Ozone, nitrogen dioxide, and PM2.5 estimation from observation-model machine learning fusion over S. Korea: Influence of observation density, chemical transport model resolution, and geostationary remotely sensed AOD

Atmospheric Environment, Journal Year: 2024, Volume and Issue: 331, P. 120603 - 120603

Published: May 23, 2024

High-resolution multi-component estimates of ground-level air pollutants are necessary for assessing their impacts to human health, agriculture, and ecosystems. We demonstrate a high-resolution fusion downscaling approach over South Korea May 2016 2021. Daily 1 km fine particulate matter (PM2.5), ozone (O3), nitrogen dioxide (NO2) concentrations calculated at ground level using random forest machine learning (ML) algorithm, with predictors including reanalysis meteorology, satellite aerosol optical depth (AOD), gridded surface fields from chemical transport models (CTM). The ML model is tested 2016, coinciding the Korea-United States Air Quality Study (KORUS-AQ) intensive field campaign, 2021, allow incorporation observations Geostationary Environment Monitoring Spectrometer (GEMS). In tests correlation coefficients (R) root mean squared errors (RMSE) relative withheld daily-averaged in 10-fold cross-validation promising: 0.93 (5.5 μg/m3), 0.90 ppbv), 0.95 (4.7 ppbv) PM2.5, O3, NO2, respectively. Relative performance assessed alternate choices predictors: (a) 80-km global Copernicus Atmosphere Service (CAMS) vs. 4-km regional Weather Research Forecasting coupled Chemistry (WRF-Chem); (b) AOD polar-orbiting Moderate Resolution Image Spectroradiometer (MODIS) Multi-Angle Implementation Atmospheric Correction (MAIAC) geostationary GEMS; (c) variations observation density. This study among very first incorporate both CTM GEMS building high resolution multiple pollution predictions Korea.

Language: Английский

---

Focus on atmospheric remote sensing and environmental change

Environmental Research Letters, Journal Year: 2025, Volume and Issue: 20(3), P. 030202 - 030202

Published: March 1, 2025

Language: Английский

---

Improved seamless mapping of surface O3 concentrations using an integrated deep learning framework

npj Climate and Atmospheric Science, Journal Year: 2025, Volume and Issue: 8(1)

Published: March 28, 2025

Language: Английский

---

Time Series Modeling of Ozone Concentration Constrained by Residual Variance in China from 2005 to 2020

Published: Jan. 1, 2024

Language: Английский

---