The vesicular layer and carbonate collars of desert soils and pavements: formation, age and relation to climate change

Abstract

The vesicular, fine-grained A horizon (Av) is the widespread, ubiquitous surficial horizon of desert soils in diverse landforms and parent materials of varying ages. Now known to form mostly through accumulation of eolian dust, recent studies show that dust accumulation and concomitant soil development are genetically linked to stone pavement formation. Changes in the magnitude of eolian activity and effective leaching related to Quaternary climatic changes are also hypothesized to have influenced the evolution of the Av horizon. Numerical modeling, geochronologic, and field/laboratory studies elucidate the nature of pedogenic processes controlling compositional evolution of Av, how the changing Av horizon increasingly influences soil infiltration and carbonate translocation and accumulation, and the control that clasts of the evolving pavement exert on pedogenic processes. Results of a model that determines soil bulk chemical composition based on mixing of estimated proportions of externally derived (eolian) material and parent materials imply that the evolution of the soil bulk composition is strongly influenced by Av horizon formation. The early development of a weakly to moderately developed Av horizon directly over gravelly parent material in late and middle Holocene soils moderately influences soil infiltration, but significant leaching of very soluble materials and some carbonate in dust are permitted. In older, Pleistocene soils, however, the texturally more mature Av and underlying, cumulic nongravelly horizons more strongly limit the rate and depth of leaching, and soil bulk composition therefore more closely approximates a simple mixture of dust and parent material. Other aspects of Av horizon development and its relations to the pavement are evaluated through studies of pavement clasts with coatings of soil carbonate, referred to as carbonate collars. Development of a numerical model that integrates soil hydrology, a CO 2 production–diffusion model, calcite kinetics and thermodynamic considerations, composition and thermal characteristics of pavement clasts and the textural and structural properties of the surface horizon provides the basis for testing a hypothesis of collar formation. Model results, combined with results of δ 13C and δ 18O analyses of collar carbonate, demonstrate how precipitation of calcite on pavement clasts and within the Av is favored at a depth much shallower than that indicated by the classic carbonate depth–climate relationship of Jenny and Leonard [Jenny, H.J., Leonard, C.D., 1935. Functional relationships between soil properties and rainfall. Soil Science 38, 363–381] and Arkley [Arkley, R.J., 1963. Calculations of carbonate and water movement in soil from climatic data. Soil Science 96, 239–248], or simulated by numerical models of carbonate accumulation. Simultaneous development of thick carbonate collars and the Av horizon requires the sustained pavement clast–Av horizon coupling for at least centuries to possibly millennia. New thermoluminescence ages also indicate that much of the Av horizon formed in the Holocene, and that it is certainly much younger than the older Pleistocene pavements. This supports the previously proposed hypothesis that increased dust flux during the Pleistocene-to-Holocene transition triggered and/or greatly accelerated Av horizon development. An understanding of the genesis of collars provides not just an understanding of how carbonate can accumulate in surface environments, but it also provides important clues into processes of pavement evolution and preservation of Av horizons during long glacial periods. The Av horizon is not merely an insignificant surficial zone of recent dust accretion; instead, its development profoundly influences the genesis of desert soils and pavements.