What is it about?
This study focuses on improving air quality predictions in two Chilean cities, Quintero and Coyhaique, which face significant pollution challenges. Quintero is near industrial plants that release sulfur dioxide (SO₂), while Coyhaique struggles with fine dust (PM₂.₅ and PM₁₀) from wood-burning for heating, worsened by its valley location and cold winters. Using data from 2016 to 2021, we combined two computer models—ARIMA and Artificial Neural Networks—to predict levels of these pollutants. We included factors like wind speed and direction to make the predictions more accurate. Our models performed well, with over 90% accuracy, helping identify when pollution might spike. This work can support better air quality management, like issuing early warnings or creating policies to reduce pollution, ultimately protecting people’s health in these regions.
Featured Image
Photo by Chris LeBoutillier on Unsplash
Why is it important?
This study is crucial because it tackles severe air pollution in Quintero and Coyhaique, two Chilean cities with unique environmental challenges. Quintero’s industrial emissions and Coyhaique’s wood-burning pollution, worsened by its valley topography, pose significant health risks. By integrating ARIMA and Artificial Neural Network models with meteorological data like wind speed, our approach achieves over 90% accuracy in predicting SO₂, PM₂.₅, and PM₁₀ levels. This timely work is unique in its hybrid modeling tailored to Chile’s diverse pollution sources and geography. It can drive impactful change by enabling early pollution alerts, informing targeted policies, and reducing health risks, offering a scalable solution for air quality management in similar regions worldwide.
Perspectives
As a researcher deeply invested in environmental challenges, I find this study particularly meaningful because it addresses air pollution in Quintero and Coyhaique, regions where communities face real health risks from industrial and residential emissions. Leading this project, I was struck by how combining ARIMA and Artificial Neural Networks allowed us to capture complex pollution patterns with high accuracy, especially by factoring in local conditions like wind and topography. This work excites me because it’s not just about numbers—it’s about providing tools for policymakers and communities to act proactively, potentially saving lives. I hope this research inspires further efforts to blend advanced modeling with local insights to tackle environmental issues globally.
Fidel Vallejo
Universidad Nacional de Chimborazo
Read the Original
This page is a summary of: Enhancing air quality predictions in Chile: Integrating ARIMA and Artificial Neural Network models for Quintero and Coyhaique cities, PLOS One, January 2025, PLOS,
DOI: 10.1371/journal.pone.0314278.
You can read the full text:
Contributors
The following have contributed to this page
