Related project notes

Convolutional Recurrent Model for Earthquake Identification and Magnitude Estimation

• Project Description: This project aims to utilize advanced machine learning techniques, specifically convolutional recurrent neural networks, to enhance the accuracy of earthquake identification and magnitude estimation. By leveraging large datasets of seismic activity, the model learns to recognize patterns and predict earthquake occurrences efficiently.

• Data Goals and Policies

  • NSF Data Management Plan Requirements:

    • Archive all research data and documentation.

    • Ensure data quality and accessibility.

    • Facilitate data reusability and preservation.

  • NEES Data Archiving Details:

    • Actors: Research teams, site personnel, and data curators.

    • Data Types: Includes sensor measurements, observations, analyses, simulations, and reports.

    • Timeframes for Archiving: Data must be uploaded at specified intervals post-experiment (1 month, 6 months, and 12 months).

    • Locations: Data stored in Project Warehouse.

    • Purpose: Enhances research impact, saves resources, facilitates knowledge transfer.

• Types of Data Managed

  • Diversity of Research Areas:

    • Structural engineering, geotechnical engineering, geophysical research, and tsunami research.

  • Data Characteristics:

    • Experimental, observational, computational, and increasingly complex in terms of sensor usage and data collection.

• Quality Assurance Components

  • Human Oversight:

    • Importance of human input and monitoring to ensure data quality.

  • Metadata Requirements:

    • Must be meaningful, consistent, accurate, and accessible.

  • Technical Standards:

    • Ensure formats are interoperable and digitally preservable.

    • Use tools for format identification and validation (e.g., JHOVE, MD5 checksums).

• Metadata Management

  • Content of Metadata:

    • Templates for sensor data, file formats.

    • Auto-completed information fields (personal names, organization details, sensor specifications).

  • Completeness Standards:

    • Adherence to research standards and NEES documentation.

    • Must complete specific milestones post-experiment for data upload.

• Re-use of Data

  • Concepts Defined:

    • Data Use: Application of research data for the intended study.

    • Data Re-use: Application for a different research purpose.

    • Data Purposing: Making existing data available for immediate use.

    • Data Re-purposing: Adapting existing data for future research needs.

NEEShub

• Project Description: NEEShub serves as a comprehensive management platform for earthquake engineering experiments, providing resources and archives related to various projects. This hub focuses on maintaining data integrity and facilitating collaboration among research teams to enhance the understanding and mitigation of earthquake risks.

• Management of Experiments:

  • Project Warehouse: Houses various experiments related to structural integrity.

    • Example Project: Mitigation of Collapse Risk in Vulnerable Concrete Buildings, Experiment-5.

    • Test Series & Specifications:

      • Test performed under cyclic lateral displacement with specific reinforcement ratios.

      • Date of experiment: February 17, 2009 to Present.

      • Specifications included: sensors used, material type, and facility information.

UCERF4

• Project Description: UCERF4 focuses on evaluating and modeling the earthquake risk in California by integrating scientific research into policy and practice. It aims to improve the understanding of seismic hazards and support planning efforts by providing reliable data and insights.

• Institution: George E. Brown, Jr. Network for Earthquake Engineering Simulation (NEES)

• License: Creative Commons Attribution 3.0 Unported License.

• Questions & Comments: Contact NEES data curators for further inquiries.

• Visual Examples: Highlight the importance of clear and detailed reporting in experiments through figures and diagrams showcased throughout the study.