V53B-01
Plio-Quaternary calc-alkaline and adakitic post-subduction volcanism in the Gulf of California, Mexico: the effects of slab tearing and continental breakoff
The Gulf of California province has recorded a complex magmatic history after the subduction of Farallon plate stopped during Late Miocene. Besides the MORB-type tholeiitic volcanism associated with the spreading segments in the southern Gulf and alkaline volcanism in the northernmost Gulf province, calc-alkaline and adakitic volcanism is also encountered in the central Gulf region, between 26.30 and 29°N. It occurs in Isla San Esteban which is located at the southeastern margin of the Pliocene extinct spreading centre of Lower Tiburon Basin. Isla San Esteban is entirely made of andesitic, dacitic and rhyolitic subaerial lava flows, domes, and pyroclastic deposits which K-Ar ages range from 4.5 to 2.5 Ma. Two magmatic series are distinguished: a medium-K calc-alkaline series which evolved from basaltic andesites to rhyolites by fractionation of plagioclase, pyroxenes, titanomagnetite and apatite; and an adakitic series including adakites of andesitic composition and dacites derived from them through separation of plagioclase and amphibole. The subduction imprint of the magmatism in Isla San Esteban, together with that of other Pliocene and Quaternary medium-K calc-alkaline volcanic centers located along the Main Gulf Escarpment, suggests the occurrence of subduction-modified mantle and possibly of slivers of oceanic crust beneath the Peninsula and the Gulf of California. Their occurrence may be linked to the development of a Late Miocene asthenospheric window south of 29°N. We propose that this slab window derived from the progressive tearing of the slab southwards from the edge of the earlier Californian asthenospheric window, between the remnant of the Magdalena plate, completely attached to the Pacific plate since 7 Ma, and the still downgoing Farallon plate. The high thermal regime linked to the opening of the Gulf of California combined with the deep asthenospheric window may have triggered the partial melting of remnants of oceanic crust as well as that of metasomatized mantle which represent the sources of adakitic and calc-alkaline magmas, respectively.
V53B-02
Variable Slab Contributions Characterize the Miocene Magmatic Record of the Central Trans- Mexican Volcanic Belt
Two distinct magmatic events, related to different petrogenetic processes, characterize the early activity of the central Trans-Mexican Volcanic Belt. These episodes are represented by the stratovolcanoes of the Palo Huerfano-La Joya-Zamorano Volcanic Complex (PH-LJ-Z; 12-10 Ma), located at 500 km from the Mesoamerican Trench, and by the Queretaro Volcanic Succession (QVS; 9-6 Ma), a mafic plateau that stratigraphically overlies the stratovolcanoes. PH-LJ-Z products are calc-alkaline andesites and dacites, while QVS rocks range from basalts to basaltic andesites. Despite their different major element contents, the two rock groups exhibit the same Mg# (40-70). Both sequences display arc-like trace element patterns; however, the stratovolcanoes present a stronger subduction signature than the QVS, and they also show compositional analogies with slab melts, such as high Sr/Y ratios, HREE depletions, and a tendency to MORB-like isotopic compositions. The two suites cannot be genetically related by simple fractional crystallization; moreover, the geochemical features of the PH-LJ-Z sequence are not compatible with lower crustal melting. Therefore, even though both rock groups have experienced contamination at different crustal levels during ascent, we attribute their primary compositional differences to distinct mechanisms of element recycling in the subduction zone: the geochemical features of PH- LJ-Z high-Mg# andesites and dacites are consistent with the interaction of silicate slab melts with mantle peridotites, while the chemical features of QVS products suggest partial melting of a fluid-fluxed mantle wedge. The modification of the subduction component from slab melts to slab fluids might be related to a variation in the thermal structure of the subduction zone during a period of prolonged sub-horizontal subduction. Indeed, the slab-melt features of PH-LJ-Z rocks, and their emplacement at a large distance from the trench, are consistent with a flat slab scenario that promotes slab melting at relatively low pressures. A protracted flat subduction would produce a gradually cooler thermal state that will hinder slab melting and instead favor slab dehydration.
V53B-03
Rhyolite genesis by slab melting and mantle interaction: an example from the Trans-Mexican Volcanic Belt
A remarkable set of rhyolitic rocks emplaced in the Trans-Mexican Volcanic Belt indicate that slab-derived siliceous melts can ascend through the mantle and crust while suffering only minor compositional modifications. These rocks were discovered in the isolated rhyolitic domes of Chalcatzingo, located at 40 km SW from Popocatepetl Volcano, in Central Mexico. New 40Ar/39Ar data on plagioclase and amphibole yielded an age of 20.7 Ma, thus revealing some of the oldest tracks of magmatic activity in the Mexican arc. The rhyolitic domes are covered by small andesitic lava flows of unknown age, that are chemically similar to the Quaternary Chichinautzin volcanic field. The rhyolitic rocks are petrographically homogeneous with a porphyritic texture made of plagioclase and amphibole set in a groundmass of plagioclase and quartz. Interestingly, the rhyolitic rocks host a large variety of crustal xenoliths: amphibolites, granulites, gabbros and gneisses. Younger andesites are also porphyritic with phenocrysts of plagioclase, two pyroxenes and minor olivine set in a microcrystalline groundmass of the same mineralogy, but they did not carry crustal xenoliths. The Chalcatzingo rhyolites (SiO2=71%) have relatively high MgO (1.2%), Mg# (60), CaO (3%), Na2O (6%) and Ni (24 ppm), but unusually low K2O (0.7%) contents. Incompatible trace element patterns are typical of arc magmas, but concentrations in the rhyolites are lower than in the andesites, and much lower than those of more typical Mexican rhyolites. In addition, Sr, Nd and Pb isotopes of the rhyolites are the most depleted compositions ever measured in the Mexican arc, completely overlapping with those of the Pacific MORB, with values that are different from the crustal xenoliths. These characteristics are inconsistent with a crystal fractionation process of a more primitive magma, and cannot be explained by direct anatexis of the continental crust. Instead, the overall geochemical features are what would be expected in melts from subducted oceanic basalt that acquired its high Mg and Ni contents through interactions with mantle peridotites, in a process that did not modify the incompatible trace element budget.
V53B-04
Petrogenesis of Cenozoic Alkaline Basalts From the Tigris Volcanic Field, NE Syria
The Tigris volcanic field of NE Syria contains large discontinuous exposures of basanitic and basaltic lava flows of Quaternary age, covering an area of about 2,700 km2, and is located near the Bitlis collision suture which forms the boundary between the Arabian and Eurasian plates. The lavas are made up of olivine (Fo78-89), clinopyroxene (salite and titan-augite), plagioclase (An57-68), and opaque phases. Chemical data show that the rocks exhibit SiO2 values ranging from 42.2 to 48.2 wt%, MgO (5.7-9.1 wt%) and are relatively enriched in TiO2 (1.7-3.2 wt%). Mg-numbers range from 0.51 to 0.62, with an average of 0.58. The rocks are enriched in incompatible trace elements such as Zr (119-231 ppm), Nb (14-43 ppm) and Y (17-22 ppm). The REE patterns are strongly fractionated [(La/Yb)N = 11], indicative of a garnet-bearing source. The 143Nd / 144Nd isotopic compositions range from 0.512803 to 0.512908, and 87Sr / 86Sr from 0.70327 to 0.70403. Such compositions are typical of those of HIMU-OIB, with a HIMU-like mantle source. Modeling indicates that the magma was produced by a small degree of partial melting (F = 1 to 2%) of a primitive, garnet-lherzolite mantle source. Elemental ratios such as K/P (4.6), La/Ta (12), La/Nb (0.90), Nb/Y (1.22), and Th/Nb (0.09), suggest that the magma was subjected to minimal crustal contamination. Shear heating at the base of the lithospheric mantle of the northern boundary of the Arabian plate, caused by a change in plate motion as the Arabian plate moved in a more easterly direction during the Plio-Quaternary may have provided the heat necessary for partial fusion and magma generation. The Neogene collision of the Arabian plate with Eurasia along the Bitlis suture resulted in reactivation (beneath the Tigris-Euphrates basin) of deep-seated fractures, along which lavas may have penetrated the crust.
V53B-05
Periodical and Spatial Compositional Variations for Flood-basalts of the Siberian Platform
Flood-basalts of the Siberian Platform (SP) have a typical spatial trend: the content of MgO decreases from north to south, while TiO2 and K2O contents increase. 'Explosivity' coefficient increases in the same direction (from ~ 11-14% to more then 50%). Volcanism of the northwestern and northern SP parts has two stages; the rest has only one stage. These stages of magma evolution can be distinguished based on lava content fluctuation, when geochemical melts changed from contaminated (at the beginning) to non-contaminated (in the end). The most complete effusive profile for the Norilsk region has the following quantitative rock content: picritoids - slightly more then 10%, trachybasalts - less then 10%, low-potassium tholeiites - more then 65%, tholeiites of high K2O content - less then 15%. Spectral amplitude for content variations attenuates by the end of time series and functions look like 'white noise'. The 'total' linear trend for the following components is negative: SiO2, Al2O3, MgO, Na2O, K2O, P2O5; TiO2, FeO* have a positive trend. Flood-basalt profile for the Putorana Plateau (Lama Lake) is typically monotonous; most petrogenic components are characterized by 'white noise' functions; only TiO2 and MgO have little correlation coefficients (0.29) for the first logs. Distribution functions are low-coherent for practically all the components. The local linear trend of distribution functions is close to ones of the Kharaelakhskaya syncline of the Norilsk region. Frequency spectra of 26 elements of the borehole profile SG-19 for the Kharaelakhskaya syncline (ARIMA method) for 4 main peaks show that both petrogenic and trace elements have both similar and greatly varying maximum content frequencies. It is important to note, that trace elements have maximums both for high-frequency and low frequency parts of time series spectra, while petrogenic components do not have these peaks. The 'extractants' of non-volatile components (boron and fluorine) have spectral frequency characteristics close to both petrogenic and ore elements. Most petrologists suggest that the above-mentioned lava profiles of both contaminated and non-contaminated melting of primary magmas resulted from two different mantle sources. Periodicity of local distribution functions of lava compositions shows that those different-level magma sources existed simultaneously. However, the composition of the lithospheric mantle substrates for different SP parts varied greatly (well-known data of lava geochemistry for the Norilsk, Maymecha-Kotuy and Putorana provinces). We constructed a numerical model for two-level melting zone above hotspots. Compositions for basic melts and the depth for the upper melt level in the lithosphere mantle above an 'asthenolens' depends on the thickness of the earth crust, the thinness of the lithosphere and the degree of its metasomatic change. This model allows us to explain evolutionary trends and variations for lava compositions of various petrochemical SP provinces, - the upper melting boundary can vary from about 70 to 120 km from the earth's surface (depending on the initial conditions); virtual dimensions of the melting zone are close to the lava shield size. However, there is no periodical eruption model for this magma system so far. This work was supported by the Russian Foundation for Basic Research (Grants: #04-05-64107; #040-05- 64322) and by the Russian Ministry for Science and Education (Grant #DSP.2.1.1.702).
V53B-06
Implications for rapid pluton emplacement via multiple discrete pulses, Lamarck Granodiorite, central Sierra Nevada batholith, California
New high-precision zircon geochronology and petrographic mapping indicate that emplacement of the Lamarck Granodiorite of the John Muir Intrusive Suite, Sierra Nevada batholith, California was punctuated by two major intervals of high magma flux ~2 m.y. apart. Existing geochronology (Coleman et al., 1995, Cont. Min. Pet.) indicates that a significant volume of the Lamarck intruded at ~92 Ma. New, nearly indistinguishable zircon ages from four mappable (at a scale of 1:10,000) pulses within the Dusy Basin area define an earlier episode of rapid intrusion around 93.9 Ma. Preservation of internal contacts at both the map and outcrop scales requires the earlier flux of magmatism included multiple discrete pulses. Map-scale (250 m to 1 km wide) intrusions within Dusy Basin are distinguished by variations in composition or foliation; whereas outcrop scale (<1m to decameter wide) intrusions are recognized on the basis of subtle differences in modal abundance (e.g. color index), enclave abundance and length/width ratio, grain size and morphology, and/or foliation. Although high-precision zircon ages (±200 ka) cannot distinguish age variation among members mapped within the early flux, relative ages between phases are apparent from field relations. Within Dusy Basin the earliest phase of the Lamarck to intrude was a leucocratic granodiorite, followed by an intermediate granodiorite, and finally a mafic quartz diorite. The Inconsolable Granodiorite, historically unassigned to an intrusive suite, flanks the eastern margin of the Lamarck. A new zircon age of 95.4 Ma (±300. ka) along with textural and structural similarities suggest the Inconsolable represents the oldest and most mafic member of the John Muir Intrusive Suite. As a whole, the Muir is temporally, aerially, texturally, and compositionally comparable to the Tuolumne Intrusive Suite to the north. Both suites intruded over an ~8 m.y. time span (95-87 Ma) starting with the oldest mafic granodiorite units and ending with assembly of K-spar megacrystic units. However, the Muir intruded as dike-like plutons along a northward progression, whereas the Tuolumne forms a nested suite. This difference may reflect variation in intrusive level of exposed suites.
V53B-07
Mixing Between Basaltic and Trachytic Magmas to Generate the Trachyte-Phonolite Suite at Suswa Volcano, Kenya Rift: Evidence from Mineral and Glass Compositions
Suswa volcano, which formed from 0.24 Ma to recent, is the southernmost central vent volcano in the Kenya Dome area. This area includes 5 trachytic volcano complexes, two of which have pantellerites and comendites as their most recent eruptions. Suswa is distinct in that pre- and syn-caldera rocks include dominantly trachytic flows and ignimbrites, whereas post-caldera flows are trachyte to phonolite. The trachyte and phonolite suites have subparallel vertical trends on a TAS diagram with the phonolites at approximately 4 weight percent lower SiO2. Basaltic (B) to basaltic trachyandesite (BTA) flows are found in the low ground between the central vent volcanoes. Based on phenocryst compositions and melt inclusion and groundmass glass analyses, we have found evidence that Suswa developed by magma mixing between B/BTA and trachytic end-members. The evidence is as follows: 1) syncaldera ignimbrite contains both plagioclase (An45Ab52Or3) and alkali feldspar (An2Ab62Or36) in the same samples. The plagioclase compositions are identical to those found in the B/BTA flows and the alkali feldspar resembles that found throughout the Suswa sequence as phenocrysts. 2) Matrix glass in the pre-caldera rocks is trachytic, similar to whole-rock compositions. Syncaldera glass compositions range from trachyandesite to trachyte. The post-caldera phonolites show a wide range in matrix glass compositions from trachyandesite to phonolite. 3) Melt inclusion glass reflects the origin of the crystals: the inclusion glass in the plagioclase from the syncaldera ignimbrite is basaltic, similar to the most mafic whole-rock compositions for the flows from the area, whereas the inclusion glass in both oxides and clinopyroxene is trachytic. In the post-caldera phonolites, the inclusion glass in the alkali feldspar is basaltic trachyandesite to phonolite, whereas in coexisting olivine, clinopyroxene, and oxides, it exhibits higher SiO2 contents and is trachyandesite to trachyte. Based on these data as well as whole-rock chemistry, we propose that injection of a B/BTA magma into the Suswa pre-caldera trachytic chamber triggered syn-caldera eruption and mingling of the two magmas. The least evolved (i.e. least Si undersaturated) post-caldera phonolites represent a more complete hybridization of the two end-members. In order to produce the more evolved phonolites, this mixing process must be followed by crystal fractionation, with alkali feldspar having a dominant role.
V53B-08
Redefining an Igneous System: Volcanic-Plutonic Links Between the Wilson Ridge Pluton and River Mountains Volcanic Section, Nevada USA
The study of an igneous system is often limited by the extent of exposure. Rarely are both the plutonic and volcanic segments of a single igneous system preserved and recognized. Just east of Las Vegas Nevada, both the volcanic (River Mountains) and plutonic (Wilson Ridge) parts of a single igneous system are exposed. The River Mountains volcanic section (12.6 to 13.4 Ma) and the Wilson Ridge Pluton (12.5 to 13.4 Ma) are linked by similarities of geochronology, structure, and geochemistry. The River Mountains volcanic section was separated from the Wilson Ridge pluton by the mid-Tertiary (~13.4 Ma) west-dipping Saddle Island detachment fault and now lies 20 km to the west of the pluton. Normal faulting and stratal rotation have exposed much of the volcano in cross section including a porphyritic quartz monzonite plug in its conduit. The Wilson Ridge Pluton is tilted to the north (~5º to 20º) resulting in nearly 20 km of continuous exposure with its base to the south and roof to the north. The connection between the the pluton and volcanic section is indicated by the tight clustering of relatively immobile trace elements Th (4.2 to 25.1 ppm), Hf (3.3 to 6.6 ppm), and Ta (0.9 to 1.7 ppm) from 88 samples from the pluton and 32 samples from the volcanic section. Furthermore, Nd and Sr isotopes indicate a single cogenetic system. The upper part of the pluton is a dike complex composed of fine-grained monzonite to granite. The largest volume of the pluton is relatively homogeneous quartz monzonite. Near its base, mafic to dioritic enclaves locally compose as much as 75% of the rock. Diorite enclaves are interpreted as xenoliths incorporated into Wilson Ridge magma during ascent. Basaltic enclaves are synplutonic mafic dikes physically disrupted by flow within the more felsic Wilson Ridge magma. The River Mountains volcanic section ranges from basalt to dacite. Mafic enclaves are present locally in dacite and may be the volcanic equivalent of those in the pluton. Chemical similarities between the mafic enclaves in the pluton and the basalt flows of the River Mountains support the link between the Wilson Ridge Pluton and River Mountains volcanic section. Ongoing research aims to model the evolution of the integrated Wilson Ridge - River Mountains igneous system by studying magma residence times from diffusion profiles in phenocrysts, U-Pb dating of Zircons, open system petrogenetic modeling, as well as a chemical and petrographic comparison of the enclaves in the pluton and volcanic section.
V53B-09
Scale of Pluton/Wall Rock Interaction in the Tuolumne Intrusive Suite, Yosemite National Park, California
The western outer unit of the concentrically zoned Tuolumne Intrusive Suite (tonalite of Glen Aulin, Kga) intruded a variety of metasedimentary wall rock units at approximately 93.5 Ma. The May Lake interpluton screen (4500 m x 550 m) is a remnant of the host rocks that includes marble, calc-silicate, quartzite, and pelitic quartzite units that are all in contact with the tonalite. The chemical and physical diversity of the rocks in the screen and their contrast with the invading pluton provide an excellent location to study pluton/wall rock interactions using chemical techniques. Chemical analyses show that contamination of the tonalite is localized within 1-2 meters of wall rock material. Outside the screen, visible wall rock material is predominantly located in an elongate zone of xenoliths (chiefly pelitic quartzite) surrounded by a hybridized fine-grained tonalite that is subparallel to the screen contact and extends from an interbedded quartzite and pelitic quartzite section of the screen into the pluton. Whole-rock geochemistry of tonalite samples near the screen contact and across the hybridized zone suggests localized contamination of various levels. Hybridized tonalite 87Sr/86Sr (i) values range from 0.7061 to 0.7141 and εNd(t) from -3.5 to -10.5, significantly different from typical Kga (87Sr/86Sr (i) = 0.7057- 0.7059 and εNd(t) ~ -3.3 ). Mixing models using the various metasedimentary wall rocks and normal tonalite as end-members show calc-silicate/tonalite and marble/tonalite mixing trends are not consistent with any of the hybridized samples. Contamination of the tonalite with quartz-rich and pelitic metasedimentary rocks occurs in two modes including, 1) bulk assimilation of quartzite and pelitic quartzite (occurring in about half of the contaminated samples), and 2) selective assimilation of partial melts of pelitic quartzite. For selective assimilation, partially melting fertile metasedimentary rocks, such as pelitic quartzite, can produce liquids similar to minimum melt granite and mixing models using experimental minimum melt compositions as the contaminant fit the remaining hybridized samples. We conclude that there are 2 distinct modes of contamination, but there is little evidence for contamination of tonalite beyond the immediate vicinity of wall rock inclusions and therefore assimilation is not a significant space creating mechanism for the Tuolumne Intrusive Suite.
V53B-10
Construction of the Tuross Head Pluton, Coastal NSW, Australia
The Tuross Head pluton, roughly 8 km in diameter, consists dominantly of tonalite with layers, lensoid zones and enclaves of diorite that define a steep basinal structure. Way up structures indicate that these rocks accumulated on an aggrading magma chamber floor, which sagged inward as accumulation occurred. Hence, rocks young inward, and coastal exposures provide stratigraphic sections with clear records of compositional variation through time. Extensive magma mingling and hybridization between tonalite and diorite is apparent from field relations and mixed populations of complexly zoned feldspars. Synmagmatic dikes are dominantly basaltic and dacitic and include composite dikes with chilled mafic pillows in a felsic host. Dikes with intermediate compositions also occur. The chemical and isotopic compositions of the dikes closely match those of the plutonic rocks and suggest they were feeders to the pluton. Isotopic compositions of dikes range widely and correlate broadly with SiO2 (eNd varies from about +8 to +3; Sri varies from 0.7032 to 0.7042). The dikes include both basaltic magmas derived from asthenosphere and dacitic magmas probably derived from Neoproterozoic- Cambrian lower crust. Intermediate dikes appear to represent magmas that mixed prior to emplacement. The Tuross Head pluton formed by multiple emplacements of compositionally varied magmas into one or more magma chambers of unknown size. Chemical and isotopic variations in the plutonic rocks appear to reflect accumulation of early-forming crystals from these contrasted and contemporaneous magmas onto an aggrading chamber floor. Although chemical variation, as viewed on Harker diagrams, might appear consistent with simple fractional crystallization, the stratigraphic occurrences of these samples indicate that tonalite compositions did not become more evolved through time. There is subtle variation upward through the stratigraphic column: the silicic host becomes more mafic, and mafic enclaves become more silicic, until they reach a zone of high strain in the magmatic state, where enclaves are so strongly attenuated that the rocks appear gneissic, and it is impossible not to sample a mixed magma. Above this zone, at the top of the Tuross section, the 'tonalitic' rocks have compositions intermediate between tonalite and diorite end-members.
V53B-11
The Volcanosedimentary and Plutonic Assemblage of the Sierra de Zacatecas, Mexico, an Early Cretaceous Volcanic Field
Based on fossil evidence, the lowermost stratigraphic unit of the Sierra de Zacatecas (SZ), La Pimienta Fm. (LPF), has been considered as Upper Triassic in age (Carnian-Norian). It is covered by the poorly described Las Pilas Diorite (here named Las Pilas volcanosedimentary sequence -LPVS) of unknown age. In this work we present the results of field work, U-Pb dating of zircons and structural analysis. From these results we propose a new interpretation on the stratigraphy, age and origin of the sequence in the SZ. The LPF is composed of schists and phylites derived from fine grained sandstones interbedded with lenses of limestone, and minor tuffs (?) and lava flows. The whole sequence shows a well developed foliation whose average pole is at 86/235.The top of LPF contains a unit of brecciated blocks of phylites and basalt in a fine- grained matrix that includes plagioclase phenocrystals and sulfides, that presumably is part of a fumarolic mound. U-Pb dated zircons from the metasandstones yield an age interval of 161-132 Ma (peak age is 132 Ma). In a gradual stratigraphic continuity, the LPF is covered by the LPVS, made up by cpx-rich basaltic massive and pillowed lava flows, interbedded with mudstone, feldesphatic wacke, greywacke, and tuffs(?). Up section the sediment content and grain size increase. Intruding the LPVS there are Northwestward trending dikes and sills of dioritic composition; also small semicircular bodies of cpx-diorite. U-Pb ages of detrital zircons from a wacke of the upper part of the LPVS yield an age interval of 141-133 Ma. Foliation and stratification are indistinguishable in the LPVS, whose average pole is at 86/244. The apparent gradual stratigraphic order, the U-Pb ages from detrital zircons, and the similar style of deformation of the LPF and LPVS stratigraphic units strongly suggest the presence of an Early Cretaceous volcanic field that is part of an island arc. Our new results deny the Upper Triassic age for LPF and open a new revision of the Mesozoic tectonic history of central Mexico.
V53B-12
Petrologic characterization of A-type granites from southern Sinai, Egypt
Samples from three locations, Sharm El-Sheikh, Wadi Yahmid, and North Nuweiba area, in southern Sinai, Egypt, have been studied to evaluate them as source rocks for potential uranium deposits. The petrographic and electron microprobe study show that the majority of the rocks are medium grain amphibole alkali feldspar granite and amphibole alkali feldspar quartz syenite. Based on the composition, the amphiboles are mainly the sodic- calcic amphiboles: ferro-barroisite and ferro-richterite. Ferrohedenbergite, biotite, and sphene appears in some samples. The alkali feldspars show perthitic structure with about equal amount of end member orthoclase and albite. The accessory minerals, which are usually included in amphiboles, are zircon, apatite, ilmenite, and magnetite. The major REE accessory phase is chevkinite. The chevkinite usually altered and changed to rutile and thorite in the altered rocks. The two oxide geothermometry give around 700 Celsius degree and oxygen fugacity -18. The reduced condition agrees with the mineral assemblages. The fine grained rocks are mainly amphibole alkali feldspar syenite and amphibole alkali feldspar quartz syenite. The amphibole is the calcic ferro- hornblende, clinopyroxene is ferro-augite. The alkali feldspars are orthoclase and albite; the two feldspars show micrographic texture. The albite has up to 1 percent CaO content. The SiO2 contents of the whole-rock chemistry range from 62 to 77 percent. The agpaitic index are larger than 1. The trace elements show an A-type granite characteristic. The rocks have low Ba, Rb, and Sr; moderately high Zr, Y and REE. The REE pattern show slightly LREE enrichment with low to medium Eu anomaly. The large Eu anomaly rocks show strong alteration. At the same time, the breakdown of amphibole and the chevkinite and zircon which it hosts releases uranium while Ti and Th are left behind. Thus, evolution of uranium source rock in granites is demonstrated by amphibole and accessory mineral alteration studies.
V53B-13
Petrological and Structural Setting of Mafic to Intermediate Dykes in the Xolapa Complex: Tierra Colorada-Acapulco Area, Southern Mexico
The south of Mexico is conformed by a mosaic of suspect terranes. Three of them, characterized by different crystalline basements (Guerrero, Mesozoic; Mixteco, Paleozoic; Zapoteco Precambrian), are bordered toward the trench by the Chatino terrane, with the Xolapa Complex representing its basement, made up of amphibolite facies metamorphic rocks, affected by repeated granitic and granodioritic intrusions, with or without deformation. Several mafic dike intrusions are associated to the felsic magmatism. Felsic intrusions in the Tierra Colorada-Acapulco area show a volcanic arc geochemical signature. They range from I-type to S-type granites with different ages since Jurassic time. All intrusives were previously dated at ~165Ma, ~130Ma, ~55-65Ma and ~30-35 Ma. Mafic dykes could be arranged in three groups according to their structural pattern, mineral association and field relationships: (1) Group one dikes intrude basement units like migmatites, and gneisses. They are made up of plagioclase-biotite-hornblende. They are foliated and open to closely folded. Commonly biotite and hornblende define foliation planes, which share the same orientation as the regional foliation domain in orthogneisses and migmatites of the Xolapa Complex (WNW-ESE). (2) Group two dikes is represented by two kinds, the first one is of intermediate composition and affected by 45-50 Ma ductile shearing near Tierra Colorada. It bears plagioclase, k-feldspar and minor quartz, as well as biotite, hornblende, garnet and pyroxene as accessory minerals. NNW- SSE trending, NNE dipping foliation is defined by biotite and hornblende. The second type is not affected by mylonitization and consists of open to close folded and foliated mafic dykes, showing abundant plagioclase and accessory minerals such as biotite, muscovite, hornblende and titanite. Muscovite often is developed as coronitic texture surrounding biotite. Hornblende defines NNW-SSE to WNW-ESE trending, NNE dipping foliation planes. This type of dikes cuts only migmatitic gneisses units. (3) Group three dikes generally consists of intermediate dykes with subvolcanic textures. They cut all units in the central and northern area, and are affected by brittle normal faulting near Tierra Colorada. This group show two mineral and textural association, plagioclase and k- feldspar, with amphibole and rutile phenocrysts, and a microcrystalline matrix of plagioclase and k-feldspar, with some xenocrysts of quartz surrounded by hornblende.
V53B-14
P-T- H2O Phase Relations of an Aleutian High-MgO Basalt: Comparison of Hydrous Experiments with Thermodynamic Models
Okmok Volcano in the Central Aleutians has erupted chemically heterogeneous lavas, ranging from voluminous, fractionated high-alumina basalts (HAB) to rarer, more primitive high-MgO basalts (HMB). These HMBs (9-12 wt.% MgO) represent near-primary magmas that may be unmodified products of mantle melting beneath the arc. Previous anhydrous studies of HMBs have constrained the P-T conditions at which the melts last equilibrated with a dry mantle peridotite assemblage. However studies have shown that the presence of a hydrous fluid within the mantle wedge has an important effect on mantle melting, causing melting to occur at higher pressures and lower temperatures than under anhydrous conditions. Previous anhydrous experimental studies on a natural HMB (ID-16) have shown that it is multiply saturated with five phases (olivine, spinel, clinopyroxene, orthopyroxene, and plagioclase) at 12 kbar and ~ 1300°C. However, it is unlikely that ID-16 equilibrated under anhydrous conditions given geophysical constraints. The goal of our experiments is to infer the P-T- H2O conditions (if any) at which ID-16 is in equilibrium with a hydrated mantle peridotite using hydrous, piston-cylinder experiments. In addition, this data combined with existing anhydrous data can be used to test the validity of thermodynamic modeling techniques (pMELTS). Hydrous experiments have been performed at water-undersaturated conditions (3-10 wt.% H2O) at temperatures ranging from 1050°C-1350°C and pressures ranging from 10-20 kbar. Preliminary results show that the liquidus at 5 wt.% H2O is at ~1300°C, ~1250°C, and ~1160°C at 20, 15, and 10 kbar, respectively. The 5 wt.% H2O liquidus is displaced down- temperature by ~150°C at 20 kbar, ~125°C at 15 kbar, and ~140°C at 10 kbar relative to the anhydrous liquidus. These results are consistent with those found using the thermodynamic modeling program pMELTS. Although we have not yet analyzed experiments at 3 and 10 wt.% H2O, pMELTS modeling as well as a second model developed by Wood (2004) predict a likely point of multiple saturation at ~14 kbar and 1200-1300°C, and ~3-5 wt.% H2O, approximately 2 kbar higher and 100°C lower than the anhydrous multiple saturation point. Data will be presented for experiments at 3 and 10 wt.% H2O as well as mineral phases and compositions.