Modeling spatial management objectives for USDA Forest Service planning in Minnesota

Research focused on improving a modeling system for forest management scheduling that tracks and values core area of mature forest while also addressing economic and ecological objectives. Test cases utilized model formulations associated with the 2004 forest plans for the Chippewa and Superior National Forests. The University of Minnesota has a long history in model development for large-scale forest management planning. Emphasis was on improving the existing system to better address many complex, interrelated, stand-level management decisions associated with landscape objectives.; A map of influence zones can be used to describe spatial interdependencies of stand-level decisions. Developing data to describe influence zones is complicated because of the many-to-many interrelationships between stands. Each influence zone represents an area of the forest that is influenced by a unique combination of stands. Forests contain many more influence zones than stands. An approach was proposed and tested for developing influence zone information. Tests found that the approach could give adequate results with computation time requirements shortened from days to a few hours compared to methods currently in use.; The University of Minnesota forest management scheduling system uses multiple model iterations. A series of overlapping subproblems are solved each iteration. Tests identified tradeoffs between computation time and loss in optimality from using smaller subproblems, smaller overlaps, fewer influence zones, and fewer stands. Early iterations of the overall solution process might emphasize computation time with loss of optimality a greater concern in later iterations. An efficiency frontier was developed describing effective model settings for applications for the Chippewa National Forest.; Core area is only one spatial management concern. Test results for both national forests found that valuing core area will also tend to produce larger patches of mature forest. Valuing core area allows patches of mature forest to move about the landscape over time. Test results show that with good planning core area production levels can be increased substantially over the long-term without large impacts on timber production. Shorter-term, it is much more difficult to increase core area levels because of the time required to change spatial arrangements.