A bi-objective mixed-integer linear programming model to optimize thinning schedules in wildfire-prone Pinus...

Multi-objective programming is frequently used in forest management research to reconcile conflicting policy objectives. Given the frequent occurrence of fire in Pinus canariensis forests in the Canary Islands, forest managers have explored various silvicultural strategies, including thinning scheduling and intensity, to mitigate the adverse effects of fire on wood supply and improve fire suppression. This study presents an integrated modelling framework for computing Pareto frontiers. The framework results from maximizing economic returns and minimizing fire suppression difficulty and was achieved using thinning scenarios and assessing fire extinction risk. Thus, the response of Pinus canariensis forests to varying thinning schedules was evaluated across a geographic gradient in Tenerife (Canary Islands, Spain), using a bi-objective mixed-integer linear programming modelling (ɛ-constrained) approach. Multiple prospective alternatives for mitigating fire risk were formulated, considering specific parameters. For different levels of the relative tolerance parameter (β), the Pareto frontier revealed limited impacts on economic return (lower than 1.275) when the suppression difficulty index (SDI) was reduced. However, the total SDI and the number of stands thinned across the planning horizon were quite sensitive to variations in the β parameter. The latter results also had significant effects on average SDI, area of stands thinned, wood flows (roundwood and energy wood), thinning cost, and total return from wood production. Given that all the points along the Pareto frontier are optimal and dominant, the decision maker must operationally select the solution to be implemented. Regardless of the solution to be chosen, our findings highlight that an effective thinning schedule in fire-prone forests substantially enhances environmental outcomes, even with relatively modest reductions in the net benefits obtained from wood harvesting.