Snow Leopard Trust and Nature Conservation Foundation, 3076/5, IV Cross Gokulam Park, Mysore, 570002, Karnataka, India.
Oecologia. 2012 Jul;169(3):581-90. doi: 10.1007/s00442-011-2237-0. Epub 2012 Jan 10.
Mountain ungulates around the world have been threatened by illegal hunting, habitat modification, increased livestock grazing, disease and development. Mountain ungulates play an important functional role in grasslands as primary consumers and as prey for wild carnivores, and monitoring of their populations is important for conservation purposes. However, most of the several currently available methods of estimating wild ungulate abundance are either difficult to implement or too expensive for mountainous terrain. A rigorous method of sampling ungulate abundance in mountainous areas that can allow for some measure of sampling error is therefore much needed. To this end, we used a combination of field data and computer simulations to test the critical assumptions associated with double-observer technique based on capture-recapture theory. The technique was modified and adapted to estimate the populations of bharal (Pseudois nayaur) and ibex (Capra sibirica) at five different sites. Conducting the two double-observer surveys simultaneously led to underestimation of the population by 15%. We therefore recommend separating the surveys in space or time. The overall detection probability for the two observers was 0.74 and 0.79. Our surveys estimated mountain ungulate populations (± 95% confidence interval) of 735 (± 44), 580 (± 46), 509 (± 53), 184 (± 40) and 30 (± 14) individuals at the five sites, respectively. A detection probability of 0.75 was found to be sufficient to detect a change of 20% in populations of >420 individuals. Based on these results, we believe that this method is sufficiently precise for scientific and conservation purposes and therefore recommend the use of the double-observer approach (with the two surveys separated in time or space) for the estimation and monitoring of mountain ungulate populations.
世界各地的高山有蹄类动物受到非法狩猎、栖息地改变、增加牲畜放牧、疾病和发展的威胁。高山有蹄类动物作为初级消费者和野生食肉动物的猎物，在草原生态系统中发挥着重要的功能作用，监测其种群数量对于保护目的非常重要。然而，目前几种估算野生有蹄类动物数量的方法要么难以实施，要么对于山区地形来说过于昂贵。因此，非常需要一种在山区地区能够对有蹄类动物数量进行严格采样且可以在一定程度上测量采样误差的方法。为此，我们结合野外数据和计算机模拟，测试了基于捕获-再捕获理论的双观察员技术相关的关键假设。该技术经过修改和适应，用于估计五个不同地点的岩羊（Pseudois nayaur）和北山羊（Capra sibirica）的种群数量。同时进行两次双观察员调查会导致种群数量低估 15%。因此，我们建议在空间或时间上分开调查。两名观察员的总体检测概率为 0.74 和 0.79。我们的调查估计了五个地点的高山有蹄类动物种群数量（±95%置信区间）分别为 735（±44）、580（±46）、509（±53）、184（±40）和 30（±14）只。发现检测概率为 0.75 足以检测到>420 只个体的种群数量变化 20%。基于这些结果，我们认为该方法对于科学和保护目的足够精确，因此建议使用双观察员方法（两次调查在时间或空间上分开）来估算和监测高山有蹄类动物的种群。
