Phylogenetic generalized linear mixed modeling presents novel opportunities for eco-evolutionary synthesis
Phylogenetic generalized linear mixed modeling presents novel opportunities for eco-evolutionary synthesis
复制标题
DOI：
10.1111/oik.08048
复制
发表时间：
2021-03-25
期刊：
影响因子：
3.4
通讯作者：
Pearse, William D.
中科院分区：
文献类型：
--
作者：
Gallinat, Amanda S.;Pearse, William D.
尽管群落生态学和进化生物学相互依存，但在群落聚集的研究中，它们很难经验性地综合起来。这主要是由于生态和进化动态的时间和空间尺度不同，从物种形成和环境过滤等大范围的过程到局部规模的过去和现在的生态位动态。系统发育广义线性混合模型(PGLMM)为解决这一问题提供了一种解决方案，它可以通过对基于特征的群落集合的演化进行建模，并跨越从大范围环境敏感性到局部尺度共生的空间范围，来随时间整合。因此，通过将生物多样性的深层和浅部驱动因素纳入单一模型，PGLMM提供了比较它们的相对强度的能力。尽管PGLMM具有独特的价值，但它的应用一直受到限制，因为统计上的进步与概念上的进步没有得到充分的匹配。最近在可获得的跨分支组合数据和系统发育工具方面的扩大，增加了概念统一的紧迫性。在这里，我们描述了PGLMM在分析群落聚集的进化和生态驱动因素方面的潜力，重点介绍了三个主要驱动因素--环境敏感性、分支内相互作用(例如竞争)和交叉分支关联(例如草食)--如何塑造历史和当今的组合。我们概述了PGLMM可以解决的三个基本问题，这些问题与上述每个驱动因素有关：1)物种的区域尺度环境反应在进化上受到限制吗？2)对过去竞争的进化反应是最小化还是增强现在的竞争？3)跨支联结在进化上受到限制吗？对于每个问题，我们将回顾概念上的进展和机会，并在补充教程中演示PGLMM的应用。我们关注PGLMM的生态和进化结果，并描述这些结果对保护和自然资源管理的价值，以使PGLMM超越统计复杂性，走向一个具有明确概念和实践目标的未来。
Despite their interdependence, community ecology and evolutionary biology have proven difficult to synthesize empirically in studies of community assembly. This is primarily due to differing temporal and spatial scales of ecological and evolutionary dynamics, ranging from broad-scale processes like speciation and environmental filtering to local-scale past and present-day niche dynamics. Phylogenetic generalized linear mixed modeling (PGLMM) offers a solution to this problem, it can be used to integrate through time by modeling the evolution of trait-based community assembly, and across space ranging from broad-scale environmental sensitivities to local-scale co-occurrences. As such, PGLMM provides the ability to compare the relative strength of deep versus shallow-time drivers of biodiversity by including them in a single model. Despite its unique value, the application of PGLMM has been limited because statistical advances have not been adequately matched by conceptual progress. Recent expansion in the availability of cross-clade assemblage data and phylogenetic tools have increased the urgency of conceptual unification. Here we describe the potential of PGLMM for parsing the evolutionary and ecological drivers of community assembly, focusing on how three major drivers - environmental sensitivities, within-clade interactions (e.g. competition), and cross-clade associations (e.g. herbivory) - shape historical and present-day assemblages. We outline three fundamental questions that PGLMM can address, linked to each of the aforementioned drivers: 1) are species' regional-scale environmental responses evolutionarily constrained? 2) Do evolved responses to past competition minimize or enhance present-day competition? 3) Are cross-clade associations evolutionarily constrained? For each question, we review conceptual advances and opportunities, and demonstrate the application of PGLMM in a supplementary tutorial. We focus on the ecological and evolutionary outcomes of PGLMM and describe the value of these outcomes for conservation and natural resource management, in order to move PGLMM beyond statistical complexities and toward a future with clear conceptual and practical goals.