Kushnir, Hadas2010-02-252010-02-252009-12https://hdl.handle.net/11299/58493University of Minnesota Ph.D. dissertation. December 2009. Major: Conservation Biology. Advisors: Dr.Craig Packer & Dr. Stephen Polasky. 1 computer file (PDF); xi, 108, appendices 1-2.Lions attacked over 1000 people in Tanzania between 1990 and 2007, killing at least two-thirds of the victims. This extreme form of human-wildlife conflict has a major impact on the lives and livelihoods of local communities and threatens lion conservation in Tanzania, home to the largest lion population in Africa. Working in the two districts with the highest number of lion attacks, Rufiji and Lindi, my research examines the problem from both ecological and human perspectives at multiple scales. Overall, I aimed to: (1) identify human, ecological, and landscape-level risk factors for lion attacks, (2) determine how people currently react to attacks and what methods they believe could help mitigate attacks, and (3) understand how people perceive attacks and how these perceptions align with reality. Chapter 1, "Using Landscape Characteristics to Predict Risk of Lion Attacks in Southeastern Tanzania," examines the problem at the attack level across both districts. Using knowledge of attack locations, land cover, and important landscape features, I was able to model attack probability and then map the modeled probability in Rufiji and Lindi districts. I also extended the model to other areas in southeastern Tanzania to determine how well the model predicts high-risk areas beyond the study districts. Such a technique has potential to predict high-risk areas for future conflict in order to pinpoint prevention efforts. Chapter 2, "Human and Ecological Risk Factors for Unprovoked Lion Attacks on Humans in Southeastern Tanzania," compares human activity patterns during attacks between the two districts and examines risk at the village level in the areas with the highest concentration of attacks in Rufiji and Lindi districts. Human activity patterns during attacks differ significantly between the two districts and in each district they match with the details of daily life the area. By comparing villages with attacks to neighboring villages without attacks, I was able to identify a number of important risk factors related to wildlife presence and daily activities. Additionally, I examined the local response to lion attacks and views on appropriate measures to prevent attacks. Knowledge about local risk factors and response to attacks, and local views on prevention measures are all critical components of formulating methods to prevent future attacks. Chapter 3, "Reality vs. Perception: How Rural Tanzanians View Risks from Man-Eating Lions," examines human-lion conflict at the level of the individual by determining how people perceive the risk of lion attacks and how well these perceptions match reality. My findings indicate that even though people tend to exaggerate their overall risk, they correctly perceive specifics related to risk. This supports the need for using multiple methodologies to assess risk perceptions because only determining overall perceptions limits findings and under-represents local knowledge. The three chapters each provide different yet important perspectives on the problem that will be useful in formulating and implementing methods to reduce lion attacks on people in southeastern Tanzania. The unique combination of methodologies and scales of investigation also provide a useful framework for studies that investigate human-wildlife conflict worldwide.en-USConflict preventionHuman-wildlifeLionsMan-eatingRisk perceptionsTanzaniaConservation BiologyThesis or Dissertation