Mapping Wildfire Risk to Transportation Infrastructure with Spatial AI
September 24, 2025 11:00 am (Central Time)
Abstract
In the 2024 I-GUIDE Spatial AI Challenge, our team built a lightweight U-Net model that maps wildfire threats along roads and railways — infrastructure often left out of wildfire risk modeling. We combined CAL FIRE threat zones, Sentinel-2 imagery, and transportation GIS data, to create a fast, spatially aware model that generates pixel-level wildfire risk maps, ready for real world use in areas like planning and disaster response.
This VCO spotlights how spatial AI can move wildfire risk modeling beyond forests and homes to focus on critical infrastructure. We’ll walk through our labeling workflow, model development, and GIS integration to show how this approach enables scalable, real-world decision support.
Speakers
Shaun Williams, CNA Corporation
Shaun Williams is a research analyst at the CNA Institute for Public Research with expertise in GIS, emergency preparedness, and infrastructure resilience. He holds a Ph.D. in Geography from Louisiana State University, where his research focused on the intersection of spatial analytics, accessibility modeling, and environmental risk. Shaun’s interests lie in applying geospatial modeling, deep learning, and data integration to improve emergency preparedness and sustainable infrastructure planning. He develops applied tools and research products that support public sector decision-making, threat assessments, supply chain resilience, and situational awareness.
Carey Whitehair-Conde, CNA Corporation
Carey Whitehair-Conde is a systems engineer at the CNA Institute for Public Research and has over 6 years of experience as an engineer and researcher. Carey specializes in supporting the integration of emerging technologies, and the development of software tools for data modeling. Prior to CNA, she worked as an engineering researcher studying nonlinear dynamics and controls.
Jeremiah Huggins, CNA Corporation
Jeremiah Huggins is a senior research specialist at the CNA Institute for Public Research and has 5 years of experience working as a remote sensing, GIS, and satellite operations specialist. At CNA, he provides expertise on Synthetic Aperture Radar, optical, infrared, and other forms of remotely sensed data in conjunction with the use of geospatial systems such as ArcGIS Pro and engineering applications such as ERDAS imagine, Socet GXP, and the European Space Agency’s Sentinel Applications Platform (SNAP). With a strong computer science background and knowledge of Python, SQL, JavaScript, R, and other languages, Jeremiah has played a fundamental role on technical projects at CNA involving AI and deep learning to analyze imagery data, conduct sentiment analysis, and extract text information. His contributions have advanced CNA's capabilities in these areas and expanded the knowledge base of the organization. In addition, Jeremiah has supported FEMA programs such as the Supply Chain Analysis Network (SCAN) and the Continuous Improvement Program (CIP).