Tanya Atwater

Implications of Plate Tectonics for the Cenozoic Tectonic Evolution of Western North America

Magnetic anomaly patterns in the northeast Pacific Ocean combined with plate theory indicate that a trench existed offshore from western North America during mid-Tertiary time and that the present episode of strike-slip motion in the San Andreas fault system originated after the cessation of subduction, not earlier than 30 m.y. ago. At present, the American and Pacific plates appear to be moving past one another parallel to the San Andreas fault at a rate of 6 cm/yr. Data concerning the late Cenozoic history of motions between these plates is inconclusive, and so 2 probable models are examined. One assumes a constant motion of 6 cm/yr throughout the late Cenozoic, whereas the other assumes that the 2 plates were fixed with respect to one another until 5 m.y. ago, at which time they broke along the San Andreas fault system and began moving at 6 cm/yr. The second model implies that the San Andreas fault took up all the motion at the boundary between the North American and Pacific plates, while the first model suggests the broader view that much of the late Cenozoic tectonic activity of western North America is related to this boundary deformation. The models make testable predictions for the distribution of igneous rocks and for the total amount and timing of deformation expected. Extrapolation of the model of constant motions to the early Cenozoic suggests an era of slightly compressional strike-slip at the edge of North America. A major change in plate motions in late Mesozoic time is suggested.

East pacific rise: hot springs and geophysical experiments.

Hydrothermal vents jetting out water at 380� � 30�C have been discovered on the axis of the East Pacific Rise. The hottest waters issue from mineralized chimneys and are blackened by sulfide precipitates. These hydrothermal springs are the sites of actively forming massive sulfide mineral deposits. Cooler springs are clear to milky and support exotic benthic communities of giant tube worms, clams, and crabs similar to those found at the Gal�pagos spreading center. Four prototype geophysical experiments were successfully conducted in and near the vent area: seismic refraction measurements with both source (thumper) and receivers on the sea floor, on-bottom gravity measurements, in situ magnetic gradiometer measurements from the submersible Alvin over a sea-floor magnetic reversal boundary, and an active electrical sounding experiment. These high-resolution determinations of crustal properties along the spreading center were made to gain knowledge of the source of new oceanic crust and marine magnetic anomalies, the nature of the axial magma chamber, and the depth of hydrothermal circulation.

Pacific-North America Plate Tectonics of the Neogene Southwestern United States: An Update

We use updated rotations within the Pacific-Antarctica-Africa-North America plate circuit to calculate Pacific-North America plate reconstructions for times since chron 13 (33 Ma). The direction of motion of the Pacific plate relative to stable North America was fairly steady between chrons 13 and 4, and then changed and moved in a more northerly direction from chron 4 to the present (8 Ma to the present). No Pliocene changes in Pacific-North America plate motion are resolvable in these data, suggesting that Pliocene changes in deformation style along the boundary were not driven by changes in plate motion. However, the chron 4 change in Pacific-North America plate motion appears to correlate very closely to a change in direction of extension documented between the Sierra Nevada and the Colorado Plateau. Our best solution for the displacement with respect to stable North America of a point on the Pacific plate that is now near the Mendocino triple junction is that from 30 to 12 Ma the point was displaced along an azimuth of ∼N60°W at rate of ∼33 mm/yr; from 12 Ma to about 8 Ma the azimuth of displacement was about the same as previously, but the rate was faster (∼52 mm/yr); and since 8 Ma the point was displaced along an azimuth of N37°W at a rate of ∼52 mm/yr. We compare plate-circuit reconstructions of the edge of the Pacific plate to continental deformation reconstructions of North American tectonic elements across the Basin and Range province and elsewhere in order to evaluate the relationship of this deformation to the plate motions. The oceanic displacements correspond remarkably well to the continental reconstructions where deformations of the latter have been quantified along a path across the Colorado Plateau and central California. They also supply strong constraints for the deformation budgets of regions to the north and south, in Cascadia and northern Mexico, respectively. We examine slab-window formation and evolution in a detailed re-analysis of the spreading geometry of the post-Farallon microplates, from 28 to 19 Ma. Development of the slab window seems linked to early Miocene volcanism and deformation in the Mojave Desert, although detailed correlations await clarification of early Miocene reconstructions of the Tehachapi Mountains. We then trace the post-20 Ma motion of the Mendocino slab window edge beneath the Sierran-Great Valley block and find that it drifted steadily north, then stalled just north of Sutter Buttes at ∼4 Ma.

Interarc spreading and Cordilleran tectonics as alternates related to the age of subducted oceanic lithosphere

Abstract Old, cold oceanic lithosphere is denser and therefore gravitationally more unstable than younger, hotter oceanic lithosphere. Hence, whereas old lithosphere will sink under its own weight, subduction of young lithosphere may require an additional force. Interarc spreading occurs or occurred recently in the western Pacific, in the southern Atlantic, and possibly in the Mediterranean, where the subducted sea floor appears to be more than 50 m.y. old, and in many cases, is more than 100 m.y. old. In most of these regions, the ease with which the old dense lithosphere sinks may have contributed to a seaward migration of the trenches, which led to interarc spreading. Cordilleran tectonics, including high mountains and broad zones of deformation, are present on the margins of the eastern Pacific where the subducted oceanic lithosphere is younger than about 50 m.y. An extra force, which we presume to be necessary to cause subduction of the young lithosphere, may be responsible for the deformation and mountains just as an extra force seems necessary to drive continental collision in Asia. The extensive early Tertiary deformation across a broad zone of western North America may be related to the long-term, continuous subduction of young lithosphere of the Farallon and Kula plates.

Microplate capture, rotation of the western Transverse Ranges, and initiation of the San Andreas transform as a low-angle fault system

Tectonic rotation of the western Transverse Ranges block is explained by capture of the partially subducted Monterey microplate by the Pacific plate at about anomaly 6 time (ca. 20 Ma). As Pacific-Monterey spreading slowed and eventually ceased, the slip vector along the gently northeast dipping subduction interface beneath the California margin changed from slightly oblique subduction to transtensional dextral transform motion. This change in slip vector and a shift of Pacific plate motion eastward along the already subducted Monterey plate interface imply that the San Andreas transform began as a system of low-angle faults that locally subjected the overriding continental margin to distributed basal shear and crustal extension. This basal shear produced the rotated western Transverse Ranges. This model helps explain the timing of initial rotation and basin formation, the sudden appearance of widely distributed transform motion well inland of the margin in early Miocene time, why the western Transverse Ranges uniquely rotated as a large coherent crustal block, and several fundamental structural characteristics of central and southern California. The model also provides major constraints on the amount of Pacific-North America strike-slip motion, the position through time of offshore oceanic plates relative to onshore geology, and a general explanation for what may happen as a spreading ridge approaches a trench and the subduction zone evolves into a transform system.

How Laramide-Age Hydration of North American Lithosphere by the Farallon Slab Controlled Subsequent Activity in the Western United States

Starting with the Laramide orogeny and continuing through the Cenozoic, the U.S. Cordilleran orogen is unusual for its width, nature of uplift, and style of tectonic and magmatic activity. We present teleseismic tomography evidence for a thickness of modified North America lithosphere <200 km beneath Colorado and >100 km beneath New Mexico. Existing explanations for uplift or magmatism cannot accommodate lithosphere this thick. Imaged mantle structure is low in seismic velocity roughly beneath the Rocky Mountains of Colorado and New Mexico, and high in velocity to the east and west, beneath the tectonically intact Great Plains and Colorado Plateau. Structure internal to the low-velocity volume has a NE grain suggestive of influence by inherited Precambrian sutures. We conclude that the high-velocity upper mantle is Precambrian lithosphere, and the lowvelocity volume is partially molten Precambrian North America mantle. We suggest, as others have, that the Farallon slab was in contact with the lithosphere beneath most of the western U.S. during the Laramide orogeny. We further suggest that slab de-watering under the increasingly cool conditions of slab contact with North America hydrated the base of the continental lithosphere, causing a steady regional uplift of the western U.S. during the Laramide orogeny. Imaged low-velocity upper mantle is attributed to hydration-induced lithospheric melting beneath much of the southern Rocky Mountains. Laramide-age magmatic ascent heated and weakened the lithosphere, which in turn allowed horizontal shortening to occur in the mantle beneath the region of Laramide thrusting in the southern Rocky Mountains. Subsequent Farallon slab removal resulted in additional uplift through unloading. It also triggered vigorous magmatism, especially where asthenosphere made contact with the hydrated and relatively thin and fertile lithosphere of what now is the Basin and Range. This mantle now is dry, depleted of basaltic components, hot, buoyant, and weak.

Magnetic study of basalts from the Mid-Atlantic Ridge, lat 37°N

An intensive paleomagnetic and rock magnetic study has been carried out on the basalt samples from the FAMOUS area of the Mid-Atlantic Ridge at lat 37 °N. In addition to the samples obtained by the submersible Alvin from the floor and walls of the central rift valley, dredge and rock-drill samples were taken on a line perpendicular to the axis of the ridge out to a distance of 30 km from the central valley. Although the rocks from the floor of the central valley were roughly uniformly magnetized, a strong decrease in remanence intensity occurred at the valley walls. This progressive decrease continued until the remanent intensity was reduced by a factor of 5 at a distance 10 km and farther from the ridge axis. The other magnetic parameters of weak-field susceptibility, median demagnetizing field, Q , and Curie temperature were also roughly constant in samples from the valley floor. However, the rise in Curie temperature, which began at the valley walls, was well correlated with the decrease in remanent intensity and Q , an increase in the median demagnetizing field, and a sharp increase and then slow decrease in susceptibility; this rise can be attributed to the low-temperature oxidation of titanomagnetite to titanomaghemite. Of the 9 vertically oriented samples that were obtained from the valley floor and inner west wall, seven were normally magnetized and two were reversely magnetized. The two reversely magnetized units from the valley floor and east wall may be from blocks of older, pre-Brunhes crust that have not yet moved out of the rift valley. Comparison of the magnetic results of the surface rocks with models of the associated magnetic anomalies suggests that the oceanic magnetic layer can be represented by a permeable zone of pillow basalts roughly one km thick and that the oxidation state of these pillows progressively increases with time.

Magnetic lineations in the Northeast Pacific

Abstract Magnetic data for the northeast Pacific are summarized and interpreted in terms of sea floor spreading. Unclear areas exist in the “disturbed zone” and west of the Juan de Fuca Ridge. The direction of spreading changed about 55 my ago. An intermediate plate was gradually destroyed as segments of the ridge neared the continent and the timing of this destruction appears related to the timing of the origin of the San Andreas fault system.

Mid-Tertiary Tectonic Transition in the Aleutian Arc

The present lack of volcanic activity in the Western Aleutian Arc correlates with the lack of underthrusting at the western Aleutian trench predictable by plate tectonic theory. A simple plate model allows one to predict that another plate (Kula plate) and a spreading ridge (Kula ridge) lay south of the arc in early Tertiary time and that the Aleutian trench consumed the ridge in the middle Tertiary. Early Tertiary volcanism in the Western Aleutians may be related to underthrusting by the Kula plate. A mid-Tertiary orogeny in the Central Aleutians and a late Tertiary orogeny in Alaska may be related to the subduction of the Kula ridge.

Are spreading centers perpendicular to their transform faults

A COMMON assumption in seafloor spreading is that mid-ocean ridge crests are aligned perpendicular to their transform faults and, hence, to their spreading directions. There are some well known exceptions to this rule, for example, the Reykjanes Ridge. Vogt et al.1 suggested that spreading systems may take one of two configurations: either a transform faultless, oblique configuration or a perpendicular one. He then assigned the Reykjanes and certain older anomaly sets to the first category and the rest, the segmented, faulted ridges to the latter. We agree with this bimodal separation of ridge types, and here we discuss only the latter, the transform-faulted ‘perpendicular’ group. We examined all available detailed data from this group, and wherever we could find a fine scale map which included both transform fault and spreading centre we measured their trends. Of eight segmented, slow-spreading centres (half-rate less than 3 cm yr−1) we did not find a case which was, in fact, perpendicular. All were 6–38° oblique, and all were oblique in the sense which shortens the connecting transform faults, that is, the configurations in Fig. 1a as opposed to those shown in Fig. 1b . Fast- and some intermediate-rate spreading centres, on the other hand, seem to be perpendicular within the errors of measurements. These results are particularly interesting for the constraints that they place upon models of spreading centres in which the ridge crest-transform fault angle is used as a measure of the relative amounts of energy dissipated by these two features as motion occurs across them.