Probabilistic forecasting for extreme NO2 pollution episodes

•A new probabilistic forecasting system is presented to predict NO2 concentrations.•While predicting the full distribution, it also outperforms other point-forecasting models.•Forecasts show good properties and peak concentrations are properly predicted.•It forecasts the probability of exceedance of thresholds, key to decision makers.•Relative forecasting importance of the variables is obtained as a by-product.

In this study, we investigate the convenience of quantile regression to predict extreme concentrations of NO2. Contrarily to the usual point-forecasting, where a single value is forecast for each horizon, probabilistic forecasting through quantile regression allows for the prediction of the full probability distribution, which in turn allows to build models specifically fit for the tails of this distribution.Using data from the city of Madrid, including NO2 concentrations as well as meteorological measures, we build models that predict extreme NO2 concentrations, outperforming point-forecasting alternatives, and we prove that the predictions are accurate, reliable and sharp. Besides, we study the relative importance of the independent variables involved, and show how the important variables for the median quantile are different than those important for the upper quantiles. Furthermore, we present a method to compute the probability of exceedance of thresholds, which is a simple and comprehensible manner to present probabilistic forecasts maximizing their usefulness.

Graphical abstractDownload high-res image (282KB)Download full-size image