Eg 100 Th Tectonics Deposits

of 40

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
40 pages
0 downs
  1097 Metallogenic Provinces in an Evolving Geodynamic Framework R OBERT K ERRICH , † Department of Geological Sciences, University of Saskatchewan, 114 Science Place, Saskatoon, Saskatchewan, Canada S7N 5E2 R ICHARD J. G OLDFARB , U.S. Geological Survey, Box 25046, MS 964, Denver Federal Center, Denver, Colorado 80225-0046, and Department of Geological Sciences, University of Colorado, 2200 Colorado Ave., Campus Box 399, Boulder, Colorado 80309 AND J EREMY P. R ICHARDS Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E3  Abstract Thermal decay of Earth resulted in decreased mantle-plume intensity and temperature and consequently agradual reduction of abundant komatiitic basalt ocean plateaus at ~2.6 Ga. In the Neoarchean, ocean crust was~11 km thick at spreading centers, and abundant bimodal arc basalt-dacite magmatic edifices were constructedat convergent margins. Neoarchean greenstone belt orogenesis stemmed from multiple terrane accretion inCordilleran-style external orogens with multiple sutures, where oceanic plateaus captured arcs by jammingsubduction zones, and plateau crust melted to generate high thorium tonalite-trondhjemite-granodiorite suites.Archean cratons have a distinctive ~250- to 350-km-thick continental lithospheric mantle keel with buoyant re-fractory properties, resulting from coupling of the buoyant residue of deep plume melting to imbricatedplateau-arc crust. In contrast, Proterozoic and younger continental lithospheric mantle is <150 km thick,denser, and less refractory and therefore easily reworked in younger orogens. The supercontinent cycle has op-erated since ~2.8 Ga: Kenorland assembled at ~2.7 Ga, Columbia ~1.8 Ga, Rodinia ~1 Ga, and Pangea ~0.3Ga. Dispersal may have been triggered by superplumes.Komatiite-hosted Ni deposits are related to plumes, where sulfide saturation resulted from crustal contam-ination. Base metal-rich volcanic rock-associated massive sulfide (VMS) deposits accumulated on thinned, frac-tured lithosphere within extensional oceanic suprasubduction environments, or back arcs, which were intrudedby anomalously hot subvolcanic sills; hence, their abundance in the Superior province of Canada (thick conti-nental lithosphere), contrasting with few in the Yilgarn craton of Australia (thick lithosphere). Orogenic golddeposits formed in sutures between accreted terranes associated with assembly of Kenorland. Diamonds werecreated by reaction of carbonate-rich asthenospheric liquids with continental lithospheric mantle at >240-kmdepth, mostly pre-2.7 Ga. They were entrained in kimberlitic to lamproitic melts related to superplume eventsat 480, 280, and ~100 Ma. Preservation of resulting mineral provinces stems from their location on stableArchean continental lithospheric mantle.Decreased plume activity after 2.6 Ga caused sea level to fall, leading to the first extensive passive-marginsequences, including deposition of phosphorites, iron formations, and hydrocarbons, during dispersal of Kenorland from 2.4 to 2.2 Ga. Deposits of Cr-Ni-Cu-PGE were generated where plumes impinged on failedrifts at the transition from thick Archean to thinner Proterozoic continental lithospheric mantle, e.g., the GreatDyke, Zimbabwe, and later at Norilsk, Russia. Paleoproterozoic orogenic belts, for example, the Trans-Hudsonorogen in North America and the Barramundi orogen in Australia, welded together the new continent of Co-lumbia. Foreland basins associated with these orogens, containing reductants (graphitic schists) in the base-ment, led to the formation of unconformity U deposits, with multiple stages of mineralization generated fromdiagenetic brines for as much as 600 m.y. after sedimentation. Plume dispersal of Columbia at 1.6 to 1.4 Ga ledto SEDEX Pb-Zn deposits in intracontinental rifts of North America and Australia, extensive belts of RapakiviA-type granites on all continents, with associated Sn veins, and Fe oxide-Cu-Au-REE deposits. All were con-trolled by rifts at the transition from thick to thin continental lithospheric mantle. Plume impingement on Ro-dinia at ~1 Ga formed extensive belts of anorogenic anorthosites and Rapakivi granites in Laurentia andBaltica, the former hosting Fe-Ti-V deposits. Sedimentary rock-hosted Cu deposits formed in intracontinentalbasins from plume dispersal of Rodinia at ~800 Ma.Iron formations and mantle plumes have common time series: Algoman type occur from 3.8 Ga to 40 Ma,granular iron formations precipitated on the passive margins of Kenorland at ~2.4 Ga, Superior-type formedon the passive margins of Laurentia, and Rapitan iron formations were created in rifts during latter stages of dispersal of Rodinia at ~700 Ma. Accordingly, such deposits are not proxies for the activity of atmospheric O 2 .Rich Tertiary placer deposits of Ti-Zr-Hf, located on the passive margins of Australia and Southern Africa, re-flect multiple cannibalistic cycles from orogens that welded Rodinia and Pangea.Orogenic Au deposits formed during Cordilleran-type orogens characterized by clockwise pressure-temper-ature-time paths from ~2.7 Ga to the Tertiary; Au-As-W and Hg-Sb deposits reflect the same ore fluids at pro-gressively shallower levels of terrane sutures. The MVT-type Pb-Zn deposits formed in foreland basins, with † Corresponding author: e-mail, ©2005 Society of Economic Geologists, Inc. Economic Geology 100th Anniversary Volume pp. 1097–1136  Historical Perspective and Scope Lindgren (1933) pioneered the concepts of both metallo-genic provinces and epochs. In the Economic Geology Fifti-eth Anniversary Volume , Turneaure (1955) synthesized globalmetallogenic provinces. He emphasized different classes of ore deposits, stable versus orogenic settings, lithologic ormagmatic associations of specific metal groupings, and therole of young mountain belts in preservation potential. Met-allogenic provinces of different ages were recognized, albeit with large age uncertainties. Primary depositional setting ver-sus replacement was, and remains, an issue. Independently,Bilibin (1968) and Smirnov (1976) documented specific litho-tectonic and age associations for various classes of metallicdeposits in the former Soviet Union. Other comparative stud-ies of major ore provinces recognized the evolving crust-man-tle system as a control on lithological associations, magmaticstyle, and types of ore deposits (Pereira and Dixon, 1965;Stanton, 1972; Hutchinson, 1981). Atlases of the distributionof metallic deposits by geologic terrane and age were com-piled by Dixon (1979) and Derry (1980).Meyer (1981) generated a global database of representativeor type metallic mineral deposits, and their age-lithotectonicassociation, in the Economic Geology Seventy-Fifth Anniver- sary Volume . He formulated the space-time distribution of metallogenic provinces in terms of two parameters: intervalsof geologic history during which specific classes of metallicdeposits formed, and changes of characteristics within a givenclass over the interval when that class formed. Meyer ob-served that trends of crustal evolution were not contempora-neous globally but did not cast his reviews in a plate tectoniccontext (Meyer, 1981, 1988).The theory of plate tectonics was established in the 1970s,supplanting the geosynclinal concept of lithotectonic associa-tions (Kay, 1951; see Sengor, 1990, for a review). Elements of the theory included: recognition of ocean-floor spreadingfrom ages of volcanic islands and transform faults (Wilson,1965; Hess, 1968) and magnetic domains (Vine andMatthews, 1963), relative to mid-ocean ridges; exponentialdecrease of heat flow orthogonal to spreading centers (Sclaterand Francheteau, 1970); and earthquake distribution at con- vergent margins (Benioff, 1964). Historical accounts of theevolution from a static to dynamic worldview are given by Uyeda (1978) and Allegré (1988).Initial hypotheses of the relationship between differentclasses of ore deposits and their plate tectonic settings wereset out by Rona (1980), Mitchell and Garson (1981), andSawkins (1984). These accounted for the distribution of someore deposit types in the Phanerozoic. However, there werelimitations: (1) at the time, genetic hypotheses for many typesof ore deposit were predicated on syngenesis; (2) where con-sensus existed on a syngenetic versus epigenetic srcin, theage of mineralization was not well constrained; (3) epochs, orsecular cycles, of metallogenic provinces were not accountedfor; and (4) extrapolation to the Precambrian met with uncer-tainties as to tectonic processes during that era. Windley (1995) compiled a concise list of metallic and nonmetallic re-sources for each era, documenting their geodynamic and ge-ologic settings. It is now generally accepted that plate tectonics operatedfrom ~3.4 Ga, albeit in some early form that likely differsfrom today, with intermittently more intense plume activity to1.9 Ga (Fyfe, 1978; Isley and Abbott, 1999). Archean craton-scale faults are commensurate with lithospheric plate interac-tions (Sleep, 1992). In addition, Cenozoic-type convergentmargin arc associations, including the presence of boninites,Mg andesites, and adakites, in Precambrian supracrustal ter-ranes require that arc-trench migration occurred (Polat et al.,2003). An alternative precept of Archean geodynamics isgiven by Hamilton (1998).Advances in geochronology have resolved many of the un-certainties in the timing of both metal deposits and metallo-genic provinces. This constraint permits evaluation of func-tional relationships between lithotectonic associations,magmatism, pressure-temperature-time (P-T-t) conditionsand fluid compositions, and geodynamic setting, concurrently resolving the syngenetic issue (e.g., Kerrich and Cassidy,1994). Based on Meyer’s (1981, 1988) compilations of thespace-time distribution of metallogenic provinces, Barley andGroves (1992) provided insights into the episodic develop-ment of distinct classes of metallic deposits as a function of thesupercontinent cycle. Geologic processes are intrinsically sto-chastic, so there is progressive uncertainty in reconstructingthe supercontinent cycle back through the Precambrian. Yet,this framework confers an elegant account for metallogenicprovinces and their episodicity from 2.7 Ga to the present. During the last 25 years there have been profound gains inknowledge as to how plate tectonics operates through time,stemming from the heuristic approach of geology as a field andanalytical science. In addition to development of the concept of the supercontinent cycle, knowledge has advanced on many fronts relevant to metal deposits, including: (1) how evolutionof lithospheric mantle controls crustal evolution (Jordan, 1988);(2) recognition of superfamilies of orogenic belts (Sengor andNatal’in, 1996); (3) the role of mantle plumes and their inter-action with lithospheric plates (Condie, 2001; Wyman and Ker-rich, 2002); (4) transitions in both plume and convergent mar-gin magmatism near the Archean-Proterozoic transition(Taylor and McLennan, 1995; Isley and Abbott, 1999); (5) de- velopment of, and processes in, convergent margins (see re- view by Richards, 2003); (6) characterization of geothermal sys-tems on land (Elder, 1981) and submarine counterparts, someof which are actively depositing sulfide minerals, such as in theLau back-arc basin (Ishibashi and Urabe, 1995; Mills and El-derfield, 1995); (7) quantification of global geochemical cycles(Jacobson et al., 2000); (8) seismic tomography (van der Hilstet al., 1998); (9) precise geochronology (Dalrymple, 1991); and(10) the fractal, or scale-invariant, nature of many geologic1098 KERRICH ET AL. 0361-0128/98/000/000-00 $6.00 1098 Phanerozoic Pb-Zn SEDEX ores localized in rifted passive continental margins containing evaporites at low latitudes. Porphyry Cu and epithermal Au-Ag deposits occur in both intraoceanic and continental margin arcs;ore fluids were related to slab dehydration, peridotite fusion, and hybridization with upper-plate crust. De-posits exposed today are largely <200 m.y.-old, given their low preservation potential in topographically ele- vated ranges.  processes, including of metallogenic provinces (Turcotte, 1992; Weinberg et al., 2004).Accordingly, in this overview, we reframe the space-timedistribution of ore deposits in terms of four interrelatedprocesses: (1) lithotectonic associations that develop in agiven geodynamic setting, (2) classes of metallogenicprovinces that develop in those associations, (3) secular varia-tions of geodynamic environments in the supercontinent-cycle framework, and (4) secular change of continental lithos-pheric mantle that influences all of the above.Evolution of near-surface conditions has also been viewedas a control on the distribution of some ore deposits throughtime, specifically those having elements with redox-sensitivesolubility, such as Fe and U. Two polarized schools of thoughtemerged and have persisted. Cloud (1972) proposed a low pO 2 in the Archean, with a transition to oxygenation of Earth’satmosphere-hydrosphere in the Proterozoic, whereas Dim-roth and Kimberly (1976) advocated Archean atmosphericpO 2 close to the present atmospheric level (PAL). More re-cently, some workers have promoted the early low pO 2 modelbased on mass-independent S isotope fractionation of atmos-pheric S gases, and a rise of atmospheric oxygen at 2.4 to 2.2Ga as the redox state of volcanic gases shifted (Farquar et al.,2000; Holland, 2002). In contrast, Ohmoto maintained thatpO 2  was within 50 percent of present atmospheric level by 4Ga, based on Fe mobility in Archean paleosols and on depo-sitional mechanisms for iron formation that are akin to thosepresently occurring in the Red and Black Seas (Ohmoto,1997, 2004a,b). Resolution of this issue is not readily tractable, as many lines of evidence may reflect local condi-tions, and it is difficult to demonstrate preservation of pri-mary signatures (e.g., Clout and Simonson, 2005). It is clearfrom molecular microfossils that the earliest photosynthesisin the Paleoarchean was anoxygenic, using bacteriochloro-phyls, whereas oxygenic photosynthesis by photosystem II, in- volving cyanobacteria, was established by the Mesoarchean(Nisbitt, 2002). This review does not further consider theissue.No modern text on ore deposits addresses recent advancesin geodynamics. Accordingly, we present a brief synthesis of geodynamic concepts as a framework for discussing mineraldeposits. The divisions between geodynamic settings usedhere reflect the preference of the authors. For example, weexplicitly recognize that there is a continuum between domi-nant plume-lithosphere interaction, where magmatic Ni-Cu-PGE deposits form; through belts of anorogenic magmatismthat host Fe-Ti-V deposits, in which plume magmas do notadvect to shallow crustal levels; and to continental rifting withsubdued plume activity, which is the setting for Fe oxide-Cu-Au-REE and sediment-hosted Cu-Co deposits. For eachmain geodynamic setting, we have selected the best charac-terized metallogenic provinces for discussion of the role of geodynamics in the formation of a class, or classes, of mineraldeposit, without necessarily including all deposit subtypes. Geodynamics Introduction Plate tectonics is a kinematic theory according to which thelithosphere, the upper layer of the Earth including crust andlithospheric mantle, is divided into a finite number of plates.The plates are torsionally, but not flexurally, rigid. Plates in-teract at divergent, convergent, and transform-fault bound-aries (Fig. 1A), as they migrate across the surface of the Earth(Isacks et al., 1968; Cox and Hart, 1986; Sengor, 1990). Platemotions are the surface reflection of the fundamental processby which heat is removed from the interior of the Earth.The oceanic and continental lithospheric plates, alsotermed the mechanical boundary layer (Fig. 2A,B), constitutethe translationally mobile upper boundary layer of the three-dimensional convection cells in the asthenospheric mantle.The core-mantle boundary (referred to as D , 2,900 km deep)is the lower boundary layer of the mantle convection cells.The boundary between upper and lower mantle (D', 670 km)is defined seismologically and reflects a mineralogical phasetransition. The upper and lower mantle probably convects in-dependently, albeit with episodic overturn, based on geo-chemical, heat flow, and seismic evidence (Stein and Hof-mann, 1994; van der Hilst et al., 1998; Butler and Peltier,2002). Heat is removed from the core and mantle to the sur-face by this convection and by plumes that rise from the core-mantle boundary, advecting through the convecting lowerand upper mantle to the surface (Davies, 1999). Heat passesfrom the convecting asthenospheric mantle through the tor-sionally rigid lithospheric plates either by conduction or by advection of magmas. Thermal boundary layers form at thetransition from convecting to convecting or convecting toconducting domains; they are present at the D core-mantleboundary, at the D' upper-lower mantle transition, and be-tween the base of the lithosphere and top of the convectingupper mantle, which is also the low-velocity zone (Fig. 2).Subducting oceanic lithospheric plates penetrate the D'upper-lower mantle boundary at 670 km, as imaged by seis-mic tomography, and probably are stored in lithosphericgraveyards at the core-mantle boundary (D ), where they aresporadically reactivated as mantle plumes. Similarly, ananomalously hot mantle plume, extending into the lowermantle, has been imaged beneath the Iceland ocean plateau(Bijward and Spakman, 1999; Kárason and van der Hilst,2000). Accordingly, there is mass as well as heat exchange be-tween the upper and lower mantles. Oceanic and continental lithosphere Schematic diagrams depicting the tectonic setting of oredeposits generally stop at the base of the deposit or the crust,the petrological seismic Mohorovic discontinuity (Moho).However, the larger context in which mineral concentrations,i.e., deposits, form should more comprehensively be consid-ered in a lithosphere-asthenosphere framework that reflectsgeodynamic settings. These in turn control the conjunction of structures, magma reservoirs, fluid reservoirs, basins, andtheir interactions (Fig. 2). Modern oceanic lithosphere has a ~6-km-thick basalticcrust and ~30- to 50-km-thick lherzolitic mantle lithospherenear ridges. The mantle lithosphere thickens to a maximumof ~70 to 100 km as it progressively cools with increasing dis-tance from the oceanic spreading axis, by accretion of under-lying asthenosphere (Fig. 2A; Keary and Vine, 1996). Com-pared to the underlying asthenosphere, oceanic lithosphereaway from ridges is relatively cool, mechanically rigid, and METALLOGENIC PROVINCES IN AN EVOLVING GEODYNAMIC FRAMEWORK  1099 0361-0128/98/000/000-00 $6.00 1099  1100 KERRICH ET AL. 0361-0128/98/000/000-00 $6.00 1100 F IG .1. A. Map of continental and oceanic lithospheric plates. Triangles signify polarity of subduction, trenches migratein the opposite direction as slabs sink approximately vertically. Length of arrows proportional to plate velocity. Red symbols= Cordilleran superfamily of orogenic belts; green symbols = continent-continent superfamily of orogenic belts. Modifiedfrom Condie (1997). B. Distribution of Archean cratons and Proterozoic and Phanerozoic terranes. After Kusky and Polat(1999). C. Thickness of continental lithospheric mantle from Artemieva and Mooney (2001).
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks