Forecasting Surface O3 in Texas Urban Areas Using Random Forest and Generalized Additive Models

Rick Pernak,Chantelle Lonsdale,Thomas Nehrkorn,Marikate Mountain,Matthew Alvarado,Jennifer Hegarty

doi:10.4209/aaqr.2018.12.0464

Rick Pernak, Chantelle Lonsdale + Show 4 more

Open Access

https://doi.org/10.4209/aaqr.2018.12.0464

Copy DOI

Abstract

ABSTRACT We developed and evaluated three types of statistical forecasting models (quantitative, probabilistic, and classification) for predicting the maximum daily 8-hour average concentration of ozone based on meteorological and ozone monitoring data for six Texas urban areas from 2009 to 2015. The quantitative and probabilistic forecasting models were generalized additive models (GAMs), whereas the classification forecast used the random forest machine learning method. We found that for the quantitative forecasting models, five of the eight predictors (the day of week, day of the year, water vapor density, wind speed, and previous day’s ozone measurement) were significant at the α = 0.001 level for all urban areas, whereas the other three varied in significance according to the location. The quantitative forecasting for the 2016 ozone season agreed well with the associated measurements (R2 of ~0.70), but it tended to under-predict the ozone level for the days with the highest concentrations. By contrast, the probabilistic forecasting models showed little accuracy in determining the probability of concentrations exceeding policy-relevant thresholds during this season. The success rate for the random forest classification models typically exceeded 75% and would likely increase if the training data sets contained more extreme events.

Highlights

Ozone (O3) at the surface can have adverse effects on public health. Lippmann (1989) summarized many studies that suggest lungs in people age quicker, lung capacity diminishes, and air flow resistance increases with sustained exposure to ozone. Devlin et al (1997) examined lung inflammation and changes in lung function and found that these afflictions are likely due to elevated ozone and that those with pre-existing respiratory conditions are more prone to experience them
We found that for the quantitative forecasting models, five of the eight predictors were significant at the α = 0.001 level for all urban areas, whereas the other three varied in significance according to the location
We developed two generalized additive models (GAMs)-based models, a quantitative forecast that predicts the numerical value of O3,MDA8 and a probabilistic model that predicts the probability that O3,MDA8 will exceed a given threshold

Summary

Introduction

Ozone (O3) at the surface can have adverse effects on public health. Lippmann (1989) summarized many studies that suggest lungs in people age quicker, lung capacity diminishes, and air flow resistance increases with sustained exposure to ozone. Devlin et al (1997) examined lung inflammation and changes in lung function and found that these afflictions are likely due to elevated ozone and that those with pre-existing respiratory conditions are more prone to experience them. Ozone (O3) at the surface can have adverse effects on public health. Lippmann (1989) summarized many studies that suggest lungs in people age quicker, lung capacity diminishes, and air flow resistance increases with sustained exposure to ozone. Studies performed on data from the United States, Canada, and Europe have linked air pollution to chronic obstructive pulmonary disease (COPD), increased hospitalization for respiratory illnesses (e.g., asthma), and lower forced expiratory volume (FEV). Anderson et al (1997) confirmed an association with ozone and other pollutants with COPD in six European cities with different climates. Sixteen Canadian cities are the subject of Burnett et al (1997), and they found that, in a population of over 12 million people over more than ten years, even low levels of surface

Methods

Results

Conclusion