A conceptual shift is overdue in geodynamics. Popular models that present plate tectonics as being driven by bottom-heated whole-mantle convection, with or without plumes, are based on obsolete assumptions, are contradicted by much evidence, and fail to account for observed plate interactions. Subduction-hinge rollback is the key to viable mechanisms. The Pacific spreads rapidly yet shrinks by rollback, whereas the subduction-free Atlantic widens by slow mid-ocean spreading. These and other first-order features of global tectonics cannot be explained by conventional models. The behavior of arcs and the common presence of forearc basins on the uncrumpled thin leading edges of advancing arcs and continents are among features indicating that subduction provides the primary drive for both upper and lower plates. Subduction rights the density inversion that is produced when asthenosphere is cooled to oceanic lithosphere: plate tectonics is driven by top-down cooling but is enabled by heat. Slabs sink more steeply than they dip and, if old and dense, are plated down on the 660 km discontinuity. Broadside-sinking slabs push all sublithosphere oceanic upper mantle inward, forcing rapid spreading in shrinking oceans. Down-plated slabs are overpassed by advancing arcs and plates, and thus transferred to enlarging oceans and backarc basins. Plate motions make sense in terms of this subduction drive in a global framework in which the ridge-bounded Antarctic plate is fixed: most subduction hinges roll back in that frame, plates move toward subduction zones, and ridges migrate to tap fresh asthenosphere. This self-organizing kinematic system is driven from the top. Slabs probably do not subduct into, nor do plumes rise to the upper mantle from, the sluggish deep mantle.

Ultrahigh-pressure (UHP) metamorphic terranes in contractional orogens reflect descent of continental crust bonded to a dense, dominantly oceanic plate to depths of 90–140 km. All recognized well-documented UHP complexes formed during Phanerozoic time. Rocks are intensely retrogressed to low-pressure assemblages, with rare relict UHP phases retained in tough, refractory host minerals. Resurrected UHP slabs consist chiefly of quartzofeldspathic rocks and serpentinites; dense mafic + ultramafic lithologies comprise <10% of exhumed masses. Associated garnet-bearing ultramafic lenses are of four general origins: type A peridotite + eclogite pods reflect premetamorphic residence in the mantle wedge; type B masses were mantle-derived ultramafic-mafic magmas that rose into the crust prior to subduction; type C tectonic lenses were present in the oceanic lithosphere prior to underflow; and type D garnet peridotites achieved their deep-seated mantle mineralogy long before—and independent of—the subduction event that produced the UHP-phase assemblages in garnet peridotite types A, B, and C. Geochronology constrains the timing of protolith, peak, and retrograde recrystallization of gneissic, ultramafic, and eclogitic rocks. Round-trip pressure-temperature ( P-T ) paths were completed in <5–10 m.y., where ascent rates approximated subduction velocities. Exhumation from profound depth involves near-adiabatic decompression through P-T fields of much lower-pressure metamorphic facies. Many complexes consist of thin, allochthonous sheets, but those in eastern China and western Norway are about 10 km thick. Ductilely deformed nappes generated in subduction zones allow heat to be conducted away as sheet-like UHP complexes rise, cooling across both upper and lower surfaces. Thicker UHP massifs also must be quenched. Ascent along the subduction channel is driven mainly by buoyancy of low-density crustal material relative to the surrounding mantle. Rapid exhumation prevents establishment of a more normal geothermal regime in the subduction zone. Lack of H 2 O impedes back reaction, whereas its presence accelerates transformation to low-P phase assemblages. Late-stage domal uplifts characterize some collisional terranes; erosion, combined with underplating, contraction, tectonic aneurysms, and/or lithospheric plate shallowing, may further elevate mid-crustal UHP terranes toward the surface.

Geologic data combined with global positioning system (GPS) and paleomagnetic data from SW China indicate that continental crust can absorb tens to perhaps at least hundreds of kilometers of horizontal shear without developing either through-going faults or obvious structures capable of accommodating shear strain. The arcuate, left-lateral Xianshuihe-Xiaojiang and Dali fault systems bound crustal fragments that have rotated clockwise around the eastern Himalayan syntaxis. The two fault systems terminate to the south, but faults reappear farther south, and these continue the GPS velocity gradient. The shear must be transmitted across the Lanping-Simao fold belt without forming through-going faults. West of the Longmen Shan, a geodetically determined velocity gradient of ∼10 mm/yr at N60°E lies in an area not marked by through-going faults. If this deformation has been active for the past 8–11 m.y., it should have accumulated ∼100 km of shear across a belt ∼100 km wide. In both regions, there are no obvious structures that are capable of accommodating the shear. Paleomagnetic data from the southern Lanping-Simao belt are interpreted to indicate an unexpected zone of left-lateral shear present (Burchfiel and Wang, 2007) where rotation of crustal material is locally more than 90° across a zone unmarked by any mapped through-going faults. In these examples, the mechanism of deformation is not obvious, but we suggest it is distributed brittle deformation at a range of scales, from closely spaced faults to cataclastic deformation. In older terranes, recognition of such zones potentially adds an unknown uncertainty to field study and tectonic analyses.

In order to investigate the development of structures at scales smaller than that of an entire belt, we examined aspects of the mechanics of thin-skinned fold-and-thrust belts in cross section using an arbitrary Lagrangian-Eulerian frictional-plastic finite-element model. A series of models, beginning with the deformation of a thick uniform layer above a thin weak layer on a fixed base, sequentially illustrates the effects of including flexural isostatic subsidence, strain-softening, multiple layers of strong and very weak materials, and finally erosion and sedimentation. These continuum models develop thin shear zones containing highly sheared material that approximate fault zones. The corresponding structures are similar to those in fold-and-thrust belts and include: far-traveled thrust sheets, irregular-roof and smooth-roof duplexes, back thrusts, pop-ups, detachment folds, fault-bend folds, break thrusts, and piggyback basins. These structures can develop in-sequence or out-of-sequence, remain active for extended periods, or be reactivated. At the largest scale, the scale of the wedge, the finite-element model results agree with critical wedge solutions, but geometries differ at the sub-wedge scale because the models contain internal structures not predicted by the critical wedge stress analysis. These structures are a consequence of: (1) the complete solution of the governing equations (as opposed to a solution assuming a stress state that is everywhere at yield), (2) the initial finite-thickness layers, (3) the spatial and temporal variations of internal and basal strength, and (4) the coupling between surface processes and deformation of the wedge. The structural styles produced in models involving feedback with surface processes (erosion and sedimentation) are very similar to those mapped in the foothills of the southern Canadian Rockies and elsewhere. Although syndeformational sediments have been removed by postorogenic erosion across the foothills belt, evidence of the interaction between surface processes and deformation is preserved in the structural style.

The North American Cordillera in Canada and Alaska has been investigated through coincident and coordinated geological, geochemical, and geophysical studies along three corridors: (1) the Lithoprobe Southern Cordillera transect, (2) the ACCRETE and Lithoprobe Slave-Northern Cordillera Lithospheric Evolution (SNORCLE) transects, and (3) the Trans-Alaska Crustal Transect (TACT) program. Seismic-reflection and refraction experiments are integral to these studies and contribute to lithospheric-scale models that enable orogen-parallel comparisons to be made. Primary observations include three points: (1) Outward-verging, crustal-scale décollements are characteristic features of the orogen. The three trans-Cordillera transects exhibit decoupling zones that dip away from the Foreland belt to the lowermost crust or Moho. These inboard décollements above an indentor or cratonic backstop extend 500–600 km downdip in the Canadian Cordillera and 250 km downdip in the Alaskan Cordillera. The active subduction megathrusts form opposing décollements and generate structures in the overriding crust that mirror those above the facing intracrustal ramps. (2) Oblique convergence resulting in significant transpressional, transtensional, and orogen-parallel motion has yielded four major transcurrent fault systems that penetrate the entire crust and are associated with tectonic boundaries. (3) Beneath the entire Canadian Cordillera, the Moho remains remarkably flat and shallow despite the variety of ages, terrane compositions, and tectonomagmatic deformations spanned by the seismic corridors. These observations indicate that the Moho is an active, near-solidus, deformation zone that represents a young, re-equilibrated crust-mantle boundary. Beneath Alaska, crustal roots are observed over the subduction zone and at the indentor wedge, but the interior of the orogen also exhibits thin crust.

Restoration of tectonic elements in the central interior of the Canadian Cordillera southward to their paleogeographic position in the Mesozoic permits comparison of data across the active orogen, recognition of the interplay between coeval lithospheric thickening and basin evolution, and new constraints on models of tectonic evolution. The onset of Middle Jurassic clastic sedimentation in the Bowser basin, on the west side of the Jurassic orogen, occurred in response to accretionary events farther inboard. Shortening and thickening of the crust between the Alberta foreland basin on the east side of the Jurassic orogen and Bowser basin on the west side resulted in an Omineca highland between the two basins and lithospheric loading that influenced their Late Jurassic–Cretaceous sedimentation. The provenance of detritus in these basins, and in the Late Cretaceous Sustut basin on the east side of the Bowser basin, reveals migration of drainage divides in the intervening Omineca highland through time. Synchronous and compatible tectonic events within the basins and evolving accretionary orogen, and in rocks of the Stikine terrane and the western margin of North America, suggest that they were kinematically connected above a lower-crust detachment, beginning in the Middle Jurassic. The Coast belt was part of this wide, dynamically linked bivergent orogen from the mid-Cretaceous to earliest Cenozoic, and the lower-crust detachment rooted near the active plate margin. Nested within the orogen, the east-vergent thin-skinned Skeena fold belt, equivalent in scale to the Rocky Mountain fold-and-thrust belt, was also linked to the detachment system.

The Mesoproterozoic Belt-Purcell Basin of the United States–Canadian Rocky Mountains formed in a complex intracontinental-rift system. The basin contained three main fault blocks: a northern half-graben, a central horst, and a southern graben. Each had distinct internal stratigraphy and mineralization that influenced Phanerozoic sedimentation; the northern half-graben and horst formed a platform with a condensed section, whereas the southern graben formed the subsiding Central Montana trough. They formed major crustal blocks that rotated clockwise during Cordilleran thrust displacement, with transpressional shear zones deforming their edges. The northern half-graben was deepest and filled with a structurally strong prism of quartz-rich sedimentary rocks and thick mafic sills that tapered toward the northeast from >15-km-thick near the basin-bounding fault. This strong, dense prism was driven into the foreland basin as a readymade, critically tapered tectonic wedge and was inverted into the Purcell anticlinorium. Erosion did not breech the Belt-Purcell Supergroup in this prism during thrusting. The southern graben was thinner, weaker, lacked mafic sills, and was engorged with sheets of granite during thrusting. It was internally deformed to achieve critical taper and shed thick deposits of syntectonic Belt-Purcell–clast conglomerate into the foreland basin. A palinspastic map of the basin combined with a detailed paleocontinental map that juxtaposes the northeastern corner of the Siberian craton against western North America indicates that the basin formed at the complicated junction of three continental-scale rift zones.

The Lewis thrust, which is >225 km long and has a maximum displacement of >80 km, is a major Foreland belt structural element in the southeastern Canadian Cordillera. We use low-temperature thermochronometry in the preserved Lewis thrust sheet stratigraphic succession to constrain variations in both paleogeothermal gradient and Lewis thrust sheet thickness immediately prior to motion on the Lewis thrust fault. Fission-track and vitrinite reflectance data combined with stratigraphic data suggest that maximum Phanerozoic burial and heating occurred in the Lewis thrust sheet during a short interval (<15 m.y.) in late Campanian time immediately prior to thrusting (ca. 75 Ma). The data suggest that the late predeformational Lewis thrust sheet paleogeothermal gradient was between ∼18 and 22.5 °C/km, which is higher than that inferred for subsequent syn- and postdeformational intervals by other studies. The inferred paleotemperatures and geothermal gradients indicate that the preserved Lewis thrust sheet stratigraphic succession was overlain by ∼4–5.5 km of additional Late Cretaceous strata that were subsequently removed by erosional denudation. We estimate that the Lewis thrust sheet was ∼12–13.5 km thick when thrusting commenced. Deposition of the Late Cretaceous succession was terminated by the onset of displacement on the Lewis thrust (ca. 75 ± 5 Ma) and was followed by intervals of erosional denudation that are constrained stratigraphically by both early Oligocene and current erosion surfaces on the Lewis thrust sheet.

Quantification of fault-related illite neomineralization in clay gouge allows periods of fault activity to be directly dated, complementing indirect fault dating techniques such as dating synorogenic sedimentation. Detrital “contamination” of gouge is accounted for through the use of illite age analysis, where gouge samples are separated into at least three size fractions, and the proportions of detrital and authigenic illite are determined using illite polytypism (1M d = neoformed, 2M 1 = detrital). Size fractions are dated using the 40 Ar/ 39 Ar method, representing a significant improvement over earlier methods that relied on K-Ar dating. The percentages of detrital illite are then plotted against the age of individual size fractions, and the age of fault-related neoformed material (i.e., 0% detrital/100% neoformed illite) is extrapolated. The sampled faults and their ages are the Absaroka thrust (47 ± 9 Ma), the Darby thrust (46 ± 10 Ma), and the Bear thrust (50 ± 12 Ma). Altered host rock along the frontal Prospect thrust gives an age of 85 ± 12 Ma, indicating that the 46–50 Ma ages are not related to a regional fluid-flow event. These ages indicate that the faults in the Snake River–Hoback River Canyon section of the Wyoming thrust belt were active at the same time, indicating that a significant segment of the thrust belt (100 km 2 +) was active and therefore critically stressed in Eocene time.

A sub–Middle Jurassic unconformity is exhumed at Swift Reservoir, in the Rocky Mountain fold-and-thrust belt of Montana. The unconformity separates late Mississippian Sun River Dolomite of the Madison Group (ca. 340 Ma) from the transgressive basal sandstone of the Middle Jurassic (Bajocian-Bathonian) Sawtooth Formation (ca. 170 Ma). North-northwest–trending, karst-widened fractures (grikes) filled with cherty and phosphatic sandstone and conglomerate of the basal Sawtooth Formation penetrate the Madison Group for 4 m below the unconformity. The fractures link into sandstone-filled cavities along bedding planes. Clam borings, filled with fine-grained Sawtooth sandstone, pepper the unconformity surface and some of the fracture walls. Sandstone-filled clam borings also perforate rounded clasts of Mississippian limestone that lie on the surface of the unconformity within basal Sawtooth conglomerate. After deposition of the overlying foreland basin clastic wedge, the grikes were stylolitized by layer-parallel shortening and then buckled over fault-propagation anticlinal crests in the Late Cretaceous–Paleocene fold-and-thrust belt. We propose that the grikes record uplift and erosion followed by subsidence as the Rocky Mountain foreland experienced elastic flexure in response to tectonic loading at the plate boundary farther to the west during the Middle Jurassic. The forebulge opened strike-parallel fractures in the Madison Group that were then karstified. The sandstone-filled karst system contributes secondary porosity and permeability to the upper Madison Group, which is a major petroleum reservoir in the region. The recognition of the fractures as pre–Middle Jurassic revises previous models that have related them to Cretaceous or Paleocene fracturing over the crests of fault-propagation folds in the fold-and-thrust belt, substantially changing our understanding of the hydrocarbon system.

New structural, metamorphic, and geochronologic data from the Clearwater complex, north-central Idaho, define the origin and exhumation history of the complex. The complex is divisible into an external zone bound by normal faults and strike-slip faults of the Lewis and Clark Line, and an internal zone of Paleoproterozoic basement exposed in two shear zone–bounded culminations. U-Pb sensitive high-resolution ion microprobe (SHRIMP) dating of metamorphic zircon overgrowths from the external zone yield zircon growth at ca. 70–72 Ma and 80–82 Ma, during peak metamorphism and before tectonic exhumation of the external zone. U-Pb SHRIMP dating of metamorphic zircon rims from the internal zone record growth at ca. 64 and between 59 and 55 Ma. The older ages record pre-extension metamorphism. The younger rim ages were derived from fractured zircons in the Jug Rock shear zone, and they document the beginning of exhumation of the internal zone along deep-seated shear zones that transported the basement rocks to the west. The 40 Ar/ 39 Ar ages record quenching of the external zone starting ca. 54 Ma and the internal zone between 53 and 47 Ma by movement along the bounding faults and internal shear zones. After ca. 47 Ma, extension was accommodated via a west-dipping detachment that was active until after ca. 41 Ma. The Clearwater complex is interpreted as an Eocene metamorphic core complex that formed in an extensional relay zone between faults of the Lewis and Clark Line.