Capturing the effects of urban drive cycles and passenger ridership on transit bus emissions and investigating the potential of emission reduction strategies

Urban transportation is a major contributor to greenhouse gas (GHG) emissions and air pollution. Worldwide, planners and policy makers are consistently encouraging individuals to reduce their reliance on private vehicles while promoting the use of public transit. While transit buses reduce per-passenger emissions of GHG and air pollutants, they generate a large amount of emissions on a vehicle basis especially when they are fuelled by conventional diesel. This research was motivated by the importance of reducing transit bus emissions which can only be achieved after understanding the factors affecting transit bus emissions and exploring the effects of various emission reduction strategies. We started with exploring the emission estimation methods currently embedded within emission inventory models followed by a validation of the most commonly used model in North America, the Motor Vehicle Emissions Simulator (MOVES), in a local context by collecting instantaneous bus speed and passenger ridership data across a variety of transit routes (downtown and suburban/highway) and bus types (standard, articulated, old, and new). We observed a lack of transit bus drive cycles in MOVES and significant differences in emissions when MOVES uses it’s embedded drive cycles to estimate emissions. To improve the estimates in MOVES, we then tested the effects of incorporating local drive cycles into the MOVES model by replacing embedded default drive cycles. A significant improvement was observed in emission estimation, showing a reduction of the average estimation error from 23% to 13%. In order to understand how emissions are affected, we analyzed the spatial and temporal variability of transit bus emissions across the island of Montreal and investigated the isolated and combined effects of different factors affecting emissions (including the level of congestion, roadway grade, passenger load, and traffic variability). The level of congestion and higher road grade were found to be the most important factors. The trade-off between total and per-passenger emissions was also analyzed under varying passenger loads. While an increasing passenger load on the bus increases emissions, we observed that the addition of each passenger influences the per-passenger emissions differently, depending on the bus occupancy. The reduction potential of using different fuels including ultra-low sulfur diesel or compressed natural gas; and of transit service operational improvements was also investigated considering a wide combination of congestion, roadway grades, and passenger ridership. We observed that the effectiveness of the improvements could vary considerably depending on the level of congestion. While compressed natural gas could achieve 8-12% GHG reduction in both congested and uncongested networks, other transit improvements such as transit signal priority and queue jumper lanes could achieve an even higher reduction. Finally, a corridor study was conducted to capture the changes in emissions as a result of the implementation of different transit service improvement strategies including smart cards, express bus service, and reserved bus lanes. Our results suggested that a reduction of 40% in GHG emissions could be possible by operating limited-stop express buses on reserved bus lanes compared to regular buses with no reserved lane. This thesis addressed critical gaps in the current knowledge of transit bus emissions in four ways: it evaluated the most commonly used emission inventory model in a local context, it demonstrated a process to embed local drive cycles into the emission model, it quantified the individual and combined effects of different factors on transit bus emissions, and it quantified the emission reduction potential of different transit improvement strategies and alternative fuels, which would be crucial when implementing emission reduction strategies or modifying existing transit facilities.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.