America. It was developed by the U.S. Environmental Protection Agency, as an improvement on the MOBILE 5 model. Ontario had relied on a Canadian version of that model, MOBILE 5C for the report. Reductions in vehicle emissions result from repairing vehicles that fail the Drive Clean test in combination with vehicle retirement, improved emissions control technologies on new vehicles and improved fuels. Emissions reduction estimates in this report focus on the benefits of emissions-related repairs, separate from emissions reductions resulting from the other factors. Repair benefit results include the performance of repairs in the year of the test as well as the carry-over benefit from vehicles repaired in previous years that continue to have reduced emissions. The report describes in detail how the total emissions benefits from repairs were estimated, and outlines the assumptions related to each of the estimates. This report also estimates emissions reductions of carbon dioxide (CO 2 ), a climate change gas. The reductions are attributed to improvements in fuel economy for 6

7 the vehicles that failed to meet emissions standards and were assumed to have been repaired or retired. Estimates of CO 2 in this report were made using the same approach as for the report. 7

8 3.0 BACKGROUND Smog is a serious health issue in southern Ontario. Scientists have found that there is no safe level for exposure to smog. Smog-related air pollution is linked to such health effects as premature death, respiratory and heart problems and bronchitis. The Ontario Medical Association (OMA) concluded that 1,900 premature deaths, 9,800 hospital admissions and 13,000 emergency room visits could be attributed to smog-causing pollutants (ozone and particulate matter) in The OMA further estimates that this costs Ontario health care system and economy more than $1 billion annually in hospital admissions, emergency room visits and absenteeism. The City of Toronto Public Health estimates that air pollution in Toronto contributes to about 1,700 premature (or early) deaths and 6,000 admissions to hospitals each year. Emissions from on-road vehicles contribute to both local and regional environmental problems: primarily smog and climate change. Vehicles are the largest single source of smog-causing pollutants in Ontario. Smog-causing emissions from vehicles include the gases NOx and HC. HC are mostly unburned gasoline and belong to a group of pollutants called volatile organic compounds (VOCs). NOx plus VOCs combine in the presence of sunlight to form ground-level ozone, a major component of smog. Vehicle emissions also contain the poisonous gas CO and air toxics such as benzene, the climate change gas carbon dioxide (CO 2 ), and microscopic particulates. Drive Clean reduces smog-causing emissions as well as other pollutants by requiring designated vehicles to have their emissions tested before their licence plates can be renewed or before transfer of ownership. The test identifies vehicles with emissions problems and requires them to be repaired. Ontario's mandatory vehicle inspection and maintenance program, Drive Clean, is an effective approach to reducing vehicular emissions. The Ministry of the Environment is committed to ensuring the Drive Clean program continues to make a positive impact on the environment and on the health of Ontarians. The primary target of the Drive Clean program is to reduce the emissions of smog-causing pollutants from vehicles in the southern Ontario smog zone where the majority of Ontario residents live and drive. As air pollution travels long distances, cutting emissions at the source benefits a large part of Ontario, as well as neighboring jurisdictions. 8

9 This report presents a summary of the analysis of light duty vehicle (LDV) and nondiesel heavy duty (NDHDV) data collected from the emissions test procedure of Ontario's Drive Clean program, during the period of January 2, 1999 to December 31, The analysis was used to: Quantify reductions of LDV emissions of nitrogen oxides (NOx), hydrocarbons (HC), and carbon monoxide (CO) from the start of Drive Clean in each Phase as percentages, and as tonnages; Quantify reductions of LDV emissions of NOx, HC and CO annually in each Phase that would have resulted without Drive Clean, and did result with Drive Clean, as percentages and as tonnages; Quantify reductions of HC and CO emissions from NDHDVs; Estimate the effect of fuel savings on reducing emissions of the climate change gas, carbon dioxide (CO 2 ). Emissions Standards Federal regulations set emissions performance standards that all new model vehicles must meet in grams of emissions per kilometre. These emissions standards are set uniformly throughout North America (note that in the United States, emissions standards are expressed as grams per mile), except California where standards are more stringent. New vehicle emissions standards have been progressively tightened to reflect the improvements in emissions control systems introduced over the last 20 years. Manufacturers demonstrate compliance by submitting samples of their new vehicles, representing every new make and model, to the comprehensive Federal Test Procedure (FTP). The FTP is a lengthy series of emissions tests performed on new vehicles, under a variety of controlled ambient conditions and driving cycles which include acceleration, deceleration and idle. Ontario sets standards limiting the permissible release of emissions from on-road vehicles. Emissions from individual vehicles increase over time with engine wear, improper or irregular maintenance, and tampering with, or failure of, the emissions control devices. Some deterioration happens even with well-maintained vehicles. Ontario's vehicle emissions standards, originally based on the FTP standards, have been converted to concentrations (parts per million or percentage), and adjusted to reflect the relative capabilities of both older and newer emissions control systems and the effects of normal engine wear. 9

10 Older vehicles, even when maintained in excellent running condition, cannot match the pollution control efficiencies of today's models, but they will be able meet the Drive Clean emissions standards as long as their engines and pollution control systems are operating properly. Conversely, when emissions control systems fail, even today s model vehicles may become gross polluters. Mandatory Emissions Testing Requirements Mandatory emissions testing requirements are found in Regulation 628 under the Highway Traffic Act. The regulation states that LDVs with a registered gross weight of 4,500 kilograms or less (mainly passenger cars, sports utility vehicles, vans and light trucks) and non-diesel HDVs with a registered gross weight greater than 4,500 kilograms must obtain an emissions test report for: - registration renewal, starting for LDVs three calendar years after the model year and every second year up to and including 19 calendar years after the model year. A Pass or Conditional Pass (using repair cost limit) report may be used to satisfy this requirement. - registration renewal, starting for NDHDVs three calendar years after the model year and every year with no age limit. A Pass report is required to satisfy this requirement; - resale: new owner requires a Pass report to purchase a licence plate sticker for the vehicle starting for one calendar year after the model year (both LDVs and non-diesel HDVs), and up to and including 19 calendar years after the model year (LDVs only; no age limit for NDHDVs). A Pass report is required to satisfy this requirement. In the biennial LDV test cycle, odd model-year vehicles are tested in an even calendar year; even model-year vehicles are tested in an odd calendar year. A conditional pass is issued when a vehicle fails a retest and qualifies for the repair cost limit. Thus, in 2002, emissions tests were mandatory for registration renewal of odd model year vehicles from 1983 to 1999, and in 2003, emissions tests were mandatory for registration renewal of even model year vehicles from 1984 to Note that the emissions Pass and Conditional Pass reports are valid for one year so that the test may be done at any time within the12 months prior to the registration expiry 10

11 date, or the application date for late renewals or ownership transfer. This means that some vehicles will be tested off cycle (e.g., many even model year vehicles due for registration renewal early in 2003 were tested in late 2002). The combined effect of oncycle and off-cycle renewals and tests for ownership transfers means that more than half of the LDVs requiring tests on the biennial cycle were actually tested in any 12- month period. Emissions Testing Drive Clean emissions testing is designed to identify those vehicles that no longer operate in compliance with acceptable emissions standards and to ensure that emissions reductions were actually achieved when these vehicles were repaired. The Drive Clean emissions test procedures are set out in Ontario Regulation 361/98, as amended, under the Environmental Protection Act. Emissions standards, emissions test methods and additional technical information are described in greater detail in the Ontario Ministry of Environment s Drive Clean Guide and the manual Standard Operating Procedures for Ontario's Drive Clean Facilities as Applied to Light Duty Vehicles and Non-Diesel Heavy Duty Vehicles. During the period of this report (1999 to 2003), several program changes were made that would improve the program effectiveness: - on July 1, 2001 the repair cost limit of $200 in place since the start of the program in Phase 1 increased to $450; - on July 1, 2002 the program was expanded to Phase 3 with a repair cost limit of $200 for the next two years, and the curb-idle test was added to the existing dynamometer test; - on January 1, 2003, the dynamometer emissions standards were tightened from initial values by 11.5 per cent and the repair cost limit of $200 in place since the start of the program in Phase 2 (January 1, 2001) increased to $450. Test Process: LDV The LDV test process (for non-diesels) consists of the following basic steps (note program changes and dates above): - a pretest check to ensure the suitability and safety of testing the vehicle on a dynamometer (non-compliant vehicles were rejected from testing at this stage and not counted as a test); verification that a catalytic converter (if one was originally installed on that vehicle) and gas cap are in place, and an advisory 11

12 check of the on-board diagnostic light for 1998 and newer vehicles (vehicle owners were advised if the light was on but their vehicles were not rejected from testing); and - a sophisticated, tamper-proof emissions test done on a chassis dynamometer (vehicle treadmill) that simulates the load applied to a vehicle as it climbs a grade of about three per cent, at a steady 40 kilometres per hour, to determine emissions levels of hydrocarbons (HC), carbon monoxide (CO), and nitrogen oxides (NOx) in the exhaust (tailpipe) stream; and - immediately following the dynamometer test, a test for emissions at curb idle (foot off the gas pedal) for HC and CO. For LDVs to pass, the concentration of all three of the polluting gases emitted from the vehicle tailpipe must be at or below the provincial emissions standards (also called cutpoints) during the dynamometer and curb idle tests. Vehicles that do not meet the emissions levels established for a particular model, make and year require repairs and retests until they receive a Pass or Conditional Pass. Test Process: All NDHDVs and Some LDVs All NDHDVs and a small percentage of LDVs (e.g., vehicles equipped with full-time, allwheel drive, certain vehicles with traction control or anti-locking braking systems, or vehicles with lowered suspension or body kits that might be damaged during the test) cannot be tested on a dynamometer. Tailpipe emissions of HC and CO from these vehicles are measured at two operating conditions: a steady 2,500 RPM and at curb idle. NOx is not measured as it results from the heat generated by a motor working to propel a vehicle, not when the vehicle is idling. While the dynamometer test better reflects on-road vehicle emissions performance, emission-related repairs for HC and CO may also result in reductions of emissions of all three gases, HC, CO and NOx. Vehicle Numbers and Retests Vehicles may be tested more than once in any given period for several reasons: - a vehicle failing to meet emissions standards requires retesting (may be retested as soon as within the same day, and a vehicle may have several retests); - a vehicle passing a test for sticker renewal would require another test for ownership transfer any time after the 12 month validity of the pass report, and vice versa; 12

13 - a vehicle obtaining a conditional pass for sticker renewal would require another test for ownership transfer, even the same day, as a full pass is required; - a vehicle coming up for licence plate sticker renewal in subsequent two-year cycles requires retesting which means that eventually almost every vehicle will have been tested more than once. - quality control and quality assurance checks (audits), and other miscellaneous tests. Vehicles were only counted once at the initial test within the given period, no matter how many other tests were done on the same vehicles. However, retests done at another facility may be entered as initial tests, if the owner does not want to indicate that the vehicle failed at another facility. Total number of tests includes all tests done to all vehicles within the given period. Repair Cost Limit (RCL) Conditional Pass Vehicle tests and repairs are conducted by Drive Clean certified emissions inspectors and repair technicians at Drive Clean Facilities that are accredited by the Ministry of the Environment. Vehicle owners who wish to qualify for the RCL must have all diagnostic and repair work carried out under the supervision of a Drive Clean certified repair technician. Vehicle owners qualifying for the RCL by having their vehicles repaired at a Drive Clean facility and failing the retest, receive a Conditional Pass. The Conditional Pass can be used to renew the vehicle registrations, but cannot be used to transfer ownership (resale). The $200 RCL was in place for the first two years in each Phase to provide vehicle owners with a grace period to make necessary emissions control system repairs. After the first two years in each program area, the RCL was increased to $450 as most vehicles can be fully repaired within this amount. The RCL feature, similar to many vehicle emissions inspection and maintenance programs in North America, was designed to lessen the financial impact of repairs on low-income vehicle owners. The RCL was increased from $200 to $450 in Phase 1 on April 1, 2002, in Phase 2 on January 1, 2003 and, in Phase 3, on July 1,

14 Test Data The emissions and vehicle data collected at Drive Clean facilities are automatically uploaded to a central computer database, known as the Drive Clean Vehicle Emissions Transaction System (VETS). The VETS information is analyzed internally for purposes of quality assurance and quality control, to detect any evidence of fraud and to assess the effectiveness of the program. Test results are accessed by the Ontario Ministry of Transportation through the Vehicle Registration System and used at vehicle licensing offices to process vehicle registrations. 14

15 4.0 ANALYSIS AND REPORTING The services of Stewart Brown Associates (SBA) were retained by the Ontario Ministry of the Environment s Drive Clean Office for evaluation of the emissions and vehicle data collected from January 1, 2002 through to December 31, SBA used vehicle emissions factors and a calculation procedure that were fully compatible with those used in the previous report, but modified to accommodate the addition of the curb-idle part of the emissions test that was introduced as an enhancement to the dynamometer testing procedure in July This report differs from the previous report with the inclusion of results from the analysis of NDHDV data, and the use of the MOBILE 6.2C vehicle emissions computer model to replace the MOBILE 5C model. With the use of the MOBILE 6.2C model it was necessary to re-assess the previous results for 1999 to 2001, for the sake of continuity with results to the end of To estimate total emissions reductions during the first five years of Drive Clean s LDV program, the consultant used two steps. 1. MOBILE 6.2C Model: Estimating the Baseline LDV Emissions Inventory The first step involved the MOBILE 6.2C vehicle emissions computer model to determine LDV emissions for each program Phase in the start-up year and in subsequent ( ongoing ) years as if Drive Clean had not been implemented ( baseline emissions ). The MOBILE 6.2C model is the Canadian version of the MOBILE 6.2 model developed by the United States Environmental Protection Agency (US EPA). This is one of a series of models developed to evaluate initiatives by U.S. states to respond to excess levels of ground level ozone in the ambient air. Under the US Federal-Aid Highway Program, to be eligible for federal funding for transportation projects, states must demonstrate that the proposed projects meet applicable federal requirements, including the Clean Air Act. The Clean Air Act is a legislative attempt to address air quality issues in local areas by designating areas as attainment (conformity) or non-attainment (non-conformity). When a zone is designated as non-attainment for any criteria pollutant, a state implementation 15

16 plan (SIP) must be developed to regulate emissions or air pollutants so that the nonattainment zones can achieve the National Ambient Air Quality Standards. A SIP may include a vehicle Inspection and Maintenance (I/M) program to achieve attainment. All U.S. jurisdictions with vehicle I/M programs submit emissions data and program design information to the US EPA as inputs to the MOBILE vehicle emissions computer model to obtain credits for their State Implementation Plans. Reporting varies state to state but are generally not geared to public performance measures (i.e. they focus on SIP reporting to the US federal government). US research indicated that the previous model, MOBILE 5, did not accurately reflect the improved emissions control technologies and stringent emissions standards of newer vehicles starting in the late 1990s, and therefore the emissions from these newer vehicles were overestimated by the model. At the same time, the MOBILE 5 model underestimated the emissions from vehicles with older emissions control technologies. The current MOBILE 6.2 model accounts for these shortcomings, and allows a greater number of vehicle classes and factors to be entered. The final result is a more accurate estimate of the vehicle emissions inventory in each program Phase for each year. The computerized MOBILE 6.2C model was used to approximate the total emissions from vehicles in the calendar year at the start of each Phase. The model was also used to approximate the emissions for each subsequent calendar year that would have been expected if Drive Clean was not operating. The model predicts ongoing reductions with the retirement of older vehicles, introduction of vehicles with cleaner emissions control technologies, and improvements in fuels. At the same time the model accounts for the increasing emissions expected with fleet growth, increasing kilometers traveled, and changes in fleet characteristics such as a trend to large trucks and sport utility vehicles in the LDV fleet. The results from running the MOBILE model form the baseline inventory, to which the actual inventory with Drive Clean (i.e., the best estimate of emissions from all vehicles each year, with all factors considered), can be compared. 2. Estimating Drive Clean LDV Emissions Reductions The second step involved Drive Clean test data to determine the change in 16

17 emissions from all vehicles once some of the vehicles had been repaired to meet Drive Clean requirements. In each calendar year initial emissions were estimated for all vehicles in each Phase area using the actual initial tailpipe emissions results for each of the gases from testing individual vehicles on the dynamometer (followed by the curb idle test starting in 2002). Results from individual vehicles, expressed as concentrations were grouped and converted to mass emissions factors (grams/kilometre). These results were added to estimates of emissions from vehicles not included in the Drive Clean program testing (i.e., those that are exempt from testing because they were too new or too old; some of these older vehicles may have been retired and would contribute to a reduction in the total emissions from one year to the next). The emissions from the newest vehicles were estimated based on their new vehicle emissions standards which is a reasonable approach because few vehicles would be expected to deteriorate while in the warranty period, which would usually cover at least the first three years. The emissions from vehicles over 20 years old were estimated from testing in the BC AirCare program which does not have an age exemption. The emissions performance of off-year exempt vehicles was interpolated from the performance of on-year vehicles that were inspected. LDV emissions reductions resulting from repairs were estimated using the actual final tailpipe emissions results for each of the gases, plus estimated emissions from pass and exempt vehicles which would have identical initial and final results. Therefore emissions reductions from repairs resulted only from the vehicles which failed the test, and following repairs, were issued a pass or conditional pass. A particular LDV is unlikely to pass or conditional pass more than once in a calendar year, but may have several fails. In calculating the emissions reductions, the first result in each calendar year was used to estimate initial emissions and the last result in each calendar year for each vehicle in the fleet was used to estimate final emissions. In cases where the initial tests and final tests were in different calendar years, the initial test results applied to that year and the final results for these vehicles were estimated for that same year, assuming that these vehicles eventually obtained passes or conditional passes in the same proportion as initial and final tests 17

18 completed in the same year. The actual final results for these vehicles were not included in the subsequent year in the analysis to prevent double counting. For each model year and vehicle type, calculations included vehicles that: - passed; - failed in each of the possible failure modes (i.e., which of the three emissions standards they were unable to meet); - passed on retest; - were granted a conditional pass; or - had not returned for a retest within the calendar year. It is impossible to know for sure how many vehicles which failed a test and had not yet returned to be retested in a particular calendar year were immediately retired from use, so it has been assumed that all of these vehicles would eventually achieve either a pass or a conditional pass. This assumption underestimates emissions reductions in the year because failing vehicles which are then retired actually achieve zero emissions (i.e., the maximum reduction possible). The assessment of the next years initial emissions will automatically exclude those vehicles that were actually retired, because only vehicles actually licenced for use require emissions tests. Initial annual emissions were calculated by multiplying the number of vehicles in the fleet (whether passing, failing, or exempt from emissions testing) by the initial condition mass emissions factor (grams per kilometer, converted from parts per million or per cent) and by the annual average distance driven for the model year. Failing is mode-specific ; some vehicles will have a failure mode with only one gas failing and others will have more than one gas that is failing this is the failure mode. Final annual emissions were calculated similarly to initial annual emissions, but the calculation was more complex as the repaired vehicles term below comprised the sum of: vehicles that did not change, those that achieved a "good" repair (pass) and those that achieved a "bad" repair (conditional pass). 18

19 Using hydrocarbons (HC) as an example, this is how the aggregate calculation was done for each parameter: where: Initial HC = Σ g/km x km/year + Σ g/km x km/year + Σ g/km x km/year passing vehicles failing vehicles exempt vehicles and: Final HC = Σ g/km x km/year + Σ g/km x km/year + Σ g/km x km/year passing vehicles repaired vehicles exempt vehicles Σ = the sum of all vehicles in the group g/km = mass emissions factor, grams per kilometre km/year = annual distance driven, kilometres per year All the initial and final results were converted from the mass emissions factor units to the inventory tonnes units. Both initial and final results include passing and exempt vehicles. The amount of inventory reduction achieved in a year by new repairs is the difference between Initial and Final values for that year. Thus, for example: Reduction on HC from New Repairs = Initial HC - Final HC The initial values represent what the inventory would be in the year if there were no more repairs (i.e, if the program requirements were suspended at the beginning of the year). The effects of repairs done in previous years, however, and the better maintenance of vehicles would continue to provide inventory reduction benefits for a number of years afterwards. This residual effect would decrease with time. The final values represent the estimated actual inventory that was achieved by the end of each year. Comparing the final values from one year to the next shows how the estimated actual inventory is decreasing. These can be compared to the initial values 19

20 from the first year of program implementation in each phase area (i.e., emissions inventory baseline without Drive Clean), to give the total reductions since implementation. The final values are also compared to the ongoing baseline inventory reductions that were modeled using the MOBILE 6.2C model to estimate annual emissions reductions. 20

21 5.0 RESULTS Pass/Failure Rate The LDV program design predicted that most vehicles would pass the initial Drive Clean test. A portion of the failed vehicles would pass after repairs, and another portion would fail again and be eligible for conditional passes, or waivers, because repair costs (serviced or recommended) at accredited facilities exceeded the repair cost limit (RCL). A further portion would be retired or transferred outside of the program area. Failure Rates = (# vehicles failed initial insp) / (# vehicles with initial insp) phase 1 phase 2 phase % % % 14.2% % 12.6% 12.9% % 11.6% 12.4% Over time, the Drive Clean program would be expected to lead to a drop in the failure rate with: - motorists better maintaining their vehicles, - motorists conducting repairs prior to the initial Drive Clean test, - improved repair technician skills, preventing, or detecting and correcting problems with vehicle emissions control systems. This downward trend in the failure rate depends on: - continued improvement of vehicle maintenance habits by vehicle owners; - continued improvement of skills by repair technicians (effective and durable repairs); - retirement of older vehicles, replaced by vehicles with improved emissions control technologies. Continuation of the downward trend in the failure rate would also depend on the influence of the Drive Clean program since: 21

22 - vehicle owners habits and repair technician skills may be maintained only so long as testing is mandatory; - vehicle retirement might be postponed in the absence of the testing requirement. Emissions Reductions All of the following percentages and figures were derived from Tables 4A, 4B and 4C, as well as Tables 6A, 6B, 6C, 6D. Phase 1: Mandatory Testing Started April 1, 1999 LDV Drive Clean emissions testing began on January 2, 1999, in the Phase 1 program area (the Greater Toronto Area and the former Hamilton-Wentworth Region, now the City of Hamilton). Note that mandatory emissions testing, as a condition for renewing LDV registrations or for ownership transfer, started in the Phase 1 program area on April 1, In 2002, 1,331,346 individual vehicles in the Phase 1 area received 1,543,622 tests. In 2003, 1,336,727 vehicles received 1,544,445 tests for a total of more than 7.2 million tests performed since April Note that 143,884 vehicles failed their initial inspection in 2003, resulting in a failure rate of 10.8%. Many of the vehicles inspected in 2003 would have been tested on three separate occasions for registration renewal: in 1999, 2001, and In Diagram 1 below, the upper dotted line indicates the level of vehicle emissions that would have been expected without Drive Clean in the year 1999 when Drive Clean started in Phase 1. This line is called the start-up baseline. The middle line indicates the MOBILE 6.2C estimate of ongoing baseline emissions without Drive Clean, showing reductions due to the other factors such as vehicle retirement, improved emissions control technologies on new vehicles and improved fuels. The bottom line indicates incremental emissions reductions with Drive Clean, (line indicates best estimate of actual emissions) due to repairs of failing vehicles as well as all of the other factors. The distance between the middle and bottom lines indicates the annual emissions reductions which resulted directly from Drive Clean repairs. While the starting year and scale would differ, the same approach is used for the other two Phases. 22

23 Diagram 1. By the end of 2003, relative to the 1999 baseline emissions at the start of testing, the combined effects of repairing vehicles that failed the Drive Clean test and the other factors described above had reduced: smog-causing emissions (HC and NOx) by an estimated cumulative total of 35.5% or 33,071 tonnes; emissions of the poisonous gas carbon monoxide (CO) by an estimated cumulative total of 27.9% or 227,126 tonnes. These estimates are indicated by the vertical arrow on the far right in Diagram 1 above, and represent the total cumulative emissions reductions relative to baseline emissions in the start-up year Without the Drive Clean Program by the end of 2003, relative to 1999 baseline emissions at the start of testing: HC and NOx would have been reduced by an estimated 20.9 % or 19,477 tonnes; CO would have been reduced by an estimated 11.9 % or 97,227 tonnes. These estimates are indicated by the vertical arrow, second from the far right in Diagram 1 above, and represent the emissions reductions that would have been expected by 2003 as a result of factors such as vehicle retirement and improved emissions control technologies and fuels, according to the Mobile 6.2C model, if Drive Clean had not been implemented. 23

24 Therefore, for the year 2003, Drive Clean repairs alone reduced: HC and NOx by 14.6% or 13,594 tonnes; CO by 16.0% or 129,899 tonnes. These values were calculated as the difference between the middle (Mobile 6.2C) and bottom lines for the year 2003, and the percentages were calculated as the proportion of the cumulative total. In other words, this calculation provides the annual emissions reductions in tonnes due to Drive Clean in the year These annual emissions reductions are illustrated in Diagram 2 below. The upper line indicates the MOBILE 6.2C estimate of ongoing baseline emissions without Drive Clean, showing reductions due to the other factors such as vehicle retirement, improved emissions control technologies on new vehicles and improved fuels. The bottom line indicates enhanced emissions reductions with Drive Clean, due to repairs of failing vehicles as well as the other factors. The distance between the middle and bottom lines indicates the emissions reductions which resulted directly from Drive Clean repairs on an annual basis. While the starting year and scale would differ, the same approach is used for the other two Phases. Diagram 2. Using the ongoing baseline from the MOBILE 6.2C model (upper line in Diagram 2, above) as the reference each year gives a snapshot of that year s emissions reduction benefits. These snapshot estimates of annual emissions reductions due to Drive Clean 24

25 for 2003 and previous years are indicated by the series of vertical arrows between the two lines in Diagram 2 (value for 1999 is not shown to simplify the illustration). These annual emissions reductions may be added together for every year of the Drive Clean program to give the overall total of all of the annual emissions reductions due to Drive Clean for the whole period from 1999 to Similarly, annual values for emissions reductions without Drive Clean can be calculated as the difference between the upper (business as usual) line, and the middle (MOBILE 6.2C line) shown in Diagram 1, for each year. Annual Emissions Reductions In 2002 the Drive Clean program reduced the 2002 baseline inventory of: - HC by 29.6% or 11,884 tonnes; - CO by 20.9% or 162,648 tonnes; - NOx by 12.5% or 5,216 tonnes; and - HC+NOx by 20.9% or 17,100 tonnes. In 2003 the Drive Clean program reduced the 2003 baseline inventory of: - HC by 30.0% or 10,648 tonnes; - CO by 18.1% or 129,899 tonnes; - NOx by 7.7% or 2,946 tonnes; and - HC+NOx by 18.5% or 13,594 tonnes. These annual emissions reductions may also be added together for every year of the Drive Clean program to give the overall total of all of the annual emissions reductions that would have been expected without Drive Clean. Then overall tonnage total of all of the annual emissions reductions without Drive Clean can be compared to the overall total of all of the annual emissions reductions due to Drive Clean to give the relative benefit of Drive Clean for the whole period from 1999 to Similar calculations can be done for Phases 2 and 3. Overall Total of Annual Emissions Reductions: Overall total of all of the annual emissions reductions due to Drive Clean for: HC and NOx amounted to 58,680 tonnes, which is 58.4% of the total of 25

26 overall total of all of the annual emissions reductions from all factors; CO amounted to 522,000 tonnes, which is 72.7% of the total of overall total of all of the annual emissions reductions from all factors. Phase 2: Mandatory Testing Started January 1, 2001 Testing of vehicles in Phase 2 of the LDV Drive Clean program became mandatory on January 1, 2001 in urban and commuter areas in southern Ontario from Windsor to Peterborough. In 2002, 805,766 individual vehicles received 944,822 tests in the Phase 2 area. In 2003, 821,346 vehicles received 955,657 tests giving a total of approximately three million tests performed since January There were 95,313 vehicles that failed their initial inspection in 2003, resulting in a failure rate of 11.6 %. Many of the vehicles would have been tested on two separate occasions for registration renewal: in 2001 and From 2001 to 2003, relative to the baseline emissions at the start of testing, the combined effects of repairing vehicles that failed the Drive Clean test and the other factors reduced: HC and NOx by an estimated cumulative total of 28.4% or 16,275 tonnes; CO by an estimated cumulative total of 22.4%, or 113,575 tonnes. Without Drive Clean for the year 2003, relative to 2001 baseline emissions at the start of testing: HC and NOx would have been reduced by 15.9%, or 9,113 tonnes; CO would have been reduced by 11.0% or 55,548 tonnes. Therefore for the year 2003, Drive Clean repairs alone reduced: HC and NOx by 14.8% or 7,162 tonnes: CO by 11.4% or 58,027 tonnes. Annual Emissions Reductions In 2002 the Drive Clean program reduced the 2002 baseline inventory of: - HC by 18.9% or 5,190 tonnes; - CO by 14.2% or 70,172 tonnes; - NOx by 10.9% or 2,877 tonnes; and - HC+NOx by 15.0% or 8,067 tonnes. 26

27 In 2003 the Drive Clean program reduced the 2003 baseline inventory of: - HC by 21.0% or 5,079 tonnes; - CO by 12.8% or 58,027 tonnes; - NOx by 8.6% or 2,083 tonnes; and - HC+NOx by 12.5% or 7,162 tonnes. Overall Total of All of the Annual Emissions Reductions: The overall total of all of the annual emissions reductions due to Drive Clean for: HC and NOx amounted to 19,195 tonnes, which is 60.4% of the total of all of the annual emissions reductions from all factors; CO amounted to 154,922 tonnes, which is 69.4% of the total of all of the annual emissions reductions from all factors. Phase 3: Mandatory Testing Started July 1, 2002 Testing of vehicles in Phase 3 of the LDV Drive Clean program became mandatory on July 1, 2002 in the expanded program area, (covering the remaining areas of the southern Ontario smog zone from Windsor to Cornwall). In 2002, 285,938 individual vehicles received 326,171 tests in the Phase 3 area. In 2003, 484,525 vehicles, mostly from even model years, received 557,145 tests, giving a total of approximately 0.8 million tests performed since July 1, There were 59,258 vehicles that failed their initial inspection in 2003, resulting in a failure rate of 12.4%. As 2002 was the baseline, or starting year, emissions were reduced only as a result of Drive Clean repairs. From 2002 to 2003, relative to baseline emissions at the start of testing, the combined effects of repairing vehicles that failed the Drive Clean test and the other factors reduced: HC and NOx by an estimated cumulative total of 18.3%, or 7,224 tonnes; CO by an estimated cumulative total of 13.7 % or 49,618 tonnes. Without the Drive Clean Program, relative to baseline emissions at the start of testing: HC and NOx would have been reduced by an estimated cumulative total of 12.3 %, or 4,860 tonnes; CO reduction would have been reduced by an estimated cumulative total of 10.4 % or 37,555 tonnes. 27

28 Therefore, in 2003, Drive Clean repairs alone reduced: HC and NOx by an estimated cumulative total of 6.0% or 2,364 tonnes; CO by an estimated cumulative total of 3.3% or 12,063 tonnes. Annual Emissions Reductions In 2002 the Drive Clean program reduced the 2002 baseline inventory of: - HC by 3.5% or 709 tonnes; - CO by 1.9% or 6,705 tonnes; - NOx by 1.8% or 344 tonnes; and - HC+NOx by 2.7% or 1,053 tonnes. In 2003 the Drive Clean program reduced the 2003 baseline inventory of: - HC by 10.4% or 1,803 tonnes; - CO by 3.7% or 12,063 tonnes; - NOx by 3.3% or 561 tonnes; and - HC+NOx by 6.8% or 2,364 tonnes. Overall Total of All of the Annual Emissions Reductions: The overall total of all of the annual emissions reductions due to Drive Clean for: HC and NOx amounted to 3,417 tonnes, which is 41.3% of the total of all of the annual emissions reductions from all factors; CO amounted to 18,768 tonnes, which is 33.3% of the total of all of the annual emissions reductions from all factors. Total for All Three Phases: The Southern Ontario Smog Zone Since Drive Clean began in 1999 in Phase 1 to the end of 2003 when all three Phases were in place in the southern Ontario smog zone, the combined effects of repairing vehicles that failed the Drive Clean test and the other factors for the year 2003, relative to the start of testing in each of the Phases reduced: HC and NOx by an estimated cumulative total of 56,570 tonnes; CO by an estimated cumulative total of 389,319 tonnes Without the Drive Clean Program for the year 2003, relative to the start of testing in each of the Phases: 28

29 HC and NOx would have been reduced by an estimated cumulative total of 33,450 tonnes; CO would have been reduced by an estimated cumulative total of 190,330 tonnes. Annual Emissions Reductions: All Phases In 2002 the Drive Clean program reduced the 2002 baseline inventories of: - HC by 20.3% or17,783 tonnes; - CO by 14.7% or 239,525 tonnes; - NOx by 9.7% or 8,437 tonnes; and - HC+NOx by 15.0% or 26,220 tonnes. In 2003 the Drive Clean program reduced the 2003 baseline inventories of: - HC by 22.8% or 17,529 tonnes; - CO by 13.4% or 199,989 tonnes; - NOx by 7.0% or 5,591 tonnes; and - HC+NOx by 14.8% or 23,120 tonnes. 29

30 Overall Total of All of the Annual Emissions Reductions: The overall total of all of the annual emissions reductions for all three Phases due to Drive Clean for: HC and NOx amounted to 81,292 tonnes, which is 57.8% of the total of the all of the annual emissions reductions from all factors; CO amounted to 696,596 tonnes, which is 69.9% of the total of all of the annual emissions reductions from all factors. These annual reductions are summarized in the Table below. Note that the tonnages for 1999 are the same on both sides of the table since 1999 is the start-up year used as the baseline so that the only emissions reductions are those due to Drive Clean. Drive Clean Reductions (tonnes) Overall Reduction from All Factors Based on Emissions in Start-up Years (tonnes) HC CO NO HC+NO HC CO NO HC+NO ,331 45,325 1,741 5,072 3,331 45,325 1,741 5, ,547 90,939 3,434 10,981 9, ,729 4,887 14, , ,817 4,617 15,899 15, ,397 7,524 23, , ,525 8,437 26,220 26, ,378 14,164 41, , ,989 5,591 23,120 37, ,319 19,245 56,570 Totals 57, ,596 23,819 81,292 93, ,149 47, ,569 Inventory Reductions The following charts show how vehicle emissions in the three Phases would have changed without Drive Clean and did change with Drive Clean. 30

31 phase 1 : HC Inventory Reductions 50,000 nventory (tonnes) ons i emissi 40,000 30,000 20,000 10,000 MOBILE 6.2C default actual, with DC calendar year phase 1 : CO Inventory Reductions 1,000,000 emissions inventory (tonnes) 800, , , ,000 MOBILE 6.2C default actual, with DC calendar year phase 1 : NOx Inventory Reductions 50,000 emissions inventory (tonnes) 40,000 30,000 20,000 10,000 MOBILE 6.2C default actual, with DC calendar year 31

32 phase 2 : HC Inventory Reductions 40,000 nventory (tonnes) emissions i 30,000 20,000 10,000 MOBILE 6.2C default actual, with DC calendar year phase 2 : CO Inventory Reductions 600,000 nventory (tonnes) ons i emissi 500, , , , ,000 MOBILE 6.2C default actual, with DC calendar year phase 2 : NOx Inventory Reductions 30,000 emissions inventory (tonnes) 25,000 20,000 15,000 10,000 5,000 MOBILE 6.2C default actual, with DC calendar year 32