Assessing the effectiveness of protected areas: An analysis of large herbivores in Tanzania

Accelerating rates of habitat destruction and overexploitation have prompted concern over how best to conserve wildlife and habitats. Protected areas are intended to protect species in the face of increasing demands for natural resources, yet much controversy surrounds the appropriateness of this conservation strategy. This dissertation addresses the effectiveness of protected areas in conserving large herbivores in Tanzania, an East African country harboring a vast diversity of megafauna. Aerial census data collected in large census zones across the country between the late 1980s and early 2000s showed that elephant increased in many census zones over a decade but the vast majority of herbivore populations showing significant changes in densities over this time period declined.; Comparisons between specific types of protected areas show that higher percentages of species fared well (increased significantly or showed no significant change) in strictly protected National Parks than in areas with partial or no protection and in heavily protected Game Reserves relative to areas with no protection. While heavy protection therefore appears to be more effective than partial or no protection, many populations declined significantly over roughly a decade even within National Parks and Game Reserves. These findings beg the question of what criteria should be used to assess whether protected areas are successfully conserving species. Indeed, this dissertation shows that comparisons between protection strategies based on three different metrics (wildlife densities, numbers of significant changes in densities over time, and magnitude of changes in densities over time) lead to different conclusions regarding whether protection is "working" in Tanzania.; The effectiveness of heavily protected areas can be compromised when wildlife wander beyond their borders and poachers wander inside. In Tanzania, the size (and variability) of herbivore populations was more closely linked with the amount of heavily protected land in a reserve than high human densities surrounding its borders. Herbivore populations did not show clear "edge effects" within protected areas; even those species declining within protected areas did not tend to fare worst close to the border. This study employs long-term monitoring data to evaluate both the success of protected areas and the factors influencing their effectiveness.