Kevin P. Furlong

Quaternary uplift astride the aseismic Cocos Ridge, Pacific coast, Costa Rica.

The Pacific coast of Costa Rica lies within the Central American forearc and magmatic-arc region that was created by northeastward subduction of the Cocos plate beneath the Caribbean plate at the Middle America Trench. From the Peninsula de Nicoya south-eastward toward the Peninsula de Osa and the Peninsula de Burica on the Panamanian border, the Middle America Trench loses its physiographic expression where it intersects the aseismic Cocos Ridge. Interaction between subduction of the buoyant, aseismic Cocos Ridge and the overriding Caribbean plate is invoked to explain the variation in rates of vertical crustal uplift along a coastal transect from Nicoya to Burica. The Pliocene and Pleistocene stratigraphic record and Holocene marine terraces and beach ridge complexes indicate that maximum rates of crustal uplift have occurred on the Peninsula de Osa, immediately landward of the aseismic Cocos Ridge. Crustal uplift rates decrease northwest toward the Peninsula de Nicoya, and to a lesser extent southwest toward the Peninsula de Burica. The late Quaternary stratigraphy on the Peninsula de Osa is subdivided into two major chronostratigraphic sequences from groupings of radiocarbon dates. Crustal uplift rates calculated from these sequences systematically decrease from 6.5 to 2.1 m/ka north-east across the peninsula. Deformation of the peninsula is modeled as uplifted and down-to-the-northeast-tilted fault blocks with an angular rotation rate of 0.03° to 0.06° per thousand years. Although less well constrained, crustal uplift rates on the Peninsula de Nicoya, 200 km to the northwest of the Peninsula de Osa, vary from <1 m/ka for Pliocene and Pleistocene sediments to 2.5 m/ka for Holocene marine terraces. In the Quepos region, 100 km to the northwest of the Peninsula de Osa, calculated uplift rates derived from incision of late Quaternary fluvial terraces range from 0.5 to 3.0 m/ka. On the Peninsula de Burica, only 60 km to the southwest of the Peninsula de Osa, calculated uplift rates range from 4.7 m/ka for a late Holocene marine terrace to 1.2 m/ka for post-late Pliocene deep-sea sediments. The variations in calculated uplift rates on the Peninsula de Osa constrain a dynamic model for subduction of the Cocos Ridge and the resulting uplift of the overriding Caribbean plate. Deflection of the Caribbean plate is modeled using various effective elastic thicknesses as the response of an elastic plate to the buoyant force of the subducted Cocos Ridge. Because the shape of the subducted end of the Cocos Ridge is unknown, two scenarios are evaluated: (1) a radially symmetric ridge with a slope similar to the slope of the flanks of the ridge and (2) a ridge where the subducted end was truncated by the Panama fracture zone. The best-fit model utilizes a truncated ridge that has been subducted during the past 0.5 m.y. ∼50 km beneath the overriding Caribbean plate, which has an effective elastic thickness of 5 km. The model predicts that the highest uplift rate should be ∼3.7 m/ka and occur on the southwest coast of the Peninsula de Osa. The rate of uplift slows considerably to the northeast and indicates that the Peninsula de Osa is tilting to the northeast, which agrees with observations in that region. The predicted uplift rate attributed to aseismic ridge subduction also decreases along the coast both north and south of the Peninsula de Osa, resulting in little uplift that can be attributed to Cocos Ridge subduction in the northwestern portions of the Peninsula de Nicoya.

Codependent histories of the San Andreas and San Jacinto fault zones from inversion of fault displacement rates

Department of Geosciences, Pennsylvania State University, 542 Deike Building, University Park,Pennsylvania 16082, USAABSTRACTThe displacement histories of the San Jacinto and southernmost San Andreas fault zonesare constrained by offset data with ages in the range of 5 Ma to 5 ka. Apparent discrep-ancies between long- and short-term average displacement rates can be reconciled with atime-variable rate model. In this model, the displacement rate on the San Andreas decel-erated from ;35 mm/yr at 1.5 Ma to as low as 9 6 4 mm/yr by 90 ka. Over this sametime period, the rate on the San Jacinto fault zone accelerated from an initial value ofzero to a rate of 26 6 4 mm/yr. The data also imply that the rate of the San Andreasfault accelerated since ca. 90 ka, from ;9 mm/yr to the modern rate of 27 6 4 mm/yr,whereas the San Jacinto decelerated from 26 6 4 mm/yr to the modern rate of 8 6 4mm/yr. The time scale of these changes is significantly longer than the earthquake cycle,but shorter than time scales characteristic of lithospheric-scale dynamics. The emergenceof the San Jacinto fault zone ca. 1.5 Ma coincided with the development of a majorrestraining bend in the San Andreas fault zone, suggesting that the formation of newsubparallel faults could be driven by conditions that inhibit displacement on preexistingfaults.Keywords: fault displacement rates, lithospheric rheology, continental dynamics, crustaldeformation.INTRODUCTIONContinental plate boundary zones are oftencharacterized by systems of faults spanningbroadly deforming regions (Fig. 1). Determin-ing how these faults accommodate relativeplate motions and the nature of their spatialand temporal interactions remains an elusivegoal. In particular, whether fault-displacementrates are constant or time variable—and if so,on what time scales—is difficult to ascertainbecause most geologically determined dis-placement rates represent averages since theage of an offset geologic or geomorphic fea-ture. In the case of time-varying rates, this av-eraging effect can lead to apparently inconsis-tent estimates depending on the ages of theoffset markers.The partitioning of Pacific–North Ameri-ca relative plate motion among the manynorthwest-oriented dextral strike-slip faults insouthern California has been the subject of in-tensive research for decades (Table 1). In thelatitude band 328Nto348N, ;70% of the totalrelative motion between these plates is accom-modated by the San Jacinto and southernmostSan Andreas faults (Fig. 1). Together theytransfer ;35 mm/yr of motion from the Im-perial to the Mojave segment of the San An-dreas fault (north of the San Jacinto–San An-dreas intersection). Previous studies havetended to concentrate either on displacementrates of individual faults or on the steady-statepartitioning of deformation between faults.However, these data can also be used to assesswhether the rate of deformation is conservedbetween these two fault zones over time, as iswidely accepted.Here we apply a technique that allows us touse time-averaged estimates of displacementrate to infer the time history of instantaneousdisplacement rates on the San Andreas andSan Jacinto fault zones. This history allows usto investigate interactions between these twofault zones and, in particular, to assess wheth-er they act in a coupled manner. As we show,interactions between these faults could haveimportant implications for geodynamics, bear-ing on issues such as the relative strengths ofupper mantle and lower- and upper-crustal lay-ers, and the degree to which the deformationfield within the upper crust is coupled to thatwithin the upper mantle. Investigations suchas ours, which use data representing signifi-cantly different periods of time, may thus pro-vide a new approach to investigating somelong-standing problems in geodynamics.DATATable 1 summarizes the geologic, geomor-phic, and geodetic observations constrainingaverage displacement rates on these two faultsover a diverse range of time scales. Each ratelisted in Table 1 represents an average over afinite interval of time. The duration of eachaveraging interval is determined by the age ofa measured offset marker, such as a distinctiverock unit, a geomorphic feature, or a geodeticmonument. With the exception of the geodeticinferences, the averaging intervals are quitelong relative to the earthquake cycle, which ison the order of hundreds of years for the SanAndreas and San Jacinto fault zones. Becausethese data record averages over long periodsof time, they should be insensitive to temporalirregularities in earthquake recurrence.Age-dependent variation in the rate data isappreciable for both faults (Table 1). For theSan Andreas, rate estimates range from ashigh as 30 mm/yr to as low as 14 mm/yr.Rates for the San Jacinto fault zone also varyby nearly a factor of two; long-term averagesare significantly larger than estimates repre-senting shorter, more recent, intervals. In or-der to compare the different observations andto constrain the history of fault displacements,we need to consider the uncertainty in eachdatum. In the geologic literature, error bars donot typically represent statistical measures ofuncertainty (e.g., standard deviations of a nor-mal distribution), but rather they usually rep-resent upper and lower bounds on possibledisplacement rates, based on complex age andoffset relationships among various markersacross the faults. Error estimates often resultin asymmetric plus-or-minus ranges. Our anal-yses of these data are based on more tradi-tional statistical measures of uncertainty. Wehave thus had to make some assumptionsabout how best to represent the reported errorbounds in terms of Gaussian error distribu-tions. Table 1 lists the standard deviations thatwe chose to best represent the allowable slipranges derived from geologic observations. Inall cases we conservatively chose error distri-butions with 95% confidence regions coincid-ing roughly with the absolute bounds reportedin the literature.ANALYSIS AND RESULTSTo explore the history of displacement rateson the San Andreas and San Jacinto faultzones implied by the averages of Table 1, weemployed the theory of smoothing splines(e.g., Matthews and Segall, 1993). The basicgoal is to use the rate averages, which repre-

The 2006-2007 Kuril Islands great earthquake sequence

The southwestern half of a ∼500 km long seismic gap in the central Kuril Island arc subduction zone experienced two great earthquakes with extensive preshock and aftershock sequences in late 2006 to early 2007. The nature of seismic coupling in the gap had been uncertain due to the limited historical record of prior large events and the presence of distinctive upper plate, trench and outer rise structures relative to adjacent regions along the arc that have experienced repeated great interplate earthquakes in the last few centuries. The intraplate region seaward of the seismic gap had several shallow compressional events during the preceding decades (notably an M_S 7.2 event on 16 March 1963), leading to speculation that the interplate fault was seismically coupled. This issue was partly resolved by failure of the shallow portion of the interplate megathrust in an M_W = 8.3 thrust event on 15 November 2006. This event ruptured ∼250 km along the seismic gap, just northeast of the great 1963 Kuril Island (M_w = 8.5) earthquake rupture zone. Within minutes of the thrust event, intense earthquake activity commenced beneath the outer wall of the trench seaward of the interplate rupture, with the larger events having normal-faulting mechanisms. An unusual double band of interplate and intraplate aftershocks developed. On 13 January 2007, an M_W = 8.1 extensional earthquake ruptured within the Pacific plate beneath the seaward edge of the Kuril trench. This event is the third largest normal-faulting earthquake seaward of a subduction zone on record, and its rupture zone extended to at least 33 km depth and paralleled most of the length of the 2006 rupture. The 13 January 2007 event produced stronger shaking in Japan than the larger thrust event, as a consequence of higher short-period energy radiation from the source. The great event aftershock sequences were dominated by the expected faulting geometries; thrust faulting for the 2006 rupture zone, and normal faulting for the 2007 rupture zone. A large intraplate compressional event occurred on 15 January 2009 (M_w = 7.4) near 45 km depth, below the rupture zone of the 2007 event and in the vicinity of the 16 March 1963 compressional event. The fault geometry, rupture process and slip distributions of the two great events are estimated using very broadband teleseismic body and surface wave observations. The occurrence of the thrust event in the shallowest portion of the interplate fault in a region with a paucity of large thrust events at greater depths suggests that the event removed most of the slip deficit on this portion of the interplate fault. This great earthquake doublet demonstrates the heightened seismic hazard posed by induced intraplate faulting following large interplate thrust events. Future seismic failure of the remainder of the seismic gap appears viable, with the northeastern region that has also experienced compressional activity seaward of the megathrust warranting particular attention.

Evolution and thickness of the lithosphere beneath coastal California

Many of the tectonic features of the California Coast Ranges are directly related to the northward migration of the Mendocino triple junction along the western edge of North America during the past 20 to 30 m.y. At the triple junction, the western edge of the North American plate slides off the (relative) northward-moving and subducting Gorda plate, leaving the thin western edge in direct contact with asthenosphere upwelling to fill the space vacated by the underlying Gorda plate. Two-dimensional, time-dependent thermal modeling of this process, constrained by teleseismic delay studies, is used to construct a map of lithospheric thickness of coastal California. Among the implications of this map are that (1) the high heat flow in the Coast Ranges can be almost entirely accounted for by the asthenospheric upwelling associated with the migrating Mendocino triple junction; (2) the general elevation and the late Cenozoic volcanism in the California Coast Ranges are responses to a zone of unusually thin (20 to 45 km) lithosphere that extends southward behind the Mendocino triple junction; and (3) the course of some segments of the San Andreas fault (in both central and southern California) appear to deviate from a deeper, more fundamental, transform boundary separating the Pacific and North American plates.

Lithospheric behavior with triple junction migration: an example based on the Mendocino triple junction

Abstract The northward migration of the Mendocino triple junction along the western margin of the North American plate produces a region of thin lithosphere in association with the growth of the San Andreas transform. Numerical modeling of the resultant thermal regime in conjunction with modeling of the mechanical response of the region indicate the role of lithospheric thinning in the tectonism along the western margin of North America. The geometry of the region of thin lithosphere is inherited from the previous subduction regime with resulting implications for the mechanical response. The elevation of the California Coast Ranges is maintained primarily by flexure from buoyant loading from below. Additionally, sporadic volcanism associated with the passage of the triple junction is a likely consequence of super solidus mantle emplaced at shallow depths after the removal of the subducting slab.

Seismic Evidence for a Lower-Crustal Detachment Beneath San Francisco Bay, California

Results from the San Francisco Bay area seismic imaging experiment (BASIX) reveal the presence of a prominent lower crustal reflector at a depth of ∼15 kilometers beneath San Francisco and San Pablo bays. Velocity analyses indicate that this reflector marks the base of Franciscan assemblage rocks and the top of a mafic lower crust. Because this compositional contrast would imply a strong rheological contrast, this interface may correspond to a lower crustal detachment surface. If so, it may represent a subhorizontal segment of the North America and Pacific plate boundary proposed by earlier thermo-mechanical and geological models.

Ephemeral crustal thickening at a triple junction: The Mendocino crustal conveyor

As the North American crust interacts with the migrating Mendocino triple junction, the crust is first significantly thickened and then equivalently thinned over a distance of a few hundred kilometers (within a time frame of 5 m.y. or less). This process of ephemeral crustal thickening is proposed to result from viscous coupling between the northward-migrating Gorda slab and the base of North America south of the triple junction. A time-dependent, thermal-mechanical finite-element model is developed to test this hypothesis of plate-boundary tectonics. Results of the numerical simulations show patterns of crustal deformation consistent with the mapped sequence of folding and faulting in the area, the observed crustal structure and triple junction regional seismicity, and localized regions of crustal extension coincident with the position of a hypothesized lower-crustal melt zone.

Ephemeral plate tectonics at the Queen Charlotte triple junction

Three plate boundaries, the Queen Charlotte transform, the Cascadia subduction zone, and the Juan de Fuca Ridge, meet in a complex triple junction offshore Vancouver Island. Some interpretations of the plate tectonics of the region have included the Explorer ridge, the Dellwood Knolls, and the Tuzo Wilson volcanic field as part of the spreading system, requiring up to three triple junctions in the region. New SeaBeam bathymetry interpreted with existing regional data sets indicate that the Dellwood Knolls and Tuzo Wilson volcanic field are not independent plate boundaries, but the result of leaky transform tectonics between two overlapping transform faults. Seismicity shows that the Explorer plate is being deformed by Pacific–North American relative motion as a new transform plate boundary forms and cuts off the Explorer ridge. The system is evolving to a single triple junction at the northern terminus of the Juan de Fuca Ridge where it meets the Nootka deformation zone. Thus the Explorer microplate, which was spawned ≈5 Ma, is an ephemeral adjustment to mechanical difficulties at the triple junction. This new model implies that the Explorer subduction zone is no longer active. Ephemeral oceanic microplates have also existed at Pacific–North American triple junctions off southern California and Baja California.

Cenozoic volcanism in the California Coast Ranges: Numerical solutions

We present the results of numerical simulations of Cenozoic volcanism in the Coast Ranges of California, associated with the northward migration of the Mendocino Triple Junction (MTJ). Three aspects of the thermal evolution and magmatism in the wake of the MTJ are investigated: (1) mantle flow and thermal perturbation in the “slabless window”; (2) pressure-release partial melting in the upwelling asthenosphere; and (3) thermally induced crustal anatexis. Numerical results show that upwelling of asthenosphere in the “slabless window” causes a drastic change in the thermal structure of the lithosphere and results in significant pressure-release partial melting in the upwelling asthenosphere. Up to 4–5 km of basaltic magma may be produced. Most of this mafic magma is likely stored near the base of the crust. Thermally induced crustal anatexis mainly occurs in the deep crust (below 20 km), where a high degree (>50%) of partial melting may be predicted. Most of the crustal melt is probably generated within 1 m.y. after the passage of the MTJ. This work also suggests that the spatial and temporal distribution of Cenozoic volcanism in the California Coastal Ranges is directly related to the variation of the relative velocity between the North America and Pacific plates in the last 30 m.y. The voluminous volcanic rocks in the northern California Coast Ranges correlate to the relatively high velocity between these two plates in the last 10 m.y. The reduced volcanism in the central Coast Ranges may be related to the slow migration of the MTJ between 10 and 20 Ma.

Three‐dimensional seismic velocity structure of the San Francisco Bay area

Seismic travel times from the northern California earthquake catalogue and from the 1991 Bay Area Seismic Imaging Experiment (BASIX) refraction survey were used to obtain a three-dimensional model of the seismic velocity structure of the San Francisco Bay area. Nonlinear tomography was used to simultaneously invert for both velocity and hypocenters. The new hypocenter inversion algorithm uses finite difference travel times and is an extension of an existing velocity tomography algorithm. Numerous inversions were performed with different parameters to test the reliability of the resulting velocity model. Most hypocenters were relocated 12 km under the Sacramento River Delta, 6 km beneath Livermore Valley, 5 km beneath the Santa Clara Valley, and 4 km beneath eastern San Pablo Bay. The Great Valley Sequence east of San Francisco Bay is 4–6 km thick. A relatively high velocity body exists in the upper 10 km beneath the Sonoma volcanic field, but no evidence for a large intrusion or magma chamber exists in the crust under The Geysers or the Clear Lake volcanic center. Lateral velocity contrasts indicate that the major strike-slip faults extend sub vertically beneath their surface locations through most of the crust. Strong lateral velocity contrasts of 0.3–0.6 km/s are observed across the San Andreas Fault in the middle crust and across the Hayward, Rogers Creek, Calaveras, and Greenville Faults at shallow depth. Weaker velocity contrasts (0.1–0.3 km/s) exist across the San Andreas, Hayward, and Rogers Creek Faults at all other depths. Low spatial resolution evidence in the lower crust suggests that the top of high-velocity mafic rocks gets deeper from west to east and may be offset under the major faults. The data suggest that the major strike-slip faults extend sub vertically through the middle and perhaps the lower crust and juxtapose differing lithology due to accumulated strike-slip motion. The extent and physical properties of the major geologic units as constrained by the model should be used to improve studies of seismicity, strong ground motion, and regional stress.