| The alarming biodiversity loss resulting from human induced habitat loss has attracted considerable interest in the biological conservation literature in the past decades. Establishing nature conservation reserves is considered as the most effective and direct way to maintain habitat quality and restore ecosystem services and therefore to protect biodiversity. Resources devoted to conservation reserves have always been scarce, however. This urges social planners to design economically and ecologically efficient reserve networks. Selecting an optimal subset of sites from among a large number of candidate sites is a critical step in this process.; In the literature, the reserve site selection problem has been addressed using heuristic and formal optimization approaches. The former approach is practical, but in general it cannot guarantee optimal solutions. On the other hand, formal optimization guarantees the most efficient resource utilization, but computational complexity may be restrictive in large scale applications especially when spatial configurations have to be taken into account in site selection.; This dissertation is built upon some recent studies which dealt with two important spatial criteria, namely reserve contiguity and minimal fragmentation, using graph theoretic approaches and linear integer programming. The optimization methods presented in those studies are modified to improve computational efficiency of the solution procedure, which may be particularly important in moderate and large scale applications. The methods developed in this study are applied to state-endangered and threatened bird species in Illinois and an efficient reserve system is determined at State level while conservation goals at regional level are taken into account. |
