EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PM2.5 Forecast in Korea using the Long Short-Term Memory (LSTM) Model

Chang-Hoi Ho, Ingyu Park, Jinwon Kim, Jae-Bum Lee

Asia-Pacific Journal of Atmospheric Sciences · 2023-11 · 59 (5) · 563-576

DOI: 10.1007/s13143-022-00293-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Asia-Pacific Journal of Atmospheric Sciences, online erschienen am 19.09.2022. Verfasst von Chang-Hoi Ho und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Earth and Environmental Sciences, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul, 08826 Republic of Korea. Die Arbeit wurde bisher 40-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Research Foundation of Korea, National Institute of Environment Research, National Institute of Meteorological Sciences.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The National Institute of Environmental Research, under the Ministry of Environment of Korea, provides two-day forecasts, through AirKorea, of the concentration of particulate matter with diameters of ≤ 2.5 μm (PM 2.5 ) in terms of four grades (low, moderate, high, and very high) over 19 districts nationwide. Particulate grades are subjectively designated by human forecasters based on forecast results from the Community Multiscale Air Quality (CMAQ) and artificial intelligence (AI) models in conjunction with weather patterns. This study evaluates forecasts from the long short-term memory (LSTM) algorithm relative to those from CMAQ-solely and AirKorea using observations from 2019. The skills of the one-day PM 2.5 forecasts over the 19 districts were 39–70% for CMAQ, 72–79% for LSTM, and 73–80% for AirKorea; the AI forecasts showed comparable skills to the human forecasters at AirKorea. The one-day forecast skill levels of high and very high PM 2.5 pollution grades are 31–98%, 31–74%, and 39–81% for the CMAQ-solely, the LSTM, and the AirKorea forecasts, respectively. Despite good skills for forecasting the high and very high events, CMAQ-solely forecasts also generate substantially higher false alarm rates (up to 86%) than the LSTM and AirKorea forecasts (up to 58%). Hence, applying only the LSTM model to the CMAQ forecasts can yield reasonable forecast skill levels comparable to the operational AirKorea forecasts that elaborately combine the CMAQ model, AI models, and human forecasters. The present results suggest that applications of appropriate AI models can greatly enhance PM 2.5 forecast skills for Korea in a more objective way.

Schlagwörter

Stichwörter der Autor:innen

Air pollutionAirKoreaArtificial intelligenceCommunity multiscale air quality (CMAQ)Long short-term memory (LSTM)PM2.5

Autor:innen und Institutionen

  • School of Earth and Environmental Sciences, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul, 08826 Republic of Korea.
  • National Institute of Meteorological Sciences, Seogwipo, Republic of Korea.
  • National Institute of Environmental Research, Incheon, Republic of Korea.

Förderung

  • National Research Foundation of Korea
  • National Institute of Environment Research
  • National Institute of Meteorological Sciences

Zitieren

APA

Ho, C. H., Park, I., Kim, J., & Lee, J. B. (2023). PM2.5 Forecast in Korea using the Long Short-Term Memory (LSTM) Model. Asia-Pacific Journal of Atmospheric Sciences, 59(5), 563-576. https://doi.org/10.1007/s13143-022-00293-2

Vancouver

Ho CH, Park I, Kim J, Lee JB. PM2.5 Forecast in Korea using the Long Short-Term Memory (LSTM) Model. Asia-Pacific Journal of Atmospheric Sciences. 2023;59(5):563-576. doi:10.1007/s13143-022-00293-2

BibTeX
@article{Ho2023,
  author = {Ho, Chang-Hoi and Park, Ingyu and Kim, Jinwon and Lee, Jae-Bum},
  title = {PM2.5 Forecast in Korea using the Long Short-Term Memory (LSTM) Model},
  journal = {Asia-Pacific Journal of Atmospheric Sciences},
  year = {2023},
  volume = {59},
  number = {5},
  pages = {563--576},
  doi = {10.1007/s13143-022-00293-2}
}
Kostenlos starten →