Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published February 2000 | public
Journal Article

Geology and geochemistry of Paleosols developed on the Hekpoort Basalt, Pretoria Group, South Africa

Abstract

The Hekpoort paleosols comprise a regional paleoweathering horizon developed on 2.224 ± 0.021 Ga basaltic andesite lavas at the top of the Hekpoort Formation of the Pretoria Group, Transvaal Supergroup, South Africa. In five separate profiles, from outcrops along road cuts near Waterval Onder and the Daspoort Tunnel and in three drill cores from the Bank Break Area (BB3, BB8, and BB14), the top of the paleosol is a sericite-rich zone. The sericite zone grades downward into a chlorite-rich zone. In core BB8 and in the road cut at the Daspoort Tunnel, we sampled the underlying or parent basaltic andesite into which the chlorite zone grades. We did not obtain samples of the parent material at Waterval Onder and in cores BB3 and BB14, but chemical analyses indicate that the chlorite and sericite zones in these profiles derive from underlying lavas similar to the ones we sampled in core BB8 and at the Daspoort Tunnel. The presence of apparent rip-up clasts of the paleosol in the overlying ironstones of the Strubenkop Formation in the cores from Bank Break makes it very unlikely that most of the alteration was a result of interactions with hydrothermal fluids. Desiccation cracks at the top of the paleosol that were filled with sand during the deposition of the overlying sediments at Waterval Onder point to a subaerial weathering origin. Very little, if any, Al, Ti, Zr, V, or Cr moved a discernible distance during weathering of any of the five profiles. The vertical distribution of Fe, Mg, Mn, Ni, and Co indicates that these elements were largely removed from the top of the soil during weathering. The overall abundance of these elements in each of the profiles indicates that a significant fraction of the complement lost from the top subsequently reprecipitated in the lower portion of the soil as constituents of an Fe^(2+)-rich smectite. The loss of Fe from the top of the soil during weathering of the Hekpoort paleosols indicates that atmospheric P_(O_2) was less than 8 x 10^(-4) atm about 2.22 Ga. Fe^(2+)-rich smectite should only precipitate during soil formation if atmospheric P_(CO_2) is less than or equal to 2 x 10^(-2) atm (Rye, Kuo, and Holland, 1995). Ca and Na were largely lost during weathering. Some Na was apparently added to the sericite zone in cores BB3, BB8, and BB14 after weathering. All five profiles are enriched in K and Rb, and most are enriched in Ba. The distribution of these elements indicates that they all were added during post-weathering hydrothermal metasomatism. Rb-Sr analysis of the paleosol at the Daspoort Tunnel indicates that metasomatism last affected that profile 1.925 ± 0.032 Ga (Macfarlane and Holland, 1991).

Additional Information

© 2000 American Journal of Science. We would like to thank John Hayes, Stein Jacobsen, and Andrew Knoll for their helpful comments on an earlier draft of this manuscript. Many thanks are owed to Mike Vollinger who ran most of the XRF major element analyses at University of Massachusetts Amherst and to Sue Hall and Sharon Farrell who performed the XRF trace element analyses at the University of the Witwatersrand. Janise Lee, Linda Feng, and Bill Croft all helped with XRD analyses. Dave Lange provided considerable assistance with the electron microprobe. Dennis Bird, Donald Lowe, and Gregory Retallack provided valuable reviews that helped define the final shape of the manuscript. Support for this project came from NASA grants NAGW-599 and NAG5-4174 and from Harvard University's Shaw Fund.

Errata

These errata have been written to correct the following in this paper (February, 2000, p. 85-142): Please substitute the following on pages 98 and 99—starting with the last paragraph on page 98: ''We have analyzed nine protolith samples in core BB8 and three protolith samples in the Daspoort Tunnel section. Element-element plots of TiO2 versus Al2O3 and Zr versus Al2O3 for samples from core BB8 and of TiO2 versus Al2O3 for samples from the Daspoort Tunnel are depicted in figures 10 and 11 respectively. In both plots, all the points representative of samples from the sericite zone (�) and from the chlorite zone (✠) fall within the range defined by the tangent lines around the parent rock (�) points. These data indicate that both soil profiles developed in place on the underlying basaltic andeiste and that they contain little or no allochthonous material. The data for the ironstone (�) in core BB8 fall well off the trends defined by the protolith and paleosol and indicate that, though it may contain rip-up clasts of the paleosol, the bulk of the ironstone is not genetically related to the paleosol.'' Please substitute the attached figure 10 for the one now on page 107 and the attached figure 14 for the one on page 110.

Additional details

Created:
August 19, 2023
Modified:
October 25, 2023