Optimal Selection of Conservation Lands at Fort Stewart Using Integer Programming

摘要

This report documents an application of linear integer programming for determining compact and ecologically valuable conservation management areas (CMAs) on a military installation with populations of at-risk animal species. Two models were developed and applied to the conservation efforts Fort Stewart, GA, involving the at-risk Gopher Tortoise (GT) and the tortoise-dependent Gopher Frog (GF). The models produced solutions that are consistent with the species conservation and military training land-use objectives at Fort Stewart. They identified suitable, compact GT habitat clusters and were able to minimize the total amount of managed areas by selecting fewer and better sites. In runs that incorporated GF requirements into the GT analysis, the model was able to determine that the optimized GT CMAs also will support a small number of GF sites. However, on runs that assumed a large required number of GF sites, the GT results changed considerably in order to incorporate the GF sites. Both the single and joint species conservation management models were solvable in a short computation time, suggesting that these models may be applied to much larger data sets without significant data processing problems. The methods introduced here may be modified for application to other species, locations, and land uses.