Investigating the impact of cold-starts on the distribution of vehicle emissions in the Greater Toronto Area

Vehicle emissions are a significant source of greenhouse gases and air quality pollutants in urban areas, yet current city-scale CO and CO₂ emission inventories may not accurately reflect real-world conditions. In Toronto, Canada, traffic emissions contribute to approximately one third of the city’s total CO₂ emissions. As a part of the Toronto Atmospheric Monitoring of Emissions (TAME) project, the goal of this work is to investigate the emission signatures of the vehicle fleet in the Greater Toronto Area (GTA) under different engine operating conditions and seasons. Specifically, we are interested in the role of ‘cold-start’ emissions (i.e. emissions during the engine and catalyst warm-up period) on the distribution of CO emissions in the GTA. As exhaust aftertreatment technologies improve for gasoline engines, air quality emissions that occur before the catalytic converter has warmed are expected to contribute an increasing, and possibly dominant, proportion of non-CO₂ vehicle emissions. Simultaneous measurements of CO and CO₂ were collected by deploying calibrated low-cost sensors in parking garages at the University of Toronto over several months. To calculate the CO and CO₂ enhancements from each vehicle emission plume, a peak-finding algorithm was developed. The ΔCO/ΔCO₂ ratio of the enhancements is used as an emission signature for each vehicle. The results from this study are compared to mobile and stationary measurement campaigns conducted by Environment and Climate Change Canada using a high precision analyzer (cavity ring-down spectrometer (CRDS)).  We assess the impact of choices made about data collection, peak-finding, and cold-start definitions. Using one approach, the range of ΔCO/ΔCO₂ ratios observed is 0.1 to 779 ppb CO/ ppm CO₂, with a median of 14 ppb CO/ ppm CO₂. On average, the cold-start emission ratios are observed to be at least 2 times greater than those of warm vehicles. These results can be used to update CO and CO₂ emission inventories to more accurately capture activity patterns and independently verify emission reductions as urban vehicle fleets transition to electric.