Gold dispersion in Andean desert environments (Atacama, Chile)

The mechanism of Au dispersion in sediments of mountainous desert environments has been studied in two different areas of the Chilean Andes. The San Pedro de Cachiyuyo placer (ca. 1800 m a.s.l.) consists of alluvial fans and ephemeral stream sediments deposited in a stable piedmont. The primary Au source is Cu–Au-bearing hydrothermal tourmaline breccia. The relief of this area is low (1850 m) and the average annual rainfall is ca. 20 mm. The La Coipa area (ca. 4000 m a.s.l.) is characterized by a rugged relief. The annual average rainfall is ca. 100 mm. The sedimentary deposits are less sorted than in San Pedro de Cachiyuyo and consist of ephemeral stream alluvium and slope deposits formed in a periglacial environment. The primary Au sources are two volcanic-hosted epithermal precious metal deposits. At San Pedro de Cachiyuyo, the halo is less than 1 km in length and the highest Au concentration are observed at the break in slope between the hillside and the piedmont. In the La Coipa area, Au was only detected by chemical analysis in the 125–63 μm and <63 μm fraction; however, the dispersion halo extends for over 10 km from the source. In both cases the geochemical signal of gold was strongest in the fraction <63 μm.     

Active skarn formation beneath Lascar Volcano, northern Chile: a petrographic and geochemical study of xenoliths in eruption products

Calcsilicate xenoliths occur in large numbers in some lavas and pyroclastic flows of Lascar Volcano. Their whole-rock major element and REE compositions indicate that the protolith was the Upper Cretaceous Yacoraite Formation, which crops out extensively in NW Argentina. The whole-rock major element compositions of the xenoliths fall into specific groups suggesting a strong geochemical zonation in the skarn zone. Three geochemical zones have been identified; (1) an outer metamorphic zone rich in wollastonite; (2) a middle zone rich in pyroxene and garnet; (3) an inner zone rich in pyroxene and magnetite. The two innermost zones have developed from the wollastonite zone by infiltration of metasomatic fluids rich in Fe, Mn, Mg, Ti and Al. Whole-rock REE patterns have not changed significantly during prograde metamorphism and metasomatism, indicating REE immobility in the altering fluids. Retrograde alteration by acid-sulphate fluids produced anhydrite skarns and secondary calcite and wilkeite veins in the wollastonite zone. The carbon and oxygen isotopic compositions of this calcite indicate that it formed by Rayleigh crystallization from a low-temperature (<200 °C) fluid containing dissolved H₂CO₃. The calculated δ¹⁸O of the water in this fluid suggests a magmatic origin whereas the calculated δ¹³C of the dissolved carbonate is consistent with derivation from rocks of the Yacoraite Formation at 350 °C. It is suggested that the magmatic acid-sulphate fluid was responsible for leaching carbonate from the surrounding carbonate rocks and redepositing it in the skarn zone. REEs were mobilized during the retrograde acid-sulphate and acid-carbonate alteration. A negative Ce anomaly associated with this carbonate and sulphate indicates high oxygen fugacities in the mineralizing fluids.     

Lascar Volcano, Northern Chile; Evidence for Steady-State Disequilibrium

Lascar Volcano, located in the Central Volcanic Zone of NorthernChile, is a medium-to high-K calc-alkaline stratovolcano. Threestages are recognized in the construction of the edifice, withactivity ranging from andesitic fissure eruptions to a large-volumedacitic pumice flow, the Soncor Flow. The eruption productsare predominantly two-pyroxene andesites and dacites, with somehornblende dacites occurring in the Soncor Flow. Disequilibriumtextures resulting from mixing between mafic and acid end-membersare recognized in all the eruption products, and geochemicalmodelling has been carried out to quantify the effects of magmamixing and fractional crystallization. A model is developedwhich explains the observed textures by the existence of a continuouslyfractionating magma chamber which receives periodic influxesof basaltic andesite. Primary anhydrite has been found in theSoncor pyroclastic flow. Anhydrite formation and sulphur dioxidedegassing are explained in terms of oxidation and quenchingof sulphide-rich mafic magma during mixing with the residentmore evolved magmas.     

Lateglacial,early and middle holocene environments,human occupation,and resource use in the Atacama (Northern Chile)

Paleoenvironmental data from the Atacama Altiplano (21°–24°S) indicate that water, vegetation and animal resources were more abundant during lateglacial and early Holocene times than today. The rate of precipitation increased above 4000 m elevation to 400–500 mm/yr compared to the present 200 mm/yr. Dry conditions prevailed below 3500 m. Evidence of Paleoindian habitation is still missing, even though there is no evidence for environmental prohibitors during lateglacial time. The early Archaic hunters (10,820 yr B.P.-ca. 8500 yr B.P.) inhabited the Altiplano (high Puna) and its western slope, where water was available due to higher river runoff from the Altiplano, and the resources in different elevation zones were accessible. Natural resources decreased significantly during the middle Archaic period (8500-ca. 5000 yr B.P.). Lakes receded to today's levels, pedogenesis in the Altiplano terminated, and human activities were restricted to the most stable sites in the Río Lao and the Río Purifica catchments north of 23°S. The less stable oases south of 23°S (Salar de Atacama and Punta Negra) were abandoned. The climatic changes are best explained by shifts of the (sub)tropical circulation. © 1994 John Wiley & Sons, Inc.     

Impact area determination of pyroclastic deposits in Lascar volcano, eruptive process in the year 1993

By means of multi-temporal analysis of satellite images and statistical algorithms, the amount of pyroclastic material deposited on Lascar volcano walls after the gravitational collapse of the eruptive column caused by the eruption that occurred on April the 19th and the 20th in 1993 was determined and quantified. For this analysis, scenes corresponding to Landsat 4 TM in 1989 and Landsat 7 ETM+ in 2001 were used. By careful selection methods such as combination of bands, unsupervised classification, and the Karhunen–Loève transform, detailed analysis of zones of change that correspond to pyroclastic deposits were made. This analysis was complemented with in situ data in order to correct and calibrate the satellite images to identify zones of 1993s eruption pyroclastic deposits. Using Kittler’s and other thresholding algorithms, a search was performed for a proper threshold to binarize the images to determine the surface area covered by the eruptive process. Matlab™ software was used both for general programming and for digital image processing.     

Neogene evaporites in desert volcanic environments: Atacama Desert, northern Chile

The Upper Miocene and Pliocene evaporite deposits of the Atacama Desert of northern Chile (Hilaricos and Soledad Formations) are among the few non‐marine evaporites in which aridity not only formed the deposits, but has also preserved them almost unaltered under near‐surface conditions. These deposits are largely composed of displacive Ca sulphate and halite together with minor amounts of glauberite, thenardite and polyhalite. However, at the base and top of these deposits, there are also beds of gypsum crystal pseudomorphs that originally formed as free‐growth forms within shallow brine bodies, rather than as displacive sediments. The halite is present as interstitial cement, displacive cubes and shallow‐water, bottom‐growth chevron crusts. Most of the calcium sulphate is presently anhydrite, pseudomorphous after gypsum, that was the primary depositional sulphate mineral. The secondary anhydrite formed under early diagenetic conditions after slight burial (some metres) resulting from the effect of strongly evolved pore brines. The anhydrite has been preserved without rehydration during late diagenetic and exhumation stages on account of the arid environment of the Atacama Desert. Both the Hilaricos and the Soledad Formations contain geochemical markers indicating that these Neogene evaporites had a largely non‐marine origin. Bromine content in the halite is very low (few p.p.m.), indicating neither a sedimentological relation with sea water nor the likelihood of direct recycling of prior marine halites. Moreover, the δ³⁴S of sulphates (+4·5‰ to +9‰) also reflects a non‐marine origin, with a strong volcanic influence, although some recycling of Mesozoic marine sulphates cannot be ruled out. δ³⁴S of dissolved sulphate from hot springs and streams in the area commonly displays positive values (+2‰ to +10‰). Leaching of oxidized sulphur and chlorine compounds from volcanoes and epithermal ore bodies, very common in the associated drainage areas, have been the main contribution to the accumulation of evaporites. The sedimentary and diagenetic evolution of the Hilaricos and Soledad evaporites (based on lithofacies analysis) provides information about the palaeohydrological conditions in the Central Depression of northern Chile during the Neogene. In addition, the diagenesis and exhumation history of these evaporites confirms the persistence of strongly arid conditions from Late Miocene until the present. A final phase of tectonism took place permitting the internal drainage to change and open to the sea, resulting in dissolution and removal of a significant portion of these deposits. Despite the extensive dissolution, the remaining evaporites have undergone little late exhumational hydration.     

Colluvial sedimentation in a hyperarid setting (Atacama Desert,northern Chile): Geomorphic controls and stratigraphic facies variability

Research on colluvial depositional systems has recently emphasized periglacial and high‐altitude settings, and the relations between Quaternary slope stratigraphy and climate change. This article examines the role of variable slope morphology, surface hydrology and microclimate in controlling colluvial sedimentation along a coastal tract of the hyperarid Atacama Desert in northern Chile. Direct accessibility of active surfaces is accompanied by uninterrupted stratigraphic exposures along the base of slopes, allowing direct comparisons between surface processes and the resulting sedimentary record. Four slope sectors are identified, based on differences in morphology and processes over active surfaces. Colluvial sedimentation is controlled by complex interactions of slope gradients and profiles, exposure to dominant winds, and potential runoff pathways, which vary considerably between different sectors. Major differences are evident between these hyperarid deposits and slope sedimentation in periglacial and temperate settings, including the complete absence of pedogenic activity and clay minerals; the volume of aeolian deposits and their role in controlling processes which redistribute sediment downslope, extending colluvial aprons; and the occurrence of runoff processes only where favoured by particular topographic configurations. Depositional surfaces range from steep talus cones, to debris‐flow‐dominated and aeolian‐dominated colluvial aprons, to an aeolian ramp subject to reworking by mass flows and flash floods. Consequently, facies associations and architectures at outcrop are highly variable and highlight the importance of spatial variations in slope morphology and processes in producing distinct, coeval colluvial stratigraphies within a single environmental context. Discrepancies between active processes and the corresponding stratigraphic signatures are also evident in some sectors; for example, preservation of alluvial and aeolian facies in stratigraphic sections does not always reflect the dominant processes over active slopes. Together with the spatial variability in processes and deposits along these slopes, this suggests that caution is required when extracting palaeoenvironmental information from analyses of colluvial successions.     

Paleowetlands and regional climate change in the central Atacama Desert, northern Chile

Widespread, organic-rich diatomaceous deposits are evidence for formerly wetter times along the margins of the central Atacama Desert, one of the driest places on Earth today. We mapped and dated these paleowetland deposits at three presently waterless locations near Salar de Punta Negra (24.5°S) on the western slope of the Andes. Elevated groundwater levels supported phreatic discharge into wetlands during two periods: 15,900 to ~ 13,800 and 12,700 to ~ 9700 cal yr BP. Dense concentrations of lithic artifacts testify to the presence of paleoindians around the wetlands late in the second wet phase (11,000?-9700 cal yr BP). Water tables dropped below the surface before 15,900 and since 8100 cal yr BP, and briefly between ~ 13,800 and 12,700 cal yr BP. This temporal pattern is repeated, with some slight differences, in rodent middens from the study area, in both paleowetland and rodent midden deposits north and south of the study area, and in lake level fluctuations on the adjacent Bolivian Altiplano. The regional synchroneity of these changes points to a strengthening of the South American Monsoon — which we term the “Central Andean Pluvial Event” — in two distinct intervals (15,900-13,800 and 12,700-9700 cal yr BP), probably induced by steepened SST gradients across the tropical Pacific (i.e., La Niña-like conditions).     

Non-biological fractionation of stable Ca isotopes in soils of the Atacama Desert, Chile

We measured Ca stable isotope ratios (δ^44/40Ca) in an ancient (2 My), hyperarid soil where the primary source of mobile Ca is atmospheric deposition. Most of the Ca in the upper meter of this soil (3.5 kmol m⁻²) is present as sulfates (2.5 kmol m⁻²), and to a lesser extent carbonates (0.4 kmol m⁻²). In aqueous extracts of variably hydrated calcium sulfate minerals, δ^44/40Ca_E values (vs. bulk Earth) increase with depth (1.4 m) from a minimum of −1.91‰ to a maximum of +0.59‰. The trend in carbonate-δ^44/40Ca in the top six horizons resembles that of sulfate-δ^44/40Ca, but with values 0.1-0.6‰ higher. The range of observed Ca isotope values in this soil is about half that of δ^44/40Ca values observed on Earth. Linear correlation among δ^44/40Ca, δ³⁴S and δ¹⁸O values indicates either (a) a simultaneous change in atmospheric input values for all three elements over time, or (b) isotopic fractionation of all three elements during downward transport. We present evidence that the latter is the primary cause of the isotopic variation that we observe. Sulfate-δ³⁴S values are positively correlated with sulfate-δ¹⁸O values (R² = 0.78) and negatively correlated with sulfate δ^44/40Ca_E values (R² = 0.70). If constant fractionation and conservation of mass with downward transport are assumed, these relationships indicate a δ^44/40Ca fractionation factor of −0.4‰ in CaSO₄. The overall depth trend in Ca isotopes is reproduced by a model of isotopic fractionation during downward Ca transport that considers small and infrequent but regularly recurring rainfall events. Near surface low Ca isotope values are reproduced by a Rayleigh model derived from measured Ca concentrations and the Ca fractionation factor predicted by the relationship with S isotopes. This indicates that the primary mechanism of stable isotope fractionation in CaSO₄ is incremental and effectively irreversible removal of an isotopically enriched dissolved phase by downward transport during small rainfall events.     

Petrology of mugearite-hawaiite: Early extrusives in the 1973 Heimaey eruption,Iceland

Rocks which erupted during the first few weeks are described and chemical analyses of rocks and microprobe analyses of minerals are presented. An outline of the eruption history and of the geology of the area is given. A gradational change in chemistry from mugearitic to hawaiitic composition is recorded. Xenoliths of hypersthene gabbro not cognate with magma were discovered in the mugearite. One xenolith has reacted with the mugearite magma at depth to form pargasitic hornblende and kaersutite. It is argued that the magma reacted with the xenoliths at a total pressure well below 8–9 kbar and at a gas pressure higher 1 kbar.     

Coastal boulder deposit as evidence of an ocean-wide prehistoric tsunami originated on the Atacama Desert coast (northern Chile)

A Late Holocene cliff-top deposit of large boulders well above the limits of modern storm waves is described from the southern coast of the Atacama Desert (northern Chile). The largest moved boulder weighs >40 t and field data point to a flood height >18·5 m above high tide level and an inland penetration greater than 284 m from the cliff edge. The minimum flow velocity needed for particle entrainment was estimated as 10·1 ms⁻¹ and the most likely processes of sediment deposition for different boulders were deduced. The boulder distribution, sorting and orientation of imbricated debris, together with the significant wave height of extreme storms reported and the occurrence of interplate earthquakes in the study area indicate that the deposit records a single event, interpreted here as a tsunami wave train rather than exceptional storm waves. The boulder field was dated to between the 13th and the 16th Centuries ce and possibly correlates with the 1420 Oei orphan tsunami, that affected the eastern coast of Japan. A magnitude of 8·8 to 9·4 has been estimated for the earthquake, which may be one of the larger events of a super-cycle of earthquakes in the southern Atacama Desert. These cycle-ending earthquakes involve large rupture areas (lengths in excess of 600 km) and highly destructive ocean-wide tsunamigenic events.     

Petrology of the 1979 eruption of Soufriere volcano,St. Vincent,Lesser Antilles

Major element, trace element, and Sr isotope data are used to study the temporal variation in the chemistry of the ejecta from the 1979 eruption of Soufriere volcano, St. Vincent, and to compare the compositions of the 1979 and 1971/2 magmas. Both the 1971/2 and 1979 products were basaltic andesites almost identical in petrography. A small temporal variation in chemistry is apparent in the 1979 samples but these cannot be related to the 1971/2 lava by fractional crystallisation of phenocryst phases, and the two eruptions may therefore have sampled different batches of magma. ⁸⁷Sr/⁸⁶Sr ratios of the two magmas were identical within analytical error.Microprobe analyses of phenocryst phases and glasses from the 1979 ejecta are presented. Clinopyroxene phenocryst cores with very high Mg/Fe ratios indicate that the basaltic andesites are products of fractionation of magnesian parents. Such magmas are represented by lavas on St. Vincent similar to the microphyric alkali picrites found to the south in Grenada. A common origin for the basaltic andesites of both islands by fractional crystallisation of picritic magmas is suggested. Dacitic glass is abundant in the groundmass of scoria blocks from the eruption. It does not represent the liquid originally in equilibrium with the phenocryst phases, but rather this liquid modified by quench crystallisation. Published interpretations suggesting that dacitic glass compositions in tephra from eruptions of the Soufriere are evidence of mixed-magma eruptions are therefore rejected.     

Pleistocene recycling of copper at a porphyry system,Atacama Desert,Chile: Cu isotope evidence

We present Cu isotope data of hypogene and supergene minerals from the Late Paleocene Spence Cu-Mo porphyry in the Atacama Desert of northern Chile. Chalcopyrite displays a restricted range of δ⁶⁵Cu values within the values reported for primary porphyry Cu sulfides (+ 0.28‰ to + 0.34‰, n = 6). Supergene chalcocite samples show heavier and remarkably homogeneous δ⁶⁵Cu values, between + 3.91‰ and + 3.95‰ (n = 6), consistent with previous models of Cu leaching and enrichment in porphyry systems. Secondary Cu minerals from the oxide zone show a wider range of composition, varying from + 1.28‰ and + 1.37‰ for chrysocolla (n = 6) to very light Cu isotope signatures reported for atacamite between -5.72‰ to -6.77‰ (n = 17). These data suggest redox cycling of Cu during supergene enrichment of the Spence Cu deposit, characterized by a first stage of supergene chalcocite formation from acidic, isotopically-heavy leach fluids of meteoric origin down-flowing in a semi-arid climate (44 to ~ 15-9 Ma). Reworking of the initial supergene copper assemblage, during the Pleistocene, by rising neutral and chlorine-rich deep formation waters under well-established hyper-arid climate conditions lead to the formation of atacamite with extremely fractionated Cu compositions. Essentially coeval chrysocolla formed by dissolution of atacamite during short episodes of wetter climatic conditions occurring in the latest Pleistocene.     

天池火山三期浮岩气孔结构局域化特征与喷发动力学初探

本文在长白山天池火山野外地质工作基础上,对天池火山黄色浮岩（50 000年前大喷发）、灰白色浮岩（千年大喷发）和黑色浮岩（八卦庙期喷发）等三期空降浮岩开展了扫描电子显微镜高精度结构分析以及浮岩气孔结构的局域化定量研究。三期浮岩洛伦茨分析表明：黄色浮岩和黑色浮岩气孔均一化程度基本一致,而灰白色浮岩均一化程度更高,反映了千年大喷发在岩浆演化程度和喷发动力学行为上的特殊性,推测可能与幔源岩浆注入地壳岩浆房的动力过程有关。气孔大小幂律分布以及气孔形态分形特征的复杂性分析表明,50 000年前大喷发的浮岩幂指数和分形维数最高,揭示了其自组织临界态程度最高,对外界环境触发最为敏感,而另外两期浮岩幂指数和分形维数呈现出标准的幂律衰减,据此我们推测三期喷发可能均处在天池火山系统逐渐稳态恢复的大周期中。     

A geomorphological approach to determining the Neogene to Recent tectonic deformation in the Coastal Cordillera of northern Chile (Atacama)

The large (≈10000 km²) and local-scale (<400 km²) geomorphologic, geomorphometric and field evidence indicates that, from the mid-Miocene onwards, the Atacama Fault System (AFS) accommodated the relative uplift of the western side of the Chilean Coastal Cordillera of the Chañaral region (southern Atacama Desert). The mean regional altitude systematically decreases eastwards crossing the AFS, independent of the lithological characteristics of the substratum cut by this system of faults. Topographic analysis reveals a more incised landscape west of the AFS that, at the local scale, is reported by the distribution of the altitudes (hypsometric curves and integrals) of tributary basins and by the presence of terraces. In the Middle and Upper Miocene, a thick (>300 m) sedimentary succession was deposited east of the AFS. The succession fills previously deep paleovalleys. And it consists of gravel, so-called “Atacama Gravels”, which passes laterally into fine-grained playa related deposits near the AFS. We interpret the deposition of this succession as a result of a blocking closure of the valley flowing from the Precordillera due to the activity on AFS. A pedimentation episode followed sediment deposition and is locally strongly re-incised by the main modern-day river valleys draining the Precordillera. Incision may result from either regional uplift of the forearc, and/or from more localized activity on the AFS. Furthermore, Recent (Quaternary?) tectonic activity on the AFS has been observed which is consistent with a localized relative uplift of the crustal block west of the AFS.     

Climate variability at the southern boundaries of the Namib (southwestern Africa) and Atacama (northern Chile) coastal deserts during the last 120,000 yr

In this study, we present grain-size distributions of the terrigenous fraction of two deep-sea sediment cores from the SE Atlantic (offshore Namibia) and from the SE Pacific (offshore northern Chile), which we ‘unmix’ into subpopulations and which are interpreted as coarse eolian dust, fine eolian dust, and fluvial mud. The downcore ratios of the proportions of eolian dust and fluvial mud subsequently represent paleocontinental aridity records of southwestern Africa and northern Chile for the last 120,000 yr. The two records show a relatively wet Last Glacial Maximum (LGM) compared to a relatively dry Holocene, but different orbital variability on longer time scales. Generally, the northern Chilean aridity record shows higher-frequency changes, which are closely related to precessional variation in solar insolation, compared to the southwestern African aridity record, which shows a remarkable resemblance to the global ice-volume record. We relate the changes in continental aridity in southwestern Africa and northern Chile to changes in the latitudinal position of the moisture-bearing Southern Westerlies, potentially driven by the sea-ice extent around Antarctica and overprinted by tropical forcing in the equatorial Pacific Ocean.     

Supergene enrichment of copper deposits since the onset of modern hyperaridity in the Atacama Desert,Chile

Supergene enrichment of Cu deposits in the Atacama Desert has played a critical role in making this the prime Cu-producing province of the world. Previously, this has been believed to have occurred exclusively over a long period from the middle Eocene to the late Miocene, which ended when climatic conditions changed from arid to hyperarid. Here, we report U-series disequilibrium ages in atacamite-bearing supergene assemblages that provide a new conceptualization on both the supergene enrichment process and the onset of extreme hyperaridity in the Atacama Desert. ²³⁰Th–²³⁴U ages of gypsum intergrown with atacamite in supergene veins from Cu deposits cluster at ~240 ka (Chuquicamata), 130 ka (Mantos Blancos, Spence), and 80 ka (Mantos de la Luna, Michilla). When coupled with previous data, these results indicate that supergene enrichment of Cu deposits did not cease after the onset of hyperaridity. We propose that supergene enrichment in the Atacama region developed in two main stages. The main phase, caused by downward circulation of meteoric waters in a semi-arid setting, was active from 45 until ~9 Ma, with a last pulse ca. 5 Ma in the southern Atacama Desert. During this phase, atacamite-bearing supergene assemblages were not preserved because atacamite requires saline water for its formation and rapidly dissolves when contacted by meteoric water. This was followed by a second stage starting at ~2–1.5 Ma and continuing until at least the late Pleistocene, when deep formation waters derived from the basement passed up through and modified the pre-existing supergene Cu oxide minerals. Atacamite has then been preserved in the prevailing hyperarid climate.     

Stable isotopic composition of soil calcite (O,C) and gypsum (S) overlying Cu deposits in the Atacama Desert,Chile: Implications for mineral exploration,salt sources,and paleoenvironmental reconstruction

Soils overlying two porphyry Cu deposits (Spence, Gaby Sur) and the Pampa del Tamarugal, Atacama Desert, Northern Chile were collected in order to investigate the extent to which saline groundwaters influence “soil” chemistry in regions with thick Miocene and younger sediment cover. Soil carbonate (calcite) was analyzed for C and O isotopes and pedogenic gypsum for S isotopes. Soil calcite is present in all soils at the Spence deposit, but increases volumetrically above two fracture zones that cut the Miocene gravels, including gravels that overlie the deposit. The C isotope composition of carbonate from the soils overlying fracture zones is indistinguishable from pedogenic carbonate elsewhere at the Spence deposit; all δ¹³C_VPDB values fall within a narrow range (1.40–4.23‰), consistent with the carbonate having formed in equilibrium with atmospheric CO₂. However, δ¹⁸O_VPDB for carbonate over both fracture zones is statistically different from carbonate elsewhere (average δ¹⁸O_VPDB = 0.82‰ vs. −2.23‰, respectively), suggesting involvement of groundwater in their formation. The composition of soils at the Tamarugal anomaly has been most strongly affected by earthquake-related surface flooding and evaporation of groundwater; δ¹³C_VPDB values (−4.28‰ to −2.04‰) are interpreted to be a mixture of dissolved inorganic C (DIC) from groundwater and atmospheric CO₂. At the Spence deposit, soils only rarely contain sufficient SO₄ for S isotope analysis; the SO₄-bearing soils occur only above the fracture zones in the gravel. Results are uniform (3.7–4.9‰ δ³⁴S_CDT), which is near the middle of the range for SO₄ in groundwater (0.9–7.3‰). Sulfur in soils at the Gaby Sur deposit (3.8–6.1‰ δ³⁴S_CDT) is dominated by gypsum, which primarily occurs on the flanks and tops of hills, suggesting deposition from SO₄-rich fogs. Sulfate in Gaby Sur deposit gypsum is possibly derived by condensation of airborne SO₄ from volcanic SO₂ from the nearby Andes. At the Gaby Sur deposit and Tamarugal anomaly, pedogenic stable isotopes cannot distinguish between S from porphyry or redeposited SO₄ from interior salars.The three sites studied have had different histories of salt accumulation and display variable influence of groundwater, which is interpreted to have been forced to the surface during earthquakes. The clear accumulation of salts associated with fractures at the Spence deposit, and shifts in the isotopic composition of carbonate and sulfate in the fractures despite clear evidence of relatively recent removal of salts indicates that transfer from groundwater is an ongoing process. The interpretation that groundwaters can influence the isotopic composition of pedogenic calcrete and gypsum has important implications for previous studies that have not considered this mechanism.