Urban form strongly mediates the allometric scaling of airshed pollution concentrations

We present allometric-scaling relationships between non-point-source emissions of air pollutants and settlement population, using 3030 urban settlements in Great Britain (home to ca. 80% of the population of that region). Sub-linear scalings (slope < 1.0; standard error on slope similar to 0.01; r(2) > 0.6) were found for the oxides of nitrogen (NOx) and microscopic airborne particles (PM10 and PM2.5). That is, emissions of these pollutants from larger cities are lower per capita than would be expected when compared to the same population dispersed in smaller settlements. The scalings of traffic-related emissions are disaggregated into a component due to under-use of roads in small settlements and a fraction due to congestion in large settlements. We use these scalings of emissions, along with a scaling related to urban form, to explain quantitatively how and why urban airshed-average air pollutant concentrations also scale with population. Our predicted concentration scaling with population is strongly sub-linear, with a slope about half that of the emissions scaling, consistent with satellite measurements of NO2 columns over large cities across Europe. We demonstrate that the urban form of a particular settlement can result in the airshed-average air pollution of that settlement being much larger or smaller than expected. We extend our analysis to predict that the likelihood of occurrence of local air pollution hotspots will scale super-linearly with population, a testable hypothesis that awaits suitable data. Our analysis suggests that coordinated management of emissions and urban form would strongly reduce the likelihood of local pollutant hotspots occurring whilst also ameliorating the urban heat island effect under climate change.