System Vicarious Calibration (SVC) is essential to meeting the radiometric uncertainty requirements of ocean colour satellite missions, and to ensure consistent, seamless, merging of data sets from successive missions to generate long-term climate-quality time series. As multiple sources of high-quality in-situ radiometric data capable of supporting primary SVC become available, the ability to reconcile gain factors derived from different sources becomes increasingly important. Differences in the atmospheric correction algorithms used across operational processing chains represent a barrier to direct comparison of platform-derived gain factors.
This work derives SVC gain factors for MODIS Aqua using in-water radiometry from 1) the BOUSSOLE buoy (Mediterranean Sea; 2003-2023), with both its multispectral and hyperspectral data records, 2) the MOBY (Hawaii; 1997-current) and, 3) the MarONet buoy (2024-current), operated in the eastern Indian Ocean. Matchups are identified within a one-hour temporal window and subject to rigorous quality filtering, yielding approximately 300 high-quality matchups for the BOUSSOLE multi-spectral record spanning 2003 to 2016. A single atmospheric correction algorithm, based on Antoine and Morel (1999), is applied consistently across all platforms to propagate in-water radiometry to top-of-atmosphere radiance, following Franz et al. (2007) for pixel selection and averaging.  The use of consistent atmospheric correction means that methodological differences between processing chains do not contribute to inter-platform gain differences but are isolated to in-water radiometry and atmospheric conditions at each platform.
The per-platform gains are applied to generate Level 2 retrievals of remote sensing reflectance and chlorophyll concentration. The multi-sensor processing pipeline is applicable to Sentinel-3 OLCI and extension to this sensor is planned. Results are discussed with implications for reconciling SVC gain factors derived from a multi-site network.
 

暂无