A tie-point zone group compaction schema for the geolocation data of S-NPP and NOAA-20 VIIRS SDRs to reduce file sizes in memory-sensitive environments. (June 2020)

• Record Type: Journal Article
• Authors: Soerensen, Anders Meier; Zinke, Stephan
• Is Part Of: Applied computing and geosciences. Volume 6(2020)
• Publication Date: 2020-06
• Subjects: European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) -- Suomi National Polar-orbiting Partnership (S-NPP) -- Visible Infrared Imager Radiometer Suite (VIIRS) -- Data compression -- Hierarchical Data Format version 5 (HDF5) -- Tie-point zones; Earth sciences -- Data processing -- Periodicals; 550.285
• DOI: 10.1016/j.acags.2020.100025
• Languages: English
• ISSNs: 2590-1974

Abstract

Abstract: The European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) disseminates weather and climate-related satellite data to its users via satellite broadcast. As part of this, the EUMETSAT Advanced Retransmission Service (EARS) provides Sensor Data Records (SDR) from the United States satellites Suomi National Polar-orbiting Partnership (S-NPP) and NOAA-20. Due to bandwidth constraints and for cost-reduction reasons for the upload to the disseminating satellite, a near-lossless compaction method has been developed for the geolocation data of the data capturing satellites' Visible Infrared Imager Radiometer Suite (VIIRS) instrument. The geolocation data are compacted by storing only the data representation for so called tie-points which form tie-point zones and are grouped in tie-point zone groups. Compression factors of 120 for M-Band, 479 for I-Band, and 154 for Day/Night Band, not using HDF5 internal compression on neither the input nor the output, and 56 for M-Band, 210 for I-Band, and 89 for Day/Night Band, using HDF5 internal compression on both the input and output, respectively, are reached. The compaction process introduces an error of about 1 ​m RMSE for the position in latitude/longitude and of about 0.001° RMSE for the angular data. The concept of the tie-point zones and the quadratic interpolation schema with coefficients analytically derived from the scanning geometry forms the main elements of the method presented. The full algorithm and mathematical background needed are presented, and the algorithm can be highly parallelized. The method can likely be applied to any cross-track scanning sensor. Highlights: Remote sensing geolocation data can be efficiently compacted. Compression factors of more than 50 can be reached. Data is stored only for tie points of tie point zones. Tie point zone schema is sensor dependent and needs specific development.