<p> The abstract is not available from PDF copy and paste.</p>

Subduction of an oceanic plate beneath the Bering Strait region resulted in the formation of an extensive magmatic belt along the margin of the region during the Cretaceous. Plutons and gneiss domes of this age are broadly distributed throughout parts of the region, in Chukotka, Russia and in Alaska. Variably deformed biotite granite and related diorite through tonalite dikes intrude all levels of the Koolen gneiss dome in NE Chukotka. Granodiorite and granite plutons and associated mafic hypabyssals form a cauldera assemblage at the eastern tip of the Okhotsk-Chukotsk volcanic belt in SE Chukotka. Mafic rocks from both localities show trace element characteristics indicative of subduction-related magmatism, including LILE enrichement and Nb and Ti depletion. All rocks from both localities classify as high-K calc-alkaline rocks and have comparably enriched REE concentrations. These characteristics, along with enriched isotope ratios, suggest derivation of the mafic rocks from enriched subcontinental mantle. At both field areas, mantle-derived magma, produced during subduction processes, intruded lower crustal levels and differentiated. (Abstract shortened by UMI.)

Examination of an ancient sedimentary basin can provide only an incomplete geologic picture, because it reveals merely the anatomy of a dead body. Physiology must be learned on a living organism, and likewise the geologist must study the present ocean floors if he is to gain a full understanding of a sedimentary basin. On the sea floor processes of transportation, deposition, and early stages in consolidation can be investigated along with the accompanying "Recent paleogeography". Viewing a sedimentary basin as a succession of sea floors, each with its overlying body of water, will improve interpretation of facies and thus correlations. Better understanding of relations between sedimentary environments and the resulting sediments and of gradual development of lithofication should materially assist in the location of the most prospective areas for petroleum exploration.

Geochemical and petrologic tools were deployed to investigate the redox state and water content of the earth's upper mantle. Study results are discussed in the context of their linkages to the atmospheric oxygen level, hydrospheric water budget and lithospheric evolution of continents. Because the partitioning of V is redox-sensitive and otherwise similar to that of Sc which is not redox sensitive, the V/Sc ratios of basalts of different ages act as a natural recorder of the redox states of the upper mantle. Through a comparison between global mid-ocean ridge basalts and Archean basalts, the fO2 of the upper mantle was inferred to have changed by no more than 0.3 log units since Archean. Combined with results from a thermodynamic model simulating the redox reactions of volcanic gases, this observation argues against the idea that the increase in oxygen in the atmosphere &sim;2.3 billion years ago was caused by redox transition in the upper mantle. Through a geochemical and petrologic study at the Feather River Ophiolite (in northern California), global water recycling rates at subduction zones were estimated based on reconstructed serpentinization depths for the oceanic lithospheric mantle. Within uncertainties, the estimated water recycling rates roughly match global volcanic dewatering rates, which suggest the hydrospheric water storage is current at steady-state. Based on water contents measured in mantle xenoliths from the Colorado Plateau and vicinity, the idea that the lithospheric mantle beneath the western North America was rehydrated by the dewatering of the flat-subducting Farallon slab is confirmed. As predicted by an updated flow law for olivine aggregates, hydration might have weakened the basal lithosphere beneath the Colorado Plateau and thus induced lithospheric thinning by &sim;15 km as a result of basal erosion. Extrapolation of the flow law to thick, cratonic lithosphere further suggests lithospheric thinning of much larger extents can occur if enough water was introduced during hydration. If so, subduction-induced hydration might have played an important role in regulating the lithospheric evolution of continents.

Magnesian granulites in Allan Hills A81005 and Dhofar 309 may possess the geochemical signature of the Moon's Feldspathic Highlands Terrane (FHT). Previous work on Apollo granulites suggests the presence of a high Mg/Fe, feldspathic, rare earth element-poor component in the lunar crust. Remote sensing data have strengthened this inference, showing such a composition is widespread in the FHT. My research extends the characterization of magnesian granulites to those in feldspathic lunar meteorites, which are essentially unstudied. I reconstruct the bulk chemistry of granulites in thin section by combining mineral proportions (retrieved from element X-ray maps via multispectral image classification software) and mineral compositions. This virtually non-destructive method is useful for rare samples and fragments in thin section. Magnesian granulites in Allan Hills A81005 and Dhofar 309 share similar major and trace element compositions. As these two meteorites are not paired, their comparable geochemistry may be representative of the average FHT.

The Great Barrier Reef (GBR) margin, located on the continental margin between Papua New Guinea and northeast Australia, is the largest extant example of a tropical mixed siliciclastic-carbonate depositional system. It is constructed by the combined input of terrigenous siliciclastic sediment delivered through riverine transport and biogenous carbonate sediment from neritic and pelagic production. This study investigates late Quaternary changes in sediment dispersal and accumulation on the slopes of this margin. Sedimentation across the GBR mixed system also serves as an important analog for understanding deposition on other extant and ancient systems and provides insight into global change, geochemical cycling, and resource management. Several concepts (e.g., reciprocal sedimentation, coeval sedimentation) have been proposed to explain spatial and temporal variations in siliciclastic and carbonate components. While these concepts are frequently used to evaluate ancient tropical mixed systems, they are rarely assessed in the Quaternary, an interval where the magnitude and timing of sea level are relatively well-constrained, and precise dating techniques can be used. These studies of the GBR mixed system integrate a full suite of data including core, seismic, and multi-beam bathymetry to gain a quantitative understanding of the GBR system and to evaluate reciprocal sedimentation concepts. Results indicate that slopes along the GBR margin have a complex depositional history. Sea level, climate, and margin physiography are all important depositional controls affecting timing, location, and mechanism of sediment dispersal. Reciprocal sedimentation can be used to predict carbonate accumulation. However, this approach must be combined with a firm understanding of sedimentary controls and processes to accurately predict siliciclastic accumulation along this margin.

Oxygen isotope analyses of corals from the Gulf of California and Gulf of Panama are calibrated with environmental parameters. Uncertainty in the isotopic composition of sea water in the Gulf of California prohibits strong correlation between temperature and $\delta\sp{18}$O. The marginal condition for coral growth in the Gulf of California prevents the construction of a long paleoclimate record. The Gulf of Panama record is strongly correlated with salinity (r$\sp2$ = 0.72) and indicates a trend toward drier conditions in Panama since 1950. Although El Nino is recorded in both records through a decrease in the annual maximum isotopic value, this signature is not unique to El Nino thereby minimizing the ability of corals to monitor this system. Strong coherence between the Gulf of Panama record and three other Panamanian records attests to the ability of corals to record regional climate variability.

During the late Cenozoic mafic lavas were emplaced contemporaneously with Basin and Range extension near Battle Mountain, Nevada. They display wide ranges in major and trace elements contents and Sr, Nd, and Pb ratios. The temporal and compositional characteristics of the basalts indicate that the effects of crustal contamination on the compositions of the younger lavas has decreased relative to the older lavas. Assimilation and fractional crystallization (AFC) modeling suggests that their compositional heterogeneity cannot be attributed solely to crustal contamination, and instead most likely reflects their derivation from a heterogeneous mantle source region. Regional variations in the trace element ratios (Rb/La and Ba/Zr) and Sr, Nd and Pb isotopic compositions of mafic lavas from the western U.S. indicate they were derived from a heterogeneous lithospheric mantle. Crustal contamination and subduction related processes appear to be unable to reproduce the range in $\sp{87}$Sr/$\sp{86}$Sr and low Rb/La of the lavas. Therefore, the variation in the basalts' isotopic ratios most likely reflects their derivation from a heterogeneous mantle source region. Plots of $\sp{87}$Sr/$\sp{86}$Sr versus $\sp{207}$Pb/$\sp{204}$Pb or $\sp{207}$Pb/$\sp{207}$Pb clearly illustrates the compositional characteristics of their source regions. The Snake River Plain and southern Sierran Province basalts display elevated $\sp{87}$Sr/$\sp{86}$Sr and varied Pb isotopic ratios indicating they were derived from a heterogeneous ancient lithospheric source with high time integrated Rb/Sr and U/Pb ratios. The low $\sp{87}$Sr/$\sp{86}$Sr and Pb isotopic ratios of the suggest the Colorado Plateau-Eastern Transition Zone basalts were derived from a heterogeneous ancient lithospheric source with low time integrated Rb/Sr and U/Pb ratios. Basalts from the Basin and Range, Oregon Plateau, and northern Sierran Province display a positive correlation between their $\sp{87}$Sr/$\sp{86}$Sr versus $\sp{208}$Pb/$\sp{204}$Pb or $\sp{207}$Pb/$\sp{204}$Pb ratios, which appears to reflect two component mixing between accreted depleted oceanic-type mantle and an enriched mantle component.

Sr, Nd, and Pb isotope systematics of lavas from the Maui Volcanic Complex (MVC) are consistent with a three-component petrogenetic mixing model. MVC shield-building (SB) lavas define linear trends on isotope-isotope plots, consistent with two-component mixing between primitive (PM) and enriched (EM) mantle components. The two-component (PM-EM) Hawaiian plume source is variable in composition during production of tholeiite magmas even within a single shield. Sr and Pb isotopic ratios of Haleakala post shield-building (PSB) lavas define a strong positively correlated array that deviates from the SB array towards an unradiogenic end-member. The PSB array may therefore result from time- and volume-dependent binary mixing between Hawaiian plume melts and a depleted (DM) mantle (i.e. MORB source) component. Several trace element ratios in Haleakala PSB lavas are correlated with isotopic compositions but not with major and trace element contents, and therefore appear to reflect changes in source composition. Trace element mixing systematics for these lavas indicate that the DM component must be a melt. The inferred PM component has chondritic ratios for several trace elements, consistent with it representing primitive mantle. The EM component may represent a part of the Hawaiian plume source that was either metasomatized or metasomatically scavenged. Alkalic cap lavas exposed in the northwest wall of Haleakala Crater display systematic, upsection geochemical variations indicative of the repetitive intrusion of discrete magma batches. Magma batches are separated by geochemical discontinuities characterized by abrupt upsection increases in incompatible element contents and commensurate drops in compatible element contents. In contrast, lava compositions within magma batches vary upsection progressively, and geochemical variations are opposite to those observed for interbatch discontinuities. Together, these geochemical variations are interpreted as resulting from the cyclic operation of a dynamic, evolving, open system magma chamber. Interbatch transitions appear to reflect periods of eruptive quiessence characterized by low magma recharge rates and relatively high degrees of crystal fractionation. Intrabatch variations appear to represent eruptive periods characterized by relatively high recharge rates, low degrees of crystal frationation, and progressive mixing of evolved rest magma with more primitive recharge magma.

The effects of fluid evolution and infiltration in the Alta stock contact aureole, Utah, were studied by evaluating chemical additions and depletions in igneous, replacement skarn, and carbonate whole-rock samples. Of the major and trace elements studied, boron proved to be the most interesting tracer of fluid, due in part to extensive borate mineralization (ludwigite and kotoite) in carbonates near the stock contact, and in part to the physical-chemical properties of boron which make it a useful element for isotopic studies and for spatial distribution analysis via alpha track mapping. Boron is enriched throughout the Alta aureole and is hosted by a variety of metamorphic and metasomatic silicate minerals (forsterite and humite group are the most significant). Aureole whole-rock B concentrations increase toward the contact, indicating that B originated in the stock and migrated down-temperature in exsolved magmatic fluids. Boron stable isotope measurements of carbonates, skarns and mineral separates range from $-$7.2 to +1.6$\perthous\ \delta\sp{11}$B. A value of ${-}6.0\perthous$ for a B-enriched igneous sill falls within this range and also indicates that the stock is the source of aureole B-enrichments.