Comparing Neural Networks and Regression Models for Ozone Forecasting

TLDR

Ground‑level ozone peaks in summer across many large cities, leading agencies to forecast daily concentrations using weather‑pollution relationships that are often nonlinear and may be better captured by neural networks. The study evaluates whether neural networks can outperform traditional regression models in daily ozone forecasting. The authors compare multiple regression and neural‑network models across several cities with diverse climate and ozone regimes to assess performance. Neural‑network models perform modestly better than regressions, with all models showing sensitivity to weather‑ozone regimes and persistence effects.

Abstract

Abstract Many large metropolitan areas experience elevated concentrations of ground-level ozone pollution during the summertime “smog season”. Local environmental or health agencies often need to make daily air pollution forecasts for public advisories and for input into decisions regarding abatement measures and air quality management. Such forecasts are usually based on statistical relationships between weather conditions and ambient air pollution concentrations. Multivariate linear regression models have been widely used for this purpose, and well-specified regressions can provide reasonable results. However, pollution-weather relationships are typically complex and nonlinear—especially for ozone—properties that might be better captured by neural networks. This study investigates the potential for using neural networks to forecast ozone pollution, as compared to traditional regression models. Multiple regression models and neural networks are examined for a range of cities under different climate and ozone regimes, enabling a comparative study of the two approaches. Model comparison statistics indicate that neural network techniques are somewhat (but not dramatically) better than regression models for daily ozone prediction, and that all types of models are sensitive to different weather-ozone regimes and the role of persistence in aiding predictions.