Mihai N. Ducea

Active foundering of a continental arc root beneath the southern Sierra Nevada in California

Seismic data provide images of crust–mantle interactions during ongoing removal of the dense batholithic root beneath the southern Sierra Nevada mountains in California. The removal appears to have initiated between 10 and 3 Myr ago with a Rayleigh–Taylor-type instability, but with a pronounced asymmetric flow into a mantle downwelling (drip) beneath the adjacent Great Valley. A nearly horizontal shear zone accommodated the detachment of the ultramafic root from its granitoid batholith. With continuing flow into the mantle drip, viscous drag at the base of the remaining ∼35-km-thick crust has thickened the crust by ∼7 km in a narrow welt beneath the western flank of the range. Adjacent to the welt and at the top of the drip, a V-shaped cone of crust is being dragged down tens of kilometres into the core of the mantle drip, causing the disappearance of the Moho in the seismic images. Viscous coupling between the crust and mantle is therefore apparently driving present-day surface subsidence.

The California Arc: Thick Granitic Batholiths, Eclogitic Residues, Lithospheric-Scale Thrusting, and Magmatic Flare-Ups

NOVEMBER 2001, GSA TODAY ABSTRACT Recent geological and geophysical data show that a significant fraction of the crust (~33 km) in the central Sierra Nevada batholith is granitic, requiring that the batholith be underlain by a significant residual mass prior to Cenozoic extension. Although batholith residua are commonly thought to be granulites, xenolith data indicate that eclogite facies residues were an important part of the California arc at depth. The arc was continuously active for >140 m.y., yet most surface and/or shallow crustal magmatism took place via two short-lived episodes: one in the Late Jurassic (160–150 Ma), and a second, more voluminous one in the Late Cretaceous (100–85 Ma). These magmatic flare-ups cannot be explained solely by increases in convergence rates and magmatic additions from the mantle. Isotopic data on xenoliths and midcrustal exposures suggest that North American lower crustal and lithospheric mantle was underthrusted beneath accreted rocks in the arc area. The Late Cretaceous flare-up is proposed to be the result of this major west dipping–lithospheric scale thrusting, an event that preceded flare-up by ~15–25 m.y. I suggest that the central part of the arc shut off at ~80 Ma because the source became melt-drained and not because of refrigeration from a shallowly subducting slab.

Buoyancy sources for a large, unrooted mountain range, the Sierra Nevada, California: Evidence from xenolith thermobarometry

Xenoliths hosted by Cenozoic volcanic flows and plugs from the Central Sierra Nevada and Eastern Sierra Nevada, Owens Valley, and Inyo Mountains were studied for petrography and thermobarometry. The Central Sierra Nevada suite consists of abundant lower crustal feldspathic granulites, garnet clinopyroxenites, and mantle-derived peridotites and garnet websterites. Mafic crustal assemblages occur down to ∼65–70 km, although below 35–40 km, they are mainly in the eclogite facies. In contrast, the Eastern Sierra Region suites show peridotitic, pyroxenitic, and harzburgitic assemblages at depths of ≥35–40 km. They define an adiabat in PT space (T ∼ 1180–1250°C), suggesting the presence of the asthenospheric upper mantle close to the base of the crust. The ultramafic mantle rocks from the Central Sierra Nevada also define an adiabatic slope in PT space, possibly an artifact of side heating from the east. There is xenolith evidence that the Sierra Nevada lost about half of its original crust on the eastern side of the range. Regardless of the actual mechanism of crustal thinning, the loss of the eclogitic lowermost crust and replacement by peridotite in the eastern Sierra Nevada is a process accompanied by a substantial density decrease (>100 kg/m^3). Overall, if the mechanism of eclogitic lowermost crust removal is viable, there are enough buoyancy sources to explain topographic differences between the Sierra Nevada and the adjacent Basin and Range, assuming isostatic equilibrium.

Origin of High Mountains in the Continents: The Southern Sierra Nevada

Active and passive seismic experiments show that the southern Sierra, despite standing 1.8 to 2.8 kilometers above its surroundings, is underlain by crust of similar seismic thickness, about 30 to 40 kilometers. Thermobarometry of xenolith suites and magnetotelluric profiles indicate that the upper mantle is eclogitic to depths of 60 kilometers beneath the western and central parts of the range, but little subcrustal lithosphere is present beneath the eastern High Sierra and adjacent Basin and Range. These and other data imply the crust of both the High Sierra and Basin and Range thinned by a factor of 2 since 20 million years ago, at odds with purported late Cenozoic regional uplift of some 2 kilometers.

Production and loss of high-density batholithic root, southern Sierra Nevada, California

Received 20 February 2002; revised 21 March 2003; accepted 1 July 2003; published 18 November 2003. [1] Eclogites are commonly believed to be highly susceptible to delamination and sinking into the mantle from lower crustal metamorphic environments. We discuss the production of a specific class of eclogitic rocks that formed in conjunction with the production of the Sierra Nevada batholith. These high-density eclogitic rocks, however, formed by crystal-liquid equilibria and thus contrast sharply in their petrogenesis and environment of formation from eclogite facies metamorphic rocks. Experimental studies show that when hydrous mafic to intermediate composition assemblages are melted in excess of 1 GPa, the derivative liquids are typical of Cordilleran-type batholith granitoids, and garnet + clinopyroxene, which is an eclogitic mineralogy, dominate the residue assemblage. Upper mantlelower crustal xenolith suites that were entrained in mid-Miocene volcanic centers erupted through the central Sierra Nevada batholith are dominated by such garnet clinopyroxenites, which are shown further by geochemical data to be petrogenetically related to the overlying batholith as its residue assemblage. Petrogenetic data on garnet pyroxenite and associated peridotite and granulite xenoliths, in conjunction with a southward deepening oblique crustal section and seismic data, form the basis for the synthesis of a primary lithospheric column for the Sierra Nevada batholith. Critical aspects of this column are the dominance of felsic batholithic rocks to between 35 and 40 km depths, a thick (� 35 km) underlying garnet clinopyroxenite residue sequence, and interlayered spinel and underlying garnet peridotite extending to � 125 km depths. The peridotites appear to be the remnants of the mantle wedge from beneath the Sierran arc. The principal source for the batholith was a polygenetic hydrous mafic to intermediate composition lower crust dominated by mantle wedge-derived mafic intrusions. Genesis of the composite batholith over an � 50 m.y. time interval entailed the complete reconstitution of the Sierran lithosphere. Sierra Nevada batholith magmatism ended by � 80 Ma in conjunction with the onset of the Laramide orogeny, and subsequently, its underlying mantle lithosphere cooled conductively. In the southernmost Sierranorthern Mojave Desert region the subbatholith mantle lithosphere was mechanically delaminated by a shallow segment of the Laramide slab and was replaced by underthrust subduction accretion assemblages. Despite these Laramide events, the mantle lithosphere of the greater Sierra Nevada for the most part remained intact throughout much of Cenozoic time. A pronounced change in xenolith suites sampled by Pliocene-Quaternary lavas to

U-Th-Pb geochronology of the Coast Mountains batholith in north-coastal British Columbia: Constraints on age and tectonic evolution

Previously published and new U-Pb geochronologic analyses provide 313 zircon and 59 titanite ages that constrain the igneous and cooling history of the Coast Mountains batholith in north-coastal British Columbia. First-order findings are as follows: (1) This segment of the batholith consists of three portions: a western magmatic belt (emplaced into the outboard Alexander and Wrangellia terranes) that was active 177–162 Ma, 157–142 Ma, and 118–100 Ma; an eastern belt (emplaced into the inboard Stikine and Yukon-Tanana terranes) that was active ca. 180–110 Ma; and a 100–50 Ma belt that was emplaced across much of the orogen during and following mid-Cretaceous juxtaposition of outboard and inboard terranes. (2) Magmatism migrated eastward from 120 to 80 (or 60) Ma at a rate of 2.0–2.7 km/Ma, a rate similar to that recorded by the Sierra Nevada batholith. (3) Magmatic flux was quite variable through time, with high (>35–50 km 3 /Ma per km strike length) flux at 160–140 Ma, 120–78 Ma, and 55–48 Ma, and magmatic lulls at 140–120 Ma and 78–55 Ma. (4) High U/Th values record widespread growth (and/or recrystallization) of metamorphic zircon at 88–76 Ma and 62–52 Ma. (5) U-Pb ages of titanite record rapid cooling of axial portions of the batholith at ca. 55–48 Ma in response to east-side-down motion on regional extensional structures. (6) The magmatic history of this portion of the Coast Mountains batholith is consistent with a tectonic model involving formation of a Late Jurassic–earliest Cretaceous magmatic arc along the northern Cordilleran margin; duplication of this arc system in Early Cretaceous time by >800 km (perhaps 1000–1200 km) of sinistral motion (bringing the northern portion outboard of the southern portion); high-flux magmatism prior to and during orthogonal mid-Cretaceous terrane accretion; low-flux magmatism during Late Cretaceous–Paleocene dextral transpressional motion; and high-flux Eocene magmatism during rapid exhumation in a regime of regional crustal extension.

Igniting flare-up events in Cordilleran arcs

High-fl ux pulses of magmatism that make up most of the exposed North American Cordilleran arcs are derived primarily from upper plate lithospheric source materials, and not the mantle wedge as most models would predict, based on a compilation of thousands of previously published Sr, Nd, and O isotopic data. Mass balance calculations show that no more than 50% of that mass can be mantle-derived. Flare-ups must have fundamentally developed simultaneously with crustal/lithospheric thickening, thus implying a connection. Subduction erosion from the trench side, and retroarc shortening from the foreland side are the main tectonic shortening processes that operate in conjunction with high fl ux magmatism during subduction, and therefore are likely triggers for fl are-up events in arc. These arcs represent the sites of crustal differentiation, and thus contribute to net continental growth, only if dense residual lower crust was returned to the convective mantle. Isotopic data shown here suggest that if convective removal of batholithic roots takes place, it must be a consequence and not a cause of episodic fl are-ups. The Altiplano-Puna Volcanic Complex in South America may be the most recent continental arc segment in flmode.

Building the Pamirs: The view from the underside

The Pamir mountains are an outstanding example of extreme crustal shortening during continental collision that may have been accommodated by formation of a thick crust—much thicker than is currently thought—and/or by continental subduction. We pre- sent new petrologic data and radiometric ages from xenoliths in Miocene volcanic rocks in the southeastern Pamir mountains that suggest that Gondwanan igneous and sedimentary assemblages were underthrust northward, buried to .50-80 km during the ear- ly stage of the India-Asia collision, and then heated and partly melted during subsequent thermal relaxation before finally being blasted to the surface. These xenoliths, the deepest crustal samples recovered from under any active collisional belt, provide direct evidence for early Cenozoic thickening of the Pamirs and lower- crustal melting during collision; the xenoliths also suggest that the present mountain range was a steady-state elevated plateau for most of the Cenozoic.

Constraints on the bulk composition and root foundering rates of continental arcs: A California arc perspective

[1]Garnetpyroxenites are the most common deep lithospheric xenolith assemblages found in Miocene volcanic rocks that erupted through the central part of the Sierra Nevada batholith. Elemental concentrations and isotope ratios are used to argue that the Sierra Nevada granitoids and the pyroxenite xenoliths are the melts and the residues/cumulates, respectively, resulting from partial melting/fractional crystallization at depths exceeding 35-40 km. The estimated major element chemistry of the protolith resembles a basaltic andesite. Effectively, at more than about 40 km depth, batholith residua are eclogite facies rocks. Radiogenic and oxygen isotope ratios measured on pyroxenites document unambiguously the involvement of Precambrian lithosphere and at least 20-30% (mass) of crustal components. The mass of the residual assemblage was significant, one to two times the mass of the granitic batholith. Dense garnet pyroxenites are prone to foundering in the underlying mantle. An average removal rate of 25-40 km 3 /km Myr is estimated for this Cordilleran-type arc, although root loss could have taken place at least in part after the cessation of arc magmatism. This rate is matched by the average subcrustal magmatic addition of the arc (∼23-30 km 3 /km Myr), suggesting that the net crustal growth in this continental arc was close to zero. It is also suggested that in order to develop a convectively removable root, an arc must have a granitoid melt thickness of at least 20-25 km. Residues of thinner arcs should be mostly in the granulite facies; they are not gravitationally unstable with respect to the underlying mantle.

Sm^Nd dating of spatially controlled domains of garnet single crystals: a new method of high-temperature thermochronology

Ganguly and Tirone [Meteorit. Planet. Sci. 36 (2001) 167^175] recently presented a method of determining the cooling rates of rocks from the difference between the core and bulk ages of a crystal, as determined by a single decay system. Here we present the first application of the method using the core and bulk ages of garnet single crystals, according to the Sm^Nd decay system, in two rock samples with contrasting cooling rates, which can be constrained independently. The samples belong to the metamorphic core complex, Valhalla, British Columbia, and the mid-crustal magmatic arc exposure of the Salinian terrane, California. We have micro-sampled the garnet crystals over specific radial dimensions, and measured the Nd isotopes of these small sample masses, as NdO þ via solid source mass spectrometry, to determine the Sm^Nd age difference between the core and bulk crystals. Using a peak metamorphic P^T condition of 8 > 1 kbar, 820 > 30‡C [Spear and Parrish, J. Petrol. 37 (1996) 733^765], the core (67.3 > 2.3 Ma) and bulk (60.9 > 2.1 Ma) ages of the British Columbian garnet sample yield a cooling rate of 2^13‡C/Myr, which is in very good agreement with the cooling rates that we have derived by modeling the retrograde Fe^Mg zoning in the same garnet, and assuming the same peak metamorphic P^T condition. Considering earlier cooling rate data derived from closure temperature vs. age relation of multiple geochronological systems [Spear and Parrish, J. Petrol. 37 (1996) 733^ 765], a cooling rate of V15^20‡C/Myr seems most reasonable for the Valhalla complex. Diffusion kinetic analysis shows that the Sm^Nd core age of the selected garnet crystal could not have been disturbed during cooling. Consequently, the core age of the garnet crystal, 67.3 > 2.3 Ma, corresponds to the peak metamorphic age of the Valhalla complex. The Salinian sample, on the other hand, yields indistinguishable core (78.2 > 2.7 Ma) and bulk (77.9 > 2.9 Ma) ages, as expected from its fast cooling history, which can be constrained by the results of earlier studies. The Sm^Nd decay system in garnet has relatively high closure temperature (usually s 650‡C); therefore, the technique developed in this paper fills an important gap in thermochronology, since the commonly used thermochronometers are applicable only at lower temperatures. Simultaneous modeling of the retrograde Fe^Mg zoning in garnet, spatially resolved Sm^Nd ages of garnet single crystals, and resetting of the bulk garnet Sm^Nd age