New methods to derive street-scale spatial patterns of air pollution from mobile monitoring

The benefits of monitoring ambient air pollution with instruments mounted to ground-based, moving platforms include increased spatial resolution of measurements and synchronous, fast-response measurements close to road sources for emissions analyses. However, these come at the cost of obtaining a suitable number of repeat visits at each location in order to achieve reliable and representative pollution estimates at the desired spatial and temporal resolution. Thus, methods that maximize the information content derived from limited repeat coverage of mobile platforms are needed in order to realize the spatial and emissions source benefits possible from mobile air pollution data collection. This work builds upon previous methods by providing generalizable approaches to quantifying sampling uncertainty that enable greater data inclusion, make sampling uncertainty an integral component of air quality findings and provide decision-makers with options to fit uncertainty analysis to their purpose. To demonstrate the uncertainty estimation methods, we analyzed mobile monitoring data collected in the Breathe London pilot project in three distinct use cases. We derived insights from two key measures of pollution: total ambient NO2 concentrations and the ratio of NOx to CO2 enhancements – a marker of the intensity of NOx pollution from emission sources. The results were useful information for London public health policymakers on street-by-street level differences in pollution, and the effects of the Ultra Low Emission Zone. The future use of these flexible uncertainty methods will allow decision-makers to best leverage the information embedded in available air pollution data