Identifying Mobile Source

Transcription

1 Identifying Mobile Source Contributions to Particulate Matter Douglas R. Lawson MARAMA Workshop on Energy and Air Quality Issues September 23, 2008 NREL is a national laboratory of the U.S. Department of Energy Office of Energy Efficiency and Renewable Energy operated by Midwest Research Institute Battelle

2 Ozone and PM 2.5 Nonattainment Counties in U.S. From:

3 Ref: Latest Findings on National Air Quality: 2001 Status and Trends

4 Northern Front Range Air Quality Study Denver Welby Monitoring Site, Winter 1996/97, PM Blame Apportionment Secondary Species 12% 25% 3% 13% 10% 10% 2% 3% 2% 16% 4% Gasoline Cold ldstartt Gasoline Hot Stabilized Gasoline High Emitter Diesel Exhaust Dust and Debris Meat Cooking Softwood Smoke Hardwood Smoke Coal Power Station Ammonium Sulfate Ammonium Nitrate

5 Washington DC PM 2.5 Apportionment Aug. '88 to Dec. ' Samples 58% 9% 21% 2% 2% 4% 2% 2% Gasoline Exhaust Diesel Exhaust Soil Incinerator Oil Combustion Sea Salt Sulfates Nitrates Ref: Kim and Hopke, JAWMA, July 2004

6 Contributions to Ambient PM Carbonaceous PM is the most important contributor to PM in U.S. urban locations What is the source of the carbonaceous PM? Is it primary or secondary, and how much of each is present? How much is from mobile and/or other source categories? How much is from spark-ignition (gasoline) vs. compression ignition (diesel)? How much is from on-road vs. off-road mobile sources?

7 DOE Comparative Toxicity Study Samples Gasoline (5 vehicles) 72 G Composition Gasoline 30 o G 30 White smoker gas. WG Black smo ker gas. BG Diesel (3 vehicles) 72 D Diesel 30 o D 30 High-emitter diesel HD Lung Inflammation per Unit Mass WG HD BG D 2-component PCA Analysis D 30 G G 30 Refs: Seagrave et al., Toxicol. Sci. 70: , 2002 Zielinska et al. J. Air Waste Man. Assoc. 54: , 2004 Hopanes and steranes, markers of crankcase oil, co-varied most closely with toxicity

8 We need to find/fix/scrap/repair the high emitters RELATIVE INFLAMMATORY EFFECT PER UNIT OF MASS G D HD BSG WSG Normal gasoline set to RELATIVE INFLAMMATORY HAZARD PER UNIT OF VEHICLE TRAVEL 60 One WSG emits as much hazard as normal gasoline vehicles! G D HD BSG WSG RELATIVE EMISSION RATES G D HD BSG WSG

9 EPA/DOE/CRC/FHWA Kansas City Gasoline Vehicle PM Emissions Study Num mber ` Model Year ons Pe ercentage of PM Emissi % of Fleet, Ranked Dirty to Clean Even though the fleet was heavily weighted toward older vehicles, only 10% of the fleet produced 63% of the PM. The true on-road fleet has much greater skewness. Median emission rate = 4 mg/mi; median rate = 14 mg/mi, even over the very aggressive UDC; these values from an older fleet.

10 PM Emissions from a Gasoline Vehicle PM from normal emitter s tailpipe = 1-2 mg/mile PM from Amy Winehouse s vehicle at 60 mph = 5 mg/mile (emitted from open window)

11 Second-by-Second Data from a 6-yr old pre-obdii SUV Unified Driving Cycle (UDC) Jeep Cherokee (No.1-4) -- 45,359 miles Speed, mph Seconds CO Emissions, mg/sec; 1.5 g/mi HC Emissions, mg/sec; 0.10 g/mi NOx Emissions, mg/sec; 0.17 g/mi PM Emissions, mg/sec; 5.78 mg/mi Black Carbon Emissions, mg/sec Seconds

12 Jack Kevorkian March 24, 2008 Out of work Will run for Congress!

13 New Program Collaborative Lubricating Oil Study on Emissions (CLOSE) Study Objective Quantify the relative contributions of fuel and engine lubricating oil to motor vehicle particulate matter (PM) and semivolatile organic compound (SVOC) emissions i through h extensive chemical and physical characterization of emissions under a variety of engine operating conditions

14 CLOSE Program Test Matrix Test Temperature 72 F (nominal) 20 F Test Lubricant Fresh Aged Fresh Aged Sample Number LD gasoline (normal PM emitter) LD gasoline (high PM emitter) LD E10 (normal PM emitter) LD E10 (high PM emitter) MD CARB diesel (normal PM emitter) MD CARB diesel (high PM emitter) Completed Comple ted MD B20 biodiesel (normal PM emitter) MD B20 biodiesel (high PM emitter) HD CNG (normal PM emitter) HD CNG (high PM emitter) HD CARB diesel (normal PM emitter) HD CARB diesel (high PM emitter) LD Driving Cycle: California Unified Cycle (LA-92) HD Driving Cycle: EPA HD Urban Dynamometer Driving Schedule (heavy duty chassis cycle)

15 What do on-road (real- world) emissions data tell us about our in-use mobile source control programs?

16 Nationwide On-Road Idle HC Emissions EPA s 1985 National Tampering Survey 6498 vehicles On average, fleet emissions increase as vehicles age. Most old cars are clean ; some new cars are dirty. Weighted eg for VMT, the high emitters are most important. New vehicles irrelevant to air quality. Adopting CA LEV standards will do little, if anything, to improve ozone air quality Ref: Lawson et al. 1996

17 Remote Sensing Data from Denver Area Speer Blvd. RS HC Emissions by Quintile 10,015 Measurements, Dec. 3, 5, 6, Average HC Emissions, % Cleanest Quintile nd Cleanest Marginal Dirtier Dirtiest Quintile (25) 1985 Model Year ) (

18 Remote Sensing Data from Denver Area Contribution to Total HC Emissions by Quintile 10,015 RS Measurements, Speer Blvd., Dec. 3, 5, 6, % of Total HC 3.0 Cleanest Quintile (25) Model Year (111 3) nd Cleanest Marginal Dirtier Dirtiest Quintile

19 The High Emitter Problem The most difficult of all the mobile source emission i problems; that s why they re still on the road. Fleet turnover has not solved the problem. New technologies have done only a little to fix the problem. I/M programs have done little to reduce on-road emissions (NRC report, 2001). Innocent until proven guilty. PM repairs are the most costly; cost waivers exceeded. For the MSATs, one high emitter = 1000 normal emitters in hot stabilized conditions

20 California Low-Emitting Vehicle (LEV) We easily forget that smog is the price of freedom of our streets from manure, and from the flies and diseases it brought. Daniel J. Boorstin ( ), Librarian of Congress from

21 What about the urban/regional ozone problem?

22 Weekend Ozone Effect Mean Sunday to mean Wednesday daytime (6 am through 3 pm) CO and NOx and peak one-hour and eight-hour ozone in the mid-atlantic region, all days, March-October