[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127194-en":3,"doc-seo-127194-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},127194,549768072016,"River Wang","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Physics-based machine learning for predicting urban air pollution using decadal time series data","Air quality significantly impacts public health and well-being, and effective policy depends on accurate pollution forecasts even in low-pollution settings such as Norway. Although LSTM models fit time-series problems, they demand large datasets and heavy computation, limiting practical use for physical air-pollution dynamics. Physics-Based Machine Learning (PBML) embeds physical laws to improve accuracy and interpretability, enabling better predictions under limited data.","Environ. Res. Commun.7(2025)051009 [https:](https://doi.org/10.1088/2515-7620/add795)[//](https://doi.org/10.1088/2515-7620/add795)[doi.org](https://doi.org/10.1088/2515-7620/add795)[/](https://doi.org/10.1088/2515-7620/add795)[10.1088](https://doi.org/10.1088/2515-7620/add795)[/](https://doi.org/10.1088/2515-7620/add795)[2515-7620](https://doi.org/10.1088/2515-7620/add795)[/](https://doi.org/10.1088/2515-7620/add795)[add795](https://doi.org/10.1088/2515-7620/add795)  \nOPENACCESS  \nRECEIVED  \n28January2025  \nREVISED  \n30April2025  \nACCEPTED FOR PUBLICATION 12May2025  \nPUBLISHED 20May2025  \nOriginal content from this work maybe used under the terms ofthe Creative CommonsAttribution4.0 licence.  \nAny further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.  \nLETTER  \nPhysics-based machine learning for predicting urban air pollution using decadal time series data  \nCog Cao1, Ramit Debnath 1,2  andR Michael Alvarez 1  \n1 California Institute of Technology, Pasadena, United States ofAmerica  \n2 University ofCambridge, Cambridge, United Kingdom [E-mail:](E-mail: rd545@cam.ac.uk)[ rd545@cam.ac.uk](E-mail: rd545@cam.ac.uk)  \nKeywords: air pollution, norway, physics-based machine learning, urban sustainability, climate impact, time-series machine learning Supplementary material for this article is available online  \nAbstract  \nAir quality significantly impacts public health and well-being. Effective air quality policies depend on accurate pollution predictions, even in countries with low pollution levels like Norway. Although LSTM networks are well-suited for time series data, they require large datasets and are computationally expensive, limiting their use for physical process data like air pollution dynamics. Physics-Based Machine Learning (PBML) integrates physical laws into models, offering more accurate and interpretable predictions for air pollution. In this study, we compare PBML, LSTM, and Linear Regression Models (LRM) to identify significant predictorsofair pollution using daily traffic, weather, and air pollution data between2009–2018from three major Norwegian cities. Ourfindings demonstrate that PBML outperforms both LSTMandLRMin predicting air pollution levels. This paper contributes to the expanding literature on PBML by offering more precise air pollution predictionson hyperlocal scales, informed bylocal trafficand weather conditions, ultimately supporting better data-driven policy decisions.  \n1. Introduction  \nExposure to high levels ofair pollution has significant negative health outcomes [1]. Concerns related to this have encouraged policymakers to consider the effectiveness oftransport and air quality policies to eliminate the adverse impact on health [2, 3] . However, this requires better forecasting and estimation ofthe link between traffic, weather, and air pollution. In practice, the scarcity ofcomprehensive and high-quality air pollution data presents alimitation to both research and policy development in many countries. Atthe sametime, accurately quantifying the health impacts ofpollution to propose an optimal air quality policy remains a challenge, even fora country like Norway, where pollution levels are already relatively low.  \nArtificial intelligence (AI) excelsin precision tasks that involve interactive impactson the environment [4] . For example, Chauetal (2022) demonstrated the effectiveness oflong-and short-term memory (LSTM) and bidirectional recurrent neural network techniques in machine learning-based weather forecasting models by tracking changes in air quality [5]. LSTM hasan independent memory unit. It can remember longer-term information fora longtime and avoid the problem ofvanishing gradients, making LSTM particularly useful for time series data [6] .  \nHowever, LSTM requires more training data to learn effectively, is computationally intensive and can be slow to train on large datasets [7], making it less suitable for weather o","cbCaicRASuLqy2EJ","https://ap.wps.com/l/cbCaicRASuLqy2EJ","pdf",358431,1,9,"English","en",105,"# Introduction\n# Background","[{\"question\":\"Why are LSTM models often difficult to apply to air pollution dynamics?\",\"answer\":\"LSTM requires large training datasets and is computationally intensive, which can slow training and limit use when modeling physical process data like air-pollution dynamics.\"},{\"question\":\"What does Physics-Based Machine Learning (PBML) add compared with standard ML?\",\"answer\":\"PBML integrates physical laws into the model, improving prediction accuracy and interpretability while remaining suitable for noisy data and mathematical constraints.\"},{\"question\":\"How does the study evaluate PBML for predicting urban air pollution?\",\"answer\":\"It compares PBML with LSTM and Linear Regression Models using daily traffic, weather, and air-pollution data from 2009–2018 across three major Norwegian cities.\"}]","Physics-based machine learning for predicting urban air pollution using decadal time series data | PDF",1785937431,23,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"physics-based-machine-learning-for-predicting-urban-air-pollution-using-decadal-time-series-data","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/physics-based-machine-learning-for-predicting-urban-air-pollution-using-decadal-time-series-data/127194/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why are LSTM models often difficult to apply to air pollution dynamics?","Question",{"text":75,"@type":76},"LSTM requires large training datasets and is computationally intensive, which can slow training and limit use when modeling physical process data like air-pollution dynamics.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What does Physics-Based Machine Learning (PBML) add compared with standard ML?",{"text":80,"@type":76},"PBML integrates physical laws into the model, improving prediction accuracy and interpretability while remaining suitable for noisy data and mathematical constraints.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the study evaluate PBML for predicting urban air pollution?",{"text":84,"@type":76},"It compares PBML with LSTM and Linear Regression Models using daily traffic, weather, and air-pollution data from 2009–2018 across three major Norwegian cities.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]