Early warning of illegal development for protected areas by integrating cellular automata with neural networks.
Land acquisition is a common practice for establishing and expanding protected areas such as wildlife refuges. However, the socioeconomic feasibility and ecological consequences of an acquisition project are rarely assessed before the project is executed. In this paper, a socio-economic-ecological model (SEELAND) was developed to simulate the socioeconomic feasibility and ecological consequences of a land acquisition project, using the Shiawassee National Wildlife Refuge in Michigan (USA) as a case study. The refuge is managed by the US Fish and Wildlife Service (USFWS) primarily for waterfowl. An adjacent area of 3035 ha has been proposed to add to the current refuge of 3680 ha. The vast majority of the proposed acquisition area is privately owned. SEELAND consists of three main components: sociological (e.g. landowners’ attitudes toward selling their land), economic (e.g. fair market value and incentives), and ecological (e.g. land-cover types, soil types, parcel sizes and locations). Simulation results indicated that most of the high-priority land was not available for purchase and the priority set by the USFWS could not be achieved. Many purchased land parcels were not connected to each other or to the existing refuge, resulting in small isolated patches, which are not good for habitat connectivity and refuge management. Furthermore, landowners’ attitudes towards selling their land significantly affected the amounts and types of land purchased. Without using incentives, less than half of the proposed acquisition area would be purchasable within the next 20 years. © 2001 Elsevier Science B.V. All rights reserved.