The coherence of ecological stability among ecosystems and across ecological scales

气候和环境变化正在对世界上所有的生态系统--海洋、淡水和陆地--产生影响。这些变化与广泛的“压力”有关，如气温上升、干旱、养分流失、杀虫剂、过度捕捞和栖息地丧失。例如，超过50%的提供饮用水、娱乐和食物的淡水水生群落在上个世纪受到有害的人为压力的威胁，包括气温上升、污染和N/P径流。在海洋社区，气候变化、酸化、过度捕捞和污染威胁着食物资源、沿海社区和碳储存能力。在陆地社区，气温上升、农业径流、栖息地碎片化、污染和更频繁的极端事件威胁到森林、草原和农业社区及其提供的服务。这一细节表明，人口、社区和生态系统功能的稳定性受到多种同时应激源的威胁。预测这些影响是很困难的。我们强调在推进对生态系统内和生态系统之间稳定性的理解方面面临的三个重大挑战。首先，有很多方法来衡量稳定性和动态性，它们在不同的生态尺度上发挥作用--一些是关于单个物种的，另一些是关于许多物种的多样性的。我们必须证明同时使用多种措施的价值，以及这些措施在不同生态尺度上的适当性。其次，我们用来评估生物体稳定性和动态性的数据--时间、空间和物种间的数量--具有必须考虑的特性，但很少被考虑。我们知道，一个物种今天和昨天的数量比今天和5年前的数量更相关。这是时间相关性。我们知道，住得近的人比住得远的人更相似。这是空间相关性。我们知道，亲缘关系密切的物种会比远亲物种更相似。这是进化上的关联。大多数稳定性指标未能考虑到真实数据的这些破坏性和众所周知的特征。第三，压力源之间可能存在交互作用。这意味着，一种压力源的效果取决于另一种压力源的存在、缺失或程度。这使得预测具有挑战性，特别是在我们不知道依赖关系的情况下。我们需要理论和数据来理解多重应激源如何以及在多大程度上影响稳定性。我们的项目旨在使用三个研究目标来处理所有这些问题。1.在多个生态尺度(如生物量/丰富度、群落结构/多样性、功能)制定改进的稳定性指标，正式纠正因空间、时间和系统发育共同变化而产生的长期认识但很少解决的问题；2.对生态系统之间的稳定性(一致性)作出预测，这些系统面临由高级模拟模型产生的多个同时应激源，这些模型允许深入了解推动稳定性的内在过程和反馈机制；和利用从陆地、海洋和淡水生态系统收集的实验和实地数据，测试模型预测和改进的稳定性指标的有效性。

Climate and environmental change is impacting on all ecosystems - marine, freshwater and terrestrial - in the world. The changes are associated with a wide range of 'stress' like rising temperature, drought, nutrient run-off, pesticides, over-harvesting and habitat loss. For example, more than 50% of freshwater aquatic communities that provide drinking water, recreation and food are threatened by detrimental anthropogenic stress in the last century, including rising temperature, pollution and N/P run-off. In marine communities, climate change, acidification, overfishing and pollution threaten food resources, coastal communities and carbon storage capacity. In terrestrial communities, rising temperatures, agricultural run-off, habitat fragmentation, pollution and more frequent extreme events threaten to forest, grassland and agricultural communities and the services they provide. This detail suggests that the stability of populations, communities and ecosystem function is threatened by multiple, simultaneous stressors. Making predictions about these effects is hard. We highlight three substantial challenges to advancing the understanding of stability within and among ecosystems. First, there are lots of ways to measure stability and dynamics and they work at different ecological scales - some are about individual species and others are about diversity of many species.. We must demonstrate the value of using simultaneously multiple measures and the appropriateness of them at different ecological scales. Second, the kind of data we use to assess the stability and dynamics of organisms - numbers in time and space and among species - have properties that must be accounted for, but rarely are. We know that the numbers of a species today and yesterday are more related than numbers today and 5 years ago. This is temporal correlation. We know that populations that live close to each other will be more alike than those far apart. This is spatial correlation. And we know that species that are closely related will be more similar than distantly related ones. This is evolutionary correlation. The majority of stability indicators have failed to accommodate these disruptive and well know features of real data. Third, there might be interactions among stressors. This means that the effect of one stressor depends on what the presence, absence or magnitude of another. This make prediction challenging, especially if we don't know about the dependencies. We need theory and data to understand how, and at what scales, multiple stressors impact stability. Our project aims to deal with all of these using three research objectives. 1. Develop improved indicators of stability, at multiple ecological scales (e.g. biomass/abundance, community structure/diversity, function), that formally rectify long recognised but rarely addressed issues arising from spatial, temporal and phylogenetic covariation;2. Make predictions about stability among ecosystems (coherence) facing multiple, simultaneous stressors produced by advanced simulation models that allow insight into the intrinsic processes and feedback mechanisms driving stability; and3. Test model predictions and the efficacy of improved indicators of stability using experimental and field collected data from terrestrial, marine and freshwater ecosystems.