Multispectral Composites and Next-Generation Advanced Satellite Imagers

Multispectral composite imagery is created by combining multiple satellite spectral bands to create a single composite image. Single bands or band differences are assigned to the Red, Green, and Blue (RGB) components and a specific linear stretching and gamma correction are applied to each component. The combination of bands allows a forecaster to assess a variety of synoptic and mesoscale processes in a single image rather than analyzing multiple single channel satellite products separately. The European Organization for the Exploitation of Meteorological Satellites (EUMETSAT) began creating multispectral composites in the early 2000s with the advent of Meteosat Second Generation (MSG) Spinning Enhanced Visible and Infrared Imager (SEVIRI). EUMETSAT developed a set of best practice recipes for multispectral composites. The best practices identify a minimum set of multispectral composites based on MSG SEVIRI. The EUMETSAT best practices recognize the need to adjust the RGB recipe when creating multispectral composites with instruments such as the Moderate Resolution Imaging Spectroradiometer (MODIS) or the Advanced Very High Resolution Radiometer (AVHRR) due to spectral differences and absorption characteristics of the channels across instrument platforms which result in inconsistencies in the composite image from sensor to sensor. Since multispectral composites are qualitative in nature and interpretation depends on identification of features based on color, inconsistencies in multispectral composites due to spectral differences and band absorption characteristics can impact interpretation and analysis.

NASA Short-term Prediction Research and Transition (SPoRT) Center began creating the EUMETSAT best practices multispectral composites with the next-generation Himawari-8 Advanced Himawari Imager (AHI) as a proxy for advanced capabilities in the Geostationary Operational Environmental Series-R (GOES-R) era. NASA SPoRT worked closely with the National Atmospheric and Oceanic Administration (NOAA) National Weather Service (NWS) Operations Proving Ground (OPG) to provide AHI single channels and capabilities for the display of multispectral composites locally and on demand within the Advanced Weather Interactive Processing System (AWIPS) as part of an evaluation to prepare forecasters for the increased number of channels in the GOES-R era. Due to spectral differences of the AHI and SEVIRI instruments, the coloring of the AHI multispectral composites differed from standard multispectral composites derived from SEVIRI. In an effort to create consistent multispectral composites across multiple instruments, provide consistent multispectral composites to forecasters for the OPG evaluation, and stay within the realm of the EUMETSAT RGB best practices, this research investigated the impact of the spectral differences between SEVIRI and AHI on creating multispectral composites through an empirical correlation analysis. As advanced satellite sensors such as AHI become available, other agencies such as Japan Meteorological Agency (JMA), Australian Bureau of Meteorology, and EUMETSAT are recognizing the need to adjust the RGB recipes to account for spectral differences such as central wavelength and spectral width in order to maintain consistency with EUMETSAT RGB best practices. An empirical correlation analysis, radiative transfer modeling, and case studies are presented to highlight the need and a methodology to adjust the EUMETSAT RGB recipes and account for spectral differences across sensors.