Close Window

Harnessing the full information content from VIIRS for active fire detection and tracking

Shane R Coffield,  NASA GSFC / UMD,  shane.coffield@nasa.gov (Presenter)
Tempest McCabe,  NASA GSFC / UMD,  tempest.mccabe@nasa.gov
Yang Chen,  University of California, Irvine,  yang.chen@uci.edu
Lesley Ott,  NASA GSFC GMAO,  lesley.e.ott@nasa.gov
Douglas Morton,  NASA GSFC,  douglas.morton@nasa.gov

Recent record-breaking fire activity in the Western US poses clear threats to humans, ecosystems, and the climate. It also increases challenges for fire managers and further motivates the need for improved fire science to track fire behavior, drivers, and impacts. In particular, there are well-known limitations to our current ability to monitor fires from space. These include smoke and cloud obscuration, omission of small or low-intensity fires, and attenuation of fire radiative power (FRP) through the atmosphere for different view angles. In this study we reexamine the VIIRS Level 1 and 2 fire imagery, considering thermal radiance data for candidate fire pixels (labeled “candidate” or “background” fires by the VIIRS fire algorithm) in addition to fire pixels already included in the Level 2 active fire product. We found that the additional candidate detections comprised 28% of all daytime detections and 12% of all nighttime detections, while not contributing significantly to false positive (commission) errors. In some cases of large fires with pyro-cumulonimbus (pyroCb) cloud obscuration, we found candidate fire detections where there was otherwise a complete gap in fire activity. Current ongoing work involves measuring fire persistence and improving FRP estimates leveraging this new information. Including the most available data over these areas of known fire activity is critical for accurate fire behavior analysis, carbon emissions estimates, and air quality modeling, both respectively and operationally.

Poster: Poster_Coffield_2-31_151_35.pdf 

Poster Location ID: 2-31

Presentation Type: Poster

Session: Poster Session 2

Session Date: Wed (May 10) 5:15-7:15 PM

CCE Program: Other

Close Window