Geographically Aware Air Quality Prediction Through CNN-LSTM-KAN Hybrid Modeling with Climatic and Topographic Differentiation

Air pollution poses a pressing global challenge, particularly in rapidly industrializing nations like China where deteriorating air quality critically endangers public health and sustainable development. To address the heterogeneous patterns of air pollution across diverse geographical and climatic...

Full description

Saved in:
Bibliographic Details
Main Authors: Yue Hu, Yitong Ding, Wenjing Jiang
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Atmosphere
Subjects:
Online Access:https://www.mdpi.com/2073-4433/16/5/513
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850257667032875008
author Yue Hu
Yitong Ding
Wenjing Jiang
author_facet Yue Hu
Yitong Ding
Wenjing Jiang
author_sort Yue Hu
collection DOAJ
description Air pollution poses a pressing global challenge, particularly in rapidly industrializing nations like China where deteriorating air quality critically endangers public health and sustainable development. To address the heterogeneous patterns of air pollution across diverse geographical and climatic regions, this study proposes a novel CNN-LSTM-KAN hybrid deep learning framework for high-precision Air Quality Index (AQI) time-series prediction. Through systematic analysis of multi-city AQI datasets encompassing five representative Chinese metropolises—strategically selected to cover diverse climate zones (subtropical to temperate), geographical gradients (coastal to inland), and topographical variations (plains to mountains)—we established three principal methodological advancements. First, Shapiro–Wilk normality testing (<i>p</i> < 0.05) revealed non-Gaussian distribution characteristics in the observational data, providing statistical justification for implementing Gaussian filtering-based noise suppression. Second, our multi-regional validation framework extended beyond conventional single-city approaches, demonstrating model generalizability across distinct environmental contexts. Third, we innovatively integrated Kolmogorov–Arnold Networks (KANs) with attention mechanisms to replace traditional fully connected layers, achieving enhanced feature weighting capacity. Comparative experiments demonstrated superior performance with a 23.6–59.6% reduction in Root-Mean-Square Error (RMSE) relative to baseline LSTM models, along with consistent outperformance over CNN-LSTM hybrids. Cross-regional correlation analyses identified PM2.5/PM10 as dominant predictive factors. The developed model exhibited robust generalization capabilities across geographical divisions (R<sup>2</sup> = 0.92–0.99), establishing a reliable decision-support platform for regionally adaptive air quality early-warning systems. This methodological framework provides valuable insights for addressing spatial heterogeneity in environmental modeling applications.
format Article
id doaj-art-74fc70e362d84da68eb5cdcb5dbc4a8f
institution OA Journals
issn 2073-4433
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Atmosphere
spelling doaj-art-74fc70e362d84da68eb5cdcb5dbc4a8f2025-08-20T01:56:20ZengMDPI AGAtmosphere2073-44332025-04-0116551310.3390/atmos16050513Geographically Aware Air Quality Prediction Through CNN-LSTM-KAN Hybrid Modeling with Climatic and Topographic DifferentiationYue Hu0Yitong Ding1Wenjing Jiang2School of Sciences, Zhejiang University of Science and Technology, Hangzhou 310023, ChinaSchool of Sciences, Zhejiang University of Science and Technology, Hangzhou 310023, ChinaFaculty of Economics and Business Administration, Babeş-Bolyai University, 400591 Cluj-Napoca, RomaniaAir pollution poses a pressing global challenge, particularly in rapidly industrializing nations like China where deteriorating air quality critically endangers public health and sustainable development. To address the heterogeneous patterns of air pollution across diverse geographical and climatic regions, this study proposes a novel CNN-LSTM-KAN hybrid deep learning framework for high-precision Air Quality Index (AQI) time-series prediction. Through systematic analysis of multi-city AQI datasets encompassing five representative Chinese metropolises—strategically selected to cover diverse climate zones (subtropical to temperate), geographical gradients (coastal to inland), and topographical variations (plains to mountains)—we established three principal methodological advancements. First, Shapiro–Wilk normality testing (<i>p</i> < 0.05) revealed non-Gaussian distribution characteristics in the observational data, providing statistical justification for implementing Gaussian filtering-based noise suppression. Second, our multi-regional validation framework extended beyond conventional single-city approaches, demonstrating model generalizability across distinct environmental contexts. Third, we innovatively integrated Kolmogorov–Arnold Networks (KANs) with attention mechanisms to replace traditional fully connected layers, achieving enhanced feature weighting capacity. Comparative experiments demonstrated superior performance with a 23.6–59.6% reduction in Root-Mean-Square Error (RMSE) relative to baseline LSTM models, along with consistent outperformance over CNN-LSTM hybrids. Cross-regional correlation analyses identified PM2.5/PM10 as dominant predictive factors. The developed model exhibited robust generalization capabilities across geographical divisions (R<sup>2</sup> = 0.92–0.99), establishing a reliable decision-support platform for regionally adaptive air quality early-warning systems. This methodological framework provides valuable insights for addressing spatial heterogeneity in environmental modeling applications.https://www.mdpi.com/2073-4433/16/5/513air quality index (AQI) predictiondeep learning hybrid model learningGaussian filteringcorrelation analysis
spellingShingle Yue Hu
Yitong Ding
Wenjing Jiang
Geographically Aware Air Quality Prediction Through CNN-LSTM-KAN Hybrid Modeling with Climatic and Topographic Differentiation
Atmosphere
air quality index (AQI) prediction
deep learning hybrid model learning
Gaussian filtering
correlation analysis
title Geographically Aware Air Quality Prediction Through CNN-LSTM-KAN Hybrid Modeling with Climatic and Topographic Differentiation
title_full Geographically Aware Air Quality Prediction Through CNN-LSTM-KAN Hybrid Modeling with Climatic and Topographic Differentiation
title_fullStr Geographically Aware Air Quality Prediction Through CNN-LSTM-KAN Hybrid Modeling with Climatic and Topographic Differentiation
title_full_unstemmed Geographically Aware Air Quality Prediction Through CNN-LSTM-KAN Hybrid Modeling with Climatic and Topographic Differentiation
title_short Geographically Aware Air Quality Prediction Through CNN-LSTM-KAN Hybrid Modeling with Climatic and Topographic Differentiation
title_sort geographically aware air quality prediction through cnn lstm kan hybrid modeling with climatic and topographic differentiation
topic air quality index (AQI) prediction
deep learning hybrid model learning
Gaussian filtering
correlation analysis
url https://www.mdpi.com/2073-4433/16/5/513
work_keys_str_mv AT yuehu geographicallyawareairqualitypredictionthroughcnnlstmkanhybridmodelingwithclimaticandtopographicdifferentiation
AT yitongding geographicallyawareairqualitypredictionthroughcnnlstmkanhybridmodelingwithclimaticandtopographicdifferentiation
AT wenjingjiang geographicallyawareairqualitypredictionthroughcnnlstmkanhybridmodelingwithclimaticandtopographicdifferentiation