The need for better resilience planning against the expanding threat of wildfires is the focus of a new paper in the ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering. Researchers Vishnupriya Jonnalagadda, Ji Yun Lee, and Abdel-Aziz Sanad began with a reality many practitioners now face: Wildfire risk is increasing in both frequency and consequence as climate conditions shift and development continues to expand into wildland–urban interface areas. Their paper “Cluster-Based Active-Learning Scenario Reduction Framework for Probabilistic Wildfire Risk Assessment” makes the case that understanding wildfire risk takes more than looking at a few individual fire scenarios. Instead, the paper points toward broader, probability-based methods that can better reflect uncertainty and the many ways a wildfire event might unfold. This is crucial in real-world settings, where decision-makers often must act quickly and with limited time, information, money, and technical resources.
Their approach offers a smarter way to work with huge amounts of wildfire data without losing sight of the most important risks. The proposed framework reduces a very large set of possible wildfire scenarios down to a smaller group of representative ones, while still capturing the bigger patterns and the rare but severe events that can cause the most damage. By applying this method, planners, local officials, and other professionals will better understand wildfire risk and make more informed choices about preparedness, mitigation, and resource allocation. The included case study demonstrates a substantial reduction in the number of scenarios needed while maintaining strong fidelity to regional burn probability and building damage patterns. Learn more about this research at https://ascelibrary.org/doi/10.1061/AJRUA6.RUENG-1840. The abstract is below.
Abstract
Wildfires pose an escalating risk to communities and infrastructure, especially in regions undergoing increased fuel dryness and temperature extremes driven by climate change, as well as continued expansion into the wildland–urban interface. Probabilistic wildfire risk assessment provides a rigorous means of quantifying potential impacts, but its application is often hindered by the high computational cost of working with hundreds of thousands of complex wildfire scenarios. This study introduces a novel scenario reduction framework tailored to the unique characteristics of wildfire hazards, which often lack standard intensity metrics and exhibit highly nonlinear, spatially distributed behavior. The proposed framework selects a subset of scenarios that best represent the spatial and statistical diversity of the full dataset, thereby greatly reducing computational costs while accounting for uncertainties. This is achieved by mapping complex wildfire scenarios into a high-dimensional feature space, enabling similarity assessments based on spatial consequence patterns rather than standard intensity metrics. A 𝑘-medoids clustering approach is then used to identify a representative subset of scenarios, while an active-learning-based outlier selection procedure incorporates rare but high-impact events without inflating computational demands. The framework was first demonstrated using a simple illustrative example to show how its performance responds to different data characteristics. To further demonstrate the practicality of the framework, it was used for wildfire risk assessment in Spokane County, WA, where the full dataset (1,000 scenarios) was reduced to 41 representative scenarios while preserving the spatial patterns of burn probability and building damage with high fidelity. The results demonstrated that the framework significantly improves computational efficiency and accuracy compared to traditional scenario reduction methods, offering a scalable and flexible tool for probabilistic wildfire risk assessment.
Learn more about using wildfire data to assess future probabilities and better channel resources in the ASCE Library: https://ascelibrary.org/doi/10.1061/AJRUA6.RUENG-1840.
