EPSCoR Research Fellows: NSF: Enhanced Predictive Understanding of Wildfire Ignitions in the Face of Changing Socioenvironmental Landscape

Wildfires have increasingly impacted social and environmental systems in many regions across the United States in recent decades, resulting in a tripling of the number of people directly exposed to wildfires from 2000-2019 in the western US. A majority of the wildfires in the western US (>60%) are human-started, many of which are preventable. Importantly, human-started wildfires account for an absolute majority of human losses, including 97% of the structures threatened across the western US, highlighting the importance of wildfire prevention. This project establishes new collaborations between Boise State University and the Universities Space Research Association (USRA) and NASA Ames Research Center (ARC) to advance the predictive understanding of wildfire ignitions, which in turn informs wildfire mitigation efforts and supports wildfire-aware growth pathways in Idaho. This work can generate transferable information and models for other states in the western US. In addition to informing the transformation of wildfire mitigation efforts from a reactive paradigm to proactive planning and implementation, this project will contribute to educational and outreach initiatives by supporting (1) a PhD student, (2) course offerings at BSc, MSc, and PhD levels, and (3) expanded collaboration between Boise State University, USRA, NASA ARC, National Interagency Fire Center, Forest Service, and US Geological Survey.Wildfire ignitions follow temporal and spatial structures, modulated through a cohort of biophysical and anthropogenic factors, enabling scientists and practitioners to predict them and devise targeted wildfire prevention strategies and response efforts. First, this project will advance the scientific understanding of climate, weather, land cover/use, topography, social, and management attributes associated with various wildfire ignitions across the western United States. Findings will illuminate (1) the similarities and differences between attributes associated with 13 different ignition causes of wildfires, (2) commonalities and distinctions between wildfire ignition attributes across the 11 states of the western US, and (3) whether or not distributions of ignition attributes are stationary. This will be accomplished through a variety of data analytics approaches, including descriptive and predictive methods. Second, this project will investigate the changing characteristics of the background social and environmental processes that shape the landscape of wildfire occurrence, and develop advanced machine-learning models to unravel future potential wildfire ignition pathways in response to climatic changes and population growth. Specifically, the project will adopt an innovative sampling strategy to represent the future social and environmental pathways and how they modulate wildfire ignitions. To that end, two general categories of possible scenarios will be investigated: (1) increasing population and expansion of wildland-urban interface, and (2) increasingly dry fuels due to background warming. The project outcomes support targeted and effective wildfire prevention strategies and inform community growth pathways that enable sustainable human coexistence with wildfire.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.