George H. Brimhall

Constitutive mass balance relations between chemical composition, volume, density, porosity, and strain in metasomatic hydrochemical systems: Results on weathering and pedogenesis

Relations characterizing the chemical, physical, and mechanical changes resulting from metasomatic hydrochemical processes are developed using mass balance models which formally link chemical composition to bulk density, mineral density, volumetric properties, porosity, and amount of deformation (strain). Rigorous analysis of aqueous solute transport effects is then made possible in a variety of porous media flow environments including chemical weathering, pedogenesis (soil formation), diagenesis, ore deposition and enrichment, and metamorphism. Application of these linear constitutive relations to chemical weathering profiles shows that immobile and locally mobile chemical elements, with masses conserved on the scale of soil profiles, can be accurately identified from analysis of appropriate data arrays and then used as natural geochemical tracers to infer the nature and extent of hydrochemical weathering processes and volume changes during pedogenesis. Assumptions commonly made in the past about the supposed immobility of certain elements, e.g., Ti and Zr, become unnecessary. Quantitative differentiation between the effects of residual and supergene fractionation is then easily made. These methods are applied to Ni-rich laterites developed by weathering of ultramafic rocks, showing that during ordinary residual enrichment, Ni is concentrated by as much as 4× protolith peridotite concentrations. This occurs simply by silicate mineral dissolution and removal of chemical elements other than Ni (e.g., Mg) with a corresponding reduction in saprolite density and increase in bulk porosity without significant deformation. In contrast, laterites with mineable concentrations of Ni which are similarly undeformed (such as the Nickel Mountain Mine in Riddle, Oregon) have experienced, in addition to residual enrichment, strong supergene enrichment by fractionation of ore elements between a leached zone from which Ni is extracted and a complementary enriched zone positioned farther along the direction of ground water flow. Soil-forming processes in podzol chronosequences developed on sandy beach terraces of the Mendocino Coast of California involved soil column collapse of 60 percent by dissolution of silicate minerals in the albic horizon of Al and Fe leaching, and 70 percent dilation (expansion) in the overlying organic-rich layer by root growth. The amount of erosion based upon paleosurface reconstructions using the excess mass of Fe, Al, Pb, Ga, and Cu in the zone of supergene enrichment (spodic horizon) below the ground water table indicates that subsurface erosion by dissolutional collapse is three times that of surficial erosion. Finally, using published chemical data for Ti, Zr, and Cr on major bauxite deposits in Australia where erosion rates are thought to be low, we infer that there may have been major amounts of dissolutional collapse to explain the upwards increase of detrital zircon and rutile in weathering profiles.

Middle Miocene climatic change in the Atacama Desert, northern Chile: Evidence from supergene mineralization at La Escondida

Geochronology and paleotopographic reconstruction of the porphyry-copper deposit at La Escondida, Chile, are used to calculate long-term erosion rates and to deduce the timing of Tertiary climatic change for a portion of the Atacama Desert region. Hypogene hydrothermal alteration and protore mineralization at La Escondida took place between 33.7 ± 1.4 and 31.0 ± 1.4 Ma based on K-Ar dating of hydrothermal biotite and sericite. Supergene weathering and copper-sulfide en-richment processes were active from 18.0 ± 0.7 to 14.7 ± 0.6 Ma based on K-Ar dating of supergene alunite, distinguished from hypogene alunite by grain size, color, and sulfur isotopic composition. Reworked lenses of volcanic ash in the vicinity of La Escondida provide useful time-stratigraphic markers at 8.7 ± 0.4, 6.5 ± 0.2, and 4.2 ± 0.2 Ma within present soil profiles. Long-term average rates of erosion are determined by these age dates and quantitative calculation of eroded leached capping thickness at La Escondida using mass-balance analysis of geochemical profiles coupled with an estimate of unmineralized lithocap thickness based on alteration petrology and fluid-inclusion geobarometry at similar deposits. The observed trend of decreasing long-term average erosion rates with time is consistent with arid to semiarid conditions in the early Miocene changing to hyperarid conditions during the middle Miocene. This climatic desiccation caused termination of significant supergene copper-suffide enrichment at La Escondida and elsewhere in the Atacama region and preservation of surficial features, including the ash horizons and the leached capping. Middle Miocene climatic desiccation in northern Chile and southern Peru was probably related to a pronounced decrease in temperature of coastal waters supplied by an ancestral Humboldt Current and an increase in upwelling intensity as the Antarctic ice cap became established at approximately 15 to 13 Ma. The Central Andes Cordillera, which now provides a rain shadow protecting the Atacama region from precipitation from the east, must have attained at least half its present elevation prior to about 15 Ma to have played a similar role in the middle Miocene.

From a black to a gray box — a mass balance interpretation of pedogenesis

We utilize chemical elements as tracers in a mass balance analysis that provides functional relationships among soil chemical composition, bulk density, and volume change in relation to parent material. These analytical functions are based on the principle of conservation of mass and include a term quantifying mass flux into/out of the soil and between horizons. We apply the technique to the oldest member of a chronosequence developed on marine terraces in northern California. It is an Alfisol that evolved from beach sand to its present state — nearly one-third of its weight is composed of secondary clay minerals — in about 240 ky. Aside from large increases in organic carbon, desilication is the dominant factor in soil evolution; 29% (50.83 g cm−2) Si was leached from the beach sand during pedogenesis. The rate of desilication is roughly 2.1 t km−2 yr−1 (0.21 g cm−2 ky−1), an order of magnitude slower than that implied by the Si denudation rate calculated for the Mattole River watershed. Weathering of primary minerals and synthesis of secondary minerals is relatively well-advanced suggesting that the rate of desilication may be declining. The local beach is composed of quartz and sodic plagioclase with smaller amounts of chlorite, mica, and kaolinite. The soil has substantially different mineralogy: sand is dominated by quartz, and clay is dominated by kaolinite/halloysite, chloritic intergrades, and gibbsite. Bases are also leached, though the total mass was much less; 57% (2.55 g cm−2) of Na in the beach sand was lost as plagioclase weathered. By focusing on elemental and mineralogical gains and losses, we emphasize the essential connection between the pedologic environment and the external hydrochemical environment.

Weathering profiles, mass-balance analysis, and rates of solute loss: Linkages between weathering and erosion in a small, steep catchment

In a headwater catchment in the Oregon Coast Range, we find that solid-phase mass losses due to chemical weathering are equivalent in the bedrock and the soil. However, the long-term rate of mass loss per unit volume of parent rock is greater in the soil than in the rock. We attribute this finding to the effects of biotic processes in the soil and to hydrologic conditions that maximize contact time and water flux through the mineral matrix in the soil. This result stems both from earlier work in which we demonstrated that rock and soil contribute equally to the solute flux and from arguments presented here that the basin is in dynamic equilibrium with respect to erosion and uplift. The silica flux of 10.7 ± 7.1 t·km−2·yr−1 from the basin is several times larger than the flux from older soils elsewhere, but comparable to the flux from sites with similar physical erosion rates. This result argues that physical denudation or uplift rates play an important role in setting the chemical denudation rate. Physical processes appear to influence chemical-weathering rates in several ways. First, they limit chemical evolution by removing material, thus setting the residence time within the weathered rock and the soil. Second, bioturbation mixes rock fragments into the more reactive soil and maintains high soil porosity, allowing free circulation of water. Because the weathering in the soil is more intense than in the rock, we argue that the chemical denudation rate will diminish where uplift rates—and, hence, physical-denudation rates—are great enough to lead to a bedrock-dominated landscape. Chemical denudation rates will increase with physical-denudation rates, but only as long as the landscape remains mantled by soil.

DEFORMATIONAL MASS TRANSPORT AND INVASIVE PROCESSES IN SOIL EVOLUTION

Soils are differentiated vertically by coupled chemical, mechanical, and biological transport processes. Soil properties vary with depth, depending on the subsurface stresses, the extent of mixing, and the balance between mass removal in solution or suspension and mass accumulation near the surface. Channels left by decayed roots and burrowing animals allow organic and inorganic detritus and precipitates to move through the soil from above. Accumulation occurs at depths where small pores restrict further passage. Consecutive phases of translocation and root growth stir the soil; these processes constitute an invasive dilatational process that leads to positive cumulative strains. In contrast, below the depth of root penetration and mass additions, mineral dissolution by descending organic acids leads to internal collapse under overburden load. This softened and condensed precursor horizon is transformed into soil by biological activity, which stirs and expands the evolving residuum by invasion by roots and macropore networks that allows mixing of materials from above.

Magmatic arc asymmetry and distribution of anomalous plutonic belts in the batholiths of California: Effects of assimilation, crustal thickness, and depth of crystallization

In order to better understand geologic fac-tors controlling pronounced regional variations in whole-rock chemistry, mineralogy, and mineral chemistry in the batholiths of California, we calculate the magmatic intensive variables f HF /f H 2 O , f HF /f HCl , and f O 2 .Regional-scale west-to-east increases in F/OH in mafic silicates, corresponding to the systematic I-WC to I-MC and I-SC progression, reflect orders-of-magnitude increase in f HF /f H 2 O attending pluton crystallization. Low f HF / f H 2 O of formation of western I-WC types is consistent with their derivation from low-fluorine source rocks in subducted oceanic slabs or the upper mantle. In contrast,higher f HF /f H 2 O of crystallization of I-MC and I-SC types to the east implies the involvement of (1) progressively greater amounts of continental crustal source material such as biotite-bearing metamorphic rocks, their unweathered sedimentary derivatives retaining F-rich mafic minerals, or their fusion products and/or (2) source materials which become more F-rich toward the continental interior. The regional distribution of I-MC and I-SC types suggests that the Precambrian craton of western North America, or derivative sediments, may extend farther north in California and be morphologically more complex than previously thought. From seemingly out-of- place occurrences of I-WC plutons on the eastern slopes of the Sierra Nevada batholith, we infer the existence of regions where Precambrian basement was thin or absent in Mesozoic time, which prevented extensive cratonal contamination of subducted slab or upper mantle-derived magmas. New methods for estimating T-f O 2 relations in the magmas demonstrate that I-SCR granites crystallize at oxygen fugacities as much as five orders of magnitude lower than those of I-WC, I-MC, and I-SC types under conditions at or below the maximum stability limit of graphite in equilibrium with a C-O-H-S gas phase. The local-scale formation of I-SCR granites in plutonic belts within specific wall-rock terranes containing highly reducing sediments or metasediments may occur by contamination of I-types with graphitic pelite or, in some cases, by the direct fusion of this reducing pelitic wall rock. The spatial distribution of I-SCR granite provinces therefore is controlled simply by wall-rock lithology. Amphibole geobarometry demonstrates a general west-to-east decrease in crystallization pressure across the Sierra Nevada batholith. In contrast, the Peninsular Ranges batholith displays a west-to-east crystallization pressure increase. The bulk of the California batholiths crystallized at pressures less than 4-5 kb and depths less than about 15-19 km. In the southern Sierra Nevada batholith, the San Gabriels, and the eastern Peninsular Ranges, however, plutons crystallized at pressures exceeding 6 kb at deep crustal levels (>23 km). Reconstruction of the pre-erosion top of the batholith shows that in an east-west cross section, the central Sierra Nevada batholith was a horizontal tabular body with an aspect ratio of at least five to one.

Quantitative geochemical approach to pedogenesis: importance of parent material reduction, volumetric expansion, and eolian influx in lateritization

Abstract Using mass balance techniques we test the prevalent view that laterite genesis is dominated by in situ residual enrichment during chemical weathering of bedrock. Through calculation of net mass fluxes through the laterite soils in Mali, West Africa, we show that residual enrichment by removal of mobile elements with a corresponding increase in bulk porosity and decrease in bulk density, contributes only a very minor fraction of the enrichment of Al, Fe, Si and Au. Instead, we demonstrate that the abundance of these elements is due to the influx and accumulation by selective retention of chemically mature detritus of local and foreign origin clearly evident in micromorphological infilling features. At the same sample depths that accumulation reaches extreme values, we show that volumetric expansion in excess of 200% has occurred locally. We infer that these spatially coincidental zones of mass influx of rock-forming metals Fe, Al, Si, and also Au with dilational hyperstrains result from a mutually reinforcing, mechanical interaction between material influx and the effects of subsurface deformational processes such as shrink-swell cycles and root growth and decay. We propose that with progressive infilling of available connected voids by illuvial microsedimentary deposits of insoluble resistate and neoformed minerals, the capacity for the combined skeleton and plasma to remain isovulumetric is exceeded. We speculate that the resultant space problem is relieved by upwards expansion towards the overlying free surface. Continued translocation and void infilling occur and are limited to the depth where the size of translocational particles is smaller than that of the connected voids. Consequently, progressive hyperstrains accumulate above this critical depth by the long-term influence of a proposed translocational wedge of chemically-resistant minerals against which numerous generations of plant roots have exerted stresses. Eclectic surficial contaminants involved are continuously derived from above leaving no indication of a relict source region within the present soil profile from which they might have been extracted. Instead, the source region is largely the existing regolith column itself which releases local material and in addition, is supplemented by deposition of colluvial detritus shed nearby by escarpment retreat

Direct dating of weathering phenomena by 40Ar39Ar and K-Ar analysis of supergene K-Mn oxides

Abstract Potassium-bearing manganese oxides, cryptomelane, K1–2(Mn3+Mn4+)8 O16 · xH2Oand hollandite, (K,Ba)1–2 (Mn3+,Mn4+)8 O16· xH2O, are often authigenically precipitated in weathering profiles. The presence of structural potassium in manganese oxides makes these minerals datable by the K-Ar and 40 Ar 39 Ar methods. Dating of these phases allows us to time the progression of oxidation fronts during weathering and pedogenic processes. Detailed age resolution of the weathering history allows us to use deep weathering profiles as land-based stratigraphic records of past climatic and geomorphic events, complementing evidence from the marine sedimentary and paleontological record. Potential problems in manganese oxide dating, such as Ar and/or K losses, excess argon, 39Ar loss by recoil during neutron irradiation, contamination by primary minerals resistant to weathering, the presence of multiple generations of manganese oxides, and proportionately low radiogenic Ar yield due to adsorbed atmospheric gases are addressed. The K-Ar and 40 Ar 39 Ar analytical results indicate that Ar and/or K losses, excess 40Ar and 39Ar recoil seem not to pose problems in manganese oxide dating. This investigation suggests that fine scale, laser-probe 40 Ar 39 Ar technique is most appropriate for dating of weathering phenomena because this technique permits us to identify contaminating phases and the presence of multiple generations of weathering minerals in the inherently complex mineral assemblage characteristic of weathering profiles. K-Ar and 40 Ar 39 Ar dating of supergene K-bearing manganese oxides formed during lateritization of Archean and Proterozoic bedrocks in the Carajâs Region, Amazonia, Brazil, indicates that weathering started before 72 ± 6 Ma. Petrographic, electron microscope and electron microprobe investigation reveal multiple generations of manganese oxide precipitation. K-Ar and 40 Ar 39 Ar laser-probe step-heating results independently confirm the presence of these various generations of manganese oxides. Age clusters at 65–69, 51–56, 40–43, 33–35, 20, 24, 12–17 Ma and zero-age (0.2 ± 0.2 Ma) suggest episodic precipitation of K-Mn oxides resulting from changing weathering conditions in the Amazon throughout the Cenozoic. K-Ar and 40 Ar 39 Ar dating of supergene cryptomelane from weathering profiles in eastern Minas Gerais, southeastern Brazil, suggests continuous weathering from 10 to 5.6 Ma ago, possibly reflecting local climatic conditions due to the proximity with the Atlantic Ocean.

Regional variations in bulk chemistry, mineralogy, and the compositions of mafic and accessory minerals in the batholiths of California

We define regional variations in mafic and accessory mineral assemblages and compositions and expand the current understanding of spatial variations in whole-rock geochemistry in the batholiths of California. In so doing, we gain new insights into the nature of magmatic source rocks and mechanisms of magma generation in volcano-plutonic arcs of active continental margins. Little-studied metaluminous to strongly peraluminous granites containing Fe-rich biotite with log(X Mg /X Fe ) F/OH and Mn in biotite and amphibole increase on a regional scale from western I-WC types to eastern I-MC and I-SC types, parallel to eastward increases in incompatible elements and decreases in compatible elements in the plutons. In contrast, the belts of western I-SCR granites and eastern I-WC quartz diorites and granodiorites disrupt the regional west-to-east systematics in both mineral and whole-rock geochemistry. Spatial variations in the Al content of amphibole are regional in scale and reflect pressures of pluton crystallization. We conclude that significant, previously unrecognized complexity exists in regional geochemical systematics in the California batholiths.

Deep lithospheric dynamics beneath the Sierra Nevada during the Mesozoic and Cenozoic as inferred from xenolith petrology

[1] Abstract: Peridotite xenoliths erupted in late Miocene basalts (� 8 Ma) in the central Sierra Nevada sample a lithosphere that is vertically stratified in terms of age and thermal history. The deeper portions (� 45– 100 km) have asthenospheric osmium isotopic compositons and possess textural and chemical evidence for cooling from >11008 to 700– 8208C. The shallower portions (<60 km) have unradiogenic Os isotopic compositions, which yield Proterozoic model ages, and contain orthopyroxenes that record temperatures as low as 6708C in their cores and heating up to 9008C on their rims. These observations suggest that the deeper xenoliths represent fragments of hot asthenosphere that upwelled to intrude and/or underplate the overlying Proterozoic lithosphere represented by the shallower xenoliths. The contrasting thermal histories between the shallow and deep xenoliths suggest that hot asthenosphere and cold lithosphere were suddenly juxtaposed, a feature consistent with the aftermath of rapid lithospheric removal or sudden intrusion of asthenospheric mantle into the lithosphere rather than passive extension. On the basis of regional tectonics and various time constraints, it is possible that this lithospheric removal event was associated with the generation of the Sierra Nevada granitic batholith during Mesozoic subduction of the Farallon plate beneath North America. Pleistocene basalt-hosted xenoliths record a different chapter in the geodynamic history of the Sierras. These xenoliths are relatively fertile, come from depths shallower than 45– 60 km, are characterized by asthenospheric Os isotopic compositions, record hot equilibration temperatures (10008– 11008C), and show no evidence for cooling. The strong contrast in composition and thermal history between the Pleistocene and late Miocene suites indicate that the post-Mesozoic lithospheric mantle, as represented by the latter, was entirely replaced by the former. The hot Pleistocene peridotites may thus represent new lithospheric additions associated witha post-Miocene lith osph eric removal event or extension. Highelevations, low sub-Moho seismic velocities, and the presence of fast velocity anomalies at 200 km depth may be manifestations of this event. If lithospheric removal occurred in the Mesozoic and Cenozoic, the observations presented here place constraints on the styles of lithospheric removal. In the Mesozoic, the lithospheric mantle was only partially removed, whereas in the Pliocene, the entire lithospheric mantle and probably the mafic lower crust were removed.