EAGER-NEON: Scaling up terrestrial plant phenology from individuals to Continental scale

植物开花、树木落叶和蚊子孵化的时间为生物体如何对其直接环境做出反应提供了一些最明显的迹象。这些标志是所有人都可以接触到的，让公众参与观察基本的生态关系。该项目结合和比较不同的数据来源，以确定为特定地点的个体收集的物候(或时间)数据在多大程度上代表更大规模的模式。例如，在加利福尼亚州的草甸上，一个物种开花较早是否代表了区域或大陆规模的开花较早？该项目将制作一系列关于植物物候学的视频，将通过东北地区分发给K-12教师，以及与BuBurst计划和自然-S笔记本有关的公民科学家。一名本科生将参与该项目，学习如何为不同类型的数据建模。该项目将聘用一名职业生涯早期的博士后研究人员，他们将接受量化建模和项目领导方面的培训。该项目的目标是通过建立一个整合物候和环境数据的分层模型框架，将个体生物体的物候观测扩展到群落、区域和大陆尺度。该框架将物候建模中的新兴技术与贝叶斯方法相结合，该方法可以容纳广泛的数据集，并允许传播不确定性。该项目将评估非生物环境、物种身份、功能类型、群落组成和景观异质性对13个国家生态观测网络(霓虹灯)站点物候的影响，并将调查个别站点的物候测量可以外推到更大区域的程度。结果将与从哈佛森林收集的长期物候数据进行校准，以验证仅使用霓虹灯数据获得的参数估计和模型输出。由此产生的建模框架将是灵活和全面的，有助于解释各种物候数据，包括来自霓虹灯的标准化地面观测、来自PhenoCam网络的近地表树冠物候数据、卫星遥感、在单个地点的长期地面观测以及来自公民科学网络的观测。

The timing of when plants flower, trees leaf out, and mosquitoes hatch provide some of the most obvious signs of how organisms respond to their immediate environments. These signs are accessible to all people, engaging the public in observing fundamental ecological relationships. This project combines and compares different sources of data to determine how well phenological (or timing) data collected for individuals at a particular location represent larger-scale patterns. Is earlier flowering of one species in a meadow in California, for example, representative of earlier flowering at regional or continental scales? The project will produce a series of videos about plant phenology that will be distributed to K-12 teachers through the Northeast as well as with citizen scientists affiliated with Project Budburst and Nature?s Notebook. An undergraduate student will participate in the project, learning to model diverse kinds of data. The project will engage an early-career postdoctoral researcher who will receive training in quantitative modeling and project leadership. The goal of this project is to scale phenology observations of individual organisms to the community, regional, and continental scales by building a hierarchical modeling framework for integrating phenological and environmental data. The framework combines emerging techniques in phenology modeling with a Bayesian approach that accommodates a wide range of datasets and allows for propagation of uncertainty. The project will evaluate the effects of the abiotic environment, species identity, functional type, community composition, and landscape heterogeneity on phenology across 13 National Ecological Observatory Network, or NEON, sites and will investigate the extent to which phenology measures at individual sites can be extrapolated to larger regions. Results will be calibrated against long-term phenological data collected from the Harvard Forest, to verify parameter estimations and model output obtained using NEON data alone. The resulting modeling framework will be flexible and comprehensive, facilitating the interpretation of diverse phenology data, including standardized ground observations from NEON, near-surface canopy phenology data from the PhenoCam network, satellite remote sensing, long-term ground observations at single sites, and observations from citizen science networks.