Understanding different characteristics of multiple LEO and GEO satellite aerosol data over East Asia during 2016 KORUS-AQ campaign

Recent multi-channel geostationary (GEO) satellite sensors such as GOCI, AHI, and ABI have provided aerosol products with high accuracy, which enables to monitor rapid diurnal variations and transboundary transport of aerosols. Because most previous aerosol studies were based on low-earth-orbit (LEO) satellite sensors such as MODIS and MISR, an understanding of similar but different characteristics between LEO and GEO aerosol measurements is newly required.

The KORUS-AQ campaign was held over Korea from 1 May to 12 June 2016. One of the main objectives of KORUS-AQ campaign encompassing ground-, airborne-, and satellite-based remote sensing is to validate measurement accuracy, improve measurement techniques and understand local and long-range transported contribution to the air quality. For satellite aerosol measurement, the GOCI Yonsei aerosol retrieval (YAER) algorithm was improved to the version 2 and operated during the campaign to monitor aerosol distribution changes as near-real-time and support air-quality forecasting through the data assimilation.

In order to evaluate satellite-retrieved 550 nm aerosol optical depth (AOD) data during the campaign, the ground-based AERONET sunphotometer data are collected from 33 East Asia sites including Korea, Eastern China, and Japan. First, in this study, the latest version of multiple LEO satellite aerosol data as MODIS Dark-Target and Deep-Blue Collection 6.1, MISR version 23, and VIIRS EDR and GEO satellite aerosol data as GOCI YAER version 2 are validated using AERONET version 3 data. Different characteristics in terms of pixel filtering methods, bias patterns, and sampling issues are also analyzed. Although the most products show a similar spatial distribution of mean AOD and instantaneous AOD scenes, the GOCI AOD shows more continuous AOD change spatiotemporally when there are transported high AOD plumes. Compared to GEO, the aerosol data of LEO satellites have temporally (revisit in 1-3 days) and spatially missing between tracks or sun-glint area. Therefore, the gap-filling approach using multi-sensor synergy using LEO and GEO sensors would be expected to generate spatiotemporally continuous and accurate aerosol distribution.