Fast, simple new method for assessing earthquake hazard
about how often earthquakes of various magnitudes might occur in the area, and how large earthquakes might get?’” Rollins says.
When one tectonic plate pushes against another, elastic strain is built up along the boundary between the two plates. The strain increases until one plate either creeps slowly past the other, or it jerks violently. The violent jerks are felt as earthquakes.
Fortunately, the gradual bending of the crust between earthquakes can be measured at the surface by studying how the earth’s surface deforms. In a previous study (done in collaboration with Caltech research software engineer Walter Landry; Don Argus of the Jet Propulsion Laboratory, which is managed by Caltech for NASA; and Sylvain Barbot of USC), Avouac and Rollins measured ground displacement using permanent global positioning system (GPS) stations that are part of the Plate Boundary Observatory network, supported by the National Science Foundation (NSF) and NASA. The GPS measurements revealed how fast the land beneath L.A. is being bent. From that, the researchers calculated how much strain was being released by creep and how much was being stored as elastic strain available to drive earthquakes.
The new study assesses whether that earthquake strain is most likely to be released by frequent small earthquakes or by one very large one, or something in between. Avouac and Rollins examined the historical record of earthquakes in Los Angeles from 1932 to 2017, as recorded by the Southern California Seismic Network, and selected the scenario that best fit the region’s observed behavior.
“Estimating the magnitude and frequency of the most extreme events, which can’t be assumed to be known from history or instrumental observations, is very hard. Our method provides a framework to solve that problem and calculate earthquake probabilities,” says Avouac.
This new method of estimating earthquake likelihood can be easily applied to other areas, offering a way to assess seismic hazards based on physical principles. “We are now refining the method to take into account the time distribution of past earthquakes, to make the forecasts more accurate, and we are adapting the framework so that it can apply to induced seismicity,” Avouac says.
— Read more in Chris Rollins and Jean‐Philippe Avouac, “A geodesy‐ and seismicity‐based local earthquake likelihood model for central Los Angeles,” Geophysical Research Letters (27 February 2019)
