A simple method for estimating the effective detection distance of camera traps
A simple method for estimating the effective detection distance of camera traps
复制标题
DOI：
10.1002/rse2.25
复制
发表时间：
2017-06-01
影响因子：
5.5
通讯作者：
Jansen, Patrick A.
中科院分区：
文献类型：
--
作者：
Hofmeester, Tim R.;Rowcliffe, J. Marcus;Jansen, Patrick A.
动物丰度的估计对于了解动物生态学至关重要。相机陷阱可用于估计陆地哺乳动物的丰度，包括难以捉摸的物种，前提是传感器的灵敏度（估计为有效检测距离（EDD））是量化的。在这里，我们展示了如何通过沿着相机视野中线在已知距离处放置标记，然后将距离采样函数拟合到动物在标记之间通过的频率，来直接从相机陷阱图像推断 EDD。如果在最大检测距离上使用至少五个间隔，则使用分箱检测距离从模拟通道得出的 EDD 估计值近似于从连续检测距离测量中获得的估计值。该方法在具有对比植被密度的两种森林类型中进行了实地测试，并以 2.5 m 的间隔设置了 5 个标记，得出了 13 种森林哺乳动物的可信 EDD 估计值。正如预期的那样，EDD 估计值与物种体重呈正相关，并且植被较密集的 EDD 值较短。我们的研究结果表明，这种简单的方法可以产生可靠的 EDD 估计。这些估计可用于纠正由于物种和栖息地之间传感器灵敏度差异而导致的采样工作差异的摄影捕获率。简化 EDD 的估计将导致相对丰度指数的偏差较小，并且还将有利于使用相机陷阱数据来估计动物密度。
Estimates of animal abundance are essential for understanding animal ecology. Camera traps can be used to estimate the abundance of terrestrial mammals, including elusive species, provided that the sensitivity of the sensor, estimated as the effective detection distance (EDD), is quantified. Here, we show how the EDD can be inferred directly from camera trap images by placing markers at known distances along the midline of the camera field of view, and then fitting distance-sampling functions to the frequency of animal passage between markers. EDD estimates derived from simulated passages using binned detection distances approximated those obtained from continuous detection distance measurements if at least five intervals were used over the maximum detection distance. A field test of the method in two forest types with contrasting vegetation density, with five markers at 2.5 m intervals, produced credible EDD estimates for 13 forest-dwelling mammals. EDD estimates were positively correlated with species body mass, and were shorter for the denser vegetation, as expected. Our findings suggest that this simple method can produce reliable estimates of EDD. These estimates can be used to correct photographic capture rates for difference in sampling effort resulting from differences in sensor sensitivity between species and habitats. Simplifying the estimation of EDD will result in less biased indices of relative abundance, and will also facilitate the use of camera trap data for estimating animal density.