Roles for fluid and/or melt advection in forming high-P mafic migmatites, Fiordland, New Zealand

A series of striking migmatitic structures occur in rectilinear networks through western Fiordland, New Zealand, involving, for the most part, narrow anorthositic dykes that cut hornblende‐bearing orthogneiss. Adjacent to the dykes, host rocks show patchy, spatially restricted recrystallization and dehydration on a decimetre‐scale to garnet granulite. Although there is general agreement that the migration of silicate melt has formed at least parts of the structures, there is disagreement on the role of silicate melt in dehydrating the host rock. A variety of causal processes have been inferred, including metasomatism due to the ingress of a carbonic, mantle‐derived fluid; hornblende‐breakdown leading to water release and limited partial melting of host rocks; and dehydration induced by volatile scavenging by a migrating silicate melt. Variability in dyke assemblage, together with the correlation between dehydration structures and host rock silica content, are inconsistent with macroscopic metasomatism, and best match open system behaviour involving volatile scavenging by a migrating trondhjemitic liquid.     

HYDROCHEMISTRY AS AN INDICATOR OF SUBGLACIAL DRAINAGE SYSTEM STRUCTURE: A COMPARISON OF ALPINE AND SUB-POLAR ENVIRONMENTS

The anion compositions (SO²₄, HCO⁻₃ and Cl⁻) of runoff from the Haut Glacier d'Arolla, Switzerland and Austre Brøggerbreen, Svalbard are compared to assess whether or not variations in water chemistry with discharge are consistent with current understanding of the subglacial drainage structure of warm- and polythermal-based glaciers. These glacial catchments have very different bedrocks and the subglacial drainage structures are also believed to be different, yet the range of anion concentrations show considerable overlap for SO²⁻₄ and HCO⁻₃. Concentrations of Cl⁻ are higher at Austre Brøggerbreen because of the maritime location of the glacier. Correcting SO²⁻₄ for the snowpack component reveals that the variation in non-snowpack SO²⁻₄ with discharge and with HCO⁻₃ is similar to that observed at the Haut Glacier d'Arolla. Hence, if we assume that the provenance of the non-snowpack SO²⁻₄ is the same in both glacial drainage systems, a distributed drainage system also contributes to runoff at Austre Brøggerbreen. We have no independent means of testing the assumption at present. The lower concentrations of non-snowpack SO²⁻₄ at Austre Brøggerbreen may suggest that a smaller proportion of runoff originates from a distributed drainage system than at the Haut Glacier d'Arolla.     

IN SITU MEASUREMENTS OF BASAL WATER QUALITY AND PRESSURE AS AN INDICATOR OF THE CHARACTER OF SUBGLACIAL DRAINAGE SYSTEMS

Continuous subglacial measurements of turbidity and electrical conductivity — two indicators of basal water quality — can be used to help characterize subglacial drainage systems. These indicators of water quality yield information that complements that provided by water pressure measurements. Quantitative attributes of subglacial drainage systems, such as water velocity and subglacial residence time, as well as qualitative behaviour — for example, spatial and temporal variations in system morphology — can be deduced using water quality measurements. Interpretation is complicated by the many potential influences on turbidity and electrical conductivity, but when these complications are appreciated a richer interpretation results. To demonstrate the utility of basal water quality measurements, observations from Trapridge Glacier, Yukon Territory, Canada were examined. The data reveal complex behaviour of the drainage system, but constraints imposed by basal water quality measurements help to clarify the nature of the subglacial flow system. The measurement and interpretation methods described and demonstrated are applicable to other glaciers. As such, they should prove useful for characterizing different subglacial drainage configurations and behaviours, thereby improving our general understanding of the hydrology and dynamics of wet-based glaciers.     

Decompressional coronas and symplectites in granulites of the Musgrave Complex, central Australia

Abstract In granulite facies metapelitic rocks in the Musgrave Complex, central Australia, reaction between S1 garnet and sillimanite involves the development in S2 of both garnet + cordierite + hercynitic spinel + biotite and hercynitic spinel + cordierite + sillimanite + biotite. The S2 assemblages occur either in coronas and symplectites, mainly around garnet, or, in rocks in which S2 is more strongly developed, as recrystallized assemblages. Ignoring the presence of biotite and ilmenite, the mineral textures can be accounted for qualitatively by a consideration of the model system FeO-MgO-Al₂O₃-SiO₂ (FMAS); the textural relationships accord with decompression accompanying the change from S1 to S2. However, since biotite and ilmenite are involved in the assemblages, the parageneses are better accounted for in terms of equilibria in the expanded model system K₂O-FeO-MgO-Al₂O₃-SiO₂-H₂-TiO₂-Fe₂O₃ (KFMASHTO), i.e. AFM + TiO₂+ Fe₂O₃. The coronas reflect the tectonic unroofing of at least part of the Musgrave Complex from peak S1 conditions of about 8 kbar to S2 conditions of about 4 kbar.     

A Holocene calcrete from North Yorkshire, England: implications for interpreting palaeoclimates using calcretes

Glacial gravels of Late Devensian Dimlington Stadial age (26 000–13 000 years BP) at West Tanfield, North Yorkshire, England, have been cemented by carbonate-rich solutions to produce a strongly indurated calcrete horizon. The low-Mg cements occur as drusy spar, needle fibres, alveolar septal structures, micrite and micropinnacles, indicative of vadose-zone cementation. Some complex pore partition structures attributed to precipitation along meniscus films also occur. These partitions separate air-dominated and water-dominated microenvironments of the vadose zone. The abundance of vadose fabrics shows that the accumulation is not a groundwater calcrete. In addition, much of the carbonate appears to have been precipitated by biological mediation. Carbon and oxygen isotopic data suggest that the carbonate did not form as a result of freezing, as has been suggested for some ‘arctic’soil carbonates. The pollen history of the area since the Devensian suggests that this calcrete precipitated at low temperatures; this contrasts with widely reported occurrences of calcrete in soils of hot arid or semi-arid regions, and suggests that palaeo-calcretes should not be used as absolute palaeoclimatic indicators. The unusual occurrence, albeit localized, of a thick calcrete under a cool and wet climate probably reflects the well-drained nature of the gravels, the abundance of CaCO₃ as limestone clasts in the gravel and a high degree of biological activity beneath a forest cover, which created a local environment favouring carbonate precipitation.     

Mbosi: An anomalous iron with unique silicate inclusions

Abstract— The Mbosi iron meteorite contains millimeter size silicate inclusions. Mbosi is an ungrouped iron meteorite with a Ge/Ga ratio >10, which is an anomalous property shared with the five-member IIF iron group, the Eagle Station pallasites and four other ungrouped irons. Neither the IIF group nor the four other ungrouped irons are known to have silicate inclusions. Chips from three Mbosi inclusions were studied, but most of the work concentrated on a whole 3.1 mm circular inclusion. This inclusion consists of a mantle and a central core of different mineralogies. The mantle is partially devitrified quartz-normative glass, consisting of microscopic crystallites of two pyroxenes and plagioclase, which are crystalline enough to give an x-ray powder diffraction pattern but not coarse enough to permit analyses of individual minerals. The core consists of silica. The bulk composition does not match any known meteorite type, although there is a similarity in mode of occurrence to quartz-normative silicate inclusions in some HE irons. Mbosi silicate appears to be unique. The bulk rare earth element (REE) pattern of the mantle is flat at ? 7×C1; the core is depleted in REE but shows a small positive Eu anomaly. The O-isotope composition of bulk silicate lies on a unit slope mixing line (parallel and close to the C3 mixing line) that includes the Eagle Station pallasites and the iron Bocaiuva (related to the IIF irons); all of these share the property of having Ge/Ga ratios >10. It is concluded that Mbosi silicate represents a silica-bearing source rock that was melted and injected into metal. Melting occurred early in the history of the parent body because the metal now shows a normal Widmanstätten structure with only minor distortion that was caused when the parent body broke up and released meteorites into interplanetary space. The cause of Ge/Ga ratios being >10 in these irons is unknown. The fact that silicates in Mbosi, Bocaiuva (related to IIF irons) and the Eagle Station trio of pallasites, all characterized by a Ge/Ga ratio >10, lie on a unit slope mixing line in the O-isotope diagram suggests that their origins are closely related. The C3 chondrites appear to be likely precursors for silicates in Mbosi, Bocaiuva and the Eagle Station pallasites.     

Uncertainties in mean discharges from two large South American rivers due to rating curve variability

Abstract

The aim of this study was to assess the effect of land application of sewage sludge on phosphorus (P) losses during intense rainfall. Three rainfall simulations (40 mm h^?1 of 30 min duration) were conducted on a field amended with sewage sludge. The overland flow water (OFW) was monitored and sampled every minute. The suspended solid, the dissolved and total phosphorus (respectively SS, TP and DP) concentrations were analysed. The forms of particulate bound P (PP) were investigated. Several results stem from this experiment: (a) sludge application induced a large increase in the DP content of the OFW; the concentrations obtained (0.15–0.57 mg l^?1) were shown to result from desorption processes from the SS; and (b) in contrast, sludge application affected neither the SS content nor the TP concentration of OFW (9.5 g mg l^?1 P, consisting of PP for 95%). However, sludge preserved the structure of soil surface and led to a 45% decrease in runoff rate (150 m³ ha^?1 collected on the test surface compared to 290 m³ ha^?1 on a reference). This indirectly reduced TP losses (2.7 kg ha^?1 on the reference surface compared to 1.4 kg ha^?1 on the test surface).