Larsen Trond H, Lopera Alejandro, Forsyth Adrian
Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ 08544, USA.
Conserv Biol. 2008 Oct;22(5):1288-98. doi: 10.1111/j.1523-1739.2008.00969.x. Epub 2008 Jun 20.
Anthropogenic disturbances such as fragmentation are rapidly altering biodiversity, yet a lack of attention to species traits and abundance patterns has made the results of most studies difficult to generalize. We determined traits of extinction-prone species and present a novel strategy for classifying species according to their population-level response to a gradient of disturbance intensity. We examined the effects of forest fragmentation on dung beetle communities in an archipelago of 33 islands recently created by flooding in Venezuela. Species richness, density, and biomass all declined sharply with decreasing island area and increasing island isolation. Species richness was highly nested, indicating that local extinctions occurred nonrandomly. The most sensitive dung beetle species appeared to require at least 85 ha of forest, more than many large vertebrates. Extinction-prone species were either large-bodied, forest specialists, or uncommon. These explanatory variables were unrelated, suggesting at least 3 underlying causes of extirpation. Large species showed high wing loading (body mass/wing area) and a distinct flight strategy that may increase their area requirements. Although forest specificity made most species sensitive to fragmentation, a few persistent habitat generalists dispersed across the matrix. Density functions classified species into 4 response groups on the basis of their change in density with decreasing species richness. Sensitive and persistent species both declined with increasing fragmentation intensity, but persistent species occurred on more islands, which may be due to their higher baseline densities. Compensatory species increased in abundance following the initial loss of sensitive species, but rapidly declined with increasing fragmentation. Supertramp species (widespread habitat generalists) may be poor competitors but strong dispersers; their abundance peaked following the decline of the other 3 groups. Nevertheless, even the least sensitive species were extirpated or rare on the smallest and most isolated islands.
诸如碎片化之类的人为干扰正在迅速改变生物多样性，但由于对物种特征和丰度模式缺乏关注，大多数研究结果难以推广。我们确定了易灭绝物种的特征，并提出了一种根据物种对干扰强度梯度的种群水平响应进行分类的新策略。我们研究了森林碎片化对委内瑞拉近期因洪水形成的33个岛屿群岛上蜣螂群落的影响。随着岛屿面积减小和岛屿隔离度增加，物种丰富度、密度和生物量均急剧下降。物种丰富度高度嵌套，表明局部灭绝是非随机发生的。最敏感的蜣螂物种似乎需要至少85公顷的森林，这比许多大型脊椎动物所需的面积还大。易灭绝物种要么体型大、是森林专家物种，要么不常见。这些解释变量互不相关，表明至少有3个灭绝的潜在原因。大型物种表现出高翼载荷（体重/翼面积）和独特的飞行策略，这可能增加它们对面积的需求。尽管森林特异性使大多数物种对碎片化敏感，但仍有一些持久的栖息地通才物种散布在基质中。密度函数根据物种密度随物种丰富度降低的变化将物种分为4个响应组。敏感物种和持久物种都随着碎片化强度增加而减少，但持久物种出现在更多岛屿上，这可能是由于它们较高的基线密度。补偿性物种在敏感物种最初减少后数量增加，但随着碎片化加剧迅速下降。超级流浪者物种（广泛分布的栖息地通才物种）可能是竞争力差但扩散能力强的物种；它们的丰度在其他3组下降后达到峰值。然而，即使是最不敏感的物种在最小和最孤立的岛屿上也会灭绝或稀少。
