资源环境科技发展态势分析平台

Anthropogenic mortality among populations of large terrestrial carnivores undermines the health of ecosystems globally, and generally increases when people respond lethally to real or perceived threats to property, including livestock. Reducing such threats through the use of non-lethal methods could therefore protect both large predators and human interests. However, the scarcity of information on the effectiveness of methods to prevent livestock predation hinders the formulation of science-based policy. We present the results of a randomized crossover experimental test of a method to prevent predation on livestock, which to our knowledge is the first such test in Latin America. By relying on a so-called "gold-standard" design, we evaluated the effectiveness of using flashing lights to deter predators. We found that light deterrents discouraged pumas (Puma concolor) but not Andean foxes (Lycalopex culpaeus) from preying on alpacas (Vicugna pacos) and llamas (Lama glama), and demonstrated that gold-standard experiments are feasible in large natural ecosystems, contradicting assumptions that people will reject placebo controls and that such systems contain too many confounding variables. Functionally effective non-lethal methods can protect wildlife, livestock, and people. Strong inference is needed for the development of sound policy concerning wildlife management, livestock husbandry, environmental conservation, and biodiversity.