Previous Articles     Next Articles

A comparison of two home range core estimators for Tibetan fox (Vulpes ferrilata)

LIU Qunxiu,WANG Zhenghuan,WANG Xiaoming   

  1. School of Life Science,East China Normal University,Shanghai 200062,China
  • Online:2009-03-08 Published:2009-03-08

两种核域估算方法在野生藏狐家域研究中的比较

刘群秀, 王正寰, 王小明   

  1. (华东师范大学生命科学学院,上海 200062

Abstract: Home range core has been defined as the area used intensively by an animal and which contains critical resources for living,such as cover,food,and if applicable,a denning or roosting location. The identification of home range core has received considerable interest among researchers,but it has proven difficult to come to consensus on the best way to quantify it in an objective and accurate way that yields biologically significant information. We documented 199 locations of 3 Tibetan foxes (Vulpes ferrilata) in Gouli Township,Dulan County,Qinghai Province during April 2006,September-October 2006,and January - April 2007. Using these locations,we estimated home range cores using both fixed kernel and harmonic mean methods based on the home range core identification principles of Seaman and Powell (1990). We compared these 2 methods to determine the better one for home range core estimation. Home range core size was influenced by total home range in both methods. Home range core estimated using the fixed kernel method was more sensitive to total
home range area than that using the harmonic mean method,and variable coefficients varied greatly. Home range core size using the fixed kernel method was negatively correlated with sample size,whereas that using the harmonic mean was positively correlated with sample size. Area not actually in the animal’s home range was more easily included in the harmonic. However,to some extent this drawback can be eliminated when home range core was estimated. The harmonic mean method reflected the center of activity more precisely,and more locations of foxes were included in the home range core estimated. For our Tibetan foxes,fixed kernel home ranges showed some more consistency with varying sample size,but harmonic mean was suggested to be the better home range core estimator.

Key words: Fixed kernel, Home range core, Home range, Harmonic mean

摘要: 核域是野生动物重点利用、含有更多居所、隐蔽场所及依赖性食物资源的区域。关于核域的确定和划分一直存在争议,且很难找到客观的并能准确展示动物生物学信息的数理统计方法。2006 年4 月,2006 年9 ~10月和2007 年1 ~4 月,我们共记录青海省都兰县沟里乡3 只藏狐的199 个活动位点,通过固定核空间(Fixed kernel)和调和平均值(Harmonic mean)模型,结合独立区域法的计算原理,估算藏狐的核域,并对两种估算方法进行比较分析,以确定更理想的核域数学模型。研究结果表明:(1)两种估算方法中核域面积均受到家域总面积的影响,固定核空间法的核域面积受家域总面积影响较大,而且变量系数不稳定; (2) 随着样本量的增加,固定核空间法估算的核域面积逐渐减小,而调和平均值法则增大,前者从数理统计上更好地描述了藏狐对家域资源的利用分布,后者更符合客观实际;(3)调和平均值法在划定家域边界时容易包括不属于家域范围的区域,而在估算核域时可以在一定程度上克服该缺陷;(4)调和平均值法能够真实地反应动物的活动中心,估算的核域中包含有更多的活动位点。因此,尽管在估算藏狐核域时固定核空间法有稳定的计算结果等优点,调和平均值法为更理想的核域估算模型。

关键词: 核域, 家域, 固定核空间, 调和平均值