IntelliO3-ts v1.0: a neural network approach to predict near-surface ozone concentrations in Germany

Abstract

Abstract. The prediction of near-surface ozone concentrations is important for supporting regulatory procedures for the protection of humans from high exposure to air pollution. In this study, we introduce a data-driven forecasting model named “IntelliO3-ts”, which consists of multiple convolutional neural network (CNN) layers, grouped together as inception blocks. The model is trained with measured multi-year ozone and nitrogen oxide concentrations of more than 300 German measurement stations in rural environments and six meteorological variables from the meteorological COSMO reanalysis. This is by far the most extensive dataset used for time series predictions based on neural networks so far. IntelliO3-ts allows the prediction of daily maximum 8 h average (dma8eu) ozone concentrations for a lead time of up to 4 d, and we show that the model outperforms standard reference models like persistence models. Moreover, we demonstrate that IntelliO3-ts outperforms climatological reference models for the first 2 d, while it does not add any genuine value for longer lead times. We attribute this to the limited deterministic information that is contained in the single-station time series training data. We applied a bootstrapping technique to analyse the influence of different input variables and found that the previous-day ozone concentrations are of major importance, followed by 2 m temperature. As we did not use any geographic information to train IntelliO3-ts in its current version and included no relation between stations, the influence of the horizontal wind components on the model performance is minimal. We expect that the inclusion of advection–diffusion terms in the model could improve results in future versions of our model.

Global impact of COVID-19 restrictions on the surface concentrations of nitrogen dioxide and ozone

“…Meteorological and atmospheric chemistry information at each of the air quality observation sites is obtained from the NASA Goddard Earth Observing System Composition Forecast (GEOS-CF) model (Keller et al, 2020). GEOS-CF integrates the GEOS-Chem atmospheric chemistry model (v12-01) into the GEOS Earth System Model (Long et al, 2015;Hu et al, 2018) and provides global hourly analyses of atmospheric composition at 25 × 25 km 2 spatial resolution, available in near-real time at https://gmao.gsfc.nasa.gov/weather_ prediction/GEOS-CF/data_access/, last access: 5 July 2020 (Knowland et al, 2020). Anthropogenic emissions are prescribed using monthly Hemispheric Transport of Air Pollution (HTAP) bottom-up emissions (Janssens-Maenhout et al, 2015), with imposed weekly and diurnal scale factors as described in Keller et al (2020).…”

MLAir (v1.0) – a tool to enable fast and flexible machine learning on air data time series

“…A very simple Python script (e.g. written in a Jupyter Notebook (Kluyver et al, 2016) or Python file) calling the MLAir package without any modification. Selected parts of the corresponding logging of the running code are shown underneath.…”

“…A more elaborate example is described in Kleinert et al (2021), who used extensions to the standard Keras library in their workflow. So-called inception blocks (Szegedy et al, 2015) and a modification of the two-dimensional padding layers were implemented as Keras layers and could be used in the model afterwards.…”

“…Due to the scientifically oriented development of MLAir starting from a specific research question (Kleinert et al, 2021), MLAir could initially only use data from the REST API of JOIN. This binding has already been revoked in the current version, however, the DefaultDataHandler still uses this data source.…”

AQ-Bench: a benchmark dataset for machine learning on global air quality metrics

“…The principle behind recent machine learning applications in ozone research is often a similar principle to the one Schultz et al (2021) described for weather data: the input data are directly mapped to a specific data product, e.g., from meteorological and past ozone measurements to the next day's maximum ozone value. In recent studies, Sayeed et al (2020) and Kleinert et al (2021) predicted regional ozone time series with convolutional neural networks and meteorological input data. Furthermore, Silva et al (2019) trained a feed-forward neural network to output ozone dry deposition at two forest measurement sites.…”