U.S. Geological Survey Final Technical Report Award No
Total Page:16
File Type:pdf, Size:1020Kb
U.S. Geological Survey Final Technical Report Award No. G12AP20066 Recipient: University of California at Santa Barbara Mapping the 3D Geometry of Active Faults in Southern California Craig Nicholson1, Andreas Plesch2, John Shaw2 & Egill Hauksson3 1Marine Science Institute, UC Santa Barbara 2Department of Earth & Planetary Sciences, Harvard University 3Seismological Laboratory, California Institute of Technology Principal Investigator: Craig Nicholson Marine Science Institute, University of California MC 6150, Santa Barbara, CA 93106-6150 phone: 805-893-8384; fax: 805-893-8062; email: [email protected] 01 April 2012 - 31 March 2013 Research supported by the U.S. Geological Survey (USGS), Department of the Interior, under USGS Award No. G12AP20066. The views and conclusions contained in this document are those of the authors, and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. Government. 1 Mapping the 3D Geometry of Active Faults in Southern California Abstract Accurate assessment of the seismic hazard in southern California requires an accurate and complete description of the active faults in three dimensions. Dynamic rupture behavior, realistic rupture scenarios, fault segmentation, and the accurate prediction of fault interactions and strong ground motion all strongly depend on the location, sense of slip, and 3D geometry of these active fault surfaces. Comprehensive and improved catalogs of relocated earthquakes for southern California are now available for detailed analysis. These catalogs comprise over 500,000 revised earthquake hypocenters, and nearly 200,000 well-determined earthquake focal mechanisms since 1981. These extensive catalogs need to be carefully examined and analyzed, not only for the accuracy and resolution of the earthquake hypocenters, but also for kinematic consistency of the spatial pattern of fault slip and the orientation of 3D fault surfaces at seismogenic depths. The San Andreas fault system in southern California has a high probability of generating a major damaging earthquake. How big, when and where such an event will occur appears to depend on subtle, second-order variations in stress, strength and fault geometry, such as fault bends, offsets, changes in fault dip, or other fault discontinuities that seem to control fault segmentation and rupture behavior. As demonstrated by recent large earthquakes, rupture along major through-going faults in southern California can be quite complex, and strongly affected by both local fault geometry and the interaction with adjacent secondary structures, such as sub-parallel fault strands, left-lateral conjugate cross faults, tear faults, and low-angle basal detachments. Identification of these features, especially at seismogenic depths, often depends on careful kinematic analysis of the earthquake hypocenters and focal mechanisms in both space and time. For this project and in close collaboration with other members of the SCEC Community Model Development Team, detailed analysis of well-determined earthquake locations and focal mechanism slip planes was used to define the geometry of active subsurface faults and to create improved, more accurate, updated digital 3D fault surface models for the SCEC Community Fault Model (CFM). Newly revised 3D models for several major fault zones, including sections of the San Andreas from Parkfield to the Salton Sea, San Jacinto, Elsinore-Laguna Salada, Agua Tibia-Earthquake Valley, Garlock, Imperial-Brawley, Landers-Joshua Tree, San Gabriel, Oak Ridge-Northridge and the San Fernando/Sierra Madre faults, as well as for several major intervening cross faults, were developed. Special attention and more detailed studies were conducted in areas of particularly complex fault geometry. This included the San Gorgonio Pass region along the San Andreas fault zone, and the Laguna Salada and adjacent Sierra Cucapah region highlighted by the 2010 M7.2 El Mayor- Cucapah sequence. These new models allow for more non-planar, multi-stranded 3D fault geometry, including changes in dip and dip direction along strike and down dip, based on the changing patterns of earthquake hypocenter and focal-mechanism nodal plane alignments, rather than projecting faults to depth assuming a constant (often vertical) dip and dip direction, as was the case for many of the previous preliminary fault models in CFM. In addition, the CFM database was reorganized with a new hierarchical name and numbering system that allows CFM to properly organize, name and number the increasing variety and complexity of modeled, multi-stranded principal slip surfaces and associated secondary faults. These new CFM fault surfaces describe a more complex pattern of interacting fault sets at depth accommodating crustal deformation in southern California, and have already provided the basis for improved dynamic rupture models, variable fault slip rate models, and correlations with associated regional topography along major active faults of the San Andreas fault system. 2 Introduction Accurate assessment of the earthquake hazards in southern California requires accurate and complete 3D fault maps. Many aspects of seismic hazard evaluation, including understanding earthquake rupture and geodetic strain, developing credible earthquake rupture scenarios, modeling geodetic and geologic fault slip rates, or predicting strong ground motion, are all strongly dependent on accurately resolving the 3D geometry of active faults at seismogenic depths [Herbert and Cooke, 2012; Lindsey et al., 2012, Oglesby and Mai, 2012; Shi et al., 2012]. In southern California, these structures can be quite complex and include both relatively high-angle strike-slip and oblique-slip faults—many of which intersect and are mapped at the surface, as well as low-to-moderate-angle faults, some of which are blind and are difficult to identify. The 1994 M6.7 Northridge earthquake was on one such blind fault and produced over $40 billion in damage. Because of this, both the Southern California Earthquake Center (SCEC) and US Geological Survey NEHRP have consistently targeted the identification and mapping in 3D of active faults in southern California as a primary research objective. The purpose of this project was to provide just such improved, more detailed, updated digital 3D fault representations of major faults in southern California for input into the online SCEC Community Fault Model (CFM) and USGS Quaternary Fault Database (Qfault). Recently, a considerable effort has been focused on developing improved catalogs of earthquake hypocenters and focal mechanism solutions for southern California (e.g., Fig.1) [e.g., Lin et al., 2007; Hauksson et al., 2012; Yang et al., 2012]. These catalogs comprise over 500,000 relocated hypocenters and nearly 200,000 well-determined earthquake focal mechanisms since 1981. These extensive catalogs need to be carefully examined and analyzed—not only for the accuracy and resolution of the earthquake hypocenters [e.g., Richards-Dinger and Ellsworth, 2001]—but also for the spatial pattern of fault slip and the orientation of 3D fault surfaces. As demonstrated by the 2010 M7.2 El Mayor-Cucapah earthquake and other events, rupture along major through-going faults in southern California can be quite complex and strongly affected by both local fault geometry and the interaction with adjacent secondary structures, such as sub-parallel fault strands, left-lateral cross faults, and basal detachments [e.g., Nicholson et al., 1986; Hudnut et al., 1989; Seeber and Armbruster, 1995; Nicholson, 1996; Carena et al., 2004; Lin et al., 2007; Nicholson et al., 2009, 2011, 2012; Hudnut et al., 2010; Fletcher et al., 2010]. Identification of these important fault features prior to a major earthquake, especially at seismogenic depths, often depends on careful analysis of microearthquake hypocenters and focal mechanisms in both space and time. To help address these issues, this project was part of an on-going, multi-year collaborative effort within SCEC to systematically upgrade and improve the Community Fault Model (CFM). Since 2010 and working in close cooperation with Andreas Plesch, John Shaw, and Egill Hauksson, we have continued to make steady and significant improvements to the CFM and its associated fault database [Nicholson et al., 2010, 2011, 2012; Plesch et al., 2010]. These improvements include more detailed and complex 3D representations of major active fault systems (Fig.1), additional detailed fault surface trace data, and finalization of our new naming and numbering scheme for CFM that allows for closer links to the USGS Qfault database. A systematic revision of CFM fault models was triggered by unexpected discrepancies between previous CFM-v.3 fault representations and the newer Qfault surface traces, as well as by the availability of the extensive relocated earthquake catalogs to better define the complex subsurface geometry of active faults. Although a draft version of CFM-v.4 was sent out for review and comment, this upgrade to CFM-v.4 is not yet complete. Many fault models in CFM-v.3 still need to be re-registered to the more detailed Qfault surface traces and, together with recent relocated hypocenters, require newer, more complex and more realistic 3D fault models for CFM. Such on-going efforts to update the 3D fault set for CFM are critical to the USGS NEHRP research priorities. This is because better knowledge of active 3D 3 fault geometry and improved 3D fault representations are fundamental to a