41EIPT NOAA Satellite Oceanographic Products Scientific Stewardship and Development at NEDSIS National Centers for Environmental Information (NCEI): Data Archival, Public Access and Production based on Machine Learning

In its role as the U.S. archive for oceanographic data, the Coasts, Oceans, and Geophysics Science (COGS) Division of the NOAA/NESDIS National Centers for Environmental Information (NCEI) provides rigorous data archiving, user-friendly data distribution, and value-added product development for both NOAA and NOAA partners’ oceanographic satellites and their other ocean and coastal data. Related satellite oceanographic data application research carried out by the NCEI/COGS Division Satellite Oceanography Group (SOG) results in a better understanding of ocean climate, an effective scientific data stewardship and generation of new value-added satellite ocean surface products for the sea surface temperature (SST), altimeter sea surface height (SSH), ocean color, ocean surface wind (OSW) and other satellite-observed oceanographic products. The satellite missions that SOG supports cover those from NOAA and NOAA partners: European Space Agency (ESA), European Organization for the Exploitation of Meteorological Satellites (EUMETSAT), Japan Meteorological Agency (JMA), Canadian Space Agency (CSA), Centre National d'Etudes Spatiales (CNES), Indian Space Research Organization (ISRO). They include, but not limited to, the OSTM/Jason-2, Jason-3 and CS, Suomi National Polar-orbiting Partnership (S-NPP), NOAA-20 and 21, GOES-16 and 18, OceanSAT-3, Meteosat-9 and 10, Himawari-9, Sentinel-1A, RADARSAT2 and RadarSat Constellation Mission (RCM), etc. In a higher NCEI archival service tier, SOG also develops tools to enable the automated collection of quality descriptive statistics in each archived data file in order to monitor the data quality and track the metadata attributes. In addition to its great contribution to the data archive and access, SOG works toward the development of the value-added products using machine learning (ML) techniques. A Random Forest (RF) model has been trained to synthesize the satellite SSH anomaly (SSHA) and in situ depth equivalent ocean bottom pressure observations from the NOAA Deep-ocean Assessment and Reporting of Tsunamis (DART) records, as well as other related observations (e.g., tide gauge data, SST, ocean surface wind and sea surface pressure), for exploring a possible improvement of the gridding of multi-satellite SSHA as the DART records and other related parameters provide a continuous daily sampling of the SSHA variations. In this poster, we will provide a brief introduction of the NCEI data archival service, public data discovery and access tools for the NOAA satellite oceanographic products and our efforts in the development of production using the ML method.