Six-year study reveals ancient quakes along 150-mile fault system in Nepal
Sean Bemis, a research scientist in the Department of Geosciences, and Elizabeth Curtiss, a Ph.D. student, were part of a collaborative project that studied earthquake history in western Nepal.
A common misconception about research is that it takes place in climate-controlled labs with microscopes, beakers, and Bunsen burners.
While that is true for many fields, obtaining geoscience data can demand fieldwork in remote, rugged terrain with potentially extreme weather conditions. These investigations may require flying across the world, hiking for days above 14,000 feet of elevation in the Himalayan mountain range during all kinds of weather, and even sacrificing personal hygiene.
“I barely ever washed my hair,” Elizabeth Curtiss, a Ph.D. student in Virginia Tech's Department of Geosciences, said. “The water is just so cold.”
Curtiss has been part of a collaborative project with researchers from several U.S. universities and one in Nepal who studied earthquake history along a 150-mile fault system in western Nepal. She twice went to Nepal, and her efforts paid off in a seismic way as she was the lead author of a paper published in Geosphere, a journal of the Geological Society of America, that spotlighted the group’s work.
The paper capped a nearly six-year project that also included Sean Bemis, a research scientist in the Department of Geosciences within the College of Science and the lead principal investigator of the project. There were many facets to the project, but he and Curtiss primarily collected data to document the timing of prehistoric earthquakes on different portions of the Western Nepal Fault System, and that data revealed a minimum of three major previously undocumented surface-rupturing earthquakes took place along the system over the past 10,000 years and perhaps as many as 14.
The value of such data is two-fold. Knowledge of the potential for earthquakes on this fault system is directly helpful to countries like Nepal and India to plan for the impact of future earthquakes. Plus, information about the pattern of past earthquakes on the Western Nepal Fault System provides a model for earthquake potential in similar tectonic regions around the world.
“We want to take what we learn on this fault system and integrate it with the broader network of faults in the Himalayas,” Bemis said. “We want to generate the types of data about earthquake sources that planners and designers can use to better engineer structures in the region to anticipate potential hazards.”
Trenching for paleoseismic evidence
Researchers have worked to decipher the tectonics of the Himalayan mountains for decades, but before the COVID-19 pandemic, a small group had come up with a project idea on the Western Nepal Fault System based on clues from preceding studies. Group members decided that they wanted to pursue more extensive study. Several grant applications had been denied, and they enlisted the help of Bemis in large part because of his expertise in paleoseismology – a field of study that determines the timing and size of prehistoric earthquakes using geological evidence.
“I hadn’t worked in Nepal before, but the Himalayan mountains are an example that all geologists study in school,” Bemis said. “I think I had a little bit of a reputation for my remote fieldwork in Alaska and that might have been what led them to reach out to me for to join the project.”
Bemis ultimately secured grant money from the National Science Foundation to fund the project and invited Curtiss to be a part of it. The group also included researchers from the University of Kansas, the University of Houston, and Tribhuvan University in Nepal. The team conducted three years of fieldwork across the fault system.
The fieldwork took place in 2019, 2021, and 2022, with Bemis going to Nepal in 2019 and 2021 and Curtiss going in 2021 and 2022. They each spent more than a month in Nepal on each excursion, with an additional week in 2021 spent in quarantine after arrival in Kathmandu while following Nepal’s COVID-19 protocols. They later braved altitude sickness, snowy passes, and once, an intense storm that trapped them in their tents for more than two days.
“It takes a lot of effort to get data there. It’s not easy to get around,” Curtiss said. “There were no roads. There’s just no way to get to those places besides just walking. To be honest, I was also reading the 'Lord of the Rings' series while we were there camping, and it sometimes just felt like the two descriptions were so similar.”
During Curtiss’ first field season, group members trekked seven or eight days just to get to the first study site, and then they, along with their porters and guides, would hand-dig a trench across a fault ‘scarp’ – landforms that are created where past earthquakes have broken the ground surface. They ended up digging seven trenches over the course of the project, with each approximately 15 to 20 feet long and 3 to 5 feet deep. Targeting sites that were approximately 15 to 20 miles apart over nearly a 100-mile stretch enabled them to collect evidence from five different fault segments of the overall fault system.
“We would dig as deep as we could get across a fault scarp and create smooth sides of the trench to expose the subsurface layers of sediments and rocks,” Curtiss said. “This allows us to make a map of the sediments across the active fault, where broken and folded layers of sediment will occur as fingerprints of past earthquakes.
“Then, to determine time when the earthquake occurred, we collect samples of organic material from the deformed sediments and from overlying undeformed layers of sediment. The sample from the deformed sediment provides a maximum earthquake age, and the same from the undeformed sediment provides a minimum age.”
Upon returning to the States, they sent the samples off to be dated. Using charcoal to conduct radiocarbon dating, researchers were able to get ages for the sediments. Then Curtiss used the sequence of earthquake ages from each trench site to explore whether each earthquake documented at individual sites could correspond with earthquakes documented at adjacent sites. By imposing different models for what may limit how far an earthquake can propagate along a fault system, she was able to constrain the number of earthquakes and estimate their possible magnitudes.
“I present a spectrum of possible earthquake scenarios,” Curtiss said. “For example, if all the fault segments always rupture together during earthquakes, our record shows that three large magnitude earthquakes happened. Alternatively, if each fault segment was rupturing individually, there were 14 or more moderate-sized earthquakes.
“Additional earthquake scenarios illustrate how earthquakes would fit in our data if certain gaps or bends in the fault system were effective at stopping earthquake ruptures. Although more data would be required to define a single earthquake scenario for past ruptures of the Western Nepal Fault System, we can confidently show that it can produce large magnitude earthquakes and show where more work can be done to further refine the earthquake history.”
Predictions require more research
The paper submitted to Geosphere does not predict or suggest imminent earthquakes on the Western Nepal Fault System or on nearby faults. But the record of past earthquake occurrence in western Nepal should encourage engineers, planners, and builders to expand their focus when constructing homes and infrastructure – much of which, such as pipelines and hydroelectric facilities, are built well away from population centers.
“The only way to be able to make data-informed forecasts about future earthquakes is through this type of investigation, where we document prehistoric earthquakes through the sedimentary and faulting record,” Bemis said. “So, one of our key goals was to establish the time since the most recent earthquake occurred and how frequently earthquakes occur, so that communities and governments can understand and plan for hazards that could impact people’s lives and infrastructure.”
Researchers ultimately need more information to make predictions. The information that this group collected represents only a start.
Both Bemis and Curtiss hope to see more work done in the region. Such work would lead to improved earthquake rupture models, but the challenges to accumulating data are undeniable.
“There’s some more analysis that I’ve thought about that I could do remotely using the existing topographical datasets,” Bemis said. “But I think to really expand our earthquake knowledge, it would require more time digging in the ground in Nepal and that’s few years away at least.
“If this was an accessible place like the San Andreas fault [in California], our work would keep us busy for years of research and many student projects. That may still happen. It’s just going to take a lot longer. We were fortunate to get funded for a first study to try to understand the basics of what’s happening along this fault system. Now, we hope our work becomes the seed data for new scientists, new ideas, and new projects.”
Curtiss agreed. She defends her Ph.D. thesis in January and hasn’t quite figured out her future, but after her experiences in Nepal, doing a deeper dive on these old earthquakes could lead to more breakthroughs and sounds appealing – despite the sacrifices required.
“It was fun,” she said of her Nepal journeys. “It’s one of those amazing experiences with a lot of memorable moments. It was a really growing experience for me.”
