Time-Series Data-Driven PM<sub>2.5</sub> Forecasting: From Theoretical Framework to Empirical Analysis

PM<sub>2.5</sub> in air pollution poses a significant threat to public health and the ecological environment. There is an urgent need to develop accurate PM<sub>2.5</sub> prediction models to support decision-making and reduce risks. This review comprehensively explores the p...

Full description

Saved in:
Bibliographic Details
Main Authors: Chunlai Wu, Ruiyang Wang, Siyu Lu, Jiawei Tian, Lirong Yin, Lei Wang, Wenfeng Zheng
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Atmosphere
Subjects:
Online Access:https://www.mdpi.com/2073-4433/16/3/292
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850204635595276288
author Chunlai Wu
Ruiyang Wang
Siyu Lu
Jiawei Tian
Lirong Yin
Lei Wang
Wenfeng Zheng
author_facet Chunlai Wu
Ruiyang Wang
Siyu Lu
Jiawei Tian
Lirong Yin
Lei Wang
Wenfeng Zheng
author_sort Chunlai Wu
collection DOAJ
description PM<sub>2.5</sub> in air pollution poses a significant threat to public health and the ecological environment. There is an urgent need to develop accurate PM<sub>2.5</sub> prediction models to support decision-making and reduce risks. This review comprehensively explores the progress of PM<sub>2.5</sub> concentration prediction, covering bibliometric trends, time series data characteristics, deep learning applications, and future development directions. This article obtained data on 2327 journal articles published from 2014 to 2024 from the WOS database. Bibliometric analysis shows that research output is growing rapidly, with China and the United States playing a leading role, and recent research is increasingly focusing on data-driven methods such as deep learning. Key data sources include ground monitoring, meteorological observations, remote sensing, and socioeconomic activity data. Deep learning models (including CNN, RNN, LSTM, and Transformer) perform well in capturing complex temporal dependencies. With its self-attention mechanism and parallel processing capabilities, Transformer is particularly outstanding in addressing the challenges of long sequence modeling. Despite these advances, challenges such as data integration, model interpretability, and computational cost remain. Emerging technologies such as meta-learning, graph neural networks, and multi-scale modeling offer promising solutions while integrating prediction models into real-world applications such as smart city systems can enhance practical impact. This review provides an informative guide for researchers and novices, providing an understanding of cutting-edge methods, practical applications, and systematic learning paths. It aims to promote the development of robust and efficient prediction models to contribute to global air pollution management and public health protection efforts.
format Article
id doaj-art-b75ef34a8ac14c7cb06b0da991deb86e
institution OA Journals
issn 2073-4433
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Atmosphere
spelling doaj-art-b75ef34a8ac14c7cb06b0da991deb86e2025-08-20T02:11:15ZengMDPI AGAtmosphere2073-44332025-02-0116329210.3390/atmos16030292Time-Series Data-Driven PM<sub>2.5</sub> Forecasting: From Theoretical Framework to Empirical AnalysisChunlai Wu0Ruiyang Wang1Siyu Lu2Jiawei Tian3Lirong Yin4Lei Wang5Wenfeng Zheng6School of Automation, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Automation, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Automation, University of Electronic Science and Technology of China, Chengdu 610054, ChinaDepartment of Computer Science and Engineering, Hanyang University, Ansan 15577, Republic of KoreaDepartment of Geography & Anthropology, Louisiana State University, Baton Rouge, LA 70803, USADepartment of Geography & Anthropology, Louisiana State University, Baton Rouge, LA 70803, USASchool of Automation, University of Electronic Science and Technology of China, Chengdu 610054, ChinaPM<sub>2.5</sub> in air pollution poses a significant threat to public health and the ecological environment. There is an urgent need to develop accurate PM<sub>2.5</sub> prediction models to support decision-making and reduce risks. This review comprehensively explores the progress of PM<sub>2.5</sub> concentration prediction, covering bibliometric trends, time series data characteristics, deep learning applications, and future development directions. This article obtained data on 2327 journal articles published from 2014 to 2024 from the WOS database. Bibliometric analysis shows that research output is growing rapidly, with China and the United States playing a leading role, and recent research is increasingly focusing on data-driven methods such as deep learning. Key data sources include ground monitoring, meteorological observations, remote sensing, and socioeconomic activity data. Deep learning models (including CNN, RNN, LSTM, and Transformer) perform well in capturing complex temporal dependencies. With its self-attention mechanism and parallel processing capabilities, Transformer is particularly outstanding in addressing the challenges of long sequence modeling. Despite these advances, challenges such as data integration, model interpretability, and computational cost remain. Emerging technologies such as meta-learning, graph neural networks, and multi-scale modeling offer promising solutions while integrating prediction models into real-world applications such as smart city systems can enhance practical impact. This review provides an informative guide for researchers and novices, providing an understanding of cutting-edge methods, practical applications, and systematic learning paths. It aims to promote the development of robust and efficient prediction models to contribute to global air pollution management and public health protection efforts.https://www.mdpi.com/2073-4433/16/3/292PM<sub>2.5</sub>deep learningbibliometric analysisLSTMtransformertime-series
spellingShingle Chunlai Wu
Ruiyang Wang
Siyu Lu
Jiawei Tian
Lirong Yin
Lei Wang
Wenfeng Zheng
Time-Series Data-Driven PM<sub>2.5</sub> Forecasting: From Theoretical Framework to Empirical Analysis
Atmosphere
PM<sub>2.5</sub>
deep learning
bibliometric analysis
LSTM
transformer
time-series
title Time-Series Data-Driven PM<sub>2.5</sub> Forecasting: From Theoretical Framework to Empirical Analysis
title_full Time-Series Data-Driven PM<sub>2.5</sub> Forecasting: From Theoretical Framework to Empirical Analysis
title_fullStr Time-Series Data-Driven PM<sub>2.5</sub> Forecasting: From Theoretical Framework to Empirical Analysis
title_full_unstemmed Time-Series Data-Driven PM<sub>2.5</sub> Forecasting: From Theoretical Framework to Empirical Analysis
title_short Time-Series Data-Driven PM<sub>2.5</sub> Forecasting: From Theoretical Framework to Empirical Analysis
title_sort time series data driven pm sub 2 5 sub forecasting from theoretical framework to empirical analysis
topic PM<sub>2.5</sub>
deep learning
bibliometric analysis
LSTM
transformer
time-series
url https://www.mdpi.com/2073-4433/16/3/292
work_keys_str_mv AT chunlaiwu timeseriesdatadrivenpmsub25subforecastingfromtheoreticalframeworktoempiricalanalysis
AT ruiyangwang timeseriesdatadrivenpmsub25subforecastingfromtheoreticalframeworktoempiricalanalysis
AT siyulu timeseriesdatadrivenpmsub25subforecastingfromtheoreticalframeworktoempiricalanalysis
AT jiaweitian timeseriesdatadrivenpmsub25subforecastingfromtheoreticalframeworktoempiricalanalysis
AT lirongyin timeseriesdatadrivenpmsub25subforecastingfromtheoreticalframeworktoempiricalanalysis
AT leiwang timeseriesdatadrivenpmsub25subforecastingfromtheoreticalframeworktoempiricalanalysis
AT wenfengzheng timeseriesdatadrivenpmsub25subforecastingfromtheoreticalframeworktoempiricalanalysis