Applied Geochemistry (v.20, #1)

In the northern part of the Indian sub-continent, the Ganga alluvial plain (GAP) feeds its weathering products to the Ganga–Brahmaputra River system, one of the world’s largest fluvial systems. The authors present a geochemical study of the GAP weathering products transported by the Gomati River (the Ganga River tributary) to understand weathering processes of an alluvial plain in a humid sub-tropical climate. A total of 28 sediment samples were collected during the monsoon season and were analysed by X-ray fluorescence spectrometry for 25 major and trace elements. Bulk chemistry of the channel, flood and suspended sediments mostly consists (>90%, >80% and >75%, respectively) of three elements; Al, Si and Fe. Major element concentrations normalised with respect to upper continental crust (UCC) show strong depletion of highly mobile elements (Na, Ca) and enrichment of immobile elements (Ti, Si). Silica enrichment in the sand fraction is probably caused by chemical weathering of feldspar. Mineral sorting during fluvial transportation acts as the single important factor that controls the geochemistry of these weathering products and also strongly influences major and trace element distribution in the individual sediment samples. Trace element (Ba, Cr, Cu, Nb, Ni, Pb, V and Zn) concentrations were strongly correlated with major element (Si, Al, Fe, Mn and K) concentrations indicating that the abundance of trace elements is controlled by the same processes that control the major element distribution in these sediments.The GAP weathering products were geochemically distinguished as arkose to litharenite in rock classification. Chemical mobility, normalised with respect to TiO2 in UCC, indicates that Si, Na, Zr, Ba and Sr, mainly derived from feldspar, muscovite and biotite, are lost during weathering. Iron and Zn remained immobile during weathering and were strongly adsorbed by phyllosilicates and concentrated in fine-grained sediment fractions. The chemical index of alteration indicates that the GAP has experienced chemical weathering of incipient to moderate intensity. The GAP weathering products also demonstrated a progressive incomplete alteration in the alluvial sequence made-up of the Himalayan-derived sediments. A model has been proposed to better understand weathering processes and products of the GAP in temporary storage of ∼50 ka in a humid sub-tropical climate.

Hydro-geochemical and isotopic fluid evolution of the Los Azufres geothermal field, Central Mexico by E. González-Partida; A. Carrillo-Chávez; G. Levresse; E. Tello-Hinojosa; S. Venegas-Salgado; G. Ramirez-Silva; M. Pal-Verma; J. Tritlla; A. Camprubi (23-39).
Hydrothermal alteration at Los Azufres geothermal field is mostly propylitic with a progressive dehydration with depth and temperature increase. Argillic and advanced argillic zones overlie the propylitic zone owing to the activity of gases in the system. The deepest fluid inclusions (proto-fluid) are liquid-rich with low salinity, with NaCl dominant fluid type and ice melting temperatures (T mi) near zero (0 °C), and salinities of 0.8 wt% NaCl equivalent. The homogenization temperature (T h)  = 325 ± 5 °C. The boiling zone shows T h  = ±300 °C and apparent salinities between 1 and 4.9 wt% NaCl equivalent, implying a vaporization process and a very important participation of non-condensable gases (NCGs), mostly CO2. Positive clathrate melting temperatures (fusion) with T h  = 150 °C are observed in the upper part of the geothermal reservoir (from 0 to 700 m depth). These could well be the evidence of a high gas concentration. The current water produced at the geothermal wells is NaCl rich (geothermal brine) and is fully equilibrated with the host rock at temperatures between T  = 300 and 340 °C. The hot spring waters are acid-sulfate, indicating that they are derived from meteoric water heated by geothermal steam. The NCGs related to the steam dominant zone are composed mostly of CO2 (80–98% of all the gases). The gases represent between 2 and 9 wt% of the total mass of the fluid of the reservoir.The authors interpret the evolution of this system as deep liquid water boiling when ascending through fractures connected to the surface. Boiling is caused by a drop of pressure, which favors an increase in the steam phase within the brine ascending towards the surface. During this ascent, the fluid becomes steam-dominant in the shallowest zone, and mixes with meteoric water in perched aquifers. Stable isotope compositions (δ 18O–δD) of the geothermal brine indicate mixing between meteoric water and a minor magmatic component. The enrichment in δ 18O is due to the rock–water interaction at relatively high temperatures. δ 13C stable isotope data show a magmatic source with a minor meteoric contribution for CO2. The initial isotopic value δ 34SRES  = −2.3‰, which implies a magmatic source. More negative values are observed for shallow pyrite and range from δ 34S (FeS2) = −4‰ to −4.9‰, indicating boiling. The same fractionation tendencies are observed for fluids in the reservoir from results for δ 18O.

The Pisa plain contains a multilayered confined aquifer made up of Pleistocene sands and gravels. The groundwater from the wells tapping these horizons are generally of poor quality: they exhibit significant TDS, relatively high Cl content and considerable hardness. During geothermal prospecting of the Pisa plain, about 80 wells ranging in depth from 20 to 250 m were sampled, and both chemical (major ions) and isotope analyses were conducted. The data collected show that TDS is strongly influenced by HCO3 and Cl, and that a 3-component mixing process affects the groundwater’s chemical composition. The end members of this mixing process have been identified as: (a) diluted HCO3 meteoric water, which enters the plain mainly from the eastern and northern sides of the study area; (b) Cl-rich water, which largely characterizes the shallow sandy horizons of the multilayered aquifer system and has been attributed to the presence of seawater, as also suggested by δ 18O data; and (c) SO4-rich groundwater, which is linked to the hot groundwater circulation within Mesozoic carbonate formations and, at first sight, seemed to affect only the gravelly aquifer. A SO4-rich water also contributes to the sandy aquifer; it probably enters the plain both laterally, from the margins of the Pisan Mountains and from depth, but promptly undergoes substantial SO4 reduction processes by bacteria. That such processes are at work is suggested both by the low SO4 and high HCO3 concentrations found in the well waters and by their C and S isotope compositions. The collected data have allowed zones with higher quality waters to be identified, which may someday be used for the local water supply.

Arsenic and other drinking water quality issues, Muzaffargarh District, Pakistan by R.T. Nickson; J.M. McArthur; B. Shrestha; T.O. Kyaw-Myint; D. Lowry (55-68).
In 49 samples of groundwater, sampled in Muzaffargarh District of south-western Punjab, central Pakistan, concentrations of As exceeded the World Health Organisation provisional guideline value, and United States Environmental Protection Agency (USEPA) Maximum Contaminant Level (MCL), of 10 μg L−1 in 58% of samples and reached up to 906 μg L−1. In this semi-arid region canal irrigation has lead to widespread water-logging, and evaporative concentration of salts has the potential to raise As concentrations in shallow groundwater well above 10 μg L−1. In fact, in rural areas, concentrations stay below 25 μg L−1 because As in the oxic shallow groundwater, and in recharging water, is sorbed to aquifer sediments. In some urban areas, however, shallow groundwater is found to contain elevated levels of As. The spatial distribution of As-rich shallow groundwater indicates either direct contamination with industrial or agricultural chemicals, or some other anthropogenic influence. Geochemical evidence suggests that pollutant organics from unconfined sewage and other sources drives reduction of hydrous ferric oxide (HFO) releasing sorbed As to shallow groundwater. The situation is slightly less clear for seven wells sampled which tap deeper groundwater, all of which were found with >50 μg L−1 As. Here As concentrations seem to increase with depth and differing geochemical signatures are seen, suggesting that As concentrations in older groundwater may be governed by different processes. Other data on parameters of potential concern in drinking water are discussed briefly at the end of the paper.

Surface sediments of the NW Aegean Sea were analyzed for clay mineral composition (94 samples), organic C and carbonate content, and major and minor elements (112 samples). Illite is the most abundant clay mineral, followed by smectite, chlorite and kaolinite. Clay minerals are preferentially deposited within the continental shelf and slope environment, due to the prevailing cyclonic circulation and the formation of flocs in the rivers’ pro-delta areas. The spatial distribution of major lithogenic elements (Si, Al, Ti, Fe) is substantially biased by the presence of relict sands located in the central-eastern part of the Thermaikos Gulf continental shelf. Biogenic elements (Ca, Sr) are noticeably pronounced on the northern flank of the North Sporades Islands. Organic C content and minor elements Cu, Zn, As and Pb are enriched markedly over the Thessaloniki Bay and Gulf suggesting substantial human impact, deriving largely from untreated or partly treated domestic and industrial effluents. The terrigenous component of the elements is preferentially contributed from the Axios River (Zn and Pb), the Aliakmon River (Cr, Co, Ni, Cu and As), and to a lesser extent from the Pinios River (V, Co, Ni and Cu); a considerable part of the metals originates in ultra-mafic and volcanic formations, which are abundant in the rivers’ catchment basins. Carbonate content exhibits great variability, with the highest values observed in biogenic sediments of the Sporades Basin. In the latter area, Mn (and Fe) oxides and hydroxides were found as coatings on calcite. The coatings were enriched in Cr, Co, Ni, Zn, As, Mo and Pb, due to adsorption or ion-exchange during early diagenesis. The combined influence of the river systems, the Thessaloniki metropolis, and the diverse sedimentological background, generates great patchiness in the distribution of major and minor elements.

Alteration of the Callovo–Oxfordian clay from Meuse-Haute Marne underground laboratory (France) by alkaline solution. I. A XRD and CEC study by S. Ramı́rez; P. Vieillard; A. Bouchet; A. Cassagnabère; A. Meunier; E. Jacquot (89-99).
The reaction of the clay fraction of the Callovo–Oxfordian hard shale formation hosting the French underground laboratory site, with high pH NaOH, KOH and Ca(OH)2 solutions has been investigated through closed system experiments at 60, 90 and 120 °C over 6, 24 and 168 h. The mineralogical composition of the run samples has been determined using X-ray diffraction (XRD) of randomly oriented powders showing the formation of different species of zeolites (analcime, chabazite, phillipsite) and Ca silicates (tobermorite, katoite). The phyllosilicates were studied using XRD of oriented preparations and cation exchange capacity measurements. Detrital or diagenetic mica and chlorite in the <2 μm fraction remain unchanged. On the contrary, the smectite and random illite–smectite mixed layer minerals are strongly reactive. The expandable layers of montmorillonite type are selectively dissolved while beidellitic ones survive or are transitionally formed.

Growth and Cu accumulation by plants grown on Cu containing mine tailings in Cyprus by Lisa Johansson; Constantinos Xydas; Nikos Messios; Eva Stoltz; Maria Greger (101-107).
The Skouriotissa Cu mine in the northern part of Cyprus has produced large amounts of mine waste. Phytoremediation could stabilise the erosion or extract the metals of this waste. The aim of this study was to find out if Pistacia terebinthus, Cistus creticus, Pinus brutia and/or Bosea cypria could grow and tolerate or maybe accumulate Cu from the mine waste containing up to 787 mg Cu (kg DW)−1. Another aim was to see if the liquid wine waste product Vinassa, containing organic acids and having a low pH, or chicken fertilizer could improve plant growth and/or Cu accumulation. The four species were planted at the mine waste site untreated or with the addition of Vinassa or chicken fertilizer as mine waste modifiers. After 3 months, shoot length growth was measured and the plants were analysed for Cu concentration. The pH and Cu concentration of the mine waste mixture in the different treatments was also measured. To find out if plants accumulated Cu to the highest extent in roots or shoot, a greenhouse study was undertaken where the plant species were cultivated for 3 weeks in Cu spiked soil. The results showed that all of the tested species survived and grew on the mine waste site, which indicates that they tolerate the high level of Cu at the mine waste site. The leaves of C. creticus had the highest Cu accumulation of all tested species. Copper accumulation varied with plant species. They seemed to have different distribution strategies for Cu: in Pistacia terebinthus and C. creticus most of the Cu was found in the roots, while B. cypria accumulated most of the Cu in the leaves. Addition of Vinassa and chicken fertilizer did not increase plant growth or Cu accumulation, but did affect the Cu distribution in B. cypria.

A previously reported methodology applied to evaluate the factors affecting the reactivity pyrite and pyritic samples under simulated weathering conditions is now utilized to analyze pyrrhotite reactivity at different environmental alteration stages. The methodology is based on the comparison of the voltammetric responses of the mineral obtained through the alteration process; changes in the electrochemical behavior are associated with changes in the surface state of the mineral and finally are related with changes of the mineral reactivity. The results of an initial study of the electrochemical behavior of pyrrothite before alteration suggest that its alteration involves the formation of 3 surface layers (in agreement with previous reports): (1) in immediate contact with pyrrhotite corresponding to a metal-deficient sulfide; (2) an intermediate layer corresponding to elemental S, and; (3) the most external layer, consisting of precipitates of Fe oxy-hydroxides, like goethite. The pyrrhotite reactivity seems to be controlled by the formation of oxidation product layers that coat and passivate the pyrrhotite surface, where the elemental S layer has most significance. The results confirm the advantage of incorporating cyclic voltammetry as an auxiliary method for acid rock drainage prediction, due to its demonstrated capacity to describe the factors that influence sulfide mineral reactivity which are not evaluated by other predictive techniques.

Controls on roxarsone transport in agricultural watersheds by B.L. Brown; A.D. Slaughter; M.E. Schreiber (123-133).
The use of the organoarsenical roxarsone, added to poultry feed to increase weight gain, results in elevated As concentrations (10–50 mg/kg) in poultry litter. This litter is used extensively as fertilizer in agricultural regions. The authors investigated the sources and sinks of As within the vadose zone of an agricultural watershed in the Shenandoah Valley of Virginia, USA, an area of intense poultry production. Batch experiments were constructed to examine adsorption and biotransformation characteristics of roxarsone within the Ap and Bt soil horizons of Frederick series soils, common in the Shenandoah Valley. Roxarsone exhibits weak adsorption to the Ap soils; however, it is rapidly biotransformed to As(V) in this soil horizon. Although the Bt horizon demonstrated strong adsorption of roxarsone and thus may act as a sink for As species, soil water data collected from lysimeters at an agricultural field site suggest that As, as As(V), is mobile in the Bt soil water. It is unclear if this mobilization is due to competitive reactions with phosphate or organic acids, also present in litter. These results have implications for As cycling within poultry-dominated watersheds. For watersheds that have experienced years of litter application, As and other litter-associated species will be attenuated in soils through adsorption to mineral surfaces, but a variety of geochemical processes, such as competitive adsorption, may allow for enhanced transport of As through the vadose zone and into aquifer systems.

Pit lakes that result from open pit mining are potential water resources or potential environmental problems, depending on lake water quality. Wall rock mineralogy can affect lake chemistry if surface water inputs and/or groundwater inputs and/or lake water in contact with submerged wall rocks react with the wall rock minerals. This study presents a mineral quantification method to measure the distribution and concentration of wall rock minerals in open pit mines, and applies the method to the Martha epithermal Au–Ag mine, Waihi, New Zealand. Heterogeneous ore deposits, like Martha, require a large number of wall rock samples to accurately define mineral distributions. X-ray diffraction analyses of 125 wall rock samples identified the most abundant minerals in the wall rocks as quartz, adularia, albite, illite, chlorite, kaolinite, pyrite and calcite. Distribution maps of these minerals defined 8 relatively homogenous areas of wall rock referred to as “mineral associations”: weakly-altered, propylitic, fresh-argillic, weathered-argillic, oxidized, potassic, quartz veins, and post-mineralization deposits. X-ray fluorescence, Leco furnace, and neutron activation analyses of 46 representative samples produced the geochemical dataset used to assign quantities of elements to observed minerals, and to calculate average mineral concentrations in each association. Thin-section petrography and calcite concentrations from Sobek acid-digestions confirm the calculated mineralogy, providing validation for the method. Calcite and pyrite concentrations allowed advanced acid–base accounting for each mineral association, identifying 3 potential acid-producing associations and one potential acid-neutralizing association. The results target areas, where detailed hydrologic and kinetic tests would be valuable in the next stage of pit lake evaluation. Detailed understanding of wall rock mineralogy will help strengthen predictions of pit lake water quality.

In monitoring a minor geochemical element in groundwater or soils, a background population of values below the instrumental detection limit is frequently present. When those values are found in the monitoring process, they are assigned to the detection limit which, in some cases, generates a probability mass in the probability density function of the variable at that value (the minimum value that can be detected). Such background values could distort both the estimation of the variable at nonsampled locations and the inference of the spatial structure of variability of the variable. Two important problems are the delineation of areas where the variable is above the detection limit and the estimation of the magnitude of the variables inside those areas. The importance of these issues in geochemical prospecting or in environmental sciences, in general related with contamination and environmental monitoring, is obvious. In this paper the authors describe the two-step procedure of indicator kriging and ordinary kriging and compare it with empirical maximum likelihood kriging. The first approach consists of using a binary indicator variable for estimating the probability of a location being above the detection limit, plus ordinary kriging conditional to the location being above the detection limit. An estimation variance, however, is not available for that estimator. Empirical maximum likelihood kriging, which was designed to deal with skew distributions, can also deal with an atom at the origin of the distribution. The method uses a Bayesian approach to kriging and gives intermittency in the form of a probability map, its estimates providing a realistic assessment of their estimation variance. The pros and cons of each method are discussed and illustrated using a large dataset of As concentration in groundwater. The results of the two methods are compared by cross-validation.

A sampling-separation method and a dynamic monitoring method were used to investigate the time-dependent reactions of H+ ions with two contrasting types of soil, variable charge soils (VCS) and constant charge soils (CCS), by directly evaluating H+ ion consumption and other relevant consequences. The results for both CCS and VCS show that H+ ion consumption, increase in positive surface charge and increase in soluble Al are all characterized by a rapid step followed by a slow one. The higher the content of free Fe oxides in the soil, the larger the increase in positive surface charge and in H+ ion consumption in the initial rapid step. This is due mainly to protonation on external surfaces. The gradual increase in positive surface charge in the slow step for the 3 VCSs is a result of H+ ion diffusion to the reactive sites of Fe–OH on internal surfaces. The very low content of free Fe oxides on internal surfaces of the 2 CCSs render a negligible increase in positive surface charge in the slow step. For the 3 VCSs, the gradual consumption of H+ ions in the slow process is the result of protonation, Al dissolution and/or transformation into exchangeable acidity. For the 2 CCSs, however, the gradual consumption is mainly the result of Al dissolution and/or transformation into exchangeable acidity. The time-dependent Al dissolution from both VCS and CCS is influenced by several factors such as mineral components, solubility and dissolution rates of the soils, and H+ ion concentration in soil suspensions.

Schwertmannite precipitated from acid mine drainage: phase transformation, sulphate release and surface properties by Jörgen Jönsson; Per Persson; Staffan Sjöberg; Lars Lövgren (179-191).
Schwertmannite precipitated from acid mine drainage at the Kristineberg Zn–Cu mine in northern Sweden has been characterised regarding elemental composition, phase transformation as a function of pH and time, SO 4 2 - release and speciation of SO 4 2 - associated with the solid. The elemental analysis gave the composition Fe8O8(OH)5.02(SO4)1.49  · 0.5H2O where approximately 1/3 of the SO 4 2 - is adsorbed to the surface. The conversion of schwertmannite to goethite at pH 9 was complete within 187 days; at pH 6, the conversion was still incomplete after 514 days. Lower pH and relatively high SO 4 2 - concentration decreased the conversion even further. Also temperature was shown to be an important parameter for this process and low temperature (+4 °C) effectively stopped the transformation at pH 3. The release of SO 4 2 - was linear with pH and X-ray photoelectron spectroscopy measurements confirming that the surface bound SO 4 2 - was released before bulk SO 4 2 - . Zeta potential measurements indicate a pHIEP of 7.2 for the schwertmannite sample. Prior to conversion into goethite, the SO 4 2 - associated with schwertmannite was indicated by attenuated total reflectance FTIR spectroscopy to be present both as bulk and surface species. Furthermore, the speciation of surface SO 4 2 - was shown to vary with pH and two predominating species were detected. As pH increases, SO 4 2 - is increasingly coordinated in an outer sphere mode whereas a stronger, possibly inner sphere, complex dominates at low pH.

Sorption and precipitation of Co(II) in Hanford sediments and alkaline aluminate solutions by Jeffrey G. Catalano; Jeffrey A. Warner; Gordon E. Brown (193-205).
Sorption and precipitation of Co(II) in simplified model systems related to the Hanford site high-level nuclear waste tank leakage were investigated through solution studies, geochemical modeling, and X-ray absorption fine structure (XAFS) spectroscopy. Studies of Co(II) sorption to pristine Hanford sediments (ERDF and Sub), which consist predominantly of quartz, plagioclase, and alkali feldspar, show an adsorption edge centered at pH ≈ 8.0 for both sediments studied, with sorption >99% above pH ≈ 9.0. Aqueous SiO2 resulting from dissolution of the sediments increased in concentration with increasing pH, though the systems remained undersaturated with respect to quartz. XAFS studies of Co(II) sorption to both sediment samples reveal the oxidation of Co(II) to Co(III), likely by dissolved O2, although this oxidation was incomplete in the Sub sediment samples. The authors propose that Fe(II) species, either in aqueous solution or at mineral surfaces, partially inhibited Co(II) oxidation in the Sub sediment samples, as these sediments contain significantly higher quantities of Fe(II)-bearing minerals which likely partially dissolved under the high-pH solution conditions. In alkaline solutions, Al precipitated as bayerite, gibbsite, or a mixture of the two at pH > 7; an amorphous gel formed at pH values less than 7. Aqueous Co concentrations were well below the solubility of known Co-bearing phases at low pH, suggesting that Co was removed from solution through an adsorption mechanism. At higher pH values, Co concentrations closely matched the solubility of a Co-bearing hydrotalcite-like solid. XAFS spectra of Co(II) sorbed to Al-hydroxide precipitates are similar to previously reported spectra for such hydrotalcite-like phases. The precipitation processes observed in this study can significantly reduce the environmental hazard posed by 60Co in the environment.

Historical deposition and fluxes of mercury in Narraguinnep Reservoir, southwestern Colorado, USA by John E. Gray; David L. Fey; Charles W. Holmes; Brenda K. Lasorsa (207-220).
Narraguinnep Reservoir has been identified as containing fish with elevated Hg concentrations and has been posted with an advisory recommending against consumption of fish. There are presently no point sources of significant Hg contamination to this reservoir or its supply waters. To evaluate potential historical Hg sources and deposition of Hg to Narraguinnep Reservoir, the authors measured Hg concentrations in sediment cores collected from this reservoir. The cores were dated by the 137Cs method and these dates were further refined by relating water supply basin hydrological records with core sedimentology. Rates of historical Hg flux were calculated (ng/cm2/a) based on the Hg concentrations in the cores, sediment bulk densities, and sedimentation rates. The flux of Hg found in Narraguinnep Reservoir increased by approximately a factor of 2 after about 1970. The 3 most likely sources of Hg to Narraguinnep Reservoir are surrounding bedrocks, upstream inactive Au–Ag mines, and several coal-fired electric power plants in the Four Corners region. Patterns of Hg flux do not support dominant Hg derivation from surrounding bedrocks or upstream mining sources. There are 14 coal-fired power plants within 320 km of Narraguinnep Reservoir that produce over 80 × 106 MWH of power and about 1640 kg-Hg/a are released through stack emissions, contributing significant Hg to the surrounding environment. Two of the largest power plants, located within 80 km of the reservoir, emit about 950 kg-Hg/a. Spatial and temporal patterns of Hg fluxes for sediment cores collected from Narraguinnep Reservoir suggest that the most likely source of Hg to this reservoir is from atmospheric emissions from the coal-fired electric power plants, the largest of which began operation in this region in the late-1960s and early 1970s.