[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126162-en":3,"doc-seo-126162-105":31,"detail-sidebar-cat-0-en-105":97},{"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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},126162,3985741905716,"Rowan","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Estimating hourly ground-level aerosols using Geostationary Environment Monitoring Spectrometer aerosol optical depth: a machine learning approach","Geostationary Environment Monitoring Spectrometer (GEMS), the first UV–visible geostationary instrument for air-quality monitoring, provides hourly daytime aerosol optical depth (AOD) over Asia since 2020. This study evaluates the first two years of GEMS AOD for estimating hourly ground-level PM10 and PM2.5 concentrations in South Korea using random forest and XGBoost models with meteorological predictors. Results reproduce observed spatial and temporal patterns but show biases at concentration extremes. Adding locally available CO and NO2 measurements improves correlations and reduces errors, and neighboring-station inputs enable estimation at ungauged sites for locations lacking ground PM monitoring.","Atmos. Meas. Tech., 18, 1471–1484, 2025 [https://doi.org/10.5194/amt-18-1471-2025](https://doi.org/10.5194/amt-18-1471-2025)[ ](https://doi.org/10.5194/amt-18-1471-2025)© Author(s) 2025 . This work is distributed under the Creative Commons Attribution 4 .0 License.  \nEstimating hourly ground-level aerosols using Geostationary Environment Monitoring Spectrometer aerosol optical depth:  \na machine learning approach  \nSungmin O 1 , Ji Won Yoon2,3 , and Seon Ki Park2,3,4  \n1Department of Electronic & AI System Engineering, Kangwon National University, Samcheok, Republic of Korea  \n2 Center for Climate/Environment Change Prediction Research, Ewha Womans University, Seoul, Republic of Korea  \n3 Severe Storm Research Center, Ewha Womans University, Seoul, Republic of Korea  \n4Department of Climate and Energy Systems Engineering, Ewha Womans University, Seoul, Republic of Korea Correspondence: Seon Ki Park ([spark@ewha.ac.kr](spark@ewha.ac.kr))  \nReceived: 16 August 2024 – Discussion started: 26 August 2024  \nRevised: 12 January 2025 – Accepted: 21 January 2025 – Published: 28 March 2025  \nAbstract. The Geostationary Environment Monitoring Spectrometer (GEMS) is the world's first ultraviolet–visible instrument for air quality monitoring in geostationary orbit. Since its launch in 2020, GEMS has provided hourly daytime air quality information over Asia. However, to date, validation and applications of these data are largely lacking. Here we evaluate the effectiveness of the first 2 years of GEMS aerosol optical depth (AOD) data in estimating ground-level particulate matter (PM) concentrations at an hourly scale. Todo so, we train random forest and XGBoost machine learning algorithms using GEMS AOD data and meteorological variables as input features, then employ the trained models to estimate PM 10 and PM2:5 concentrations in South Korea. The model-estimated PM concentrations capture the spatial and temporal variations observed in ground-based measurements well, showing strong correlations. However, they exhibit noticeable biases at the extremes, with a tendency to overestimate concentrations at lower PM levels and underestimate them at higher PM levels. Incorporating locally available data, such as carbon monoxide and nitrogen dioxide measurements, into the model training further enhances performance, improving correlations and reducing errors. Moreover, we demonstrate the feasibility of using machine learning models with neighbouring station data to estimate PM concentrations at ungauged locations where ground PM measurements are not available. Our results will serve as a reference to aid the evaluation of future GEMS AOD retrieval al-  \ngorithm improvements and also provide initial guidance for data users.  \n1 Introduction  \nThe adverse impacts of particulate matter (PM) on human health are well known. Exposure to high PM concentrations can cause serious health risks such as cancers, respiratory diseases, and cardiovascular diseases (Chen and Hoek, 2020 ; Kim and Kim, 2020 ; Ciabattini et al., 2021 ; Moreno-Ríoset al., 2022) . PM can also have a harmful effect on ecosystems through deposition of PM and its subsequent uptake by plants (Rai, 2016 ; Roy et al., 2024) . Accordingly, in many countries, it is mandatory to control ambient PM concentrations, and regular PM concentration measurements are key to designing appropriate policies to constrain the presence of PM. Given this background, the number of air quality monitoring stations has been growing worldwide; however, these ground-based measurement stations are often concentrated in city areas only and sparsely distributed to provide spatially continuous data (Martin et al., 2019) .  \nIn contrast, satellite observational data, with their broad spatial coverage, can potentially be used to improve air quality monitoring (including PM) on a regional to global scale. In this context, the Geostationary Environment Monitoring Spectrometer (GEMS) on board the Geostationary Korea Multi-Purpos","cbCaijiLP4pK8g2t","https://ap.wps.com/l/cbCaijiLP4pK8g2t","pdf",2446876,10,1,14,"English","en",105,"# Abstract\n# 1 Introduction\n## Health impacts of particulate matter\n## Role of satellite observations\n## Focus on GEMS aerosol optical depth","[{\"question\":\"What data and methods are used to estimate hourly ground-level PM concentrations?\",\"answer\":\"The study trains random forest and XGBoost models using GEMS aerosol optical depth (AOD) and meteorological variables as input features, then estimates PM10 and PM2.5 concentrations hourly over South Korea.\"},{\"question\":\"How well do the model estimates match ground-based measurements?\",\"answer\":\"The estimates capture spatial and temporal variations well and show strong correlations with ground-based observations, indicating effective performance for many conditions.\"},{\"question\":\"What problems remain in the estimates, and how are they improved?\",\"answer\":\"Noticeable biases occur at concentration extremes, with overestimation at lower PM levels and underestimation at higher PM levels. Incorporating locally available CO and NO2 measurements into training improves performance and reduces errors.\"},{\"question\":\"Can the approach estimate PM where no ground monitoring station exists?\",\"answer\":\"Yes. The study demonstrates feasibility of using machine learning with neighboring station data to estimate PM concentrations at ungauged locations lacking direct ground PM measurements.\"}]","Estimating hourly ground-level aerosols using Geostationary Environment Monitoring Spectrometer aerosol optical depth: a machine learning approach | PDF",1785903476,35,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":92,"head_meta":94,"extra_data":96,"updated_unix":29},"estimating-hourly-ground-level-aerosols-using-geostationary-environment-monitoring-spectrometer-aerosol-optical-depth-a-machine-learning-approach","",{"@graph":37,"@context":91},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/estimating-hourly-ground-level-aerosols-using-geostationary-environment-monitoring-spectrometer-aerosol-optical-depth-a-machine-learning-approach/126162/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-25","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83,87],{"name":74,"@type":75,"acceptedAnswer":76},"What data and methods are used to estimate hourly ground-level PM concentrations?","Question",{"text":77,"@type":78},"The study trains random forest and XGBoost models using GEMS aerosol optical depth (AOD) and meteorological variables as input features, then estimates PM10 and PM2.5 concentrations hourly over South Korea.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"How well do the model estimates match ground-based measurements?",{"text":82,"@type":78},"The estimates capture spatial and temporal variations well and show strong correlations with ground-based observations, indicating effective performance for many conditions.",{"name":84,"@type":75,"acceptedAnswer":85},"What problems remain in the estimates, and how are they improved?",{"text":86,"@type":78},"Noticeable biases occur at concentration extremes, with overestimation at lower PM levels and underestimation at higher PM levels. Incorporating locally available CO and NO2 measurements into training improves performance and reduces errors.",{"name":88,"@type":75,"acceptedAnswer":89},"Can the approach estimate PM where no ground monitoring station exists?",{"text":90,"@type":78},"Yes. The study demonstrates feasibility of using machine learning with neighboring station data to estimate PM concentrations at ungauged locations lacking direct ground PM measurements.","https://schema.org",{"og:url":53,"og:type":93,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":95,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":98},[99,103,107,111,116,121,126,129,134,137,140],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},"Exam",70,"exam",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},5,"Comic",60,"comic",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},6,"Technology",50,"technology",{"id":122,"doc_module":4,"doc_module_name":47,"category_name":123,"show_sort_weight":124,"slug":125},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":127,"slug":128},30,"research-report",{"id":130,"doc_module":4,"doc_module_name":47,"category_name":131,"show_sort_weight":132,"slug":133},9,"Religion & Spirituality",20,"religion-spirituality",{"id":132,"doc_module":4,"doc_module_name":47,"category_name":135,"show_sort_weight":132,"slug":136},"World Cup","world-cup",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":20,"slug":139},"Lifestyle","lifestyle",{"id":141,"doc_module":4,"doc_module_name":47,"category_name":142,"show_sort_weight":112,"slug":143},19,"General","general"]