Uncovering Spatio-temporal Air Pollution Exposure Patterns During Commutes to Create an Open-Data Endpoint for Routing Purposes

TLDR: In this paper, high-resolution air pollution models are used to distinguish between routes of higher and lower air pollution concentrations, allowing active road users to choose an alternative route to lower their air pollution exposure.

Abstract: Air pollution is difficult to detect with human senses. It is to a large extent out of sight and out of sense, while causing a burden on our economy, our health and our environment. A relevant illustration of this is the exposure to air pollution during commutes. The air pollution commuters are exposed to remains to a considerable extent a hidden geography, with, for example, a lack of available reliable information regarding the on-the-road concentrations of several air pollutants. This research aims to unravel, to the best possible extent, spatio-temporal air pollution patterns (active) commuters are exposed to. Cyclists and pedestrians can be unaware that they commute in polluted air. They often travel close to motorised traffic, resulting in high exposure to several air pollutants, which have elevated levels on the road due to vehicular emissions. Significantly higher concentrations of particulate matter (<2.5 µm), black carbon and nitrogen dioxide were found on roads with high-traffic intensities than on roads with less traffic, cycling highways or separated cycle lanes. The amplitude of the concentration differences between routes depends on both temporal factors, such as the season, the day of the week, or the time of day, and spatial factors, such as the traffic’s density, the footpath or cycle lane’s location, the architectural makeup (e.g. street canyons) and the meteorological conditions. Using high-resolution air pollution models, it is possible to distinguish between routes of higher and lower air pollution concentrations, allowing active road users to choose an alternative route to lower their air pollution exposure. However, on-the-road concentrations displayed by the Belgian ATMO-Street model are often considerably underestimated, especially for routes with high levels of motorised traffic. In general, for air pollution models to distinguish between routes, a minimum spatial-model resolution of 10 m2 including street configuration effects (e.g. street canyons) is desired. For temporal resolution, static seasonal-hourly raster model data, calculated from a previous year’s hourly data, are sufficient to make a scientifically sound distinction between alternative routes regarding exposure to air pollution. Those tools are a great help in uncovering the spatio-temporal pollution patterns (active) commuters are exposed to and also provide relevant insights to reduce the health and economic burden of air pollution, which is unseen to a large extent and of which most people are not aware. Additional research using microscale measurement setups to further unravel gradients in air pollutant concentrations and further reveal reliable estimates of on-the-road concentrations of those pollutants is recommended.