Centro de Biologia Ambiental, Faculdade de Ciências da Universidade de Lisboa, Lisboa, Portugal.
PLoS One. 2013 Sep 2;8(9):e73619. doi: 10.1371/journal.pone.0073619. eCollection 2013.
Widespread farmland abandonment has led to significant landscape transformations of many European mountain areas. These semi-natural multi-habitat landscapes are important reservoirs of biodiversity and their abandonment has important conservation implications. In multi-habitat landscapes the adaptation of communities depends on the differential affinity of the species to the available habitats. We use nested species-area relationships (SAR) to model species richness patterns of bird communities across scales in a mountain landscape, in NW Portugal. We compare the performance of the classic-SAR and the countryside-SAR (i.e. multi-habitat) models at the landscape scale, and compare species similarity decay (SSD) at the regional scale. We find a considerable overlap of bird communities in the different land-uses (farmland, shrubland and oak forest) at the landscape scale. Analysis of the classic and countryside SAR show that specialist species are strongly related to their favourite habitat. Farmland and shrubland have higher regional SSD compared to oak forests. However, this is due to the opportunistic use of farmlands by generalist birds. Forest specialists display significant regional turnover in oak forest. Overall, the countryside-SAR model had a better fit to the data showing that habitat composition determines species richness across scales. Finally, we use the countryside-SAR model to forecast bird diversity under four scenarios of land-use change. Farmland abandonment scenarios show little impact on bird diversity as the model predicts that the complete loss of farmland is less dramatic, in terms of species diversity loss, than the disappearance of native Galicio-Portuguese oak forest. The affinities of species to non-preferred habitats suggest that bird communities can adapt to land-use changes derived from farmland abandonment. Based on model predictions we argue that rewilding may be a suitable management option for many European mountain areas.
农田的广泛废弃导致许多欧洲山区的景观发生了重大变化。这些半自然的多生境景观是生物多样性的重要储存库，其废弃对保护具有重要意义。在多生境景观中，群落的适应取决于物种对可用生境的不同亲和力。我们使用嵌套物种面积关系 (SAR) 来模拟葡萄牙西北部山区景观中鸟类群落的跨尺度物种丰富度模式。我们在景观尺度上比较了经典 SAR 和乡村 SAR（即多生境）模型的性能，并比较了区域尺度上的物种相似性衰减 (SSD)。我们发现不同土地利用方式（农田、灌丛和橡树林）的鸟类群落有很大的重叠。经典 SAR 和乡村 SAR 的分析表明，特化种与它们最喜欢的栖息地密切相关。农田和灌丛的区域 SSD 高于橡树林。然而，这是由于机会主义地利用农田的一般性鸟类造成的。森林特化种在橡树林中表现出显著的区域更替。总的来说，乡村 SAR 模型对数据的拟合度更好，表明生境组成决定了跨尺度的物种丰富度。最后，我们使用乡村 SAR 模型预测了四种土地利用变化情景下的鸟类多样性。农田废弃情景对鸟类多样性的影响很小，因为模型预测完全失去农田在物种多样性损失方面的影响，不如原生加利西亚-葡萄牙橡木林的消失那么严重。物种对非首选生境的亲和力表明，鸟类群落可以适应由农田废弃引起的土地利用变化。基于模型预测，我们认为重新引入野生动物可能是许多欧洲山区的一种合适的管理选择。
