Particulate matter and nitrogen dioxide in the air have a major impact on our health. In recent months, VITO experts updated one of the statistical models behind the Dutch air monitoring network. The RIO model calculates background concentrations of air pollution on the basis of measurements, and for many years it has also been the basis for public information on current air quality in Belgium.
Measuring air quality without monitoring stations
VITO developed the RIO model 13 years ago. RIO enables the creation of areawide maps with so-called large-scale concentrations. Bino Maiheu: “The model uses an interpolation scheme. Concentrations measured at specific locations are linked to land use, which allows RIO to take into account the local character of air pollution. Trend functions are developed from the correlation between land use and concentration levels, as a result of which information on air quality in areas without monitoring stations can be provided.”
Particulate matter and nitrogen
“The Belgian Interregional Environment Agency (IRCEL) has been using the RIO method for years to display on its website the air quality throughout Belgium every hour of the day. It specifically concerns large-scale concentrations of particulate matter and nitrogen dioxide, presented on maps with a resolution of 4x4 kilometres. In a wide-ranging comparative study of air quality models, the RIO model emerged as the most accurate for Flanders. Which is why the official projections for air quality in Flanders are checked against the RIO model.”
In addition to the Netherlands, other countries are also enthusiastic about the RIO model. It will soon be deployed in Poland, and VITO is looking at opportunities in China.
Monitoring network in the
The Dutch National Institute for Public Health and the Environment (RIVM) has been using RIO since 2009. Dennis Mooibroek: “VITO had the original model expanded and adapted to make it usable in the Netherlands. But a lot of changes have taken place since then in the National Air Quality Monitoring Network: some measurement stations have closed or moved, new ones have opened, and the concentrations of most substances have decreased in recent years. In addition, data from partner monitoring networks in Amsterdam and Rotterdam had not yet been integrated. In the meantime, improvements to the model itself have also been made. Which is why we requested an update based on the current concentrations and measurement stations. This will soon be operational.”
