A combined Y/Ho, high field strength element (HFSE) and Nd isotope perspective on basalt weathering, Deccan Traps, India

High-precision high field strength element (HFSE: Zr, Hf, Nb, Ta, Th, U, W, Mo), Y/Ho, and Nd isotope chemostratigraphy of two contrasting Deccan Traps weathering profiles - an ancient, deeply weathered laterite, and a younger (Quaternary), more moderately weathered saprolite - are used to reconstruct different aspects of basalt weathering. Precision of the HFSE analyses is demonstrated through a report of the long-term concentrations and ratios determined in United States Geological Survey (USGS) and Geological Survey of Japan (GSJ) basalt rock standards (BHVO-1, BHVO-2, BIR-1, BCR-2, JB-2). The oxyanion-forming members (U, Mo, W) are the most mobile of the considered HFSE group. Extreme loss of W, far exceeding those of U and Mo during certain stages of basalt alteration, is first reported here. The ability to strongly fractionate Mo and W during weathering may contribute to solving the unresolved mass imbalance between the crustal and marine inventories of W. By contrast, Zr, Hf, Nb, and Ta preserve the ratios of the parent basalt in the profiles due to their limited mobility; these are therefore of great potential value in reconstructing basalt flow stratigraphy and correlating lava flows in weathered flood basalt provinces. Of the HFSE, Th is not a good choice as a conservative element because it is strongly susceptible to addition of aeolian dust; this is evidenced by strong excursions in Th/Nb that are correlatable with alkali element enrichment and less radiogenic Nd-143/Nd-144 ratios. The chemical fingerprints of dust were identified in a paleo-flow top of the saprolite profile, suggesting that dust accumulation occurred during periods of quiescence between basaltic eruptions. During protracted exposure and laterite development, the magnitude to which dust overprints the basalt chemistry increases substantially as evident from much less radiogenic Nd isotope ratios and higher Th/Nb ratios in the Bidar profile relative to the protolith basalt. Attempts at quantifying the magnitude of dust accumulation in the laterite based on Th enrichment indicate a mass fraction of greater than 0.5 when the dust is assumed to have the chemistry of average upper continental crust. Although mixing models between the basalt and assumed dust composition cannot unambiguously constrain the dust source, the Nd isotope mixture preserved in the laterite points to a relatively young crustal dust source (e.g., similar to loess in composition) rather than the Precambrian shield rocks in the vicinity of the Deccan Traps. The contrasting topologies of dust-derived Nd and dust-derived Th in the laterite appears to record both physical transport of dust (Th) through permeable laterite horizons as well as transport by chemical dissolution and precipitation (Nd) at an inferred paleo-water table and in deep saprolite zones. Yttrium and Ho fractionate substantially during all observed stages of weathering, with Y/Ho ratios ranging from 26.5 to 21.9 in the moderately weathered saprolite profile and from 30.2 to 14.7 in the laterite profile. The single strongly superchondritic Y/Ho ratio of 30.2 in the laterite is restricted to a sample at depth, and appears to fingerprint the deposition of REE derived from dissolution higher in the profile. Decrease in the Y/Ho ratio relative to the protolith basalt (24.4-24.7) in both profiles inversely correlates with chemical weathering indices, and suggests that Y/Ho ratios have significant potential as a silicate weathering proxy. Consequently, suspended vs. dissolved river loads may record the differing behaviour of these elements during weathering. (C) 2014 Elsevier B.V. All rights reserved.