Current Focus

The work of the London pilot is currently focused on running test cases, many of which centre on the pollution episodes of Summer 2003, during which levels of PM10, NO2 and O3 throughout London and the South-East of England exceeded European limits. Research suggests that dust from Africa, the still air that predominated during much of this period, and the fact that 2003 was quite a warm Summer all played an important role in the occurrence of these episodes. As a general rule pollution is less likely to accumulate in windy conditions than when air is still, as pollutants are subject to significantly less disruption in the latter case.

Testing the Toolkit

The group is also testing the GENESIS toolkit, providing feedback on its performance, and making suggestions on how it could be improved. Version 1 of the toolkit is now available, and the group is now testing the generic tools created by the IT partners. The key criteria by these tools are being assessed is first of all useability. The tools are intended not just for technical specialists, but researchers across a number of thematic areas, hence their accessibility is a key issue. As version 1 is the first release of the software it is also being rigorously tested for any bugs or other problems. The software will then be tried out on the group's data sets, after which it can be used in other areas.

Born in Bradford

While a large amount of data has been generated by the London pilot, the group is still keen to gather more information from throughout Britain, and to pursue further research into the links between environment and health. Imperial College, a key partner within GENESIS, are closely involved in the Born in Bradford project, a groundbreaking study that will follow the lives of over 10,000 babies born in the West Yorkshire city over the next 20 years. While Born in Bradford is very much a long-term project, some of the data gathered from the study could potentially be used in GENESIS.

Pilot project overview

The London air quality pilot is developing and evaluating a health application tool based on GMES geo-spatial air quality mapping services, developed for the local scale with input data at high time and high spatial resolution. The tool is being developed to take into account time activity patterns, including the time spent in different micro-environments, at home, travelling, at work and the ambient environment, for example, either on the basis of individual level time activity data that might be obtained from purpose-designed surveys, or by simulating time activity stochastically from group-level, statistical data.

The pilot focuses on modelling impacts on air pollution exposures, and associated modification in health risks to air pollution, of the London Low Emissions Zone (LEZ). It builds on existing technologies, the airTEXT air quality forecasting and alerting system for London, www.airtext.info and the STEMS (Space-Time Exposure Modelling System) which is being developed from a research tool into a practical tool for use in the GENESIS system.

The London Low Emissions Zone (LEZ)

The LEZ aims to deter the most polluting vehicles from operating in London by applying a charge to those vehicles that do not meet the LEZ emissions standards. Introduced in February 2008, the LEZ applies to vehicles that are over 3.5 tonnes and buses and coaches over five tonnes with more than eight seats. The charge is implemented using automatic number plate recognition technology via fixed and mobile cameras over the whole of Greater London (within the M25). The LEZ requires these vehicles to meet certain emissions standards; from 2010 a standard EURO III for particulate to a standard EURO IV in 2012. The use case assesses and compares the exposure and health impact with and without the LEZ in London.

State of the art

Estimates of exposure and health impacts are usually made on the basis of single value representative annual average concentrations and total population. In this first phase the pilot has gone beyond that and uses a spatially varying concentration field and a population distribution to calculate an improved estimate of exposure and hence health impacts of the LEZ.

The final system will use improved temporal information on emissions to improve the estimate of annual average concentration and will have the capability of using individual or probabilistic time activity for residents.

The first phase: exposure assessment in relation to the London LEZ

In the first phase, the calculation of exposure was carried out in ArcGIS Geographical Information System. The residential density data was an ArcGIS data set (based on Great Britain Ordnance Survey data) of residential postcode locations covering the study area. The London area has a total of 148,203 postcode point locations for 2003, the year modeled. Each point has a record of the residential head count to be used as a basis for subsequently weighting the predicted exposures. There is a unique record for each point location (postcode) along with the number of households, and the total population, across all households associated with each location.

Air pollution exposures were determined at each postcode location for fine particles (PM₁₀) and nitrogen dioxide (NO₂) as they have been cited as the main pollutants of concern in relation to long-term health impacts of air pollution.

 

Predicted absolute difference in NO₂ exposures at postcodes following implementation of the LEZ

Modelled effects of the policies on traffic variables were substantial, resulting in PM₁₀ and NO₂ emission decreases of about 30% (at street level) and 10% (city average level) respectively. At the city area level modelled changes in concentration of NO₂ and PM₁₀ related to the policy interventions were modest (< 1 µg/m³). Predicted changes were larger for NO₂ than for PM₁₀. The residential location data were intersected with the modelled air pollution data to derive exposure profiles.

The distribution of the change in population exposures to NO₂ following the LEZ.

Health impacts of the LEZ

Health impacts of the LEZ were calculated at the city level, using relative risk estimates suggested by the World Health Organisation for mortality and hospital admissions in relation to long-term air pollution that will be applied in STEMS.

The modelled impact of the LEZ in London on mortality is shown as a preliminary example application in the table. Further work will be undertaken in phase 2 to develop tools to look at smaller area variations in these health impacts and apply a range of risk assessments for mortality and hospital admissions.

Phase 2

In Phase 2 an epidemiological study will analyse associations between air pollution and cardio-respiratory health (mortality and/or hospital admissions), using the modelled exposures and health data from the SAHSU (Small Area Health Statistics Unit) database.

Available expertise on personal and micro-environmental exposure data for a sample of individuals will be used and then extended to the vulnerable groups of interest. The pilot will also benefit from using data gathered in an ongoing exposure assessment in Bradford, a town in the north of England.

Interview with application pilot responsible Christine McHugh

While air quality in London has improved over the last 50 years, pollution episodes still have a significant impact on public health. Christine McHugh, the leader of the GENESIS Air Quality Pilot on London city, explains how her work in monitoring and modelling pollutants will improve air quality forecasts

Patrick Truss: What are the main areas you are focusing on in the London Pilot?

Christine McHugh: We calculate air quality concentrations, from which we calculate the exposure of the population or a particular sub-group. From that we can look at health impacts: for example the change in mortality or morbidity, ie death and illness rates.
We chose to focus on London because we currently run the airTEXT forecasting system,  twice a day it calculates an air quality forecast system for London. Therefore there is a system currently in operation, and there are currently 6,000 users of that system receiving alerts for high or moderate air quality.
So that existed already but the principles are extremely transferable and we may do an experiment in a city outside London during the course of the GENESIS project.

PT: So is your work partly about modifying the airTEXT system?

CM: Yes, we are modifying it to pick up the latest information to have greater,¦ higher time resolution at input. So for instance we aim to pick up the latest traffic data,  in an ideal world we would be able to pick that up in near real-time, that might not be possible, it might be possible to refresh the fleet data on an annual basis. We calculate twice a day, and then users can select whether they receive the forecast in the morning or the evening.

PT: Does this take differences between micro-environments into account? Is air quality in a confined space very different to air quality in an open area for example?

CM: Yes. However, there is not a lot of information on exactly how they differ. Under the London Air Quality Pilot there will be experiments to investigate that a little bit further, so people will go around with monitors on their backs,  portable monitors, and travel on different modes of transport and so forth to try and gain more information about the differences between air quality in a confined space and air quality in open areas.

PT: Are you looking at the overall composition of the air or analysing each pollutant separately?

CM: On the whole the impact of each of the regulated pollutants,  nitrogen dioxide (NO2), ozone (O3) and particulate matter (PM10),  is usually looked at separately. Although those pollutants interact in the process of being generated, they are looked at separately in the air quality guidelines.
The air quality guidelines we work to look at each separately. So if one is very high we would issue an alert to say there is bad pollution, while if another is very high we would issue an alert to say there is bad pollution. There probably is an interaction, but the national and EU air quality guidelines don't look at that interaction.
From an air quality regulation point of view it is the particulate levels in London that are causing concern. The EU requires cities and countries to comply with air quality guidelines, but across certain areas of Europe that has not been achieved,  the issue of exceeding PM limits is quite widespread across the continent. In London this has happened mainly by busy roads such as Marylebone Road and Euston Road.

PT: Is weight of traffic the issue?

CM: Traffic is definitely an issue, while background level is also an issue. Things that you can not control locally actually contribute to exceeding the air quality guidelines. The low-emissions zone (LEZ) in London was established to tackle this issue and investigations into improving air quality with respect to national and EU guideline levels are ongoing.

PT: Are you looking at the immediate impact of air quality on health, or more over the longer term?

CM: It could be either. For example in 2003, when there were pollution episodes, that gave rise to some additional admissions to hospital. So we are looking at those kinds of pollution episodes, but we are also looking at long-term effects, so changes over a year.
The overall goal of GENESIS is to make air quality data available to users in general, whoever is interested and whoever will find it interesting. If the users want to use it in a comparative way then they will be able to.

PT: Who are the main users? Is GENESIS intended primarily for healthcare professionals?

CM: Yes, it is intended primarily for healthcare professionals. For instance it could be planners for the emergency service. If you ran it in a mode where you get a forecast of air quality and were able to say, this is going to result in an increase in hospital admissions, then you would know in advance to prepare staff, or to supply certain medicines to GP surgeries.
More generally it might be healthcare professionals, and anybody responsible for public well-being, so local authorities. One of the intermediate outputs will be exposure, and that could be used by members of the public, pressure groups, groups of activists who are interested in air quality, or again by local authorities.

PT: And I understand you use satellite data to generate your forecasts as well?

CM: Yes. Satellite data is used in one of the inputs to our forecasts, so our forecasts pick up regional forecasts. If you have a look at our website, www.airtext.info, then you will see that our forecasts are actually produced to more or less street scale. So you can see the differences between pollution levels near to the road, in open spaces, central London versus outer London and that type of thing.
There is a combination of modelling and monitoring in our project. In terms of the modelling we use one technology,  called ADMS-Urban, to model all the pollutants together. However, there are different monitors for each pollutant.

PT: Do you integrate this data with long-term weather forecasts?

CM: The daily forecasts are short-term. We take short-term weather forecasts, and short-term predictions of background regional concentration, and then prepare a forecast for the next 36 hours. We can run these for longer terms to assess, for instance, annual averages and long term health impacts, but they would tend to use historical meteorological data.

PT: Have you established any links with policy makers who might want to use your data to inform environmental policy?

CM: The GLA (Greater London Authority) have supported airTEXT with some grants in the past. They are fully aware of the forecasting system; one of the uses that we are trying out ourselves, and that might interest local authorities, is to investigate the impact of the London LEZ zone on health, mortality and morbidity.
We are looking at the air quality and health impact of the LEZ, and what the situation would be if the LEZ had not been put in place. So GENESIS can help policy-makers.