Global gaps in trait data for terrestrial vertebrates
Global gaps in trait data for terrestrial vertebrates
复制标题
DOI：
10.1111/geb.13184
复制
发表时间：
2020-10-11
影响因子：
6.4
通讯作者：
Newbold, Tim
中科院分区：
文献类型：
--
作者：
Etard, Adrienne;Morrill, Sophie;Newbold, Tim
目的性状数据越来越多地被用于研究全球变化对生态群落结构和功能的影响。尽管陆地脊椎动物的性状数据整理数量越来越多，但迄今为止还没有对数据存在的差距和偏差进行全球评估。在这里，我们评估陆生脊椎动物性状数据是否在分类学上，空间和遗传学上有偏见。位置全球。存在时间段。主要分类群研究陆生脊椎动物。方法我们编制了7个生态特征和量化的覆盖率的比例，估计是可用的。对于一个物种，我们将完整性定义为各个特征中非缺失值的比例。我们评估覆盖率和完整性是否在不同类别之间存在差异，并检查性状数据中的系统发育偏差。为了研究空间偏差，我们测试了范围较广的物种是否比范围较窄的物种具有更完整的性状数据。此外，我们测试是否物种丰富的地区，这是最关心的保护，采样比物种贫乏的地区。结果哺乳动物和鸟类即使在物种丰富的地区也能很好地采样。对于爬行动物和两栖动物（爬虫类），除了与栖息地有关的变量（两栖动物）之外，只有身体大小呈现高覆盖率（>80%）。爬虫类动物的其他特征取样很少。在一些物种丰富的地区和某些分支中，这种短缺特别严重。在所有类别中，地理上较罕见的物种具有较不完整的性状信息。主要结论性状信息是平均在一些最多样化的地区和地理上罕见的物种，这两个生物多样性保护的关键。特质数据的差距可能会阻碍我们进行大规模分析的能力，而偏见可能会影响外推的有效性。解决这个问题的短期办法是使用插补技术估计缺失的性状数据，而长期和更有力地填补现有差距需要持续的数据收集工作。
Aim Trait data are increasingly being used in studies investigating the impacts of global changes on the structure and functioning of ecological communities. Despite a growing number of trait data collations for terrestrial vertebrates, there is to date no global assessment of the gaps and biases the data present. Here, we assess whether terrestrial vertebrate trait data are taxonomically, spatially and phylogenetically biased. Location Global. Time period Present. Major taxa studied Terrestrial vertebrates. Methods We compile seven ecological traits and quantify coverage as the proportion of species for which an estimate is available. For a species, we define completeness as the proportion of non-missing values across traits. We assess whether coverage and completeness differ across classes and examine phylogenetic biases in trait data. To investigate spatial biases, we test whether wider-ranging species have more complete trait data than narrow-ranging species. Additionally, we test whether species-rich regions, which are of most concern for conservation, are less well sampled than species-poor regions. Results Mammals and birds are well sampled even in species-rich regions. For reptiles and amphibians (herptiles), only body size presents a high coverage (>80%), in addition to habitat-related variables (amphibians). Herptiles are poorly sampled for other traits. The shortfalls are particularly acute in some species-rich regions and for certain clades. Across all classes, geographically rarer species have less complete trait information. Main conclusions Trait information is less available on average in some of the most diverse areas and in geographically rarer species, both of which crucial for biodiversity conservation. Gaps in trait data might impede our ability to conduct large-scale analyses, whereas biases can impact the validity of extrapolations. A short-term solution to the problem is to estimate missing trait data using imputation techniques, whereas a longer-term and more robust filling of existing gaps requires continued data-collection efforts.