A mathematical model developed at the School of Computing is forecasting air quality across a number of European cities

Founded in 1992, the GMSMA has built a complex air quality simulation system that is at the leading edge in meteorology, environmental physics and chemistry. The system is now in use and is forecasting air quality in the cities where the model has been deployed. After forecasting (it usually takes the system a day to make a 72-hour forecast), OPANA transmits this information through the latest communication systems (GPRS, WAP…) to street-level information panels or to the Internet.

The system outputs an air quality indicator based on five urban pollutants: sulphur dioxide (SO2), nitrogen dioxide (NO2), particulate matter (PM10), ozone (O3) and carbon monoxide (CO). Air quality in the area under observation is defined by the worst of the partial indicators of each pollutant, which is known as the global air quality indicator. The indicator values range from 0 to >150, and the higher the indicator is the worse the air quality is. The indicator value 0 is equivalent to a zero concentration of pollutant, whereas the value 100 represents the pre-established limit as of which the population should be warned of the potential risks.

A region’s air quality is influenced by the geographical distribution of emission sources, the quantity of emitted pollutants and the physical and chemical processes taking place in the atmosphere. The climatology and terrain influence the dispersion and transportation processes.

The forecasting system developed by the GMSMA takes into account all these variables. The system comprises an emissions model, a meteorological model, a transportation model, a photochemical model and a deposition model.

Measuring stations

Air quality is measured directly at stations located in different parts of the cities, but this information is confined to the space around the station. After calibration with the measuring stations, the models can produce maps and information about the whole region.

The emissions model (MM5-CMAQ-EMIMO) used by the GMSMA, which is OPANA’s mainstay, covers anthropogenic emissions from traffic, industry, households and the services sector with a spatial resolution of 1 km and a time resolution of 1 hour, respectively. It also accounts for biogenic emissions (primarily isoprenes and monoterpenes) from trees and vegetation.

The goal of the forecasting system is to provide users and environmental authorities with 24- to 72-hour air quality forecasts that can be drawn on to then take steps, in line with specially designed models, to reduce emissions and comply with the limits set out in European Air Quality Directives.

This is a complex process, as, in the case of ozone for example, a reduction in NOx emissions could lead to a significant increase in ozone levels in some parts of the city and its surroundings on the next day.

Urban applications

Originally applied in the cities of Madrid, Leicester and Bilbao, it has now been deployed in other cities, like Las Palmas de Gran Canaria in the Canary Islands, as well as in Asturias and Andalusia.

A data collection algorithm gathers information for the forecasts from ground emission stations (first 24 hours). This algorithm automates the processing of the observed information for use in the forecasts and has led to a statistically significant improvement in the results.

OPANA is a real-time air quality forecasting tool. OPANA offers mesoscale domains, is easy to configure and is flexible enough to accept additional information to improve the forecasting system. However, the tool can only be operated by experts, and, in almost all applications, the service is provided over the Internet. The GMSMA is responsible for routine system operation.

Environmental impact studies and industrial forecasts

Apart from air quality forecasting, the model also has the potential to conduct environmental impact studies. OPANA has been used to run environmental impact studies on the Txingudi and San Sebastián incinerators, as well as power stations for Unión Fenosa, Endesa, Cepsa, EHL, Electrabel and others. The system is also capable of forecasting the impact of industrial plants, like the ACECA power station and Portland Valderrivas cement works, on air quality.