Seismic Nuclear Explosion Monitoring Panel Discussion at DDDAS Workshop
Explore the discussions and advancements in Seismic Nuclear Explosion Monitoring, including the challenges, motivations, and use of machine intelligence. Get insights from experts at the panel discussion and learn about the research and development programs in this crucial field. Stay informed about the efforts to enhance global security through innovative technologies.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
DDDAS Workshop: Earth Planets, Climate and Life DDDAS Workshop: Earth Planets, Climate and Life Session VII Session VII Panel Discussion: Seismic Nuclear Explosion Monitoring
Agenda Agenda Background and Context Introduction of Panel Members Problem Description (Drs. Alan Poffenberger and Brian Pope, Air Force Technical Applications Center) Guided Discussion / Questions & Answers
Nuclear Explosion Monitoring (NEM) Nuclear Explosion Monitoring (NEM) Massive and increasing data volumes
Motivation Motivation The Air Force Technical Applications Center (AFTAC) monitors global sensor networks to detect and identify nuclear events for the United States. The Air Force Research Lab (AFRL) NEM R&D program has a strong focus on improved and automated processing for an exponentially growing volume of seismic data. photo courtesy AFTAC
Machine Intelligence in Nuclear Explosion Machine Intelligence in Nuclear Explosion Monitoring (MINEM) R&D Program Monitoring (MINEM) R&D Program Mandate: exploit the promise of machine intelligence in NEM to achieve automatic processing good enough to trust. Structure: AFRL funds fundamental R&D through a flexible contracting agreement with Applied Research Associates (ARA) and multiple team member organizations. Timeline: MINEM launched in November 2020 with a unique project to develop a strategic R&D plan. The first set of R&D projects began in early 2022. Term of the MINEM agreement ends in November 2025.
Todays Panel Members Today s Panel Members Dr. Eli Baker (Air Force Research Laboratory) Prof. Karianne Bergen (Brown University) Prof. Abdullah Mueen (U. New Mexico) Dr. Alan Poffenberger (Air Force Technical Applications Center) Dr. Brian Pope (Air Force Technical Applications Center) Dr. Sai Ravela (MIT) Dr. Delaine Reiter (Applied Research Associates, Inc.) Dr. William Rodi (retired MIT, independent consultant) Dr. Jesse Williams (Global Technology Connection)
United States National Data Center Automated Data Processing Challenges A. Poffenberger, B. Pope Oct 2022
Processing is Driven by Requirements AFTAC s data processing is driven by our challenging requirements Event Detection/Identification Thresholds Which drives: Low signal detection thresholds for station processing Low Event Definition Criteria for network processing Multiple analysis timelines and thorough analyst review 8
Comparisons of automated vs final results: Station Processing Relative contribution to final Eval1 bulletin 53% of arrivals in final bulletin are detected automatically Signal Detection 63% of automated onset times re- timed by analysts Signal Onset Determination 56% of signal measurements adjusted by analysts Signal Measurement 9
Automatic vs Reviewed Bulletins, 2018 Automated: N Analyst: 1.08 x N Fraction in Auto Bulletin % Events Overlap: 62.5%/57.9% EVAL1 NDEF 11
Conclusion Meeting the US NDC threshold requirements means processing orders of magnitude more data, signals, and events An increase to the number of analysts is unlikely; R&D is needed to relieve the analyst burden for seemingly mundane tasks such as onset time and A/T measurements for which improved automation is needed This is in addition to R&D needed for network processing, such as improved association algorithms, additional regional magnitude scales and event identification methodologies 12
Discussion? 13
Themes & Issues for Discussion What does AI/ML bring to the table in relation to "standard" approaches -- inverse problems, signal processing, statistical inference etc.?
Themes & Issues for Discussion Themes & Issues for Discussion What are some important challenges in seismic nuclear explosion monitoring that are amenable to AI/ML or other DDDAS-style solutions?
Themes & Issues for Discussion Themes & Issues for Discussion Will AI/ML need to be informed by theory" and user on the loop" to be successful?
Themes & Issues for Discussion Themes & Issues for Discussion How can dynamic data-driven approaches be used to continually improve monitoring from past bulletins, especially given the paucity of new data?
Themes & Issues for Discussion Themes & Issues for Discussion How can dynamic data-driven approaches be used to continually improve monitoring from past bulletins, especially given the paucity of new data?
Themes & Issues for Discussion Themes & Issues for Discussion What synergies between general seismic monitoring and nuclear explosion monitoring are useful to explore? Are there concepts from other DDDAS efforts that could improve MINEM outcomes?
