AVIRIS-3 Rapid Response to January 2025 Los Angeles Wildfires

Wildfires in wildland-urban interfaces (WUIs) are a growing concern due to their devastating impact on human communities and ecosystems. Low-latency impact assessment is critical for wildfire response, yet immediate access to fire-affected communities can be limited. Here, we demonstrate that unmixing char/ash fractional cover using imaging spectroscopy data can support rapid structure damage assessment. Using acquisitions over the 2025 Eaton Fire burn scar collected by the Airborne Visible/Infrared Imaging Spectrometer 3 (AVIRIS-3), we demonstrate that a generic spectral endmember library is sufficient for binary structural damage classification between undamaged and destroyed buildings with an accuracy of 86.3%. Incorporating locally collected Eaton Fire endmembers into the library improves the accuracy by 1.3%. This demonstrates the feasibility of assessing structure damage from airborne imaging spectroscopy to rapidly inform post-fire response and recovery efforts in WUI communities.