All websites, smart phone apps and air pollution information systems managed by King’s College London have now been amended to reflect the recent changes to the Daily Air Quality Index (DAQI).

The update implemented several minor changes to the index, to ensure consistency with new methods used in the calculation of European Union Limit Values. The update also clarifies the methodology used for calculating the index and confirms that the index should only be applied to measurements made by EU reference equivalent methods.

As part of the update King’s were asked to repeat the analysis originally undertaken for COMEAP to generate new predictive triggers for the revised index. Since the availability of EU reference measurements for PM10 and PM2.5 has increased since the time of the COMEAP analysis, revision of the triggers has been undertaken on an extended dataset of EU reference equivalent PM10 and PM2.5 concentrations from the AURN, and from London and neighbouring networks from 2004 to 2011. Approximately 385,000 site days of PM10 concentrations were modelled for each possible trigger for each band, and approximately 73,000 site days were modelled in the same way for PM2.5. When tested with this extended dataset, only marginal changes in the triggers were required, as expected from the revisions to the index, confirming the robustness of this predictive approach.

It is important to note that this revised index applies retrospectively to the beginning of 2012. All archived daily bulletins since 1st January 2012 have therefore been regenerated. All web statistical tools also reflect this new index and the redundant index originally used in 2012 is no longer available.

We remain the only UK data provider to have fully implemented the DAQI, displaying only EU reference equivalent particulate measurements, and including the use of triggers to display predicted indices in near real time for particles (daily mean) and ozone (running 8 hour mean) at each monitoring site. We continue to use the index’s daily mean averaging period for PM10 and PM2.5 uniformly across all public reporting, i.e in forecasts, near real time data, and historical statistical analysis.

Item date 28/05/2013