Utah Black Bear Multi-Site Capture-Recapture 2004 - 2011

See Pederson et al. 2012 (https://doi.org/10.2192/URSUS-D-10-00029.1) for specific methods about sampling design, data collection, processing, and genetic analysis. Resulting capture histories and trap locations were formatted for input into the oSCR package in R: We formatted the individually identified black bear capture-recapture data from each site into multi-session (3 - 6 years) capture histories that retained spatial information on individual (i) detections at hair collection corrals (j = 16, 25) across sampling occasions (k = 4) such that yijk ~ Bernoulli(pij). The detection model component defines probability of detection of an individual at a particular trap (pij) as a function of distance from the individual’s activity center (si) to that trap having location xj. We used the half-normal model pij = p0 * exp(-dist(xj,si)2/2σ2) where p0 is the baseline encounter probability, x is the location of trap j, s is the activity center of individual i, and σ is the spatial scale parameter, sigma, determining the rate of decrease in detection probability in regards to the distance between xj and si. We have uploaded a ReadMe document that details each column in the provided trap data frame (tdf) and encounter data frame (edf) .csv files.有关采样设计、数据采集、数据处理以及遗传分析的具体方法，请参见Pederson等人2012年的研究（https://doi.org/10.2192/URSUS-D-10-00029.1）。最终得到的捕获历史记录与陷阱位置数据，均按照R语言中oSCR包（oSCR package）的输入格式进行了整理：我们将各研究站点经个体识别的美洲黑熊捕获-再捕获(capture-recapture)数据，整理为多时段（3~6年）的捕获历史记录，保留了个体(i)在采样时段(k=4)内于毛发收集围栏(j=16、25)处被检测到的空间信息，满足yijk服从以pij为参数的伯努利(Bernoulli)分布。检测模型分量将个体在特定陷阱处的检测概率(pij)定义为个体活动中心(activity center，si)至该陷阱位置(xj)的距离的函数。我们采用半正态模型(half-normal model) $p_{ij} = p_0 cdot exp(- ext{dist}(x_j,s_i)^2/(2sigma^2))$，其中$p_0$为基线偶遇概率，$x_j$为陷阱j的空间位置，$s_i$为个体i的活动中心，$sigma$为空间尺度参数（sigma），用于表征检测概率随$x_j$与$s_i$之间距离增大而下降的速率。我们已上传一份ReadMe文档，详细说明了本次提供的陷阱数据框（tdf）与偶遇数据框（edf）的.csv格式文件中各列的具体信息。