51ÁÔÆæÈë¿Ú

Project concept infographic: Decarbonization by reducing activity (VkT) in parallel to electrification 

Partner workshop at SFU, 2026

Explore the Insights from Our July 2024 Workshop! Click on the image to download the "What We Heard" report, based on feedback from our community partners.

This project was featured by the Civics Innovation Lab!

Click on the image to check out the newsletter.  

An output of traffic model - the difference between morning rush hour traffic volume between the current state and future extreme exapnsion of SkyTrain in Metro Vancouver

Agent-based Travel Demand Model

ASIM (Activity Simulator) is an advanced agent-based transportation modelling platform developed by the UBC Integrated Transport Research (UiTR) Lab at UBC Okanagan.

Applied to Metro Vancouver, ASIM creates a detailed virtual representation of approximately 2.7 million people, simulating the daily activities and travel patterns of individual residents for the 2023 base year. It models when and where people undertake activities, where they travel, how they travel, and how their individual travel decisions interact with the transportation network.

By combining detailed individual-level travel behaviour with large-scale transportation network simulation, ASIM provides a powerful platform for testing transportation policies, infrastructure investments, emerging technologies, and future scenarios in a realistic virtual environment before implementation in the real world.

This figure illustrates the spatial distribution of LDV activity across dissemination areas (DAs). Regional PM2.5 emissions intensity varies from under 10 g/km²/day in rural areas to nearly 9,000 g/km²/day in dense urban centers. In Metro Vancouver, baseline LDV emissions heavily concentrate along major transportation infrastructure, specifically the Trans-Canada Highway corridor and the urban cores of Vancouver, Burnaby, and Surrey.

Vehicle and Traffic Emissions Model (MOVES / UTEC)

The mobile-source (vehicle) emissions model is built on the U.S. EPA’s regulatory-grade MOVES framework and has been customized to reflect the Canadian context. A version of the model is available for public use through uteclab.com. The model uses traffic-model outputs as its primary activity input and integrates vehicle registration data to develop a refined, locally representative fleet composition. This information can be further enhanced using traffic-camera imagery to improve characterization of the on-road fleet. The model accounts for multiple emission processes, including running exhaust, idling, cold and hot starts, and non-exhaust emissions such as tire and brake wear particles. Emissions are also sensitive to vehicle-specific power, which is calculated using traffic speed and operating conditions for each roadway link between intersections. The resulting output is a bottom-up, link-based emissions inventory with high spatial and temporal resolution. In the next step, this mobile-source inventory is combined with emissions from non-mobile sources to develop the comprehensive emissions inventory used for subsequent air-quality modeling.

A map of planetary boundary layer height generated by WRF for the Lower Mainland coastal domain.

Model 3 - Meteorology and Weather model using WRF

Our framework uses the Weather Research and Forecasting (WRF) model to generate the high-resolution meteorological fields required for regional air quality simulations. Because Metro Vancouver features complex shorelines and steep mountain terrain, standard weather models often struggle to capture local wind patterns.

We employ a three-tiered nested domain configuration scaling down from a 9-km grid covering the Pacific Northwest to a localized 1-km grid over Metro Vancouver. To ensure optimal performance across seasons, we conducted a rigorous sensitivity analysis evaluating multiple planetary boundary layer, land-surface, radiation, and microphysics parameterizations. By integrating high-resolution land-use data with these season-specific physics configurations, the WRF setup accurately simulates both summer sea-breeze ventilation and winter stagnation episodes. This tailored meteorological foundation ensures that subsequent pollutant transport, mixing, and chemical transformation processes are simulated with high fidelity.

Spatial distribution of selected pollutant concentrations averaged over the dissemination areas of the Metro Vancouver Regional District.

Model 4 - Chemical Transport Model using CMAQ

We utilize the Community Multiscale Air Quality (CMAQ) model to simulate the atmospheric chemistry and transport of pollutants across the region. Driven by our optimized WRF meteorology and comprehensive bottom-up emissions inventories, CMAQ tracks the life cycle of critical criteria pollutants including fine particulate matter (PM_2.5), nitrogen dioxide (NO₂), and ozone (O₃).

The system incorporates advanced gas-phase and aerosol chemistry mechanisms to capture secondary pollutant formation, deposition, and regional transport pathways. By resolving concentration gradients at a 1-km scale, this modeling platform allows us to evaluate how localized transportation policies, emerging technologies, and behavioral shifts will impact neighborhood-level air quality.

Estimated annual all-cause mortality attributable to acute NOâ‚‚ exposure across Metro Vancouver in 2017. The spatial pattern reflects the combined influence of NOâ‚‚ concentrations and baseline mortality rates, illustrating substantial local variation in estimated health impacts

Human Health Risk Assessment (HHRA) model

HHRA model links spatially and temporally resolved air-pollution concentrations with population data, baseline health rates, and epidemiological concentration-response relationships to estimate the health impacts of air pollution across Metro Vancouver. Using high-resolution WRF-CMAQ air pollution estimates, the model can quantify attributable health outcomes and benefits of potential transportation and emission-reduction interventions, while examining how impacts vary across locations and over time.

A key feature of the framework is its ability to move beyond the aggregated, regional health impact average or total and examine the fine-level spatial and temporal patterns of estimated impacts. The example shown here maps the annual all-cause mortality attributable to acute NOâ‚‚ exposure in 2017. The results reveal substantial spatial heterogeneity: areas with higher NOâ‚‚ exposure and higher baseline mortality rates experience larger estimated health impacts. This provides additional information for evaluating where and when intervention benefits occur and whether those benefits are evenly distributed, supporting more health- and equity-informed planning and policy decisions.