Eocene melting of subducting continental crust and early uplifting of central Tibet: Evidence from central-western Qiangtang high-K calc-alkaline andesites, dacites and rhyolites

Changes in oceanic O–Sr isotopic compositions and global cooling beginning in the Eocene are considered to have been caused by the uplift of the Tibetan Plateau. The specific timing and uplift mechanism, however, have long been subjects of debate. We investigated the Duogecuoren lavas of the central-western Qiangtang Block, which form the largest outcrops among Cenozoic lavas in northern-central Tibet and have widely been considered as shoshonitic. Our study demonstrates, however, that most of these lavas are high-K calc-alkaline andesites, dacites and rhyolites. Moreover, they are characterized by high Sr (367–2472 ppm) and Al₂O₃ (14.55–16.86 wt.%) and low Y (3.05–16.9 ppm) and Yb (0.31–1.48 ppm) contents and high La/Yb (27–100) and Sr/Y (48–240) ratios, similar to adakitic rocks derived by partial melting of an eclogitic source. They can be further classified as either peraluminous and metaluminous subtypes. The peraluminous rocks have relatively high SiO₂ (> 66 wt.%) contents, and low MgO (< 1.0 wt.%), Cr (4.94–23.3 ppm) and Ni (2.33–17.0 ppm) contents and Mg^# (20–50) values, while the metaluminous rocks exhibit relatively low SiO₂ (55–69 wt.%) contents, and high MgO (1.41–6.34), Cr (25.7–383 ppm), Ni (14.13–183 ppm) and Mg^# (46–69) values, similar to magnesian andesites. ⁴⁰Ar/³⁹Ar and SHRIMP zircon U–Pb dating reveal that both peraluminous and metaluminous adakitic rocks erupted in the Eocene (46–38 Ma). Paleocene–Early Miocene thrust faults and associated syn-contractional basin deposits in the Qiangtang Block suggest that this region was undergoing crustal shortening within a continent during the Eocene. The low ε_Nd (− 2.81 to − 6.91) and high ⁸⁷Sr/⁸⁶Sr (0.7057–0.7097), Th (11.2–32.3 ppm) and Th/La (0.23–0.88) values in the Duogecuoren adakitic rocks further indicate that they were not derived by partial melting of subducted oceanic crust. Taking into account tectonic and geophysical data and the compositions of xenoliths in Cenozoic lava in northern-central Tibet, we suggest that the peraluminous adakitic rocks were most probably derived by partial melting of subducted sediment-dominated continent of the Songpan-Ganzi Block along the Jinsha suture to the north at a relatively shallow position (the hornblende + garnet stability field), but the metaluminous adakitic rocks likely originated from the interaction between peraluminous adakitic melts generated at greater depths (the garnet + rutile stability field) and mantle. Because the Duogecuoren adakitic rocks must have originated from a garnet-bearing (namely, eclogite facies) source, Eocene continental subduction along the Jinsha suture caused the thickening of the Qiangtang crust. Given that crustal thickening generally equates with elevation, the uplift of the Central Tibetan Plateau probably began as early as 45–38 Ma, which provides important evidence for tectonically driven models of oceanic O–Sr isotope evolution during global cooling and Asian continental aridification beginning in the Eocene.     

High-precision 40Ar/39Ar sanidine geochronology of ignimbrites in the Mogollon-Datil volcanic field,southwestern New Mexico

⁴⁰Ar/³⁹Ar age spectra have been obtained from 85 sanidine separates from 36 ignimbrites and one rhyolitic lava in the latest Eocene-Oligocene Mogollon-Datil volcanic field of southwestern New Mexico. Of the 97 measured age spectra, 94 yield weighted-mean plateau ages each giving single-spectrum 1 precision of±0.25%–0.4% (±0.07–0.14 Ma). Replicate plateau age determinations for eight different samples show within-sample 1 precisions averaging ±0.25%. Plateau ages from multiple (n=3–8) samples of individual ignimbrites show 1 within-unit precision of ±0.1%–0.4% (±0.04–0.13 Ma). This within-unit precision represents a several-fold improvement over published K-Ar data for the same ignimbrites, and is similar to the range of precisions reported from single-crystal laser fusion studies. A further indication of the high precision of unit-mean ⁴⁰Ar/³⁰Ar ages is their close agreement with independently established stratigraphic order. Two samples failed to meet plateau criteria, apparently due to geologic contamination by older feldspars. Effects of minor contamination are shown by six other samples, which yielded slightly anomalous plateau ages. ⁴⁰Ar/³⁹Ar plateau ages permit resolution of units differing in age by 0.5% (0.15 Ma) or less. This high resolution, combined with paleomagnetic studies, has helped to correlate ignimbrites among isolated ranges and has allowed development of an integrated timestratigraphic framework for the volcanic field. Mogollon-Datil ignimbrites range in age from 36.2 to 24.3 Ma. Ignimbrite activity was strongly episodic, being confined to four brief (<2.6 m.y.) eruptive episodes separated by 1–3 m.y. gaps. Ignimbrite activity generally tended to migrate from the southeast toward the north and west.     

Geochemistry and tectonic setting of Tertiary volcanism in the Güvem area, Anatolia, Turkey

Detailed field mapping in the Güvem area in the Galatia province of NW Central Anatolia, Turkey, combined with K–Ar dating, has established the existence of two discrete Miocene volcanic phases, separated by a major unconformity. The magmas were erupted in a post-collisional tectonic setting and it is possible that the younger phase could be geodynamically linked to the onset of transtensional tectonics along the North Anatolian Fault zone. The Early Miocene phase (18–20 Ma; Burdigalian) is the most voluminous, comprising of over 1500 m of potassium-rich intermediate-acid magmas. In contrast, the Late Miocene volcanic phase (ca. 10 Ma; Tortonian) comprises a single 70-m-thick flow unit of alkali basalt. The major and trace element and Sr–Nd isotope compositions of the volcanics suggest that the Late Miocene basalts and the parental mafic magmas to the Early Miocene series were derived from different mantle sources. Despite showing some similarities to high-K calc-alkaline magma series from active continental margins, the Early Miocene volcanics are clearly alkaline with higher abundances of high field strength elements (Zr, Nb, Ti, Y). Crustal contamination appears to have enhanced the effects of crystal fractionation in the petrogensis of this series and some of the most silica-rich magmas may be crustal melts. The mantle source of the most primitive mafic magmas is considered to have been an asthenospheric mantle wedge modified by crustally-derived fluids rising from a Late Cretaceous–Early Tertiary Tethyan subduction zone dipping northwards beneath the Galatia province. The Late Miocene basalts, whilst still alkaline, have a Sr–Nd isotope composition indicating partial melting of a more depleted mantle source component, which most likely represents the average composition of the asthenosphere beneath the region.     

The Taylor Creek Rhyolite of New Mexico: a rapidly emplaced field of lava domes and flows

The Tertiary Taylor Creek Rhyolite of southwest New Mexico comprises at least 20 lava domes and flows. Each of the lavas was erupted from its own vent, and the vents are distributed throughout a 20 km by 50 km area. The volume of the rhyolite and genetically associated pyroclastic deposits is at least 100 km³ (denserock equivalent). The rhyolite contains 15%–35% quartz, sanidine, plagioclase, ±biotite, ±hornblende phenocrysts. Quartz and sanidine account for about 98% of the phenocrysts and are present in roughly equal amounts. With rare exceptions, the groundmass consists of intergrowths of fine-grained silica and alkali feldspar. Whole-rock major-element composition varies little, and the rhyolite is metaluminous to weakly peraluminous; mean SiO₂ content is about 77.5±0.3%. Similarly, major-element compositions of the two feldsparphenocryst species also are nearly constant. However, whole-rock concentrations of some trace-elements vary as much as several hundred percent. Initial radiometric age determinations, all K–Ar and fission track, suggest that the rhyolite lava field grew during a period of at least 2 m.y. Subsequent ⁴⁰Ar/³⁹Ar ages indicate that the period of growth was no more than 100 000 years. The time-space-composition relations thus suggest that the Taylor Creek Rhyolite was erupted from a single magma reservoir whose average width was at least 30 km, comparable in size to several penecontemporaneous nearby calderas. However, this rhyolite apparently is not related to a caldera structure. Possibly, the Taylor Creek Phyolite magma body never became sufficiently volatile rich to produce a large-volume pyroclastic eruption and associated caldera collapse, but instead leaked repeatedly to feed many relatively small domes and flows.The new ⁴⁰Ar/³⁹Ar ages do not resolve preexisting unknown relative-age relations among the domes and flows of the lava field. Nonetheless, the indicated geologically brief period during which Taylor Creek Rhyolite magma was erupted imposes useful constraints for future evaluation of possible models for petrogenesis and the origin of trace-element characteristics of the system.     

Tectonic implications of the age, composition, and orientation of lamprophyre dikes, Navajo volcanic field, Arizona

Fifty-one lamprophyre dikes of the Navajo volcanic field are randomly oriented in contrast to dikes of similar age from adjacent tectonic provinces. KAr ages obtained on nine of these dikes indicate that they were emplaced between 28 and 19 m.y. ago. Initiation of dike emplacement was approximately coeval with the initial demarcation of the Colorado Plateau in the mid-Tertiary. Chemical analyses obtained on nine dike samples show a wide range in composition. Silica contents range from 33.6% in a phlogopite katungite to 55.9% in a felsic minette. Two of the samples have Na₂O > K₂O indicating that sodic volcanism was occurring concurrently with the dominant potassic volcanism.Published results of high-pressure experimental studies of katungite suggest that this magma was derived from K-metasomatised mantle. Because this potassium metasomatism produces a positive volume change, we propose a model that relates uplift of the Colorado Plateau to metasomatism, magma genesis, and dike emplacement.     

40Ar/39Ar analyses of phlogopite nodules and phlogopite-bearing peridotites in South African kimberlites

⁴⁰Ar/³⁹Ar analyses have been made on phlogopite-bearing peridotite nodules from Bultfontein and phlogopite nodules from Du Toitspan, Kimberley area, South Africa. Neither definite plateau nor isochron age could be obtained due to the occurrence of an excess⁴⁰Ar in phlogopite. However, the extrusion age of a phlogopite nodule from Du Toitspan has been estimated to be about 86 m.y. from the combination of the youngest⁴⁰Ar/³⁹Ar age in the intermediate temperature fraction with Rb/Sr age data reported for this area.Excess⁴⁰Ar correlates with K-derived³⁹Ar in some phlogopites suggesting that it is trapped in K- or K-similar sites and has been incorporated during phlogopite formation.The occurrence of large amounts of excess⁴⁰Ar in phlogopite suggests that it was not formed at a shallow depth.     

Paleomagnetism, geochronology, and magnetic mineralogy of Devonian dikes from the Kola alkaline province (NE Fennoscandian Shield)

The new paleomagnetic data on forty dikes and two intrusive plutons of Devonian age located in different parts of the Kola Peninsula, which have not been previously covered by systematic paleomagnetic studies, are reported. We describe the results of the rock magnetic, petrographic, and microprobe investigations of the Devonian dikes and present their isotopic ages (⁴⁰Ar/³⁹Ar, stepwise heating). Within the studied area, almost all the Devonian dikes, metamorphic Archaean-Proterozoic complexes of the Fennoscandian Shield, and Proterozoic dikes have undergone low-temperature hydrothermal-metasomatic alteration, which resulted in the formation of new magnetic minerals with a secondary (chemical) component of magnetization. The comparison of the paleomagnetic poles indicates the Early Jurassic age of the secondary component. We suggest that regional remagnetization event was caused by endogenic activity genetically related to the formation of the Barents Sea trap province 200–170 Ma ago. On the basis of the obtained data, the preliminary Devonian paleomagnetic pole of the East European Platform is determined.     

Ar/Ar ages of tephras intercalated in astronomically tuned Neogene sedimentary sequences in the eastern Mediterranean

We present new ⁴⁰Ar/³⁹Ar data for sanidine and biotite derived from volcanic ash layers that are intercalated in Pliocene and late Miocene astronomically dated sequences in the Mediterranean with the aim to solve existing inconsistencies in the intercalibration between the two independent absolute dating methods. ⁴⁰Ar/³⁹Ar sanidine ages are systematically younger by 0.7-2.3% than the astronomical ages for the same ash layers. The significance of the discrepancy disappears except for the upper Ptolemais ashes, which reveal the largest difference, if an improved full error propagation method is applied to calculate the absolute error in the ⁴⁰Ar/³⁹Ar ages. The total variance is dominated by that of the activity of the decay of ⁴⁰K to ⁴⁰Ar (∼70%) and that the amount of radiogenic ⁴⁰Ar_p in the primary standard GA1550 biotite (∼15%). If the ⁴⁰Ar/³⁹Ar ages are calculated relative to an astronomically dated standard, the influence of these parameters is greatly reduced, resulting in a more reliable age and in a significant reduction of the error in ⁴⁰Ar/³⁹Ar dating.Astronomically calibrated ages for Taylor Creek Rhyolite (TCR) and Fish Canyon Tuff (FCT) sanidine are 28.53±0.02 and 28.21±0.04 Ma (±1 S.E.), respectively, if we start from the more reliable results of the Cretan A1 ash layer. The most likely explanation for the large discrepancy found for the younger Ptolemais ash layers (equivalent to FCT of 28.61 Ma) is an error in the tuning of this part of the sequence.     

Evolution of the Bayramiç magmatic complex, northwestern Anatolia

During the Oligocene–Middle Miocene period widespread magmatic activity developed in Western Anatolia, following the continental collision between the Sakarya continent and the Tauride–Anatolide platform. This produced both intrusive and extrusive rocks, which appear to be associated in space and time, as exemplified from the Bayramiç area. In the Bayramiç area, the magmatic activity started with the intrusion of the Evciler granite, and the coeval lower volcanic association. This was followed by the development of the upper volcanic association. These rock groups form collectively the Bayramiç magmatic complex, which was generated under an on-going north–south compressional regime. The Bayramiç magmatic complex has a subalkaline composition, displaying a calcalkaline trend. Trace elements and REE contents resemble to island-arc and collision-related magmas. According to the isotope values the Bayramiç magmatic complex was derived from the magmas of lithospheric mantle origin, which were later contaminated, while passing through the thick continental crust, in a post-collisional tectonic setting, during the Oligocene–Early Miocene period. The latest product of the magmatism is the Late Miocene–Pliocene basalt lavas. Their geochemical properties are clearly different from the Oligocene–Early Miocene magmatic rocks. The basalts were generated when the north–south compression gave way to the north–south extensional regime.     

K-Ar geochronology of the late cenozoic volcanic rocks of the Cordillera Occidental, southernmost Peru

Twenty-four K-Ar radiometric ages are presented for late Cenozoic continental volcanic rocks of the Cordillera Occidental of southernmost Perú (lat. 16° 57′–17° 36′S). Rhyodacitic ignimbrite eruptions began in this transect during the Late Oligocene and continued episodically through the Miocene. The development of andesitic-dacitic strato volcanoes was initiated in the Pliocene and continues to the present.The earliest ignimbrite flows (25.3–22.7 Ma) are intercalated in the upper, coarsely-elastic member of the Moquegua Formation and demonstrate that this sedimentary unit accumulated in a trough, parallel to Andean tectonic trends, largely in the Oligocene. More voluminous ash-flow eruptions prevailed in the Early Miocene (22.8–17.6 Ma) and formed the extensively preserved Huaylillas Formation. This episode was coeval with a major phase of Andean uplift, and the pyroclastics overlie an erosional surface of regional extent incised into a Paleogene volcano-plutonic arc terrain. An age span of 14.2–8.9 Ma (mid-Late Miocene) is indicated for the younger Chuntacala Formation, which again comprises felsic ignimbrite flows, largely restricted to valleys incised into the pre-Huaylillas Formation lithologies, and, at lower altitudes, an extensive aggradational elastic facies. The youngest areally extensive ignimbrites, constituting the Sencca Formation, were extruded during the Late Miocene.In the earliest Pliocene, the ignimbrites were succeeded by more voluminous calcalkaline, intermediate flows which generated numerous large and small stratovolcanoes; these range in age from 5.3 to 1.6 Ma. Present-day, or Holocene, volcanism is restricted to several large stratovolcanoes which had begun their development during the Pleistocene (by 0.7 Ma).The late Oligocene/Early Miocene (ca. 22–23 Ma) reactivation of the volcanic arc coincided with a comparable increase in magmatic activity throughout much of the Cordilleras Occidental and Oriental of the Central Andes.     

Geochemical variation with time in the Cenezoic volcanic rocks of southwest Hokkaido, Japan

Abundances of major and trace elements were determined for the Tertiary volcanic rocks from SW Hokkaido. The Late Miocene to Pliocene volcanic rocks of this region show geochemical features similar to those of the Quaternary rocks, that is, K/Si, Th/Si and LREE/HREE ratios increasing across the arc, east to west, from the Pacific to the Japan Sea side. In contrast, the Early Miocene volcanic rocks, which are geographically restricted to the Japan Sea coast, are distinct from all later volcanics and show “within-plate” characteristics — in particular, high concentrations of HFS elements. The Quaternary basalts have low Hf/Yb ratios and Hf contents, whereas the Early Miocene basalts are high in Hf/Yb and Hf, similar to Hawaiian alkali basalts. The compositional variation with time may result from the progressive depletion of incompatible HFS elements in the mantle source. Th/Yb ratios increase from Early Miocene to Quaternary, possibly reflecting increase in the LIL element contribution to the mantle source during that time.     

The significance of scarp retreat for cenozoic landform evolution on the Colorado Plateau,U.S.A.

Since the beginning of the Cenozoic period several hundreds of metres of the sedimentary cover have been removed from the Colorado Plateau. Palaeoclimatic considerations show that the Colorado Plateau has been dominated by dry climates throughout the Cenozoic with the possible exception of the early Palaeocene. Today in the still prevailing arid climate, which strongly accentuates differences in rock resistance, the relief shows a structurally controlled cuesta scarp topography in the slightly deformed strata of alternating resistance. In examining whether the denudational efficiency of scarp retreat was sufficient to account for the wide erosional gaps in the sedimentary cover, rates of scarp retreat were determined by using the information of dated volcanic material and by applying a new method, which calculates the amount of retreat from the width of beheaded valleys of known age. Rates of retreat range from 0·5 to 6·7 km my^?1. The results show that the rates of retreat are controlled by the thickness and resistance of the caprocks. A model of Cenozoic scarp retreat demonstrates that the rates of recession calculated for the scarps in Upper Cretaceous rocks were sufficient to bring them into their present positions from the centre of the Monument Uplift on the central Colorado Plateau. The late Eocene positions of the cliffs in the Early Tertiary formations give an indication of their maximal extent. After the Upper Cretaceous sediments had been removed from the uplifts, erosion cut through successively older rocks, and activated scarps in stratigraphically lower positions. Scarp retreat can operate simultaneously and independently at different levels, which enabled this erosional mechanism to remove great proportions of the sedimentary cover of the Colorado Plateau during the Cenozoic.     

Subvolcanic structure of the central dike swarm associated with the ring complexes in the Shitara district,central Japan

The Subvolcanic structure of the central dike swarm associated with the Miocene Otoge ring complex and the Shitara igneous complex, central Japan, has been reconstructed. The central dike swarm was supplied from several aligned magma reservoirs. Flow lineations observed at the margin of the dikes converge towards a region that is regarded as a magma reservoir about 1–2 km below present sea level. The minimum diameter of the magma reservoir corresponds to the width of the central dike swarm, estimated to be about 3–4 km. The inferred magma reservoir of the Otoge ring complex, may have a zoned structure, as suggested by the flow lineations of dikes and the arrangement of cone sheets. Felsic magma occupied the upper part, about 1–2 km below present sea level, and basic magma the lower part, deeper than 2 km. The centre of the Shitara igneous complex is interpreted to be composed of several other shallow magma reservoirs. The distribution pattern in plan view of the central dike swarm is summarized from the frequency of dikes (defined by the number of dikes per kilometre in the direction normal to the trend of the dike swarm) and the variations of the different properties of individual dikes along the dike swarm. It has a plane of symmetry normal to the dike swarm above the magma reservoir. The patterns critical to a general understanding of the dike formation are:

Triassic warm subduction in northeast Turkey: Evidence from the Ağvanis metamorphic rocks

Within the Tethyan realm, data for the subduction history of the Permo–Triassic Tethys in the form of accretionary complexes are scarce, coming mainly from northwest Turkey and Tibet. Herein we present field geological, petrological and geochronological data on a Triassic accretionary complex, the A?vanis metamorphic rocks, from northeast Turkey. The A?vanis metamorphic rocks form a SSE–NNW trending lozenge‐shaped horst, ～20 km long and ～6 km across, bounded by the strands of the active North Anatolian Fault close to the collision zone between the Eastern Pontides and the Menderes–Taurus Block. The rocks consist mainly of greenschist‐ to epidote‐amphibolite‐facies metabasite, phyllite, marble and minor metachert and serpentinite, interpreted as a metamorphic accretionary complex based on the oceanic rock types and ocean island basaltic, mid‐ocean ridge basaltic and island‐arc tholeiitic affinities of the metabasites. This rock assemblage was intruded by stocks and dikes of Early Eocene quartz diorite, leucogranodiorite and dacite porphyry. Metamorphic conditions are estimated to be 470–540°C and ～0.60–0.90 GPa. Stepwise ⁴⁰Ar/³⁹Ar dating of phengite–muscovite separates sampled outside the contact metamorphic aureoles yielded steadily increasing age spectra with the highest incremental stage corresponding to age values ranging from ～180 to 209 Ma, suggesting that the metamorphism occurred at ≥ 209 Ma. Thus, the A?vanis metamorphic rocks represent the vestiges of the Late Triassic or slightly older subduction in northeast Turkey. Estimated P–T conditions indicate higher temperatures than those predicted by steady state thermal models for average subduction zones, and can best be accounted for by a hot subduction zone, similar to the present‐day Cascadia. Contact metamorphic mineral assemblages around an Early Eocene quartz diorite stock, on the other hand, suggest that the present‐day erosion level was at depths of ～14 km during the Early Eocene, indicative of reburial of the metamorphic rocks. Partial disturbance of white‐mica Ar–Ar age spectra was probably caused by the reburial coupled with heat input by igneous activity, which is probably related to thrusting due to the continental collision between Eastern Pontides and the Menderes–Taurus Block.     

The Llullaillaco volcano, northwest Argentina: construction by Pleistocene volcanism and destruction by sector collapse

Llullaillaco is one of a chain of Quaternary stratovolcanoes that defines the present Andean Central Volcanic Zone (CVZ), and marks the border between Chile and Argentina/Bolivia. The current edifice is constructed from a series of thick dacitic lava flows, forming the second tallest active volcano in the world (6739 m). K–Ar and new biotite laser ⁴⁰Ar/³⁹Ar step-heating dates indicate that the volcano was constructed during the Pleistocene (≤1.5 Ma), with a youngest date of 0.048±0.012 Ma being recorded for a fresh dacite flow that descends the southern flank. Additional ⁴⁰Ar/³⁹Ar measurements for andesitic and dacitic lava flows from the surrounding volcanic terrain yield dates of between 11.94±0.13 Ma and 5.48±0.07 Ma, corresponding to an extended period of Miocene volcanism which defines much of the landscape in this region. Major- and trace-element compositions of lavas from Llullaillaco are typical of Miocene–Pleistocene volcanic rocks from the western margin of the CVZ, and are related to relatively shallow-dipping subduction of the Nazca plate beneath northern Chile and Argentina.Oversteepening of the edifice by stacking of thick, viscous, dacitic lava flows resulted in collapse of its southeastern flank to form a large volcanic debris avalanche. Biotite ⁴⁰Ar/³⁹Ar dating of lava blocks from the avalanche deposit indicate that collapse occurred at or after 0.15 Ma, and may have been triggered by extrusion of a dacitic flow similar to the one dated at 0.048±0.012 Ma. The avalanche deposits are exceptionally well preserved due to the arid climate, and prominent levées, longitudinal ridges, and megablocks up to 20-m diameter are observed.The avalanche descended 2.8 km vertically, and bifurcated around an older volcano, Cerro Rosado, before debouching onto the salt flats of Salina de Llullaillaco. The north and south limbs of the avalanche traveled 25 and 23 km, respectively, and together cover an area of approximately 165 km². Estimates of deposit volume are hampered by a lack of thickness information except at the edges, but it is likely to be between 1 and 2 km³. Equivalent coefficients of friction of 0.11 and 0.12, and excess travel distances of 20.5 and 18.5 km, are calculated for the north and south limbs, respectively. The avalanche ascended 400 m where it broke against the western flank of Cerro Rosado, and a minimum flow velocity of 90 m s⁻¹ can be calculated at this point; lower velocities of 45 m s⁻¹ are calculated where distal toes ascend 200 m slopes.It is suggested that the remaining precipitous edifice has a high probability for further avalanche collapse in the event of renewed volcanism.     

2.8-Ma ash-flow caldera at Chegem River in the northern Caucasus Mountains (Russia), contemporaneous granites, and associated ore deposits

Diverse latest Pliocene volcanic and plutonic rocks in the north-central Caucasus Mountains of southern Russia are newly interpreted as components of a large caldera system that erupted a compositionally zoned rhyolite-dacite ash-flow sheet at 2.83 ± 0.02 Ma (sanidine and biotite ⁴⁰Ar/³⁹Ar). Despite its location within a cratonic collision zone, the Chegem system is structurally and petrologically similar to typical calderas of continental-margin volcanic arcs. Erosional remnants of the outflow Chegem Tuff sheet extend at least 50 km north from the source caldera in the upper Chegem River. These outflow remnants were previously interpreted by others as erupted from several local vents, but petrologic similarities indicate a common origin and correlation with thick intracaldera Chegem Tuff. The 11 × 15 km caldera and associated intrusions are superbly exposed over a vertical range of 2,300 m in deep canyons above treeline (elev. to 3,800 m). Densely welded intracaldera Chegem Tuff, previously described by others as a rhyolite lava plateau, forms a single cooling unit, is > 2 km thick, and contains large slide blocks from the caldera walls. Caldera subsidence was accommodated along several concentric ring fractures. No prevolcanic floor is exposed within the central core of the caldera. The caldera-filling tuff is overlain by andesitic lavas and cut by a 2.84 ± 0.03-Ma porphyritic granodiorite intrusion that has a cooling age analytically indistinguishable from that of the tuffs. The Eldjurta Granite, a pluton exposed low in the next large canyon (Baksan River) 10 km to the northwest of the caldera, yields variable K-feldspar and biotite ages (2.8 to 1.0 Ma) through a 5-km vertical range in surface and drill-hole samples. These variable dates appear to record a prolonged complex cooling history within upper parts of another caldera-related pluton. Major W-Mo ore deposits at the Tirniauz mine are hosted in skarns and hornfels along the roof of the Eldjurta Granite, and associated aplitic phases have textural features of Climax-type molybdenite porphyries in the western USA. Similar ⁴⁰Ar/³⁹Ar ages, mineral chemistry, and bulk-rock compositions indicate that the Chegem Tuff, intracaldera intrusion, and Eldjurta Granite are all parts of a large magmatic system that broadly resembles the middle Tertiary Questa caldera system and associated Mo deposits in northern New Mexico, USA. Because of their young age and superb three-dimensional exposures, rocks of the Chegem-Tirniauz region offer exceptional opportunities for detailed study of caldera structures, compositional gradients in volcanic rocks relative to cogenetic granites, and the thermal and fluid-flow history of a large young upper-crustal magmatic system.     

利用地震光照成像法研究青藏高原东南缘岩石圈间断面结构及其动力学意义

为了进一步认识青藏高原东南缘的构造演化等动力学问题以及该区域的深部孕震机理,本文使用位于该区域内的中国地震科学台阵探测项目的台站所记录到的远震P波波形数据,采用地震光照成像法获取了岩石圈间断面的结构,并讨论了该方法的准确性和稳定性。研究结果显示,青藏高原东南缘的岩石圈西薄东厚,其中:滇缅泰地块腾冲火山附近最薄,约为60 km,其较薄的岩石圈可能是软流圈地幔物质上涌造成的;扬子地块岩石圈厚度从四川盆地向南逐渐减薄,特别是四川盆地下方最厚,可达190 km左右;滇缅泰地块腾冲火山下方150 km深度左右探测到明显的间断面,该间断面可能是腾冲火山原始岩浆源的位置即岩浆源。本研究所得结果 “印支地块与滇缅泰地块结构的连续性” 进一步为印度板块的推挤作用造成腾冲火山低速物质向东溢出的结论提供了地震学证据。此外,研究区域最北端的剖面显示,峨眉山大火成岩省的内带在50—250 km深度范围及其上方地壳内存在明显的局部高速异常,其不均匀分布特征可能与二叠纪火山喷发过程中岩浆底侵及中新生代以来多期次构造活动有关。      

Development of continental lithospheric mantle as reflected in the chemistry of the Mesozoic Appalachian Tholeiites, U.S.A.

Geochemical analyses of dikes, sills, and volcanic rocks of the Mesozoic Appalachian Tholeiite (MAT) Province of the easternmost United States provide evidence that continental tholeiites are derived from continental lithospheric mantle sources that are genetically and geochronologically related to the overlying continental crust. Nineteen olivine tholeiites and sixteen quartz tholeiites from the length of this province, associated in space and time with the last opening of the Atlantic, display significant isotopic heterogeneity: initial ε_Nd = +3.8 to −5.7; initial ⁸⁷Sr/⁸⁶Sr= 0.7044−0.7072; ²⁰⁶Pb/²⁰⁴Pb= 17.49−19.14; ²⁰⁷Pb/²⁰⁴Pb= 15.55−15.65; ²⁰⁸Pb/²⁰⁴Pb= 37.24−39.11. In PbPb space, the MAT define a linear array displaced above the field for MORB and thus resemble oceanic basalts with DUPAL Pb isotopic traits. A regression of this array yields a secondary PbPb isochron age of ≈ 1000 Ma (μ₁ = 8.26), similar to Sm/Nd isochrons from the southern half of the province and to the radiometric age of the Grenville crust underlying easternmost North America. The MAT exhibit significant trace element ratio heterogeneity (e.g., Sm/Nd= 0.226−0.327) and have trace element traits similar to convergent margin magmas [e.g., depletions of Nb and Ti relative to the rare earth elements on normalized trace element incompatibility diagrams, Ba/Nb ratios (19–75) that are significantly greater than those of MORB, and low TiO₂ (0.39–0.69%)].Geochemical and geological considerations very strongly suggest that the MAT were not significantly contaminated during ascent through the continental crust. Further, isotope and trace element variations are not consistent with the involvement of contemporaneous MORB or OIB components. Rather, the materials that control the MAT incompatible element chemistry were derived from subcontinental lithospheric mantle. Thus: (1) the MAT/arc magma trace element similarities; (2) the PbPb and Sm/Nd isochron ages; and (3) the need for a method of introducing an ancient (> 2−3 Ga) Pb component into subcontinental mantle that cannot be much older than 1 Ga leads to a model whereby the MAT were generated by the melting of sediment-contaminated arc mantle that was incorporated into the continental lithosphere during arc activity preceding the Grenville Orogeny (≈ 1000 Ma).     

Ar/Ar dating of the Bandelier Tuff and San Diego Canyon ignimbrites, Jemez Mountains, New Mexico: Temporal constraints on magmatic evolution

The Jemez Mountains volcanic field (JMVF), located in north-central New Mexico, has been a site of basaltic to rhyolitic volcanism since the mid-Miocene with major caldera forming eruptions occurring in the Pleistocene. Eruption of the upper Bandelier Tuff (UBT) is associated with collapse of the Valles Caldera, whereas eruption of the lower Bandelier Tuff (LBT) resulted in formation of the Toledo Caldera. These events were previously dated by K-Ar at 1.12 ± 0.03 Ma and 1.45 ± 0.06 Ma, respectively. Pre-Bandelier explosive eruptions produced the San Diego Canyon (SDC) ignimbrites. SDC ignimbrite “B” has been dated at 2.84 ± 0.07 Ma, whereas SDC ignimbrite “A”, which underlies “B”, has been dated at 3.64 ± 1.64 Ma. Both of these dates are based on single K-Ar analyses.⁴⁰Ar/³⁹Ar dating of single sanidine crystals from these units indicates revision of the previously reported dates. Isochron analysis of 26 crystals from the UBT gives a common trapped ⁴⁰Ar/³⁶Ar component of 304.5, indicating the presence of excess ⁴⁰Ar in this unit, and defines an age of 1.14 ± 0.02 Ma. Isochron analysis of 26 crystals from the LBT indicates an atmospheric trapped component and an age of 1.51 ± 0.03 Ma. An age of 1.78 ± 0.04 Ma, based on the weighted mean of 5 individual analyses, is indicated for SDC ignimbrite “B”, whereas 3 analyses from SDC ignimbrite “A” give a weighted mean age of 1.78 ± 0.07 Ma. Evidence for xenocrystic contamination in the SDC ignimbrites comes from analyses of a correlative air-fall pumice unit in the Puye Formation alluvial fan giving ages of 1.75 ± 0.08 and 3.50 ± 0.09 Ma. The presence of xenocrysts in bulk separates used for the original K-Ar analyses could account for the significantly older ages reported.Geochemical data indicate that SDC ignimbrites are early eruptions from the magma chamber which evolved to produce the LBT, as compositions of SDC ignimbrite “B” are virtually identical to least evolved LBT samples. Differentiation during the 270-ka interval between eruption of SDC ignimbrite “B” and the LBT produced an array of high-silica rhyolite compositions which were erupted to form the LBT. Mixed pumices associated with eruption of the LBT indicated an influx of more mafic magma into the system which produced shifts in some incompatible trace-element ratios. Lavas and tephras of the Cerro Toledo Rhyolite record the geochemical evolution of the Bandelier magma system during the 370-ka interval between eruption of the LBT and the UBT.The combined geochronologic and geochemical data place the establishment and evolution of the Bandelier silicic magma system within a precise temporal framework, beginning with eruption of the SDC ignimbrites at 1.78 Ma, and define a periodicity of 270–370 ka to ash-flow eruptions in the JMVF. These intervals are comparable to those in other multicyclic caldera complexes and are a measure of the timescales over which substantial fractionation of large silicic magma bodies occur.     

Geochronological evidence for pervasive Miocene weathering,Minas Gerais,Brazil

⁴⁰Ar/³⁹Ar laser incremental‐heating analyses of 22 individual grains of supergene cryptomelane from three weathering pro?les, up to 400 km apart, in the Rio Doce valley and Barbacena regions at Minas Gerais, Brazil, show that the formation of weathering pro?les in these regions is contemporaneous, suggesting a strong weathering event in the Middle to Late Miocene (10–8 Ma). The preservation of these Miocene samples at or near the present surface suggests that either erosion rates have been very low in the region since the Miocene or that a much thicker weathering mantle was present in the region originally. Assuming a constant thickness of weathering pro?les in the region throughout the Tertiary, we may calculate weathering front propagation rates of 4–8 m Myr^?1 during the past 10 Ma. Copyright © 2004 John Wiley & Sons, Ltd.