Blake John G, Loiselle Bette A
Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, FL, United States of America.
Department of Wildlife Ecology & Conservation and Center for Latin American Studies, University of Florida, Gainesville, FL, United States of America.
PeerJ. 2018 Jan 9;6:e4241. doi: 10.7717/peerj.4241. eCollection 2018.
Terrestrial mammals are important components of lowland forests in Amazonia (as seed dispersal agents, herbivores, predators) but there are relatively few detailed studies from areas that have not been affected by human activities (e.g., hunting, logging). Yet, such information is needed to evaluate effects of humans elsewhere. We used camera traps to sample medium to large-sized terrestrial mammals at a site in lowland forests of eastern Ecuador, one of the most biologically rich areas in the world. We deployed cameras on two study plots in forest at Tiputini Biodiversity Station. Sixteen cameras were arranged 200 m apart in a 4 × 4 grid on each plot. Cameras were operated for 60 days in January-March, 2014-2017, for a total of 3,707 and 3,482 trap-days on the two plots (Harpia, Puma). A total of 28 species were recorded; 26 on Harpia and 25 on Puma. Number of species recorded each year was slightly greater on Harpia whereas overall capture rates (images/100 trap-days) were higher on Puma. Although most species were recorded on each plot, differences in capture rates meant that yearly samples on a given plot were more similar to each other than to samples on the other plot. Images of most species showed a clumped distribution pattern on each plot; was the only species that did not show a clumped distribution on either plot. Images at a given camera location showed no evidence of autocorrelation with numbers of images at nearby camera locations, suggesting that species were responding to small-scale differences in habitat conditions. A redundancy analysis showed that environmental features within 50 or 100 m of camera locations (e.g., elevation, variation in elevation, slope, distance to streams) accounted for significant amounts of variation in distribution patterns of species. Composition and relative importance based on capture rates were very similar to results from cameras located along trails at the same site; similarities decreased at increasing spatial scales based on comparisons with results from other sites in Ecuador and Peru.
陆生哺乳动物是亚马孙低地森林的重要组成部分（作为种子传播者、食草动物、食肉动物），但在未受人类活动（如狩猎、伐木）影响的地区，相关的详细研究相对较少。然而，评估人类活动在其他地区的影响需要此类信息。我们利用相机陷阱对厄瓜多尔东部低地森林中的中大型陆生哺乳动物进行采样，该地区是世界上生物多样性最丰富的地区之一。我们在蒂普蒂尼生物多样性站的森林中的两个研究地块上部署了相机。每个地块上，16台相机以4×4的网格排列，彼此相距200米。相机在2014年至2017年的1月至3月运行60天，两个地块（哈皮亚、美洲狮地块）的总陷阱天数分别为3707天和3482天。共记录了28个物种；哈皮亚地块记录到26种，美洲狮地块记录到25种。每年在哈皮亚地块记录到的物种数量略多，而美洲狮地块的总体捕获率（图像数/100陷阱天）更高。尽管大多数物种在每个地块上都有记录，但捕获率的差异意味着给定地块上的年度样本彼此之间比与另一地块上的样本更相似。大多数物种的图像在每个地块上都呈现出聚集分布模式；是唯一一个在两个地块上都没有呈现聚集分布的物种。给定相机位置的图像与附近相机位置的图像数量之间没有自相关的证据，这表明物种对栖息地条件的小尺度差异做出了反应。冗余分析表明，相机位置50米或100米范围内的环境特征（如海拔、海拔变化、坡度、距溪流的距离）在物种分布模式的变化中占很大比例。基于捕获率的组成和相对重要性与同一地点沿小径设置的相机的结果非常相似；与厄瓜多尔和秘鲁其他地点的结果相比，随着空间尺度的增加，相似性降低。
