{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T06:01:55Z","timestamp":1772690515061,"version":"3.50.1"},"reference-count":44,"publisher":"MDPI AG","issue":"21","license":[{"start":{"date-parts":[[2023,10,31]],"date-time":"2023-10-31T00:00:00Z","timestamp":1698710400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"State Key Laboratory of Geo-Information Engineering","award":["2023-04-13"],"award-info":[{"award-number":["2023-04-13"]}]},{"name":"State Key Laboratory of Geo-Information Engineering","award":["AR2204"],"award-info":[{"award-number":["AR2204"]}]},{"name":"Chinese Academy of Surveying and Mapping Basic Research Fund Program","award":["2023-04-13"],"award-info":[{"award-number":["2023-04-13"]}]},{"name":"Chinese Academy of Surveying and Mapping Basic Research Fund Program","award":["AR2204"],"award-info":[{"award-number":["AR2204"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Accurate and reliable prediction of air pollutant concentrations is important for rational avoidance of air pollution events and government policy responses. However, due to the mobility and dynamics of pollution sources, meteorological conditions, and transformation processes, pollutant concentration predictions are characterized by great uncertainty and instability, making it difficult for existing prediction models to effectively extract spatial and temporal correlations. In this paper, a powerful pollutant prediction model (STA-ResConvLSTM) is proposed to achieve accurate prediction of pollutant concentrations. The model consists of a deep learning network model based on a residual neural network (ResNet), a spatial\u2013temporal attention mechanism, and a convolutional long short-term memory neural network (ConvLSTM). The spatial\u2013temporal attention mechanism is embedded in each residual unit of the ResNet to form a new residual neural network with the spatial\u2013temporal attention mechanism (STA-ResNet). Deep extraction of spatial\u2013temporal distribution features of pollutant concentrations and meteorological data from several cities is carried out using STA-ResNet. Its output is used as an input to the ConvLSTM, which is further analyzed to extract preliminary spatial\u2013temporal distribution features extracted from the STA-ResNet. The model realizes the spatial\u2013temporal correlation of the extracted feature sequences to accurately predict pollutant concentrations in the future. In addition, experimental studies on urban agglomerations around Long Beijing show that the prediction model outperforms various popular baseline models in terms of accuracy and stability. For the single-step prediction task, the proposed pollutant concentration prediction model performs well, exhibiting a root-mean-square error (RMSE) of 9.82. Furthermore, even for the pollutant prediction task of 1 to 48 h, we performed a multi-step prediction and achieved a satisfactory performance, being able to achieve an average RMSE value of 13.49.<\/jats:p>","DOI":"10.3390\/s23218863","type":"journal-article","created":{"date-parts":[[2023,10,31]],"date-time":"2023-10-31T12:53:32Z","timestamp":1698756812000},"page":"8863","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":15,"title":["Prediction of Pollutant Concentration Based on Spatial\u2013Temporal Attention, ResNet and ConvLSTM"],"prefix":"10.3390","volume":"23","author":[{"given":"Cai","family":"Chen","sequence":"first","affiliation":[{"name":"School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing 102616, China"},{"name":"Chinese Academy of Surveying and Mapping, Beijing 100830, China"}]},{"given":"Agen","family":"Qiu","sequence":"additional","affiliation":[{"name":"Chinese Academy of Surveying and Mapping, Beijing 100830, China"}]},{"given":"Haoyu","family":"Chen","sequence":"additional","affiliation":[{"name":"Jiangsu Provincial Surveying and Mapping Engineering Institute, Nanjing 210013, China"}]},{"given":"Yajun","family":"Chen","sequence":"additional","affiliation":[{"name":"China Electronics Standardization Institute, Beijing 100007, China"}]},{"given":"Xu","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing 102616, China"}]},{"given":"Dong","family":"Li","sequence":"additional","affiliation":[{"name":"School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing 102616, China"}]}],"member":"1968","published-online":{"date-parts":[[2023,10,31]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"105622","DOI":"10.1016\/j.knosys.2020.105622","article-title":"Predicting concentration levels of air pollutants by transfer learning and recurrent neural network","volume":"192","author":"Fong","year":"2020","journal-title":"Knowl.-Based Syst."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1341","DOI":"10.1007\/s10098-019-01709-w","article-title":"Air pollution prediction by using an artificial neural network model","volume":"21","author":"Maleki","year":"2019","journal-title":"Clean Technol. Environ. Policy"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"142543","DOI":"10.1016\/j.scitotenv.2020.142543","article-title":"Cost of economic growth: Air pollution and health expenditure","volume":"755","author":"Chen","year":"2021","journal-title":"Sci. Total Environ."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"106143","DOI":"10.1016\/j.envint.2020.106143","article-title":"Ensemble-based deep learning for estimating PM 2.5 over California with multisource big data including wildfire smoke","volume":"145","author":"Li","year":"2020","journal-title":"Environ. Int."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"105972","DOI":"10.1016\/j.asoc.2019.105972","article-title":"A novel combined forecasting system for air pollutants concentration based on fuzzy theory and optimization of aggregation weight\u2014ScienceDirect","volume":"87","author":"Yang","year":"2020","journal-title":"Appl. Soft Comput."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"104600","DOI":"10.1016\/j.envsoft.2019.104600","article-title":"Constructing a PM2.5 concentration prediction model by combining auto-encoder with Bi-LSTM neural networks","volume":"124","author":"Zhang","year":"2019","journal-title":"Environ. Model. Softw."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"118017","DOI":"10.1016\/j.eswa.2022.118017","article-title":"RCL-Learning: ResNet and convolutional long short-term memory-based spatiotemporal air pollutant concentration prediction model","volume":"207","author":"Zhang","year":"2022","journal-title":"Expert Syst. Appl."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Li, D., Liu, J., and Zhao, Y. (2022). Prediction of Multi-Site PM2.5 Concentrations in Beijing Using CNN-Bi LSTM with CBAM. Atmosphere, 13.","DOI":"10.3390\/atmos13101719"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Wu, X., Zhang, C., Zhu, J., and Zhang, X. (2022). Research on PM2.5 Concentration Prediction Based on the CE-AGA-LSTM Model. Appl. Sci., 12.","DOI":"10.3390\/app12147009"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"2412","DOI":"10.1109\/TKDE.2019.2954510","article-title":"Deep Air Quality Forecasting Using Hybrid Deep Learning Framework","volume":"33","author":"Du","year":"2019","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Huang, C.J., and Kuo, P.H. (2018). A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities. Sensors, 18.","DOI":"10.3390\/s18072220"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Li, D., Liu, J., and Zhao, Y. (2022). Forecasting of PM2.5 Concentration in Beijing Using Hybrid Deep Learning Framework Based on Attention Mechanism. Appl. Sci., 12.","DOI":"10.3390\/app122111155"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"144507","DOI":"10.1016\/j.scitotenv.2020.144507","article-title":"A novel Encoder-Decoder model based on read-first LSTM for air pollutant prediction","volume":"765","author":"Zhang","year":"2021","journal-title":"Sci. Total Environ."},{"key":"ref_14","unstructured":"Byun, D.W., and Ching, J.K.S. (1999). Science Algorithms of the EPA Models-3 Community Multiscale Air Quality (CMAQ) Modeling System."},{"key":"ref_15","unstructured":"Zhu, B., Akimoto, H., and Wang, Z. (2005). AGU Fall Meeting Abstracts, American Geophysical Union."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"100269","DOI":"10.1016\/j.bdr.2021.100269","article-title":"FDN-learning: Urban PM2.5-concentration Spatial Correlation Prediction Model Based on Fusion Deep Neural Network","volume":"26","author":"Zou","year":"2021","journal-title":"Big Data Res."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"20050","DOI":"10.1109\/ACCESS.2019.2897028","article-title":"A Novel Combined Prediction Scheme Based on CNN and LSTM for Urban PM;2.5; Concentration","volume":"7","author":"Qin","year":"2019","journal-title":"IEEE Access"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Moursi, A.S.A., El-Fishawy, N., Djahel, S., and Shouman, M.A. (2022). Enhancing PM2.5 Prediction Using NARX-Based Combined CNN and LSTM Hybrid Model. Sensors, 22.","DOI":"10.3390\/s22124418"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"824","DOI":"10.1016\/j.procs.2018.05.094","article-title":"Evolving Differential evolution method with random forest for prediction of Air Pollution","volume":"132","year":"2018","journal-title":"Procedia Comput. Sci."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"776","DOI":"10.1080\/10962247.2017.1292968","article-title":"Forecasting of particulate matter time series using wavelet analysis and wavelet-ARMA\/ARIMA model in Taiyuan, China","volume":"67","author":"Zhang","year":"2017","journal-title":"J. Air Waste Manag. Assoc."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"103208","DOI":"10.1016\/j.jece.2019.103208","article-title":"Prediction of air pollution index (API) using support vector machine (SVM)","volume":"8","author":"Leong","year":"2019","journal-title":"J. Environ. Chem. Eng."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"5515103","DOI":"10.1155\/2021\/5515103","article-title":"Longer Time Span Air Pollution Prediction: The Attention and Autoencoder Hybrid Learning Model","volume":"2021","author":"Tu","year":"2021","journal-title":"Math. Probl. Eng."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"119348","DOI":"10.1016\/j.envpol.2022.119348","article-title":"Deep neural networks for spatiotemporal PM 2.5 forecasts based on atmospheric chemical transport model output and monitoring data","volume":"306","author":"Kow","year":"2022","journal-title":"Environ. Pollut."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"101168","DOI":"10.1016\/j.apr.2021.101168","article-title":"PM 2.5 concentrations forecasting in Beijing through deep learning with different inputs, model structures and forecast time","volume":"12","author":"Yang","year":"2021","journal-title":"Atmos. Pollut. Res."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"118011","DOI":"10.1016\/j.apenergy.2021.118011","article-title":"A hybrid model for multi-step coal price forecasting using decomposition technique and deep learning algorithms","volume":"306","author":"Zhang","year":"2022","journal-title":"Appl. Energy"},{"key":"ref_26","first-page":"15","article-title":"A Spatiotemporal Prediction Framework for Air Pollution Based on Deep RNN","volume":"4","author":"Fan","year":"2017","journal-title":"Remote Sens. Spat. Inf. Sci."},{"key":"ref_27","unstructured":"Chung, J., Gulcehre, C., Cho, K.H., and Bengio, Y. (2014). Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling. arXiv."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"997","DOI":"10.1016\/j.envpol.2017.08.114","article-title":"Long short-term memory neural network for air pollutant concentration predictions: Method development and evaluation","volume":"231","author":"Li","year":"2017","journal-title":"Environ. Pollut."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"125341","DOI":"10.1016\/j.jclepro.2020.125341","article-title":"Regional air quality forecasting using spatiotemporal deep learning","volume":"283","author":"Abirami","year":"2021","journal-title":"J. Clean. Prod."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"486","DOI":"10.1016\/j.chemosphere.2018.12.128","article-title":"Long short-term memory\u2014Fully connected (LSTM-FC) neural network for PM 2.5 concentration prediction","volume":"220","author":"Zhao","year":"2019","journal-title":"Chemosphere"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Li, S., Xie, G., Ren, J., Guo, L., Yang, Y., and Xu, X. (2020). Urban PM2.5 Concentration Prediction via Attention-Based CNN\u2013LSTM. Appl. Sci., 10.","DOI":"10.3390\/app10061953"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"264","DOI":"10.1016\/j.scitotenv.2014.07.051","article-title":"A hybrid model for PM2.5 forecasting based on ensemble empirical mode decomposition and a general regression neural network","volume":"496","author":"Zhou","year":"2014","journal-title":"Sci. Total Environ."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"525","DOI":"10.1080\/15435075.2021.1875474","article-title":"A Novel Short-Term Photovoltaic Power Forecasting Approach based on Deep Convolutional Neural Network","volume":"18","author":"Korkmaz","year":"2021","journal-title":"Int. J. Green Energy"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"114513","DOI":"10.1016\/j.eswa.2020.114513","article-title":"Multi-hour and multi-site air quality index forecasting in Beijing using CNN, LSTM, CNN-LSTM, and spatiotemporal clustering","volume":"169","author":"Yan","year":"2020","journal-title":"Expert Syst. Appl."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"125027","DOI":"10.1016\/j.energy.2022.125027","article-title":"Ash determination of coal flotation concentrate by analyzing froth image using a novel hybrid model based on deep learning algorithms and attention mechanism","volume":"260","author":"Yang","year":"2022","journal-title":"Energy"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"348","DOI":"10.1016\/j.neucom.2020.04.110","article-title":"Interpretable spatio-temporal attention LSTM model for flood forecasting","volume":"403","author":"Ding","year":"2020","journal-title":"Neurocomputing"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"123231","DOI":"10.1016\/j.jclepro.2020.123231","article-title":"Multi-step ahead forecasting of regional air quality using spatial-temporal deep neural networks: A case study of Huaihai Economic Zone","volume":"277","author":"Zhang","year":"2020","journal-title":"J. Clean. Prod."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"107691","DOI":"10.1016\/j.envint.2022.107691","article-title":"Multi-step forecast of PM2.5 and PM10 concentrations using convolutional neural network integrated with spatial\u2013temporal attention and residual learning","volume":"171","author":"Zhang","year":"2023","journal-title":"Environ. Int."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"187","DOI":"10.1016\/j.carbon.2022.08.031","article-title":"Single Pd atoms anchored graphitic carbon nitride for highly selective and stable photocatalysis of nitric oxide","volume":"200","author":"Hu","year":"2022","journal-title":"Carbon"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Huang, F., Li, X., Wang, C., Xu, Q., Wang, W., Luo, Y., Tao, L., Gao, Q., Guo, J., and Chen, S. (2015). PM2.5 Spatiotemporal Variations and the Relationship with Meteorological Factors during 2013-2014 in Beijing, China. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0141642"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Ma, Z., Chen, C., Meng, X., Li, W., and Zhang, C. (2021). Short-term Effects of Different PM2.5 Thresholds on Daily All-cause Mortality in Jinan, China. Preprint.","DOI":"10.1038\/s41598-022-09057-4"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"413","DOI":"10.1016\/j.envint.2014.08.016","article-title":"Spatial and temporal variations of six criteria air pollutants in 31 provincial capital cities in China during 2013\u20132014","volume":"73","author":"Wang","year":"2014","journal-title":"Environ. Int."},{"key":"ref_43","unstructured":"Bahdanau, D., Cho, K., and Bengio, Y. (2014). Neural Machine Translation by Jointly Learning to Align and Translate. arXiv."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Woo, S., Park, J., Lee, J.Y., and Kweon, I.S. (2018, January 8\u201314). CBAM: Convolutional Block Attention Module. Proceedings of the European Conference on Computer Vision, Munich, Germany.","DOI":"10.1007\/978-3-030-01234-2_1"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/21\/8863\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T21:14:53Z","timestamp":1760130893000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/21\/8863"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,10,31]]},"references-count":44,"journal-issue":{"issue":"21","published-online":{"date-parts":[[2023,11]]}},"alternative-id":["s23218863"],"URL":"https:\/\/doi.org\/10.3390\/s23218863","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,10,31]]}}}