Project Grant 2219975

This $803,970 project grant from the National Science Foundation Office of Advanced Cyberinfrastructure will support the development of a scalable real-time streaming analytics and machine learning framework for geoscience and hazards research. A collaboration between the University of Colorado, University of Oregon, Rutgers University, and UNAVCO will create a data framework to enable generalized real-time streaming analytics and machine learning using over 1,500 sensors from the EarthScope and Ocean Observatories Initiative networks. The framework will aggregate diverse data streams, apply machine learning algorithms like convolutional neural networks to detect natural hazard events as they occur, and support rapid analysis and understanding of earthquake, volcanic, and tsunami data. Initial use cases will focus on the Cascadia subduction zone and Yellowstone area. Products will include an extensible, transportable architecture running in Docker to process data on laptops, clusters and the cloud. Outreach through collaborative online resources is intended to make datasets and computational resources more accessible. This project aligns with the goals of the National Science Foundation Computer and Information Science and Engineering program to advance computing and cyberinfrastructure for discovery and innovation.