Research
Bicycle-based air pollution measurements
We monitor air pollution using a mobile, bicycle-based monitoring platform to measure various aspects of particulate air pollution (e.g., particle number concentration, black carbon, PM2.5, and particle size distribution). We use these measurments to (1) describe typical concentrations in various urban environments, (2) develop spatial models of air quality, and (3) advance our understanding of the spatiotemporal patterns of exposure.
A news clip describing these measurements can be seen here, a newspaper article here, and another article here.
We recorded video for a portion of the monitoring runs in a Minneapolis study. The video clip below shows short clips of footage for segments of the sampling routes in low and high air pollution environments. We ranked all black carbon measurements by percentile and binned by quintle; then we randomly pulled clips that had a minimum of 30 consecutive seconds in one quintile. The video shows compiled clips in each quintile (increasing from low to high black carbon concentrations).
Land use regression models of urban air quality
We are working to develop new methods to allow for better temporal resolution in land use regression modeling. An example is the use of mobile monitoring to develop hourly land use regression models rather than long-term (i.e., annual average) models. An example of this work can be seen in this journal article.
We are also working to (1) automate mobile data collection for use in air quality modeling and (2) develop near real-time prediction as part of a larger effort to develop air quality models for use in health effects studies (link).
Measuring and modeling non-motorized traffic
Measurements and models of bicycle and pedestrian traffic are needed to better plan for active travel. We have worked to collect long-term counts of mixed-mode traffic on urban trails in Minneapolis, MN and Blacksburg, VA. Research project pages (Minneapolis, Blacksburg), a video, and journal articles (here and here).
As part of the Mid-Atlantic Transportation Sustainability University Transportation Center (MATS-UTC), we have systematically designed a monitoring campaign to describe non-motorized traffic patterns for the entire transportation network in Blacksburg, VA. Our approach involves a two stage process: (1) siting a long-term reference network of automated counters and performing short-duration counts (~1 week) to estimate AADT on ~10% of the street segments in Blacksburg and (2) developing regression models based on land use and characteristics of the street network to estimate AADT at locations where counts were not collected.
We are also working with MATS-UTC to collect and archive bicycle and pedestrian counts for ~20 major US cities. We will use these counts to develop direct-demand models of bicycle and pedestrian traffic across the country. The research project page is here.
Designing healthy, clean neighborhoods
A widely studied aspect of designing healthy cities focuses on increasing physical activity. A less studied area is how to mitigate exposure to hazards during active travel (e.g., air pollution). A key question is how development patterns impact air pollution and health as well as greenhouse gas emissions.
A journal article describing a risk assessment of air pollution and physical inactivity for different neighborhoods in Los Angeles and Minneapolis. Journal articles that estimate the built environment's impacts on motor vehicle greenhouse gas emissions are here and here. Additional manuscripts on this topic are currently in review.