Incorporating long-term spatial and temporal effects in a state space model to evaluate land management alternatives for imperiled species conservation in Alabama
Type of Degreethesis
Forestry and Wildlife Sciences
MetadataShow full item record
Conservation decisions are often made based on expected outcomes without full consideration of costs, likelihood of achieving targeted results, or spatial and temporal consequences of their implementation. We used structured decision making to establish management objectives and compile information for development of a decision support tool to evaluate management alternatives on ten state-owned lands in Alabama. The identified problem was to determine how to best manage state lands to enhance primary functions of properties while improving habitat for imperiled wildlife. We developed a heuristic state space model to predict consequences on wildlife species of 11 management alternatives implemented over a 100-year planning horizon. Management objectives, alternatives, and costs were elicited from land managers and included combinations of actions that affect land cover type and structure on uplands, floodplains, and wildlife openings. We used a matrix of land cover transition rates describing natural and human-induced processes to predict the cost of management, user (e.g. hunters and hikers) preferences, and wildlife responses based on likelihood of land cover change. We derived user preferences from surveys of potential users. Wildlife responses were predicted from occupancy rates estimated using field data for a suite of focal species representing each imperiled species. We estimated the utility of each alternative based on the average of the scaled outcomes for each objective on each state park and wildlife management area. The preferred alternative varied among study areas based on their intended use. For example, removing existing wildlife openings within upland pine provided greater utility for imperiled species on wildlife management areas within the coastal plain. Our model is capable of incorporating additional objectives and trade-offs and would be useful for evaluating alternatives at landscape scales in similar regions. Additionally, we make recommendations for monitoring programs and user preference surveys that may reduce the ambiguity of management decisions.