Petrology and Emplacement of Reversely Zoned Gabbro-Diorite Plutons in the Smartville Complex, Northern California

Gabbroic plutons are part of the intrusive substructure of theSmartville Complex, a late Jurassic, rifted, ensimatic arc locatedin the northern Sierra Nevada of California. The plutons rangefrom unzoned, equant bodies of olivine gabbro less than 1 kmin diameter to elongate intrusions up to 25 km in length thatare reversly zoned from olivine gabbro cores to quartz dioriterims. The felsic rocks dip inward beneath the mafic core, indicatingthat this zoning reversal continues to depth. The zoned plutonshave relatively shallow keels. We interpret the reversed zoningas an emplacement feature, analogous to the compositional zoningin a zoned tephra sheet. It formed as a result of tapping analready zoned, deeper level magma chamber. Whether the originalzoning of the magma was concentric or stratiform cannot be readilydeduced. During emplacement, considerable amounts of cumulaterocks were mobilized. The mineralogy and geochemistry of the reversely zoned plutonsindicate that they contain two suites of rocks: a cumulate suiterepresented by olivine gabbro and olivine clinopyroxenite anda differentiated suite of non-cumulate olivine gabbros, gabbronorites,and diorites that lie along a compositional continuum and approximateliquid compositions. Plagioclase and olivine compositions inthe Smartville Complex cumulate suite are identical to thosein modern arc cumulates and are characteristic of the arc cumulatesuite. The differentiated rocks form a compositionally continuousseries that is geochemically very similar to a differentiatedsuite of arc tholeiitic basalts and andesites. Fractionationmodeling indicates that removal of mineral phases found in thecumulate gabbros from the mafic members of the differentiatedsuite can produce the lithologic variation seen in the zonedplutons. Plutons such as those in the Smartville Complex indicatethat there is a genetic link between cumulate rocks and a basalt-andesitefractionation trend in arcs, supporting the hypothesis thatarc andesites form by crystal fractionation. The gabbroic plutonsand related Alaska-type ultramafic complexes contain ultramaficcumulates that can rectify the discrepancy between the cumulatemode predicted by fractionation models and the observed modeof gabbroic cumulates in arcs.     

Hydrous Basaltic Andesites Associated with Minette and Related Lavas in Western Mexico

The convergent margin of western Mexico is uniquely characterizedby a volcanic front of lamprophyric and related lavas located{small tilde}70 km closer to the Middle America trench thanthe main axis of andesitic volcanism. This front, defined bysmall volcanic centers ranging in age from {small tilde}1 kato 3 Ma, contains several lava types: minette, absarokite, leucitite,spessartite, and kersantite, all without feldspar phenocrysts.Many of the lavas contain hydrous phenocrysts; they are enrichedin potassium and other incompatible elements, and they are moreoxidized relative to the andesitic suite of the main axis. Intimatelyassociated are flows of basaltic andesite of comparable volume.They range in composition from 53 to 58 wt.%SiO₂, have 5–9wt.%MgO and contain phenocrysts of olivine, sparse augite, andvarying amounts of plagioclase. Their alkali contents are typicalof calc-alkaline varieties, with average Na₂O and K₂O concentrationsof 4?2 and 1?1 wt.% respectively. The basaltic andesites oftencontain olivine of unusually high forsterite content, reflectingcrystallization under oxidizing conditions, and they have oxygenfugacities up to 3?3 log units above the Ni-NiO buffer. Manifestationsof high water contents are (1) the ubiquitous occurrence ofgroundmass olivine rather than orthopyroxene, and (2) the suppressionof plagioclase as an early crystallizing phase. Both featuresreflect the role of water in reducing the activity of silicain the melt. The progressive influence of water during crystallizationis also seen in the continuum between the two intermediate lavatypes, basaltic andesite and kersantite, as plagioclase is suppressedand hornblende is stabilized in the phenocryst assemblage. Thus,despite the absence of hydrous minerals in the basaltic andesites,their phenocryst assemblages reveal the influence of substantialamounts of water, and thereby show a genetic link to the variouslamprophyric lavas.     

Analysing laser‐scanned digital terrain models of gravel bed surfaces: linking morphology to sediment transport processes and hydraulics

The grain‐scale topography of a sediment surface is a key component of a fluvial system, affecting aspects including sediment transport, flow resistance and ecology. However, its effect is hard to quantify because of the need for grain‐scale elevation data from in situ fluvial gravel surfaces which are difficult to collect. The sediment surface properties are, therefore, commonly estimated as a function of the sediment grain‐size distribution; however, because of additional factors, such as grain packing and shape, there is not necessarily a unique relationship between the two. A new methodology has been developed that uses terrestrial laser scanning to collect grain‐scale topographic data from in situ fluvial gravel surfaces, from which digital terrain models are created. This paper investigates methods of analysing such digital terrain models, and possible sedimentological interpretations that can be drawn from the analysis. Eleven digital terrain models from exposed gravel surfaces in two contrasting rivers (the River Feshie and Bury Green Brook) were analysed by calculating: the distribution of surface elevations, semivariograms, surface inclinations, surface slopes and aspects and grain orientation. The distribution of surface elevations and surface slope and aspect analysis were found to be most informative. In the River Feshie, grain‐size was interpreted as being a dominant control on sediment surface structure and gravel imbrication was identified. In Bury Green Brook, the location of the digital terrain models within the riffle–pool sequence was the dominant control on surface structure and grain orientation. Such digital terrain models therefore provide a new approach to measuring and quantifying the topography of fluvial sediment surfaces.     

Dehydration Melting and Water-Saturated Melting of Basaltic and Andesitic Greenstones and Amphibolites at 1, 3, and 6. 9 kb

The melting relations of five metamorphosed basalts and andesites(greenstones and amphibolites), collected from the late JurassicSmartville arc complex of California, were investigated experimentallyat 800–1000? C and 1, 3, and 6. 9 kb. Dehydration-melting(no water added) experiments contained only the water structurallybound in metamorphic minerals (largely amphiboles). They yieldedmildly peraluminous to metaluminous granodioritic to trondhjemiticmelts (Na/K is a function of starting composition) similar inmajor element composition to silicic rocks in modern oceanicarcs. The dehydration melts are water-undersaturated, with,and coexist with the anhydrous residual solid (restite) assemblageplagioclase + orthopyroxene + clinopyroxene + magnetite ? ilmen-ite,with plagioclase constituting 50% of the restite mode. In thedehydration-melting experiments at 3 kb the onset of meltingoccurred between 850 and 900 ? C, as amphibole and quartz brokedown to yield pyroxenes plus melt. Total pressure is greaterthan in the dehydration-melting experiments and has little effecton melt composition or phase relations. In the water-saturated (water added, so that experiments, meltsformed at 3 kb and above are strongly peraluminous, rich inCa and poor in Fe, Mg, Ti, and K. Their compositions are unlikethose of most silicic igneous rocks. These melts coexist withthe amphibole-rich, plagioclase-poor restite assemblage amphibole+ magnetite ? clinopyroxene ? plagioclase ? ilmenite. The highlyaluminous nature of the melts and the plagioclase-poor natureof the restite both reflect the substantial contribution ofplagioclase (along with quartz) to melts in high-pressure water-saturatedsystems. Water pressure equals P_toul in the water-saturatedexperiments and has a profound effect on both melt compositionand phase relations. At 1 kb, the water-saturated experimentsyielded melt and mineral products with some characteristicsof the dehydration-melting experiments (no amphibole at highT), and some characteristics of the 3-kb, water-saturated experiments(amphibole plus melt coexisting at lower T, elevated Al, loweredFe). As pressure is increased from 3 to 6. 9 kb, the stabilityfields of both plagioclase and clinopyroxene decrease relativeto amphibole and the Al contents of the melts increase. These experiments have important implications for the petrogenesisof low-K silicic rocks in arcs. First, dehydration melting isa viable mechanism for the formation of these rocks; water-saturatedmelting is not. Second, because of the influence of rock compositionon melt composition, low-grade metamorphic and hydrothermalprocesses that alter the alkali contents and Na/ K in arc basementterranes may have a direct impact on the petrogenesis of silicicmagmas in arcs, particularly the formation of extremely low-Ktrondhjemites. Third, the experiments predict that anhydrous,pyroxene- and plagioclase-rich ‘granulitic’ restiteassemblages should develop as a result of partial melting inarc terranes. Such assemblages occur in at least two deeplyeroded arc complexes.     

Gabbroic Pegmatite Intrusions, Iberia Abyssal Plain, ODP Leg 173, Site 1070: Magmatism during a Transition from Non-volcanic Rifting to Sea-floor Spreading

On the Iberia Abyssal Plain (Ocean Drilling Program Site 1070),gabbroic pegmatites and related rocks (127 ± 4 Ma, U–Pbzircon) intrude upper mantle that was subsequently exposed andserpentinized during Early Cretaceous non-volcanic rifting.The pegmatites include a 3–4 m dike or sill (the ‘main’pegmatite), numerous dikelets of 1–5 cm thickness, andclasts within the overlying ophicalcite breccia. Exclusive ofrodingitization, the main pegmatite contains 40–70% calcicandesine, 25–35% kaersutitic amphibole (Mg# 60–70),5–25% augite (Mg# 70–80) and 1–2% ilmenite.The dikelets are more magnesian (Mg# up to 82 in kaersutiteand 88 in augite). Most indications are that the high Mg#s inthe dikelets reflect igneous compositions. Isotopic and elementalchemistry indicate that the pegmatite-forming melt was enrichedin incompatible elements relative to normal mid-ocean ridgebasalt, but not as enriched as Azores basalts. The amphibole-bearingplagioclase peridotites of the Iberia Abyssal Plain are an appropriatesource for the pegmatite melts. A combination of decompressionaccompanying unroofing and heating from the upwelling asthenospherebeneath the developing rift caused P–T conditions in theamphibole-bearing lithosphere to exceed the dehydration-meltingsolidus (     

Evolution of a large Miocene growth structure in the upper plate of the Whipple detachment fault, northeastern Whipple Mountains, California

Miocene sedimentary and volcanic rocks in the north-eastern Whipple Mountains, California, and the north-western Aubrey Hills, Arizona, accumulated in the upper plate of the Whipple detachment fault during regional extension and slip on the detachment. Miocene rocks in this area can be divided into three sequences: (1) pre-18.5-Ma dominantly volcanic rocks; (2) the 18.5-Ma Peach Springs Tuff; and (3) post-18.5-Ma dominantly sedimentary rocks. Important stratigraphic markers in sequence 3 include a 100- to 14–0-m-thick basalt unit and the voluminous War Eagle landslide, both of which correlate across Lake Havasu from the north-east Whipple Mountains to the Aurbrey Hills. We divide clastic sedimentary rocks of sequence 3 into three informal members: (3a) conglomerate and sandstone stratigraphically beneath the basalt; (3b) conglomerate and sandstone above the basalt and below the War Eagle landslide; and (3c) conglomerate and sandstone that overlie the War Eagle landslide. Detailed stratigraphic analysis and field mapping reveal dramatic south-westward thickening of member 3b strata, from about 50 m in the Aubrey Hills to over 1500 m in the north-east Whipple Mountains. In the north-east Whipple Mountains, this thick dipping section is overlain by the War Eagle landslide along a major angular unconformity; in the Aubrey Hills the base of the War Eagle landslide is roughly parallel to bedding dips of underlying strata. The above stratigraphic relationships can be explained by syndepositional growth of a rollover monocline by progressive tilting of the hangingwall above a master listric normal fault (Whipple detachment fault). This phase of upper-plate deformation began shortly after deposition of the basalt and ended prior to emplacement of the War Eagle landslide. Interbedded breccias low in member 3b, about 100 m above the basalt, record the first appearance of mylonitic detritus in the section. Growth of this upper-plate rollover was thus initiated at about the same time (shortly after deposition of the basalt) that the lower plate of the Whipple detachment fault was first exposed at the earth's surface by tectonic denudation and large-scale crustal uplift. These events are interpreted to record initiation of a secondary breakaway fault on the north-east flank of the growing Whipple detachment dome shortly after deposition of the basalt at about 14.5 (±1.0) Ma.     

Effects of P, f(O2) and Mg/Fe Ratio on Dehydration Melting of Model Metagreywackes

We present results of dehydration melting experiments [3–15kbar, 810–950C f(O₂) QFM (quartz-fayalite-magetite)and Ni-NiO] on two Fe-rich mixtures of biotite (37%), plagioclaseAn₃₈ (27%), quartz (34%) and ilmenite (2%), which differ onlyin their biotite compositions (mg-number 23 and 0.4). Dehydrationmelting of metagreywackes of constant modal composition generatesa wide range of melt fractions, melt compositions and residualassemblages, through the combined effects of pressure, Fe/Mgratio and f(O₂). Crystallization of garnet is the chief controlon melting behavior, and is limited by two reactions: (1) thebreakdown of garnet + quartz to orthopyroxene + plagioclaseat low P, and (2) the oxidation of garnet to magnetite + anorthite+ quartz (enstatite), which is sensitive to both f(O₂) andP. Because of these reactions, melting of Mg-rich metagreywackesis rather insensitive to f(O₂) but strongly sensitive to P;the converse is true for Fe-rich metagreywackes. Garnet crystallizationrequires that plagioclase break down incongruently, liberatingalbite. This increases the Na₂O content of the melts and enhancesmelt production. Thus, melting of metagreywacke in a reducingdeep-crustal environment (with garnet stable) would producemore, and more sodic, melt than would garnet-absent meltingof the same source material in a relatively oxidizing, shallow-crustalenvironment. KEY WORDS: anatexis; metasediments; gneisses; granites; garnet ^*Corresponding author. Telephone: 706-542-2394; fax: 706-542-2425; e-mail: alpatino{at}uga.cc.uga.edu     

Variable Impact of the Subducted Slab on Aleutian Island Arc Magma Sources: Evidence from Sr, Nd, Pb, and Hf Isotopes and Trace Element Abundances

Major and trace element compositions and Sr, Nd, Pb, and Hfisotope ratios of Aleutian island arc lavas from Kanaga, Roundhead,Seguam, and Shishaldin volcanoes provide constraints on thecomposition and origin of the material transferred from thesubducted slab to the mantle wedge. ⁴⁰Ar/³⁹Ar dating indicatesthat the lavas erupted mainly during the last