|
The Visible Infrared Imaging Radiometer Suite (VIIRS) ocean color (OC) Environmental Data Records (EDR) (or
Level-2 data) are one of the most important product sets derived from the visible and near-infrared (NIR) moderate
resolution (M) bands of VIIRS Sensor Data Records (SDR or Level-1B) data. NOAA Interface Data Processing System
(IDPS) only produces the operational SDR from Raw Data Records (RDR or Level-0 data) in the forward mode (no
reprocessing). As the ocean color EDR are highly sensitive to the quality of the SDR, while the forward processed SDR
by nature have large uncertainty, VIIRS SDR need to be reprocessed from RDR periodically with improved instrument
calibration algorithm and look-up tables (LUTs), primarily the F-factors LUTs (F-LUTs), for ocean color EDR. The
NOAA VIIRS ocean color team has been building the capability of processing the RDR to SDR efficiently for VIIRS
entire mission as well as any selected areas with any selected time periods. In this paper, we describe our effort to
develop two approaches in VIIRS RDR to SDR data processing. The first one is to process the raw RDR data into SDR
using an internal offline-processing tool, which is based on Joint Polar Satellite System (JPSS) Algorithm Development
Library (ADL) with updated calibration LUTs. This tool allows one to select the IDPS/ADL version and input
auxiliary/ancillary data for the VIIRS RDR to SDR reprocessing. The second approach, however, is for the case where
only the F-LUTs are updated (most cases). We have developed a ratio approach, which is based on the linear relationship
between the SDR radiance/reflectance and the F-factors. This approach requires only about one hundredth computational
effort compared to the ADL approach and can also significantly reduce the storage requirements. With the ratio
approach, the updated SDR are generated from the old SDR and only the bands that are interested are processed.
|
