William E. Lukens

A data-driven spline model designed to predict paleoclimate using paleosol geochemistry

Paleosols (fossil soils) are abundant in the sedimentary record and reflect, at least in part, regional paleoclimate. Paleopedology thus offers a great potential for elucidating high resolution, deep-time paleoclimate records. However, many fossil soils did not equilibrate with climate prior to burial and instead dominantly express physical and chemical features reflective of other soil forming factors. Current models that use elemental oxides for climate reconstruction bypass the issue of soil-climate equilibration by restricting datasets to narrow ranges of soil properties, soil-forming environments and mean annual precipitation (MAP) and mean annual temperature (MAT). Here we evaluate a data-driven paleosol-paleoclimate model (PPM1.0) that uses subsoil geochemistry to test the ability of soils from wide-ranging environments to predict MAP and MAT as a joint response with few initial assumptions. The PPM1.0 was developed using a combined partial least squares regression (PLSR) and a nonlinear spline on 685 mineral soil B horizons currently forming under MAP ranging from 130 to 6900 mm and MAT ranging from 0 to 27 °C. The PLSR results on 11 major and minor oxides show that four linear combinations of these oxides (Regressors 1-4), akin to classic oxide ratios, have potential for predicting climate. Regressor 1 correlates with increasing MAP and MAT through Fe oxidation, desilication, base loss and residual enrichment. Regressor 2 correlates with MAT through temperature-dependent dissolution of Na- and K-bearing minerals. Regressor 3 correlates with increasing MAP through decalcification and retention of Si. Regressor 4 correlates with increasing MAP through Mg retention in mafic-rich parent material. The nonlinear spline model fit on Regressors 1 to 4 results in a Root Mean Squared Error (RMSEMAP) of 228 mm and RMSEMAT of 2.46 °C. PPM1.0 model simulations result in Root Mean Squared Predictive Error (RMSPEMAP) of 512 mm and RMSPEMAT of 3.98 °C. The RMSE values are lower than some preexisting MAT models and show that subsoil weathering processes operating under a wide range of soil forming factors possess climate prediction potential, which agrees with the state-factor model of soil formation. The nonlinear, multivariate model space of PPM1.0 more accurately reflects the complex and nonlinear nature of many weathering processes as climate varies. This approach is still limited as it was built using data primarily from the conterminous USA and does not account for effects of diagenesis. Yet, because it is calibrated over a broader range of climatic variable space than previous work, it should have the widest array of potential applications. Furthermore, because it is not dependent on properties that may be poorly preserved in buried paleosols, the PPM1.0 model is preferable for reconstructing deep time climate transitions. In fact, previous studies may have grossly underestimated paleo-MAP for some paleosols.

Reconstructing pH of Paleosols Using Geochemical Proxies

Soil pH is essential for understanding weathering, nutrient availability, and biological-edaphic relationships. However, standard pH measurement on soils requires friable material, thereby excluding most paleosols. In this article, bulk geochemical proxies for pH are developed for soil B horizons using indices that track pH-dependent mineralogical transformations. Geochemical relationships within a continental-scale modern soil data set (n=619) reveal a close association between pH and log-transformed CaO and little influence of refractory oxides on pH. These results guided the formulation of three geochemical indices that consist of ratios of Fe2O3, TiO2, and Al2O3 to CaO, herein referred to as FeCa, TiCa, and AlCa. After careful screening for anthropogenic influence, pedotransfer functions relating each index to pH were derived using sigmoidal regressions on a calibration data set (n=305). Each index has similar predictive capacity for pH (r2=0.70–0.74, root mean square error=0.83–0.88). The models were cross-validated on an external testing set (n=130), which returned root mean square prediction errors (RMSPEs) similar to regression results (RMSPE=0.81–0.86). While soil pH shows a significant correlation with mean annual precipitation, partial correlation analysis of FeCa, TiCa, AlCa, and a number of widely used paleosol weathering indices revealed that the relationship between B horizon composition and pH is significant, even when climate is held constant. This finding implies that bulk geochemical indices used in pedotransfer functions for climate primarily track pH, which in turn responds to climate state. A case study is examined, where the pH transfer functions were applied on a succession of Triassic alluvial paleosols that experienced a large range of soil-forming conditions. Reconstructed pH values closely track interpreted vegetation, climate, and pedogenesis. These pedotransfer functions offer a new pathway to estimate an ecologically significant parameter in deep-time Critical Zones.

Paleopedology as a Tool for Reconstructing Paleoenvironments and Paleoecology

Soils form as a product of physical, chemical, and biological activity at the outermost veneer of Earth’s surface. Once buried and incorporated into the sedimentary record, these soils, now paleosols, preserve archives of ancient climates, ecosystems, and sedimentary systems. Paleopedology, the study of paleosols, includes qualitative interpretation of physical characteristics and quantitative analysis of geochemical and mineralogical assays. In this chapter, the paleosol macroscopic, micromorphological, mineralogical, and geochemical indicators of paleoecology are discussed with emphasis on basic analytical and interpretative techniques. These data can reveal a breadth of site-specific interpretations of vegetation, sedimentary processes, climatic variables, and durations of landscape stability. The well-known soil-forming factors are presented as a theoretical framework for understanding landscape-scale soil evolution through time. Vertical and lateral patterns of stacked paleosols that appear in the rock record are discussed in order to address practical approaches to identifying and describing paleosols in the field. This chapter emphasizes a robust multi-proxy approach to paleopedology that combines soil stratigraphy, morphology, mineralogy, biology, and chemistry to provide an in-depth understanding of paleoecology.

The Early Miocene Critical Zone at Karungu, Western Kenya: An Equatorial, Open Habitat with Few Primate Remains

Early Miocene outcrops near Karungu, Western Kenya, preserve a range of fluvio-lacustrine, lowland landscapes that contain abundant fossils of terrestrial and aquatic vertebrates. Primates are notably rare among these remains, although nearby early Miocene strata on Rusinga Island contain a rich assemblage of fossilized catarrhines and strepsirrhines. To explore possible environmental controls on the occurrence of early Miocene primates, we performed a deep-time Critical Zone (DTCZ) reconstruction focused on floodplain paleosols at the Ngira locality in Karungu. We specifically focused on a single stratigraphic unit (NG15), which preserves moderately developed paleosols that contain a microvertebrate fossil assemblage. Although similarities between deposits at Karungu and Rusinga Island are commonly assumed, physical sedimentary processes, vegetative cover, soil hydrology, and some aspects of climate state are notably different between the two areas. Estimates of paleoclimate parameters using paleosol B horizon elemental chemistry and morphologic properties are consistent with seasonal, dry subhumid conditions, occasional waterlogging, and herbaceous vegetation. The reconstructed small mammal community indicates periodic waterlogging and open-canopy conditions. Based on the presence of herbaceous root traces, abundant microcharcoal, and pedogenic carbonates with high stable carbon isotope ratios, we interpret NG15 to have formed under a warm, seasonally dry, open riparian woodland to wooded grassland, in which at least a subset of the vegetation was likely C4 biomass. Our results, coupled with previous paleoenvironmental interpretations for deposits on Rusinga Island, demonstrate that there was considerable environmental heterogeneity ranging from open to closed habitats in the early Miocene. We hypothesize that the relative paucity of primates at Karungu was driven by their environmental preference for locally abundant closed canopy vegetation, which was likely absent at Karungu, at least during the NG15 interval if not also earlier and later intervals that have not yet been studied in as much detail.