Shaikh Arifuzzaman
University of Nevada-Las Vegas, NV·Earth & Environmental Science
Earthquakes kill people, wreck infrastructure and disrupt communities, yet many of the faults capable of producing them are mapped poorly or not at all. Fault mapping today relies heavily on expert interpretation of images and field evidence, which is slow and inconsistent from region to region.
- Fields
Earth & Environmental Science
Computer Science
- Topics
active faults
fault mapping
neotectonics
- Grants
- 1 known grant, through Sep 2029
What they work on
- Develop an artificial intelligence framework that maps and interprets active fault networks by integrating multiple data sources, and that characterizes fault-system dynamics across regions with contrasting tectonic settings.
- Build physically informed learning models that embed geophysical and geological constraints so that inferred fault properties remain consistent with known Earth physics rather than resting on image features alone.
- Incorporate interpretable, uncertainty-aware learning so fault-system maps carry confidence estimates and explanations, moving beyond simple feature detection.
- Benchmark the resulting workflows against well-characterized fault systems and release reproducible tools and datasets for community use.
More about the research
This project treats fault mapping as a machine learning problem and asks whether artificial intelligence can identify and interpret active fault networks more accurately and more uniformly worldwide by combining evidence from many different data sources.
Why it matters
Hazard maps, building codes and insurance decisions all depend on knowing where active faults are, and large parts of the world lack that information. Consistent, uncertainty-aware AI mapping could extend fault inventories to under-studied regions and improve earthquake hazard estimates for the communities living on them. Releasing open tools and datasets would let other groups apply the same methods to their own regions.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Folarin Kolawole
Columbia University, NY·Earth & Environmental Science
Test whether fault motion along the reactivating Davie Fracture Zone varies along its length because of inherited bends and offsets, or stays uniform through local rotations of the stress field.
Ahmed Elbanna
University of Illinois Urbana-Champaign, IL·Earth & Environmental Science
Build an extensible software framework and science gateway that gives a broad user community access to physics-based earthquake rupture forecast models without requiring local expertise or computing resources.
Lucy Flesch
Purdue University, IN·Earth & Environmental Science
Build a regionally consistent, multi-plate velocity field for the Arctic by combining GPS measurements with mid-ocean ridge spreading rates that record motion over hundreds of thousands of years.
The school
University of Nevada-Las Vegas: 4 labs on PI Pages, $118M a year on research, 241 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.