Jacques Christopher N, Jenks Jonathan A, Grovenburg Troy W, Klaver Robert W
Department of Natural Resource Management, South Dakota State University, Brookings, South Dakota, United States of America.
U. S. Geological Survey, Iowa Cooperative Fish and Wildlife Research Unit, Iowa State University, Ames, Iowa, United States of America.
PLoS One. 2015 Dec 2;10(12):e0144026. doi: 10.1371/journal.pone.0144026. eCollection 2015.
Increased understanding of the influence of habitat (e.g., composition, patch size) and intrinsic (e.g., age, birth mass) factors on survival of neonatal pronghorn (Antilocapra americana) is a prerequisite to successful management programs, particularly as they relate to population dynamics and the role of population models in adaptive species management. Nevertheless, few studies have presented empirical data quantifying the influence of habitat variables on survival of neonatal pronghorn. During 2002-2005, we captured and radiocollared 116 neonates across two sites in western South Dakota. We documented 31 deaths during our study, of which coyote (Canis latrans) predation (n = 15) was the leading cause of mortality. We used known fate analysis in Program MARK to investigate the influence of intrinsic and habitat variables on neonatal survival. We generated a priori models that we grouped into habitat and intrinsic effects. The highest-ranking model indicated that neonate mortality was best explained by site, percent grassland, and open water habitat; 90-day survival (0.80; 90% CI = 0.71-0.88) declined 23% when grassland and water increased from 80.1 to 92.3% and 0.36 to 0.40%, respectively, across 50% natal home ranges. Further, our results indicated that grassland patch size and shrub density were important predictors of neonate survival; neonate survival declined 17% when shrub density declined from 5.0 to 2.5 patches per 100 ha. Excluding the site covariates, intrinsic factors (i.e., sex, age, birth mass, year, parturition date) were not important predictors of survival of neonatal pronghorns. Further, neonatal survival may depend on available land cover and interspersion of habitats. We have demonstrated that maintaining minimum and maximum thresholds for habitat factors (e.g., percentages of grassland and open water patches, density of shrub patches) throughout natal home ranges will in turn, ensure relatively high (>0.50) neonatal survival rates, especially as they relate to coyote predation. Thus, landscape level variables (particularly percentages of open water, grassland habitats, and shrub density) should be incorporated into the development or implementation of pronghorn management plans across sagebrush steppe communities of the western Dakotas, and potentially elsewhere within the geographic range of pronghorn.
深入了解栖息地（如组成、斑块大小）和内在因素（如年龄、出生体重）对新生叉角羚（Antilocapra americana）存活的影响，是成功开展管理计划的前提条件，特别是在涉及种群动态以及种群模型在适应性物种管理中的作用方面。然而，很少有研究提供量化栖息地变量对新生叉角羚存活影响的实证数据。在2002年至2005年期间，我们在南达科他州西部的两个地点捕获了116只新生叉角羚并为其佩戴了无线电项圈。在我们的研究过程中记录了31只叉角羚的死亡情况，其中郊狼（Canis latrans）捕食（n = 15）是主要死因。我们使用Program MARK中的已知命运分析来研究内在因素和栖息地变量对新生叉角羚存活的影响。我们生成了先验模型，并将其分为栖息地效应和内在效应。排名最高的模型表明，新生叉角羚的死亡率可以通过地点、草地百分比和开阔水域栖息地得到最佳解释；当草地和水域在50%的出生地家域范围内分别从80.1%增加到92.3%以及从0.36%增加到0.40%时，90天存活率（0.80；90%置信区间 = 0.71 - 0.88）下降了23%。此外，我们的结果表明草地斑块大小和灌木密度是新生叉角羚存活的重要预测因素；当灌木密度从每100公顷5.0个斑块下降到2.5个斑块时，新生叉角羚的存活率下降了17%。排除地点协变量后，内在因素（即性别、年龄、出生体重、年份、分娩日期）并非新生叉角羚存活的重要预测因素。此外，新生叉角羚的存活可能取决于可用的土地覆盖和栖息地的散布情况。我们已经证明，在整个出生地家域范围内维持栖息地因素的最小和最大阈值（如草地和开阔水域斑块的百分比、灌木斑块的密度），反过来将确保相对较高（>0.50）的新生叉角羚存活率，特别是在与郊狼捕食相关的方面。因此，景观水平变量（特别是开阔水域、草地栖息地的百分比以及灌木密度）应纳入南达科他州西部鼠尾草草原群落以及叉角羚地理分布范围内其他可能地区的叉角羚管理计划的制定或实施过程中。
