Deciphering Nature's Seismograph
How Sediments Record Past Earthquakes and Inform Future Hazard Assessments
Earthquakes can be catastrophic events, damaging structures and endangering lives. Thus, the ability to assess risks of future earthquakes is of enormous benefit to society. To do so, USGS scientists are increasingly turning to the past.
People have been recording seismic activity for centuries. To assemble a detailed earthquake history of an area and understand how faults may behave in the future, however, scientists need to go further back in time—from several hundred to many thousands of years ago.
Paleoseismology is the study of earthquakes that pre-date modern instrumentation. For example, the 1700 Cascadia earthquake occurred before seismometers existed to measure it; the timing and magnitude was inferred from historical evidence such as sunken coastlines, salt-killed forests and written records of an "orphan tsunami" that reached Japan. Paleoseismology incorporates such evidence to help scientists understand the likelihood and risks of future earthquakes.
The USGS has long been involved with deep-sea fault mapping, which typically uses a combination of sonar and sediment sampling to study faults offshore, often far beneath the seafloor. Applying similar techniques to lake sediments is a relatively new and growing field.
"Trenching allows us to determine [using chemical analysis of sediment] when a specific fault moved, but to get an idea of when an entire area experienced earthquake shaking —in other words, the timing of regional earthquakes, not necessarily tied to a specific fault—we can look to ocean and lake sediments," said Kate Scharer, Research Geologist with the USGS Earthquake Science Center. "The methods are complementary; both provide needed data to understand earthquake hazards."
"Compared to the seafloor, lake sediment is usually less disturbed, and in some cases almost pristine, with layers going back thousands of years," said Danny Brothers, Research Geologist at the PCMSC.
"Researchers have long studied lake sediments in a paleoclimatic context, carbon-dating pollen grains and other organic material to better understand past climates. Only recently have researchers been looking to these sediments to study earthquakes."
Although techniques exist to reconstruct seismic histories of megathrust and crustal earthquakes, determining the history—and potential recurrence—of intraslab earthquakes is trickier, as they often occur deep beneath the surface and leave little evidence in the geologic record.
Lake paleoseismology offers a solution. To assemble a long and detailed seismic history of a given area from turbidites in lake sediment, Brothers and colleagues needed to determine the minimum amount of shaking necessary to create those turbidites.
In Alaska, for example, the team used the magnitude-7.0 earthquake that occurred near Anchorage in November 2018 as a calibration point, taking sediment core samples and "compressed high intensity radar pulse" (CHIRP) sonar profiles from lakes at varying distances from the earthquake’s epicenter.
"Just like how trenching provides evidence about the timing of earthquakes along a fault onshore, CHIRP data is the equivalent to an 'acoustic trench' in places underwater that we can't physically dig through," said Brothers.
"Analyzing sediment cores allows us to ground-truth the layers we see in the CHIRP data; we're looking for stratigraphic layers that have been offset or deformed during earthquakes on a discrete fault."
Sediment cores are taken back to the PCMSC laboratory in Santa Cruz for analysis. Using sophisticated tools such as a CT scanner (similar to ones used in medical settings, but more detailed) and a multi-sensor core logger , researchers can "see" into the cores, discerning physical and chemical properties and identifying possible turbidites in the sediment layers. To determine approximate ages for the layers, samples are taken from sections of the core and sent elsewhere to dedicated labs for radiocarbon dating.
Splitting sediment cores in the PCMSC Coring Lab
“Seismometers and oral histories can tell us a lot about earthquakes that have occurred in modern history, but to get a better understanding of how a given fault might act in the future, we need to go back further into the seismic record,” said Drake Singleton, a USGS Research Geologist working with Brothers.
“Determining the minimum amount of shaking required to create turbidites is one way we can look at those older earthquakes.”
"Within the USGS, a lot of work is done in the Earthquake Hazards Program to develop a seismic hazard map for the nation, and this lake paleoseismology work contributes to that," said Peter Haeussler, Research Geologist with the USGS Alaska Science Center.
"The goal is to understand how often these seismic events occur—particularly the deep intraslab earthquakes—and thus create better seismic maps to predict hazard risks."
Haeussler's paleoseismic work in Alaska started in the 2000s, focusing mainly on marine paleoseismology. He turned his attention to lakes in 2015, when a team of Belgian researchers joined him in Alaska to study lake sediment cores for seismic clues. This technique was more widespread in Europe at the time, and the results from that 2015 project convinced Haeussler and others that lake paleoseismology could be applied widely, not just across Alaska but the entire country—even in the less tectonically active eastern U.S.
Lacustrine paleoseismology fieldwork in Virginia and Missouri
"Out West, researchers are looking in areas of known seismicity to get a more refined record of past earthquake activity. In the East, we're looking at whole regions of the country where there is little to no earthquake history at all," said Research Geophysicist Thomas Pratt with the USGS Geologic Hazards Science Center in Colorado.
In Virginia, Research Geologist Jessica Rodysill with the USGS Florence Bascom Geoscience Center is using lacustrine paleoseismology to characterize previously unknown or "cryptic" faults, looking in the soft sediments of lakes and reservoirs for tell-tale evidence of seismic events such as the 2011 magnitude-5.8 Mineral Earthquake .
"We're using similar techniques as the scientists studying lakes out West, and for the same reasons: to determine recurrence intervals of fault movement and assess seismic hazard risks," said Rodysill. "In our case, we're actively exploring areas with almost no paleoseismic records and looking for evidence of previously unstudied faults."
Back at the Pacific Coastal and Marine Science Center, fieldwork for summer 2022 is already being planned. The recent work in Alaska, Utah, and Washington was a success, said Brothers, and the equipment performed beyond the team's expectations.