MRI: Development of an Instrument for Student and Faculty Research on Multimodal Environmental Observations

该仪器专注于动态沼泽地生态系统的生态研究，提供了一个平台来汇总和比较许多环境数据集。它还评估了算法对生态模型输出的影响。该工具利用了佛罗里达大沼泽地试点研究的经验和区域途径。预期的结果很可能构成计算和生态学领域最先进的新发现的重大进展。通过更好地理解生态系统动力学中的时空协同作用，科学家将能够预测复杂的生态系统反应，并指导风险评估、规划和解决方案的制定。有能力的研究将产生基于证据的科学，对土地管理决策有用。该工具是一项试点，可扩展到其他区域以及国家或全球范围。所有软件和fiu生成的数据都将是开源的。该工具支持需要集成大型多模态异构数据集的研究，以应对当前数据驱动的生态系统研究和工具所面临的挑战，包括不同数据的缩放；通用全局模型和算法；校准;遥感数据的获取呈指数级增长。具体启用的研究活动涉及时空数据的多维模糊逻辑；多模态图像分析；还有机器学习。该系统包括一个数据存储库、一个多模态生态系统数据聚合器以及分析和可视化工具。现有的大型全局数据集存储库不允许合并大量多维的、历史的本地获取的数据集。该平台提供全面的本地数据集成及其与全球数据的关联。虽然仪器将容纳精心策划的环境数据集合，但其独特的服务将超越单纯的管理，内置数据挖掘，模糊逻辑推理，图像分析和可视化工具，开发和接口，以增强学生和教师研究人员的能力。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Focusing on ecological research in the dynamic Everglades Ecosystem, this Instrument provides a platform to aggregate and compare many environmental data sets. It also evaluates the effect of algorithms on ecological model outputs. The instrument leverages experiences and regional access to pilot studies on the Florida Everglades. The expected outcomes are likely to constitute a significant advance in state-of-the-art, novel discoveries in the fields of computing and ecology. By developing a better understanding of spatial and temporal synergies in ecosystem dynamics, scientists will be able to predict complex ecosystem responses and guide risk assessment, planning, and solution development. The enabled research will produce evidence-based science, useful in land-management decisions. The Instrument is a pilot, expandable to other regions and to the national or global scale. All software and FIU-produced data will be open-sourced.This Instrument supports research that requires the integration of large multimodal heterogeneous datasets to address the challenges with current data-driven ecosystems research and instruments, including scaling of different data; generic global models and algorithms; calibration; and exponentially-growing acquisition of remotely sensed data. Specific enabled research activities involve multi-dimensional fuzzy logic on spatiotemporal data; multimodal image analysis; and machine learning. The system comprises a data repository, a Multimodal Ecosystem Data Aggregator, and analytical and visualization tools. Existing large repositories of global datasets do not readily allow the incorporation of vast amounts of multidimensional, historical locally-acquired datasets. The platform provides comprehensive local data integration and its correlation to global data. While the Instrument will house a well-curated collection of environmental data, its unique service will transcend mere curation, with built-in tools for data mining, fuzzy-logic reasoning, image analysis, and visualization -- developed and interfaced to empower both student and faculty researchers.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.