Enhancing Fire Investigations through Multi-Satellite Data Fusion

Abstract:

To improve the depiction of gas emission into the troposphere associated with the fire outbreaks, we investigate the enhancement of CrIS fire information using VIIRS data in a fusion approach. CrIS ammonia (NH3) profiles at 13.5 km field of view (FOV) resolution are transformed to 750 meter pixels using VIIRS split infrared window data. The product fusion approach contains two main steps: k-d tree (nearest neighbor) search uses VIIRS bands 15 and 16 at 10.8 and 12.0 µm respectively at pixel (HIRES) and FOV (LORES) spatial resolution radiance data. For each HIRES pixel the best matches (in radiance space) of LORES FOVs are found. VIIRS HIRES data is averaged to represent LORES data collocated with the CrIS sounder FOVs. MUSES (MUlti-SpEctra, MUlti-SpEcies, MUlti-SensorsSEnsors) trace gas retrievals using CrIS data are then averaged over these 5 best LORES matches to provide a HIRES trace gas product. Examples of NH3 fields from several California fires in 2018 and 2019 are presented. Further research is underway to determine if fusion is able to provide important information on wildfire detection and plume characteristics that benefits operational fire management.