Improved Methodology for GOES-R ABI Data Resampling for Hot Spot Detection and Fire Monitoring

The advanced baseline imager (ABI) on the Geostationary Operational Environmental Satellite (GOES)-R series includes both visible and infrared spectral bands that are used to locate and retrieve subpixel characteristics of fires. The spatial and temporal resolution of ABI imagery makes it a critical asset to NOAA’s real-time fire intelligence capabilities. The ground processing for ABI produces radiance imagery—the GOES-R ABI L1b data product—on a fixed grid such that every pixel in the image represents the same location on Earth for every product. This simplifies image-to-image geolocation issues for users and is achieved through an operational resampling process that is beneficial and optimized for many ABI products. The operational resampling algorithm is not, however, optimized for the detection of point sources, such as fires. This work details an effort focused on maximizing the extraction of information content from ABI imagery for fire detection and monitoring using the native detector sample observations. A methodology was developed to produce an optimized resampling algorithm consistent with the same fixed grid and operational format as the L1b data while preserving information observed for point sources and fire signals. The differences between the operational and optimized resampled imagery are explored, including radiometric, spatial, and temporal fidelity. This approach underscores a pathway to optimize the observation capabilities of ABI and future geostationary instruments (such as the GXI planned to be part of NOAA’s GeoXO program) for fire monitoring and potential enhancements to a critical Earth surface product from geostationary monitoring orbit.