American Plate Boundary Zone Research Articles

Investigating the Effect of Mantle Flow and Viscosity Structure on Surface Velocities in Alaska Using 3‐D Geodynamic Models

AbstractWe utilize 3‐D finite element geodynamic models, incorporating long‐term kinematic estimates of upper plate motion, to better understand the roles that viscosity structure and mantle tractions play in generating plate motions and continental interior deformation in Alaska. Surface deformation in the Pacific‐North American plate boundary zone in Alaska and northwest Canada is strongly influenced by the complex interactions between flat‐slab subduction, gravitational collapse and mantle tractions. Predictions of long‐term tectonic block motion derived from recent Global Positioning System datasets (GPS) reveal surface motions atypical of other continental convergent plate boundary zones. Specifically, in northern and northwestern Alaska, southeastward motion is observed directed back toward the plate boundary. Geodynamic models that incorporate a southeastward directed long‐wavelength mantle traction of ∼2.5–3.8 MPa best replicate surface velocities in Alaska. These mantle tractions, in conjunction with the collision of the Yakutat microplate, appear to drive the uplift and deformation in the Mackenzie Mountains. Furthermore, the extent of the northward motion in southern and central Alaska is controlled by the location of the leading edge of the Yakutat flat slab.

Morphotectonic study of the Greater Antilles

The first morphotectonic model of the Greater Antilles is presented. The model is adjusted to the current dynamics between the Caribbean and North American plates. It is mainly elaborated by Rantsman’s methodology. We determined 2 megablocks, 7 macroblocks, 42 mesoblocks, 653 microblocks and 1264 nanoblocks. They constitute a set of active blocks under rotation, uplifting and tilting movements. A total of 11 active knots of faults and 8 cells are the main articulation areas. The largest seismogenetic structures in the Northern Caribbean are an array of the active fault segments. The majority of them are in the Caribbean-North American Plate Boundary Zone, the Hispaniola has the most complex neotectonic structure–associated with the central axis of the morphotectonic deformations in the region.

Tectonic inversion in the Caribbean-South American plate boundary: GPS geodesy, seismology, and tectonics of theMw6.7 22 April 1997 Tobago earthquake

On 22 April 1997 the largest earthquake recorded in the Trinidad-Tobago segment of the Caribbean-South American plate boundary zone (Mw 6.7) ruptured a shallow (~9 km), ENE striking (~250° azimuth), shallowly dipping (~28°) dextral-normal fault ~10 km south of Tobago. In this study, we describe this earthquake and related foreshock and aftershock seismicity, derive coseismic offsets using GPS data, and model the fault plane and magnitude of slip for this earthquake. Coseismic slip estimated at our episodic GPS sites indicates movement of Tobago 135 ± 6 to 68 ± 6 mm NNE and subsidence of 7 ± 9 to 0 mm. This earthquake was anomalous and is of interest because (1) its large component of normal slip and ENE strike are unexpected given the active E-W dextral shearing across the Caribbean-South American plate boundary zone, (2) it ruptured a normal fault plane with a low (~28°) dip angle, and (3) it reactivated and inverted the preexisting Tobago terrrane-South America ocean-continent (thrust) boundary that formed during early Tertiary oblique plate convergence.

Structural Geometry of a Sector of the Colorado River Delta, Baja California, Mexico, Based on Seismic Reflections

A structural study in the SW section of the Colorado River delta using seismic reflection data is presented. The study area is located along the Cerro Prieto transform fault, which extends from the northern Gulf of California through the Mexicali Valley and is an active fault within the Pacific-North American plate boundary zone. The research was supported by a database of five seismic profiles with a total length of 215 km, collected in the early 80’s by Petroleos Mexicanos. The results show a high density of faults, most of which are buried by sediments. Within the Cerro Prieto fault zone, several faults were identified, such as: Palmas, Mesa, and Pangas Viejas, until now unknown. In addition, even though the Indiviso fault was investigated and superficially identify prior to this work, herein mapped at depth. West of the Cerro Prieto fault zone lies the Las Tinajas basin, bound by the Dunas and Saldana faults and by the Montague basin to the southeast. The deformation zone along the plate boundary is 18-km-wide, stretching from the Cerro Prieto fault in the east to the Pangas Viejas fault in the west. The orientations of the faults are NW–SE, and if projected from the southern side of the Sierra Cucapah southward, the faults tend to join the Cerro Prieto fault. In the Las Tinajas basin, the acoustic basement is deeper than 5,000 m. Some of the largest vertical displacements generated by the 2010 7.2-Mw El Mayor-Cucapah earthquake occurred southeast of the epicenter and coincided with the location of the Pangas Viejas Fault, which is buried by sediments. Before this event, seismic activity was very low, and no structures were known in the area. In this paper, we demonstrate that there are at least seven major faults that may now pose a high seismic hazard.

Receiver function imaging in strongly laterally heterogeneous crust: Synthetic modeling of BOLIVAR data

We resolve a large (∼20 km) discrepancy in Moho depth determined from PdS receiver functions (RFs) and from active source seismic profiling in the complex Caribbean-South American plate boundary zone in eastern Venezuela. As part of the BOLIVAR experiment 20 broadband stations were deployed along an active source profile to record teleseisms. Using the extremely heterogeneous crustal model obtained from active source data, we generated 2D finite-difference elastic wave synthetics and from them calculated receiver functions and CCP stacks. We compare the observations with synthetic sections that have been spatially sampled at 0.25 km to 40 km. The densely sampled synthetics show that several events in the field data that were originally interpreted as the Moho are multiple reflections within sedimentary basins. Where the Moho has the steepest dip under the plate boundary the CCP stacks fail to image the Moho well, regardless of the density of spatial sampling. A suitable spatial sampling criterion for clearly imaging the lower crust and Moho is to overlap Fresnel zones by 50% at Moho depth, which for the 1 Hz receiver functions examined here, requires an instrument spacing of 15–20 km, with the actual field data density ranging from 20 km to 100 km.

Tectonics, basin subsidence mechanisms, and paleogeography of the Caribbean-South American plate boundary zone

Using a mega-regional dataset that includes over 20,000 km of on- and offshore 2D seismic lines and 12 wells, we illustrate three different stages of fault formation and basin evolution in the Caribbean arc-South American continent collisional zone. Transpressional deformation associated with oblique collision of the Caribbean arc migrates diachronously over a distance of ∼1500 km from western Venezuela in Paleogene time (∼57 Ma) to a zone of active deformation in the eastern offshore Trinidad area. Each diachronous stage of pre-, syn-, and post-collisional basin formation is accompanied by distinct patterns of fault families. We use subsidence histories from wells to link patterns of long-term basinal subsidence to periods of activity of the fault families. Stage one of arc-continent collision Initial collision is characterized by overthrusting of the south- and southeastward-facing Caribbean arc and forearc terranes onto the northward-subducting Mesozoic passive margin of northern South America. Northward flexure of the South American craton produces a foreland basin between the thrust front and the downward-flexed continental crust that is initially filled by clastic sediments shed both from the colliding arc and cratonic areas to the south. As the collision extends eastward towards Trinidad, this same process continues with progressively younger foreland basins formed to the east. On the overthrusting Caribbean arc and forearc terranes, north-south rifting adjacent to the collision zone initiates and is controlled by forward momentum of southward-thrusting arc terranes combined with slab pull of the underlying and subducting, north-dipping South American slab. Uplift of fold-thrust belts arc-continent suture induces rerouting of large continental drainages parallel to the collisional zone and to the axis of the foreland basins. Stage two This late stage of arc-continent collision is characterized by termination of deformation in one segment of the fold-thrust belt as convergent deformation shifts eastward. Rebound of the collisional belt is produced as the north-dipping subducted oceanic crust breaks off from the passive margin, inducing inversion of preexisting normal faults as arc-continent convergence reaches a maximum. Strain partitioning also begins to play an important role as oblique convergence continues, accommodating deformation by the formation of parallel, strike-slip fault zones and backthrusting (southward subduction of the Caribbean plate beneath the South Caribbean deformed belt). As subsidence slows in the foreland basins, sedimentation transitions from a marine underfilled basin to an overfilled continental basin. Offshore, sedimentation is mostly marine, sourced by the collided Caribbean terranes, localized islands and carbonate deposition. Stage three This final stage of arc-continent collision is characterized by: 1) complete slab breakoff of the northward-dipping South American slab; 2) east-west extension of the Caribbean arc as it elongates parallel to its strike forming oblique normal faults that produce deep rift and half-grabens; 3) continued strain partitioning (strike-slip faulting and folding). The subsidence pattern in the Caribbean basins is more complex than interpreted before, showing a succession of extensional and inversion events. The three tectonic stages closely control the structural styles and traps, source rock distribution, and stratigraphic traps for the abundant hydrocarbon resources of the on- and offshore areas of Venezuela and Trinidad.

The Bayamo Earthquake (Cuba) of the 18 October 1551

Using contemporary and original documents from the Archivo General de Indias it has been possible to complete the data for the 18 October 1551 earthquake in Cuba. The seism took place at midday, approximately. It had foreshocks and aftershocks. In Bayamo, 7 inhabitants were injured, and the town was severely affected. Maximum seismic intensity was IX degrees on the MSK scale, and the area of perceptibility is estimated at 40,000 km2. Liquefaction processes and soil type in Bayamo contributed to the damage. This locality is in the Eastern region of the island, and continues to suffer the most and the strongest seismic events. The epicenter was in the southern marine area of the western segment of Oriente trough (19.6 N 77.8 W, h = 15 km, Ms = 6.6), where there is a crossing of faults, and neotectonics and focal mechanisms are affected by transtension, although the Bartlett-Cayman region’s tendency to left-lateral strike-slip movement is maintained, in the Caribbean and North American plate boundary zone.

Observations and interpretation of fundamental mode Rayleigh wavefields recorded by the Transportable Array (USArray)

Broadband recordings of the dense Transportable Array (TA) in the western United States provide unparalleled detailed images of long‐period seismic surface wavefields. With 400 stations spanning most of the western United States, wavefronts of fundamental mode Rayleigh waves may be visualized coherently across the array at periods ≳40 s. In order to constrain the Rayleigh wave phase velocity structure in the western United States, I assemble a data set of vertical component seismograms from 53 teleseismic events recorded by the TA from April 2006 to October 2007. Complex amplitude spectra from these recordings at periods 27–100 s are interpreted using the multiplane wave tomographic method of Friederich and Wielandt (1995) and Pollitz (1999). This analysis yields detailed surface wave phase velocity and three‐dimensional shear wave velocity patterns across the North American plate boundary zone, elucidating the active processes in the highly heterogeneous western U.S. upper mantle.

Morphotectonic study of Hispaniola

Geomorphological analysis; aerial photographs; and geomorphologic, geological, geophysical, topographical, and field studies show that the morphology of Hispaniola can be linked to lateral variations in the geometry and tectonism of the Caribbean-North American Plate Boundary Zone. Three main categories of the relief were established: territorial units (1 megablock, 2 macroblocks, 14 mesoblocks, 209 blocks, 401 microblocks, and 527 nanoblocks), morphostructural alignments (4 first-, 1 second-, 12 third-, and 30 fourth-rank), and 16 knots between morphostructural alignments (second-to fourth-rank). The main seismic activity is concentrated on the first-and second-rank lineaments, and some important epicenters are located in the vicinity of the lineament intersections. The origin of the earthquakes in the vicinity of such knots can be explained by the forcing/pushing of macroblocks northeastward. The existence of earthquakes along the main lineaments may be explained by tension or compression in a restraining bend zone. From the current study, it appears that earthquake occurrence in Hispaniola is related with the stress concentrations in the vicinity of morphotectonic zones. A seismotectonic interpretation of Hispaniola is shown where three zones exist, each of them with a different active level and dimensions.

Chronology of Cenozoic tectonic events in western Venezuela and the Leeward Antilles based on integration of offshore seismic reflection data and on-land geology

Newly acquired, deep-penetration Broadband Onshore-Offshore Lithospheric Investigation of Venezuela and the Antilles Arc Region seismic reflection data from offshore western Venezuela (Bonaire Basin) and around the Leeward Antilles are combined with existing geologic and geophysical data sets to examine the chronology of Late Cretaceous–Cenozoic tectonic events in this part of the Caribbean–South American plate boundary zone. These tectonic events have controlled the maturation and structural trapping of known hydrocarbons in the offshore Bonaire Basin and the adjacent onland Falcn Basin. We infer three tectonic phases that are constrained using these combined data sets. (1) The late Eocene–early Oligocene, north-south opening of the 3–6-km (1.8–3.7-mi)-thick Falcn-Bonaire Basin occurred along east-west–striking normal fault systems that have locally been inverted by later tectonic phases. These Paleogene normal faults rifted the Upper Cretaceous arc crust and Paleogene marine depositional sequences within the offshore Bonaire Basin. (2) Northwest-striking normal faults crosscut the older normal faults of the Bonaire Basin and Leeward Antilles and form deep, submarine rifts that contain up to 4 km (2.5 mi) of sedimentary fill and form deep-water channels between islands of the Leeward Antilles. Offshore well data and age of onshore sediments in the Falcn Basin indicate that this second phase of rifting occurred mainly during the late Oligocene to early Miocene and remains active to the present. (3) Inversion of the subaerial Falcn Basin commenced during the middle Miocene. This inversion phase is reflected in the present-day pattern of an east-northeast–trending fold-thrust belt that can be traced over 200 km (124 mi) along strike in the Falcn Basin. A second offshore fold-thrust belt (La Vela) can be traced over a distance of 175 km (108 mi) along strike and parallel to the northeast-trending Falcn Basin coast. Restoration of imbricate thrusts seen on seismic lines perpendicular to the La Vela fold-thrust belt indicates a minimum of 7 km (4.3 mi) of northeast-southwest–directed, thin-skinned shortening. Geochemical work indicates that source rocks for scattered occurrences of hydrocarbons in the Falcn Basin and its coastal zone are Paleogene and Miocene marine shale. Reservoir rocks are Tertiary marine sandstone and shale deposited in Paleogene rifts formed during the first tectonic phase in the late Eocene to early Oligocene. Structural traps were formed by thrusting during the second tectonic phase in the late Oligocene to early Miocene.

Escape tectonics and the extrusion of Alaska: Past, present, and future

The North Pacific Rim is a tectonically active plate boundary zone, parts of which may be characterized as a laterally moving orogenic stream. Crustal blocks are transported along large-magnitude strike-slip faults in western Canada and central Alaska toward the Aleutian–Bering Sea subduction zones. Throughout much of the Cenozoic, at and west of its Alaskan nexus, the North Pacific Rim orogenic Stream (NPRS) has undergone tectonic escape. During transport, relatively rigid blocks acquired paleomagnetic rotations and fault-juxtaposed boundaries while flowing differentially through the system, from their original point of accretion and entrainment toward the free face defined by the Aleutian–Bering Sea subduction zones. Built upon classical terrane tectonics, the NPRS model provides a new framework with which to view the mobilistic nature of the western North American plate boundary zone.

Calcite and quartz microstructural geothermometry of low-grade metasedimentary rocks, Northern Range, Trinidad

The Northern Range of Trinidad is a mountainous exposure of deformed metasedimentary rocks with Mesozoic depositional ages and Tertiary metamorphic ages located in the Caribbean–South American plate boundary zone. These rocks lack precise mineralogical indicators of metamorphic grade. Using temperature-sensitive quartz and calcite microstructures, and fission track data, we identify, describe, quantify, and map a nearly continuous spectrum of relatively low (>l50°C) to relatively high (250–400°C) deformation temperatures from east to west across the range. Average estimated rock exhumation rates also increase systematically along this trend. The analysis illuminates a sharp, east–west-trending, major thermal discontinuity (i.e. a fault) along the southern boundary of the range. Kinematic analysis of faults and shear bands adjacent to this boundary and modeling indicate that the fault probably dips southward at 80–85° and accommodated normal dip-slip at rates that decreased from west to east. Earlier experimental studies at high strain rates illustrated a strong strain rate dependency for quartz microstructural transitions. Our work indicates a possible strong temperature control for quartz microstructural transitions at natural strain rates.

Southern Alaska as an example of the long-term consequences of mountain building under the influence of glaciers

Southern Alaska is a continent-scale region of ongoing crustal deformation within the Pacific–North American plate boundary zone. Glaciers and glacial erosion have dictated patterns of denudation in the orogen over the last ∼5 Myr. The orogen comprises three discrete topographic domains from south to north, respectively: (1) the Chugach/St. Elias Range; (2) the Wrangell Mountains; and (3) the eastern Alaska Range. Although present deformation is distributed across the orogen, much of the shortening and uplift are concentrated in the Chugach/St. Elias Range. A systematic increase in topographic wavelength of the range from east to west reflects east-to-west increases in the width of a shallowly dipping segment of the plate interface, separation of major upper plate structures, and a decrease in the obliquity of plate motion relative to the plate boundary. Mean elevation decays exponentially from ∼2500 to ∼1100 m from east to west, respectively. Topographic control on the present and past distribution of glaciers is indicated by close correspondence along the range between mean elevation and the modern equilibrium line altitude of glaciers (ELA) and differences in the modern ELA, mean annual precipitation and temperature across the range between the windward, southern and leeward, northern flanks. Net, range-scale erosion is the sum of (1) primary bedrock erosion by glaciers and (2) erosion in areas of the landscape that are ice-marginal and are deglaciated at glacial minima. Oscillations between glacial and interglacial climates controls ice height and distribution, which, in turn, modulates the locus and mode of erosion in the landscape. Mean topography and the mean position of the ELA are coupled because of the competition between rock uplift, which tends to raise the ELA, and enhanced orographic precipitation accompanying mountain building, which tends to lower the ELA. Mean topography is controlled both by the 60° latitude and maritime setting of active deformation and by the feedback between shortening and uplift, glacial erosion, and orographic effects on climate accompanying mountain building.

Transpression, displacement partitioning, and exhumation in the eastern Caribbean / South American plate boundary zone

The Caribbean/South American plate boundary zone in northeastern Venezuela is a transpressive orogenic belt consisting from north to south of a nascent subduction zone (South Caribbean deformed belt), a volcanic arc (Leeward Antilles arc), a “hinterland” with high‐pressure (P)/low temperature (T) metamorphic rocks (Cordillera de la Costa belt), and a southern nonmetamorphic, foreland fold and thrust belt (Serranía del Interior). The geometry, style, and orientation of mid‐Cretaceous to Tertiary synmetamorphic deformation structures (D1) in the hinterland are compatible with formation in a right‐oblique subduction or collision zone in which displacement partitioning has occurred. Late Oligocene to Recent right‐oblique convergence resulted in the emplacement of the arc and hinterland on the passive South American margin and the formation of the foreland fold and thrust belt (D2); the displacements between the Caribbean and South American plates are partitioned as well. Both D1 and D2 deformations are diachronous: they are older in the west and younger in the east and related to the eastward passage of the Caribbean plate with respect to South America. The ascent, decompression, and exhumation of the high‐P/low‐T metamorphic rocks occurred in two stages: the first in the Cretaceous by arc‐parallel extension (D1) and the second in Neogene time by thrusting (D2) and subsequent erosion.

Paleomagnetism of Eocene Basalts of Mayreau, West Indies: Implications for Contrasts in Tectonic Rotation in the Southeastern Caribbean

Paleomagnetic directions were determined from 29 sites on the Grenadine island of Mayreau. Most data come from the middle Eocene Mayreau Basalt, which has been interpreted to be oceanic crust of the Grenada Basin uplifted in the half-horst of the southern Lesser Antilles arc platform. The objective was to assess the existence in the Mayreau Basalt of anomalies in paleomagnetic declination and inclination relative to Eocene South America and to other terranes in the Caribbean-South American plate-boundary zone. Primary and early deuteric magnetic directions were extracted that yielded site-means with α95 ≤ 15° from 15 sites in the Mayreau Basalt. Demagnetization removed a pervasive north-down component of low unblocking temperature. Tests demonstrated that the Eocene rocks were not remagnetized during Neogene island-arc magmatism and that the magnetization occurred before folding. The mean of site-means for the Mayreau Basalt is not different at the 95% confidence level from expected Eocene directions at Mayreau, from either North or South America. Therefore, there has been no net vertical-axis tectonic rotation or latitudinal transport of the Mayreau Basalt relative to the continents. Moreover, the eastward translation of the Grenada Basin as part of the Caribbean plate relative to the American plates must have been about a Euler pole that is geographically nearly 90° from the Grenada Basin. This result contrasts markedly with the large declination anomalies relative to South America in Tobago, Aruba, and other localities in the plate-boundary zone bordering northern South America. If the declination anomaly in Tobago is Eocene or younger and results from vertical-axis rotation in the plate-boundary zone, the difference in motions between Tobago and Mayreau must have been taken up at the southern Grenada Basin deformation front, with right-oblique subduction of Grenada Basin lithosphere and large rotation of the Tobago terrane.

A preliminary paleomagnetic pole for mid-Cretaceous rocks from Tobago: further evidence for large clockwise rotations in the Caribbean-South American plate boundary zone

The island of Tobago occupies the eastern end of the central (igneous) belt of the Caribbean-South American plate boundary zone. Volcaniclastic sediment of the Albian (∼ 100 Ma) Tobago Volcanic Group and dikes of similar age within it were sampled in two homoclinal sections with different attitudes. The mean of virtual geomagnetic poles for 12 sites (25.4°N, 24.1°E, A95 = 4.2°) is well defined, pre-tilting and apparently reliable, yet is far removed from a similar age reference pole for South America. Five other paleomagnetic studies of Cretaceous rocks from widely separated localities farther west in the plate boundary zone yield pole positions that are remarkably similar to the Tobago pole. Poles obtained from the Guajira Peninsula of Colombia, the islands of Aruba and Bonaire, and the Caribbean Mountains of Venezuela are among those that agree with the Tobago result. The paleolatitudes for study areas within the plate boundary zone are consistent with an origin on or near South America, yet the poles throughout the zone are rotated roughly 90°. Dextral relative motion between the Caribbean and South American plates was probably responsible for the rotation.

Spectral analysis of gravity anomalies and the architecture of tectonic wedging, NE Venezuela and Trinidad

We have analyzed the spectral content of free air gravity anomalies in the Caribbean‐South American plate boundary zone in order to determine better the near‐surface (0–120 km) distribution of crustal and upper mantle elements which give rise to the unusual gravity field of this region. The plate boundary zone in northeastern Venezuela and Trinidad is the site of the world's sea level continental minimum of Bouguer gravity anomalies, yet the region is also one of mild topography (mean value 43 m, maximum 1200 m). We find the mean depths to interfaces of significant density contrast at a variety of depths for portions of the plate boundary zone. We interpret interfaces at 30–35 km and 32 km beneath the Guyana Shield and the Aves Ridge, respectively, to be the Moho. Other shallow interfaces (5–14 km) are most likely sediment cover‐basement contacts in the Maturin foreland basin and southern Grenada Basin. Deeper interfaces (54–63 km) we associate with loaded and downwarped continental and oceanic South American lithosphere. The deepest boundaries, at depths of 89–120 km, may be related to detached or detaching oceanic lithosphere overridden by continental South America. We use our results to test the tectonic wedging model of the plate boundary zone recently published by Russo and Speed (1992). We find that the tectonic wedging model adequately describes many of the structural boundaries inferable from our analysis of gravity anomalies but that the model must be modified to include a thinner Guyana Shield crust.

Oblique collision and tectonic wedging of the South American continent and Caribbean terranes

Our studies of the surface structures, seismicity, and gravity anomalies of the Caribbean-South American plate-boundary zone of northeast Venezuela and Trinidad show that the crustal and upper lithospheric structure and kinematics of this zone are complex and differ markedly along strike. We propose a model of the zone9s lateral transition from near Trinidad west to Caracas, emphasizing three concepts: (1) the steepening, detachment, and then sinking away of Atlantic oceanic lithosphere that is (was) attached to the northern edge of continental South America; (2) the wedging seaward and imbricate thickening of the northern margin of the continent in the region of the detaching Atlantic slab; the continental wedge overrides the slab and is overridden by terranes attached to the Caribbean plate; and (3) that the structural and kinematic transitions are a consequence of the progressive oblique collision between continental South America and the overriding Caribbean terranes.

Extension and transtension in the plate boundary zone of the northeastern Caribbean

We propose that the Caribbean (Ca)‐North American (NA) plate boundary zone (pbz) from the Puerto Rico Trench to the Venezuelan Basin from Mona Canyon east has been in left‐transtension over the last 15–20 ma. A boundary‐normal component of extension occurs throughout the pbz and is a principal cause of the Puerto Rico Trench. Such extension is due to WNW velocity of NA‐Ca and the northward pullaway of NA from its S‐dipping slab, which is below Puerto Rico. Strike slip motion may be taken up among terranes in the pbz by rigid CCW rotation and by oblique slip at their boundaries. Rotation of the largest terrane, Puerto Rico‐Virgin Islands (PRVI), has caused such major structures as the Muertos thrust and Anegada Passage. The model implies NA‐Ca velocity estimated from Cayman transforms is more accurate than that from slip vectors from seisms in the NA slab.

Effects of density contrasts on the orientation of stresses in the lithosphere: Relation to principal stress directions in the Transverse Ranges, California

The influence of stresses arising from horizontal density contrasts on the orientation and relative magnitudes of principal stresses in an otherwise uniform lithospheric stress field is investigated. A simple model is constructed, in which a local deviatoric stress due to a density anomaly, embedded within or just below the lithosphere, and a regionally constant deviatoric stress field are each approximated by biaxial tensors. The net stress field is obtained from the sum of the two. Both the relative magnitudes of principal stresses and the magnitude of the angular difference in principal stress direction of the summed tensor compared with that obtained in the absence of buoyancy forces depend on two parameters. The first is the ratio τ/τ′, where τ is a measure of the magnitude of the regional deviatoric stress and τ′ is the magnitude of the stress arising from buoyancy forces associated with the density anomaly. The second parameter is the angle between the trend of the density anomaly and the direction of maximum compressional stress that obtains in the absence of any perturbation by the local buoyancy forces. The directions of the principal axes of the total stress field are found to differ by up to 90° from those of the reference stress field. The model is applied to the Transverse Ranges, California, where the observed 23° difference in orientation of principal horizontal compressive stress compared with the principal compressive stress direction in central California constrains the predicted value of τ/τ′ to be approximately −0.4. This is consistent with an independently calculated range of τ/τ′ in which τ′ is inferred from seismological constraints on the magnitude of density variations underneath the Transverse Ranges and τ is inferred from observations of heat flow along the San Andreas fault in central California. The agreement between the two estimates of τ/τ′ supports the hypothesis that the observed differences in horizontal principal stress orientation in California can be explained by the combined influence of a local negative buoyancy force under the Transverse Ranges and a regional stress field associated with transcurrent deformation within the Pacific‐North American plate boundary zone. The observed counterclockwise angular difference in principal horizontal stress direction in the Transverse Ranges compared with central California implies that the plane of maximum right lateral shear stress is also rotated counterclockwise relative to that in central California. This supports the possibility that the “big bend” in the San Andreas fault may be a consequence of the negative buoyancy forces acting in the Transverse Ranges, and not the cause of Transverse Ranges formation, as has often been assumed.