DOI · DOI-0155 · record R2607

Machine Learning for automatic fracture mapping and rock identification [2024 INV#WO0000000109499]

Department of the Interior · USGS

Overview

Development stage: DeployedSource: c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.
High-impact designation: Not high-impactSource: c) Not high-impact
Impact justification: Not reported
Start date: 2025-03-01 (day precision)Source: 3/1/2025
Withheld from public reporting: NoSource: a) No

Mission

Topic area: Science
Operational functions: No function tag matched Derived by keyword rules; see methodology

Problem the AI is intended to solve

Machine learning algorithms are being used to improve detection and characterization of fault surface geometries using the spatial patterns of earthquake locations. We have improved our ability to generate long, continuous, fault surface representations and have implemented non-planar machine learning based fitting approaches.

Expected benefits

Improved fault geometries that advance our ability to accurately model earthquake hazards.

System outputs

The outputs are 3D fault models that are meshed at the user's specified resolution, fault model quality metrics, and a fully 3D render of the fault surfaces.

Technology

AI classification: Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…
System name(s): Not reported
Custom-developed code: No
Public source code: code.usgs.gov ↗

Sourcing

How it was built: Developed in-houseSource: b) Developed in-house
Vendor (as reported): Not reported
Vendors (standardized): None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII: No
Privacy Impact Assessment: Not reported
Authorization to Operate: No
Demographic features: k) None of the above
Training and evaluation data: We used publicly available earthquake catalogs from northern California, specialized high-resolution aftershock catalogs, an existing fault models to evaluate model performance.
Federal Data Catalog: Not reported