Atmospheric measurement techniques · 2024 · 17 citations · 59 references
Natural SciencesComputer ScienceEarth SciencesAtmospheric ScienceRemote SensingSensor FusionAerosol Remote SensingAtmospheric Sounding From SatellitesRemote Sensing Image AnalysisSatellite Data ValidationSatellite InstrumentationSatellite SensorsAerosol Measurement TechniquesAerosol MonitoringAerosol Optical Depth RetrievalData FusionGeostationary Satellite MonitoringMulti-satellite Data FusionPixel-level FusionStatistical Data Fusion
Abstract. Data fusion of aerosol optical depth (AOD) datasets from the second generation of the Geostationary Korea Multi-Purpose Satellite (GEO-KOMPSAT-2, GK-2) series was undertaken using both statistical and deep neural network (DNN)-based methods. The GK-2 mission includes an Advanced Meteorological Imager (AMI) aboard GK-2A and a Geostationary Environment Monitoring Spectrometer (GEMS) and Geostationary Ocean Color Imager II (GOCI-II) aboard GK-2B. The statistical fusion method, maximum likelihood estimation (MLE), corrected the bias of each aerosol product by assuming a Gaussian error distribution and accounted for pixel-level uncertainties by weighting the root-mean-square error of each AOD product for every pixel. A DNN-based fusion model was trained to target AErosol RObotic NETwork (AERONET) AOD values using fully connected hidden layers. The MLE and DNN AOD outperformed individual GEMS and AMI AOD datasets in East Asia (R = 0.888; RMSE = −0.188; MBE = −0.076; 60.6 % within EE for MLE AOD; R = 0.905; RMSE = 0.161; MBE = −0.060; 65.6 % within EE for DNN AOD). The selection of AOD around the Korean Peninsula, which incorporates all aerosol products including GOCI-II, resulted in much better results (R = 0.911; RMSE = 0.113; MBE = −0.047; 73.3 % within EE for MLE AOD; R = 0.912; RMSE = 0.102; MBE = −0.028; 78.2 % within EE for DNN AOD). The DNN AOD effectively addressed the rapid increase in uncertainty at higher aerosol loadings. Overall, fusion AOD (particularly DNN AOD) showed improvements with less variance and a negative bias. Both fusion algorithms stabilized diurnal error variations and provided additional insights into hourly aerosol evolution. The application of aerosol fusion techniques to future geostationary satellite projects such as Tropospheric Emissions: Monitoring of Pollution (TEMPO), Sentinel-4, and Geostationary Extended Observations (GeoXO) may facilitate the production of high-quality global aerosol data.
59
AERONET—A Federated Instrument Network and Data Archive for Aerosol Characterization
B. N. Holben, T. F. Eck, I. Slutsker et al. · Remote Sensing of Environment · 1998 · 8.5K citations · Full text
The MODIS Aerosol Algorithm, Products, and Validation
L. A. Remer, Yoram J. Kaufman, D. Tanré et al. · Journal of the Atmospheric Sciences · 2005 · 3.4K citations · Full text
The Collection 6 MODIS aerosol products over land and ocean
R. C. Levy, S. Mattoo, L. A. Munchak et al. · Atmospheric measurement techniques · 2013 · 2.4K citations · Full text