Land‐use change modifies the spatial structure of terrestrial landscapes, potentially shaping the distribution, abundance and diversity of remaining species assemblages. Non‐human primates can be particularly vulnerable to landscape disturbances, but our understanding of this topic is far from complete. Here we reviewed all available studies on primates' responses to landscape structure. We found 34 studies of 71 primate species (24 genera and 10 families) that used a landscape approach. Most studies (82%) were from Neotropical forests, with howler monkeys being the most frequently studied taxon (56% of studies). All studies but one used a site‐landscape or a patch‐landscape study design, and frequently (34% of studies) measured landscape variables within a given radius from the edge of focal patches. Altogether, the 34 studies reported 188 responses to 17 landscape‐scale metrics. However, the majority of the studies (62%) quantified landscape predictors within a single spatial scale, potentially missing significant primate–landscape responses. To assess such responses accurately, landscape metrics need to be measured at the optimal scale, i.e. the spatial extent at which the primate–landscape relationship is strongest (so‐called ‘scale of effect’). Only 21% of studies calculated the scale of effect through multiscale approaches. Interestingly, the vast majority of studies that do not assess the scale of effect mainly reported null effects of landscape structure on primates, while most of the studies based on optimal scales found significant responses. These significant responses were primarily to landscape composition variables rather than landscape configuration variables. In particular, primates generally show positive responses to increasing forest cover, landscape quality indices and matrix permeability. By contrast, primates show weak responses to landscape configuration. In addition, half of the studies showing significant responses to landscape configuration metrics did not control for the effect of forest cover. As configuration metrics are often correlated with forest cover, this means that documented configuration effects may simply be driven by landscape‐scale forest loss. Our findings suggest that forest loss (not fragmentation) is a major threat to primates, and thus, preventing deforestation (e.g. through creation of reserves) and increasing forest cover through restoration is critically needed to mitigate the impact of land‐use change on our closest relatives. Increasing matrix functionality can also be critical, for instance by promoting anthropogenic land covers that are similar to primates' habitat.

中文翻译：

---

灵长类动物对景观结构反应的全球评估

土地利用变化改变了陆地景观的空间结构，可能影响剩余物种组合的分布、丰度和多样性。非人类灵长类动物可能特别容易受到景观干扰的影响，但我们对这个话题的理解还远未完成。在这里，我们回顾了所有关于灵长类动物对景观结构反应的现有研究。我们发现了对 71 种灵长类动物（24 属和 10 科）使用景观方法的 34 项研究。大多数研究（82%）来自新热带森林，吼猴是最常研究的分类群（56%的研究）。除了一项研究外，所有研究都使用了场地景观或斑块景观研究设计，并且经常（34% 的研究）测量了距焦点斑块边缘给定半径内的景观变量。共，34 项研究报告了对 17 个景观尺度指标的 188 个响应。然而，大多数研究 (62%) 在单一空间尺度内量化了景观预测因子​​，可能遗漏了重要的灵长类动物景观反应。为了准确评估这些反应，景观指标需要在最佳尺度上进行测量，即灵长类动物与景观关系最强的空间范围（所谓的“效应尺度”）。只有 21% 的研究通过多尺度方法计算了效应尺度。有趣的是，绝大多数不评估效应规模的研究主要报告了景观结构对灵长类动物的无效影响，而大多数基于最佳规模的研究发现了显着的反应。这些重要的反应主要是对景观构成变量而不是景观配置变量。特别是，灵长类动物通常对增加森林覆盖率、景观质量指数和基质渗透性表现出积极的反应。相比之下，灵长类动物对景观配置的反应较弱。此外，显示对景观配置指标的显着响应的研究中有一半没有控制森林覆盖的影响。由于配置指标通常与森林覆盖率相关，这意味着记录的配置影响可能只是由景观规模的森林损失驱动的。我们的研究结果表明，森林丧失（而非破碎化）是灵长类动物的主要威胁，因此可以防止森林砍伐（例如 通过建立保护区）和通过恢复增加森林覆盖率是减轻土地利用变化对我们近亲的影响的迫切需要。增加矩阵功能也很重要，例如通过促进类似于灵长类动物栖息地的人为土地覆盖。