Basalt Granodiorite
Total Page:16
File Type:pdf, Size:1020Kb
Lunar and Planetary Science XXXI 1460.pdf SILICA MOBILITY DURING SURFICIAL PROCESSES ON MARS. Scott M. McLennan, Department of Geosciences, State University of New York at Stony Brook, Stony Brook, NY, 11794-2100, USA ([email protected]). Introduction: The mobile nature of silica during als and glass [e.g. 9-12]. low temperature surficial processes on Earth is appar- During weathering of basalt SiO2/Al2O3 ratios may ent from the common occurrence of sedimentary silica decrease by as much as 50% in the most altered part deposits, quartz veins, silcretes, quartz overgrowths of weathering profiles [11]. This degree of silica loss and replacements in sedimentary rocks and the com- is comparable to that expected from weathering of mon depletion of silica in soils. This is in spite of the more evolved granitic rocks [e.g., 13] but differs in fact that crystalline quartz is not particularly soluble that silica loss during basalt weathering occurs more under most near surface conditions. Accordingly, during early stages of weathering whereas it occurs sedimentary silica is largely derived from the break- during later stages of weathering of granitic rocks down of other silicates (e.g., feldspar, amphiboles, (Fig. 1). olivine, pyroxene) and glass during low temperature processes, pressure solution during compaction and creep, as well as a variety of diagenetic and metamor- 0 phic reactions during high pressure and temperature -10 conditions. 3 O For Mars, the issue of near surface silica mobility 2 -20 Granodiorite /Al has received scant attention. There is wide agreement 2 that a variety of sedimentary processes have affected -30 the Martian surface and a number of workers have Basalt recognized that silica may play a role during alteration -40 (e.g., [1-4]). For example, Burns [1] proposed the Baynton Basalt possibility of “hydrous iron oxide - silica deposits” in % Change SiO -50 (Nesbitt & Wilson, 1992) Toorongo Granodiorite the Argyre and Hellas basins and some models of the (Nesbitt & Markovics, 1997) Martian hydrological cycle predict significant -60 40 50 60 70 80 90 100 amounts of sedimentary silica [3]. However, evidence Chemical Index of Alteration (CIA) of significant silica deposits at or near the Martian surface has been inadequately explored. Fig. 1. Change in SiO2/Al2O3 during progressive weather- ing in southeastern Australia. Note substantial loss of silica The question of sedimentary silica on Mars is of during basalt weathering and that silica is lost during early considerable importance for a variety of reasons. Mo- weathering stages. Data from [11, 13]. bility of silica during surficial processes would signifi- cantly constrain hydrological models. Surprisingly high SiO2 contents of rocks found at the Pathfinder 20 Meltwater / site have proven perplexing but generally have been 10 Meteoric Water interpreted in terms of igneous processes. There is T = 0°C Cl = 5-10ppm 3 0 some question, however, as to whether the chemistry O of Pathfinder rocks are best interpreted as being sedi- 2 -10 /Al mentary in origin [5]. In either case, the presence of 2 -20 secondary silica phases would severely constrain any -30 interpretation of these rock compositions. Finally, silica dust is a well known health hazard [6] and per- -40 haps as little as 0.1% quartz in Martian dust could -50 % Change SiO Palagonite severely limit human exploration of Mars [7]. -60 Clay Alteration Products Low Temperature Silica Mobility in Basalt Ter- -70 ranes: Analyses of SNC meteorites and Martian rocks 0.0010 0.0015 0.0020 0.0025 87Sr/86Sr 87Sr/86Sr and soils and general understanding of planetary evo- clay - glass lution indicate that magmatic rocks on Mars are Fig. 2. Change in SiO /Al O of co-existing palagonite and dominantly mafic to ultramafic [8]. Alteration of ter- 2 2 3 clay during low temperature alteration of Icelandic basaltic restrial basalt across a broad range of open system glass. The change in the Sr isotopic composition between conditions, from weathering to hydrothermal, is com- clay and glass is taken as a measure of alteration. Data monly accompanied by loss of large amounts of SiO2 from [12]. mainly due to the dissolution of unstable mafic miner- Lunar and Planetary Science XXXI 1460.pdf SILICA MOBILITY ON MARS: S. M. McLennan Palagonitization of basalt is a commonly proposed quartz . Thus silica could be found in a variety of process for the Martian surface. On Earth, silica is geological settings, including sedimentary chert de- typically very mobile during such processes. An in- posits (iron formations of Burns [1]?), veins, fracture structive example comes from the palagonitization of fillings, mineral overgrowths, encrustations on rock Icelandic basaltic glass taking place at very low tem- surfaces, silcretes, authigenic mineral grains, or dust perature during interaction with meltwater and mete- coatings. oric water having very low dissolved solids [12]. In The presence of significant amounts of silica this example, SiO2/Al2O3 ratios are decreased by as would have important implications for interpreting much as 60% in both palagonite and altered clays, imaging, remote sensing and chemical data that have compared to the parental glass (Fig. 2). been returned from Mars. To date quartz has not been Evidence for Low Temperature Alteration of identified from the various spectroscopic methods and Martian Basalt: Chemical relationships among Mar- limited mapping by the thermal emission spectrometer tian rocks and soils are consistent with a significant (TES) restricts quartz to less than about 5% in the non-silicate secondary mineralogy. At the Pathfinder regions mapped [20], a value that is not inconsistent site, MgO (and possibly FeOT) correlate positively with the constraints discussed here. with SO3 and Cl, suggesting that a significant fraction Perhaps the most significant implication of the of the Mg in soils is present as sulfate and chloride possibility of sedimentary silica is that it leaves open minerals [14-17]. It is also generally agreed that the question of whether the high silica contents of much of the iron in Martian surficial deposits is pre- Pathfinder rocks could have an origin that is funda- sent as secondary oxide minerals such as nanophase mentally unrelated to igneous petrogenesis. If Path- oxides, hematite and maghemite [2,18,19]. Chemical finder rocks are of sedimentary origin then silica data are also consistent with fractionation of iron rich phases could constitute a fundamental component of phases (e.g., hematite, Fe-sulfate) in Martian soils and the rock. On the other hand, if Pathfinder rocks are rocks, likely during sedimentary transport [17]. indeed of igneous origin, high silica abundances could The significance of these relationships is that they be related to an occurrence as vein fillings or various provide strong indirect evidence that silica was mobile types of surface coatings. Even a minor surficial silica during surficial processes. The most likely source of component on the rocks could have an undue influ- Mg and Fe is from alteration of mafic minerals that ence on the APXS chemical results and is not entirely reacted to form these non-silicate minerals. Accord- inconsistent with the spectral data [21]. ingly, simple mass balance suggests that there is a likelihood that substantial amounts of SiO2, also de- References: [1] Burns, R.G. (1993) GCA, 57, rived from these alteration reactions, would be avail- 4555-4574. [2] Bell III, J.F. (1996) Geochem. Soc. able to form discrete sedimentary silica phases, Spec. Publ. 5, 359-380. [3] Catling, D. C. (1999) including non-crystalline or crystalline opals, JGR, 104, 16453-16469. [4] McLennan, S.M. (1999) microcrystalline quartz (e.g., chalcedony) or quartz. 9th Goldschmidt Conf., LPI Contrib. 971, 194-195. The amount of silica that could be produced is [5] Dreibus, G. et al. (1999) EOS, 80, F616. [6] Saf- difficult to estimate with confidence but an upper limit fiotti, U. et al. (1993) Rev. Mineral., 28, 523-554. [7] of about 17% of altered material can be calculated that Meloy, T. P. (1999) 5th Int. Mars Conf., Abst. #6138. in turn would translate into a lower value for the [8] McSween Jr., H.Y. & Treiman, A.H. (1998) Rev. overall Martian surface. Mineral., 36, 6/1-6/53. [9] Dove, P. M. and Rimstidt, Silica Cycle near the Martian Surface: The ul- J. D. (1994) Rev. Mineral., 29, 259-308. [10] timate fate of silica is also difficult to constrain be- Brantley, S. L. and Chen, Y. (1995) Rev. Mineral., cause little is known about the nature of the Martian 31, 119-172. [11] Nesbitt, H.W. & Wilson, R.E. hydrological cycle over geological history. Among the (1992) AJS, 292, 740-777. [12] Daux, V. et al. (1994) critical questions are the fluid/rock ratios, water chem- GCA, 58, 4941-4954. [13] Nesbitt, H. W. and Mark- istry, and the degree to which the systems are open to ovics, G. (1997) GCA 61, 1653-1670. [14] Clark, transport away from the site of alteration. Near altera- B.C. (1982) JGR, 87, 10059-10067. [15] Rieder, R. et tion zones, silicon may combine with aluminum (from al. (1997) Science, 278, 1771-1774. [16] McSween, plagioclase) and various cations to form palagonite H. Y. and Keil, K. (2000) GCA, in press. [17] and a variety of clays (e.g., saponite, nontronite, sepio- McLennan, S. M. (2000) GRL, submitted. [18] Bell, lite, montmorillonite, etc.). However silica that is lost J. F., III et al. (2000) JGR, in press. [19] Madsen, to near-surface waters could be transported consider- M.B. (1999) JGR, 104, 8761-8779. [20] Christensen, able distance and precipitated as a separate silica P. R. (1998) Science, 279, 1692-1698. [21] McSween phase, that could range anywhere through the spec- Jr., H. Y. et al. (1999) JGR, 104, 8679-8715. trum from amorphous silica through to crystalline .