Thomas K. Rockwell

The Surface Rupture and Slip Distribution of the 17 August 1999 İzmit Earthquake (M 7.4), North Anatolian Fault

The 17 August 1999 Izmit earthquake occurred on the northern strand of the North Anatolian fault zone. The earthquake is associated with a 145-km-long surface rupture that extends from southwest of Duzce in the east to west of Hersek delta in the west. Detailed mapping of the surface rupture shows that it consists of five segments separated by releasing step-overs; herein named the Hersek, Karamursel-Golcuk, Izmit-Sapanca Lake, Sapanca-Akyazi, and Karadere segments from west to east, respectively. The Hersek segment, which cuts the tip of a large delta plain in the western end of the rupture zone, has an orientation of N80°. The N70°-80°E-trending Karamursel-Golcuk segment extends along the linear southern coasts of the Izmit Gulf between Karamursel and Golcuk and produced the 470-cm maximum displacement in Golcuk. The northwest-southeast-striking Golcuk normal fault between the Karamursel-Golcuk and Izmit-Sapanca segments has 2.3-m maximum vertical displacement. The maximum dextral offset along the Izmit-Sapanca Lake segment was measured to be about 3.5 m, and its trend varies between N80°E and east-west. The Sapanca-Akyazi segment trends N75°-85°W and expresses a maximum displacement of 5.2 m. The Karadere segment trends N65°E and produced up to 1.5-m maximum displacement. The Karadere and Sapanca-Akyazi segments form fan-shape or splaying ruptures near their eastern ends where the displacement also diminished.

Paleoseismology of the Johnson Valley, Kickapoo, and Homestead Valley Faults: Clustering of Earthquakes in the Eastern California Shear Zone

Paleoseismic data from 11 trenches at seven sites excavated across the southern Johnson Valley, Kickapoo, and Homestead Valley faults that ruptured in the 1992 Landers earthquake, as well as the northern Johnson Valley fault which did not fail in 1992, indicate that the return period for large surface rupturing events in this part of the eastern California shear zone is in the range of 5-15 ka. The inferred slip rates, based on their respective recurrence intervals, are in the range of 0.2-0.6 mm/yr for each of the faults studied. A previous large earthquake ruptured the southern Johnson Valley and Kickapoo faults about 5 ka B.P. The northern Johnson Valley fault also failed at about this time at 5.8 ka B.P. and may have been part of the same rupture. In contrast, the penultimate large earthquake that we identify on the Homestead Valley fault occurred about 15 ka B.P., much earlier than other faults involved in the 1992 rupture. From these observations, combined with paleoseismic work by others after the 1992 earth- quake, it appears that previous events along the southern Johnson Valley and Kick- apoo faults were different than those of 1992 and may have involved other fault segments. It has been over 5 ka since the most recent rupture on the northern Johnson Valley fault. Therefore, it is surprising that it did not fail in the 1992 rupture. From our observations, dextral shear appears to be distributed across the entire eastern California shear zone, with individual faults taking only a small proportion of the overall slip. Release of this regional strain appears to occur in temporal clusters of large (?) earthquakes, with the 1992 event apparently the most recent of a sequence of late Holocene (0-1 ka) earthquakes that have ruptured the nine faults we have trenched in the southwestern Mojave desert. Previous clusters of earthquake activity occurred in the early (8-9 ka) and middle (5-6 ka) Holocene, and possibly the latest Pleistocene (15 ka).

Quaternary soils and dust deposition in southern Nevada and California

Eoliandustconstitutesmuchofthepedogenic material in late Pleistocene and Holocene soils of many arid regions of the world.Comparisonofthecompositionsand influx rates of modern dust with the eolian component of dated soils at 24 sites in southern Nevada and California yields informationon(1)thecompositionandinflux rate of dust in late Pleistocene and Holocene soils, (2) paleoclimate and its effects on the genesis of aridic soils, especially with regard to dustfall events, (3) the timing and relative contribution of dust from playa sources versus alluvial sources, and (4) the effects of accumulation of dust in soil horizons. The<2mmfractionsofAandBhorizons of soils formed on gravelly alluvial-fan deposits in the study area are similar to moderndustingrainsize,contentofCaCO3and salt,majoroxides,andclaymineralogy;thus, they are interpreted to consist largely of dust. The major-oxide compositions of the shallow soil horizons are nearly identical to that of the modern dust, but the compositions of progressively deeper horizons approachthatoftheparentmaterial.Theclay mineralogyofmoderndustatagivensiteis similar to that of the Av horizons of nearby Holocene soils but is commonly different from the mineralogies of deeper soil horizonsandoftheAvhorizonsofnearbyPleistocenesoils.Theseresultsareinterpretedto indicate that dust both accumulates and is transformed in Av horizons with time. Changes in soil-accumulation rates provide insights into the interplay of paleoclimate,dustsupply,andsoil-formingprocesses. Modern dust-deposition rates are more than large enough to account for middle and late Holocene soil-accumulation rates at nearly all sites. However, the early Holocene soil-accumulation rates in areas near late Pleistocene pluvial lakes are much higher than modern rates and clearly indicate a dust-deflation and -deposition event that caused rapid formation offine-grained shallow soil horizons on uppermost Pleistocene and lower Holocene deposits. We interpret late Pleistocene soil-accumulation rates to indicate that dust-deposition rates were low during this period but that increased effective moisture during the late Wisconsinan favored translocation of clay andCaCO3fromnearthesurfacetodeeper inthesoilprofile.Pre‐latePleistocenerates are very low in most areas, mainly due to a pedogenic threshold that was crossed when accumulations of silt, clay, and CaCO3 began to inhibit the downward transport of eolian material, but in part due to erosion.

Prospects for Larger or More Frequent Earthquakes in the Los Angeles Metropolitan Region

Far too few moderate earthquakes have occurred within the Los Angeles, California, metropolitan region during the 200-year-long historic period to account for observed strain accumulation, indicating that the historic era represents either a lull between clusters of moderate earthquakes or part of a centuries-long interseismic period between much larger (moment magnitude, Mw, 7.2 to 7.6) events. Geologic slip rates and relations between moment magnitude, average coseismic slip, and rupture area show that either of these hypotheses is possible, but that the latter is the more plausible of the two. The average time between Mw 7.2 to 7.6 earthquakes from a combination of six fault systems within the metropolitan area was estimated to be about 140 years.

Community Fault Model (CFM) for Southern California

We present a new three-dimensional model of the major fault systems in southern California. The model describes the San Andreas fault and associated strike- slip fault systems in the eastern California shear zone and Peninsular Ranges, as well as active blind-thrust and reverse faults in the Los Angeles basin and Transverse Ranges. The model consists of triangulated surface representations (t-surfs) of more than 140 active faults that are defined based on surfaces traces, seismicity, seismic reflection profiles, wells, and geologic cross sections and models. The majority of earthquakes, and more than 95% of the regional seismic moment release, occur along faults represented in the model. This suggests that the model describes a comprehen- sive set of major earthquake sources in the region. The model serves the Southern California Earthquake Center (SCEC) as a unified resource for physics-based fault systems modeling, strong ground-motion prediction, and probabilistic seismic hazards assessment.

Chronology and rates of faulting of Ventura River terraces, California

New evidence concerning the chronology of four late Pleistocene terraces of the Ventura River near Oak View, California, that are vertically offset and tilted by reverse and reverse flexural-slip faults provides a means of estimating rates of fault movement and downcutting by the Ventura River. Radiocarbon ages of charcoal contained within terrace deposits date two of the terraces. Correlation of soils that have developed on terrace deposits and extrapolation of the rate of vertical displacement of the Arroyo Parida-Santa Ana fault (not a flexural-slip fault) date the others. Resulting age estimates for the four main late Pleistocene terraces are Qt5b = 30,000 Qt6a = 38,000, Qt6b = 54,000, and Qt6c = 92,000 yr B.P. Vertical-slip rates on flexural-slip faults range from <0.3 to 1.1 mm/yr and apparently are related to the rate, form, and mechanics of folding of the north limb of the Ayers Creek–Canada Larga syncline. Average rates of downcutting of the Ventura River for several intervals during the late Pleistocene, estimated from the chronology and relative elevation of river terraces north of the Arroyo Panda–Santa Ana fault upstream from the zone of faulting, vary from ∼0.5 to 1.3 mm/yr. The range in rates probably reflects variations in local uplift as well as adjustments to changing eustatic sea level, climatic conditions, and/or regional deformation of the western Transverse Ranges. The average rate of downcutting of the Ventura River north of the zone of flexural-slip faulting is ∼0.8 mm/yr, compared with 1.2 to 2.2 mm/yr in the deformation zone immediately to the south. This apparently indicates that during tectonic deformation there is an approximate balance between the rate of uplift due to faulting and folding and the rate of downcutting by the fluvial system.

Lateral offsets on surveyed cultural features resulting from the 1999 İzmit and Düzce earthquakes, Turkey

Surveys of offset linear cultural features that cross the surface ruptures of the 17 August and 12 November 1999 earthquakes on the North Anatolian fault in Turkey yield slip values as large as or larger than those recorded by near-field measurements in the same areas immediately after the earthquake. Using long, linear alignments of trees, fence lines, walls, and canals, we demonstrate as much as a 2-m increase in observed slip over the initial field measurements. On an average, we observed about 15% of the total lateral slip as off-fault deformation, with values ranging from 0% to 40% of the total slip. Part of this lateral deformation is accom- modated by bending or drag in a zone as much as 30 m in width, although usually the zone varies between 5 and 20 m in width. This supports the idea that substantial nonbrittle, off-fault deformation is associated with ruptures in areas of alluvial fill. Alternatively, there may have been substantial afterslip, although none has been recognized on postearthquake repaired structures. Our observations suggest that post- earthquake measurements of fault slip, using tape measures on offset geomorphic features, may underestimate the actual amount of slip in that event, where the non- linearity of the offset feature does not allow recognition of the warping. Surveys of multiple tree lines within groves of poplar trees, planted in straight lines across the fault prior to the earthquake, show surprisingly large lateral varia- tions. In one grove, slip increases by nearly 1.8 m, or 35% of the maximum measured value, over a lateral distance of nearly 100 m. This and other observations along the 1999 ruptures suggest that the lateral variability of slip observed from displaced geomorphic features in many earthquakes of the past may represent a combination of (1) actual differences in slip at the surface and (2) the difficulty in recognizing distributed nonbrittle deformation.

Crusader castle torn apart by earthquake at dawn, 20 May 1202

The Crusader castle of Vadum Jacob, an outpost overlooking the Jordan River, was deformed during a destructive earthquake triggered by motion along the Dead Sea Transform. The M >7 earthquake occurred at dawn, 20 May 1202, and offset the castle walls by 1.6 m. This exceptional precision in dating and estimating displacement was achieved by combining accounts from primary historical sources, by excavating the Dead Sea Transform where it bisects the castle, and by dating faulted archaeological strata. The earthquakes of October 1759 and/or January 1837 may account for the remaining 0.5 m out of a total 2.1 m of offset. Our study exploits the potential embodied in interdisciplinary historical-archaeological-geological research and illustrates how detailed histories of seismogenic faults can be reconstructed.

Dynamic pedogenesis: New views on some key soil concepts, and a model for interpreting quaternary soils

Abstract Inasmuch as soils are open systems and rarely, if ever, achieve equilibrium with their environments, it is philosophically sound to view all soils as expressing varying levels of polygenesis as that term has been redefined. Soil genesis and resultant morphology may then be viewed in a comprehensive framework of soil evolution that consists of two linked pathways, one developmental and the other regressive, that reflect interactions of both exogenous and endogenous vectors (vectors are factors, processes, and conditions of pedogenesis). Following this philosophy, a model of pedogenesis is framed in an evolutional paradigm that emphasizes the integrated effects of dynamic and passive pedogenic vectors in directing pathways and in controlling rates of soil genesis through time. The dynamic vectors include energy and mass fluxes, frequencies of soil wetting and drying, water table dynamics, organisms, feedback processes, and pedoturbation. The passive vectors include parent materials, soil chemical environment, stability of geomorphic surfaces, and various evolved soil properties and conditions (accessions). Both sets of vectors vary spatially, and the dynamic vectors, more so than passive vectors, fluctuate through time. The model is expressed as S=f(D,P, dD dt , dD dt ) where S is the state of the soil or degree of profile evolution, D is the set of dynamic vectors, P is the set of passive vectors, and dD dt and dP dt denote change through time t. The model helps explain the apparent minimal development and regressed character of some old soils and the rapid and strong development of some young ones.

Quaternary rate of folding of the Ventura Avenue anticline, western Transverse Ranges, southern California

Upper Quaternary terraces of the Ventura River, California, are uplifted, tilted, and folded over the Ventura Avenue anticline. Rates of uplift and tilting have decreased since inception of the structure over the past 200 ka. Assuming that the chronology, based on amino-acid racemization, 14C dates, and soils correlation, is approximately correct, then the minimum possible average rate of uplift in the axial region of the fold has decreased from ∼14 mm/yr to 2 mm/yr during the past 200 ka. Interval rates of uplift for the periods 200 ka to 80 or 105 ka, 80 or 105 ka to 30 ka, and 30 ka to present are, respectively, about 20 mm/yr, 9 mm/yr, and 5 mm/yr. The rate of tilting shows a similar trend, decreasing from ∼5.8 urad/yr, 2.5 urad/yr, and 1.2 urad/yr for the same time intervals, respectively. Based on the mechanics of flexural slip folds in stratified sedimentary rocks, these data suggest that the rootless Ventura Avenue anticline is a fold that has been shortening at a relatively constant rate of about 9 mm/yr since its inception.