Characterization of the Goethite-Hematite Ratio in Modern

Lehigh University Lehigh Preserve

Theses and Dissertations

2017 Characterization of the Goethite-Hematite ratio in Modern and Ancient Soils in the Mid-Atlantic Region as a Paleoprecipitation Proxy Laura Audrey Thomasina Markley Lehigh University

Follow this and additional works at: http://preserve.lehigh.edu/etd Part of the Environmental Sciences Commons

Recommended Citation Markley, Laura Audrey Thomasina, "Characterization of the Goethite-Hematite ratio in Modern and Ancient Soils in the Mid-Atlantic Region as a Paleoprecipitation Proxy" (2017). Theses and Dissertations. 2711. http://preserve.lehigh.edu/etd/2711

This Thesis is brought to you for free and open access by Lehigh Preserve. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of Lehigh Preserve. For more information, please contact [email protected].

Characterization of the Goethite-Hematite ratio in Modern and Ancient Soils in the Mid- Atlantic Region as a Paleoprecipitation Proxy

Laura Markley

A Thesis

Presented to the Graduate and Research Committee

of Lehigh University

in Candidacy for the Degree of

Master of Science

Earth and Environmental Sciences

Lehigh University

May 2017

Thesis is accepted and approved in partial fulfillment of the requirements for the Master of Science in Earth and Environmental Sciences.

Characterization of the Goethite-Hematite ratio in Modern and Ancient Soils in the Mid- Atlantic Region as a Paleoprecipitation Proxy Laura Markley

Date Approved

Stephen Peters, Ph.D., Thesis Advisor

Frank Pazzaglia, Ph.D., Committee Member

Kenneth Kodama, Ph.D., Committee Member

Dave Anastasio, Ph.D., Chairperson of Department

iii

ACKNOWLEDGEMENTS

Many thanks to my thesis advisor, Steve Peters, for all of his advisement and support throughout these past 2 years. His guidance, knowledge, and enthusiasm has molded me into a much better scientist and I am deeply indebted to him for bestowing upon me some of his lab and key-figure making skills. Thank you for countless edits, discussions, and suggestions.

Committee Members, Ken Kodama and Frank Pazzaglia. Ken’s patience and knowledge in developing these magnetic methods for soils despite numerous issues and Frank’s immaculate memory of all things related to soils are unmatched. I owe both of you my gratitude for your intellectual discourse and suggestions throughout this process.

Soil Team Members, Matt McGavick and Cora Summerfield for assistance with data collection and analysis as well as providing me with data they collected for their respective projects. Collaborating with both Matt and Cora has been a pleasure.

Past Soil Team Members, Jordan Dykman, Taylor Cummins, and Johanna Blake for their soil, data, and well-organized excel documents.

Tammy Rittenour from Utah State University for OSL and IRSL dates.

Mark Carter from the USGS for assistance in the field and intellectual discourse.

George for fixing the basement furnace at a moment’s notice during a busy week.

Udo Schwertmann and Ethan Hyland for essential background information for this study.

For financial support, I thank the GIC and Earth and Environmental Science Department at Lehigh.

TABLE OF CONTENTS

List of Figures ...... vii List of Tables ...... xi Abstract ...... 1 1.0 Introduction ...... 3 1.1 Intro to Paleoprecipitation Proxies ...... 3 1.2 Intro to Soils ...... 3 1.3 Intro to Iron Oxide Formation ...... 4 1.4 Goethite-Hematite (G/H) and Mean Annual Precipitation (MAP) ...... 7 2.0 Site Location & Characterization ...... 9 2.1 Soil Chronology ...... 10 3.0 Soil Processing and Methodology ...... 14 3.1 Soil Sampling and Description ...... 14 3.2 Particle Size Distribution Analysis ...... 14 3.3 Bulk Elemental Analysis & Weathering Intensity ...... 14 3.4 Iron Oxide Crystallinity...... 15 3.5 Magnetic Goethite and Hematite Abundance ...... 16 3.6 Application of Hyland et al.’s G/H to MAP Proxy ...... 18 3.7 Application of Sheldon et. al’s CIW to MAP Proxy ...... 19 4.0 Results ...... 19 4.1 Soil Characterization ...... 19 4.1.1 Middle/Late Miocene ...... 24 4.1.2 Middle Pleistocene ...... 25 4.1.3 Late Pleistocene ...... 29 4.2 Predicted Mean Annual Precipitation (MAP) ...... 34 5.0 Discussion ...... 36 5.1 Predicted Paleoprecipitation with Time ...... 36 5.2 Comparison with other Indices...... 38 5.3 Deposition of Sediments vs. Onset of Pedogenesis ...... 40 5.4 Presence of Gleying ...... 41 5.5 Influence of Organic Matter ...... 42 5.6 Influence of Hydrology and Water Table Fluctuations ...... 44

5.7 Impacts of Parent Material Composition on G/H ...... 44 5.8 Total Dithionite-Extractable (DCB) Iron Threshold ...... 46 5.9 Use of Magnetic Methods to Determine G/H...... 47 6.0 Conclusions ...... 49 References ...... 52 Appendices ...... 58 Appendix 1. Extended Procedures ...... 58 Particle Size Distribution Analysis ...... 58 Bulk Elemental Analysis ...... 58 Iron Oxide Crystallinity...... 59 Appendix 2. Particle Size Distribution Analysis (PSDA) Data ...... 60 Appendix 3. FeO/FeD Values ...... 63 Appendix 4. Major Elements (% By Mass) ...... 66 Appendix 5. Sample Demagnetization ...... 67 Appendix 6. G/H, CIW, & Predicted MAP ...... 75 Supplementary Figures ...... 77 Vita ...... 80

LIST OF FIGURES

Figure 1: Diagram modified from Stucki et al. (1988), showing the formation of Fe(III) oxide precipitates, goethite and hematite, as a function of different climatic factors and processes. Fe(III) oxide precipitate formation in soils is a competitive pathway and is dependent on climatic variables such as: rate of Fe release, pH, amount of organic matter, and, most importantly, soil temperature and moisture. Hematite formation is preferred in environments with higher rates of Fe release, pH values between 3-8, and higher soil temperature. Goethite formation is preferred in conditions contrasting those of hematite, as well as in areas with higher organic matter and soil moisture...... 5

Figure 2: Schematic figure of two soil profiles consisting of typical soil horizons and a parent material. The tropical soil profile on the left shows a red, highly rubified B horizon indicating more hematite formation. The temperate soil profile on the right shows a yellowish brown B horizon indicating preferential formation of goethite...... 6

Figure 3: Linear relationship between mean annual precipitation (MAP) and the goethite-hematite ratio (G/H) in modern soils worldwide (Hyland et al. 2015)...... 7

Figure 4: Soil profile from Emmaus Kame, Pennsylvania showing a brown, Holocene soil overlying an older red alluvial-colluvial deposit...... 8

Figure 5: Geographic location of sample sites in the mid-Atlantic region. Samples within Pennsylvania include Millheim Narrows (MHN), Emmaus Kame (EK), and the Penn State Agricultural Center (PS) sites. The Bryn Mawr (BM) site is located within Cecil County, Maryland, and additional sites (Pit F, BB1, AF1, LZ1, Site D, AH2, and BB2) are located in Louisa, Virginia and are designated the “South Anna River” soils...... 9

Figure 6: Stratigraphic context of site locations, including interpreted or approximate age for soil pits, see Table 1 for chronological methods and numerical ages ...... 12

Figure 7: Illustration showing the stepwise demagnetization of soil samples for the determination of the magnetic G/H. The magnetic moment of each soil sample is saturated with an applied field of 5T prior to demagnetization. The preceding stepwise removal of magnetite and goethite from the overall magnetic moment allows for the calculation of G/H given the moment at M100 mT and M125C...... 17

Figure 8: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio / 10, and chemical index of weathering (CIW)/100 with depth for the Bryn Mawr formation in Cecil County, Maryland. Note the scale change for the G/H...... 24

vii

Figure 9: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for the Emmaus Kame soil pit in Pennsylvania...... 25

Figure 10: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for Pit F in Virginia, note that the PSDA exceeds a cumulative % of 100 in the 2Btb2 horizon, indicating possible error with determination of the clay percentage...... 26

Figure 11: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for the Millheim Narrows soil pit in Pennsylvania...... 27

Figure 12: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for the AH2 soil pit in Virginia...... 28

Figure 13: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for the Penn State Agricultural Center soil pit in Pennsylvania...... 29

Figure 14: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for the BB2 soil pit in Virginia...... 30

Figure 15: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for the AF1 soil pit in Virginia...... 31

Figure 16: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for the BB1 soil pit in Virginia...... 32

Figure 17: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for the LZ1 soil pit in Virginia...... 33

Figure 18: From left to right: particle size distribution (PSDA), tau plot, iron oxide crystallinity, goethite/hematite (G/H) ratio, and chemical index of weathering (CIW)/100 with depth for Site D in Virginia...... 34

Figure 19: Predicted Mean Annual Precipitation (MAP) value based on an average G/H value for each B horizon(s) plotted on Hyland et al’s linear function (2015). Red,

viii

yellow, and orange X’s show the estimated present day MAP values for each state based on 1981-2010 Normals (Virginia (VA), Pennsylvania (PA), and Maryland (MD) (Durre et al., 2012; Durre et al., 2012; Durre and Squires, 2015). The top right arrow indicates the existence of a point off scale at a G/H of ~26.7 and predicted MAP ~ 26,887 for the Bryn Mawr (BM) soil pit in Maryland. The majority of samples plot at dryer (lower MAP) conditions compared to the present...... 36

Figure 20: Mean annual precipitation (MAP), based on Hyland et al.’s linear function relating MAP and G/H (2015), with time. Ranges in predicted MAP are a result of changes in the G/H within the B horizon of each soil profile. The time window indicates a maximum time of ~100 KA to equilibrate soil conditions (Harden, 1982). MAP in the Middle to Late Miocene is significantly higher than in present day conditions, while the Early to Late Pleistocene is dryer and exhibits a lower MAP. Due to possible error associated with inherited hematite and water level fluctuations within the soils, the younger South Anna River Soils (bottom right) should not be used for comparison between present and past precipitation conditions. Present precipitation conditions are estimated at ~1100 mm/yr based on 1981-2010 Normals for MA, VA, and PA (Durre et al., 2012; Durre et al., 2012; Durre and Squires, 2015)...... 37

Figure 21: Predicted mean annual precipitation (MAP) from Hyland et al’s G/H proxy (2015) and Sheldon et al.’s CIW proxy (2005) for all soil pits, excluding BM in Maryland. Samples are taken throughout the B horizon. The two proxies do not show an agreement between predicted MAP and indicate that: the CIW proxy may be over predicting MAP, the G/H may be under predicting MAP, or these soils, or a subset of these soils, do not meet the necessary criteria for application of one or either proxy. . 38

Figure 22: Predicted mean annual precipitation (MAP) from Hyland et al’s G/H proxy (2015) and Sheldon et al.’s CIW proxy (2002) for all soil pits, excluding BM in Maryland. Data points reflect an average MAP value of samples within the B horizon that most closely meet the G/H proxy criteria (Table 2). A large disagreement between the two proxies remains and can possibly be attributed to the limited applicability of the CIW proxy...... 39

Figure 23: Distribution of goethite around a root channel within a hematitic soil (Stucki et al., 1988)...... 42

Figure 24: Plot showing the G/H and total dithionite-extractable iron (Fed) on log scales for B horizon samples for 11 soil sites from Virginia, Maryland, and Pennsylvania. Samples cluster by state. A G/H development threshold value of 1000 ppm total secondary iron was chosen based on the location of unreliable G/H values from the South Anna River soil samples, which plot below 1000 ppm. Samples from Maryland, Pennsylvania, and Pit F from Virginia all plot at or above the threshold...... 46

Figure 25: Stratigraphic column of the Bryn Mawr formation in Cecil County, Maryland, with depth in meters...... 77 ix

Figure 26: Stratigraphic soil column of the Emmaus Kame soil pit in Pennsylvania with depth in meters...... 78

Figure 27: Stratigraphic soil column of the Penn State Ag. Center soil pit in Pennsylvania with depth in meters...... 79

LIST OF TABLES

Table 1: Summary table of location and chronology of soil pits...... 11

Table 2: Summary table of data for each soil profile used to calculate MAP...... 35

Table 3: Predicted MAP using both the CIW and G/H proxy for three soil profiles that best met both proxies’ criteria...... 40

ABSTRACT

The goethite/hematite (G/H) ratio in modern soils around the world has a strong, positive linear relationship with mean annual precipitation (MAP) (Hyland et al., 2015). If this relationship holds true in paleosols, then the G/H could be a reliable proxy of paleoclimate conditions recorded in soils. The Holocene, brown soils in the mid-Atlantic region are observably different than the older, red paleosols buried meters below, suggesting they formed in response to different environmental conditions, and are ideally suited to explore the G/H paleoclimate proxy.

Soil development occurs in the critical zone, where parent material is altered by physical, chemical, and biological processes. These soils can be preserved on the landscape, and record paleoprecipitation conditions. MAP at the time of soil development can be encoded in elemental composition of pedogenic minerals, such as iron oxides.

MAP controls the relative formation of goethite and hematite at the time of soil formation, resulting in the creation of the G/H ratio. Iron oxide minerals form in soils and give soils a characteristic color based on the relative abundance of different iron oxide minerals.

Goethite and hematite are the two most commonly occurring of the iron (III) oxides and are formed under competing environmental conditions, with goethite (FeOOH) preferring high soil moisture, whereas hematite (Fe2O3) prefers low soil moisture.

This work applies the relationship between G/H and MAP established by Hyland et al.

(2015) to reconstruct MAP for 11 soils in the mid-Atlantic region of the US using novel magnetic methods to characterize G/H abundance. The particle size distribution, iron oxide

crystallinity, and bulk elemental analysis are characterized for each soil profile. Interpreted

MAP values indicate relative wetter or drier conditions through geologic time. Compared to present day conditions, the Middle to Late Miocene experienced a wetter climate than present, whereas the Middle and Late Pleistocene had lower MAP or experienced drier conditions than present. In comparison to the Middle Pleistocene, the Late Pleistocene shows relatively wetter conditions.

MAP values obtained from the G/H proxy were compared to the Chemical Index of

Weathering (CIW) proxy from Sheldon et al. (2002). The CIW proxy relates MAP with the weathering of feldspar minerals and accumulation of clays. Disagreement between the

G/H and CIW proxies indicate possible over or under prediction of MAP in some soils.

The influence of inheritance from parent material, organic matter leaching of secondary iron oxides, water level fluctuations and gleying, total dithionite-extractable (DCB) iron, and time of pedogenesis are discussed as they control the encoding of G/H in the soil profile.

1.0 Introduction

1.1 Intro to Paleoprecipitation Proxies Precipitation reconstruction is important for constraining future trends in rainfall, which directly impact regional water availability. A variety of methods and proxies, such as tree- ring records, have been employed to determine paleoprecipitation conditions (Cook et al.,

2004). However, tree-ring records only encompass the past 2,000 years and are not applicable over longer time scales. The soil record may provide a valuable proxy for recording average rainfall over the time of soil formation and have become increasingly recognized as a rich archive of past climate conditions, which has resulted in the evolution of paleopedology (Pelletier et al., 2015; Sheldon, 2009; Tabor and Myers 2015; Retallack,

2013).

1.2 Intro to Soils Soil formation occurs in the critical zone and is controlled by terrestrial climate conditions, which act in concert to form a soil through a parent material (Jenny, 1941). As parent material is altered into soil, the elemental composition of the parent is altered via chemical weathering and secondary mineral formation (Buggle et al. 2011). Chemical weathering results in changes in elemental composition of pedogenic minerals, such as iron oxides, within the B horizon in response to precipitation (Tabor and Myers, 2015). If a soil is buried, common in a subsiding basin or at the toe of a hillslope that is accumulating colluvium, this record is removed from the soil forming environment and can be preserved in the landscape. These buried soils are called paleosols and provide a geochemical record of their climate of formation. Additionally, soils at the surface may provide a record of modern and recent past rainfall trends depending on their time of soil pedogenesis. 3

Paleosols and soils will present the core data set that will be explored in this study and focus is placed on the formation of iron oxides, mainly goethite and hematite, in the B horizon in response to precipitation. The Bt horizon forms over long periods of time, averaging climate conditions during the time of formation. Changes in iron oxide composition within the Bt reflect equilibrium processes and should provide a record of long-term, rather than short or extreme, trends in precipitation (Sheldon and Tabor, 2009).

1.3 Intro to Iron Oxide Formation Soil color can be partly attributed to the relative abundance of different iron oxide minerals, particularly the two most abundant: hematite (5R – 2.5YR), and goethite (2.5Y – 7.5YR)

(Schwertmann, 1988). Hematite and/or goethite formation depends on environmental conditions during pedogenesis. Dissolved Fe(III) ions in solution originate from the hydrolysis and oxidation of primary Fe(II) silicates from the weathering of parent material.

The resulting Fe(III) ions in solution can then take competing pathways toward the formation of either goethite or hematite depending on the environmental conditions present during the time of soil formation (Figure 1) (Stucki et al., 1988).

Hematite (α-Fe2O3) causes the rubification, or degree of redness, in soils. The formation of hematite is highly dependent on climate regime and is favored in soils with: high soil

temperature, free drainage, pH values 7-8, and low organic matter content (Figure 1).

Hematite is usually found in tropical to subtropical soils and is lacking in temperate climates (Fischer and Schwertmann, 1975; Schwertmann and Taylor, 1989; Schwertmann and Fitzpatrick, 1992; Figure 2).

(Hematite)

(Goethite)

TropicalSub / Tropical Temperate Figure 2: Schematic figure of two soil profiles consisting of typical soil horizons and a parent material. The tropical soil profile on the left shows a red, highly rubified B horizon indicating more hematite formation. The temperate soil profile on the right shows a yellowish brown B horizon indicating preferential formation of goethite.

Goethite (α-FeOOH) formation results in a yellowish-brown (10YR) soil color in contrast to the redder color associated with hematite. The formation of goethite is favored in cool and moist climate regimes and predominates in areas with a low rate of iron supply, higher water activity, lower temperatures, and more extreme pH conditions (either higher or lower

than the natural spectrum) (Figure 1) (Yapp, 2001; Schwertmann and Taylor, 1977;

Torrent et al., 1980; Schwertmann et al. 1982).

The formation of iron (III) oxides in soils responds to climate and provides a record of

environmental change that is encoded in the relative abundance of goethite and hematite

and results in an overall goethite-hematite ratio (G/H) (Figure 2). This ratio can be

quantified for a specific soil using the magnetic properties of each mineral.

1.4 Goethite-Hematite (G/H) and Mean Annual Precipitation (MAP) A previous study by Hyland et al. (2015) revealed that there is a robust positive linear

Figure 3: Linear relationship between mean annual precipitation (MAP) and the goethite-hematite ratio (G/H) in modern soils worldwide (Hyland et al. 2015).

relationship between the magnetic goethite/hematite (G/H) ratio in modern soils and mean

annual precipitation (MAP) (Figure 3). The observed relationship between MAP and G/H

in soils may provide a method to determine paleoprecipitation conditions over the time span of soil development.

In this paper, we apply Hyland’s function to buried and surficial soils of known ages in the mid-Atlantic region to further investigate the application of the G/H proxy through geologic time. The mid-Atlantic region is one area that exhibits visible differences in soil rubification between buried paleosols and modern soils, which may indicate variation in iron oxide formation caused by changes in precipitation Figure 4: Soil profile from Emmaus Kame, Pennsylvania showing a brown, Holocene soil (Figure 4). In order to validate the estimated overlying an older red alluvial-colluvial deposit.

MAP from the G/H proxy, results are compared to the Chemical Index of Weathering (CIW or CIA-K) proxy from Sheldon et al. (2002), which relates the weathering of feldspar minerals and accumulation of clays with MAP.

2.0 Site Location & Characterization

This study targets paleosols and soils representing a range of parent materials, topographic,

age, and modern climate conditions (Fig. 5). Samples are assembled from sites that have

been previously excavated and described in Pennsylvania (Dykman, 2015; Cummins

2015), as well as two sites in Maryland and Virginia (Pazzaglia et al. 1997; Summerfield,

2015; McGavick, 2017) that are new to this investigation. The three paleosols in

Pennsylvania at the Penn State Ag Center, Millheim Narrows, and Emmaus, are all buried

by younger deposits and soils and thereby mostly cutoff from modern soil forming

conditions. In contrast, the soils and paleosols in Maryland and Virginia are at the land

surface, and directly influenced by the modern soil forming environment.

Parent materials for these soils vary regionally. The Bryn Mawr (BM) soil in Maryland has a parent material composed predominantly of quartzose sandy gravel originating from fluvial-deltaic deposition. The presence of bald cypress tree pollen, typically found in modern warm climates along the U.S. gulf coast, indicate a potentially wetter paleoenvironment during the time of soil formation (Pazzaglia et al. 1993).

Soil sites in Pennsylvania exhibit two different parent materials. The Millheim Narrows

(MHN) and Penn. State Agricultural Center (PS) soils are both developed through alluvial fans draining ridges underlain by the Bald Eagle Sandstone. The Emmaus Kame (EK) site, a Pre-Illinoian glaciation kame delta, is formed through alluvium sourced from the Bryam

Gneiss that underlies South Mountain.

The South Anna River soils in Louisa, Virginia are all composed of similar alluvial parent material derived from saprolite and residuum sourced in the deeply chemically weathered metamorphic rocks of the Virginia piedmont Locally, the bedrock units contributing to this alluvium are the Ordovician-Silurian age Chopawamsic Formation and Ellisville

Pluton. The Chopawamsic Formation varies in composition between a garnet-chlorite schist and felsic to mafic composition metavolcanic rocks (Spears et al., 2013). The

Ellisville Pluton is a granodiorite to diorite dated at 443.7 +/- 3.3 Ma (U-Pb zircon; Hughes and others, 2013; Spears et al., 2013).

2.1 Soil Chronology A range of geochronologic data including optically stimulated and infra-red stimulated luminescence (OSL and IRSL), terrestrial cosmogenic nuclide (TCN) profile dating,

biostratigraphy, and magnetostratigraphy are used to bracket soil and paleosol ages (Table

1).

et al.et

1993 1997

Citation

personal personal personal

University

Pazzaglia,

Utah State

Harkins, N., N., Harkins,

unpublished.

Kodama and

Kirby, Kirby, E. and

Ciolkosz

communication.

communication,

Pazzaglia et al., al., Pazzagliaet

OSL OSL OSL

IRSL IRSL IRSL IRSL

Method

FeO/FeD

Chronological Chronological

Biostratigraphy,

soil in McElhatten, soilin McElhatten,

Paleomagreversal

Pollenassemblage

10Be date of10Be similar date

Magnetostratigraphy, Magnetostratigraphy,

0.9 1.6 0.8

- -

- - - - -

0.9 0.6 0.9

0.85 1.0 0.75

Depth(m)

44 ka 44

31.6; 31.6; 11.3;

23.6 ka 23.6 ka 51.7 ka 4.86

21.2 ka 21.2 ka 16.5 ka 11.4

- -

- - -

Age

years)

Middle Middle

reversal

Early Late Early

(10^4 years) (10^4

60.7+/ 61.4+/

Late Middle orLate Middle Miocene (Late

Serravallian to

391.8+/

97.2+/ 75.9+/ 64.3+/

earlyTortonian)

Late Pleistocene LatePleistocene abovepaleomag

149.2+/ 119.3+/ 18.09+/

< < soilka,780 unit

Pleistocene (10^5 Pleistocene(10^5

LZ1

AF1

BB1 BB2

AH2

(PS) (EK)

Pit F Bryn

(BM)

Penn

Mawr Mawr

Kame

SiteD

(MHN)

Center

SoilPit

Millheim

Narrows Narrows

StateAg

Emmaus

77.4820 77.9339 75.4795 78.0331 76.0282

- - - - -

Longitude

Latitude

40.9204 40.7118 40.5489 37.9711 39.6200

: Summary table of location and chronology of soilof locationpits. chronology and of : table Summary

Louisa,

Virginia

Location

Maryland

River Soils) River

(South Anna

Pennsylvania CecilCounty, Table 1 Table

Figure 6: Stratigraphic context of site locations, including interpreted or approximate age for soil pits, see Table 1 for chronological methods and numerical ages

Paleomagnetism assigns ages to soils around 1 mya. The last major polarity reversal was at the Matuyama-Brunhes boundary (MBB), which occurred at 780 ka (Shackleton et al.,

1990; Spell and McDougal, 1992; Tauxe et al., 1996). Measuring the polarity of a given soil sample within the paleosols, such as that in Emmaus Kame, can indicate whether the soil acquired its magnetism during the normal polarity Brunhes chron or the reversed

Matuyama chron. The Emmaus Kame deposit showed a reversed polarity in its lower alluvium deposit, indicating deposition during the Matuyama chron and an age greater than

780 KA (Kodama and Pazzaglia, unpublished; Table 1; Figure 6).

The seven sites from Virginia, designated the South Anna River sites, are dated using OSL or IRSL techniques. OSL/IRSL samples are collected at a depth of approximately 1 meter in each soil profile to obtain an age of each river terrace. Sample preparation and OSL/IRSL measurements were carried out at the Luminescence Laboratory at Utah State University and returned ages for deposits ranging from Middle to Late Pleistocene (Table 1; Figure

6).

Soil chronologies for remaining soil pits in Pennsylvania and Maryland are obtained using cosmogenic dating, iron oxide crystallinity, and biostratigraphy, respectively. Cosmogenic

(10Be) dating is obtained from an analysis of a soil pit from a similar site, located in

McElhatten, PA, with an analogous soil to the Millheim Narrows soil pit in Pennsylvania

(Kirby and Harkins, unpublished). The age of the Penn. State Agriculture Center pit in

Pennsylvania is constrained using its stratigraphic context and color compared to similar soils in the region, in addition to a relative age obtained from iron oxide crystallinity

(Ciolkosz et al. 1993). The Bryn Mawr formation in Cecil County Maryland is thought to 13

be late Miocene in age based on pollen assemblage (Pazzaglia et al., 1997; Table 1; Figure

6).

3.0 Soil Processing and Methodology

3.1 Soil Sampling and Description In the field, soils are classified and described based on NRCS soil description taxonomy, including: color (redness), texture, composition, and structure. Soils are sampled using soil pits or auger holes. Soil pits are located at mid-slope, where alluvium accumulates to bury older soils. Soil pits are sampled at 10 cm intervals from the base to the top of the pit.

Auger holes are drilled and significant textural or color changes are sampled with depth.

3.2 Particle Size Distribution Analysis Particle size distribution analysis (PSDA) is initiated by separating the <2mm fraction by wet sieving and saturating with dispersant. Organic content is determined by wet digestion using 30% Hydrogen Peroxide. Finer grain sizes is determined by continued wet sieving and settling using the NRCS pipette method (Appendix 1; Soil Survey Staff, 2014).

3.3 Bulk Elemental Analysis & Weathering Intensity Bulk elemental analysis is determined using a sub sample of <2mm soil ball milled into a fine powder. Samples are fluxed with lithium metaborate, heated in a furnace at 1000°C, and subsequently dissolved in nitric acid. Samples are then diluted before analysis using inductively coupled plasma mass spectrometry (ICP-MS) (Appendix 1; Cummins, 2015).

Tau ratio values are determined using Brimhall and Dietrich (1987):

퐶푗,푤 퐶푖,푝 휏푖푗 = · – 1 (1) 퐶푗,푝 퐶푖,푤

where C is the concentration (mol/m3) of immobile (i) or mobile (j) elements in weathered

(w) or parent (p) material. All calculations are referenced to zirconium, which is assumed to be immobile. Positive tau values indicate enrichment of an element, whereas negative values show depletion (Ma et al., 2011).

An additional method of quantifying chemical weathering is through calculation of the

CIW. In comparison to the chemical index of alteration (CIA), the CIW does not include potassium (CIA-K), which is carried in minerals that do not consistently weather with exposure during pedogenesis (Harnois, 1988). CIW changes from the parent material to the soil can be large or small depending on the mineralogy of the parent rock and the degree of weathering (Sheldon and Tabor, 2009). The results of the bulk elemental analysis can be used to calculate the chemical index of weathering (CIW) using the following equation and cation concentrations in moles (Harnois, 1988):

퐴푙 퐶퐼푊 = 100 푥 (2) 퐴푙+퐶푎+푁푎

The chemical index of weathering (CIW) values are expressed as CIW/100 to plot values as a range from 0 – 1, with higher values indicating accumulation of aluminum (Al) and depletion of sodium (Na) and calcium (Ca).

3.4 Iron Oxide Crystallinity Measuring iron oxide crystallinity provides an insight into the relative time of soil development within a specific profile given that secondary iron oxide formation is dependent on weathering processes over time (Dykman, 2015; Ciolkosz et al., 1993). The ratio of Feo (oxalate-extractable iron, representing ferriydrite and iron associated with 15

organic matter) to Fed (dithionite extractable iron) in soils decreases with the duration of pedogenesis due to increasing iron oxide crystallinity (Appendix 1; Dykman, 2015; Lair et al., 2009).

Dithionite-citrate-bicarbonate (DCB)-extractable iron (Fed) is determined using the method of Mehra and Jackson (1960) at 80°C in a hot block. Ammonium oxalate extractable iron is determined using the procedure of McKeague and Day (1996). Both iron extractions are diluted for analysis on the ICP-MS.

3.5 Magnetic Goethite and Hematite Abundance Goethite and hematite abundance is determined using the response of the magnetic properties of the minerals (magnetite, hematite, and goethite) within the samples. Goethite and hematite abundance is determined for samples taken from the B-horizon. Given there can be multiple B-horizons or varying soil rubification within a B-horizon, samples are taken throughout the thickness of the B-horizon to ensure measurements throughout are consistent.

The G/H of soil samples is determined using the magnetic properties of the minerals magnetite, goethite, and hematite, present in the sample. The approach utilizes the low coercivity of magnetite and the Neel temperature of goethite (125°C) for the step-wise removal of their magnetism from the samples. Soil samples (<2 mm) are tamped using a glass rod tightly into 2.0 mL polypropylene microcentrifuge tubes which are pre-cut to a length of 2 mm. The soil sample is then sealed with 2-3 drops of liquid water glass (sodium silicate 40%) and allowed to dry for 2 days to prevent grain movement and loss of sample during analysis. 16

Once dried, each sample tube is measured for its NRM (natural remanent magnetization) using a 2-G Enterprises superconducting rock magnetometer. After determination of the

NRM for each soil sample, an ASC Scientific Model IM-10-30 Impulse Magnetizer is used to apply a field of 5 T to each sample prior to re-measuring samples in the magnetometer

(MIRM). The sample’s acquisition of this IRM (isothermal remanent magnetization) will magnetically saturate all magnetic mineral present. Each sample is then loaded individually 17

into the magnetometer and AF (alternating field) demagnetization at 100 mT is used to randomize the magnetization of any low coercivity magnetic minerals, most likely magnetite. Samples are then re-measured (M100 mT) prior to thermal demagnetization.

This measurement quantifies the hematite and goethite in the sample. Samples are loaded into the sample boat (15 – 20 samples at a time) to undergo thermal demagnetization in an

ASC Scientific Model TD-48 5C thermal specimen demagnetizer. Samples are heated for an hour at increasing temperatures from 70°C to 125°C. Samples are kept at 125°C for 5 to 10 minutes to prevent overheating and melting of polypropylene tubes, while also heating all samples throughout to the Neel temperature of goethite thus removing the magnetization carried by goethite. Samples are then cooled until they reach room temperature, after which they are measured in the superconducting magnetometer. This last measurement will obtain the signature of only hematite (M125°C). The G/H ratio is then calculated using the formula contained in Figure 7.

After heating, samples are inspected for any signs of grain movement or loss. Samples showing any indication of grain movement are measured 3-5 times to determine if the magnetization, both intensity and direction, of the samples shows any change between measurements. Any sample with a significant change in the measurements are set aside and re-run with the next sample batch using a newly packed soil tube.

3.6 Application of Hyland et al.’s G/H to MAP Proxy The G/H values obtained through magnetic measurements are applied to Hyland’s linear function:

퐺 푀퐴푃 (푚푚 푦푟−1) = 1003.5 ∗ + 49.3 (3) 퐻

This proxy may be applied to soils that do not exhibit: 1) Evidence of significant pedoturbation, 2) Weakly developed soils or channel-proximal sediments, and 3) Soils that have experienced substantial burial or postburial gleying (Hyland et al., 2015; Mack et al.,

1993; Kampf and Schwertmann, 1983; Torrent et al., 2010; Gualtieri and Venturelli, 1999;

Bjorlykke, 2010).

3.7 Application of Sheldon et. al’s CIW to MAP Proxy A function between MAP and CIW (Sheldon et al, 2002) is applied to soil samples in the

B horizon for comparison to the G/H proxy:

푃 (푚푚 푦푟−1) = 221.1 푒0.0197 (퐶퐼푊) (4) where CIW is determined using results from the bulk elemental analysis (Eq. 2). This proxy can be applied to: 1) The Bw or Bt horizon in soils, 2) Soils with a 5-8 CIW value difference between the parent material weathering ratio value and the B horizon value, 3) Soils without near surface carbonates or evaporite minerals, 4) Soils not located in hillslope or montane settings, and 5) Non lateritic soils (Sheldon et al., 2002).

4.0 Results

4.1 Soil Characterization The results of the particle size distribution analysis (PSDA) (Appendix 2), bulk elemental analysis (Appendix 4), iron crystallinity analysis (Appendix 3), G/H abundance

(Appendix 5, 6), and CIW/100 calculation (Appendix 6) are plotted in Figures 8-18 and are organized by decreasing soil age, with BM (Bryn Mawr) as the oldest (late Middle to 19

early Late Miocene) and Site D as the youngest (Late Pleistocene). PSDA results are shown as a cumulative % out of 100 with depth.

The Bryn Mawr (BM) formation (Figure 8) has different characteristics from the other soil profiles, given its much older age (late Middle to Early Late Miocene). The PSDA shows a relatively high abundance of gravel (>2mm) and sand and low percentage of silt and clay within the first 550m. The lower BC horizon has an increase in sand content compared to upper horizons. The tau plot, referenced to the lower gravel-sand parent in the BC horizon, shows enrichment of aluminum (Al) and magnesium (Mg) within the upper horizons and slight enrichment of silicon (Si). Iron (Fe), sodium (Na), and potassium (K) are all depleted, with some horizons showing local enrichment.

The BM soil had the lowest iron oxide crystallinity ratios, highest G/H values, and variable

CIW throughout its profile. Iron oxide crystallinity (Feo / Fed) is low in the BM soil, and decreases with depth. The maximum Feo / Fed value is 0.0757. Relative to other soil profiles, the BM soil shows the lowest Feo / Fed, with values close to 0. G/H values are the highest in the BM soil, ranging from 9.77 in a highly rubified, gravel-rich section of the profile to ~35 in the Btb3 horizon. Since low iron was suspected in other B horizons, the

Btb3 horizon and rubified horizon were targeted for sampling for the G/H. CIW/100 values show a gradual decrease with depth from 0.608 in the uppermost sample to 0.250 in the lowest sample or parent material.

The Pennsylvania soil pits, which range in age from Middle Pleistocene (EK, MHN) to

Late Pleistocene (PS), are plotted in Figures 9, 11, and 13 and have variable particle size distributions. Clay content for the upper horizons at EK and all of the PS profile show an increase with depth, whereas MHN shows almost no change with depth. MHN horizons have little to no clay content and are dominated by the >2mm and sand fractions compared to EK and PS. EK has an increase in clay content with depth until the beginning of the BC horizons, which show a decrease in clay content with depth.

MHN and PS have an enrichment of iron (Fe) either throughout their profile (PS) or within a specific horizon with depth (MHN). EK and PS both show similar low Feo / Fed with depth, ranging from 0.067 – 0.32. MHN shows larger Feo / Fed values, ranging from 0.57 –

1.50. Potassium (K), magnesium (Mg), and silicon (Si) are generally depleted within these soil profiles. Tau plots show no measureable sodium (Na) in any profile, except for an enrichment within the upper horizon of PS.

Goethite and hematite abundance (G/H) determined from magnetic methods and CIW values vary between soil profiles and within the B horizon. EK has consistent values

(0.226) within the 2Bt2 horizon, but shows variation within the lower BC horizons. MHN has varying G/H in its 2B horizon (0.212 and 0.337). The PS pit shows the greatest G/H values in Pennsylvania with the largest range (0.506 – 0.859). CIW/100 values for all profiles are close to 1 and range from 0.609 to 0.949. The lowest value occurs in PS and is localized to the BE horizon, with other horizons showing values around or exceeding 0.83.

The Virginia soil pits located on the South Anna River range in age from Middle

Pleistocene (Pit F, AH2) to Late Pleistocene (AF1, BB1, BB2, LZ1, Site D) and are plotted in Figures 10, 12, and 14 - 18. PSDA results for most soils show an increase in clay content and a general decrease in larger grain sizes (>2mm, sand) with depth, aside from Pit F. All pits excluding Pit F show little to no >2mm fraction. Pit F shows an increase in the >2mm fraction with depth before the BC horizon, with as much as ~50 cumulative % in the 2Btb2 horizon.

Tau plots are referenced to the lowest sample taken from Site D, which is assumed to be the most unweathered parent out of the South Anna River soils. This was chosen based on the varying composition of parent material in the region and difficulty in attributing the parent to one rock formation. All profiles excluding Site D and Pit F show a depletion in iron (Fe) and aluminum (Al) and an enrichment in silicon (Si). Some profiles show a local enrichment in deeper horizon of specific elements, including sodium (Na), aluminum (Al), magnesium (Mg) and iron (Fe). Pit F shows different trends, with depletion in all elements in the upper horizons and progressive enrichment with depth. Site D shows a depletion in all elements, aside from low enrichment of silicon (Si) in the uppermost horizon.

Iron oxide crystallinity (Feo / Fed) decreases with depth for all pits excluding LZ1 and Site

D. Values range from a maximum of 0.914 in LZ1 to 0.161 in Site D. On average, LZ1

(0.775) and Site D (0.547) have the highest Feo / Fed and AH1 (0.260) and AF1 (0.341) have the lowest.

Goethite and hematite abundance (G/H) and CIW/100 varies with depth for the Virginia soils and within the B horizons. BB2 has a small range in G/H values throughout the entire profile, whereas AH2, AF2, BB1, LZ1, Pit F, and Site D show large variations in values at different depths and within different B horizons. Higher values tend to occur in Bw or

Bg/Btg horizons. The chemical index of weathering (CIW/100) is either very low (~0) or variable within soil profiles. AH2, BB1, BB2, LZ1, and Site D have consistent values around 0. Profiles AF1 (0.216 – 0.705) and Pit F (0.434 – 0.813) have highly variable

CIW/100, with values for Pit F increasing with depth.

4.1.1 Middle/Late Miocene Bryn Mawr (BM)

4.1.2 Middle Pleistocene Emmaus Kame (EK)

Pit F

Millheim Narrows (MHN)

AH2

4.1.3 Late Pleistocene Penn State Agricultural Center (PS)

BB2

AF1

BB1

LZ1

Site D

4.2 Predicted Mean Annual Precipitation (MAP) Hyland et al.’s (2015) function was used to convert the goethite-hematite ratio (G/H) to mean annual precipitation (MAP) using an average G/H value for each B horizon(s) (Table

2). B horizons unaltered by diagenesis were selected when possible, but could not be avoided for sites LZ1 and MHN. Average G/H values range from 0.062 in a Virginia soil

(AH2) to 26.745 in the oldest soil profile in Maryland (BM). Predicted MAP values range

from 111 mm/yr to 26,887 mm/yr according to the linear function of Hyland et al. (2015;

Figure 19).

Table 2: Summary table of data for each soil profile used to calculate MAP. Soil Pit Horizon(s) # Samples Average G/H Predicted MAP (mm/yr) AF1 B 2 0.110 160 AH2 Bt1, Bt2 3 0.062 111 BB1 Bt2, 2Btb 3 0.153 202 BB2 Bt2, 2Btb 2 0.068 118 BM Btb3, Btb4 3 26.745 26,887 EK 2Bt2 2 0.226 276 LZ1 Btg 2 0.101 151 MHN 2B (local gley) 2 0.274 325 Pit F 2Btb2 4 0.438 488 PS 4Bt1b, 4Bt2b, 4Bt3b 4 0.688 739 Site D Bt1, Bt2 2 0.082 131

The average G/H and predicted MAP (mm/yr) are plotted on Hyland et al.’s linear function

(Figure 1) to compare modern estimated MAP values with values derived from older paleosols and modern soils (Figure 18). Most paleosol or soil-derived MAP values plot dryer than modern conditions.

2000

1800

1600

1400

1200 VA

1000 MD PA

800

600 Mean annual precipitation (mm / yr) / (mm precipitation annual Mean 400

200

0 0 0.5 1 1.5 2 Goethite-hematite ratio (G/H)

Figure 19: Predicted Mean Annual Precipitation (MAP) value based on an average G/H value for each B horizon(s) plotted on Hyland et al’s linear function (2015). Red, yellow, and orange X’s show the estimated present day MAP values for each state based on 1981-2010 Normals (Virginia (VA), Pennsylvania (PA), and Maryland (MD) (Durre et al., 2012; Durre et al., 2012; Durre and Squires, 2015). The top right arrow indicates the existence of a point off scale at a G/H of ~26.7 and predicted MAP ~ 26,887 for the Bryn Mawr (BM) soil pit in Maryland. The majority of samples plot at dryer (lower MAP) conditions compared to the present.

5.0 Discussion

5.1 Predicted Paleoprecipitation with Time The age of each soil is used to interpret MAP with time in the mid-Atlantic region (Figure

20). Ranges in predicted MAP values are shown as a reflection of changes in the G/H within the B horizon. The time of soil development and the pedogenic maturation of the

G/H is presented as a range beginning at the age of the parent material and lasting for ~100

KA, depending on the uncertainty of the chronological method (Harden, 1982).

MAP values through the time of soil development indicate relative wetter or drier conditions through geologic time, but may not be an accurate reflection of precise predicted

MAP. The extreme MAP value found for BM (~27,000 mm/yr) indicates wetter conditions

than present day, but is not a reasonable estimate of MAP in Maryland during the Middle to Late Miocene (See Section 5.5). Soils in the Middle Pleistocene indicate lower MAP, or dryer conditions, relative to present day. In comparison to the Middle Pleistocene, the Late

Pleistocene shows relatively wetter conditions, but still dryer in comparison to present. The younger South Anna River soils, which developed during the Late Pleistocene to modern day, were not included in the determination of MAP trends with time because of possible diagenetic impacts on the G/H from water table fluctuations and inheritance of hematite from the parent material (See Sections 5.6 & 5.7).

5.2 Comparison with other Indices

1600 AH2 AF1 1400 BB1 BB2 1200 LZ1 PitF 1000 SiteD MHN EK 800 PS 1:1 600

Linear (1:1) G/H Predicted Predicted MAP G/H 400

200

0 0 200 400 600 800 1000 1200 1400 1600 CIW Predicted MAP

Predicted MAP results from Hyland’s G/H proxy (2015) are compared to results using the

CIW proxy from Sheldon et al. (2005) (Figure 21). The two proxies produce different predicted MAP for the same soil samples and overall do not exhibit a 1:1 relationship.

Figure 21 shows that either the CIW proxy is over predicting MAP, the G/H is under predicting MAP, or a particular soil is not appropriate for one of the indices.

1600 AH2 AF1 1400 BB1 BB2 1200 LZ1 PitF 1000 SiteD MHN EK 800 PS 1:1 600

Linear (1:1) G/H Predicted G/HPredicted MAP

400

200

0 0 200 400 600 800 1000 1200 1400 1600 CIW Predicted MAP

Figure 12: Predicted mean annual precipitation (MAP) from Hyland et al’s G/H proxy (2015) and Sheldon et al.’s CIW proxy (2002) for all soil pits, excluding BM in Maryland. Data points reflect an average MAP value of samples within the B horizon that most closely meet the G/H proxy criteria (Table 2). A large disagreement between the two proxies remains and can possibly be attributed to the limited applicability of the CIW proxy.

Figure 22 shows average MAP values from samples within the B horizon that most closely meet the criteria for the G/H proxy (See Section 3.6). There is still a strong disagreement between the CIW and G/H proxy, particularly at sites with a higher predicted MAP. Part of this disagreement stems from an inability to meet the criteria of the CIW proxy (See

Section 3.7). Many of the soils, particularly soils with CIW values around 0 or 100, did not 39

exhibit a 5-8 value change in CIW from the parent (Sheldon et al., 2002). Of all soil profiles presented, only three (Pit F, AF1, BM) met criteria for both the G/H and CIW proxy (Table

3).

Table 3: Predicted MAP using both the CIW and G/H proxy for three soil profiles that best met both proxies’ criteria. Site Name Horizon Depth CIW Predicted MAP G/H Predicted MAP AF1 B 30 251 141 AF1 B 60 888 178 PitF 2Btb2 80 1,031 671 PitF 2Btb2 80 1,031 535 PitF 2Btb2 110 1,097 319 PitF 2Btb2 110 1,097 430 BM Btb3 250 221 35,404

Large discrepancies are found between predicted MAP for all three sites and cannot be fully explained. However, both proxies, principally the CIW proxy, could be affected by aeolian and aerosolic inputs, which would impact bulk elemental and iron oxide chemistry.

Additionally, the CIW proxy will over predict at higher values because of its limited MAP range and saturation at higher values (>1600 mm/yr) and the G/H will under predict, or provide a minimum estimate of MAP, in soils that exhibit highly rubified, oxidized horizons inconsistent with climate conditions. This oxidization will convert iron oxides into hematite and will lower the G/H with time.

5.3 Deposition of Sediments vs. Onset of Pedogenesis The amount of time for stabilization of the G/H in these soils was chosen based on findings that soil properties tend to stabilize or become less variable following 105 years of soil development (Harden, 1982). Many of the soil ages are based on dating of soil units

stratigraphically below the sampled soils and are most indicative of the earliest time that soil development and subsequent formation of the iron oxides could have started.

In the case of the South Anna River soils in Louisa, Virginia, parent ages and the onset of pedogenesis are more refined by OSL and IRSL results, but the pedogenic maturation of the G/H in the Late Pleistocene deposits is uncertain. Soil development and stabilization would not be likely until incision of the South Anna River was significant enough to limit periodic flooding of river terrace deposits. The youngest river terrace soils, Site D and LZ1, are presently river proximal and likely experiencing flooding during high stage. The oldest river terrace, Pit F (~400 KA), has been long removed from any influence of the river and shows G/H values consistent with other Middle Pleistocene soils (Figure 20). However, the remaining terrace deposits are all Late Middle Pleistocene to Late Pleistocene in age and may still remain in a state of continued pedogenesis. AH2 likely experienced a wide range of precipitation regimes given its window of development from the late Middle

Pleistocene to the Late Pleistocene. Virginia soils with ages in the Late Pleistocene (AF1,

BB1, BB2, LZ1, Site D) may not have yet reached stability based on their more recent age of development (<100 KA).

5.4 Presence of Gleying The presence of gleying in the B horizon should be given special consideration when applying the G/H proxy (Hyland et al. 2015). Postburial gleying can result in the precipitation of secondary goethite, resulting in higher MAP estimates (Hyland et al., 2015;

Peppe et al., 2009). Two sites (MHN, LZ1) in this study exhibited either local or substantial gleying within the B horizon, particularly at depth, which could result in higher G/H values.

However, MAP estimates from MHN are consistent with other soil pits and do not appear influenced by local gleying. The slight change in G/H values within MHN are consistent with a change in the CIW values (Figure 11). Changes in the G/H may be a reflection of variability in the B horizon, which has been shown in other pits (Figure 10, 13, 16), or soil response to variability in precipitation, rather than the influence of gleying.

Two other sites in Pennsylvania which were excluded from this study exhibited substantial gleying and a large range of G/H values within their B horizon, resulting in unreliable estimates of

MAP. When gleying is present and cannot be avoided, care should be taken to determine the impact of gleying on the G/H by Figure 23: Distribution of goethite around a root channel within a hematitic soil (Stucki et al., 1988). sampling vertically and horizontally throughout the B horizon. If the G/H is highly variable, the soil profile may not provide reliable estimates of MAP. G/H changes within a soil profile should be accompanied with an analysis of changes in elemental composition or weathering with depth for comparison.

5.5 Influence of Organic Matter Organic matter is an important factor influencing iron (III) oxide formation in soils (Figure

3). Because of the variability of organic matter type and distribution with pedoclimate, or

the microclimate within a soil, there is an inherent relationship between goethite and hematite distribution in soils with organic matter (Stucki et al., 1988). At a smaller scale, the influence of organic matter can be seen around root channels (Figure 23). One possible explanation for an anti-hematitic environment created by organic matter has been suggested by Schwertmann (1971). Organic matter acts as a complexing agent of inorganic

Fe (III) cations, resulting in a lower activity of ions in solution. This lower activity prevents the solubility product of ferrihydrite, but not the lower solubility product of goethite, from being met. Since ferrihydrite is needed to form hematite (Figure 3), only goethite will be able to form in environments with high organic matter. Additionally, organic matter often creates a reducing environment in poorly drained soils that reduces hematite to Fe(II), inhibiting its formation and leaving behind goethite (Fey, 1983). Both of these processes are important to consider when evaluating soils with high G/H values.

This dataset presents one specific soil location, BM in Cecil County, Maryland, that exhibits G/H values in its B horizon ~27, resulting in unreasonably high MAP estimates

(~27,000 mm/yr). It is likely that the BM soil had high organic matter at the time of iron oxide formation, resulting in the preferential formation of goethite over hematite and a high

G/H. Based on a palynology study of the Bryn Mawr Formation, the paleoenvironment was similar in climate to the gulf coastal plain, evidenced by the appearance of bald cypress pollen (Pazzaglia et al., 1997). Though the G/H in the BM soils does not translate to an accurate MAP, it does provide further evidence to support an organic-rich, wet terrestrial environment during the Middle to Late Miocene in Maryland.

5.6 Influence of Hydrology and Water Table Fluctuations Previous studies have shown the influence of topography and, by extension, hydrology, on the formation of goethite and hematite (Stucki et al., 1988; Curi and Franzmeier, 1984;

Santana, 1984; Adams and Kassim, 1984). Poorly drained soils are more likely to form goethite than well drained soils due to an increase in wetness and the creation of a reducing environment. Fluctuations in the water table caused by channel proximity and periodic flooding of soils can also result in artificially elevated G/H (Hyland et al. 2015; Torrent et al., 2010). Periodic flooding would inhibit the dehydration step required to form hematite

(Figure 3) and may result in dissolution of existing hematite under certain conditions.

The South Anna River soils in this study are formed on the treads of fluvial terraces, which experienced water table fluctuations during deposition as the river has flooded and incised through time. Once terraces are formed, pedogenesis sets in on well drained terrace treads upland, but deposits in the lowland may receive additional deposition from adjacent hillslopes or remain periodically flooded. However, the G/H of these soils are relatively low compared to other soil sites in the Mid-Atlantic region, indicating there was not preferential formation of goethite (Table 2) or goethite formation was outweighed by the formation or inheritance of hematite (See 5.7).

5.7 Impacts of Parent Material Composition on G/H Parent lithologies that have easily weatherable iron, such as igneous rock, will preferentially form ferrihydrite, the precursor to hematite. This preferential formation of ferrihydrite will lead to soils higher in hematite content on igneous rock bodies (Stucki et al., 1988; Figure 3). Therefore, inheritance of hematite from soils developed on igneous

rocks must be considered when applying the G/H proxy. The South Anna River soils in

Virginia are all formed through saprolite and residuum reworked from the surrounding landscape, which notably includes weathering inputs from the Ellisville and Green Springs plutons. A study of these plutons showed a hematite composition ranging from 2.87% in the Ellisville to 6.38% in the Green Springs (Wilson, 2001).

Inherited hematite from the parent material would result in decreased G/H ratios and low predicted MAP in comparison to other soils from the same time period, which is seen in all Late Pleistocene soils from the South Anna River (Figure 20). Interestingly, Pit F exhibits G/H values consistent with other Middle Pleistocene soil pits and appears unaffected by inherited hematite from its parent material. One possible explanation for this is the time of soil development. Pit F has had more time to develop and weather from its parent material, thus overprinting the inherited hematite, as evidenced by higher CIW values throughout the profile (Figure 16).

5.8 Total Dithionite-Extractable (DCB) Iron Threshold

100000 AH2 AF1 BB1 BB2 BM EK LZ1 MHN Pennsylvania Pit F PS Site D 10000

Maryland

G/H Development Threshold

1000 Total DCB Iron (Fed) (ppm) (Fed) Iron DCB Total

Virginia Less Developed (G/H Unreliable)

100 0.01 0.1 1 10 Goethite-Hematite Ratio

The reliability of the G/H is impacted by the total amount of DCB iron (Fed) that forms during pedogenesis (Figure 24). DCB iron includes amorphous and crystalline phases, such as magnetite, ferrihydrite, goethite, and hematite. As a soil develops, it will accumulate DCB iron with time depending on the amount of weatherable minerals containing iron. Crystalline phases of iron, such as hematite, can also be inherited from the parent material and contribute to the overall amount of DCB iron. In a soil with low total

DCB iron, variations in MAP can significantly impact the G/H value, whereas soils with 46

high total DCB iron will only shift the G/H value if the MAP conditions persist over longer time scales. Therefore, soils with higher total DCB iron will produce a more stable estimate of long-term MAP.

Caution is warranted when the G/H proxy is applied in soils with low DCB iron, as the lower iron concentrations allow significant changes in the G/H value on short time scales.

A threshold value of 1000 mg/kg (dashed line, Figure 24) was chosen for this study based on the distinction between Pit F and the younger South Anna River soils (AH2, AF1, BB1,

BB2, LZ1, and Site D). The younger soils were shown to be underdeveloped and produce unreasonable MAP estimates, whereas Pit F had developed sufficiently beyond the parent material to produce plausible estimations of MAP. Soil pits from Pennsylvania and

Maryland plot well above this threshold value. This threshold should be considered in concert with other variables impacting the G/H when applying the proxy to produce estimations of MAP in paleosols.

5.9 Use of Magnetic Methods to Determine G/H Magnetic methods have been increasingly employed to quantify magnetic mineral abundance in soils and paleosols due to the cost and difficulty associated with other techniques, such as X-Ray diffraction (XRD) (Kampf and Schwertmann, 1983), diffuse reflectance spectroscopy (DRS; Balsam et al., 2004; Torrent et al., 2007; Zhang et al.,

2007; Lyons et al., 2014; Hu et al., 2015), or Mössbauer spectroscopy (Carter-Stiglitz et al., 2006). Magnetic determination of goethite and hematite abundance has many advantages, including a lower cost, higher sensitivity at low volumes, and the ability to

analyze more samples in a shorter period of time, but a standard protocol is yet to be defined

(Maxbauer, 2016).

In this paper, we applied a new method of determining G/H that has an advantage over other methods used, such as IRM (Isothermal Remanent Magnetization) which identifies goethite and hematite primarily using their coercivity (e.g., Hyland et al. 2015). Minor amounts of cation substitution, primarily with aluminum, can occur in goethite and alter their coercivity, making it difficult to accurately determine their magnetic abundance (Liu et al., 2006; Roberts et al., 2006). This new method distinguishes goethite and hematite with thermal demagnetization, rather than coercivity, in light of this possible issue. Despite this advantage, there are some inherent limitations in using the presented magnetic methods to determine G/H. The methods employed in this study assumes that all magnetic minerals present will be saturated at an applied field of 5T. However, high field (as much as >57 T;

Rochette et al., 2005) may be required to fully saturate goethite. Additionally, soils often contain high concentrations of superparamagnetic grain sizes, which would not retain remanence or appear using the analyses presented in this paper (Maher, 1998; Guyodo et al., 2006; Till et al., 2015).

Despite the possible error associated with magnetic determination of goethite and hematite abundance via these methods, we present a viable alternative to mass-based estimates of

G/H, which present their own complications. XRD methods were attempted earlier in this study to compare magnetic abundance of G/H with those obtained from a mass based method (e.g., Hyland et al. 2015). Complications arise in using XRD methods to determine abundance of goethite and hematite in soils. Soils often have abundant constituents, such 48

as clay, silica, and other minerals, that create noisy XRD patterns. Noisy patterns cause difficulty in curve-matching and quantitative analysis of goethite and hematite peaks. This issue is often addressed by chemically treating the clay fraction of a soil sample with NaOH to remove the 1:1 layer silicates and gibbsite prior to XRD analysis (Kamp and

Schwertmann, 1983). This method, especially if employed on many soil samples, can be costly and time consuming compared to magnetic methods.

6.0 Conclusions

The application of the G/H to MAP proxy on Miocene to Late Pleistocene paleosols in the mid-Atlantic region provides insights into the reliability and usage of this proxy to constrain past terrestrial climate conditions through time. This study makes the following conclusions:

 MAP values through the time of soil development indicate relative wetter or drier

conditions through geologic time, but may not be an accurate reflection of precise

predicted MAP. The extreme MAP value found for BM (~26,000 mm/yr) indicates

wetter conditions than present day, but is not a reasonable estimate of MAP in

Maryland during the Middle to Late Miocene. Soils in the Middle Pleistocene

indicate lower MAP, or dryer conditions, relative to present day. In comparison to

the Middle Pleistocene, the Late Pleistocene shows relatively wetter conditions, but

still dryer in comparison to present.

 The CIW and G/H proxy produce dissimilar MAP for the same samples. The CIW

proxy over predicts at higher values due to its limited MAP range and the G/H

under predicts or provides a minimum MAP estimate in soils exhibiting oxidation

of iron oxides into hematite.

 The timing of deposition of sediments and the onset of pedogenesis should be

considered when constraining the time of paleosol formation. A variety of

chronological methods can be used to constrain soil ages, but pedogenic maturation

of the G/H could take up to 105 years and may represent integration of MAP through

a window of time (Harden, 1982).

 Characterization of the G/H in paleosols can result in extreme or inaccurate results

in predicted MAP under specific conditions. The presence of abundant organic

matter at the time of soil formation will preferentially form goethite and result in

high estimated MAP values. Gleying and water table fluctuations can also result in

elevated G/H values in some soils. Inheritance of hematite from the parent material

can result in lower G/H values and unreliable MAP estimates. However, with

increasing soil development, soils can overprint inherited hematite and reliably

reflect climatic controls on iron oxide formation. Sampling for the G/H proxy

should be avoided in gleyed or modern river-proximal soils.

 A total dithionite-extractable iron (Fed) threshold for application of the G/H on

paleosols has been proposed as 1000 mg/kg. The G/H proxy should not be used if

total DCB iron is low enough to allow for significant changes in the G/H on short

time scales.

 The determination of goethite-hematite abundance using magnetic methods has

many advantages over mass based methods. The new method presented in this

paper relies on the determination of goethite abundance using its Neel temperature

rather than coercivity, which can be effected by cation substitution (Liu et al., 2006;

Roberts et al., 2006).

The magnetic method discussed in this study could be used to further refine the relationship between G/H and MAP in modern soils and provide a more accurate quantification of goethite-hematite abundance in soil samples. Future work should apply the G/H to MAP proxy to paleosols in other regions to determine its reliability as a paleoprecipitation indicator in other areas and in different soil types, taking into consideration the total DCB iron threshold and other complicating factors in the G/H.

REFERENCES

Abrajevitch, A., Van der Voo, R., Rea, D.K., 2009. Variations in relative abundance of goethite and hematite in Bengal Fan sediments: Climatic vs. diagnetic signals. Marine Geology 267, 191-206. Adams, W.A., and Kassim, J.K., 1984. Iron oxyhydroxides in soils developed from lower Palaeozoic sedimentary rocks in mid Wales and implications for some pedogenetic processes. J. Soil Sci., 35, 117-126. Balsam, W.L., Ellwood, B.B., Ji, J., Williams, E.R., Long, X., El Hassani, A., 2011. Magnetic susceptibility as a proxy for rainfall: worldwide data from tropical and temperate climate. Quat. Sci. Rev., 30, 2732-2744. Bjorlykke, K., 2010, Heat transport in sedimentary basins, in Bjorlykke, K., ed., Petroleum Geoscience: From Sedimentary Environments in Rock Physics: New York, Springer, 253-260. Brimhall, G.H., Dietrich, W.E., 1987. Constitutive mass balance relations between chemical composition, volume, density, porosity, and strain in metasomatic hydrochemical systems: results on weathering and pedogenesis. Geochimica et Cosmochimica Acta, 51(3), 567-587. Buggle, B., Glaser, B., Hambach, U., Gerasimenko, N., Markovic, S., 2011. An evaluation of geochemical weathering indices in loess – paleosol studies. Quaternary International, 240 (2011), 12-21. Carter-Stiglitz, B., Banerjee, S.K., Gourlan, A., Oches, E., 2006. A multi-proxy study of Argentina loess: marine oxygen isotope stage 4 and 5 environmental record from pedogenic hematite. Palaeogeogr. Palaeoclimatol. Palaeoecol. 239, 45–62. http://dx.doi.org/10.1016/j.palaeo.2006.01.008. Ciolkosz, E.J., Walman, W.J., Thurman, N.C., 1993. Iron and Aluminum in Pennsylvania, Agronomy series (USA). Cook, E.R., Woodhouse, C.A., Eakin, C.M., Meko, D.M., Stahle, D.W., 2004. Long-Term Aridity Changes in the Western United States. Science, 306, 5698, 1015-1018. Costantini, E.A.C., Lessovaia, S., Vodyanitskii, and Yu., 2006. Using the analysis of iron and iron oxides in paleosols (TEM, geochemistry and iron form) for the assessment of present and past pedogenesis. Quaternary International, 156-157 (2006) 200- 211. Cummins, T., 2015. Geochemistry and pedogenesis of three late Cenozoic paleosols in east central Pennsylvania. Bachelor Thesis, Lehigh University. Curi, N., and Franzmeier, D.P., 1984. Toposequence of Oxisols from the central plateau of Brazil. Soil Sci. Soc. Am. J., 48, 341-346.

Durre, I., M. F. Squires, R. S. Vose, X. Yin, A. Arguez, and S. Applequist, 2012: NOAA's 1981-2010 U.S. Climate Normals: Monthly Precipitation, Snowfall, and Snow Depth. Journal of Applied Meteorology and Climatology, 52, 2377– 2395, doi:10.1175/JAMC-D-13-051.1 Durre, I., M. F. Squires, R. S. Vose, A. Arguez, S. Applequist, and X. Yin, 2012: Computational Procedures for the 1981-2010 Normals: Precipitation, Snowfall, and Snow Depth. NCDC Report, 10 pp. Durre, I., and M. F. Squires, 2015: White Christmas? An Application of NOAA's 1981- 2010 Daily Normals. Bulletin of the American Meteorological Society, 96, 1853- 1858, dx.doi.org/10.1175/BAMS-D-15-00038.1 Dykman, J., 2015. Alignment and divergence of pedologic, geomorphic, and geochemical data for Critical Zone hillslope soils in central PA.Bachelor Thesis, Lehigh University. Eppes, M.C., 2011. Laboratory Instruction Manual, Department and Geography and Earth Sciences Soil Laboratory. Fey, M.V., 1983. Hypothesis for the pedogenic yellowing of red soil materials. Techn. Commun., Dept. Agric. Fisheries, Rep. South Africa, 18, 130-136. Fischer, W.R. and Schwertmann, U., 1975. The formation of hematite from amorphous iron (III) hydroxide. Clays and Clay minerals, 23, 33-37. Guyodo, Y., LaPara, T.M., Anschutz, A.J., Penn, R.L., Banerjee, S.K., Geiss, C.E., Zanner, W., 2006. Rock magnetic, chemical and bacterial community analysis of a modern soil from Nebraska. Earth Planet. Sci. Lett. 251, 168–178. http://dx.doi.org/10.1016/j.epsl.2006.09.005. Harnois, L. 1988. The CIW Index: a new chemical index of weathering. Sedimentary Geology 55, 319-322. Hughes, K.S., Hibbard, J.P., and Miller, B.V., 2013, Relationship between Ellisville pluton and Chopawamsic fault; Establishment of significant Late Ordovician faulting in the Appalachian piedmont of Virginia: American Journal of Sicence, 313, 584-612. Hu, P., Liu, Q., Heslop, D., Roberts, A.P., Jin, C., 2015. Soil moisture balance and magnetic enhancement in loess–paleosol sequences from the Tibetan Plateau and Chinese Loess Plateau. Earth Planet. Sci. Lett. 409, 120–132. Hyland, E.G., Sheldon, N.D., Van der Voo, R., Badgley, C., and Abrajevitch, A., 2015. A new paleoprecipitation proxy based on soil magnetic properties: implications for expanding paleoclimate reconstructions. Geological Society of America Bulletin, doi: 10.1130/B31207.1. Jenny, Hans, 1941. Factors of Soil Formation: A System of Quantitative Pedology. Dover Publications, Inc. New York.

Kirby, E. and Harkins, N., Personal Communication, Unpublished. Kodama, K. and Pazzaglia, F.J., Personal Communication, Unpublished. Kruiver, P.P., Dekkers, M. J., and Heslop, D., 2001. Quantification of magnetic coercivity components by the analysis of acquisition curves of isothermal remanent magnetization: Earth and Planetary Science Letters, 189, 269-276, doi: 10.1016/S0012-821X(01)00367-3. Lair G.J., Zehetner, F., Hrachowitz, M., Franz, N., Maringer, F.J., Gerzabek, M.H., 2009. Dating of soil layers in a young floodplain using iron oxide crystallinity. Quartnary Geochronology, 4(3), 260-266. Liu, Q., Yu, Y., Torrent, J., Roberts, A.P., Pan, Y., Zhu, R., 2006. Characteristic low temperature magnetic properties of aluminous goethite [α-(Fe, Al)OOH] explained. J. Geophys. Res. B: Solid Earth 111, 1–12. Lyons, R., Tooth, S., Duller, G.A.T., 2014. Late Quaternary climatic changes revealed by luminescence dating, mineral magnetism and diffuse reflectance spectroscopy of river terrace paleosols: a new form of geoproxy data for the southern African interior. Quat. Sci. Rev. 95, 43-59. Ma, L., Jin, L. and Brantley, S.L. (2011) Geochemical behaviors of different element groups during shale weathering at the Susquehanna/Shale Hills Critical Zone Observatory. Applied Geochemistry, 26, S89-S93. Mack, G.H., James, W.C., and Monger, H.C., 1993, Classification of paleosols: Geological Society of America Bulletin, 105, 129-136, doi:10.1130/0016- 7606(1993)105<0129:COP>2.3.CO;2. Maher, B.A., 1998. Magnetic properties of modern soils and Quaternary loessic paleosols: paleoclimatic implications. Palaeogeogr. Palaeoclimatol. Palaeoecol. 137, 25–54. http://dx.doi.org/10.1016/S0031-0182(97)00103-X. Maxbauer, D.P., Feinberg, J.M., and Fox, D.L., 2016. Magnetic mineral assemblages in soils and paleosols as the basis for paleoprecipitation proxies: A review of magnetic methods and challenges. Earth-Science Reviews, 155, 28-48. McGavick, M., 2017. Extrinsic terrace formation processes and tectonically driven river incision along the South Anna River, Virginia. M.S. Thesis, Lehigh University. McKeague, J.A. and Day, J.H. (1966). Dithionite-and oxalate-extractable Fe and Al as aids in differentiating various classes of soils. Canadian Journal of Soil Science, 46(1), 13-22. Mehra, O.P. and Jackson, M.L. (1960). Iron oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate. In Proc. 7th nat. Conf. Clays., Vol. 5, 317-327

Nimmo, J. R. 2004, Porosity and Pore Size Distribution, in Hillel, D., ed. Encyclopedia of Soils in the Environment: London, Elsevier, 3: 295. Pazzaglia, F.J., Robinson, R.A.J., and Traverse, A., 1997. Palynology of the Bryn Mawr Formation (Miocene): insights on the age and genesis of Middle Atlantic margin fluvial deposits. Sedimentary Geology, Vol. 108, 19-44. Pelletier, J.D. et al., 2015, Forecasting the response of Earth’s surface to future climatic and land use changes: A review of methods and research needs, Earth’s Future, 3, 220-251, doi: 10.1002/2014EF000290. Peppe, D.J., Evans, D.A.D., and Smirnov, A.V., 2009, Magnetostratigraphy of the Ludlow Member of the Fort Union Formation (Lower Paleocene) of the Williston Basin in North Dakota: Geological Society of America Bulletin, 121, 65-79, doi:10.1130/B26353.1. Retallack, G. J., 2013. Soils of the Past: An Introduction to Paleopedology, Springer, doi: 10.1002/9780470698716 Roberts, A.P., Liu, Q., Rowan, C.J., Chang, L., Carvallo, C., Torrent, J., Horng, C.S., 2006. Characterization of hematite (α-Fe2O3), goethite (α-FeOOH), greigite (Fe3S4), and pyrrhotite (Fe7S8) using first-order reversal curve diagrams. J. Geophys. Res. Solid Earth 111, 1–16. http://dx.doi.org/10.1029/2006JB004715. Rochette, P.,Mathé, P.E., Esteban, L., Rakoto, H., Bouchez, J.L., Liu, Q., Torrent, J., 2005. Nonsaturation of the defect moment of goethite and fine-grained hematite up to 57 Teslas. Geophys. Res. Lett. 32, 1–4. http://dx.doi.org/10.1029/2005GL024196. Santana, D.P., 1984. Soil formation in a toposquence of Oxisols from Patos De Minas region, Minas Gerais State, Brazil. Ph.D. Thesis, Purdue Univerity. Schwertmann, U., 1988. Occurrence and formation of iron oxides in various pedoenvironments. See Stucki et al. 1988, pp. 267-308. Schwertmann, U., 1971. Transformation of hematite to goethite in soils. Nature, 232, 274- 285. Schwertmann, U. and Fitzpatrick, R.M., 1992. Iron minerals in surface environment. Catena Supplement 21, 7-30. Schwertmann, U. and Taylor, R.M., 1977. Iron oxides. Minerals in Soil Environments. Soil Sci. Soc. Of America, 145-180. Schwertmann, U. and Taylor, R.M., 1989. Iron oxides. Minerals in Soil Environments, second ed. Soil Science Society of America Book Series, 1, 279-438 (Chapter 8). Schwertmann, U., Murad E., and Schulze, D.G., 1982. Is there Holocene reddening (Hematite formation) in soils of axeric temperature areas? Geoderma, 27, 209-223.

Schulze, D.G. 1981. Identification of soil iron-oxide minerals by differential x-ray diffraction. Soil Science Society of America Journal, 45, 437-440. Shackleton, N.J., Berger, A., Peltier, W.R., 1990. An alternative astronomical calibration of the Lower Pleistocene timescale based on ODP site 677. Transactions of the Royal Society of Edinburgh 81, 251-261. Sheldon, N.D., Retallack, G.J., and Tanaka, S., 2002. Geochemical Climofunctions from North American Soils and Application to Paleosols across the Eocene-Oligocene Boundary in Oregon. The Journal of Geology, 110, 687-696. Sheldon, N.D. and Tabor, N.J., 2009. Quantitative paleoenvironmental and paleoclimatic reconstruction using paleosols, Earth-Science Reviews, doi: 10.1016/j.earscirev.2009.03.004 Soil Survey Staff. 2014. Kellogg Soil Survey Laboratory Methods Manual. Soil Survey Investigations Report No. 42, Version 5.0. R. Burt and Soil Survey Staff (ed.). U.S. Department of Agriculture, Natural Resources Conservation Service. Spears, D. B, Evans, N. H., and Gilmer, A. K., 2013, Geologic map of the Pendleton quadrangle, Virginia: Virginia Department of Mines, Minerals and Energy, Division of Geology and Mineral Resources, 2013 STATEMAP deliverable, 1:24,000-scale geologic map. Spell, T.L., McDougall, I., 1992. Revisions to the age of the Brunhes/Matuyama boundary and the Pleistocene geomagnetic polarity timescale. Geophysical Research Leterrs 19, 1182-1184. Stucki, J.W., Goodman, B.A., and Schwertmann, U., eds. 1988. Iron in Soils and Clay Minerals. Dordrecht, The Netherlands: Reidel. Summerfield, C., 2015. Pedogenesis and geochronology of the Bryn Mawr Formation (Late Miocene?) in Cecil County, MD, a possible analogue environment of future mid-Atlantic climates. Bachelor Thesis, Lehigh University. Tabor, N.J. and Timothy, S.M., 2015. Paleosols as Indicators of Paleoenvironment and Paleoclimate. Annu. Rev. Earth Planet. Sci. 2015. 43, 333-361. Tauxe, L., Herbert, T., Shackleton, N.J. Kok, Y.S., 1996. Astronomical calibration of the Matuyama-Brunhes boundary: consequences for magnetic remanence acquisition in marine carbonates and the Asian loess sequences. Earth and Planetary Science Letters 140, 133-146. Till, J.L., Guyodo, Y., Lagroix, F., Morin, G., 2015. Goethite as a potential source of magnetic nanoparticles in sediments. 43, 75–78. http://dx.doi.org/10.1130/G36186.1. Torrent, J., Liu, Q.S., Blomendal, J., Barron, V., 2007. Magnetic enhancement and iron oxides in the upper Luochuan loess-paleosol sequence, Chinese Loess Plateau. 71.

Torrent, J., Liu, Q., and Barron, V., 2010, Magnetic susceptibility changes in relation to pedogenesis in a Xeralf chronosquence in northwestern Spain: European Journal of Soil Science, v. 61, p. 161-173, doi: 10.1111/j.1365-2389.2009.01216.x. Torrent, J., Schwertmann, U., Fechter, H., and Alferez, F., 1983. Quantitative relationships between soil color and hematite content. Soil Science, 136, 354-358. Torrent, J., Schwertmann, U., and Schulze, D.G., 1980. Iron oxide mineralogy of some soil of two river terrace sequences in Spain. Geoderma, 23, 191-208. Wilson, J.R., 2001. U/Pb Zircon Ages of Plutons from the Central Applachians and GIS- based assessment of Plutons with Comments on Their Regional Tectonic Significance. M.S. Thesis, Virginia Polytechnic Institute and State University. Yapp, C., 2001. Rusty relics of earth history: iron(III) oxides, isotopes, and surficial environments. Annu. Rev. Earth Planet. Sci., 29, 165-199. Zhang, Y.G., Ji, J., Balsam, W.L., Chen, J., 2007. High resolution hematite and goethite records from ODP 1143, South China sea: co-evolution of monsoonal precipitation and El Nino over the past 600,000 years. Earth Planet. Sci. Lett. 264, 136-150.

APPENDICES

Appendix 1. Extended Procedures Particle Size Distribution Analysis Particle size distribution analysis (PSDA) is performed on every soil sample. Soil samples (100 grams) are wet sieved using a -1 ϕ sieve to separate the >2mm fraction. Ten grams of the <2mm fraction is placed in a 100 mL beaker and wetted with deionized water before adding 5 ml of 30% hydrogen peroxide under a fume hood. When frothing of the sample slows, the mixture is placed onto a hot plate to further the reaction. If needed, a second addition of 5 ml of 30% hydrogen peroxide is added. The sides of the beaker are rinsed with deionized water before placing them into an oven set at 90° Celsius to dry the sample and remove remaining hydrogen peroxide. The samples are then re-weighed to determine the percent organics. A dispersing agent (10 mL) is added to the sample to disaggregate remaining fine particles. The dispersing agent is made by dissolving 53.25 g of sodium pyrophosphate (Na4P2O7) and 4.24 g of sodium carbonate (Na2CO310H2O) into one liter of H2O then gently heating and stirring the solution (Eppes, M. C., 2011). The sample is placed in a shaker for at least 8 hours. The sample is sieved through a 4ϕ sieve into a 1500 mL Fleaker to separate the sand fraction from the silt and clay. The sand is then dried, weighed, and set aside. Deionized water is added to bring the volume of the Fleaker containing the silt and clay fraction to 1000 mL. The Fleaker is then shaken until all particles are in suspension and left sitting for 8 hours. A 25 mL pipette is then used to take a fraction (25 mL) of the suspended clay from the Fleaker at a depth of 10 cm below the water surface (Soil Survey Staff, 2014). The pipetted sample of clay and water is added to a pre-weighed tin to be dried and weighed. The percentage of the >2mm fraction is found by dividing its weight by the total dry weight of the initial sample. The following equation is used to determine the percentage of sand separated from the 10 gram sample:

푀푎푠푠 표푓 푠푎푛푑 Sand % = ( ) · (1 - >2mm fraction) (5) 푀푎푠푠 표푓 10푔 푠푎푚푝푙푒

The percent clay is calculated using an equation from the USDA NRCS Soil Sample Methods:

푅푊 ∗퐶퐹 Clay % = (100* ( 2 )) · (1 - >2mm fraction) (6) 푇푊 where RW is the residual mass in grams of the < 2mm fraction, CF is the 100 mL volume divided by the dispended pipet volume (25 mL), and TW is the total weight of the oven dried sample in grams (Cummins, 2015).

Bulk Elemental Analysis Bulk elemental analysis is done using 10 gram sub-samples. The samples are wet sieved through a -1ϕ sieve to remove the >2mm fraction and the sample is dried. A small sample of dry soil is placed into a ceramic ball mill to be crushed into a fine powder for 5 minutes. Approximately 0.2 grams of each sample is mixed with 0.9 grams of lithium metaborate to 58

be shaken in a small glass vial. The mixture is heated in a furnace at 1000°C for 8 minutes, quenched, and subsequently dissolved in 100 mL of 5% nitric acid. After the sample has completely dissolved, it is diluted before analysis with the inductively coupled plasma mass spectrometer (ICP-MS) (Cummins, 2015). The accuracy of the ICP-MS values are tested by comparison to pre-determined USGS pulverized rock standards. Tau ratio values are determined using (Brimhall and Dietrich, 1987):

퐶푗,푤 퐶푖,푝 휏푖푗 = · – 1 (1) 퐶푗,푝 퐶푖,푤 where C is the concentration (mol/m3) of immobile (i) or mobile (j) elements in weathered (w) or parent (p) material. All calculations are referenced to zirconium, which is assumed to be immobile. Positive tau values indicate enrichment of an element, whereas negative values show depletion (Ma et al., 2011).

Iron Oxide Crystallinity

Dithionite-citrate-bicarbonate (DCB)-extractable iron (Fed) is determined using the method of Mehra and Jackson (1960). A 0.5 gram subsample of each selected soil sample is taken and combined with 22.5 mL of 0.3 M sodium citrate and 2.5 mL of 1M sodium bicarbonate in a 50 mL centrifuge tube. The tubes are placed in a preheated heating block at 80°C until the mixture reached the same temperature. 0.5 grams of sodium dithionite is added to the mixture while stirring consistently for one minute, followed with 15 minutes of intermittent stirring. The samples are checked after 15 minutes for reaction completion. If the reaction is not complete, additional doses of sodium dithionite are added. The samples are then diluted for analysis on the ICP-MS (Cummins, 2015).

Ammonium oxalate extractable iron (Feo) follows the procedures of McKeague and Day (1996). A 0.1 subsample of each selected soil sample are combined with 6 mL of 0.2 M ammonium oxalate in a 15 mL centrifuge tube. The mixture is shielded from light during the entire process. It is then sealed and shaken for 4 hours. The sample are then set aside to settle and be diluted for analysis on the ICP-MS (Cummins, 2015).

Appendix 2. Particle Size Distribution Analysis (PSDA) Data

Location Field Site Sample Depth (cm) % >2mm % Sand % Silt % Clay MHN 65 65 81.681 11.375 5.287 1.656 MHN 160 160 52.623 37.075 6.968 3.335 Millheim MHN 210 210 42.536 38.900 18.080 0.485 Narrows MHN 240 240 42.265 45.397 7.857 4.480 MHN 300 300 45.631 42.023 10.427 1.919 EK 0 0 8.504 26.103 46.924 18.469 EK 10 10 14.450 22.526 47.941 15.083 EK 30 30 17.985 20.730 45.532 15.753 EK 40 40 14.979 20.064 40.046 24.911 EK 70 70 10.818 18.604 33.641 36.937 EK 100 100 14.987 18.404 30.824 35.785 EK 110 110 12.804 20.320 28.519 38.357 EK 130 130 8.774 23.876 26.637 40.712 Emmaus EK 160 160 17.559 23.292 27.488 31.662 Kame EK 180 180 10.150 27.415 28.806 33.628 EK 210 210 25.781 50.824 13.807 9.589 EK 230 230 29.001 43.446 17.292 10.261 EK 250 250 35.178 30.252 18.917 15.653 EK 290 290 32.641 33.203 24.053 10.103 EK 320 320 34.272 42.296 18.810 4.621 EK 350 350 42.776 30.728 22.641 3.855 Pennsylvania EK 380 380 44.355 36.739 15.300 3.606 EK 400 400 38.832 41.305 16.109 3.754 PS 15 15 12.824 15.694 49.495 21.987 PS 30 30 15.950 13.082 49.651 21.316 PS 40 40 28.826 8.490 38.779 23.905 PS 60 60 15.549 9.758 38.156 36.537 PS 100 100 18.795 11.234 36.745 33.226 PS 120 120 19.559 11.033 37.267 32.141 PS 145 145 14.243 14.344 37.210 34.202 PS 185 185 18.374 22.198 33.462 25.966 PS 195 195 17.267 25.308 30.828 26.597 Penn Stat PS 220 220 23.813 19.576 34.334 22.277 Ag. Center PS 225 225 10.551 18.083 38.987 32.380 PS 235 235 13.222 9.988 37.572 39.218 PS 245 245 14.502 33.471 6.335 45.692 PS 250 250 0.527 0.626 49.487 49.360 PS 265 265 0.000 0.366 34.392 65.243 PS 305 305 0.000 0.472 47.517 52.010 PS 310 310 0.083 1.382 50.054 48.481 PS 340 340 0.056 0.580 47.386 51.978 PS 350 350 0.101 0.386 52.592 46.921

Location Field Site Sample Depth (cm) % >2mm % Sand % Silt % Clay BM1 0 39.164 36.760 14.152 9.923 BM2 50 38.037 32.530 16.162 13.271 BM3 100 46.125 27.794 10.683 15.399 BM4 150 40.239 36.182 11.381 12.198 BM5 200 38.596 39.963 9.858 11.583 BM6 250 43.360 34.904 11.708 10.027 BM7 300 42.707 41.909 4.362 11.022 Cecil County, Bryn Mawr BM8 350 43.706 35.842 5.589 14.863 Maryland BM9 400 38.019 37.239 11.498 13.243 BM10 450 37.122 15.257 36.913 10.708 BM11 500 39.648 32.781 18.098 9.473 BM12 550 5.850 59.253 19.186 15.711 BM16 280 5.081 72.153 8.377 14.389 BM17 825 29.773 48.600 12.640 8.986 KP 1100 4.661 57.169 7.290 30.879 AH1-Ap-0 0 18.012 55.982 7.750 18.257 AH1-Bw-10 10 16.346 54.477 11.621 17.557 AH1 AH1-Bt-30 30 14.808 39.235 37.958 8.000 AH1-Bt-50 50 5.816 35.397 40.743 18.044 AH1-Bt-60 60 29.201 38.367 33.105 -0.674 AH2-A-0 0 10.160 53.912 14.046 21.882 AH2-Bw-10 10 8.296 39.923 26.747 25.033 AH2 AH2-Bt1-30 30 9.026 39.869 33.449 17.656 AH2-Bt2-50 50 10.257 26.047 39.289 24.407 AH2-Bt2-60 60 20.823 38.198 39.642 1.337 AF1-Ap-20 20 0.959 41.288 13.302 44.450 AF1-B-30 30 3.351 39.413 19.171 38.064 AF1 AF1-B-60 60 2.527 37.489 31.446 28.538 AF1-Bg-95 95 2.495 27.253 46.365 23.887 AF2-Ap-0 0 17.922 49.679 17.730 14.669 South Anna AF2-Bt1-10 10 6.643 48.962 20.318 24.077 River, AF2 AF2-Bt2-30 30 1.161 25.476 51.631 21.731 Virginia AF2-Bt2-48 48 1.203 13.598 70.988 14.212 AF2-2Btb-55 55 0.063 10.507 74.092 15.338 BB1-Ap-0 0 1.657 58.516 9.842 29.986 BB1-Bw-10 10 3.229 52.651 9.494 34.627 BB1-Bt-20 20 1.418 45.676 19.591 33.314 BB1 BB1-Bt2-35 35 2.061 45.134 26.259 26.546 BB1-Bt2-50 50 1.437 29.999 42.644 25.920 BB1-2Btb-65 65 1.168 29.310 45.608 23.914 BB2-AE-10 10 5.135 62.828 4.321 27.717 BB2-Bw-40 40 9.107 50.879 14.757 25.257 BB2 BB2-Bt2-70 70 5.108 41.556 35.936 17.399 BB2-Bt2-77 77 8.491 39.307 41.517 10.685 LF2 0.306 4.670 56.326 38.698 LF LF3 0.150 2.705 48.734 48.411 LF4 1.172 18.800 52.283 27.745

Location Field Site Sample Depth (cm) % >2mm % Sand % Silt % Clay LZ1-A-0 0 5.304 40.166 18.464 36.066 LZ1-E-10 10 1.780 29.030 14.732 54.458 LZ1 LZ1-Bw-30 30 0.333 14.647 42.881 42.139 LZ1-Btg-50 50 0.132 15.105 51.254 33.509 LZ1-Btg-80 80 0.424 27.515 32.960 39.102 PitF-O-0 0 37.624 48.326 10.516 3.534 PitF-Bw-20 20 19.539 45.006 12.456 22.999 PitF-Be-30 30 36.350 35.953 29.098 -1.401 PitF-2Btb1-40 40 25.392 22.934 60.552 -8.878 PitF PitF-2Btb1-60 60 33.347 37.015 47.137 -17.499 South Anna PitF-2Btb2-80 80 49.142 35.503 48.520 -33.165 River, PitF-2Btb2-110 110 0.291 22.781 63.426 13.502 Virginia PitF-BC-120 120 0.000 33.411 45.483 21.106 PitF-BC-150 150 0.000 36.110 43.297 20.593 SiteD-A-0 0 1.166 45.940 11.828 41.066 SiteD-Bw-38 38 0.394 26.347 26.867 46.392 SiteD-Bt1-48 48 0.251 30.665 27.941 41.143 SiteD SiteD-Bt2-64 64 1.230 23.796 35.579 39.395 SiteD-Bg-83 83 0.396 27.205 35.967 36.432 SiteD-BC-100 100 0.063 19.525 40.699 39.713 SiteD-BC-121 121 -0.039 15.163 44.472 40.405 LZ3 LZ3 N/A 0.172 64.558 8.434 26.836 Site E Site E - B 0.130 20.234 54.047 25.589

Appendix 3. FeO/FeD Values

FeO FeD Location Field Site Sample Depth (cm) FeO/FeD (mg/kg) (mg/kg) MHN 65 65 5885 6851 0.859 MHN 160 160 2443 4308 0.567 Millheim MHN 210 210 21312 14248 1.496 Narrows MHN 240 240 5842 6570 0.889 MHN 300 300 4133 5979 0.691 EK 0 0 12597 60244 0.209 EK 10 10 4140 36497 0.113 EK 30 30 6079 51714 0.118 EK 40 40 3489 48431 0.072 EK 70 70 8787 75695 0.116 EK 100 100 7109 81828 0.087 EK 130 130 9089 95773 0.095 EK 160 160 6751 57266 0.118 Emmaus EK 180 180 4854 61615 0.079 Kame EK 210 210 8704 54609 0.159 EK 230 230 5753 44562 0.129 EK 250 250 8477 72698 0.117 EK 290 290 8325 58661 0.142 EK 320 320 5287 50985 0.104 EK 350 350 11975 126906 0.094 Pennsylvania EK 380 380 5640 66970 0.084 EK 400 400 5027 75683 0.066 PS 15 15 5897 18330 0.322 PS 30 30 5553 19752 0.281 PS 40 40 4456 23069 0.193 PS 60 60 7308 27228 0.268 PS 100 100 6402 27307 0.234 PS 120 120 5064 26859 0.189 PS 145 145 4749 28979 0.164 PS 185 185 4249 23739 0.179 PS 195 195 4878 23693 0.206 Penn Stat PS 220 220 3557 25796 0.138 Ag. Center PS 225 225 4626 29189 0.158 PS 235 235 7805 31953 0.244 PS 245 245 7779 37111 0.210 PS 250 250 3026 27237 0.111 PS 265 265 3493 35757 0.098 PS 305 305 3457 30144 0.115 PS 310 310 3311 24930 0.133 PS 340 340 3361 30081 0.112 PS 350 350 4397 25842 0.170

FeO FeD Location Field Site Sample Depth (cm) FeO/FeD (mg/kg) (mg/kg) BM1 0 168 2214 0.076 BM2 50 179 4315 0.041 BM3 100 371 7550 0.049 BM4 150 220 4701 0.047 BM5 200 86 2853 0.030 BM6 250 118 3634 0.032 BM7 300 105 1739 0.060 Cecil County, Bryn Mawr BM8 350 136 3013 0.045 Maryland BM9 400 206 4460 0.046 BM10 450 158 4511 0.035 BM11 500 152 9777 0.016 BM12 550 120 4349 0.028 BM16 280 101 1977 0.051 BM17 825 102 6677 0.015 KP 1100 109 3900 0.028 AH1-Ap-0 0 40 194 0.205 AH1-Bw-10 10 85 206 0.414 AH1 AH1-Bt-30 30 146 849 0.172 AH1-Bt-50 50 268 896 0.299 AH1-Bt-60 60 135 651 0.208 AH2-A-0 0 112 291 0.384 AH2-Bw-10 10 337 591 0.570 AH2 AH2-Bt1-30 30 221 707 0.313 AH2-Bt2-50 50 326 911 0.358 AH2-Bt2-60 60 240 965 0.249 AF1-Ap-20 20 51 132 0.384 AF1-B-30 30 83 198 0.418 AF1 AF1-B-60 60 95 290 0.326 AF1-Bg-95 95 111 470 0.236 AF2-Ap-0 0 78 333 0.234 AF2-Bt1-10 10 91 368 0.248 AF2 AF2-Bt2-30 30 182 891 0.204 South Anna AF2-Bt2-48 48 260 1356 0.192 River, AF2-2Btb-55 55 296 1343 0.221 Virginia BB1-Ap-0 0 51 62 0.820 BB1-Bw-10 10 60 81 0.748 BB1-Bt-20 20 118 219 0.536 BB1 BB1-Bt2-35 35 136 348 0.391 BB1-Bt2-50 50 180 659 0.273 BB1-2Btb-65 65 303 716 0.423 BB2-AE-10 10 47 97 0.486 BB2-Bw-40 40 71 193 0.369 BB2 BB2-Bt2-70 70 206 586 0.351 BB2-Bt2-77 77 172 609 0.282 LF2 238 1217 0.196 LF LF3 311 1853 0.168 LF4 461 2289 0.201 LZ1-A-0 0 218 238 0.914 LZ1-E-10 10 234 314 0.745 LZ1 LZ1-Bw-30 30 448 726 0.618 LZ1-Btg-50 50 439 592 0.742 LZ1-Btg-80 80 380 445 0.854

Depth FeO FeD Location Field Site Sample FeO/FeD (cm) (mg/kg) (mg/kg) PitF-O-0 0 110 166 0.665 PitF-Bw-20 20 90 208 0.431 PitF-Be-30 30 197 609 0.324 PitF-2Btb1-40 40 394 1121 0.351 PitF PitF-2Btb1-60 60 286 957 0.299 PitF-2Btb2-80 80 413 1012 0.408 PitF-2Btb2-110 110 515 1661 0.310 PitF-BC-120 120 566 1821 0.311 South Anna PitF-BC-150 150 487 1260 0.386 River, SiteD-A-0 0 144 176 0.819 Virginia SiteD-Bw-38 38 67 419 0.161 SiteD-Bt1-48 48 276 611 0.452 SiteD SiteD-Bt2-64 64 431 650 0.663 SiteD-Bg-83 83 461 956 0.483 SiteD-BC-100 100 508 786 0.646 SiteD-BC-121 121 606 995 0.609 LZ3 Site E - B N/A 506 786 0.645 Site E LZ3 58 42 1.377

Appendix 4. Major Elements (% By Mass)

90Zr

PPM

454.85000 492.10000 257.80000 211.35000 365.70000 211.00000 428.33973 718.25322 647.42672 601.50620 393.40978 344.46070 275.34163 360.98672 159.34641 111.43004 155.70685 184.07052 214.61175 169.83441 203.03667 238.02628 122.43803 150.00000

56Fe

2.18500 5.76000 1.88050 3.56750 2.07000 3.93237 3.37082 3.51423 4.11947 5.41729 5.21033 6.04521 4.84852 4.14590 4.88179 4.61809 7.12314 8.03532 7.12936 7.55426 7.29812 5.21094 2.02000

2.48300

55Mn

0.02444 0.19705 0.02531 0.05570 0.05000 0.07188 0.05588 0.04697 0.03640 0.02554 0.02449 0.02846 0.02081 0.01488 0.18230 0.18948 0.16895 0.15889 0.23949 0.19127 0.09628 0.16848

0.03242

47Ti

0.29800 0.33495 0.35990 0.17675 0.29150 0.14000 0.38668 0.43649 0.43564 0.41085 0.42916 0.41817 0.37999 0.38561 0.29266 0.22881 0.27458 0.33572 0.42362 0.41101 0.38416 0.28183 0.23682 0.32900

BDL

44Ca

0.02432 0.05315 0.00166 0.21000 0.21036 0.24039 0.25851 0.24866 0.24442 0.19058 0.19987 0.16615 0.13378 0.14948 0.11760 0.09178 0.15309 0.11232 0.10588 0.12043 0.16979 1.38600

0.01814

BDL BDL BDL BDL BDL BDL BDL

43Ca

0.01255 0.08478 0.07496 0.19776 0.20914 0.26734 0.16460 0.10827 0.03705 0.07113 0.00122 0.09756 0.07944 0.04419 0.03225 1.38600

0.21000

39K

0.70750 0.76200 1.65950 0.18990 0.85400 6.03000 1.74496 1.70412 1.77437 1.43621 1.10558 0.97849 1.09419 1.03573 1.05731 1.08815 1.27071 1.48063 1.70246 2.03586 2.21960 1.17491 1.51739 1.22000

29Si

32.10500 23.53500 23.01500 28.62500 23.22000 24.29223 27.87397 27.94185 25.43641 23.78876 23.13635 21.95555 23.55762 20.47852 25.24483 24.40389 21.72782 23.04603 23.03213 21.61767 25.36071 22.99114 22.24000

31.46000

27Al

3.94100 3.64850 7.36000 2.44500 5.03000 5.73524 4.94369 5.56262 6.05009 8.08221 7.83100 8.97809 8.22508 7.24537 4.70898 5.07938 6.26075 5.08769 5.30604 5.79380 3.44605 3.24095 3.82900

14.36000

24Mg

0.38630 0.36135 0.71350 0.12725 0.37040 0.66000 0.24668 0.17395 0.19752 0.19196 0.25739 0.25519 0.32805 0.28481 0.30436 0.27337 0.29548 0.32849 0.33151 0.53419 0.56822 0.26670 0.29369 0.59700

BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL

23Na

0.02688 1.04600

0.07000

65 10 30 40 70

160 210 240 300 350 100 130 160 180 210 230 250 290 320 350 380 400 (cm)

Depth

Parent Parent

70cm50k -

Sample

EK_0_BM_FLUX

MHN MHN 160cm 50k MHN 210cm 50k MHN 240cm 50k MHN 300cm 50k

EK_10_BM_FLUX EK_30_BM_FLUX EK_40_BM_FLUX EK_70_BM_FLUX

EK_100_BM_FLUX EK_130_BM_FLUX EK_160_BM_FLUX EK_180_BM_FLUX EK_210_BM_FLUX EK_230_BM_FLUX EK_250_BM_FLUX EK_290_BM_FLUX EK_320_BM_FLUX EK_350_BM_FLUX EK_380_BM_FLUX EK_400_BM_FLUX

MHN MHN 60

EK EK EK EK EK EK EK EK EK EK EK EK EK EK EK EK EK

MHN MHN MHN MHN MHN MHN

Byram Byram

Granite

Field Site Field

Hornblende

Pennsylvania

Location

90Zr

PPM

63.72840 96.25000 92.20000 55.85000 58.35000 35.85000 70.15000 79.45000 89.90000 92.40000 42.00000 65.60000

202.25436 286.31757 269.30691 263.49301 254.63696 216.25055 550.22691 281.05051 283.69714 251.13367 257.09391 214.30963 163.65176 110.21483 154.97967 125.71485 134.51553 106.50853 211.00000 250.90000 104.00000 104.45000

56Fe

2.96133 3.45754 3.67355 4.00737 3.81328 7.70691 2.93214 2.92124 3.07922 3.64299 3.68616 3.96574 2.83698 3.88877 3.08789 2.84054 3.02349 2.58547 2.07000 1.72050 2.70350 3.86900 2.64500 1.64750 2.02400 0.92250 1.63150 2.73300 2.13000 5.71000 2.33900 1.10500 3.21600

2.73078

55Mn

0.00001 0.00010 0.00104

0.23792 0.09942 0.05814 0.02452 0.06716 0.07607 0.11925 0.08894 0.10750 0.06149 0.04250 0.05480 0.04633 0.01994 0.01644 0.01290 0.01814 0.02211 0.01635 0.05000 0.00662 0.00255 0.00290 0.00120 0.00013 0.00096 0.00088 0.00240 0.00156 0.00112 0.00259

- - -

47Ti

0.40331 0.43278 0.39739 0.36242 0.36356 0.36182 0.74874 0.35746 0.38820 0.38326 0.40539 0.32404 0.31070 0.16818 0.25379 0.18601 0.17839 0.18056 0.17032 0.14000 0.40630 0.34505 0.26165 0.21645 0.16595 0.14605 0.10855 0.17820 0.21110 0.18815 0.30525 0.19045 0.13685 0.18355

44Ca

0.01990 0.01451 0.01451 0.23570 0.32200 0.07115 0.78600 0.24920 0.38135 0.47575 0.28155 0.74800 0.62700 0.79400

0.33796 0.37181 0.44531 0.27320 0.26861 0.26481 0.66420 0.24149 0.23128 0.29147 0.33375 0.22921 0.18647 0.23948 0.21637 0.22305 0.27179 0.26683 0.22863 0.21000

------

43Ca

0.10653 0.19147 0.32933 0.19517 0.15059 0.14567 0.38432 0.22231 0.20014 0.18013 0.26371 0.21192 0.18866 0.18475 0.16315 0.16301 0.15955 0.23487 0.22537 0.21000 2.49150 3.27150 2.96900 3.38600 3.16950 3.77450 3.32100 3.23350 3.19500 3.49500 4.35250 3.86950 3.19700 3.65650

39K

0.24270 0.37880 0.55650 0.51950 0.77800 1.13000 0.50500 0.66100 0.85900 0.71200 0.85150 1.10600 1.03650

0.98803 1.24240 2.35037 1.69697 1.83028 1.85428 2.35271 1.06973 1.10261 1.42050 1.47252 1.71553 1.78830 1.26335 1.75115 1.42740 1.22387 1.20248 1.13852 6.03000 0.27320

------

29Si

26.10669 29.16574 25.32078 27.27963 25.85447 25.49761 62.68290 27.76719 26.61773 28.33678 30.16675 25.86907 23.41957 25.58035 23.25680 25.64461 28.08839 24.84254 26.44762 23.22000 39.63500 32.29500 35.19000 34.58500 29.54500 25.81500 17.82500 25.53000 29.70500 20.53500 21.31500 21.96500 17.20000 24.65500

27Al

4.17762 4.58566 5.79944 6.18189 6.84340 7.04370 4.72125 4.56467 5.30792 6.24328 6.75988 7.19192 6.06409 7.50775 6.70885 6.34906 6.20463 5.67900 3.15950 4.30100 3.22000 2.94600 2.45750 2.12200 1.66250 2.71050 4.16200 3.23600 6.39500 3.45400 1.97600 0.97350

13.10642 14.36000

24Mg

0.36418 0.49568 0.46491 0.58127 0.56853 0.95379 0.33578 0.32797 0.40201 0.45016 0.46502 0.50506 0.48730 0.65527 0.54117 0.47917 0.54020 0.49234 0.66000 0.16625 0.10195 0.06745 0.04782 0.03884 0.02725 0.02511 0.02967 0.05115 0.01848 0.05040 0.04293 0.00815 0.00581

0.30759

BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL

23Na

0.05835 1.66451 0.43444 0.07000 0.30585 0.15465 0.03421 0.14580 0.20940 0.04730 0.15600 0.39915 0.24525 0.01616 0.48165 0.25480 0.12330 0.38555

15 30 40 60 50

100 120 145 185 195 220 225 235 245 250 265 305 310 340 350 400 100 150 200 250 280 300 350 400 450 500 550 825

Depth (cm) Depth

flux flux flux flux flux flux flux flux flux

flux flux flux flux flux

------

- - - - -

Parent

BM1 BM2 BM3 BM4 BM5 BM6 BM7 BM8 BM9

Sample

PS_15_FX PS_30_FX PS_40_FX PS_60_FX

BM16 BM10 BM11 BM12 BM17

PS_100_FX PS_120_FX PS_145_FX PS_185_FX PS_195_FX PS_220_FX PS_225_FX PS_235_FX PS_245_FX PS_250_FX PS_265_FX PS_305_FX PS_310_FX PS_340_FX PS_350_FX

Sand

PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS

BM BM BM BM BM BM BM BM BM BM BM BM

Parent

Field Site Field

Gravel Bald Eagle SS Bald

Maryland County, Cecil

Location

90Zr

PPM

3.71258 4.56775 3.39834 2.56783 3.00156 4.23212 5.06459 4.25127 4.07660 2.22938 2.81532 1.55022 1.39313 1.62779 1.69685 4.23212 4.53584 3.94838 5.37606 5.29940 4.68377 4.49722 4.23212 3.74627 2.81259

56Fe

0.01867 0.02007 0.03909 0.03877 0.02835 0.05210 0.00902 0.00988 0.02718 0.04893 0.04049 0.04711 0.05184 0.05533 0.05457 0.05210 0.01314 0.01948 0.02724 0.03432 0.04367 0.04292 0.05210 0.01462 0.03771

55Mn

0.00050 0.00069 0.00039 0.00033 0.00045 0.00046 0.00029 0.00027 0.00023 0.00018 0.00015 0.00014 0.00019 0.00022 0.00023 0.00046 0.00092 0.00044 0.00047 0.00047 0.00046 0.00046 0.00046 0.00043 0.00026

47Ti

0.00670 0.00820 0.00729 0.00610 0.00623 0.00708 0.00575 0.00595 0.00616 0.00514 0.00416 0.00422 0.00469 0.00526 0.00517 0.00708 0.00674 0.00718 0.00791 0.00804 0.00769 0.00808 0.00708 0.00605 0.00672

BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL

44Ca

BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL

43Ca

0.01621 0.01662 0.02410 0.01763 0.02664 0.02303 0.02303 0.02005 0.02325 0.01818 0.02668 0.02648 0.03524 0.02303 0.00162

39K

BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL

29Si

0.26287 0.27714 0.24768 0.21671 0.22723 0.23095 0.23290 0.22475 0.23732 0.17978 0.20114 0.19276 0.13252 0.16548 0.19317 0.23095 0.29785 0.22077 0.28046 0.25689 0.24859 0.25211 0.23095 0.27242 0.18286

27Al

0.04019 0.03296 0.06699 0.08059 0.06419 0.06633 0.01778 0.01790 0.04579 0.07580 0.07138 0.07560 0.08135 0.09403 0.09866 0.06633 0.02882 0.03428 0.04314 0.05245 0.05600 0.06472 0.06633 0.03934 0.07506

24Mg

0.00309 0.00270 0.00326 0.00376 0.00391 0.00336 0.00296 0.00250 0.00312 0.00384 0.00343 0.00351 0.00324 0.00329 0.00382 0.00336 0.00191 0.00206 0.00247 0.00271 0.00272 0.00311 0.00336 0.00742 0.00642

23Na

0.01641 0.01508 0.01883 0.02009 0.02475 0.01846 0.01976 0.01713 0.01636 0.01983 0.01572 0.01801 0.01595 0.02024 0.02525 0.01846 0.01233 0.01627 0.01526 0.01365 0.01598 0.02017 0.01846 0.03252 0.02388

0 0 0

10 30 50 80 20 30 40 60 80 38 48 64 83

121 110 120 150 121 100 121

N/A

Depth (cm) Depth

0 0 0

- -

20 30 48 64

10 30 50 80 40 60 80 38 83

------

LZ3

A A

121 110 120 150 121 100 121

- -

------

Bw Bw

Btg Btg

Bt1 Bt2

- -

- BC BC BC BC BC BC

- -

LZ1

------

PitF

LZ1

Site E Site

2Btb1 2Btb1 2Btb2

SiteD

- - -

PitF

2Btb2

LZ1

LZ1 LZ1

PitF

Sample

PitF PitF

SiteD

SiteD SiteD

SiteD SiteD SiteD SiteD

PitF PitF PitF

PitF

LZ1 LZ1 LZ1 LZ1 LZ1 LZ3

PitF PitF PitF PitF PitF PitF PitF PitF PitF

SiteD SiteD SiteD SiteD SiteD SiteD SiteD

Site E Site

Field Site Field

SiteD (Parent) SiteD (Parent) SiteD

Virginia River, Anna South

Location

90Zr

PPM

3.61918 3.55754 2.86821 2.48899 2.76911 4.23212 4.26421 3.48585 3.76291 3.21484 3.24456 4.23212 3.54809 3.64022 2.81933 3.21225 4.23212 3.97671 4.12447 2.98448 1.84393 1.33958 4.23212 4.32401 2.99790 3.74345 3.59590 3.13839 2.81285 4.23212 4.57216 3.82964 3.34337 3.19366 4.23212 1.06735 1.07862 0.93240

56Fe

0.01366 0.01608 0.03756 0.03885 0.03257 0.05210 0.02116 0.02775 0.03567 0.04225 0.04452 0.05210 0.00719 0.01125 0.01650 0.02474 0.05210 0.01564 0.01895 0.03992 0.05850 0.06360 0.05210 0.00718 0.00608 0.01374 0.02189 0.03822 0.04263 0.05210 0.00622 0.01064 0.02509 0.02923 0.05210 0.07002 0.07980 0.08466

55Mn

0.00042 0.00042 0.00025 0.00025 0.00028 0.00046 0.00049 0.00040 0.00046 0.00035 0.00033 0.00046 0.00045 0.00032 0.00019 0.00017 0.00046 0.00035 0.00036 0.00023 0.00015 0.00015 0.00046 0.00046 0.00034 0.00036 0.00036 0.00035 0.00031 0.00046 0.00043 0.00029 0.00024 0.00022 0.00046 0.00033 0.00030 0.00040

47Ti

0.00930 0.00972 0.00830 0.00765 0.00837 0.00708 0.01034 0.00854 0.00894 0.00843 0.00852 0.00708 0.00571 0.00659 0.00511 0.00581 0.00708 0.00543 0.00596 0.00592 0.00547 0.00505 0.00708 0.00548 0.00448 0.00603 0.00653 0.00653 0.00599 0.00708 0.00520 0.00570 0.00507 0.00534 0.00708 0.00816 0.00537 0.00527

BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL

44Ca

BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL

43Ca

0.02303 0.02110 0.01347 0.01122 0.00668 0.00519 0.02303 0.00030 0.00015 0.02303 0.02303 0.03516 0.02521 0.02191 0.02249 0.02368 0.02550 0.02303 0.00603 0.00626 0.00592 0.00139 0.02303

39K

BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL BDL

29Si

0.29695 0.29798 0.24615 0.21937 0.23967 0.23095 0.30272 0.26118 0.26149 0.23863 0.23697 0.23095 0.26190 0.29517 0.24841 0.26643 0.23095 0.27769 0.27762 0.21237 0.15895 0.13600 0.23095 0.36530 0.26923 0.30748 0.29436 0.25839 0.24836 0.23095 0.31985 0.28008 0.25098 0.24275 0.23095 0.12444 0.14080 0.14977

27Al

0.01276 0.01859 0.04615 0.04746 0.04307 0.06633 0.02054 0.02956 0.03957 0.04553 0.04514 0.06633 0.01646 0.02419 0.03508 0.05043 0.06633 0.02515 0.03132 0.06670 0.09585 0.10415 0.06633 0.01172 0.01040 0.02357 0.03585 0.05190 0.05821 0.06633 0.01125 0.02049 0.04563 0.05214 0.06633 0.11945 0.12050 0.08799

24Mg

0.00239 0.00281 0.00315 0.00384 0.00373 0.00336 0.00161 0.00179 0.00201 0.00184 0.00194 0.00336 0.00208 0.00254 0.00270 0.00308 0.00336 0.00308 0.00314 0.00330 0.00351 0.00394 0.00336 0.00076 0.00071 0.00123 0.00145 0.00173 0.00192 0.00336 0.00155 0.00179 0.00219 0.00236 0.00336 0.00319 0.00282 0.00289

23Na

0.01279 0.01869 0.01321 0.01786 0.01906 0.01846 0.01612 0.01543 0.01921 0.01457 0.01795 0.01846 0.01552 0.01552 0.01247 0.01684 0.01846 0.01685 0.01593 0.01696 0.01673 0.01906 0.01846 0.00859 0.01318 0.01101 0.01060 0.00953 0.00996 0.01846 0.02027 0.01603 0.01352 0.01762 0.01846 0.02582 0.02140 0.02361

0 0 0 0

10 30 50 60 10 30 50 60 20 30 60 95 10 30 48 55 10 20 35 50 65 10 40 70 77

121 121 121 121 121 121

Depth (cm) Depth

0 0 0 0

- - - -

30 50 60 20 30 60 95 55 10 20 65 10 40

10 30 50 60 10 10 30 48 35 50 70 77

------

LF2 LF3 LF4

121 121 121 121 121 121

------

B B

Ap Ap Ap

- -

Bt Bt Bt

- - -

Bw Bw

Bt1 Bt2 Bt2 Bt1 Bt2 Bt2 Bt2 Bt2 Bt2 Bt2

- -

BC BC BC BC BC BC

------

- - - 2Btb - 2Btb - -

AH2

- -

AF1 AF1

AF2

BB1

AH1

BB1

AF1

AH1 AH1 AH1

AF1

BB2

BB1 BB2

Sample

AF2 AF2 AF2

AH1 AH2

BB1 BB1 BB2 BB2

AH2 AH2 AH2

AF2

BB1

SiteD SiteD SiteD SiteD SiteD SiteD

LF LF LF

Site

AF1 AF1 AF1 AF1 AF2 AF2 AF2 AF2 AF2

BB1 BB1 BB1 BB1 BB1 BB1 BB2 BB2 BB2 BB2

AH1 AH1 AH1 AH1 AH1 AH2 AH2 AH2 AH2 AH2

Field Field

SiteD SiteD SiteD SiteD SiteD SiteD SiteD

(Parent) (Parent) (Parent) (Parent) (Parent) (Parent)

jjj Virginia River, Anna South

Location

Appendix 5. Sample Demagnetization

Sample Name Demag Type Demag Level Moment Core Declination Core Inclination AF1-30 NRM 0 2.28E-06 288.5 -45.7 AF1-30 IRM 0 0.001197 114.5 -87.2 AF1-30 mT 100 0.0005099 243.6 -84.7 AF1-30 C 120 0.000467 205.5 -84.3 AF1-60 NRM 0 1.22E-06 86.9 16.4 AF1-60 IRM 0 0.0007357 110.2 -85.1 AF1-60 mT 100 0.0003102 29.2 -85 AF1-60 C 120 0.0002749 91.9 -85 AF1-95 NRM 0 5.99E-07 129.7 58.5 AF1-95 IRM 0 0.000363 324.8 -82.2 AF1-95 mT 100 0.0001619 282.4 -81.7 AF1-95 C 120 0.0001177 274.8 -82 AH2-10 NRM 0 0.001014 97 13.4 AH2-10 IRM 0 0.01198 33 -70.2 AH2-10 mT 100 0.002273 286.5 -85.6 AH2-10 C 120 0.002097 220.5 -84.6 AH2-30 NRM 0 0.0008324 336.2 -45.7 AH2-30 IRM 0 0.01656 176.3 -69.3 AH2-30 mT 100 0.002036 182.5 -85.5 AH2-30 C 120 0.001954 174.8 -85.6 AH2-30 C 120 0.001956 175.6 -85.6 AH2-50 NRM 0 5.90E-06 83.1 -20.2 AH2-50 IRM 0 0.003429 145.3 -84.9 AH2-50 mT 100 0.001611 254 -84.7 AH2-50 C 120 0.001538 193.5 -83.1 AH2-60 NRM 0 0.0008325 116.2 -8.5 AH2-60 IRM 0 0.008761 142.6 -74.7 AH2-60 mT 100 0.001728 294.8 -84.1 AH2-60 C 120 0.001576 130.5 -82.3 BB1-10 NRM 0 4.10E-06 26.1 -81.7 BB1-10 IRM 0 0.002157 267.6 -88.3 BB1-10 mT 100 0.0009194 235 -88.3 BB1-10 C 120 0.0008515 293.2 -88.6 BB1-20 NRM 0 1.44E-06 132.7 -0.3 BB1-20 IRM 0 0.001616 200.3 -87.6 BB1-20 mT 100 0.0004956 62.7 -89.6 BB1-20 C 120 0.0004204 86.4 -89.6 BB1-35 NRM 0 5.58E-06 220 -44 BB1-35 IRM 0 0.001358 347.2 -84.6 BB1-35 mT 100 0.0004712 267.6 -82.2 BB1-35 NRM 0 1.63E-06 29.8 -60.6 BB1-35 IRM 0 0.001126 236.3 -89.1 BB1-35 mT 100 0.0004405 234.8 -88.1 BB1-35 C 120 0.0004041 153.4 -88.2 BB1-50 NRM 0 3.03E-06 195.2 -45.1 BB1-50 IRM 0 0.001168 304.7 -88.7 BB1-50 mT 100 0.000572 290.5 -84.9 BB1-50 C 120 0.0002837 183.1 -84.6 BB1-50 C 120 0.0005041 254.2 -86.5 BB1-50 NRM 0 2.24E-06 181.4 -21 BB1-50 IRM 0 0.001328 357.9 -86.3 BB1-50 mT 100 0.0006621 1.6 -85.4 BB1-50 C 125 0.0005568 303.9 -86.8 BB1-50 C 125 0.0005581 305.1 -86.9 BB1-65 NRM 0 1.76E-06 194.5 -58.9 BB1-65 IRM 0 0.001677 337.2 -88 BB1-65 mT 100 0.001148 214.7 -86.1 BB1-65 C 120 0.0009741 161.5 -86.7 BB2-40 NRM 0 3.45E-06 200.7 35.3 BB2-40 IRM 0 0.002128 340.3 -85.8 70

Sample Name Demag Type Demag Level Moment Core Declination Core Inclination BB2-40 mT 100 0.001087 251.3 -85.1 BB2-40 C 120 0.001019 209.3 -82.9 BB2-70 NRM 0 5.18E-06 32.8 17.6 BB2-70 IRM 0 0.001629 355.7 -85.8 BB2-70 mT 100 0.0008557 359.8 -86.3 BB2-70 C 120 0.0007987 44.8 -87.3 BB2-77 NRM 0 0.000011 200 9.3 BB2-77 IRM 0 0.001648 331.7 -83.4 BB2-77 mT 100 0.0007595 108.5 -85.6 BB2-77 C 120 0.0007128 216.3 -84.3 BM-16 NRM 0 2.07E-06 312.1 -85.3 BM-16 IRM 0 0.001605 307.2 -87.4 BM-16 mT 100 0.001466 310.5 -87.6 BM-16 C 125 0.0001361 305.7 -86.5 BM6 NRM 0 1.55E-06 126.9 -68.1 BM6 IRM 0 0.003422 173.9 -85.4 BM6 mT 100 0.003283 173.1 -85 BM6 C 125 0.0000995 295.5 -86.5 BM6 C 125 0.000099 309.9 -86.5 BM6 C 125 0.0000986 316.2 -86.3 BM6 NRM 0 9.67E-07 129.3 -57.5 BM6 IRM 0 0.004462 309.3 -88.8 BM6 mT 100 0.004287 302.8 -88.9 BM6 C 125 0.0001089 215.2 -87.6 BM6 C 125 0.0001079 217.8 -87.7 BM6-2 NRM 0 1.82E-06 154.5 -37.1 BM6-2 IRM 0 0.003227 326.2 -88.4 BM6-2 mT 100 0.0028 334.1 -88.5 BM6-2 C 125 0.0000846 300.2 -89.3 EK-130 NRM 0 0.000048 254.9 8.4 EK-130 IRM 0 0.02792 183.8 -75.2 EK-130 mT 100 0.004404 172.1 -81.4 EK-130 C 125 0.004094 16.8 -89.1 EK-130 C 125 0.004089 52.5 -89 EK-130 C 125 0.004116 22.7 -88.9 EK-130 NRM 0 0.0001407 37.1 -27.2 EK-130 IRM 0 0.01505 344 -88.1 EK-130 mT 100 0.002719 104.9 -89.2 EK-130 C 125 0.002218 163.4 -86 EK-130 C 125 0.002214 166.8 -86 EK-160 NRM 0 0.0000337 47.7 -6.6 EK-160 IRM 0 0.01098 330 -87.4 EK-160 mT 100 0.002453 313.7 -86.2 EK-160 C 125 0.002001 133.5 -85.9 EK-160 C 125 0.002002 133.7 -86 EK-210 NRM 0 0.0003872 52.9 -64.1 EK-210 IRM 0 0.01904 287.8 -85 EK-210 mT 100 0.00306 235.7 -89.4 EK-210 C 120 0.002571 160.5 -86.9 EK-230 NRM 0 0.001707 103.7 4.4 EK-230 IRM 0 0.02752 22.4 -84 EK-230 mT 100 0.004332 285.9 -84.1 EK-230 C 120 0.003874 309.5 -87.6 EK-250 NRM 0 0.0001095 254.1 32 EK-250 IRM 0 0.03982 81.6 -83.5 EK-250 mT 100 0.005587 41.5 -87.5 EK-250 C 120 0.005016 276.1 -84.4 EK-290 NRM 0 0.0005658 96.2 -12 EK-290 IRM 0 0.09178 277.2 -85 EK-290 mT 100 0.01199 255.8 -84.5 EK-290 C 120 0.01099 178.6 -81.6 EK-320 NRM 0 0.0115 246.1 26 71

Sample Name Demag Type Demag Level Moment Core Declination Core Inclination EK-320 IRM 0 0.08147 242 -72.5 EK-320 mT 100 0.007834 254.6 -85.7 EK-320 C 120 0.006934 150.1 -81.4 EK-350 NRM 0 0.0007249 19.8 31.3 EK-350 IRM 0 0.1393 156 -65.5 EK-350 mT 100 0.007605 212.5 -89.3 EK-350 C 120 0.006777 124.1 -86.1 EK-350 C 120 0.006794 127.5 -86.2 LF2 NRM 0 3.90E-06 155.8 -18 LF2 NRM-T 0 2.76E-06 291.3 -11.6 LF2 IRM 0 0.001767 141.8 -82.2 LF2 mT 100 0.001475 134.7 -83.1 LF2 C 120 0.0009337 210.1 -82.1 LF2 C 120 0.0009289 10.4 -81.5 LF2 C 120 0.0009313 13.7 -80.5 LF3 NRM 0 7.78E-06 144.7 -55.6 LF3 IRM 0 0.001576 101.8 -84.7 LF3 mT 100 0.00117 186.2 -87.4 LF3 C 120 0.0007913 348.8 -86 LF3 C 120 0.0007892 75.1 -84.5 LF3 C 120 0.0007894 74.6 -84.5 LF3 C 120 0.0007894 75.2 -84.5 LF4 NRM 0 0.0000288 104.8 -42.2 LF4 IRM 0 0.03886 251.8 -87.8 LF4 mT 100 0.007696 167.2 -88.1 LF4 C 120 0.007282 112.5 -88.5 LF4 C 120 0.007268 20.4 -87 LF4 C 120 0.007273 24.8 -86.8 LZ1-30 NRM 0 5.43E-07 168.1 -71.3 LZ1-30 NRM 0 9.36E-07 171 56.6 LZ1-30 IRM 0 0.0006827 254.1 -84.2 LZ1-30 mT 100 0.0002558 200.9 -83.3 LZ1-30 C 120 0.0002205 101.3 -82.4 LZ1-30 C 120 0.0002093 107.2 -83.7 LZ1-30 C 120 0.0002029 108.7 -83.1 LZ1-50 NRM 0 6.52E-07 82 -30.2 LZ1-50 IRM 0 0.0004323 143.9 -83.3 LZ1-50 mT 100 0.0001959 36.4 -82.3 LZ1-50 C 120 0.0001848 321.9 -84 LZ1-80 NRM 0 1.45E-06 305 -49.8 LZ1-80 IRM 0 0.0005794 238.6 -86 LZ1-80 mT 100 0.000257 222.4 -85.6 LZ1-80 C 120 0.0002249 118 -87 LZ1-80 C 120 0.0002245 117.1 -87 MHN-210 NRM 0 2.11E-06 211.4 -17.2 MHN-210 IRM 0 0.0005217 106.3 -89 MHN-210 mT 100 0.0001491 99 -88.3 MHN-210 C 125 0.000123 65.2 -87.8 MHN-210 C 125 0.0001231 61.3 -88.2 MHN-210 C 125 0.0001232 62.3 -88.3 MHN-240 NRM 0 0.0000169 215.6 -50 MHN-240 IRM 0 0.003043 221.2 -86.8 MHN-240 mT 100 0.000621 221 -86.7 MHN-240 C 125 0.0004646 222.1 -88 PitF_110 NRM 0 0.0000265 176.1 -58.3 PitF_110 IRM 0 0.003329 251.6 -85.7 PitF_110 mT 100 0.002329 301.5 -85.9 PitF_110 C 120 0.001689 242.3 -89.4 PitF_120 NRM 0 0.0000208 105.2 -43.8 PitF_120 IRM 0 0.0035 127.4 -84.9 PitF_120 mT 100 0.002748 71.4 -88.1 PitF_120 C 120 0.002074 350.8 -88.2 72

Sample Name Demag Type Demag Level Moment Core Declination Core Inclination PitF_120 C 120 0.002057 128.6 -85.4 PitF_150 NRM 0 0.0000199 165 -35.4 PitF_150 IRM 0 0.00421 26 -82.4 PitF_150 mT 100 0.002773 112 -87.8 PitF_150 C 120 0.002359 11.8 -87.3 PitF_80 NRM 0 0.000012 55.9 -42.2 PitF_80 IRM 0 0.001516 80.7 -87.2 PitF_80 mT 100 0.0009627 290.9 -87 PitF_80 C 120 0.0005945 176.9 -87.4 PitF-110-2 NRM 0 0.0000131 165 -23.4 PitF-110-2 IRM 0 0.003714 207.2 -84.2 PitF-110-2 mT 100 0.002688 201.9 -83.3 PitF-110-2 C 125 0.002119 260.9 -87.1 PitF-60 NRM 0 8.56E-06 222.5 -63 PitF-60 IRM 0 0.002097 187.9 -87.7 PitF-60 mT 100 0.001303 185 -87.6 PitF-60 C 125 0.0009982 238.3 -86.7 PitF-80-2 NRM 0 8.89E-06 218.5 -74.2 PitF-80-2 IRM 0 0.001915 299.9 -87.5 PitF-80-2 mT 100 0.001153 321.8 -80.1 PitF-80-2 C 125 0.0007771 177.9 -89.2 PS-265 NRM 0 3.86E-06 181.6 -49.5 PS-265 IRM 0 0.0005532 68.6 -88.7 PS-265 mT 100 0.0002108 94 -87.7 PS-265 C 125 0.0001134 12.6 -88.7 PS-305 NRM 0 4.03E-06 234.7 -61 PS-305 IRM 0 0.0006923 175.7 -84.1 PS-305 mT 100 0.0001387 174.8 -84 PS-305 C 125 0.0000921 329.9 -87.7 PS-310 NRM 0 4.92E-06 159.1 -56.1 PS-310 IRM 0 0.0006942 213 -84.3 PS-310 mT 100 0.0001435 208.4 -84.7 PS-310 C 125 0.0000822 338.6 -85.6 PS-340 NRM 0 2.94E-06 195.1 -62.1 PS-340 IRM 0 0.0004232 167.4 -87.6 PS-340 mT 100 0.0001174 153.6 -87.7 PS-340 C 125 0.0000716 287.3 -83.1 PS-340 C 125 0.0000716 287.4 -83.1 SiteD-100 NRM 0 8.39E-06 198.7 -23.8 SiteD-100 IRM 0 0.0007511 318.7 -87.5 SiteD-100 mT 100 0.0003342 234.8 -83.5 SiteD-100 C 120 0.0002843 177.4 -83.3 SiteD-121 NRM 0 2.13E-06 332 -77.3 SiteD-121 IRM 0 0.0009512 195.1 -87 SiteD-121 mT 100 0.0005022 204.7 -87.5 SiteD-121 C 120 0.0004032 93.3 -86.8 SiteD-38 NRM 0 2.00E-06 261.4 -41.7 SiteD-38 IRM 0 0.001232 189.3 -87.2 SiteD-38 mT 100 0.0005438 323.7 -88.2 SiteD-38 C 120 0.0005033 310 -89.2 SiteD-38 C 120 0.0005024 313 -89.2 SiteD-48 NRM 0 0.0000133 329.2 3.7 SiteD-48 IRM 0 0.003429 128.5 -81.4 SiteD-48 mT 100 0.0005191 242.5 -86.5 SiteD-48 C 120 0.000475 157.5 -83.2 SiteD-64 NRM 0 1.52E-06 260.3 -50 SiteD-64 IRM 0 0.001098 255.4 -83.9 SiteD-64 mT 100 0.0004495 222.5 -82.7 SiteD-64 C 120 0.0004199 183.2 -83.5 SiteD-64 C 120 0.0004204 184.7 -83.6 SiteD-83 NRM 0 3.80E-07 121.1 -53.9 SiteD-83 IRM 0 0.0009344 96.7 -88.8 73

Sample Name Demag Type Demag Level Moment Core Declination Core Inclination SiteD-83 mT 100 0.0004612 306.7 -88.8 SiteD-83 C 120 0.000419 301.5 -87.5 SiteD-83 C 120 0.0004192 302.7 -87.4

Appendix 6. G/H, CIW, & Predicted MAP

Site Average CIW Predicted G/H Predicted State Depth CIW Magnetic G/H Name G/H MAP MAP MHN 65 100.0000 1585.44 MHN 160 100.0000 1585.44 MHN 210 100.0000 0.2122 1585.44 262.24 MHN 240 100.0000 0.3366 1585.44 387.11 MHN 300 100.0000 1585.44 EK 0 100.0000 1585.44 EK 10 99.8291 1580.11 EK 30 98.9825 1553.97 EK 40 99.1712 1559.76 EK 70 98.3764 1535.53 EK 100 98.2305 1531.12 EK 130 98.0311 0.2259 1525.12 275.97 EK 160 98.6681 0.2259 1544.38 275.98 EK 180 99.0022 1554.58 EK 210 99.4721 0.1902 1569.03 240.16 EK 230 99.0643 0.1182 1556.48 167.94 EK 250 99.9868 0.1138 1585.03 163.53 EK 290 98.7231 0.0910 1546.05 140.61 EK 320 99.0003 0.1298 1554.52 179.55 EK 350 99.4881 0.1222 1569.53 171.91 Pennsylvania EK 380 100.0000 1585.44 EK 400 98.3820 1535.70 PS 15 96.7499 1487.11 PS 30 68.7689 856.94 PS 40 88.7930 1271.35 PS 60 97.9150 1521.63 PS 100 98.5374 1540.41 PS 120 98.6243 1543.05 PS 145 98.0605 1526.00 PS 185 96.9217 1492.15 PS 195 97.1275 1498.21 PS 220 97.7622 1517.06 PS 225 97.2299 1501.24 PS 235 97.9292 1522.06 PS 245 98.2614 1532.05 PS 250 97.9864 1523.78 PS 265 98.5554 0.8589 1540.95 911.21 PS 305 98.3876 0.5060 1535.87 557.04 PS 310 98.3332 0.7457 1534.22 797.65 PS 340 97.5103 0.6397 1509.55 691.20 PS 350 97.3931 1506.07 BM1 0 60.7719 732.03 BM2 50 64.2985 784.69 BM3 100 61.1849 738.01 BM4 150 54.5458 647.53 BM5 200 50.7635 601.04 BM6 250 44.9252 38.3664 32.0969 35.2317 535.74 35404.27 Cecil County, BM16 280 40.6918 9.7715 492.87 Maryland BM7 300 50.5697 598.75 BM8 350 63.0144 765.09 BM9 400 57.6587 688.48 BM10 450 64.6220 789.71 BM11 500 54.2826 644.18 BM12 550 46.1998 549.36 BM17 825 25.0111 361.89 AH2 0 0.0900 221.49 75

Site Average CIW Predicted G/H Predicted State Depth CIW Magnetic G/H Name G/H MAP MAP AH2 10 0.2027 0.0839 221.98 133.52 AH2 30 0.3252 0.0420 222.52 91.41 AH2 50 0.6262 0.0475 223.84 96.93 AH2 60 0.7967 0.0964 224.60 146.08 AF1 20 47.4759 563.34 AF1 30 6.5334 0.0919 251.47 141.48 AF1 60 70.5602 0.1284 887.71 178.16 AF1 95 21.6388 0.3755 338.63 426.15 BB1 0 0.0309 221.23 BB1 10 0.0382 0.0797 221.27 129.32 BB1 20 0.0995 0.1789 221.53 228.80 BB1 35 0.1474 0.0901 221.74 139.69 BB1 50 0.2026 0.1891 221.98 239.08 BB1 65 0.2110 0.1785 222.02 228.45 BB2 10 0.1719 221.85 BB2 40 0.3013 0.0667 222.42 116.27 BB2 70 0.7076 0.0714 224.20 120.92 BB2 77 3.3159 0.0655 236.03 115.05 LZ1 0 0.2290 222.10 South Anna River, LZ1 10 0.1833 221.90 Virginia LZ1 30 0.2567 0.1601 222.22 209.95 LZ1 50 0.4212 0.0601 222.94 109.58 LZ1 80 0.2225 0.1427 222.07 192.53 PitF 0 43.3891 519.77 PitF 20 47.1010 559.20 PitF 30 70.4623 886.00 PitF 40 76.5122 998.15 PitF 60 79.4669 0.3053 1057.98 355.72 PitF 80 78.1520 0.6193 0.4837 0.5515 1030.92 602.76 PitF 110 81.2970 0.3789 0.2685 0.3237 1096.82 374.16 PitF 120 79.8311 0.3250 1065.59 375.41 PitF 150 76.9050 0.1755 1005.91 225.41 SiteD 0 0.1329 221.68 SiteD 38 0.1364 0.0805 221.69 130.05 SiteD 48 0.2192 0.0928 222.06 142.47 SiteD 64 0.1817 0.0705 221.89 120.04 SiteD 83 0.1954 0.1007 221.95 150.37 SiteD 100 0.1698 0.1755 221.84 225.43 SiteD 121 0.2660 0.2455 222.26 295.70

SUPPLEMENTARY FIGURES

Figure 25: Stratigraphic column of the Bryn Mawr formation in Cecil County, Maryland, with depth in meters.

Figure 26: Stratigraphic soil column of the Emmaus Kame soil pit in Pennsylvania with depth in meters.

Figure 27: Stratigraphic soil column of the Penn State Ag. Center soil pit in Pennsylvania with depth in meters.

VITA

Laura Markley went to high school at Norwich Free Academy. After finishing high school in 2011, she went on to study Environmental Earth Science with a minor in Geographic Information Systems at Eastern Connecticut State University (ECSU). She graduated Summa Cum Laude as a university honors scholar from ECSU in 2015.