CAREER: Advancements in Spatio-temporal Modeling and Education in Support of NEON and Large-scale and Long-term Ecological Research

科学界正在进入这样一个时代，在这个时代，开放获取的丰富数据环境为理解区域到大陆范围内生态过程的空间和时间复杂性提供了非同寻常的机会。国家生态观测网(NEON)和地球数据观测网(DataONE)的倡议体现了对收集、开发和分发数据和工具以促进大规模和长期科学的投资。这些，以及类似的举措，代表着未来科学发现方式的范式转变。这一职业奖将开发理论、方法、软件和教学方面的进步，使当前和未来的科学家和教育工作者能够通过以下方式得出关于大型和复杂生态系统的有效推断：吸收不同的来源和类型的数据；适应空间和时间相关性以满足统计模型假设并改进预测推理；通过在大范围内进行精细空间尺度预测来划分和传播不确定源；以及进行扩展以有效地利用海量数据集中的信息。这项研究将开发新的灵活的时空建模框架，以评估霓虹灯在气候变化、土地利用、入侵物种、生物地球化学、生物多样性、生态水文学和传染病领域的重大挑战。尽管提出的方法是由与霓虹灯任务相关的实质性问题推动的，时空数据建模的潜在进步将在基本目标相同的领域得到应用，如公共和环境健康、气象学、工程学和地球科学--使用新的发现来帮助改善社会。建议的教育目标将使学生能够探索他们特定的研究兴趣，利用复杂的数据建立新的理解，并学会合作来应对各自学科内和跨学科的挑战和机遇。该奖项将开发和提供几项综合教育活动，包括：i)开发和实施跨学院的地理和生态信息学本科和研究生学位课程；ii)应用环境数据建模的本科高级课程；iii)研究生级别的课程，侧重于分层贝叶斯时空建模中的更高级主题；iv)在密歇根州西南部13个学区的23所K-12学校丰富科学教学；以及五)研究生研究专题讨论会侧重于环境数据分析的当代主题，将吸引来自多个机构的学生和专家参与，并为毕业生提供一个分享他们的研究、建立网络、获得专门技能和了解霓虹灯数据产品的机会。教育活动将允许未来和当前的科学家以创新的方式扩展自己，并在问题上进行合作。

The scientific community is moving into an era where open-access data-rich environments provide extraordinary opportunities to understand the spatial and temporal complexity of ecological processes at regional to continental scales. Investment to collect, develop, and distribute data and tools to further large-scale and long-term science is exemplified by the National Ecological Observatory Network (NEON) and Data Observation Network for Earth (DataONE) initiatives. These, and similar initiatives, represent a paradigm shift in the way future scientific discovery will occur. This Career award will develop theoretical, methodological, software, and instructional advancements that will allow current and future scientists and educators to draw valid inference about large and complex ecological systems by: assimilating disparate sources and types of data; accommodating spatial and temporal dependence to satisfy statistical model assumptions and improve predictive inference; partitioning and propagating sources of uncertainty through fine spatial scale predictions over large domains, and; scaling to effectively exploit information in massive datasets. The research will develop new flexible spatio-temporal modeling frameworks tailored to enable assessment of NEON's Grand Challenges in the areas of climate change, land use, invasive species, biogeochemistry, biodiversity, ecohydrology, and infectious diseases. Although development of the proposed methods is motivated by substantive questions related to NEON's mission, potential advancements in spatio-temporal data modeling will find use in fields such as public and environmental health, meteorology, engineering, and geosciences where the fundamental goal is the same -- use new findings to help improve society.The proposed educational objectives will enable students to explore their particular research interests, exploit complex data to build new understanding, and learn to collaborate to address challenges and opportunities within and across their respective disciplines. The award will develop and deliver several integrative education activities including: i) the development and implementation of cross-college undergraduate and graduate degree programs in Geo- and Eco-Informatics; ii) an undergraduate senior-level course in applied environmental data modeling; iii) a graduate-level course focused on more advanced topics in hierarchical Bayesian spatio-temporal modeling; iv) enrichment of science instruction in 23 K-12 schools in 13 districts in southwestern Michigan, and; v) graduate research symposia focused on contemporary topics in environmental data analysis that will engage students and experts from multiple institutions and serve as an opportunity for the graduates to share their research, network, garner specialized skills, and learn about NEON data products. Education activities will allow future and current scientists to extend themselves in innovative ways and collaborate on problems.