Gross polluters and vehicles' emissions reduction

Abstract: Vehicles' emissions produce a significant share of cities' air pollution, with a substantial impact on the environment and human health. Traditional emission estimation methods use remote sensing stations, missing vehicles' full driving cycle, or focus on a few vehicles. We use GPS traces and a microscopic model to analyse the emissions of four air pollutants from thousands of private vehicles in three European cities. We find that the emissions across the vehicles and roads are well approximated by heavy-tailed distributions and thus discover the existence of gross polluters, vehicles responsible for the greatest quantity of emissions, and grossly polluted roads, which suffer the greatest amount of emissions. Our simulations show that emissions reduction policies targeting gross polluters are way more effective than those limiting circulation based on a non-informed choice of vehicles. Our study contributes to shaping the discussion on how to measure emissions with digital data.

• The paper employs a microscopic emissions model with GPS data to reveal that the top 10% of vehicles contribute nearly 40-50% of CO2 emissions.
• It identifies distinct distribution patterns, with London’s emissions fitting a stretched exponential model while Rome and Florence follow truncated power law distributions.
• Simulation scenarios show that targeting electrification to just 1% of gross polluters can achieve emissions reductions equivalent to electrifying 10% of vehicles at random.

The paper "Gross polluters and vehicles' emissions reduction" introduces a novel framework to assess vehicle emissions using GPS data and a microscopic emissions model. This research is conducted by analyzing data from thousands of private vehicles in three major European cities: Greater London, Rome, and Florence. The cities were chosen for their distinct characteristics in terms of urban size and road network density. The methodological approach adopted in this study allows for a detailed evaluation of emissions across various geospatial and temporal dimensions, thereby addressing the limitations of traditional emission estimation methods.

Key Findings:

1. Emissions Distribution: The study identifies that emissions follow heavy-tailed distributions both across vehicles and roads, leading to the identification of gross polluters and grossly polluted roads. Specifically, a small percentage of vehicles contribute disproportionately to the air pollution, with the top 10% of gross polluters responsible for nearly 40-50% of CO$_2$ emissions, depending on the city.
2. Patterns Across Cities: The emissions distribution patterns vary across cities. For example, the London emissions distribution across vehicles is best approximated by a stretched exponential model, while Rome and Florence exhibit truncated power law distributions. This suggests variations in urban design and driving behaviors significantly influence emission patterns.
3. Impact of Driver Behavior: The analysis reveals a strong correlation between vehicular emissions and predictable travel behavior. Gross polluters tend to exhibit more regular travel patterns, highlighted by lower mobility entropy scores. This finding indicates that emissions are less related to erratic travel behaviors, but rather associated with consistent, high-emission activities.
4. Simulation Scenarios: The paper explores various emissions reduction scenarios through electrification and remote working. It finds that targeting gross polluters for electrification is significantly more effective in reducing emissions than non-discriminatory approaches. Simulating the transition to electric vehicles for just 1% of the most polluting vehicles can lead to emission reductions equivalent to electrifying 10% of vehicles chosen at random.
5. Implications for Policy: The results underscore the potential of data-driven policymaking to efficiently target emissions reduction strategies. By focusing on gross polluters, cities can achieve significant reductions in emissions with a fraction of the effort compared to broader, less targeted approaches.

Methodological Contributions:

• Data Utilization: The use of detailed GPS trajectories allows the computation of instantaneous speed and acceleration, providing a granular view of emissions in real-time and offering a comprehensive approach over spatial and temporal dimensions.
• Emissions Model: The microscopic emissions model employed in the study uses detailed vehicle movement data to provide estimates of various pollutants such as CO$_2$, NO$_x$, PM, and VOCs.
• Urban Structure Effects: The study highlights how different city structures and driving habits influence emissions, suggesting a tailored approach to emissions reduction based on urban characteristics and specific vehicular patterns.

Conclusion:

This research provides a significant advancement in understanding vehicular emissions patterns and lays the groundwork for effective policy measures targeting specific emission sources in urban environments. By integrating digital data and sophisticated modeling, it supports decision-makers in crafting informed, data-driven strategies that align with sustainable development goals and urban well-being improvements.