V43A-1102
Determining the Age and Cooling History of the World's Largest Layered Intrusion: U-Pb Zircon-Rutile Geochronology of the Merensky Reef, Bushveld Complex, South Africa
The Bushveld Complex, South Africa, is the world's largest layered intrusion and host to the majority of the known resources of platinum group elements, chromium and vanadium. Thermal modeling [1] and Ar-Ar [2] studies suggest rapid emplacement and cooling of the intrusion, however precise crystallization and cooling ages have been difficult to obtain. In a previous contribution [3], we provided a chemical abrasion ID-TIMS U-Pb Concordia age for zircon of 2054.3 ± 2.5 Ma (2sd, decay-constant errors included) from a sample of the platiniferous Merensky Reef (West Mine, Rustenburg Section). This was interpreted as the crystallization age of the sample, corresponding to conditions at or above the solidus of the Bushveld magmas (>950°C). The sample also contains low-U (1-4 ppm) rutile, which occurs as acicular grains with biotite and as inclusions or overgrowths on chromite. New U-Pb data from seven concordant multi-grain fractions of rutile yield a Concordia age of 2054.6 ± 4.9 Ma (2sd, decay-constant errors included) and date the timing of closure to Pb diffusion in rutile (600-700°C). Overlapping U-Pb Concordia ages for co-existing zircon and rutile indicate extremely rapid initial cooling of the intrusion (1000°C/Myr). Based on a comparison with existing geochronology, emplacement of the Bushveld Complex and other intrusions from the widespread Bushveld magmatic event across the northern Kaapvaal Craton is restricted to an interval of several million years at ca. 2054 Ma. Estimated magma emplacement rates are 0.1-1 km3/yr, which is characteristic of the high output rates of flood basalts, and indicates that the Bushveld layered intrusion and other products of the Bushveld magmatic event represent the high-level intrusive roots of a major Paleoproterozoic large igneous province. [1] Cawthorn and Walraven (1998) J. Petrol. 39, 1669-1687; [2] Nomade et al. (2004) J. Geol. Soc. London 61, 411-420; [3] Scoates and Friedman (2006) AGU 87(52), V31D-0611.
V43A-1103
Paleomagnetic and AMS Results from the Early Eocene Shonkin Sag and Square Butte Laccoliths, north-central Montana
The Shonkin Sag (SS) and Square Butte (SB) laccoliths, part of the Cenozoic Montana alkaline province in north- central Montana, are shallow level, mafic layered intrusions representing different stages in the formation and growth of laccoliths. Both bodies exhibit thin, finer grained shonkonitic sills that emanate subhorizontally for 100s of meters into the Upper Cretaceous Eagle Sandstone. The SS laccolith interior is about 70 m thick; the SB laccolith is over 480 m thick. Both intrusions are strongly differentiated into a lower shonkonite overlain by syenite. The main shonkonitic part of the SB laccolith exhibits a uniform remanence direction comparable to the time averaged expected early Eocene field direction (e.g., D = 344°, I = 68°, α95= 2.1°, N = 35 samples distributed over 8 sites). In contrast, the lower shonkonitic part of the SS exhibits a remanence direction that differs significantly from the expected field direction (e.g., D = 300°, I = 70°, α95= 10°, N = 22 sites). Interestingly, the upper highly differentiated syenitic part of the SS reveals ChRM directions that differ from the main part of the laccolith (D = 005°, I = 55°, α95= 7°, N = 12 samples from one site and D = 017°, I = 77°, α95= 1.8°, N = 10 samples from a second site). The difference in remanence directions possibly reflects differing magma solidification times and sub- solidus cooling rates for the two rock compositions. Anisotropy of magnetic susceptibility (AMS) as well as anhysteretic remanent magnetization (AARM) data are being obtained from 69 sites sampled throughout the SS and 51 sites sampled along the northern and southeastern parts of the SB. The main body of the SS reveals dominantly subhorizontal and prolate AMS fabrics that vary strongly and abruptly in orientation. Toward the interior of the SB laccolith, the shonkonite exhibits a strong magmatic fabric defined by isolated, subhorizontal syenitic blebs trapped in shonkonite. AMS results from this part of the laccolith show a strong oblate fabric, with a vertical least principal susceptibility axis. Our field observations and analytical results suggest early laccolith development via rapid pulses of low volume magma input during lateral sill growth. Continued growth during laccolith development involved higher volume magma input, and rapid vertical thickening allowing for gravity-controlled differentiation.
V43A-1104
Primary magnetic fabric in an ultrabasic body embedded in high-pressure felsic granulite from the Moldanubian Zone, European Variscides: resistance to exhumation-related medium- pressure granulite-amphibolite facies metamorphism
About 11 x 5 m large body of serpentinized peridotite-clinopyroxenite, enclosed in felsic granulite, occurs in a quarry near the village of Bory (Moldanubian Zone of Western Moravia). The granulite is characterized by the presence of lighter quartz and feldspar-rich bands-layers (up to 10 cm width) alternating with gray stripes consisting of feldspars, quartz, garnet and kyanite. The presence of biotite and sillimanite, that together with quartz ribbons and relic kyanite define the foliation, depends on the degree of retrogression in amphibolite facies conditions. The ultramafic rocks show various degree of serpentinization of olivine and orthopyroxene and of amphibolization of clinopyroxene. The high-temperature textural modification of the clinopyroxenite is characterized by formation of exsolution lamellae of orthopyroxene and garnet in clinopyroxene or recrystallization of large (up 10 mm) clinopyroxene crystals into fine-grained matrix with garnet. The anisotropy of magnetic susceptibility (AMS) was used to investigate the fabrics of both the ultrabasic body and the host granulite. Through investigating low-field variation of the AMS, one could resolve the AMS into the field-independent component due to mafic silicates and pure magnetite in the ultrabasic body and due to mafic silicates in the host granulite and the field-dependent component due to Fe-Ti oxides in the ultrabasic body and due to pyrrhotite in granulite. The field-independent and field-dependent components are coaxial both within ultrabasite and within granulite. On the other hand, they show different orientations in ultrabasite and granulite. In granulite, the magnetic foliation is roughly parallel to the metamorphic schistosity and banding that show relatively stable orientation throughout the entire quarry. The magnetic lineation is rather scattered, but still relatively near the mesoscopic lineation defined by mineral alignment. The AMS ellipsoids are mostly oblate to very oblate, but some of them are also prolate. The degree of AMS is variable, ranging from low to relatively high. In the ultrabasic body, the magnetic foliation is relatively scattered in spatial orientation, but mostly oriented differently than that in granulite. The magnetic lineation is also scattered, but still relatively well defined spatially. Again, its orientation is mostly different than that of granulite. The shapes of the AMS ellipsoid are variable, ranging from very prolate to clearly oblate. The degree of AMS is relatively high, slightly higher than that in granulite. Consequently, the magnetic fabric in the ultrabasic body is different from that in the host granulite even though they experienced at least partially common structural history. The componental movements forming the granulite fabric were evidently not strong enough to strongly overprint the magnetic fabric of ultrabasite. The ultrabasite is therefore strong enough to maintain its pre-metamorphism fabric even at such high temperatures and strong pressure that are characteristic of granulite facies metamorphism.
V43A-1105
Static vs dynamic processes in mafic magma chambers: AMS and image analysis in the Stillwater Complex, Montana
Most mafic intrusions are regarded as magma chambers where various processes have led to the formation of a layering. Some of these processes are static (e.g., crystal settling) while others are dynamic (magma flow due to emplacement and/or convection). The 3-D analysis of the petrofabric allows to constrain the nature of the main layering forming processes, yet only few methods allow to gain this information rapidly. We present a new template study, based on the Stillwater mafic rocks, where magnetic fabric and image analysis techniques have been combined to resolve the 3-D petrofabric. The Stillwater Complex, Montana, is an Archean mafic-ultramafic intrusion that consists of 5 main units: the basal series, ultramafic series, and lower, middle and upper banded series. This study focuses on surface oriented samples (pyroxene-rich gabbronorites, plagioclase rich anorthosites, and olivine containing troctolites) collected near the J-M reef, the main PGE deposit of the complex. 3-D digital image analysis is performed on 3 mutually perpendicular, large oriented thin-sections to define the shape-preferred orientation (SPO) of major minerals. Two of the SPO principal axes (X and Z) correspond respectively to lineation and perpendicular to foliation. The layering is defined macroscopically in the field. Preliminary results show that whereas a well-defined mineral foliation is systematically present in all samples, a mineral lineation is present in only a few samples. The magnetic susceptibility of a rock is an anisotropic property defined by the ratio between induced and inducing magnetization of a sample. It is distinct from the paleomagnetic directions of a sample. In mafic rocks, the anisotropy of magnetic susceptibility (AMS) results from the lattice preferred orientation (LPO) of paramagnetic ferromagnesian silicates, and from the shape preferred orientation (SPO) of ferromagnetic oxides and sulfides. The AMS data is used to calculate the magnetic anisotropy ellipsoid with 3 principal axes where K1≥K2≥K3. Preliminary results show that the mean susceptibility (Km) varies from 13 x 10-6 [SI] to 40000 x 10-6 [SI] with an average of 825 x 10-6 [SI], and suggests that some weak samples are dominated by diamagnetic minerals, and strong samples have a significant ferromagnetic contribution in addition to a paramagnetic contribution. The degree of magnetic anisotropy (Pj) ranges from 1.012 to 1.797 with an average of 1.103. The shape parameter (T) ranges from 0.968 to - 0.714 with an average of 0.450 and suggests that most fabrics are oblate. The combination of the two methods (image analysis and AMS) has allowed to identify 2 fundamental types of magmatic fabrics: (1) oblate fabrics resulting mainly from static magmatic processes in pure-shear (crystal settling, crystal mush compaction); (2) prolate fabrics resulting mainly from dynamic magmatic processes in pure-shear/simple shear (magma flow, crystal mush slumping).
V43A-1106
The Magnetic Stratification of Mafic Magma Chambers: Natural Examples and Numerical Models
Mafic plutons commonly display a prominent petrographic and geochemical layering attributed to differentiation in a magma chamber. This layering occurs at scales from several hundreds of meters in large intrusions down to scales of a few mm at the hand specimen scale. The processes responsible for development of such layering are numerous and include, both internal processes, such as crystal settling, flow seggregation and compaction, and external processes such as magma recharge, magma extraction or magma mixing. An additional distinction can be made between mafic magma chambers that operated as a closed system and those that operated as an open system. Several recent investigations have been conducted on borehole cores through the Great Dyke (Zimbabwe), the Bushveld Complex (South Africa), the Stillwater Complex (Montana), the Sonju Lake intrusion (Minnesota) and the Insizwa sill (South Africa). We present a synthesis of magnetic results on these five intrusions. The magnetic properties of mafic rocks are dominated by the presence of multi-domain magnetite in most layers. Pyrrhotite, the only ferromagnetic sulfide, is restricted to sulfide-rich reefs (Merensky, J-M) or to massive sulfide deposits at the base of the intrusion. Single-domain magnetite (characterized by an inverse magnetic fabric) occurs preferentially in cumulate layers and in the peridotitic basal layers. All intrusions display remarquable variations of their magnetic properties (magnetic susceptibility, degree of anisotropy, shape factor, AMS, magnetic remanence, magnetic saturation, high field magnetic susceptibility) across the layering. The significance of these magnetic zones varies depending on the nature of variations from zone to zone. For example, a drop in magnetic susceptibility does not necessarily correspond to a new magma batch, unless it is also associated with a change in AMS directions. In most cases, the magnetic susceptibility and the total iron content (estimated from high field measurements) increase towards the top of these intrusions suggesting a general iron enrichment. Numerical petrologic models for tholeiitic magma compositions predict absolute magnetic susceptibility variations across a statically-differentiated, closed system intrusion (no convection). The effects of sequential magma recharge and discrete magma extraction are investigated from the magnetic properties point of view. Conversely, these models can be used to interpret the magnetic stratification pattern in natural examples and test for its possible causes.
V43A-1107
Ilmenite-Rich Layers and Sequences in Regressive Intervals in the Bjerkreim Layered Series and Their Relation to Magma Mixing During Episodes of Chamber Replenishment.
The Bjerkreim-Sokndal Layered Intrusion and the closer massif-type anorthosite plutons make up the Egersund- Farsund Igneous Province emplaced in Rogaland south Norway. The layered Intrusion on the base of the rock composition is divided in three main portions: anorthosite and leuconorite, troctolite and leuconorite, mangerite and qtz-mangerite. The main aim of the present work is to characterize the boundary between Megacyclic Units (MCU) II and III along the North-Eastern side of the Intrusion in order to construct a model for mixing in the magma chamber that led to the crystallization of the ilmenite-rich sequences at the base of MCU III. In the Bjerkreim Layered Series ilmenite- rich layers and sequences are rare and restricted to the regressive zones at the bases of MCU III and IV. Cumulates elsewhere in the Layered Series contain limited amounts of ilmenite, as expected during the cotectic crystallization of plagioclase, orthopyroxene and ilmenite. In a restricted area along the eastern flank of the Bjerkreim lobe of the intrusion, ilmenite-rich cumulates form thin, strongly modally-layered sequences near the base of MCU III. Here, they are closely associated with rather massive leuconorite, rare troctolite and a sulphide- bearing layer of melanorite or ilmenite orthopyroxenite. Jensen et al. (2000) inferred that the ilmenite orthopyroxenite or melanorite that marks the base of MCU III was related to the initiation of replenishment and the consequent hybridization of inflowing jotunite (hypersthene monzodiorite) magma and resident magmas. At this time the floor of the chamber was characterised by a deep axial trough and a much shallower eastern flank limited to the north by a pronounced transverse ridge. The inflowing magma was envisaged as being less dense than the lower part of the differentiated and stratified resident magma and orthopyroxene and ilmenite crystallised from an extensive layer of hybrid magma some distance above the temporary floor of the chamber. Dilute suspensions of pyroxene, ilmenite and subordinate plagioclase were postulated to have been carried from the layer of hybrid magma to the floor of the chamber in intermittent plumes. The layer of hybrid magma thickened as replenishment and mixing in the buoyant plume continued and it eventually reached the floor of the chamber. Replenishment culminated in the crystallisation of troctolite (plagioclase-olivine-ilmenite-magnetite cumulate) on the deeper parts of the chamber floor while the crystallisation of lower-temperature norite (plagioclase-hypersthene-ilmenite cumulate) continued on the elevated portions. The results of a more detailed investigation of the stratigraphy, cryptic layering and whole-rock major and trace- element compositions of the cumulates forming the basal portion of MCU III on the north-eastern flank of the Bjerkreim lobe shows a striking correlation between the mineral composition and the stratigraphic distribution of cumulus ilmenite and orthopyroxene. Ilmenite–rich layers and sequences are always characterised by more primitive mineral compositions than the enclosing leuconorite. This is clear evidence that cumulates at the base of MCU III crystallised from two different magmas. Furthermore the shape of the floor of the magma chamber seems to have played a fundamental role in the formation of the ilmenite-rich layers since they are preserved only on the shallow eastern flank of the intrusion. Enrichment in ilmenite in excess of normal cotectic proportions appears to be a consequence of either the sequence of crystallisation in the hybrid magma or mineral sorting in descending plumes.
V43A-1108
Interacting Convective Processes in Kilauea Iki Lava Lake, Hawaii
Kilauea Iki lava lake formed in 1959 as a closed magma chamber of 40 million m3 of picritic magma. Repeated drilling and sampling of the lake allows recognition of processes of magmatic differentiation, and places time restrictions on the periods when they operated. Two processes, double-diffusive convection and finger diapirism, occurred because melt density decreases as olivine crystallization and re-equilibration proceeds, until after plagioclase begins to crystallize. Finger diapirism, described in previous work, occurred from 1961 to 1971 and affected most the lava lake between depths of 13 to 94 m. The period of inferred double- diffusive convection occurred between mid-1962 and 1964 and affected only the most olivine-poor part of the lava lake. Recent re-evaluation of petrographic and chemical data refine our understanding of this second process. The overall variation of bulk MgO content with depth in Kilauea Iki is an S-curve, consistent with gravitative redistribution of the abundant olivine phenocrysts present in the erupted lava. The olivine-poor zone (MgO <11 weight percent) is a sill-like volume found between depths of 21 to 43 m in the lake. This zone is bisected by a median layer containing more and slightly coarser olivine phenocrysts, which has an MgO content 2 weight percent higher than the minimum in the layers above and below. This configuration, not achievable by gravitative settling, suggests that the olivine-poor zone at some point contained a two-layer convective system. The upper and median layers of the olivine-poor zone contain a sparse population of augite microphenocrysts (0.2-0.4 mm in length), often in monomineralic clusters (1-3 mm in length), while the lower layer contains only olivine. Plagioclase and other phases occur only in the groundmass in all samples. If the layers developed before groundmass crystallization began, then the assemblage in the upper layer was olivine + augite, and was olivine-only in the lower. Because melt density decreases as temperature decreases in this part of the crystallization range and because the lava lake was strongly cooled from above, the conditions for double- diffusive convection, with splitting of the melt column into layers, were met. Core samples and temperature data obtained by drilling the lake in mid-1962 and late 1967 constrain the period of double-diffusive convection to the first half of that period. The process ceased without shifting the position of the median olivine-enriched layer downward, suggesting that it was very brief. Finger diapirism, already active in Kilauea Iki, was volumetrically more important, and passed through both layers. This overlapping process may have ended the broader convective process by reducing the thermal gradient that drove it. Although double- diffusive convection was a minor process in Kilauea Iki, it did occur in this closed magma system.
V43A-1109
Modeling of Convection Driven "Weather" Patterns in Bushveld Type Magma Chambers via Computational Fluid Dynamics
Well established and well benchmarked multi-physics finite element codes which are commercially available (e.g., ANSYS, Inc) for modeling (amongst things) fluid flow are proving extremely useful in their application to problems in earth sciences. These codes incorporate the following physical processes: variable viscosity, phase change (e.g., freezing, melting), the impact of latent heat, assimilation, transport of suspended crystal load, porous media flow, chemical evolution, etc. These codes have seen extensive and successful application to fluid environments more complicated than or even similar to magmatic melts and have allowed 3D modeling… with much of their complexity… of the convective cooling and freezing history of magma chambers simulating the Bushveld Complex. Although constrained by much greater viscosities to appropriately diminished velocities, these models yield analogs similar to patterns seen in lakes, oceans and atmospheres: e.g., stratifications, currents, vortices, formation of clouds of crystals, precipitation, etc. The viscosities employed in the Bushveld models are taken to be strongly dependent on temperature (this implicitly includes particulate content and evolving chemistry, etc), increasing over eleven orders of magnitude during cooling. The initial stages of cooling are typified by highly turbulent flow which rapidly settles into more orderly and symmetric forms as the magma cools and "sets" on reaching approximately 65% crystal content. The inclusion of suspended load leads to the build up of stratifications within the model, the lowest in the chamber of primitive composition and the highest (i.e., at the top) of the most evolved composition. This is a natural consequence of the components of highest melting point crystallizing out first, followed by material of lower melting point, etc. Similar to ice crystals gathering along the boundaries of stratified layers in the atmosphere to form thin clouds, primocrysts will be driven by dispersive pressure into the shear zones that define the boundaries between stratifications. The final architecture of the chamber appears to be established quite early and then settles down to simmer and eventually stew in its own juices and any others that may flow in afterwards from surrounding country rock. These models replicate much of the features of the Bushveld Complex. This includes layering. Appeal to well known engineering derived relationships and established research in other disciplines dealing with stratification and layering in fluids supports the numerical inferences.
V43A-1110
Infiltration Metasomatism in Stillwater Anorthosite
The Stillwater Complex is a layered gabbroic intrusion that has been extensively studied because it hosts the bulk of North America’s known reserves of PGE’s (Platinum-Group Elements). Infiltration metasomatism is a process involving the movement of magma and/or fluids into partially molten rock creating a disequilibrium condition. It has been proposed to be the fundamental process responsible for the concentration of PGE’s in the Stillwater Complex and in other layered intrusions. Immediately below the contact between Anorthosite zone II (AN-II) and the overlying troctolites of Olivine-Bearing zone V (OB-V) in the Stillwater Complex is a 15-m-thick section in which the anorthosite differs both texturally and compositionally from that which composes the bulk of AN-II. The rocks in the upper anorthosite section are nearly monomineralic and have plagioclase grains that are 2-3 times smaller than those found in the lower anorthosites, which also contain up to 10% interstitial pyroxene. To further document this discontinuity and to attempt to determine whether it was formed by infiltration metasomatism or as a result of a cumulus replenishment event, samples were collected along a vertical traverse of this transition zone at Picket Pin Mountain and analyzed by electron microprobe and LA-ICP-MS. Analysis of plagioclase shows that the medium-grained anorthosites at the top of AN-II are chemically more similar to the overlying troctolites than they are to the coarse-grained anorthosites below. Plagioclase texture and zoning profiles also show that the zone at the top of AN-II is more favorably comparable with the overlying troctolites than the anorthosites forming the bulk of AN-II. The close chemical and textural resemblance of the upper anorthosites to the troctolites, in conjunction with evidence of compaction and fluid involvement, suggest that the infiltration metasomatism model better elucidates the nature of this discontinuity.
V43A-1111
Extreme differentiation in single samples from the olivine-rich zone of the Palisades Sill, NY and NJ
The Palisades Sill, NY and NJ, has long been the focus of much attention due to its potential contributions to the understanding of igneous differentiation. Many early authors, including N.L. Bowen, thought of the intrusion as an ideal example of crystal settling because of an olivine-rich zone that occurs close to the base of the sill. Later investigations demonstrated that at least three geochemically distinct batches of magma were emplaced at different levels within this body, one of which produced the olivine-rich zone. Here we report the results of our detailed electron probe study of 27 olivine grains, 134 plagioclase grains, and 181 pyroxene grains from four samples of the olivine-rich zone. In one of the rocks plagioclase ranges from An84 (large grain; core) to An0.5 (interstitial plagioclase), whereas olivine ranges from Fo81 - Fo69. These ranges are comparable to our other samples. The plagioclase range is far greater than what is shown by the Skaergaard (An25 - An69) and Bushveld (An30 - An80) intrusions. Though the sill has long been known to be vertically differentiated, such extreme differentiation in individual thumb-sized samples was not expected. The implication is that interstitial melt pockets were so well enclosed (i.e., zero permeability) by the surrounding crystals that they were sealed off from further communication with larger volumes of melt that may have existed at a higher levels in the sill at that time. That is, compaction and associated filter pressing was not efficient in removing the interstitial melt. It is unlikely that such a situation could occur if the olivine-rich zone formed early from a large body of magma because the load pressure of the overlying crystals and melt would be too great to avoid compaction. A better explanation is that the olivine-rich zone was emplaced as a separate magma batch (as suggested by several previous workers based on bulk rock geochemistry) in a mostly solidified magma-crystal mush. In this setting, the interstitial melt would not be able to move through the mostly solidified mush and would have been well encapsulated by the surrounding crystals.
V43A-1112
The origin of halide melt phases in layered intrusions, and their significance to platinum-group element mobility
Fluid and melt inclusions are preserved within pegmatite bodies and cumulus minerals within mafic-ultramafic layered intrusions that host economic concentrations of the platinum-group elements (e.g., Bushveld Complex, South Africa; Stillwater Complex, Montana). The inclusions indicate that the earliest volatile phase to have exsolved from the crystallizing intrusions was a relatively anhydrous carbonic fluid (CO2-dominated). As crystallization proceeded, volatiles became increasingly water-rich and saline, consistent with the relative saturation limits of carbonic and aqueous fluids in mafic silicate liquids, and the partitioning behavior of Cl in fluid-melt systems. Previously unreported, the latest stage volatiles in the layered intrusions were halide melts (slightly hydrous molten salts) of relatively simply composition (NaCl with minor KCl or CaCl2) with salinities in excess of 90 wt% eq. NaCl or CaCl2. These volatiles were trapped at minimum temperatures of 760-800oC, near the eutectic temperature for water-saturated granitic liquid at moderate crustal pressures. Trace element analysis of the salt melt inclusions by laser ablation ICP-MS (ETH Zürich) show that they contain no detectable concentrations of ore and accessory metals. This is in contrast to the earlier, lower salinity volatiles which contain ppm-concentrations of Pt, Pd, As, Bi, Sb as well as abundant S and base metals. Heterogeneous entrapment of late-stage silicate melt and halide melt provides unambiguous evidence for the coexistence of both phases. However, experimental constraints on the nature of exsolved volatiles from mafic or felsic silicate liquids suggest that the halide melt phases cannot represent an exsolved phase from that coexisting silicate liquid, since this would require unrealistically high (initial) Cl:H2O ratios for the parental silicate liquid (> 9 for a granitic residue). Analysis of rhyodacitic silicate melt inclusions that coexist with the halide melt inclusions show that the coeval silicate melts had Cl:H2O ratios of only 0.1 to 0.2. Similarily, the salt melt phases could not have evolved via the removal of H2O by crystallization of hydrous magmatic minerals (e.g., biotite, apatite) since their modal abundances in the intrusions are very low. The most plausible explanation for the halide melt phases involves the "dehydration" of an initially lower- salinity aqueous fluid. This may have occurred by the reaction of the aqueous fluid with nominally-anhydrous minerals such as pyroxene, or by the late-stage alteration of cumulus minerals to hydrous mineral assemblages. Through the use of conventional hydrothermal experimental techniques, it can be shown that the reaction of a volumetrically-minor CaCl2-rich aqueous fluid phase (20 wt% eq. CaCl2) with the assemblage diopside-enstatite-quartz at near-solidus conditions (700oC, 0.4 kbar) results in the formation of tremolite by the reaction of H2O with the initially anhydrous mafic mineral assemblage. The resulting salinity of the dehydrated saline phase, trapped as synthetic inclusions in quartz, was > 96 wt% eq. CaCl2, consistent with the water-poor nature of the salt melt inclusions from the intrusions. The results of this study indicate that, through the loss of H2O, metal-bearing aqueous volatiles in layered intrusions may precipitate metals as they are dehydrated to form salt melt phases. Metal precipitation may occur as amount of free H2O in the volatile phase necessary to hydrate metal complexes decreases. This precipitation mechanism challenges the conventional magmatic hypothesis for platinum-group element deposit formation in layered intrusions.
V43A-1113
Rates of Thermal and Chemical Evolution of Magmas in a Cooling Magma Chamber Beneath Rishiri Volcano, Japan
Rates of magmatic processes in a magma chamber were investigated for alkali basalt and trachytic andesite lavas from Rishiri Volcano, northern Japan. Pre-eruption magmatic history of the lavas has been investigated by detailed petrology and geochemistry (e.g. Kuritani et al., 2005), and it has been shown that these lavas represent a series of magmas evolved by assimilation and fractional crystallization in the same magma reservoir. In addition, the eruption ages of the basalt and andesite lavas have also been estimated to be 29.3 ka and 20.2 ka, respectively (Kuritani et al., 2006; 2007), suggesting that the timescale for the parental basaltic magma to have evolved to the daughter andesitic magma was ~9.1 kyr. In this study, thermal and chemical evolution of the Rishiri magma chamber was modeled using mass and energy balance considerations, as well as quantitative constraints obtained from petrologic and geochemical observation on the lavas, in order to estimate timescales of magmatic processes. A model developed in this study considers a thermal structure of a magma chamber to be determined by conductive heat balance between the surrounding crust and the magma chamber, in which the main molten part of the magma chamber (main magma) is cooled by thermal convection. The model assumes a sill-like magma body, in which heat loss from the vertical walls of the reservoir is negligible (i.e. one-dimensional problem). The magma body is divided into three regions: the roof boundary layer, the main magma body, and the floor boundary layer. Heat transfer through the boundary layers and the surrounding crust is assumed to occur by conduction. The main magma is cooled solely by thermal convection, and is homogeneous in temperature and chemical composition. Following e.g. Turner (1979) and Kerr et al. (1990), the Nusselt-Rayleigh relationship is utilized to express the dimensional heat flux from a convecting liquid layer. The heat flux is further constrained by an observed relationship between estimated magmatic temperatures and bulk-rock chemical composition (K2O contents) of the lavas. In order to close the governing equations, the K2O contents are linked to parameters used in the equations by utilizing a mass balance model of Kuritani et al. (2005). Using the timescale of the magmatic evolution of ~9.1 kyr, the thickness of the magma body was estimated to have been ~1.7 km. The calculations show that, in the early stage of the evolution, the magma cooled at relatively high rate (>0.1 E#8249;C/year), and the cooling rate decreased with time. The rate of chemical evolution has also decreased progressively with time. Convective heat flux from the main magma body exceeded 2 W/m2 when the magma was basaltic, and the intensity diminished exponentially with magmatic evolution. Rate of convective melt exchange (compositional convection) between the main magma and mush melt also decreased progressively with time, from ~1 m/year to ~0.01 m/year, as the magmas evolved from basaltic to andesitic compositions.
V43A-1114
Sulfide Minerals in the Bushveld Complex
Debate on the origin of the major platinum-group element (PGE) deposits of the Bushveld, Stillwater and other layered intrusions range from conventional magma mixing-sulfide saturation models in the resident magma chamber to those that suggest that the PGE and S may have been introduced by upward-percolating fluids moving through the solidifying crystal pile. This study looked at details of the petrographic associations and compositions of sulfides in the Bushveld Complex away from the ore zones to see if they can help elucidate degassing and vapor transport in that intrusion. In fresh assemblages where the silicate have not been altered to a variety of low temperature minerals, the sulfide assemblages are typically composed of chalcopyrite, pyrrhotite, pentlandite, pyrite and a variety of other trace phases that are locally important. Textural and modal abundance suggests these assemblages formed by unmixing from high temperature monosulfide phase(s) on cooling. Throughout much of the complex pyrrhotite + pyrite –bearing assemblages are the most common. However, in the Lower and Lower Critical zones pyrite is locally absent and pyrrhitite is associated with a Ti-free magnetite. The pyrrhotite in these oxides-bearing assemblages is characterized by lower S contents as compared with the pyrite-bearing assemblages. These observations imply that the oxide-bearing assemblages formed as the rock cooled along a lower fS2/fO2 path, and are consistent with S loss to degassing interstitial fluids in the rocks beneath the Merensky Reef.
V43A-1116
Magma Intrusion, Deformation: the Importance of Crustal Layering.
The Campi Flegrei caldera (CF) is a volcano-tectonic depression, between Naples and the volcanic islands of Ischia and Procida, and is a highly populated area (about 400000 people). Since the last eruption (Monte Nuovo, 1538 A.D.) the CF caldera suffered notable unrest episodes, including large ground deformations, seismic swarms and increases in the degassing activity. The caldera had been continuously subsiding (at about 1.5 cm per year) from 1538 till 1969. A substantial ground uplift, more than 1 m of deformation, occurred in the period 1969-1972 and, after a small subsidence of about 30 cm after 1972, a very strong uplift occurred in the period 1982-1984 (about 1.8 m), with subsequent partial recovery. Superposed on the still continuing subsidence are some short uplift phases (mini-uplifts during 1989, 1994, 2000, 2004-2006); ground level remains about 2.5 m above pre-1970 levels at the town of Pozzuoli. Early papers on the 1982-1984 CF unrest usually invoked magmatic intrusion to explain observed ground deformation and gravity changes. Later papers invoked fluid intrusions or hybrid sources (including both magmatic and hydrothermal components). Some authors suggested the key role of freely slipping ring faults on the deformation pattern. We show that crustal layering plays a key role in deformation (horizontal and vertical displacement) pattern as well as gravity changes. Using a layered model appropriate for the CF caldera (based on seismically derived estimates of the P wave speed for the crust) we are able to better fit all deformation (horizontal and vertical displacement) and gravity data for the 1982-1984 large uplift with a horizontal crack source intruded by silicate melts. The same source is shared by the most recent (2004-2006) mini-uplift. In the case of the 1982-1984 large uplift, we also investigate whether the presence of a deeper deflating supply reservoir (as suggested for Uzu volcano in Japan) is required by the data.
V43A-1117
Does Chaotic Micro-mixing of Magmas Undermine the Melt Inclusion Paradigm in Igneous Petrology?
Recent research on magma mixing systems has shown that the mixing process is governed by chaotic dynamics and that this process is responsible for the generation of fractal structures that propagate within the magmatic mass from the meter to the micrometer length-scale. Laser Ablation ICP-MS trace element analyses have been performed on rock samples with evidence of chaotic mixing phenomena. Results indicate that trace elements with similar values of diffusion coefficient display good correlations in inter elemental plots whereas, as the difference between diffusion coefficients increases, the correlation is progressively lost. In addition, a large variability of REE patterns is observed, with the remarkable feature of the presence of positive and negative Eu anomalies occurring at short length scale, of the order of few mm. Given the chaotic nature of magma mixing structures, the mixing process has been simulated by coupling a chaotic advection and a chemical diffusion numerical scheme by considering several trace elements with variable diffusivities. Simulations indicate that such a model explains with good approximation the variable correlations among trace elements observed in natural samples. In addition, the same patterns of REE observed in natural samples, including the occurrence of positive and negative Eu anomalies at short length scale, are observed indicating that a chaotic advection/diffusion dynamic system is a suitable model to explain natural data. Results presented in this contribution indicate that at the micrometric length-scale small volumes of magmas are strongly influenced by the coupled action of chemical diffusion and chaotic flow fields and, hence, they do not represent magmas 'de facto' present in the magmatic system because their compositions may have experienced a 'diffusive fractionation' process. These results may have important petrogenetic implications. For instance, if such melts were trapped as melt inclusions, they would provide misleading information about melt compositions. It is suggested that the approach of studying the degree of correlation among trace elements may be a possible method to test if melt inclusion compositions, commonly used as petrogenetic indicators, display evidence of such a 'diffusive fractionation' process.
V43A-1118
Immiscibility of Fluid Phases at Magmatic-hydrothermal Transition: Formation of Various PGE-sulfide Mineralization for Layered Basic Intrusions
Fluid inclusions in quartz of the Merensky Reef (Bushveld Complex, South Africa) and the Chineisky Pluton (Transbaikal Region, Russia) were studied using cryometry, microthermometry, Raman-spectroscopy, LA ICP- MS, scanning electronic microscopy, gas-chromatography and isotopic methods. This allowed us to document some examples of fluid phase separation resulting in formation of different types of PGE-sulfide mineralization for layered basic intrusions. The results obtained show at least three generations of fluid separated from boiling residual alumosilicate intercumulus liquid of the Merensky Reef. The earliest fluid phase composed of homogenous high-dense methane and nitrogen gas mixture was identified in primary gas and co-existing anomalous fluid inclusions from symplectitic quartz. The next generation, heterophase fluid, composed of brines containing a free low-dense (mostly of carbon dioxide) gas phase, was observed in primary multiphase and coexisting gas-rich inclusions of miarolitic quartz crystals. The latest generation was also a heterophase fluid (low salinity water-salt solution and free low-dense methane gas phase) found in primary water-salt and syngenetic gas inclusions from peripheral zones of miarolitic quartz crystals. For the Chineisky Pluton reduced endocontact magmatogene fluids changed to oxidized low salinity hydrothermal fluids in exocontact zone. This resulted in formation of sulfide-PGE enrichment marginal zones of intrusion. The results obtained give us a possibility to suggest that: 1) Fluid phase separation is a typical feature of magmatogene fluids for layered basic intrusions. 2) Reduced fluids can extract and transport substantial PGE and sulfide concentrations. 3) Oxidation of reduced fluids is one of the most important geochemical barriers causing abundant PGE minerals and sulfides precipitation. This in turn results in both formation of PGE reefs or enriched contact zones of layered basic intrusions. This work was supported by the Ministry for Russian Science and Education, Grant DSP.2.1.1.702, by RFBR Grants ## 07-05-00685, 07-05-00803, Grant VMTK-2007 IGM SB RAS.
V43A-1119
Geochemical, Mineralogical, and Textural Characterization of the Beacon Sill, McMurdo Dry Valleys, Antarctica
The Beacon sill, part of the Jurassic Ferrar Large Igneous Province, is a 150 meter thick, macroscopically uniform diabase sill intruding the Beacon sandstone at Beacon Heights. It is part of a complex that includes the underlying Basement and Peneplain sills, and the overlying Kirkpatrick basalts. Although these other units have been the subject of intense research, the Beacon sill, perhaps because of its apparent homogeneity, has not been examined in detail. In this study, variations in bulk-rock chemistry, modal mineralogy, and texture are measured in a complete section, sampled at 5 m intervals, through the sill. The textural variations, as quantified using CSDs, indicate that despite the generally homogenous nature of the diabase in the sill, this intrusion was inflated in a pulsatile fashion. In particular, small but distinguishable reversals in the overall coarsening-inwards trend mark the interfaces between partially solidified magma from an earlier injection pulse and fresh magma from a reinjection event. Further evidence for this model is provided by mineralogical and geochemical anomalies consistent with redistribution of interstitial liquid across the reinjection horizon. Thermal modeling will be used to constrain the frequency of reinjection events based on the spatial distribution of textural and compositional anomalies.
V43A-1120
The Magma Chamber Simulator: An Interactive Computer Program for Modeling the Chemical and Physical Evolution of Complex Magmatic Systems
The Magma Chamber Simulator (MCS) is an interactive graphical computational tool for modeling the chemical and physical evolution of a complex magmatic system defined here as a magma body-wallrock couple. The tool seamlessly integrates the phase equilibria capabilities of MELTS (Ghiorso & Sack 1995) with trace element and isotope conservation equations based on the energy constrained (EC-RAFC) approach of Spera & Bohrson (2001, 2002, 2004) and Bohrson & Spera (2001, 2003). Evaluation of trace element behavior includes partitioning among coexisting solid, melt and fluid phases (Spera et al. 2007). Each MCS simulation provides a thermodynamically based description of the chemical and energetic state of a magma body and its surrounding wallrock as it evolves along a complex P-T-X path. The MCS is designed to address closed-system fractional or equilibrium crystallization as well as open-system processes such as assimilation of wallrock partial melt, stoping of wallrock blocks, recharge, and eruption. Input includes initial compositions and masses of magma, wallrock, and recharge magma and mineral-melt-fluid partition coefficients for trace elements. Compositions (major, trace element, isotopes) and abundances of solids and melt, and thermodynamic and physical properties of the system (e.g., viscosity, density, volume fraction fluid) are output. Because users may be interested in a subset of the large database generated for each simulation, the MCS offers an efficient method of presenting results tailored to the user's interest. The MCS also includes a graphing tool that allows direct comparison between model results and data for a particular natural system. Because the graphical display is updated after each step, the user can compare model vs. natural data in real-time and modify MCS input. Modifications are accommodated through an interactive module that allows the user to stop the simulation, return to previous states, modify a subset of input parameters (e.g., recharge magma composition), and continue the simulation. The potential of the MCS will be highlighted by examining results of a case study in which mafic magma undergoes assimilation-fractional crystallization in typical upper crust. Data for compositions of initial magma and wallrock provide the starting point for a MELTS analysis that incorporates energy conservation, such that, for each step of the simulation, energy liberated from the magma provides constraints on the degree of melting in wallrock; thus, the compositions of associated anatectic melt and residual solids are known. Some user-defined proportion of anatectic melt is incorporated into the magma, and the MCS provides physical and chemical information about the contaminated magma and the wallrock. Trace element analysis involves calculation of bulk partition coefficients for elements of interest using fluid-solid-melt partition coefficients estimated from experimental or theoretical data. Because the mineral assemblage and mass fraction of fluid for each step are output, calculated bulk partition coefficients reflect compositional and other dependencies, and the behavior of elements that are partitioned into a separate fluid phase is also quantified. Addition of the stoping, recharge, and eruption functions will expand the capabilities of the MCS, thereby providing a comprehensive modeling tool to address mechanistic questions about crustal magmatic processes.
V43A-1121
Mineralogy and Texture of the Peneplain Sill, McMurdo Dry Valleys, Antarctica
The Peneplain sill, part of the Jurassic Ferrar Large Igneous Province, is a 335 m thick diabase sill in the Dry Valleys, Antarctica. Mineralogical and textural data were collected for a 335 meter transect through the sill at Pandora's Spire, Solitary Rocks. Variations in the modal abundance of plagioclase, pyroxene, opaque minerals, and granophyric intergrowth through the Peneplain sill were determined by point counting. The mineralogical composition of the rocks varies from ~54% plagioclase, ~34% pyroxene, and ~10% granophyre near the lower contact to ~51% plagioclase, ~40% pyroxene, and ~7% granophyre at 145 m (approximately the center of the sill). In the upper half of the sill, the composition returns to values close to those at the lower contact, ~56% plagioclase, ~30% pyroxene, and ~12% granophyre. In summary the modal mineralogy is very homogeneous through the sill. There is a slight (<5%) increase in pyroxene abundance with a more significant decrease in granophyre abundance near the center of the sill. Textural variations in the Peneplain sill were quantified using plagioclase crystal size distributions (CSDs). CSDs for the Peneplain sill reveal a high population density of small (~0.1 mm) crystals, and a low population density of large (~1.0 mm) crystals near the contacts. Towards the center of the sill, the population density of small crystals decreases slightly while the population density of large crystals remains nearly constant. There is a slight irregularity in the population density trend in the small size classes at a height of ~50 m, but the significance of this is unclear. Texturally, mineralogically, and geochemically, the Peneplain sill is remarkably homogeneous. Many other apparently homogeneous intrusions are seen to be heterogeneous upon closer inspection. This sill therefore provides a unique opportunity to study an important end-member of intrusion petrogenesis.