Harmonized tropospheric NO2 retrieval from LEO and GEO constellations and its application to air quality monitoring

Abstract

Satellite-based remote sensing has significantly advanced our understanding of global tropospheric nitrogen dioxide (NO2) over the past decades. Low-earth-orbiting (LEO) satellite instruments, such as GOME, SCIAMACHY, OMI, GOME-2, and TROPOMI, have provided valuable observations of NO2 distributions and trends worldwide. However, due to their orbital characteristics, LEO satellites offer observations only once per day at a specific local time. To overcome this temporal limitation, geostationary (GEO) satellite missions have been launched, including GEMS (Asia), TEMPO (North America), and Sentinel-4 (Europe). Although GEO platforms have limited spatial coverage, they offer high temporal resolution with multiple observations per day, enabling detailed monitoring of diurnal NO2 variability driven by emission dynamics and atmospheric chemistry. Given the complementary nature of LEO and GEO observations—LEO providing daily global coverage and GEO offering hourly snapshots over targeted regions—a combined approach offers a powerful opportunity to enhance NO2 monitoring capabilities. However, synergistic use of LEO and GEO data remains challenging due to differences in their retrieval algorithms, including variations in stratosphere-troposphere separation techniques, cloud corrections, and auxiliary inputs. In this study, we develop a harmonized retrieval system for tropospheric NO2 columns applicable to both LEO and GEO satellite measurements. The DLR NO2 retrieval algorithm for LEO and GEO constellations (NO2_LGC) consists of three key steps: (1) spectral retrieval of total NO2 slant columns, (2) separation of slant columns into stratospheric and tropospheric contributions using the Copernicus Atmosphere Monitoring Service (CAMS) forecast model data, and (3) conversion of tropospheric slant columns to tropospheric vertical columns using air mass factors (AMFs). Following Seo et al. (2024), this retrieval approach is consistently applied to current LEO (e.g., TROPOMI) and GEO (e.g., GEMS and TEMPO) observations, using identical forecast models, cloud corrections, and auxiliary datasets. This harmonized retrieval algorithm for the LEO+GEO constellation provides more consistent NO2 data records from TROPOMI, GEMS and TEMPO, and will be extended to future GEO and LEO missions, including Sentinel-4 and Sentinel-5. Furthermore, we demonstrate that combining harmonized LEO and GEO NO2_LGC retrievals enhances NO₂ analysis in regions with high pollution levels as well as in areas affected by pollution outflows.