Cloud filling of oceanic chlorophyll-a concentration remote sensing products by DINEOF methodology

Ocean color data is a crucial tool for monitoring and understanding optical, biological, and ecological phenomena in aquatic environments. The measurement of chlorophyll-a concentration effectively facilitates the monitoring of ocean eutrophication and primary productivity. Traditional in-situ observations of the ocean are constrained by temporal and spatial scales, rendering them incapable of providing large-scale continuous monitoring of ocean color data. Satellite remote sensing has the potential to facilitate global ocean monitoring and offer extensive long-term observations of chlorophyll-a concentration products. The FY-3 Meteorological Satellite Medium Resolution Spectral Imager (MERSI) has been providing global ocean color products since 2008. However, achieving continuous daily spatial coverage of the global oceans with a single instrument is impractical due to several constraints, including the instrument's scanning width, zenith angle, solar flares, and cloud cover. In contrast, the U.S. SNPP and NOAA-20 Visible Infrared Imaging Radiometer (VIIRS) instruments offer extensive global marine aquatic products. We investigate a data-interpolating empirical orthogonal function method to composite chlorophyll-a concentration images derived from VIIRS and FY-3 MERSI satellites, aiming to produce daily global chlorophyll-a datasets. The processes of outlier detection and removal significantly enhance the overall efficacy of this interpolation technique. Chlorophyll-a data at both Level 2 and Level 3 have been utilized and reprocessed to recover missing information from cloudy images. This study demonstrates that the proposed methodology effectively addresses gaps in chlorophyll-a concentration data resulting from geophysical factors associated with limitations in inversion algorithms—such as elevated sensor zenith angles.