Air quality forecasting in madrid using long short-term memory networks

Abstract

European and Spanish legislation set hourly limits for Nitrogen Dioxide, NO2, that are enforced with traffic restrictions. In this context it is important to warn the citizens in advance, which can only be done if the NO2 levels are forecasted. In this paper we propose a deep learning based air quality forecasting system that uses air quality and meteorological data to produce NO2 forecasts up to 24 h with a root mean squared error, RMSE, of 10.54 µg/m3. We also compare our results with the model based system CALIOPE.