Recent developments

The project has recently achieved the status of experimentation readiness, which is a very important step forward. The GENESIS system - the specifications of which are based on requirements from the pilots and other potential application areas - has been implemented and integrated based on a number of earlier results from predecessor projects.

The pilots are now in a position to do some important experiments, and to this end have prepared a suitable amount of test, environmental and health data, as well as algorithms with which to analyse this data. The thematic pilots are now getting acquainted with the GENESIS system, and will soon start the first round of experiments.

Thomas Holzer-Popp explains the importance of his work measuring air quality in BavariaThe first version of GENESIS will be used in this round, and the thematic pilots will provide feedback to the technology developers on its performance, which will then lead to a new set of specifications for an upgraded version 2.1 of the GENESIS system. This will then be made available later in the project for a second round of experimentation.

Second round of experimentation

A number of functionalities within the GENESIS system which were not implemented within version 1 will be implemented in version 2. Now that a concrete solution is available it will be easier for the thematic pilots to really look at how their requirements and specifications have been implemented within the system, and to make clear recommendations on what needs to be changed or adjusted. This will also have a significant impact on the GENESIS system and how it can be used. On the thematic side the researchers themselves will get better at using the system, hence the second round of experimentation will go into more depth than the first.

New and improved functionalities

By the second round of experimentation the researcher's ability to use the functionalities within the system will have significantly improved. It will also be possible to see where there are gaps in the system, and to identify necessary functionalities for which it had not previously been possible to put forward clear requirements. This means that at least some functionalities will be improved or added to the system, so that it really meets the needs of the thematic pilots within GENESIS. There is still more time before the second round of experimentation to do more analysis on possible correlations between environmental and health data, and thus to develop a refined version of this first test index.

Project Overview

Air Quality Overview

Air Quality Thematic activity focuses on developing and evaluating health applications based on GMES geo-spatial air quality mapping services. Within two specific pilots, a range of spatial scales (local to regional), target groups (public, persons suffering from cardio-vascular and pulmonary diseases, epidemiologists) and delivery modes (web GIS, assessment reports) are assessed. For both pilots, a service provider utilizes its own capacities to provide specific environmental information products. A health partner assures access to anonymous personal health data (acquired from health insurances, hospitals, ongoing health studies, patient surveys) and provides epidemiological expertise. Together, they analyse environment-health correlations and apply them in user-friendly and targeted service prototypes. The health partner is also the user in these cases being responsible for either patient consulting (a hospital) or epidemiological research.

The focus of this sub-project will be on evaluating/demonstrating IT tools. For its pilot 1 (Bavaria), this means that statistical analysis results will be preliminary in a sense that work will be conducted with limited demonstration amounts of data for demonstration purposes, to allow preparation and testing of IT tools as a basis for future analysis of large, dispersed databases. The full statistical significance of thematic results will therefore only be reached after GENESIS project by large-scale application of the IT tools developed and tested within GENESIS. In order to avoid duplication of ENHIS project activities (working on a European health and environment system mainly based on correlations of aggregated in situ data with aggregated health data) and because of the coarse horizontal resolution / large spatial gaps of in situ data the focus is put on exploitation of geospatial data being made available from GMES services. In pilot 2 (London), the very high spatial accuracy (street level) will allow to achieve full thematic significance with the available data amounts.

For the Bavarian pilot additional support is achieved through the project "health weather" developing a specific health index for pulmonary diseases" funded by the ministry for environment, health and consumer protection of the state of Bavaria. This project will provide additional health data (cohort study in Munich) and validation of the derived health index (assessing its correlation with medical data) and will be the first user of GENESIS tools, which will allow handling of large data amounts from distributed databases. This project will run from 01.06.2008 - 31.05.2011. Participants of the "health weather" project are the same partners as in the GENESIS air quality pilot 1 (LMU and DLR), so that full exchange of results between this project and GENESIS is guaranteed. During presentation of results to end users it is intended to show joint results of both projects. Epidemiology and Mapping This pilot is focused on developing integrated health indeces and their mapping for public information.

Objective

Derivation and application of an integrated health index for pulmonary diseases based on environmental, meteorological and medical data

Specific Objectives

Develop and apply an integrated health index suitable for public information on pulmonary diseases by analysing combined meteorological (day and night temperatures, humidity) and air quality (nitrogen dioxide, particulate matter, ozone) data to prepare a comprehensive analysis of conditions causing potentially negative acute health effects.

Support medical expert users in visual screening and data preparation for offline statistical analysis of environment-health correlations to derive integrated health indeces for a specific disease, population group and region.Provide maps of the integrated health index for pulmonary diseases to citizens and medical staff during a test episode over all Bavaria including the neighbouring Austrian region.

Fig. 1: The figure shows example environment datasets which will be used in the pilot together with further environmental and health datasets (left: satellite night surface temperature, right: modelled nitrogen dioxide concentration near the surface during the morning displayed in a geographical information system).

Background

This pilot will make use of the large PROMOTE-2 expertise in developing services responding to user needs and will utilize satellite and model services (particulate matter, tropospheric NO2, ozone) and products together with further air quality and meteorological models and a large archive of atmospheric data. This will be complemented with medical expertise of Munich university hospital and access to large quantities of medical data from call centers and health insurances.

Interview with sub-project leader Thomas Holzer-Popp

A number of GMES (Global Monitoring for Environment and Security) projects have been established to make satellite data on the environment usable for the European public and decision-makers. However, this work faces a number of significant scientific and technical challenges, an area of particular interest to Thomas Holzer-Popp, leader of a GENESIS sub-project looking at air quality in Bavaria, London and Nice

Patrick Truss: How is air quality measured? Is it based on the levels of certain specific substances, or is it more about its overall composition?

Thomas Holzer-Popp: The first key point to make is that when it comes to legal compliance air quality is measured by stations on the ground – so-called in situ measurements. Measurements are made of a number of regulated substances, so their concentrations are the scientific information, then for the convenience of the lay public a kind of index is calculated by comparing a number of these pollutant concentrations versus their legal thresholds. This is then converted into a value that can be shown on an easy colour scale.

PT: This is with reference to European guidelines on air quality?

T H-P: Yes, a number of directives are in force in Europe which have been translated into national legislation by Member States. So there are thresholds, and if they are exceeded then certain measures have to be taken, one of which could be informing the public for example.
If the threshold is exceeded a certain number of times then the relevant authorities have to produce further plans on how they will reduce the concentration of these pollutants.

PT: Which pollutants have a particularly bad impact on air quality?

T H-P: The key issue in Central and Western Europe is fine particulate matter in winter, in summer ozone and nitrogen oxides levels are often a concern, while sulphur dioxide and carbon monoxide are major pollutants in Eastern Europe, and they are also measured and controlled.
The impact of these pollutants on human health varies; some people are susceptible to only some of these substances, and when you want to calculate an index to judge the overall air quality you have to include as many of these pollutants as you can measure at a station.
The concentration of these pollutants also varies throughout the year. Typically particulate matter levels are higher in winter and ozone in summer, then there are also regional variations.
Meteorological conditions always have an impact on air quality; if you have stagnant weather then pollutants can accumulate. Another major issue with ozone is that in the diurnal circle it is typically destroyed by a chemical reaction with nitrogen dioxides. So if ozone is moved away from traffic sources, where the level of these nitrogen dioxides is lower, then it can prevail for longer. So the ozone smoke in recreational areas is sometimes more severe than in nearby locations with heavy traffic.

PT: What are the major factors affecting air quality?

T H-P: It's a mix of local emissions - for example from traffic, house-heating systems and industry - and also long-range transport. No region is really independent of upwind regions that may also export pollutants - or precursor substances to pollutants. This is really on the continental level - so pollutants are transported from the US into northern and central Europe, and Europe itself exports further East. In the Mediterranean there are several pollution export routes, either into Europe or from Europe.

PT: Are precursor substances pollutants in themselves, or are they just indicative of likely poor air quality in future?

T H-P: A very complex slew of chemical reactions occurs in the atmosphere. Smog is induced by sun illumination since the chemical reactions need UV light - so there is a wide range of these precursors, and some of them are also critical to health. However, most of them are short-lived, and whether they produce smog depends on the balance between these precursors.

PT: Is Bavaria itself quite an environmentally varied region?

T H-P: There is a clear meteorological contrast between North and South Bavaria, and in terms of local emissions of pollutants it's of course even more scattered, depending on the main local areas of habitation, traffic routes and industrial regions.

PT: Are you measuring air quality at several locations within Bavaria?

T H-P: We cover the whole of Bavaria with satellite / model based services. By regional level I'm referring to the typical pixel of the satellite measurement, or the typical Grid cell of a model that we employ, which is around 1 kilometre x several kilometres, or even larger. That's suitable for looking at regional issues; we can see contrasts between different regions of Bavaria and then match these with health data right down to the district or zip-code areas.
We also aim to develop indices which not only compare pollutant concentrations with legal thresholds, but which are also linked to the occurrence of health issues. This can then be used as an information resource or warning index for people from specific population groups with specific illnesses.
We are doing correlation studies and statistical analysis of the links between environmental situations and health issues. In this we combine meteorological conditions - for example rapid increases or decreases in temperature - and pollution information with data on its impact on health, such as whether it causes stress to elderly people already affected by asthmatic or pulmonary diseases. Stagnant air conditions can lead to the accumulation of pollutants. This together with rising temperatures can for example be critical for certain people who may need to take extra medicine or to avoid exposing themselves to pollutants; this is the purpose of providing them with clear information on air quality.

PT: Will accurate historical data help you understand the impact of air quality on health?

T H-P: We need good statistics to derive specific indices for a specific target group on a specific illness, so require a large number of measurement points. Therefore we need several years of data to be able to do such statistical analysis.
This is what we call the diagnosis mode, which is an important part of GENESIS applications. The application mode - when we produce a daily forecast, a daily information bulletin map with a health index - only requires the actual environment data, not historical data. But it's based on a wider and deeper analysis that has to be done before.
Statistical analysis also deals with uncertainties in the data. The less accurate the data is then the more data you need to make up for the uncertainties. We have to live with quite substantial uncertainties in the data - satellite retrieval is an indirect measurement while model output depends on a number of underlying data sets, in which there are uncertainties and inaccuracies.