NSF Postdoctoral Fellowship in Biology for FY 2011

该行动资助了2011财年美国国家科学基金会生物学博士后研究奖学金，生物学与数学和数学的交叉。该奖学金支持在生物学与数学和地球化学交叉的宿主实验室中为研究员提供研究和培训计划。Cynthia Chang的研究和培训计划的标题是“使用分层贝叶斯模型来了解火山爆发后的社区聚集并预测对全球变化的反应”。“该研究金的主办机构是华盛顿大学西雅图分校，赞助科学家是Janneke Hille Ris Lambers博士和Peter Guttorp博士。了解自然群落在重大干扰后如何聚集，可以对生态系统随时间的动态性质提供独特的见解。在植物群落中，扩散策略、生活史和生长性状是影响群落相互作用的决定性因素。随机因素，如随机机会，也可以决定社区如何聚集。这两种类型的相互作用影响群落物种多样性的演替过程中。贝叶斯分层建模具有独特的能力，将确定性和随机因素，多尺度和相互作用。然而，生物学的洞察力是建立有意义的模型所必需的。基于一个30年的大火山爆发后的植物定植，组成和性状分布的数据集，这项研究正在生成模型，进一步我们的社区组装过程后，一个重大干扰的理解，以及预测未来的社区对全球变化的反应基于过去的关系。培训目标包括发展统计和分析技术的专业知识。教育推广包括开发本科和高中课程，使用圣海伦山？的研究作为一个案例研究，并涉及研究生和本科生的研究。更广泛的影响包括使决策者了解这一生态系统和类似生态系统的特点，以便今后进行土地管理。预测模型可以作为一个理论框架，了解生态群落如何应对其他重大干扰以及预期的全球变化。这项研究的结果正在通过科学和大众讲座，与大自然保护协会和美国林务局的会议以及同行评审期刊上的出版物进行传播。

This action funds an NSF Postdoctoral Research Fellowship in Biology for FY 2011, Intersections of Biology and Mathematical and Math. The fellowship supports a research and training plan in a host laboratory for the Fellow at the intersection of biology with mathematics and geochemistry. The title of the research and training plan for this fellowship to Cynthia Chang is "Using hierarchical Bayesian modeling to understand community assembly after an eruption and predict responses to global change." The host institution for this fellowship is the University of Washington, Seattle and the sponsoring scientists are Drs. Janneke Hille Ris Lambers and Peter Guttorp.Understanding how natural communities assemble following major disturbances gives unique insight into the dynamic nature of ecosystems through time. In plant communities, dispersal strategy, life history, and growth traits are deterministic factors that influence community interactions. Stochastic factors, such as random chance, can also dictate how communities assemble. Both types of interactions influence community species diversity over the course of succession. Bayesian hierarchical modeling has the unique ability to incorporate deterministic and stochastic factors, multiple scales, and interactions. Biological insight, however, is necessary to build meaningful models. Based upon a 30-year dataset on plant colonization, composition, and trait distribution after a major volcanic eruption, this research is generating models that further our understanding of community assembly processes after a major disturbance, as well as predict future community response to global change based on past relationships.Training objectives include developing expertise in statistical and analytical techniques. Educational outreach includes developing undergraduate and high school course curricula that use Mount St. Helen?s research as a case study and involving graduate and undergraduate students with research. The broader impacts include informing policy makers for future land management of the characteristics of this and similar ecosystems. The predictive models can serve as a theoretical framework for understanding how ecological communities may respond to other major disturbances as well as anticipated global change. The findings of this research are being disseminated through scientific and popular talks, meetings with The Nature Conservancy and U.S. Forest Service, and publications in peer-reviewed journals.