The legacy of crystal-plastic deformation in olivine: high-diffusivity pathways during serpentinization

Crystal-plastic olivine deformation to produce subgrain boundaries composed of edge dislocations is an inevitable consequence of asthenospheric mantle flow. Although crystal-plastic deformation and serpentinization are spatio-temporally decoupled, we identified compositional readjustments expressed on the micrometric level as a striped Fe-enriched ( [`(X)]<sub>\textFe</sub> \bar{X}_{\text{Fe}}  = 0.24 ± 0.02 (zones); 0.12 ± 0.02 (bulk)) or Fe-depleted ( [`(X)]_\textFe \bar{X}_{\text{Fe}}  = 0.10 ± 0.01 (zones); 0.13 ± 0.01 (bulk)) zoning in partly serpentinized olivine grains from two upper mantle sections in Norway. Focused ion beam sample preparation combined with transmission electron microscopy (TEM) and aberration-corrected scanning TEM, enabling atomic-level resolved electron energy-loss spectroscopic line profiling, reveals that every zone is immediately associated with a subgrain boundary. We infer that the zonings are a result of the environmental Fe²⁺Mg₋₁ exchange potential during antigorite serpentinization of olivine and the drive toward element exchange equilibrium. This is facilitated by enhanced solid-state diffusion along subgrain boundaries in a system, which otherwise re-equilibrates via dissolution-reprecipitation. Fe enrichment or depletion is controlled by the silica activity imposed on the system by the local olivine/orthopyroxene mass ratio, temperature and the effect of magnetite stability. The Fe-Mg exchange coefficients K_\textD^{\textAtg/\textOl} K_{\text{D}}^{{{\text{Atg}}/{\text{Ol}}}} between both types of zoning and antigorite display coalescence toward exchange equilibrium. With both types of zoning, Mn is enriched and Ni depleted compared with the unaffected bulk composition. Nanometer-sized, heterogeneously distributed antigorite precipitates along olivine subgrain boundaries suggest that water was able to ingress along them. Crystallographic orientation relationships gained via electron backscatter diffraction between olivine grain domains and different serpentine vein generations support the hypothesis that serpentinization was initiated along olivine subgrain boundaries.     

Multiple ways of producing intermediate and silicic rocks within Thingmúli and other Icelandic volcanoes

Major and trace element compositions of rocks and coexisting phenocrysts of the Thingmúli volcano suggest a revision of the existing models for the formation of intermediate and silicic melts in Iceland. The new data define two compositional tholeiitic trends with a significant gap between them. A high-iron trend (HFe) contains 6–14 wt% total FeO in silicic rocks with c. 1 wt% MgO, as well as sodic plagioclase and hedenbergite phenocrysts. A low-iron trend (LFe) contains 3–5 wt% FeO at c. 1 wt% MgO, which is typical of Iceland but higher than MORB compositions. The most evolved phenocrysts of the LFe trend do not reach iron-rich end members. The HFe trend is interpreted as a result of fractional crystallization; numerical modelling using the MELTS algorithm suggests that crystallization took place under redox conditions constrained to one-log unit below the fayalite-magnetite-quartz oxygen buffer (FMQ-1). The LFe trend is explained by a combination of mixing between rhyolite and ferrobasalt, assimilation of hydrated crust and fractional crystallization under higher redox conditions (FMQ). The two trends and the gap are best defined in a plot of Mg# versus SiO₂ that is useful to unravel petrogenetic processes. For example, intermediate and silicic rocks of the Holocene volcanic systems of spreading rifts (e.g. Krafla), propagating rifts (e.g. Hekla) and off-rifts (Öræfajökull) also fall into high- and low-iron fields and outline a gap similar to Thingmúli. The identification of two compositional trends in erupted intermediate and silicic volcanic products shows that processes in the deep roots of single volcanic systems are highly diverse and likely controlled by local variations in the thermal gradients and the extend of hydrothermal alteration. Generalizations about the relationship between the compositions of intermediate and silicic rocks and plate tectonic setting, therefore, should be avoided.     

Phosphate removal from water by naturally occurring shale,sandstone, and laterite: The role of iron oxides and of soluble species

Aqueous phosphate removal by three geomaterials from Ivory Coast was evaluated to determine their potential application as low-cost phosphate adsorbents in wastewater treatment. Batch experiments showed that phosphate uptake strongly depended on pH. Laterite and sandstone dissolution was less pronounced compared to shale. A correlation between concentrations of aqueous cation species released from shale and phosphate uptake was observed. The kinetics were well described using the pseudo-second-order model. Isotherms displayed a saturation level on shale, while phosphate uptake continuously increased for laterite and sandstone. The removal efficiency decreased in the following ranking order: laterite > sandstone > shale. Laterite was also the most efficient adsorbent in column experiments. The high phosphate removal efficiency of laterite (8.3 mg PO₄ g^?1) was attributed to the presence of superparamagnetic low grain sizes of goethite. Laterite is a particularly promising material for further investigation in wastewater treatment technology such as constructed wetlands.     

Fluid composition and evolution in coesite-bearing rocks (Dora-Maira massif,Western Alps): implications for element recycling during subduction

Fluid inclusions and F, Cl concentration of hydrous minerals were analysed in the coesite-pyrope quartzite, the interlayered jadeite quartzite and their country-rock gneiss from the Dora-Maira massif using a combination of microthermometry, Raman spectrometry, synchrotron X-ray microfiuorescence and electron microprobe analysis. Three populations of fluid inclusions were recognized texturally and can be related to distinct metamorphic stages. A low-salinity aqueous fluid occurs in the retrogressed country gneiss and as late secondary inclusions in jadeite quartzite and chloritized pyrope. An earlier secondary population is found in matrix quartz of the jadeite- and pyro-pe-quartzites. This population can be related to the early decompression and so to incipient breakdown of garnet into phlogopite-bearing assemblages. The inclusion fluid is highly saline (up to 84 wt% equivalent NaCl) and contains Na, Ca, Fe, Cu and Zn as major cations. In pyrope quartzite, additional K was found in these brines, which locally coexist with CO₂-rich inclusions. The oldest fluid inclusions are preserved in kyanite grains included in fresh pyrope and in pyrope itself. In pyrope, all inclusions have decrepitated and contain magnesite, an Mg-phosphate, sheet-silicate(s), a chloride and an opaque phase, with no fluid preser ved. In contrast, the kyanite inclusions in pyrope preserve primary H₂O-CO₂ low-salinity fluid inclusions, probably owing to the low compressibility of the kyanite inclusions and host garnet. In spite of in-situ re-equilibration, these inclusions can be interpreted as relics of the dehydration fluid that attended pyrope growth. These correlations between textural and chemical fluid inclusion data and metamorphic stages are consistent with the fluid composition calculated from the halogen content of different generations of phlogopite and biotite. The preservation of different fluid compositions, both in time and space, is evidence for local control and possibly origin of the fluids, in agreement with isotopic data. These results, in particular the absence of CO₂ in the jadeite quartzite, are best interpreted in terms of a fluid-melt system evolution. With increasing metamorphism, partitioning of H₂O, Na, Ca, Fe and heavy metals into melt (jadeite quartzite) and Mg, Na/K, F, CO₂ and P(?) into a residual aqueous fluid can account for depletion in Na, Ca and Fe of the pyrope quartzite. During the retrograde path, a _{H 2 O} rose as melt crystallized, generating the two populations of hypersaline and water-rich fluids that were highly reactive to pyrope. The process of fluid-melt interaction envisioned here coupled with models of melt extraction in subduction zones provides an attractive opportunity for the instantaneous ( < 1 Ma) and selective transport of elements between a downgoing slab and the overlying mantle wedge.     

Isotopic fractionation of Cu in tektites

Tektites are terrestrial natural glasses of up to a few centimeters in size that were produced during hypervelocity impacts on the Earth’s surface. It is well established that the chemical and isotopic composition of tektites is generally identical to that of the upper terrestrial continental crust. Tektites typically have very low water content, which has generally been explained by volatilization at high temperature; however, the exact mechanism is still debated. Because volatilization can fractionate isotopes, comparing the isotopic composition of volatile elements in tektites with those of their source rocks may help to understand the physical conditions during tektite formation.Interestingly, volatile chalcophile elements (e.g., Cd and Zn) seem to be the only elements for which isotopic fractionation is known so far in tektites. Here, we extend this study to Cu, another volatile chalcophile element. We have measured the Cu isotopic composition for 20 tektite samples from the four known different strewn fields. All of the tektites (except the Muong Nong-types) are enriched in the heavy isotopes of Cu (1.98 < δ⁶⁵Cu < 6.99) in comparison to the terrestrial crust (δ⁶⁵Cu ≈ 0) with no clear distinction between the different groups. The Muong Nong-type tektites and a Libyan Desert Glass sample are not fractionated (δ⁶⁵Cu ≈ 0) in comparison to the terrestrial crust. To refine the Cu isotopic composition of the terrestrial crust, we also present data for three geological reference materials (δ⁶⁵Cu ≈ 0).An increase of δ⁶⁵Cu with decreasing Cu abundance probably reflects that the isotopic fractionation occurred by evaporation during heating. A simple Rayleigh distillation cannot explain the Cu isotopic data and we suggest that the isotopic fractionation is governed by a diffusion-limited regime. Copper is isotopically more fractionated than the more volatile element Zn (δ^66/64Zn up to 2.49‰). This difference of behavior between Cu and Zn is predicted in a diffusion-limited regime, where the magnitude of the isotopic fractionation is regulated by the competition between the evaporative flux and the diffusive flux at the diffusion boundary layer. Due to the difference of ionic charge in silicates (Zn²⁺ vs. Cu⁺), Cu has a diffusion coefficient that is larger than that of Zn by at least two orders of magnitude. Therefore, the larger isotopic fractionation in Cu than in Zn in tektites is due to the significant difference in their respective chemical diffusivity.     

Temporal and geochemical evolution of the Cenozoic intraplate volcanism of Zealandia

In order to constrain better the distribution, age, geochemistry and origin of widespread Cenozoic intraplate volcanism on Zealandia, the New Zealand micro-continent, we report new ⁴⁰Ar/³⁹Ar and geochemical (major and trace element and Sr–Nd–Hf–Pb isotope) data from offshore (Chatham Rise, Campbell and Challenger Plateaus) and onland (North, South, Auckland, Campbell, Chatham and Antipodes Islands of New Zealand) volcanism on Zealandia. The samples include nephelinite, basanite through phonolite, alkali basalt through trachyte/rhyolite, and minor tholeiite and basaltic andesite, all of which have ocean island basalt (OIB)-type trace element signatures and which range in age from 64.8 to 0.17 Ma. Isotope ratios show a wide range in composition (⁸⁷Sr/⁸⁶Sr = 0.7027–0.7050, ¹⁴³Nd/¹⁴⁴Nd = 0.5128–0.5131, ¹⁷⁷Hf/¹⁷⁶Hf = 0.2829–0.2831, ²⁰⁶Pb/²⁰⁴Pb = 18.62–20.67, ²⁰⁷Pb/²⁰⁴Pb = 15.54–15.72 and ²⁰⁸Pb/²⁰⁴Pb = 38.27–40.34) with samples plotting between mid-ocean-ridge basalts (MORB) and Cretaceous New Zealand intraplate volcanic rocks.Major characteristics of Zealandia's Cenozoic volcanism include longevity, irregular distribution and lack of age progressions in the direction of plate motion, or indeed any systematic temporal or spatial geochemical variations. We believe that these characteristics can be best explained in the context of lithospheric detachment, which causes upwelling and melting of the upper asthenospheric mantle and portions of the removed lithosphere. We propose that a large-scale seismic low-velocity anomaly, that stretches from beneath West Antarctica to Zealandia at a depth of > 600 km may represent a geochemical reservoir that has been in existence since the Cretaceous, and has been supplying the upper mantle beneath Zealandia with HIMU-type plume material throughout the Cenozoic. In addition, the sources of the Cenozoic intraplate volcanism may be at least partially derived through melting of locally detached Zealandia lower lithosphere.     

The 2020 glacial lake outburst flood process chain at Lake Salkantaycocha (Cordillera Vilcabamba,Peru)

Glacial lakes represent a threat for the populations of the Andes and numerous disastrous glacial lake outburst floods (GLOFs) occurred as a result of sudden dam failures or dam overtoppings triggered by landslides such as rock/ice avalanches into the lake. This paper investigates a landslide-triggered GLOF process chain that occurred on February 23, 2020, in the Cordillera Vilcabamba in the Peruvian Andes. An initial slide at the SW slope of Nevado Salkantay evolved into a rock/ice avalanche. The frontal part of this avalanche impacted the moraine-dammed Lake Salkantaycocha, triggering a displacement wave which overtopped and surficially eroded the dam. Dam overtopping resulted in a far-reaching GLOF causing fatalities and people missing in the valley downstream. We analyze the situations before and after the event as well as the dynamics of the upper portion of the GLOF process chain, based on field investigations, remotely sensed data, meteorological data and a computer simulation with a two-phase flow model. Comparison of pre- and post-event field photographs helped us to estimate the initial landslide volume of 1–2 million m³. Meteorological data suggest rainfall and/or melting/thawing processes as possible causes of the landslide. The simulation reveals that the landslide into the lake created a displacement wave of 27 m height. The GLOF peak discharge at the dam reached almost 10,000 m³/s. However, due to the high freeboard, less than 10% of the lake volume drained, and the lake level increased by 10–15 m, since the volume of landslide material deposited in the lake (roughly 1.3 million m³) was much larger than the volume of released water (57,000 m³, according to the simulation). The model results show a good fit with the observations, including the travel time to the uppermost village. The findings of this study serve as a contribution to the understanding of landslide-triggered GLOFs in changing high-mountain regions.

     

Une dépression piézométrique naturelle en hausse au Sahel (Sud-Ouest du Niger)A rising piezometric depression in the Sahel (southwestern Niger)

In southwest Niger, the Continental Terminal water table displays a natural hollow shape about 10 m in depth over an area of 4000 km². A 10-year survey of this hollow aquifer has shown that current recharge is above $\text{20 }\text{mm}\text{yr}{}^{\mathrm{?1}}$. The water table has risen continuously since the 1950–1960s as a result of land clearance. This shows a disequilibrium in the aquifer balance. The long-term recharge rate is estimated by radioisotopes to be around $\text{1 }\text{mm}\text{yr}{}^{\mathrm{?1}}$. This figure fits with the only possible origin of the piezometric depression, i.e. evapotranspiration losses in its centre. To cite this article: G. Favreau et al., C. R. Geoscience 334 (2002) 395–401.     

Advection in plankton models with variable elemental ratios

As ocean biogeochemical models evolve to permit the elemental composition of plankton populations and dissolved organic matter to vary, each element is normally assigned a separate state variable, which is advected and mixed independently of the others. In a population of cells with varying elemental quotas, the proper currency of the advection operator is subpopulations of similar cells. The spatial gradient in total C, N, or P summed over the spectrum of such subpopulations is identical to that calculated for the population means, so treating the various elements as independent should generally be a valid approximation. However, errors can arise in high-order advection schemes with nonlinear corrector terms, which are not additive across the subpopulations. Some numerical examples indicate that these errors are relatively small [O(10⁻³–10⁻⁴)] but can be as high as O(10⁻²) in certain cases. As grid resolution varies, the error scales approximately to the Courant number.     

Gas-Phase Formation of Water-Soluble Organic Compounds in the Atmosphere: A Retrosynthetic Analysis

Atmospheric particles contain a myriad of organiccompounds, including many multifunctional,water-soluble organic compounds. Many of thesecompounds are postulated to be secondary of origin. This work investigates the possible precursors ofseveral classes of multifunctional, water-solublesecondary organic compounds by analyzing the pathwaysleading to their formation, based on known gas-phasereactions. The analysis is termed `retrosynthetic'due to the backward direction of the analysis, fromproducts to precursors. Pathways for multi-functionalcompounds were generated combinatorially, consideringthe formation of one functional group at a time.Many multifunctional organic compounds with carboxylicacid, carbonyl, and hydroxy functional groups werefound to be first- or second-generation products ofcommon anthropogenic and biogenic volatile organiccompounds such as alkanes, alkenes, aromatics, andcyclic alkenes. The estimated yields of water-solubleorganic compounds from primary precursors ranged fromless than 1% to over 10%, based on stoichiometricconsiderations. The SOA formation index, whichcombines the concepts of yields and rates, was used tocompare the feasibility of the retrosyntheticpathways. Many of the candidate pathways involve theisomerization reaction of alkoxy radicals andoxygenated intermediate products such asmonocarboxylic acids and hydroxyaldehydes.