Evaluation of ocean color data processing schemes for VIIRS sensor using in-situ data of coastal AERONET-OC sites

摘要

In the processing of Ocean Color (OC) data from sensor data recorded by Visible Infrared Imaging Radiometer Suite (VIIRS) aboard JPSS-Suomi satellite, NASA Ocean Biology Processing Group (OBPG) is deriving a continuous temporal calibration based on the on-board calibration measurements for the visible bands, and then reprocessing the full mission to produce a continuously calibrated sensor data record (SDR) product. In addition, a vicarious calibration during SDR to OC Level-2 processing is applied. In the latest processing the vicarious calibration is derived from the Marine Optical Buoy (MOBY) data, whereas in the initial processing it was derived from a sea surface reflectance model and a climatology of chlorophyll-a concentration. Furthermore, NASA has recently reprocessed the OC data for the entire VIIRS mission with lunar-based temporal calibration and updated vicarious gains. On the other hand, in fulfilling the mission of the U.S. National Oceanic and Atmospheric Administration (NOAA), the Interface Data Processing Segment (IDPS) developed by Raytheon Intelligence and Information Systems, for the processing of the environmental data products from sensor data records, has gained beta status for evaluation. As these processing schemes continue to evolve, monitoring the validity and assessments of the related VIIRS ocean color products are necessary, especially for coastal waters, to evaluate the consistency of these processing and calibration schemes. The ocean color component of the Aerosol Robotic Network (AERONET-OC) has been designed to support long-term satellite ocean color investigations through cross-site measurements collected by autonomous multispectral radiometer systems deployed above water. As part of this network, the Long Island Sound Coastal Observatory (LISCO) near New York City and WaveCIS in the Gulf of Mexico expand those observational capabilities with continuous monitoring as well as (for the LISCO site) additional assessment of the hyper-spectral properties of coastal waters. In the investigations carried out over a one and half year period dataset of VIIRS, based on the data from two coastal AERONET-OC sites, it has been observed that the VIIRS sensor captures well the seasonal and temporal variations in the nLw data, exhibiting significant correlation with in-situ data (R = 0.929 and 0.985 for LISCO and WaveCIS respectively). For the WaveCIS site, VIIRS nLw data retrievals are seen to be enhanced with each incremental adjustments of vicarious and calibration procedures. However, that is not the case for the LISCO site which exhibits more frequent occurrences of negative water-leaving radiances, while underestimation in VIIRS nLw data is further exacerbated. Strong consistency between the time-series nLw data retrieved from the VIIRS and MODIS sensors was also observed.