Data from: Should scientists be required to use a model-based solution to adjust for possible distance-based detectability bias?

The most popular method used to gain an understanding of population trends or of differences in bird abundance among land condition categories is to use information derived from point counts. Unfortunately, various factors can affect one’s ability to detect birds, and those factors need to be controlled or accounted for so that any difference in one’s index among time periods or locations is an accurate reflection of differences in bird abundance and not differences in detectability. Avian ecologists could use appropriately sized fixed-area surveys to minimize the chance that they might be deceived by distance-based detectability bias, but the current method of choice is to use a modeling approach that allows one to account for distance-based bias by modeling the effects of distance on detectability or occupancy. I challenge the idea that modeling is the best approach to account for distance-based effects on the detectability of birds because the most important distance-based modeling assumptions can never be met. The use of a fixed-area survey method to generate an index of abundance is the simplest way to control for distance-based detectability bias and should not be universally condemned or be the basis for outright rejection in the publication process.

当前用于解析种群动态或不同土地状况类别间鸟类丰富度差异的主流方法，是依托点计数法（point counts）获取相关观测信息。遗憾的是，诸多环境因素会干扰鸟类的可检测性，因此必须对这些干扰因素加以控制或纳入分析考量，唯有如此，不同时段或调查点位间得到的种群指数差异，才能准确反映鸟类丰富度的真实变化，而非由可检测性差异所导致的假象。鸟类生态学家本可采用规模适配的固定面积调查法，尽可能规避因基于距离的可检测性偏差而产生的观测误判，但当前学界首选的解决方案却是借助建模手段：通过模拟距离对鸟类可检测性或占用率（occupancy）的影响，来抵消距离带来的检测偏差。笔者对此提出质疑：建模法并非解决距离相关可检测性偏差的最优方案，因为此类距离建模的核心前提假设往往永远无法得到满足。采用固定面积调查法生成鸟类丰富度指数，是控制基于距离的可检测性偏差最为简便有效的途径，该方法不应在学术发表流程中被全盘否定，亦不应成为直接拒稿的依据。