Gordon G. Goles

Geochemistry and petrogenesis of a basalt-benmoreite-trachyte suite from the southern part of the Gregory Rift, Kenya

In the southern Gregory Rift valley a series of transitional basalt, ferrobasalt, and benmoreite flows (1.65–1.4 Myr) is overlain by flood trachyte lavas (1.3–0.9 Myr). Mass balance calculations for major element compositions of rocks of this suite and their phenocrysts and microphenocrysts suggest that the ferrobasalts and benmoreites formed from magma resembling the most primitive basalt by closed system fractionation of plagioclase, clinopyroxene, olivine, titanomagnetite, and apatite. The trachytes formed from evolved magmas largely by alkali feldspar fractionation. Estimates of phenocryst and liquid densities and Rayleigh-law modelling of trace element contents support these conclusions. From Rayleigh-law modelling, we derived a set of effective distribution coefficients. Partial melting of crustal rocks or volatile transfer processes had no significant effect on the petrogenesis of this suite. The duration of the eruptive cycle, cooling time calculations, and mass balance calculations suggest that fractionation occurred in a magma reservoir with volume of at least 3 × 104 km3 during an interval of about 0.8 Myr. Temperatures during fractionation probably ranged from about 1200 °C to 900 °C, and pressures may have been roughly 5 to 8 Kb. We suggest that rift development was accompanied by large-scale injection of basaltic magma and dilation of the crust, extensive fractionation, preferential eruption of low-density and fluid trachytic flood lavas, and by several episodes of normal faulting.

Re-Os Isotopic Evidence for Early Differentiation of the Martian Mantle

Variations in the short-lived systems of 182Hf-182W and 146Sm-142Nd in the SNC meteorites indicate an early isolation of, and subsequent inefficient mixing between, mantle reservoirs in Mars. Correlations of eW and e142Nd with initial γOs are consistent with the Re-Os isotopic systematics of these meteorites being set during the earliest differentiation history of Mars. Contamination by a juvenile Martian crust may have affected Zagami Os isotopic systematics but successful contamination models combining Nd and Os systematics, are inconsistent with such a process affecting the isotopic compositions of the shergottite lherzolites (EETA 77005, LEW 88516, Y 793605). At least two long-lived mantle reservoirs, and possibly three, are required to explain the observed systematics. One mantle reservoir (NC Group), represented by Nakhla and Chassigny, has a projected present day γOs of −5.4 ± 2.6. Another mantle reservoir represented by the shergottite lherzolites and possibly Zagami, has a present day γOs of about +4. This represents a 2 to 3% enrichment in Re/Os relative to the primitive mantle estimated for the Earth (+1.6). A third mantle reservoir may be represented by DaG 476, having a nearly chondritic γOs coupled with very high e143Nd of around +40. The isotopic systematics of these reservoirs may be linked to development of cumulate crystal piles in a Martian magma ocean and variable amounts of late stage intercumulus melt (following Borg et al., 1997). In this model, fractional crystallization of olivine and possibly other phases with slightly subchondritic Re/Os, from a solidifying magma ocean, resulted in a lower Re/Os ratio in the NC Group source cumulates, and a resultant low γOs. Later cumulates or evolved melts crystallized with higher Re/Os ratios to produce the shergottite mantle reservoir(s), and hence, consequent higher γOs. Crystallization of the Martian magma ocean followed earliest core formation, as indicated by the correlation of eW with e142Nd and initial γOs.

Provenance interpretation of quartz by scanning electron microscope–cathodoluminescence fabric analysis

We used a cathodoluminescence (CL) detector attached to a scanning electron microscope (SEM) to study patterns of variable-intensity CL in quartz grains from a variety of igneous, metamorphic, sedimentary, and shock-deformed (meteorite-impact) rocks. Distinctive fabrics in quartz grains revealed by SEM-imaged differential CL include zoning, healed fractures, complex shears, planar features (shocked quartz), dark CL streaks and patches, indistinct, mottled texture, and nondifferential (low-contrast) CL. Zoning is common in volcanic quartz and some plutonic quartz. Zoned plutonic quartz is distinguished from volcanic quartz by the presence of closed fractures and dark CL streaks and patches. Metamorphic quartz displays either an indistinct, mottled texture, or nearly uniform (nondifferential) CL. Quartz from rocks severely deformed by tectonism displays a complex pattern of multiple, small-scale shears. Quartz from meteorite-impact sites and some system boundaries is characterized by intricate patterns of planar features, presumably created by shock metamorphism. Thus, the SEM-CL fabric-analysis technique provides a rapid method for distinguishing quartz from a variety of source rocks.

Is Quartz Cathodoluminescence Color a Reliable Provenance Tool? A Quantitative Examination

ABSTRACT We examined cathodoluminescence (CL) colors of quartz by using red (590-780 nm), green (515-590 nm), and blue (380-515 nm) optical filters interfaced with a cathodoluminescence (CL) detector attached to a scanning electron microscope (SEM). SEM/CL images taken through these filters were captured digitally and transferred to a computer. Luminescence intensities (luminosities) of the images were measured by using available commercial software. Measured luminosities of these CL images are directly related to relative intensities of red, green, and blue CL emissions. Luminosity data were then used to construct plots that display relative luminosities of the CL images acquired through the red, green, and blue filters. An unfiltered CL image of each quartz grain, generated by photons with wavelengths ranging from 200-700 nm, was also acquired. By subtracting the numerical luminosity values of the images acquired through the red, green, and blue filters from the luminosity value of the unfiltered image, the contribution to total luminosity provided by CL emission in the near ultraviolet (UV) was calculated. The CL colors of quartz from a variety of volcanic, plutonic, and metamorphic rocks and hydrothermal deposits were examined. Volcanic quartz phenocrysts have the most restricted CL color range, with strongest emission intensity in the blue wavelength band. CL colors of plutonic quartz overlap those of volcanic phenocrysts but extend over a broader range to include quartz that displays higher intensity of red emission. CL emission in hydrothermal (vein) quartz is similar to that in plutonic quartz, although some hydrothermal quartz exhibits stronger green-CL emission than does plutonic quartz. The CL colors of metamorphic quartz exhibit the widest variation, overlapping the color fields of both volcanic and plutonic quartz and extending further into the red. CL emission in the near UV makes a significant contribution ( 5-85 percent) to the total luminosity of SEM/CL images, particularly images of plutonic quartz. Because of overlap in the CL color ranges of volcanic, plutonic, metamorphic, and hydrothermal quartz, unambiguous identification of quartz provenance on the basis of CL color alone is problematic. It is difficult to distinguish between volcanic and some plutonic quartz, and between some plutonic and hydrothermal quartz, or to distinguish magmatic quartz from metamorphic quartz that exhibits blue CL color. Only metamorphic quartz that exhibits moderately strong red emission appears distinguishable (on the basis of color) from quartz of other origins. Our work thus suggests that CL color is not a reliable indicator of quartz provenance.

Geochemistry of high-grade eclogites and metarodingites from the Central Alps

Analyses for major, minor, and trace element contents of metamorphosed, variably rodingitized mafic rocks demonstrate substantial removal of Na and as much as three-fold gains in Ca as a consequence of rodingitization. Modest declines in Si and Fe can be explained in terms of dilution effects. Losses in K and Ba do not correlate with Ca% and may have been caused by an alteration process not related to the rodingitization. The Ca-metasomatism was not accompanied by a gain in Sr. The relative contents of Ti, Zr, Hf, Y, Co, Sc, and heavy REE show no readily detectable changes, despite the rodingitization (±other alteration) and subsequent metamorphisms, namely, eclogite facies (T≧800° C, P≧ 20kbar) followed by amphibolite facies, sillimanite zone. Protoliths were tholeiitic basalt or diabase, and gabbro, with trace element contents indicative of a spreading center origin. Trace element and REE patterns indicate low-pressure fractionation of this magma, with plagioclase stable. This petrogenesis is consistent with prior conclusions on the shallow crustal origin of the protolith of the eclogite-metarodingite-garnet lherzolite suite in the Cima Lunga-Adula nappe, Central Alps. Based on their bulk chemical composition, the mafic rocks in this suite could be the equivalent of Mesozoic ophiolitic rocks in the more external parts of the Alps.

A Miocene subcontinental plume in the Pacific Northwest: geochemical evidence

Most of the recent discussions on the number and kinds of components which can be distinguished in the Columbia River Basalt Group (CRBG) magmas have used arguments developed from isotopic evidence. In this paper, we consider relative contents of excluded trace elements, interpreted by means of spidergrams and abundance ratios, and compare data from the CRBG and related lavas with those of selected oceanic basalts. Although many recent models for CRBG petrogeneses do not include a plume component, spidergrams for the American Bar flows of the Imnaha Basalt of the CRBG have broad humps at Ba, Th, and Nb (Ta). This feature suggests that an enriched mantle plume component is present in the American Bar magmas, an inference which is weakly confirmed by 87Sr/86Sr data. This plume may have been identical to that presently beneath Yellowstone. The inferred hot spot track lies almost directly east-west, and requires a vector of motion of the Pacific Northwest part of the North American plate significantly different from that inferred by Minster and Jordan (1978) [31]. Our proposed vector for the North American plate agrees, however, with that suggested by the Anahim Belt, a volcanic lineament in southern British Columbia that has been proposed as a hot spot track. This proposed vector requires alterations in current models for Pacific and North American plate movement during the Cenozoic.

Evaluating crustal contamination in continental basalts: the isotopic composition of the Picture Gorge Basalt of the Columbia River Basalt Group

Crustal contamination of basalts located in the western United States has been generally under-emphasized, and much of their isotopic variation has been ascribed to multiple and heterogeneous mantle sources. Basalts of the Miocene Columbia River Basalt Group in the Pacific Northwest have passed through crust ranging from Precambrian to Tertiary in age. These flows are voluminous, homogenous, and underwent rapid effusion, all of which are disadvantages for crustal contamination while en route to the surface. The Picture Gorge Basalt of the Columbia River Basalt Group erupted through Paleozoic and Mesozoic oceanic accreted terranes in central Oregon, and earlier studies on these basalts provided no isotopic evidence for crustal contamination. New Sr, Nd, Pb, and O isotopic data presented here indicate that the isotopic variation of the Picture Gorge Basalt is very small, 87Sr/86Sr=0.70307–0.70371, ɛNd=+7.7-+4.8, δ18O=+5.6±6.1, and 206Pb/204Pb=18.80–18.91. Evaluation of the Picture Gorge compositional variation supports a model where two isotopic components contributed to Picture Gorge Basalt genesis. The first component (C1) is reflected by low 87Sr/86Sr, high ɛNd, and nonradiogenic Pb isotopic compositions. Basalts with C1 isotopic compositions have large MgO, Ni, and Cr contents and mantle-like δ18O=+5.6. C1 basalts have enrichments in Ba coupled with depletions in Nb and Ta. These characteristics are best explained by derivation from a depleted mantle source which has undergone a recent enrichment by fluids coming from a subducted slab. This C1 mantle component is prevalent throughout the Pacific Northwest. The second isotopic component has higher 87Sr/ 86Sr and δ18O, lower ɛNd, and more radiogenic Pb isotopic compositions than C1. There is a correlation in the Picture Gorge data of Sr, Nd, and Pb isotopes with differentiation indicators such as decreasing Mg#, and increasing K2O/TiO2, Ba, Ba/Zr, Rb/Sr, La/Sm, and La/Yb. Phase equilibrium and mineralogical constraints indicate that these compositional characteristics were inherited in the Picture Gorge magmas at crustal pressures, and thus the second isotopic component is most likely crustal in origin. Mixing and open-system calculations can produce the isotopic composition of the most evolved Picture Gorge flows from the most primitive compositions by 8 to 21% contamination of isotopic compositions similar to accreted terrane crust found in the Pacific Northwest. Therefore, in spite of the disadvantages for crustal contamination and their narrow range in isotopic compositions, the process controlling isotopic variation within the Picture Gorge Basalt is primarily crustal contamination. We suggest that comprehensive analyses for basaltic suites and careful consideration of these data must be made to test for crustal contamination, before variation resulting from mantle heterogeneity can be assessed.

Comments on petrogeneses and the tectonic setting of Columbia River basalts

Abstract Data on major and a few trace element contents for specimens of five formational-scale units of the Columbia River Basalt Group are reviewed and compared with those for other flood basalts and the Chyulu basalts of Kenya. Flows of the Grande Ronde Formation, which constitutes more than two-thirds by volume of the Columbia River Basalt Group, in several respects resemble calc-alkaline rocks, and they plot in the field for magmas erupted at destructive plate margins on a Th Hf Ta diagram. These features of Grande Ronde flows may be related to the anomalous setting of the Columbia River Basalt Group as a whole. Calc-alkaline characteristics are exhibited by other flood basalts but rarely to the degree shown by Grande Ronde flows. Few if any of the Columbia River basalts represent primary magmas. Rather, they generally have been affected by multi-stage fractionation, and perhaps also by interaction with lower crustal rocks. Their compositions also may reflect partial melting of fragments of subducted lithospheric slabs.

Geochemical and petrological characterization of ash samples from cascade range volcanoes

Abstract Twenty-nine samples of volcanic ash from the Pacific Northwest were analyzed by instrumental neutron activation techniques, with the aim of distinguishing among ashes from different sources. Preliminary results of petrographic studies of 42 ash or pumice samples are also reported. Geochemical characteristics of Mazama ash are defined, and problems induced by winnowing of crystalline material during transport and by weathering are discussed. Contents of La, Th, and Co, and La Yb ratios are shown to be good discriminants. Data on refractive indices and on proportions of crystalline materials also aid in distinguishing among the various volcanic ashes studied. Ash and pumice found in archaeological contexts at Fort Rock Cave, Paisley Cave, Wildcat Canyon, and Hobo Cave are all from Mount Mazama, presumably from the culminating cruption of 7000 years ago.

Some constraints on the origin of phonolites from the Gregory Rift, Kenya, and inferences concerning basaltic magmas in the Rift System

Plateau-type phonolites of the Gregory Rift represent magmas with densities of about 2.3 g/cm3. These magmas must have erupted soon after their formation. The average depth of the bases of the approximately 100 to 300 magma chambers from which they erupted was between about 10 and about 24 km. The average vertical extent of the magma chambers was between about 3 and about 8 km. The aggregate volume of phonolitic magma formed beneath the Rift in Miocene times probably lies in the range of 0.5 to 1×105 km3. Both the crystal fractionation model and the anatectic model for formation of this volume of phonolitic magma require the presence of a large reservoir of basaltic magma, probably picritic in character, with a volume of at least 5×105 km3 and perhaps as much as 20×105 km3. This reservoir presumably is now part of the dense basic intrusive complex along the Rift axis. The Miocene and Pliocene episodes of basaltic volcanism in this region may be related to eruption of evolved liquids from this reservoir.