Current animal tracking studies are most often based on the application of external geolocators such as GPS and radio transmitters. While these technologies provide detailed movement data, they are costly to acquire and maintain, which often restricts sample sizes. Furthermore, deploying external geolocators requires physically capturing and recapturing of animals, which poses an additional welfare concern. Natural biomarkers provide an alternative, non-invasive approach for addressing a range of geolocation questions and can, because of relatively low cost, be collected from many individuals thereby broadening the scope for population-wide inference. We developed a low-cost, minimally invasive method for distinguishing between local versus non-local movements of cattle using sulfur isotope ratios (δ34S) in cattle tail hair collected in the Greater Serengeti Ecosystem, Tanzania. We used a Generalized Additive Model to generate a predicted δ34S isoscape across the study area. This isoscape was constructed using spatial smoothers and underpinned by the positive relationship between δ34S values and lithology. We then established a strong relationship between δ34S from recent sections of cattle tail hair and the δ34S from grasses sampled in the immediate vicinity of an individual’s location, suggesting δ34S in the hair reflects the δ34S in the environment. By combining uncertainty in estimation of the isoscape, with predictions of tail hair δ34S given an animal’s position in the isoscape we estimated the anisotropic distribution of travel distances across the Serengeti ecosystem sufficient to detect movement using sulfur stable isotopes. While the focus of our study was on cattle, this approach can be modified to understand movements in other mobile organisms where the sulfur isoscape is sufficiently heterogeneous relative to the spatial scale of animal movements and where tracking with traditional methods is difficult.

中文翻译：

---

使用尾毛中的生物标记跟踪动物运动：一种根据硫等景物进行动物地理定位的新方法。


当前的动物追踪研究通常基于外部地理定位器的应用，例如 GPS 和无线电发射器。虽然这些技术提供了详细的运动数据，但获取和维护成本高昂，这通常会限制样本大小。此外，部署外部地理定位器需要物理捕获和重新捕获动物，这会带来额外的福利问题。天然生物标志物提供了一种替代的、非侵入性的方法来解决一系列地理定位问题，并且由于成本相对较低，可以从许多个体中收集，从而扩大了全人群推断的范围。我们开发了一种低成本、微创方法，利用坦桑尼亚大塞伦盖蒂生态系统中收集的牛尾毛中的硫同位素比率 (δ34S) 来区分牛的本地和非本地移动。我们使用广义加性模型来生成整个研究区域的预测 δ34S 等景观。该等景观是使用空间平滑器构建的，并以 δ34S 值和岩性之间的正关系为基础。然后，我们在最近的牛尾毛切片中的 δ34S 与在个体位置附近采样的草中的 δ34S 之间建立了密切的关系，表明头发中的 δ34S 反映了环境中的 δ34S。通过将等景观估计的不确定性与给定动物在等景观中的位置的尾毛 δ34S 的预测相结合，我们估计了整个塞伦盖蒂生态系统的移动距离的各向异性分布，足以使用硫稳定同位素检测运动。 虽然我们研究的重点是牛，但可以修改这种方法来了解其他移动生物体的运动，其中硫等景观相对于动物运动的空间尺度具有足够的异质性，并且用传统方法进行跟踪很困难。