Electronic Encyclopedia of Earthquakes System Developer

Project Description:
Intern(s): Brion Vibber
Mentor(s):

Mark Benthien, University of Southern California

Seismic Sleuths Flash Project: Mercalli Intensity Activity

Project Description: I am helping Robert deGroot in reworking Seismic Sleuths for future distribution. My focus will mainly be on doing the interactive computer portion of the new Seismic Sleuths using Macromedia Flash MX. Currently, I am learning Macromedia Flash as I work on the first of several interactive activities. The current activity that I am working on is a computer-based Mercalli Intensity Scale activity -- studnets will be able to select a zipcode region and then color it according to what they believe the intensity is based on a description of what was felt there. There are other projects proposed for the future including animations for P and S waves, the “Building a Model Wall” activity in Seismic Sleuths (4.2, Seismic Sleuths), an animation of the BOSS model, and a “Finding the Epicenter” activity as well.
Intern(s): Glenn Song
Mentor(s):

Mark Benthien, University of Southern California

Researcher

Project Description:
Intern(s): Steve Shim
Mentor(s):

Mark Benthien, University of Southern California

Analysis of Regional Seismicity in two areas related to the Pacific-North American Transform Plate boundary

Project Description: The aim of this study was to examine the spatial and temporal distribution of seismicity in two areas along the Pacific-North American transform plate boundary in southern California. Using data from the Southern California Earthquake Center online database (http://www.scecdc.scec.org/catalog-search.html), we created 3-dimensional plots of the yearly seismic activity in an area north of the San Andreas Fault and east of Cajon Pass surrounding Silverwood lake (N latitude 34.34 S latitude 34.25, E longitude –117.235 W longitude –117.4375), as well as an area just south of the southern termination of the Clark Fault extending south along the western margin of the Salton Trough (N latitude 33.3 S latitude 33, E longitude –115.75 W longitude –116.25). The data examined spanned nearly two decades from August 1, 1983 to August 1, 2001 and were grouped in seventeen plots from August 1 of one year to August 2 of the following year using the RockWorks 2002 and RockPlot3D programs. Analysis of the Salton Trough region displays persistent microseismicity in the area just below the southern termination of the Clark fault, suggesting a blind extension of the fault farther south than previously mapped. This idea is consistent with the earlier hypothesis of Janecke et. Al. 2001. The data also shows strong correlation between activity along the southern Clark extension and nearby seismicity such as the Superstition hills event. Analysis of the Silverwood Lake data shows startling aseismicity in that area, despite numerous visible faults apparently aligned along the principle stress tensors in the region. The comparison of the two areas over a concurrent 18 year period raises interesting questions about the seismic behavior of the entire southern California region.
Intern(s): Elliot Aguilar
Mentor(s):

James Evans, Utah State University

Towards Development of a Vertical Motion Database for Southern California

Project Description: Towards the development and population of a database on long-term vertical velocities, we began compilation of information on the ages and elevations of marine and fluvial terraces in coastal southern California, and rock uplift information derived from low-temperature thermochronology. Surface uplift rates are calculated as the difference between a terrace’s current elevation and the elevation at which it formed, relative to today, divided by the terrace age. We identify three primary sources of uncertainty: terrace age, elevation of the marker feature (ie. paleo-shoreline), and elevation of the reference frame (paleo-sea level or river channel level). For terrace age, we compiled all relevant information including radiometric age (C-14, U-series or other age data, including all isotopic ratios), amino-stratigraphic information, faunal assemblage zoogeographic data including inferences from southern and northern extra-limital species, and other age data (cosmogenic, OSL, TL dates). Elevation data are compiled from primary sources and include both surveyed and topographic elevation information on the shore platform, paleo-shoreline or terrace tread. Estimates of paleo-sea level are potentially the most problematic, as California observations appear somewhat at odds to estimates from Barbados and New Guinea. We have developed three alternative sea level models that we test against the terrace observations, all of which agree at the 0.1 mm/yr level. All terrace information has been entered into a Geographic Information System database (Arc/INFO), and geo-referenced into a common coordinate system. We have included available data on modern vertical uplift rates, including continuous Global Positioning System data from the Southern California Integrated GPS Network, Electromagnetic Distance Measurements (EDM) from the USGS, and tide gauge data from NOAA. An experimental web site has been developed to distribute the vertical uplift information to the SCEC community via a map-driven interface utilizing Arc Internet Map Server. Currently, all data can be viewed, printed, and queried through the web interface. Advanced modeling objects, allowing SCEC community members to calculate vertical uplift rates for specific terraces using user-specified sea level curves, or denudation rates with specified thermal models, and to compare long-term and recent uplift rates from multiple data sets are under development.
Intern(s): Nicholas Campagna
Mentor(s):

Mike Oskin, University of California, Davis
Nathan Niemi, University of Michigan

Effective Seismic Risk Communication for a SCEC Target Audience: Are Safety Elements Too Shaky?

Project Description: We are conducting a study of the type and level of earthquake hazard mitigation efforts employed by Orange County cities. Results will provide an overview of local mitigation practices and identify areas where seismic risk communication activities may be most effective. The study is focused on evaluating the effectiveness of previous SCEC activities and products in communicating seismic risk at the municipal level. Orange County is well suited for this study because it contains diverse sociologic, geologic, and seismic conditions. Orange County is particularly at risk because it ranks second in California counties by total population, and approximately 40% of the housing stock was built before 1970, which is prior to substantial upgrades in seismic building practices. Using HAZUS methodology, the CDMG (CGS) estimated Orange County's expected annualized total loss due to earthquake activity to be among the highest in the state. Our study focuses on cities because they represent a key component of risk communication and mitigation. A substantial amount of policy implementation, compliance, and enforcement occurs at the municipal government level, so it is important to understand how cities utilize seismic hazard information to mitigate risk. We are conducting a cross-sectional survey of Orange County's 34 Safety Elements and related documents (Technical Background Reports, EIRs and MEAs). These documents identify hazards to public welfare and provide guidance for local decisions on zoning, subdivisions and permitting. To date, we have compiled available documents for all 34 cities, and conducted a preliminary assessment by tabulating the dates of Safety Element adoption, and reviewing text and references in the Seismic Safety portions of the Safety Element. Our preliminary observations reveal substantial variation in the treatment of seismic hazard assessment, planning, and mitigation among Orange County cities. Six cities have not revised or created Safety Elements since the founding of SCEC in 1991, and therefore do not utilize any SCEC products. The remaining 28 cities have revised or updated their Safety Elements since 1991. At least 5 of these cities cite SCEC products directly as a source of seismic hazard information. The other cities have not fully utilized SCEC products, and may be the good targets for future seismic risk communication and mitigation efforts.
Intern(s): Kristin Iriarte
Mentor(s):

Lisa Grant Ludwig, University of California, Irvine

Moment Release from Paleoseismic Histories: Interaction between the Eastern California Shear Zone, San Andreas, San Jacinto and Elsinore Faults

Project Description: Analysis of earthquake sequences in southern California and other regions at timescales of 1000 years suggests triggering or suppression of earthquakes from elastic stress changes is a common phenomena. Elastic stress changes from an individual earthquake depend on its magnitude, the orientation and sense of slip on the fault and the strength of the brittle crust. Elastic stress changes appear to affect nearby faults for decades, in contrast to visco-elastic stress changes that extend for much greater distances away from the fault that slipped in an earthquake (i.e. the width of the plate boundary in southern California). Visco-elastic stress transfer may also occur over long time periods (e.g. a century or more). As a summer SCEC intern working with Karl Mueller and Tom Rockwell, I compiled all published and many unpublished paleoseismic records for major strike slip faults in southern California in order to begin to assess the following questions.

1. Can clusters of earthquakes that occur over periods of centuries be related to elastic or viscoelastic stress transfer models, given uncertainties about rupture endpoints and the amount of slip that occurred during the event? In short, I seek to extend the moment release backward in time from historical records and the period of seismograph networks.
2. Do lower order patterns in moment release exist and do these patterns occur at timescales appropriate for elastic or viscoelastic stress triggering or suppression? If this is not the case, then what other processes may occur to govern or affect moment release at 1000 year or greater timescales.

Compilation of paleoseismic data and construction of event plots for strike slip faults across the Southern California plate boundary suggests the following.

1. Clusters of earthquakes separated by much longer time intervals is more the rule, than the exception for strike slip faults in southern California.
2. Little long-term correlation appears to exist between the Mojave and Indio segments of the San Andreas fault (SAF RI=100’s of yrs) and Eastern California Shear Zone (ECSZ RI=1000’s of yrs).
3. Events along the San Jacinto fault (SJF) occur at higher frequency than the ECSZ, but still have higher RI than the SAF. Interestingly, slip rates on the SAF and SJF appear to be inversely correlated (SAF slows in last millennia, SJF speeds up).
4. The Elsinore fault ruptures at longer RI (~500 yr) than either the SJF or SAF.
5. The determination of slip rates may hold little hope for understanding modern cycles of moment release, given the remarkable variation in rate for some faults, such as the Garlock.

Given these results, it appears that other processes, besides simple elastic or viscoelastic stress transfer govern first-order patterns of moment release in southern California. Subsequent efforts will be aimed at modeling stress transfer of higher second order patterns (e.g. interaction between faults that release moment in clusters at less than 1000 year timescales) and attempting to develop concepts for transfer of plate motions from the upper mantle and lower crust with brittle moment release along discrete faults. Initial forays at millennial and greater timescales will be aimed at testing models of brittle faults that are underlain by similarly oriented, but plastic shear zones vs. brittle faults offset horizontally at the 250o geotherm (i.e. the onset of crystal plasticity) from nearby plastic shear zones vs brittle faults that overlie a smoothly shearing upper mantle and lower crust.
Intern(s): Ryan Tolene
Mentor(s):

Karl Mueller, University of Colorado, Boulder

3000 Years of Ground-Rupturing Earthquakes int he Anza Seismic Gap, San Jacinto Fault, Southern California: Time to Shake it Up?

Project Description: An array of excavations at Hog Lake (HL) on the San Jacinto fault (SJF) near Anza, California exposes 3000 years of stratigraphy containing evidence for 9-10 surface ruptures. The site, an ephemeral pond that has recorded successive periods of deposition and desiccation, lies astride the principal active strand of the SJF. Evidence of surface ruptures includes abrupt upward truncation of faults and fissures, angular unconformities and disconformities due to folding, liquefaction, and growth strata on the down-thrown (NE) side of the fault. The most recent event (MRE) is preserved at 0.5 m depth and is overlain by undisturbed, stratified sediment and a well-developed surface soil. Reported observations of no surface rupture at HL resulting from the 1918 earthquake, analysis of the historical earthquake record for the region, and the absence of pollen from introduced plant species overlying the event horizon all suggest that this event is pre-1850 and most likely prehistoric. We acquired an initial suite of 25 C-14 ages on seeds, charcoal and shell to compare results from different media. We found that the dates on seeds collected from individual strata yielded results consistent with stratigraphic relations with no apparent reworking or recycling. We acquired an additional 25 dates on seeds to determine the ages of the past six surface ruptures. These dates show that the Anza segment has ruptured six times in the past 1000 years, with event ages at about AD 1760, 1630, 1360, 1290, 1230, and 1020. Prior to about 1000 AD, which is the age of the oldest section on the downthrown side of the fault, it appears that either there were fewer earthquakes in the preceding 2000 years, as we recognize only four additional events, or that the record is incomplete using only the stratigraphy exposed on the southwest side of the fault. Additional dating is being conducted to further date these earlier events. These observations indicate that the time between surface ruptures at Anza is relatively short, ranging between 60 and 270 years with an average recurrence interval of about 190 years. As it has been about 240 years since the most recent event, it appears that the Anza segment may be ready to shake, rattle and roll.
Intern(s): Danielle Verdugo
Mentor(s):

Thomas Rockwell, San Diego State University

Finite Element and Boundary Element Benchmarks for the Post-Seismic Deformation

Project Description: We present numerical solutions to several benchmark problems illustrating instantaneous and delayed deformation of the Earth's crust caused by large earthquakes. The first set of benchmarks describes visco-elastic relaxation due to a strike-slip fault. To facilitate comparisons between various numerical methods, we assume a geometrically simple source (a finite rectangular dislocation with a constant slip) in a semi-infinite medium. One of the major limitations of the finite element (FE) models is the treatment of the remote boundary conditions. We tackle this problem by simulating a visco-elastic response of a Maxwellian half-space underlying an elastic layer using meshes with an increasing node spacing away from the source, and infinite elements at the external boundaries of the computational domain. We use a menu-driven pre-processor APMODEL to specify the problem geometry, and generate the input files for the finite element codes. At present, APMODEL allows one to generate input files for the 3-D finite element codes ABAQUS and TEKTON. The fully relaxed solution for the strike-slip fault problem is also obtained using a boundary element (BE) approach. The BE solution uses an elastic half-space model with an imposed stress-free boundary condition at the brittle-ductile transition. The stress-free boundary condition is enforced by adding stress-controlled boundary elements at the brittle-ductile interface. The FE and BE solutions are in good agreement, indicating that our benchmark is robust. Other benchmark problems include the case of a dip-slip fault, effects of gravity, and poro-elasticity. We also compare the efficiency and accuracy of several BE codes, including DIS3D, Interact, and POLY3D. Our numerical benchmarks, and the pre-processing software (e.g., interactive mesh generators, converters of input files between different BE codes) may be of use for researchers studying the quantitative aspects of earthquake-induced deformation.
Intern(s): Dave Huang
Mentor(s):

Yuri Fialko, University of California, San Diego

Three-Dimensional Geometry of Buried Fold Scarps Associated with Ancient Earthquakes on the Puente Hills Blind Thrust Fault: Preliminary Observations

Project Description: The Puente Hills thrust fault (PHT) is a large blind thrust fault that extends east-west beneath the heart of the metropolitan Los Angeles region (Shaw and Shearer, 1999; Shaw et al., 2003). Christofferson (2002; in prep.) and Dolan et al. (2003) identified four buried fold scarps associated with large (M7), ancient earthquakes on the PHT beneath the City of Bellflower, in northern Orange County. One of the major outstanding questions regarding this research concerns the subsurface, three-dimensional geometry of these buried scarps. Specifically, we want to determine the extent to which the subsurface geometry of these scarps is controlled by tectonic versus fluvial processes. In order to begin addressing these questions, we drilled a north-south transect of hollow-stem, continuously cored boreholes across the buried fold scarps. This new borehole transect, which comprises six, 20-m-deep boreholes, was drilled parallel to, and ~ 100 m west of, the original Carfax Avenue transect of Christofferson (2002) and Dolan et al. (2003). We have just begun analysis of the cores, and have reached no final conclusions regarding these new data. Our preliminary examination, however, does allow us to note that the overall pattern of progressive southward thickening of sedimentary units observed in the Carfax borehole transect extends westward to the new transect. Several key sedimentary contacts that are traceable laterally between the two transects: (1) occur at approximately the same depths at all locations along both transects; and (2) define buried fold scarps similar to those used to identify ancient earthquakes in the Carfax transect. These observations confirm that the buried scarps are primarily tectonic, rather than fluvial features. Moreover, the lateral continuity of these features indicates that they provide useful paleoseismologic information.
Intern(s): Jeffrey Hoeft
Mentor(s):

James Dolan, University of Southern California