Graduation date: 2007This paper presents a framework for analyzing efficient spatial allocation of forest management efforts - fuel treatment and harvest - under the risk of fire. The framework integrates a fire behavior model and a spatially-explicit stochastic dynamic optimization model. I investigate the effects of spatial interaction across plots during forest fires - in particular a spatial externality - on efficient allocation of fuel management efforts. This spatial externality is captured in a spatial endogenous risk framework - decisions affect a risk distribution over space. By solving computationally for a number of initial bioeconomically heterogeneous landscapes, general insight into implementing spatial allocation of fuel management efforts is derived. Sensitivity analyses are conducted to evaluate various economic and physical environments and their impact on the optimal solutions. Because the optimal spatial allocation of fuel management efforts depends on a spatial distribution of "value" and "risk", the decision generated from a conventional dynamic stand level (or aspatial forest level) model or a simple dynamic spatial model can be suboptimal. Numerical solutions demonstrate that the risk of value loss by fire damage does not necessarily shorten the optimal rotation age because of spatial externalities. These analyses also show that a land manager faces spatially explicit trade-offs in deciding how to investment in fuel treatments.