Loading…

Controls on rind thickness on basaltic andesite clasts weathering in Guadeloupe

A clast of low porosity basaltic andesite collected from the B horizon of a soil developed on a late Quaternary volcaniclastic debris flow in the Bras David watershed on Basse-Terre Island, Guadeloupe, exhibits weathering like that observed in many weathered clasts of similar composition in other tr...

Full description

Saved in:
Bibliographic Details
Published in:Chemical geology 2010-09, Vol.276 (3), p.129-143
Main Authors: Sak, Peter B., Navarre-Sitchler, Alexis K., Miller, Christine E., Daniel, Christopher C., Gaillardet, Jerome, Buss, Heather L., Lebedeva, Marina I., Brantley, Susan L.
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A clast of low porosity basaltic andesite collected from the B horizon of a soil developed on a late Quaternary volcaniclastic debris flow in the Bras David watershed on Basse-Terre Island, Guadeloupe, exhibits weathering like that observed in many weathered clasts of similar composition in other tropical locations. Specifically, elemental profiles measured across the core–rind interface document that primary minerals and glass weather to Fe oxyhydroxides, gibbsite and minor kaolinite in the rind. The earliest reaction identified in the core is oxidation of Fe in pyroxene but the earliest reaction that creates significant porosity is plagioclase dissolution. Elemental loss varies in the order Ca ≈ Na > K ≈ Mg > Si > Al > Fe ≈ P ≫ Ti, consistent with the relative reactivity of phases in the clast from plagioclase ≈ pyroxene ≈ glass > apatite > ilmenite. The rind surrounds a core of unaltered material that is more spherical than the original clast. The distance from the core–rind boundary to a visually prominent rind layer, L, was measured as a proxy for the rind thickness at 36 locations on a slab cut vertically through the nominal center of the clast. This distance averaged 24.4 ± 3.1 mm. Maximum and minimum values for L, 35.8 and 20.6 mm, were observed where curvature of the core–rind boundary is greatest (0.12 mm − 1) and smallest (0.018 mm − 1) respectively. Extrapolating from other rinds in other locations, the rate of rind formation is estimated to vary by a factor of about 2 (from ∼ 4 to 7 × 10 − 14 m s − 1) from low to high curvature. The observation of a higher rate of rind formation for a higher curvature interface is consistent with a diffusion-limited model for weathering rind formation. The diffusion-limited model predicts that, like rind thickness, values of the thickness of the reaction front ( h) for a given reaction, defined as the zone over which a parent mineral such as plagioclase completely weathers to rind material, should also increase with curvature. Values of h were quantified as a function of interface curvature using bulk chemical analysis (500 < h < 2000 μm). Values of h were also quantified by measuring loss of matrix glass and increase in porosity as a function of curvature. In contrast to rind thickness, h shows no consistent increase with curvature. This contradiction is attributed to the mm-scale roughness of the interface which is related to phenocryst grain size. Therefore, the overall rind formation rate is strongly affec
ISSN:0009-2541
1872-6836
DOI:10.1016/j.chemgeo.2010.05.002