Marjorie S. Schulz

CHEMICAL WEATHERING IN A TROPICAL WATERSHED, LUQUILLO MOUNTAINS, PUERTO RICO : I. LONG-TERM VERSUS SHORT-TERM WEATHERING FLUXES

Abstract The pristine Rio Icacos watershed in the Luquillo Mountains in eastern Puerto Rico has the fastest documented weathering rate of silicate rocks on the Earth’s surface. A regolith propagation rate of 58 m Ma−1, calculated from iso-volumetric saprolite formation from quartz diorite, is comparable to the estimated denudation rate (25–50 Ma−1) but is an order of magnitude faster than the global average weathering rate (6 Ma−1). Weathering occurs in two distinct environments; plagioclase and hornblende react at the saprock interface and biotite and quartz weather in the overlying thick saprolitic regolith. These environments produce distinctly different water chemistries, with K, Mg, and Si increasing linearly with depth in saprolite porewaters and with stream waters dominated by Ca, Na, and Si. Such differences are atypical of less intense weathering in temperate watersheds. Porewater chemistry in the shallow regolith is controlled by closed-system recycling of inorganic nutrients such as K. Long-term elemental fluxes through the regolith (e.g., Si = 1.7 × 10−8 moles m−2 s−1) are calculated from mass losses based on changes in porosity and chemistry between the regolith and bedrock and from the age of the regolith surface (200 Ma). Mass losses attributed to solute fluxes are determined using a step-wise infiltration model which calculates mineral inputs to the shallow and deep saprolite porewaters and to stream water. Pressure heads decrease with depth in the shallow regolith (−2.03 m H2O m−1), indicating that both increasing capillary tension and graviometric potential control porewater infiltration. Interpolation of experimental hydraulic conductivities produces an infiltration rate of 1 m yr−1 at average field moisture saturation which is comparable with LiBr tracer tests and with base discharge from the watershed. Short term weathering fluxes calculated from solute chemistries and infiltration rates (e.g., Si = 1.4 × 10−8 moles m−2 s−1) are compared to watershed flux rates (e.g., Si = 2.7 × 10−8 moles m−2 s−1). Consistency between three independently determined sets of weathering fluxes imply that possible changes in precipitation, temperature, and vegetation over the last several hundred thousand years have not significantly impacted weathering rates in the Luquillo Mountains of Puerto Rico. This has important ramifications for tropical environments and global climate change.

Chemical weathering rates of a soil chronosequence on granitic alluvium: I. Quantification of mineralogical and surface area changes and calculation of primary silicate reaction rates

Mineral weathering rates are determined for a series of soils ranging in age from 0.2–3000 Ky developed on alluvial terraces near Merced in the Central Valley of California. Mineralogical and elemental abundances exhibit time-dependent trends documenting the chemical evolution of granitic sand to residual kaolinite and quartz. Mineral losses with time occur in the order: hornblende > plagioclase > K-feldspar. Maximum volume decreases of >50% occur in the older soils. BET surface areas of the bulk soils increase with age, as do specific surface areas of aluminosilicate mineral fractions such as plagioclase, which increases from 0.4–1.5 m2 g−1 over 600 Ky. Quartz surface areas are lower and change less with time (0.11–0.23 m2 g−1). BET surface areas correspond to increasing external surface roughness (λ = 10–600) and relatively constant internal surface area (≈ 1.3 m2 g−1). SEM observations confirm both surface pitting and development of internal porosity. A numerical model describes aluminosilicate dissolution rates as a function of changes in residual mineral abundance, grain size distributions, and mineral surface areas with time. A simple geometric treatment, assuming spherical grains and no surface roughness, predicts average dissolution rates (plagioclase, 10−17.4; K-feldspar, 10−17.8; and hornblende, 10−17.5 mol cm−1 s−1) that are constant with time and comparable to previous estimates of soil weathering. Average rates, based on BET surface area measurements and variable surface roughnesses, are much slower (plagioclase, 10−19.9; K-feldspar, 10−20.5; and hornblende 10−20.1 mol cm−2 s−1). Rates for individual soil horizons decrease by a factor of 101.5 over 3000 Ky indicating that the surface reactivities of minerals decrease as the physical surface areas increase. Rate constants based on BET estimates for the Merced soils are factors of 103–104 slower than reported experimental dissolution rates determined from freshly prepared silicates with low surface roughness (λ < 10). This study demonstrates that the utility of experimental rate constants to predict weathering in soils is limited without consideration of variable surface areas and processes that control the evolution of surface reactivity with time.

Demonstration of significant abiotic iron isotope fractionation in nature

Field and laboratory studies reveal that the mineral ferrihydrite, formed as a result of abiotic oxidation of aqueous ferrous to ferric Fe, contains Fe that is isotopically heavy relative to coexisting aqueous Fe. Because the electron transfer step of the oxidation process at pH >5 is essentially irreversible and should favor the lighter Fe isotopes in the ferric iron product, this result suggests that relatively heavy Fe isotopes are preferentially partitioned into the readily oxidized Fe(II)(OH) x (aq) species or their transition complexes prior to oxidation. The apparent Fe isotope fractionation factor, α ferrihydrite- water , depends primarily on the relative abundances of the Fe(II) (aq) species. This study demonstrates that abiotic processes can fractionate the Fe isotopes to the same extent as biotic processes, and thus Fe isotopes on their own do not provide an effective biosignature.

The role of disseminated calcite in the chemical weathering of granitoid rocks

Accessory calcite, present at concentrations between 300 and 3000 mg kg−1, occurs in fresh granitoid rocks sampled from the Merced watershed in Yosemite National Park, CA, USA; Loch Vale in Rocky Mountain National Park CO USA; the Panola watershed, GA USA; and the Rio Icacos, Puerto Rico. Calcite occurs as fillings in microfractures, as disseminated grains within the silicate matrix, and as replacement of calcic cores in plagioclase. Flow-through column experiments, using de-ionized water saturated with 0.05 atm. CO2, produced effluents from the fresh granitoid rocks that were dominated by Ca and bicarbonate and thermodynamically saturated with calcite. During reactions up to 1.7 yr, calcite dissolution progressively decreased and was superceded by steady state dissolution of silicates, principally biotite. Mass balance calculations indicate that most calcite had been removed during this time and accounted for 57–98% of the total Ca released from these rocks. Experimental effluents from surfically weathered granitoids from the same watersheds were consistently dominated by silicate dissolution. The lack of excess Ca and alkalinity indicated that calcite had been previously removed by natural weathering. The extent of Ca enrichment in watershed discharge fluxes corresponds to the amounts of calcite exposed in granitoid rocks. High Ca/Na ratios relative to plagioclase stoichiometries indicate excess Ca in the Yosemite, Loch Vale, and other alpine watersheds in the Sierra Nevada and Rocky Mountains of the western United States. This Ca enrichment correlates with strong preferential weathering of calcite relative to plagioclase in exfoliated granitoids in glaciated terrains. In contrast, Ca/Na flux ratios are comparable to or less than the Ca/Na ratios for plagioclase in the subtropical Panola and tropical Rio Icacos watersheds, in which deeply weathered regoliths exhibit concurrent losses of calcite and much larger masses of plagioclase during transport-limited weathering. These results indicate that the weathering of accessory calcite may strongly influence Ca and alkalinity fluxes from silicate rocks during and following periods of glaciation and tectonism but is much less important for older stable geomorphic surfaces.

The effect of temperature on experimental and natural chemical weathering rates of granitoid rocks

The effects of climatic temperature variations (5–35°C) on chemical weathering are investigated both experimentally using flow-through columns containing fresh and weathered granitoid rocks and for natural granitoid weathering in watersheds based on annual solute discharge. Although experimental Na and Si effluent concentrations are significantly higher in the fresh relative to the weathered granitoids, the proportional increases in concentration with increasing temperature are similar. Si and Na exhibit comparable average apparent activation energies (Ea) of 56 and 61 kJ/mol, respectively, which are similar to those reported for experimental feldspar dissolution measured over larger temperature ranges. A coupled temperature–precipitation model, using an expanded database for solute discharge fluxes from a global distribution of 86 granitoid watersheds, produces an apparent activation energy for Si (51 kJ/mol), which is also comparable to those derived from the experimental study. This correlation reinforces evidence that temperature does significantly impact natural silicate weathering rates. Effluent K concentrations in the column study are elevated with respect to other cations compared to watershed discharge due to the rapid oxidation/dissolution of biotite. K concentrations are less sensitive to temperature, resulting in a lower average Ea value (27 kJ/mol) indicative of K loss from lower energy interlayer sites in biotite. At lower temperatures, initial cation release from biotite is significantly faster than cation release from plagioclase. This agrees with reported higher K/Na ratios in cold glacial watersheds relative to warmer temperate environments. Increased release of less radiogenic Sr from plagioclase relative to biotite at increasing temperature produces corresponding decreases in 87Sr/86Sr ratios in the column effluents. A simple mixing calculation using effluent K/Na ratios, Sr concentrations and 87Sr/86Sr ratios for biotite and plagioclase approximates stoichiometric cation ratios from biotite/plagioclase dissolution at warmer temperatures (35°C), but progressively overestimates the relative proportion of biotite with decreasing temperature. Ca, Mg, and Sr concentrations closely correlate, exhibit no consistent trends with temperature, and are controlled by trace amounts of calcite or exchange within weathered biotite. The inability of the watershed model to differentiate a climate signal for such species correlates with the lower temperature dependence observed in the experimental studies.

Chemical weathering of a soil chronosequence on granitoid alluvium: II. Mineralogic and isotopic constraints on the behavior of strontium

The use of strontium isotopes to evaluate mineral weathering and identify sources of base cations in catchment waters requires an understanding of the behavior of Sr in the soil environment as a function of time. Our approach is to model the temporal evolution of 87Sr/86Sr of the cation exchange pool in a soil chronosequence developed on alluvium derived from central Sierra Nevada granitoids during the past 3 Ma. With increasing soil age, 87Sr/86Sr of ammonium-acetate extractable Sr initially decreases from values typical of K-feldspar to those of plagioclase and hornblende and then remains constant, even though plagioclase and hornblende are absent from the soils after approximately 1 Ma of weathering. The temporal variation of 87Sr/86Sr of exchangeable Sr is modeled by progressively equilibrating Sr derived from mineral weathering and atmospheric deposition with Sr on exchange sites as waters infiltrate a soil column. Observed decreases in quartz-normalized modal abundances of plagioclase, hornblende, and K-feldspar with time, and the distinct87Sr/86Sr values of these minerals can be used to calculate Sr flux from weathering reactions. Hydrobiotites in the soils have nearly constant modal abundances, chemistry, and 87Sr/86Sr over the chronosequence and provide negligible Sr input to weathering solutions. The model requires time and soil horizon-dependent changes in the amount of exchangeable Sr and the efficiency of Sr exchange, as well as a biologic cycling term. The model predicts that exchangeable Sr initially has 87Sr/86Sr identical to that of K-feldspar, and thus could be dominated by Sr leached from K-feldspar following deposition of the alluvium. The maximum value of 87Sr/86Sr observed in dilute stream waters associated with granitoids of the Yosemite region is likewise similar to that of the K-feldspars, suggesting that K-feldspar and not biotite may be the dominant source of radiogenic Sr in the streams. This study reveals that, when attempting to use Strontium isotopes to identify sources of base cations in catchment waters and biomass, both preferential leaching of Sr from minerals during incipient soil development and changing Sr exchange efficiency must be considered along with chemical contributions due to mineral dissolution.

Chemical weathering in a tropical watershed, Luquillo Mountains, Puerto Rico III: Quartz dissolution rates

The paucity of weathering rates for quartz in the natural environment stems both from the slow rate at which quartz dissolves and the difficulty in differentiating solute Si contributed by quartz from that derived from other silicate minerals. This study, a first effort in quantifying natural rates of quartz dissolution, takes advantage of extremely rapid tropical weathering, simple regolith mineralogy, and detailed information on hydrologic and chemical transport. Quartz abundances and grain sizes are relatively constant with depth in a thick saprolite. Limited quartz dissolution is indicated by solution rounding of primary angularity and by the formation of etch pits. A low correlation of surface area (0.14 and 0.42 m2 g−1) with grain size indicates that internal microfractures and pitting are the principal contributors to total surface area. Pore water silica concentration increases linearly with depth. On a molar basis, between one and three quarters of pore water silica is derived from quartz with the remainder contributed from biotite weathering. Average solute Si remains thermodynamically undersaturated with respect to recently revised estimates of quartz solubility ( 17–81 μM). Etch pitting is more abundant on grains in the upper saprolite and is associated with pore waters lower in dissolved silica. Rate constants describing quartz dissolution increase with decreasing depth (from 10−14.5–10−15.1 mol m−2 s−1), which correlate with both greater thermodynamic undersaturation and increasing etch pit densities. Unlike for many aluminosilicates, the calculated natural weathering rates of quartz fall slightly below the rate constants previously reported for experimental studies (10−12.4–10−14.2 mol m−2 s−1). This agreement reflects the structural simplicity of quartz, dilute solutes, and near-hydrologic saturation.

Variations in the Fine-Scale Composition of a Central Pacific Ferromanganese Crust: Paleoceanographic Implications

A 47- to 60-mm-thick Fe-Mn crust from Horizon Guyot (water depth 1800–1780 m), central Pacific, was used to evaluate the potential of crusts as recorders of Neogene paleoceanographic and paleoclimatic conditions. The chemical composition was determined by microprobe for 16 elements from a polished thin section. Three analyses were made per millimeter and averaged to give the composition of each millimeter. The age of the crust was determined by measuring the strontium isotope composition of the crust and comparing it with the Tertiary seawater curve. The crust represents 18.5 m.y. of growth of Fe and Mn oxyhydroxides. The crust is composed of alternating botryoidal and laminated layers. The botryoidal layers formed during the same time intervals that widespread Neogene deep-sea hiatuses were forming in bottom sediments. The botryoidal layers represent growth during times of intensified deepwater flow, whereas the laminated intervals represent more quiescent conditions. The correspondence between the botryoidal layers and the Neogene hiatuses is so strong that we were able to choose a variable growth rate model over a constant growth rate model for the crust. Chemical changes in the crust take two forms. The first is represented by broad changes in the composition defined chiefly by fourth-order polynomial fits to the chemical profiles of each element with depth in the crust. The second is high-frequency changes in composition. The broader changes occurred primarily at about 15, 11.5, 7.4, 6.4, 5.2, and 4.6 Ma, which may correlate with major changes in paleoceanographic circulation and development of ice caps at the poles. The periods of the high-frequency changes may reflect climatic changes that resulted from orbital forcing. These high-frequency changes may correspond to the high-order eccentricity periods of 3.47, 2.04, and 1.31 m.y.

Central Pacific Cobalt-Rich Ferromanganese Crusts: Historical Perspective and Regional Variability

Interelement correlations coupled with X-ray mineralogy, chemical analyses, and Q-mode factor analysis define five phases that compose ferromanganese crusts: δ-MnO2, Fe-phases including Fe oxyhydroxide and Fe silicate, detrital aluminosilicate, biogenic phosphate, and biogenic-nonphosphate. This last phase is derived primarily from the dissolution of biogenic carbonate and silica. These five phases are characterized by the following elements respectively: Co, Mn, Ni, Pb; Fe, Si, As; Al, Cr, Si, Ti, K; P, Ca; and Cu, Ni, Ba, Zn, Cd. The distribution of the δ-MnO2 phase is related to latitude and is controlled by the equatorial zone of higher productivity, which produces a strong and extensive oxygen-minimum zone. The iron phases are similarly distributed, but the overall variability is not as great as for the δ-MnO2 phase. The detrital phase has the inverse distribution to the δ-MnO2 phase and is greatest at higher latitudes, especially in areas beneath the trade-wind belt. The detrital phase is composed of eolian debris and volcanogenic debris eroded from submarine outcrops by bottom currents. The phosphate phase is not clearly distributed with latitude, longitude, or the equatorial zone of high biological productivity. The biogenic-nonphosphate phase is most prominent in the equatorial zone of high biological productivity.

Low sulfur content in submarine lavas: an unreliable indicator of subaerial eruption

Low S content (< 250 ppm) has been used to identify subaerially erupted Hawaiian and Icelandic lavas. Large differences in S content of submarine-erupted lavas from different tectonic settings indicate that the behavior of S is complex. Variations is S abundance in undegassed, submarine-erupted lavas can result from different source compositions, different percentages of partial melting, and crystal fractionation. Low S concentrations in highly vesicular submarine lavas suggest that partial degassing can occur despite great hydrostatic pressure. These processes need to be evaluated before using S content as an indicator of eruption depth.