Biology Department, University of New Brunswick at Saint John, P.O. Box 5050, Saint John, New Brunswick, E2L 4L5, Canada.
Ottawa Carleton Institute of Biology, University Ottawa, Ottawa, Ontario, K1N 6N5, Canada.
Ecology. 2018 Nov;99(11):2592-2604. doi: 10.1002/ecy.2514. Epub 2018 Oct 12.
Effects of species diversity on population and community stability (or more precisely, the effects of species richness on temporal variability) have been studied for several decades, but there have been no large-scale tests in natural communities of predictions from theory. We used 91 data sets including plants, fish, small mammals, zooplankton, birds, and insects, to examine the relationship between species richness and temporal variability in populations and communities. Seventy-eight of 91 data sets showed a negative relationship between species richness and population variability; 46 of these relationships were statistically significant. Only five of the 13 positive richness-population variability relationships were statistically significant. Similarly, 51 of 91 data sets showed a negative relationship between species richness and community variability; of these, 26 were statistically significant. Seven of the 40 positive richness-community-variability relationships were statistically significant. We were able to test transferability (i.e., the predictive ability of models for sites that are spatially distinct from sites that were used to build the models) for 69 of 91 data sets; 35 and 31 data sets were transferable at the population and community levels, respectively. Only four were positive at the population level, and two at the community level. We conclude that there is compelling evidence of a negative relationship between species richness and temporal variability for about one-half of the ecological communities we examined. However, species richness explained relatively little of the variability in population or community abundances and resulted in small improvements in predictive ability.
几十年来，人们一直在研究物种多样性对种群和群落稳定性的影响（或者更准确地说，是物种丰富度对时间变异性的影响），但在自然群落中，还没有对理论预测进行大规模测试。我们使用了包括植物、鱼类、小型哺乳动物、浮游动物、鸟类和昆虫在内的 91 个数据集，来检验物种丰富度与种群和群落时间变异性之间的关系。在 91 个数据集中，有 78 个数据集显示物种丰富度与种群变异性之间呈负相关关系；其中 46 个关系具有统计学意义。在 13 个正相关的丰富度-种群变异性关系中，只有 5 个具有统计学意义。同样，在 91 个数据集中，有 51 个数据集显示物种丰富度与群落变异性之间呈负相关关系；其中 26 个关系具有统计学意义。在 40 个正相关的丰富度-群落变异性关系中，有 7 个具有统计学意义。我们能够对 91 个数据集中的 69 个数据集进行可转移性（即模型对空间上与建立模型的地点不同的地点的预测能力）的检验；分别有 35 个和 31 个数据集在种群和群落水平上是可转移的。在种群水平上仅有 4 个是正相关的，在群落水平上仅有 2 个是正相关的。我们的结论是，在我们研究的大约一半生态群落中，物种丰富度与时间变异性之间存在着很强的负相关关系。然而，物种丰富度对种群或群落丰度的变异性解释相对较少，并且仅能略微提高预测能力。
