Understanding post-fire vegetation using computer vision

One of my earliest projects at the FOCAL in UC Davis, explored how vegetation recovers after wildfire. We developed a computer vision pipeline to transform large-scale, multi-year drone imagery into geospatial vegetation maps, enabling landscape-level change analysis. We found minimal change 3-4 years post-fire,  with key changes being shrub (re)establishment and tree resprouting. Our findings demonstrate the viability of scalable drone-based approaches for tracking vegetation change in fire-prone landscapes. Read preprint here (under review by Ecosphere).

We used data from three reserves in the University of California Natural Reserve System (UCNRS) including Blue Oaks Ranch Reserve, Hastings and Quail Ridge Reserve, all impacted by the 2020 wildfires and primarily a mix of oak woodland and grassland habitats.