Paleoseismicity Research: From Trench to Earthquake Hazard Map

View fullsize

by Carol Hasenberg

Dr. Ashley Streig, Assistant Professor in the Geology Department of Portland State University and NSF Postdoctoral Research Fellow, gave a talk to GSOC on February 12 about her paleoseismology research on the San Andreas Fault in California. 

Streig began by describing the seismicity of the United States looking at a USGS earthquake probability map (ie., acceleration contours with a 2% chance of exceedance in 50 years). These maps are updated by the USGS every few years, and are used to develop seismic resistant design regulations and other earthquake mitigation efforts. Looking at the map, Streig compared the seismicity of Oregon to that of California, where California has a bit higher shaking hazard. She described the data needed to generate such a map as a preface to describing her research on the San Andreas Fault.

In generating the earthquake probability map, seismologists consider the earthquakes that are generated on faults that can affect the map area. Data needed to analyze shaking hazard associated with the faults are the average earthquake displacement along the fault, the rupture length, the timing of earthquakes on the fault and the style of deformation associated with the earthquakes. The magnitude of individual quakes can be determined from the first two of these data. Paleoseismicity, the study of the geologic record of past earthquakes, combined with historic earthquake records give seismologists this data so they can determine the earthquake shaking hazard.

Streig also mentioned paleoseismicity research efforts done in Oregon to determine the seismicity of the Cascadia Subduction Zone with which the audience may have been familiar. She discussed the research on turbidite deposits that has been conducted by Dr. Chris Goldfinger of Oregon State University. Using underwater coring techniques, Goldfinger obtained a stratigraphic record of all the turbidite (ie., underwater landslide) deposits in the underwater canyons of the Oregon coast, and by dating and correlating the deposit records, assembled a comprehensive record of large subduction zone earthquakes on the Oregon coast during the last several thousand years. Other types of paleoseismic records along the Cascadia Subduction Zone that have been researched include fossil remains from ghost forests and salt marsh subsidence.

Getting back to her research on the San Andreas fault, which is exposed on dry land, Streig discussed her use of trenching equipment to expose fault rupture zones in stream valleys along the fault zone. Low lying land accumulates sediment which can be cross-sectioned by trenching and its geologic history interpreted. A number of possible cross section scenarios were presented in a diagram and discussed by Streig. An ideal cross section shows a fault which has offset stratigraphy delineated by a clean fault line which is somewhat rounded at the top of the fault scarp which was exposed and eroded immediately after the quake. A colluvial wedge of material eroded from the material above the scarp will further emphasize the depiction of the effect of a single quake.

Streig’s research in the Hazel Dell site in the Santa Cruz Mountains was interpreted as showing the results of 3 significant earthquakes happening since the land was settled and logged by the first Spanish settlers.

Several conditions can complicate the picture for the interpretation, such as the occurrence of liquefaction in wet, loose alluvial soils during the quake. Liquefied sediment will appear folded and wavy in the cross section, and sand volcanoes can be seen cutting through the liquefied sediment to a patch of sand on the historic surface at the time of the quake.  The San Andreas is a strike/slip fault, and fault displacement will have a significant horizontal component. The rupture from an earthquake may be just be a crack which fills with colluvium in this case.

Streig’s research in the Hazel Dell site in the Santa Cruz Mountains was interpreted as showing the results of 3 significant earthquakes happening since the land was settled and logged by the first Spanish settlers. The settlement strata were identified by the presence of logged stumps which the researchers were able to carbon date as having lived into the late 1700’s. She was able to use the stratigraphic and historic records to date the earthquakes as having occurred in 1838, 1890, and the 1906 San Francisco quake. Correlation with other sites gave a rupture length of about 100km for the 1838 quake, 25km for the 1890 quake, and 470km for the 1906 quake. Streig compared this to the previously accepted seismicity model for the region of one event similar to the 1906 quake occurring approximately every 200 years. Clearly at least 2 modes of strain relief behavior is occurring on this section of the San Andreas fault.

Streig described some further research she has done on the southern San Andreas fault near Frazier Mountain. She had a broad site which was bracketed by two parallel fault expressions, called a releasing stepover by seismologists. In this large site they used a cone penetrometer to determine the stratigraphy over the bulk of the site. Infill stratigraphy from 3 earthquakes were determined from the testing. The most recent earthquake (1857) with 5 m of slip shows 1 meter of sag across the stepover. Older earthquakes had nearly equal dimensions to this one. Streig’s conclusion was that either the earthquakes were similar in size or some other physical constraints in the site can only accommodate a specific amount of slip.

The siting of the Meers Fault test trench was accomplished with the help of LIDAR imagery which can show the fault characteristics in great detail, and has been a great tool for seismologists since the 1980’s.

Streig then described some recent work she has done on the Meers Fault in Oklahoma. This fault is 35 km long and generates earthquakes up to M6.5. This may not seem to be impressive when compared to the large fault zones in the far west, but earthquake waves transmitting through the continental craton of the Midwest and East travel much farther in distance because of the more solid base rock. Also, wastewater is being injected underground in the area and some of these injection wells are near the end of the fault zone. It is feared that the wells may increase the slip on the fault and up the potential magnitude to M7.2.

The siting of the Meers Fault test trench was accomplished with the help of LIDAR imagery which can show the fault characteristics in great detail, and has been a great tool for seismologists since the 1980’s. The results from the trench shows at least 2 earthquakes have occurred in last 2900 yrs.

Streig’s work is an important part of the research being done to educate scientists, engineers, legislators and citizens of the potential hazards of earthquakes. Through her work and the work of similar researchers hazards can be quantified and plans can be made to keep our dwellings and lifelines resistant to damage.