SOFIA DEVELOPMENT ASSOCIATION was founded as non-profit organization with the decision 348
of the Sofia Municipal Council to work for the sustainable development of the capital.

Predict the Air Pollution by Geo-locations in Sofia

Scientists from Faculty of Economics and Business Administration at Sofia University presented study for prediction Air Pollution by Geo-location in Sofia at the beginning of the Sofia Innovation Camp. The Air Pollution Predict Model was developed with the JRC financial and expert support.
This study is organized in cooperation with the Science meets Regions event (Sofia InnoCamp) on Energy and Climate Change and Social Effects of Climate Change, organized by Sofia Municipality (Sofia Development Association) in March 2019. The premise here is that the study would foster the debate on Sofia’s air quality - one of the biggest environmental issues of the city in later years.

The ultimate objective of this analysis is to deliver forecast for the next 24 hours per station using data from citizen and official science measurements. Firstly, there is bias correction of citizen science measurements, checked against the official measurement stations. The official comply with the EU directives on air quality monitoring can be used for regulatory purposes, but are limited in number (only 5 in the whole city). Citizen science stations have very good coverage of the city, but may carry instrumental biases – due to different measurement methods, different interaction with meteorology, etc.

Secondly a prediction model for next-day forecast of PM10 is built, using additional factor from meteorological parameters (from a weather forecast) and topography satellite data. When there are qualitative prediction results, the data is mapped over geo-locations of Sofia.

The authors of the model Assoc. Prof. Boryana Pelova and Assoc. Prof. Angel Marchev shared that “There is a lack of confidence from a certain group of citizens to the data from the official sources. At the same time, the popularity of the civil air quality sensors system is based mainly on the fact that the data is locally oriented (the quarter of the neighborhood) and they give to residents of Sofia some local context and understanding. It is important to emphasize that no one challenges the immense technical superiority of official measurement stations over civil network sensors, but the popular opinion is that the five official stations do not meet the needs of citizens to predict time and place.”

Also, the Prediction model uses additional meteorological data (weather forecast) and topographic satellite data. When the qualitative results are predicted, the data are mapped in geo-locations of Sofia.

"These are well-spent money," said Charlina Vicheva, Deputy Director of the DG Joint Research Center, European Commission (JRC) after hearing the presentation of the Air Pollution Forecasting Model in Sofia through geo-locations. The data and methodology are already published on and can be used by everyone.


Go to top