Making air pollution model predictions that are both extensive and detailed
So far it has been very difficult to make air pollution model predictions that are both extensive (for whole countries) and detailed (provide estimates of air pollution street by street).
King’s solution to this problem is to couple the USEPA’s Community Scale Air Quality model and CERC’s ADMS roads model together using our highly efficient modelling methods, run on the NERC Archer Supercomputer .
We have called the combined model CMAQ-urban, which is described in detail in Beevers et al., 2012 and funded by the NERC Traffic Pollution & Health in London project. The model predicts air pollution at 19.2 trillion points for each year - that's a lot of data! and so these are averaged to give maps like the one below, which represents annual average NO2 concentrations for 2011.
So why go to all this trouble? Well we now have a single model that can provide detailed human exposure estimates for STEAM and COPE), and within the Traffic Pollution & Health in London project itself. It can also be used to predict the future, so we’re using it to predict the change in air pollution in 2035 and 2050 for a NIHR project . Under different future scenarios of CO2 reduction we will look at the co-benefits of air pollution and climate change policies.
Finally, we’re using results from CMAQ-urban in a forecast system for mobile phone applications, so that people can get a detailed look at their personal exposure. Have a look around the air pollution map and see what it’s like where you live. You'll notice higher concentrations tend to be nearer roads and city centres, where the red colours indicate levels close to or above the 40 µg/m³ annual NO2 limit value.
One way coupling of CMAQ and a road source dispersion model for fine scale air pollution predictions. Atmospheric Environment 59 (2012) 47-58
SD Beevers, N Kitwiroon, ML. Williams, DC. Carslaw, 2012. Environmental Science & Technology
Dr Sean Beevers
0207 848 4009