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

点击翻译按钮获取中文摘要

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.