|
Substantial efforts were expended in primarily four areas during 1999. Three of the these were planned activities along the San Andreas fault, in Los Angeles basin, and in peripheral areas of high interest, as described in last years RFP. The fourth relates to SCEC involvement in post earthquake surface rupture studies conducted after three M7-7.5 earthquakes that occurred in the fall of 1999.
San Andreas Project SCEC funded six paleoseismic projects along the San Andreas fault this past year that significantly clarifies our understanding of the late Holocene behavior of California's master fault. From south to north, these projects are in the Salton Trough (Seeber), at Burro Flats (Sieh/Yule), Plunge Creek (McGill), Elizabeth Lake (Dolan), Frazier Mountain (Rockwell/Lindvall), Wallace Creek (Sieh/Rubin), and Cholame Valley (Arrowsmith/Grant).
Deformational style in the Salton Trough was modeled by combining geologic and microseismicity data and considering both rotational and irrotational models. The rotational models fit the data better and support the idea that significant seismic production in the Salton Trough region can be attributed to the cross-faults. This supports paleoseismic observations of fewer than expected earthquakes on the southern San Andreas fault during the past 1-2 ka. Model results also suggest between 7-10 mm/yr of spreading in the Brawley seismic zone.
At Burro Flats, trenching continued by deepening existing exposures and developing new ones. The new trenches confirm earlier results and extend the record of large surface ruptures back to about 3500 years, with 5-7 large surface ruptures occurring over this time period. The last large surface rupture may have been nearly 900 years ago, although continuing work this fall has potentially identified a younger strand. In any case, it appears that large earthquakes occur less frequently at Burro Flats than on sections of the San Andreas fault to the north and south. It also appears that this section of the fault is ripe for failure.
To the north on the San Bernardino section of the fault, trenching continued at the Plunge Creek site. The existing trench was deepened, and a new one was opened. New 14C dates confirm that the most recent event is between A.D. 1440 and 1640, and that no surface rupture occurred at this site in 1812 or the 1690/1700 event observed at Pitman Canyon. This puts strong limits on the southern extent of both these events, and suggests a longer return period than at Pitman Canyon or Wrightwood, but shorter than at Burro Flats.
Along the 1857 rupture near Ft. Tejon, new results at the Frazier Mountain site confirm that the penultimate (pre-1857) rupture occurred between A.D. 1400-1650. This is probably event V at Pallet Creek, and may be the same as the most recent event at Plunge Creek. These results place an upper bound on the possible rupture length of the 1812 earthquake at 200 km, based on the Plunge Creek and Frazier Mountain results. Evidence for 1812 is present at Pallet Creek, Wrightwood and Pitman Canyon, allowing for a minimum rupture length of 70 km.
Finally, new excavations in the northern Carrizo Plain, south of Cholame, exposed evidence for four events since about A.D. 1100. This is inconsistent with prior work in the Carrizo plain farther south at Bidart where five events were documented for the same time period. Thus, some of the Bidart events may have been smaller. Work also continues near Bidart to resolve this, and the new northern Carrizo site has the potential of resolving slip on buried gravelly stream channels, so work will also continue there.
One of the more provocative observations from this new trenching is that the most recent earthquake appears to be small relative to the penultimate event, based on the extent of damage to the sedimentary section and width of faulting. The most recent event was assumed to be 1857, but these observations suggest that 1857 is the penultimate event in the northern Carrizo Plain. If true, then the most recent surface rupture may have been the 1881 "Parkfield" event rupturing the Cholame segment due to its increased stress from high 1857 slip in the Carrizo Plain. Current work is focussing on using pollen of introduced species, such as Eucalyptus, to try to better date the sediment involved in these two events and resolve if the most recent event is 1857 or possibly 1881. Work also continues in reinterpreting the early geodetic surveys of the area to see if sufficient data can resolve post 1855 slip. The conclusions of this study will rewrite the recent history of the central San Andreas fault south of Parkfield, and possibly revise our expectations for a Parkfield type event triggering an M7 along the Cholame segment expect in the near future.
Figure 4 is a summary correlation diagram that depicts the current state of paleoseismic knowledge along the San Andreas fault in southern California. Annual updates of this and similar diagrams guide efforts in resolving the most pressing and outstanding issues on segmentation, slip in prior events, repeatability, and likely future behavior.
The Los Angeles Basin Several focussed studies on tasks in the Los Angeles basin further resolved the seismic potential of several major faults. New work on the Sierra Madre fault (Dolan) confirm the long return period for this important fault, and place the most recent earthquake at greater than 8,000 years ago. A minimum of 11 m of thrust slip was also resolved to have occurred between 8 ka and 23 ka, although there is not yet evidence on how many events produced this slip. These results are consistent with the Lincoln Park trench results (Rubin et al., 1998) that demonstrate two events in the past 15 ka, producing about 10 m of slip. The Sierra Madre fault apparently produces large infrequent earthquakes, and the most recent event was 8000-9000 years ago. Combined with its late Pleistocene to present slip rate of 0.5-1 mm/yr (Walls et al., 1998), a near-future large earthquake on this fault would not be a surprise.
New work on the Verdugo fault (Dolan) limits the uplift rate to be quite low. A series of bucket auger borings exposed well-weathered soil (>50-100 ka) on granitic bedrock to at least 75 m south of the topographic break in slope. These observations indicate that either the main fault is deeply buried to the south, or the rate of vertical slip is very low. In the latter case, there is the possibly that the recent behavior of the Verdugo fault has been predominantly strike-slip, so more work is needed to resolve the faults kinematics. In the eastern Los Angeles basin, considerable progress was made on resolving the complex structural and kinematic history of the Elysian Park and Puente Hills thrust systems. Detailed work along the Coyote Hills uplifts (Yeats/Bjorkland) shows that the south side of the hills are bound by a high-angle (60-79oN) fault with about 1.3 km of left-lateral slip that accrued during the Miocene. These same faults (shown on Wright's 1991 cross-sections in Figure 9 of his report as the Norwalk fault) were reactivated as reverse faults during the Pliocene, based on deformation of the Fernando Formation. Of note is that these faults deform the Pico strata but not the San Pedro Formation, so are no longer active. Thus, the "Norwalk fault", as mapped, is not a potential seismic source. Current uplift of the Puente Hills is now accommodated by the blind Puente Hills thrust (Shaw and Shearer, 1999), which has produced growth of folds seen in the San Pedro Formation as well as the overlying aquifers (Yeats).
A seismic transect was produced across the Los Angeles basin and San Pedro bay, and includes further resolving the architecture of potential blind thrust sources (Shaw). This is the first time such a contiguous transect has been produced across the basin, allowing for analysis of individual structures in a more regional context.
Elsewhere in the basin, significant new work was accomplished on the Hollywood fault (Rockwell/Lindvall/Yeats). Subsurface work has further defined a pull-apart basin between the Santa Monica and Hollywood faults (Yeats), compatible with left-lateral lateral Quaternary slip. Near subsurface work on secondary faults in West Hollywood near the pull-apart basin (Rockwell/Lindvall) show that most have normal or normal-oblique senses of slip, also consistent with left-slip across a transtensional basin. Furthermore, identification of marine terraces in the Hollywood Hills (Rockwell/Lindvall) and in the basin on the south side of the fault (Ponti) indicate a differential uplift rate across the Hollywood fault of only 0.1-0.15 mm/yr over the past several hundred thousand years.
Other Significant Project Results Work was completed on the structural interpretation of the uplift of the onshore San Joaquin Hills (Grant et al., 1999). New work indicates that the most recent uplift event occurred after A.D. 1635 and produced an average of 1.7 m along the San Joaquin Hills, with a maximum uplift of as much as 3.5 m (Grant). There is limited evidence that this event may have produced a tsunami. Continued structural work on the offshore faults in this area (Shaw and Mueller) has led to the recognition of a more extensive zone of thrusting adjacent to the offshore Newport-Inglewood-Rose Canyon fault zone. Work continue to define whether this represents a significant seismic source or is secondary to the strike-slip deformation.
In the Santa Barbara region, work was completed on the earthquake hazard of the Santa Barbara fold belt (Keller/Gurrola), where all potential seismic sources are believed now to be identified. The rate of coastal uplift has been determined for the entire Santa Barbara coastline, and magnitudes for potential earthquakes have been suggested. Trenches and borings were completed that showed three events cutting the 30-40 ka Stage 3 marine terrace and its sediments. The marine abrasion platform is deformed by about 15 m of vertical structural relief, and it appears that recent deformation is blind. The Rincon Creek fold, which is propagating west into City of Carpenteria, may hold the potential for developing a complete and detailed time-history of folding. Finally, the controversy on the structural interpretation of the Mission Ridge anticline was solved by identification of the backlimb, which supports the presence of a blind thrust beneath Mission Ridge in Santa Barbara.
Studies on the effects of pressure solution were also undertaken (Vermilye and Seeber) continue exploring the significance of this process on seismic hazard assessment. They demonstrate that pressure solution records the effects of stress, strain, and fluid activity in the evolution of an active and young compressional structure. They also demonstrate that the strains contributed by pressure solution are significant. Specific results and current lines of research include: 1) Distributed grain-scale shortening fabrics and solution cleavage accommodate shortening in the earliest stages of compression, before the onset of folding. Thus, pressure solution seems to be the main mechanism for strain in the most external portions of the orogenic belt at the leading edge of the accretionary wedge. 2) Pressure solution continues to operate in parallel to other deformation processes after the initial layer-parallel phase. This is suggested by the correlation between cleavage orientation and current stress/strain directions. Progressive pressure-solution strain, however, may continue to develop the original layer-parallel cleavage fabric. 3) We have determined that 5-23% of the total strain is accommodated by grain-scale dissolution. This is a minimum for aseismic strain because additional strain is taken up by removal of material on cleavage planes and by aseismic slip on faults, particularly by numerous small faults. Our current task is to quantify the latter. And 4) The density of small faults characterized by aseismic mineral fiber growth increases in the vicinity of the San Cayetano fault. This may reflect a combination of enhanced fluid flow, pore pressure, and shear stress near the fault, possibly at the time of large ruptures. Studies to compare detailed profiles of cleavage plane density, grain boundary strain, and aseismic fault slip across the fault are planned.
Finally, work continued on a listric thrust model (Sorlien/Seeber) for the Santa Monica and Channel Islands thrusts. Correct modeling of the geometry of these thrusts is required to correctly assess the inferred slip rates and seismic hazard.
The SCEC Earthquakes SCEC geologists joined with scientists from the US Geological Survey to map and measure slip associated with the surface rupture of the August 17, 1999 Mw7.4 Izmit earthquake. SCEC also joined with geologists from the US Geological Survey and the California Division of Mines and Geology to map the surface rupture associated with the October 16 Mw7.1 Hector Mine earthquake. Finally, SCEC geologists also participated in mapping the surface rupture and measuring slip produced from the Mw7.1 Duzce earthquake of November 12, 1999. The results from each of these studies is summarized below.
The August 17, 1999 Izmit Earthquake in Turkey The August 17 Izmit earthquake was produced by rupture of a 120 km-long portion of the North Anatolia fault in northwestern Turkey. This earthquake is significant for a number of reasons. First, the event was forecast by Stein et al. (1996) based on the prior activity of the North Anatolian fault this century, and understanding its current and past seismic activity may provide valuable insights into fault behavior that may be applied back to active faults in the United States. Second, the North Anatolia fault is similar to the San Andreas fault of southern California, so further understanding of fault segmentation and the role that step-overs play in limiting extent of rupture are important issues in assessing seismic hazard in California and elsewhere. Six SCEC geologists (Rockwell, Dolan, Dawson, Hartleb, Christofferson, Tucker, and Langridge) combined with geologists from the USGS (Fumal and Stenner), IGPG (Armijo, Meyer, and de Chabalier), and ITU (Barka, Akyuz, Cakir, Altunel, Sunal, Dikbas, Zkaya, and Yerli) to map the surface rupture and determine its slip distribution. SCEC geologists took the lead on the eastern Duzce segment, and joined with the USGS team on the Sakarya (Sapanca to Akyaza) segment. It is along the Sakarya segment that the maximum slip for the earthquake was measured at 5.2 m.
Several papers involving SCEC scientists are being presented at the fall AGU, and a summary paper is being prepared by Barka at ITU. The USGS also released a Circular on the results of the combined work (including seismology and engineering) that includes SCEC scientists as contributors.
The October 16, 1999 Hector Mine Earthquake Close on the heels of the Izmit earthquake, southern California was visited by it own M7.1 earthquake on the Lavic Lake fault on October 16, 1999. SCEC geologists (22 personnel altogether) joined with USGS and CDMG geologists to map and determine slip distribution for this important earthquake. The earthquake occurred on a previously unnamed fault in the Bullion Mountains, and produced about 40 km of surface rupture. Of this, most of the slip and moment release was along 33 km of the Lavic Lake fault, and maximum slip was measured at 5.25 m. Thus, this appears to have been a high stress drop (>100 bars) event. The remaining slip occurred on the Bullion Mountain and several other minor faults, and expressed generally small displacements suggesting that much or most of it was induced slip. In addition to the mapping, we deployed up to eight teams of surveyors to document near and far field displacement to compare to the field measurements of slip. The preliminary results were written up for SRL (in press) and combine geologic, seismologic and geodetic results.
The November 12, 1999 Duzce Earthquake Following close on the heels of Augusts earthquake, the remaining piece of the Duzce fault ruptured between the east end of the August rupture at Eftan Golu eastward for about 35 km to near Bolu with oblique normal slip. About 3 m of right slip was recorded along with substantial dip slip. Six SCEC scientists (Hartleb, Akyuz, Langridge, Dawson, Lindvall, Rockwell) participated in post-earthquake studies; Hartleb and Akyuz focused on mapping slip distribution, whereas Dawson, Langridge, Lindvall and Rockwell surveyed offsets by EDM and searched for viable trench sites). Part of this second effort was supported independent of SCEC.
|
system
Southern California Earthquake Center
@ 