The characteristic earthquake model is based on the idea that over time, rupture segments and earthquake displacements at a point along a fault are fairly constant so all large earthquakes are similar in size and extent. This model has dominated fault behavior theory for the San Andreas fault because it generally agrees with several aspects of historic data. However, to test fault behavior models, long (1000 year) records of earthquake ages and slip along a fault are needed but an approximately 200 km gap in paleoseismic sites near the Big Bend prevents correlation of earthquakes along the southern San Andreas fault. In this project, research collaboration between Appalachian State University, the University of Oregon and the U.S.Geological Survey is filling this gap at site near Frazier Mountain in southern California. The goals of this investigation are to (1) extend the record of earthquakes to greater than 1000 years, (2) determine slip per event by detailed structural mapping of offset layers, and (3) develop a chronostratigraphic framework that combines traditional radiocarbon dating with paleoclimate indicators to provides additional temporal attributes for correlating earthquakes between sites. These data will contribute to earthquake rupture scenarios for the past 1500 years along the southern San Andreas fault for comparison with fault behavior models and add a dataset to the small pool of studies that permit tests of earthquake recurrence models and comparison of slip rates on different time scales.
Despite its fame, the San Andreas fault is poorly understood. For example, while sufficient data exist to forecast about a 59% chance of an earthquake larger than M6.7 on the southern San Andreas fault in the next 30 years, the portion of the fault that will rupture and the amount of slip that will be produced is unknown, in part due to a spatial gap in the available paleoseismic records. Paleoseismologic studies investigate recent deposits and offsets produced by large earthquakes to determine the age and displacement of past earthquakes. This study will result in paleoseismic data from this gap and which will be used to refine models of fault behavior and earthquake recurrence. Because these models underpin probabilistic seismic hazard analyses required by Federal and State agencies and which set insurance rates, building codes, and allocation of seismic hazard mitigation resources, adding this critical data will further scientific understanding of faults and ultimately reduce societal risk from earthquakes.
When large magnitude earthquakes occur on the San Andreas fault, the ground along the fault is torn, folded, and cracked. When soil and dirt subsequently covers the disrupted ground surface, evidence of past earthquakes is preserved underground. The goal of this project was to understand how often large earthquakes occurred in the last 2000 years by looking for evidence of this past destruction of the ground surface in buried sediments. To accomplish this goal, we excavated a network of trenches across an area of the San Andreas fault near Frazier Mountain, California. We photographed and mapped the layers of dirt, soil, and sand, searching for layers that were disrupted by pre-historic earthquakes. Organic material in the layers was dated with radiocarbon methods so that we can determine the age of past earthquakes. Our work shows that large, ground-rupturing earthquakes occur on the San Andreas fault at this location more often than previously thought: between ~500 AD and 1857 AD (an historic earthquake on the fault) there have been at least 11 earthquakes. This makes the average time between earthquakes only ~120 years, rather than the ~200 year average that was previously determined by a smaller study. Currently, we are working to determine the magnitude of the earthquakes recorded at the Frazier site by modeling deformation of sediments.
