Machine Learning Techniques for Spatio-Temporal Air Pollution Prediction to Drive Sustainable Urban Development in the Era of Energy and Data Transformation

Accurately predicting PM2.5 is a crucial task for protecting public health and making policy decisions. In the meanwhile, it is also a challenging task, given the complex spatio-temporal patterns of PM2.5 concentrations. Recently, the utilization of graph neural network (GNN) models has emerged as a promising approach, demonstrating significant advantages in capturing the spatial and temporal dependencies associated with PM2.5 concentrations. In this work, we collected a comprehensive dataset spanning 308 cities in China, encompassing data on seven pollutants as well as meteorological variables from January 2015 to September 2022. To effectively predict the PM2.5 concentrations, we propose a graph attention recurrent neural network (GARNN) model by taking into account both meteorological and geographical information. Extensive experiments validated the efficiency of the proposed GARNN model, revealing its superior performance compared to other existing methods in terms of predictive capabilities. This study contributes to advancing the understanding and prediction of PM2.5 concentrations, providing a valuable tool for addressing environmental challenges.

A Graph Attention Recurrent Neural Network Model for PM2.5 Prediction: A Case Study in China from 2015 to 2022

Pan,

Liu,

2024

Particulate Matter in the Eastern Slovakia Region: Measurement, Monitoring, and Evaluation

Kirešová,

Guzan,

Sobota

et al. 2024

The paper focuses on the measurement of PM and other meteorological parameters in a small region of central Europe—eastern Slovakia and northeastern Hungary. Due to the increasing availability of sensors measuring not only PM, but also temperature, humidity, pressure, VOC, NOx, and CO2, new possibilities arise in terms of comparing (mainly in terms of correlation) PM and the other measured parameters, thus generating a large amount of data for evaluation. The correlations found are typical for inland conditions, thus able to map other regions of the world. The presented measurements can also be used to predict the evolution of PM with alerts for people with respiratory diseases, or in virtual reality using a digital twin of a humanoid robot.

Spatiotemporal Analysis of Urban Growth and PM2.5 Concentrations in Sylhet, Bangladesh

Rahman,

Meng,

Mathews

et al. 2024

Environmental impacts of urbanization, such as increased air pollution, pose significant challenges for developing countries. This study examines land cover changes and their relationship with particulate matter 2.5 (PM2.5) concentrations in two upazilas of Bangladesh from 2001 to 2019 using GIS and remote sensing techniques. Results show significant urban expansion (i.e., increase in built-up area) in both upazilas, corresponding with increasing PM2.5 levels. Linear regression reveals correlations between land cover types and PM2.5 levels. Mixed forests and waterbodies tend to be negatively associated with PM2.5 concentrations; on the other hand, built-up and barren land show a positive correlation with PM2.5. The most significant increase in built-up land and PM2.5 was in Madhabpur, indicating urgent environmental and health issues. This study emphasizes the critical role of sustainable urban planning and environmental conservation in mitigating urbanization’s adverse effects on air quality, advocating for preserving natural landscapes to maintain ecological balance, protecting urban health, and providing policymakers with insights to develop strategies addressing urban expansion and air pollution.