• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Molecular and isotopic indicators of paleoenvironmental change in low-organic-carbon soils with applications to Pleistocene archaeological sites in Greece, Algeria, and Ethiopia

ProQuest Dissertations and Theses, 2009
Dissertation
Author: Melanie Amber Everett
Abstract:
Reconstruction of terrestrial paleoenvironments in ancient soils (paleosols) is analytically challenging if the organic carbon content is less than ~0.5% (low-organic-carbon soils). Lipid biomarkers derived from vascular plant waxes, specifically n -alkane homologues C25 through C 33 , combined with stable-isotopic analyses of bulk soil organic matter (SOM) and pedogenic carbonates can enhance paleoenvironmental interpretations of paleosols. This dissertation presents three case studies of low-organic-carbon paleosols for which a combined approach is employed in paleoecosystem assessment. The first study reconstructs Plio-Pleistocene paleoenvironments at the archaeological locality of El-Kherba in Ain Hanech, Algeria. The abundance of in -alkanes and preference for odd-carbon numbered molecules combined with stable-carbon-isotopic analyses of bulk SOM and stable-carbon- and stable-oxygen-isotopic analyses of pedogenic carbonates indicate that a coastal Mediterranean ecosystem existed at El-Kherba. The second case study focuses on Pleistocene paleosols from the upper Busidima Formation at the early hominid site of Gona, Ethiopia. Using data similar to those for the Algerian study, molecular and stable-isotopic analyses indicate that vegetation varied through time in the upper Busidima as a result of lateral migrations of the paleo-Awash River through the Awash Valley. The third case study considers vegetation and rapid climate fluctuations as recorded by a stacked sequence of gravels and terra rossa paleosols at the Morphi Pleistocene site in Epirus, Greece. Vegetation changes are closely coincident with regional glacial and interglacial climate fluctuations recorded for western Epirus. Taken together, these three biogeochemical studies demonstrate the potential of biomarker analyses to form the basis of detailed paleoenvironmental reconstructions for low-organic-carbon paleosols.

viii Table of Contents

Page Chapter 1. Molecular and stable-isotopic approaches to reconstructing terrestrial paleoenvironments as recorded in low-organic-carbon paleosols

A. Introduction 1 B. Soil factors and preservation of soil organic matter 2 C. Lipid biomarkers in soil organic matter 5 D. Paleoenvironmental n-alkane indicators 8 E. Biomarker contributions to stable-carbon-isotopic analyses of bulk organic matter 10 F. Conclusions 13 G. Figures 15 H. References Cited 17 Chapter 2. Overview of paleosol case studies A. Introduction 21 B. Plio-Pleistocene paleosols at the El-Kherba archaeological locality, Ain Hanech, Algeria 22 C. Pleistocene paleosols in the upper Busidima Formation, Gona, Ethiopia 24 D. Late Pleistocene terra rossa paleosols at Morphi, Epirus, Greece 26 E. References Cited 29 Chapter 3. Plio-Pleistocene paleoenvironmental change inferred from biogeochemical evidence at the archaeological site of Ain Hanech, Algeria

A. Introduction 31 B. Background 32 C. Materials and Methods 43 D. Results and Interpretations 46 E. Discussion 54 F. Conclusions 58 G. Tables 60 H. Figures 64 I. References Cited 72 Chapter 4. Molecular and stable-isotopic approaches to paleoenvironmental reconstruction of Pleistocene paleosols from the upper Busidima Fm., Gona, Ethiopia

A. Introduction 79 B. Background 81 C. Materials and Methods 95 D. Results and Interpretations 94 E. Discussion 105 F. Conclusions 112 G. Tables 114 H. Figures 117

i x I. References Cited 125 Chapter 5. A late Pleistocene record of paleovegetation response to glacial cycles as recorded in terra rossa paleosols at the Middle Paleolithic site of Morphi, Greece

A. Introduction 132 B. Background 133 C. Materials and Methods 145 D. Results and Interpretations 148 E. Discussion 154 F. Conclusions 158 G. Tables 160 H. Figures 165 I. References Cited 173 Appendices 179

x Tables

Page 3.1. Summary statistics for stable carbon and oxygen isotope analyses of El-Kherba pedogenic carbonates. 60 3.2. Summary statistics for total organic carbon and stable carbon isotope analyses of El-Kherba paleosol organic matter samples. 60 3.3. Abundances of n-alkane biomarkers in the El-Kherba sediment samples. 61 3.4. Standard indices for n-alkane biomarkers in the El-Kherba sediment samples. 62 3.5 Environmental indices for n-alkane biomarkers in the El-Kherba sediment samples. 63 4.1. Abundances of n-alkane biomarkers and major contaminants in the upper Busidima paleosol samples. 114 4.2. Summary statistics for total organic carbon and stable carbon isotopic analyses of bulk organic matter in upper Busidima Fm. paleosols. 114 4.3. Summary statistics for stable carbon and oxygen isotopic analyses of pedogenic carbonates from the upper Busidima Fm. 115 4.4. Standard indices for n-alkane biomarkers in the upper Busidima paleosol samples. 115 4.5 Environmental indices for n-alkane biomarkers in the upper Busidima paleosol samples. 116 5.1. Maturity Stages for Bt horizons of Quaternary paleosols in Greece. 160 5.2. Collection data for Morphi sediment samples. 160 5.3. Abundances of n-alkane biomarkers in the Morphi sediment samples. 161 5.4 Morphi sediment sample total carbon, total organic carbon, and bulk organic matter stable carbon isotopic values. 162 5.5 Standard indices for n-alkane biomarkers in the Morphi sediment samples. 163 5.6 Environmental indices for n-alkane biomarkers in the Morphi sediment samples. 164

xi Figures

Page 1.1 Descriptive shorthand for labeling master soil horizons and descriptions of characteristics of each horizon. 15 1.2 Chemical structures of compounds discussed in Chapter 1. 16 3.1. Geographical location of the Ain Hanech site and study area. 64 3.2. Composite lithographic profile of the Ain Hanech Formation and associated geochronology. 65 3.3. Composite microstratigraphic profile for El-Kherba, showing paleosol horizons and major pedological features. 66 3.4. #

$(#)" 67 3.5 #

" 67 3.6. El-Kherba pedogenic carbonate stable carbon and stable oxygen isotopic values plotted against the major paleosol intervals. 68 3.7. n-Alkane molecular abundances plotted for C 25 :C 33 homologues for major El-Kherba paleosol horizons. 69 3.8. Bulk organic matter stable carbon isotopic values and biomarker vegetation index Q grass/plant plotted against the El-Kherba microstratigraphic profile. 70 3.9. Pedogenic carbonate stable carbon and stable oxygen isotopic values from Ain Hanech compared to those of contemporaneous East African Plio-Pleistocene sites. 70 3.10. El-Kherba pedogenic carbonate stable carbon and stable oxygen isotopic values plotted against pedogenic carbonates from modern climatic zones. 71 4.1. Geologic overview map of the Gona study area, showing the aerial extent of the Busidima North exposures considered in this study. 117 4.2 Chart showing the thickness and ages of major geologic formations present at Gona and their associated dominant depositional environments. 118 4.3. Composite stratigraphic section of the upper Busidima Fm., showing major tuffs and vertical placement of paleosol and pedogenic carbonate sample locations. 119 4.4. Illustrations of savanna ecosystem classifications shown with composite ranges of pedogenic carbonate and soil organic matter stable carbon isotopic values. 120 4.5. #

#&&" 121 4.6. #

" 121 4.7. n-Alkane molecular abundances plotted for C 25 :C 33 homologues relative to major contaminant DEHP in upper Busidima Fm. paleosol samples. 122

xii 4.8. Stable carbon isotopic values for pedogenic carbonate and paleosol organic matter, and biomarker vegetation index Q grass/plant

values plotted for each stratigraphic unit. 123 4.7. Pedogenic carbonate stable carbon and oxygen isotopic values from the upper Busidima Fm. compared to those from contemporaneous East African Pleistocene sites. 124 5.1. Map showing the geographic location of Morphi relative to natural features and major city centers in Epirus, Greece. 165 5.2 Diagram showing the Morphi site formation processes. 165 5.3. Microstratigraphic profile of the Morphi site showing the position of samples analyzed in this study and designations of major paleosols. 166 5.4. Correlation table showing the relationship between fragmentary glacial and periglacial sequences in the Pindus Mountains, the continuous lacustrine parasequence in nearby Lake Ioannina, and the paleosols in the Morphi section. 167 5.5. #

%%. 168 5.6. #

" 168 5.7. n-Alkane molecular abundances plotted against C 25 :C 33

homologues for non-pedogenic deposits at the Morphi site. 169 5.8 n-Alkane molecular abundances plotted against C 25 :C 33

homologues for pedogenic deposits at the Morphi site. 170 5.9.

#

&(!&*!  '% " 171 5.10. n-Alkane molecular abundances plotted against C 25 :C 33

homologues for soil horizons within paleosol P7 at the Morphi site. 172 5.11. Average chain length (ACL), biomarker vegetation index Q grass/plant , and bulk organic matter stable carbon isotopic values for Morphi sediment samples. 172

xiii Appendices

Page A. Field photographs of El-Kherba (Ain Hanech, Algeria) microstratigraphic profile. 180 B. Stable carbon and oxygen isotope values for pedogenic carbonate nodules collected from El-Kherba (Ain Hanech, Algeria). 181 C. X-ray diffraction analyses of selected El-Kherba (Ain Hanech, Algeria) pedogenic carbonate nodules. 182 D. Total carbon, total organic carbon and stable carbon isotope results for paleosol organic matter samples collected from El Kherba (Algeria). 183 E. Concentrations of n-alkane homologues extracted from El-Kherba (Ain Hanech, Algeria) paleosol samples. 184 F. Molecular abundance data for n-alkanes extracted from El- Kherba (Ain Hanech, Algeria) paleosols. 185 G. Aerial photograph of the upper Busidima Fm. (Gona, Ethiopia) showing the locations of soil pits where paleosol and pedogenic carbonate sample were collected. 186 H. Upper Busidima Fm. (Gona, Ethiopia) sections showing the vertical positions of soil pits where paleosol and pedogenic carbonate samples were collected. 187 I. Field photographs of major geologic features in the upper Busidima Fm. (Gona, Ethiopia). 188 J. Upper Busidima Fm. paleosol samples collected during the 2004 field season, with associated stratigraphic package and pedologic descriptions. 189 K. Concentrations of n-alkane homologues extracted from Upper Busidima Fm. (Gona, Ethiopia) paleosol samples. 193 L. Molecular abundances of n-alkane homologues extracted from upper Busidima Fm. (Gona, Ethiopia) paleosols. 194 M. Total carbon, total organic carbon content and stable carbon isotope results for paleosol bulk organic matter samples collected from upper Busidima Fm. (Gona, Ethiopia). 195 N. X-ray diffraction analyses of 6N HCl-acidified and non-acidified paleosol samples from the Upper Busidima Fm. (Gona, Ethiopia). 196 O. Stable carbon and oxygen isotope results for pedogenic carbonate nodules collected from the Upper Busidima Fm. (Gona, Ethiopia). 197 P. X-ray diffraction analyses of selected pedogenic carbonates from the Upper Busidima Fm. (Gona, Ethiopia). 198 Q. Field photographs of the Morphi site (Epirus, Greece) and surrounding area. 199 R. Concentrations of n-alkane homologues extracted from sediments at Morphi (Epirus, Greece). 200

xiv S. Molecular abundance data for n-alkane homologues extracted from sediment samples from Morphi (Epirus, Greece). 201

1 Chapter 1. Molecular and stable-isotopic approaches to reconstructing terrestrial paleoenvironments as recorded in low-organic-carbon paleosols

A. Introduction Soils are unique amongst terrestrial deposits as they form in response to subaerial conditions that are environmentally mediated and, thus, act as an archive for biological and physical microenvironments. Soil organic matter (SOM) derived from in situ animal, plant, fungal, and bacterial decomposition provides a direct link to biomass present during soil formation and is often employed as an indicator of terrestrial environments. Despite its value in modern terrestrial studies, the low preservation potential of SOM in ancient soils or paleosols has discouraged its use as a paleoenvironmental indicator. For the Quaternary terrestrial record, of which paleosols comprise a large fraction, the inability to reliably utilize SOM as a paleoenvironmental indicator has greatly limited the scope and scale of terrestrial environmental reconstruction. Reconstruction of terrestrial paleoenvironments is crucial to addressing questions regarding land-based organismal evolution, ecology, and climate, yet it remains a difficult task due to low preservation potential of organic environmental indicators in most terrestrial settings. In particular, autochthonous indicators useful in determining local environments, such as biogenic remains of subaerial ecosystems, are often removed or degraded in terrestrial settings via surficial processes. As one solution to the problem of low SOM preservation in paleosols, stable-carbon-isotopic analyses of SOM and associated proxies are rapidly becoming standard practice in terrestrial geologic studies and have been used in numerous settings to provide environmental context for terrestrial deposits. Environmental reconstruction via stable-carbon-isotopic analysis is based on carbon-isotopic fractionation in terrestrial higher plants due to different photosynthetic pathways that have evolved in response to varied climate and soil conditions (Ehleringer, 1991). The potential to utilize low amounts of organic matter from soils and paleosols for stable-carbon-isotopic analysis has improved considerably due to advances in analytical instrumentation and clean-handling techniques. Using these techniques, soils that contain less than ~0.5% total organic carbon (low-organic-carbon soils) still retain

2 enough organic matter to reconstruct vegetation types present during soil formation, assuming that SOM is representative of original plant biomass. A major drawback of stable-carbon-isotopic analysis of SOM, however, is that the technique does not identify or quantify sources of organic carbon contributing to SOM. Furthermore, stable-carbon- isotopic analyses cannot distinguish between vegetation types in plant communities that use the same photosynthetic pathway. Molecular (compound-specific) analyses have the potential to provide a necessary analytical context for stable-isotopic analyses of low-organic-carbon soils and paleosols. Molecular analyses of SOM allow identification and quantification of biological compounds (biomarkers) derived from soil biomass, including contributions from plants, fungi, and various microbial organisms. Certain compound classes of biomarkers are known to survive in sediments for hundred millions of years and are relatively resistant to microbial and diagenetic degradation (Eglinton & Eglinton, 2008). Lipids derived from plant waxes have been particularly effective in reconstruction of both paleovegetation and climate patterns from loess, lacustrine and marine sediments. As such, molecular analyses have excellent potential as organic paleoenvironmental indicators for low- organic-carbon soils and paleosols. Although molecular characterization of organic matter has been used to examine relative contributions of higher plant matter to marine sediments and loess sequences, it has rarely been employed in paleoenvironmental reconstructions of low-organic-carbon soils and paleosols. This chapter reviews the applicability of molecular and stable isotopic analyses as a combined approach to paleoenvironmental reconstruction of low- organic-carbon soils and paleosols.

B. Soil factors and preservation of soil organic matter The utility of soils in reconstructing terrestrial environments derives from properties related to major soil-forming factors:

S = f cl,o,r,p,t ( ) Equation 1.1

3 where S is soil; cl, climate; o, organisms; r, topography; p parent material; and t, time (Jenny, 1941). These soil-forming factors highlight differences between soils and other terrestrial deposits: (1) soils reflect both physical and biological environments; (2) soils record environments specific to soil location (local or microenvironments); and (3) soils are products of weathering over time and, therefore, record time-averaged environments. Climate is a key determinant of soil characteristics, as precipitation and ambient temperature control rates of chemical weathering. Climate also plays a large role in determining the types of macro- and micro-organisms contributing to soil formation. Vegetation, in particular, is considered the most important biotic factor in soil formation and can affect both soil morphology and soil chemistry (Birkeland, 1999). The climate- vegetation-soil relationship shown here:

highlights the roles soils play in both influencing local vegetation types and as archives of climatic and vegetation change. According to this same relationship, SOM derived primarily from in situ vegetation also reflects characteristics of the climate and soil. The relationships of climate, vegetation, and soil inherent to soil formation are bases for interpreting paleoenvironmental change using both paleosols and ancient SOM. The potential for SOM in soils to reliably record environmental change is dependent upon the temporal scale of change and unbiased preservation of the organic matter. Soils, by definition, form as a time-averaged response to environmental conditions. Contribution of organic matter to soils is also time-averaged, such that organic matter in soils represents the total accumulated and preserved biotic input to soil over the period of formation. In cases where environmental fluctuations are on a smaller temporal scale than that of pedogenic development, environmental reconstruction using soils and SOM is not likely to record those changes. If, however, the temporal scale of

4 climatic or vegetation change is equal to or greater than the duration of soil formation, environmental reconstruction using soils or SOM as indicators is more likely to be successful. Time-averaged indicators, such as SOM, are also useful in cases of environmental change when rates of organic matter turnover are slow. For example, in forested or tundra ecosystems in the northern or high latitudes, changes to new vegetation may be gradual and soil properties related to the former vegetation might persist for thousands of years (Birkeland, 1974). Long periods of soil development (100s – 1000s years) increase the probability of documenting slow vegetation turnovers, provided that enough time has elapsed to overprint evidence of previous vegetation (Retallack, 1988). Preservation of SOM in soils or paleosols is dependent upon numerous inter- related factors including climate, vegetation biomass, microbial activity, parent material, soil mineralogy, and diagenetic effects (in paleosols). These effects of these factors are most simply expressed in soil horizonation. Horizonation or vertical morphological differences in soils due to weathering is an excellent predictor of the vertical distribution of organic matter in soils. Typical master horizons in soils consist of an O horizon, an A horizon, an E horizon, a B horizon, a K horizon, and a C horizon, in order from the surface to the base of a soil profile (Figure 1.1). Climate and vegetation control the thickness and maturity of soil horizons, such that soils developed in temperate regions supporting dense vegetation (e.g., Spodosols or Alfisols) will develop thicker O and A horizons than soils forming in arid regions with little vegetation, which typically develop thin or even non-existent A horizons (e.g., Aridisols) (Birkeland, 1999). O horizons and A horizons preserve the greatest abundance of organic matter (~10% to 30% organic matter) and thus yield SOM that is most representative of organic biomass during soil formation (Wang et al., 2000). In contrast to O and A horizons, E, B and K horizons typically preserve low amounts of organic matter due to leaching and accumulation of minerals (Birkeland, 1999). Low-organic-carbon soils or paleosols result from climatic and vegetational conditions that either inhibit formation of O and A horizons or encourage loss of O and A horizons through increased surface erosion, typically leaving behind organic-poor B horizons. Environments leading to formation of low-organic-carbon soils are found throughout the world and are the predominant conditions in areas such as the Southwest

5 region of the United States, southern regions of South America, sub-Saharan Africa, Eurasia and Australia. Low preservation potential of SOM in these soils necessitates use of appropriate preparative and analytical techniques to determine primary sources of SOM and to interpret environments from SOM-based indicators, such as stable-carbon- isotopic analyses. Certain SOM-derived molecular compounds have high survivability in comparison with other fractions of organic matter (refractory compounds) and have been shown to remain in low-organic-carbon soils for millions of years (Bada, 1991; Eglinton & Eglinton, 2008). Lipid biomarkers, as particularly resistant biomolecules, can provide both source-specific characterization of SOM as well as additional environmental information when used in combination with stable-carbon-isotopic analyses of SOM.

C. Lipid biomarkers in soil organic matter SOM produced mainly by plant productivity and respiration as well as microbial aerobic and anaerobic respiration and decomposition includes biological compounds (biomarkers), which are distinctive biomolecules derived from living organisms. Animals, vascular plants (plants with vascular tissues for circulation of resources), and microbes present in sediments during the period of soil formation affect the composition of SOM through production of biomolecules and biopolymers and selective loss of metabolizable compounds. Molecular analysis of these biochemicals offers a means of characterizing major sources of soil organic carbon as well as reconstructing vegetation present during soil formation. Approximately half of the input of organic matter to soils is in situ debris and exudates from root systems of plants, although falling debris also contributes (Hedges & Oades, 1997). Roots introduce organic matter by dying or sloughing tissue, rapid turnover of root hairs, and exuding mucilage and soluble biochemicals directly into the surrounding soil. A key issue in biomarker analyses of low-organic-carbon soils is distinguishing between local higher plant matter in ancient soils and plant matter introduced through modern root systems (Goodfriend, 1999; Liu & Huang, 2008). Plant waxes, which cover surfaces of leaves and stems, provide a means of characterizing subaerial vegetation present during soil formation and have been used as indicators of terrestrial higher plant input in ancient sediments (Eglinton & Hamilton, 1967; Huang et

6 al., 2001; Huang et al., 1999; Liu & Huang, 2008; Simoneit, 1977; Xie et al., 2003; Zhang et al., 2008). Plant waxes are thin, waxy layers that coat stems, leaves, flowers, and fruits of most plants, sometimes imparting a bluish-white cast to surfaces on which they occur (Eglinton & Hamilton, 1967). Functions of plant waxes are to help minimize water loss from the plant, as well as to protect against insects, bacteria, fungi, and other damaging agents. Plant waxes may constitute a fraction of a percent to several percent of the dry weight of a plant, depending upon plant species and climate (Eglinton & Hamilton, 1967). They are heterogeneous in chemical composition, being mainly composed of long, straight-chain alkanes, alkanols, and alkanoic acids with 24-36 carbon atoms (Eglinton & Eglinton, 2008) (see Figure 1.2):

n - alkane CH 3 (CH 2 ) n CH 3 n - alkanol CH 3 (CH 2 ) n CH 2 OH n - alkanoic acid CH 3 (CH 2 ) n CHO 2 H

As long-chain lipids, waxes have low water solubility, negligible volatility (for compounds containing greater than ~20 carbon atoms), chemical inertness and resistance to biodegradation - all characteristics that improve preservation potential for lipid biomarker compounds (Eglinton & Eglinton, 2008). Sources of lipid biomarkers in soils include algae, fungi, bacteria and archaea, in addition to vascular plants. Chain-length distributions of lipid biomarkers are useful in determining whether microbial or vascular plant organic matter predominates SOM in low-organic-carbon soils. Biomarkers diagnostic of terrestrial vascular-plant waxes include C 26 -C 32 n-alkanoic acids and alkanols with even/odd chain length predominance, and C 27 -C 33 n-alkanes with odd/even preference (Eglinton & Hamilton, 1967; Kolattukudy, 1969). The characteristic odd/even carbon-number distribution is a consequence of the universal polyketide (acetate, malonate – see Figure 1.2) biosynthetic pathway (Kolattukudy, 1969). The odd numbered n-alkanes are formed by the loss of a single carbon from precursor even-numbered n-alkanoic acids (Kolattukudy, 1969). Components with shorter chain lengths (< C 24 ) are derived primarily from algal, fungal,

7 and bacterial contributions (Canuel & Martens, 1993; Gong & Hollander, 1997; Meyers, 1997; Volkman et al., 1998; Volkman et al., 1980). The relative input from vascular plants versus other organisms to lipid biomarkers is typically expressed in terms of the Carbon Preference Index (CPI):

CPI = 2 C n [ ]

( ) C n-1 [ ] + C n+1 [ ]

Equation 1.2

where n for n-alkanoic acids and alkanols is equal to 26 and 28, and n for n-alkanes is equal to 27, 29, and 31. An indigenous CPI value less than 3 in immature soils or non- polluted (petroleum-based materials) sediments indicates significant algal or bacterial input to total lipid abundance (Feakins et al., 2007). Concentrations of other lipid classes, such as phospholipid fatty acids (PLFAs), may also signal the presence of an active microbial community in soils, although PLFAs decompose quickly and are, thus, primarily of use in modern soils (e.g., Boschker & Middleburg, 2002; Bouillon & Boschker, 2006; Kaneda, 1991; Kramer & Gleixnre, 2008; Patra et al., 2008; Zink et al., 2008). Lipids derived from vascular plant waxes are easily transported by winds and may represent vegetation from large geographic areas (Simoneit, 1977). Significant eolian input to modern soils may be gauged by the presence of hydrocarbons derived from aerosolized petroleum residues. The contribution of petroleum residues to modern soils is calculated using the index C w/n , which is based on the assumption that straight-chain hydrocarbons derived from fossil fuels have a CPI value of 1:

C w/n [ ] = 0.5 C n1 [ ] + C n+1 [ ] ( ) Equation 1.3

where [C w/n ] represents the concentration of molecular fossils originating from higher plants and C n , C n-1 and C n+1 are homologous compounds in samples which represent the total abundance of exotic hydrocarbon and vascular plant sources (Schneider et al., 1983; Sicre et al., 1987; Simoneit et al., 1991; Xie et al., 2000; Zhang et al., 2008). Determining contributions of eolian-derived lipids to paleosols is more difficult. Often

8 an assumption must be made based upon the geographic, depositional, and pedogenic context of paleosols as to whether the primary lipid source is in situ or eolian-derived vegetation. In several cases, the index C w/n has been used to discriminate between autochthonous molecular fossils originating from higher plants and allochthonous, eolian- derived plant waxes (Xie et al., 2002; Zhang et al., 2008). This application of the C w/n

Full document contains 220 pages
Abstract: Reconstruction of terrestrial paleoenvironments in ancient soils (paleosols) is analytically challenging if the organic carbon content is less than ~0.5% (low-organic-carbon soils). Lipid biomarkers derived from vascular plant waxes, specifically n -alkane homologues C25 through C 33 , combined with stable-isotopic analyses of bulk soil organic matter (SOM) and pedogenic carbonates can enhance paleoenvironmental interpretations of paleosols. This dissertation presents three case studies of low-organic-carbon paleosols for which a combined approach is employed in paleoecosystem assessment. The first study reconstructs Plio-Pleistocene paleoenvironments at the archaeological locality of El-Kherba in Ain Hanech, Algeria. The abundance of in -alkanes and preference for odd-carbon numbered molecules combined with stable-carbon-isotopic analyses of bulk SOM and stable-carbon- and stable-oxygen-isotopic analyses of pedogenic carbonates indicate that a coastal Mediterranean ecosystem existed at El-Kherba. The second case study focuses on Pleistocene paleosols from the upper Busidima Formation at the early hominid site of Gona, Ethiopia. Using data similar to those for the Algerian study, molecular and stable-isotopic analyses indicate that vegetation varied through time in the upper Busidima as a result of lateral migrations of the paleo-Awash River through the Awash Valley. The third case study considers vegetation and rapid climate fluctuations as recorded by a stacked sequence of gravels and terra rossa paleosols at the Morphi Pleistocene site in Epirus, Greece. Vegetation changes are closely coincident with regional glacial and interglacial climate fluctuations recorded for western Epirus. Taken together, these three biogeochemical studies demonstrate the potential of biomarker analyses to form the basis of detailed paleoenvironmental reconstructions for low-organic-carbon paleosols.