Senckenberg Biodiversity and Climate Research Centre, Frankfurt am Main, Germany.
DOF-BirdLife Denmark, Copenhagen V, Denmark.
J Anim Ecol. 2018 Jul;87(4):1034-1045. doi: 10.1111/1365-2656.12829. Epub 2018 Apr 23.
The effects of different environmental drivers on the changes in species' population abundances can be difficult to disentangle as they often act simultaneously. Researchers have built statistical models that include environmental variables (such as annual temperature) or species attributes (such as a species' temperature preference), which are assumed to detect the impacts of specific drivers (such as climate change). However, these approaches are often applied separately or, if combined, not explicitly compared. We show the complementary insights gained by applying both these approaches to a community dataset on Danish terrestrial birds. We use our analysis to compare the relative importance of climate change and agricultural land-use change for the abundance changes within the community between 1983 and 2013. Population models were fitted to the community data of species' annual abundances with predictors comprising: species attributes (species' temperature and habitat preferences), environmental variables (climatic and agricultural land-use change variables) or both. Relationships between species' abundances and environmental variables were used to identify the drivers associated with average abundance changes of species in the community. Relationships between species' abundances and their attributes were used to understand the drivers causing interspecific variation in abundance changes. Warmer winters were positively associated with community-level abundances, and warm-adapted species had more positive abundance changes than cold-adapted ones. Agricultural land-use area was negatively associated with community-level abundances, and birds using a high proportion of meadow and habitat specialists had more negative abundance changes than birds using other habitats and habitat generalists. Effect sizes of environmental variables were larger for agricultural land-use change while those of species attributes were larger for climate change. The environmental data approach suggested that agricultural land-use change has decreased the average abundances of species in the community, affecting total community size while the species attribute-based approach suggested that climate change has caused variation in abundance among species, affecting community composition. Environmental variables and species attributes that are hypothesized to link to specific drivers can be used together to provide complementary information on the impacts of different drivers on communities.
不同环境驱动因素对物种种群数量变化的影响可能难以区分，因为它们通常同时作用。研究人员构建了包含环境变量（如年温度）或物种属性（如物种对温度的偏好）的统计模型，这些变量被认为可以检测特定驱动因素（如气候变化）的影响。然而，这些方法通常是单独应用的，或者如果结合使用，也没有明确比较。我们展示了通过将这两种方法应用于丹麦陆地鸟类的群落数据集来获得互补的见解。我们使用我们的分析来比较气候变化和农业土地利用变化对 1983 年至 2013 年间群落内物种数量变化的相对重要性。使用物种年丰度的群落数据拟合种群模型，预测因子包括：物种属性（物种对温度和栖息地的偏好）、环境变量（气候和农业土地利用变化变量）或两者兼有。物种丰度与环境变量之间的关系用于确定与群落中物种平均丰度变化相关的驱动因素。物种丰度与其属性之间的关系用于了解导致物种丰度变化的种间差异的驱动因素。较暖的冬季与群落水平的丰度呈正相关，而适应温暖气候的物种的丰度变化比适应寒冷气候的物种更为积极。农业土地利用面积与群落水平的丰度呈负相关，而使用高比例草地和栖息地专家的鸟类的丰度变化比使用其他栖息地和栖息地一般的鸟类更为消极。环境变量的效应大小对农业土地利用变化较大，而物种属性的效应大小对气候变化较大。环境数据方法表明，农业土地利用变化减少了群落中物种的平均丰度，影响了群落的总体规模，而基于物种属性的方法表明，气候变化导致了物种之间丰度的变化，影响了群落的组成。假设与特定驱动因素相关的环境变量和物种属性可以一起使用，以提供有关不同驱动因素对群落影响的互补信息。
