Detection and characterization of biomass burning in the GOES-R era

Geostationary weather satellites have been used operationally to detect and characterize wildfires and anthropogenic biomass for the last decade. The Wildfire Automated Biomass Burning Algorithm (WF_ABBA) developed at the Cooperative Institute for Meteorological Satellite Studies (CIMSS) at UW-Madison has been used with data from GOES-8 through GOES-15, Meteosat-8/-9, and MTSAT-1R/-2, with support for other satellites in the international geostationary constellation in development. The increased spatial and temporal resolution afforded by GOES-R ABI, as well as the additional infrared channels, make the GOES-R Fire Detection and Characterization product a substantial step forward in the development of geostationary fire detection and characterization. This fire data finds use in aerosol and smoke modeling as well as real-time monitoring of active fires; both sets of users will see benefits from the higher data rate and low latency provided by ABI, improving detection confidence and the quality of derived characteristics. Additionally, research is underway to utilize the new capabilities of ABI for fire detection and characterization, including the use of additional, previously unavailable channels in the algorithm. Development of the ABI Fire Detection and Characterization algorithm has led to pioneering work in the modeling of satellite fire detection, primarily with respect to understanding how sub-pixel fires with their high-contrast against the background are measured by a sensor whose footprint is much larger than the fires being detected as well as how resampling of satellite data changes the quality of fire detection and characterization. Both producers and users of satellite detected and characterized fire data benefit from this enhanced understanding of how the observations themselves drive the error budget for fire detection and characterization.