Detecting Critical Nodes in Forest Landscape Networks to Reduce Wildfire Spread

This document presents a network optimization approach designed to strategically place forest fuel treatments, such as prescribed burns, to curtail the spread of wildfires. By representing a landscape as a network of connected patches, the authors use a modified Critical Node Detection model to identify and remove key nodes to effectively fragment fire pathways. Their methodology incorporates directional fire spread probabilities derived from stochastic simulations, allowing the model to account for real-world factors like wind patterns and topography rather than relying solely on local fuel loads. The study further refines the planning process by introducing constraints on the spatial contiguity and timing of treatments, ensuring that recommended fuel breaks are logistically feasible and safe to implement. The researchers demonstrate through a case study in Kootenay National Park that this optimized partitioning is significantly more effective at reducing landscape connectivity than traditional strategies that focus only on individual high-risk sites.