Do we need high temporal resolution modelling of exposure in urban areas? A test case.
Roadside concentrations of harmful pollutants such as NOx are highly variable in both space and time. This is rarely considered when assessing pedestrian and cyclist exposures. We aim to fully describe the spatio-temporal variability of exposures of pedestrians and cyclists travelling along a road at high resolution. We evaluate the value added of high spatio-temporal resolution compared to high spatial resolution only. We also compare high resolution vehicle emissions modelling to using a constant volume source. We highlight conditions of peak exposures, and discuss implications for health impact assessments. Using the large eddy simulation code Fluidity we simulate NOx concentrations at a resolution of 2 m and 1 s along a 350 m road segment in a complex real-world street geometry including an intersection and bus stops. We then simulate pedestrian and cyclist journeys for different routes and departure times. For the high spatio-temporal method, the standard deviation in 1 s concentration experienced by pedestrians (50.9 μg.m-3) is nearly three times greater than that predicted by the high-spatial only (17.5 μg.m-3) or constant volume source (17.6 μg.m-3) methods. This exposure is characterised by low concentrations punctuated by short duration, peak exposures which elevate the mean exposure and are not captured by the other two methods. We also find that the mean exposure of cyclists on the road (31.8 μg.m-3) is significantly greater than that of cyclists on a roadside path (25.6 μg.m-3) and that of pedestrians on a sidewalk (17.6 μg.m-3). We conclude that ignoring high resolution temporal air pollution variability experienced at the breathing time scale can lead to a mischaracterization of pedestrian and cyclist exposures, and therefore also potentially the harm caused. High resolution methods reveal that peaks, and hence mean exposures, can be meaningfully reduced by avoiding hyper-local hotspots such as bus stops and junctions.