Resource management agencies worldwide seek to efficiently restore habitat and identification of effective restoration designs is a key step in this process. This collaborative research project included evaluation of alternative restoration plans for a large site on lower Prairie Creek in Redwood National Park, located in northwestern California. Prairie Creek is a critical regional hotspot for Coho Salmon (a threatened species) and for aquatic biodiversity in general and conservation of fish is a primary goal of this restoration project. Forest Service scientists and their collaborators formulated and applied a spatially explicit, individual-based model of Coho Salmon. One key feature of the model is its ability to explicitly incorporate temporal heterogeneity in environmental conditions: effective restoration projects provide habitats where animals can be continuously successful over lengthy time-spans. Key advances by the Forest Service’s Pacific Southwest Research Station this year included development of methods to assign model-specific habitat parameters in spatially explicit simulations of restoration alternatives. Results from these simulations are informing the selection of specific restoration plan and contributing to its design. This project is important because of the weakness of current approaches to forecasting restoration consequences, which rely on simplistic conceptual models with hidden assumptions that focus on static conditions rather than environmental regimes.