Development of a non-linear mixed spectral finite difference model for turbulent boundary-layer flow over topography

Further development of the non-linear mixed spectral finite difference (NLMSFD) model of turbulent boundary-layer flow over topography is documented. This includes modifications and refinements to the solution procedure, the incorporation of second-order turbulence closures to the model and the three-dimensional extension of the model. Based on these higher order closures, linear limitations, boundary-layer approximation and non-linear effects are discussed. The impact of different turbulence closures on the prediction of the NLMSFD model is also demonstrated. Furthermore, sample results for 3D idealized topography (sinusoidal) are presented. The parameterization of drag over small-scale topography is also addressed.     

Nd,Pb and Sr isotopic data from the Napak carbonatite-nephelinite centre,eastern Uganda: an example of open-system crystal fractionation

Nd, Pb and Sr isotopic data from nephelinite lavas from the Tertiary nephelinite-carbonatite complex of Napak, eastern Uganda, show large isotopic variations that can only be attributed to open-system behaviour. Possible explanations of the data include mixing between nephelinitic melts derived from an isotopically heterogeneous mantle, or interaction between a HIMU melt and mafic granulites. In both models crystal fractionation, involving olivine and clinopyroxene, played an important role. Major element chemistry, textural evidence and isotopic data from clinopyroxene phenocrysts from the olivine-bearing nephelinites, suggest that the pyroxenes did not crystallize from their host liquids. The isotopic data from the clinopyroxene phenocrysts support an interpretation of crystal fractionation in an open magma system that was undergoing continuous isotopic change. This study emphasises the importance of using combined isotopic data from both whole rock and mineral phases to interpret the evolutionary history of a single eruptive centre.     

Role of melt during deformation in the deep crust

Deformation in the deep crust is strongly influenced by the presence of melt. Injected melt (or magma) weakens the crust because strain will tend to localize where melt is present. The amount of strain a pluton may accommodate is dependent on the length of time it takes for a pluton to crystallize and the strain rate. For plutons that intrude into rocks which are near the solidus temperature of the melt, crystallization times can be quite long (> 1Myr).
Partial melting of deep crustal rocks can lead to melt-enhanced embrittlement. This occurs because the volume change for most melting reactions is positive. Therefore, when the rate of melt production outpaces the rate at which melt can leave the system, the melt pressure increases. Eventually, the melt pressure may become sufficiently high that the melting rocks behave in a brittle fashion and fracture.
Conjugate sets of dilatant shear fractures filled with melt occur in migmatite from the Central Gneiss belt (Canada); this suggests that melt-enhanced embrittlement occurred in these rocks. An expression which relates the magnitude of differential stress to the angle between conjugate dilatant shear fractures is derived. Assuming that migmatite has a small tensile strength, differential stresses are ≤ 20 MPa in migmatitic rocks at the time melt-enhanced embrittlement occurs. The occurrence of melt-enhanced embrittlement shows that a switch in deformation mechanism from plastic flow to cataclasis is possible in the deep crust during melting. Furthermore, repeated episodes of melt-enhanced embrittlement in migmatitic rocks may be an efficient mechanism for extracting melt from partially melted terrains.     

Reintroduction of gold, other chalcophile elements and LILE during retrogression of depleted granulite, Tromøy, Norway

Several metamorphic and magmatic events were associated with the Sveconorwegian orogeny ( 1.1 Ga) that affected the Bamble shear belt, south Norway. Among the most important were (1) granuliteto amphibolite-facies metamorphism of the entire belt; (2) intrusion of mafic dykes (“hyperites”), followed by their localized metamorphism to amphibolite; (3) intrusion of granitic sheets, which in granulite terrane caused rehydration and retrogression. Previous studies have shown that rocks from across the belt metamorphosed to amphibolite and granulite grade during event (1) are strongly depleted in Au, Sb and As; while some granulites are also depleted in Rb, Cu and S. Gold and chalcophile elements were also lost during (2) amphibolization of hyperites. This paper is concerned with event (3), at which time the granulites had been uplifted and cooled. H₂O---CO₂ fluids, at least in part derived from the granitic magma, caused localized rehydration and retrogression of granulite. This was accompanied by the variable reintroduction of the same elements (Au, As, Sb, Cu, S and Rb) that had earlier been depleted. The nature of the reintroduction of Au and other elements has relevance to the origin of mesothermal gold deposits.     

The petrology and origin of coals from the lower Carboniferous Mattson formation, southwestern district of Mackenzie, Canada

Petrographic analyses were carried out on thin coals and coaly sediments from the Lower Carboniferous Mattson Formation at Clausen Creek and Jackfish Gap-Yohin Ridge in the northern part of the Liard Basin, northern Canada. The composition and optical characteristics indicate that the coals are high-volatile bituminous B, predominantly sapropelic (canneloid) and accumulated subaquatically.The coals are dominantly composed of inertinite-rich and exinite-rich durities with subsidiary inertites and clarodurites; vitrite is minor and liptite is rare. The inertinite-rich microlithotypes are dominated by semifusinite, but micrinite, semimacrinite and ?resino-inertinites are abundant. Sporinite, comprising megaspores, crassispores, tenuispores and miospores, is the dominant liptinite maceral with subsidiary cutinite and minor alginite. Except for pyrite, mineral matter is minimal.Three populations of telocollinite are observed: a low-reflectance variety (I), commonly associated with micrinite (as vitrinertite), displays weak brown fluorescence and a reflectance some 0.4-0.5% lower than type II; type II is non-fluorescing telocollinite, with intermediate reflectance (0.67-0.74% R_om), it occurs as vitrite and is also associated with micrinite; and a higher-reflectance telocollinite (III), having no fluorescence or association with micrinite, has variable reflectance (0.74-0.8% R_om) implying higher oxidation or gelification levels.The abundance of semimacrinite, macrinite and ?resino-inertinites in inertites and durites (I) suggests that much of the peat accumulated subaquatically. Furthermore, fluorescing vitrinite and an abundance of micrinite (derived by oxidation or coalification of bituminite), suggest that the coal accumulated under anaerobic conditions. The predominance of semifusinite in humic laminae and micrinite in sapropelic layers suggests extensive surface or near-surface oxidation of the peat. Oxidised sporinites suggest that they were wind-borne.Depositional environment is interpreted as marginal marine, perhaps in shallow lakes in the middle to upper delta plain. Peat accumulations probably began subaquatically at the oxygen-hydrogen sulphide interface, but periodic subaerial exposure and natural oxidation gave rise to the high inertinite coals. Upper Mattson coals are interbedded with algal laminites and probably accumulated in a lagoonal setting.     

Age estimations based on amino acid racemization: Reply to comments of J.F. Wehmiller

Determining geologic ages of fossils by amino acid racemization techniques is often difficult because of the uncertainties in assumptions about diagenetic temperatures. Two kinetic model methods have been employed. Method 1, used by us, assumes that racemization of amino acids in the bivalve mollusk Saxidomus giganteus from Willapa Bay, Washington, follows linear kinetics. Ages are calculated by means of first-order kinetic equations. Method 2, used by Wehmiller, involves an empirical non-linear kinetic model Method 1 is simpler in concept and more easily applied. Wehmiller claims that ambiguities in paleotemperature arise when method 1 is used and that these ambiguities can be reconciled by the use of method 2. We show that application of method 1 can also provide reasonable temperature histories and leads to age estimates that are consistent with the geologic history of the sedimentary deposits at Willapa Bay.     

Torsional instability in structures

Torsional instability under harmonic excitation was detected in structures with cubic softening member stiffness. A variational procedure was introduced to perform the analysis for the general problem formulation. Single-storey structures with zero eccentricity were studied first; previous findings were checked and corrected, and the behaviour was examined in greater detail. Two-storey structures with large eccentricity were also examined, and further regions of torsional instability were detected. Time history analysis was then employed to quantify the behaviour of various typical structures. An unbounded growth in rotation or column displacement was not observed. Rather, the instability appeared to be related to the jump in response which exists in single degree-of-freedom softening oscillators subjected to harmonic excitation. The practical significance of torsional instability is therefore rather weak, although prudent practice would recommend its avoidance.     

Stability of composite river banks

The stability of a river bank depends on the balance of forces, motive and resistive, associated with the most critical mechanism of failure. Many mechanisms are possible and the likelihood of failure occurring by any particular one depends on the size, geometry and structure of the bank, the engineering properties of the bank material, the hydraulics of flow in the adjacent channel and climatic conditions. Rivers flowing through alluvial deposits often have a composite structure of cohesionless sand and gravel overlain by cohesive silt/clay. Bank erosion occurs by fluvial entrainment of material from the lower, cohesionless bank at a much higher rate than material from the upper, cohesive bank. This leads to undermining that produces cantilevers of cohesive material. Upper bank retreat takes place predominantly by the failure of these cantilevers. Three mechanisms of failure have been identified: shear, beam and tensile failure. The stability of a cantilever may be analysed using static equilibrium and beam theory, and dimensionless charts for cantilever stability constructed. Application of the charts requires only a few simple measurements of cantilever geometry and soil properties. In this analysis the effects of cracks and fissures in the soil must be taken into account. These cracks seriously weaken the soil and can invalidate a stability analysis by affecting the shape of the failure surface. Following mechanical failure, blocks of soil must be removed from the basal area by fluvial entrainment if rapid undermining and cantilever generation are to continue. Hence, the rate of bank retreat is fluvially controlled, even though the mechanism of failure of the upper bank is not directly fluvial in nature. This cycle of bank erosion: undermining, cantilever failure and fluvial scour of the toe, operates over several flood events and has important implications for river engineering, channel changes, and the movement of sediment through fluvial systems.