Characterizing Fire and Fire Atmospheric States from Space Using Collocated Hyperspectral Infrared Sounding and Narrow-band Imagery

Under global climate change, wildfires play an increasingly important role in ecosystems. Satellite-borne instruments provide a unique vantage point to study fire, among which, the synergy of narrow-band imagers and hyperspectral infrared sounders enable the observation and monitoring of both the fire characteristics and the accompanied atmospheric changes with a multi-decadal global dataset. Therefore, it is important to develop methods that combine the advantage of the high spatial resolution from the imagers and the high spectral resolution from sounders for fire applications and studies. In this study, we use pixel-scale collocated instantaneous measurements of the Visible Infrared Imaging Radiometer Suite (VIIRS) and the Cross Track Infrared Sounder (CrIS), onboard NASA’s Suomi National Polar-orbiting Partnership (SNPP) satellite, to characterize both the fire and the atmospheric conditions before, during, and after fire. Two months of satellite observations over the Southwest United States and the Amazonia regions when large fires occurred in the regions of interest (October 2017 and August 2020) are used. It is found that CrIS displays high sensitivity to fire with spatial extent covering as small as only 1% of CrIS field of view (FOV), especially when averaged fire radiative power (FRP) larger than 1 megawatt during the night or 5 megawatts during the day. The principal components of the CrIS spectra during, and immediately before/after the fire events are used to quantify the spectral signature of infrared sounders to atmospheric temperature, humidity, and trace gases corresponding to fire characteristics. This study demonstrates that collocated imager and infrared sounder observations combined with the principal component analysis provide a valuable tool to efficiently identify and monitor fires and associated changes to the atmospheric states with greatly reduced data volume while retaining the full spectral information.