Towards Advanced Sentinel-3 Near Real Time (NRT) L2 synergy aerosols capabilities - On-going Day 2 & Day 3 developments

As an operational user-driven Earth observation satellite agency, EUMETSAT is the reference European provider of Near Real Time (NRT - < 3h from the sensing time) Level 2 (L2) aerosol satellite observations from a constellation combining both Low Earth Orbit (LEO), with Metop / Sentinel-3 and EPS-SG, and GEOstationary (MSG & MTG). Primary users are operational air quality and climate services. Notably, for several years, EUMETSAT has closely interacted with the Copernicus Atmospheric Monitoring Service (CAMS) and provided expertise to support the uptake of all its observations into the modelling and assimilation processes.

With two multi-spectral optical sensors and observations acquired at a high spatial resolution at 10:00, Sentinel-3 is the main Copernicus mission mandated by the European Commission (EC) to provide a high quality of Aerosol Optical Depths (AODs) at global coverage during morning overpass time for the long future. As such, EUMETSAT is mandated since 2014 by its Member States and Copernicus to develop, enhance and ensure the Copernicus NRT Sentinel-3 Aerosol product. This is the 2^nd European product delivering Aerosol Optical (AOD) after Metop with PMAP (Polar Multi-sensor Aerosol optical Properties). Since 2020, it is derived from the OSSAR-CS3 (Optimized Simultaneous Surface Aerosol Retrieval for Copernicus Sentinel-3) algorithm, jointly specified & developed by EUMETSAT scientific experts & the Swansea University team led by Prof. Dr. Peter North (Chimot et al., 2021). Furthermore, EUMETSAT intensively works with the European Centre for Medium-Range Weather Forecasts (ECMWF) by exchanging the necessary expertise to support the future operational assimilation by the Copernicus Atmospheric Monitoring Service (CAMS), as done with PMAP.

The latest evolution, called Collection 3.1, compiles several major developments: 1) a sophisticated classification mask (called Naïve probabilistic) leading to enhanced coverage over all waters, and better cloud, snow, dust / ash, and dark vs. inland waters distinction, 2) more accurate 1^st guess of land vegetation reflectance, 3) revised underlight scattering caused by Ocean Colour features by using the Sentinel-3 Level 2 OLCI water reflectance, 4) improved constraints for bare soils in the dual-angular land model, and 4) a new Quality Indicator (QI) system integrated in the L2 product to guide further the users.

Based on lessons learned, EUMETSAT is now leading a major redesign of this algorithm, structured in two steps: first a Day-2 solely based on SLSTR, and a Day-3 intended as a NRT synergy (OLCI + SLSTR) to be based on the future Sentinel-3 NRT L1C under study and generated by EUMETSAT (De Bartolomei, 2014). As such, enhanced aerosol typing and further understanding of interaction with clouds are being investigated. This redesign also covers the development of Aerosol Layer Height (ALH) retrieval from the OLCI O2-A bands.

This presentation will summarise an extensive set of validation results including match-up with ground-based AERONET, inter-comparison of time series with MODIS/VIIRS from NASA and NOAA, and exchanges with the oeprational user community. Then, it will share the details of the plans of developments of the NRT Sentinel-3 Synergy and the major algorithm redesign items under specification, implementation, and testing.