JOURNAL OF SEDIMENTARY RESEARCH,VOL. 75, NO. 3, MAY, 2005, P. 339–349 Copyright  2005, SEPM (Society for Sedimentary Geology) 1527-1404/05/075-339/$03.00 DOI 10.2110/jsr.2005.027 DISTINGUISHING CLIMATE IN THE SOIL RECORD USING CHEMICAL TRENDS IN A VERTISOL CLIMOSEQUENCE FROM THE TEXAS COAST PRAIRIE, AND APPLICATION TO INTERPRETING PALEOZOIC PALEOSOLS IN THE APPALACHIAN BASIN, U.S.A. STEVEN G. DRIESE, 1 LEE C. NORDT, 1 WARREN C. LYNN, 2 CYNTHIA A. STILES, 3 CLAUDIA I. MORA, 4 AND LAWRENCE P. WILDING 5 1 Department of Geology, Baylor University, Waco, Texas 76798-7354, U.S.A. 2 National Soil Survey Laboratory, National Soil Survey Center, 100 Centennial Mall North, Lincoln, Nebraska 68508-3866, U.S.A. 3 Department of Soil Science, University of Wisconsin, Madison, Wisconsin 53706-1299, U.S.A. 4 Department of Earth and Planetary Sciences, University of Tennessee, Knoxville, Tennessee 37996-1410, U.S.A. 5 Department of Soil and Crop Sciences, Texas A & M University, College Station, Texas 77843-2474, U.S.A. ABSTRACT: A suite of Vertisols (clay-rich soils with high shrink-swell potential) were examined across a climosequence (climatic transect) in twelve soil pits from the Coast Prairie of Texas in order to determine if mean annual precipitation (MAP) exerts a control on the chemistry of these soils, and if the observed chemical trends are useful for inter- preting paleoclimate records of paleoVertisols in the geologic record. The precipitation regime of the climosequence spans a range between 144 and 86 cm/year, with moisture regimes classified as udic, udic– ustic, ustic, and aridic–ustic, in a general northeast to southwest di- rection. Other soil-forming factors, such as soil age (35–40 ka), par- ent material (fluviodeltaic Beaumont Formation of late Pleistocene age), landscape (low-relief coastal plain), and vegetation (prairie or mixed woody shrubs), are relatively constant across the climosequence. Climate-sensitive chemical proxies of MAP identified include dithionite citrate-extractable Fe (Fe dith ), acid oxalate-extractable Fe (Fe oxal ), CaCO 3 equivalent (CaCO 3equiv ), S, and ammonium acetate-extractable Na, K, and Mg (Na acet ,K acet , and Mg acet , respectively), which vary across the climosequence because of differences in effective depths of leaching and intensity of wetting and drying cycles. These standard USDA wet-chemical climate proxies are related to bulk (oxide or ele- ment) chemistry of soils and paleosols measured using XRF, which supports the use of geochemical climate proxies for interpreting the paleoclimate records of paleoVertisols. Application of the chemical in- dex of alteration minus potash (CIA-K) geochemical climofunction to late Mississippian paleosols from the Appalachian basin of the eastern U.S. demonstrates evidence for a shift from a lower to a higher MAP paleoclimate that is consistent with previous paleoclimate models and with observed morphological changes in the paleosols. We advocate actualistic research using bulk chemistry of soils and paleosols as a means of obtaining soil information useful for interpreting paleosols in the geological record. INTRODUCTION Vertisols (clay-rich soils with high shrink-swell potential) occur in many climatic settings and have distinctive pedogenic characteristics, including prominent bowl-like structures (gilgai), randomly oriented planar slicken- sides, and pedogenic carbonate and Fe-Mn oxide nodules (Lynn and Wil- liams 1992; Coulombe et al. 1996a, 1996b; Soil Survey Staff 1998). An- cient Vertisol equivalents (i.e., paleoVertisols) have been widely identified in geologic successions, including Paleozoic rocks of the Appalachian Ba- sin in the eastern U.S., and effectively preserve morphologic, micromor- phologic, and geochemical characteristics formed within their paleopedo- genic settings (Driese and Foreman 1992; Driese and Mora 1993; Caudill et al. 1996; Mora and Driese 1999; Driese et al. 1992; Driese et al. 2000; Driese et al. 2003; Stiles et al. 2001), even when buried to relatively great depths (to 10 km; Caudill et al. 1997; Mora et al. 1996; Mora et al.1998). This preservation potential, combined with the dependence of some aspects of Vertisol chemical properties on climate, including the total Fe content of pedogenic Fe-Mn nodules and depth to calcic horizon (Stiles et al. 2001), total mass flux of elements on a total-soil basis (Stiles et al. 2003a), Ti/Zr contents vs. depth (Stiles et al. 2003b), and mass-balance on reagent-ex- tractable soil chemistry using different particle size fractions (Nordt et al. 2004), collectively suggest that Vertisol geochemistry may be employed as a paleoclimate proxy. In order to fully utilize this paleoclimate proxy for interpreting the geologic (paleosol) record, one must first examine climate- dependent chemical trends in Quaternary (Pleistocene–Holocene) Vertisols formed in well-defined pedogenic settings. OBJECTIVES Our primary objective is to interpret regional variations in soil chemistry of a Quaternary Vertisol climosequence, i.e., a transect of Vertisols in which major soil-forming factors, with the exception of climate (of which mean annual precipitation (MAP) is one component of climate), are held constant (Stiles 2001; Stiles et al. 2001, 2003a; Stiles et al. 2003b), and to test the applicability of derived climate proxy measures for interpreting the paleoclimate record from ancient (lithified) Vertisols preserved as paleo- sols. MAP is one of the most significant of the soil-forming factors that influence pedogenesis (Birkeland 1999), and we hypothesize that soil con- stituents are therefore leached to varying depths or mobile to varying de- grees depending upon MAP and intensity of wetting and drying cycles (i.e., climosequence trends are manifested by MAP-dependent patterns in soil chemistry). A second objective is to show geoscientists how to use the standard soil wet-chemical characterization data that are readily available from the Na- tional Soils Database maintained by the Natural Resources Conservation Service (NRCS) of the U.S. Department of Agriculture (USDA) to interpret chemical patterns in modern soils considered as analogs for paleosols. We also demonstrate how standard total soil analysis (X-ray fluorescence anal- yses of bulk soil expressed as wt% element or oxide) can be related to soil wet-chemical data, and, by extrapolation, related to geochemical patterns observed in paleosols. A third objective is to test whether or not the geochemical proxy for paleoprecipitation estimation proposed recently by Sheldon et al. (2002) has predictive capability for the Texas Vertisol climosequence, and by in- ference, for paleoclimate interpretations based upon the geochemistry of Paleozoic paleoVertisols from the Appalachian basin of the eastern U.S. Sheldon et al.’s (2002) proxy, defined as the ‘‘chemical index of alteration minus potash’’ (CIA-K), was developed using total soil (oxide or element) analyses from a suite of USDA Mollisols, Alfisols, and Ultisols, but did not specifically include Vertisols. Finally, we hope that our study illustrates how valuable linkages can be established when geologists work together with soil scientists, and with the NRCS in general. With this study we show that the strengthening of col- laborations between the two research groups can yield valuable information for interpreting the climate records of both modern and ancient soils. LOCATION OF STUDY AREA AND METHODS Texas Vertisols The Coastal Prairie region of Texas is the ideal choice for examining a climosequence through modern Vertisols for a variety of reasons. It rep-