US Geological Survey seeks enhanced AI function to watch for earthquakes

The U.S. Geological Survey has started market research work to look at acquiring software that uses artificial intelligence and machine learning techniques to help operators monitor seismic activity, including earthquakes.

USGS’ National Earthquake Information Center operates detection algorithms in near-real time that help ingest data to inform the creations of earthquake information products. Both AI-driven and non-AI driven algorithms are part of that setup and work together across different systems.

The center is interested in acquiring new software to help refresh its AI and ML tools in order to incorporate more true real-time functionality, USGS said in a sources sought notice posted Tuesday.

As the notice says, AI/ML technologies have rapidly advanced in recent years and that movement is driving USGS to take a new look at what is out there.

NEIC’s primary challenges on the technology front cover the need to derive scientific data products from continuous seismic data acquisition systems, analysis systems, data quality tools, and other system health and monitoring systems.

The request for information breaks out software development work into five major areas:

• Develop and maintain real-time seismic data streaming and detection systems
• Modernize and optimize detection algorithms
• Integrate and advance AI/ML capabilities
• Build and integrate monitoring, archiving and distribution systems
• Support scientific research and innovation

Responses to the RFI are due by 4:30 p.m. Eastern time on July 24.