본문 바로가기

Quantifying Multi-pollutant Co-exposure via Deep Learning-Based Simultaneous Prediction Using Geostationary Satellite Data

문서 유형
연구출판물 · 학술지 논문
등재
Nature IndexSCIEScopus
저널 순위
JCR Q1Scopus Q1
저널
ENVIRONMENTAL SCIENCE & TECHNOLOGY60(12) · pp.9319 · 2026
피인용
Web of Science 2Scopus 1
FWCI (Scopus)
0.84
JCR IF
JCR 2024 11.3
ISSN
0013-936X

저자

  1. Kang, Eunjin제1저자Ulsan Natl Inst Sci & Technol, Dept Civil Urban Earth & Environm Engn, Ulsan 44919, South Korea
  2. Jung, SihunUlsan Natl Inst Sci & Technol, Dept Civil Urban Earth & Environm Engn, Ulsan 44919, South Korea
  3. Im, Jungho교신저자Ulsan Natl Inst Sci & Technol, Dept Civil Urban Earth & Environm Engn, Ulsan 44919, South Korea
  4. Choi, HyunyoungUlsan Natl Inst Sci & Technol, Dept Civil Urban Earth & Environm Engn, Ulsan 44919, South Korea
  5. Hwang, SoominUlsan Natl Inst Sci & Technol, Dept Civil Urban Earth & Environm Engn, Ulsan 44919, South Korea
  6. 유철희Pusan Natl Univ, Major Smart City, Busan 46241, South Korea
  7. Marshall, Julian D.Univ Washington, Dept Civil & Environm Engn, Seattle, WA 98195 USA
  8. Kim, DeoksuKorea Inst Ocean Sci & Technol, Marine Nat Disaster Res Dept, Busan 49111, South Korea
  9. Kim, JhoonYonsei Univ, Dept Atmospher Sci, Seoul 03722, South Korea
  10. Kim, Sang-MinNatl Inst Environm Res, Environm Satellite Ctr, Climate & Air Qual Res Dept, Incheon 22689, South Korea

초록

Simultaneous prediction of multiple air pollutants is essential for quantifying human co-exposure and evaluating the health impacts of pollutant mixtures. However, spatial and temporal gaps in geostationary satellite observations, chemical transport models, and ground-based monitoring networks hinder accurate hourly assessments of multi-pollutant dynamics. Here, we present Deep Learning for Multiple Air Pollutant analysis (DeepMAP), a deep learning framework that simultaneously predicts six major air pollutants-PM10, PM2.5, O3, NO2, CO, and SO2-at hourly resolution. DeepMAP demonstrated robust performance across multiple pollutants and generalized well to unseen regions. The framework accurately captured dynamic high-concentration co-pollution episodes during March 2021, with normalized RMSE values below 0.36 for all pollutants. DeepMAP revealed that PM10-PM2.5 co-exceedance was the most frequent across East Asia (91 days/year), followed by PM10-PM2.5-NO2 (42), PM2.5-O3 (18), and PM10-PM2.5-O3 (12). Hotspots for PM10-PM2.5-NO2-O3 co-exceedance were identified over the North China Plain, East China, and South Korea, where the regional annual totals reached 24, 19, and 15 days, respectively. A novel co-exposure index further identified three distinct hotspot regions where the contribution of NO2 was approximately twice that observed elsewhere. Our findings provide a high-resolution, data-driven framework for characterizing multi-pollutant co-exposure and identifying regional priorities for air quality management and public health protection.

키워드

multi-task learningdeeplearningco-exposureair pollutiongeostationarysatelliteAEROSOL OPTICAL DEPTHAIR-POLLUTIONHEALTH IMPACTSCHINAQUALITYNO2

식별자

Scopus EID
2-s2.0-105034759998
WoS UID
001720002700001