Extreme event ecology needs proactive funding
Extreme event ecology needs proactive funding
复制标题
极端事件生态需要积极的资助
DOI：
10.1002/fee.2569
复制
发表时间：
2022
影响因子：
10.3
通讯作者：
McDowell, William H
中科院分区：
文献类型：
--
作者：
Patrick, Christopher J;Hensel, Enie;Kominoski, John S;Stauffer, Beth A;McDowell, William H
野火、飓风和洪水等极端事件的频率和强度都有所增加。问题不再是这些事件是否会发生，而是何时何地发生（Stott 2016），对基本生态系统服务产生不利影响，包括清洁水，可收获材料和碳封存。在某些情况下，野火等极端事件可能对人口和生态系统产生积极影响。管理这些影响需要了解环境背景以及生态系统和干扰特征如何驱动系统响应（Hogan等人，2020）。然而，对生态极端事件研究的资助，例如通过美国国家科学基金会（NSF）的RAPID计划，通常是反应性的。事件前的数据，快速的先决条件，通常是缺乏或只有零星的，极端事件的案例研究往往出现在现有的长期研究地点的偶然干扰。这种反应性的随机方法已经播种的文献与计划外的个案研究，描述个别事件。虽然对荟萃分析有用（例如帕特里克等人，2022），但此类研究提供的时空推断和预测能力有限。优先考虑极端事件的研究和对干扰生态学基本概念的实证检验是至关重要的（Aoki et al. 2022）。虽然美国国家科学基金会是合乎逻辑的美国资助机构支持这类工作，我们的作者不知道任何资助模式在美国国家科学基金会（或其他美国联邦机构）的积极主动，协调，假设驱动的研究在时空尺度需要有效地研究未来的自然事件。因此，新的资助机制是必要的，这些机制以新颖的方式将现有项目的联合收割机元素结合起来，为研究人员提供填补关键知识空白的灵活性。要加深我们对极端事件的驱动因素和对地球各种生态系统的影响的理解，就需要对实地的概念框架进行精心策划的测试。这种机械的、经验性的研究将需要：（1）在适合于检验先验假设的地点收集事件前的数据;（2）在足以解决季节和年际动态、事件前时期、随机扰动事件和事件后恢复时期的时间尺度内从实验阵列收集数据并维护数据;（3）在地理上不同的地点进行复制，以确保研究包括受影响和未受影响地点的比较。网络实验和足够长时间的监测对这种方法都至关重要。网络研究可以提供强有力的推理，是设计未来极端事件调查的有效方法。围绕未来事件规划干扰研究具有固有风险，因为无法保证研究中心在研究期间会受到干扰。但是，这种风险可以大大降低。首先，在多个地理位置不同的研究中心工作会增加一个或多个研究中心在研究期间受到影响的可能性。其次，使用历史干扰频率数据来选择干扰发生机会最高的位置，进一步增加了研究地点受到影响的可能性。例如，沿着沿着美国大陆海岸线有三个飓风热点，可以作为持续飓风研究网络的站点（Landsea and富兰克林，2013）：北卡罗来纳州的恐怖角、佛罗里达南部和路易斯安那州中部海岸（图1）。在过去20年中的任何一个特定的五年间隔内，这三个热点地区的一个或多个地点都有100%的机会受到...
Extreme events such as wildfires, hurricanes, and floods have increased in frequency and intensity. It is no longer a question of if, but rather when and where these events will occur (Stott 2016), with adverse impacts on essential ecosystem services including clean water, harvestable materials, and carbon sequestration. In some cases, extreme events such as wildfires may have positive impacts on populations and ecosystems. Managing these impacts requires understanding how environmental context as well as ecosystem and disturbance characteristics drive system responses (Hogan et al. 2020). However, funding for ecological extreme events research, such as through the US National Science Foundation’s (NSF’s) RAPID program, is typically reactive. Pre-event data, a RAPID prerequisite, are typically lacking or only sporadically available, and case studies of extreme events often arise from chance disturbances at existing long-term research sites. This reactive stochastic approach has seeded the literature with unplanned case studies describing individual events. While useful for meta-analyses (eg Patrick et al. 2022), such studies provide limited spatiotemporal inference and predictive capacity. Prioritizing the study of extreme events and empirically testing fundamental concepts in disturbance ecology is paramount (Aoki et al. 2022). Although NSF is the logical US funding agency for supporting this type of work, we–the authors–are unaware of any funding model at NSF (or other US federal agencies) for proactive, coordinated, hypothesis-driven research at the spatiotemporal scales needed to effectively study future natural events. Therefore, new funding mechanisms are necessary, ones that combine elements of existing programs in novel ways to provide researchers the flexibility to fill critical knowledge gaps. Advancing our understanding of the drivers and effects of extreme events on Earth’s diverse ecosystems requires carefully planned tests of conceptual frameworks in the field. Such mechanistic, empirical studies will necessitate:(1) collection of pre-event data at locations ideal for testing a priori hypotheses;(2) data collection from and maintenance of experimental arrays over timescales sufficient to resolve seasonal and interannual dynamics, pre-event periods, stochastic disturbance events, and post-event recovery periods; and (3) replication across geographically distinct locations to ensure that studies include comparison of impacted and unimpacted sites. Networked experiments and monitoring over sufficient time periods are both critically important to this approach.Networked studies can provide powerful inference and are an efficient way to design investigations of future extreme events. Planning a disturbance study around a future event is inherently risky, as there is no guarantee that a study site will be disturbed during the study period. However, this risk can be greatly reduced. First, working at multiple, geographically distinct study sites increases the probability that one or more sites will be affected during a study period. Second, using historical disturbance frequency data to select locations with the highest chance of a disturbance occurring further increases the probability that a study site will be impacted. For example, there are three hurricane hotspots along the continental US coastline that could serve as sites for a sustained hurricane research network (Landsea and Franklin 2013): Cape Fear in North Carolina, southern Florida, and the central Louisiana coast (Figure 1). During any given five-year interval in the past 20 years there was a 100% chance that one or more sites within these three hotspots would be impacted by …
影响因子：
10.1
作者：
通讯作者：
--