Crystallization of the West Farrington Pluton, North Carolina, U.S.A.

The West Farrington pluton in the North Carolina Piedmont isconcentrically zoned from gabbro-diorite near the chilled marginsto leucogranodiorite in the center. A crystallization modelfor the West Farrington pluton has been derived utilizing chemical,petrographic, field, and experimental data. The model involvessimple in situ fractional crystallization from the margins inward,with minimal contamination, crystal settling and floating, ormetasomatism. Rocks of the pluton can be considered as mixturesof early crystallizing minerals (liquidus or near-liquidus phases)and crystallized interstitial liquids. Relative percentage ofentrapped pore liquid increased with increasing degree of crystallization. The original tonalite magma began crystallizing Fe-Ti spinels,plagioclase, and hornblende within a short temperature interval.Crystallization of these minerals controlled fractionation trends.The initial water content in the magma was probably 2–3per cent; under such conditions water saturation would havebeen reached after about 60 per cent of the magma crystallized,assuming crystallization in the lower epizone at about 2000bars total pressure.     

The Cretaceous-Tertiary boundary on the cape fear arch,North Carolina,U.S.A.

Petrologic and faunal study of a 72.5 m continuous corehole drilled in southeastern North Carolina has provided an opportunity to study a relatively uninterrupted vertical sequence across the Cretaceous-Tertiary boundary. The following stratigraphic sequence occurs; upper middle Maastrichtian Peedee Formation, −65.8 m to −51.8 m below mean sea-level (BMSL), upper middle Maastrichtian Rocky Point Member of the Peedee Formation, −51.8 m to −27.4 m BMSL, and middle to upper (?) Eocene Castle Hayne Limestone, −27.4 m to −15.2 m BMSL (base of casing).The Peedee Formation consists of moderately indurated, very fine to fine, sandy foraminiferal biomicrite and sandy biomicrite. Silt-size zoned dolomite rhombohedra form up to 30% of the upper Peedee Formation and are most abundant where bioturbation is common. A diverse and well-preserved foraminiferal fauna indicates a middle to outer continental shelf environment.The Rocky Point Member conformably overlies typical Peedee Formation lithology and consists of well-indurated sandy, fossiliferous biomicrite that grades upward into sandy, pelecypod biomicrosparite, and finally pelecypod biomicrudite. The Peedee Formation and the Rocky Point Member represent an overall shallowing-upward sequence with the upper surface forming the Cretaceous-Tertiary boundary.The Castle Hayne Limestone disconformably overlies the Rocky Point Member and consists of lithoclast-bearing, bryozoan-molluscan biomicrudite grading upward into bryozoan biomicrudite. The Castle Hayne Limestone was deposited in an open, normal salinity environment between 30 m and 100 m in water depth.     

Mercury contamination of active channel sediment and floodplain deposits from historic gold mining at Gold Hill,North Carolina,USA

A reconnaissance investigation of mercury contamination associated with historic gold mining in North Carolina, USA, revealed high concentrations of mercury in channel and floodplain sediments downstream from the Gold Hill mining district. The most intense period of mining activities in this region occurred in the 1840s and 1850s when mercury amalgamation was used to recover fine gold particles from milled ore. This paper evaluates mercury concentrations measured in active channel sediments and two cores recovered from historic floodplain deposits of the lower portion of Dutch Buffalo Creek. Mercury concentrations in these cores range from 0.01 to 2.2 mg/kg, with maximum concentrations more than 35 times background levels. A later peak in copper concentrations is associated with the operation of a large copper mine between 1899 and 1906. Following the most intense periods of mining, both mercury and copper concentrations decrease upcore to constant levels of about twice pre-mining background concentrations. Results suggest that vertical trends in mercury and other trace metals provide a useful tool for interpreting rates of historic floodplain sedimentation in the Piedmont of North Carolina.     

Mercury pollution in two gold mining areas of the Brazilian Amazon

Gold has been exploited intensively in the Brazilian Amazon during the past fifteen years using garimpo methods (small-scale gold mining). In this study, two gold mining areas were investigated, the municipalities of Pocone and Alta Floresta located in the state of Mato Grosso. Central Brazil. The elemental mercury (Hg) used in amalgamating the gold, the final stage of the ore dressing process, has caused abnormal Hg concentrations in waterways. This has occurred principally in the Amazon region, where most of the ore prospected is alluvial. Background levels of metals were determined by analyzing sediments and soils located upstream of the anthropogenic inputs and unaffected by mining activities. The study aimed to evaluate the pollution level in sediments and soils, taking into account drainage waters directly affected by gold mining. ‘Geoaccumulation indexes’ (Igeo) of Hg in sediments from both study areas were used to assesses the pollution level in the aquatic environment. The geoaccumulation indexes of Hg in sediments of the Bento Gomes River in Pocone indicate a relatively high degree of pollution at some sites, even reaching class 4 (1.85 mg/kg). However, when the river reaches the Pantanal swamp, Hg concentrations drop considerably to 0.30 mg/kg. This drop seems to be due to accumulation of metals in the sediments of a lake (sampling site PG-24), which retains most of the sediments transported by the Bento Gomes River. Accumulation of metals in the lake also occurred for Cu, Pb, Zn, Fe and Mn. In the region of Alta Floresta, total Hg concentrations in sediments of the Teles Pires River were studied in the grain size fractions < 74 μm and > 74 μm. Hg concentrations in bottom sediments of this river were higher than those found in the Pocone region, with increases of 1.5 to 30 times the background, and thus reaching an Igeo up to class 5.     

A geochemical atlas of North Carolina, USA

A geochemical atlas of North Carolina, U.S.A., was prepared using National Uranium Resource Evaluation (NURE) stream-sediment data. Before termination of the NURE program, sampling of nearly the entire state (48,666 square miles of land area) was completed and geochemical analyses were obtained.The NURE data are applicable to mineral exploration, agriculture, waste disposal siting issues, health, and environmental studies. Applications in state government include resource surveys to assist mineral exploration by identifying geochemical anomalies and areas of mineralization. Agriculture seeks to identify areas with favorable (or unfavorable) conditions for plant growth, disease, and crop productivity. Trace elements such as cobalt, copper, chromium, iron, manganese, zinc, and molybdenum must be present within narrow ranges in soils for optimum growth and productivity. Trace elements as a contributing factor to disease are of concern to health professionals. Industry can use pH and conductivity data for water samples to site facilities which require specific water quality.The North Carolina NURE database consists of stream-sediment samples, groundwater samples, and stream-water analyses. The statewide database consists of 6,744 stream-sediment sites, 5,778 groundwater sample sites, and 295 stream-water sites. Neutron activation analyses were provided for U, Br, Cl, F, Mn, Na, Al, V, Dy in groundwater and stream water, and for U, Th, Hf, Ce, Fe, Mn, Na, Sc, Ti, V, Al, Dy, Eu, La, Sm, Yb, and Lu in stream sediments. Supplemental analyses by other techniques were reported on U (extractable), Ag, As, Ba, Be, Ca, Co, Cr, Cu, K, Li, Mg, Mo, Nb, Ni, P, Pb, Se, Sn, Sr, W, Y, and Zn for 4,619 stream-sediment samples. A small subset of 334 stream samples was analyzed for gold.The goal of the atlas was to make available the statewide NURE data with minimal interpretation to enable prospective users to modify and manipulate the data for their end use. The atlas provides only very general indication of geochemical distribution patterns and should not be used for site specific studies. The atlas maps for each element were computer-generated at the state's geographic information system (Center for Geographic Information and Analysis [CGIA]). The Division of Statistics and Information Services provided input files. The maps in the atlas are point maps. Each sample is represented by a symbol generally corresponding to a quartile class. Other reports will transmit sample and analytical data for state regions. Data are tentatively planned to be available on disks in spreadsheet format for personal computers.During the second phase of this project, stream-sediment samples are being assigned to state geologic map unit names using a GIS system to determine background and anomaly values. Subsequent publications will make this geochemical data and accompanying interpretations available to a wide spectrum of interdisciplinary users.     

Geochemistry and evolution of the Wolf River Batholith,a late Precambrian Rapakivi Massif in North Wisconsin,U.S.A.

The Wolf River Batholith is an anorogenic rapakivi massif in central and northeastern Wisconsin with an age of 1.5 Ga. The Batholith has alkaline affinities and consists of biotite granite and biotite-hornblende adamellite with minor occurrences of quartz syenite and older monzonite and anorthosite. The batholith is part of a major Late Precambrian (1.4–1.5 Ga) magmatic event of continental proportions, represented by separate intrusions extending from Labrador to southern California (Silver et al., 1977).The major and trace element composition (Li, Rb, Sr, Ba, and REE) of 40 samples from the anorthosite, monzonite, and rapakivi granite and adamellite plutons precludes a comagmatic (although not cogenetic) model between all three rock units. However, the monzonite may be related to the anorthosite alone by fractional crystallization of plagioclase, orthopyroxene, clinopyroxene, and apatite. Alternatively, the monzonite may be a separate parent melt or a hybrid associated with the granite and adamellite plutons. The high REE content of the monzonite precludes it from being related to the rapakivi granite and adamellite plutons as a source material, a residuum, or a cumulate.A major portion of the Batholith is an undifferentiated intrusive sequence ranging from older rapakivi granite to younger adamellite. The compositions of these plutons suggest a crustal fusion origin at intermediate to lower levels of the crust (25–36 km). The trace element data are consistent with partial fusion of tonalitic to granodioritic source material.During crystallization and emplacement into the upper crust (less than 4 km), 55–70% fractionation of two feldspars, biotite and hornblende from one of the granite plutons produced a small volume of differentiated granitic melt high in Si, Fe/Mg, Rb, Li, and REE (except Eu), and low in Ca, Mg, Al, Ca/Na, Sr, Ba, and K/Rb and with a large negative Eu anomaly. Presumed associated cumulate material ranges from silica-poor quartz monzonite and quartz syenite.The chemical and mineralogical similarity between the Wolf River Batholith and younger magmatic analogs associated in continental break-up (Nigerian younger granites, White Mountain magma series, and the peralkaline volcanics of the Red Sea Region) are suggestive but not conclusive of an extensional tectonic setting. A preliminary tectonic model suggests that the 1.4–1.5 Ga event is in response to thermal doming in an extensional regime leading to continental separation in the western Cordillera (pre-Belt) and extensive crustal fusion with no rifting or separation across the North American Craton.     

Soil geochemistry of Mother Lode-type gold deposits in the Hodson mining district, central California, U.S.A.

The Hodson mining district is in the westernmost foothills of the Sierra Nevada in California, about 17 km west of the town of Angels Camp. This district is part of the West Gold Belt, which lies about 12–16 km west of, and generally parallel to, the better known Mother Lode Gold Belt in central California. The district produced several million dollars worth of Au between about 1890 and 1940.     

Geochemical exploration methods for gold in areas with mountain glaciation in Siberia,U.S.S.R.

In the mountainous areas of Siberia intensive destruction of ore deposits occurs as a result of Pleistocene mountain glaciation and, as a consequence, element dispersion trains form in the eroded material and in the stream sediments. In these areas, cirques and glaciated valleys are incised as much as 250–600 m into ore deposits. Geochemical anomalies in the stream sediments in the glaciated valleys form as water washes through the till and lacustrine sediments. The surface erosion processes result in the removal of fine gold from talus-covered cirque walls from where it is deposited in channels and lakes. Secondary dispersion halos may either be localized entirely within cirques, or they may be found in the drift within the outwash or glaciolacustrine environments downstream of the cirques. Intermittent accumulations of gold are specifically concentrated in glacially-fed stream sediments in those glaciated valleys characterized by smooth, flat bottoms and abundant lakes: gold in such cases tends to be concentrated in the lake muds. After the lakes have silted up, gold will be found in the fluvial deposits. In general, the complete elemental composition of the ore under discussion (Au, Ag, Pb and Zn) is reflected in only short dispersion trains within the fluvial and lacustrine deposits. Gold anomalies, on the other hand, are extensive (up to 6 km) in the stream sediments within the glaciated valleys and will occur sorbed onto hydrous iron oxides and clays, as well as in the form of flattened (thin) gold particles. In the glaciated valleys, the gold geochemical anomalies in moraine, lacustrine and stream sediments properly reflect the mineralization in the drainage basin.     

Mercury interactions in a simulated gold placer

A simulated stream sediment bed was constructed in a laboratory to determine whether dissolved Hg could be transported through sediment and deposited as amalgam on Au grains. Metallic Hg was placed in a sump at one end of a tank filled with gravel (quartz sand, granules, and pebbles), and Au grains were buried in the gravel at the other end. Water was circulated in a continuous closed loop over the Hg and through the gravel that included the Au grains for more than 850 days. The Hg content of the water increased from nil at the beginning to approximately 0.5 μg/l after approximately 22 days. The Hg content on the rims of Au grains went from nil to approximately 0.2 wt.% over 22 days reaching a maximum 0.48 wt.% Hg after 14 days. Subsequent measurements indicated a persistent decrease of the maximum Hg on Au grains to ⩽0.19 wt.% Hg at 552 days and to ⩽0.05 wt.% at 851 days. Deposition of Hg on the Au grains indicates that amalgams can form without actual contact of Hg on Au in stream sediments. Why Hg first deposited on Au and then dissolved from the Au is unknown, but a paucity of microbiota early in the experiment and subsequent development of microbiota that could facilitate dissolution of Hg is suspected. The simulated Au placer, with its coarse sediments and free water flow, is analogous to streams that have measurable (>0.2 μg/l) Hg in the water and no amalgams on Au grains within the sediments. An example of a Au mining region with similar water concentrations of Hg would be the Amazon Basin, although information on the presence of amalgam rims on Au grains is lacking, as for most regions. Lower but still measurable (⩽0.55 μg/l Hg) concentrations of Hg in stream water and a lack of amalgams on Au grains occur in Au placers near Talladega, Alabama. The opposite case would be streams with less-than-measurable (<0.2 μg/l) Hg in the water but amalgams on Au grains, where conditions are less aerated and Hg would be more likely to remain in the substrate. This situation is analogous to Au placers in the North Carolina Piedmont (South Mountains, Robbins, and High Point), where Hg is not detected in stream water (<0.2 μg/l) and Au grains possess amalgamated rims. Mercury concentrations in the air over the tank (41–465 ng/m³) varied inversely with barometric pressure (1012–1033 mb @ SL), with a positive response to light, which is consistent with the work of other researchers. The positive photo effect on Hg concentration in the air was obvious at lower barometric pressures (∼ave. 1015 mb @ SL) but subdued or nonexistent at higher barometric pressures (∼ave 1025 mb @ SL). Mercury concentrations were as much as 3 times as high during daylight hours compared to nighttime concentrations over the tank at relatively low barometric pressures. The Hg content of the water remained relatively low (<1 μg/l) through the first 200 days and then abruptly increased where it oscillated between 4 and 17 μg/l to the end of the experiment (851 days). Meanwhile, the rate of loss of Hg from the tank averaged approximately 1 μg/cm² day with a high of 1.54 μg/cm² day. Apparently the release of Hg is little affected by the Hg content of the water, as long as a minimum amount of Hg is maintained in the water. The release rate of Hg from the tank experiment is approximately 10 times higher than those reported by other workers but probably represents a maximum due to ideal, oxidizing, high water flow conditions in the tank.     

Sheet flow in a salt-marsh basin,North Inlet,South Carolina

The magnitude of sheet flow in a small South Carolina salt marsh was evaluated by comparing the storage curve for the basin based on topographic data with that based on current measurements in the main channel. The results indicate that on spring tides more than 50% of the volume of water that enters the basin enters as sheet flow over the grassy flats of the marsh. The current data also suggest that the actual storage curve may be a hysteresis loop with separate flood and ebb segments.     

Orgin of a precambrian ultramafic intrusion in southeastern Wyoming,U.S.A.

The Preacher Creek ultramafic body is a small, lenticular-shaped intrusion of peridotite exposed in an ancient (2400–2500 m.y.), regionally metamorphosed terrane in southeastern Wyoming, U.S.A. The central core of the body consists of clinopyroxene-olivine peridotite and is surrounded by a marginal rim of peridotite in which orthopyroxene and plagioclase also occur as primary minerals. Alteration of primary minerals is relatively minor. However, at the contact with the country rock progressive alteration of the peridotite to actinolite and chlorite is locally severe. Chemical study of the effect of this alteration indicates that no more than minor changes from the primary composition of the peridotite have occurred. Petrographic studies reveal accumulative textures characteristic of stratiform complexes. Cryptic layering in the body is indicated by partial chemical analyses of the two major primary minerals, olivine and clinopyroxene.The body is inferred to have formed within the earth's crust by fractional crystallization and gravity accumulation of mafic minerals from a gabbroic magma that was differentiating along a tholeiitic trend. Subsequent to (or during) crystallization the body was remobilized, folded, and emplaced in its present site as a tectonic intrusion. Ultramafic intrusions with chemical, mineralogical, and structural features similar to the Preacher Creek body may be best explained perhaps as crystalline accumulates formed in volcanic magma chambers.     

Mineral reactions in a calc-mica schist from Gassetts,Vermont, U.S.A.

A petrographic and electron microprobe study of an interbedded calc-mica schist from Gassetts, Vermont reveals the complexities of Fe-Mg crystalline solution and gradients in X _H2O/X _CO2 during regional metamorphism. The common association of microcline+diopside ± zoisite formed from biotite +calcite+quartz may be produced by continuous (Fe-Mg) exchange reactions, despite the implied metastability in the CaO-K₂O-MgO-Al₂O₃-SiO₂-H₂O-CO₂ system. Different assemblages are produced in a reaction zone between carbonate and pelite beds of varying thickness. This illustrates the importance of exhaustion of fluid buffering assemblages due to mineral reactions proceeding on a local scale. Siliceous dolomites at the same metamorphic grade have produced significantly different assemblages to the interlayered carbonatepelite sequence.     

Evidence for episodic cementation and diagenetic recording of seismic pumping events,North Coles Levee,California, U.S.A.

Bands of calcite and dolomite cements alternating with zones of nearly carbonate-free sand occur in the Stevens sandston aat North Coles Levee, San Joaquin, Valley, California. Temperatures calculated from O isotopes suggest that the calcite cement bands were emplaced episodically as a result of repeated injections of hot water from deeper in the section. Burial analysis suggests that these cements precipitated from 7 Ma to the present over the temperature range of 45 to ∼95°C.Carbon isotope data suggest that the C in the cements is a mixture derived from two sources, detrital shell material (δ¹³C(PDB)≈) and CO₂ liberated from maturing kerogen (δ¹³C ≈ −24). Plots of δ¹³C vs time and depth of crystallization show that the cementation sequence was: (1) dolomite cements, possibly concretionary, precipitated at depths <1–2 km and at temperatures <45°C; (2) calcite cements with δ¹³C(PDB) values as low as −13, crystallized from depths between 1220 and 1820 m (4000 and 6000 ft) and at temperatures between 45 and 80°C; (3) calcite cements with δ¹³C(PDB) values approaching zero and calculated temperatures of crystallization up to the present reservoir temperature of 95±3°C.A log of δ¹³C vs calculated depth of crystallization correlates with the stratigraphic column at North Coles Levee. If the correlation is valid the light δ¹³ in each cement sample can be tied to its source. A model based on this interpretation suggests that the early, light C was derived from maturing kerogen in the Kreyenhagen Formation (Eocene) as it passed through the oil window between 4 and 5 Ma. The subsequent passage of younger sediments with less organic material produced correspondingly smaller amounts of light CO₂ which was reflected in the relatively heavier C isotopes in the later cements.It is suggested that the epidsodic injections of hot water carried dissolved gases and minerals, principally calcite, upward from rocks as deep as 2–3 km below the Stevens sandstone and reprecipitated the calcite in more permeable zones in the rock. Degassing of CO₂ from rising pore waters likely triggered the precipitation and accounts for the relatively large volumes of cement. The Sibson model for seismic pumping of pore fluids is considered a likely explanation for the observed cementation.     

Kinematics of fault-related folding in a duplex,lost river range,Idaho, U.S.A.

The Doublespring duplex, located in the Lost River Range of Idaho, is a Sevier age fault-related fold complex in massive limestones of the Upper Mississippian Scott Peak Formation. Folds within the duplex closely resemble fault-bend fold geometrics, with open interlimb angles and low-angle bed cut-offs. Narrow, widely spaced, bedding-parallel shear zones with well-developed pressure solution cleavage alternate with massive, relatively undeformed layers on fold limbs. Shear zones are developed only on the limbs of anticlines, and have similar but unique morphologies in each of three different folds. Incremental strain histories reconstructed from antitaxial fibrous overgrowths and veins within the shear zones constrain the kinematics of folding. Shear zones experienced distributed bedding-parallel simple shear (flexural flow) towards pins near axial surfaces, while adjacent massive layers experienced rotation through an externally fixed extension direction. The absence of footwall synclines and morphological differences in shear zones from adjacent folds suggest that faulting preceded folding. Kinematic histories of folds that have experienced different translational histories are identical, and are not compatible with strain histories predicted from previous kinematic models of fault-bend folding. Shear zone development and fiber growth is instead interpreted to have occurred during low amplitude fixed-hinge buckling in response to initial resistance to translation of the thrust sheet. Fault-bend folding with mobile axial surfaces occurred with translation of the thrust sheets once the initial resistance to translation was overcome and resulted in no penetrative strain.     

Pedoarchaeology of Cactus Hill,a sandy Paleoindian site in southeastern Virginia,U.S.A.

Cactus Hill is located in the Virginia Coastal Plain on a terrace above the Nottoway River. The site has a record of occupation that spans the Holocene and also offers evidence of humans late in the Pleistocene before Clovis time. Soil investigations identified several deposit types, and demonstrated that multisequal eolian sands forming the site's primary core are arrayed in spatially and temporally discrete horizons. Resting atop an ancient paleosol, the earliest sand stratum (19,540 ± 70 ¹⁴C yr B.P.) is marked by a conspicuous but culturally sterile buried surface horizon. Eolian sand above this surface supports another sequum in which Clovis and underlying “Blade” artifacts are associated with a fainter surface horizon and pronounced subsoil lamellae. Early Archaic and successively younger artifacts occur above the Clovis level in a more weakly developed uppermost sequum. This soil and cultural stratigraphy, together with considerations of regional topography, demonstrate that the landscape has evolved incrementally since about the last glacial maximum. © 2004 Wiley Periodicals, Inc.     

新疆北部准噶尔-东天山地区金矿床的硫铅碳同位素地球化学及对金成矿作用的指示

以位于西准噶尔的阔尔真阔腊、布尔克斯岱、哈图、包古图金矿；东准噶尔的科克萨依、野马泉金矿；东天山的康古尔，石英滩，红石金矿为研究重点，着重探讨了S、Pb、C同位素在本区特定构造地质条件下对金成矿作用的示踪。结果表明。主成矿期矿化剂元素硫来源于深部，有少量地壳物质混染，金属元素铅主要来源于地幔及下地壳，少量浅部铅混染。反映了本区金的来源以深源为主的特征，各类金矿矿化剂元素碳为无机碳与有机碳不同程度的混合，表明地层有机质不同程度地参与了金的运移与卸载成矿，这是本区金成矿的重要特色。本区年轻的基底、深大断裂、火山活动及地层中有机质的参与等地质条件成就了本区不同类型金矿床的成矿。     

Ammonium halos in lithogeochemical exploration for gold at the Horse Canyon carbonate-hosted deposit,Nevada, U.S.A.: use and limitations

Rock samples taken from the oxidized, open pit (to depths of 55 m) of the carbonate-hosted Horse Canyon gold deposit contain no NH₄⁺. Samples from the unoxidized zone at the bottom of the pit in the “protore”, however, contain as much as 1.44% NH₄⁺ (1.12% N); the presence of NH₄⁺ was confirmed by mass spectrometry. From these results it is concluded that samples from any oxidized environment (e.g. outcrops, soils) likely will have lost all, or most, of any NH₄⁺ originaly present in minerals such as buddingtonite of NH₄⁺-containing clays. This limits the use of NH₄⁺ halos in lithogeochemical exploration programs. The most reliable application of lithogeochemical studies based on NH₄⁺ halos will likely be based on samples from drill cores and other situations where the samples have always been in a reducing environment.Analysis for NH₄⁺ by IR spectroscopy, the method preferred by most previous workers, is subject to several inherent problems, the main one being the presence of the CO₃²⁻ ion whose absorbance at about 1430 cm⁻¹ coincides with that used to determine NH₄⁺. The removal of CO₃²⁻ and other phases (e.g. organic matter) which may cause analytical difficulties by chemical treatment is not recommended because it is not possible to predict the effect of such treatment on any one of several possible minerals which may contain NH₄⁺. It is suggested that the most efficient method of analysis of rock samples for NH₄⁺ (or N which can be assumed to represent NH₄⁺) will be to analyze the samples after heating to 550°C, at which temperature all organic NH₄⁺ (and N) is removed by oxidation. Any remaining NH₄⁺ (N) determined on this residue can be assumed to originate in NH₄⁺-containing mineral structures.     

Mercury concentrations in surface waters from fluvial systems draining historical precious metals mining areas ink southeastern U.S.A.

This study evaluates several southern Appalachian Piedmont mining districts for Hg contamination in surface waters and determines potential relationships between Hg discharged from historical mining operations and site-specific physical factors. Water samples were collected from 3 fluvial systems that drain areas where Hg was used to amalgamate Au from ore during the 19th century. Each of the fluvial systems exhibit similar physical characteristics such as climate, vegetation, and rock type. Total Hg (Hg_T) determinations were made using cold vapour atomic fluorescence spectroscopy techniques. Concentrations of Hg_T in the southern Appalachian Piedmont range from 1–3 ng l⁻¹ in waters of the Arbacoochee, Alabama, and South Mountains, North Carolina, Mining Districts to 13 ng l⁻¹ in waters of the Dahlonega Mining District in Georgia. The correlation between Hg_T and total suspended solids (TSS) at the southern Appalachian sites was good with a coefficient of determination (r²) of 0.82. A clear trend between environmentally-available Fe (Fe_E) and Hg_T (r²=0.86) was also evident. The correlation between Hg_T and Fe_E most likely reflects similarities in the mechanisms that control the aqueous concentrations of both metals (i.e., the particle-reactive nature of the two elements), allowing for the sorption of Hg onto Fe-oxyhydroxides. Hence, increased loads of TSS from erosional events are probably responsible for higher stream water Hg_T concentrations. Vegetation at these sites, which is heavy due to the warm, humid climate of the SE, may help reduce the total amount of Hg released from contaminated mining sites to the rivers by controlling erosion, hence, decreasing the input of contaminated particles into streams and rivers.These southern Appalachian mining sites used Hg amalgamation techniques similar to those used in other precious metals mining districts, such as the highly contaminated Comstock Au–Ag district in Nevada, yet Hg_T concentrations are orders of magnitude lower; This difference in concentration between the southern Appalachian districts and the Comstock district may correlate to the relative amounts of Hg that were used in each. However, other variables were evaluated to determine if physio-chemical differences such as climate could influence Hg_T concentrations in surface waters of the two areas.     

Iron deposition as acidic groundwater encounters carbonates in the alluvium of Pinal Creek,Arizona, U.S.A.

In a column experiment, acidic groundwater from Pinal Creek Arizona, a Cu mining area, was eluted through a composited alluvial sample obtained from a core that had been removed from a well downgradient of the acidic groundwater. The minerals present in typical grains and flakes in the alluvium before and after the elution were determined by X-ray diffraction (XRD), scanning electron microscopy, and energy dispersive multichannel analyses (EDX). The concentrations of Fe, Ti, Mn, Si, Al, Na, Ca, K, Mg and S in these grains and flakes and in their microcrystalline surface coatings were measured by EDX.In addition to magnetite, hematite, and Fe-Ti oxides, Fe was most concentrated in micas (especially biotite-like flakes) and in the microcrystalline coatings. The measured elements in these microcrystalline coatings were primarily K, Fe, Al, and Si. The microcrystalline coatings on the mica flakes also contained Mg. The approximate l:3 Mg:Si atomic ratios (ARs)_of the biotite-life flake both before and after the elution would suggest that the Fe deposited during the elution had not substituted for Mg in these flakes. As a result of the elution, assuming no loss of Si, the averaged recorded Fe:Si AR of the microcrystalline coatings increased from (0.46 to 0.58):3.00.Iron deposition on the typical grains and flakes may relate to the presence of Fe in the particle on which it is deposited or to the presence of Fe in the microcrystalline surface coatings before elution. The data here are not sufficient for a statistical evaluation, but elution caused the following trends: (l) The Fe:Si AR increased in the (K,Fe,Al,Si)-microcrystalline surface coatings; (2) For the mica flakes, there was more than a 2-fold increase in the Fe:Si AR for the microcrystalline surface coatings of the Fe-rich biotite-like flakes but no measurable increase of the Fe:Si AR for the microcrystalline surface coatings of the muscovite-like flakes that contained 3–5 times less Fe; (3) Also for the biotite-like flakes, the increase in Fe:Si AR was greater in the flakes that had a higher Fe:Si AR; (4) The Fe deposition on the Fe-rich microcrystalline surface coatings of the feldspar was much greater than on the Fe-poor, beige quartz and feldspar grains that, prior to elution, had only CaS0₄ microcrystalline coatings; and (5) No Fe was deposited on Fe-poor grains with no microcrystalline surface coating.     

Balance of S in a constructed wetland built to treat acid mine drainage, Idaho Springs, Colorado, U.S.A.

The wetland constructed at the Big Five Tunnel in Idaho Springs, Colorado was designed to remove, passively, heavy metals from acid mine drainage. In optimizing the design of such a wetland, an improved understanding of the chemical processes operating there was required, particularly SO₄²⁻ reduction and sulfide precipitation. For this purpose, field and laboratory data were collected to study the balance of S in the system. Field data collected included water analyses of the mine drainage and wetland effluents and measurements of H₂S gas emissions from the wetland. The concentration of sulfide in the wetland effluent ranged from 10⁻⁴ to 10⁻³ mol/l. The average rates of H₂S emission from the surface of the substrate were 150 nmol/cm²/d in the summer and 0.17 and 0.35 nmol/cm²/d in the winter. This maximum estimated loss of sulfide was not significant in reducing the amount of sulfide available for precipitation with metals. Sequential extraction experiments for S on wetland substrates showed that acid volatile sulfides (AVS) increased with time in the wetland substrate. A serum bottle experiment was conducted to study the S balance in the Big Five wetland by quantitatively measuring the amount of S in different phases as microbial SO₄²⁻ reduction progressed. The increase in AVS reasonably balanced the decrease in SO₄²⁻ concentration in the experiment, suggesting that the decrease in SO₄²⁻ concentration represented the amount of SO₄²⁻ reduced and that nearly all of the sulfide produced was precipitated as AVS. Sulfide precipitation was determined to be the primary metal removal process in the wetland system and amorphous FeS is the primary iron sulfide formed in the substrate.