Abstract: |
Quantitative models of how wildfire responds to climate, ecosystem, and land management activities at multiple spatial and temporal scales provide critical information for optimizing vegetation management, allocating pre-suppression and suppression resources across time and space, and predicting spatial and temporal spillovers across ecosystems and ownerships. When expressed in statistical forms useful for economic modeling, the resulting wildfire production functions provide mechanisms for evaluating economic trade-offs among alternative wildfire interventions. In this presentation, we outline alternative methods expressing and estimating wildfire production functions at a variety of scales. We consider the advantages, limitations, and complementarities of alternative functional forms for wildfire production. These include the following: size-frequency distributions; single-equation and multiple-equation systems as descriptions of wildfire area, damages, or risk; ignition count models; extreme-value functions; Heckman-type ignition probability-area models; and spatial point process models. We conclude by discussing how these approaches can be incorporated into optimal intervention simulations at fire or landscape scales. |
