Optimising fuel treatments over time and space

Abstract

Fuel treatments have been widely used as a tool to reduce catastrophic wildland fire risks in many forests around the world. However, it is a challenging task for forest managers to prioritise where, when and how to implement fuel treatments across a large forest landscape. In this study, an optimisation model was developed for long-term fuel management decisions at a landscape scale. Using a simulated annealing algorithm, the model optimises locations and timing of fuel treatments, while considering changes in forest dynamics over time, fire behaviour and spread, values at risk, and operational feasibility. The model employs the Minimum Travel Time algorithm in FlamMap and the Fire and Fuels Extension to the Forest Vegetation Simulator to assess spatial and temporal effects with and without fuel treatments. The objective function is set to minimise total expected loss from a landscape due to wildfires throughout the planning horizon. The model was applied to a 14 000-ha study landscape located on the west side of the Bitterroot Valley in Montana. Comparisons between the optimised and random solutions show that the model was able to strategically locate and schedule fuel treatments to efficiently reduce expected loss from the landscape.