Sj Edwards

Whole-rock geochemistry of gabbros from the Southwest Indian Ridge: constraints on geochemical fractionations between the upper and lower oceanic crust and magma chamber processes at (very) slow-spreading ridges

Whole-rock major and trace element compositions are presented for a suite of gabbroic samples formed at the Southwest Indian Ridge with the aim of constraining magmatic processes at ultra slow-spreading centres. The gabbros, together with subordinate basalts, dolerites and peridotites were collected from a not, vert, similar700 km2 area around Atlantis Bank, adjacent to the Atlantis II Fracture Zone during cruise JR31 of the RRS James Clark Ross. The large area sampled, the abundance of gabbro, and the recovery of samples representative of all other levels in the oceanic lithosphere, allow an average lower crustal composition to be estimated. The estimated composition is not sufficiently primitive in terms of Mg/Fe or compatible trace element abundances (Ni, Cr) for the bulk crust to be in equilibrium with the mantle. This is probably due to compositional modifications within the mantle during melt extraction, although crystallisation within the crust in an unsampled area, either towards the segment centre or along a flow line, cannot be ruled out. Gabbro compositions show evidence for being mixtures of cumulate crystals and significant proportions of basalt, with the proportion of each end-member dependent on the distribution coefficient of the element in question. This suggests that the concept of ‘trapped melt’ cannot be used to understand the origin of these compositions and consideration of the migration of interstitial liquid within a crystal mush is necessary. The fractionation of incompatible elements between the upper and lower crust correlates with the observed degree of variation in trace element concentrations in basalts from the same spreading segment. This suggests that interstitial liquids are, in part, extracted from the crystal mush and mixed back into subsequently erupted basalts leading to their compositional modification.

New insights concerning the influence of water during the formation of podiform chromitite

Podiform chromitites are pod-like bodies dominated by chromian spinel and enveloped by dunite. Most major bodies are Cr rich and occur in the harzburgitic mantle section and the mantle-crust transition zone of ophiolitesformed in supra-subduction zone environments. Current models for the origin of podiform chromitites require a combination of crystal fractionation and melt-peridotite reaction as basaltic melt migrates through the upper mantle. Although such models in general account for large accumulations of chromian spinel and the field and textural relationships in and around podiform bodies, they do not adequately deal with the problem of the low solubility of Cr in basaltic melts, which begs the questions: How does Cr enter melt and why is it transported to a restricted zone in a melt conduit where accumulation of chromian spinel takes place? The present study suggests that the answer may lie with water and its ability to depolymerize the silica network of the melt. The formation of most, if not all, melts capable of forming podiform chromitites requires the involvement of water in the zone of melt generation. Water not only promotes partial melting of refractory peridotites, it also dissolves in the resulting melt. This hydrous melt will have a greater abundance of octahedral sites than its anhydrous equivalent, and this modification promotes greater solubility of Cr in the hydrous melt owing to the high octahedral site preference energy of Cr3+-the major ion controlling Cr solubility. Cr is held in the melt until the melt and peridotite reaction raises the Si content and extent of polymerization of the melt (so that the concentration of octahedral sites is reduced), at which point the Cr becomes insoluble and chromian spinel precipitates. Growth of chromian spinel takes place in an open system in which there is a continuous supply of Cr-rich melt during a melting episode. An examination of the partial-melting history of peridotites preserved in the Troodos ophiolite, Cyprus, indicates that Cr behaves as an incompatible element during the hydrous melting of refractory peridotites. The highest Cr contents of melts that separate from these peridotites are rarely seen in the extrusive sequences of the Troodos ophiolite, suggesting that these melts fractionated substantial volumes of chromian spinel prior to eruption.

A baseline survey of the distribution and origin of platinum group elements in contemporary fluvial sediments of the Kentish Stour, England

The extensive anthropogenic use of the platinum group elements (PGE: Ru, Rh, Pd, Os, Ir and Pt) is the main factor responsible for the widespread dispersion of these elements throughout the environment. Significant quantities of the PGE enter fluvial systems via road runoff, storm drains and wastewater and sewage treatment systems, and may accumulate in fluvial sediments by physical and chemical processes. A baseline survey of contemporary fluvial sediments in the Kentish Stour river, east Kent, England, has been undertaken to document the sources and distribution of anthropogenic PGE in an attempt to constrain some of the physical and chemical parameters that may influence the distribution of these elements. The geology of the catchment of the Kentish Stour is dominated by carbonate and silicate sedimentary rocks, and the river passes through urban and rural land and receives inputs of waste from sewage works. Nine sedimentary rocks, 4 motorway-runoff sediments and 22 river sediments were analysed for PGE by NiS fire assay preconcentration and ICP–MS. The highest element abundances occur in the motorway-runoff sediments (maximum total PGE content of 55 ng/g), whilst the lowest values were recorded in the sedimentary rocks, where some samples contain PGE at concentrations below the limit of detection. The total PGE content of the river sediments ranged from 0.4 to 10.8 ng/g. The distribution and variation in concentrations and ratios of the PGE in the contemporary fluvial sediments of the Kentish Stour correspond strongly with land-use changes (urban versus rural) and with points of discharge from sewage works. The absence of a pure catalytic converter signature in the river sediments, however, indicates that source signatures may be mixed in sewage works or that PGE may chemically fractionate in the fluvial environment.

MELT-PERIDOTITE INTERACTIONS IN SHALLOW MANTLE AT THE EAST PACIFIC RISE : EVIDENCE FROM ODP SITE 895 (HESS DEEP)

Abstract Ocean Drilling Program (ODP) Leg 147 recently drilled at Site 895 in Hess Deep (eastern Equatorial Pacific), where a structurally dissected section of the East Pacific Rise (EPR) is preserved, and intersected the mantlecrust transition zone of a fast-spreading centre for the first time. The core from Hole 895D (latitude 2°16.645ʹN longitude 101°26.777ʹW)revealed that harzburgite is predominant over dunite in the top section of the Hole, but the reverse relationship is found lower in the section where dunite is closely associated with gabbroic rocks (gabbro and troctolite). Texture, mineralogy and mineral chemistry suggest a two-stage evolution for harzburgite preserved at the transition zone. Harzburgite with a porphyroclastic texture was produced by partial melting of peridotite to, or beyond the clinopyroxene-out phase boundary before or during asthenospheric (>1000°C flow, which suggests a higher degree of mantle melting than normally expected below mid-ocean ridges. Subsequently, basaltic melt(s) interacted with this refractory harzburgite (olivine + orthopyroxene + spinel), which resulted in dissolution of orthopyroxene, re-equilibration and formation of olivine and spinel, and formation of clinopyroxene ± plagioclase, this is manifested as a progressive conversion of harzburgite to gabbroic rock through an intermediate dunite. At low melt/peridotite ratios, harzburgite was refertilised as the plagioclase component of the melt completely reacted with the peridotite matrix to produce clinopyroxene-spinel intergrowths and Al enrichment in ferromagnesian minerals. At high ratios, orthopyroxene completely dissolved incongruently, plagioclase appeared, and spinel was partially to completely resorbed; this produced olivine-bearing and olivine-free gabbroic rocks. Residual minerals in peridotites adjacent to gabbroic zones were enriched in Fe and Ti and depleted in Al.

Mineralogical controls on storage of As, Cu, Pb and Zn at the abandoned Mathiatis massive sulphide mine, Cyprus

Abstract The Mathiatis massive sulphide deposit in Cyprus was a low-grade (0.3% Cu), three million ton ore body of pyrite and minor chalcopyrite occurring within basaltic lavas of the Troodos ophiolite. Cessation of mining in 1987 left a deep open pit surrounded by large heaps of spoil, which are undergoing oxidation and leaching. The aim of this study was to determine the mineralogical controls on the storage and potential remobilization of As, Cu, Pb and Zn within, and from, mine spoil heaps. Most of the spoil samples collected, and related materials (stream sediment, reaction zone between a boulder of massive pyrite and calcareous chert, salt crusts on stream beds), are enriched in As (27-220 ppm), Cu (110-400 ppm), Pb (10-140 ppm) and Zn (290 - 12,000 ppm) relative to both the basalt and calcareous chert (As 4-10 ppm, Cu 20-76 ppm, Pb 3-6 ppm, Zn 39-200 ppm). Arsenic, Cu, Pb and Zn in the spoil and related materials are associated with Fe(-Al-S)-O, Fe(-Al-Mg)-S-O, Al(-Mg-Fe)-S-O and Mg(-Al-Fe)-S-O phases (the brackets represent minor components of less than 20 wt.% within the phases). Chemical extraction work using CaCl2 suggests that Cu, Zn and to some extent, As, are potentially more soluble than Pb. This is corroborated by the very high total concentrations of Cu and Zn in both the secondary salt crusts and the reaction zone material, high CaCl2-extractable As, Cu and Zn in the salt crusts, and aqueous data for the Mathiatis mine area collected for a European Union LIFE report. This may have implications for ecosystem health and water quality in the Mathiatis area and areas of similar mineralogy and climate world wide.

Contrasting mantle sequences exposed in the Lewis Hills massif: Evidence for the early, arc-related history of the Bay of Islands ophiolite

A major fracture zone is preserved within the Lewis Hills massif (Bay of Islands ophiolite). It separates arc-type crust in the west from 17 m.y. younger, backarc-related oceanic crust in the east. We link mantle peridotites occurring in the Springers Hill locality to the arc crust and use their petrologic characteristics to define the early history of the arc. The Springers Hill mantle displays a weak shape fabric, high Cr/(Cr + Al) ratios in spinel (>60), very low, chondrite-normalized, heavy rare earth element abundances in clinopyroxene (HREE N < 1), and scattered calcic amphibole. It contains two important chromitite deposits and numerous pyroxenite dikes, some of which have a boninitic composition; and it is intruded by a late, large ultramafic body. The Springers Hill mantle peridotites are modeled by 26% melting of a primitive mantle fluxed by a slab melt or sediment (open-system melting with <1% total addition). The melting was outlasted by melt and fluid migration that caused middle rare earth element enrichment through chromatographic effects. This late infiltration event was fabric-destroying by causing orthopyroxene breakdown and recrystallization. In the absence of a clearly identifiable old basement for the arc, and given the extreme degree of depletion of the mantle peridotites underlying the arc, we interpret the Springers Hill mantle as related to upwelling and extension in an infant arc setting. Such a primitive arc setting is supported by strong geochemical similarities of the Springers Hill mantle to peridotites exposed in the Mariana and Bonin forearc systems. In a later event, the infant arc was rifted to give way to mantle upwelling, backarc formation, and associated final exhumation of the Springers Hill mantle within an extensional fracture-zone setting. The exhumation was probably associated with renewed melting at depth and melt and fluid migration in the Springers Hill mantle.

A review of the petrology of harzburgites at Hess Deep and Garrett Deep: implications for mantle processes beneath segments of the East Pacific Rise

Abstract In recent years a unique set of samples of uppermost mantle at the mantle-crust transition zone have been collected from two different environments along the fast-spreading East Pacific Rise (EPR): a ‘normal’ spreading segment (represented by samples from Hess Deep) and the end of a spreading segment where the EPR meets the Garrett transform fault (represented by samples from Garrett Deep). A review of the petrology of harzburgites from the two sites demonstrates that these rocks were produced by partial melting of adiabatically upwelling mantle and, subsequently, at the top of the mantle, they were impregnated by reactive and crystallizing mid-ocean ridge basaltic (MORB) melts. Despite this similar history, non-impregnated harzburgites at Garrett Deep have a more fertile spinel chemistry than those at Hess Deep, which is consistent with reduced partial melting of shallow mantle as a transform fault is approached — the ‘transform fault effect’. The extent of reaction between melt and harzburgite during the impregnation event suggests that melt arrived in the uppermost mantle in a highly reactive state because along the adiabatic path it had been highly channelled in spatially restricted conduits. This implies that mantle upwelling below the EPR was, and presumably still is, dominantly two dimensional (sheet-like). Within this framework, the chemical evolution of MORB melt below fast-spreading ridges will be significantly affected by melt-periodotite reaction only when melt reaches the uppermost mantle and mantle-crust transition zone, where along-axis transport of melt may also be important. Although the harzburgites from Hess Deep and Garrett Deep formed and evolved beneath different parts of different first-order segments of the EPR, the petrology of these rocks presents the best analogue available for defining real variations in mantle processes along a single first-order ridge segment in a fast-spreading environment.

Disaster risk reduction for natural hazards: Putting research into practice

Disaster risk reduction for natural hazards: Putting research into practice Stephen J. Edwards a c , Carina J. Fearnley a d , Catherine J. Lowe a e & Emily Wilkinson b f a Aon Benfield UCL Hazard Research Centre, Department of Earth Sciences, University College London, Gower Street, London, WC1E 6BT, UK b Department of Geography, University College London, Pearson Building, Gower Street, London, WC1E 6BT, UK c Now at: Aon Benfield UCL Hazard Centre, Institute for Risk and Disaster Reduction, Department of Earth Sciences, University College London, Gower Street, London, WC1E 6BT, UK d Aberystwyth University, Now at: Institute of Geography and Earth Sciences, Aberystwyth, SY23 3DB, UK e Now at: Aon Benfield, 55 Bishopsgate, London, EC2N 3BD, UK f Now at: Overseas Development Institute, 111 Westminster Bridge Road, London, SE1 7JD, UK

The importance of assessing variability in the distribution of anthropogenic palladium, platinum and rhodium in fluvial sediments.

The distribution and behaviour of anthropogenic platinum-group elements (PGE) in river systems has as yet attracted little attention. Aquiring good quality data in PGE analysis poses a serious challenge to the environmental chemist, not in the least due to the low concentrations at which the PGE are currently found in the environment. Using the River Stour in Kent, U.K. as a case study, the work presented in this paper focuses specifically on the spatial and temporal variability of PGE in fluvial sediments. The analysis of river bed sediments shows high variability both on a catchment scale and on a local scale within sample sites. Moreover, high relative standard deviation points to a heterogeneous distribution of PGE within sediment samples, which is characteristic of the elements and is commonly known as the nugget effect. If unassessed, analytical data are unlikely to be truly representative and the interpretation of PGE data could be inherently flawed. For this reason this paper suggests a need to include thorough quantification and reporting of this variability as a matter of practice in the acquisition of environmental PGE data.

The 1831 eruption of Babuyan Claro that never happened: has the source of one of the largest volcanic climate forcing events of the nineteenth century been misattributed?

The 1831 eruption of Babuyan Claro in the Philippines is regarded as one of the most significant volcanic climate forcing events of the nineteenth century. Modern databases have assigned the eruption a VEI of 4? and Magnitude of 4.7. Our analysis of historical sources, however, suggests that there was no such eruption in 1831 and that this date is the result of a misinterpretation of a traveller’s account which had been taken to be the primary source. We therefore suggest that the 1831 eruption is a false event. In this case, one or more eruptions elsewhere must have been responsible for producing the climate-impacting stratospheric sulphate aerosol in 1831. Our results reveal the need to re-evaluate the hazard assessment of Babuyan Claro volcano and also, potentially, the quantitative treatment of the 1831 stratospheric sulphate aerosol in climate models. The Babuyan Claro example discussed in this paper therefore reinforces a call for the careful analysis of primary historical sources in volcanology.