Department of Biology, Faculty of Science, Memorial University of Newfoundland, St. John's, NL, Canada.
Glob Chang Biol. 2018 Aug;24(8):3849-3861. doi: 10.1111/gcb.14169. Epub 2018 May 4.
Land-cover and climate change are two main drivers of changes in species ranges. Yet, the majority of studies investigating the impacts of global change on biodiversity focus on one global change driver and usually use simulations to project biodiversity responses to future conditions. We conduct an empirical test of the relative and combined effects of land-cover and climate change on species occurrence changes. Specifically, we examine whether observed local colonization and extinctions of North American birds between 1981-1985 and 2001-2005 are correlated with land-cover and climate change and whether bird life history and ecological traits explain interspecific variation in observed occurrence changes. We fit logistic regression models to test the impact of physical land-cover change, changes in net primary productivity, winter precipitation, mean summer temperature, and mean winter temperature on the probability of Ontario breeding bird local colonization and extinction. Models with climate change, land-cover change, and the combination of these two drivers were the top ranked models of local colonization for 30%, 27%, and 29% of species, respectively. Conversely, models with climate change, land-cover change, and the combination of these two drivers were the top ranked models of local extinction for 61%, 7%, and 9% of species, respectively. The quantitative impacts of land-cover and climate change variables also vary among bird species. We then fit linear regression models to test whether the variation in regional colonization and extinction rate could be explained by mean body mass, migratory strategy, and habitat preference of birds. Overall, species traits were weakly correlated with heterogeneity in species occurrence changes. We provide empirical evidence showing that land-cover change, climate change, and the combination of multiple global change drivers can differentially explain observed species local colonization and extinction.
土地覆盖和气候变化是物种分布变化的两个主要驱动因素。然而，大多数研究全球变化对生物多样性的影响的研究都集中在一个全球变化驱动因素上，通常使用模拟来预测未来条件下生物多样性的反应。我们对土地覆盖和气候变化对物种出现变化的相对和综合影响进行了实证检验。具体来说，我们研究了 1981-1985 年至 2001-2005 年间北美鸟类的局部繁殖地的定居和灭绝是否与土地覆盖和气候变化有关，以及鸟类的生活史和生态特征是否解释了观察到的物种出现变化的种间变异。我们拟合逻辑回归模型来检验物理土地覆盖变化、净初级生产力变化、冬季降水、夏季平均温度和冬季平均温度对安大略省繁殖鸟类局部定居和灭绝的概率的影响。对于 30%、27%和 29%的物种，气候变化、土地覆盖变化以及这两个驱动因素的组合模型分别是局部繁殖地定居的最佳模型。相反，对于 61%、7%和 9%的物种，气候变化、土地覆盖变化以及这两个驱动因素的组合模型分别是局部灭绝的最佳模型。土地覆盖和气候变化变量的定量影响在鸟类物种之间也有所不同。然后，我们拟合线性回归模型来检验鸟类的平均体重、迁徙策略和栖息地偏好是否可以解释区域定居和灭绝率的变化。总体而言，物种特征与物种出现变化的异质性相关性较弱。我们提供了实证证据，表明土地覆盖变化、气候变化以及多个全球变化驱动因素的组合可以不同程度地解释观察到的物种局部繁殖地的定居和灭绝。
