Understanding and quantifying motor vehicle emissions with vehicle specific power and TILDAS remote sensing

Abstract

Motor vehicles are one of the largest sources of air pollutants worldwide. Despite their importance, motor vehicle emissions are inadequately understood and quantified. This is due in part to large variations in individual vehicle emissions with changing operating conditions, and to significant differences between vehicles. To better relate emissions with operating conditions, a new parameter termed "specific power" (SP) is presented. SP is the instantaneous tractive power per unit vehicle mass. This parameter has three main advantages: it can be calculated from roadside measurements, it captures most of the dependence of light-duty vehicle emissions on driving conditions, and it is directly specified in emissions certification cycles. The dependence of CO, HC, and NOx emissions on SP is better than on several other commonly used parameters, such as speed, acceleration, power, or fuel rate. Using SP as the basic metric allows meaningful comparisons to be made between data from different remote sensing sites, dynamometer driving cycles, and emission models. Modem U.S. vehicles are likely to operate under commanded enrichment when SP exceeds the maximum value on the Federal Test Procedure (-22 kW/Metric Ton). This may allow transient high emissions to be screened out during future remote sensing campaigns. Remote sensing can address the problem of inter-vehicle differences by quickly and cheaply measuring the emissions of large numbers of vehicles. Here, a tunable infrared laser differential absorption spectrometer (TILDAS) remote sensor was used to gather the first on-road measurements of N20 and N02, and the first high precision measurements of NO. NO was detected with a sensitivity of 5 ppm, which allowed even Ultra Low Emission Vehicles to be measured. On-road accuracy was demonstrated by comparing the TILDAS results with the on-board measurements of a heavy-duty diesel truck (HDDT). The remote sensor could operate with an optical path length of 88 meters, more than five times that of competing instruments. The NO and N20 emission distributions of passenger cars (PCs) and light-duty trucks (LDTs) were found to be highly skewed, while the NO emission distribution for HDDTs was not. N20 emissions from PCs and LDTs are estimated to contribute between 0.5% and 0.9% to U.S. greenhouse gas emissions.