Mark Cloos

Lithospheric buoyancy and collisional orogenesis: Subduction of oceanic plateaus, continental margins, island arcs, spreading ridges, and seamounts

The sizes of continental blocks, basaltic oceanic plateaus, and island arcs that would cause collisional orogenesis when they enter a subduction zone are calculated in an analysis based upon the assumption of local isostasy and the assumption that plate subduction is primarily driven by the negative buoyancy of the lithosphere. Buoyancy analysis indicates that the bulk density contrast between 80-m.y.-old oceanic lithosphere capped by a 7-km-thick basaltic crust and the less dense underlying asthenosphere is on the order of 0.04 gm/cm 3 . Oceanic lithosphere that is ∼10 m.y. old is the youngest that is more dense than the asthenosphere and hence inherently susceptible to subduction. Subduction zone metamorphism causes the crustal layer of basalt/gabbro to transform into more dense amphibolite and eclogite. Where eclogite formation is extensive, the descending oceanic lithosphere increases in bulk density by as much as 0.04 gm/cm 3 . Lithosphere that is 100 km thick with a 30-km-thick granitic continental crust resists Subduction because it is ∼0.09 gm/cm 3 less dense than the asthenosphere. Contrasts in lithospheric bulk density (crust + mantle) of 3 are the difference between whether subduction is nearly inevitable (as for normal ocean crust) or greatly resisted (as for thick, ancient continents). Collisional orogenesis is defined as a plate interaction of the sort that causes a rearrangement of plate motions, generally with the initiation of a new subduction zone and the creation of mountains. Buoyancy analysis indicates that only bodies of continental and oceanic island are crust that are > ∼15 km thick make the lithosphere buoyant enough to jam a subduction zone. Oceanic island arc complexes built upon ocean crust typically must be active for more than ∼20 m.y. to attain crustal thicknesses so that their attempted subduction causes collisional orogenesis. Oceanic plateaus where basaltic crust as much as ∼17 km thick caps 100-km-thick lithosphere are inherenty subductable and actually less buoyant than normal oceanic lithosphere following subduction metamorphism. Basaltic plateaus must have crustal thicknesses >∼30 km to typically cause collisional orogenesis during subduction. Short subducting seamounts (

Subduction-channel model of prism accretion, melange formation, sediment subduction, and subduction erosion at convergent plate margins: 2. Implications and discussion

Many geological and geophysical investigations, particularly the Deep Sea Drilling Project, have shown that convergent plate margins are highly diverse features. For example, at some sites of subduction, such as the Lesser Antilles, the bedded sediment atop the incoming oceanic plate is extensively offscraped, whereas at others, such as Mariana, not only is the incoming sediment completely subducted beneath crystalline rock but portions of the overriding plate are undergoing subduction erosion. Earthquakes indicate wide variations in stress distribution within and between sites of plate convergence. Many ancient accretionary complexes include tracts of intensely-deformed subduction melange that contain blocks of mafic greenstones. Some contain bodies of thoroughly recrystallized blueschist that were uplifted from depths of 20 to 30 km. A comprehensive model for convergent plate margins must explain these and numerous other observations. Although the still widely cited imbricatethrust model for prism accretion qualitatively explains some observations at subduction zones, it does not account for many others, such as deep sediment subduction and subduction erosion.The subduction-channel model postulates essentially the same basic mechanics for all convergent plate margins that have attained a quasi-steady state (typically reached after about 20 Ma of subduction at speeds of 10 to 20 km Ma−1). It assumes that the subducting sediment deforms approximately as a viscous material once it is dragged into a relatively thin shear zone, or subduction channel, between the downgoing plate and the overriding one. It predicts the overall movement patterns of the sediment deforming within the channel and near its inlet, accounts for most of the observed features at convergent plate margins, and quantifies the processes of sediment subduction, offscraping, and underplating, and the formation of subduction melange. The predicted variations in tectonic behavior depend upon such site-specific variables as the speed of subduction, the supply of sediment, the geometry of the descending plate, and the topography and structure of the overriding block.

Thrust-type subduction-zone earthquakes and seamount asperities: A physical model for seismic rupture

A thrust-type subduction-zone earthquake of M{sub W} 7.6 ruptures an area of {approximately}6,000 km{sup 2}, has a seismic slip of {approximately}1 m, and is nucleated by the rupture of an asperity {approximately}25km across. A model for thrust-type subduction-zone seismicity is proposed in which basaltic seamounts jammed against the base of the overriding plate act as strong asperities that rupture by stick-slip faulting. A M{sub W} 7.6 event would correspond to the near-basal rupture of a {approximately}2-km-tall seamount. The base of the seamount is surrounded by a low shear-strength layer composed of subducting sediment that also deforms between seismic events by distributed strain (viscous flow). Planar faults form in this layer as the seismic rupture propagates out of the seamount at speeds of kilometers per second. The faults in the shear zone are disrupted after the event by aseismic, slow viscous flow of the subducting sediment layer. Consequently, the extent of fault rupture varies for different earthquakes nucleated at the same seamount asperity because new fault surfaces form in the surrounding subducting sediment layer during each fast seismic rupture.

Cenozoic tectonics of New Guinea

Major hydrocarbon discoveries have been made in eastern and westernmost New Guinea, and there is great potential for additional discoveries. Although the island is a type locality for arc-continent collision during the Cenozoic, the age, number, and plate kinematics of the events that formed the island are vigorously argued. The northern part of the island is underlain by rocks with oceanic island arc affinities, and the southern part is underlain by the Australian continental crust. Based on regional sedimentation patterns, it is argued herein that the Cenozoic tectonic history of the island involves two distinct collisional orogenic events.The first Cenozoic event, the Peninsular orogeny of Oligocene age (35–30 Ma), was restricted to easternmost New Guinea. Emergent uplifts that shed abundant detritus resulted from the subduction of the northeastern corner of the Australian continent beneath part of the Inner Melanesian arc. This collision uplifted the Papuan ophiolite and formed the associated mountainous uplift that was the primary source of siliciclastic sediments that largely filled the Aure trough. Between the Oligocene and Miocene, the paleogeography of the region was similar to present-day New Caledonia. The continental crust under central and western New Guinea remained a passive margin.The second event, the Central Range orogeny, began in the latest middle Miocene, when the bulldozing of Australian passive-margin strata first created emergent uplifts above a north-dipping subduction zone beneath the western part of the Outer Melanesian arc. The cessation of carbonate shelf sedimentation and widespread initiation of siliciclastic sedimentation on top of the Australian continental basement is dated at about 12 Ma. This collision emplaced the Irian ophiolite and created the present mountainous topography forming the spine of the island.

Shear-zone thickness and the seismicity of Chilean- and Marianas-type subduction zones

Chilean-type convergent margins have many large (M > 7.6) earthquakes, whereas Marianas-type ones do not. This dichotomy is enigmatic if the plate interface is viewed as a thin frictional decollement, whereas it becomes understandable if it is viewed as a relatively thick, sediment-filled shear zone, which thins or thickens arcward depending on subduction speed and sediment supply. Chilean-type margins have thick trench fills, and their shear zones generally thin arcward from inlets as much as several thousand metres high, the most pronounced thinning being located near backstops. Tall (up to several kilometres) seamounts are subducted essentially intact to relatively great depths and confining pressures before jamming into the roof of the channel and becoming seismogenic asperities. Their near-basal ruptures can generate large thrust-type earthquakes, mainly concentrated in seismic fronts near backstops. Marianas-type margins, in contrast, have thin trench fills, and their shear zones generally thicken arcward from inlets that can be as little as 300 m high. Seamounts are truncated near the inlet at low confining pressures and generate only small earthquakes. After passing the inlet, they do not touch the roof and therefore cannot generate large earthquakes. A similar mechanism may explain seismic gaps at sediment-poor regions of subduction zones.

Bubbling Magma Chambers, Cupolas, and Porphyry Copper Deposits

Porphyry copper deposits are the major source of copper and significant sources of molybdenum, gold, and other metals. They are associated with the near-surface intrusion of small stocks of intermediate composition. They can form when H2O-unsaturated magma is emplaced into wall rock that is cool enough that steep lateral thermal gradients create a narrow solidification front. At depths less than ∼4 km, cooling and crystallization cause fluid saturation to occur within sidewall magma that is mobile because it contains less than ∼25% suspended crystals. After a sufficient volume of bubbles forms, mobile sidewall magma buoyantly rises instead of sinking. The bubbles expand as they decompress, and at depths of ∼2 km they become large enough to rise on their own. separate from the upwelled magma, and charge the cupola at the top of the stock with magmatic fluid. The partially degassed magma sinks into the interior of the stock. Upwelling of saturated sidewall magma entrains deeper-seated, nearly saturated magma, which decompresses and saturates as it rises. As the system cools, the depth of H2O saturation and sidewall upwelling increases. Bubbles of copper-rich fluid are generated where the saturation front extends to depths of ∼6 km or more. Overall, the system is cooling, but the upward advection of heat maintains the cupola region at roughly constant position for the life of convective upwelling along the sidewalls. Porphyry copper ore deposits can form where draining of the fluid pocket beneath a cupola is steady and a large volume of magma is cycled through the system. Magma in the stock that escapes to intrude commonly has a porphyritic texture because crystal growth is enhanced, and nucleation is suppressed when the magma is H2O saturated. Porphyry copper deposits of common size can form during the solidification of large stocks. Super-giant porphyry copper deposits can form where the saturation front propagates from a stock into an underlying batholithic chamber with a magma volume on the order of 1000 km3 and a top at depths of 10 to 15 km.

Landward-dipping reflectors in accretionary wedges: Active dewatering conduits?

Landward-dipping seismic reflectors (LDRs) within the accretionary wedge at convergent plate margins have generally been interpreted as tilted-bedding or thrust-fault surfaces (e.g., Mexico, northern Japan). This interpretation accords with models in which trench-axis sediments are scraped off the descending plate to form a series of imbricate thrust sheets. However, intensely disturbed melanges with a regionally extensive landward-dipping foliation are found beneath slope sediments in many ancient and active subduction complexes. This observation, combined with the fact that large volumes of fluid are released during dewatering of subducting and previously accreted sediments, leads to the proposal that some of the seismic reflectors are porous, laterally extensive fractureways filled with dewatering fluids derived from deeper underthrust water-rich sediments. These fractureways trend subparallel to the regional foliation of the accreted melange wedge. The contrast in acoustic impedance between the high-porosity fluid-filled fractureways and the melange wall rocks is enhanced whenever methane and/or carbon dioxide exsolve from rising waters. Ancient LDRs may be recognized where mineralized veins trend subparallel to the regional foliation of melange belts generated during subduction accretion. 42 references, 2 figures.

Pliocene-Pleistocene asymmetric unroofing of the Irian fold belt, Irian Jaya, Indonesia: Apatite fission-track thermochronology

The Central Range of New Guinea is an orogenic belt up to 100 km wide and more than 1300 km long with numerous peaks over 3 km high. The orogen was generated since the middle Miocene by the collisional underthrusting of the northern passive margin of the Australian continent into a north-dipping subduction zone beneath the Melanesian island arc. South of Puncak Jaya, the highest peak in New Guinea, Pliocene–Pleistocene uplift and unroofing have exposed Tertiary carbonates and Mesozoic clastic rocks of a rift and passive-margin sequence above unmetamorphosed Paleozoic strata and greenschist-facies slates probably of Precambrian age. Eight magmatic apatite samples from Pliocene intrusions of intermediate composition in the Gunung Bijih (Ertsberg) mining district at the crest of the Central Range in Irian Jaya, Indonesia, yield pooled fission-track ages ranging between 3.7 ± 0.9 and 2.0 ± 0.3 Ma (±1σ). Long etchable mean track lengths (>14 µm) and narrow track length distributions indicate rapid cooling, as expected for shallow-level intrusions. These track-length data and the observation that the Grasberg pluton was emplaced into its own volcanic cover indicate that <2 km of material has been eroded from the top of this part of the range since the Pliocene. Seven detrital apatite samples from the Triassic-Jurassic Tipuma and the Carboniferous–Permian Aiduna Formations and apatite from two igneous dikes exposed halfway down the southern slope of the range yield pooled fission-track ages ranging between 2.7 ± 0.7 and 2.0 ± 0.5 Ma (±1σ). Etchable mean track lengths of ≈9–12 µm and wide track length distributions indicate slower cooling compared to the Pliocene intrusions. Complete resetting of provenance fission-track ages in detrital apatite requires burial deeper than ≈4 km. Unroofing rates are ≈1.7 km/m.y. on the southern flank of the range but >0.7 km/m.y. at the crest. These fission-track data combined with local and regional geologic relationships indicate that unroofing of the Irian fold belt occurred at least two and a half times faster, and perhaps more than five times faster, on the lower slope than at the crest. This difference is probably due to regional weather patterns that result in up to ≈ 11 m/yr of orographically induced precipitation on the southern slope of the Irian Central Range but only ≈3 m/yr near the highest elevations.

Pliocene Cu-Au-Bearing Igneous Intrusions of the Gunung Bijih (Ertsberg) District, Irian Jaya, Indonesia: K-Ar Geochronology

Nine distinct potassic intermediate igneous bodies have been identified during detailed surface and underground mapping of