作者: John F. Benson
关键词: Sampling (statistics) 、 Biology 、 Scale (map) 、 Habitat 、 Statistics 、 Euclidean distance 、 Sensitivity (control systems) 、 Range (statistics) 、 Selection (genetic algorithm) 、 Sample size determination
摘要: Summary Animal habitat selection analyses often rely on comparisons of use and availability to infer selection. Random locations are commonly used assess despite inefficiency potential uncertainty associated with random sampling. Herein, I propose a systematic approach estimate reduce sampling error computing time GIS-based estimation using locations. I Euclidean distance analysis (EDA) as model technique demonstrate the sensitivity use-availability insufficient evaluate proposed approach. re-analysed data from previous study Florida panthers (Puma concolor coryi) compared results in which distance-based (i.e. expected distance) was estimated range sample sizes locations, also systematically. My that distances statistical EDA based can be unreliable low arbitrary numbers points, vary if increasing points used, obtained systematically at greater sufficient sampling). The efficiently measures by making calculations all possible specified resolution, across scale interest. Thus, it eliminates due is considerably faster. The improves rigour efficiency animal ensures repeatability for practical theoretical applications.