ﬂ.
.: . .12.,
t x J ... A

5.62: ﬁrhsuxﬁréﬁv . , V , . . .
A V V . z , V . . V . . . . V irratﬁchhi‘f

. :v J... z 5.

 

 

_ LIBRARY
Michigan State
University

 

 

 

This is to certify that the
thesis entitled

Homblende Etching as an Indicator of
Soil Development and
Relative Weathering Among Spodosols

presented by

Leslie Renee Mikesell

has been accepted towards fulﬁllment
of the requirements for the

 

 

MS. degree in Geological Sciences
jeré/aﬂvZ'A—ZJZ, Zy/é/

 

Major Professor's Signature

/0 £964 {mag/43.4) a;

 

Date

MSU is an Afﬁrmative Action/Equal Opportunity Institution

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

I DATE DUE

DATE DUE

DATE DUE

 

MW

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIFiC/DateDue.p65-p.15

 

HORNBLENDE ETCHING AS AN INDICATOR OF SOIL DEVELOPMENT
AND RELATIVE WEATHERING AMONG SPODOSOLS

By

Leslie Renee Mikesell

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Geological Sciences

2002

ABSTRACT

HORNBLENDE ETCHING AS AN INDICATOR OF SOIL DEVELOPMENT
AND RELATIVE WEATHERING AMONG SPODOSOLS

By

Leslie Renee Mikesell

Homblende etching holds great promise as a relative weathering indicator and as
another soil development indicator. This research was designed to test hornblende
denticulation amplitude as a weathering indicator, by comparing it to a standard
weathering indicator (quartz/ feldspar ratios), and to determine its reliability as an
indicator of soil development by comparing it to four soils in various stages of
development. The natural weathering of hornblende produces etching, which results in a
“denticulated” margin. These features gradually and systematically increase in depth and
size. In this study, hornblende denticulation amplitude measurements exhibited a clear
trend showing more extensively weathered grains up proﬁle, and therefore should be
considered an indicator of weathering. Homblende denticulation amplitude is a record of
weathering, which perhaps more accurately reﬂects weathering intensity rather than
duration. The denticulation amplitude measurements also mimicked the soil
development order of the four soils showing that it will also serve as a relative
weathering indicator among soils of the same age. Due to the fact that it correlated with
the rank order of soil development, it can also be used as a soil development indicator. In
this study, mineral weathering appears to be related to the degree of soil development in

soils representing different degrees of development but of the same age.

ACKNOWLEDGMENTS

There are many people who have inﬂuenced my educational choices, and to those
who steered me toward geology, I am eternally grateful.

I want to thank my committee, Dr. Michael Velbel, Dr. Randy Schaetzl, and Dr.
Gary Weissmann, for guiding me through this process and teaching me more than just
geology; I also learned about the educational process and about myself. Thank you for
your patience and encouragement, and for sharing your knowledge, expertise, and time
with me as I worked toward my goals. A special thanks goes to my advisor, Dr. Velbel,
for taking me on as a student after I attended a small, liberal arts college and then took
time off for several years before pursuing this degree. I truly appreciate what you did for
me. Thank you, Dr. Schaetzl, for your help collecting samples in the ﬁeld. I know I
could not have done that without you.

I would like to thank my family for all of their love and support. I am so
appreciative of my wonderful parents, Lowell and Betty Moore, who instilled in me a
love of education. I am grateful for support from my fabulous sister, Dawn Smithson,
who thinks more highly of me than she probably should, and her wonderful husband,
Scott, who is a great addition to our family. And last, but certainly not least, I want to
thank my awesome husband, Marc, whose unconditional love and unwavering support
are what saw me through some rough days during this endeavor. I cannot express in
words what you mean to me. I will always be grateful that you are in my life.

Special thanks go to many others who also helped me. Many thanks go to Beth

Weisenbom, who helped me more than she even knows. You taught me a lot about

iii

science, dedication, detemiination, and work ethics. I also want to thank Ewa
Danielewicz at the Center for Advanced Microscopy who deserves all the credit for the
great SEM images in this thesis. Thanks to Bruce Knapp of the USDA for his help in the
ﬁeld ﬁnding the sites to sample. I am also so glad to have met Dr. Lina Patino who has
been such an encouragement and friend. I want to express my appreciation to the
Geology Ofﬁce and Library staff for the great help they have been. Thank you, all for
your support.

A special thanks also goes to Dr. Max Reams of Olivet Nazarene University. Dr.
Reams is the one who awoke in me the love of geology, nurtured it, and then directed me
to Michigan State University to allow it to continue to grow. Thank you.

I want to thank the Department of Geological Sciences for ﬁnancial support
through two years of teaching assistantships. Also, this material is based upon work
supported by the National Science Foundation under Grant No. 9819148 and Grant No.

6799-00.

“Can you fathom the mysteries of God? ...speak to the earth, and it will teach you, ...for
does it not know that the hand of the Lord has done this?” Selections from Job 11 and

12.

iv

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................ viii

LIST OF FIGURES ...................................................................................................... ix

KEY TO SYMBOLS OR ABBREVIATIONS ............................................................ xvii

INTRODUCTION ........................................................................................................ 1
Homblende Etching as a Relative Age Indicator .................................................... 1
Quartz/Feldspar Weathering Ratio ......................................................................... 2
Soil Development .................................................................................................... 3
Research Questions ................................................................................................. 4
Purpose of Study ..................................................................................................... 4

CHAPTER 2

Homblende Weathering ................................................................................................ 6
General Description of Homblende ........................................................................ 6
Homblende Distribution ......................................................................................... 6
Weathering Environment ........................................................................................ 7
Homblende Weathering Rates ................................................................................ 8
Homblende Weathering Features ........................................................................... 9
Homblende Weathering in Soil .............................................................................. 15

CHAPTER 3

Quartz and Feldspar Weathering .................................................................................. 18
Mineral Persistence ................................................................................................. 1 8
Quartz and Feldspar ................................................................................................ 20
Quartz/Feldspar Ratio ............................................................................................. 22

CHAPTER 4

Spodosols ...................................................................................................................... 24
Podzolization ........................................................................................................... 24
Factors Contributing to Podzolization .................................................................... 25
The Spodosol Proﬁle ............................................................................................... 27
Extractable Iron and Aluminum in Spodosols ........................................................ 28

CHAPTER 5

Study Area .................................................................................................................... 31
Site Location and Description ................................................................................. 31
Site Selection Criteria ............................................................................................. 31

CHAPTER 6

Methods ......................................................................................................................... 37
Sample Preparation ................................................................................................. 37
Homblende Denticulation Amplitude Determination ............................................. 38

Homblende Classification Scheme ................................................................... 38
Unexpected Etching Features ........................................................................... 39
Quartz/Feldspar Ratio Determination ..................................................................... 49
Iron and Aluminum Extractions .............................................................................. 50
Laboratory-Produced Etching Features ........................................................... 50

CHAPTER 7

Results ........................................................................................................................... 69
Soil Classiﬁcation ................................................................................................... 69
Homblende Denticulation Amplitude ..................................................................... 81
Quartz/Feldspar Ratios ............................................................................................ 85
Extractable Iron and Aluminum .............................................................................. 85

CHAPTER 8

Discussion ..................................................................................................................... 91
Homblende Denticulation Amplitude ..................................................................... 91
Horizon vs. Homblende Denticulation Amplitude ................................................. 92
Inﬂuence of pH on Dissolution Rates ..................................................................... 94
Factors that Inﬂuence pH ........................................................................................ 95

Kalkaska Heavy Mineral Grains and Clay ....................................................... 96
Quartz/Feldspar Ratios as a Weathering Indicator ................................................. 107
Grain Size and Quartz/Feldspar Ratios ................................................................... 108
Summary ................................................................................................................. 1 10

CHAPTER 9

Conclusions ................................................................................................................... 1 12
Homblende Denticulation Amplitude ..................................................................... 112
Comparison of Homblende Denticulation Amplitude and Soil Development ....... 112
Comparison of Homblende Denticulation Amplitude and

Quartz/Feldspar Ratios ...................................................................................... 112
Comparison of Soil Development and Quartz/Feldspar Ratios .............................. 113
Weathering Versus Soil Development .................................................................... 113
Homblende as an Age Indicator ............................................................................. 114

APPENDICIES

Appendix A
Denticulation Amplitude Measurements Raw Data Table ..................................... 117

Appendix B
Quartz and Feldspar Raw Data Table .................................................................... 118

vi

Appendix C
Extractable Iron and Aluminum ............................................................................. 119

REFERENCES ............................................................................................................. 1 25

vii

LIST OF TABLES

Table l. ........................................................................................................................... 66
Pedon Data for Laboratory-Produced Alteration Features

Table 2. ........................................................................................................................... 70
Physical Properties of Pedons

Table 2 (continued) .......................................................................................................... 71
Physical Properties of Pedons

Table 3. ........................................................................................................................... 72
Chemical Properties of Pedons

Table 4. ........................................................................................................................... 83
Denticulation Amplitude Results

Table 5. ............................................................................................................................ 86
Quartz/Feldspar Ratios

Table 6. ........................................................................................................................... 89

Fe and Al Extraction Calculations

viii

LIST OF FIGURES
Images in this thesis are presented in color.

Figure 1. ........................................................................................................................... 1
Scanning electron micrograph showing naturally weathered hornblende with etch
pits, denticulated margin, and “teeth” parallel to the c-axis. Arrow indicates c-axis
direction. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

Figure 2. ........................................................................................................................... 11
Scanning electron micrograph of crystallographically controlled etch pits, which
form parallel to the c-axis in the direction of cleavage. Arrow indicates c-axis
direction. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

Figure 3. .......................................................................................................................... 12
Petro graphic micrograph of “sawtooth” termination on a naturally weathered
hornblende grain. Image captured with a Pixera digital imaging system.

Figure 4. .......................................................................................................................... 12
Scanning electron micrograph of weathered hornblende showing ferruginous
microboxwork (B) separated from denticulated hornblende (D) by void space with

pendants (P) projecting from the microboxwork into the void space (Figure 6 in
Velbel, 1989).

Figure 5. .......................................................................................................................... l3
Photomicrograph showing hornblende has been completely removed, leaving only
ferruginous boxwork (B) and pendants (P), or “negative pseudomorph” (Figure 9 in
Velbel, 1989).

Figure 6. .......................................................................................................................... 13
Scanning electron micrograph showing ferruginous boxwork (B) with neoformation
pendants (P) (Figure 8 in Velbel, 1989).

Figure 7. .......................................................................................................................... 14
Scanning electron micrograph of enlarged lens-shaped etch pits in naturally
weathered hornblende. Image taken at the Center for Advanced Microscopy at
Michigan State University by Leslie Mikesell and Ewa Danielewicz.

Figure 8. .......................................................................................................................... l4

Scanning electron micrograph of end-to-end coalescence of etch pits forming
grooves or striations in naturally weathered hornblende (Figure 3 in Velbel, 1989).

Figure 9. .......................................................................................................................... 19
Mineral-Stability Series in Weathering proposed by Goldich (1938, Table 18).

ix

Figure 10. ........................................................................................................................ 32
A map of counties in lower Michigan with Kalkaska County highlighted.

Figure 11. ........................................................................................................................ 32
A map of Kalkaska County townships with research sites indicated. Sites 1, 2, and
3 are in Blue Lake Township, while Site 4 is in Bear Lake Township.

Figure 12. ........................................................................................................................ 33
Map of the morainic system (shown in black) of Michigan with a portion of the
Outer Port Huron moraine and the study area for this research in north-central
Michigan is indicated in red. (This map is public domain, per a personal
communication with Dr. Randy Schaetzl.)

Figure 13. ........................................................................................................................ 35
Map showing the ice movement over the Great Lakes region during the Wisconsin
glacial stage. The arrows indicate ice ﬂow and the dashed lines indicate ice lobe
boundaries. The Outer Port Huron moraine was deposited by the Lake Michigan
Lobe, which ﬂowed southward from the Superior east area of the Canadian Shield
(modiﬁed from Dworkin et al., 1985).

Figure 14. ........................................................................................................................ 40
Diagrammatic etching scale showing progress of etching on broken (left) and
rounded (right) faces. Depth of etching is shown by indentation in grain margin and
is highlighted by shadow (Figure 2 in Locke, 1979).

Figure 15. ........................................................................................................................ 41
Class O. Petrographic micrograph of a hornblende grain with no apparent etching.
Image captured with a Pixera digital imaging system.

Figure 16. ........................................................................................................................ 41
Class 1. Petrographic micrograph of a hornblende grain with etching between 0pm
— 4pm. Image captured with a Pixera digital imaging system.

Figure 17. ........................................................................................................................ 42
Class 2. Petrographic micrograph of a hornblende grain with etching between 4pm
— 8pm. Image captured with a Pixera digital imaging system.

Figure 18. ........................................................................................................................ 42
Class 3. Petrographic micrograph of a hornblende grain with etching between 8pm
— 12 um. Image captured with a Pixera digital imaging system.

Figure 19. ........................................................................................................................ 43
Class 4. Petrographic micrograph of a hornblende grain with etching between

12 um — 16pm. Image captured with a Pixera digital imaging system.

Figure 20. ........................................................................................................................ 43
Class 5. Petrographic micrograph of a hornblende grain with etching between
16pm — 20pm. Image captured with a Pixera digital imaging system.

Figure 21. ........................................................................................................................ 44
Class 6. Petrographic micrograph of a hornblende grain with etching between
20pm — 24pm. Image captured with a Pixera digital imaging system.

Figure 22. ........................................................................................................................ 44
Class 7. Petrographic micrograph of a hornblende grain with etching between
24pm -— 28 um. Image captured with a Pixera digital imaging system.

Figure 23. ........................................................................................................................ 45
Class 8. Petrographic micrograph of a hornblende grain with etching between
28pm — 32 um. Image captured with a Pixera digital imaging system.

Figure 24. ........................................................................................................................ 46
Petrographic micrographs of weathered hornblende grains with holes. Arrows point
to holes. Images captured with a Pixera digital imaging system.

Figure 25. ........................................................................................................................ 47
Petrographic micrographs of weathered hornblende grains with fractures. Images
captured with a Pixera digital imaging system.

Figure 26. ........................................................................................................................ 48
Petrographic micrographs of weathered hornblende grains with both holes and
fractures. Images captured with a Pixera digital imaging system.

Figure 27. ........................................................................................................................ 53
Scanning electron micrograph of an untreated hornblende grain. Image taken at the
Center for Advanced Microscopy at Michigan State University by Leslie Mikesell
and Ewa Danielewicz.

Figure 28. ........................................................................................................................ 53
Scanning electron micrograph of the indicated area on Figure 27 showing smooth
surfaces with no etching features.

Figure 29. ........................................................................................................................ 54
Scanning electron micrograph of a hornblende grain treated with sodium
hypochlorite. Image taken at the Center for Advanced Microscopy at Michigan
State University by Leslie Mikesell and Ewa Danielewicz.

Figure 30. ........................................................................................................................ 54
Scanning electron micrograph of the indicated area on Figure 29 showing smooth
surfaces with no etching features.

xi

Figure 31. ........................................................................................................................ 55
Scanning electron micrograph of a hornblende grain treated with sodium citrate-
dithionite. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

Figure 32. ........................................................................................................................ 55
Scanning electron micrograph of the indicated ‘Area 1’ on Figure 31 showing
laboratory-produced alteration features.

Figure 33. ........................................................................................................................ 56
Scanning electron micrograph of the indicated ‘Area 2’ on Figure 31 showing large
laboratory-produced alteration feature.

Figure 34. ........................................................................................................................ 57
Scanning electron micrograph of a hornblende grain treated with sodium citrate-
dithionite. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

Figure 35. ........................................................................................................................ 57
Scanning electron micrograph of the indicated area on Figure 34 showing
laboratory-produced alteration at the edges.

Figure 36. ........................................................................................................................ 58
Scanning electron micrograph of a hornblende grain treated with sodium citrate-
dithionite. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

Figure 37. ........................................................................................................................ 58
Scanning electron micrograph of the indicated area on Figure 36 showing
laboratory-produced alteration on surfaces.

Figure 38. ........................................................................................................................ 61
Stereo pair 1. Polaroid pictures of scanning electron images were used to estimate
the depth of the laboratory-produced etch pits on the hornblende grains that were
treated with sodium citrate-dithionite. The feature in the left picture is centered and
straight; the right picture is tilted 3 degrees. Images were taken at the Center for
Advanced Microscopy at Michigan State University by Leslie Mikesell and Ewa
Danielewicz.

Figure 39. ........................................................................................................................ 62
Stereo pair 2. Polaroid pictures of scanning electron images were used to estimate
the depth of the laboratory-produced etch pits on the hornblende grains that were
treated with sodium citrate-dithionite. The feature in the left picture is centered and
straight; the right picture is tilted 3 degrees. Images were taken at the Center for
Advanced Microscopy at Michigan State University by Leslie Mikesell and Ewa
Danielewicz.

xii

Figure 40. ........................................................................................................................ 63
Stereo pair 3. Polaroid pictures of scanning electron micrographs were used to
estimate the depth of the laboratory-produced etch pits on the hornblende grains
that were treated with sodium citrate-dithionite. The feature in the left picture is
centered and straight; the right picture is tilted 3 degrees. Images were taken at the
Center for Advanced Microscopy at Michigan State University by Leslie Mikesell
and Ewa Danielewicz.

Figure 41. ........................................................................................................................ 64
Stereo pair 4. Polaroid pictures of scanning electron micrographs were used to
estimate the depth of the laboratory-produced etch pits on the hornblende grains
that were treated with sodium citrate-dithionite. The feature in the left picture is
centered and straight; the right picture is tilted 3 degrees. Images were taken at the
Center for Advanced Microscopy at Michigan State University by Leslie Mikesell
and Ewa Danielewicz.

Figure 42. ........................................................................................................................ 73
The Kalkaska pedon is the most strongly expressed Spodosol in the development
sequence of this study. Photo taken by Dr. Randy Schaetzl, Professor, Dept. of
Geography, Michigan State University.

Figure 43. ........................................................................................................................ 74
Selected information from the Soil Proﬁle Description Sheet of the Kalkaska pedon.

Figure 44. ........................................................................................................................ 75
The strong Rubicon pedon is the second best expressed Spodosol in the
development sequence of this study. Picture taken by Dr. Randy Schaetzl,
Professor, Dept. of Geography, Michigan State University.

Figure 45. ........................................................................................................................ 76
Selected information from the Soil Proﬁle Description Sheet of the strong Rubicon
pedon.

Figure 46. ........................................................................................................................ 77

The weak Rubicon pedon is the third best expressed Spodosol in the development
sequence of this study. Photo taken by Dr. Randy Schaetzl, Professor, Dept. of
Geography, Michigan State University.

Figure 47. ........................................................................................................................ 78
Selected information from the Soil Proﬁle Description Sheet of the weak Rubicon
pedon.

Figure 48. ........................................................................................................................ 79

The Grayling pedon is the least strongly expressed Spodosol in the development
sequence of this study. Photo taken by Dr. Randy Schaetzl, Professor, Dept. of
Geography, Michigan State University.

xiii

Figure 49. ........................................................................................................................ 80
Selected information from the Soil Proﬁle Description Sheet of the Grayling pedon.

Figure 50. ........................................................................................................................ 82
Petrographic micrograph of the denticulated margins of a naturally weathered
hornblende grain. Image captured with a Pixera imaging system.

Figure 51. ........................................................................................................................ 84
Denticulation amplitude vs. horizon plot for each of the pedons in this study. These
plots show the change in the mean denticulation amplitude measurement with
change in horizon. Error bars represent 1 standard deviation.

Figure 52. ........................................................................................................................ 87
Horizon vs. Quartz/Feldspar Ratio. Numbers were used as horizon designations
instead of traditional horizon nomenclature because each pedon has various horizon
orders. This plot shows the change in the quartz/ feldspar ratio with depth.

Figure 53. ........................................................................................................................ 98
Scanning electron micrograph of a weathered heavy mineral grain from the A
horizon of the Kalkaska pedon with very little weathering product. This product is
an aluminosilicate. Image taken at the Center for Advanced Microscopy at
Michigan State University by Leslie Mikesell and Ewa Danielewicz.

Figure 54. ........................................................................................................................ 98
Energy-dispersive spectrograph of the weathering product indicated the by circle in
Figure 53. The high Si content with A1 and 0 led to the determination that it is an
aluminosilicate. The Au is the signature of the coating used to prepare the grain.

Figure 55. ........................................................................................................................ 99
Scanning electron micrograph of a weathered heavy mineral grain from the A
horizon of the Kalkaska pedon with organic material present. Image taken at the
Center for Advanced Microscopy at Michigan State University by Leslie Mikesell
and Ewa Danielewicz.

Figure 56. ........................................................................................................................ 99
Energy-dispersive spectrograph of the organic material indicated by the circle in
Figure 55 showing high C content. The Au is the signature of the coating used to
prepare the grain.

Figure 57. ........................................................................................................................ 100
Scanning electron micrograph of a weathered heavy mineral grain from the A
horizon of the Kalkaska pedon with tube-like organic material present. Image taken
at the Center for Advanced Microscopy at Michigan State University by Leslie
Mikesell and Ewa Danielewicz.

xiv

Figure 58. ........................................................................................................................ 100
Scanning electron micrograph of tube-like organic material indicated by the circle
in Figure 57.

Figure 59. ........................................................................................................................ 101
Energy-dispersive spectrograph of tube-like organic material indicated by the circle
in Figure 57 showing high C content. The Au is the signature of the coating used to
prepare the grain.

Figure 60. ........................................................................................................................ 102
Scanning electron micrograph of a weathered heavy mineral grain from the E
horizon of the Kalkaska pedon with more weathering product than the A horizon.
This product is an aluminosilicate. Image taken at the Center for Advanced
Microscopy at Michigan State University by Leslie Mikesell and Ewa Danielewicz.

Figure 61. ........................................................................................................................ 102
Energy-dispersive spectrograph of the weathering product indicated the by circle in
Figure 60. The high Si content with Al and 0 led to the determination that it is an
aluminosilicate. The Au is the signature of the coating used to prepare the grain.

Figure 62. ........................................................................................................................ 103
Scanning electron micrograph of a weathered heavy mineral grain from the Bhs
horizon of the Kalkaska pedon with more weathering product and organic material
than the E horizon. This weathering product is an aluminosilicate. Image taken at
the Center for Advanced Microscopy at Michigan State University by Leslie
Mikesell and Ewa Danielewicz.

Figure 63. ........................................................................................................................ 103
Energy-dispersive spectrograph of the weathering product indicated the by circle in
Figure 62. The high Si content with Al and 0 led to the determination that it is an
aluminosilicate. The Au is the signature of the coating used to prepare the grain.

Figure 64. ........................................................................................................................ 104
Scanning electron micrograph of a weathered heavy mineral grain from the B51
horizon of the Kalkaska pedon with an extensive coating of weathering product.
This product is an aluminosilicate. Image taken at the Center for Advanced
Microscopy at Michigan State University by Leslie Mikesell and Ewa Danielewicz.

Figure 65. ........................................................................................................................ 104
Energy-dispersive spectrograph of the weathering product indicated by the circle in
Figure 64. The high Si content with Al and 0 led to the determination that it is an
aluminosilicate. The Au is the signature of the coating used to prepare the grain.

XV

Figure 66. ........................................................................................................................ 105
Scanning electron micrograph of a weathered heavy mineral grain from the B81
horizon of the Kalkaska pedon with string-like features that were determined to not
be organic due to the EDS having the same signature as the weathering product.
Image taken at the Center for Advanced Microscopy at Michigan State University
by Leslie Mikesell and Ewa Danielewicz.

Figure 67. ........................................................................................................................ 105
Scanning electron micrograph of string-like features indicated by the circle in
Figure 66.

Figure 68. ........................................................................................................................ 106

Energy-dispersive spectrograph of the string-like features indicated the by circle in
Figure 66. The features were determined to not be organic due to the EDS having
the same signature as the weathering product. The Au is the signature of the
coating used to prepare the grain.

xvi

KEY TO SYMBOLS OR ABBREVIATIONS

um .................. micrometer

AAO ............... acidiﬁed ammonium-oxalate
AAS ................ atomic absorption spectrometer
CD .................. sodium citrate-dithionite

EDS ................ energy-dispersive spectrograph

F ed and AL; ..... iron and aluminum extracted with sodium citrate-dithionite
Fe0 and A10 ..... iron and aluminum extracted with acidiﬁed ammonium-oxalate

Fep and Alp ..... iron and aluminum extracted with sodium pyrophosphate

PP ................... sodium pyrophosphate

SEM ............... scanning electron micrograph
vs. ................... versus

yr B.P. ............ years before present

xvii

INTRODUCTION

Homblende Etching as a Relative Age Indicator

Homblende etching holds great promise as a relative weathering indicator.
Homblende, a member of the amphibole class of minerals, is common and abundant on
the earth’s surface (Blackburn and Dennen, 1988). Natural weathering of hornblende
corrodes the hornblende in a crystallographically controlled manner, resulting in etch pits
onthe surface of the mineral and a “sawtooth” or “denticulated” margin on which the
“teeth” are parallel to the c-axis (e.g., Bemer et al., 1980; Bemer and Schott, 1982;
Velbel, 1987, 1989) (Figure 1). Progressive weathering causes these features to
gradually and systematically increase in depth and size from a well-deﬁned starting point
(Locke, 1979, 1986; Birkeland, 1978), the Antarctic (Ugolini and Bull, 1965; Everett,

1971; Locke, 1979, 1986). Using this knowledge, hornblende etching has been applied

/

 

20pm

Figure 1. Scanning electron micrograph showing naturally weathered hornblende with
etch pits, denticulated margin, and “teeth” parallel to the c-axis. Arrow indicates c-axis
direction. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

as a relative age indicator of soils in the Rocky Mountains of the United States (Hall and
Heiny, 1983; Hall and Martin, 1986; Hall and Michaud, 1988), the Canadian Arctic
(Everett, 1971), and Italy (Andrews and Miller, 1980). This was accomplished by
statistically comparing the depth of hornblende denticulation, or denticulation amplitude,
in soils and sediments of different ages (Locke, 1979). Homblende from the older sites
exhibited a greater degree of etching, as measured by greater amplitude, than the
hornblende from the younger sites (Locke, 1979). Thus, it was concluded that
hornblende could be used to determine relative age.

In the previously mentioned studies, the researchers assumed that time of
exposure, or duration of weathering, was the primary factor that inﬂuenced denticulation
amplitude. These studies were based on the idea that etch features enlarge over time
without taking into consideration other factors that may affect this enlargement. Factors,
other than time, that may signiﬁcantly contribute to this weathering process include:
climate, amount of water present, pH of soil, presence and abundance of clay, and
presence of organic matter (White and Brantley, 1995). Previous work has only
emphasized the relationship between hornblende denticulation and the duration of
weathering. However, hornblende denticulation amplitude is a cumulative record of
weathering, which reﬂects both the weathering intensity and the weathering duration.
Thus, denticulation amplitude measurements might serve as a relative weathering

indicator among soils of the same age but differing degrees of development.

Quartz/Feldspar Weathering Ratio

The quartz/feldspar ratio is a measure used to determine the degree of mineral

weathering based on the persistence of quartz compared to that of feldspar (e. g.,
Birkeland, 1974; Muhs 1982; Bockheim et al., 1996). From Goldich’s (1938) study, we
know that a sediment will become depleted in feldspar with respect to quartz as
weathering continues because feldspar is less persistent than quartz. Since the
hornblende denticulation amplitude may also indicate the degree of mineral weathering,
these two measurements can provide information regarding the degree of mineral

weathering.

Soil Development

Many researchers have assumed that mineral weathering and soil development
occur simultaneously. Due to this assumption, or poor word choice, the terms
“weathering” and “soil development” have often been used interchangeably in the
literature (e. g., Ruhe, 1956; Locke, 1979; Andrews and Miller, 1980). However,
weathering and soil development are two different but inter-related processes (e. g.,
F anning and Fanning, 1989). Weathering is the alteration of rock and minerals, at or near
the Earth’s surface, due to environmental conditions (Bland and Rolls, 1998). On the
other hand, soil development is the amount of pedologic change, including horizonation,
that has taken place in a parent material (Birkeland, 1999) and is a function of climate,
biota, topography, parent material, and time (Jenny, 1941). It is possible that soil
development and weathering will correspond to each other, and that a measure of one
may proxy for a measure of the other, but the two measures are not necessarily providing
the same information about the soil. If hornblende denticulation amplitude can produce

the same trend with regards to relative weathering as soil development does with regards

to relative proﬁle development, hornblende denticulation amplitude could be used as a

proxy for soil development.

Research Questions

Soil development is a function of several factors; these include climate, biota,
topography, parent material, and time (Jenny, 1941). The inﬂuence of any one factor is
studied by examining a sequence of soils in which all but one of these soil-forming
factors are held as constant as possible, so that only one factor varies in the sequence
(Birkeland, 1999). Using this approach, the degree of hornblende weathering, as
measured by denticulation amplitude, has been used to determine relative age of soils on
glacial parent materials in studies where time or soil age is the only soil-forming factor
allowed to vary (e.g., Hall and Heiny, 1983; Locke, 1979, 1986; Birkeland, 1978).
However, the inﬂuence of soil-forming factors other than time on hornblende weathering
remains to be examined. As Fanning and Fanning (1989) point out, soil-development
involves a variety of processes, of which weathering is only one, and some aspects of soil
development can proceed without progress in mineral weathering, and vice versa. Thus,
a number of questions relating hornblende weathering to soil development remain. These
include: 1) is hornblende weathering (as one type of mineral weathering) related to or
independent of soil development? and 2) is hornblende denticulation amplitude

comparable to other measures of mineral weathering?

Purpose of Study

This research was designed to examine the relationship between mineral

weathering in general, homblende weathering in particular, and soil development. In
order to examine these relationships, four soils, similar in age but at various stages of
development, from northern lower Michigan were sampled for this study. Denticulation
amplitude measurements were taken from each horizon in every pedon. A plot of
horizon vs. denticulation amplitude was then constructed to determine if any correlations
could be made. A quartz/feldspar ratio was also determined for each horizon, and a
horizon vs. quartz/ feldspar ratio plot was then constructed and examined to determine if
correlations could be drawn from these plots. If it is determined that correlations may be
drawn between hornblende denticulation amplitude, or quartz/feldspar ratios, and soil
development, it would provide researchers with other analytical tools that might be useful

in situations where traditional analysis do not yield conclusive results.

CHAPTER 2

Homblende Weathering

General Description of Homblende

Homblende is the most common mineral species of the amphibole group. It is a
ferromagnesian mineral in the inosilicate mineral class (Blackburn and Dennen, 1988),
having the general composition of
[(Na,K)0-lCa2(Mg,Fe2+,Fe3+,Al)5Sim,5Alo,5-2022(OH)2] (Brantley and Chen, 1995).
Inosilicates are characterized by silica tetrahedra that are linked in double chains by
sharing alternately two and three oxygen atoms per tetrahedron (Brantley and Chen,
1995)

Blackburn and Dennen (1988) describe hornblende as black to dark green, and it
generally appears columnar, bladed, or ﬁbrous, but can also be granular massive. The
crystallography of hornblende is monoclinic, 2/m. Its crystals are perfect prismatic {110}
at 56° and 124°, with good cleavage also at 56° and 124°. Homblende has subconchoidal

to uneven fracture, with a hardness of 5 - 6, and a speciﬁc gravity of 2.9 - 3.3.

Homblende Distribution

Homblende is widely distributed in igneous rocks and is often associated with
quartz, feldspar, and pyroxene. It is also a common component of metamorphosed
basaltic rocks (Blackburn and Dennen, 1988). Goldich (193 8) found hornblende to be
more persistent than olivine, pyroxene, or calcic plagioclase during weathering, but less

persistent than micas, alkali feldspars, or quartz. Because it is common in both igneous

and metamorphic rocks, it is abundant in transported material as well. Several studies
have found that it is the most abundant of the heavy minerals in soils formed in a variety
of parent materials, including hornblende schist in central Texas (Stahnke, 1965, as cited
in Allen and Hajek, 1989), loess-mantled till in southwestern Iowa (Ruhe, 1956), glacial
till in southwestern Montana (Hall and Martin, 1986), glacial till in Ontario, Canada
(Dreimanis et al., 1957), glacial till in the northeastern United States (Johnson and Chu,
1983, as cited in Allen and Hajek, 1989), soils in the Great Lakes region (Dworkin et al.,
1985), glacial till in south-central Michigan (Haile-Mariam and Mokma, 1996), and
sandy outwash material of northern Michigan (Haile-Mariam and Mokma, 1995).
Because hornblende is common and widespread, better understanding of its weathering is

potentially widely applicable.

Weathering Environment

Soil environments that are well drained and leached are more conducive to
weathering than environments that are poorly drained (Cremeens et al., 1992; Haile—
Mariam and Mokma, 1996,1995). Bemer (1978) found that ﬂushing water through rock,
sediment, or soil removed dissolved weathering products so that they are no longer in
contact with the mineral, thus increasing the dissolution rate. He went on to state that
with continuous ﬂushing, a limit could be reached where the ﬂushing rate is no longer a
controlling factor, but the mineral reactivity controls the rate. This occurs when
sufﬁcient water passes through the rock or soil to remove any weathering product, at
which point the factor that controls the dissolution rate is the reactivity of the mineral. In

the unsaturated zone of a soil, ﬂushing controls the dissolution rate due to an insufﬁcient

amount of water. Therefore, the most inﬂuential climatic factors on hornblende etching

are effective precipitation and inﬁltration of unfrozen water (Locke, 1979).

Homblende Weathering Rates

Weathering rates have been determined for naturally weathered hornblende by
studying the mean maximum etching depth, or MMED, (Locke, 1979, 1986) and etch pit
distribution (Cremeens et al., 1992; MacInnis and Brantley, 1993). These rates were
generally found to be 101 — 104 times slower than the experimentally determined rates
(MacInnis and Brantley, 1993; White et al., 1996). The general discrepancy between
natural weathering rates, and experimentally determined rates, is discussed in detail by
Velbel (1993a). Brantley et al. (1993) provide possible explanations for the difference
between weathering rates determined in the laboratory and those determined by naturally
weathered hornblende, such as: lack of understanding of reaction mechanisms,
experimental error, inaccuracies in reactive surface area estimation, inaccuracies in
wetted surface area estimation, temperature variability, and biological processes.
Another possible reason for the discrepancy in these rates is the changing dissolution
rates in the natural environment due to rate constants that decrease with time (White et
al., 1996; Hall and Martin, 1986). Murakami et al. (1998) also found a discrepancy
between natural rates and laboratory determined rates of weathering in anorthite due to
the decreasing weathering rate in nature. However, they concluded that these rates
decreased due to the saturation effect caused by secondary minerals between grains. Not
only are weathering rates slower in nature relative to laboratory experiments, but they

also differ between the upper part of the proﬁle where the rate is rapid, and the lower part

of the proﬁle where the rate is slow (Markos, 1975).

Several possible factors that could affect the dissolution rate of minerals in the
natural environment include: precipitation of dissolved silica, iron precipitates, and
organic and inorganic coatings on grain surfaces (Dove, 1995). Nugent et a1. (1998)
found thin, hydrous, patchy natural coatings of amorphous and crystalline aluminosilicate
on feldspar grains that were only visible under atomic force microscopy. They concluded
that these coatings were partially inhibiting dissolution and thus contributing to the
discrepancy between natural and laboratory determined weathering rates. However,
Velbel (1993b) argued that examination of naturally weathered silicate minerals reveals
etch pits, which are features produced when chemical reactions at the interface control
the weathering rate. The presence of etch pits on surfaces beneath clay coatings or oxide
and/or hydroxide products indicate diffusion through a surface layer is not rate-limiting.
Similarly, Bemer and Schott (1982) found that cation-depleted layers on soil grains were
too thin to be considered diffusion-limiting, and therefore weathering is controlled by
surface chemistry and not by diffusion through a surface layer. However, precipitates
and alteration products could slow dissolution rates by blocking or altering ﬂow paths,
and thus decrease the amount of water that reaches the mineral (Eggleton, 1986, as cited

by Brantley and Chen, 1995; Velbel, 1993b; Murakami et al., 1998).

Homblende Weathering Features
Natural weathering of hornblende forms etch pits on the surface of the grain.
These dissolution voids may be seen using a petrographic microscope. Velbel (1989)

noted that hornblende weathering progresses through a sequence of features that indicate

the extent of weathering. The earliest feature of weathering (under oxidizing conditions)
is the presence of ferruginous material along grain boundaries or in fractures within the
grain observed in thin section. This chemical precipitation of goethite, gibbsite, or
kaolinite produces a boxwork structure. The more extensively weathered hornblende
exhibits crystallographically controlled etch pits, which form parallel to the c-axis in the
direction of cleavage (Figure 2). This feature is formed by the selective weathering at
crystal defects and dislocations (Bemer et al., 1980; Velbel, 1993b). More advanced
weathering produces a sharp denticulated margin, also known as cockscomb termination,
hacksaw termination (e.g., Bemer et al., 1980; Bemer and Schott, 1982; Velbel, 1987,
1989, 1993b), or “sawtooth” termination (Velbel, 1989) (Figure 3). These denticulations
result from side-by-side coalescence of lenticular, or lens-shaped, etch pits (Bemer et al.,
1980; Bemer and Schott, 1982; Velbel, 1993b). Velbel (1989) goes on to state that even
more advanced weathering produces hornblende remnants that occupy dissolution
cavities bounded by ferruginous boxworks (Figure 4). These are void spaces that lack
weathering product, either in the cavity or on the hornblende surface. If the parent
hornblende is removed through dissolution, a “negative pseudomorph” or
alveoporomorph (terminology of Delvigne, 1998) is left (Figures 5 and 6). An interesting
feature of hornblende weathering observed by Velbel (1989) was pendants of ferruginous
material that were separated from the hornblende remnants by void space (Figure 4).
This separation suggests that the pendants formed by dissolution-reprecipitation, or
neoformation.

Homblende etch pits progress through a sequence of stages (Bemer et al., 1980;

Bemer and Schott, 1982; Cremeens et al., 1992). These stages begin with the formation

10

of one or more lens-shaped etch pits that enlarge with weathering (Figure 7). The etch
pits then coalesce either end-to-end or side-by-side. End-to-end coalescence results in
the formation of longitudinal grooves or striations (Figure 8). It is the side-by-side
coalescence of etch pits that form the denticulated margin (Figure 3). These tooth-like
denticulations are unweathered hornblende, which are adjacent to pits that formed by the
dissolution of the hornblende.

Etch pits are crystallographically controlled dissolution voids, produced by
interface-limited weathering mechanisms (Velbel, 1993b, and references contained
therein). Bemer and Schott (1982) put forth the idea that this mechanism consists of H”
attacking the mineral at a point of excess energy, primarily at a dislocation outcrop,
which is accompanied by Fe oxidation in the outermost atomic layers. This selective
attack also occurs at crystal defects (Bemer et al., 1980). However, this does not

penetrate inward over time, but produces lens-shaped etch pits (Bemer and Schott, 1982).

 

200nm

Figure 2. Scanning electron micrograph of crystallographically controlled etch pits,
which form parallel to the c-axis in the direction of cleavage. Arrow indicates c-axis
direction. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

11

 

ti, ‘

20pm

 

Figure 3. Petrographic micrograph of “sawtooth” termination on a naturally weathered
hornblende grain. Image captured with a Pixera digital imaging system.

 

Figure 4. Scanning electron micrograph of weathered hornblende showing ferruginous
microboxwork (B) separated from denticulated hornblende (D) by void space with
pendants (P) projecting from the microboxwork into the void space (Figure 6 in Velbel,
1989).

 

Figure 5. Photomicro graph showing hornblende has been completely removed, leaving
only ferruginous boxwork (B) and pendants (P), or “negative pseudomorph” (Figure 9 in
Velbel, 1989).

 

Figure 6. Scanning electron micrograph showing ferruginous boxwork (B) with
neoformation pendants (P) (Figure 8 in Velbel, 1989).

 

Figure 7. Scanning electron micrograph of enlarged lens-shaped etch pits in naturally
weathered hornblende. Image taken at the Center for Advanced Microscopy at Michigan
State University by Leslie Mikesell and Ewa Danielewicz.

 

Figure 8. Scanning electron micrograph of end-to-end coalescence of etch pits forming
grooves or striations in naturally weathered hornblende (Figure 3 in Velbel, 1989).

Cremeens et al. (1992) suggested that etch pits uniform in size and shape might be
associated with strain due to exsolution lamellae. They also concluded irregularly shaped

and spaced pits might be associated with microcracks and fractures.

Homblende Weathering in Soils

The weathering of hornblende can produce different weathering features and
products depending on environmental conditions. Etch pit size and shape could be an
indicator of the environment in which they formed. Cremeens et a1. (1992) studied the
etch pits in orthoclase and pyriboles. They observed that in poorly drained soils the
minerals had small pits but a large number of them. This suggested a slower growth rate
and presumably less ﬂushing with fresh water. Greater coalescence of pits was observed
in well drained soils. These pits were larger, irregularly shaped, and elongated,
suggesting a greater grth rate, which they attributed to increased ﬂushing of fresh
water, which would remove the weathering products and reduce contact time of the
primary mineral with its weathering products.

In a study conducted in Maryland, the weathering products of hornblende, such as
Fe oxides and oxyhydroxides and clay minerals, were dependant upon the intensity of
leaching in the proﬁle (Cleaves, 1974). Cleaves (1974) also found that in a poorly
drained, structured saprolite developed on a quartz-plagioclase-hornblende gneiss,
hornblende weathered to montrnorillonite. It weathered to ferric oxides in a well drained,
structured saprolite developed on the same quartz-plagioclase-hornblende gneiss parent
material. He also noted that the weathering products of hornblende in a well drained

saprolite (which developed on an epidote amphibolite) included kaolinite, halloysite,

15

goethite and other ferric oxides, and isotropic amorphous material. Velbel (1989) found
that hornblende weathered to goethite, gibbsite, and kaolinite in an intensely leached
weathering proﬁle on a plagioclase-hornblende gneiss of the Carroll Knob maﬁc complex
of North Carolina. Other weathering products of hornblende include: iron-rich
montmorillonite (Walker et al., 1967), vermiculite, chlorite (Stephen, 1952, as cited by
Huang, 1989), beidellite (Goldich, 1938), and smectite (Rice et al., 1985).

In the soil, hornblende is most commonly found in the sand and silt size fraction
where it constitutes a signiﬁcant portion of the heavy minerals (Ruhe et al., 1966; Locke,
1979; Allen and Hajek, 1989). Haile-Mariam and Mokma (1995, 1996) studied the
mineralogy of two ﬁne-loamy hydrosequences in south-central Michigan, and two sandy
Spodosol hydrosequences also in Michigan. In all the pedons studied, they found that the
percent of hornblende increased with depth, as the percent of quartz decreased with
depth. They attributed this trend to the persistence of quartz compared to other minerals.

In addition to mineral depletion trends, hornblende etching trends have been
observed in soils. Locke (1979) conducted a study on the etching of ﬁne sand sized
hornblende collected from a coarse-grained glacial deposit from Bafﬁn Island, Canada.
He found that etching decreased logarithmically with increasing depth in the proﬁle. He
also noted that the rate of etching at any give depth decreased logarithmically with
increasing soil age. He proposed that precipitation and unfrozen water inﬁltration
inﬂuenced the etching rate as a function of depth and age. Based on these observations,
Locke (1979) felt that hornblende etching showed great promise as a relative dating

technique.

16

In a later study, Locke (1986), found that soils of similar age exhibit trends where
hornblende etching was greatest near the surface and decreased with depth. This trend
was interpreted to be due to a decrease in soil wetting events at depth. He also noted that
hornblende etching showed signiﬁcant differences in soils of similar age that formed
under different climatic regimes. This difference in hornblende etching was attributed to
differential rainfall. This precipitation theory was supported by Hall and Horn (1993),
who studied soils of two chronosequences, one each from the Wind River Range in
western Wyoming and the Tobacco Root Range in southwestern Montana, U.S.A. They
found faster weathering rates in the soils at higher elevations due to increased orographic

precipitation.

17

CHAPTER 3

Quartz and F eldspar Weathering

Mineral Persistence

Many studies have been conducted to determine the stability of minerals by
comparing the mineral composition of the fresh parent material to its weathering product
(e.g., Goldich, 1938; White et al., 1996; Birkeland, 1999). This comparison allowed
researchers to quantify the relative depletion of a mineral from a system. Less stable
minerals are preferentially depleted. Using relative depletion data, weathering sequences
have been developed that rank minerals according to their stability (Goldich, 1938;
Pettijohn, 1941, as cited in Bland and Rolls, 1998; Ollier and Pain, 1996).

Samuel Goldich (193 8) developed the weathering sequence that is still the
standard today. Goldich (193 8) developed what he called a mineral-stability series in
weathering by conducting extensive analyses on fresh and weathered materials from
several sites. He studied a fresh amphibolite from the Black Hills of South Dakota and
found it to be 77 percent hornblende while its residual material was only 7 percent
hornblende. He concluded that homblende was undergoing extensive alteration to
beidellite or related clay minerals.

Goldich (193 8) also studied the Morton Gneiss in Minnesota. He concluded that
the heavy minerals of this granite gneiss had been largely destroyed and that the
remaining minerals represented the relatively resistant or more stable minerals. The least
stable minerals were plagioclase, hornblende, epidote, titanite, and apatite, whereas

ilmenite-magnetite, biotite, microcline, and orthoclase were more resistant. Quartz and

18

zircon were the most stable minerals, with even these showing signs of being altered.

Additionally, Goldich (193 8) analyzed the Medford diabase from eastern
Massachusetts. He found an absolute loss of heavy minerals as a result of weathering.
He also noted the relative abundances of heavy minerals in the fresh and weathered
diabase suggested that the rate of weathering differed for various minerals. From this
diabase, Goldich ranked the minerals from least to most stable: chlorite, pyroxene,
epidote, hornblende, biotite, apatite, and magnetite-ilmenite.

From this classic study, Goldich (1938) proposed a mineral-stability series for
weathering in which the minerals were ranked from least stable to most stable as:
olivine, augite, hornblende, biotite, potassium feldspar, muscovite, and quartz (Figure 9).
The plagioclase series is separate from the maﬁc series with sodic plagioclase being more
stable than calcic plagioclase. As a group, the maﬁc minerals are less stable than felsic

minerals.

 

MINERAL-STABILITY SERIES IN WEATHERING

Olivine
Calcic plagioclase
Augite
Calcic-alkalic plagioclase
Homblende Alkali-calcic plagioclase
Alkalic plagioclase
Biotite
Potash feldspar
Muscovite
Quartz

 

 

 

Figure 9. Mineral-Stability Series in Weathering proposed by Goldich (1938, Table 18).

19

Despite its historical precedence, Velbel (1999) points out that the term “stability
series” is a misnomer. The series proposed by Goldich (1938) and the many that
followed (e.g., Pettijohn, 1941, as cited in Bland and Rolls, 1998; Dryden and Dryden,
1946; Nickel, 1973; Morton, 1984; Bateman and Catt, 1985; Allen and Hajek, 1989;
White et al., 1996; Velbel, 1999) rank the persistence of different minerals. These
rankings are due to the kinetics of the reactions, not the thermodynamic stability of the

mineral. A more accurate term for these series should be “mineral persistence series”.

Quartz and F eldspar

Two relatively abundant minerals that are commonly compared, and which were
chosen for this study, are quartz and feldspar. A soil will become depleted in feldspar,
with respect to quartz, because feldspar is less resistant to weathering than quartz (e. g.,
Goldich, 1938; Ollier and Pain, 1996; Birkeland, 1999).

Quartz, with the formula of SiOz, is a member of the silica group of tectosilicates.
The basic building block of silicate minerals is the tetrahedron, where a Si4+ is
surrounded by four 02' ions. The tectosilicates are those silicate minerals in which all
oxygens are shared with adjacent tetrahedra to form a network with the net charge of zero
(Blackburn and Dennen, 1988). Blackburn and Dennen (1988) provide a summary of
crystallography, physical properties, and occurrence of quartz. Quartz forms hexagonal
prisms with pyramidal termination and can have either leﬁ- or right-handed forms. It is
usually found massive or as crystals. Quartz has no cleavage and no parting. It has
irregular to conchoidal fracture. Quartz has a hardness of 7 and a speciﬁc gravity of 2.65.

Quartz is one of the most abundant minerals and is an important constituent in many

20

igneous, metamorphic, and sedimentary rocks. It is a major component in some rocks,
such as sandstone and quartzite. Quartz is often the sole constituent in veins and is
almost universally found in soils and sediments. Ollier and Pain (1996) note that quartz
is often the last mineral to crystallize from a magma, and due to this lower temperature of
crystallization, it is fairly stable at earth surface temperatures. Although quartz is
extremely resistant to chemical weathering, physical weathering by wind or water can
round quartz grains. Due to its resistance to chemical weathering, quartz can be a
standard by which the weathering of other grains can be measured. Ollier and Pain
(1996) state that due to its persistence, the proportion of quartz increases up proﬁle
because as other minerals are removed higher in the proﬁle, quartz remains.

Feldspars are aluminosilicate minerals, with the alkali feldspars having the
general formula (K,Na)(AlSi3Og) and the plagioclase feldspars having the general
formula (Na,Ca)(Al,Si)4Og (Blackburn and Dennen, 1988). Blackburn and Dennen
(1988) provide a summary of the feldspars. The distinction between the alkali feldspars
and the plagioclase feldspars is not always clear-cut due to solid solution. However, the
alkali feldspars are monoclinic, or, if triclinic, have more K than Ca. The plagioclase
feldspars are triclinic with less K than Na or Ca. Due to the solid solution, a wide range
of chemical composition is permissible. Twinning is very common among the feldspars,
occuning as simple or repeated twins. The feldspars make up almost 60% of igneous and
metamorphic rocks, which constitute over 90% of the earth’s crust. Feldspars are found
in the light fraction of soils and have a speciﬁc gravity of 2.6. Ollier and Pain (1996)
note that feldspars alter rapidly due to the pronounced cleavage that probably allows

rapid penetration of water. Feldspars often alter to kaolin, but other secondary minerals,

21

such as mica and allophane, can also be formed.

Quartz/Feldspar Ratio

Quartz/ feldspar ratios examine the relationship between a resistant mineral and a
non-resistant mineral. In the soil environment, feldspar is more quickly leached than
quartz (Nesbitt et al., 1997). Nesbitt et al. (1997) describe this relationship in a
simpliﬁed model that has plagioclase and K-feldspar leached at approximately the same
rate, which is 100 times faster than quartz. When dissolution begins, the soil quickly
reaches silica saturation with respect to quartz. This saturation limit causes quartz to stop
dissolving, however feldspar and plagioclase continue dissolving. Generally, silica
saturation with respect to K-feldspar is reached next allowing only plagioclase to
continue dissolving. They also note that the soil environment is critical in these reactions
because the amount of organic acids present determines the amount of mineral
dissolution.

Studies have shown that the feldspar content of a soil decreases in the upper
horizons due to an increasing of weathering intensity (e. g., Nesbitt et al., 1997; Allen and
Haj ek, 1989). Birkeland (1974) stated that this observation is one way of determining
weathering rates. Rate determination is accomplished by establishing the ratio of a
resistant mineral (e. g. quartz) to a nonresistant mineral (e. g. feldspar) in the parent
material and comparing that to the ratio of the same minerals in a horizon of known age.
This gives the rate at which the less resistant mineral is depleted. This is the technique
Ruhe (1956) used in Iowa when estimating that the Wisconsin surface had weathered

6800 years, the soil of the late-Sangamon surface at least 13,000 years, and the soil

22

related to the Yarmouth-Sangamon surface 105 years or more.

Relative age may also be determined by using quartz/ feldspar ratios (Dorronsoro
and Alonso, 1994). Dorronsoro and Alonso (1994) concluded soils of different ages
would have different quartz/ feldspar ratios under conditions that hold other soil forming
factors constant. The soil with the larger ratios, indicating more depletion of feldspar, is
the older soil, thus providing relative age.

Quartz/feldspar ratios also serve as a relative weathering indicator (Ruhe, 1956;
Nesbitt and Markovics, 1997). Ruhe (1956) noted that a comparison of the weathering
ratios for each site in his study indicated the intensity of weathering was greatest in the
older surface. The older surface had ratios with higher values, which suggested there
were greater amounts of resistant minerals relative to less resistant minerals. These
conclusions indicate the quartz/feldspar ratio could be used as a relative weathering
indicator.

Quartz/ feldspar ratios have also been used to determine parent material uniformity
due to the stability of quartz and abundance of both quartz and feldspar (Wilding et al.,
1977; Osei, 1992).

Lithological discontinuities can be detected using quartz/feldspar ratios (Price et
al., 1975; Osei, 1992). Price et a1. (1975) were able to distinguish the boundary between
an overlying loess and the underlying residuum by constructing a depth vs.
quartz/ feldspar ratio plot for three proﬁles in their research area. They noted there was a
fairly uniform quartz/feldspar ratio to a depth of about 100cm at which point there was a
large increase in the ratio. They concluded this large increase indicated a lithological

discontinuity.

23

CHAPTER 4

Spodosols

Podzolization

Podzolization is the pedogenic process by which Fe, Al, and organic carbon are
translocated down the soil proﬁle from the O, A and E horizons to an illuvial horizon.
The driving force behind this process is the percolation of acidic water through the soil
proﬁle. Inﬁltrating water becomes acidic as it passes through the decaying leaf litter of
the O horizon. This process produces Podzol or Spodosol proﬁles (DeConinck, 1980;
Buurman and van Reeuwijk, 1984; Schaetzl and Isard, 1991; Mokma and Evans, 2000;
Schaetzl, 2002).

The podzolization process can be summarized in the following four steps: (1)
organic material on the surface decays and fulvic acids are produced which are carried by
the water as it percolates through the proﬁle, (2) organic matter then complexes with and
mobilizes F e3 + and Al3+ cations, (3) these mobile cations are translocated from the O, A
and E horizons to the B horizon where (4) the Fe and A1 are deposited (Mokma and
Evans, 2000).

The fulvic acids, percolating through the soil with the water, acidify the upper
horizons. This acidic environment causes intense weathering of the minerals, which
release cations; of particular interest are Fe and Al. The Fe and A1 are then translocated
down the proﬁle (Mokma and Evans, 2000). There are two current theories to explain
the translocation of Fe and Al during podzolization. One theory states that the fulvic

acids chelate the Fe and A1 to form organometallic complexes that are then translocated

24

in percolating water. Once the complexes reach the water table, or a less permeable
layer, microbes break down the fulvic acid thus causing the precipitation of the Fe and Al
(DeConinck, 1980; Buurman and van Reeuwijk, 1984). The other theory is the “proto-
imogolite” theory. This theory was developed by researchers who observed that Fe and
Al exist in amorphous, inorganic compounds such as imogolite and allophane.

According to this view, the Fe and A1 are translocated into the B horizon as proto-
imogolite ﬁrst, and then organic matter precipitates onto the Fe and Al (Farmer et al.,
1980; Farmer, 1982; Anderson et al., 1982). Some aspect of each theory, and possibly a
combination of the two, might better explain the translocation of Fe and A1 during
podzolization rather than either theory independently (Wang et al., 1986; J akobsen, 1991;

Schaetzl, 2002).

Factors Contributing to Podzolization

Several factors contribute to the rate of podzolization. These factors include: (1)
climate, especially snowmelt inﬁltration, (2) parent material texture, (3) parent material
composition, and (4) type of vegetation in the region.

Climate affects the rate of podzolization by determining not only the amount of
precipitation, but just as importantly, it determines whether or not that precipitation
inﬁltrates. Schaetzl and Isard (1996) observed that regions that received little snow
during the winter had a slower rate of podzolization than regions that received enough
snow to accumulate a deep winter-long snowpack. The soil under deep snowpack is
insulated and does not freeze often, unlike soil that has little snow accumulation (Isard

and Schaetzl, 1998). The insulated soil is impacted by a high amount of inﬁltration of the

25

snowmelt during spring, due to the quantity of meltwater available and the permeability
of the non-frozen soil. The frozen soil does not allow for inﬁltration during spring until it
has thawed, by which time there may be little or no snowmelt remaining to enter the
proﬁle (Schaetzl and Isard, 1996). Because long, continuous inﬂuxes of meltwater
contribute to the rate of podzolization, climate is an important factor.

Parent material texture is a signiﬁcant control on podzolization. The best
expressed Spodosol proﬁles are in coarse-textured soils (Gardner and Whiteside, 1952;
Messenger et al., 1972; Vance et al., 1986; Schaetzl and Isard, 1991). The coarse texture
results in more rapid permeability and higher inﬁltration rate of the acidic water.
Podzolization is inhibited by high clay contents. Mokma and Evans (2000) stated that
clay might slow or even stop the translocation of Fe and Al by adsorbing the Fe and Al
cations, thus making them immobile. Due to this, the E and B horizons will not thicken
quickly and the B horizons will redden slowly. Another way clay interrupts the process
is by weathering so quickly in the acidic water that it releases more cations than the
system can handle. The few chelating acid molecules get overwhelmed and immobilized,
stopping the translocation. For these reasons, clay in the soil parent material must be
minimal or weathered extensively by acidic water prior to podzolization.

In the best expressed Spodosols, the parent material is sandy and composed
primarily of quartz with lesser amounts orthoclase, plagioclase, pyroxene, amphibole,
mica, and opaque minerals (Haile-Mariam and Mokma, 1995; Mokma and Evans, 2000).
Carbonate minerals are notably few in the parent material. Any carbonate minerals
present will react with the percolating water to neutralize the acid and slow the rate of

podzolization (Mokma and Evans, 2000). Therefore, most of the carbonates must be

26

weathered and leached from the system for podzolization to progress.

Vegetation also contributes to the rate of podzolization. Schaetzl (2002) noted
that certain types of coniferous vegetation, such as pine and hemlock, accelerate
podzolization. The leaf-litter under conifers is composed of needles, which are more acid
than the leaf-litter under broadleaf trees. Decay of the conifer needles produces more of
the acids needed for podzolization. For this reason, Spodosols are associated with

coniferous and coniferous-deciduous forests.

The Spodosol Proﬁle

Spodosols have four major horizons (Mokma and Evans, 2000). The O and/or A
horizon is dark brown to black due to the high organic content. The process of
podzolization leaves a gray to ashy white eluvial zone, or E horizon, from which Fe and
Al have been removed. The B horizon is below the A horizon and above the B horizon,
which is a dark reddish brown illuvial zone where the Fe and A1 have accumulated. The
illuvial zone is a spodic (Bs or Bhs) horizon. There may be several B subhorizons; those
directly below the B horizon are usually reddest and darkest becoming more yellow down
proﬁle. The underlying C horizon usually has an even more yellow hue than the lowest
B horizon.

As a Spodosol develops, the E horizon becomes whiter while the Bs or Bhs
horizon continues to darken and thicken. The POD index is a numerical index of
Spodosol development based on this pathway of development (Schaetzl and Mokma,
1988). This number is calculated using the differences of Munsell hues between the

eluvial and illuvial horizons. Due to the requirement of an eluvial horizon being present

27

in the soil, a POD index cannot be calculated for a pedon lacking an E horizon. If both
eluvial and illuvial horizons are present, redder hues in the B horizons, and more B
subhorizons, will result in a higher POD index. A higher number indicates a more

strongly developed Spodosol.

Extractable Iron and Aluminum in Spodosols

Iron and aluminum exist in soils in various pedogenic forms. When identifying a
spodic horizon, the Fe and Al content must be determined because the horizon is
characterized by the accumulation of translocated Fe and A1 (McKeague and Day, 1965;
Soil Survey Staff, 1999). Three extractants are commonly used in this determination.
Sodium citrate-dithionite (CD) is used because it primarily extracts “free” Fe, denoted by
Fed, from pedogenic iron oxides, and “free” Al, denoted by Ald, from aluminum oxides
(Birkeland, 1999; Weisenbom, 2001). The term “free” is used to describe compounds
having a single species of coordinating cation (Jackson et al., 1986; Weisenbom, 2001).
The term is also used for an iron oxide that can be reduced and dissolved by a dithionite
treatment (Anonymous, 1996; Weisenbom, 2001). Acidiﬁed ammonium-oxalate (AAO)
extracts “active” Fe, denoted by Fee, and “active” Al, denoted by A10, from noncrystalline
hydrous oxides. The term “active” is given to oxides that are small in size, have a high
surface area, and a high degree of reactivity (Loeppert and Inskeep, 1996; Weisenbom,
2001). AAO also extracts Al from noncrystalline, or amorphous, aluminosilicates such
as allophane and imogolite and other weathering products (Farmer et al., 1983). F eO
selects Fe mainly from ferrihydrite (F65H0804H20), and is such a good test that its value

may be multiplied by 1.7 to estimate the amount of ferrihydrite present (Parﬁtt and

28

Childs, 1988; Birkeland, 1999). The third extractant is sodium pyrophosphate (PP). It
extracts Fe, denoted by Fep, and Al, denoted by Alp, from organically-bound complexes
and, to a lesser degree, noncrystalline hydrous oxides. It has been reported and long
assumed that these three extractants do not extract Fe and Al from crystalline minerals
(Mehra and Jackson, 1960; Ross and Wang, 1993; McKeague et al., 1971).

In addition to analyzing the extraction data directly, various calculations and
formulations may be made from the raw data. These calculations indicate the form of Fe
and Al that is found in each horizon (e. g., crystalline or noncrystalline, complexed or
not). The ratio of AAO / CD, or “activity ratio” (Blume and Schwertmann, 1969),
measures the amount of pedogenic Fe that is amorphous, plus fen'ihydrite (Birkeland,
1999). It approximates the relative crystallinity of “free” oxides. A low value of Feo/

F ed indicates Fe oxides with a high degree of crystallization or aging (Blume and
Schwertmann, 1969; Weisenbom, 2001), while a high value from this ratio indicates Fe is
in a more “active” noncrystalline form that is complexed either organically or
inorganically (Weisenbom, 2001 ).

The molar ratio of Fep / A1p is the difference between organically complexed,
noncrystalline Fe and A1 (Barrett, 1995; Gustafsson et al., 1995). Ratio values less than
one signify there is more organically complexed, noncrystalline Al than Fe in that
horizon (Weisenbom, 2001), while a ratio of greater than 10.5 could indicate that
imogolite is not likely to form due to insufﬁcient Al (Huang, 1991).

The noncrystalline, organically complexed forms of Fe and Al versus the
noncrystalline, inorganic complexed forms of Fe and Al in a horizon may be estimated by

(Fe0 — Fep) / Fep and (A10 — Alp) / Alp (Wang et al., 1986; Wang and Kodama, 1986;

29

Schaetzl, 1992). Higher calculated values signify that inorganic complexes contribute
more than organic complexes to the state of Fe and Al in the soil. Conversely, lower
calculated values emphasize the role of organic complexes over inorganic complexes.
The relative amount of crystalline “free” oxides compared to noncrystalline “free”
oxides may be determined by the difference between the CD value and the AAO value
(McKeague et al., 1971; Parﬁtt and Childs, 1988). When the CD values are larger than
the AAO values, crystalline oxides are present (Parﬁtt and Childs, 1988). Thus, this
calculation can, under favorable circumstances, be used to estimate the Fe in goethite,
lepidocrocite, and hematite (Parﬁtt and Childs, 1988; McKeague et al., 1971). The value
is negative when the AAO extracts a large amount of “active” noncrystalline oxides and
organically bound noncrystalline forms, and which exceeds the CD extracted “free”

value. Thus, a low, or negative, value indicates that the “free” oxides are predominantly

in noncrystalline form (McKeague et al., 1971).

30

CHAPTER 5

Study Area

Site Location and Description

The study area for this research was located in the northern Michigan county of
Kalkaska (Figure 10). Three of the four sites chosen for this study were within Blue
Lake Township; the fourth site was in neighboring Bear Lake Township (Figure 11). All
of the sites were located on the sandur east of the Outer Port Huron moraine (Figure 12).
Three soil series are present at the four sampling sites; these are the Grayling, Rubicon,

and Kalkaska series.

Site Selection Criteria

The four sites used in this research were chosen based on several criteria intended
to constrain the soil-forming factors, which are: climate, organics or biota, relief or
topography, parent material, and time (Jenny 1941, as cited in Birkeland, 1999). Two of
these factors (climate and organics) are active (dynamic, in that they can change with
time) in the soil forming process and really cannot be constrained. Even though these
factors cannot be fully constrained, their effects were constrained, and between—site
variation in these factors was held to a minimum by choosing sites that were in close
proximity to one another. The passive factors (relief, parent material, and time) were
accounted for by holding them as constant as possible among the sites.

Due to the higher elevations of the moraine, this region is within a major lake

effect snowbelt (Schaetzl, 2002). As previously stated, in order to minimize the effects

31

Kalkaska County

 

 

Figure 10. A map of counties in lower Michigan with Kalkaska County highlighted.

 

 

 

 

 

 

 

 

Silt‘ l
. ‘ T ' r? t ' 3
Rapid ' . T Cdld I Sitcl' "1
iii 1 . w . . '
atwat r River I I . .31“ We
MW .- if - ..Tb bwnshib
» . . at, r at. I- i l- i A 51103
i I i; i i I - i l l ..

MW'Qurokaalka TTYTY ii“?

Township Township 3m 4‘
VII “..1 ‘ntrN‘mMnme:

t i . . i t .. i .‘t. I
Boardman I Orange: 2 bliver ' ; i : ‘
prnshrp 7 Township”? Tawnship ' ”
Springﬁeld barlleld , T
Tovtnship . frownship I

. . N

 

 

 

 

Figure 11. A map of Kalkaska County townships with research sites indicated. Sites 1,
2, and 3 are in Blue Lake Township, while Site 4 is in Bear Lake Township.

32

 

 

 

 

 

Z

 

Figure 12. Map of the morainic system (shown in black) of Michigan with a portion of
the Outer Port Huron moraine and the study area for this research in north-central

Michigan is indicated in red. (This map is public domain, per a personal communication
with Dr. Randy Schaetzl.)

33

of climate, the distance between sites was kept to a minimum.

The soil-forming factor of vegetation could not be fully constrained. Jack pine
(Pinus banksiana) was primarily found on the Grayling, red (P. resinosa) and white pine
(P. strobus) were found on the Rubicon, and hardwood, such as sugar maple (Acer
saccharum) and yellow birch (Betula allegheniensis), was found on the Kalkaska sites.
The type of vegetation present could be due to the soil’s capacity to hold water, which
increases with more strongly developed soils (Shetron, 1974; Mokma and Evans, 2000).
It could also be due to the history of ﬁre (Schaetzl, 1994, 2002; Barrett, 1997).
Vegetation-soil co-occurrences in the region are widespread (Mokma and Vance, 1989).
The vegetation patterns are temporal and shift, so it cannot be deﬁnitively stated whether
vegetation is acting as a dependent or independent variable in soil development in this
region (Schaetzl, 2002).

Relief was constrained by choosing sites where the ground slope at the pit was 0-
3%. Due to the relief, all the sites are well drained with no water table.

Differences in the parent material were constrained by choosing sites that were all
located on the sandur east of the Outer Port Huron moraine (Figure 12). This moraine
was formed by the Lake Michigan Lobe, whose ice was mainly from the Superior east
area of the Canadian Shield and ﬂowed southwestward across a region north of Lake
Huron, then into southern Michigan (Figure 13) (Dworkin et al., 1985). The sediments in
this area are almost all glacioﬂuvial sand or gravel with some glaciolacustrine silt and
clay; till is very rare (Blewett et al., 1993). All of the sediments were deposited by
glacial meltwater during the retreat of the Port Huron readvance. Some of the sediments

may have been reworked by meteoric runoff and post-glacial streams (Schaetzl, 2002).

34

 

. .r
K Canadian Sh' Id
Lake Superior L
Superior '
._ /

. . “A, ._ 9 Grenville
Mi I \5
a

/
\ / // Lake Erie /

\ / '
i'l\ l by/

 

 

 

 

 

 

 

Figure 13. Map showing the ice movement over the Great Lakes region during the
Wisconsin glacial stage. The arrows indicate ice ﬂow and the dashed lines indicate ice
lobe boundaries. The Outer Port Huron moraine was deposited by the Lake Michigan
Lobe, which ﬂowed southward from the Superior east area of the Canadian Shield
(modiﬁed from Dworkin et al., 1985).

35

By choosing sites on the same deposit and in close proximity to one another, age was also
constrained. The Outer Port Huron moraine contains some of the youngest Late
Wisconsin deposits in southern Michigan (Blewett, 1990). The age of the morainic
system was estimated at 12,960 i 350 yr B.P. (Blewett et al., 1993; Schaetzl, 2002).
Based on the above-mentioned criteria, four soils from the Spodosol order were
chosen (Figure 11). Site 1 is a soil from the Kalkaska series, which is a strongly
developed Spodosol. Sites 2 and 3 are from the Rubicon series, which is less deve10ped
than the Kalkaska. Site 2 is a strong Rubicon, while site 3 is a weak Rubicon. Site 4 is

from the Grayling series and is the least developed of the soils sampled.

36

CHAPTER 6

Methods

Sample Preparation

The samples that were used in this research were gathered from four soil pits dug
in Kalkaska County (Figures 10 and 11). The pits were dug with a backhoe, and the soils
classiﬁed based on ﬁeld observations. Approximately 4kg of soil were taken from each
identiﬁed pedogenic horizon. The samples were air dried and sieved through an 8mm
sieve to remove gravel and larger organics such as twigs and leaves. The samples were
then split to obtain 1kg of sample. A soil splitter was used to ensure the sample used in
testing was representative of the whole. The grain size distribution was determined for
each pedon by Laura Beeman using particle size analysis, following standard laboratory
procedures (Soil Survey Division Staff, 1993) (Schaetzl, 2002). The soil pH for each
horizon was measured with a model #720A Orion pH meter using a 2:1 slurry of distilled
water to soil. The clays were dispersed by soaking and shaking the sample overnight in a
solution of sodium hexametaphosphate ([NaPO3]6-Na2CO3) and distilled water. The
sample was then wet sieved through a 53 um sieve. The samples were then placed in a
bath of 3% sodium hypochlorite (NaOCl) to remove organic matter, rinsed with distilled

water, and dried in an oven at 105°C. The samples were then sieved to obtain the ﬁne

sand, or 125 um-250um, size fraction; this was the size fraction used for this research

because the heavy minerals are concentrated in this size fraction (following Locke, 1979;

Hall and Horn, 1993).

37

Homblende Denticulation Amplitude Determination

The heavy fraction of the ﬁne sand, which includes hornblende, was separated
from the cleaned sample by heavy liquid separation. Approximately 42g of each sample
were suspended in approximately 50mL of sodium polytungstate
(3Na2WO4-9WO30HZO), which had been mixed with distilled water (following the
manufacturer’s instructions) to obtain a speciﬁc gravity of 2.95. Homblende, having a
speciﬁc gravity of 2.9 — 3.3 (Blackburn and Dennen, 1988), sank to the bottom of the
separatory funnel and was easily extracted.

The denticulation amplitude was then determined for 300 hornblende grains from
each horizon. This quantity was chosen because of the high conﬁdence limit it provides
while allowing a large number of samples to be analyzed in a timely manner (Van der
Plas and Tobi, 1965; Pettijohn et al., 1973). The hornblende grains were placed on a
glass slide and viewed under a petrographic microscope at 200x. Immersion oil with a
refractive index of 1.5 was used to aid in the viewing of the grains at this magniﬁcation
(hornblende has a refractive index of 1.679 - 1.698 (Klein and Hurlbut, 1999)). To
measure the denticulation amplitude, a graded ocular was used in which one scale

division equals 4pm.

Homblende Classiﬁcation Scheme

Based on amplitude, the grains were assigned a classiﬁcation that was modeled
after the classiﬁcation scheme of Locke (1979) (Figure 14). Grains that showed no
apparent etching were assigned Class 0 (Figure 15). Class 1 indicates etching was

present but was less than 4pm in depth (Figure 16). Class 2 indicates the grains with

38

etching greater than 4pm but less than 8pm (Figure 17). Class 3 had etching greater than
8pm but less than 12pm, and so on (Figures 18 — 23). For statistical analysis the mean
value of the depth of etching in each class was used. Class 0 was assigned a value of 0.
Class 1 was assigned a value of 2pm; in class 2 the value was 6pm; and in class 3, 10pm;
and so on. These values were then used to calculate a weighted average of denticulation
amplitude for each horizon. This was accomplished by multiplying the class value times
the number of grains in that class, summing these products, then dividing that sum by the

total number of grains counted, which was 300 per horizon.

Unexpected Etching Features

Unexpected etching features were observed while examining grains from the A
and E horizons of the Kalkaska, strong Rubicon, and weak Rubicon pedons. On the
average, 1 - 3 grains from each of these horizons had dissolution holes through the
middle of them (Figure 24). These holes may reﬂect hornblende dissolution around
inclusions, and/or removal of inclusions. Other grains ﬁ'om these same horizons, except
for the E horizon of the weak Rubicon pedon, had dissolution features that caused the
grains to appear as though they were fracturing (Figure 25). Dissolution at multiple
dislocations arrayed in a plane could also produce this corrosion form: side-to-side
coalescences of en echelon etch pits is known to form such features (Bemer and Schott,
1982; Velbel, 1989). Several grains exhibited both the dissolution hole and fracture
feature (Figure 26). A method for measuring these features was not taken into
consideration when designing this study simply because of the lack of knowledge at the

time with regard to such features. The research was designed for measurements to be

39

 

Figure 14. Diagrammatic etching scale showing progress of etching on broken (left) and
rounded (right) faces. Depth of etching is shown by indentation in grain margin and is
highlighted by shadow (Figure 2 in Locke, 1979).

40

 

20pm

Figure 15. Class 0. Petrographic micrograph of a hornblende grain with no apparent
etching. Image captured with a Pixera digital imaging system.

 

20pm

Figure 16. Class 1. Petrographic micrograph of a hornblende grain with etching between
0pm — 4pm. Image captured with a Pixera digital imaging system.

41

 

—'

 

20pm

Figure 17. Class 2. Petrographic micrograph of a hornblende grain with etching between
4pm — 8pm. Image captured with a Pixera digital imaging system.

 

 

20pm

Figure 18. Class 3. Petrographic micrograph of a hornblende grain with etching between
8pm — 12pm. Image captured with a Pixera digital imaging system.

42

 

20pm

Figure 19. Class 4. Petrographic micrograph of a hornblende grain with etching between
12pm - 16pm. Image captured with a Pixera digital imaging system.

 

20pm

Figure 20. Class 5. Petrographic micrograph of a hornblende grain with etching between
16pm - 20pm. Image captured with a Pixera digital imaging system.

43

 

 

20pm

Figure 21. Class 6. Petrographic micrograph of a hornblende grain with etching between
20pm — 24pm. Image captured with a Pixera digital imaging system.

 

 

20pm

Figure 22. Class 7. Petrographic micrograph of a hornblende grain with etching between
24pm — 28pm. Image captured with a Pixera digital imaging system.

44

 

20pm '

Figure 23. Class 8. Petrographic micrograph of a hornblende grain with etching between
28pm - 32pm. Image captured with a Pixera digital imaging system.

45

 

20pm

Figure 24. Petrographic micrographs of weathered hornblende grains with holes. Arrows
point to holes. Images captured with a Pixera digital imaging system.

46

 

    

20pm

 

Figure 25. Petrographic micrographs of weathered hornblende grains with fractures.
Images captured with a Pixera digital imaging system.

47

 

 

20pm

Figure 26. Petrographic micrographs of weathered hornblende grains with both holes and
fractures. Images captured with a Pixera digital imaging system.

48

taken of the denticulated edge of the hornblende in order to determine amplitude and
subsequent classiﬁcation. However, when the amplitude of the denticulated edge was
measured and the classiﬁcation was determined, these grains were most likely placed in a
class that underestimated the extent of their weathering. These grains were assigned, on
the average, to classes 1-3 based on denticulation amplitude of the grain margin; yet, the
amount of dissolution that had to occur in order to produce holes and fracture features is
greater than the classiﬁcation 1-3 would indicate. However, the decision was made to
classify these grains based on the original research design. Due to the small number of
grains encountered with either of these features, it is doubtful that the results are skewed
by virtue of using this classiﬁcation scheme. Perhaps future studies should take these
dissolution features into consideration when designing the research in order to utilize or
develop a classiﬁcation scheme that would better accommodate grains with these unusual

features.

Quartz/Feldspar Ratio Determination

The degree of mineral weathering was determined by quantifying the
quartz/ feldspar ratios in each of the horizons. Approximately 25g of the cleaned, ﬁne
sand size fraction of the sample was used from each horizon to prepare thin sections. The
samples were impregnated with epoxy and allowed to harden to produce a cube that was
then sliced, mounted on microscope slides, and ground to a thickness of approximately
30pm. The thin sections were stained with Alizarin red to help distinguish plagioclase
feldspar from quartz. However, both plagioclase and K-feldspar were counted as feldspar

for the ratio determination. Point counting was performed on these thin sections to obtain

49

quartz/ feldspar ratios. Only quartz and feldspar grains were counted, other minerals and
void spaces were disregarded. A total of 300 grains were counted per horizon. Again,
the quantity of 300 was chosen for the optimum combination of accuracy and speed (Van

der Plas and Tobi, 1965; Pettijohn et al., 1973).

Iron and Aluminum Extractions

The three extracts that were used in this research are: sodium citrate-dithionite
(CD), acidiﬁed ammonium-oxalate (AAO), and sodium pyrophosphate (PP). Extractions
were performed on the illuvial and eluvial horizons only; the C horizons were not
analyzed. The methods, which were used to perform these extractions, were from
Weisenborn (2001), which were modiﬁcations of methods by Ross and Wang (1993) and
Loeppert and Inskeep (1996). The modiﬁcations were necessary to obtain a sufﬁcient
amount of supernatant to yield detectable and measurable amount of Fe and Al when
analyzed by atomic absorption spectrophotometer (AAS).

The supematants were analyzed for extractable Fe and Al on the ﬂame
apparatus of the Perkin-Elmer 5100 PC AAS in the Geochemistry Laboratory at
Michigan State University. Weisenborn (2001) developed the method for this analysis by
modifying the methods from the Soil Survey Laboratory Staff (1996) and Anonymous

(1989). The results were reported in percentages of the analyte cation.

Laboratory-Produced Etching Features

Two points around which this research was designed involve chemical analysis of

soils to determine Fe and A1 accumulation, and measurement of hornblende denticulation

50

amplitude. It was thought that CD could be employed in both of these analyses. CD
could be one of the extractants used to determine Fe and Al accumulation, and it could be
used to “clean up” the hornblende grains to prepare them for visual examination (Jackson
et al., 1986). Jackson et al. (1986) describe CD as a good method of Fe removal because
it is a strong reducing agent, and has a chelating agent to sequester F e2+ and F e3 + ions.
The CD system uses sodium citrate as a chelating agent, sodium dithionite (N a23204) as a
reducing agent, and sodium bicarbonate (NaHCO3, pH 7.3) as a buffer.

During the design of this thesis, the question arose as to the degree to which CD, a
strong reducing agent, would attack the hornblende, a ferromagnesian mineral. If CD
would reduce and remove any crystalline Fe, it could duplicate or enhance the natural
etch-features that were going to be measured visually. Using a petrographic microscope,
it is possible to discern and measure features that are approximately 1pm in length. Due
to this level of detail, and the importance to this study of accurate amplitude
measurements, the question of laboratory-produced features on hornblende grains from
soils could not be ignored. In order to answer this question fully it was decided that CD,
along with the other laboratory pre—treatments, would be tested on hornblende grains to
determine whether or not they would produce any artiﬁcial alteration features.

Homblende from the Michigan State University mineralogy teaching collection
was used as the control to test the different pre-treatments. The hornblende was ﬁrst
ground with a mortar and pestle then sieved to obtain a ﬁne sand grain size fraction of
125 um - 250nm. The sample was then divided into thirds for treatment. The ﬁrst third
of the sample was not treated. The remainder of the sample was exposed to a clay

dispersant ((NaP03)6 + NazCO3) in order to follow the standard sample cleaning process

51

(Soil Survey Division Staff, 1993). The sample was slowly shaken over night in the
solution then rinsed with distilled water and dried in an oven at 85°C. The sample was
then split in half for the remaining two treatments. One half was treated with sodium
hypochlorite (NaOCl) to simulate the removal of organic matter. Bleach, with 6%
available NaOCl, was mixed with distilled water to a concentration of 3% available
NaOCl. The hornblende grains were placed in the bleach for approximately 8 hours,
rinsed with distilled water, and then dried. The ﬁnal portion of the sample was treated in
sodium citrate-dithionite (Na3C6H507 + IOHZO + Na2S204). The hornblende grains were
placed in the solution and shaken for 12 hours then placed in the refrigerator for 6 hours.
The grains were then rinsed with distilled water and dried.

A J EOL J SM-6400V scanning electron microscope (SEM) was used to view the
hornblende grains to search for alteration features caused by the pre-treatrnents. The
control hornblende grains, which were untreated, showed smooth freshly exposed
surfaces and no etching features (Figures 27 and 28). The grains that were treated with
sodium hypochlorite also had smooth surfaces (Figures 29 and 30). There were no signs
of alteration or weathering features on these grains. The grains that were treated with
sodium citrate-dithionite all contained etch pits and laboratory-produced alteration
features (Figures 31 and 32), some of them very large (Figure 33). The edges of the
grains were not sharp, but showed signs of alteration (Figures 34 and 35). The surfaces
on these grains were not smooth, but showed dissolution features (Figures 36 and 37).
The hornblende grains treated with CD were the only grains to show any sign of

laboratory-produced alteration.

52

 

200nm

Figure 27. Scanning electron micrograph of an untreated hornblende grain. Image taken
at the Center for Advanced Microscopy at Michigan State University by Leslie Mikesell
and Ewa Danielewicz.

 

Figure 28. Scanning electron micrograph of the indicated area on Figure 27 showing
smooth surfaces with no etching features.

53

 

 

200nm

Figure 29. Scanning electron micrograph of a hornblende grain treated with sodium
hypochlorite. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

. ,L‘ .
”ﬂirting

’ 392'?“ .‘l

 

 

20pm

Figure 30. Scanning electron micrograph of the indicated area on Figure 29 showing
smooth surfaces with no etching features.

54

Area 1

Area 2

 

Figure 31. Scanning electron micrograph of a hornblende grain treated with sodium
citrate-dithionite. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

 

Figure 32. Scanning electron micrograph of the indicated ‘Area 1’ on Figure 31 showing
laboratory-produced alteration features.

55

 

20pm

Figure 33. Scanning electron micrograph of the indicated ‘Area 2’ on Figure 31 showing
large laboratory-produced alteration feature.

56

”I. {13' ii

0
I
I, II“ All“
{MN

9m i
it
iiilll'l-""

i
.jll

 

 

200nm

Figure 34. Scanning electron micrograph of a hornblende grain treated with sodium

citrate-dithionite. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

,_.
/.v
’.' I: I”.

,, err-2y:

 

Figure 35. Scanning electron micrograph of the indicated area on Figure 34 showing
laboratory-produced alteration at the edges.

57

 

Figure 36. Scanning electron micrograph of a hornblende grain treated with sodium
citrate—dithionite. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

 

5 mm

Figure 37. Scanning electron micrograph of the indicated area on Figure 36 showing
laboratory-produced alteration on surfaces.

58

The laboratory-produced alteration features raised two major concerns about the
results of this study. The ﬁrst concern was whether the laboratory-produced features are
discemable at the petrographic microscope scale. This was a problem because the
petrographic microscope would be used to classify the hornblende according to a scheme
based on the denticulation amplitude. If the laboratory-produced features were mistaken
for natural features and then measured as such, the grain would be classiﬁed incorrectly.

Due to these laboratory-produced alteration features, I was also concerned was
about the validity of the data from the atomic absorption spectrophotometry (AAS).
These data could be skewed by mineral dissolution, which would release Fe and Al into
the supernatant for measurement. The values obtained would not accurately reﬂect the
“free” Fe and A1 but would also measure the Fe and Al from dissolved hornblende.

It was clear from Figures 31 - 37 that hornblende dissolved in the presence of CD
and released Fe and Al into solution. It had to be determined how much Fe and Al was
released per a given volume of hornblende, and whether or not it was a large enough
amount to skew the results from the AAS.

In order to estimate how much Fe and Al was released, a representative
hornblende formula was adopted. The hornblende that is found in Michigan soils was
glacially transported from a source area in Canada (Dworkin et al., 1985). There are
several studies that provide chemical analyses of hornblende from the Grenville Orogen
of Ontario, Canada, which is part of the source area for the sediments in this study
(Cureton et al., 1997; Cosca et al., 1991). Averaging the values provided, a
representative hornblende formula was derived:

K01 INaO.42C31.83 (Mgr .47P‘eﬁostCﬁz.06'1‘i0.07M110.02/3\10.75)Vi (A11.69Si6.31)iv 022F0.02(0H)1.98

59

From this average formula, the molecular weight of hornblende was calculated to be
906.28 g/mol. With a speciﬁc gravity of 3.1 g/cm3 (Klein and Hurlbut, 1999), the molar
volume of hornblende is 292.35 cm3/mol. Iron in this formula is 149.68 g/mol, or 16.5%
of the total weight. It occupies a volume of 48.28 cm3/mol. Aluminum is 65.84 g/mol, or
7.3% of the total weight, and occupies 21.29 cm3/mol.

The volume of hornblende that can be dissolved in the presence of CD, during the
time allotted in the methodology, had to be determined. Stereo pairs were made using
pictures from the SEM to estimate the depth of the laboratory-produced etch pits (Figures
38 — 41). The stereo pairs were made by taking two Polaroid pictures of the feature of
interest. The feature was centered on the screen, and then the ﬁrst picture was taken.

The stub was tilted 3 degrees then re-centered, and another picture was taken. By placing
the pictures side-by-side and using stereoscopes, the depth of etch pits was estimated. An
average depth for all etch pits viewed was 0.5 pm — 1pm.

Several assumptions were made before continuing with the calculations. First, it
was assumed, for the sake of calculation, that the grains were all spherical. This makes
the volume of the grain much easier to calculate. Also, due to the fact that the natural soil
grains of this thesis study are from glacial deposits that have been reworked by
meltwater, they tend to have a rounded, sub-spherical shape. The second assumption is
that the grains are all 200nm in diameter. This number was chosen because it is an
intermediate value for the ﬁne sand size fraction (125 um — 250nm) used in the study.
The ﬁnal assumption is that a shell lum thick was dissolved from each grain. One

micrometer is the largest value in the range of 0.5 pm — 1.0um, which was observed as

the average depth for laboratory-produced features. By using 1pm, the amount of

60

l» l" I l-' I'l

 

.‘V . ‘ ”a...“ m‘ .
E01. 1 SHTWMI , ‘ , 1 r 1:; U I", KB ,. in 1:71 LI .4: mm

Figure 38. Stereo pair 1. Polaroid pictures of scanning electron images were used to
estimate the depth of the laboratory-produced etch pits on the hornblende grains that were
treated with sodium citrate-dithionite. The feature in the left picture is centered and
straight; the right picture is tilted 3 degrees. Images were taken at the Center for
Advanced Microscopy at Michigan State University by Leslie Mikesell and Ewa
Danielewicz.

61

 

Figure 39. Stereo pair 2. Polaroid pictures of scanning electron images were used to
estimate the depth of the laboratory-produced etch pits on the hornblende grains that were
treated with sodium citrate-dithionite. The feature in the left picture is centered and
straight; the right picture is tilted 3 degrees. Images were taken at the Center for
Advanced Microscopy at Michigan State University by Leslie Mikesell and Ewa
Danielewicz.

62

I.” "j- [Ir I’ll

 

Figure 40. Stereo pair 3. Polaroid pictures of scanning electron micrographs were used
to estimate the depth of the laboratory-produced etch pits on the hornblende grains that
were treated with sodium citrate-dithionite. The feature in the left picture is centered and
straight; the right picture is tilted 3 degrees. Images were taken at the Center for
Advanced Microscopy at Michigan State University by Leslie Mikesell and Ewa
Danielewicz.

63

 

Figure 41. Stereo pair 4. Polaroid pictures of scanning electron micrographs were used
to estimate the depth of the laboratory-produced etch pits on the hornblende grains that
were treated with sodium citrate-dithionite. The feature in the left picture is centered and
straight; the right picture is tilted 3 degrees. Images were taken at the Center for
Advanced Microscopy at Michigan State University by Leslie Mikesell and Ewa
Danielewicz.

material added to solution was over-estimated to give an upper limiting value.

In order to estimate the amount of Fe and Al added to solution by the dissolution
of hornblende, several calculations were made. The molecular weight and the
assumptions stated above were used to calculate the total volume of a 200nm diameter
sphere of hornblende to be 2.35x10'6cm3 with a mass of 7.30x10'6g. When a 1 pm shell
was subtracted to simulate dissolution, the volume of the sphere dropped to 2.32x10‘6cm3
with a new mass of 7.19x10'6g. The volume loss was 3.6x10'80m3 and the mass loss was
1.12x10'7g. This was a 1.53% loss of both volume and mass.

In order to determine if this volume and weight loss would affect AAS values
from the soils, the percentage of hornblende in a soil sample also needed to be
determined. This percentage would be the portion of the soil that would dissolve in CD,
thus adding Fe and Al to solution. The percent of hornblende in the soil samples
collected for this thesis study was determined by performing grain counts (Table 1).
Using this percent value, calculations were made to determine the amount of Fe and Al
that could be added to solution from hornblende for each horizon. These calculated
values were then compared to the actual values of Fe and Al that were obtained by AAS
(Table 1). The amount of Fe and Al that could be in solution due to CD dissolution of
hornblende was negligible. According to calculations, only 3.6x10'8cm3 or 1.12x10'7g
would be lost when a 1pm shell of hornblende dissolved from a 200nm sphere. This
would release 1.8x10'8g of Fe and 8.0x10'9g of Al per 7.30x10'6g of hornblende. The
amount of Fe and Al that was measured using AAS was signiﬁcantly higher than the
amount of Fe and Al that could be added by the dissolution of hornblende. The AAS

measurements were one order of magnitude larger than the calculated values for 13 of the

65

Table l. Pedon Data for Laboratory-Produced Alteration Features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pedon Soil Sample Homblende Fe (%) in Fe (%) Al (%) in Al (%)
i‘lUtiLuu weight (%) in ﬁne dry sample due to CD dry sample due to CD
(g) grain sample AA value dissolution AA value dissolution
Kalkaska
A 1.54 no data 0.067 no data 0.015 no data
E 1.59 2.5 0.052 0.010 _ 0.013 0.004
Bhs 1.52 3.7 0.748 0.014 0.145 0.006
le 1.52 2.2 0.160 0.008 0.183 0.004
B82 1.53 1.5 0.054 0.006 . 0.059 0.003
Bw 1.50 no data 0.045 no data 0.045 no data
Strong Rubicon
A 1.56 no data 0.085 no data 0.027 no data
E 1.52 2.2 0.080 0.008 r 10.017 0.004 '
le 1.50 6.5 0.696 0.025 0.264 0.011 ‘
Bs2 1.53 4.7 0.328 0.018 ,_ 0.179 70.008
Weak Rubicon
A 1.60 no data 0.040 no data 0.017 no data
E 1.50 3.3 0.036 0.012 0.008 0005
B51 1.59 3.3 0.286 0.013 0.176 0.006
B82 1.64 4.0 0.049 0.017 0.065 0.007 3
Grayling
A 1.58 4.3 0.059 0.017 """ii‘jb‘fimwmdi'dtﬁmi
Bs 1.52 5.3 0.108 0.020 0.111 ,. 0.009 .5

n

 

66

Blue: Al comparisons

Yellow: Fe comparisons

Orange: value below valid detection limit ,

comparisons. It was two orders of magnitude larger for 6 of the comparisons. In the
remaining 6 comparisons, the AAS measurements were higher than the calculated values,
but not signiﬁcantly higher. Thus, I concluded that the Fe and Al from the sesquioxide
coatings on the soil particles were the dominant contributor to the measured values from
the AAS.

The original concern that needed to be addressed was whether the laboratory-
produced features are discemable under a petrographic microscope. This concern, in the
context of this thesis study, became moot. The grains used for denticulation amplitude
measurements were not cleaned of sesquioxide coating prior to examination. That was
deemed unnecessary because the coatings were thin enough that they did not affect the
measurements taken. However, for future studies, these laboratory-produced features are
a source of potential error that should be addressed. It has been reported that CD does
not extract Fe and Al from crystalline minerals (Mehra and Jackson, 1960; McKeague et
al., 1971; Ross and Wang, 1993). Ross and Wang (1993) stated that this treatment
removes the noncrystalline iron oxides, organically bound Fe and Al, as well as the ﬁnely
divided hematite, goethite, lepidocrocite, and ferrihydrite. Mehra and Jackson (1960)
tested CD against three other methods of iron oxide removal to determine its
effectiveness. The other three methods were: (1) sodium sulﬁde-oxalic acid (NaZS-
NH4Cl-oxalic acid), at a pH of 3.5-10 (Truog et al., 1937), (2) sodium dithionite in acid
system (Na2SZO4-HZO), at a pH of 3.5 (Deb, 1950), and (3) zinc-ammonium oxalate (Zn-
oxalic acid-NH4 oxalate), at a pH of 3.6 (Haldane, 1956). It was shown that these three
methods attack iron-rich silicate minerals. In their conclusion, Mehra and Jackson (1960)

said of CD extractant, “Above all, the treatment has almost no destructive effect on iron

67

silicate clay mineral.” My study, however, showed that CD did produce alteration
features on hornblende grains that were discemable and measurable under a petrographic
microscope. Therefore, it is not recommended to clean hornblende samples with CD

when the objective is to measure denticulation amplitude or other etching features.

68

CHAPTER 7

Results

Soil Classiﬁcation

The samples from the four sites were analyzed in the laboratory for various
physical properties (Table 2) and chemical properties (Table 3). Using these data, the
pedons were then classiﬁed based on Keys to Soil Taxonomy (Soil Survey Staff, 1998).
Site 1, Kalkaska, was classiﬁed as a sandy, mixed, frigid Typic Haplorthod (Schaetzl,
2002) (Figures 42 and 43). This site was the most strongly expressed soil studied, and
the only pedon with a Bhs horizon. Site 2, strong Rubicon, is a sandy, mixed, frigid Entic
Haplorthod (Schaetzl, 2002) (Figures 44 and 45). Site 3, weak Rubicon, is a sandy,
mixed, frigid Typic Haplorthod (Schaetzl, 2002) (Figures 46 and 47). Site 4, Grayling,
was classiﬁed as a sandy, mixed, ﬁigid Entic Haplorthod (Schaetzl, 2002) (Figures 48
and 49). This was the least expressed soil studied, and it lacked an E horizon. Based on
the laboratory tests, the Grayling soil classiﬁed as a Rubicon. The site, however, was
mapped and ﬁeld classiﬁed as a Grayling. So, in order to minimize confusion, I will
continue to refer to site 4 as a Grayling soil. There was a trend in soil development that
increased from southeast to northwest with the most strongly developed soil, the
Kalkaska, being the farthest northwest. That site is in close proximity to Starvation Lake,
where local residents report it is not uncommon to have 60+ cm of snow on the ground
during winter. The snOWpack decreases slightly to the southeast where the lesser-

developed soils are located (Schaetzl, 2002).

69

Table 2 . Physical Properties of Pedons.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Horizon Depth Munsell Grain Size T C
(cm) Color C f IVC le s [Ms lFs WFSIS [C
(moist) per cent by weight
Kalkaska POD Index of 1.5
Oi 0-3
A 3-16 7.5YR2/0 0.0 0.7 13.0 52.8 18.5 3.3 11.6 0.0 sand
E 16-24 5YR4/3 0.0 0.5 15.5 54.1 14.8 2.6 12.4 0.0 sand
Bhs 24-31 5YR2.5/2 0.0 1.7 12.1 50.2 17.1 2.2 13.2 3.5 c.sand
le 31-58 7.5YR4/6 0.0 0.8 11.0 58.0 21.9 3.4 4.9 0.0 sand
Bs2 58-85 5YR 5/8 4.0 1.9 16.0 62.5 16.2 0.8 2.5 0.0 sand
Bw 85—128 10YR 4/6 4.0 0.8 7.3 59.3 29.3 1.5 1.8 0.0 sand
2E/Bt 128-160+ 10YR 4/6(E) 8.0 1.1 14.1 58.7 21.6 2.6 2.0 0.0 sand
7.5YR 4/4(Bt)
Strong Rubicon POD Index of 0
Oi 0-5
A 5-9 7.5YR 3/2 2.0 1.4 14.3 52.4 19.0 3.2 9.7 0.0 sand
E 9-18 7.5YR4/2 2.0 1.3 14.5 52.4 18.8 3.3 8.2 1.4 sand
le 18-28 5YR4/6 6.0 1.8 16.0 50.2 18.0 4.5 8.8 0.7 sand
BS2 28-48 7.5YR 4/6 6.0 2.3 11.7 51.6 22.2 5.7 6.5 0.0 sand
BC 48-79 10YR 5/6 8.0 2.6 16.5 60.4 17.9 0.7 2.0 0.0 sand
C 79—170 10YR 6/4 8.0 1.4 14.3 59.4 21.1 2.4 1.4 0.0 sand
2C 170+ 10YR 6/3 0.0 0.8 10.0 69.5 17.9 1.1 0.7 0.0 sand
gay

C f - Coarse fragments >2mm

VC 3 - Very coarse sand 1.0mm - 2.0mm

C s — Coarse sand 0.5mm - 1.0mm

M s - Medium sand 0.25mm - 0.5mm
F s - Fine sand 0.125mm - 0.25mm

VF s - Very ﬁne sand 0.053mm - 0.125mm
S - Silt 2pm - 50pm
C - Clay <2um
T C - Texture Class
c. sand - coarse sand

70

 

 

 

Table 2 (continued). Physical Properties of Pedons.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Horizon Depth Munsell Grain Size T C

(cm) Color C f IVC sIC 5 IM 5 IF 8 IVF sIS IC

(moist) per cent by weight

Weak Rubicon POD Index of 2
Oi 0-4
A 4-9 N 2/0 0.0 0.7 19.7 64.2 11.6 0.6 3.3 0.0 sand
E 9-20 7.5YR 4/2 0.0 0.7 17.5 62.0 13.7 1.2 3.7 1.2 sand
le 20-41 5YR 3/4 0.0 1.0 14.0 63.7 15.4 1.2 4.8 0.0 sand
882 41-62 10YR 4/6 0.0 0.2 10.3 71.5 16.1 0.7 0.1 1.0 sand
BC 62-126 10YR 5/4 1.0 0.1 3.0 65.0 30.7 0.8 0.5 0.0 sand
C 126-165+ 10YR 6/4 1.0 0.2 2.6 49.5 46.0 1.4 0.0 0.4 sand
Grayling POD Index of 0
Oi 0-7
A 7-11 7.5YR 3/2 0.0 0.4 12.1 63.8 13.9 0.8 9.0 0.0 sand
Bs 11-36 7.5YR 4/6 0.0 0.2 14.7 64.7 15.8 1.0 3.6 0.0 sand
BC 36-62 10YR 5/6 0.0 1.0 15.3 67.0 15.7 0.4 0.7 0.0 sand
C 62-116 10YR 5/4 2.0 0.2 14.7 72.4 11.9 0.4 0.4 0.0 sand
2C 116-140 10YR 5/3 6.0 1.8 16.4 59.0 20.4 1.3 1.1 0.0 sand
3C 140-165+ 10YR 5/3 0.0 0.3 11.5 62.8 23.5 1.3 0.6 0.0 sand
Key

C f - Coarse fragments >2mm

VC 8 - Very coarse sand 1.0mm - 2.0mm

C s - Coarse sand 0.5mm - 1.0mm
M s - Medium sand 0.25mm - 0.5mm

F s - Fine sand 0.125mm - 0.25mm

VF s - Very ﬁne sand 0.053mm - 0.125mm
S - Silt 2pm - 50pm
C - Clay <2um
T C - Texture Class

71

 

 

 

Table 3. Chemical Properties of Pedons.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Horizon Depth pH Fed Fep Fe0 Ald Alp

(cm) per cent of dry soil
Kalkaska
Oi 0-3
A 3-16 4.7 0.07 0.02 0.01 0.01 0.01 0.01
E 16-24 4.2 0.05 0.01 0.01 0.01 0.01 0.01
Bhs 24-31 4.0 0.75 0.17 0.18 0.15 0.14 0.14
851 31-58 6.1 0.16 0.07 0.04 0.18 0.20 0.22
B52 58-85 6.6 0.05 0.03 0.02 0.06 0.07 0.10
Bw 85-128 6.7 0.05 0.01 0.01 0.04 0.06 0.08
2E/Bt 128-160+ 6.9
Best Expressed Rubicon
Oi 0-5
A 5-9 5.4 0.09 0.03 0.02 0.03 0.03 0.03
E 9—1 8 6.1 0.08 0.02 0.03 0.02 0.03 0.03
le 18-28 5.7 0.70 0.26 0.17 0.26 0.30 0.30
B32 28-48 6.4 0.33 0.08 0.08 0.18 0.20 0.28
BC 48-79 7.1
C 79-170 7.2
2C 170+ 7.6
Least Ex ressed Rubicon
Oi 0-4
A 4-9 4.0 0.04 0.01 0.01 0.02 0.02 0.02
E 9-20 6.1 0.04 0.01 0.01 0.01 0.01 0.01
le 20-41 6.1 0.29 0.06 0.06 0.18 0.18 0.24
B52 41-62 6.9 0.05 0.02 0.01 0.06 0.07 0.13
BC 62-126 7.2
C 126-165+ 7.5

 

 

 

 

 

 

 

 

 

Gra '

0-7
7-1 1

1 1-36
36-62

62-116
116-140
140-165+

72

 

 

 

 

 

 

Figure 42. The Kalkaska pedon is the most strongly expressed Spodosol in the
development sequence of this study. Photo taken by Dr. Randy Schaetzl, Professor,
Dept. of Geography, Michigan State University.

73

Site Name: Kalkaska Pit

Location: Sand Lake Road

Slope at pit (%): 0

Parent material: Outwash, coarse at depth

Drainage class: Well drained

Water table (cm): none

Landform type: Outwash plain

Elevation: 394m

Erosion: none

Evidence of bioturbation: none

Vegetation at site: Sugar maple, Red pine, Yellow birch

Map: Starvation Lake Quadrangle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Horizon Horizon Moist Color Structure Consistence
Depth (cm) (Munsell) grade
Oi 0 - 3
A 3 — 16 7.5YR 2/0 weak granular moist v friable
E 16 -— 24 5YR 4/3 weak granular moist v friable
Bhs 24 — 31 5YR 2.5/2 medium subangular blocky moist ﬁ'iable
B51 31 - 58 7.5YR 4/6 medium subangular blocky moist friable
882 58 — 85 7.5YR 5/8 weak subangular blocky moist v friable
Bw 85 — 128 10YR 4/6 weak subangular blocky moist v friable
2E/Bt 128 — 160+ 10YR 5/3 E weak subangular blocky moist v ﬁiable
7.5YR 4/4 Bt

 

Figure 43. Selected information from the Soil Proﬁle Description Sheet of the Kalkaska

pedon.

74

 

 

Figure 44. The strong Rubicon pedon is the second best expressed Spodosol in the
development sequence of this study. Picture taken by Dr. Randy Schaetzl, Professor,
Dept. of Geography, Michigan State University.

75

Site Name: Strong Rubicon Pit

Location: N. Blue Lake Road across from Hassley Road (private)

Slope at pit (%): 0

Parent material: Outwash

Drainage class: Excessively well drained

Water table (cm): none

Landform type: Outwash Plain

Elevation: 384m

Erosion: none

Evidence of bioturbation: none

Map: Starvation Lake Quadrangle

Vegetation at site: White pine stumps, reindeer moss, White birch, Pin cherry, Aspen

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Horizon Horizon Moist Color Structure Consistence
Depth (cm) (Munsell) grade

Oi 0 - 5

A 5 — 9 7.5YR 3/2 weak granular moist v friable
E 9 — 18 7.5YR 4/2 weak subangular blocky moist v friable
le 18 — 28 5YR 4/6 weak subangular blocky moist v friable
Bs2 28 - 48 7.5YR 4/6 weak subangular blocky moist v friable
BC 48 — 79 10YR 5/6 8 grain subangular blocky moist friable
C 79 — 170 10YR 6/4 3 grain subangular blocky moist v friable
2C 170+ 10YR 6/3 5 grain subangular blocky moist v friable

 

Figure 45. Selected information from the Soil Proﬁle Description Sheet of the strong
Rubicon pedon.

76

 

 

Figure 46. The weak Rubicon pedon is the third best expressed Spodosol in the
development sequence of this study. Photo taken by Dr. Randy Schaetzl, Professor,
Dept. of Geography, Michigan State University.

77

Site Name: WeakrRubicon Pit Map: Fredrick Quadrangle
Location: Deward Road on west side

Slope at pit (%): 0 Landform type: Terrace

 

 

 

 

 

Parent material: Outwash Elevation: 355m
Drainage class: Excessively well drained Erosion: none
Water table (cm): none Evidence of bioturbation: none

Vegetation at site: Red we, Jack pine, White pine, White spruce, Red maple, Pin cherry
Blueberry, reindeer moss

 

 

 

 

 

 

 

 

Horizon Horizon Moist Color Structure Consistence
Depth (cm) (Munsell) grade

Oi 0 - 4

A 4 -— 9 N 2/0 weak granular moist v friable
E 9 — 20 7.5YR 4/2 weak medium granular moist v friable
B81 20 — 41 5YR 3/4 weak subangular blocky moist friable
Bs2 41 — 62 10YR 4/6 8 grain subangular blocky moist v friable
BC 62 — 126 10YR 5/4 5 grain subangular blocky moist v friable
C 126 — 165+ 10YR 6/4 s grain subangular blocky moist v friable

 

 

 

 

 

 

 

Figure 47. Selected information from the Soil Proﬁle Description Sheet of the weak
Rubicon pedon.

78

 

Figure 48. The Grayling pedon is the least strongly expressed Spodosol in the
development sequence of this study. Photo taken by Dr. Randy Schaetzl, Professor,
Dept. of Geography, Michigan State University.

79

Site Name: GraylingPit

Location: Goose Creek Road across from Berkley Road

Slope at pit (%): 0

Parent material: Outwash

Drainage class: Excessively well drained
Water table (cm): none

Vegetation at site: Jack pine, blueberry, planted Red pine, reindeer moss

Landform type: Terrace in Goose Creek Valley

Elevation: 351m

Erosion: none

Evidence of bioturbation: none

Map: Lake Margrethe Quadrangle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Horizon Horizon Moist Color Structure Consistence
Depth (cm) (Munsell) grade
Oi 0-7
A 7 — 11 7.5YR 3/2 weak subangular blocky moist loose
Bs 11 — 36 7.5YR 4/6 weak subangular blocky moist v ﬁiable
BC 36 — 62 10YR 5/6 s grain subangular blocky moist loose
C 62 — 116 10YR 5/4 3 grain subangular blocky moist loose
2C 116 — 140 10YR 5/3 s grain subangular blocky moist loose
3C 140 — 156+ 10YR 5/3 s grain subangular blocky moist loose
Deep 3C 280 — 300+ 10YR 5/3 8 grain subangular blocky moist loose

 

Figure 49. Selected information from the Soil Proﬁle Description Sheet of the Grayling

pedon.

80

 

Homblende Denticulation Amplitude

Three hundred hornblende grains from each horizon were examined under a
petrographic microscope to determine the denticulation amplitude along the margin of the
grain perpendicular to the c-axis (Figure 50). A weighted average of denticulation
amplitude and the standard deviation were then calculated for each horizon (Table 4, raw
data in Appendix A). The weighted average for each horizon was calculated by
multiplying the number of grains in each class of that horizon by the mean amplitude
measurement for that class, adding these values together, then dividing the sum by 300,
which is the number of grains measured in each horizon. These values were used to
create a horizon vs. denticulation amplitude plot for each site (Figure 51). All four sites
exhibited the same trend within each proﬁle. The denticulation amplitude increased up
proﬁle; denticulation amplitudes were minimal in the C horizons, but were larger in the
shallower soil horizons. The range of denticulation amplitude, which is the difference
between the highest weighted average and the lowest weighted average in each proﬁle,
increased as the strength of soil development increased. The Grayling pedon had a range
of 0.39pm - 3.24pm or 2.85pm, and the weak Rubicon was 0.57pm - 5.79pm or 5.22pm.
The strong Rubicon had a range of 1.15 pm - 7.16pm or 6.01 pm, while the Kalkaska
pedon was 1.31 um-8.21um or 6.90pm. The average denticulation amplitude
measurement for each individual pedon was calculated by averaging the weighted
horizon values (Table 4). The average denticulation amplitude for the Grayling was
1.53pm, the weak Rubicon was 2.56pm, the strong Rubicon was 3.09pm, and the
Kalkaska was 3.70pm. All four plots exhibit a break in slope directly above the

uppermost B horizon. In the Grayling the break was between the A and 851 horizons.

81

 

20pm

Figure 50. Petrographic micrograph of the denticulated margins of a naturally weathered
hornblende grain. Image captured with a Pixera imaging system.

82

Table 4. Denticulation Amplitude Results

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Soil & Weighted Standard Maximum Minimum Pedon Pedon Pedon

Horizon Amplitude Deviation Amplitude Amplitude Weighted Highest Lowest
Average Measure Measure Average Average Average

Kalkaska 3.70 8.21 1.31

A 8.21 4.10 26 2

E 7.28 4.06 22 2

Bhs 3.39 2.47 14 2

le 2.79 1.92 14 2

882 2.67‘ 1.82 14 2

Bw 2.20 1.39 10 O

2E/Bt 1.74 2.00 14 0

Deep 2C 1.31 1.43 10 0

Strong Rubicon 3.09 7.16 1.15

A 7.16 4.53 30 2

E 5.84 4.00 26 2

B51 2.43 1.44 14 2

B32 1.94 0.87 6 0

BC 1.77 1.15 10 0

C 1.36 1.20 6 0

2C 1.15 1.61 10 0

Weak Rubicon 2.56 5.79 0.57

A 5.79 4.17 26 2

E 4.73 2.99 14 2

B5] 2.28 1.26 14 2

Bs2 2.04 0.63 6 0

BC 1.65 0.99 6 0

C 0.89 1.15 6 0

Deg) C 0.57 1.07 6 0

Gr yling 1.53 3.24 0.39

A 3.24 2.20 14 2

Bs 2.23 0.97 6 0

BC 1.75 1.15 6 0

C 1.42 1.38 10 0

2C 1.14 1.57 10 0

3C 0.53 1.12 6 0

Deep 3C 0.39 0.93 6 0

 

 

 

 

 

 

83

I

IL._-o ﬂ

 

 

 

Strong Rubicon

 

 

 

 

 

Horizons

 

 

 

 

 

 

 

012345678910111213
012345678910111213

Mean Denticulation Amplitude (urn)

 

Mean Denticulation Amplitude (um)

 

 

 

 

 

 

 

Weak Rublcon

 

 

 

 

 

Horizons
Horizons

 

 

 

 

 

 

 

Y T

3456789101112” 012345678910111213

 

0 1 2

 

Mean Denticulation Amplitude (um) Mean Denticulation Amplitude (um)

 

 

 

 

 

 

Figure 51. Denticulation amplitude vs. horizon plot for each of the pedons in this study.

These plots show the change in the mean denticulation amplitude measurement with
change in horizon. Error bars represent 1 standard deviation.

84

 

In the other three plots the break was below the E and above the uppermost B horizon.

Quartz/Feldspar Ratios

Horizon vs. quartz/feldspar ratio plots were constructed for each site (Table 5 and
Figure 52, raw data in Appendix B). The quartz/feldspar values for the deepest horizons
for all sites plot at roughly the same point. The two more weakly developed soils, the
Grayling and the weak Rubicon, had similar trends, with no systematic vertical change in
the quartz/ feldspar ratios. This differs from the two more strongly developed soils, the
strong Rubicon and the Kalkaska, which plotted similarly to one another exhibiting an

irregular but unmistakable increase in the quartz/feldspar ratios up proﬁle.

Extractable Iron and Aluminum

Three extractants were used in this study to determine where Fe and Al were
enriched or depleted, and to determine the form of the Fe and Al in each horizon. Fed
data, which indicate the presence of organically complexed Fe as well as “free” Fe, was
at maximum in the horizon immediately below the B horizon in the Kalkaska pedon and
both Rubicon pedons (Table 3). The Grayling pedon does not exhibit an E horizon, but
the Fed is at maximum abundance in the Bs horizon, which is the uppermost B horizon.
Ald data mimic the Fed trend in each proﬁle, except in the Kalkaska pedon. Ald is at
maximum abundance in the B51 horizon, which is the second B horizon, not the
uppermost B horizon.

The PP extractant (F ep, Alp), which extracts Fe and A1 from organic complexes,
shows a maximum abundance of both Fe and Al in the uppermost B horizon for all

pedons (Table 3).

85

Table 5. Quartz / Feldspar Ratios

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Soil & Quartz / Feldspar
Horizon Ratios
Kalkaska

A 15.67
E 16.65
Bhs 8.38
881 11.00
B52 5.98
Bw 9.34
2E/Bt 7.33
Deep 2C 6.32
Strong Rubicon

A 24.00
E 11.50
851 16.65
852 10.54
BC 16.65
C 9.71
2C 6.14
Weak Rubicon

A 5.67
E 7.82
851 6.32
852 4.45
BC 6.69
C 4.56
Deep C 7.11
Ms

A 5.25
B5 6.50
BC 6.14
C 8.38
2C 10.54
3C 8.68
Deep 3C 7.57

 

86

 

 

Quartz/Felds par Ratios

 

 

 

 

 

 

 

 

 

l
“g 2
‘1: 5
a 3
E a
,2 4 + Kalkaska
“g, H -I— Strong Rubicon
g 5 + Weak Rubicon
3 r + Grayling
£2 6
O
N -
'8 7
I:

8 _

0 5 10 15 20 25 30

Quartz/Felds par Ratio

 

 

 

Figure 52. Horizon vs. Quartz/Feldspar Ratio. Numbers were used as horizon
designations instead of traditional horizon nomenclature because each pedon has various
horizon orders. This plot shows the change in the quartz/feldspar ratio with depth.

87

F 60 and A10 data, which are a measure of the “active” Fe and A1 of noncrystalline
hydrous oxides, are at maximum in the uppermost B horizon in each pedon, except the
Kalkaska pedon where A10 is at maximum in the B51 horizon (Table 3).

Various calculations were made with the extraction data (Table 6). These
calculations provide more information as to the nature of the Fe and Al in each horizon
(Appendix C). The activity ratio is calculated by AAO / CD (F e0 / Fed) and indicates the
amount of crystalline Fe oxides in the soil. A low ratio indicates a large amount of
crystalline Fe oxides. The A horizon of the Kalkaska pedon had the lowest ratio, which
was calculated to be 0.14. The other ratios were moderately high between 0.20 and 0.40,
with the highest ratio being 0.40 from the 852 horizon of the Kalkaska pedon.

The calculation that indicates whether the “free” oxides are in a crystalline form
or noncrystalline form is CD — AAO (Fed — F eo and Ald -— A10) (Table 6). Low values
from this calculation indicate the oxides are primarily in a noncrystalline form. The
values for Fed — FeO are moderately low and range from 0.03 to 0.57. The values for Ald
- Al0 are negative, except for one positive value of 0.01 in the Bhs horizon of the
Kalkaska pedon.

The molar ratio of Fep / Alp signiﬁes the difference between Fe and Al in the
organically complexed, noncrystalline form (Table 6). Ratio values less than 1.0 signify
that there is more organically complexed, noncrystalline Al than Fe in the horizon. The
Kalkaska pedon has values greater than 1.0 in the A, E, and Bhs horizons, while the BSI ,
BS2 and Bw horizons have values less than 1.0. The strong Rubicon has a value equal to
1.0 in the A horizon, while the E, B51, and 852 horizons are less than 1.0. However, the

weak Rubicon has a value equal to 1.0 in the B horizon, while the A, B51, and 852

88

Table 6. Fe and Al Extraction Calculations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Horizon FeO/Fed Fed-Fe0 Ald-Alo Few/Alp (Fe(,-Fep)/Fep (Al(,-Al,,)/Alp
percent of dry soil

Kalkaska

A 0.14 0.06 0 2.00 -0.50 0
E 0.20 0.04 0 1.00 0 0
Bhs 0.24 0.57 0.01 1.20 0.06 0
B5] 0.25 0.12 -0.04 0.35 -0.43 0.10
852 0.40 0.03 -0.04 0.43 -0.33 0.43
Bw 0.20 0.04 -0.04 0.17 0 0.33
Strong Rubicon

A 0.22 0.07 0 1.00 -0.33 0
E 0.38 0.05 -0.01 0.67 0.50 0
B51 0.24 0.53 —0.04 0.86 -0.35 0
Bs2 0.24 0.25 -0.01 0.04 0 0.40
Weak Rubicon

A 0.25 0.03 0 0.50 0 0
E 0.25 0.03 0 1.00 0 0
851 0.21 0.23 -0.06 0.33 0 0.33
852 0.20 0.04 -0.07 0.29 -0.50 0.86
Grayling

A 0.33 0.06 -0.01 1.67 -0.60 0
Bs 0.27 0.07 -0.06 0.54 -0.57 0.31

 

 

 

 

 

 

 

89

 

 

 

 

 

horizons are less than 1.0. The Grayling pedon has a value greater than 1.0 in the A
horizon and a value less than 1.0 in the B5 horizon.

The calculations of (FeO — Fep) / Fep and (A10 — Alp) / Alp indicate whether the Fe
and A1 are organically bound or inorganically bound in a noncrystalline form (Table 6).
Higher values suggest the inorganically bound complexes contribute more to this
measurement, while lower values indicate the organically bound complexes are more
dominant. In the Kalkaska pedon, the Fe values are low and range from —0.5 to 0.06.
While the Al values in the Kalkaska are zero in the A, E, and Bhs horizons, in the
remaining three horizons they range from 0.1 to 0.43. The strong Rubicon has Fe values
that are either zero or negative, except in the E horizon where the value is 0.5. The Al
values for this pedon are zero in the A, E, and B51 horizons, while the B52 is 0.4. The
weak Rubicon has Fe values of zero, except in the B52 horizon where the value is —0.5.
The Al values for this horizon are zero except for the B51 and B52 horizons where the
values are 0.33 and 0.86 respectively. The Grayling pedon has Fe values that are

negative and Al values that are zero and 0.31.

90

CHAPTER 8

Discussion

Homblende Denticulation Amplitude

The weighted average, maximum, and minimum denticulation amplitude of
hornblende were determined for each horizon within each pedon (Table 4). Homblende
grains were more weathered, as determined by the denticulation amplitude
measurements, at the top of the soil proﬁle than at the bottom of the proﬁle in all four
sites. The average was smallest in the deepest C horizon sampled and largest in the A
horizon for every pedon. The Grayling C horizon average was 0.39pm compared to its A
horizon of 3.24pm. The weak Rubicon averages were 0.57pm in the C horizon and
5.79pm in the A horizon. The strong Rubicon averages were 1.15pm and 6.01pm in C
and A horizons respectively, while comparable Kalkaska data were 1.31pm in the C
horizon compared to 8.21pm in the A horizon.

Standard deviations were calculated on the weighted averages for each horizon
within each pedon (Table 4). The standard deviation, which indicates the spread in the
data, was largest in the A horizon and became progressively smaller with depth in the
proﬁle for each pedon. This trend is to be expected since the more extensively weathered
hornblende is in the upper horizons.

While the homblende denticulation amplitude showed an obvious trend within
each proﬁle, the extent of weathering varied between the proﬁles. This variability
directly corresponded with the strength of soil development. The more strongly

developed soils had the more extensively weathered hornblende. This was determined by

91

comparing the minimum, maximum, and average value of denticulation amplitude for
each proﬁle, which was calculated from the weighted averages for each horizon (Table
4). The minimtun and maximum values of the pedons followed a progression that
increased with increasing strength of soil deve10pment. The lowest weighted amplitude
average for the Grayling, weak Rubicon, strong Rubicon, and Kalkaska were 0.39pm,
0.57pm, 1.15pm, and 1.31pm respectively. The highest weighted averages were 3.24pm,
5.79pm, 7.16pm, and 8.21pm respectively. The proﬁle average denticulation amplitude
also corresponded with strength of soil development (Table 4). The average
denticulation amplitude for the Grayling was 1.53um, the weak Rubicon was 2.56pm, the
strong Rubicon was 3.09pm, and the Kalkaska was 3.70pm. The ranges of denticulation
amplitude, as well as the average amplitudes, all directly correspond with the strength of

soil development.

Horizon vs. Homblende Denticulation Amplitude

A plot of horizon vs. weighted denticulation amplitude average was constructed
for the soil proﬁles (Fig. 51). All four plots exhibit a break in slope directly above the
uppermost B horizon. In the Grayling the break was between the A and B51 horizons. In
the other three plots, the break was below the E and above the uppermost B horizon. A
possible explanation for this break in slope is the presence of two acid systems at work in
the soil. Ugolini et a1. (1977) suggested that soluble organic acids control the pH and the
anionic balance in the upper portion of the soil, whereas the lower portion is dominated
by a bicarbonate system. They found mobile fulvic acids were reduced by 60 to 70

percent within the B horizons. The uppermost B horizon delineated the lower boundary

92

of the biopedological compartment that starts at the top of the canopy. The lower
compartment extended downward from the uppermost B horizon, and was in contact with
groundwater. The pH increased at the boundary between these two systems, which could
account for the varying degree of etching.

In the Grayling soil there was a distinct increase in pH from 4.4 to 6.1 between
the A and Bs horizons (Table 3), which was the same point at which the horizon vs.
denticulation amplitude plot had a distinct break in slope (Figure 51). The pH increased
from 4.0 to 6.1 between the A and E horizons in the weak Rubicon soil (Table 3).
However, this was not where the horizon vs. denticulation amplitude plot had a break in
slope (Figure 51). The break for this plot was between the E and 351 horizons, where the
pH held constant at 6.1. The strong Rubicon actually had a decrease in pH from 6.1 to
5.7 (Table 3) between the E and B51 horizons (Figure 51) where the plot had a break in
slope, and then the pH did increase to 6.4 in the B52 horizon. The Kalkaska horizon vs.
denticulation amplitude plot exhibited a similar trend to that of the strong Rubicon. In
the Kalkaska the pH decreased from 4.2 to 4.0 (Table 3) between the E and Bhs horizons
where there was a break in slope (Figure 51). However, the pH did increase to 6.1 in the
B51 horizon.

There was an increase in pH down proﬁle in all four pedons. Only in the
Grayling pedon did the increase in pH correspond with the break in slope of the plot.
This pH increase did not, however, directly correspond with the break in slope for the
other three pedons; the pH increase occurred one horizon below the break in slope. The

data did not show strong, convincing evidence that the boundary between the two

93

systems of acid was the cause for the break in slope on each of the horizon vs.

denticulation amplitude plots, however it may be inﬂuential.

Inﬂuence of pH on Dissolution Rates

It was observed that the pH in the upper section of the pedon, above the distinct
change in pH, was between 4 and 5, while the pH in the lowest sample taken was
between 6 and 7.5 (Table 3). The difference in pH between the upper and lower intervals
was roughly 2 units in all four sites. The hornblende in the upper proﬁle, with the lower
pH, was 6 to 10 times more deeply etched than the hornblende in the lower proﬁle, with
the higher pH. All of the grains have been at their present location for the same amount
of time, so the hornblende in the upper section etched 6 to 10 times faster than the
hornblende in the lower section. These empirical data were compared with experimental
data from literature. Sverdrup (1990, as cited in Brantley and Chen, 1995) determined a
dissolution rate for hornblende, having the composition of
Ca1_7Mg3,5Fel_3A11,4Si6,9Ozz(OH)2, at various values of pH. At a pH of 4.4, a dissolution
rate of 1.58x10'l4 moles inosilicate/cmz/sec was measured. At a pH of 6.5, the
dissolution rate was 2.00x10’IS moles inosilicate/cmz/sec. Given these two rates,
dissolution at a pH of 4.4 is 7.9 times faster than dissolution at a pH of 6.5. This value
lies within the range observed in the empirical data where dissolution rates were 6 to 10
times faster at a pH of 4 to 5 than at a pH of 6 to 7.5. This consistency supports the idea
that the variation in hornblende etching in the upper proﬁle is more strongly inﬂuenced

by variation in pH than in age.

94

Factors that Inﬂuence pH

It has been concluded by many researchers that pH does inﬂuence the dissolution
rate of primary minerals (Brantley and Chen, 1995, and all references therein). However,
there are many factors that inﬂuence the pH of a soil, and thus ultimately inﬂuence the
dissolution rate of primary minerals.

The range in the dissolution rate of hornblende observed in the empirical data
could be explained by natural conditions that vary among the pedons, such as organic
matter and the presence of clay. Organic matter inﬂuences the soil environment in a
number of ways. The accumulation of trace metals in the soils has been attributed to the
scavenging of organic humic and fulvic acids based on the observed association between
organic-rich soils and high metal values (Hoffman and Fletcher, 1981). It is accepted that
organic matter strongly inﬂuences the mobility of Fe and Al by complexation and thus
providing a means of transport for the Fe and Al ions (e. g., Antweiler and Drever, 1983;
Mokma and Evans, 2000; Chadwick and Graham, 2000; Ollier and Pain, 1996;
Birkeland, 1999).

Several researchers have studied the effects of organic matter on pH (e. g.,
Antweiler and Drever, 1983; Mast and Drever, 1986; Birkeland, 1999). Antweiler and
Drever (1983) found that the presence of organic matter could lower the pH of the pore
water, and thus inﬂuence the dissolution rates of minerals. In their study they noted that
the pH of percolating water ranged from 4.3 to 6.5. Laboratory experiments were
conducted on samples from their study area, in the absence of organic compounds, and
yielded a pH range of 7.3 to 9.5. In the laboratory samples the Fe and Al concentrations

were below detection limits because Fe and A1 are not soluble in that range of pH values.

95

They also noted that organic matter breaks down to form several different acids that react
differently in the soil environment. Acids that were found in their study include:
humic/fulvic, oxalate, acetate, and formate acids. Propionate, succinate, and malonate
were also tentatively identiﬁed. Therefore, they concluded that the breakdown of

primary minerals is strongly inﬂuenced by the presence of organic matter.

Kalkaska Heavy Mineral Grains and Clay

Clay may also have an inﬂuence on the dissolution rate of primary minerals. Not
only does clay inhibit the pH from lowering in soils (Bloom, 2000; Churchman, 2000),
pedogenic clay may attach to grains, ﬁll pore space, and alter water ﬂow patterns
(Chadwick and Graham, 2000; Goldberg et al., 2000; Radcliffe and Rasmussen, 2000).

Understanding that clay may have an inﬂuence on the dissolution rate of primary
minerals, heavy mineral grains from the Kalkaska A, E, Bhs, and B51 horizons were
viewed in a scanning electron microscope (SEM) in order to View any clay that might be
present. The Kalkaska pedon was chosen because of the very distinct pH change from
4.0 to 6.1 that occurs between the Bhs and B51 horizons (Table 3). Roughly 15 grains
from each horizon were rinsed to remove dust, and heavy liquid separated to make the
selection process easier. In order to observe the clay coatings, or coverings, the clay and
sesquioxides were not removed. Energy-dispersive spectroscopy (EDS) was used to
identify the weathering product.

The grains above the change in pH were markedly different from the grains below
the pH change. These differences included: organic material, weathering product

chemistry, and weathering product abundance. In horizon A, with a pH of 4.7, there is

96

very little weathering product present (Figure 53), however, the weathering product that
is present was identiﬁed as an aluminosilicate, most likely kaolin (A12SizOs(OH)4)
(Figure 54). The determination was based on elements identiﬁed by EDS. Organics are
also present on these grains as determined by EDS analyses that show a high C peak
(Figures 55 and 56). Tube-like features (Figures 57 and 58) were also scanned with EDS
(Figure 59) and determined to be organic material due to the high C content. In the E
horizon, with a pH of 4.2, there is more clay present than in the A horizon, but less
organic matter (Figure 60). The weathering product in this horizon was identiﬁed as an
aluminosilicate (Figure 61). The grains from Bhs horizon, with a pH of 4.0, have notably
more clay than the B horizon, but the grains are not completely coated (Figure 62). This
weathering product was also an altuninosilicate (Figure 63). There is also more organic
matter in this horizon than in the E horizon (Figure 62). The le horizon has a pH of 6.1
and is the uppermost horizon below the distinct change in pH. In this horizon, the grains
are covered extensively with clay (Figure 64), which was determined to be an
aluminosilicate (Figure 65). String-like features were observed in this horizon (Figures
66 and 67), however EDS did not show them to contain carbon (Figure 68). They all had
the same EDS signature as the weathering product, which was identiﬁed as an
aluminosilicate. The identity of these features is unknown.

Several observations were made among the horizons. Tungsten was detected with
the aluminosilicate in the Bhs and B51 horizons (Figures 63, 65, 68). It most likely
sorbed to the clay from the heavy liquid, which was sodium polytungstate, during the
laboratory procedure to separate the heavy minerals from the sample. Also, upon

examining the EDS data, it was noticed that Al increased in abundance with depth

97

 

 

200nm

Figure 53. Scanning electron micrograph of a weathered heavy mineral gain from the A
horizon of the Kalkaska pedon with very little weathering product. This product is an
aluminosilicate. Image taken at the Center for Advanced Microscopy at Michigan State
University by Leslie Mikesell and Ewa Danielewicz.

 

4000

”ﬂacan

 

 

15.360

 

 

 

Figure 54. Energy-dispersive spectrogaph of the weathering product indicated the by
circle in Figure 53. The high Si content with Al and 0 led to the determination that it is
an aluminosilicate. The Au is the signature of the coating used to prepare the gain.

98

 

100nm

Figure 55. Scanning electron microgaph of a weathered heavy mineral gain from the A
horizon of the Kalkaska pedon with organic material present. Image taken at the Center
for Advanced Microscopy at Michigan State University by Leslie Mikesell and Ewa
Danielewicz.

 

5000

INJCOO
O

 

 

"‘ 9“ nu Ca Ti 90
n F- I K , a“ A J

 

 

 

Figure 56. Energy-dispersive spectrogaph of the organic material indicated by the circle
in Figure 55 showing high C content. The Au is the signature of the coating used to
prepare the gain.

99

 

 

200nm

Figure 57. Scanning electron microgaph of a weathered heavy mineral gain from the A
horizon of the Kalkaska pedon with tube-like organic material present. Image taken at
the Center for Advanced Microscopy at Michigan State University by Leslie Mikesell
and Ewa Danielewicz.

 

 

50pm

Figure 58. Scanning electron microgaph of tube-like organic material indicated by the
circle in Figure 57.

100

 

3000 ‘

an

amazon
m
m

 

 

, N "a u Ca
3 1J Bu Fe Qu

I “' ‘“ “ W ., [L ““

0 l A Au
I.) luv

.000 14.400

 

 

 

Figure 59. Energy-dispersive spectrogaph of tube-like organic material indicated by the
circle in Figure 57 showing high C content. The Au is the signature of the coating used
to prepare the gain.

101

 

200nm

Figure 60. Scanning electron microgaph of a weathered heavy mineral grain from the B
horizon of the Kalkaska pedon with more weathering product than the A horizon. This
product is an aluminosilicate. Image taken at the Center for Advanced Microscopy at
Michigan State University by Leslie Mikesell and Ewa Danielewicz.

 

7000

Inacon

 

 

le 16.875

 

 

 

 

Figure 61. Energy-dispersive spectrogaph of the weathering product indicated the by
circle in Figure 60. The high Si content with A1 and 0 led to the determination that it is
an aluminosilicate. The Au is the signature of the coating used to prepare the gain.

102

Organic A
material

 

 

200nm

Figure 62. Scanning electron microgaph of a weathered heavy mineral grain from the
Bhs horizon of the Kalkaska pedon with more weathering product and organic material
than the B horizon. This weathering product is an aluminosilicate. Image taken at the

Center for Advanced Microscopy at Michigan State University by Leslie Mikesell and

Ewa Danielewicz.

 

4000I

I it

unJCOn

 

 

 

4..
I0 . 000 Rev 1:. .520

 

 

 

Figure 63. Energy-dispersive spectrogaph of the weathering product indicated the by
circle in Figure 62. The high Si content with A1 and 0 led to the determination that it is
an aluminosilicate. The Au is the signature of the coating used to prepare the gain.

103

 

 

200m

Figure 64. Scanning electron microgaph of a weathered heavy mineral gain from the
351 horizon of the Kalkaska pedon with an extensive coating of weathering product.
This product is an alurrrinosilicate. Image taken at the Center for Advanced Microscopy
at Michigan State University by Leslie Mikesell and Ewa Danielewicz.

 

I n 3 = o n

 

 

 

ktV 15.360

 

 

Figure 65. Energy-dispersive spectrogaph of the weathering product indicated by the
circle in Figure 64. The high Si content with A1 and 0 led to the determination that it is
an aluminosilicate. The Au is the sigrature of the coating used to prepare the gain.

104

 

 

200nm

Figure 66. Scanning electron microgaph of a weathered heavy mineral grain from the
BSI horizon of the Kalkaska pedon with string-like features that were determined to not
be organic due to the EDS having the same signature as the weathering product. Image
taken at the Center for Advanced Microscopy at Michigan State University by Leslie
Mikesell and Ewa Danielewicz.

ilill|j :‘

(.11

Jill" In I ' iimj

U“ ‘;H'r
an

 

Figure 67. Scanning electron microgaph of string-like features indicated by the circle in
Figure 66.

105

 

3000

Si

Or'DCOf‘!

RI

 

 

 

 
 

I?

 

o
I"°°° 15.360

 

 

 

Figure 68. Energy-dispersive spectrograph of the string-like features indicated the by
circle in Figure 66. The features were determined to not be organic due to the EDS
having the same signature as the weathering product. The Au is the signature of the
coating used to prepare the grain.

106

(Figures 54, 61, 63, 65). This is to be expected since Al is more soluble at low pH
values, allowing it to move more deeply in the proﬁle (Mokma and Evans, 2000). The
mobility of the A1 allowed aluminosilicates to form lower in the pedon. A marked
increase of pedogenic clay on gains, which was observed between the E and Bhs
horizons, could indicate altered water ﬂow paths that would affect the dissolution rate of
primary mineral (e. g., Dove, 1995; Nugent etal., 1998). The gain size distribution data
support this qualitatively observed increase in clay in the Bhs horizon (Table 2). The Bhs
horizon of the Kalkaska pedon is 3.5 percent clay, while the other horizons have 0.0
percent clay. It is at this same point, between the E and Bhs horizons of the Kalkaska
pedon, that the break in slope is observed in the horizon vs. denticulation amplitude plot
(Fig. 51); meaning the dissolution rate above the Bhs horizon is faster than the rate in the
Bhs horizon and below. Determining the exact effect clay has on the dissolution rate of
primary minerals is beyond the scope of this study, however, the presence of clay does

seem to inﬂuence the dissolution rate of primary minerals in soils.

Quartz/Feldspar Ratios as a Weathering Indicator

The quartz/ feldspar ratio was chosen to determine relative weathering. For this
study, a horizon vs. quartz/feldspar ratio plot was constructed for each of the four pedons
(Figure 52). A base, or initial, value was determined by calculating a quartz/feldspar
ratio for the deepest sample taken. The deepest horizons for all sites plot at roughly the
same point. This indicated that the parent material for the four sites is fairly uniform with
respect to the quartz and feldspar content. The divergence in the plots from that point

was assumedly due to soil development. The two less developed soils, the Grayling and

107

the weak Rubicon, plotted similarly, and had near vertical trends; the quartz/feldspar
ratios were relatively uniform up-proﬁle. Upper horizons in these two pedons were not
depleted in feldspar with respect to quartz, indicating they were not extensively
weathered. The two more strongly developed soils, the strong Rubicon and the Kalkaska,
also plotted similarly to one another. They both had higher ratios up-proﬁle than the ﬁrst
two plots, and they both had many breaks in slope producing a zigzag pattern. The larger
ratio value was caused by a depletion of feldspar with respect to quartz, indicating that

these two pedons were more extensively weathered than the Grayling and weak Rubicon.

Grain Size and Quartz/Feldspar Ratios

The multiple breaks in slope, or zigzag pattern, on the horizon vs. quartz/feldspar
ratio plot may be due to the gain size used for point counting. Point counts were
performed on thin sections that contained only the ﬁne sand size fraction (125 um-
250pm) of the soil. During the design of this research, it was decided to use the ﬁne sand
size fraction throughout the study because several studies show that heavy minerals,
which include hornblende, are concentrated here (e.g., Locke, 1979; Hall and Horn,
1993). Quartz and feldspar are present in this size fraction, however, they are light
minerals and their abundance may not be accurately represented in this size fraction
alone. The point counting did produce data with an obvious overall trend, but the zigzag
pattern may be due to choosing a narrow range and small gain size for point counting.
Feldspar grains may break apart along cleavage planes when weathering, producing more
and smaller gains than the original. These smaller gains would be concentrated in this

size fraction and would not be an accurate depiction of the mineral’s abundance in the

108

whole soil. Quartz, on the other hand, is more persistent than feldspar and may not break
up into smaller gains during weathering. This could lead to skewed ratios. The trend of
the plot is an accurate depiction for the overall trend of the quartz/feldspar ratios in each
pedon. The zigzag pattern, however, may be the result of using only the ﬁne sand size
fraction of the soil when point counting.

The A horizon of the Kalkaska pedon does not have the highest quartz/feldspar
value, which is different than expected. The zigzag pattern of the slope has the Kalkaska
and strong Rubicon plots crisscross each other with the Rubicon having the highest value
in the A horizon (Figure 52). In examining the data, it was observed that the strong
Rubicon has two anomalies in the physical data (Table 2). The pH values for this pedon
are more variable than the pH values of the other three pedons. The Kalkaska, weak
Rubicon, and the Grayling pedons all have pH values that exhibit a clear trend; that being
a trend of decreasing pH values from the A to the E or an upper B horizon, and then
increasing with depth. The pH values for the strong Rubicon do not exhibit a clear trend.
This pedon also is different from the other three in its gain size distribution. It is the
only pedon to have coarse fragnents in the upper horizons. The other pedons have
coarse fragments in their lower horizons; however, the strong Rubicon has the highest
percent of coarse fragnents in those horizons. These two anomalies may indicate the
composition of the parent material for the strong Rubicon is slightly different than the
parent material of the other three soils. A different composition in the parent material
would yield different quartz/feldspar ratios. One observation that makes this explanation
questionable is that the deepest samples from all four pedons plot at roughly the same

point. However, the anomalous gain size distribution alone could yield these

109

differences. The strong Rubicon has a higher abundance of the coarse fragments than the
other pedons; however, the point counts were performed on the ﬁne sand size fraction of
the soil. By using a narrow range of gain sizes, this test could give erroneous data for
this pedon. Due to the close proximity of the pedons to one another in the ﬁeld, and the
clear and strong trends in the analytical data for both the hornblende denticulation
amplitude and the Fe and Al mobility, it was concluded that the quartz/feldspar ratio
trends are due to the difference in gain size and not a difference in composition of the
parent material.

Changes in the quartz/feldspar ratios indicate bulk mineralogy change. Rai and
Kittrick (1989) suggested soils are short-term entities relative to the time required for the
dissolution of minerals to the extent that changes in bulk chemistry could be observed.
The age of these pedons, approximately 13,000 yr B.P., may simply be too young for this
index to yield conclusive trends and results. It is not known which contributes more to
the ambiguity in these results, the age of the soil or the grain size fraction used in point
counﬁng.

While these plots do not contradict the order of soil development as assessed in
the ﬁeld, they do not support it either. This mineral weathering indicator simply
distinguishes between the two least developed pedons and the two more strongly
developed pedons. A clear ordering of the four pedons cannot be demonstrated using

quartz/ feldspar ratios.

Summary

Homblende denticulation amplitude measurements and soil development both

110

indicate the same trend. They suggest the order of 5011 development among the pedons,
in increasing strength, is the Grayling pedon, the weak Rubicon pedon, the strong
Rubicon pedon, and the Kalkaska pedon. They both show the same clear, strong trends
within each pedon and the same convincing trends among pedons. The quartz/feldspar
ratios results did not exhibit strong trends due to ambiguity in the data, which was most
likely caused by the young age of the soil or the gain size fraction used for point
counting. These ratios did, however, indicate that weatherable silicates were less
depleted from the less developed Grayling and weak Rubicon pedons, and more depleted

from the more developed strong Rubicon and Kalkaska pedons.

111

CHAPTER 9

Conclusions

Homblende Denticulation Amplitude

The hornblende denticulation amplitude measurements were plotted and used to
conclude that mineral weathering increased up proﬁle. The plots also showed clear
trends that were used to determine a relative weathering order for the pedons in this

study.

Comparison of Homblende Denticulation Amplitude and Soil Development

The hornblende denticulation amplitude measurements and the soil development
trends were similar both within and among the four pedons. There was a deﬁnite trend of
the mineral weathering and soil development being less in the lower proﬁle and
progessively increasing in intensity up proﬁle. Also, the mineral weathering trend
among the pedons mimicked the order of soil development. From these trends, it can be
concluded that, as these soils developed, the minerals weathered, and thus weathering and

soil development are related and not independent of one another in these soils.

Comparison of Homblende Denticulation Amplitude and Quartz/Feldspar Ratios
Quartz/feldspar ratios resulted in trends similar to, but not as strong as, those of
the denticulation amplitude measurements. The ratios decreased up proﬁle indicating

feldspar was weathering more in the upper proﬁle than in the lower proﬁle. This follows

112

the denticulation amplitude trend, but due to the zigzag nature of the ratio plot, the trend

is not strong or clear.

Comparison of Soil Development and Quartz/Feldspar Ratios

Attempting to use quartz/ feldspar ratios as indicators of soil development may be
inappropriate for soils as young as the pedons in this study, which are approximately
13,000 yr B.P. The ratio of quartz/feldspar relies on the depletion of feldspar as
compared to quartz, which is not a rapid process. The plots made in this study showed
the two least developed soils to be similar, and the two strongest developed soils to be
similar to each other but different than the ﬁrst two. The results did not provide a

convincing relative weathering order for the pedons.

Weathering Versus Soil Development

The terms weathering and soil development have occasionally been used
interchangeably. This study shows that mineral weathering and soil development do
proceed in tandem on the soils of the study area. However, the terms mineral weathering
and soil development reﬂect two different processes. Bland and Rolls (1998) deﬁned
weathering as, “the alteration by chemical, mechanical, and biological processes of rock
and minerals, at or near the Earth’s surface, in response to environmental conditions.”
Birkeland (1999) deﬁnes soil development as, “a qualitative measure of the amount of
pedologic change that has taken place in the parent material.” These two processes,
although usually occurring simultaneously and/or in parallel in regolith, should be named

and referred to correctly.

113

Homblende as an Age Indicator

Homblende denticulation amplitude, as an age indicator, may not be appropriate
to use on older soils. The hornblende in older soils may have reached a steady state
denticulation morphology so there would be no signiﬁcant difference in the denticulation
amplitude (Hall and Michaud, 1988). Steady state conditions in older soils may render
this tool useless for determining relative weathering. Denticulation amplitude
measurements may be useless as an age indicator also due to steady state. Age
determination of a soil, or relative ages of multiple soils, would be impossible by
denticulation amplitude alone once steady state has been reached. Without signiﬁcant
differences in amplitudes from one soil to another, an age determination cannot be made.

As previously stated, the quartz/ feldspar ratio may be an inappropriate soil
development indicator for younger soils because of the time requirements needed for a
region to show a depletion in feldspar. However, hornblende denticulation amplitude
measurements and quartz/ feldspar ratios complement each other. Where the
quartz/feldspar ratio is not a good development indicator for younger soils, the
denticulation amplitude measurements yield convincing conclusions. The opposite is
also true, where the denticulation amplitude measurements may become useless due to
steady state in older soils, the quartz/ feldspar ratios may exhibit strong trends.

Homblende etching has been studied by many researchers and has been deemed a
suitable tool for relative age determination (Clayton, 1979; Hall and Michaud, 1988; Hall
and Heiny, 1983; Hall and Martin, 1986; Locke, 1979, 1986). This study, however,
demonstrates that soils formed in deposits of the same age can have signiﬁcant

differences in denticulation amplitudes from one pedon to another. The differences

114

between the end members, the Grayling and Kalkaska pedons, is great enough that a
researcher might conclude they were of different ages if hornblende amplitude was used
as a relative age indicator in this setting. Homblende weathers to produce etch features in
a predictable manner that is easily recognizable. However, by using these features alone,
it is impossible to distinguish intensity of weathering from duration of weathering.
Because of this fact, concluding the etch features are due to duration of weathering is an

assumption at best and simply wrong at worst.

115

APPENDICIES

116

APPENDIX A

Denticulation Amplitude Measurements Raw Data Table

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pedon and Class 0 Class 1 ClassZ Class 3 Class4 Class 5 Class 6 Class 7 Class 8 Tot Avg Std
Horizon MAO MA2 MA6 MA 10 MA 14 MA 18 MA22 MA26 MA 30 Dev
Kalkaska

A 0 33 132 88 35 8 3 l 0 300 8.21 4.10
E 0 61 128 76 27 5 3 0 0 300 7.28 4.06
Bhs O 216 67 14 3 0 0 0 O 300 3.39 2.47
B51 0 256 34 9 1 0 0 0 0 300 2.73 1.92
852 0 259 33 7 1 0 O 0 0 300 2.67 1.82
Bw 24 251 23 2 O 0 0 0 0 300 2.20 1.39
2E/Bt 101 177 14 7 1 0 0 0 0 300 1.74 2.00
Deep2C 129 159 11 l 0 0 0 0 0 300 1.31 1.43
Pedon Avg 3.69
Strong Rubicon

A 0 67 137 61 21 8 4 1 1 300 7.16 4.53
E 0 112 120 44 19 3 l 1 0 300 5.84 4.00
BS] 0 272 25 2 1 O 0 0 0 300 2.43 1.44
B52 25 267 8 0 0 O 0 0 0 300 1.94 0.87
BC 55 236 8 1 0 0 0 O 0 300 1.77 1.15
C 110 183 7 0 0 0 0 0 0 300 1.36 1.20
2C 159 128 10 3 0 0 0 0 O 300 1.15 1.61
Pedon Avg 3.09
Weak Rubicg

A 0 117 117 44 14 4 3 l O 300 5.79 4.17
E 0 139 124 30 7 0 0 0 0 300 4.73 2.99
851 0 283 14 2 1 0 0 0 0 300 2.28 1.26
B52 6 288 6 0 0 0 0 0 0 300 2.04 0.63
BC 62 233 5 0 0 0 0 0 0 300 1.65 0.99
C 175 121 4 0 0 0 0 0 0 300 0.89 1.15
DeepC 223 73 4 O 0 O 0 0 0 300 0.57 1.07
Pedon Avg 2.56
(3423:1198

A 0 219 70 10 1 0 0 0 0 300 3.24 2.20
B5 2 280 18 0 0 0 0 0 O 300 2.23 0.97
BC 59 230 11 0 0 0 0 0 0 300 1.75 1.15
C 111 178 10 1 0 0 0 0 0 300 1.42 1.38
2C 163 121 15 1 0 0 0 0 O 300 1.14 1.57
3C 233 61 6 0 0 0 0 0 0 300 0.53 1.12
Deep 3C 248 49 3 0 0 0 0 0 0 300 0.39 0.93
Pedon Avg 1.53

Key:

MA - Mean amplitude for that class
Tot - Total gains measured

117

Avg - Weighted average
Std Dev - Standard deviation

APPENDIX B

Quartz and Feldspar Raw Data Table

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Soil & Quartz Feldspar Total Ratios
Horizon Counts Counts Counts

Kalkaska

A 282 18 300 1 5.67
E 283 1 7 300 16.65
Bhs 268 32 300 8.38
B81 275 25 300 11.00
B82 257 43 300 5.98
Bw 271 29 300 9. 34
2E/Bt 264 36 300 7.33
Deep 2C 259 41 300 6.32
Strong Rubicon

A 288 12 300 24.00
E 276 24 300 1 1.50
B81 283 17 300 16.65
BS2 274 26 300 10. 54
BC 283 1 7 300 16.65
C 272 28 300 9.71
2C 258 42 300 6.14
Weak Rubicon

A 255 45 300 5.67
E 266 34 300 7.82
BS] 259 41 300 6.32
B82 245 55 300 4.45
BC 261 39 300 6.69
C 246 54 300 4.56
DeeLC 263 37 300 7.1 1
Ms

A 252 48 300 5.25
BS 260 40 300 6.50
BC 258 42 300 6.14
C 268 32 300 8.38
2C 274 26 300 10.54
3C 269 31 300 8.68
Deep 3C 265 35 300 7.57

 

 

 

 

 

118

 

APPENDIX C

Extractable Iron and Aluminum

Podzolization is the process by which Spodosols are developed. During this
process, Fe and A1 are translocated from the O and A horizons to the B horizons. This
process leaves an eluvial zone, or E horizon, which is depleted in Fe and Al, and
produces an accumulation of Fe and Al in the B horizons. One measure of strength of
soil development is the difference in the amount of Fe and A1 in the A and E horizons
from that in the B horizons. To determine the amount of Fe and Al that has translocated,
Fe and Al were chemically extracted from each of these horizons. In a well developed
soil, there should be a trend that exhibits low amounts of Fe and Al in the A and E
horizons and a higher amount of Fe and A1 in the B horizons, typically with the highest
amount in the uppermost B horizon.

Iron and aluminum are found in crystalline, complexes (organic and inorganic),
and noncrystalline or amorphous forms in the soil. Three extractions were performed
which were designed to test for Fe and Al in these various states. The ﬁrst extractant was
sodium citrate-dithionite (CD), which extracted the “free” Fe and A1. This value
represents the total pedogenic crystalline, noncrystalline, and organically bound Fe and
Al. The results are denoted as F ed and Ald. The second extractant was acidiﬁed
ammonium-oxalate (AAO), which is good for indicating the presence of noncrystalline,
or amorphous, Fe and Al. The results are denoted as Feo and A10. The third extractant

was sodium pyrophosphate (PP), which is a good indicator for the presence of organically

119

bound Fe and Al. The results are denoted as F ep and Alp. These values yield a fairly
accurate depiction of the movement of Fe and Al within the pedon.

The F ed, which indicates the presence of “free” Fe in crystalline, noncrystalline,
and organically bound complexes, has its maximum abundance in the horizon
immediately below the B horizon in the Kalkaska pedon and both Rubicon pedons (Table
3). This maximum in the uppermost B horizon indicates a translocation of Fe down
proﬁle in each pedon. The Grayling pedon does not exhibit an B horizon, but the Fed is at
maximum abundance in the B5 horizon showing Fe is being eluviated from the A
horizon. The Ald mimics the F ed trend in each pedon except in the Kalkaska. The Ald
value is highest in the B51 horizon, which is the second B horizon, not the uppermost B
horizon. It is common, and almost expected, to have the maximum value for A1 occur
lower in the pedon than the maximum value for Fe because of the solubility of A1 at
lower pH values (Mokma and Evans, 2000). These trends show a downward movement
of both “free” Fe and Al in all four pedons.

The Feo, which indicates the “active” Fe in noncrystalline hydrous oxides, is at
maximum abundance in the uppermost B horizon, and has a much lower abundance in
the upper horizons of all four pedons (Table 3). The A10, which indicates amorphous
aluminosilicates, trend is the same as the Ald trend in that it is highest in the uppermost B
horizon in each pedon, except in the Kalkaska pedon where it is highest in the B51
horizon. There is a downward movement of “active” Fe and Al in all four pedons.

The F ep, which indicates the organically bound, or complexed, Fe, is highest in

the uppermost B horizon, and is lower in abundance in the upper horizons (Table 3). The

120

Alp trend is the same as the A10 and the Ald. The results from this extractant also suggest
a downward movement of Fe and Al.

In examining the data directly, as individual extractions, an accumulation of Fe
and Al in the B horizons was observed (Table 3). In all four pedons, the lower values for
Fe and A1 are in the A and E horizons. The higher values for Fe and A1 are in the
uppermost B horizon, or in the second B horizon in the case of the Kalkaska pedon, and
then the values decrease from that point with depth. Generally, Fed > Fep > Feo, which
indicates that Fe exists as “free” oxides, either crystalline or noncrystalline, in all four
pedons. Also, in all four pedons Alo > Alp > Ald and suggests that Al is primarily
“active” noncrystalline.

Various calculations may be made using these values that provide more
information about the state of the Fe and Al in the horizon.

The “activity ratio”, or F eo / Fed, for all four pedons yields moderately low values
(Table 6). This indicates that Fe is predominantly crystalline in the pedons.

The relative amount of crystalline “free” oxides compared to noncrystalline “free”
oxides is calculated by CD — AAO (Fed — Fe0 or Ald — A10) (Table 6). Fed — Feo is
positive for every horizon in each pedon indicating that crystalline Fe oxide is the
primary form of “free” Fe. The Ald — Al0 value is either zero or negative. One exception
to this is the Bhs horizon of the Kalkaska, which is still a low value at 0.01. This
indicates that “free” Al is primarily noncrystalline in all four pedons.

The molar value of F ep / Alp signiﬁes whether Fe or A1 is more prevalent in the
noncrystalline organic complexes (Table 6). Low ratio values indicate more of the

complexes are Al than Fe. In the Kalkaska pedon, the A, E, and Bhs horizons have

121

Fep/Alp values of one or greater, signifying that Fe complexes are equal to or more
prevalent than A] complexes in the upper portion of the pedon; while Pep/Alp in the B51,
B52 and Bw horizons are less than one, indicating that Al complexes are more prevalent
in the lower portion of the pedon. This trend could be explained by the solubility and
subsequent mobility of A1 at these pH values. The A horizon of the strong Rubicon has a
Fep/Alp value of one, indicating the number of Fe complexes is roughly equal to Al
complexes, while the E, 851 , and 832 horizons have organic complexes mainly of Al.
The weak Rubicon has Al complexes more prevalent than Fe complexes in all horizons
except the B horizon, where they are roughly equal. The Grayling pedon has mainly Fe
complexes in the A horizon and predominantly Al complexes in the B5 horizon. There is
an overall trend among the four pedons of higher values in the upper pedon, indicating
more Fe complexes than Al complexes, and decreasing values in the lower pedon,
indicating Al complexes are more prevalent.

The form of the noncrystalline complexes, whether organic or inorganic, can be
estimated by (Fe0 — Fep) / F ep and (A10 — Alp) / A1p (Table 6). For this calculation, lower
values indicate organic complexes are predominant, while higher values indicate
inorganic complexes are more prevalent. The values for Fe are zero or negative,
suggesting Fe is predominantly organically complexed. This is true for all horizons
except in the Kalkaska Bhs horizon where it is 0.06, and in the strong Rubicon where it is
0.50. However, these values are considered low (Wang and Kodama, 1986), indicating
Fe is organically complexed in these horizons as well. The values for A1 are zero or
positive in all horizons. Even with the values being positive, they are still considered low

ratio values, indicating Al is also primarily organically complexed.

122

 

To summarize, the extraction data indicates that Fe and Al in crystalline,
noncrystalline, and complexed forms are all moving down the proﬁle. In all four pedons,
the Fe is predominantly in “free” crystalline form, meaning it is in Fe oxides. When it is
complexed, it is organically complexed. The Al, in all four pedons, is predominantly
“active” noncrystalline and organically complexed. Where “free” Al is present, it is also
noncrystalline.

It was also observed that the extraction data support the strength of soil
development order of the pedons (Table 3 and 5). The Kalkaska pedon exhibits strong,
clear trends with maximum values higher than those of any other pedon. This pedon is
the only one in which the horizon of maximum Al abundance is below that of maximum
Fe abundance, indicating a more strongly developed soil. The strong Rubicon pedon
exhibits clear trends, and the maximum values are lower than those of the Kalkaska. The
weak Rubicon pedon has trends that are not as clear as the ﬁrst two pedons, and the
maximum values are lower as well. The Grayling pedon does not exhibit an E horizon,
indicating it is weakly developed, and so only the A and Bs horizons could be analyzed.
These data do, however, show a downward movement of Fe and A1. The extraction data

do correlate to the strength of soil development order.

123

 

REFERENCES

124

REFERENCES

Allen, BL. and BF. Hajek. 1989. Mineral occurrence in soil environments. In: J .B.
Dixon and SB. Weed (eds), Minerals in Soil Environments, 2nd ed. SSSA Book
Series 12199-264.

Anderson, H.A., Berrow, M.L., Farmer, V.C., Hepburn, A., Russell, J .D., and AD.
Walker. 1982. A reassessment of podzol formation processes. J. Soil Sci. 33: 125-
136.

Andrews, J.T. and G.H. Miller. 1980. Dating Quaternary deposits more than 10,000
years old. In: R.A. Cullingford, D.A. Davidson, and J. Lewin (eds), Timescales in
Geomorphology. John Wiley & Sons Ltd., New York, New York: 263-287.

Anonymous. 1989. Methods for determination of inorganic substances in water and
ﬂuvial sediments. In: F ishman, M.J., and LC. Friedman (eds.), Techniques of
water-resources investigations of the United States Geological Survey. USDI
Laboratory Analysis 5, Chapter A1, US Govt. Printing Ofﬁce, Washington, DC.
595p.

Anonymous. 1996. Glossary of soil science terms. Soil Sci. Soc. Am., Madison, WI.
134p.

Antweiler, RC. and J .I. Drever. 1983. The weathering of a late Tertiary volcanic ash:
Importance of organic solutes. Geochim. et Cosmochim. Acta 47:623-629.

Barrett, LR. 1995. A stump prairie landscape in northern Michigan: Soils, forest
vegetation, logging, and ﬁre. Ph.D. dissertation. Michigan State University, East
Lansing, Michigan. 206p.

Barrett, LR. 1997. Podzolization under forest and stump prairie vegetation in northern
Michigan. Geoderma. 78:37-58.

Bateman, RM. and J.A. Catt. 1985. Modiﬁcation of heavy mineral assemblages in
English coversands by acid pedochemical weathering. Catena. 12:1-21.

Bemer, RA. 1978. Rate control of mineral dissolution under earth surface conditions.
Am. J. Sci. 278:1235—1252.

Bemer, R.A., Sjoberg, E.L., Velbel, M.A., and MD. Korm. 1980. Dissolution of
pyroxenes and arnphiboles during weathering. Science 207: 1205-1206.

Bemer, RA. and J. Schott. 1982. Mechanism of pyroxene and amphibole weathering.
II. Observations of soil gains. Am. J. Sci. 282:1214-1231.

125

Birkeland, P.W. 1974. Pedology, weathering, and geomorphological research. Oxford
University Press, Inc., New York, New York, 285p.

Birkeland, P.W. 1978. Soil development as an indication of relative age of Quaternary
deposits, Bafﬁn Island, N.W.T., Canada. Arct. Alp. Res. 10:733-747.

Birkeland, P.W. 1999. Soils and geomorphology, 3rd ed. Oxford University Press, New
York, New York, 430p.

Blackburn, W.H. and W.H. Dennen. 1988. Principles of Mineralogy. Wm. C. Brown
Publishers, Dubuque, Iowa. 413p.

Bland, W. and D. Rolls. 1998. Weathering: An introduction to the scientiﬁc principles.
Oxford University Press Inc., New York, New York, 271p.

Blewett, W. 1990. The glacial geomorphology of the Port Huron complex in
northwestern southern Michigan. Ph.D. dissertation, Department of Geogaphy,
Michigan State University, East Lansing, MI.

Blewett, W., Winters, HA, and R.L. Rieck. 1993. New age control on the Port Huron
moraine in northern Michigan. Phys. Geog. 14:131-138.

Bloom, RR. 2000. Soil pH and pH buffering. In: Summer, M.E., (ed.) Handbook of
Soil Science. CRC Press, Boca Raton, Florida. pp. B-333-B-352.

Blume, HP. and U. Schwertmann. 1969. Genetic evaluation of proﬁle distribution of
aluminum, iron, and manganese oxides. Soil Sci. Soc. Am. Proc. 33:438-444.

Bockheim, J.G., Marshal, J.G., and HM. Kelsey. 1996. Soil-forming processes and rates
on uplifted marine terraces in southwestern Oregon, USA. Geoderma. 73:39-62.

Brantley, S.L., Blai, A.C., Cremeens, D.L., MacInnis, I., and R.G. Darrnody. 1993.
Natural etching rates of feldspar and hornblende. Aquatic Sciences 55(4):262-
272.

Brantley, S.L. and Y.Chen. 1995. Chemical weathering rates of pyroxenes and
amphiboles. In: A.F. White and S.L. Brantley (eds), Chemical Weathering Rates
of Silicate Minerals. Min. Soc. Am., Rev. Mineral. 31:119-172.

Buurman, P. and LP. van Reeuwijk. 1984. Proto-imogolite and the process of podzol
formation: A critical note. J. Soil Sci. 35:447-452.

Chadwick, DA. and RC. Graham. 2000. Pedogenic processes. In: Summer, M.E., (ed.),
Handbook of Soil Science. CRC Press, Boca Raton, Florida. pp. E-41-E-75.

126

Churchman, G]. 2000. The alteration and formation of soil minerals by
weathering. In: Summer, M.E., (ed.), Handbook of Soil Science. CRC Press, Boca
Raton, Florida. pp. F -3-F -76.

Clayton, J .L. 1979. Nutrient supply to soil by rock weathering. p. 75-96. In: Impact of
Intensive Harvesting on Forest Nutrient Cycling, State University of New York,
College of Environmental Science and Forestry, Syracuse, New York.

Cleaves, ET. 1974. Petrologic and chemical investigation of chemical weathering in
maﬁc rocks, eastern piedmont of Maryland: Maryland Geological Survey
Report of Investigations No. 25, 28p.

Cosca, M.A., Essene, E.J., and JR. Bowman. 1991. Complete chemical analyses of
metamorphic homblendes: Implications for normalizations, calculated HzO
activities, and therrnobarometry. Cotnrib. Mineral. Petrol. 108:472-484.

Cremeens, D.L., Darmody, R.G., and L.D. Norton. 1992. Etch-pit size and shape
distribution on orthoclase and pyriboles in a loess catena. Geochim. et
Cosmochim. 56:3423-3434.

Cureton, J .S., Van der Pluijm, BA, and E.J. Essene. 1997. Nature of the Elzevir-
Mazinaw domain boundary, Grenville Orogen, Ontario. Can. J. Earth Sci. 34:976-
991.

Deb, B.C. 1950. The estimations of free iron oxide in soils and clays and their removal. J.
Soil Sci. 1:212-220.

DeConinck, F. 1980. Major mechanisms in formation of spodic horizons. Geoderma
24:101-128.

Delvigne, J .E. 1998. Atlas of micromorphology of mineral alteration and weathering.
Mineral. Ass. Can, Ottawa, Ontario. 494p.

Dorronsoro, C. and P. Alonso. 1994. Chronosequence in Almar River ﬂuvial-terrace soil.
Soil Sci. Soc. Am. J. 58:910-925.

Dove, RM. 1995. Kinetic and thermodynamic controls on silica reactivity in weathering
environments. In: A.F. White, and S.L. Brantley (eds.), Chemical Weathering
Rates of Silicate Minerals. Min. Soc. Am., Rev. Mineral. 31:236-290.

Dreimanis, G.H., Reavely, G.H., Cook, R.J.B., Knox, KS, and F.J. Moretti. 1957. Heavy
mineral studies in till of Ontario and adjacent areas. J. Sed. Pet. 27(2): 148-161.

Dryden, L. and C. Dryden. 1946. Comparative rates of weathering of some heavy
minerals. J. Sed. Pet. 16:91-96.

127

Dworkin, S.I., Larson, G.J., and G.W. Monaghan. 1985. Late Wisconsinan ice-ﬂow
reconstruction for the central Great Lakes region. Can. J. Earth Sci. 22:935-940.

Eggleton, RA. 1986. The relation between crystal structure and silicate weathering rates.
In: S.N. Colman, and DP. Dethier (eds.), Rates of Chemical Weathering of Rocks
and Minerals. Academic Press, Inc., Orlando, Floridaz21-40.

Everett, KR. 1971. Soils of the Meserve Glacier area, Wright Valley, South Victoria
Land, Antarctica. Soil Sci. 112:425-438.

Fanning, D.S. and M.C.B. Farming. 1989. Soil morphology, genesis, and classiﬁcation.
Wiley. New York, New York. 395p.

Farmer, V.C., Russell, J .D., and ML. Berrow. 1980. Irnogolite and proto-imogolite
allophane in spodic horizons: Evidence for a mobile aluminum silicate complex in
podzol formation. J. Soil Sci. 31:673-684.

Farmer, V.C. 1982. Signiﬁcance of the presence of allophane and imogolite in podzol Bs
horizons for podzolization mechanisms: A review. Soil Sci. Plant Nutr. 28:571-
578.

Farmer, V.C., Russell, J .D., and B.F.L. Smith. 1983. Extraction of inorganic forms of
translocated A1, Fe and Si from a podzol Bs horizon. J. Soil Sci. 34:571-576.

Gardner, DR. and ER Whiteside. 1952. Zonal soils in the transition region between the
Podzol and Gray-Brown Podzolic regions in Michigan. Soil Sci. Soc. Am. Proc.
16:137-144.

Goldberg, S., Lebron, I., and D.L. Suarez. 2000. Soil colloidal behavior. In: Summer,
M.E., (ed.) Handbook of Soil Science. CRC Press, Boca Raton, Florida. pp.
B-195-B-239.

Goldich, SS. 1938. A study in rock-weathering. J. Geo. 46:17-58.

Gustafsson, J .P., Bhattacharya, P., Bain, D.C., Fraser, AR, and W.J. McHardy. 1995.
Podzolisation mechanisms and the synthesis of imogolite in northern Scandinavia.
Geoderma. 66:167-184.

Haile-Mariam, S. and D.L. Mokma. 1995. Mineralogy of two sandy Spodosol
hydrosequences in Michigan. Soil Survey Horizons, Winter2121-l32.

Haile-Mariam, S. and D.L. Mokma. 1996. Mineralogy of two ﬁne-loarny
hydrosequences in south-central Michigan. Soil Survey Horizons, Summerz65-74.

Haldane, AD. 1956. Determination of free iron oxide in soils. Soil Sci. 82:483-489.

128

 

Hall, RD. and J .S. Heiny. 1983. Glacial and postglacial physical stratigaphy and
chronology, North Willow Creek and Cataract Creek drainage basins, eastern
Tabacco Root Range, southwestern Montana, U.S.A. Arct. Alpine Res. 15(1):19
-52.

Hall, RD. and RE. Martin. 1986. The etching of hornblende gains in the matrix of
alpine tills and periglacial deposits. In: S.M. Colman, and DP. Dethier (eds.),
Rates of Chemical Weathering of Rocks and Minerals. Academic Press, Inc.,
Orlando, F lorida:101-128.

Hall, RD. and D. Michaud. 1988. The use of hornblende etching, clast weathering, and
soils to date alpine glacial and periglacial deposits: A study from southwestern
Montana. Geo. Soc. Am. Bul. 100:458-467.

Hall, RD. and LL. Horn. 1993. Rates of hornblende etching in soils in glacial deposits
of the northern Rocky Mountains (Wyoming-Montana, U.S.A.): Inﬂuence of

climate and characteristics of the parent material. Chem. Geo. 105:17-29.

Hoffman, SJ. and WK. Fletcher. 1981. Organic matter scavenging of copper, zinc,

molybdenum, iron and manganese, estimated by a sodium hypochlorite extraction

(pH 9.5). J. Geochem. Explor. 15:549-562.

Huang, RM. 1989. Feldspars, olivines, pyroxenes, and amphiboles. In: J.B. Dixon, and
SB. Weeds (eds.), Minerals in Soil Environments. SSSA Book Series 1:975-
1050.

Huang, P.M. 1991. Ionic factors affecting the formation of short-ranged aluminosilicates.

Soil Sci. Soc. Am. J. 55:1172-1180.

Isard, SA. and R.J. Schaetzl. 1998. Effects of winter weather conditions on soil freezing
in southern Michigan. Phys. Geog. 19:71-94.

Jackson, M.L., Lim, CH, and L.W. Zelazny. 1986. Oxides, hydroxides, and
aluminosilicates. p.101-150. In: A. Klute (ed.), Methods of Soil Analysis. Part 1.
Physical and Mineralogical Methods. SSSA Book Series No. 9(1), SSSA and
ASA, Madison, Wisconsin.

J akobsen, B.H. 1991. Multiple processes in the formation of subarctic podzols in
Greenland. Soil Sci. 152:414-426.

Jenny, H. 1941. Factors of soil formation. McGraw-Hill, New York, New York, 281p.

Johnson, LE. and CH. Chu. 1983. Mineralogical characterization of selected soils from
northeastern United States. Pennsylvania Ago. Expr. Sta. Bull. 847p.

129

1.0.1. '. ”.4 It?!
i.

Klein, C. and CS. Hurlbut, Jr. 1999. Manual of mineralogy, 21St ed. John Wiley & Sons,
Inc., New York, New York. 681p.

Locke, W.W. III. 1979. Etching of hornblende grains in Arctic soils: An indicator of
relative age and paleoclimate. Quat. Res. 11(2):197-212.

Locke, W.W. III. 1986. Rates of hornblende etching in soils on glacial deposits, Bafﬁn
Island, Canada. In: S.M. Colman, and DP. Dethier (eds), Rates of Chemical
Weathering of Rocks and Minerals. Academic Press, Orlando, Florida: 129-145.

Loeppert, RH. and WP. Inskeep. 1996. Iron. p. 639-663. In: Bartels, J .M. (ed.),
Methods of Soil Analysis. Part 3. Chemical Methods. SSSA Book Series No.
5(3), SSSA and ASA, Madison, Wisconsin.

MacInnis, IN. and S.L. Brantley. 1993. Development of etch pit size distributions on
dissolving minerals. Chem. Geo. 105231-49.

Markos, G. 1975. Rates and kinetics of plagioclase dissolution in late Quaternary tills in
Colorado and Wyoming. p. 1190. Geo. Soc. Am. Abstr. with Pro. 7(7). Annual
Meeting.

Mast, MA. and J .I. Drever. 1986. Effect of organic chelates on the dissolution rate of
oligoclase. p. 685. Geo. Soc. Am. Abstr. with Prog. 99‘h Annual Meeting
and Expo.

McKeague, J .A. and J .H. Day. 1965. Dithionite- and oxalate-extractable Fe and Al as
aids in differentiating various classes of soils. Can. J. Soil Sci. 46:13-22.

McKeague, J .A., Brydon, J .E., and NM. Miles. 1971. Differentiation of forms of
extractable iron and aluminum in soils. Soil Sci. Soc. Am. Proc. 35:33-38.

Mehra, OR and ML. Jackson. 1960. Iron oxide removal from soils and clays by a
dithionite-citrate system buffered with sodium bicarbonate. In: E.
Ingerson (ed.), Clays and Clay Minerals. Monogaphy No. 5. Earth Science
Series. Pergamon Press, New York, New York. p. 317-327.

Messenger, A.S., Whiteside, ER, and AR. Wolcott. 1972. Climate, time, and organisms
in relation to Podzol development in Michigan sands: I. Site descriptions and
microbiological observations. Soil Sci. Soc. Am. Proc. 36:633-638.

Mokma, D.L. and CV. Evans. 2000. Spodosols. In: Summer, M.E., (ed.) Handbook of
Soil Science. CRC Press, Boca Raton, Florida. pp. E-307-E-321.

Mokma, D.L. and GP. Vance. 1989. Forest vegetation and origin of some spodic
horizons, Michigan. Geoderma 43:311-324. Not seen by this author.

130

Morton, AC. 1984. Stability of detrital heavy minerals in Tertiary sandstones from the
North Sea Basin. Clay Min. 19:287-308.

Muhs, DR. 1982. A soil chronosequence on Quaternary marine terraces, San Clemente
Island, California. Geoderma. 28:257-283.

Murakami, T., Kogure, T., Kadohara, H., and T. Ohnuki. 1998. Formation of secondary
minerals and its effect on anorthite dissolution. Am. Mineral. 83:1209-1219.

Nickel, E. 1973. Experimental dissolution of light and heavy minerals in comparison
with weathering and intrastratal dissolution. Contributions to Sed. 1:1-68.

Nesbitt, H.W., Fedo, C.M., and GM. Young. 1997. Quartz and feldspar stability, steady
and non-steady-state weathering and petrogenesis of silicate sands and munds. J.
Geol.105:173-191.

Nebsitt, H.W. and G. Markovics. 1997. Weathering of ganodioritic crust, long-tenn
storage of elements in weathering proﬁles, and petrogenesis of siliciclastic
sediments. Geochim. et Cosmochim. Acta 61 :1653-1670.

Nugent, M.A., Brantley, S.L., Pantano, CG, and RA. Maurice. 1998. The inﬂuence of
natural mineral coatings on feldspar weathering. Nature 395:588-590.

Ollier, C. and C. Pain. 1996. Regolith, soils and landforms. John Wiley & Sons, Ltd.,
West Sussex, England, 316p.

Osei, BA. 1992. Sand mineralogy and related properties of some soils in south-westem
Nigeria. Commun. Soil Sci. Plant Anal. 23:1815-1826.

Parﬁtt, R.L. and CW. Childs. 1988. Estimation of forms of Fe and A1: A review, and
analysis of contrasting soils by dissolution and Moessbauer methods. Aust. J. Soil
Res. 26:121-144.

Pettijohn, F.J. 1941. Persistence of heavy minerals and geologic age. J. Geo. 49:610-625.

Pettijohn, F .J ., P.E. Potter, and R. Siever. 1973. Sand and Sandstone. Springer-Verlag,
New York, New York, 618p.

Price, T.W., Blevins, R.L., Bamhisel, R.I., and H.H. Bailey. 1975. Lithologic
discontinuities in loessial soils of southwestern Kentucky. Soil Sci. Soc. Am.
Proc. 39:94-98.

Radcliffe, DE. and TC. Rasmussen. 2000. Soil water movement. In: Summer, M.E.,

(ed.), Handbook of Soil Science. CRC Press, Boca Raton, Florida. pp. A-87-A-
127.

131

Rai, D. and J .A. Kittrick. 1989. Mineral equilibria and the soil system. In: J .B. Dixon
and SB. Weed (eds.), Minerals in Soil Environments, 2nd ed. SSSA Book Series
1:161-198.

Rice, Jr., T.J., Buol, S.W., and SB. Weed. 1985. Soil-saprolite proﬁles derived from
maﬁc rocks in the North Carolina piedmont: 1. Chemical, morphological, and
mineralogical characteristics and transformations. Soil Sci. Soc. Am. J. 49:171-
178.

Ross, G.J. and C. Wang. 1993. Extractable Al, Fe, Mn, and Si. In: Carter, M.R. (ed.),
Soil Sampling and Methods of Analysis. Can. Soc. Soil. Sci., Boca Raton,
Florida. p. 239-246.

Ruhe, R.V. 1956. Geomorphic surfaces and the nature of soils. Soil Sci. 82(6):441-455.

Ruhe, R.V., R.B. Daniels, and J .G. Cady. 1966. Landscape evolution and soil formation
in southwestern Iowa. US. Dept. Agr. Tech. Bul. 1349. 242p.

Schaetzl, R.J. and D.L. Mokma. 1988. A numerical index of podzol and podzolic soil
development. Phys. Geog. 9:232-246.

Schaetzl, R.J. and SA. Isard. 1991. The distribution of Spodosol soils in southern
Michigan: A climatic interpretation. Annals Assoc. Am. Geogs. 81 :425-442.

Schaetzl, R.J. 1992. Beta spodic horizons in podzolic soils (Lithic Haplorthods and
Haplohumods). Pedologie. 42:271-287.

Schaetzl, R.J. 1994. Changes in O horizon mass, thickness and carbon content following
ﬁre in northern hardwood stands. Vegetatio 115:41-50.

Schaetzl, R.J. and SA. Isard. 1996. Regional-scale relationships between climate and
strength of podzolization in the Great Lakes region, North America. Catena
28:47-69.

Schaetzl, R.J. 2002. A spodosol-entisol transition in northern Michigan: Climate or
vegetation? Soil Sci. Soc. Am. J. 66(4):1272-1284.

Shetron, S.G. 1974. Distribution of free iron and organic carbon as related to available
water in some forested sandy soils. Soil Sci. Soc. Am. Proc. 38:359-362.

Soil Survey Division Staff. 1993. Soil survey manual. USDA Handbook No. 18. US
Govt. Printing Ofﬁce, Washington, DC. 437p.

Soil Survey Laboratory Staff. 1996. Soil survey laboratory methods manual. USDA-SCS,
National Soil Survey Center. Soil Survey Investigations Report. 42, Version 3.0,
National Soil Survey Center, Lincoln, Nebraska. 693p.

 

132

Soil Survey Staff. 1998. Keys to soil taxonomy, 8th ed. Natural Resources Conservation
Services, US Dept. Agri. 326p.

Soil Survey Staff. 1999. Soil taxonomy: A basic system of soil classiﬁcation for making
and interpreting soil surveys, 2nd ed. USDA Agic. Handbook 436. US Govt.
Printing Ofﬁce, Washington, DC. 870p.

Stahnke, CR. 1965. A study of mineral transformations and weathering processes
occurring during the genesis of two soils developed from gneiss and schist in
Llano Co., Texas. M.S. thesis. Texas Tech University, Lubbock. TX.

Stephen, 1. 1952. A study of rock weathering with reference to the soils of the Malvem
Hills. J. Soil Sci. 3:20-33.

Sverdrup, H.U. 1990. The kinetics of base cation release due to chemical weathering.
Lund University Press.

Truog, E., Taylor, J .R., Jr., Pearson, R.W., Weeks, M.E., and R.W. Simonson. 1937.
Procedure for special type of mechanical and mineralogical soil analysis. Soil Sci.
Soc. Am. Proc. 1:101-112.

Ugolini, RC. and C. Bull. 1965. Soil development and glacial events in Antarctica.
Quatemaria 7:251-269.

Ugolini, F .C., Minden, R., Dawson, H., and J. Zachara. 1977. An example of soil
processes in the Abies Amabilis zone of central Cascades, Washington. Soil Sci.
124(5):291-302.

Van der Plas, L. and AC. Tobi. 1965 . A chart for judging the reliability of point
counting results. Am. Jour. Sci. 263287-90.

Vance, G.F., Mokma, D.L., and SA. Boyd. 1986. Phenolic compounds in soils of
hydrosequences and developmental sequences of Spodosols. Soil Sci. Soc. Am. J.
50:992-996.

Velbel, MA. 1987. Rate-controlling factors in the weathering of some ferromagnesian
silicate minerals. Transactions, 13th Congess of the International Society of Soil
Science, Hamburg, Germany 621107-1118.

Velbel, MA. 1989. Weathering of hornblende to ferruginous products by a dissolution-
reprecipitation mechanism: Petrogaphy and stoichiometry. Clays and Clay Min.
37:515-524.

Velbel, M.A. 1993a. Constancy of silicate-mineral weathering-rate ratios between natural

and experimental weathering: implications for hydrologic control of differences in
absolute rate. Chem. Geol. 105289-99.

133

Velbel, M.A. 1993b. Formation of protective surface layers during silicate-mineral
weathering under well-leached, oxidizing conditions. Am. Mineralogist 78:405-
414.

Velbel, MA. 1999. Bond strength and the relative weathering rates of simple
orthosilicates. Am. J. Sci. 299:679-696.

Walker, T.R., Ribbe, RH, and RM. Honea. 1967. Geochemistry of hornblende alteration
in Pliocene red beds, Baja California, Mexico. Geo. Soc. Am. Bul. 78:1055-1060.

Wang, C. and H. Kodama. 1986. Pedogenic imogolite in sandy Brunisols of eastern
Ontario. Can. J. Soil Sci. 66:135-142.

Wang, C., McKeague, J .A., and H. Kodama. 1986. Pedogenic imogolite and soil
environments: Case study of Spodosols in Quebec, Canada. Soil Sci. Soc. Am. J.
50:711-718.

Weisenborn, B.N. 2001. A model for fragipan evolution in Michigan soils. M.S. thesis.
Michigan State University, East Lansing, Michigan.

White, AF. and S.L. Brantley (eds). 1995. Reviews in Mineralogy, Volume 31:
Chemical weathering rates of silicate minerals. Min. Soc. Am. Washington, DC.
583p.

White, A.F., Blum, A.E., Schulz, M.S., Bullen, T.D., Harden, J ., and ML. Peterson.
1996. Chemical weathering rates of a soil chronosequence on ganitic alluvium: I.
Quantiﬁcation of mineralogical and surface area changes and calculation of
primary silicate reaction rates. Geochim. et Cosmochim. Acta 60:2533-2550.

Wilding, L.P., Smeck, NE, and LR. Drees. 1977. Silica in soils: Quartz, cristobalite,

tridymite, and opal. In: Dinauer, R.C. (ed.), Minerals in Soil Environments. Soil
Sci. Soc. Am., Madison, Wisconsin: 471-540.

134