Inequities in Canadian Traffic and Congestion Exposure Under Socioeconomic Pathways
This study develops a machine-learning framework to project traffic volume and congestion across Canada through 2100 under five socioeconomic scenarios, revealing that future congestion will be unevenly distributed with persistent inequities exposing minority groups to significantly higher traffic levels than non-minority populations.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Traffic is more than just a daily frustration; it is a fundamental force shaping how people access jobs, healthcare, and fresh air. When roads are clogged, the consequences ripple outward, slowing down economies and filling the air with pollutants that harm human health. These problems are not distributed equally. In many places, the people who suffer the most from traffic fumes and delays are often those with the least power to change their situation, such as low-income families or racial minorities. Understanding exactly where and why these traffic patterns form is difficult because the road network is vast, constantly changing, and often poorly mapped in detail. For decades, planners have relied on surveys that capture only a tiny fraction of drivers or models that are too coarse to see the specific bottlenecks affecting a single neighborhood. Without a clear, high-resolution picture of the entire system, it is hard to know where to build new roads, where to improve public transit, or how to protect vulnerable communities from the growing burden of congestion.
A team of researchers at the University of British Columbia has tackled this challenge by building a new way to map traffic across all of Canada, looking far into the future. Instead of guessing based on limited data, they created a computer framework that learns from real-world observations to predict how traffic will behave on nearly every major road in the country, from the bustling streets of Toronto to the remote highways of the Yukon. They did this by training a machine-learning system on millions of actual traffic counts, teaching it to recognize the patterns that drive vehicles onto the road. The system considers a wide range of factors, including how many people live nearby, where jobs are located, the shape of the land, and even the type of road itself. By feeding this system different scenarios of how Canada's population and economy might grow or shrink over the next eighty years, the researchers were able to simulate how traffic volume and congestion will evolve under various possible futures.
The results reveal a landscape of traffic that is deeply uneven. Currently, the vast majority of vehicles are concentrated in metropolitan areas, where they crowd onto major highways and arterial roads. The study projects that while some roads will see their traffic levels stabilize or even decline, others will become increasingly clogged, particularly along the major corridors connecting Canada's largest cities. The severity of this congestion depends heavily on which path society takes. In one scenario where development is fragmented and economic growth is slow, traffic might ease in some eastern provinces. In another scenario driven by rapid economic expansion and fossil fuel use, traffic volumes could surge dramatically, turning many more road segments into gridlock by the end of the century. The simulations show that these changes will not happen everywhere at once; instead, the heaviest impacts will be felt in specific urban corridors, while rural connecting roads may remain relatively free-flowing.
Perhaps the most striking finding concerns who bears the burden of this traffic. The researchers calculated how much traffic each person is exposed to simply by living near a road, using a buffer zone that represents the distance where air pollution from vehicles remains a significant health risk. They found that this exposure is not shared equally. In both cities and towns, racialized minority groups face significantly higher levels of traffic exposure than the non-minority population. For example, in non-metropolitan areas, Arab and West Asian populations experience nearly twice the traffic exposure of their neighbors. In urban centers, the "Other Minority" category faces a quarter more exposure than non-minorities. These disparities persist even when the researchers accounted for income levels and education, suggesting that where people live is strongly tied to their race and ethnicity, placing them in closer proximity to the noise and fumes of heavy traffic.
The study also identified the specific factors that drive traffic volume, offering clues for how to manage it. In cities, the number of people living within a short distance is the single biggest predictor of how many cars will be on the road. In rural areas, however, the location of a road relative to the north-south population gradient matters more than immediate population density. Employment levels also play a major role, particularly for the major highways that connect cities. The researchers noted that while some of these drivers, like population growth, are difficult to control, others, such as transit accessibility and traffic speeds, can be influenced by policy. By pinpointing exactly which roads and which communities are most likely to face severe congestion in the future, this work provides a detailed map for planners. It suggests that to build a fairer and more sustainable transportation system, interventions must be targeted not just at the busiest highways, but specifically at the neighborhoods where vulnerable populations are already paying the highest price for the country's reliance on cars.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.