Designing compact and contiguous reserve networks with a hybrid heuristic algorithm

Conflicting opinions from environmental advocates and economic interests on the best strategy for management of public lands often leaves land managers in a difficult position. Since ecosystem sustainability is in the long-term interest of each group, the establishment of nature reserves could simultaneously address both views. To promote sustainability, fragmentation of existing natural habitats should be avoided, since it is commonly recognized as being disruptive to the species adapted to these habitats. Therefore, when designing an efficient nature reserve, the compactness and contiguity of the land reserved is an essential consideration. A new formulation of the reserve selection problem is presented that explicitly addresses these issues; specifically, the model minimizes a weighted combination of compactness and contiguity measures subject to constraints on the minimal representation of each habitat class. Motivated by the ongoing reserve efforts in the large and diverse Klamath–Siskiyou region of southwestern Oregon and northwestern California, common heuristic search techniques are implemented and results compared on various simulated test problems. From these findings a new heuristic is developed that reduces solution time and increases solution quality. When applied to the Klamath–Siskiyou region, results are promising.